CHAPTER 3 ASTM DISTILLATION, TRUE BOILING POINT DISTILLATION, AND EQUILIBRIUM FLASH VAPORIZATION RELATIONSHIPS FOR PETROLEUM FRACTIONS 3-0 INTRODUCTION ASTM and true boiling poi tillations are used to define the volatility characteristics of petroleum fractions and other complex mixtures Both are batch distllations which differ mainly in the degree of fractionation obtained during the distillation. ‘Atmospheric ASTM distillations are run in an Engler flask. No packing is employed, and reflux results only from heat losses through the neck of the flask. ASTM distillations are more widely used than TBP distillations because the former are simpler, less expensive, require Jess sample, and require only approximately one-tenth as much time. Furthermore, ASTM distillations are standardized, whereas TBP distillations vary appreci- ably in procedure and apparatus. ASTM distillation methods in use today are: ASTM Method D86: This method is used for the dis- tillation of motor gasolines, aviation gasolines, aviation turbine fuels, naphthas, kerosines, gas oils, distillate fuel oils, and similar petroleum products. It is carried out at atmospheric pressure. An exposed thermometer is used, and temperatures are reported without stem corrections. ASTM Method D216: This method is used for the di tillation of natural gasoline and is performed at atmo- spheric pressure. ASTM Method D1160: This method is used for heavy petroleum products which can be vaporized partially or completely at a maximum liquid temperature of 750 F at absolute pressures down to 1 mm Hg and condensed as liquids at the pressure of the test. Itis carried out at pressures between 1 mm Hg and 760 mm Hg, absolute. ‘Temperatures are measured with a thermocouple. ASTM Method D2887: Simulated distillation (SD) by gas chromatography appears to be the most simple, reproducible, and consistent method to describe the boiling range of a hydrocarbon fraction unambiguously This method is applicable to all petroleum fractions with a final boiling point of 1000 F or less. The method is also limited to samples having a boiling range of at least 100 F. Figure 3-0.1 shows a typical relation be- tween ASTM D86 and ASTM D2887 distillations for a naphtha fraction. It should be noted that ASTM D86 distillations are plotted in volume percent while simulated distillations are plotted in weight percent. 1987 (TBP) analytical dis- In ASTM distilations there may be a residue left in the distillation equipment as well as a difference be- tween the volume of the original charge and the sum of the distillate and residue. This difference is usually termed “Joss” and is generally thought of as volatile components of the charge which have not been recon- densed. For preparation of an ASTM distillation for conversion to a TBP or equilibrium flash vaporization (EFV), the percent distilled at the reported tempera~ ture is the sum of the distillate collected and the loss. When heated sufficiently hot, petroleum fractions undergo thermal cracking. Although’ a function of chemical composition, the amount and severity of ther- mal cracking increases with increasing boiling point, contact time, pressure and temperature. Previous edi- tions of this chapter included a correction for eracking for observed ASTM D86 temperatures above 475 F. In this revision, no correction for cracking is included. 800 700 * 600 500 400 TEMPERATURE 300 200 0 2 40 60 80 100 PERCENT VAPORIZED Soutce: Edniser and Pollock, Chem. Eng. Progr, 4495 (1948) Figure 3-0.1—Typical ASTM D86 and Simulated Distillation Curves for a Naptha Petroleum Fraction at3-0 ‘TBP distillation are performed in columns with 15 t0 100 theoretical plates at relatively high reflux ratios (ie., 5to 1 or greater). The high degree of fractionation in these distillation gives accurate component distribu- tions for mixtures. The lack of a standardized apparatus ‘and operational procedure is a disadvantage, but the variations between various laboratories are small be- cause a close approach to perfect component separation is usually achieved. The EFV curve is also a plot of temperature against percent by volume of liquid distilled, with the total ‘vapor in equilibrium with the unvaporized liquid at a ‘constant pressure. Each point on the EFV curve repre- sents a separate equilibrium experiment. The number of equilibrium experiments needed to define all por- tions of the EFV curve varies with the shape of the curve. Normally, at least five such experiments are re~ quired. Figure 3-0.2 also shows the EFV curves of a naphtha-kerosine blend at atmospheric and several superatmospheric pressures. ‘The tedious procedures necessary to obtain experi- mental EFV data have given impetus to the develop- ment of correlations for predicting EFV data from the analytical ASTM and TBP disilations. Both empirical and theoretical prediction methods have been devel- oped ‘Additional figures are presented for predicting the properties of the products of an equilibrium flash vaporization. Specifically, Figure 3B4.1 should be used to estimate the overhead product ASTM distillation curve, Figure 3B4.2 to estimate the bottoms product distillation curve, and igure 3B4.3 to estimate the APT eravities of both products Users are emphatically cautioned against relying heav- ily on results obtained from these correlations. Because of a lack of standardization and other inherent inade- ‘quacies in the methods, the existing ASTM, TBP, and EFV data are not sufficiently precise or consistent to develop good correlations. ‘The correlations ofthis chapter were developed using data for hydrocarbon stocks and fractions which in- cluded many components and exhibited smooth dis- fillation curves. The correlations do not apply to mix- tures of few compounds with widely different boiling points. Some published methods of predicting EFV data from the analytical ASTM and TBP distillations are given in references 7 through 9, 21 through 28, 39, and 40. These methods are usually based on dividing the complex mixtures into a series of narrow-boiling cuts or fractions, each characterized by its mean or average boiling point. Each cut is viewed as representing an 32 individual or pseudo component in a simulated mult component mixture. ‘A fecent approach to characterizing wide boiling range petroleum fractions is to use the method of continuous thermodynamics (12, 13, 77). This method provides a continuous distribution function for the Property desired. While such methods are still in the developmental stages, they show promise for carrying ‘out phase equilibrium calculations of undefined, multi- component mixtures. ‘A schematic diagram of the interconversion proce- dures is shown in Figure 3-0.3. Correlations are sum- marized in Table 3-0.4, Correlations in this chapter are empirical in nature and are arranged according to the various paits between ASTM, TBP, SD, and EFV relations. Section 3A1 ASTM-TBP (Atmospheric) Section 32. ASTM-TBP (Subatmospheric) Section 3A3 ASTM-SD (Atmospheric) Section 3B1 ASTM-EFV (Atmospheric) Figure 3-0.2—ASTM, True Boiling Point, anc Equilibrium Flash Vaporization Distillation Curves for a ‘Naphtha-Kerosine Blend 1987Section 382 ASTM-EFV (Subatmospheric) Section 383 ASTM-EFV (Superatmospheric) Section 385 TBP-EFV — (Subatmospheric) ‘TBP distillation data obtained at one pressure may be converted to another pressure by Procedure 344.1 ASTM distillation data may be similarly converted (as described in ASTM Method D1160) but with somewhat reduced reliability. The preferred procedure for con- verting ASTM distillation data taken at one pressure to that at another pressure is to obtain the corresponding ‘TBP data, convert these by Procedure 3A4-1, and then reconvert to ASTM distillation data. The procedure is ‘outlined for 10 mm Hg to 1 atmosphere in Table 3-0.4 Lamb and Sitar (45) theoretically and experimentally analyzed the conversion of distillation data from one pressure to another. They presented a modified Cox chart for petroleum fractions ASTM apparatus and procedures have changed over the years. TBP apparatus and procedures have never been standardized. Reproducibilities of ASTM and 30 ‘TBP distillations are a function of the distillation curve slope, and differences often become as high as 30 F. In addition, a large portion of the experimental data used to construct the correlations was obtained by non- standard test methods. Equilibrium flash data, while probably more repro- ducible than ASTM and TBP data, were obtained using different types of apparatus and many variations in pro- cedure. Procedure 3C1.1 is a computer method for flash cal- culations and estimation of equilibrium K-Values for petroleum fractions using a modified Soave-Redlich- Kwong (71) equation of state. This procedure is be- lieved to be the most accurate of the methods presented. Nore: A report which documents the basis upon which the original material in this chapter was selected has been pub- lished by the American Petroleum Institute as Documentation Report No. 3-66 and No. 2,3-86. Procedure 3D1.1 is docu- ‘mented in Documentation Report No. 2-81 Table 3-0.4—Summary of Correlations for Handling Distillation Data Data Available Pressure, ‘Type ‘mm Hg ‘Type ASTM D2887 (SD) 160 ASTM D86 ‘ASTM D6 760 TBP ‘ASTM D1160 10 ‘TBP. ASTM D1160 10 TEP ASTM D116 10 ASTM DB6, ‘ASTM 86 160 EFV ‘ASTM D1160 10 EFV EFV 160 EFV TBP 10 TBP ‘TBP 7160 EFV TEP 10 EFV EFV Subatmo- —- EFV spheric ASTM D2887 (SD) 160 TBP. ASTM D286? (SD) 760 EFV ASTM D1160 1 ‘TeP. ASTM D160. 1 ASTM D86 ‘ASTM D1160. 100 TBP. ‘ASTM D1160 100 ASTM D86 Conversion Data Desired Reet Pressure, Steps in ‘mm Hg Fig. 30.3 760 3 160 1 10 2 160 24 160 241 760 A 10 6 Superatmo. 7 spheric 760 4 760 fs 10 8 (Steps 5, 4 preferred if ASTM 86 available) Other Subatmo- 4 spheric 160 3,1 760 3s 160 4.24 760 4,204,1 760 424 760 aa Note: All ASTM D86 temperatures at 760 mm Hg are observed values. 1987 38