Implementation Pitfalls for Carbonate Mineral Dissolution – a Technical Note
Fiona J. Weiss, Leon Keim, Kai Wendel, Holger Class · 2025
Read Implementation Pitfalls for Carbonate Mineral Dissolution – a Technical Note by Fiona J. Weiss, Leon Keim, Kai Wendel, Holger Class (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-05225.txt.
Introduction
Weiss, Keim, Wendel, and Class flagged a database error in this technical note from twenty twenty-five. Modelers of underground water and rock rely on rate tables for how minerals dissolve. One widely used table lists carbonate dissolution rates from a classic source. The note shows one parameter was reported against the wrong quantity.
The reaction order was tied to carbon dioxide pressure instead of the acid form actually meant. That slip stretched predicted reaction times by a wide margin. The team reran a calcite dissolving test in two simulators. With the parameter fixed, the simulations matched lab data about ten times better. The note urges routine checks of simulations against experiments.
Catalog record
| CREATOR | Fiona J. Weiss, Leon Keim, Kai Wendel, Holger Class |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 14 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-05225
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Implementation Pitfalls for Carbonate Mineral Dissolution – a Technical Note Fiona J. Weiss, Leon Keim, Kai Wendel, Holger Class Abstract In systems with slow reaction kinetics, such as mineral dissolution processes, chemical equilibrium cannot be assumed and an accurate understanding of reaction rates is essential; discrepancies in parameter reporting can greatly affect simulation results. This technical note identifies an issue with the reporting of rate parameters for carbonate mineral dissolution in a widely used database for reactive transport modeling based on Palandri and Kharaka 2004 . This misrepresentation leads to a considerable overestimation of reaction timescales. Using the simulators Reaktoro and DuMuX, we simulated a simple calcite dissolution batch test and compared the results to experimental data. By adjusting the parameter to align with established literature, we demonstrate an improved fit between simulated and experimental data. Discrepancies in reaction timescales were reduced by an order of magnitude, emphasizing the importance of regular validation of simulations with experimental data. [orcid=0000-0002-0852-1500] [orcid=0000-0003-2456-1809] [orcid=0000-0003-1871-1540] [orcid=0000-0002-4476-8017] keywords Geochemical modeling ,Reactive transport modeling ,Reaction kinetics ,Reaction rate parameters ,Carbonate mineral dissolution † † credit: Simulation, Visualization, Writing (original draft, review and editing) † † credit: credit Leon † † credit: credit Kai † † credit: credit Holger † † corresponding: Corresponding author 1 Introduction Chemical equilibrium often cannot be assumed in systems with slow reaction kinetics such as mineral dissolution processes, thus understanding the rates of mineral dissolution and precipitation is essential. Particularly for processes like CO 2 sequestration in geologic formations, where injected CO 2 disturbs the system far from equilibrium, the slow kinetics of mineral-CO 2 -water interactions dictate the re-equilibration timescales and the rates at which primary minerals dissolve and secondary minerals precipitate. A commonly used reaction rate model, along with the necessary parameters for minerals frequently studied, can be found in a USGS report by Palandri and Kharaka 2004 . This report serves as the foundation for a widely used database for computational programs in the reactive transport modeling community. The parameters for the Palandri and Kharaka 2004 model were derived from experimental studies reported in older literature. However, we identified an issue with the reporting of rate parameters for carbonate minerals. Specifically, the authors reported the reaction order n of the carbonate mechanism with respect to P({}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}) , when it should have been reported with respect to {}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast} , to be consistent with the literature used to derive these parameters [ 1 , 2 , 9 , 10 ] . This oversight leads to a considerable overestimation of reaction timescales. To address this issue, we propose an adjustment of the relevant section of the Palandri and Kharaka 2004 report. 2 Methods For carbonate minerals, three elementary mechanisms are involved in the dissolution reaction [ 2 , 9 ] ; namely the acid mechanism, the carbonate mechanism, and the neutral mechanism with forward reaction rate parameters k_{1} , k_{2} , and k_{3} and backward reaction rate parameters k_{-1} , k_{-2} , and k_{-3} , respectively. \begin{split}&{}\text{MCO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}\penalty\hskip 2.15277pt+\hskip 2.15277pt{}\text{H}{\vphantom{\text{X}}}^{\text{+}}{}\mathrel{\hbox to0.0pt{\raisebox{0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 9.0mu{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{1}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu\hphantom{{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{\hskip 0.45209pt--\hskip 0.45209pt1}}}}\mkern 5.0mu$}}}}$}\hss}\raisebox{-0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 5.0mu\hphantom{{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{1}}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{\hskip 0.45209pt--\hskip 0.45209pt1}}}\mkern 9.0mu$}}}}$}}{}{}\text{M}{\vphantom{\text{X}}}^{\text{2+}}\penalty\hskip 2.15277pt+\hskip 2.15277pt{}\text{HCO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}{\vphantom{\text{X}}}^{\text{\hskip 0.90417pt--\hskip 0.90417pt}}\\ &{}\text{MCO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}\penalty\hskip 2.15277pt+\hskip 2.15277pt{}{}\text{H}{\vphantom{\text{X}}}_{\smash[t]{\text{2}}}\text{CO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}^{\ast}{}\mathrel{\hbox to0.0pt{\raisebox{0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 9.0mu{}\mathrm{k}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu\hphantom{{}\mathrm{k}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{-2}}}}\mkern 5.0mu$}}}}$}\hss}\raisebox{-0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 5.0mu\hphantom{{}\mathrm{k}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu{}\mathrm{k}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{-2}}}\mkern 9.0mu$}}}}$}}{}{}\mathrm{M}{\vphantom{\mathrm{X}}}^{\mathrm{2+}}{}+{}{}{}\mathrm{2}\,\mathrm{HCO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}{\vphantom{\mathrm{X}}}^{\mathrm{-}}\\ &{}\text{MCO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}{}\mathrel{\hbox to0.0pt{\raisebox{0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 9.0mu{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu\hphantom{{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{\hskip 0.45209pt--\hskip 0.45209pt3}}}}\mkern 5.0mu$}}}}$}\hss}\raisebox{-0.94722pt}{$\mathrel{\mathop{\makebox[0.0pt]{\to}}\limits^{\mkern 5.0mu\hphantom{{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}}\mkern 9.0mu}_{\makebox{\raisebox{3.76735pt}[0.0pt]{$\scriptstyle\mkern 9.0mu{}\text{k}{\vphantom{\text{X}}}_{\smash[t]{\text{\hskip 0.45209pt--\hskip 0.45209pt3}}}\mkern 9.0mu$}}}}$}}{}{}\text{M}{\vphantom{\text{X}}}^{\text{2+}}\penalty\hskip 2.15277pt+\hskip 2.15277pt{}\text{CO}{\vphantom{\text{X}}}_{\smash[t]{\text{3}}}{\vphantom{\text{X}}}^{\text{2\hskip 0.90417pt--\hskip 0.90417pt}}\end{split} (1) where {}\mathrm{M} represents the metal ion. The total forward and backward reaction rates R_{f} and R_{b} are obtained as \begin{split}R_{f}&=k_{1}\,a_{{}\mathrm{H}{\vphantom{\mathrm{X}}}^{\mathrm{+}}}+k_{2}\,a_{{}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast}}+k_{3}\\ R_{b}&=k_{-1}\,a_{{}\mathrm{M}{\vphantom{\mathrm{X}}}^{\mathrm{2+}}}\,a_{{}\mathrm{HCO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}{\vphantom{\mathrm{X}}}^{\mathrm{-}}}+k_{-2}\,a_{{}\mathrm{M}{\vphantom{\mathrm{X}}}^{\mathrm{2+}}}\,a_{{}\mathrm{HCO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}{\vphantom{\mathrm{X}}}^{\mathrm{-}}}^{2}+k_{-3}\,a_{{}\mathrm{M}{\vphantom{\mathrm{X}}}^{\mathrm{2+}}}\,a_{{}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}{\vphantom{\mathrm{X}}}^{\mathrm{2-}}}\\ &=k_{4}\,a_{{}\mathrm{M}{\vphantom{\mathrm{X}}}^{\mathrm{2+}}}\,a_{{}\mathrm{HCO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}{\vphantom{\mathrm{X}}}^{\mathrm{-}}}\end{split} (2) The empirical rate parameters k_{1} , k_{2} , k_{3} , and k_{4} are either stated in literature directly [ 2 ] , or are described by the equation \log\,k_{i}=a_{i}\,(1/T)+b_{i} (3) where T is the temperature, and a_{i} and b_{i} can be found in literature [ 1 , 9 ] . A semi-empirical rate model that can fit the experimental data reasonably well without explicit knowledge of the backward rate coefficients is given in Palandri and Kharaka 2004 as \begin{split}R=-A&\left[k_{\text{acid}}^{298.15\text{K}}\,e^{-\frac{E_{1}}{R}\left(\frac{1}{T}-\frac{1}{298.15\text{K}}\right)}\,a_{{}\mathrm{H}{\vphantom{\mathrm{X}}}^{\mathrm{+}}}^{n_{1}}\,\left(1-\Omega^{p_{1}}\right)^{q_{1}}\right.\\ &\left.+k_{\text{neutral}}^{298.15\text{K}}\,e^{-\frac{E_{2}}{R}\left(\frac{1}{T}-\frac{1}{298.15\text{K}}\right)}\,\left(1-\Omega^{p_{2}}\right)^{q_{2}}\right.\\ &\left.+k_{\text{carbonate}}^{298.15\text{K}}\,e^{-\frac{E_{3}}{R}\left(\frac{1}{T}-\frac{1}{298.15\text{K}}\right)}\,a_{{}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast}}^{n_{3}}\,\left(1-\Omega^{p_{3}}\right)^{q_{3}}\right]\end{split} (4) where A is the mineral surface area, E_{i} is the activation energy of mechanism i , R is the gas constant, T is the temperature, a_{j} is the activity of species j , and \Omega is the mineral saturation index. n_{i} , p_{i} , and q_{i} are dimensionless empirical parameters. Note that for 25°C, the exponential term becomes unity and the relation for the forward reaction rate ( 2 ) can be obtained. Also note that the indices of the neutral mechanism and of the carbonate mechanism are switched in ( 4 ). The rate parameters k_{\text{acid}}^{298.15\text{K}} , k_{\text{neutral}}^{298.15\text{K}} , and k_{\text{carbonate}}^{298.15\text{K}} can be obtained from literature [ 1 , 2 , 9 , 10 ] ; however, a unit conversion may be necessary. As an example, we demonstrate this derivation for calcite. [ 9 ] yields the reaction orders n_{1}=1 with respect to {}\mathrm{H}{\vphantom{\mathrm{X}}}^{\mathrm{+}} for the acid mechanism and n_{3}=1 with respect to {}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast} for the carbonate mechanism, as well as the empirical rate parameters a_{1}=-444.0 , b_{1}=0.198 , a_{2}=-2177.0 , b_{2}=2.84 , a_{3}=-317.0 , and b_{3}=-5.86 ; thus, using equation ( 3 ) and T=298.15 K \begin{split}&\log\,k_{1}=-1.3\Rightarrow k_{1}=10^{-1.3}\,\frac{\text{mmol}}{\text{cm}^{2}\,\text{s}}=10^{-0.3}\,\frac{\text{mol}}{\text{m}^{2}\,\text{s}}\\ &\log\,k_{2}=-4.46\Rightarrow k_{2}=10^{-4.46}\,\frac{\text{mmol}}{\text{cm}^{2}\,\text{s}}=10^{-3.46}\,\frac{\text{mol}}{\text{m}^{2}\,\text{s}}\\ &\log\,k_{3}=-6.92\Rightarrow k_{3}=10^{-6.92}\,\frac{\text{mmol}}{\text{cm}^{2}\,\text{s}}=10^{-5.92}\,\frac{\text{mol}}{\text{m}^{2}\,\text{s}}\end{split} (5) These values for k_{1} , k_{2} , and k_{3} align reasonably well with \log\,k_{\text{acid}}^{298.15\text{K}}=-0.30 , \log\,k_{\text{carbonate}}^{298.15\text{K}}=-3.48 , and \log\,k_{\text{neutral}}^{298.15\text{K}}=-5.81 stated in Palandri and Kharaka 2004 . The method can be applied similarly to the other carbonate minerals mentioned herein. 3 Demonstration and Validation To illustrate the impact of using the model with parameters as stated by Palandri and Kharaka 2004 , or any database implementing this data, we simulated a simple calcite dissolution batch test based on the experiment conducted in Plummer and Wigley 1976 . The test was simulated using Reaktoro [ 6 ] , a commonly used computational tool for modeling chemically reactive systems, and the results were compared to the original experimental data from run 7 in Plummer and Wigley 1976 . Figure 1 shows that implementing the reaction order n of the carbonate mechanism with respect to P({}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}) results in a considerable overestimation of reaction timescales; the time to reach saturation is approximately 20 times faster than observed experimentally. Conversely, implementing the reaction order n with respect to {}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast} aligns well with the experimental data, demonstrating the validity of the proposed adjustment. Similar observations can be made for the species molality of {}\mathrm{Ca}{\vphantom{\mathrm{X}}}^{\mathrm{2+}} in water. Implementing the original model from [ 9 ] in the software DuMuX [ 5 ] provides additional support for our claim. This approach also yields results consistent with our proposed implementation in Reaktoro and experimental data. The slight difference in outcomes between Reaktoro and DuMuX can be explained by the different underlying approaches; Reaktoro’s kinetics solver minimizes the Gibbs energy, while DuMuX solves the electroneutrality equation. Figure 1: Comparison of the originally reported values in Palandri and Kharaka 2004 , our proposed adjustment, simulation of the model from Plummer et al. 1978 in DuMuX and experimental data from Plummer and Wigley 1976 . Evidently, reaction time scales are significantly overestimated by the original report. 4 Conclusion This technical note demonstrates that the original parameters reported in Palandri and Kharaka 2004 , using the reaction order n with respect to P({}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}) , very significantly overestimate reaction timescales. Adjusting the reaction order to reference {}{}\mathrm{H}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{2}}}\mathrm{CO}{\vphantom{\mathrm{X}}}_{\smash[t]{\mathrm{3}}}^{\ast} is required and