Academia.eduAcademia.edu

Figure 4 – uploaded by Luqman Buba Umdagas

See full PDFdownloadDownload figure
Figure 4. Model of heat transfer to LNG boiling on water (a) start of boil-off (b) on set of film boiling [3] sizes. Notwithstanding, this assumption is undoubtedly true particularly in the case when the release time is negligible compared to the spreading time [3]. LNG was modelled as pure liquid methane, which upon release, was assumed to be at its boiling point, the pressure was assumed atmospheric, and water temperature was assumed to be 293K through out the period of vaporization, as the unconfined body of water is considerably large with adequate mixing, hence sufficient amount of heat is supplied to the cryogen. The amount of LNG released is 50,000 kg. The model for pool spreading during an LNG spill on water includes a relative density term showing that the LNG layer displaces the water downward to some extent as expressed in Equation 1 [8].

Figure 4 Model of heat transfer to LNG boiling on water (a) start of boil-off (b) on set of film boiling [3] sizes. Notwithstanding, this assumption is undoubtedly true particularly in the case when the release time is negligible compared to the spreading time [3]. LNG was modelled as pure liquid methane, which upon release, was assumed to be at its boiling point, the pressure was assumed atmospheric, and water temperature was assumed to be 293K through out the period of vaporization, as the unconfined body of water is considerably large with adequate mixing, hence sufficient amount of heat is supplied to the cryogen. The amount of LNG released is 50,000 kg. The model for pool spreading during an LNG spill on water includes a relative density term showing that the LNG layer displaces the water downward to some extent as expressed in Equation 1 [8].

Related Figures (12)

Natural gas is comprised primarily of methane with up to about 10% ethane and lesser quantities of nitrogen, propane, and higher aliphatic hydrocarbons (Figure 1). The boiling temperature at one bar is in the range of 111 K (depending upon composition) and the density is about 420-470 kg/m? [2]. When natural gas is produced from the earth, it includes many other molecules, like ethane (used for manufacturing), propane (which we commonly used for backyard grills) and butane (used in lighters). We can find natural gas in Nigeria and around the world by exploring for it in the earth’s crus and then drilling wells to produce it. The composition of natural gas varies considerably depending on the origin of the gas and geographical location, as well as geological structure of the gas reserve [3].
Natural gas is comprised primarily of methane with up to about 10% ethane and lesser quantities of nitrogen, propane, and higher aliphatic hydrocarbons (Figure 1). The boiling temperature at one bar is in the range of 111 K (depending upon composition) and the density is about 420-470 kg/m? [2]. When natural gas is produced from the earth, it includes many other molecules, like ethane (used for manufacturing), propane (which we commonly used for backyard grills) and butane (used in lighters). We can find natural gas in Nigeria and around the world by exploring for it in the earth’s crus and then drilling wells to produce it. The composition of natural gas varies considerably depending on the origin of the gas and geographical location, as well as geological structure of the gas reserve [3].
Liquefied natural gas (LNG) is natural gas that has been cooled to the point that it condenses to a liquid, which occurs at a temperature of approximately -256°F (-161°C) and at atmospheric pressure through a complex cryogenic process called liquefaction. Liquefaction reduces the volume by approximately 600 times thus making it more economical to transport between continents in specially designed ocean vessels, whereas traditional pipeline transportation systems would be less economically attractive and could be technically or politically infeasible. Figure 2 illustrates the economic advantage of LNG transport over other options. Figure 2. Natural gas transportation costs, modified from [3]
Liquefied natural gas (LNG) is natural gas that has been cooled to the point that it condenses to a liquid, which occurs at a temperature of approximately -256°F (-161°C) and at atmospheric pressure through a complex cryogenic process called liquefaction. Liquefaction reduces the volume by approximately 600 times thus making it more economical to transport between continents in specially designed ocean vessels, whereas traditional pipeline transportation systems would be less economically attractive and could be technically or politically infeasible. Figure 2 illustrates the economic advantage of LNG transport over other options. Figure 2. Natural gas transportation costs, modified from [3]
Upon delivery to a final destination the LNG is pumped into either another cryogenic storage facility or is channelled to a regasification unit where it is vaporized back to gaseous form, and subsequently pumped through pipeline system that leads to an ultimate market, as illustrated in the LNG value chain in Figure 3. Thus, LNG technology makes natural gas available throughout the world [4]. Currently, imported LNG is commonly 95% - 97% methane, with the remainder a combination of ethane, propane, and other heavier gases and it varies with location around the globe [5]. LNG vapour is colourless, odourless, and non-toxic, and is considered flammable. LNG vapour typically appears as a visible white cloud, because its cold temperature condenses water vapour present in the atmosphere. The lower and upper flammability limits of methane are 5.5% and 14% by volume at a temperature of 25°C [6].
Upon delivery to a final destination the LNG is pumped into either another cryogenic storage facility or is channelled to a regasification unit where it is vaporized back to gaseous form, and subsequently pumped through pipeline system that leads to an ultimate market, as illustrated in the LNG value chain in Figure 3. Thus, LNG technology makes natural gas available throughout the world [4]. Currently, imported LNG is commonly 95% - 97% methane, with the remainder a combination of ethane, propane, and other heavier gases and it varies with location around the globe [5]. LNG vapour is colourless, odourless, and non-toxic, and is considered flammable. LNG vapour typically appears as a visible white cloud, because its cold temperature condenses water vapour present in the atmosphere. The lower and upper flammability limits of methane are 5.5% and 14% by volume at a temperature of 25°C [6].
It could be seen from the spreading profiles in Figure 5 that when 50,000 kg of LNG is spilled on water surface, the at radius continues to increases until it reaches a maximum about 95 s corresponding to a radius of 80 m. Simultaneously with the pool spreading, the spilled cryogen undergoes boiling and vaporization processes due to the heat transfer from the underlying water. Consequently, the amount of the cryogen spilled continues to decrease until it has all vaporized in about 120 s as shown on the vaporization rate profile in Figure 6. This, while neglecting all external disturbances, like the effects of wind current and wave actions. Furthermore, the turbulence effects were also neglected here. The result obtained here was in congruence with the works of [10] and [13]; the former spi me sec led 50,000 kg and the latter spilled 70,000 kg of liquid thane, the liquid pool completely vaporized in about 120 onds and 170 s respectively. Figure 6. Rate of LNG pool vaporization
It could be seen from the spreading profiles in Figure 5 that when 50,000 kg of LNG is spilled on water surface, the at radius continues to increases until it reaches a maximum about 95 s corresponding to a radius of 80 m. Simultaneously with the pool spreading, the spilled cryogen undergoes boiling and vaporization processes due to the heat transfer from the underlying water. Consequently, the amount of the cryogen spilled continues to decrease until it has all vaporized in about 120 s as shown on the vaporization rate profile in Figure 6. This, while neglecting all external disturbances, like the effects of wind current and wave actions. Furthermore, the turbulence effects were also neglected here. The result obtained here was in congruence with the works of [10] and [13]; the former spi me sec led 50,000 kg and the latter spilled 70,000 kg of liquid thane, the liquid pool completely vaporized in about 120 onds and 170 s respectively. Figure 6. Rate of LNG pool vaporization
The comparison of the mass vaporised shown in Figure 7 shows good agreement between the numerical simulation and the experimental values for all runs. The maximum deviations of the simulations results obtained compared to the experimental results were 7.0%, 10% and 12% for R-47, R-46 and R-44 respectively. Figure 7. Comparison of simulation runs using turbulence factors with Valencia-Chavez [2] experimental data
The comparison of the mass vaporised shown in Figure 7 shows good agreement between the numerical simulation and the experimental values for all runs. The maximum deviations of the simulations results obtained compared to the experimental results were 7.0%, 10% and 12% for R-47, R-46 and R-44 respectively. Figure 7. Comparison of simulation runs using turbulence factors with Valencia-Chavez [2] experimental data
The results of the model simulation of this work that accounts for effects of turbulence was compared with the experimental values from [2]. The experimental values tremendously uunderestimates mass boil-off compared to the numerical simulation of this work as can be seen from Figure 9. This can be attributed to the fact that the equations used to determine the heat transfer coefficient are in quiescent conditions and as such the heat flux reaching the water surface was low. Consequently, that under underestimates the heat transfer to the LNG pool. It can also be looked
The results of the model simulation of this work that accounts for effects of turbulence was compared with the experimental values from [2]. The experimental values tremendously uunderestimates mass boil-off compared to the numerical simulation of this work as can be seen from Figure 9. This can be attributed to the fact that the equations used to determine the heat transfer coefficient are in quiescent conditions and as such the heat flux reaching the water surface was low. Consequently, that under underestimates the heat transfer to the LNG pool. It can also be looked
incorporating the turbulence in runs R-46, R-47, R-6 [2]. t can be seen from Figure 8 that considering the effects of turbulence, for R-46, the cryogen pool evaporated in 40 s, giving about 200% overestimation of evaporation time if turbulence affects were neglected (120 s). Similarly, for both R-6 and R-47, the pool evaporated in about 60 s, giving about 100% overestimation of evaporation time for both runs. Figure 8. Turbulence factors for Valencia Chavez experimental runs
incorporating the turbulence in runs R-46, R-47, R-6 [2]. t can be seen from Figure 8 that considering the effects of turbulence, for R-46, the cryogen pool evaporated in 40 s, giving about 200% overestimation of evaporation time if turbulence affects were neglected (120 s). Similarly, for both R-6 and R-47, the pool evaporated in about 60 s, giving about 100% overestimation of evaporation time for both runs. Figure 8. Turbulence factors for Valencia Chavez experimental runs
Boe’s [16] experiment is simi ar to that of Valencia Chavez [2], in that both made significant efforts to eliminate turbulence effects. The the experimental series is 1.5 s. In Boe's experimental runs were selec kg To = 298K) and LCH-15 (Mo which are fairly analogous to Val average spill time of the present work, two of ted: LCH-13 (Mo =0.29 = 0.28 kg, To = 314); encia-Chavez's [2] runs that was simulated in terms of spilled, but only differ with the amount of cryogen higher initial water temperatures of 298 K and 313.5 K respectively. The two selected experimental runs were simulated using the resultant turbulence factors, and the results were then compared with the experimental values. Both experimental conditions were simulated using the corresponding turbulence factors. The results were compared with the experimental values as shown in Figure 10. Figure 10. Mass of liquid cryogen remaining with time for LCH13
Boe’s [16] experiment is simi ar to that of Valencia Chavez [2], in that both made significant efforts to eliminate turbulence effects. The the experimental series is 1.5 s. In Boe's experimental runs were selec kg To = 298K) and LCH-15 (Mo which are fairly analogous to Val average spill time of the present work, two of ted: LCH-13 (Mo =0.29 = 0.28 kg, To = 314); encia-Chavez's [2] runs that was simulated in terms of spilled, but only differ with the amount of cryogen higher initial water temperatures of 298 K and 313.5 K respectively. The two selected experimental runs were simulated using the resultant turbulence factors, and the results were then compared with the experimental values. Both experimental conditions were simulated using the corresponding turbulence factors. The results were compared with the experimental values as shown in Figure 10. Figure 10. Mass of liquid cryogen remaining with time for LCH13
Figure 11. Mass of liquid cryogen remaining with time for LCH15
Figure 11. Mass of liquid cryogen remaining with time for LCH15
4.3.3. Comparison with Zubairu & Velissa’s Numerical Experiment Figure 12. Comparison of model simulation to Zubairu & Velissa’s [13] numerical experiment The most probable source of error responsible for the observed disagreement between the model prediction of experimental set-up. As can be seen through the walls of the boiling ce even evident as the observed misma lw this work and the experiment of Boe [16] could be attributed to heat loss due to poor thermal insulation of their from Figure 10 and 11, the model under predicts the experimental boil-offs. This also might be due to the extra heat supp ied to the cryogen hich led to higher boil-off of the cryogen than the predicted values. This is tch and Boe’s [16] experiment was hig ner between the model for LCH-I3 which the initial temperature of water was 20°C.
4.3.3. Comparison with Zubairu & Velissa’s Numerical Experiment Figure 12. Comparison of model simulation to Zubairu & Velissa’s [13] numerical experiment The most probable source of error responsible for the observed disagreement between the model prediction of experimental set-up. As can be seen through the walls of the boiling ce even evident as the observed misma lw this work and the experiment of Boe [16] could be attributed to heat loss due to poor thermal insulation of their from Figure 10 and 11, the model under predicts the experimental boil-offs. This also might be due to the extra heat supp ied to the cryogen hich led to higher boil-off of the cryogen than the predicted values. This is tch and Boe’s [16] experiment was hig ner between the model for LCH-I3 which the initial temperature of water was 20°C.
Figure 13. Comparison of model simulation to Conrado & Vesovic [10] numerical experiment Conrado & Vesovic [10] studied the influence o hemical composition on vaporisation of L G and LPG yn unconfined water surfaces. Using their numerica xperimental conditions, the mod ensitivity of LNG c f liquid spill was studied. In the id ssumed to complete naintaining constant omposition on the rate o ly boil in the film boiling el was run, and _ the f vaporizatior ealized case, LNG wa: regime while heat transfer coefficien yf changes in the composition of vapour film. hey simulated the s vind action. In the simulation o ‘onditions considered by [10], calm pills with and without th t independen Furthermore e presence 0 f this work using the water surface (no winc ffect) was considered as wind effects and wave action: vere neglected. Figure 13 compares the simulation result o: his work with that of [10]. We can see that the mas: oil-off of LNG in our simulation is higher than that of [10 \dditionally, it could be seen that there was about 30 : lifference in the vaporization time of the spilled LNG.
Figure 13. Comparison of model simulation to Conrado & Vesovic [10] numerical experiment Conrado & Vesovic [10] studied the influence o hemical composition on vaporisation of L G and LPG yn unconfined water surfaces. Using their numerica xperimental conditions, the mod ensitivity of LNG c f liquid spill was studied. In the id ssumed to complete naintaining constant omposition on the rate o ly boil in the film boiling el was run, and _ the f vaporizatior ealized case, LNG wa: regime while heat transfer coefficien yf changes in the composition of vapour film. hey simulated the s vind action. In the simulation o ‘onditions considered by [10], calm pills with and without th t independen Furthermore e presence 0 f this work using the water surface (no winc ffect) was considered as wind effects and wave action: vere neglected. Figure 13 compares the simulation result o: his work with that of [10]. We can see that the mas: oil-off of LNG in our simulation is higher than that of [10 \dditionally, it could be seen that there was about 30 : lifference in the vaporization time of the spilled LNG.
Related topics:
Turbulence modeling
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved