Page 308 - Enhanced Oil Recovery in Shale and Tight Reservoirs
P. 308

Spontaneous imbibition                                       281


              according to Eq. (10.1); initially the fluid imbibes into larger pores and later
              into smaller pores; therefore, the imbibition velocity becomes slower at late
              time. Another fact that can cause slow imbibition rate at late time is that the
              pressure inside the core builds up; this built pressure serves as a resistance to
              the imbibition. The late flat portion of the curve is not caused or represented
              by diffusion as Shen et al. (2016) interpreted.
                 Again from Eq. (10.1), we can see that the imbibition velocity into a
              smaller pore is lower than that into a larger pore; as the imbibition takes a
              longer time (longer l), the imbibition velocity becomes lower. Tagavifar
              et al.’s (2019) simulation results also approve this fact. Yang et al.’s (2016)
              experimental data showed that imbibition velocities were lower for lower
              porosity and permeability cores; their experimental velocities were lower
              than what the theory (Eq. 10.1) predicted; the lower their porosity and
              permeability, the lower their experimental velocity was, compared with
              the theoretical velocity.
                 Real experimental data in log(V) versus log(t) may not show the slope of
              0.5. Hu et al. (2012) suggested the change of the slope of the curve repre-
              sented the change in pore connectivity. Cai and Yu (2011) suggested the
              slope change was caused by pore tortuosity. Yang et al. (2016) suggested
              that the slopes reflected the pore distribution and pore connectivity, with
              the early slope reflected macropores (>50 nm), the middle and late slopes
              reflected meso- and micropores, as shown in Fig. 10.1.
                 For type “B,” the linearity appears in the early time, suggesting relatively
              high permeability and good pore connectivity. The macropores are well-
              developed, and the pore size distribution is of a single-peak type. For type
              “S,” the initial position has an “arc-shaped tail” which suggests n i > 0.5 at
              the early imbibition stage and good pore connectivity. The macropores
              and mesopores are well developed, and the pore size distribution is of the
              two-peak-type. For type “A,” the arc-shaped and convex behavior suggests
              a low initial time exponent (n i < 0.5) and poor pore connectivity; the late
              time exponent (n L ) appears to be above 0.1 which suggests well-developed
              meso/micropores. The pores are narrowly distributed. For type “M,” with
              complex multiporosity feature, the initial imbibition rate becomes lower,
              which suggests microfractures are embedded in the rock matrix, representing
              good-connected microfractures to poor-connected matrix pores. The macro-
              pores, mesopores, and micropores are developed, and the pore size distribu-
              tion is of the multipeak type. Here macropores are >50 nm diameter,
              mesopores are between 2 and 50 nm, and micropores are <2 nm, according
   303   304   305   306   307   308   309   310   311   312   313