logKcalc vignette 1 2 3 4 5

Choosing the water model (DEW)

This vignette shows the conversion of a GWB data file to use the Deep Earth Water (DEW) model.

This vignette was compiled on 2023-09-15 with logKcalc 0.1.1-2 and CHNOSZ 2.0.0-25.

The Deep Earth Water (DEW) model includes correlations for H2O and revised thermodynamic parameters of aqueous species for estimating thermodynamic properties to pressures of up to 6 GPa (Sverjensky et al., 2014a). This model was used to predict a significant stability region for organic ions, such as acetate and propionate, at high pressures corresponding to upper mantle conditions (Sverjensky et al., 2014b).

Let us use the DEW model in CHNOSZ in order to calculate equilibrium constants for dissociation reactions in a GWB thermodynamic data file. Separate commands are needed to change the water model and load the parameters of the revised Helgeson-Kirkham-Flowers (HKF) equations for aqueous species that are taken from the DEW spreadsheet.

reset()
water("DEW")
if(packageVersion("CHNOSZ") >= "1.4.0") add.OBIGT("DEW") else add.obigt("DEW")
# modOBIGT(c("addSUPCRT", "steam"))

Note that other changes to the database to add the properties of steam and minerals from SUPCRT92 (Johnson et al., 1992) are not needed for this calculation.

The thermo_12elements.tdat data file includes aqueous CH4, CO2, HCO3-, CO3‑‑, CH3COO- (acetate) and HCH3COO (acetic acid), which will be updated during the conversion. We can use the ispecies argument to add other species from OBIGT, in particular formic and propanoic acid and their anions, which were predicted to be major species by Sverjensky et al. (2014b). For completeness, we include other aqueous organic species, although they are not expected to appear in the diagrams. However, we do not include glycinate (added in the DEW 2019 dataset) to maintain backwards compatibility with CHNOSZ versions before 1.4.0.

major <- c("formic acid", "formate", "propanoic acid", "propanoate")
o1 <- c("benzene", "diglycine", "diketopiperazine")
o2 <- c("ethane", "ethanol", "ethylene")
o3 <- c("glutamate", "glutamic acid", "H-glutarate", "glutaric acid")
#o4 <- c("glutamine", "glycinate", "glycine", "glycolate", "glycolic acid")
o4 <- c("glutamine", "glycine", "glycolate", "glycolic acid")
o5 <- c("H-succinate", "hexane", "isobutane")
o6 <- c("lactate", "lactic acid", "leucine", "methanol")
o7 <- c("propane", "propanol", "toluene", "urea")
ispecies <- info(c(major, o1, o2, o3, o4, o5, o6, o7))

In the GWB thermodynamic data files, pressure is not an independent variable, but is expanded in terms of the principal temperatures (Bethke and Farrell, 2020). In order to set up calculations to compare different pressures, we choose a series of closely spaced temperatures that approximate a constant temperature of 600 °C. The pressures are set to 0.5, 1, 1.5, 2, 2.5, 3, 4, 5 GPa, converted to units of bar.

T <- 593:600
P <- c(5000, 10000, 15000, 20000, 25000, 30000, 40000, 50000)

Now we perform the conversion. To calculate the extended term parameter of the Debye-Hückel equation at high pressure (see Manning et al., 2013; Dick, 2019), we use the bgamma method. This does not affect the activity diagrams shown below, but is important for speciation calculations that may be possible with GWB.

infile <- system.file("extdata/thermo_12elements.tdat", package = "logKcalc")
outfile <- file.path(tempdir(), "logKcalc_vig4.tdat")
logKcalc(infile, outfile, T = T, P = P, ispecies = ispecies, DH.method = "bgamma", maxprint = 10)

Scripts and output of GWB runs.

This script for the Act2 program plots the speciation of carbon as a function of oxygen fugacity (fO2) and pH. The activity of HCO3- is set to 0.03, which is close to the concentration of carbon in the fluid calculated for a model eclogite at 600 °C and 5 GPa (Supplementary Information of Sverjensky et al., 2014b). So as not to exceed the maximum pressure accepted by the Act2 program, the script sets a pressure of 5000 bar. This pressure affects the water limits and stability fields of gases but not the log K values for reactions between aqueous species, which were calculated for higher pressures as described above. To make these diagrams, we turn off the water limits and gas and mineral fields in Act2.

The first diagram shows a small stability field for formate at 0.5 GPa (5 kbar). This field is present both for the DEW model and for the water model from SUPCRT92 (Johnson et al., 1992) with the default OBIGT database in CHNOSZ, which has parameters of the organic acids from Shock (1995). (The GWB data file was created as shown above except the water and add.OBIGT commands were omitted and all pressures were set to 5000 bar, the maximum for the revised HKF equations without the DEW model.)

The next pair of diagrams is for a pressure of 5 GPa (50 kbar) using the DEW model. In the first, propanoate and formate are predicted to predominate at some redox and pH conditions. In the second, the formation of both of these species is suppressed; the diagram shows that no other organic species can predominate at equilibrium.

The presence of predominance fields for propanoate and formate is consistent with the model shown in Figure 3 of Sverjensky et al. (2014b), where these species are the most abundant C-bearing species in the fluid at 600 °C. It is unclear why these species are not present in the logfO2–pH diagram in Figure 1 of that paper, which instead shows a large predominance field for acetate. However, it may not be surprising that we have trouble reproducing some results of Sverjensky et al. (2014b); according to a comment in the DEW spreadsheet the acetate parameters were revised after that publication, on January 26th, 2016.

Comparison of equilibrium constants

To check the accuracy of the data in OBIGT and equations in CHNOSZ, let’s compare the calculations with the DEW spreadsheet. The dissociation reactions of the species in the GWB file are written in terms of the basis species HCO3-, H2O, O2(aq), and H+. We can use logKcomp to retrieve the equilibrium constants calculated for the reactions at 600 °C and 5 GPa, which is the 8th T,P pair. For comparison, the 2019 version of the DEW spreadsheet was used to calculate log K for the same reactions and conditions. HCO3- is not included; because it is a basis species, its dissociation reaction has a log K of zero by definition.

species <- c("CH4", "CO2", "CO3--", "formate", "CH3COO-", "propanoate")
# logK values calculated with CHNOSZ
calc <- logKcomp(outfile, iTP = 8)
cvals <- calc[species]
# logK values calculated with the DEW spreadsheet
dvals <- c(48.4582, -3.7302, 5.3468, 12.5707, 48.6838, 77.5725)
logKs <- t(data.frame(dvals, cvals, dvals - cvals))
colnames(logKs) <- paste0("`", species, "`")
rownames(logKs) <- c("spreadsheet", "CHNOSZ", "difference")
knitr::kable(logKs)
CH4 CO2 CO3-- formate CH3COO- propanoate
spreadsheet 48.4582 -3.7302 5.3468 12.5707 48.6838 77.5725
CHNOSZ 48.4259 -3.7277 5.3432 12.5623 48.6513 77.5206
difference 0.0323 -0.0025 0.0036 0.0084 0.0325 0.0519
ratios <- t(data.frame(dvals / cvals))
colnames(ratios) <- paste0("`", species, "`")
rownames(ratios) <- "ratio"
knitr::kable(ratios)
CH4 CO2 CO3-- formate CH3COO- propanoate
ratio 1.000667 1.000671 1.000674 1.000669 1.000668 1.00067

The differences between the CHNOSZ calculation and DEW spreadsheet are roughly proportional to the magnitude of log K. The ratios are close to 1.000666, the ratio between the values of the gas constant (R) used in CHNOSZ (roughly 1.9872 cal mol-1 K-1, converted from 8.314463 J mol-1 K-1) and in the DEW spreadsheet (1.9858775 cal mol-1 K-1). These small differences have a negligible effect on the diagrams shown above.

References

Bethke CM, Farrell B. 2020. The Geochemist’s Workbench® Release 14 GWB Reference Manual. Champaign, Illinois: Aqueous Solutions, LLC.

Dick JM. 2019. CHNOSZ: Thermodynamic calculations and diagrams for geochemistry. Frontiers in Earth Science 7: 180. doi: 10.3389/feart.2019.00180

Johnson JW, Oelkers EH, Helgeson HC. 1992. SUPCRT92: A software package for calculating the standard molal thermodynamic properties of minerals, gases, aqueous species, and reactions from 1 to 5000 bar and 0 to 1000°C. Computers & Geosciences 18(7): 899–947. doi: 10.1016/0098-3004(92)90029-Q

Manning CE, Shock EL, Sverjensky DA. 2013. The chemistry of carbon in aqueous fluids at crustal and upper-mantle conditions: Experimental and theoretical constraints. Reviews in Mineralogy and Geochemistry 75(1): 109–148. doi: 10.2138/rmg.2013.75.5

Shock EL. 1995. Organic acids in hydrothermal solutions: Standard molal thermodynamic properties of carboxylic acids and estimates of dissociation constants at high temperatures and pressures. American Journal of Science 295(5): 496–580. doi: 10.2475/ajs.295.5.496

Sverjensky DA, Harrison B, Azzolini D. 2014a. Water in the deep Earth: The dielectric constant and the solubilities of quartz and corundum to 60 kb and 1,200 °C. Geochimica et Cosmochimica Acta 129: 125–145. doi: 10.1016/j.gca.2013.12.019

Sverjensky DA, Stagno V, Huang F. 2014b. Important role for organic carbon in subduction-zone fluids in the deep carbon cycle. Nature Geoscience 7(12): 909–913. doi: 10.1038/ngeo2291