OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties

OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties OIL SPILL SCIENCE chapter 4 – measurement of oil physical properties

Chapter Measurement of Oil Physical Properties Bruce Hollebone Chapter Outline 4.1 Introduction 4.2 Bulk Properties of Crude Oil and Fuel Products 4.3 Hydrocarbon Groups 4.4 Quality Assurance and Control 63 63 73 77 4.5 Effects of Evaporative Weathering on Oil Bulk Properties Appendix 4.1 78 85 4.1 INTRODUCTION During any uncontrolled release of oil, the properties of the spilled oil, including the bulk physical property changes due to weathering, must be immediately available, so that models can be used to predict the environmental impacts of the spill and guide the selection of various remediation alternatives Unfortunately, the properties routinely measured by oil producers and refiners are not the ones that spill responders need to know most urgently Questions important to responders include the following: l l l l l l the physical properties of the oil and how these change over time how the compositional and bulk property changes affect an oil’s behavior and fate whether emulsions will form whether the oil is likely to submerge the hazard to on-site personnel during cleanup the oil toxicity to marine or aquatic organisms 4.2 BULK PROPERTIES OF CRUDE OIL AND FUEL PRODUCTS The physical properties of the almost limitless variety of crude oils are generally correlated with aspects of chemical composition Some of these key Oil Spill Science and Technology DOI: 10.1016/B978-1-85617-943-0.10004-8 Copyright Ó 2011 Elsevier Inc All rights reserved 63 64 PART | III Oil Analysis and Remote Sensing properties for determining the fate and behavior of oil and petroleum products in the environment are viscosity, density, specific gravity (density relative to water), flash point, pour point, distillation, and interfacial tension These properties for the oils are listed in Table 4.1 Viscosity is the resistance to flow in a liquid The lower the viscosity, the more readily the liquid flows The viscosity of an oil is a function of its composition; therefore, crude oil has a wide range of viscosities For example, the viscosity of Federated oil from Alberta is mPa$s, while a Sockeye oil from California is 45 mPa$s at 15 C In general, the greater the fraction of saturates and aromatics and the lower the amount of asphaltenes and resins, the lower the viscosity As oil weathers, the evaporation of the lighter components leads to increased viscosity As with other physical properties, viscosity is affected by temperature, lower temperatures giving higher viscosities For most oils, the viscosity varies approximately exponentially with temperature Oils that flow readily at high temperature can become a slow-moving, viscous mass at low temperature In terms of oil spill cleanup, viscous oils not spread rapidly, not penetrate soils readily, and affect the ability of pumps and skimmers to handle the oil The dynamic viscosity of an oil can be measured by a viscometer using a variety of standard cup-and-spindle sensors at controlled temperatures Density is the mass of a unit volume of oil, usually expressed as grams per millilitre (g/mL) or, equivalently, as kilograms per cubic metre (kg/m3) It is used by the petroleum industry to grade light or heavy crude oils Density is also important because it indicates whether a particular oil will float or sink in water As the density of water is 1.0 g/mL at 15 C and the density of most oils ranges from 0.7 to 0.99 g/mL, oils typically float on water As the density of seawater is 1.03 g/mL, even heavier oils will usually float on it Only a few bitumens have densities greater than water at higher temperatures However, as water has a minimum density at 4 C and oils will continue to contract as temperature decreases, heavier oils, including heavy crudes and residual fuel oils, may sink in freezing waters Furthermore, as density increases as the light ends of the oil evaporate off, a heavily weathered oil, long after a spill event, may sink or be prone to overwashing, where the fresh oil, immediately after the spill, may have floated readily A related measure is specific gravity, an oil’s density relative to that of water As the densities of both water and oil vary differently with temperature, this quantity can be highly variable The American Petroleum Institute (API) uses the specific gravity of petroleum at 50 F (15.56 C) as a quality indicator for oil Pure water has an API gravity of 10 Oils with progressively lower specific gravities have higher API gravities Heavy, inexpensive oils have less than 25 API; medium oils are 25 to 35API; and light commercially valuable oils are 35 to 45API API gravities generally vary inversely with viscosity and asphaltene content Interfacial tensions are the net stresses at the boundaries between different substances They are expressed as the increased energy per unit area (relative to the bulk materials), or equivalently as force per unit length The ‘Standard Chapter | Intermediate Fuel Oil Bunker C Crude Oil Emulsion 1,000 to 15,000 10,000 to 50,000 20,000 to 100,000 0.88 to 1.00 0.94 to 0.99 0.96 to 1.04 30 to 50 10 to 30 10 to 20 to 15 10 to 15 10 to 30 15 to 30 25 to 30 25 to 35 N/A 80 to 100 >100 >80 to 20 >50 Property Units Gasoline Diesel Light Crude Viscosity m.Pa$s 0.5 to 50 Density g/mL 0.72 0.84 0.78 to 0.88 50 to 65 35 to 40 27 27 API Gravity Interfacial Tension mN/m Heavy Crude 50 to 50,000 Flash Point  C À35 55 to 65 À30 to 30 À30 to 60 Pour Point  C N/A À60 À55 to À30 to 30 À10 to 10 0.95 to 1.0 Measurement of Oil Physical Properties TABLE 4.1 Typical Oil and Fuel Properties at 15 C 65 66 PART | III Oil Analysis and Remote Sensing International (SI)’ units for interfacial tension are milliNewtons per meter (mN/ m) Surface tension is thought to be related to the final size of a slick The lower the interfacial tension of oil with water, the greater the extent of spreading and thinner terminal thickness of oil In actual practice, the interfacial tension alone does not apparently account for spreading behavior; environmental effects and other effects seem to be dominant The flash point of an oil is the temperature at which the vapor over the liquid can be ignited A liquid is considered to be flammable if its flash point is less than 60 C Flash point is an important consideration for the safety of spill cleanup operations Gasoline and other light fuels can ignite under most ambient conditions and therefore are a serious hazard when spilled Many freshly spilled crude oils also have low flash points until the lighter components have evaporated or dispersed On the other hand, Bunker C and heavy crude oils generally are not flammable when spilled The pour point of an oil is the temperature at which no flow of the oil is visible over a period of seconds from a standard measuring vessel The pour point of crude oils ranges from À60 C to 30 C Lighter oils with low viscosities generally have lower pour points As oils are made up of hundreds of compounds, some of which may still be liquid at the pour point, the pour point is not the temperature at which an oil will no longer pour The pour point represents a consistent temperature at which an oil will pour very slowly and therefore has limited use as an indicator of the state of the oil For example, waxy oils can have a very low pour point, but may continue to spread slowly at that temperature and can evaporate to a significant degree 4.2.1 Density and API Gravity The density of an oil sample, in g/mL, is best measured using a digital density meter following American Society for Testing and Materials (ASTM) method D 5002.1 The instrument is calibrated using air and distilled, deionized water Acoustically measured densities must be corrected for sample viscosity, as specified by the instrument manufacturer API gravity (API 82) is calculated using the specific gravity of an oil at 60 F (15.56 C).2 The oil density at 15.56 C can be estimated by exponential extrapolation from the higher (THi) and lower (TLo) data points, if necessary This is converted to specific gravity by division by the density of water at 15.5 C, using the following equation: s:g:15:56 ¼ rTHi exp h .À Á THi À TLo i À Á  THi À 15:56 þ In rTHi rðH2 OÞ15:56 In rTHi À In rTLo (1) where s.g.15.56 is the specific gravity of the oil or product at 15.56 C (60 F), rTLo and rTHi are the measured oil densities at TLo and THi, respectively, and Chapter | Measurement of Oil Physical Properties 67 r(H2O)15.56 is the density of water at 15.56 C The API gravity is then determined using the formula (API 82): À Á API ¼ 141:5= s:g:15:56 À 131:5 (2) 4.2.2 Dynamic Viscosity The dynamic viscosity of an oil sample, in mPa$s or cP, is measured using an enclosed spinning cup viscometer using standard NV and SV1 cup-and-spindle sensors.3 Check standards of pure ethylene glycol and glycerine can be conveniently used to validate the NV and SV1 methods, respectively From a qualitative observation of the oil, either the NV or the SV1 sensor is chosen to measure the sample The NV sensor is used for oils with viscosities below 100 mPa$s, and the SV1 sensor, for oils above 70 mPa$s to 10,000 mPa$s For oils with higher viscosity, measurements must be made on cone and plate or parallel plate instruments (see below) For both cases using the rotary viscometer, the measurement cup is filled with a sample to the edge or the rotating surface The sensor is mounted onto the instrument, and the sample volume is adjusted to the proper level The sample is allowed to equilibrate until the sample temperature probe stabilizes at the measurement temperature and remains stable for minutes Samples and sensors are kept chilled at the appropriate temperature prior to use For the NV sensor, the rotational shear rate is set at 1,000/s, the SV1 sensor at 50/s If the oil is observed to be non-Newtonian, single samples are run at shear rates of 1/s, 10/s, and 100/s In all cases, the sensors are ramped up to speed over a period of minutes The viscosity is measured for a subsequent minutes, sampled once per second The viscosity reported is that at time zero of the second, constant-shear rate interval This may be obtained by the mean of the constant-shear rate interval data or by linear fit to the time-viscosity series if friction-heating has occurred during the measurement For Newtonian samples, triplicate measurements are averaged and the mean is reported as the absolute or dynamic viscosity For non-Newtonian samples, viscosities are reported for each of the three shear rates Viscosities above 50,000 mPa$s are measured on a parallel plate rheometer with an air bearing Measurement for most oils can be performed with a 35 mm plate/plate geometry at a gap of mm between plates A stress sweep in forced oscillation mode at Hz performed over an appropriate range will determine the stress independent regions A creep test can then be performed at a stress value selected in the stable “sol” range of flow response for the material This provides the zero shear viscosity value 4.2.3 Surface and Interfacial Tensions Surface and interfacial tensions, in mN/m, are normally determined by one of two methods The de No€ uy ring is a common technique, used by many laboratories, 68 PART | III Oil Analysis and Remote Sensing and has been codified as ASTM method D 971.4 It depends on accurate measurement of the maximum force that a platinum ring can exert on the surface of a liquid before detachment A second emerging technique that shows much promise for improved speed and accuracy is the pendant/rising drop method, which depends on shape calculations of a droplet of oil in air or water.5,6 The values that are important for spill responders include the oil/air, oil/ water, and the oil/seawater interfacial tensions The oil/air interfacial tension is often called surface tension As interfacial tensions are temperature dependent, it is often convenient to determine these quantities for several temperatures Two measurements at freezing, 0 C, and at ambient temperature, 25 C, allow for a wide range of interpolated values Measurement at 50 F/15 C also allows determination of common marine temperatures € Ring Determination of Interfacial Tensions De Nouy A measurement apparatus specific to the de No€ uy ring test is required Manual machines are common, but automated systems are now available that make measurements much quicker and repeatable All measurement equipment, rings, measurement vessels, transfer, and storage containers must be scrupulously clean before measurement Surface and interfacial tension measurements are very sensitive to contamination by organic chemicals or salts For sample/air surface tensions, the instrument is zeroed with the measurement ring in the air A small amount of sample, approximately 15 mL, is poured into a vessel of sufficient diameter that the wall effects on the meniscus not affect the area through which the ring will pass The ring is dipped into the sample to a depth of no more than mm and is then pulled up such that it is just visible on the surface of the liquid The system is allowed to rest for 30 seconds The measurement is initiated, terminating when the upward pulling force on the ring just balances the downward force exerted by the liquid The apparent surface tension, sAPP, is recorded For sample/water and sample/brine interfacial tensions, the ring is zeroed in the sample at a depth of not more than mm The ring is removed and cleaned A volume of water or brine is dispensed into the measurement vessel The ring is dipped mm into the aqueous phase A small volume of sample is carefully poured down the side of the vessel wall, with great care taken so as to disturb the aqueous/oil interface as little as possible The overlying layer should be at least mm thick The ring is then raised to the bottom on the interface, and the system is allowed to rest for exactly 30 seconds The measurement is started, and the apparent interfacial tension is recorded, sAPP, when the force balance is reached The apparent surface tension is corrected for mass of the upper phase lifted by the ring during measurement using the Zuidema and Waters6 correction: s  1:452sAPP 1:679 ỵ 0:04534 (3) s ẳ sAPP 0:7250 ỵ C2 D À dÞ R=r Chapter | Measurement of Oil Physical Properties 69 where s is the interfacial tension, sAPP is the instrument scale reading, C is the ring diameter, D is the density of the lower phase, d is the density of the upper phase, R is the radius of the du No€ uy ring, and r is the radius of the ring wire As these measurements depend on temperature, samples, aqueous phases and glassware should be kept at the measurement temperature for a minimum of 30 minutes before a determination is made Pendant/Rising Drop Determination of Interfacial Tensions In this test, the interfacial tension is determined by calculation with comparison to the shape of a drop hanging from the end of a needle A camera is used to photograph a picture of a drop hanging from a needle The digital picture is analyzed by software; then a parameterized curve shape is developed, from which the surface tension is calculated.6 In the case of a liquideliquid interfacial tension, the surrounding fluid must be clear, so that a good image may be generated For oil in water, this requires that the oil be suspended in water However, as most oils are less dense than water, the rising oil bubble, rather than the pendant drop, must be measured In this case, the image is inverted in software and, instead of the force of gravity, the buoyant force, determined as the fraction of gravity based on the specific gravity of the oil is used: b ẳ grwater roil ị=rwater (4) where b is the buoyant force, g is the acceleration due to gravity, rwater is the density of water at the measurement temperature, and roil is the oil density 4.2.4 Flash Point The flash point of an oil product can be determined by several methods, depending on the oil product and the quantity available Lower viscosity products, including light fuel oils and most fresh crudes, are measured by the Tag closed-cup method This follows ASTM method D 1310.7 Though accurate, the Tag method uses a comparatively large volume of oil, 50 to 70 mL Smaller volumes, 1e2 mL, can be measured by ASTM D6450.8 The practical working range of these two methods is e10 C to approximately 100 C With subambient cooling, using dry ice baths and/or liquid nitrogen baths, much lower flash point temperatures can be measured, but this is often not necessary for emergency response considerations Heavier products, including intermediate and heavy fuel oils, can be measured by a Pensky-Martins analyzer, following ASTM D 93.9 As with the Tag method, this method uses 50e70 mL of crude oil Smaller volumes can be used with the newer method ASTM D7094, which uses only mL of oil.10 The working range for these heavier type tests is approximately 50 C to 225 C 70 PART | III Oil Analysis and Remote Sensing The standard test material for assuring quality control for a lowertemperature flash point apparatus historically has been para-xylene; however, heavier normal alkane standards, n-decane, n-undecane, n-tetradecane, and n-hexadecane have also been found to be suitable and offer a wider range of test temperatures.11 4.2.5 Pour Point The pour point of an oil sample, in degrees Celsius, can only be determined by following ASTM method D 97.12 Sample aliquots are poured into ASTMapproved jars, stopped and fixed with ASTM-certified thermometers The temperature regime described in the standard is critical; particularly in waxy oils, with high normal alkane contents, a crust of waxy crystals can form on the surface of the oil as it cools The ASTM D 97 heating and cooling process for oil is designed to ensure that the formation of these microstructures does not interfere with reproducible measurement of the pour point 4.2.6 Sulphur Content The mass fraction of atomic sulphur in oil is conveniently determined using X-ray fluorescence closely following ASTM method D 4294.13 In brief, the method is as follows: approximately g of oil is weighed out into standard 31 mm XRF cells The sealed cells are then measured in an XRF spectrometer The spectrometer response is calibrated using a series of certified reference material standards Spectra should be corrected for interference by chlorine by subtraction, based on a calibration curve established by the certified reference materials Matrix effects, X-ray absorption by the base oil, can be corrected by subtraction of a spectrum of an oil free of sulphur, such as a mineral or lubricating oil 4.2.7 Water Content The mass fraction of water in oil or an emulsion, expressed as a percentage, is best determined by Karl Fischer titration, using ASTM method D 4377.14 The Karl Fischer reaction is an amine-catalyzed reduction of water in a methanolic solution: CH3 OH ỵ SO2 ỵ RN/ẵRNHỵ ỵ ẵSO3 CH3 2RN ỵ H2 O ỵ I2 ỵ ẵRNHỵ ẵSO3 CH3 /ẵRNHỵ ẵSO4 CH3 ỵ 2ẵRNHỵ IÀ (5) The amine, RN, or mixture of amines is proprietary to each manufacturer An aliquot of approximately g of oil is accurately weighed, then introduced to the reaction vessel of the autotitrator A solution of 1:1:2 (by volume) mixture of methanol:chloroform:toluene is used as a working fluid Chapter | Measurement of Oil Physical Properties 71 4.2.8 Evaluation of the Stability of Emulsions Formed from Brine and Oils and Oil Products Water-in-oil emulsions are formed in 2.2-liter fluorinated vessels on an endover-end rotary mixer at a rotational speed of 50 RPM.15,16 600 mL of salt water (3.3% w/v NaCl) is placed in each mixing vessel 30 mL of oil is added to each vessel for a 1:20 oil:water ratio The vessels are sealed and placed in the rotary mixer such that the cap of each mixing vessel follows, rather than leads, the direction of rotation The rotary mixer is kept in a temperature-controlled cold room at 15 C The vessels and their contents are allowed to stand for approximately hours before rotation begins, then mixed continuously for 12 hours At the conclusion of the mixing time, the emulsions are collected from the vessels for measurement of water content, viscosity, and the complex modulus The emulsions are stored at 15 C for one week, then observed for changes in physical appearance Water content for the emulsions should be determined The Karl-Fischer titration method works well for all types of emulsion and watereoil mixtures The complex modulus of the emulsion is measured on a rheometer using a 35 mm plate-plate geometry A stress sweep is performed in the range 100 to 10,000 mPa in the oscillation mode at a frequency of Hz The complex modulus value in the linear viscoelastic region is reported 4.2.9 Evaluation of the Relative Dispersability of Oil and Oil Products This method determines the relative ranking of effectiveness for the dispersibility of an oil sample by to a dispersant test mixture It is used either to determine the effectiveness of a dispersant product for a standard crude oil or to test the dispersability of a crude oil against a standard dispersant This method follows ASTM F 2059 closely.17 A premix of 1:25.0 dispersant:oil is made up by adding oil to 100 mg of dispersant (approximately 2.50 mL of oil in total) Six ASTM-standard swirling conical flasks modified with side spouts, containing 120 mL of 33& brine, are placed into an incubator-shaker An aliquot of 100 mL of premix is added to the surface of the liquid in each flask, care being taken not to disturb the bulk brine The flasks are mechanically shaken at 20.0 C with a rotation speed of 150 rpm for exactly 20 minutes The solutions are allowed to settle for 10 minutes Using the side spout, 30 mL of the oil-in-water phase is transferred to a 250 mL separatory funnel, first clearing the spout by draining mL of liquid The 30 mL aliquot is extracted with 3Â5 mL of 70:30 (v:v) dichloromethane:pentane, collected into a 25 mL graduated cylinder 72 PART | III Oil Analysis and Remote Sensing A Gas Chromatograph-Flame Ionization Detector (GC/FID) is used to determine the oil concentration in the solvent A 900 mL aliquot of the 15 mL solvent extract is combined with 100 mL of internal standard (200 ppm of 5-aandrostane in hexane) in a crimp-top injection vial and shaken well The total petroleum hydrocarbon content of the sample is quantified by the internal standard method using the total resolved peak area and the average hydrocarbon response factor over the entire analytical range: RPH ¼ ATOTAL =AI:S: =RRF  20  15  120=30=0:9 (6) where RPH is the resolved petroleum hydrocarbon (mg/mL), ATOTAL is the total resolved peak area, AI.S is the internal standard peak area, and RRF is the relative response factor for a series of alkane standards covering the analytical range The method is calibrated using a series of six oil-in-solvent mixtures prepared from the premix for each oil The volume of premix dispersant/oil solution for each standard is selected to represent a percentage efficiency of the dispersed oil The volume of the premix is then carefully applied to the surface of the brine in a shaker flask and shaken exactly as one of the samples, as described previously Upon removal from the shaker however, the entire contents of the flask is transferred to the separatory funnel This is extracted with  20 mL of 70:30 (v:v) dichloromethane:pentane and made up to 60 mL Chromatographic quantitation is then performed using the formula: RPH ¼ ATOTAL =AI:S: =RRF  20  60  120=120=0:9 (7) The RPH values as a function of % effectiveness for the calibration standards are plotted The sample RPH values are then used to determine the percentage effectiveness of the dispersant Note that these effectiveness percentages are not expected to correlate to real-world dispersabilities It is important to remember that these values are relative rankings only 4.2.10 Adhesion to Stainless Steel Adhesion to stainless steel is useful to responders in order to judge the “stickiness” of oil to certain drum skimmer configurations Environment Canada has developed a quantitative test for this purpose.18,19 An analytical balance is prepared by hanging an ASTM method D standard penetrometer needle from the balance hook and allowing the apparatus to stabilize and tare Approximately 80 mL of oil sample is poured into a 100 mL beaker The beaker is elevated until the oil reaches the top of the stainless steel needle Care is taken not to coat the brass segment of the needle The needle rests for 30 seconds immersed in the oil The beaker is lowered until the needle is clear of the oil The system is left undisturbed, closed inside a draft shield After 30 minutes, the weight of the oil adhering to the needle is recorded The Chapter | Measurement of Oil Physical Properties 73 mass of the oil divided by the surface area of the needle is the adhesion of the oil in g/cm2 Typically, four measurements are taken for each oil sample and the mean reported as the final value 4.3 HYDROCARBON GROUPS The fate and behavior of crude oils and petroleum products are strongly determined by their chemistries The main constituents of oils can be grouped into four categories: saturated hydrocarbons (including waxes), aromatics, resins, and asphaltenes Saturates: A group of hydrocarbons composed of only carbon and hydrogen with no double bonds or aromaticity They are said to be “saturated” with hydrogen They may by straight-chain (normal), branched, or cyclic Typically, however, the group of “saturates” refers to the aliphatics generally including alkanes, as well as a small amount of alkenes The lighter saturates, those less than ~C18, make up the components of an oil most prone to weathering The larger saturates, generally those heavier than C18, are termed waxes Aromatics: These are cyclic organic compounds that are stabilized by a delocalized p-electron system They include such compounds as BTEX (benzene, toluene, ethylbenzene, and the three xylene isomers), polycyclic aromatic hydrocarbons (PAHs, such as naphthalene), and some heterocyclic aromatics such as the dibenzothiophenes Benzene and its alkylated derivatives can constitute several percent in crude oils PAHs and their alkylated derivatives can also make up as much as a percent in crude oils Resins: This is the name given to a large group of polar compounds in oil They include heterosubstituted aromatics (typically oxygen- or nitrogencontaining PAHs), acids, ketones, alcohols, and monoaromatic steroids Because of their polarity, these compounds are more soluble in polar solvents than the nonpolar compounds, such as waxes and aromatics, of similar molecular weight Asphaltenes: A complex mixture of very large organic compounds that precipitate from oils and bitumen by natural processes For the purposes of this method, asphaltenes are defined as the fraction that precipitates in n-pentane The separation of petroleum and its products into these four characteristic groups is known as fractionation The quantification of the groups is often referred to as SARA analysis, an acronym of the characteristic groups: saturates, aromatics, resins, and asphaltenes Historically, many techniques have been used to perform this separation, including distillation, solvent precipitation (ASTM D6560)20, treatment with strong acids (ASTM D2006)21, adsorption (ASTM D2007 and D4124)22,23, and thin-layer chromatography.24 For reviews of the methods, see Speight and Becker.24-26 While excellent methods for the determination of the SARA groups have been developed using thin-layer chromatograph (TLC), there has been continuing interest in alternate 74 PART | III Oil Analysis and Remote Sensing test methods based on solvent separation and adsorption techniques.22-24 Gravimetric methods are typically based on the solubilities of the groups in n-pentane, hexane/benzene, and methanol.3 Such methods can rely on gravimetric determinations of all components, including the saturate and aromatic groups However, the drawback of such methods is that they contain significant volatile components This is particularly true of crude oils and lighter fuels More sophisticated methods rely on a combination method involving determination of the saturate and aromatic fractions by gas chromatography, an adaptation of total petroleum hydrocarbon methods, while gravimetrically determining the nonvolatile resin and asphaltene components.27,28 Resin and Asphaltene Gravimetric Determination A 100 mL quantity of n-pentane is added to a preweighed sample of approximately g of oil The flask is shaken well and allowed to stand for 30 minutes.27 The sample is filtered through a 0.45 mm membrane using a minimum of rinsings of n-pentane The precipitate is allowed to dry, then weighed The weight of the precipitate as a fraction of the initial oil sample weight is reported as the percentage asphaltenes The filtrate from the precipitation, the “maltene” fraction, is recovered and made up to 100 mL with n-pentane A 15 g, a cm diameter column of activated silica gel is prepared The top of the column is protected by a cm layer of sodium sulphate A mL aliquot of the maltene fraction is loaded onto the column A 60 mL volume of 1:1 (v:v) benzene:hexane is eluted through the column and discarded A 60 mL volume of methanol, followed by a 60 mL volume of dichloromethane, are eluted through the column and combined The methanol/dichloromethane fractions are reduced by rotary evaporation and blown down to dryness under nitrogen The mass fraction of this dried eluent, compensating for the volume fraction used, is reported as the percentage of resins in the sample Resin and Asphaltene Thin-Layer Chromatography Determination While no standard method for this technique exists, it has the advantages over the gravimetric methods of being much faster, requiring much less oil or product and being more reproducible It has the disadvantage of requiring a sophisticated instrument, a TLC with a flame ionization detector (FID) A TLC that quantifies analytes developed on silica gel-coated glass rods, such as the Iatroscan Mark 6, is necessary for this method Briefly, an aliquot of sample dissolved in dichloromethane at a concentration of mg/mL is spotted at a point, the origin, near one end of a rod, the foot of the rod The rods are then developed by immersion of the feet into a series of solvents to separate the four hydrocarbon groups The origin points must remain above the liquid surface, but the feet of the rods must be immersed sufficiently to cause solvent to travel up the rods by capillary action Chapter | Measurement of Oil Physical Properties 75 The first solvent used is n-hexane to develop the saturates Toluene develops the aromatics Finally, a 95% dichloromethane, 5% methanol mixture is used to develop the resins The asphaltenes remain at the spotting origin The hydrocarbon groups that are not quantified by this method, the saturates and aromatics, are removed by pyrolysis A known standard is then applied to the chromarod and then quantified using an FID and an internal standard A sample of octadecanol at mg/mL concentration is a convenient internal standard This is spotted on the rod just prior to measurement, on the part of the rod pyrolyzed to remove the saturate and aromatic fractions The development of the chemicals on the rods critically depends on the conditions The rods must be developed in tanks to control the vapors in atmosphere Also, temperature and humidity must remain as consistent as possible in order to achieve reproducible results When drying after each development, the rods must rest in a controlled humidity chamber Resin and asphaltene contents are determined as follows: %Resin ¼ CIS  VIS  AR =AIS (8) %Asphaltene ¼ CIS  VIS  AA =AIS (9) where: CIS: Internal standard concentration VIS: Internal standard volume AIS: Internal standard area from TLC integration AR: Resin area from TLC integration AA: Asphaltene area from TLC integration Note that while saturate and aromatic fractions are separated by the development process and could, in principle, be measured by TLC-FID, the drying process between development stages requires significant evaporation This level of evaporation is significant enough to remove most of the volatile components, which includes a large fraction of both saturates and aromatics (but not the resins or asphaltenes) For this reason, this TLC-FID method is not suitable for saturate or aromatic determination Saturate and Aromatic Chromatographic Determination This method is adapted and simplified from a previously published method for crude oil and petroleum product determination.28 An 80 mg/mL solution of oil is prepared in hexane A 3.0 g column of activated silica-gel is prepared, topped with 0.5 cm anhydrous sodium sulphate The column is conditioned with 20 mL of hexane An amount of 200 mL of the oil solution, approximately 16 mg of oil, is quantitatively transferred onto the column using an additional mL of hexane to complete the transfer The eluent is also discarded Just prior to exposure of 76 PART | III Oil Analysis and Remote Sensing the sodium sulphate to the air, 12 mL of hexane is added to the column The eluent is labeled fraction “F1.” F1 is considered to contain all the saturates, including the waxy components in the oil The column is then eluted with 15 mL of 1:1 (volume:volume) benzene/ hexane or dichloromethane/hexane The eluent is collected and labeled fraction “F2.” F2 is considered to contain the aromatic compounds in the oil, including the BTEX compounds, other alkylated benzene species, PAHs, and the alkylated PAH homologues Half of fractions F1 and F2 are combined This composite fraction is labeled “F3.” This fraction is used for analysis of total petroleum hydrocarbons (TPH) All the three fractions are concentrated under dry nitrogen The fractions are then spiked with the internal standard, 100 mL of 200 ppm 5-a-androstane, and made up with hexane to mL The analysis for total petroleum hydrocarbons and saturates is performed by high-resolution capillary GC/FID using the following conditions: Column: Carrier Gas: Injection volume: Injector temperature: Detector temperature: Oven program: 30 m  0.32 mm ID HP DB5-HT fused silica column (0.10 mm film thickness); Helium, 3.0 mL/min, constant flow; 1.0 mL; 290 C; 325 C; 40 C for minutes, followed by 25 C/minute to a final temperature of 340 C, then held for 15 minutes The total run time is 29 minutes To calculate the concentration of hydrocarbons in each fraction, the area response attributed to the petroleum hydrocarbons must be determined This area includes all of the resolved peaks and unresolved “hump.” This total area must be adjusted to remove the area response of the internal standards and GC column bleed Column bleed is the reproducible baseline shift that occurs during the oven cycle of the GC To determine this area, a hexane blank injection is analyzed before and after every 10 samples to determine the baseline response The integration baseline is then set at a stable reproducible point just before the solvent peak This baseline area for the blank run is subtracted from the actual sample run The total areas of the chromatograms of F1, F2, and F3 are obtained by integration of all peaks, corrected by removal of the baseline The area response attributable to the internal standard is calculated The F3 fraction is used to calculate the TPH values for the oil.28 The F1 and F2 fractions are used to calculate the total saturate (TSH) and total aromatic (TAH) contents Note that TPH should be within 10% of TSH ỵ TAH Chapter | Measurement of Oil Physical Properties 77 As not all the oil is passed through the GC column, a simple sum of TSH, TAH, resin, and asphaltene contents will not equal 100% This missing portion of the oil, which does not precipitate or get analyzed by the GC method, is approximated by proportionally dividing it into the saturate and aromatic portions Thus the saturate content of the oil is commuted using: % Saturates ẳ TSH=TSH ỵ TAHị1 % Asphaltenes À % ResinsÞ (10) Likewise, the aromatic content is computed using: % Aromatic ẳ TAH=TSH ỵ TAHị1 % Asphaltenes À % ResinsÞ (11) Note that the asphaltene and resin contents may be determined by either gravimetric or TLC-FID method described earlier For crude oils or products with high water content, it is necessary to dry the sample prior to the gravimetric determination of the hydrocarbon group contents If a Karl-Fischer water content determination can be made, then the composition of the original product can be reported and adjusted for the observed water content If not, the values should be reported as for dried product only 4.4 QUALITY ASSURANCE AND CONTROL Most of the physical property methods described here rely on a single instrument and involve a simple measurement with little sample manipulation.28 For these methods, the instruments are calibrated as directed by the manufacturer or the appropriate ASTM method with chemical and/or gravimetric standards as appropriate In addition, instrumental and operator performance should be monitored by periodic measurement of check standards A control chart should be kept for each procedure, for the check or performance standard measurements The check standard measurements are monitored closely Failure of the check standard measurement to fall within the smaller of either a historical 95% confidence limit or the appropriate ASTM required repeatability should result in an investigation of the procedure This typically includes required instrument maintenance, cleaning, recalibration, and measurement of the check standard until the desired precision and accuracy is reached The chromatographic methods described here, including the dispersability tests and the hydrocarbon group analysis, involves significant sample preparation, followed by a measurement by gas chromatography Such techniques require a higher level of effort to maintain quality assurance Check or surrogate samples of either pure materials or certified reference standards should be processed in the same manner as the samples Calibration should be accomplished with a second, separate set of certified reference materials Internal standards should also be certified reference materials from reputable suppliers Surrogate recovery, calibration stability, and internal standard response control charts should all be checked regularly to ensure procedure 78 PART | III Oil Analysis and Remote Sensing and measurement accuracy Chromatograms should be checked to ensure that chromatographic quality, including good peak shape, baseline drift, column bleed, sample carryover, and chromatographic resolution are within acceptable limits 4.5 EFFECTS OF EVAPORATIVE WEATHERING ON OIL BULK PROPERTIES Long experience has shown that the physical characteristics and chemical fingerprint of a crude oil can change greatly over the course of a spill incident These changes have a profound effect on the fate, behavior, and effects of an oil in the environment The oil may transmute to other states, evaporating, dissolving in water, or condensing to a semisolid residue, each new state having unique behaviors and eventual fates In order to aid in the estimation and prediction of spill behavior, it is useful to know not only the characteristics of the fresh crude oil, but also those of oils at different stages of “weathering” in the environment Previous work has shown that immediately after a spill, the dominant process of oil weathering is evaporation The following discussion focuses on the effects of evaporative weathering on changes of oil physical properties and chemical compositions 4.5.1 Weathering When oil is spilled, on either water or land, a number of transformation processes operate on the oil In general, there are two types of transformation processes: the first is weathering, and the second is a group of processes (including spreading, movement of oil slicks, and sinking and ove-washing) related to the movement of oil in the environment Weathering and movement processes overlap, with weathering strongly influencing how oil moves in the environment and vice versa These processes depend very much on the type of oil spilled and the weather conditions during and after the spill Thoroughly understanding the behavior of spilled oil in the environment is extremely important for development of oil spill models Today’s sophisticated spill models combine the latest information on oil fate and behavior with computer technology to predict where the oil will go, what state it will be in, and when it gets there “Weathering” is the term referring to a combination of a wide variety physical, chemical, and biological processes of a spilled oil in the environment The weathering processes include evaporation, emulsification, natural dispersion, dissolution, microbial degradation, photo-oxidation, and other processes such as sedimentation, and oil-suspended particle interactions Weathering has a very significant effect on most bulk oil properties Unlike the chemical compositions, however, where environmental parameters only affect the rate and type of weathering, bulk properties of the oil are also highly variable depending on the physical conditions The most important of these is Chapter | Measurement of Oil Physical Properties 79 temperature, but other factors such as pressure and the materials with which the oil is in contact also play a role As an oil loses mass and changes in composition, several general trends in physical property changes can be observed: l l l Density increases approximately linearly with increasing weathering Density decreases approximately linearly with temperature Viscosity increases with increasing weathering, but a simple functional relationship is not easy to develop Viscosity increases approximately exponentially with decreasing temperature Surface and interfacial tensions tend to increase slightly with increasing weathering 4.5.2 Preparing Evaporated (Weathered) Samples of Oils A common technique for simulating weathering in the laboratory is evaporation While this is only one of the possible processes in the natural environment, it is probably the dominant one for most spills, particularly in the first few hours or days following a spill A laboratory oil-weathering technique by rotary evaporation allows for convenient preparation of artificially weathered oils with varying degrees of weight loss A typical oil-weathering system consists of a rotary evaporator The bath temperature of the evaporator should be variable from 20 C to 100 C Ỉ 0.5 C The rotation speed should be continuously variable from 10 to 135 rpm The following evaporation procedure is used to evaporate oils: (1) The water bath is brought to a temperature of 80 C (2) The empty rotary flask is weighed, and no more than one-third the volume of the rotary flask in oil is added and the flask reweighed (3) The flask is mounted on the apparatus and the flask partially immersed in the water bath and spun at high speed, at least 120 rpm A constant flow of air through the apparatus should be maintained by a vacuum pump (4) At set intervals, the sample flask is removed and weighed It is convenient to prepare two to three weathered samples for each type of oil measured With a moderate flow rate through the instrument, a duration of 48 hours evaporation will come close, within to 10%, to simulating the eventual final state of an oil in the environment Intermediate fractions of approximately one- and two-thirds of the 48-hour loss by weight will simulate approximately the condition of the oil after a few hours to days and a few days to weeks of natural evaporation The exact time taken to prepare these intermediate fractions is determined by estimation from the measured fractional mass-loss as a function of time for the 48-hour sample The fraction mass-loss is calculated as: % weathering ẳ mi mf ị=mi À me Þ x 100% (12) 80 PART | III Oil Analysis and Remote Sensing where % weathering is the percentage evaporative mass-loss over the 48-hour period, mi is the initial mass of the flask and oil, mf is the final mass of the flask and oil, and me is the mass of the empty flask A graph of % weathering as a function of time is plotted using the interval weighing data The times for one-third (t1/3) and two-thirds (t2/3) of the 48-hour mass loss are interpolated from a time-weathering graph Typical times for t1/3 range from 30 minutes to hours, for t2/3, to 12 hours This technique allows for precise control of the evaporative weight loss for a target oil and can be directly correlated to bulk property and compositional changes of the weathered oil By tracking weight loss as a function of time, an equation for predicting evaporation can be found Also, from this same graph, it is possible to determine a point at which the evaporation rate is sufficiently slow that the oil may be considered to have achieved the maximum evaporative loss likely to be observed under the conditions of a marine spill 0.94 0.94 Cook Inlet 2003 vs.T Density (g/mL) 0.92 Cook Inlet 2003 0.92 34.4% 0.90 Density (g/mL) (a) 25.0% 0.88 11.4% 0.86 Fresh 0.84 0.82 vs.W(%) 0.90 0.88 0.86 °C 15 °C 0.84 30 °C 0.82 10 15 20 25 30 0.80 35 10 Temperature (°C) (b) 1.02 1.02 Platform Elly 40 Platform Elly 1.00 13.3% 0.98 7.9% 4.6% 0.96 Density (g/mL) 1.00 Density (g/mL) 30 vs.W(%) vs.T 0.98 °C 0.96 Fresh 0.94 20 Weathering (%) 10 15 20 25 Temperature (°C) 30 35 15 °C 30 °C 0.94 10 12 14 Weathering (%) FIGURE 4.1 Density versus temperature and weathering for a light (Cook Inlet) (a) and heavy (Platform Elly) (b) crude oil Chapter | 81 Measurement of Oil Physical Properties 4.5.3 Quantifying Equation(s) for Predicting Evaporation The evaporation kinetics are determined for each oil by measuring the weight loss over time from a shallow dish.30,31 Approximately 20 g of oil is weighed into a 139 mm petri dish The oil weight is recorded by an electronic balance accurate to 0.01 g at set intervals and collected on a computer logging system Measurements are conducted in a climate-controlled chamber at 15 C Temperatures are monitored by a digital thermometer The evaporation period can last from a few days for light oils to weeks for heavier products The time versus weight-loss data series are fitted to a set of simple equations The best curve-fit is chosen as the equation for predicting evaporation Effects of Evaporative Weathering on Crude Oil Density Densities of oils typically increase approximately to 10% as oil weathers Cook Inlet, a light oil, changes from 0.84 g/mL to 0.91 g/mL at 30 C (see 10000 10000 Cook Inlet 2003 vs.W(%) 1000 100 Viscosity (mPas) Viscosity (mPas) Cook Inlet 2003 vs.T 34.4% 25.0% 10 1000 100 10 11.4% Fresh 10 15 20 25 30 °C 15 °C 30 °C 35 10 1e+7 30 40 1e+7 Platform Elly vs.T 1e+5 13.3% 1e+4 7.9% 4.6% 1e+3 Fresh 1e+2 1e+6 Viscosity (mPas) 1e+6 Viscosity (mPas) 20 Weathering (%) Temperature (°C) Platform Elly vs.W(%) 1e+5 1e+4 1e+3 °C 15 °C 30 °C 1e+2 10 15 20 25 Temperature (°C) 30 35 10 12 14 Weathering (%) FIGURE 4.2 Viscosity versus temperature and weathering for light (Cook Inlet) and heavy (Platform Elly) crude oils 82 PART | III Oil Analysis and Remote Sensing Figure 4.1a), while Platform Elly, a very heavy crude oil, has a fresh density of 0.9531 g/mL and increases to 0.9843 g/mL in its most weathered state at 30 C (Figure 4.1b) From Figure 4.1, it can be seen that, to a first approximation, 32 31 Cook Inlet 2003 vs T o/a 30 34.4% 29 25.0% 28 11.4% 27 26 Fresh 25 Surface Tension (oil/air) (mN/m) Surface Tension (oil/air) (mN/m) 32 10 15 20 25 30 34 o/w vs.T 32 30 28 26 34.4% 25.0% 24 11.4% Fresh 22 10 15 20 25 30 o/b 32 vs.T 30 28 26 24 34.4% 25.0% 11.4% 22 Fresh 20 10 15 20 25 Temperature (°C) 30 30 °C 28 27 26 25 30 °C 15 °C °C 35 10 20 30 40 34 vs.W(%) o/a 32 30 28 26 °C 15 °C 24 30 °C 22 35 34 °C 15 °C 29 35 Interfacial Tension (oil/water) (mN/m) Interfacial Tension (oil/3.3%brine) (mN/m) Interfacial Tension (oil/water) (mN/m) Cook Inlet 2003 vs.W(%) o/a 30 24 24 Interfacial Tension (oil/3.3%brine) (mN/m) 31 10 20 30 40 34 vs.W(%) 32 o/b 30 28 26 24 °C 15 °C 22 30 °C 20 10 20 30 40 Weathering (%) FIGURE 4.3 Surface and interfacial tensions as a function of temperature and weathering for Cook Inlet (2003) Chapter | Measurement of Oil Physical Properties 83 density increases linearly with increasing mass-loss and decreasing temperature Better extrapolations can be made from log-log extrapolations of both quantities Note that the uncertainties in density are very small: Ỉ0.0002 g/mLdapproximately part in 5,000 Effects of Evaporative Weathering on Crude Oil Viscosity In contrast to most other physical properties, the viscosity of an oil can change by orders of magnitude with weathering and changes in temperature For example, the viscosity of Cook Inlet (2003) changes from 5.8 mPa s to 67.0 mPa s at 30 C (see Figure 4.2), while fresh Platform Elly has a viscosity of 1070 mPa s, and reaches 52280 mPa s in the most weathered fraction (Figure 4.3) As can be seen from the logarithm of viscosity is roughly inversely linear with temperature, but the effects of weathering on viscosity are more complex Uncertainties in viscosity are Æ5% Effects of Evaporative Weathering on Crude Oil Surface and Interfacial Tensions Surface and interfacial tensions have no simple quantitative relationships in general to either the degree of weathering or the temperature Surface tensions however, not vary greatly from oil to oil; values from 25 mN/m to 32 mN/m are typical for almost all types of oil Interfacial tensions for oil/water and oil/ 3.3% brine are often marginally lower than the corresponding oil/air surface tension Oil/brine interfacial tensions are usually somewhat higher than the corresponding oil/(pure) water values Typical values for both range from 18 mN/m to 32 mN/m Surface and interfacial tensions tend to decrease with temperature and increase with weathering Care should be taken not to overinterpret the significance of surface and interfacial tension values; however, the errors on these measurements are relatively large, Ỉ15%, and the relative variations of the values are fairly small REFERENCES ASTM D 5002 Standard Test Method for Density and Relative Density of Crude Oils by Digital Density Analyzer Conshohocken, PA: American Society for Testing and Materials (ASTM); 2009 API 82 American Petroleum Institute (API), Petroleum Measurement TablesdVolume XI/XII West Conshohocken, PA: American Society for Testing and Materials; 1982 Jokuty P, Fingas M, Whiticar S Oil Analytical Techniques for Environmental Purposes AMOP 1994;245 ASTM D 971 Standard Test Method for Interfacial Tension of Oil Against Water by the Ring Method West Conshohocken, PA: American Society for Testing and Materials; 2009 Jokuty P, Fingas M, Whiticar S, Fieldhouse B A Study of Viscosity and Interfacial Tension of Oils and Emulsions, Manuscript Report EE-153, Ottawa, ON: Environment Canada, 1995 84 PART | III Oil Analysis and Remote Sensing Song B, Springer J Determination of Interfacial Tension from the Profile of a Pendant Drop Using Computer-aided Image Processing Colloid Interface Sci 1996;64 ASTM D1310 Standard Test Method for Flash Point and Fire Point of Liquids by Tag Open-Cup Apparatus West Conshohocken, PA: American Society for Testing and Materials; 2007 ASTM D 6450 , Standard Test Method for Flash Point by Continuously Tester West Conshohocken, PA: American Society for Testing and Materials; 2009 ASTM D 93 American Society for Testing and Materials (ASTM), Standard Test Method for Flash Point by Pensky-Martens Closed Cup Tester West Conshohocken, PA: American Society for Testing and Materials; 2009 10 ASTM D 7094 Standard Test Method for Flash Point by Modified Continuously Closed Tester West Conshohocken, PA: American Society for Testing and Materials; 2009 11 Montemayor RG, Rogerson JE, Colbert JC, Schiller SB Reference Verification Fluids for Flash Point Determination J Test Eval 1999;27 12 ASTM D 97 Standard Test Method for Pour Point of Petroleum Oils West Conshohocken, PA: American Society for Testing and Materials; 2009 13 ASTM D 4294 Standard Test Method for Sulfur in Petroleum Products by Energy-Dispersive X-ray Fluorescence Spectroscopy West Conshohocken, PA: American Society for Testing and Materials; 2009 14 ASTM D 4377 Standard Test Method for Water in Crude Oils by Potentiometric Karl Fischer Titration West Conshohocken, PA: American Society for Testing and Materials; 2009 15 Fingas M, Fieldhouse B, Mullin J Studies of Water-in-oil Emulsions: Stability and Oil Properties AMOP 1998;1 16 Fingas M, Fieldhouse B Studies on Crude Oil and Petroleum Product Emulsions: Water Resolution and Rheology Colloids Surf A 2009;67 17 ASTM F 2059 Standard Test Method for Laboratory Oil Spill Dispersant Effectiveness Using the Swirling Flask West Conshohocken, PA: American Society for Testing and Materials; 2007 18 Jokuty P, Whiticar S, McRoberts K, Mullin J Oil Adhesion TestingdRecent Results AMOP 1996;9 19 ASTM D Standard Test Method for Penetration of Bituminous Materials West Conshohocken, PA: American Society for Testing and Materials; 2009 20 ASTM D 6560 Standard Test Method for Determination of Asphaltenes (Heptane Insolubles) in Crude Petroleum and Petroleum Products West Conshohocken, PA: American Society for Testing and Materials; 2006 21 ASTM D 2006 Method of Test for Characteristic Groups in Rubber Extender and Processing Oils by the Precipitation Method (Withdrawn 1975) West Conshohocken, PA: American Society for Testing and Materials; 1965 22 ASTM D 2007 American Society for Testing and Materials (ASTM), Standard Test Method for Characteristic Groups in Rubber Extender and Processing Oils and Other PetroleumDerived Oils by Clay-Gel Absorption Chromatographic Method West Conshohocken, PA: American Society for Testing and Materials; 2007 23 ASTM D 4124 Standard Test Methods for Separation of Asphalt into Four Fractions West Conshohocken, PA: American Society for Testing and Materials; 2006 24 Barman BN Hydrocarbon-Type Analysis of Base Oils and Other Heavy Distillates by ThinLayer Chromatography with Flame-Ionization Detection and by the Clay-Gel Method J Chromat Sci 1996;219 25 Speight JG The Chemistry and Technology of Petroleum New York: Marcel Dekker; 2007 26 Becker JR Chapter 13, Asphaltene Test Methods, Crude Oil Waxes, Emulsions and Asphaltenes Tulsa, OK: Penn Well Publishing Co; 1991 Chapter | 85 Measurement of Oil Physical Properties 27 Hollebone B, Wang Z, Landriault M, Smith P A New Method for the Determination of the Hydrocarbon Groups in Oils: Saturates, Aromatics, Resins, and Asphaltenes (SARA) AMOP 2003;31 28 Wang ZD, Fingas M, Li K Fractionation of ASMB Oil, Identification and Quantitation of Aliphatic Aromatic and Biomarker Compounds by GC/FID and GC/MSD (Parts I and II) J Chromat Sci 1994;361 29 Environment Canada, Oil Properties Database, http://www.etc-cte.ec.gc.ca/databases/ OilProperties/oil_prop_e.html, accessed May 2010 30 Fingas M The Evaporation of Oil Spills AMOP 1995;43 31 Fingas M Modeling Evaporation Using Models That Are Not Boundary-Layer Regulated J Haz Mat 2004;27 APPENDIX 4.1 Table A4.1 gives the environmentally relevant properties of selected crude oils TABLE A4.1 Environmentally-Relevant Properties of Selected Crude Oils29 Alaska North Slope Prudhoe Bay, Alaska, USA Arabian Light Saudi Arabia Mississippi Brent Canyon Blend Federated Block 807 Gulf of North Mexico, Sea, United Alberta, Louisiana, USA Kingdom Canada West Texas Intermediate Texas USA 0 C 0.8777 0.8776 0.8472 0.8413 0.9310 0.8594 15 C 0.8663 0.8641 0.8351 0.8293 0.9461 0.8474 30.89 31.30 37.8 38.9 17.5 34.38 0 C 23.2 32.6 16 10 88.1 19.2  11.5 13 4.8 8.6 mN/m C 27.3 27.2 28.0 27.3 28.8 27.4  26.4 26 25.5 25.8 28.2 26.0 mN/m C 26.7 23.5 25.7 18.7 24.4 19.3  23.6 23.8 22.7 15.9 24.1 15.8 Oil-sea water Interfacial tension mN/m C 22.5 21.3 24.9 17.6 26.0 18.8  20.2 21.6 22.5 16.2 26.6 15.6 Flashpoint  C