H2S solubility

[ehsan]

hey dears
i have problem with h2s solubility in pure water
and I ask you to help me please
my phreeqc is attachment
thanks

Code: [Select]

SOLUTION 1
PHASES 1
REACTION 1
    H2S(g)     1;0 23*1
GAS_PHASE 1
    -fixed_volume
volume 5
    H2S(g)    0
h2o(g) 0
REACTION_TEMPERATURE 1
    40
INCREMENTAL_REACTIONS True
GAS_PHASE_MIX 1
1 0.2
USER_GRAPH 1
-headings 40C
-plot_tsv_file h2s2.tsv
 -axis_titles "Pressure, in atmospheres" "H2S concentration"   
 -axis_scale x_axis 0 500
 -axis_scale y_axis 0 4
 -connect_simulations false
 10 graph_y TOT("S(-2)")
 20 graph_x pressure
end
USE gas_phase 1
USE reaction 1
USE solution 1

REACTION_TEMPERATURE 2
    60
USER_GRAPH 1
-headings 60C
end
USE gas_phase 1
USE reaction 1
USE solution 1
REACTION_TEMPERATURE 3
    90
USER_GRAPH 1
-headings 90C
end


[dlparkhurst]

I don't see anything wrong with your calculation. The extent of the graph will change when you use GAS_PHASE_MIX to make total gas volume 1 L as opposed to 5 L, but the basic curve will be the same.

There are not that many parameters involved in the calculation. I think the Peng-Robinson parameters should be relatively well known--critical temperature, critical pressure, and acentric factor.  You can check in the literature. I think there may be a reference for H2S in
"Equations for calculating hydrogeochemical reactions of minerals and gases such as CO2 at high pressures and temperatures",CAJ Appelo, DL Parkhurst, VEA Post - Geochimica et Cosmochimica Acta, 2014

H2S(aq) is by far the dominant species, and being uncharged, there is no activity coefficient effect available unless the medium has other ions contributing to the ionic strength.

That leaves only the Henry's law equilibrium constant as a function of temperature. Using the analytical expression compared to the Van't Hoff seems to bracket the solubility. sit.dat has a different delta_H for H2S(g) (the analytical expression is equivalent to Van't Hoff).

So, you are on your own to decide which parameters could be adjusted, or how your data compares to other experiments.

[ehsan]

excuse me dr
can you send to me a few phreeqc file of H2S solubility please?
because i'm not have a reference for this

[dlparkhurst]

You will need to search the literature. One reference in CAJ Appelo, DL Parkhurst, VEA Post - Geochimica et Cosmochimica Acta, 2014 is the following:

Peng-Robinson predictions for hydrocarbons, CO2, N2, and H2 S with pure water and NaCI brine
I Søreide, CH Whitson - Fluid Phase Equilibria, 1992 - Elsevier

[ehsan]

thanks dr
i have three question more
one : How does phreeqc analyze analytic parameter in PHASES when we commented it with a #?  in code
two : Why do break occur in the solubility diagram when dissolved? in attachment pic
three : Where can I find Pitzer parameters(a0,...,a5) for H2S?

Code: [Select]

# DATABASE c:\phreeqc\database\llnl.dat # has S5- etc species
# Compare expt data of H2S solubility at high H2S pressure and
#   Peng-Robinson calc's with initial solution equilibria.
PHASES
H2S(g)
H2S  =  H+ + HS-
log_k    -7.97
-delta_h   9.1
# -analytic -9.7354e+001 -3.1576e-002 1.8285e+003 3.7440e+001 2.8560e+001
-T_c  373.2
-P_c  88.2
-Omega 0.1

SOLUTION 1
 pH 7 charge
 S(-2) 1# H2S(g) 0
SELECTED_OUTPUT; -reset false; -file H2S_71.prn
USER_PUNCH
  1 sc$ = chr$(59)
 10 for i = 1.1 to 220 step 5
 20   punch 'SOLUTION 1' + sc$ + ' -temp 71' + sc$ + ' pH 3 charge' + EOL$ + 'S(-2) 1 H2S(g) ' + str$(log10(i)) + EOL$ + '-pressure ' + str$(i) + EOL$ + 'END' + EOL$
 30 next i
END

SELECTED_OUTPUT 1; -active false
PRINT; -reset false; -echo_input true
USER_GRAPH
 -plot_csv_file H2S_71C.dat
 -axis_titles "Pressure / atm" "H2S / (mol/kgw)" " phi_H2S"
 -axis_scale x_axis 0 210
 -initial_solutions true
 10 x = PR_P("H2S(g)")
 20 graph_x x # tot("Na")
 30 graph_y tot("S(-2)")
 40 graph_sy PR_phi("H2S(g)") #mol("H2S") / tot("S")
 -end
INCLUDE$ H2S_71.prn
END

[dlparkhurst]

one: https://phreeqcusers.org/index.php/topic,1787.msg6087.html#msg6087
two: It is related to the Peng-Robinson parameters.
three: I have no information

[ehsan]

thanks Dr for your guidance.
How should i delete the break in this code?

[dlparkhurst]

I have not dug deeply into the Peng-Robinson theory, but I think you should. I think the curves are probably representing real phenomena, but that is for you to find out.

[dlparkhurst]

Did you look at the reference Søreide and Whitson, figure 12?

[ehsan]

yes that is h2s in vapor phase.
and in low temperature have a break.

[ehsan]

hi dears
my code works for H2S in pure water but not worked in brine, how can i fixed and it worked for brine?

Code: [Select]

# DATABASE c:\phreeqc\database\llnl.dat # has S5- etc species
# Compare expt data of H2S solubility at high H2S pressure and
#   Peng-Robinson calc's with initial solution equilibria.
PHASES
H2S(g)
H2S  =  H+ + HS-
log_k    -7.97
-delta_h   9.1
# -analytic -9.7354e+001 -3.1576e-002 1.8285e+003 3.7440e+001 2.8560e+001
-T_c  373.2
-P_c  88.2
-Omega 0.1

SOLUTION 1
 pH 7 charge
 S(-2) 1 H2S(g) 0
SELECTED_OUTPUT; -reset false; -file H2S_71.prn
USER_PUNCH
  1 sc$ = chr$(59)
 10 for i = 1.1 to 220 step 5
 20   punch 'SOLUTION 1' + sc$ + ' -temp 71' + sc$ + ' pH 3 charge' + EOL$ + 'S(-2) 1 H2S(g) ' + str$(log10(i)) + EOL$ + '-pressure ' + str$(i) + EOL$ + 'END' + EOL$
 30 next i
END

SELECTED_OUTPUT 1; -active false
PRINT; -reset false; -echo_input true
USER_GRAPH

 -axis_titles "Pressure / atm" "H2S / (mol/kgw)" " phi_H2S"
 -axis_scale x_axis 0 210
 -initial_solutions true
 10 x = PR_P("H2S(g)")
 20 graph_x x # tot("Na")
 30 graph_y tot("S(-2)")
 40 graph_sy PR_phi("H2S(g)") #mol("H2S") / tot("S")
 -end
 INCLUDE$ H2S_71.prn
END


[dlparkhurst]

You are not giving me much to work with here. How does it "not work" for brines? The file does not run? The results do not match the experimental data (which you do not provide)?

Your file is for pure water, which runs.

You may want to contact Tony Appelo. I think he has been working on H2S solubility.

[ehsan]

when i add the ion salt in the solution keyword and run phreeqc, The result obtained for h2s solubility with pure water, no for h2s solubility with brine.

Code: [Select]

PHASES
H2S(g)
H2S  =  H+ + HS-
log_k    -7.97
-delta_h   9.1
# -analytic -9.7354e+001 -3.1576e-002 1.8285e+003 3.7440e+001 2.8560e+001
-T_c  373.2
-P_c  88.2
-Omega 0.1
SOLUTION 1
 pH 8.1 charge
    Na        0.46907
    Mg        0.05282
    Ca        0.01028
    K         0.01021
    Sr        9e-005
    Cl        0.54588
    S(6)      0.02824
    Br        0.00084
    C(4)      0.00027
    F         6.8e-005
 S(-2) 1 H2S(g) 0
SELECTED_OUTPUT; -reset false; -file H2S_90.prn
USER_PUNCH
  1 sc$ = chr$(59)
 10 for i = 1.1 to 220 step 5
 20   punch 'SOLUTION 1' + sc$ + ' -temp 90' + sc$ + ' pH 3 charge' + EOL$ + 'S(-2) 1 H2S(g) ' + str$(log10(i)) + EOL$ + '-pressure ' + str$(i) + EOL$ + 'END' + EOL$
 30 next i
END

SELECTED_OUTPUT 1; -active false
PRINT; -reset false; -echo_input true
USER_GRAPH
 -plot_csv_file H2S_90C.tsv
 -axis_titles "Pressure / atm" "H2S / (mol/kgw)" " phi_H2S"
 -axis_scale x_axis 0 210
 -initial_solutions true
 10 x = PR_P("H2S(g)")
 20 graph_x x # tot("Na")
 30 graph_y tot("S(-2)")
 40 graph_sy PR_phi("H2S(g)") #mol("H2S") / tot("S")
 -end
 INCLUDE$ H2S_90.prn
END

[dlparkhurst]

Your SELECTED_OUTPUT/USER_PUNCH script is defining 'SOLUTION 1;' for each simulation, which redefines solution 1 as pure water; the other solutes are not defined.

If your solution is a brine, you probably meant for the concentrations to be "-units mol/kgw". By default, units are mmol/kgw.

You do not have many adjustable parameters. With llnl.dat, the -llnl_gamma parameter does not affect the uncharged aqueous species H2S(aq). You can define -CO2_llnl_gamma, which will add an ionic strength dependence (bdot), however, the value is the same for all uncharged species for which it is defined, otherwise it is zero.

With phreeqc.dat and other ion-association models, you can adjust the bdot parameter for H2S(aq) independently, which would allow some adjustment of solubility with ionic strength. Default is 0.1.

With pitzer.dat, there are specific interaction parameters for uncharged species, which would allow variation of solubility dependent on the ionic medium.

I suggest you print the output file and look at it carefully.

Code: [Select]

PHASES
H2S(g)
H2S  =  H+ + HS-
log_k    -7.97
-delta_h   9.1
# -analytic -9.7354e+001 -3.1576e-002 1.8285e+003 3.7440e+001 2.8560e+001
-T_c  373.2
-P_c  88.2
-Omega 0.1
SOLUTION_SPECIES
HS- + H+ = H2S
-log_k 6.994
      -gamma 0 0.1
-delta_h -5.30 kcal
-analytical  -11.17  0.02386  3279.0
-dw 2.1e-9
-Vm  7.81  2.96  -0.46 # supcrt
END
USER_GRAPH
 #-plot_csv_file H2S_90C.tsv
 -axis_titles "Pressure / atm" "H2S / (mol/kgw)" " phi_H2S"
 -axis_scale x_axis 0 210
 -initial_solutions true
 10 x = PR_P("H2S(g)")
 20 graph_x x # tot("Na")
 30 graph_y tot("S(-2)")
 40 graph_sy PR_phi("H2S(g)") #mol("H2S") / tot("S")
 -end
END
SOLUTION 1
    units     mol/kgw
    temp      90
    pH        8.1 charge
    Na        0.46907
    Mg        0.05282
    Ca        0.01028
    K         0.01021
    Sr        9e-005
    Cl        0.54588
    S(6)      0.02824
    Br        0.00084
    C(4)      0.00027
    F         6.8e-005
    S(-2)     1        H2S(g) 0
SELECTED_OUTPUT 1; -reset false; -file H2S_90.prn
USER_PUNCH 1
10 FOR i = 1.1 TO 220 STEP 5
20 s$ = s$ + 'USE solution 1' + EOL$
30 s$ = s$ + 'EQUILIBRIUM_PHASES 1' + EOL$
40 s$ = s$ + 'H2S(g) ' + STR$(LOG10(i)) + ' 10' +  EOL$
50 s$ = s$ + 'END' + EOL$
60 NEXT i
70 PUNCH s$
END
SELECTED_OUTPUT 1; -active false
INCLUDE$ H2S_90.prn


[ehsan]

thank you Dr. for your guidance