Modeling the Release, Spreading, and Burning of LNG, LPG, and Gasoline on Water. David W. Johnson and John B. Cornwell

Transcription

1 Modeling the Release, Spreading, and Burning of LNG, LPG, and on Water David W. Johnson and John B. Cornwell Printed in Journal of Hazardous Materials Vol. 140, Issue 3: pp February 0, 007 Quest Consultants Inc th Avenue N.W. Noran, Oklahoa Telephone: Fax: E-ail: info@questconsult.co URL: QUEST

2 Modeling the Release, Spreading, and Burning of LNG, LPG, and on Water David W. Johnson* and John B. Cornwell Quest Consultants Inc., 908 6th Avenue NW, Noran, OK USA Current interest in the shipent of liquefied natural gas (LNG) has renewed the debate about the safety of shipping large volues of flaable fuels. The size of a spreading pool following a release of LNG fro an LNG tank ship has been the subject of nuerous papers and studies dating back to the id-1970s. Several papers have presented idealized views of how the LNG would be released and spread across a quiescent water surface. There is a considerable aount of publicly available aterial describing these idealized releases, but little discussion of how other flaable fuels would behave if released fro siilar sized ships. The purpose of this paper is to deterine the whether the odels currently available fro the United States Federal Energy Regulatory Coission (FERC) can be used to siulate the release, spreading, vaporization, and pool fire ipacts for aterials other than LNG, and if so, identify which aterial-specific paraeters are required. The review of the basic equations and principles in FERC s LNG release, spreading, and burning odels did not reveal a critical fault that would prevent their use in evaluating the consequences of other flaable fluid releases. With the correct physical data, the odels can be used with the sae level of confidence for aterials such as LPG and gasoline as they are for LNG. 1.0 Introduction Current interest in the shipent of liquefied natural gas (LNG) has renewed the debate about the safety of shipping large volues of flaable fuels. The size of a spreading pool of LNG following a release fro an LNG tank ship has been the subject of nuerous papers and studies dating back to the id-1970s [1,, 3, 4]. In 004, the Federal Energy Regulatory Coission (FERC) contracted with ABS Consulting to identify appropriate consequence analysis ethods for estiating flaable vapor and theral radiation hazard distances for potential releases of LNG fro tank ships during transit and while at berth. The results of the study were initially published on May 13, 004. After review by FERC staff and public coents, a revised consequence analysis ethodology was issued by FERC on June 18, 004. On June 9, 004, the NOTICE OF AVAILABILITY OF DETAILED COMPUTATIONS FOR THE CONSEQUENCE ASSESSMENT METHODS FOR INCIDENTS INVOLVING RELEASES FROM LIQUEFIED NATURAL GAS CARRIERS, Docket No. AD , was issued [5]. Since June 9, 004, these coputational ethods have becoe the de facto standard for evaluating the release, spreading, and pool fire hazards presented by the arine transport of LNG in the United States. Copyright 006, Quest Consultants Inc., 908 6th Avenue N.W., Noran, Oklahoa 73069, USA. All rights reserved. Copyright is owned by Quest Consultants Inc. Any person is hereby authorized to view, copy, print, and distribute docuents subject to the following conditions. 1. Docuent ay be used for inforational purposes only.. Docuent ay only be used for non-coercial purposes. 3. Any docuent copy or portion thereof ust include this copyright notice. -1- QUEST

3 There has been uch public coent regarding the odels recoended by the study. In general, the odels suffer fro a lack of large-scale validation. This is not an uncoon proble for any of the coputational ethods that can be used to perfor consequence analysis for LNG releases. To date, no large-scale (i.e., larger than 35 diaeter pools) LNG release, pool spreading, dispersion, or fire testing experients have been perfored. LNG has been transported by ship for the past forty years. During this sae tie period, shipents of other flaable fuels, such as liquefied petroleu gas (LPG), crude oil, gasoline, and diesel fuel, have also been ade, and with uch higher frequencies. In any cases, the ships that transport these flaable fuels are as large or larger than LNG ships currently in use (18,000 to 145,000 3 ). As in the case of large-scale LNG experients on water, there have not been any large-scale experients involving LPG, gasoline, or other flaable fuels on water. While reviewing the liquid release, spreading, and burning odels incorporated into the FERC LNG odel, it is worth investigating whether this odel can be odified so that it can be used to predict the consequences associated with releases of other flaable fuels on water..0 Brief Discussion of the FERC Consequence Analysis Models The FERC consequence analysis odels copute the rate of release fro a hole in a containent vessel, the subsequent spreading of the released liquid, the evaporation of the liquid, and, if the liquid is ignited, fire radiation resulting fro the flae colun. The odel writes out a tie-vapor rate-radius file suitable for use by the DEGADIS vapor dispersion progra, but the FERC odel does not link to the DEGADIS progra..1 Release Rate The rate of release of LNG fro a containent vessel is coputed using the following orifice forula. Q= C ρ A gh (1) d l h where : Q = ass flow rate, kg / s Cd = dischargecoefficient, assued tobe ρl = density of LNG, kg / Ah = area of hole, g = gravitational constant, 9.81 / s H = liquid height above hole, Equation (1) iplies a circular sooth-edged hole and a non-flashing liquid at the orifice. While it can be argued that a hull breach in a arine LNG carrier would likely not be circular or sooth-edged, the actual shape of the breach cannot be definitively deterined beforehand. Thus, the circular sooth-edged hole is a reasonable starting point for flow calculations. A ore serious liitation to Equation (1) is the neglection of the ullage pressure in the tank. A typical ullage pressure of 0.14 kg/c ( psi) can increase the effective liquid height by QUEST

4 . Liquid Spreading The spreading of LNG on a water surface follows the technique developed by Webber [6]. The governing equation is: [ ] = r Φ r/ t 4 g h/ r C () where : r = pool radius, t = tie, s gr = g( ρw ρl) / ρw, / s 3 ρw = density of liquid substrate, kg / Φ = coefficient that is a function of h f / h h f = pool height at leading edge, h = ean pool height, C = frictional resistance force, / s f f As originally proposed, the frictional resistance ter, C f, was for a liquid in direct contact with the substrate. This assuption was questioned since it is known that LNG fors a thin vapor fil when in contact with a war surface. The frictional resistance ter was eventually changed in the final ABS report (June, 004) to account for the foration of a low friction vapor fil. Solution of Equation () involves an iterative approach, adding and subtracting LNG fro the pool at each tie step due to the spillage and vaporization that is taking place. As currently ipleented, Equation () is useful only for cryogens that for a vapor fil. For liquids that do not for vapor fils, the original developent by Webber [6] can be used..3 Liquid Vaporization on Open Water Liquid is reoved fro the LNG pool by vaporization. Vaporization occurs due to two echaniss: 1) heat transfer fro the substrate surface to the pool and ) heat transfer fro the flae to the pool if the pool is ignited. Several theoretical odels for coputing the heat transfer fro a large water surface are explored in the ABS report. However, the FERC odel assues a constant value of heat transfer for LNG based on whether or not the pool is burning. For a non-burning LNG pool, a constant heat transfer flux of 85 kw/ (0.167 kg/ -s) is used. For a burning LNG pool, a constant heat transfer flux of 143 kw/ (0.8 kg/ -s) is used. The burning vaporization flux includes heat transfer fro the liquid substrate and fro the flae. These values were chosen based on a review of available inforation..4 Fire Radiation The FERC odel coputes fire radiation fro a burning LNG pool using a solid flae odel. Radiation fro the flae to a receptor can be coputed using the following relationship. -3- QUEST

5 q= τ F E (3) where : q τ = incident radiant flux, kw / = atospheric transissivity, the fraction of energy leaving the flae that reaches the receptor F = view factor between flae and receptor E = surface eissive power of flae, kw / For LNG, E is defined as 65 kw/ in the odel. The atospheric transissivity, τ, is coputed with: [ P x] 0.09 τ =.0 / (4) water where : P = partial pressure of water vapor in air, pascals water x = line of sight distance fro point on flae to receptor, The view factor, F, is found using the following forula. ( ) ( ) F = cos β cos β / π d da (5) 1 1 S where : β and β = angles between line joining flae and receptor and 1 line noral to flae surface or receptor surface d = distance between flae surface eleent and receptor, da 1 = sall area on flae surface, The integral found in Equation (5) is evaluated over the entire surface of the flae. The physical diensions of the flae (flae length, clear length of flae, flae tilt angle, etc.) are coputed using the ethods developed by Rew [7] and AIChE (000) [8] as outlined in the revised consequence analysis ethodology issued by FERC on June 18, 004. The ethodology suffers fro a lack of validation for large LNG fires. The largest fire for which experiental data is available is approxiately 35 in diaeter [9]. Further, the algorith for deterining the flae length predicts that the length increases at a rate proportional to D Inforation Required to Predict the Release, Spread, and Burning Behavior of Liquids Using the FERC Models. Using published inforation regarding the FERC odels, Table.1 suarizes the aterial properties required for the use of these odels. The FERC odels were used to calculate the hazard zones associated with releases of LNG (as CH 4 ), LPG (as C 3 H 8 ), and gasoline (as n-octane, C 8 H 18 ). Since the copositions of the three selected aterials vary fro source to source, pure coponent properties were used in the calculations. Properties for these three aterials are presented in Table QUEST

6 Table.1 Specific Material Inforation Required for FERC Models Property Description Units Used to Copute CH carbon to hydrogen ratio none fire radiation E s flae surface flux kw/ fire radiation h fg liquid heat of vaporization J/kg liquid spreading k v theral conductivity of vapor W/( K) liquid spreading b burning vaporization flux kg/( s) liquid spreading, fire radiation w olecular weight kg/kg-ol liquid spreading, fire radiation δ vapor fil thickness for cryogenic liquids liquid spreading µ l viscosity of liquid Pa s liquid spreading µ v viscosity of vapor Pa s liquid spreading ρ v density of vapor kg/ 3 fire radiation ρ l density of liquid kg/ 3 release rate, liquid spreading σ liquid surface tension N/ liquid spreading Table. Material Properties Used in Calculations Value Property Units CH 4 C 3 H 8 n-octane CH none E s kw/ h fg 509,33. 45, ,60. J/kg k v W/( K) b kg/( s) w kg/kg-ol δ 6.3e e-05* 3.e-05* 0.0* 0.0 µ l 1.e-04.0e e-04 Pa s µ v 4.4e e e-06 Pa s ρ v kg/ 3 ρ l kg/ 3 σ N/ * range of vapor fil thickness used to evaluate the effect of this variable. For liquids with fil thickness of 0.0, the initial ABS/FERC report (May 13, 004) spreading odel was used in this evaluation -5- QUEST

7 Material properties were deterined at one atosphere pressure and the noral boiling point of each aterial. n-octane properties were deterined at one atosphere pressure and 91.4 K (65 F). Theral conductivities were evaluated at the average teperature between the noral boiling point and the assued seawater teperature of 91.4 K (65 F). CH 4 (LNG) aterial properties were the sae as used by FERC. 3.0 Results Calculations were ade using the FERC odels suarized in Section for two release scenarios. For each release scenario, the aterial properties listed in Table. were used. Releases were assued to occur over seawater at 91.4 K (65 F). Pool spreading was assued to occur in a radial direction without obstacles (i.e., circular pools). 3.1 Scenario 1 - Equal Release Volues In this scenario, the liquid is assued to be spilling fro a cargo tank containing 5,000 3 of liquid, of which 1,500 3 of the subject liquid is above the waterline. In all cases, the liquid pool was assued to ignite upon release. The liquid head above the waterline was assued to be 13. A 1- hole at the waterline was assued. For the C 3 H 8 releases, vapor fil thicknesses of 6.3x10-5, 3.x10-5, and 0.0 were used to deonstrate the odel sensitivity to this paraeter. Table 3.1 suarizes the results of the coputations. Ties, radii, and distances in all tables are rounded to the nearest 5 (s,, ). Figure 3.1 shows the ass release rate for each aterial. As would be expected for the case of equal volues and equal liquid heads, the ass rate of release of gasoline, the highest density fluid, is greater than for LPG or LNG. The underlying assuption that the flaable aterial is ignited upon release results in the radius of the burning pool varying with tie, as shown in Figure 3.. The burning gasoline pool grows to a larger axiu diaeter than the LNG and LPG pools. This is priarily due to the lower vaporization rate of the gasoline when copared to LNG and LPG. For equal volue releases, the gasoline pool grows to a pool with twice the diaeter of the LNG pool. As can be seen fro the results in Table 3.1 and Figure 3.3, the choice of the LPG vapor fil thickness does not significantly influence the results. Figure 3.4 shows the radiant theral flux vs. distance for each aterial. These calculations were ade at each pool s axiu radius. Reviewing Figures 3. and 3.4 and the odel inputs show that the cobination of larger pool radius, lower flae height, and lower radiant surface flux for gasoline results in a siilar radiant ipact as the LNG and LPG pool fires with their saller pool radii, taller flae heights, and higher radiant surface fluxes. Using 5 kw/ as a coon radiant ipact liit, Table 3.1 shows that the gasoline and LNG radiant ipacts extend 60 eters fro the center of the pool, while the LPG ipacts are slightly saller. This analysis shows that for equal volues of these flaable aterials, the ipacts fro an expanding burning pool are nearly identical. -6- QUEST

8 LNG LPG Release Rate, kg/sec Tie, sec Figure 3.1 Equal Volue Releases Release Rate vs. Tie for 1- Diaeter Hole LNG LPG (3.e-05 fil thickness) 15 Pool Radius, Tie, s Figure 3. Equal Volue Releases Pool Radius vs. Tie for 1- Diaeter Hole -7- QUEST

9 e-05 3.e-05 No fil 15 Pool Radius, Tie, sec Figure 3.3 Equal Volue Releases Pool Diaeter vs. Tie for LPG at Different Fil Thicknesses LNG LPG 5 Flux, kw/ Distance fro Center, Figure 3.4 Equal Volue Releases Radiant Flux vs. Distance for 1- Diaeter Hole -8- QUEST

10 3. Representative Cargo Releases This scenario is based on tank ship copartents that are ore representative of actual tank ships in use today: a 15,000 3 LNG ship with 5,000 3 cargo tanks for LNG (of which 1,500 3 of the liquid is above the waterline), an 80,000 3 insulated LPG ship with 0,000 3 cargo tanks for LPG (of which 7,415 3 of the liquid is above the waterline), and an 85,000 3 Panaax refined products tank ship with 8,000 3 cargo tanks for gasoline (of which,460 3 of the liquid is above the waterline). Due to differing liquid densities, these assuptions result in an initial liquid height above the hole of 13 for LNG, 9.6 for LPG, and 8 for gasoline. For each cargo tank, a 1- hole at the waterline was assued. For the C 3 H 8 releases, a vapor fil thicknesses of 3.x10-5 was used. Table 3. suarizes the results of the coputations. Ties, radii, and distances in all tables are rounded to nearest 5 (s,, ). The trade offs apparent in the equal volue releases are present in the representative cargo releases. A review of Table 3. and Figure 3.5 shows the radius of the LNG pool is saller than the radius of the gasoline pool, even though the total volue of LNG released is five ties larger. This is partly due to the nature of the spreading odel. The gasoline pool continues to spread and burn until the fuel is exhausted. The LNG and LPG pools spread and burn in a siilar anner, but the fil thickness algorith in the FERC odel allows the pool to shrink instead of breaking up as discussed by Otteran []. A review of Table 3. and Figure 3.6 shows that a,460 3 release of gasoline and a 1,500 3 release of LNG result in siilar radiant ipacts for the higher flux levels. When reviewing the results for the lower (i.e., 5 kw/ ) radiant flux level, the ipacts fro the LNG fire extend further. This is priarily due to the taller flae height associated with the LNG fire. 4.0 Suary A review of the basic equations and principles in FERC s LNG release, spreading, and burning odels did not reveal a critical fault that would prevent using the to evaluate the consequences of other flaable fluid releases. With the correct physical data, the odels can be used with the sae level of confidence for aterials such as LPG and gasoline as they are for LNG. The liitations identified in the FERC odel also apply to the other aterials. The ost iportant result of this review is the identification of the behavior of flaable fluids under siilar circustances using the sae odeling tools. The variation in spreading rates, vaporization rates, and flae heights for three coonly shipped flaable aterials show that the ore volatile aterial, LNG, will not produce a significantly larger radiant ipact than a saller release of a lower volatility aterial such as gasoline when releases fro typical cargo containers under siilar conditions are evaluated. A ore detailed study considering other factors including hole size, cargo coposition and inventory ay yield different results. -9- QUEST

11 Table 3.1 Equal Release Volue Results for 1-Meter Diaeter Hole Material Fil Thick. () Total Volue Released ( 3 ) Tie to Epty (s) Tie to Evaporate All Liquid (s) Max. Radius () Distance in eters to Radiant Flux in kw/ at a Wind Speed of 9 /s LNG (CH 4 ) 6.3x10-5 1,500 3,070 3, x10-5 1,500 3,070 3, LPG (C 3 H 8 ) 3.x10-5 1,500 3,070 3, ,500 3,070 3, (C 8 H 18 ) 0.0 1,500 3,070 3, QUEST Material Fil Thick. () Table 3. Representative Cargo Copartent Results for 1-Meter Diaeter Hole Total Volue Released ( 3 ) Tie to Epty (s) Tie to Evaporate All Liquid (s) Max. Radius () Distance in eters to Radiant Flux in kw/ at a Wind Speed of 9 /s LNG (CH 4 ) 6.3x10-5 1,500 3,070 3, LPG (C 3 H 8 ) 3.x10-5 7,415,115, (C 8 H 18 ) 0.0,

12 LNG LPG 15 Pool Radius, * * Inventory Depleted Tie, s * Figure 3.5 Representative Cargo Volues Pool Radius vs. Tie for 1- Diaeter Hole LNG LPG 5 Flux, kw/ Distance fro Center, Figure 3.6 Representative Cargo Volues Radiant Flux vs. Distance for 1- Diaeter Hole -11- QUEST

13 References [1] D.R. Blackore, J.A. Eyre, and G.G. Suers, Dispersion and Cobustion Behaviour of Gas Clouds Resulting fro Large Spillages of LNG and LPG onto the Sea. Transactions of the Institute of Marine Engineers (TM), Vol. 94, Paper 9, 198 [] B. Otteran, Analysis of Large LNG Spills on Water Part 1: Liquid Spread and Evaporation. Cryogenics, August, [3] P. K. Raj, Models for Cryogenic liquid Spill Behavior on Land and Water. Journal of Hazardous Materials, Vol. 5, [4] P. Shaw and F. Briscoe, Evaporation fro Spills of Hazardous Liquids on Land and Water. Safety and Reliability Directorate (SRD), United Kingdo Atoic Energy Authority, SRD R 100, May, [5] U.S. Federal Energy Regulatory Coission (FERC), Notice of Availability of Detailed Coputations for the Consequence Assessent Methods for Incidents Involving Releases fro Liquefied Natural Gas Carriers, published June 9, 004, FERC Docket No. AD [6] C.J.H. van den Bosch and R.A.P.M Weterings, eds., Methods for the Calculation of Physical Effects (TNO Yellow book), TNO, The Hague, The Netherlands, Third edition [7] P. J. Rew and W. G. Hulbert, Developent of Pool Fire Theral Radiation Model, HSE Contract Research Report No. 96/1996, Health and Safety Executive, [8] Center for Cheical Process Safety of the Aerican Institute of Cheical Engineers, Guidelines for Cheical Process Quantitative Risk Analysis, Second Edition, ISBN X, New York, 000. [9] P.J. Waite, R.J. Whitehouse, and E.B. Winn, The Spread and Vaporization of Cryogen Liquids on Water, Journal of Hazardous Materials, Vo. 8, pp , QUEST