Experimental study of polymer injection enhanced oil recovery in homogeneous and heterogeneous porous media using glass-type micromodels

Experimental study of polymer injection enhanced oil recovery in homogeneous and heterogeneous... In water flooding process, volumetric sweep efficiency and oil recovery can be enhanced using polymer to increase the viscosity of water. As a result, polymer flooding has higher recovery as compared to water flooding due to front stability and reduction of fingering problem. In this research work, a set of polymer flooding runs were carried out using glass-type micromodels. The micromodels were fabricated to have homogeneous and heterogeneous flow patterns. They were positioned horizontally and saturated with a heavy crude oil sample taken from an Iranian oil field before starting the injection. Three commercial polymers were used in this study. Whole process was photographed continuously with a high-resolution camera to monitor the displacement of polymer solution in the micromodels. As a part of this study, the effect of different param- eters including polymer solution concentration, injection flow rate and heterogeneity on performance of polymer flooding was investigated. On top of the regular homogeneous and heterogeneous flow patterns used in this study, a heterogeneous flow pattern mimicking sandstone reservoirs was created based on the image of a thin section of a sandstone (outcrop) and polymer front movement was observed during injection. Keywords Polymer injection · Glass-type micromodel · Sweep efficiency Introduction adsorb on rock surface (Sheng 2013; Sheng et  al. 2015; Buchgraber 2008; Barati 2011). Polymer is a large molecule built up by the addition of small When polymer is added to water (brine), its viscosity repeating units (monomers) (Allcock et al. 2003). In general, will be increased. Therefore, polymer injection can lead to the polymers used in the EOR process are divided into two changes in fractional flow and mobility ratio and also has groups of synthetic polymer and biopolymer. The unique fluid diversion effect. As a result, polymer can help with features of each category of these polymers have caused reduction of viscous fingering and improving of water injec- advantage and disadvantage. Synthetic polymers have tion profile that will lead to an improvement in sweep effi- affordable prices, appropriate viscosity in fresh water and ciency (Buchgraber 2008; Barati 2011; Sorbie 1991; Need- acceptable adsorption on the rock surface. The disadvantage ham and Doe 1987; Chang 1978). Nevertheless, stability of of this type of polymer can be attributed to the sensitivity polymer is an issue. Many researchers have focused on the to flow rate and shear degradation. This type of polymer parameters that affect the stability of polymers in different has low efficiency in high-salinity water. Biopolymers show conditions. The factors of chemical, mechanical and biologi- excellent performance against high-salinity water and shear cal degradation cause instability of the polymers. Oxidation degradation but they can be sensitive to bacterial degrada- and ferric ions are the factors affecting the chemical degra- tion in low-temperature reservoirs. Biopolymer does not dation. Reducing the amount of oxygen in the water in con- 3+ 2+ tact with the polymer prevents the generation of Fe, Fe 2− and free radical O ions making the polymer solution to be * Jalal Foroozesh more stable and prevents chemical degradation. Mechani- jalal.foroozesh@gmail.com; jalal.foroozesh@utp.edu.my cal degradation occurs when polymer molecules break 1 down due to shear stress. Flow rate has a significant impact Department of Petroleum Engineering, Universiti Teknologi on preventing this phenomenon and should be optimized. PETRONAS, Seri Iskandar, Malaysia Vol.:(0123456789) 1 3 Journal of Petroleum Exploration and Production Technology Biological degradation is more likely to occur in case of used. Glass micromodels are quite useful tools to study the biopolymers, but under a few conditions; synthetic polymers active recovery mechanisms during polymer injection and can also be degraded biologically. Using some additives the effective parameters in oil recovery (Danesh et al. 1987). can solve these problems (Sheng et al. 2015; Sorbie 1991). Due to the transparency, the whole process can be directly Polymer injection in reservoirs will face some limitations. observed and photographed and by using a digital image Reservoir water salinity, reservoir temperature, type of for- analysis (DIA), the photos that are taken during the process, mation and reservoir permeability are factors influencing the can be analyzed for any pore level mechanism and also to performance of polymer injection. For example, in the case estimate the recovery factor. of polymer incompatibility with the reservoir fluid salinity The challenges ahead when using micromodels are to cre- and the ions contained therein, the viscosity of the polymer ate a porous media that can represent a real porous media. solution is not sufficiently achieved and its effectiveness Many different pore patterns have been created in micro- is greatly reduced (Sheng et al. 2015; Needham and Doe models, each of which has its advantage and disadvantage. 1987). The high temperature of the reservoir can break down Figure 1 shows some of these micromodels (Danesh et al. the molecular chains of polymers and turns them into mono- 1987; Romero-Zeron and Kantzas 2007; Buchgraber et al. mers. If these phenomena happen, the viscosity of the solu- 2011; Farzaneh et al. 2012; Hamedi Shokrlu and Babadagli tion of water and polymer will be reduced. In most reports, 2015; Howe et al. 2015; Bahari Moghaddam and Rasaei a temperature tolerance of 93 °C for both types of polymer 2015; Manlowe and Radke 1990). (synthetic polymer and biopolymer) is acceptable. It is worth In the laboratory, a lot of experiments have been done on noting that the temperature of 93 °C does not guarantee the the polymer injection using micromodels. Heshmati et al. stability of the polymer over a long period of time (Taber (2007) used a micromodel with two different permeabil- et al. 1997a, b). Most successful polymer injections have ity layers to study the effect of permeability heterogeneity been in sandstone formations. The use of this technology in on the performance of the polymer flooding. After poly - carbonate formations requires further investigation. Hence, mer injection, a 17% increase in recovery was reported at the type of reservoir rock and the mineralogy of the rock is best (Heshmati et al. 2007). Emami et al. (2008) employed important (Jewett and Schurz 1970). The polymer is com- a micromodel with five-spot injection pattern to examine posed of large molecules. Therefore, the use of polymer in the impact of local and global heterogeneity (similar to lay- rocks with small pore–throat size is not recommended. This ered reservoirs) on recovery. The results of the experiments will cause the pore–throat to block. Pore–throat size will showed that the maximum recovery achieved when layers control the rock permeability. Therefore, rocks with low are perpendicular to mean flow direction. Also, the slope of permeability as a result of having small pore size cannot be the micromodel strongly affects the efficiency of the polymer a good candidate for polymer injection (Jewett and Schurz injection process (Emami Meybodi et al. 2008). Hematpur 1970). et al. (2011) studied the effect of polymer injection in the Field practice of polymer flooding is reported in the liter - presence of low-viscosity oil using a micromodel setup. The ature. The use of polymer as an additive for injection into oil results showed that using the polymer of hydrolyzed poly- reservoirs began in 1967 and has quickly become popular. acrylamide (HPAM) had the best performance under test So far, many fields have been under polymer injection. These conditions (Hematpour et al. 2011). Maghzi et al. (2011) include ‘Taber south field’, ‘North Burbank unit’, ‘Brelum used a micromodel setup with five-spot injection pattern unit reservoir’ and ‘Vernon’. In all cases, the polymer injec- to investigate the injection of the nanoparticles with poly- tion has been effective. According to reports, except for a mer. An increase of 10% in recovery was the result of the few cases, the injected polymer type was synthetic. An addi- experiments conducted by these researchers (Maghzi et al. tional average recovery of 7% has been reported. Another 2011). Wegner and Ganzer (2013) compared the results of notable case is the formation type that the polymer has been polymer flooding in a micromodel with numerical simu- injected into it. Most formations under polymer injection lation (Wegner and Ganzer 2013). Yousefvand and Jafari have been sandstone rocks (Clampitt and Reid 1975; Lozan- (2015) investigated the polymer efficiency in the presence ski and Martin 1970; Shaw and Stright 1975; Rowalt 1973). of nanosilica particles. The investigation was done in a Experimental study is an important step in the investiga- micromodel using reservoir oil and brine (Yousefvand and tion of polymer injection. Core flooding is commonly used Jafari 2015). Sedaghat et al. (2015) investigated the perfor- for experimental study. Core is one of the best representa- mance of alkaline surfactant polymer (ASP) injection for tives of the reservoir at the surface, but having native res- heavy oil recovery using fractured five-spot micromodels. ervoir’s core is not easy. Also, due to the lack of visualiza- Hydrolyzed polyacrylamide showed a better performance tion of the core flooding in cases where the flow behavior as compared to other polymers (Sedaghat et al. 2015). Rock is necessary, this porous media cannot be a good candidate. et al. (2017) investigated the behavior of viscoelastic poly- To overcome these problems, visual micromodels have been mers in a porous media using a glass micromodel setup. The 1 3 Journal of Petroleum Exploration and Production Technology Fig. 1 Micromodel pore patterns two factors of salinity and mechanical degradation had the industry as an additive to water-based muds for mainly fluid high impact on the results presented by these researchers. loss control. The results obtained by these researchers showed that there The first step was to fabricate a glass micromodel which is a relationship between shear thickening and elastic flow was a relatively time-consuming procedure. The steps instability (Rock et al. 2017). involved in the fabrication process were first using a piece It is inferred that the performance of polymer flooding of mirror with appropriate dimensions and then removing can be dependent on reservoir type and characteristic and the paint coated on the backside until the mercury coat- micromodel setups can be used to study it. Therefore, in ing was exposed. Then, we covered a layer of plastic lami- this research, a number of micromodel setups with differ - nate on the mercury coating and the pattern in black and ent heterogeneities of pore patterns have been fabricated white printed on a paper, overlaid on the laminate surface. to study the performance of three different polymers in an Ultraviolet (UV) exposure was used to polymerize the non- Iranian oil field. pattern area (resistant to acid) and then by means of nitric acid, the non-polymerized (pattern) area was removed (dis- solved) and then the hydrofluoric acid was used to etch the Experimental procedure and materials pattern area. Cyclic submersion in hydrofluoric acid for a certain time can lead to a certain depth of etching. It should The primary objective of the undertaken series of experi- be mentioned that all the steps were taken under complete ments was to investigate the effect of different parameters caution. A number of micromodel setups with homogene- such as polymer solution concentration, polymer type and ous and heterogeneous flow patterns were built and used for injection flow rate on oil recovery. Hence, three different experimental study of polymer injection. Figure 2 shows a types of commercial polymer were used to prepare the poly- homogenous flow pattern used for micromodel fabrication. mer solution with different concentrations and were injected The heterogeneous micromodels will be explained later on. at different flow rates. The specifications of the polymers The next step was to saturate the micromodel setup with are given in Table 1. These polymers are used in drilling the crude oil without having any connate water saturation. Table 1 The specifications of Polymer Commercial name Provider Polymer family the polymers used in this study type no. ® ® 1 DRISPAC Superlo Chevron Phillips Chemical Company Polyanionic cellulose 2 DRISPAC Regular Chevron Phillips Chemical Company Polyanionic cellulose 3 DRISCAL D Chevron Phillips Chemical Company Polyacrylamide (syn- thetic polymer) 1 3 Journal of Petroleum Exploration and Production Technology Pore volume The depth of etched glass (pores) created in micromodel was measured by Micrometer tool. Therefore, by multiplying the depth by pore area, pore volume was determined as 0.068 cc. A magnified view of the etched glass is shown in Fig.  4. Fig. 2 Homogenous flow pattern Permeability The crude oil used in this study was taken from one of the Iranian oil fields which is located in the southern part Another important property of glass micromodel which of Iran and has an API of 19.8 (relatively heavy). The should be determined was absolute permeability. This could saturation process was done very carefully to avoid any be achieved by gathering the flow rate data and the related penetration of crude oil through the non-porous media, injection pressure. So, the flow rate was set and let the sys- rather than etched porous media. Figure 3 shows a sample tem to reach the constant (stabilized) pressure die ff rence. By of saturated flow pattern with crude oil. knowing the pressure difference (outlet pressure is atmos- After the flow pattern is saturated, the experiment was pheric pressure), the absolute permeability can be calculated started. Quizix pump was used in this study to inject the by Darcy’s law. The permeability of homogenous pattern polymer solution into the micromodel at a very precise rate measured was around 20 Darcy. close to real velocity of fluid in the reservoir. Results and discussion The physical properties of micromodel Homogeneous flow pattern As one of the important stages of the experiment, the prop- erties of the generated glass micromodel were determined. Eec ff t of polymer solution concentration on oil recovery Some of these properties, namely pore volume and poros- ity were determined by means of computer software using Three polymer solutions with different concentrations (1000, digital image analysis (DIA) process. Some other proper- 1500 and 2250 ppm) of polymer type 1 were prepared and ties such as permeability were measured experimentally injected at the rate of 0.0002  cc/min to the oil-saturated by steady-state flow test after stabilization of differential homogenous micromodel to study the effect of polymer con- pressure (ΔP) across the setup. It should be noted that the centration on oil recovery. Figure 5 shows the effect of poly - micromodels used here are water-wet as made by glass mer solution concentration on oil recovery for polymer type 1. and no aging process has been carried out to change their original wettability. Porosity Using DIA process, the areal porosity was determined and with regard to uniform distribution of etching depth, the porosity was calculated and reported as a total porosity. The pattern was saturated with oil and then the picture was taken and analyzed to determine the porosity. Using image analysis, the total porosity estimated was 0.36 (36%). Fig. 4 A magnified view of the etched glass (blue and red colors refer Fig. 3 Sample of homogenous flow pattern saturated with crude oil to water and oil phases, respectively) (dark color is oil; white color is grain) 1 3 Journal of Petroleum Exploration and Production Technology 2250 ppm 1000 ppm 1500 ppm 00.1 0.20.3 0.40.5 0.60.7 Pore Volume Injected(fraction) Fig. 5 Comparison between oil recoveries by three different concentrations of polymer solutions of polymer type 1 (flow rate = 0.0002 cc/min) Table 2 Measured physical properties of water and polymer solutions of 1000 pm and it is 4.15 cp at concentration of 2250 pm at room temperature (T = 24 °C) resulting in a better sweep efficiency. In micromodel setup, it is not possible to investigate the vertical conformance and Material Viscosity (cp) Density (g/cm ) sweep due to the nature of the equipment but the role of poly- Water 0.94 1.01 mer injection and gel treatment in conformance control in oil Polymer type 1 (c = 1000 ppm) 1.44 1.01 reservoirs has been discussed in the literature (Kantzas et al. Polymer type 1 (c = 1500 ppm) 1.98 1.01 1999; Suleimanov and Veliyev 2016). Polymer type 1 (c = 2250 ppm) 4.15 1.02 Polymer type 2 (c = 1500 ppm) 7.14 1.01 Polymer type 3 (c = 1500 ppm) 3.24 1.01 Eec ff t of injection flow rate on oil recovery In these series of experiments, three different flow rates were tested: 0.0002, 0.0005 and 0.0008  cc/min for the As it can be seen in Fig. 5, for 150% increase in poly- mer concentration in solution (from 1500 to 2250 ppm), the polymer type 1 and 0.0002 and 0.0005 cc/min for poly- mer types 2 and 3, all at fixed concentration of 1500 ppm. recovery has increased around 7%. Enhancement of polymer concentration increases solution viscosity resulting in front As it can be observed, in each step, the flow rate has been increased by a factor of 2.5. Figure 7 shows the effect of stability and minimizes fingering and channeling problem, which lead to higher oil recovery. injection flow rate for the polymer type 2. As it can be seen in Fig. 7, at lower flow rate, the polymer As it can be observed in Fig. 5, the lower concentration (1000 ppm) solution had the fastest breakthrough among all displacement in the model is more piston-like while at high injection rate, it is having unstable front with less piston- at 0.54 pore volume (PV) injected, while the other solutions had later breakthrough at 0.58 and 0.62 PV injected for 1500 like displacement. Similar results were observed for polymer types 1 and 3. This is because, at low injection rates, fluid and 2250 ppm, respectively. Lower concentration translates to lower viscosity (which is shown in Table 2) and lower has time to distribute areally to have a more piston-like flow. It should be noted that sensitivity of oil recovery to poly- sweep efficiency (see Fig.  6). It should be mentioned that for all our experiments, once there was no more oil production, mer injection rate during polymer flooding is also affected by low injectivity of polymer solution compared to water the injection was ceased shortly. Figure 6 shows the profile of injected polymer in micro- injection (Van den Hoek et al. 2012). Loss of injectivity especially in low-permeability formations is a serious prob- model for low and high concentrations at three different pore volumes. As it can be seen, at high concentration of 2250 ppm, lem in polymer flooding projects, which can occur because of large molecule size and high viscosity of polymer solu- the polymer displacement is more towards positon-like as compared to the profile of low concentration of 1000 ppm. tions. In real reservoirs, high injection rates can cause the bottom-hole pressure to rise up and if not controlled, This could sweep more area occupied by oil. Table 2 shows that the viscosity of polymer type 1 is 1.44 cp at concentration exceed the fracture pressure and cause formation damage. 1 3 Oil Recovery (%) Journal of Petroleum Exploration and Production Technology Fig. 6 Front stability for low-concentration (1000 ppm) polymer solution and high-con- centration (2250 ppm) polymer solution at pore volumes of 0.2, 0.5 and 1, polymer type 1, injection rate; 0.0002 cc/min, injection point on the left side (brown color: oil, whitish blue color: polymer) Fig. 7 Comparison between flooded pattern at breakthrough time for polymer type 2 for flow rates 0.0002  cc/min (lower side picture) and 0.0005 cc/min (upper side picture) both at concentration = 1500 ppm. Injection point on the left side (brown color: oil, white color: polymer) Comparison between water injection and polymer injection the displacement of water and polymer in the micromodel at three different pore volumes. Figure  8 clearly illustrates Water injection was done to compare its performance with the better areal sweep efficiency and also less residual oil polymer injection. Water flooding is one of the most favora- trapped in the pores and throats resulted from polymer flood- ble secondary recovery methods due to the availability of ing as compared to water injection leading to a higher oil water (especially in offshore operation) and the relatively recovery factor. low cost and complexity compared to the other EOR meth- ods. Nevertheless, the mobility of water needs to be con- Comparison of performance of the three polymer types trolled especially in layered and heterogeneous reservoirs to have a good conformance control. Adding polymer to water Figure 9 compares the displacement of polymer types 1, 2 will increase the viscosity as it was discussed earlier and will and 3 in the micromodel at the same pore volume injected. increase the volumetric sweep efficiency. Figure  8 compares They all have the fixed concentration of 1500 ppm. It can be 1 3 Journal of Petroleum Exploration and Production Technology Fig. 8 Flooded patterns at equal pore volumes injected for water flooding and polymer flooding, Polymer type 1, injection rate = 0.0002 cc/min, concentration = 1500 ppm (brown color: oil, whitish blue color: polymer and blue color: water) Fig. 9 Comparison between displacement of polymer type 1, 2 and 3 at fixed concentration of 1500 ppm (identical pore volume injected) 1 3 Journal of Petroleum Exploration and Production Technology observed that the displacement of polymer type 2 is more side layer has lower permeability compared to the upper piston-like (i.e., stable displacement) as compared to two side layer which has higher permeability. These different other polymers. Next, polymer type 3 is displacing like a permeabilities have been obtained by controlling the size piston. This could be explained by considering the viscos- of the pores and throats in each layer. This is one of the ity of polymer solutions presented in Table 2. As it given in heterogeneities that can be found in layered and segregated Table 2, the solution of polymer type 2 has largest viscosity reservoirs. and after that polymer type 3 has a higher viscosity as com- As it can be seen in Fig. 10, the polymer is injected from pared to polymer type 1. These viscosity values have led to the left and has been mainly advanced in bottom side with a better mobility ratio of polymer type 2 resulting in a more higher permeability. Although the upper layer has lower stable displacement. permeability as compared to lower layer, its area is large causing polymer to move through it too. It should be noted Heterogeneous flow patterns that the permeabilities of micromodels are generally large making the upper layer to also have a large permeability (in Real oil reservoirs are complex set of different heterogenei- order of several Darcys) but smaller than the permeably of ties. But, in the majority of the routine reservoir engineering the lower side layer. That means, the contrast between layers’ calculations, this heterogeneity is not considered. In routine permeability may not cause much difference in flow behavior core-flooding experiments also the effect of heterogeneity through different layers. It is worth mentioning that, due to cannot be considered due to the small scale of core plugs. the horizontal positioning of the micromodel pattern, there Hence, inclusion of heterogeneity in glass micromodels is not any effect of gravity involved. enables us to create different kind of heterogeneity and to visualize the polymer front movement and behavior from Faulted heterogeneous flow pattern the beginning to the end of the experiment. In the following section, the results of polymer flooding in the artificially In this flow pattern, the aim was to simulate the fault effect created heterogeneous patterns will be discussed. It should (faulted layer) on flow condition in real reservoirs dur - be mentioned that type of heterogeneity has been inherited ing polymer injection. There are two layers in this pattern from some real large-scale heterogeneities, which frequently which have different permeability (high and low) which are can be found in subsurface reservoirs. faulted in the middle point of the flow pattern (normal fault) as shown in Fig. 11. As it can be seen, mimicking a faulted Two-layered flow pattern reservoir and due to displacement of zones, high- and low- permeable zones are juxtaposed. This flow pattern includes two layers with different perme- As it can be observed in Fig.  11, when the polymer ability (layers are separated by a red dashed line). The lower front has reached the fault, it was diverted through the Fig. 10 Polymer front advancement in two-layered heterogeneous flow pattern, injection point on the right, upper side picture after 0.2 PV injected, lower side picture after 0.4 PV injected (brown color is oil, yellow color is grain and whitish blue color is polymer solution) 1 3 Journal of Petroleum Exploration and Production Technology high-permeable zone on the top-right side of the figure. It Flow pattern fabricated using a sandstone thin section should be noted that the stream line was not 1-D (in x-direc- tion) anymore and polymer breakthrough will potentially Using the picture of a thin section from an outcropped sand- take longer time comparing to the 1-D flow. In real reser - stone, this flow pattern was fabricated. The primary objec- voirs, faults can change the flow direction and affect the tive was to simulate more real porous media to study fluid volumetric sweep efficiency. flow. It should be noted that this thin section is a small sec- tion of a rock which heterogeneity cannot be captured at this Tilted zone heterogeneous flow pattern scale. However, tortuosity, dead-end pore effect and irreg- ular pore and throat shapes still can be observed to some In this heterogeneous flow pattern, tilted zones have been extent. Figure 13 shows the micromodel with sandstone o fl w placed adjacent to each other with low and high permeabil- pattern that is saturated with oil. ity, alternatively as shown in Fig. 12. Figure  14 shows a magnified image of polymer solu- Polymer front movement in this flow pattern is zig-zag tion–oil interface inside this flow pattern. shape due to the permeability contrast between tilted zones. As it is observed in this, due to the high viscosity of In this flow pattern, when polymer front has reached low- polymer solution comparing to water injection which was permeable zone, it has tried to move forward (1-D) longitu- shown before in Fig. 8, there is less residual oil trapped dinally while in high-permeable zone, the movement is both in the corners of pores and specifically in throats lead- longitudinally and transverse (2-D). ing to a better displacement efficiency with minimum Fig. 11 Polymer front movement in quadratic flow pattern, polymer is injected from the left (brown color is oil, light yellow color is grain and blue color is polymer solution) Fig. 12 Polymer front movement in tilted-stripped layered flow pattern (high- and low-permeable layers alternate each other; injection from right side; blue dashed lines are zone borders; L low-permeable zone, H high-permeable zone) Fig. 13 Oil–saturated sandstone flow pattern (grains in white color) 1 3 Journal of Petroleum Exploration and Production Technology Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. References Allcock HR, Lampe FW, Mark JE (2003) Contemporary polymer chemistry, Third edn, Pearson Education, New York Bahari Moghaddam M, Rasaei MR (2015) Experimental study of the fracture and matrix effects on free-fall gravity drainage with micromodels. SPE J 20:Issue 02 Fig. 14 Oil–polymer solution interface in sandstone flow pattern dur - Barati N (2011) Study of polymer flooding performance in the pres- ing polymer flooding (polymer solution in light blue and oil in brown ence of nanoparticles. Master Thesis, Petroleum University of colors) Technology Buchgraber M (2008) An enhanced oil recovery micromodel study with associative and conventional polymers. Diploma Thesis, University of Leoben, Austria oil-bypassing during the injection. This figure shows the Buchgraber M, Clemens T, Castanier LM, Kovscek A (2011) A efficiency of polymer flooding in systems with more real- microvisual study of the displacement of viscous oil by polymer istic flow patterns. solutions. SPE Reserv Eval Eng 14:Issue 03 Chang HL (1978) Polymer flooding technology yesterday, today, and tomorrow. J Petrol Technol 30:Issue 08 Clampitt RL, Reid TB (1975) An economic polymer flood in the North Burbank Unit, Osage County, Oklahoma. Fall meeting of Summary and conclusions the Society of Petroleum Engineers of AIME, 28 September–1 October, Dallas Danesh A, Peden JM, Krinis D, Henderson GD (1987) Pore level A set of polymer flooding experiments was performed in visual investigation of oil recovery by solution gas drive and manually fabricated glass homogeneous and heterogene- gas injection. SPE annual technical conference and exhibition, ous micromodels in horizontal positioning having 1-D or 27–30 September, Dallas, Texas Emami Meybodi H, Kharrat R, Ghazanfari MH (2008) Effect of het- 2-D flow direction. The micromodel setup was saturated erogeneity of layered reservoirs on polymer flooding: an exper - with a relatively heavy Iranian crude oil. The oil recov- imental approach using 5—spot glass micromodel. Europec/ ery results were compared with oil recovery from water EAGE conference and exhibition, 9–12 June 2008, Rome, Italy flooding under the same initial conditions. Three different Farzaneh SA, Dehghan AA, Ghazanfari MH, Kharrat R (2012) A comparative study on WAS, SWAS, and solvent-soak scenarios polymer types were used during this experimental study. applied to heavy-oil reservoirs using five-spot glass micromod- The polymer solution concentration and injection rate els. J Can Pet Technol 51:Issue 05 were changed to account for their effects on oil recovery. Hamedi Shokrlu Y, Babadagli T (2015) Pore-scale investigation of The results showed a better performance for polymer type phase distribution and residual-oil development during second- ary and tertiary solvent injection. SPE Reserv Eval Eng 18:Issue 2. It was also observed that at higher polymer concentra- tions and lower injection rates, a better sweep efficiency Hematpour H, Mardi M, Edalatkhah S, Arabjamaloe R (2011) can be achieved by the polymer. Visualization of polymer Experimental study of polymer flooding in low-viscosity oil front inside the micromodel showed a more piston-like using one-quarter five-spot glass micromodel. Pet Sci Technol 29:1163–1175 movement of solution as compared to the front in pure Heshmati M, Mahdavi H, Haghighi M, Torabi M (2007) Visualiza- water injection. Heterogeneity, as another studied param- tion of polymer flooding in heterogeneous system using glass eter in this research, had an important role in oil recovery. micromodels. J Sci 332:25–32 Several artificially made heterogeneities were introduced Howe AM, Clarke A, Mitchell J, Staniland J, Hawkes LA (2015) Visualising surfactant EOR in core plugs and micromodels. SPE into glass micromodels aiming to simulate the real het- Asia Pacific enhanced oil recovery conference, 11–13 August, erogeneity in the reservoirs. As a result, swept area was Kuala Lumpur, Malaysia significantly affected by heterogeneity since it affects the Jewett RL, Schurz GF (1970). Polymer flooding—a current appraisal. front movement. Therefore, macroscopic heterogeneities JPT 22:675–684 Kantzas A, Burger D, Pow M, Cheung V, Mourits F (1999) Design such as layered or faulted reservoirs and also existence strategies for improved conformance using polymer gels. Pet of any high- or low-permeable zones must be taken into Soc Can. https ://doi.org/10.2118/99-13-18 account before any polymer injection project to achieve Lozanski WR, Martin P (1970) Taber South—Canada’s first polymer the best EOR performance. flood. J Cdn Pet Tech 1970:99 1 3 Journal of Petroleum Exploration and Production Technology Maghzi A, Mohebbi A, Kharrat R, Ghazanfari MH (2011) Pore- Sheng JJ, Leonhardt B, Azri N (2015) Status of polymer-flooding tech- scale monitoring of wettability alteration by silica nanoparticles nology. J Can Pet Technol 54:Issue 02 during polymer flooding to heavy oil in a five-spot glass micro- Sorbie KS (1991) Polymer-improved oil recovery, 1st ed. CRC Press, model. Transp Porous Med 87:653–664 Inc, Boca Raton Manlowe DJ, Radke CJ (1990) A pore-level investigation of foam/oil Suleimanov BA, Veliyev EF (2016) Nanogels for deep reservoir con- interactions in porous media. SPE Reserv Eng 5:Issue 4 formance control. Soc Pet Eng. https ://doi.org/10.2118/18253 Needham RB, Doe PH (1987) Polymer flooding review. J Pet Technol 4-MS Rock A, Hincapie RE, Wegner J, Ganzer L (2017) Advanced flow Taber JJ, Martin FD, Seright RS (1997a) EOR screening criteria revis- behavior characterization of enhanced oil recovery polymers ited—part 2: applications and impact of oil prices. SPE Reserv using glass–silicon–glass micromodels that resemble porous Eng 12:Issue 03 media. SPE Europec featured at 79th EAGE conference and Taber JJ, Martin FD, Seright RS (1997b) EOR screening criteria exhibition, 12–15 June, Paris, France revisited—part 1: introduction to screening criteria and enhanced Romero-Zeron LB, Kantzas A (2007) The ee ff ct of wettability and pore recovery field projects. SPE Reserv Eng 12:Issue 03 geometry on foamed-gel-blockage performance. SPE Reserv Eval Van den Hoek P, Mahani H, Sorop T, Brooks D, Zwaan M, Sen S, Eng 10:Issue 02 Saadi F (2012) Application of injection fall-off analysis in poly - Rowalt RJ (1973) A case history of polymer water flooding—Bre- mer flooding. Soc Pet Eng. https ://doi.org/10.2118/15437 6-MS lum field unit. Paper SPE 4671 presented at the SPE-AIME 48th Wegner J, Ganzer L (2013) Numerical analysis of polymer micro- annual fall meeting, Las Vegas, Sept. 30–Oct. 3, 1973 model flooding experiments. In: Hou M, Xie H, Were P (eds) Sedaghat M, Mohammadzadeh O, Kord S, Chatzis I (2015) Pore-level Clean energy systems in the subsurface: production, storage and experimental investigation of ASP flooding to recover heavy oil conversion. Springer series in geomechanics and geoengineering. in fractured five-spot micromodels. EUROPEC 2015, 1–4 June, Springer, Berlin Madrid Yousefvand H, Jafari A (2015) Enhanced oil recovery using polymer/ Shaw RA, Stright DH Jr (1975) Performance of the Taber South Poly- nanosilica. Procedia Mater Sci 11:565–570 mer Flood. Paper CIM 75–41 presented at the CIM 26th Annual Technical Meeting, Banff, June 11–13, 1975 Publisher’s note Springer Nature remains neutral with regard to Sheng JJ (2013) Chapter 3—polymer flooding—fundamentals and field jurisdictional claims in published maps and institutional affiliations. cases. Enhanc Oil Recov Field Case Stud 2013:63–82 1 3 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Petroleum Exploration and Production Technology Springer Journals

Experimental study of polymer injection enhanced oil recovery in homogeneous and heterogeneous porous media using glass-type micromodels

Free
11 pages

Loading next page...
 
/lp/springer_journal/experimental-study-of-polymer-injection-enhanced-oil-recovery-in-HeFqZW99qa
Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Earth Sciences; Geology; Industrial and Production Engineering; Energy Technology; Offshore Engineering; Industrial Chemistry/Chemical Engineering; Monitoring/Environmental Analysis
ISSN
2190-0558
eISSN
2190-0566
D.O.I.
10.1007/s13202-018-0492-x
Publisher site
See Article on Publisher Site

Abstract

In water flooding process, volumetric sweep efficiency and oil recovery can be enhanced using polymer to increase the viscosity of water. As a result, polymer flooding has higher recovery as compared to water flooding due to front stability and reduction of fingering problem. In this research work, a set of polymer flooding runs were carried out using glass-type micromodels. The micromodels were fabricated to have homogeneous and heterogeneous flow patterns. They were positioned horizontally and saturated with a heavy crude oil sample taken from an Iranian oil field before starting the injection. Three commercial polymers were used in this study. Whole process was photographed continuously with a high-resolution camera to monitor the displacement of polymer solution in the micromodels. As a part of this study, the effect of different param- eters including polymer solution concentration, injection flow rate and heterogeneity on performance of polymer flooding was investigated. On top of the regular homogeneous and heterogeneous flow patterns used in this study, a heterogeneous flow pattern mimicking sandstone reservoirs was created based on the image of a thin section of a sandstone (outcrop) and polymer front movement was observed during injection. Keywords Polymer injection · Glass-type micromodel · Sweep efficiency Introduction adsorb on rock surface (Sheng 2013; Sheng et  al. 2015; Buchgraber 2008; Barati 2011). Polymer is a large molecule built up by the addition of small When polymer is added to water (brine), its viscosity repeating units (monomers) (Allcock et al. 2003). In general, will be increased. Therefore, polymer injection can lead to the polymers used in the EOR process are divided into two changes in fractional flow and mobility ratio and also has groups of synthetic polymer and biopolymer. The unique fluid diversion effect. As a result, polymer can help with features of each category of these polymers have caused reduction of viscous fingering and improving of water injec- advantage and disadvantage. Synthetic polymers have tion profile that will lead to an improvement in sweep effi- affordable prices, appropriate viscosity in fresh water and ciency (Buchgraber 2008; Barati 2011; Sorbie 1991; Need- acceptable adsorption on the rock surface. The disadvantage ham and Doe 1987; Chang 1978). Nevertheless, stability of of this type of polymer can be attributed to the sensitivity polymer is an issue. Many researchers have focused on the to flow rate and shear degradation. This type of polymer parameters that affect the stability of polymers in different has low efficiency in high-salinity water. Biopolymers show conditions. The factors of chemical, mechanical and biologi- excellent performance against high-salinity water and shear cal degradation cause instability of the polymers. Oxidation degradation but they can be sensitive to bacterial degrada- and ferric ions are the factors affecting the chemical degra- tion in low-temperature reservoirs. Biopolymer does not dation. Reducing the amount of oxygen in the water in con- 3+ 2+ tact with the polymer prevents the generation of Fe, Fe 2− and free radical O ions making the polymer solution to be * Jalal Foroozesh more stable and prevents chemical degradation. Mechani- jalal.foroozesh@gmail.com; jalal.foroozesh@utp.edu.my cal degradation occurs when polymer molecules break 1 down due to shear stress. Flow rate has a significant impact Department of Petroleum Engineering, Universiti Teknologi on preventing this phenomenon and should be optimized. PETRONAS, Seri Iskandar, Malaysia Vol.:(0123456789) 1 3 Journal of Petroleum Exploration and Production Technology Biological degradation is more likely to occur in case of used. Glass micromodels are quite useful tools to study the biopolymers, but under a few conditions; synthetic polymers active recovery mechanisms during polymer injection and can also be degraded biologically. Using some additives the effective parameters in oil recovery (Danesh et al. 1987). can solve these problems (Sheng et al. 2015; Sorbie 1991). Due to the transparency, the whole process can be directly Polymer injection in reservoirs will face some limitations. observed and photographed and by using a digital image Reservoir water salinity, reservoir temperature, type of for- analysis (DIA), the photos that are taken during the process, mation and reservoir permeability are factors influencing the can be analyzed for any pore level mechanism and also to performance of polymer injection. For example, in the case estimate the recovery factor. of polymer incompatibility with the reservoir fluid salinity The challenges ahead when using micromodels are to cre- and the ions contained therein, the viscosity of the polymer ate a porous media that can represent a real porous media. solution is not sufficiently achieved and its effectiveness Many different pore patterns have been created in micro- is greatly reduced (Sheng et al. 2015; Needham and Doe models, each of which has its advantage and disadvantage. 1987). The high temperature of the reservoir can break down Figure 1 shows some of these micromodels (Danesh et al. the molecular chains of polymers and turns them into mono- 1987; Romero-Zeron and Kantzas 2007; Buchgraber et al. mers. If these phenomena happen, the viscosity of the solu- 2011; Farzaneh et al. 2012; Hamedi Shokrlu and Babadagli tion of water and polymer will be reduced. In most reports, 2015; Howe et al. 2015; Bahari Moghaddam and Rasaei a temperature tolerance of 93 °C for both types of polymer 2015; Manlowe and Radke 1990). (synthetic polymer and biopolymer) is acceptable. It is worth In the laboratory, a lot of experiments have been done on noting that the temperature of 93 °C does not guarantee the the polymer injection using micromodels. Heshmati et al. stability of the polymer over a long period of time (Taber (2007) used a micromodel with two different permeabil- et al. 1997a, b). Most successful polymer injections have ity layers to study the effect of permeability heterogeneity been in sandstone formations. The use of this technology in on the performance of the polymer flooding. After poly - carbonate formations requires further investigation. Hence, mer injection, a 17% increase in recovery was reported at the type of reservoir rock and the mineralogy of the rock is best (Heshmati et al. 2007). Emami et al. (2008) employed important (Jewett and Schurz 1970). The polymer is com- a micromodel with five-spot injection pattern to examine posed of large molecules. Therefore, the use of polymer in the impact of local and global heterogeneity (similar to lay- rocks with small pore–throat size is not recommended. This ered reservoirs) on recovery. The results of the experiments will cause the pore–throat to block. Pore–throat size will showed that the maximum recovery achieved when layers control the rock permeability. Therefore, rocks with low are perpendicular to mean flow direction. Also, the slope of permeability as a result of having small pore size cannot be the micromodel strongly affects the efficiency of the polymer a good candidate for polymer injection (Jewett and Schurz injection process (Emami Meybodi et al. 2008). Hematpur 1970). et al. (2011) studied the effect of polymer injection in the Field practice of polymer flooding is reported in the liter - presence of low-viscosity oil using a micromodel setup. The ature. The use of polymer as an additive for injection into oil results showed that using the polymer of hydrolyzed poly- reservoirs began in 1967 and has quickly become popular. acrylamide (HPAM) had the best performance under test So far, many fields have been under polymer injection. These conditions (Hematpour et al. 2011). Maghzi et al. (2011) include ‘Taber south field’, ‘North Burbank unit’, ‘Brelum used a micromodel setup with five-spot injection pattern unit reservoir’ and ‘Vernon’. In all cases, the polymer injec- to investigate the injection of the nanoparticles with poly- tion has been effective. According to reports, except for a mer. An increase of 10% in recovery was the result of the few cases, the injected polymer type was synthetic. An addi- experiments conducted by these researchers (Maghzi et al. tional average recovery of 7% has been reported. Another 2011). Wegner and Ganzer (2013) compared the results of notable case is the formation type that the polymer has been polymer flooding in a micromodel with numerical simu- injected into it. Most formations under polymer injection lation (Wegner and Ganzer 2013). Yousefvand and Jafari have been sandstone rocks (Clampitt and Reid 1975; Lozan- (2015) investigated the polymer efficiency in the presence ski and Martin 1970; Shaw and Stright 1975; Rowalt 1973). of nanosilica particles. The investigation was done in a Experimental study is an important step in the investiga- micromodel using reservoir oil and brine (Yousefvand and tion of polymer injection. Core flooding is commonly used Jafari 2015). Sedaghat et al. (2015) investigated the perfor- for experimental study. Core is one of the best representa- mance of alkaline surfactant polymer (ASP) injection for tives of the reservoir at the surface, but having native res- heavy oil recovery using fractured five-spot micromodels. ervoir’s core is not easy. Also, due to the lack of visualiza- Hydrolyzed polyacrylamide showed a better performance tion of the core flooding in cases where the flow behavior as compared to other polymers (Sedaghat et al. 2015). Rock is necessary, this porous media cannot be a good candidate. et al. (2017) investigated the behavior of viscoelastic poly- To overcome these problems, visual micromodels have been mers in a porous media using a glass micromodel setup. The 1 3 Journal of Petroleum Exploration and Production Technology Fig. 1 Micromodel pore patterns two factors of salinity and mechanical degradation had the industry as an additive to water-based muds for mainly fluid high impact on the results presented by these researchers. loss control. The results obtained by these researchers showed that there The first step was to fabricate a glass micromodel which is a relationship between shear thickening and elastic flow was a relatively time-consuming procedure. The steps instability (Rock et al. 2017). involved in the fabrication process were first using a piece It is inferred that the performance of polymer flooding of mirror with appropriate dimensions and then removing can be dependent on reservoir type and characteristic and the paint coated on the backside until the mercury coat- micromodel setups can be used to study it. Therefore, in ing was exposed. Then, we covered a layer of plastic lami- this research, a number of micromodel setups with differ - nate on the mercury coating and the pattern in black and ent heterogeneities of pore patterns have been fabricated white printed on a paper, overlaid on the laminate surface. to study the performance of three different polymers in an Ultraviolet (UV) exposure was used to polymerize the non- Iranian oil field. pattern area (resistant to acid) and then by means of nitric acid, the non-polymerized (pattern) area was removed (dis- solved) and then the hydrofluoric acid was used to etch the Experimental procedure and materials pattern area. Cyclic submersion in hydrofluoric acid for a certain time can lead to a certain depth of etching. It should The primary objective of the undertaken series of experi- be mentioned that all the steps were taken under complete ments was to investigate the effect of different parameters caution. A number of micromodel setups with homogene- such as polymer solution concentration, polymer type and ous and heterogeneous flow patterns were built and used for injection flow rate on oil recovery. Hence, three different experimental study of polymer injection. Figure 2 shows a types of commercial polymer were used to prepare the poly- homogenous flow pattern used for micromodel fabrication. mer solution with different concentrations and were injected The heterogeneous micromodels will be explained later on. at different flow rates. The specifications of the polymers The next step was to saturate the micromodel setup with are given in Table 1. These polymers are used in drilling the crude oil without having any connate water saturation. Table 1 The specifications of Polymer Commercial name Provider Polymer family the polymers used in this study type no. ® ® 1 DRISPAC Superlo Chevron Phillips Chemical Company Polyanionic cellulose 2 DRISPAC Regular Chevron Phillips Chemical Company Polyanionic cellulose 3 DRISCAL D Chevron Phillips Chemical Company Polyacrylamide (syn- thetic polymer) 1 3 Journal of Petroleum Exploration and Production Technology Pore volume The depth of etched glass (pores) created in micromodel was measured by Micrometer tool. Therefore, by multiplying the depth by pore area, pore volume was determined as 0.068 cc. A magnified view of the etched glass is shown in Fig.  4. Fig. 2 Homogenous flow pattern Permeability The crude oil used in this study was taken from one of the Iranian oil fields which is located in the southern part Another important property of glass micromodel which of Iran and has an API of 19.8 (relatively heavy). The should be determined was absolute permeability. This could saturation process was done very carefully to avoid any be achieved by gathering the flow rate data and the related penetration of crude oil through the non-porous media, injection pressure. So, the flow rate was set and let the sys- rather than etched porous media. Figure 3 shows a sample tem to reach the constant (stabilized) pressure die ff rence. By of saturated flow pattern with crude oil. knowing the pressure difference (outlet pressure is atmos- After the flow pattern is saturated, the experiment was pheric pressure), the absolute permeability can be calculated started. Quizix pump was used in this study to inject the by Darcy’s law. The permeability of homogenous pattern polymer solution into the micromodel at a very precise rate measured was around 20 Darcy. close to real velocity of fluid in the reservoir. Results and discussion The physical properties of micromodel Homogeneous flow pattern As one of the important stages of the experiment, the prop- erties of the generated glass micromodel were determined. Eec ff t of polymer solution concentration on oil recovery Some of these properties, namely pore volume and poros- ity were determined by means of computer software using Three polymer solutions with different concentrations (1000, digital image analysis (DIA) process. Some other proper- 1500 and 2250 ppm) of polymer type 1 were prepared and ties such as permeability were measured experimentally injected at the rate of 0.0002  cc/min to the oil-saturated by steady-state flow test after stabilization of differential homogenous micromodel to study the effect of polymer con- pressure (ΔP) across the setup. It should be noted that the centration on oil recovery. Figure 5 shows the effect of poly - micromodels used here are water-wet as made by glass mer solution concentration on oil recovery for polymer type 1. and no aging process has been carried out to change their original wettability. Porosity Using DIA process, the areal porosity was determined and with regard to uniform distribution of etching depth, the porosity was calculated and reported as a total porosity. The pattern was saturated with oil and then the picture was taken and analyzed to determine the porosity. Using image analysis, the total porosity estimated was 0.36 (36%). Fig. 4 A magnified view of the etched glass (blue and red colors refer Fig. 3 Sample of homogenous flow pattern saturated with crude oil to water and oil phases, respectively) (dark color is oil; white color is grain) 1 3 Journal of Petroleum Exploration and Production Technology 2250 ppm 1000 ppm 1500 ppm 00.1 0.20.3 0.40.5 0.60.7 Pore Volume Injected(fraction) Fig. 5 Comparison between oil recoveries by three different concentrations of polymer solutions of polymer type 1 (flow rate = 0.0002 cc/min) Table 2 Measured physical properties of water and polymer solutions of 1000 pm and it is 4.15 cp at concentration of 2250 pm at room temperature (T = 24 °C) resulting in a better sweep efficiency. In micromodel setup, it is not possible to investigate the vertical conformance and Material Viscosity (cp) Density (g/cm ) sweep due to the nature of the equipment but the role of poly- Water 0.94 1.01 mer injection and gel treatment in conformance control in oil Polymer type 1 (c = 1000 ppm) 1.44 1.01 reservoirs has been discussed in the literature (Kantzas et al. Polymer type 1 (c = 1500 ppm) 1.98 1.01 1999; Suleimanov and Veliyev 2016). Polymer type 1 (c = 2250 ppm) 4.15 1.02 Polymer type 2 (c = 1500 ppm) 7.14 1.01 Polymer type 3 (c = 1500 ppm) 3.24 1.01 Eec ff t of injection flow rate on oil recovery In these series of experiments, three different flow rates were tested: 0.0002, 0.0005 and 0.0008  cc/min for the As it can be seen in Fig. 5, for 150% increase in poly- mer concentration in solution (from 1500 to 2250 ppm), the polymer type 1 and 0.0002 and 0.0005 cc/min for poly- mer types 2 and 3, all at fixed concentration of 1500 ppm. recovery has increased around 7%. Enhancement of polymer concentration increases solution viscosity resulting in front As it can be observed, in each step, the flow rate has been increased by a factor of 2.5. Figure 7 shows the effect of stability and minimizes fingering and channeling problem, which lead to higher oil recovery. injection flow rate for the polymer type 2. As it can be seen in Fig. 7, at lower flow rate, the polymer As it can be observed in Fig. 5, the lower concentration (1000 ppm) solution had the fastest breakthrough among all displacement in the model is more piston-like while at high injection rate, it is having unstable front with less piston- at 0.54 pore volume (PV) injected, while the other solutions had later breakthrough at 0.58 and 0.62 PV injected for 1500 like displacement. Similar results were observed for polymer types 1 and 3. This is because, at low injection rates, fluid and 2250 ppm, respectively. Lower concentration translates to lower viscosity (which is shown in Table 2) and lower has time to distribute areally to have a more piston-like flow. It should be noted that sensitivity of oil recovery to poly- sweep efficiency (see Fig.  6). It should be mentioned that for all our experiments, once there was no more oil production, mer injection rate during polymer flooding is also affected by low injectivity of polymer solution compared to water the injection was ceased shortly. Figure 6 shows the profile of injected polymer in micro- injection (Van den Hoek et al. 2012). Loss of injectivity especially in low-permeability formations is a serious prob- model for low and high concentrations at three different pore volumes. As it can be seen, at high concentration of 2250 ppm, lem in polymer flooding projects, which can occur because of large molecule size and high viscosity of polymer solu- the polymer displacement is more towards positon-like as compared to the profile of low concentration of 1000 ppm. tions. In real reservoirs, high injection rates can cause the bottom-hole pressure to rise up and if not controlled, This could sweep more area occupied by oil. Table 2 shows that the viscosity of polymer type 1 is 1.44 cp at concentration exceed the fracture pressure and cause formation damage. 1 3 Oil Recovery (%) Journal of Petroleum Exploration and Production Technology Fig. 6 Front stability for low-concentration (1000 ppm) polymer solution and high-con- centration (2250 ppm) polymer solution at pore volumes of 0.2, 0.5 and 1, polymer type 1, injection rate; 0.0002 cc/min, injection point on the left side (brown color: oil, whitish blue color: polymer) Fig. 7 Comparison between flooded pattern at breakthrough time for polymer type 2 for flow rates 0.0002  cc/min (lower side picture) and 0.0005 cc/min (upper side picture) both at concentration = 1500 ppm. Injection point on the left side (brown color: oil, white color: polymer) Comparison between water injection and polymer injection the displacement of water and polymer in the micromodel at three different pore volumes. Figure  8 clearly illustrates Water injection was done to compare its performance with the better areal sweep efficiency and also less residual oil polymer injection. Water flooding is one of the most favora- trapped in the pores and throats resulted from polymer flood- ble secondary recovery methods due to the availability of ing as compared to water injection leading to a higher oil water (especially in offshore operation) and the relatively recovery factor. low cost and complexity compared to the other EOR meth- ods. Nevertheless, the mobility of water needs to be con- Comparison of performance of the three polymer types trolled especially in layered and heterogeneous reservoirs to have a good conformance control. Adding polymer to water Figure 9 compares the displacement of polymer types 1, 2 will increase the viscosity as it was discussed earlier and will and 3 in the micromodel at the same pore volume injected. increase the volumetric sweep efficiency. Figure  8 compares They all have the fixed concentration of 1500 ppm. It can be 1 3 Journal of Petroleum Exploration and Production Technology Fig. 8 Flooded patterns at equal pore volumes injected for water flooding and polymer flooding, Polymer type 1, injection rate = 0.0002 cc/min, concentration = 1500 ppm (brown color: oil, whitish blue color: polymer and blue color: water) Fig. 9 Comparison between displacement of polymer type 1, 2 and 3 at fixed concentration of 1500 ppm (identical pore volume injected) 1 3 Journal of Petroleum Exploration and Production Technology observed that the displacement of polymer type 2 is more side layer has lower permeability compared to the upper piston-like (i.e., stable displacement) as compared to two side layer which has higher permeability. These different other polymers. Next, polymer type 3 is displacing like a permeabilities have been obtained by controlling the size piston. This could be explained by considering the viscos- of the pores and throats in each layer. This is one of the ity of polymer solutions presented in Table 2. As it given in heterogeneities that can be found in layered and segregated Table 2, the solution of polymer type 2 has largest viscosity reservoirs. and after that polymer type 3 has a higher viscosity as com- As it can be seen in Fig. 10, the polymer is injected from pared to polymer type 1. These viscosity values have led to the left and has been mainly advanced in bottom side with a better mobility ratio of polymer type 2 resulting in a more higher permeability. Although the upper layer has lower stable displacement. permeability as compared to lower layer, its area is large causing polymer to move through it too. It should be noted Heterogeneous flow patterns that the permeabilities of micromodels are generally large making the upper layer to also have a large permeability (in Real oil reservoirs are complex set of different heterogenei- order of several Darcys) but smaller than the permeably of ties. But, in the majority of the routine reservoir engineering the lower side layer. That means, the contrast between layers’ calculations, this heterogeneity is not considered. In routine permeability may not cause much difference in flow behavior core-flooding experiments also the effect of heterogeneity through different layers. It is worth mentioning that, due to cannot be considered due to the small scale of core plugs. the horizontal positioning of the micromodel pattern, there Hence, inclusion of heterogeneity in glass micromodels is not any effect of gravity involved. enables us to create different kind of heterogeneity and to visualize the polymer front movement and behavior from Faulted heterogeneous flow pattern the beginning to the end of the experiment. In the following section, the results of polymer flooding in the artificially In this flow pattern, the aim was to simulate the fault effect created heterogeneous patterns will be discussed. It should (faulted layer) on flow condition in real reservoirs dur - be mentioned that type of heterogeneity has been inherited ing polymer injection. There are two layers in this pattern from some real large-scale heterogeneities, which frequently which have different permeability (high and low) which are can be found in subsurface reservoirs. faulted in the middle point of the flow pattern (normal fault) as shown in Fig. 11. As it can be seen, mimicking a faulted Two-layered flow pattern reservoir and due to displacement of zones, high- and low- permeable zones are juxtaposed. This flow pattern includes two layers with different perme- As it can be observed in Fig.  11, when the polymer ability (layers are separated by a red dashed line). The lower front has reached the fault, it was diverted through the Fig. 10 Polymer front advancement in two-layered heterogeneous flow pattern, injection point on the right, upper side picture after 0.2 PV injected, lower side picture after 0.4 PV injected (brown color is oil, yellow color is grain and whitish blue color is polymer solution) 1 3 Journal of Petroleum Exploration and Production Technology high-permeable zone on the top-right side of the figure. It Flow pattern fabricated using a sandstone thin section should be noted that the stream line was not 1-D (in x-direc- tion) anymore and polymer breakthrough will potentially Using the picture of a thin section from an outcropped sand- take longer time comparing to the 1-D flow. In real reser - stone, this flow pattern was fabricated. The primary objec- voirs, faults can change the flow direction and affect the tive was to simulate more real porous media to study fluid volumetric sweep efficiency. flow. It should be noted that this thin section is a small sec- tion of a rock which heterogeneity cannot be captured at this Tilted zone heterogeneous flow pattern scale. However, tortuosity, dead-end pore effect and irreg- ular pore and throat shapes still can be observed to some In this heterogeneous flow pattern, tilted zones have been extent. Figure 13 shows the micromodel with sandstone o fl w placed adjacent to each other with low and high permeabil- pattern that is saturated with oil. ity, alternatively as shown in Fig. 12. Figure  14 shows a magnified image of polymer solu- Polymer front movement in this flow pattern is zig-zag tion–oil interface inside this flow pattern. shape due to the permeability contrast between tilted zones. As it is observed in this, due to the high viscosity of In this flow pattern, when polymer front has reached low- polymer solution comparing to water injection which was permeable zone, it has tried to move forward (1-D) longitu- shown before in Fig. 8, there is less residual oil trapped dinally while in high-permeable zone, the movement is both in the corners of pores and specifically in throats lead- longitudinally and transverse (2-D). ing to a better displacement efficiency with minimum Fig. 11 Polymer front movement in quadratic flow pattern, polymer is injected from the left (brown color is oil, light yellow color is grain and blue color is polymer solution) Fig. 12 Polymer front movement in tilted-stripped layered flow pattern (high- and low-permeable layers alternate each other; injection from right side; blue dashed lines are zone borders; L low-permeable zone, H high-permeable zone) Fig. 13 Oil–saturated sandstone flow pattern (grains in white color) 1 3 Journal of Petroleum Exploration and Production Technology Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creat iveco mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. References Allcock HR, Lampe FW, Mark JE (2003) Contemporary polymer chemistry, Third edn, Pearson Education, New York Bahari Moghaddam M, Rasaei MR (2015) Experimental study of the fracture and matrix effects on free-fall gravity drainage with micromodels. SPE J 20:Issue 02 Fig. 14 Oil–polymer solution interface in sandstone flow pattern dur - Barati N (2011) Study of polymer flooding performance in the pres- ing polymer flooding (polymer solution in light blue and oil in brown ence of nanoparticles. Master Thesis, Petroleum University of colors) Technology Buchgraber M (2008) An enhanced oil recovery micromodel study with associative and conventional polymers. Diploma Thesis, University of Leoben, Austria oil-bypassing during the injection. This figure shows the Buchgraber M, Clemens T, Castanier LM, Kovscek A (2011) A efficiency of polymer flooding in systems with more real- microvisual study of the displacement of viscous oil by polymer istic flow patterns. solutions. SPE Reserv Eval Eng 14:Issue 03 Chang HL (1978) Polymer flooding technology yesterday, today, and tomorrow. J Petrol Technol 30:Issue 08 Clampitt RL, Reid TB (1975) An economic polymer flood in the North Burbank Unit, Osage County, Oklahoma. Fall meeting of Summary and conclusions the Society of Petroleum Engineers of AIME, 28 September–1 October, Dallas Danesh A, Peden JM, Krinis D, Henderson GD (1987) Pore level A set of polymer flooding experiments was performed in visual investigation of oil recovery by solution gas drive and manually fabricated glass homogeneous and heterogene- gas injection. SPE annual technical conference and exhibition, ous micromodels in horizontal positioning having 1-D or 27–30 September, Dallas, Texas Emami Meybodi H, Kharrat R, Ghazanfari MH (2008) Effect of het- 2-D flow direction. The micromodel setup was saturated erogeneity of layered reservoirs on polymer flooding: an exper - with a relatively heavy Iranian crude oil. The oil recov- imental approach using 5—spot glass micromodel. Europec/ ery results were compared with oil recovery from water EAGE conference and exhibition, 9–12 June 2008, Rome, Italy flooding under the same initial conditions. Three different Farzaneh SA, Dehghan AA, Ghazanfari MH, Kharrat R (2012) A comparative study on WAS, SWAS, and solvent-soak scenarios polymer types were used during this experimental study. applied to heavy-oil reservoirs using five-spot glass micromod- The polymer solution concentration and injection rate els. J Can Pet Technol 51:Issue 05 were changed to account for their effects on oil recovery. Hamedi Shokrlu Y, Babadagli T (2015) Pore-scale investigation of The results showed a better performance for polymer type phase distribution and residual-oil development during second- ary and tertiary solvent injection. SPE Reserv Eval Eng 18:Issue 2. It was also observed that at higher polymer concentra- tions and lower injection rates, a better sweep efficiency Hematpour H, Mardi M, Edalatkhah S, Arabjamaloe R (2011) can be achieved by the polymer. Visualization of polymer Experimental study of polymer flooding in low-viscosity oil front inside the micromodel showed a more piston-like using one-quarter five-spot glass micromodel. Pet Sci Technol 29:1163–1175 movement of solution as compared to the front in pure Heshmati M, Mahdavi H, Haghighi M, Torabi M (2007) Visualiza- water injection. Heterogeneity, as another studied param- tion of polymer flooding in heterogeneous system using glass eter in this research, had an important role in oil recovery. micromodels. J Sci 332:25–32 Several artificially made heterogeneities were introduced Howe AM, Clarke A, Mitchell J, Staniland J, Hawkes LA (2015) Visualising surfactant EOR in core plugs and micromodels. SPE into glass micromodels aiming to simulate the real het- Asia Pacific enhanced oil recovery conference, 11–13 August, erogeneity in the reservoirs. As a result, swept area was Kuala Lumpur, Malaysia significantly affected by heterogeneity since it affects the Jewett RL, Schurz GF (1970). Polymer flooding—a current appraisal. front movement. Therefore, macroscopic heterogeneities JPT 22:675–684 Kantzas A, Burger D, Pow M, Cheung V, Mourits F (1999) Design such as layered or faulted reservoirs and also existence strategies for improved conformance using polymer gels. Pet of any high- or low-permeable zones must be taken into Soc Can. https ://doi.org/10.2118/99-13-18 account before any polymer injection project to achieve Lozanski WR, Martin P (1970) Taber South—Canada’s first polymer the best EOR performance. flood. J Cdn Pet Tech 1970:99 1 3 Journal of Petroleum Exploration and Production Technology Maghzi A, Mohebbi A, Kharrat R, Ghazanfari MH (2011) Pore- Sheng JJ, Leonhardt B, Azri N (2015) Status of polymer-flooding tech- scale monitoring of wettability alteration by silica nanoparticles nology. J Can Pet Technol 54:Issue 02 during polymer flooding to heavy oil in a five-spot glass micro- Sorbie KS (1991) Polymer-improved oil recovery, 1st ed. CRC Press, model. Transp Porous Med 87:653–664 Inc, Boca Raton Manlowe DJ, Radke CJ (1990) A pore-level investigation of foam/oil Suleimanov BA, Veliyev EF (2016) Nanogels for deep reservoir con- interactions in porous media. SPE Reserv Eng 5:Issue 4 formance control. Soc Pet Eng. https ://doi.org/10.2118/18253 Needham RB, Doe PH (1987) Polymer flooding review. J Pet Technol 4-MS Rock A, Hincapie RE, Wegner J, Ganzer L (2017) Advanced flow Taber JJ, Martin FD, Seright RS (1997a) EOR screening criteria revis- behavior characterization of enhanced oil recovery polymers ited—part 2: applications and impact of oil prices. SPE Reserv using glass–silicon–glass micromodels that resemble porous Eng 12:Issue 03 media. SPE Europec featured at 79th EAGE conference and Taber JJ, Martin FD, Seright RS (1997b) EOR screening criteria exhibition, 12–15 June, Paris, France revisited—part 1: introduction to screening criteria and enhanced Romero-Zeron LB, Kantzas A (2007) The ee ff ct of wettability and pore recovery field projects. SPE Reserv Eng 12:Issue 03 geometry on foamed-gel-blockage performance. SPE Reserv Eval Van den Hoek P, Mahani H, Sorop T, Brooks D, Zwaan M, Sen S, Eng 10:Issue 02 Saadi F (2012) Application of injection fall-off analysis in poly - Rowalt RJ (1973) A case history of polymer water flooding—Bre- mer flooding. Soc Pet Eng. https ://doi.org/10.2118/15437 6-MS lum field unit. Paper SPE 4671 presented at the SPE-AIME 48th Wegner J, Ganzer L (2013) Numerical analysis of polymer micro- annual fall meeting, Las Vegas, Sept. 30–Oct. 3, 1973 model flooding experiments. In: Hou M, Xie H, Were P (eds) Sedaghat M, Mohammadzadeh O, Kord S, Chatzis I (2015) Pore-level Clean energy systems in the subsurface: production, storage and experimental investigation of ASP flooding to recover heavy oil conversion. Springer series in geomechanics and geoengineering. in fractured five-spot micromodels. EUROPEC 2015, 1–4 June, Springer, Berlin Madrid Yousefvand H, Jafari A (2015) Enhanced oil recovery using polymer/ Shaw RA, Stright DH Jr (1975) Performance of the Taber South Poly- nanosilica. Procedia Mater Sci 11:565–570 mer Flood. Paper CIM 75–41 presented at the CIM 26th Annual Technical Meeting, Banff, June 11–13, 1975 Publisher’s note Springer Nature remains neutral with regard to Sheng JJ (2013) Chapter 3—polymer flooding—fundamentals and field jurisdictional claims in published maps and institutional affiliations. cases. Enhanc Oil Recov Field Case Stud 2013:63–82 1 3

Journal

Journal of Petroleum Exploration and Production TechnologySpringer Journals

Published: Jun 1, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off