Research

Paper

A Comparative Model-Analysis on Sulphide Bio-oxidation with Different Electron Acceptors

01

Summary

What the paper reports, in its own words.

AI Enhanced· 72%Resource Recovery

Abstract

Sulphide (H2S, HS- and S2-) is an undesired by-product of biogas production processes. This modelling work in Aquasim was carried out to study three parallel processes related to sulphide in AD processes: 1) H2S liquid-gas mass transfer; 2) Acid-base equilibrium; and 3) Sulphide oxidation with three different electron acceptors; nitrate, oxygen, and a biotic anode with a given potential. Multiplicative Monod (biotic processes) and Nernst-Monod kinetics (bioelectrochemical process) provide the basis for the sulphide bio-oxidation processes. At the current stage, the model can be used to study sulphide bio-oxidation and the effect of relevant parameters, including initial biomass concentration, uptake rates, temperature, and pH. The model can be improved further by implementing anaerobic microbial processes as competing reactions. With the proposed improvements, the model can be a useful tool for calculating the chemical dosage or electrode potential required for sulphide removal. These calculations can be based on both the concentration of H2S(g) in the headspace (ppm) often available at full-scale plants and the concentration of sulphide (HS-(liq)) in effluent streams from the plants.

AI summary

72% confidence

This paper presents a mechanistic Aquasim model for H2S removal from anaerobic digesters via bio-oxidation using three electron acceptors: nitrate, oxygen, and a bioanode. The model integrates liquid–gas mass transfer, acid–base equilibrium, and Monod/Nernst–Monod kinetics. Sensitivity analysis reveals maximum uptake rate (km⁰) as the dominant control on sulphide removal rate, while anode potential exhibits a plateau effect above 0.1–0.15 V. The work is primarily a modeling framework study with limited experimental validation, designed to predict H2S(g) headspace concentration and estimate dosage/time requirements for target sulphide levels in full-scale biogas plants.

Key insights

  • Threshold Maximum microbial uptake rate km⁰ is the most sensitive parameter controlling SHS⁻ concentration; low km⁰ values (2.5–5 mole S mole⁻¹ X d⁻¹) cause sulphide accumulation and high H2S(g) volatilization regardless of kinetic model used.

    So what. Operators should prioritize enriching or inoculating with high-activity sulphide-oxidizing cultures or increase active biomass concentration rather than relying on electron acceptor dosage alone.

  • Threshold Anode potential exhibits a plateau effect: increasing local potential from −0.1 to +0.3 V improves sulphide oxidation rate only up to 0.1–0.15 V; beyond this, Nernst–Monod kinetics collapse to single Monod kinetics and further potential increase yields no benefit.

    So what. Bioelectrochemical sulphide oxidation systems need not operate above 0.1–0.15 V; higher potentials waste energy without improving removal kinetics under substrate-limited conditions.

  • Trade-off pH has a much stronger effect on H2S(g) partial pressure than temperature: increasing pH from 6.8 to 8.0 reduces PH2S to 12% of the original value, whereas temperature change from 15 to 55 °C changes PH2S by only 77% ratio.

    So what. pH control is a more cost-effective operational lever than temperature management for reducing H2S emissions from biogas; raising pH to 8.0 is far more effective than heating.

  • Trade-off Continuous low-dose electron acceptor supply (1.68 kg d⁻¹ O₂ or 2.60 kg d⁻¹ NO₃⁻) reaches target SHS⁻ = 0.1 mM in 65–76 days without initial shock dosing, whereas 1.5× dosage reduces time to ~22 days, showing strong dose–response nonlinearity.

    So what. Gradual dosing strategies avoid inhibition and allow biomass acclimation; operators can trade off startup time against chemical cost and microbial stress.

  • Performance Oxygen achieves target sulphide removal 1.7× faster than nitrate (2.4 vs. 3.9 days with initial dosing; 65 vs. 76 days continuous) and requires lower total chemical input (35.84 kg initial + 1.68 kg d⁻¹ vs. 55.55 kg + 2.60 kg d⁻¹), but the model does not account for inhibition or competing reactions.

    So what. Oxygen is kinetically superior for H2S removal in anaerobic digesters, but real-world performance may differ due to unmodeled inhibition; experimental validation is needed before full-scale deployment.

Contradictions1

Tensions the paper itself points to. Each side quotes the passage it comes from.

  • Within the paper Model predicts anode potential has lower impact on bioelectrochemical sulphide oxidation than expected, yet anode-based removal is fastest in case study

    One side

    Anode potential exhibits low sensitivity in sulphide concentration (SHS⁻) and plateau effect above 0.1–0.15 V

    “the uptake rate of sulphide increases with an increase in the potential up to a certain threshold (here: 0.1-0.15 V)... Consequently, increasing the potential further would not improve the oxidation rate”
    Other side

    Anode achieves target sulphide removal in 1.7 days vs. 2.4 days (oxygen) and 3.9 days (nitrate), making it fastest pathway

    “the simulation results show that the target concentration is reached within 3.9 days with nitrate, 2.4 days with oxygen, and 1.7 days with an anode potential of 0.1 V”

    Authors’ reading. Authors acknowledge this in Section 3.4: 'the anode potential has a lower impact on the simulated bioelectrochemical sulphide oxidation rate than expected.' The discrepancy arises because the Nernst–Monod term saturates at low potential, making kinetics substrate-limited; the anode's speed advantage comes from stoichiometry and biomass kinetics, not potential control. Authors note this as a model limitation requiring further investigation.

Opportunity areas6

Specific gaps in this paper, with an experiment that would close each.

  • High Model does not test or compare different initial biomass concentrations (Xeet, Xsob, Xsnb) systematically; only one base case (0.1 mM) and one sensitivity run (1 mM) are shown

    Section 3.1.1 states 'The effect of the uptake rate is more prominent with a lower concentration of active biomass (results not included)' and Section 3.3 mentions 'by increasing the initial biomass from 0.1 to 1 mM, the target concentration of 0.1 mM sulphide was reached in less than 60 days (results not included).' This suggests biomass concentration is a critical control but is not systematically characterized.

    Try. Vary initial active biomass concentration from 0.01 to 1.0 mM for each electron acceptor pathway; measure time to reach SHS⁻ = 0.1 mM and steady-state removal rate as function of biomass.

  • High Model assumes chemolithoautotrophic growth only; does not include heterotrophic competition or inhibition by high nitrate/oxygen concentrations, which are known to occur in real AD systems

    Section 3.4 states: 'To improve the model, other anaerobic microbial processes and competing reactions should be implemented. These processes can affect the oxidation rates through competition for substrates and inhibition.' Authors also note 'supplying high initial concentrations is unrealistic, as the microorganisms can be negatively affected by too high dosages of nitrate or oxygen' but do not model this.

    Try. Add inhibition kinetics (e.g., competitive or non-competitive) for high NO₃⁻ and O₂ concentrations; test km⁰ and Ks values at [NO₃⁻] > 5 mM and [O₂] > 2 mM to identify inhibition thresholds.

  • High Anode potential implementation uses simplified Nernst–Monod with single electron transfer (n=1) and no biofilm thickness, mass transfer, or competing electron sinks; model does not capture potential-dependent biofilm dynamics

    Section 3.4 explicitly states: 'To improve the model implementation of bioelectrochemical sulphide oxidation, incorporation of different stoichiometry and other bioanode-related processes would be valuable. This includes biofilm thickness, mass transfer limitations, competing reactions at the bioanode surface, and electron transfer limitations (due to different electron sinks).' The plateau effect at 0.1–0.15 V may be an artifact of oversimplification.

    Try. Implement biofilm-based model with variable thickness and electron transfer resistance; test anode potential from −0.2 to +0.5 V with measured biofilm conductivity and compare to experimental MES data.

  • Medium Model does not test C/N ratio or carbon source availability, which are critical for heterotrophic growth and may compete with chemolithoautotrophic sulphide oxidation in real AD

    Paper assumes chemolithoautotrophs only (Section 2, assumption 2) and does not model organic carbon or heterotrophic pathways. In real biogas digesters, organic matter is abundant and may support heterotrophic denitrifiers or aerobic heterotrophs that compete with sulphide oxidizers.

    Try. Add heterotrophic denitrification and aerobic heterotrophy pathways; vary influent COD/NO₃⁻ ratio from 1 to 10 and measure sulphide removal rate and biomass composition.

  • Medium Model does not predict or test N2O or NO₂⁻ accumulation during nitrate-based sulphide oxidation; only assumes complete reduction to N₂

    Stoichiometric equations 10–11 assume 100% conversion to N₂ with no intermediate accumulation. Real denitrifying systems often produce N2O, especially under low C/N or high NO₃⁻ conditions. No sensitivity analysis on N2O yield or NO₂⁻ accumulation is provided.

    Try. Implement branched denitrification pathway with NO₃⁻ → NO₂⁻ → NO → N₂O → N₂; vary [NO₃⁻] from 0.1 to 5 mM and measure N2O and NO₂⁻ accumulation as function of SHS⁻ and pH.

  • High Model validation is entirely absent; no experimental data from real biogas plants or lab reactors are used to calibrate or validate predictions

    Section 3.4 concludes: 'calibration and validation of the model with data from real plants will improve the model further.' Authors acknowledge 'different kinetic constants and parameters can be found in the literature related to the sulphide routes and the specific microbial groups. The variation will affect the simulation results.' No comparison to published experimental data is provided.

    Try. Conduct batch or continuous-flow experiments with real biogas digestate or enriched sulphide-oxidizing cultures; measure SHS⁻, SNO₃⁻, SO₂, and biomass over time; fit Monod and Nernst–Monod parameters; validate model predictions against independent dataset.

Generated by MESSAI extraction pipeline · review against source PDF

02

Extracted values

Every value pulled from the paper, with its source, so each number can be checked against the text.

n = 57 extracted values

Select a citation to see the quote, page and confidence behind a value.

4 of 57 extracted rows modelable

  • 14 not canonicalised (no canonical slug)
  • 26 no SI value
  • 10 outside physical bounds
  • 3 failed a physics check
ParameterValueUnitConfidenceSourceNotesModelable
C/N Ratio50—70%
extracted
Excluded: no SI value
Ratio50/70%
extracted
"ratio is 50/"Excluded: not canonicalised (no canonical slug)
pH0.1-60%
extracted
"otic sulphide oxidation processes."Excluded: failed a physics check
Temperature96,500Outside physical range, check source°C60%
extracted
"omass on uptake of sulphide. Used for"Excluded: outside physical bounds
Mode30-60%
extracted
"ed that there is no proton"Excluded: no SI value
Mode452-60%
extracted
"‘Biogas upgrading and"Excluded: no SI value
Mode454-60%
extracted
"43;3372-3377. doi:10.1021/es802809m"Excluded: no SI value
Mode12,000-60%
extracted
"cell (MFCs-MEC) coupled system’."Excluded: no SI value
Mode1,170-60%
extracted
"cell (MFCs-MEC) coupled system’."Excluded: no SI value
Mode2-60%
extracted
"‘Biogas upgrading and"Excluded: no SI value
Mode424-60%
extracted
") with sulfate reduction for a very high"Excluded: no SI value
Temperature55°C60%
extracted
") to account for dissociation,"Modelable
Temperature35°C60%
extracted
"oxygen, and varied"Modelable
Temperature15°C60%
extracted
") to account for dissociation,"Modelable
pH10-60%
extracted
"chapter 3.2). The concentration of biomass, 𝑋𝑖,"Modelable
pH1,170Outside physical range, check source-60%
extracted
"res, C. I., and Rittman, B. E. (2007)"Excluded: outside physical bounds
pH15Outside physical range, check source-60%
extracted
"0-10000 ppm H2S"Excluded: outside physical bounds
pH100Outside physical range, check source-60%
extracted
"ogas reactor with a total"Excluded: outside physical bounds
pH20Outside physical range, check source-60%
extracted
"to the modelled AD reactor."Excluded: outside physical bounds
pH0-60%
extracted
"on (AD)"Excluded: failed a physics check
pH50Outside physical range, check source-60%
extracted
"on (AD)"Excluded: outside physical bounds
pH30Outside physical range, check source-60%
extracted
"on (AD)"Excluded: outside physical bounds
pH288Outside physical range, check source-60%
extracted
"1.0∙10-4"Excluded: outside physical bounds
pH1-60%
extracted
"of the first dissociation"Excluded: failed a physics check
pH168Outside physical range, check source-60%
extracted
"res, C. I., and Rittman, B. E. (2007)"Excluded: outside physical bounds
Showing 25 of 57

Methodology

What was the system and how was it run

Methodology

Methodology

The authors developed a mechanistic model in Aquasim integrating three coupled processes: (1) H2S liquid–gas mass transfer via Henry's law with temperature-dependent constants, (2) acid–base equilibrium of H2S/HS⁻ with temperature-corrected pKa, and (3) biological sulphide oxidation via multiplicative Monod kinetics (nitrate, oxygen) or Nernst–Monod kinetics (bioanode). Sensitivity analysis identified critical parameters; acid–base and air–water equilibrium calculations explored pH and temperature effects. A case study compared the three pathways under base conditions (SHS⁻ = 1.5 mM, biomass = 0.1 mM, target = 0.1 mM) with and without initial electron acceptor dosing.

System configuration

FieldValue
reactor typeBES
chamber config1-chamber

Microbial kinetics

2 constants
OrganismSubstrateParameterValueUnitConf.
mixed culture (sulphide-oxidising microbes)(unspecified)decay_rate0.048per day100%
mixed culture (sulphide-oxidising microbes)sulphideyield0.03mole S_X mole S⁻¹100%

Pipeline

Raw extractions, artifacts, quality signal

Extraction quality

honesty signal

Calibrate your trust in the extracted content against these signals — they're how the pipeline rated its own output for this paper.

Parameter completeness

13%

Fraction of expected parameters extracted

Data completeness

40%

Overall extraction coverage across fields

AI confidence

72%

Model self-reported confidence

Mean parameter confidence

50%

Average confidence across PaperParameter rows surfaced above

Last extraction:vpipeline-v2FULL_EXTRACTIONcompleted57 parameters surfaced

review against source PDF before citing · scores 0–1

03

Chart explorer

Move the inputs this paper used and see how the predicted outcome shifts.

04

Explore model

The reactor in 3D, with flow you can switch on and off.

05

AI insights

Contradictions, gaps and opportunities found by comparing this paper with the corpus.

06

Compare similar papers

The nearest papers by system, materials and organisms, with their outcomes side by side.

07

Source PDF

The paper as published.

No 3D model is mapped to this paper yet. The values extracted from it are listed under Extracted values.

57
Extracted values
2022
Year

Keywords

  • Electron acceptor
  • Nernst equation
  • Anode
  • Anaerobic respiration
  • Biomass (ecology)
  • Biogas

Cite

Vibeke B. Karlsen, Gamunu Samarakoon, C. Dinamarca (2022). A Comparative Model-Analysis on Sulphide Bio-oxidation with Different Electron Acceptors. Linköping Electronic Conference Proceedings. https://doi.org/10.3384/ecp192014