71
The immobilization of the bubble surface is obviously not the determining factor,
and other physical processes might influence the rise velocity.
A reason for different rise velocities of equally sized bubbles has been given by
Tomiyama et al. (2002). They explained that the initial conditions during the bubble
formation force the bubbles on either a zigzag or a helical rise path. Bubbles rising
on the helical path are much faster than those on the zigzag path because the latter
are accelerating and decelerating repeatedly, while those on the helical path are
moving with a more or less constant velocity. Initial shape deformations lead to different sphericities of the bubbles. Nearly spherical bubbles show an acceleration
and deceleration behavior and are therefore quite slow. More ellipsoidal bubbles (of
the same volume-equivalent diameter) rise on a helical trajectory and are consequently much faster. Due to the random intensity of bubble deformation, the measured rise velocities are distributed over a quite large range (Fig. 5.1). The results of
Laqua et al. (2016) agree very well with the correlation between sphericity and
terminal rise velocity given by Tomiyama et al. (2002). It can therefore be concluded that the initial shape deformation and the resulting rise trajectories of the
bubbles are the governing factors for the terminal rise velocities (Laqua et al. 2016).
During an actual blowout, the trajectories of the bubbles are influenced by the high
level of turbulence and collisions with other bubbles and droplets. However, either
a helical or a zigzag trajectory will establish once the bubbles leave the highturbulence high-particle-density regime. Again, the well-known correlations for
clean and contaminated systems seem to provide a good estimate for the upper and
lower boundaries of the bubble rise velocities, respectively.
5.4 Oil Droplet Behavior: Theoretical and Experimental
Insights
For the investigation of the rise velocities of crude oil droplets, a high-pressure
experimental facility with a volume of 35 mL and optical access via a windowed
slot is applied at Hamburg University of Technology (TUHH) in Hamburg, Germany.
The droplets are generated using a 1/16″ capillary with an inner diameter d N of
0.18 mm leading to spherical to slightly ellipsoidal droplets with volume-equivalent
diameters between 1.8 and 2.5 mm. During their ascent the droplets are captured
with a high-speed camera against the light. Deionized water and artificial seawater
are used as continuous phase, while Louisiana Sweet Crude (LSC) oil serves as
dispersed phase. The temperature of 20 °C and pressure levels between 0.1 and
15.1 MPa are monitored to ensure reproducible experimental conditions. From the
gray value distribution of the images, the droplet displacement over time is detected
automatically using a self-written Java script, and the terminal rise velocities of the
bubbles are calculated. All results are in good accordance with the empirical correlation of Grace et al. (1976), which can therefore be applied for the calculation of
crude oil droplet rise velocities in a quiescent water phase. The correlation slightly
5 Behavior of Rising Droplets and Bubbles: Impact on the Physics of Deep-Sea…
The immobilization of the bubble surface is obviously not the determining factor,
and other physical processes might influence the rise velocity.
A reason for different rise velocities of equally sized bubbles has been given by
Tomiyama et al. (2002). They explained that the initial conditions during the bubble
formation force the bubbles on either a zigzag or a helical rise path. Bubbles rising
on the helical path are much faster than those on the zigzag path because the latter
are accelerating and decelerating repeatedly, while those on the helical path are
moving with a more or less constant velocity. Initial shape deformations lead to different sphericities of the bubbles. Nearly spherical bubbles show an acceleration
and deceleration behavior and are therefore quite slow. More ellipsoidal bubbles (of
the same volume-equivalent diameter) rise on a helical trajectory and are consequently much faster. Due to the random intensity of bubble deformation, the measured rise velocities are distributed over a quite large range (Fig. 5.1). The results of
Laqua et al. (2016) agree very well with the correlation between sphericity and
terminal rise velocity given by Tomiyama et al. (2002). It can therefore be concluded that the initial shape deformation and the resulting rise trajectories of the
bubbles are the governing factors for the terminal rise velocities (Laqua et al. 2016).
During an actual blowout, the trajectories of the bubbles are influenced by the high
level of turbulence and collisions with other bubbles and droplets. However, either
a helical or a zigzag trajectory will establish once the bubbles leave the highturbulence high-particle-density regime. Again, the well-known correlations for
clean and contaminated systems seem to provide a good estimate for the upper and
lower boundaries of the bubble rise velocities, respectively.
5.4 Oil Droplet Behavior: Theoretical and Experimental
Insights
For the investigation of the rise velocities of crude oil droplets, a high-pressure
experimental facility with a volume of 35 mL and optical access via a windowed
slot is applied at Hamburg University of Technology (TUHH) in Hamburg, Germany.
The droplets are generated using a 1/16″ capillary with an inner diameter d N of
0.18 mm leading to spherical to slightly ellipsoidal droplets with volume-equivalent
diameters between 1.8 and 2.5 mm. During their ascent the droplets are captured
with a high-speed camera against the light. Deionized water and artificial seawater
are used as continuous phase, while Louisiana Sweet Crude (LSC) oil serves as
dispersed phase. The temperature of 20 °C and pressure levels between 0.1 and
15.1 MPa are monitored to ensure reproducible experimental conditions. From the
gray value distribution of the images, the droplet displacement over time is detected
automatically using a self-written Java script, and the terminal rise velocities of the
bubbles are calculated. All results are in good accordance with the empirical correlation of Grace et al. (1976), which can therefore be applied for the calculation of
crude oil droplet rise velocities in a quiescent water phase. The correlation slightly
5 Behavior of Rising Droplets and Bubbles: Impact on the Physics of Deep-Sea…