Extended Hydrocarbon Analyses using VMGSim’s PIONA Characterization
Herbert Loria - VMG Calgary
The VMGSim’s PIONA (Paraffins, Iso- Paraffins, Olefins, Naphthenes and Aromatics) Characterization technique along with the Oil Source unit operation can be used to create component slates that represent extended hydrocarbon analyses. The PIONA Characterization in VMGSim is flexible enough to encode in the slate compounds known chemical characteristics of the analyses ranging from simple properties such as molecular weight and density to vapour pressure data. The Oil Source unit operation is a tool that generates the best combination of PIONA characterization compounds that match the compositional information and physical properties of a hydrocarbon fluid.
Extended hydrocarbon analyses can be used to characterize hydrocarbon fluids and generate a representative compositional slate in VMGSim’s Oil Source unit operation. These analyses can be found in this unit operation expressed as Carbon Number (Cn) and PIONA Compositional Analyses.
The Cn Compositional Analysis combines the composition of low molecular weight pure compounds (commonly known as Light Ends) and laboratory defined Carbon number groups. These compositions are obtained through a chromatographic analysis where the heavy hydrocarbons are lumped into a single group called hydrocarbon “plus” fraction (Cn+). These types of analyses are quite popular in the oil and gas industry because they require smaller samples, less time and less cost than a distillation curve analysis; these assays are also becoming prevalent in the tight fluids production industry.
The PIONA Compositional Analyses contain the PIONA distribution per carbon number in addition to the light ends definition. These types of analyses are not very common but they are very useful when characterizing product streams, especially gasoline and diesel fuels.
Extended Hydrocarbon Analysis in VMGSim: Carbon Number (Cn) Compositional Analysis
Entering the Data
The first step to use this type of analysis is to add the pure components acting as Light Ends in the analysis to the current Property Package. To add a Cn Analysis in the Oil Source unit operation, open the unit operation and check the Carbon Number (Cn) Compositional Analysis box and select Cn as the type from the Laboratory Analyses frame in the Summary tab:
This will open the Cn Analysis tab that contains the Options and Cn Analysis frames:
The Options frame contains the analysis basis and the definition of the Cn group ranges. The basis can be entered in Mass (Default), Mole or Volume (Standard Liquid Volume) basis. The Carbon number ranges can be defined by the actual Carbon number or by Boiling points.
The Cn Distribution Settings frame contains the Carbon number distribution limits and, a toggle to define if the light ends will participate in the Cn distribution. The default starting and plus Cn’s are 7 and 20, respectively. The Add Light Ends to Cn Distribution check box defines if pure light ends will participate in the Cn Distribution definition.
If the Cn Groups from the Cn Distribution are defined by boiling point ranges, then an extra toggle will be shown to decide if default values for the boiling points are to be used. The default Carbon number boiling point ranges are based on tables presented in the work of Katz and Firoozabadi . If the toggle is turned off, then the user can provide a particular definition for the boiling point ranges.
The Total Distribution frame shows the entered Light Ends and Cn Distribution fractions, if the total fraction does not add up to 100%, then a Warning message is displayed and a Normalize Analysis button will appear to fix the entered fractions.
The Cn Analysis frame contains grids where the bulk and Cn+ properties can be entered as well as the tables where the compositional distribution is added. These experimental variables used to be in the Summary tab in VMGSim 9.0 but for convenience they are now showed in the Cn Analysis tab (VMGSim 9.5).
The compositional analysis is added in the Light Ends and Cn Distribution tables. The Light Ends table contains all the pure components available in the current case except for Water (the Cn Analysis should be always in dry basis). Most of the light ends components are low carbon number hydrocarbons in addition of some low molecular weight gases, like CO2, H2 or N2. The Cn Distribution table has the carbon number and boiling point ranges (in case the option is selected) for which a reported distribution exists; note that the last Cn represents the plus fraction.
If the Add Light Ends to Cn Distribution check box is on and the component slate contains pure components with carbon numbers higher than the Starting Cn, then these pure components will not be shown in the Light Ends table, instead they will take part of the Cn Distribution calculations.
Comments on Cn Analysis Calculations
The Oil Source unit operation internally creates component groups based on their normal boiling point and assign a mass distribution based on a Gamma probability function, the parameters of the distribution can be seen in the inside the Regressed Parameters frame from the Settings tab. The mass distribution among components of the same PIONA species is controlled by the Splits of each group than can also be seen in the Regressed Parameters frame.
In the specific case of the Cn Analysis, the pure component light ends do not participate in the internal groups distribution and their compositions are directly passed to the final composition slate. Each internal group has a minimum and a maximum carbon number and an initial and ending boiling point; these boundaries are used to normalize the mass content of each group based on the information given in the Cn Distribution table and the definition of the Carbon Number group ranges, in this way the compositions among Carbon number groups are always matched.
It is important to note that preferred basis to input the Cn Analysis data is Mass since all the internal calculations are mass based; if the data is given in Mole or Volume basis, hydrocarbon standard molecular weights and liquid densities at 60 F from Katz and Firoozabadi  are used to interconvert the basis. In addition, the Mass basis is more reliable because the laboratory data is measured in this basis and converted to mole and volume basis based on calculated molecular weights and liquid densities that are not always reported in the laboratory analysis. As an example, the following note can usually be seen in typical laboratory extended analysis and illustrates the previous comments:
“The compositional values are based upon a measured mass fraction, and assume a total hydrocarbon recovery from the chromatographic system."
If bulk physical properties are given in the Cn Analysis and they are to be matched, then they have to be entered in the Bulk Properties frame. In addition to these properties; any other provided experimental data or analyses can be matched (PIONA analysis, distillation, density and viscosity curves); however, is not typical for a laboratory to report both compositional or property curves and extended analysis.
In tight fluids extended analyses is common to report saturation pressures at a specific temperature, this property can be added to the regression when the Oil Source Application is Tight Fluid.
The following example will go through the addition of a Cn Analysis in the Oil Source, the Cn Analysis used in this example comes in three different bases, as a way to illustrate the Oil Source capabilities, the analysis will be entered in those three bases and properties will be regressed to show that identical results can be expected.
The following is a mock-up Cn Analysis for a light to medium oil:
Start by selecting Advanced Peng-Robinson as the Property Package and SI as the Unit System, then add the pure components from the list of the Cn Analysis:
Observe that Water is part of the added compounds but it does not participate in the Cn Analysis, it was added just to illustrate that Water can be part of the flowsheet compounds but it is not used in the Cn Analysis.
Open the PIONA Slate characterization and create a slate based on the following parameters, these parameters were chosen in order to guarantee a good resolution in each carbon number group based on the Cn Analysis.
Note that the Olefins and Dehydrated Aromatics groups were eliminated from the slate because these groups are not typical found in an oil feed, they can be added to the slate (and they will have zero composition) but just for the sake of the example they were not selected.
Now, add an Oil Source unit operation to the flowsheet and open it, check the Oil / Refinery Application and the Cn Compositional Analysis box from the Laboratory Analyses frame; the Cn Analysis tab will appear and then enter the analysis data in Mass basis from the previous table:
Go to the Settings tab and uncheck the Olefins box from the PIONA Family Inclusion frame since olefins will not participate in the calculation:
At this moment the composition slate has been calculated based on the default Gamma distribution and split parameters from the Settings tab, it can be seen that the pure components mass fractions have been honoured in the slate.
Go to the back to the Cn Analysis tab and add the bulk physical properties of the analysis (molecular weight and liquid density at 60 F) in the Bulk Properties frame:
Now the parameters of the Oil Source unit operation can be regressed to match the experimental variables, before that go to the Settings tab and change the Regression Tolerance to 1.00E-03 in order to have a better match of the properties:
Click the Regress Parameters button (at the bottom of the unit operation) and observe the results:
It can be seen that the bulk molecular weight and liquid density at 60 F match perfectly. Now repeat the previous steps in two more Oil Sources and add the Cn Analysis from the original table in mole and volume basis in each one. The results after regression are shown in the next figures.
In order to compare the slates from the three Oil Sources with different Cn Analysis basis, a TBP distillation curve can be calculated for each one of them. To do this connect a Material stream to each Oil Source output and then connect a Distillation curve unit operation to the streams.
Now, we can compare the TBP’s from the three calculated Oil Sources. The following figure shows overlapping of the resulting TBP’s from each Oil Source:
Observe the overlapping of the TBP’s meaning that the three Oil Sources produced very similar slates for the same Cn Analysis in different basis, the small differences are due to the interconversion to mole and volume basis.
This example can be found in the Documentation folder of VMGSim in: C:\Program Files (x86)\VMG\VMGSimPkg\Documentation\Manual Examples\Oil Data Regressions\Oil Source\PIONAOilSourceCnAnalysisExample.vmp
Extended Hydrocarbon Analysis in VMGSim: PIONA Compositional Analysis
Entering the Data
The first step to use the PIONA Compositional Analysis in the Oil Source is to add the pure components present in the analysis to the current Property Package. To add a PIONA Compositional Analysis in the Oil Source unit operation, open the unit operation and check the Carbon Number (Cn) Compositional Analysis box and select PIONA as the Cn Compositional Analysis Type:
Once the option is selected, a new tab called PIONA Analysis becomes available.
The Options frame contains the analysis basis and plus Carbon numbers per PIONA family. The basis can be entered in Mass (Default), Mole or Volume (Standard Liquid Volume) basis, the default plus Cn’s are 10 for each family.
The Total Distribution frame shows the entered non-hydrocarbon and PIONA fractions, if the total fraction does not add up to 100%, then a warning message is displayed and a Normalize Analysis button will appear to fix the entered fractions.
The PIONA Cn Analysis frame contains tables where the compositional distribution per carbon number is added, the Non-Hydrocarbons table contains all non-hydrocarbon pure components available in the current case. Most of these components are low molecular weight gases, like CO2, H2 or N2. The Paraffins, Iso-Paraffins, Olefins, Naphthenes and Aromatics tables contain components (pure or pseudo components) that are part of these respective families. Here, the corresponding composition per carbon number and family will be specified.
Comments on PIONA Compositional Analysis Calculations
In the specific case of the PIONA Compositional Analysis, the specified components in the PIONA Analysis tab do not participate in the internal group distributions and their compositions are directly passed to the final composition slate, i.e. only the components lumped into the Cn+ fraction will be distributed by the Oil Source computations.
If bulk physical properties are given in the PIONA and they are to be matched, then they have to be entered in the Bulk Experimental Variable frame from the Summary tab. In addition to these properties; any other provided experimental data or analyses can be matched (distillation, density and viscosity curves). Any other property that is not available in the Oil Source can be also matched and added to the regression using the Custom Regression Case button.
The following example will go through the addition of a PIONA Cn Analysis in the Oil Source. The following experimental information corresponds to a low octane number heavy naphtha cut studied by Stratiev et al .
Naphtha PIONA Cn Characterization 
Naphtha ASTM D86 Distillation curve :
The naphtha specific gravity and Research Octane Number (RON) reported in  are 753 kg/m3 and 42, respectively.
Start by selecting Advanced Peng-Robinson as the Property Package and SI as the Unit System. Open the PIONA Slate characterization and create a slate based on the following parameters.
Now, add an Oil Source unit operation to the flowsheet and open it, check the Oil / Refinery Application, check the Cn Compositional Analysis box from the Laboratory Analyses frame and select PIONA as the Cn Compositional Analysis Type, the PIONA Analysis tab will become available.
Go to the PIONA Analysis tab and add the compositional information based on the given experimental analysis. Note that the non-hydrocarbons table is not shown since this type of components do not take part of this case.
Go to the Summary tab and add the given specific gravity and distillation curve by checking their respective boxes:
Add the distillation curve data in the Distillation Curve tab:
Observe that although the calculated standard liquid density is close to the experimental value, the calculated distillation curve does not match the experimental one because we still need to regress the Oil Source parameters. Before doing the regression there is still one experimental variable that we need to add: the Research Octane Number (RON). RON is not part to the Oil Source experimental variables; therefore, a Custom Regression case must be built to add this variable to the regression.
Click the Custom Regression Case button and this will bring the Model Regression form.
Then, go to the Equilibrium Results tab from the Oil Source and click the Extended Results box, scroll down the list and add the RON to the regression case using the left click menu:
Add the RON experimental value (42) using the Exp. Data tab from the Model Regression:
Now click on the Run button of the Model Regression to start the regression process, once the tuning is done click on the Set Regressed Vals as Specs button and observe the results.
Observe that now the standard liquid density, RON and Distillation Curve have a very close match with respect to the experimental data.
 Riazi, M. R. Characterization and Properties of Petroleum Fractions. West Conshohocken, PA: ASTM International, 2005
 Katz, D. L., Firoozabadi, A. Predicting the Behavior of Condensate/Crude Oils Systems using MethaneIntaraction Coeffiicients. Journal of Petroleum Technology, 30, 11, 1649-1655, 1978
 Stratiev, D., Tzingov, T., Shishkova, I., Pavlova, A. and Ivanova, P. Evaluationof Feasible Ways for Refinery Naphtha Streams Processing. 44th International Petroleum Conference, Bratislava, Slovak Republic, September 2009