Page 244 - Characterization and Properties of Petroleum Fractions - M.R. Riazi
P. 244

T1: IML
                           QC: IML/FFX
              P2: IML/FFX
  P1: IML/FFX
                        AT029-Manual-v7.cls
            AT029-Manual
                                           August 16, 2007
  AT029-05
                                                          17:42
         224 CHARACTERIZATION AND PROPERTIES OF PETROLEUM FRACTIONS
                                                              the following form [68]:
         set of equations:

                     B T = mX + B I                                         sat          B + P/P ce
                                                                       V = V   1 − C ln
                                                                                        B + P sat /P ce
                      m = 1492.1 + 0.0734P + 2.0983 × 10 −6  P 2
                                                                              1     ∂V       C
                                 5
                      X = B 20 − 10 /23170                             κ =−  V sat  ∂ P  =  (BP ce + P)


        (5.129)                                                                       T
                                    −3
                 log B 20 =−1.098 × 10 T + 5.2351 + 0.7133ρ o                           1/3          2/3
                                                              (5.132)  B =−1 − 9.070217τ  + 62.45326τ
                                   3
                     B I = 1.0478 × 10 + 4.704 P − 3.744 × 10 −4 P 2                     4/3
                                                                           − 135.1102τ + eτ
                         + 2.2331 × 10 −8 P 3
                                                                       C = 0.0861488 + 0.0344483ω
         where B T is in bar and ρ is the liquid density at atmospheric                                   2
                             o
                       3
         pressure in g/cm . In the above equation T is absolute tem-    e = exp 4.79594 + 0.250047ω + 1.14188ω
         perature in kelvin and P is the pressure in bar. The average
                                                                       τ = 1 − T/T c
         error from this method is about 1.7% except near the critical
         point where error increases to 5% [59]. This method is not  where V sat  is the saturation molar volume and P sat  is the sat-
                                                     o
         recommended for liquids at T r > 0.95. In cases that ρ is not  uration pressure at T. V is liquid molar volume at T and P
         available it may be estimated from Eq. (5.121) or (5.127). Al-  and κ is the isothermal bulk compressibility defined in the
         though this method is recommended for petroleum fractions  above equation (also see Eq. 6.24). T c is the critical temper-
         but it gives reasonable results for pure hydrocarbons (≥C 5 )  ature and ω is the acentric factor. P ce is equivalent critical
         as well.                                             pressure, which for all alcohols was near the mean value of
          For light and medium hydrocarbons as well as light  27.0 bar. This value for diols is about 8.4 bar. For other series
         petroleum fractions the Tait-COSTALD (corresponding state  of compounds P ce would be different. Garvin found that use of
         liquid density) correlation originally proposed by Hankinson  P ce significantly improves prediction of V and κ for alcohols.
         and Thomson may be used for the effect of pressure on liquid  For example, for estimation of κ of methanol at 1000 bar and
         density [66]:                                                                              −7   −1
                                                              100 C, Eq. (5.132) predicts κ value of 7.1 × 10  bar , which
                                                                 ◦
                                            	 −1              gives an error of 4.7% versus experimental value of 6.8 × 10 −7
                                      B + P                      −1

        (5.130)     ρ P = ρ P o 1 − C ln                      bar , while using P c the error increases to 36.6%. However,
                                     B + P o                  one should note that the numerical coefficients for B, C, and e
                                                              in Eq. (5.132) may vary for other types of polar liquids such
         where ρ P is density at pressure P and ρ P o is liquid density at
                            o                                 as coal liquids.
         reference pressure of P at which density is known. When ρ P o  Another correlation for calculation of effect of pressure on
         is calculated from the Rackett equation, P = P sat  where P sat
                                           o
         is the saturation (vapor) pressure, which may be estimated  liquid density was proposed by Chueh and Prausnitz [69] and
         from methods of Chapter 7. Parameter C is a dimensionless  is based on the estimation of isothermal compressibility:
         constant and B is a parameter that has the same unit as pres-  ρ P = ρ P o[1 + 9β(P − P )] 1/9
                                                                                  o
         sure. These constants can be calculated from the following
         equations:                                             β = α 1 − 0.89 ω exp 6.9547 − 76.2853T r + 191.306T r 2


                                                                            √
                  B                    1/3               2/3                  3          4
                     =−1 − 9.0702 (1 − T r)  + 62.45326 (1 − T r)   − 203.5472T + 82.7631T r
                                                                              r
                  P c
                      − 135.1102 (1 − T r) + e (1 − T r) 4/3        V c   Z c
        (5.131)                                                 α =     =
                                                                    RT c  P c
                                                      2
                   e = exp 4.79594 + 0.250047ω + 1.14188ω
                                                              (5.133)
                   C = 0.0861488 + 0.0344483ω
                                                              The parameters are defined the same as were defined in Eqs.
                                                              (5.131) and (5.132). V c is the molar critical volume and the
         where T r is the reduced temperature and ω is the acentric
         factor. All the above relations are in dimensionless forms.  units of P, V c , R, and T c must be consistent in a way that
         Obviously Eq. (5.130) gives very accurate result when P is  PV c /RT c becomes dimensionless. This equation is applicable
         close to P ; however, it should not be used at T r > 0.95. The  for T r ranging from 0.4 to 0.98 and accuracy of Eq. (5.134) is
                 o
         COSTALD correlation has been recommended for industrial  just marginally less accurate than the COSTALD correlation
         applications [59, 67]. However, in the API-TDB [59] it is rec-  [67].
         ommended that special values of acentric factor obtained
         from vapor pressure data should be used for ω. These val-  Example 5.8—Propane has vapor pressure of 9.974 bar at 300
                                                                                                    L
         ues for some hydrocarbons are given by the API-TDB [59].  K. Saturated liquid and vapor volumes are V = 90.077 and
                                                                V
                                                                            3
         Application of these methods is demonstrated in Example  V = 2036.5cm /mol [Ref. 8, p. 4.24]. Calculate saturated liq-
         5.8. The most recent modification of the Thomson method  uid molar volume using (a) Rackett equation, (b) Eqs. (5.127)–
         for polar and associating fluids was proposed by Garvin in  (5.129), (c) Eqs. (5.127) and (5.130), and (d) Eq. (5.133).
                              --`,```,`,``````,`,````,```,,-`-`,,`,,`,`,,`---
   Copyright ASTM International
   Provided by IHS Markit under license with ASTM             Licensee=International Dealers Demo/2222333001, User=Anggiansah, Erick
   No reproduction or networking permitted without license from IHS  Not for Resale, 08/26/2021 21:56:35 MDT
   239   240   241   242   243   244   245   246   247   248   249