Assay Curves - Technical Information
General Information
For many petroleum streams the composition is not completely known in terms of defined components. Therefore laboratory assay curves must be used to represent the streams with pseudocomponents (boiling point cuts) for which the necessary thermophysical properties can be estimated.
Assay curves may include laboratory distillation, gravity, molecular weight, predefined special properties such as pour point and viscosity, and user defined special properties. Assay curves may be supplied on either on LV% or weight % distilled basis. The minimum information required to characterize any stream is a laboratory distillation and the average gravity or Watson K Factor.
For many petroleum streams, the composition of the lightest portion is known from a chromatographic analysis. These known components may be supplied as light ends data and used directly to characterize the front portion of the distillation curve.
Note: Properties of light end components depart significantly from the values predicted by PETRO component characterization. For this reason, it is strongly recommended that petroleum components should have a carbon number of 6 or more. To model species with lower carbon numbers, use fully defined components, such as those found in the component data libraries.
The number of pseudocomponents to use in characterizing a petroleum stream is defined with a TBP Cutpoint set, where the number of components is defined for one or more temperature intervals on the TBP curve. The pseudocomponents for all petroleum assay streams which use a given TBP Cutpoint set are averaged to produce a single set of blend components which are used to represent all assay streams in the cutpoint set. PRO/II allows multiple cutpoint sets to be used in any problem to define multiple blends of pseudocomponents. This is important when petroleum streams are dissimilar and one set of blend components is not adequate to represent all streams. See Assay Stream Blending Options for more information on blending of streams.
Development of the TBP Distillation Curve
The first step in characterization of any assay stream is development of the TBP (true boiling point) curve at 760 mm Hg. Laboratory curves may be supplied as TBP, ASTM D86, ASTM D1160, and ASTM D2887 at any laboratory pressure. See Assay Distillation Types for further information on the test procedures.
The laboratory distillation curve must undergo curve fitting to provide the standard volume percents distilled which are used by the various conversion methods. Furthermore, fitting may be necessary to extrapolate the curve data to cover the entire range for the stream. The fitting procedures are described in Assay Curve Fitting Procedures.
TBP distillations are almost always reported by the laboratory on a 760 mm Hg basis. Should the TBP be reported at another pressure, API Technical Data Book Procedure 5A1.13 is used to adjust the temperatures to a 760 mm Hg basis.
ASTM D1160 distillations are usually reported at 760 mm Hg, however, the API Technical Data Book method to convert D1160's to TBP's is based on a pressure of 10 mm Hg. Therefore, D1160 distillations are first translated to a 10 mm Hg basis with API Procedure 5A1.13 and then converted to a TBP at 10 mm Hg with API Figure 3A2.1 (which is based on a method devised by Edmister and Okamoto). The 10 mm Hg TBP is translated to a 760 mm Hg TBP with API Procedure 5A1.13.
Several options are provided for conversion of ASTM D86 distillations to TBP's at 760 mm Hg. These options are discussed in Distillation Curve Interconversions.
ASTM D2887 simulated distillations correspond closely to TBP distillations and it is recommended that these distillations be entered as TBP distillations for gas oil and heavier streams. The default interconversion method in PRO/II is procedure 3A3.1 from the 1994 API Technical Data Book. This procedure converts D2887 cut temperatures to TBP temperatures using empirically-derived correlations.
An earlier version of procedure 3A3.1, published in the 1987 API Technical Data Book, implements a two-step conversion. It first converts ASTM D2887 curves to ASTM D86 curves using another method from the 1987 API Technical Data Book. The D86 curve then is converted to a TBP curve. This method still is available in PRO/II, but no longer is the default. See ASTM D2887/ ASTM D86 Conversion.
Reference: Edmister, W.C. and Okamoto, K.K., "Equilibrium Flash Vaporization Correlations for Heavy Oils at Subatmospheric Pressures", Petroleum Refiner, Vol. 38, No. 9, p 271 (1959).
Pseudocomponent Normal Boiling Points
The TBP curve is divided into pseudocomponents using the cutpoint ranges in the TBP Cutpoint set which is selected for processing the assay. See Assay Main Window for information on cut point sets. The TBP curve is divided into cutpoint ranges, using the temperature intervals and number of components per interval defined by the TBP cutpoint set. All components within a temperature interval are given identical temperature ranges.
For streams with defined light end components, the defined light ends are used in place of pseudocomponents to characterize the front portion of the TBP curve. An option is available which adjusts the total light ends to match the TBP curve at the mid-volume percent of the highest boiling light end component. See Assay Lightends Data for more information on handling of light ends.
By default the cutpoint range for each pseudocomponent is volumetrically integrated to determine its average normal boiling point. Alternatively, you may select the midpoint method to determine the average normal boiling point for each pseudocomponent. The same cutpoint range (% on TBP) is used for the pseudocomponent gravity and molecular weight determination. All other needed component properties are derived from the pseudocomponent normal boiling point, gravity and molecular weight.
Pseudocomponent Gravities
If a gravity curve is provided for the assay, it is used in the following manner:
If data are not given up to 95 LV or weight percent, quadratic extrapolation is used to generate an estimated 100 % gravity. Pseudocomponent gravities are next determined by integration of the gravity curve over the pseudocomponent cutpoint ranges.
The stream average gravity is compared to the supplied average. If they do not agree the estimated 100% gravity is adjusted as needed. If the supplied gravity curve includes the 95 % point or higher, the entire gravity curve is normalized until the stream average is met. Obviously it is best to estimate and supply the 95 % point so that the gravity curve correction is spread over the entire curve instead of only the last component.
If a gravity curve is not given, PRO/II generates the curve using the Assay Characterization Options selected for the problem. See Gravity Curve Generation Methods for more information on gravity curve generation.
Pseudocomponent Molecular Weights
If a molecular weight curve is provided it will be extrapolated and interpolated quadratically as needed to cover the entire range. The molecular weights for the pseudocomponents are determined by integration of the molecular weight curve over the cutpoint ranges. When the stream average molecular weight is also provided, the curve is adjusted to meet the stream average. All adjustment is put in the last component when the 100 % point is not supplied. The adjustment is spread through the entire curve when the 100 % point is supplied. No adjustments are made when the stream average value is not given.
If a molecular weight curve is not given, PRO/II generates the curve using the Assay Characterization Options selected for the problem. See Pseudocomponent Molecular Weight Estimation Methods for further details.
Note: Molecular weight is the most difficult property to predict accurately and should be supplied whenever possible for the most accurate characterization, especially for cracked stocks, synthetic fuels, tar sands, etc.
Pseudocomponent Critical Properties and Ideal Gas Enthalpies
These properties are needed for the various thermophysical methods. See Pseudocomponent Criticals, Ideal Gas Enthalpy Estimation Methods for further details.
Blending of Pseudocomponents
As previously mentioned, all assay streams within a problem which use the same TBP Cutpoint set are blended to produce a set of common blend components. All streams in the TBP Cutpoint set are then represented with the blend components. See Blending of Pseudocomponents for details on the blending process.
Related topics