Friday, 24 April, 2026
Speciation and Propagation
Speciation, kinetics, and transport are not independent processes: a contaminant does not migrate the same way whether it is in a mobile or adsorbed form, nor does it degrade at the same rate depending on local redox conditions. This page details how IsoFind couples these three dimensions within the simulation engine, focusing on scenarios where this coupling qualitatively changes the predicted outcome.
The Coupling Problem
A pure transport model assumes the contaminant is chemically inert: its only evolution is its displacement. A reactive model without transport assumes the contaminant remains stationary and only evolves chemically. Reality combines both, and the coupling produces behaviors that neither simple model can capture alone.
The most demonstrative example is a groundwater chromium plume. Cr(VI) is soluble and mobile, moving with the flow. Cr(III) is poorly soluble and strongly adsorbed, remaining almost stationary. A single chromium source can therefore produce two radically different plumes depending on local redox conditions. The transformation of Cr(VI) to Cr(III) via microbial or abiotic reduction is thus a massive plume attenuation mechanism—provided it is effective in the zones the plume traverses.
The ability to couple speciation, kinetics, and transport is what sets IsoFind’s 3D Visualization apart from classical simulation tools. Local parameters (pH, Eh, organic matter) modulate kinetics cell by cell, not just at a global scale.
Cell-by-Cell Redox Speciation
For compounds whose behavior depends on redox state, the engine calculates the distribution between species in each cell at every time step based on local pH and Eh. This distribution is derived from Eh-pH diagrams tabulated in the reference database, adjusted according to the concentrations of complexing species present.
| Element | Dominant Species | Primary Control |
|---|---|---|
| Chromium | Mobile Cr(VI), Adsorbed Cr(III) | Eh primarily, pH secondary |
| Iron | Mobile Fe(II), Fe(III) precipitated as oxyhydroxides | Eh, pH, alkalinity |
| Arsenic | Adsorbed As(V) oxyanion, less adsorbed/mobile As(III) | Eh, presence of Fe oxyhydroxides |
| Uranium | Soluble U(VI) uranyl, insoluble U(IV) uraninite | Eh, carbonate complexation |
| Manganese | Soluble Mn(II), solid Mn(IV) oxides | Eh, pH, microbial activity |
| Selenium | Highly mobile Se(VI), less mobile Se(IV), insoluble Se(0) | Eh, presence of Fe(II) |
Kinetic Coupling
For organic compounds, the degradation kinetic constant $k$ is not uniform across the domain: it depends on the locally dominant pathway, which is itself a function of geochemical conditions. The engine therefore calculates an effective $k$ per cell using several information sources.
| Information Source | Contribution to Local $k$ Calculation |
|---|---|
| Molecular Reference Database | List of possible pathways, reference $k$, favorable conditions |
| Local Geochemical Conditions | Interpolated pH, Eh, dissolved O₂, nitrates, sulfates, iron, methane |
| Local Temperature | Arrhenius correction applied to the reference $k$ |
| Local Organic Matter | Proxy for potential microbial biomass |
The result is a spatially variable $k$ field that can vary tenfold between an aerobic zone rich in organic carbon and an anaerobic oligotrophic zone. This variability generates the plume shapes observed in the field: rapid degradation zones at the aerobic periphery and central persistence zones under more reducing conditions.
Nexus Bridge Overrides
To explicitly control the choice of pathway or fractionation mode, the simulation request accepts several fields that bypass the Bridge's auto-selection. These overrides are useful when the site context contradicts what the Bridge would automatically deduce, or for testing alternative hypotheses.
| Request Field | Role | Typical Values |
|---|---|---|
| nexus_process_hint | Forces the process category to query in the fractionation database | redox, adsorption, biological, dissolution, precipitation, evaporation |
| prefer_fractionation_mode | Forces the fractionation mode | equilibrium (Rayleigh) or kinetic |
| oxidation_state_initial & _final | Refines the query for elements requiring speciation | e.g., Cr(VI) → Cr(III) |
| csia_pathway | Explicit degradation pathway for a molecule (overrides auto-selection) | Exact name from molecule_degradation_pathways.pathway_name |
| csia_element | Preferred isotopic element for tracking | C (default), N, Cl, H |
Parent-Metabolite Chains in Simulation
When the reference database defines a parent-metabolite chain for the simulated compound, the engine propagates the parent and its daughter products simultaneously. The coupling works in a cascade: the degradation of the parent at rate $k_{parent}$ produces a metabolite, which in turn degrades at rate $k_{metabolite}$. For chlorinated solvents, this cascade can include four stages (PCE to TCE to cis-DCE to VC to ethylene) simulated in parallel.
Each metabolite has its own transport parameters: different $K_d$, specific kinetics, and distinct isotopic signatures. The simulation thus produces a stack of 3D concentration maps—one per compound in the chain—allowing visualization of where each compound dominates spatially.
To interpret a cascading plume, the most informative visualization mode is the semi-transparent overlay of isosurfaces for the parent and primary metabolites. This immediately shows whether the parent and its products coexist (active degradation) or are spatially segregated (clear degradation front between the source and peripheral zones).
Isotopic Signature Propagation
The isotopic signature is propagated in parallel with concentration. In every cell where degradation occurs, isotopic enrichment is calculated using the Rayleigh equation with the $\epsilon$ of the locally dominant pathway. The final field is a 3D map of expected $\delta$ values, directly comparable to CSIA data measured on project samples.
This comparison is one of the most critical uses of simulation: a model that quantitatively reproduces not only concentrations but also residual isotopic signatures is far more robust than one that only fits concentrations. Two radically different models may produce the same concentrations, but they will rarely yield the same signatures.
Non-linear Adsorption and Memory Effects
In cases where a contamination source was active for a long time before stopping, the solid matrix may have accumulated a significant amount of adsorbed contaminant. This mass then gradually desorbs, feeding the plume for potentially long periods after source termination. IsoFind represents this phenomenon with an optional Langmuir isotherm, which is more realistic than the linear $K_d$ model for high concentrations.
$$q = \frac{q_{max} \cdot K \cdot C}{1 + K \cdot C}$$
Where $q$ is the adsorbed mass per mass of solid, $q_{max}$ is the maximum matrix capacity, $K$ is an affinity constant, and $C$ is the concentration in solution. For $C$ much smaller than $1/K$, the Langmuir isotherm reduces to linear $K_d$; for large $C$, it saturates. This saturation explains why high concentrations can migrate faster than a classical $K_d$ model would predict.
Precipitation and Coprecipitation
For certain elements, the dominant mechanism is not adsorption but precipitation as a distinct solid phase. Reduced chromium precipitates as Cr(OH)₃ oxyhydroxide, ferrous iron oxidizes into Fe(III) oxyhydroxides, and reduced uranium precipitates as uraninite. These precipitations are treated as mass sinks in transport, with a rate controlled by local supersaturation.
Coprecipitation—where an element is scavenged by the precipitation of another phase—is also modeled for common cases: As coprecipitated with Fe oxyhydroxides, trace metals with CaCO₃, etc. These mechanisms explain why metallic plumes often terminate at sharp redox interfaces without requiring biological degradation.
Coupling Limitations
- Speciation-transport coupling assumes local thermodynamic equilibrium at each time step. For slow reactions, this hypothesis may be invalid; the engine then offers a re-equilibration kinetic with a configurable rate.
- Representation via a molecular reference database assumes the tabulated pathways cover the actual mechanisms at the site. Unlisted pathways (exotic microbial strains, unique conditions) are not accounted for.
- Small-scale heterogeneities (fissures, micropores, aggregates) that generate non-Fickian behaviors are not resolved by the regular grid scheme.
- Interactions between multiple co-occurring contaminants (adsorption competition, mutual kinetic inhibition) are only modeled when explicitly declared.
Coupling produces impressive results, but uncertainty regarding input parameters (especially degradation rates $k$ and speciation constants) remains the limiting factor. Presenting results with a sensitivity analysis is more honest than presenting a single simulation as absolute truth.
Learn More
- Engine Principles: Basic equations without coupling.
- Degradation Pathways: Favorable conditions for each pathway.
- Redox Speciation: General framework for Cr, Fe, As, and U equilibria.
- Case Study: Cr(VI) Plume: Practical application of coupling.