Distillation Data Research Articles

Modeling Simulated Distillation. Influence of Catalyst Addition upon Model Parameters

Simulated distillation data of distillates produced by hydroconversion of a vacuum residue have been described as a two-component cumulative normal distribution function. The influence of catalyst addition upon calculated parameters of the model indicates that distillate fraction is produced through a consecutive mechanism. A modified three-component model has then been applied. It reveals that catalyst limits the extent of the first bond cleavage step and reduces the yield of light distillate. Furthermore, the distribution of hydrogen consumption is strongly affected by catalyst addition, which favors hydrogen incorporation in distillate.

Modeling Simulated Distillation: A Tool for the Evaluation of Hydroconverted Petroleum Residues

Simulated distillation, commonly applied to the characterization of complex products issued from the conversion of heavy petroleum fractions, provides information concerning the yield of distinct classes of distillable products. A simple modeling of simulated distillation data has been applied to several distillates obtained by hydroconversion of a vacuum residue. This model gives access to characteristic parameters which allows a description of this complex mixture in terms of two distinguishable components. The lighter one is produced through two simultaneous mechanisms, involving the rupture of substituted carbon positions and a symmetrical cleavage of the resulting molecules. The remaining component is representative of the initial feed and is preserved in a slightly modified state up to high level of conversion.

An efficient dynamic model for batch distillation

Abstract A novel, rigorous and efficient solution technique for multicomponent batch distillation modelling equations is proposed. Model predictions using the technique are shown to be in close agreement with experimental batch distillation data for a ten sieve tray, 15 cm diameter column separating ethanol and water. The results also show improved accuracy over commercially available programs for batch distillation. The method incorporates rigorous dynamic energy blances as well as accurate representation of both tray hydraulics and non-ideal mass transfer. The technique is based on a functional approximation for liquid enthalpy and makes a difficult-to-calculate temperature derivative implicit in other terms in the equations, eliminating the need for iterative solution techniques. The numerical efficiency of the method permits its utilization in model-based optimization and control calculations. The modelling approach is applicable to both batch and continuous dynamic distillation models.

Analytical steam distillation model for thermal enhanced oil recovery processes

Steam distillation is one of several mechanisms contributing to oil recovery during thermal enhanced oil recovery (EOR) operations. As such, it is of interest to predict the quantity and character of distillate under conditions encountered during such operations. This paper presents an easy-to-use, multicomponent analytical model to predict steam distillation yield of a given heavy oil under specified thermal EOR conditions. Required input is the simulated distillation (SIMDIS) trace of the oil, which is utilized to break the heavy oil into the desired number of pseudocomponents. Steam distillation efficiency is the only adjustable parameter. Experimental steam distillation data obtained with South Belridge oil were used to validate the preformance of the model.

Preparation of bitumen from oil sand by ultracentrifugation

Ultracentrifugation was investigated as a means to obtain solvent-free bitumen from oil sand. The bitumen from three oil sands of varying grades was separated by placing the sands in specially designed tubes and centrifuging for 2 h at 198 000 g at 20 °C. For all grades of oil sand, approximately 70% of the bitumen was recovered. The recovered bitumen was compared to the residual remaining on the sand, and to that extracted by the conventional Soxhlet technique. The ultracentrifuged bitumen contained some emulsified water and a small amount of fine solids. The solvent-extracted material was water-free, but contained a small amount of residual solvent and fine solids. The ultracentrifuge caused some fractionation of the bitumen, resulting in a product slightly enriched in asphaltene components compared to the solvent-extracted material. The residual bitumen remaining on the sand was correspondingly slightly depleted in asphaltenes. However, as evidenced by gas Chromatographic simulated distillation data, ultracentrifugation did retain the light (180–220 °C) components of the bitumen which were lost during the solvent removal step following solvent extraction. Other analyses such as density, viscosity and elemental composition verified that ultracentrifugation resulted in some fractionation of bitumen components.

Experimental Studies on the Interaction Between the Initial Liquid Distribution and the Performance of Structured Packings

Abstract Distillation and hydraulic test data are reported which give insight into the effect of the quality of initial liquid distribution on the performance of two corrugated sheet metal structured packings, differing considerably in surface area and texture. As expected, the packing with the larger surface (Montz-Pak BS-450) is more sensitive to initial liquid maldistribution. From the packing with smaller surface (Ralu-Pak 250 YC) a good performance may be expected even with only 40 drip points/sq m.

Methods for predicting the physical changes in oil spilt at sea

Recently developed mathematical expressions describing oil slick behaviour have been compared with data obtained from various experimental oil spills. The report presents the equations used and shows how the results from some sixteen different spills compare with the predictions. These comparisons have shown that the algorithms generally give good predictions of behaviour. The models require detailed oil property data, particularly asphaltene contents, distillation data and density-composition relationships. Fortunately, such information is now readily available for a wide range of crude oils.

Hydrocarbon composition of crude oil from Lam Bank

The authors discuss the crude oil from a new offshore field called the Lam Bank in the Caspian Sea. A segregated commercial crude was distilled and the distillation data is shown. In order to determine the content of n-paraffins, the naphthenic-paraffinic part of the narrow cuts was subjected to adsorptive separation on CaA zeolite. Owing to the high contents of naphthenic and isoparaffinic hydrocarbons and the low content of aromatic hydrocarbons in the distillate part, this crude can be used to produce high-quality fuels and oils by the use of the dewaxing processes.

Calculation of High Temperature Crude-Oil/Water/Vapor Separations Using Simulated Distillation Data

Summary High-temperature crude-oil/water/vapor separation takes place in steamflooding and in-situ combustion processes. It also takes place in hydrocarbon recovery from deep volatile oil and condensate reservoirs. A practical procedure that uses the Holland and Welch method and sirilulated distillation data was developed to calculate crude-oil/water/vapor separations at 387 and 456°F [197 and 236°C]. The overhead yields obtained from the calculations were expressed as a function of the steam distillation factor. The Vw/Voi results were compared with laboratory crude-oil steam distillation data. The approach satisfactorily predicted the overhead yields of 13 out of 16 crude oils with an average error of 12% (±3.6% in yield). This is within experimental error of crude-oil steam distillation. Twelve pseudocomponents of crude oils were selected and characterized with simulated distillation data for the calculations. The physical properties of the pseudocomponents were determined from existing correlations and from matching laboratory steam distillation data. The use of simulated distillation data eliminates the uncertainty and assumptions normally involved in the selection of crude oil pseudocomponents with U.S. Bureau of Mines distillation data, and thus improves the reliability of the proposed computational approach. The proposed approach eliminates the need to conduct experimental steam distillation tests if simulated crude oil distillation data are available. It is easy and fast to calculate the overhead yields ahd densities without use of the equation of state (EOS) and uncertain pseudocomponent critical properties and interaction parameters. The proposed approach will provide useful information for the numerical simulation and design of thermal recovery processes and for the prediction of hydrocarbon recovery from high-temperature volatile oil and condensate reservoirs.

Application of the “cooled needle” technique to split and splitless sampling onto capillary columns

AbstractCooled needle sampling using syringes was applied to splitless injection and to simulated distillation analyses. Slight changes of the construction of the previous device are also described. The changes concern the temperature profile within the injector and especially the vaporization insert. Even with the low carrier gas flow through the injector during splitless injection, discrimination by component volatility can be avoided. Precision and accuracy of simulated distillation data obtained with split sampling also can be improved by the cooled needle technique.

Determination of vapor pressure/molecular weight correlations from droplet evaporation data

A rapid method is presented for the determination of vapor pressure/molecular weight correlations for homologous hydrocarbon mixtures. By comparing droplet evaporation data to discretized theoretical continuum mass flux rates which employ two parameter exponential vapor pressure/molecular weight correlations, best fit pressure correlation parameters are determined. The present technique is applied to the multicomponent oil mixture, 100 pale oil, which is modeled as a fifteen component system. Postulating the mixture is thermodynamically ideal, best fit pressure correlations are determined which result in less than 5% error between theory and experiment. It is shown the hydrocarbon vapor pressure calculated from the derived pressure correlation is in satisfactory agreement with pressures calculated from distillation data and the Clausius Clapeyron relation.

Steam Distillation of Crude Oils

Abstract The objectives of this investigation were to generate crude oil steam distillation data for the prediction of phase behavior in steamflood simulation and to correlate the steam distillation yields for a variety of crude oils. Thirteen steam distillation tests were run on 10 crude oils ranging in gravity from 9.4 to 37°API (1.004 to 0.840 g/cm3). In each test the crude was steam distilled sequentially at about 220, 300, 400, and 500°F (104, 149, 204, and 260°C). The cumulative steam distillation yields at 400°F (204°C) ranged from about 20 to 55 vol %. Experimental results showed that crude oil steam distillation yields at steamflood conditions are significant, even for heavy oils. The effects of differences in steam volume throughput and steam temperature were taken into account when comparing yields for different crudes or repeat runs on the same crude. Steam distillation yields show a high correlation with crude oil API gravity and wax content.

Carbon-13 vapor pressure isotope effect in CHF3

The vapor pressure difference between 12CHF3 and 13CHF3 has been measured by means of a cryogenic distillation under total reflux at 11 temperatures between 161.2 and 204.6 °K. A second-order extension of Cohen’s column kinetics has been tested. The reduced partition function ratio obtained from the distillation data has led to a plot of T ln(fc/fg) vs 1/T, which displays a steep negative slope and points to a crossover at a temperature above those of the present experiment. The steep negative slope is due to strong external–internal interactions in liquid fluoroform, and the predicted crossover is a consequence of a blue shift in the C–H stretching frequency upon condensation of fluoroform.

Murphree and vaporization efficiencies in multicomponent distillation

A quantitative comparison between Murphree and vaporization efficiencies is presented based on ternary distillation data for the systems acetone/methanol/ethanol, acetone/benzene/chlorobenzene, benzene/toluene/mxylene and n-hexane/methylcyclopentane/benzene. The influence of experimental errors on calculated values of Murphree efficiencies is also analysed. It is shown that the vaporization efficiency model fails to describe the behaviour of distillation plates and it is suggested that Murphree's model gives a more useful representation of the behaviour of distillation columns.

The Determination of Vapor Pressures of Low Viscosity Lubricants from Distillation Data

The vapor pressure of low viscosity aliphatic hydrocarbon lubricants is shown to be determined with good accuracy from data obtained from a modification of the ASTM Distillation Test D86-67. This test gives the initial boiling point of a lubricant from which the vapor pressure at any other temperature can be calculated by use of an empirical equation relating the boiling point to the vapor pressure. Two other equations were derived, one relating the viscosity of an aliphatic hydrocarbon lubricant to the vapor pressure at any temperature, and the other extending the use of this distillation method to calculation of the vapor pressures of other types of lubricants as, for example, esters, alcohols, and acids. The data presented indicate that this method is superior to the isoteniscope method in accuracy and ease of operation. Presented at the 25th ASLE Annual Meeting in Chicago, May 4–8, 1970

Change in the distillation of automobile gasolines by addition of light hydrocarbons

1. To meet requirements with regard to distillation data and saturated vapor pressure of winter and northern grades of automobile gasolines, we may use as the light components of the fuel butane, isopentane, isomerizate, and the light fractions of straight-run and gas naphtha, or various mixtures of these components. 2. The authors have plotted the decrease in the 10 and 50% distillation temperatures of base gasolines with differing distillation data versus the amount of butane and 23–62 and 34–67°C light hydrocarbon fractions added. These graphs enable us to calculate the necessary amounts of components for obtaining northern, winter, and summer grades of automobile gasolines with distillation data in line with requirements. 3. The paper gives data characterizing the change in the IBP of base gasolines due to compounding with butane and other light hydrocarbon fractions. 4. It is shown that butane has little effect, and a mixture of C4, C5, and C6 fractions virtually no effect, on the decrease in the 90% distillation temperature and the FBP of base gasolines. 5. Addition of 1% of butane, isopentane, the straight-run 23–62°C fraction, and the 34–67°C gas-naphtha fraction increases the saturated vapor pressures of base gasolines (with saturated vapor pressure of 100 to 400mm Hg) by 35–40, 14–7, 12–5.5, and 9-3.5 mm Hg, respectively. 6. A lower limit of the saturated vapor pressure of winter grades of automobile gasolines of 500 mm Hg may be obtained if the 10% distillation temperature is not more than 50°C.

Observations concerning the determination of porosities in graphites

The porosities of a number of graphites were determined by mercury porosimetry and helium intrusion methods, and in addition, vapor pressures and mercury distillation data were obtained on graphite specimens intruded with mercury in a porosimeter. All of the results were in agreement with the conclusions that mercury porosimetry can cause damage to graphite specimens above some critical pressure range, and that this damage can cause the results of porosimetry measurements to be misleading.

106. Observations concerning the determination of porosity in graphites

Abstract The porosities of a number of graphites were determined by mercury porosimetry and helium intrusion methods, and in addition, vapor pressures and mercury distillation data were obtained on graphite specimens intruded with mercury in a porosimeter. All of the results were in agreement with the conclusions that mercury porosimetry can cause damage to graphite specimens above some critical pressure range, and that this damage can cause the results of porosimetry measurements to be misleading.

Road Antiknock Properties of Gasolines in Correlation with their Volatility Characteristics (II)

Out of samples taken up in the previous report, four samples were selected for further studies. One sample of premium grade is representative of a gasoline, the road octane rating of which is comparable to its research octane rating, while two other samples of premium grade indicate significant road depreciations and the fourth sample of regular grade shows little depreciation in road octane rating. The latter three gasolines were found to be composed of very poor antiknock constituents concentrated in the lighter fraction.Effective volatility of these gasolines under the condition of a full throttle acceleration as used when testing road octane number was estimated from their ASTM distillation curves.The composition of the vapor and the liquid residue in the carburetter system was calculated by the flash vaporization method from the fractional distillation data and the effective volatility. The research octane numbers of the vapor and the residue were also calculated.The air-fuel mixture entering the cylinder must contain the residual liquid in the form of entrainment. The weight of entrainment per 100 weights of intake air is defined as "entrainment ratio", and the role of the entrainment was established by the theoretical calculation. The entrainment plays important roles in two ways. As the heavy ends of these gasolines have very high octane values, they contribute to the antiknock property of the cylinder mixture, and as a greater part of TEL is calculated to be concentrated in the residual liquid, the entrainment should act as a carrier of the antiknock compound.Rise of the experimental road octane number with engine speed by modified borderline method is attributed to the increase of entrainment ratio rather than the effect of aromaticity of the sample gasoline.

Possible mechanisms in thermal polymerization of vegetable oils. II. Polymer formation

SummaryFrom molecular distillation data, and information obtained by hydrogenation in dilute solution, extents of reaction have been calculated for the polymerization of linseed, safflower, tung, and oiticica oils. The hydrogenation data indicate that more than one double bond per molecule can be consumed during polymerization.The kinetic order of the polymerization of nonconjugated oils increases from an initial value of unity to approach a value of two. With conjugated oils, second‐order characteristics were maintained during the course of the bodying. An explanation of this difference in reaction order based on monomer disappearance is presented. Rate constants and overall energies of activation were calculated and compared with those obtained from viscosity measurements. Agreement between the two sets of values was obtained. Oiticica oil, which gives curved viscosity plots that cannot be resolved, has been treated successfully in this manner.Intraglyceride reactions have been studied, and differences in behavior between nonconjugated and conjugated oils at high and low polymerization temperatures have been observed.