Page 104 - Hybrid Enhanced Oil Recovery Using Smart Waterflooding
P. 104

96      Hybrid Enhanced Oil Recovery using Smart Waterflooding

            60                                  700        0.14
            50                    A, Ca/Na=0    600        0.12        A
           Recovery factor (%OOIP)  40  B, Ca/Na=0.0175  400 Pressure drop (mbar)  krw  0.08  LSS
                                                                       B
                                                500
                                                           0.10
                                                                       C
                                  C, (Ca+Mg)/Na=0.033
            30
                                                300
                                                           0.06
            20
            10                                  200        0.04            LSW
                                                100
                                                           0.02
            0                                   0          0.00
              0       5       10      15      20
                           PV injected                         0.0   0.2    0.4   0.6    0.8   1.0
                                                                             SW
          FIG. 4.25 Recovery and pressure drop in the experiments
          of secondary low-salinity waterflood and tertiary low  FIG. 4.26 Endpoint relative permeabilities to water in the
          salinityeaugmented surfactant flood. (Credit: From  experiments of secondary low-salinity waterflood and
          Hosseinzade Khanamiri, H., Baltzersen Enge, I., Nourani, M.,  tertiary low salinityeaugmented surfactant flood. (Credit:
          Stensen, J.Å., Torsæter, O., & Hadia, N. (2016). EOR by low  From Hosseinzade Khanamiri, H., Baltzersen Enge, I.,
          salinity water and surfactant at low concentration: Impact of  Nourani, M., Stensen, J.Å., Torsæter, O., & Hadia, N. (2016).
          injection and in situ brine composition. Energy and Fuels,  EOR by low salinity water and surfactant at low
          30(4), 2705e2713. https://doi.org/10.1021/acs.energyfuels.  concentration: Impact of injection and in situ brine
          5b02899.)                                     composition. Energy and Fuels, 30(4), 2705e2713. https://
                                                        doi.org/10.1021/acs.energyfuels.5b02899.)
            Zhang, Yu, Cheng, and Lee (2015) formulated a va-
          riety of microemulsions without alkali and investigated  10e24 wt%. The optimal salinity is determined as
          the phase behavior and IFT reduction of the microemul-  19 wt%. The salinity less than 10 wt% shows the Win-
          sions. Conventionally, the alkali addition decreases the  sor type Ⅰ. The formulation of Winsor type Ⅱ is not
          optimal salinity of surfactant solution, but it can intro-  observed in the range of salinities to be tested. For the
          duce risks for chemical EOR process. A number of for-  surfactant mixture of mixing ratio of 2:1, which has po-
          mulations of microemulsions are prepared using the  tential to aggregate, Winsor type Ⅲ is observed in the
          anionic, cationic, zwitterionic, and nonionic surfac-  range of 0e10 wt% salinity. The optimal salinity is
          tants. Mixtures of different surfactant types have a risk  determined to be 3 wt%. In contrast to the previous sur-
          to precipitate because of strong electrical interaction.  factant mixture, Winsor type Ⅱ is formulated above
          However, the mixture with appropriate mixing ratio is  10 wt%. No Winsor type Ⅰ is observed. These phase
          able to produce benefits without precipitation problem.  behavior tests reveal that the mixing ratio controls the
            Firstly, the surfactant mixture between cationic and  salinity condition determining microemulsion type. In
          ionic surfactants is investigated through phase behavior  addition, the mixing ratio significantly changes optimal
          test of microemulsion and IFT measurement. In the  salinity (Fig. 4.27). Additional experiments investigate
          phase behavior test, the mixing ratio between anionic  another surfactant mixture using different anionic and
          and cationic surfactants varies from 4:1 to 1:4. The  same cationic surfactants. The experiments using the
          cloudy solutions are observed with mixing ratios of 2:  two different surfactant mixture solutions indicate that
          1, 1:1, and 1:2. The mixing ratios either higher than 2:  the properties of surfactant mixture are closely related
          1 or smaller than 1:2 show the clear transparency of so-  to the selected surfactant type.
          lution. Using the mixing ratios of 4:1 and 2:1, surfactant  Secondly, the formation of surfactant mixture using
          solutions are investigated in terms of salinity-  anionic and zwitterionic surfactants is investigated.
          dependent volume fraction of microemulsion type.  Because the zwitterionic surfactant exhibits both posi-
          For the surfactant mixture with the mixing ratio of 4:  tive and negative charges, it behaves like anionic surfac-
          1, anionic surfactant is predominant over cationic sur-  tant at high pH and cationic surfactant at low pH.
          factant. The phase behavior test evaluates the microe-  Surfactant mixture is prepared adapting the mixing ratio
          mulsion type in the range of 0e24 wt% salinity. The  between anionic and zwitterionic surfactants as 2:1. In
          Winsor type Ⅲ of microemulsion appears in range of  the range of 0e18 wt% salinity, only Winsor type Ⅲ is
          high salinity conditions including optimal salinity,  observed at pH 2 condition. In addition, the volume
   99   100   101   102   103   104   105   106   107   108   109