oEPA
          United States
          Environmental Protection
          Agency
          Industrial Environmental Research
          Laboratory
          Research Triangle Park NC 27711
EPA-6OO/7-79-097
April 1979
The Solubility of Acid
Gases in Methanol

Interagency
Energy/Environment
R&D Program Report

-------
                  RESEARCH  REPORTING SERIES


Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental  technology Elimination  of  traditional  grouping  was consciously
planned to foster technology transfer  and a maximum interface m related fields.
The nine series are:

    1.  Environmental Health  Effects Research

    2  Environmental Protection Technology

    3.  Ecological Research

    4.  Environmental Monitoring

    5.  Socioeconomic Environmental  Studies

    6.  Scientific and Technical Assessment Reports (STAR)

    7.  Interagency Energy-Environment Research and Development

    8.  "Special" Reports

    9.  Miscellaneous Reports

This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH  AND  DEVELOPMENT series Reports .n this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program These studies relate to EPA's mission to protect the public
health  and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of  domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of  the transport of energy-related pollutants and their health and ecological
effects; assessments  of, and development of,  control  technologies for energy
systems; and integrated assessments of a wide'range of energy-related environ-
mental issues.
                        EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for  publication. Approval does not  signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.

This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                 EPA-600/7-79-097

                                           April 1979
The Solubility of Acid
   Gases in Methanol
                  by

    J.K. Ferrell, R.W. Rousseau, and D.G. Bass

        North Carolina State University
      Department of Chemical Engineering
        Raleigh, North Carolina 27650
            Grant No. R804811
        Program Element No. EHE623A
       EPA Project Officer: N. Dean Smith

   Industrial Environmental Research Laboratory
     Office of Energy, Minerals, and Industry
       Research Triangle Park, NC 27711
               Prepared for

   U.S. ENVIRONMENTAL PROTECTION AGENCY
      Office of Research and Development
           Washington, DC 20460

-------
                             ABSTRACT
     A coal gasification/gas cleaning facility has been constructed
at North Carolina State University as part of a study funded by the
Environmental Protection Agency.  A major part of this facility in-
volves absorption and stripping of acid gas constituents, and the
description of operations to carry out these processes requires in-
formation on the equilibrium behavior of the constituents with the
system's solvent.  Two approaches may be used to obtain this inform-
ation:  measuring extensive equilibrium data or developing a thermo-
dynamic model that uses limited data to predict equilibrium behavior.
The latter approach is suggested.  The system chosen for study re-
flects current opinion concerning the most important candidate for
industrial acid gas removal systems.

     A thermodynamic model was developed to predict phase equilibrium
in the methanol-carbon dioxide-nitrogen-hydrogen sulfide system based
on parameters determined from binary vapor-liquid equilibrium data
available in the literature.  An experimental apparatus for obtaining
high pressure vapor-liquid equilibrium data was used to obtain data
for the methanol-carbon dioxide system at -15.0°C and at pressures
froin_5 to 16 atmospheres.  Results showed excellent agreement with
published data.  Equilibrium data were obtained for the methanol-car-
bon dioxide-nitrogen-hydrogen sulfide system at -15.0°C and pressures
ranging from 8 to 35 atmospheres.  Experimental liquid phase composi-
tions and temperatures were used to predict vapor compositions and
totel pressures.  Predicted values showed an average deviation from
expfrimental data of 21 percent for vapor compositions and 10 percent
for total pressures indicating the possibility of ternary effects not
acccjnted for by the model.

-------
                         TABLE OF CONTENTS



                                                               Page




ABSTRACT  	   ii




FIGURES  	   iv



TABLES  	    v






1.  INTRODUCTION  	    "1



2,  FUNDAMENTAL EQUATIONS  	    5



3.  MODEL DEVELOPMENT  	   10



4.  EXPERIMENTAL EQUIPMENT AND PROCEDURE  	   34




5.  EXPERIMENTAL RESULTS  	   38



6.  SUMMARY OF RESULTS  	   44






LITERATURE CITED  	   45




NOMENCLATURE  	   48

-------
                                 FIGURES

Number                                                                Page
   1.   Solubility of gases in methanol 	   3
   2.   P-x Diagram for CH3OH(1 )-C02(2) 	  20
   3.   P-x Diagram CH3OH(1)-H2S(2)  	  22
   4.   P-x Diagram for CH-OHO )-N9(2)  	  25
   5.   P-x Diagram for C02(l)-H2$(2)   	  27
   6.   P-x Diagram for C02(1)-N2(2)  	  30
   7.   P-x Diagram for H2S(1)-N2(2)  	  32
   8.   Experimental  Equipment  	   35
   9.   Calibration  apparatus  	   37
 10.   P-x  Diagram  Experimental CH3OH(1)-C02(2) Data  	   39

-------
                                  TABLES


Number                                                                  Page

   1.   Composition of raw gas  from coal  gasification  	   6

   2.   Binary equilibrium data from literature   	   7

   3.   Relationships  of binary Margules  parameters  to  multicomponent
            parameters   	  11

   4.   Binary interaction constants for  Redlich-Kwong  equation   	  13

   5.   Pure  component parameters   	  13

   6.   Constants  for  Equation  (12d)   		  16

   7.   Margules parameters   	  17

   8.   Calculated  and equilibrium  data for methanol-carbon
            dioxide at  298°K  	  18

   9.   Calculated  and equilibrium  data for methanol-carbon
            dioxide at  258°K  	  19

  10.   Calculated  and equilibrium  data for methanol-carbon
            dioxide at  243°K  	  19

  11.   Calculated  and experimental  equilibrium data for
            methanol-hydrogen  sulfide at 258°K   			  21

  12.   Calculated  and experimental  equilibrium data for
            methanol-hydrogen  sulfide at 248°K   	  21

  13.   Calculated  and experimental equilibrium data for
            methanol-hydrogen  sulfide at 273°K   		  23

  14.   Calculated  and experimental equilibrium data for
            methanol-nitrogen  at 298°K  	  24

  15.   Calculated  and experimental equilibrium data for
           methanol-nitrogen  at 310°K  	  24

  16.   Calculated and experimental equilibrium data for
            carbon dioxide-hydrogen sulfide at 293°K  	  26

  17.  Calculated and experimental equilibrium data for
           carbon dioxide-hydrogen sulfide at 277.6°K 	  26

  18.  Calculated and experimental equilibrium data for
           carbon dioxide-nitrogen at 273°K  	  29

-------
                               TABLES (Contd)


Number
                                                                        Page
  19.  Calculated and experimental equilibrium data for
            carbon dioxide-nitrogen at 273°K  	  29

  20.  Calculated and experimental equilibrium data for
            hydrogen sulfide-nitrogen at 300.1°K  	  31

  21.  Calculated and experimental equilibrium data for
            hydrogen sulfide-nitrogen at 277.7°K  	  31

  22.  Comparison of model and experimental equilibrium data
            for carbon dioxide-methanol at 258°K  	 --  40

  23.  Experimental vapor-liquid equilibrium data for mettianol-carboi
            dioxide-nitrogen-hydrogen sulfide at -15.0°C	  41

  24.  Comparison of calculated and experimental pressures and gas
            compositions for CCL-H?S-N?-methanol  	 --  42
                                      VI

-------
                             SECTION 1

                             INTRODUCTION


      North  Carolina State University, under the sponsorship of the
 Environmental  Protection Agency,  Industrial Environmental Research
 Laboratory  at  Research Triangle Park, is engaged in a research pro-
 ject  to study  the environmental effects of coal gasification.  So
 that  the results of this research will represent what might be ex-
 pected from full-scale gasification operations, a highly flexible
 and complete coal gasification and gas cleaning facility has been
 constructed for the project.  The overall objective of the project
 is to characterize completely the gaseous and condensed phase emis-
 sions from  typical coal gasification-gas cleaning processes and to
 determine how  emissions depend upon various process parameters.

     The North Carolina State University facility consists of a con-
 tinuous fluidized bed gasifier, devices for removing participates,
 condensables,  and soluble matter from the raw synthesis gas, and an
 acic gas removal system (AGRS).  The gasifier operates at pressures
 up to 100 psig, has a capacity of 50 Ib coal/hr, and can run with
 eitfer steam-02 or steam-air feed mixtures to produce roughly 25 SCFM
 of crude synthesis gas.  The AGRS is modular in design so that alter-
 native absorption processes may be evaluated with a minimal amount of
 system modification.

     The primary purpose of an AGRS is to remove most of the sulfur
 containing  compounds and, for some applications, the carbon dioxide.
 The processes  available to accomplish this employ either chemical or
 physical  absorption.  In chemical  absorption, acid gas species react
 with alkaline compounds or solvents; the solvent is regenerated by
 heating.   One of the problems with processes employing chemical ab-
 sorption for gas cleanup is that some of the crude gas components such
 as carbonyl  sulfide often react with the solvents to form decomposi-
 tion products which prevent complete regeneration of the solvent.  Pro-
 cesses employing physical  absorption capitalize on the differing solu-
 bilities  of various gases in liquids.   Polar liquids are best suited
 for these processes because they have a high degree of selectivity be-
 tween acid gases and the valuable  gas components.   Methanol serves as
 the solvent in Lurgi's proprietary Rectisol  process.  It is a good sol-
 vent for hydrogen sulfide,  carbonyl  sulfide, carbon disulfide and car-
 bon dioxide.  As shown by plots of equilibrium constants in Figure 1,
methanol  will  absorb hydrogen sulfide preferentially to carbon dioxide.

     The  economics and desirability of these two types of gas cleanup
processes depend heavily on their  particular application.   However, the
use of methanol is particularly attractive and economical  for cleanup
systems used in conjunction with coal  gasification facilities for some
applications (1).   This has been demonstrated  by Lurgi in  their design
and successful  operation of the Rectisol  process.

     Because of its immediate importance,  methanol  is  the  first solvent
that will  be studied in the NCSU gas cleaning  test facility and solubility
data are needed to evaluate the AGRS performance.

-------
  Figure 1.  Solubility  of gases in methane 1.
 x"103
o
 ii
 O,
    io2
                             hydrogen
                                    nitrogen
                   carbon  monoxide
                         temperature *F

      -ISO   -1OO   -5O     O      5O     1OO     ISO

-------
      In the facility the crude gas consists mainly of nitrogen, car-
bon monoxide, carbon dioxide, hydrogen, ethane, water, methane, hydro-
gen sulfide, other sulfur gases, and small amounts of higher hydrocar-
bons.  The gas  is dehydrated, compressed to 500 psi, cooled, and fed
to a  packed absorber column, where it is contacted with cold methanol.
Carbon dioxide, hydrogen sulfide and other sulfur gases are absorbed
from  the crude  gas which is then discharged to a flare.  Methanol en-
ters  the absorber at about -30°F.  The circulating solvent flows from
the bottom of the absorber through a pressure reduction valve to a
flash tank at 15-25 psig.  Liquid solvent leaving the flash tank is re-
generated by stripping carbon dioxide and hydrogen sulfide with nitrogen.

     A careful  search of the literature determined that the data and
correlations required for an accurate analysis of the methanol-based
process are not available.   Binary vapor-liquid equilibrium data for
some  of the system components were located, but there appears to be
very  little multicomponent data.  It is clear that experience and pro-
prietary data or rough estimates based on ideal solution behavior have
been used in the design of existing facilities.  It is not feasible to
meet the requirements of the current study using either of the above ap-
proaches; without accurate vapor-liquid equilibrium data it will be al-
most  impossible to analyze the operation of the AGRS.

      It is not  practical to develop an exhaustive data base for multi-
component vapor-liquid equilibrium behavior.  An approach that has been
used satisfactorily to describe systems similar to those encountered
here  is to develop a thermodynamic model that uses binary equilibrium
data to predict multicomponent behavior.

     The_purpose of this study was to begin developing a useful model
for multicomponent equilibrium behavior and to obtain binary vapor-liquid
equilibrium data for systems consisting of carbon dioxide, nitrogen, hy-
drogen sulfide and methanol.  The range of temperatures, pressures and
compositions that have been considered in these studies include those
found in typical acid gas removal systems.  Both theoretical and experi-
mental parts of this effort have been developed to a point that it seems
appropriate to describe the progress and point out directions for future
efforts.  This report does  that in its following sections.
                             SECTION 2

                       FUNDAMENTAL EQUATIONS


     A typical composition of gases to be treated in acid gas removal
systems is shown in Table 1.   The nature of these components indicates
that both gas and liquid solutions containing these materials will  not
behave ideally at the system conditions; simple thermodynamic expressions
can not be used to describe their equilibrium behavior with methanol.   It
is necessary, therefore, either to take extensive multicomponent equili-
brium data or to develop thermodynamic models using binary and perhaps

-------
 Table  1.   Composition of  Raw Gas from Coal Gasification (3).





    Component                With 02 (Vol. %)      With Air (Vol. %)




 carbon dioxide, CCu



 carbon monoxide, CO



 hydrogen, H?



methane, CH.



nitrogen, N~



argon, A



hydrogen sulfide,  HLS



carbonyl  sulfide,  COS
31.
15.
45.
8.
0.2
0.2
6800 ppmv
400 ppmv
10.
19.
22.
0 5
48.
. . .
0.5
150 ppmv

-------
 limited  multi component  equilibrium data.  The  latter approach was used
 in  this  research.

      Systems  consisting  of C02,  H2S,  N;> and methanol were chosen for
 study.   An  extensive  search   of the  literature for equilibrium pro-
 perties  of  binary mixtures formed by  methanol  and these three major
 acid  gas components is  summarized in  Table 2.  While some miilticompon-
 ent data are  available  in the literature, it often is not in the range
 of  conditions encountered in the AGRS processes or does not consider
 one or more of  the significant components of the AGRS mixtures.


 Models and  Correlations

      The objective of thermodynamic models and correlations is to deve-
 lop expressions relating temperature, pressure and compositions in co-
 existing phases at equilibrium.  For  example,  given a system pressure
 and a liquid  phase composition,  a thermodynamic model could be used to
 determine the system temperature and  composition of a gas phase in equi-
 librium  with the liquid.  To obtain useful expressions relating these
 quantities  it is necessary to begin with the fundamental thermodynamic
 relationship which states that the fugacities  of a component in coexist-
 ing phase are equal at equilibrium; for vapor-liquid equilibrium
     Two methods can be used to express these fugacities in terms of
systam temperature, pressure and composition of each phase.  The first
follows from an exact thermodynamic treatment of the system to give
either of the following equations:
                       VT.V.n. ,- " T-
                        i       J/i


                       '  \            RT.
                                      r            ]               (3)


To use Equation 2 a pressure explicit equation of state must be avail-
able and it must describe the system over the limits of the integration.
Equation 3 requires a volume explicit equation of state.  This technique
can !)e and is used to estimate gas phase fugacities but, since no single
equa ;ion of state can be used to describe a mixture from ideal gas condi-
tions  to liquid conditions at elevated pressures, a second approach is
used to estimate liquid fugacities.

     Liquid fugacities may be defined in terms of their deviation from
ideal  solution behavior, as given by Raoult's or Henry's laws.  The work-
ing equation is

-------
Table 2.  Binary equilibrium data from literature
Pressure
Binary Range, Atm
co2
co2
co2
co2
co2
co2
N2-
H2S
H2S
co2
co2
N2-
N2-
N2-
N2-
- CH3OH
- CH3OH
- CH3OH
- CH3OH
- CH3OH
- CH3OH
CH3OH
- CH2OH
- CH3OH
- H2S
- H2S
H2S
H2S
co2
co2
2,33
1 ,30
16,43
2,60
7,69
5,80
35,63
2,10
1
10,89
7,82
3,136
11,204
24,143
37,125
Temperature
Range, °C Source
0,-SO
185, -50
60,15
25
0,75
40,25
60,15
0.-25
30, -30
89,0
91, -48
-45, -73
71, -17
15, -20
25, -40
Yorizane et al (4)
Bezdel and Teodorovi:h (5)
Hemmaplardh and King (6)
Katayama et al (7)
Krichevakii and Lebedova (8)
Ohgaki and Katayama (9)
Hemmaplardh and Kin
-------
 where YI defines the deviation of the solution from ideal behavior
 and     f." 1S the fugacity of the reference state at the system con-
 ditions.  If the ideal solution is defined in terms of Raoult's law
 then
 When this definition of the reference state is used

      Y.J ->• 1.0  as  x. •+ 1.0.


 A major difficulty associated  with using the Raoult's law definition
 of the ideal  state is evaluating  p* for components having a critical
 temperature  below the system  temperature.   In  this case a hypotheti-
 cal  value of the  vapor pressure can be  estimated  by extrapolation  of
 P* vs.  T expressions.

      When the ideal  solution  is defined  in  terms  of Henry's law


      fi   = ^fi  ^inf  =  Hi (HenrY's  constant)                      (6)


 Using  this definition  of the reference  state T. -> 1.0 as  x   + 0
 0  Connell  (16)  has noted severe difficulties ir\ using Henry's law  to
 define  the reference  state for multicomponent  systems.  Therefore,
 Kaoult  s  law  has  been  used in this  research.

     Although significant progress  has been made  in developing tech-
 niques  to predict activity coefficients from solution  theory, it is
 almost always necessary to rely on  experimental data  for  the develop-
 ment of suitable correlative and predictive routines.  The  van Laar,
 Margules, Wilson  (17) and UNIQUAC  (18) equations  have  been  used with
 varying degrees of success  to correlate activity  coefficients.   In most
 cases fairly accurate predictions  of multicomponent system  behavior can
 be obtained by using parameters evaluated by fitting these equations to
data for each binary mixture making up the system.

-------
                              SECTION 3

                          MODEL DEVELOPMENT


      The  primary  constituents  of methanol -based  acid  gas  removal  sys-
 tems  are  COz,  ^S,  N£  and methanol.   A description  of the phase  equi-
 librium behavior  of quaternary mixtures  of these components  is neces-
 sary  for  a  quantitative  analysis of  absorber  and stripper operations
 in  these  processes.

      The  initial  approach in  this  investigation  was patterned after
 that  reportedly used industrially  for process  development and design
 studies with high pressure nonideal  systems  (19).  A  modified Redlich-
 Kwong equation was  used  to calculate  fugacities  of  the qaseous compon-
 ents, the four suffix  Margules  form  of the Wohl  equation  was used to
 correlate activity  coefficients,  and  the Chao-Seader  equatior. (20) was
 used  to estimate  the fugacity  of the  reference state  (pure liquid at
 the system  conditions).

     The Wohl equation gives the  activity  coefficient of  a component
 in a mul ticomponent solution as
,  = 4 I   I   Zx.x.x  Piik  -  I   I   I   I  x.x x x, ;•:
1     j = l  k=l  £-1  J k l  1Jk£   i = l  j = l k=l f.= l   n  J k ''  ljk'
                                                                  (7)


where the values of 3--kf are related to binary Margules constants and
a ternary constant;   J  " the Margules constants are defined for a
binary mixture by the expression

In Y- = x.2 [A.. + 2(A.. - A.. - D..) x. + 3D., x.2]              (8)
   Yi    J    iJ      Ji    ij    iJ   i     ij  i                l  '

Table 3 gives the definition of 8,-jk(, in terms of the binary Margules
constants.  Following the observation of Adler, Friend and Pigford (21),
the ternary constants C*-^ were set equal to zero.  Nonlinear least
squares techniques can be used to estimate the Margules constants in
Equation 8.  Adler, Ozkardesh and Schreiner (22) showed that the tem-
perature dependence of these parameters follow either a directly pro-
portional or inversely proportional relationship with temperature.

     The Redl ich-Kwong equation of state as modified by Prausnitz and
Chueh (23) was used with Equation 2 to estimate vapor phase fugacity
coefficients.   This equation of state is described in the following re-
lationships.

-------
Table 3.   Relationships  of  binary  Margules  parameters to multicompon-
          ent parameters.
                                      Q
 Combination                           ijkfc

i = j = k = s,                           0

i = j = if k                 A^/4

i = * t J = k                 (A.. + A...  -  D...)/6

J = *.* M i                 C(Akj , AIJ + A.k1)  -
where
           = Aik
     C*.,  = ternary constant
      i JK
      ,  _

         "

               R T
                  lL                                            Ob)
       a-  ?      y.y.a.-                                       (9c)
       «    /.    L  J iJ -\   t i
                  ?   2  5
                 R  T
                                                                 (9d)

-------
                TT
                        ?  9  R
                         T
                           U                                  (9e)
           4T
                                                              (9g)
             (v
 All  of the mixing rules were taken from Prausnitz (24)  except (9g)
 which came from Tarakad and Darmer (25).   The binary constant,  k-j-,
 is independent of composition, temperature and density.   Table  4
 lists the values k-jj used in this study and their sources.   The dimen-
 sionless pure component constants are characteristic of  a particular
 substance and are calculated from volumetric data for the pure  satur-
 ated vapor.   All pure component parameters used are listed  in Table  5.

      An expression for the vapor phase fugacity coefficient  of  a com-
 ponent in a  mixture of n components  is obtained by combining Equations
 2  and 9.
Table 4.  Binary interaction constants  for  Redlich-Kwong equation

                                  k
  System                           ij                           Source

CH3OH - C02                      0.063                           (25)

CH3OH - H2S                      0.14                            (25)

CH3OH - N2                       0.14                            (25)

C02 - H2S                        0.08                            (24)

C02 - N2                         0.048                           (25)

HS - N                          0.105                           (12)
                                  10

-------
Table 5.
Pure component parameters
Component a ^b
CH3OH
co2
H2S
N?
0.8451
0.4470
0.4340
0.4290
0.2075
0.0911
0.0882
0.0870
*
Tc
(°K)
512.6
304.2
373.2
126.2
*
Pc
(atm)
79.9
72.8
88.2
33.5
*
Vc
(cc/mole)
118.0
94.0
98.5
87.5
*
0)
0.559
0.225
0.100
0.040
 t
  From Prausnitz  (24)  except  values  for  methanol  which  were calculated
  from volumetric  data of Eubank  (27).

  From Reid,  et al.  (26).
                                n

                             2   ^  y   a
            -  m
          ,                 .
          K.       V-D    V-D      RT     h
                 DT1.5.2      T" "  V?b"  "     RT
                 RT    b
The molar volume of the mixture, V, is found by solving  Equation 9 for
the largest real root.

     Various equations were used to calculate the fugacities of the
reference state, chosen to be the pure liquid.  The exact thermodynamic
relationship given in Equation 5 was used for methanol.  The vapor pres-
sure for methanol was calculated from the Antoine equation.

log [P*(mm Hg)] = 7.87862 - 1473.11/[T(°K) - 43.247]              (11)


The Antoine constants used were determined from experimental vapor pres-
sure data from Eubank (27) for temperatures from -40°C to +30°C.  The
molar volume of methanol was calculated from a corresponding states cor-
relation of Chueh and Prausnitz (Reid, Prausnitz and Sherwood (26)).
                                11

-------
                   9Z  N(P-P*)  ,/q
     - P  = P[l +-±- - 1 '                              02
             s
N =  (1.0 - 0.89u)[exp(6.9547 - 76.2853 Tr + 191.3060 Tp


     - 203.5472 T^3 + 82.8731 Tr4)]                         (12a
                                                              2b)
pc = l/Vc                                                   02c)


Vr(j) = a.. + b. Tr + c. T2 + d.. T^ + 6j/Tr + f .  ln(l-Tr)    (12d)


The constants for Equation (12d) are given in Table 6.  Using the
above correlation for liquid molar volume and evaluating the integral
in Equation 5,  the following expression is obtained for the pure li-
quid reference fugacity.

                       P            9Z  N(P-P.*)

V • P,* V «P (gRT^V «' - -^ -- >8/9 - 1])  ("I
                        si ci           ci

     For carbon dioxide and hydrogen sulfide, a  three parameter reduc-
ed states correlation was used (28)

    f°
log _i.= i0g v(°) + a, log v(1)                              (14)


log v(0) = Bo + B1 Pr + B2 Pr2 - log Pp                     (14a)


B  = -20.651608 + 84.517272 Tr - 15.376424 1^ + 152.65216 Tf3

     - 84.899391 T 4 + 24.84688 T^ - 2.9786581  T^6         (14b)
                                12

-------
If0.8>Tr>0.3,
                                          ,0.286-
                                                                (He)
If 0.9  >  Tr >_ 0.8,  B1  =  0.321895 Tr  -  0.184316
If 1.8 >.Tr  >  0.9,  B1  =  58.16962  -  326. 54444 T
    + 775.11716 Tr  - 1006.8122 Tr




    +• 87.677429 T 6 - 9.2517986 T 7
                                      773.32667 T 4 - 351.56938 T 5
Table 6. Constants for equation (12d)
a . h . r . d . e .
1
0
1
2
JL _i
0.11917 0.009513
0.98465 -1.60378
-0.55314 -0.15793
_1
0.21091
1.82484
-1.01601
JL _1
-0.06922 0.07480
-0.61432 -0.34546
0.34095 0.46795
f .
JL
-0.084476
0.087037
-0,239938
If 0.8 >  Tr >_ 0.3  ,  B2  =  0



If 0.9 >  T  > 0.8  ,  B0  =  0.0549369  (0.8  - Tj
          r —        c                      '


If 1.0 >  T  ^0.9  ,  B2  =  0.673344 x  10"3 -  0.685226 x  10
                                                        -2
 If T   >_ 1.0  , B2 = 0.72203901 - 2.7182597 Tr + 3.984423 T^


                   - 2.8712448 T 3 + 1.0202739 T 4 - 0.14314712
log
         --  log ^
                   (1
                                      log y
                                           (1)
                                       3P
                                                                  (14d)
                                                                  (He)
                                  13

-------
          "log vn c    = - 660.08698 + 7766.7774  T  -  40007.379  T  '
               U. b                               r              r

                        + 116582.6 Tr3 - 209756.24  Tr4


                        + 238673.14 T 5 - 167856.45 T 6
                                     r               r

                        + 66762.602 Tr7 - 11504.984 Tr8         (14f)
 rp~  - =  -  0.28997623  +  0.96418856 Tr  -  1.3344703 Tp


            +  0.82575807  T  3  - 0.18939410 T 4                  (14n)
                         r               r                   v  J/


      For  nitrogen, the Chao-Seader equation with adjusted parameters
was used  (29).


    f°
log -p1 =  2.7365534 - 1. 981831 0/T  - 0.51487289 Tr

       +  0.042470988 T 2 - 0.002814385 Tr3 + (-0.029474696


       +  0.021495843 T) P  - log P  + <*.
     To use the techniques described, activity coefficients were evalu-
ated from literature x-y-P-T data on the six possible binary mixtures
of the four component system:  methanol-C02, methanol-^S, methanol-N2,
C02- H2S, C02-N2, and H2S-N2-  Using the references listed in Table 3,
Margules parameters and their temperature dependence were determined us-
ing a non-linear least squares procedure as described by Bass (30).  The
results of these evaluations are summarized in Table 7 and discussed in
the following paragraphs.   Details of computer programs are given by
Bass (30).
                                  14

-------
Table 7.  Margules parameters
Binary
CH3OH - C02
CH3OH - H2S
CO, - H0S
l C-
C02 - N2
H2S - N2
(°K)
298.2
258.2
293.2
253.3
300.1
1.1672
1.0694
0.6607
1.2832
0.6165
A-i
0.594C
0.6608
0.5141
0.4359
0.7885
CH3OH -
                            n
                             i
                          0.5056

                          1.0068

                          0.5991

                          2.5135

                         -0.1280
298.2   0.6317   0.7683   1.6583
                                                   Temperature
                                                   dependence

                                                 A" = A1  T]/T2

                                                 A" = A1  T,/T2

                                                 A" = A1  T2/T]

                                                 A" = A1  T]/T2

                                                 A" = A1  (T-,

                                                 D" = D1  (Tg/^

                                                 A" = A1  T,/T2
                                                 D" = D'  T2/T1
     Model parameters for the methanol-carbon dioxide system were de-
termined from the 298°K data of Katayama et al.  (7).  The model fit is
illustrated in Table 8.  Predictions are compared with the 258°K and
243°K_data of Yorizane et al. (4) in Tables 9 and 10.  Vapor phase com-
positions at the latter two temperatures were not available for com-
parison.  Average deviations of the predicted equilibrium pressure from
measured values increase with further extrapolations from 298°K but
plots of the model  (Figure 2) appear to be asymptotic to the experimen-
tal values at higher concentrations.

     Model parameters for the methanol-hydrogen  sulfide system were de-
termined from the 258°K data of Yorizane et al .  (4); the fit is shown
in Table 11.  Vapor compositions were not given  and the concentration
of methanol was reported to be too low to be measured.  For the pur-
poses of determining activity coefficients, the  vapor composition of
methanol was assumed to be 0.0001.  Extrapolations to 273°K and 248°K
show an excellent fit to the experimental data except for high hydrogen
sulfide concentrations at 273°K where the assumed vapor composition is
expected to be most in error.  Figure 3 graphically compares the model
to Yorizane's data.
                                15

-------
Figure  2.  P-x Diagram for CH3OH(1)-C02(2!
 60
 50
 40
 30
 2O
  10
                                                  DATA OF



                                                  YORIZANE




                                                  A    258'K
                                                  DATA  OF




                                                  KATAYAMA



                                                  •    298'K
                0.2
O.4        O.6        O.8         l.O
                               16

-------
 Table  11.  Calculated and  experimental  equilibrium data for metha-
           nol-hydrogen  sulfide  at  258°K.
0.165                2 0
0.231                3.0
0.298                3 4
0.367                4.2
0.403                4.4
0.490                5.0
0.585                5.4
0.662                5.8

Average percent deviation in P = 3.7.
Table 12.
0.203
0.290
0.327
0.465
0.582
0.733
Calculated and experimental equilibrium data for metha-
nol-hydrogen sulfide at 248°K.

          P
           exp
         (atm)

          2.0
          2.5
          3.0
          3.4
          4.0
          4.3
Average percent deviation in P = 31
                               17

-------
Figure  3.  P-x Diagram CH,OH(•)-H?S(2).
                        \J      (—
10  r
 8
               0.2
                        DATA  OF
                        YORIZANE
                        •  273-K

                        A  258 -K

                        •  248 K
                        -  MODEL

__l	1	1	>
 0.4         0.6        0.8         l.O

-------
Table 13.   Calculated and experimental  equilibrium data for metha-
           nol-hydrogen sulfide at 273°K.
0.092                2.0
0.199                4.0
0.329                6.0
0.453                7.5
0.484                8.0
0.608                9.1
0.743                9.8
0.840               10.0

Average percent deviation in P = 7.8.
     High pressure vapor-liquid equilibrium data for the methanol-nitro-
gen system are scarce.   Model  parameters for this system were determined
from the 298°K data of Hemmaplardh and King (6) with the fit illustrated
in Table 14.  As shown in Table 15 extrapolations to 310°K produced good
agreement between predicted and experimental vapor phase compositions but
poor agreement between predicted and experimental pressures.  Figure 4
shows how the model fits both  sets of data.  More data are needed before
an improved fit can be obtained.  However, the region of primary inter-
est is for pressures at or below 35 atmospheres where the model predic-
tions are expected to be more  accurate.

     Model parameters for the  carbon dioxide-hydrogen sulfide system were
determined from the 293°K data of Bierlein and Kay (10); the resulting
fit is illustrated in Table 16.  Extrapolations were made to 277.6°K for
comparison with data from Sobocinski and Kurata  (11) with the resulting
fit shown in Table 17 and Figure 5.
                                 19

-------
Table 14. Calculated and experimental equil
ibrium data
for metha-
nol-nitrogen at 298°K.

X2
0.0241
0.0295
0.0483
0.0575
Average
P
exp
(atm)
35.8
35.1
49.8
62.8
Pcalc
(atm)
32.3
37.7
52.8
58.4
percent deviation
DP
(atm)
-3.5
2.6
3.0
-4.4
: in P,
y2
exp
0.9939
0.9938
0.9952
0.9958
7.6; in y,
y2
calc
0.9941 0
0.9934 -0
0.9954 0
0.9957 -0
0.02.

Dy
.0002
.0004
.0002
.0001


Table 15


X2
0.0355
0.0539
0.0559
0.0611
Average
. Calcul
nol-ni
exp
(atm)
35.0
43.7
47.7
48.6
ated and
experimental equil
ibrium data
for metha-
trogen at 310°K.
Pcalc
(atm)
40.0
52.1
53.2
55.8
percent deviation
DP
(atm)
5.0
8.4
5.5
7.2
: in P,
y?
L.
0.9891
0.9909
0.9914
0.9915
15; in y,
y?
^calc
0.9902 0.
0.9919 0.
0.9921 0.
0.9923 0.
0.09.

Dy_
0011
0010
0007
0003


Table 16
. Calcul
ated and
xide-hydrogen

X2
0.980
0.910
0.821
0.715
0.580
0.437
0.261
0.180
exp
(atm)
20.0
25.0
30.0
35.0
40.0
45.0
50.0
52.0
calc
(atm)
19.4
26.4
31.1
34.7
38.9
43.4
48.2
49.7
experimental equil
sulfide
DP
(atm)
-0.6
1.4
1.1
-0.3
-1.1
-1.6
-1.8
-2.3
at 293°K.
y2
exp
0.869
0.685
0.550
0.457
0.370
0.290
0.189
0.130
ibrium data

y?
^calc
0.851
0.633
0.533
0.462
0.381
0.293
0.192
0.147
for carbon dio-


Dy_
-0.018
-0.052
-0.017
0.005
0.011
0.003
0.003
0.017
Average percent deviation:   in  P,  3.4;  in y, 4.1.
                                20

-------
    Figure 4.   P-x  Diagram for CH-OH(1)-N9(2).
    60
    50
    40
<   30
Q.
    20
    10
                        298
                                              31O -
DATA  OF

HEMMAPLARDH

•  298 'K

•  31O 'K

- MODEL
  '           f
                  0.02       O.04       0.06        O.O8
                              21

-------
 Table 17.   Calculated  and experimental  equilibrium  data  for  cartx n
            dioxide-hydrogen sulfide  at  277.6°K.


           Pexp         calc     DP       y~        y~
  ^2_      (atm)       (atm)    (atm)       exp     __caj£       Dy_

 0.967      13.61      14.60     0.99      0.855      0.774     -0.081
 0.915      17.0       18.14     1.14      0.660      0.621     -0.039
 0.850      20.41      20.88     0.47      0.540      0.532     -0.008
 0.757      23.0       23.50     0.50      0.470      0.458     -0.012
 0.605      27.22      27.07    -0.15      0.355      0.363       0.008
 0.485      30.0       29.89    -0.11      0.214      0.29?       0.078
 0.340      33.0       33.15     0.15      0.210      0.210       0
 0.290      34.02      34.15     0.13      0.185      0.184     -0.001

 Average percent deviation:   in  P, 2.5;  in y, 7.3.
     Model parameters for  the carbon dioxide-nitrogen system were de-
termined from the 253°K data of  Kaminishi and Toriuini (15).  Extra-
polations to 273°K were compared to data from the same source.  These
results are shown in Tables 18 and 19 and in Figure 6.  The best fit
was obtained at pressures  below 80 atmospheres.

     Model parameters for  the hydrogen sulfide-nitrogen system we e
determined from the 300.1°K data of Besserer and Robinson  (13).  , x-
trapolation to 277.7°K was fit best by having the Margules constants
proportional to temperature to the 1.7 power.  The resultinc fits are
shown in Tables 20 and 21  and Figure 7.

     Poor fits were observed for both the methanol-nitrogen and car-
bon dioxide-nitrogen systems.  Data were not available over a wide con-
centration range for either system.  For instance, in the carbon diox-
ide-nitrogen system, data  at low nitrogen concentrations allowed an ac-
curate determination of fy] but the lack of data at low carbon dioxide
concentrations prevented an accurate determination of An 2-  Thus, the
model fits the experimental data more closely at low nitrogen concen-
trations anc1 consequently  at low pressures.   This problem was even more
pronounced in the methanol-nitrogen system where the data were even
more limited and had significant scatter.

     Good fits were obtained for the methanol-carbon dioxide, metha-
nol-hydrogen sulfide and carbon dioxide-hydrogen sulfide systems.
Among these three systems, the average percent deviation in pressure
was 6.6 and in vapor phase composition was 2.3.  In the absence of tern-
ary effects, it is reasonable to assume that the model will provide a
good estimate of total  pressure and vapor phase composition for a mix-
ture consisting of these three components.  Predictions for the four
component system containing nitrogen is expected to be less reliable.

-------
    Figure 5.   P-x Diagram for CCUl )-H9S(2).
    60
    50
    40
   30
<
tL
   20
   10  -
DATA OF
SOBOCINISKI, BIERLEIN
•   293 'K

•   277.6 -K

-   MODEL
                0-2         0.4        0.6        0.8
                    1.0
                              23

-------
Table 18. Calculated and experimental equil
bon-dioxide-nitrogen at 253°K.
p P

X2
0.023
0.059
0.095
0.170
0.210
0.246
Average
exp
( a trn )
36.5
52.0
67.5
100.5
115.8
125.7
percent
calc
(atm)
33.7
51.1
63.5
86.5
101.2
117.2
deviation:
DP
(atm)
-2.8
-0.9
-4.0
-14.0
-14.6
-8.5
in P,
y
2exp
0.382
0.509
0.568
0.596
0.582
0.557
8.1 ; in y,
ibrium data for car-
V^
calc
0.411
0.555
0.609
0.668
0.692
0.713
14.

Dy
-0 029
0.046
0 041
0.072
0 1 10
0.156

Table 19. Calculated and experimental equilibrium data
bon-dioxide-nitrogen at 273°K.

ii
0.076
0.133
0.188
0.215
exp
(atm)
73.2
95.0
108.7
112.3
Pcalc
(atm)
77.9
100.3
122.8
135.5
OP
(atm)
4.7
5.3
14.1
23.2
y
exp
0.353
0.395
0.377
0.366

calc
0.400
0.444
0.471
0.487
for car-

D^
0.047
0.049
0.094
0.121
Average percent deviation:   in P,  n;  in y,  21
                               24

-------
 Figure 6.   P-x Diagram for C02(1)-N2(2).
140
120
1OO
80
 60
 4O
 20
        253 K
                      DATA OF
                     KAMINISHt

                     •   253-K

                     •   273-K

                     -   MODEL


—I	1	1	1	I
 0.2        0.4        O.6        0.8        1.0
                           25

-------
Table 20. Calculated and experimental
sulf ide-nitrogen at 300. 1°K

X2
0.0071
0.0153
0.0239
0.0317
0.0401
0.0495
0.0570
0.0663
0.0762
0.0844
0.0957
Average
exp
(atm)
34.7
50.7
69.1
86.1
101.9
119.0
135.7
152.9
170.7
187.2
204.1
calc
(atm)
33.3
49.6
66.7
82.1
98.8
117.3
131.9
149.9
168.7
183.9
204,4
percent deviation:
DP
(atm)
-1.4
-1.1
-2.4
-4.0
-3.1
-1.7
-3.8
-3.0
-2.0
-3.3
0.3
in P, 2.4;
equilibrium data for hydrogen
y2
e*£
0.332
0.488
0.576
0.628
0.653
0.674
0.684
0.691
0.695
0.694
0.693
in y, 3.6.

calc
0.352
0.514
0.598
0.642
0.672
0.692
0.703
0.711
0.716
0.719
0.720


Dy_
0.020
0.026
0.022
0.014
0.019
0.018
0.019
0.020
0.025
0.025
0.027

Table 21.


X2
0.0035
0.0102
0.0186
0.0241
0.0314
0.0377
Calculated and experimental
sulf ide-nitrogen at 277.7.
P P
exo calc DP
(atm) (atm) (atm)
17.1 19.7 2.6
33.7 35.7 2.0
52.1 55.9 3.8
67.4 69.0 1.6
85.1 86.2 1.1
102.1 101.0 -1.1
equi 1 ibrium

y y
exp
0.271
0.582
0.698
0.743
0.769
0.791
data

9
calc
0.383
0.622
0.726
0.760
0.788
0.804
for hydrogen


Dy
0.112
0.040
0.028
0.017
0.019
0.013
Average percent deviation:   in  P,  5.5;  in y, 9.8.
                                  26

-------
    Figure 7.  P-x  Diagram for  H9S(1)-M0(2).
   TOO
                                    277.7
    80
    60
0.
   40
   20
                         DATA  OF
                         BESSERER

                         •   3O0.1 K

                         •   277.7 'K

                         -  MODEL

-I	1	1	I	(
 0.01        0.02       0.03        0.04       0.05
                             27

-------
  It  is difficult to estimate the error involved as it was not possible
  to  test the binary predictions for all of the nitrogen containing sys-
  tems in the concentration and pressure range of interest.

      To test predictions of the equilibrium behavior of C02-HpS-N2- me-
  thanol  mixtures it was necessary to obtain x-y-P-T data experimentally.
 The apparatus and procedures used to do this are described in the fol-
  lowing  section.  Results of the experimental study are discussed in
 Section 5.
                              SECTION 4

                EXPERIMENTAL EQUIPMENT AND PROCEDURE


      The experimental  apparatus used in obtaining high  pressure  vapor
 liquid  equilibrium data is shown in Figure 8.   The stainless  steel
 565-ml  equilibrium cell  was equipped with baffles and agitated with  a
 magnetic stirrer.   Pressures were measured using  a 16 inch  Heise gauge
 graduated in  0.5  psi  increments up to 85 atm.   The gauge  had  a guaran-
 teed  accuracy of  0.1  percent of full  scale.   Its  calibration  was check-
 ed using an Ashcroft  type  1327  portable deadweight tester.  Tempera-
 tures were measured using  a copper-constantan  thermocouple  and a digi-
 tal thermocouple  meter  calibrated against known temperatures.  Refrig-
 eration  was provided  with  a Harris  industrial  freezer.  Control  of the
 bomb  temperature  to within 0.1°C was  accomplished using a fan, heater,
 and Thermistemp temperature controller.   All valves were  teflon  packed
 and rated for high pressure use.

      Methanol  used was  Fisher Spectranalyzed Qy  with a stated purity  of
 99.95 percent.  It was  checked  for  water  content  using a gas  chromato-
 graph and dried when  necessary  with 3A  molecular  sieves to  less  than
 0.05  percent  water.   Coleman grade  carbon  dioxide with a stated  purity
 of 99.99 percent  and  ultra pure carrier  grade  nitrogen with a stated
 purity of 99.999  percent supplied by  Air  Products  and Chemicals  were
 used.  A mixture of 1.5 percent hydrogen  sulfide  in nitrogen  supplied
 by Linde Division of  the Union  Carbide  Corporation was used.

     Analysis  of the samples was  done on  a Tracor  Model  550 gas  chro-
matograph equipped with a  thermal conductivity  cell, temperature pro-
grammer  and heated gas sampling  valve.  A  3 meter  long by 0.0032 meter
diameter  column packed with Porapak QS was used to achieve the separa-
tions.   Compositions were  determined  using Southern Analytical"s Supe-
grator 3 digital  integrator and  a Leeds and Northrop strip chart re-
corder.    For the methanol-carbon dioxide system,  the chromatograph was
calibrated by mixing the components at below atmospheric pressures in
the 10-5, vessel shown in Figure  9.  Concentrations were  determined from
pressure measurements.  For the multicomponent system,  calibrations were
done by  injection  of known amounts of each component.
                                 28

-------
                Figure  8.  Experimental  Equipment.
r\3
       4-way  valve
                          temperature
                          controller
digital
thermocouple
                                                                                 pressure  gauge
       CO,
N
                      Mix
                           heater
                                                                       capillary
                                                                       tubing
                                                                                            heated
                                                                                            sample
                                                                                            containers
                                                                                           freezer
                                                                                        equilibrium cell
                                                                                        magnetic stirrer

-------
Figure  9.  Calibration apparatus.
TO  G.C.
 SEPTUM
                                     FLANGE
                                      9.61 VESSEL
                                              MANOMETER
                                              GAS SOURCE
                                           VACUUM  PUMP
                                      MAGNETIC
                                      STIRRER
                     30

-------
     Before each series of runs, the equilibrium cell  was filled with
approximately 250-ml of methanol and then purged until no air or water
could be detected leaving the cell.   The cell  was then pressurized
and brought to the desired temperature.   The gases of interest were
added to achieve the desired pressure.   The contents were agitated for
at least six hours and then allowed  to sit unagitated for at least
twelve hours prior to sampling.   Samples were allowed to expand through
the capillary tubing into the evacuated  sample containers.  Sampling
was done quickly and the cell pressure was seldom disturbed by more
than 0.14 atm.  The vapor sample container was pressurized to approxi-
mately 1.4 atm with helium and then  both containers were heated to 140°C
for five hours.  The pressures in the containers were monitored to in-
sure that they did not approach  the  vapor pressure of methanol.  The
contents of each container were  analyzed a minimum of three times using
the gas chromatograph.
                             SECTION 5

                        EXPERIMENTAL RESULTS
Methanol-Carbon Dioxide
     For the purpose of verifying the proposed experimental procedure,
vapor liquid equilibrium data were obtained for the carbon, dioxide-metha-
nol system.  The data are compared with those of Yorizane et al.  (4) in
Figure 10 and to the model predictions in Table 22.  From Figure  10, it
is evident that the liquid phase concentration and pressure measurements
show good agreement with those of Yorizane et al.  (4).   Table 22  shows
the average percent deviation between calculated and experimental pres-
sures to be 8.8, which is identical  to that calculated  from Yorizane1s
data.  The vapor phase concentration of methanol is higher than expected
and shows significant scatter.  It is believed that liquid methanol be-
came entrained with the vapor leaving the equilibrium cell during sampl-
ing.  In these experiments pressures were measured to within 0.3  atm.,
temperatures within 0.1°C and compositions within  2 percent.  These bi-
nary carbon dioxide-methanol vapor-liquid equilibrium data demonstrate
the utility of the equipment and experimental procedure in obtaining and
analyzing liquid samples.  Using the present vapor sampling procedure
unusually high and unpredictable concentrations of methanol can be ex-
pected in the vapor phase.  Further experiments should  be conducted in
an apparatus modified to improve vapor sampling.  At temperatures less
than or equal to -15°C and at pressures greater than approximately 10
atm., the assumption that the vapor is methanol free introduces an er-
ror of less than 1 percent.   If this sampling technique does not appre-
ciably affect equilibrium, the equipment can be used to obtain data for
a multicomponent system to serve as a check of the thermodynamic model.
                                  31

-------
    Figure 10.   P-x Diagram Experimental CH_OH(1)-C02(2)  Data.
  20
   16
5
    8
T = 258-K




• DATA  OF  YORIZANE





• DATA-THIS INVESTIGATION




- MODEL
                 O.2        O.4        O.6         O.8
                    l.O
                                32

-------
Table 22.   Comparison of model  and experimental  equilibrium data  for
           carbon dioxide-methanol at 258°K.

          p         p
           exp       calc
  X2      (atm)     (atm)
0.085      5.03      5.76     0.73     0.936     0.997     0.061
0.149      8.45      9.24     0.79     0.995     0.998     0.003
0.227     12.13     12.84     0.71     0.980     0.999     0.019
0.349     16.30     17.21     0.91     0.912     0.999     0.087

Average percent deviation:  in P, 8.8;  in y, 4.6.
Multicomponent Equilibrium Data

     Experimental equilibrium data were obtained for C02~HoS-N2
nol mixtures at -15°C.   These data are presented in Table 23. ^Preci-
sion of the analytical  equipment varied among the system components:
mole fractions of the most concentrated component could be determined
to 1.5»; N2 and CC^ in the liquid could be determined to 2  and 81-, re-
spectively; methanol  and C02 in the vapor to within Zc: and 1«, respec-
tively.  Confidence in hydrogen sulfide compositions was good at high
pressures, but at low pressures a decrease in sensitivity of the ther-
mal conductivity detector caused ^S determinations to became less pre-
cise.

     Equilibrium temperature and liquid composition were used with the
algorithm BULP (described by  Bass  (30) to calculate system pressure
and gas composition at equilibrium.  Comparisons between calculated and
experimental results are given in Table 24.  The average deviation be-
tween calculated and experimentally measured pressures is 10.81".  Con-
siderable differences between experimental and calculated gas composi-
tions may also be noted.  In general, pressure predictions are more ac-
curate at low pressures while vapor composition predictions are more
accurate at high pressures.  The average percent deviation in pressure
is only slightly worse than the 6.6 parcent value calculated for all
binaries used in determining and checking the model parameters.  How-
ever, the percent deviation in vapor compositions is much worse.  While
not conclusive, it is believed that tnis unusually large discrepancy is
due to ternary effects among the components carbon dioxide, hydrogen
sulfide and methanol.  Calculations made ignoring ternary effects show
vapor compositions for carbon dioxide and hydrogen sulfide to be larger
than the experimental values.  Thus, gas cleanup systems will likely be
overdesigned if calculations are based on vapor-liquid equilibrium pre-
dictions made by considering only binary interactions.
                                33

-------
Table 23.  Experimental vapor-liquid equilibrium data for ir,ethano:-carbon dioxide-nitrogen-
          hydrogen sulfidc at -15.0°C.
P(atm)
8.23
9.37
9.75
19.9
27.9
34.6
*CH3OH
0.945
0.927
0.929
0.909
o.i:>-
0.920
Xco2
0.052
U . w u y
0.057
0.033
0.0/0
0.055
X
0.002
r\ A A t)
\J • W «^
0.003
0.007
0.010
0.013
XH2S
0.001
0.001
0.001
0.001
0.001
0.002
ycH3c:;
0.003
0.002
0.002
0.002
0
0
yco2
0.319
0.276
0.270
0.134
0.101
0.035
\
O.C77
0. 722
0.726
0.863
0.0 j'd
0.913
!!2S
0.001
-
O.UD1
0.001
0.001
0.002

-------
Table 24.  Comparison of  calculated  and experimental  pressures  and  gas  compositions  for C00  - H0S  -  N0  -
           methanol                                                                        *-     <•      i
      P(atm)
                 yCH3OH
                                               'CO,
exp
calc
                DP
exp   calc    Dy
exp   calc
exp   calc    Dy
                                                                                        exp   calc     Dy
 8.23   7.77   0.46

 9.37  10.77  -1.40

 9.75  10.66  -0.91

19.9   19.1    0.8

27.9   23.8    4.1

34.6   28.8    5.8
              0.003  0.002  0.001


              0.002  0.002  0


              0.002  0.002  0


              0.002  0.001  0.001


              0      0.001 -0.001


              0      0.001 -0.001
                                          0.319  0.493   -0.174


                                          0.276  0.466   -0.190


                                          0.270  0.459   -0.189

                                          0.134  0.327   -0.193


                                          0.101   0.234   -0.133


                                          0.085  0.188   -0.103
                                           0.677  0.502  0.174


                                           0.722  0.053  0.192


                                           0.726  0.537  0.189


                                           0.863  0.863  0.192


                                           0.898  0.898  0.134


                                           0.913  0.913  0.103
                                           0.001  0.004 -0.003


                                                  0.003   -

                                           0.001  0.003 -0.002


                                           0.001  0.002 -0.001


                                           0.001  0.001  0


                                           0.002  0.002  0

-------
                              SECTION 6

                          SUMMARY OF RESULTS
 1.   A thermodynamic model  was  developed  which  successfully  corre-
     lated  binary vapor-liquid  equilibrium  data for  the methanol-hy-
     drogen  sulfide, methanol-carbon  dioxide, carbon  dioxide-hydrogen
     sulfide and nitrogen  hydrogen  sulfide  systems.   The  lack of ade-
     quate data prevented  an  adequate correlation  for the methanol-ni-
     trogen and carbon dioxide-nitrogen systems.

 2.   Experimental vapor-liquid  equilibrium  data  were  obtained for the
    methanol-carbon dioxide  system at -15.0°C which  showed  excellent
    agreement in pressure  and  liquid phase composition measurements
    to literature data.  Vapor compositions were  unusually  high in
    methanol.

3.  Vapor phase composition  and total pressure  predictions made using
    the thermodynamic model  showed poor agreement with experimental
    results.  Predicted vapor compositions for carbon dioxide and hy-
    drogen sulfide were larger than the experimental values.  It is
    postulated that this is the result of  ternary effects among this
    system components.

4.  Models that do not require experimental data on ternary effects
    should be tested.

5.  Additional  modifications to the experimental equilibrium cell  are
    needed to improve the vapor sampling  technique.
                               36

-------
                         LITERATURE CITED
 1.  Ranke, Gerhard,  "The Rectisol Process for the Selective Removal
    of C02 and Sulfur Compounds  from  Industrial Gases," Chemical
    Economy and Engineering Review, 4, 25 (1972).

 2.  Morrison, James  A., "Technical Data Manual  for  EPA Gas Cleaning
    Facility Located at North Carolina State University," EPA  Con-
    tract Mo. 68-02-2601, 1977.

 3.  Fisch, E. J. and J. A. Sykes, "Synthetic Fuel Gas Purification
    Using Shell Treating Processes,"  ACS Meeting, Dallas, Texas,
    1973.

 4.  Yorizane, Masahiro, S. Sadamoto,  H. Masuoka, and Y. Eto, "Gas
    Solubilities in  Methanol at  High  Pressure," Kogyo Kagaku Zashi.
    72,  2174-2177  (1969).

 5.  Bezdel, L. S.  and V. P. Teodorovich, "The Solubilities of  Carbon
    Dioxide Hydrogen Sulfide, Methane, and  Ethylene in Methanol  at
    Low  Temperatures," Gazovaia  Promsnlenmost,  (Moscow),  8,  38-43
    (1958).

 6.  Hemmaplardh, B.  and A. D. King, Or., "Solubility of Methanol  in
    Compressed Nitrogen, Argon Methane, Ethylene, Ethane, Carbon
    Dioxide and Nitrous Oxide.   Evidence for Association  of  Carbon
    Dioxide with Methanol  in the Gas  Phase," The J. of  Phys.  Chem.,
    76,  2170-2175  (1972).

 7,  Katayama, Takashi, 0.  Kazunari, G. Maekawa, M.  Goto  and  T. Nagano,
    "Isothermal Vapor-Liquid Equilibria of  Acetone-Carbon Dioxide and
    Methanol  Carbon  Dioxide  at  High Pressures," J.  of  Chem.  Enq. of
    Japan. 8_, 89-92  (1975).

 8.  Krichevskii,  I.  R. and  E. S. Lebedeva,  "Solubilities  of  Nitrogen
    and  Carbon Dioxide  in  Methanol  Under Pressure," Zhurnal  Fizicheskoi
     Khitrni,  21_, 715  (1947).

 9.  Ohgaki,  Kazunari and Takashi Katayama,  "Isothermal  Vapor-Liquid
    Equilibrium Data for Binary Systems  Containing  Carbon Dioxide at
    High Pressures:   Methanol-Carbon  Dioxide,  n-Hexane-Carbon  Dioxide,
    and  Benzene-Carbon  Dioxide  Systems," J. of  Chem.  Eng. Data, 21,
    53-55 (1976).

10.   Bierlein,  James  A.  and Webster  B. Kay,  "Phase   Equilibrium Properties
    of System Carbon Dioxide-Hydrogen Sulfide," ]&_E_C_,  45_, 618-624 (1953).

11.   Sobocinski, D. P.  and  Fred  Kurata, "Heterogeneous  Phase Equilibria
     of the Hydrogen  Sulfide-Carbon  Dioxide System," AIChE J. 5, 545-551
     (1959).

12.   Kalra,  Harish, Thamra  R. Krishman, and Donald  B. Robinson, "Equili-
     brium-Phase  Properties of Carbon  Dioxide-n Butane and Nitrogen-Hydro-
     gen  Sulfide  Systems  at Subambient Temperature," J. of Chem. and Eng.
     Data, 21_, 222-225 (1976).


                                  37

-------
 13.   Besserer, George J., and Donald B.  Robinson, "Equilibrium Phase
      Properties of Nitroqen-Hydrogen Sulfide System," Chem.  and Enq
      Data, 20_, 157-161  (1975).

 14.   Arai, Yasuhiko,  Gen-Ichi Kamim'shi,  and Shozaburo Saito,  "The  Ex-
      perimental Determination of the p-v-T-X Relations for the Carbon
      Dioxide-Nitrogen and the Carbon Dioxide-Methane Systems," Chem
      Eng.  of Japan, 4_,  113-122  (1971).

 15.   Kamim'shi, Genichi  and  Tatsuro  Toriumi, "Gas-Liquid  Equilibrium
      Under Pressures,  VI  Vapor-Liquid  Phase  Equilibrium in the C02-H2,
      C02-N2  and C02-02  Systems,"  Kogyo Kagaku  Zasshi,  69^,  175-178  (1966).

 16.   O'Connell,  John  P.,  "Some  Asoects of Henry's Constants  and Unsym-
      metric  Convention  of Activity Coefficients," ACS  Symposium Serjie^
      60, Phase  Equilibria and Fluid  Properties  in the  Chemical  Industry,
      1977.

 17.   Wilson, G.  M., "Vapor-Liquid Equilibrium.   XI.  A New Expression
      for the Excess Free  Energy of Mixing,"  J.  Am. Chem. Soc.,  86,  127
      (1964).

 18.   Abrams, D.  C. and J.  M.  Prausnitz, "Statistical Thermodynamics  of
      Liquid Mixtures:  A  New  Expression for  the Gibbs  Energy of Partly
      or Completely Miscible Systems," AIChE  J.,  21_,  116  (1975).

 19.   Adler, Stanley B., Calvin  E. Spencer, Hal  Ozkardesh,  and  Chia-Ming
      Kuo,  "Industrial Uses of Equations of State:  A State-of-the Art
      Review," ACS Symposium Series 60, Phase Equilibria and  Fluid Pro-
      perties in  the Chemical  Industry, 1977.                       ~~

 20.   Chao, K. C. and J. D. Seader, "A General Correlation  of Vapor-Liquid
      Equilibria  in Hydrocarbon  Mixtures," AIChE  J. 7,  598-605  (1961).

 21.   Adler, S.  B., Z. Friend, and R. L. Pigford,  "Application  of the Wohl
      Equation to Ternary  Liquid-Vapor Equilibria," AIChE J.  12, 629-637
      (1966).

 22.   Adler, S.  B., H.  Ozkardesh, and VI. C. Schreiner,  "These Equations
      Predict Equilibria," Hydrocarbon Processing, 47,  145-153  (1968).

 23.   Prausnitz, J.  M.  and P.  L. Chueh, Computer  Calculations for High
      Pressure Vapor-Liquid Equilibria, Prentice  Hall,  New  Jersey, 1968.

24.   Prausnitz, J.  M., Molecular Thermodynamics  of Fluid-Phase  Equilibria.
      Prentice Hall, New Jersey, 1969.~

25.  Tarakad, Ramanathan, and R. ?.  Danner,  "An  Improved Corresponding
     States Method for Polar  Fluids:   Correlation of Second Virial Coef-
     ficients," AIChE J. 23_,  685-595 (1977).

26.  Reid,  Robert C.,  J. M. Prausnitz and T.  K.  Sherwood, The Properties
     of Gases and Liquids, third edition, McGraw  Hill  Book Company, New
     York,  1977.

-------
27.  Lubank  P  T., "A Review of Volumetric, Thermodynamic, and other
     Physical Properties for Methanol," CEP Symposium Series-Methanol
     Technology and Economics. 98., 16-23 (1970).	

23.  Robinson, R  L., Jr. and Kwang-Chu Chao, "A Correlation of Vapori-
     zation Equilibrium Ratios for Gas Processing Systems," I&EC Pro-
     cess Design and Development. l_p_, 221-229 (1971).       	

29.  Erbar, John H. and Wayne C.  Edmister, "New Constants for Chao-Seader
     Correlation for N2,  H2S, and C02," Oklahoma State University, Still-
     water, Oklahoma, personal communication (1978).

30.  Bass, D.  G.,  "The Solubility-of-Acid Gases  in Methanol," M. S
     Thesis,  North Carolina State University at  Raleigh, 1978.
                                39

-------
                            NOMENCLATURE

DP         -         Pcalc ' Pexp
DV         =         ycalc " yexp
f.         =         fugacity of i
f o        =         fugacity of i  in reference state

H          =         Henry's constant for i
 i
k          =         binary constant, Equation 9f
 ij
n          =         moles of i
 i
p          =         pressure
p          =         reduced pressure
 r
p          =         critical pressure of i
 ci
p *        =         vapor  pressure of i

R          =         gas  constant
7          =         temperature
T          =         reduced temperature
  r
v          =         molar  volume
V          =         critical  volume

V  L        =         liquid molar  volume  of  i
  i
v          =         mole fraction i  in  gas  phase
y\

 Greek Symbols
                     activity  coefficient of i
                     fugacity  coefficient of i
                      fugacity  coefficient of saturated i
                      Pitzer accentric factor
                      dimensionless pure component constants, Table 6
                                 40
Yi

-------
                                TECHNICAL REPORT DATA
                          (Please read Inuructions on the reverse before completing}
 1 REPORT NO.
  EPA-600/7-79-097
                                                      3. RECIPIENT'S ACCESSION NO.
4. TITLE ANDSUBTITLE
 The Solubility of Acid Gases in Methanol
                                                     5. REPORT DATE
                                                      April 1979
                                                      6. PERFORMING ORGANIZATION CODE
7~~AUTHOR(S)
j.K.Ferrell, R.W.Rousseau, and D.G.Bass
                                                     8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
 North Carolina State University
 Department of Chemical Engineering
 Raleigh, North Carolina 27650
                                                      10. PROGRAM ELEMENT NO.
                                                      E HE 62 3 A
                                                      11. CONTRACT/GRANT NO.

                                                      Grant No. R804811
•12. SPONSORING AGENCY NAME AND ADDRESS
 EPA, Office of Research and Development
 Industrial Environmental Research Laboratory
 Research Triangle Park, NC 27711
                                                      13. TYPE OF RfcEQRT AND PEJ
                                                      Final: 10/76 - 9/78
                                                                        RIOD COVERED
                                                      14. SPONSORING AGENCY CODE
                                                       EPA/600/13
15. SUPPLEMENTARY NOTES jERL_RTppro.ectofficer
2708.
                                                        Smith ,  MD- 61 , 919/541-
16 ASSTRACT The report describes a thermodynamic model developed to predict phase-
equilibrium behavior in a methanol/carbon-dioxide/nitrogen/hydrogen-sulfide sys-
tem based on parameters determined from binary vapor/liquid equilibrium data
available in the literature. Model predictions are compared with actual experimen-
tal data. Predicted values show an average deviation from experimental data of 21%
for vapor compositions and 10% for total pressures, indicating the possibility of
ternary effects not accounted for by the model.  (The model is to be used in conjunc-
tion with a coal-gasification/gas-cleaning facility constructed at North Carolina
State University as part of a study funded by the EPA. The overall objective  of the
project is to characterize completely the gaseous and condensed phase emissions
from typical coal-gasification/gas-cleaning processes and to determine how  emis-
sions depend on process parameters. To describe and evaluate operations involved
in the removal of acid  gas constituents from the crude synthesis gas, information is
needed concerning the  equilibrium behavior of these constituents with the  particular
solvent used in the removal unit; hence, the model.)
 7.
                             KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                          b,IDENTIFIERS/OPEN ENDED TERMS
                                                                    COSATi Field/Gioup
 Pollution
 Carbinols
iSour Gas
 Solubility
1 Mathematical Models
 Thermodynamics
                      Coal Gasification
                      Gas Purification
Pollution Control
Stationary Sources
Methanol
Acid Gases
13B
07C
2 ID
07D
12A
20M
07A,13H
"3 DlSTFt UUTION STA TEMENT
 Unlimited
                                          19. SECURITY CLASS (This Report/
                                          Unclassified
                        21. NO. OF PAGES
                             46
                                          20. SECURITY CLASS (This page/
                                          Unclassified
                                                                  22. PRICE
EPA Form 2220-1 (9-73)

-------