Paper · MFC
Dual-chamber microbial fuel cell with denitrifying biocathode for the treatment of cassava processing wastewater
Summary
What the paper reports, in its own words.
Abstract
ABSTRACT The aim of this study was to assess the feasibility of using cassava wastewater as a substrate for dual-chamber microbial fuel cells (MFCs) operating with denitrifying biocathodes. Two configurations related to the ion exchange membrane used were evaluated: one with an anion exchange membrane (MFC-A) and the other with a cation exchange membrane (MFC-C). Both bioreactors were operated in sequential batch mode. Furthermore, a low-cost platform based on Arduino technology was also proposed to enable continuous measurement and recording of voltage data from the MFCs. The highest voltage values were observed in the first days of MFC operation, with readings reaching approximately 350 mV (0.41 W·m 3) and gradually decreasing after 100 days of operation to 243 mV (0.20 W·m 3) and 125 mV (0.05 W·m 3) for the MFC-A and MFC-C, respectively (mean values for the last 20 days of operation). In both MFCs, the chemical oxygen demand reduction and nitrogen removal were over 98% after reactor stabilization, with no noticeable nitrite accumulation. The experimental results indicated superior performance when MFC was equipped with an anion exchange membrane. The results presented here demonstrate the feasibility of using cassava wastewater as a viable substrate for MFCs equipped with a denitrifying biocathode, allowing for efficient wastewater treatment and simultaneous electricity generation.
AI summary
72% confidenceThis study evaluated dual-chamber MFCs with denitrifying biocathodes for cassava processing wastewater treatment, comparing anion exchange membrane (AEM) and cation exchange membrane (CEM) configurations. Both systems achieved >98% COD and nitrogen removal in sequential batch mode, with MFC-A (AEM) generating 0.20 W·m⁻³ at 243 mV steady-state versus MFC-C (CEM) at 0.05 W·m⁻³ at 125 mV. No nitrite or ammonia accumulation occurred. Coulombic efficiency remained very low (~1.5–5%), and power declined over 100 days despite stable treatment performance across three organic loading rates (1.33–2.67 g COD·L⁻¹·d⁻¹).
Key insights
Trade-off AEM reduced internal resistance by 84% (22.3 vs 41.0 ohms) and minimized pH splitting between chambers, yielding 4× higher power density (0.20 vs 0.05 W·m⁻³) than CEM despite identical substrate and loading.
So what. Specify AEM over CEM in dual-chamber denitrifying MFCs; pH splitting is a critical design constraint that directly couples to power output and should be monitored continuously.
Threshold Increasing organic loading rate from 1.33 to 2.67 g COD·L⁻¹·d⁻¹ and proportional nitrate loading (0.13 to 0.26 g NO₃⁻-N·L⁻¹·d⁻¹) did not degrade COD removal (>99% maintained in MFC-A) or nitrogen removal (>99%), but power generation remained capped at ~0.20 W·m⁻³ in MFC-A.
So what. Do not assume higher loading will improve power; the MFC architecture itself limits current, not substrate availability. Optimize HRT and membrane properties instead of pushing OLR.
Mechanism Coulombic efficiency collapsed from ~5% in early operation to 1.5% by end in MFC-A, despite stable >99% COD and >99% NO₃⁻ removal, indicating that most electron flow was diverted to competing processes (methanogenesis, heterotrophic denitrification, sulfate reduction).
So what. Treat coulombic efficiency and treatment efficiency as decoupled metrics in wastewater MFCs; low CE does not invalidate treatment but signals microbial competition. Reduce substrate surplus and external resistance to recover CE.
Performance No nitrite or ammonia accumulation was detected throughout operation (below limit of quantification in influent and effluent), indicating complete denitrification to N₂ and no nitrification in the cathode despite high NO₃⁻ loading (973 mg NO₃⁻-N·L⁻¹ inlet).
So what. Autotrophic denitrifying biocathodes can handle high nitrate loads without N₂O or NO₂⁻ slip; this is a key advantage over heterotrophic denitrification. Confirm N₂ production via gas chromatography in future work.
Mechanism Voltage declined slowly but steadily in MFC-A (350→243 mV over 100 days) while MFC-C stabilized quickly (350→125 mV by day 50), suggesting different membrane fouling or biofilm maturation kinetics between AEM and CEM.
So what. Monitor voltage trajectory shape, not just steady-state value, to diagnose membrane degradation. CEM may foul faster but reach a stable plateau; AEM may degrade continuously.
Contradictions1
Tensions the paper itself points to. Each side quotes the passage it comes from.
Within the paper High treatment efficiency contradicts low coulombic efficiency in the same system
One sideCOD removal >99% and NO₃⁻ removal >99% in MFC-A
“In MFC-A, COD reduction levels above 99% were achieved after the 35th day of operation...In terms of nitrate removal, around 99% removal was observed after initial stabilization.”
Other sideCoulombic efficiency only 1.5% by end of operation in MFC-A
“In MFC-A, which performed slightly better, values around 5% were observed only in the first days of operation, gradually dropping to 1.5% by the end of the operation period.”
Authors’ reading. Authors reconcile by noting that high treatment is achieved via complementary processes (methanogenesis, heterotrophic denitrification, sulfate reduction) in parallel with electrogenic autotrophic denitrification. Coulombic efficiency measures only the electron flow through the electrode, not total nitrogen removal. This is not a true contradiction but a decoupling of two metrics in wastewater treatment MFCs.
Opportunity areas6
Specific gaps in this paper, with an experiment that would close each.
High No measurement of N₂ production or N₂O accumulation; only NO₃⁻ and NO₂⁻ were tracked. Autotrophic denitrification pathway and stoichiometry remain unconfirmed.
Paper claims 'complete denitrification' and 'no noticeable nitrite accumulation' but does not verify the final product is N₂ gas. N₂O slip would be a critical environmental concern in nitrogen cycling MES.
Try. Gas chromatography or isotope labeling (¹⁵N) to quantify N₂, N₂O, and NO production across the three loading phases; measure at cathode outlet gas phase.
High Coulombic efficiency was calculated only from nitrate removal, not from COD oxidation in the anode. Electron balance between anode and cathode is not reported.
Authors state 'coulombic efficiency was calculated using the method described by Logan et al. (2006)' but do not specify whether this was based on anode COD removal or cathode NO₃⁻ removal. The 1.5% value is suspiciously low and may indicate unaccounted electron sinks (H₂ evolution, methanogenesis).
Try. Measure anode effluent for residual COD and calculate electron balance: electrons from anode oxidation vs. electrons consumed at cathode (NO₃⁻ reduction). Quantify H₂ and CH₄ in gas phase.
High No investigation of cathode potential or anode potential; only cell voltage (V) and external resistance (R) were controlled. Electrochemical operating window is not defined.
Paper does not report cathode or anode potentials vs. reference electrode. Without this, it is impossible to determine if the system is operating in the autotrophic denitrification window or if heterotrophic denitrification is dominant. Potential windows for different N-cycling pathways are not established.
Try. Install Ag/AgCl reference electrodes in both chambers and measure anode and cathode potentials vs. SHE across the three loading phases. Correlate potential windows to N-species profiles and coulombic efficiency.
Medium Sequential batch mode with daily exchange volume (EV) was used, but no comparison to continuous flow or fed-batch operation. HRT effect on power and treatment is not isolated.
Table 1 shows HRT varied from 7.5 to 3.75 d as OLR increased, but these are confounded. No single-variable study of HRT on denitrification kinetics or power output.
Try. Fix OLR at 1.33 g COD·L⁻¹·d⁻¹ and vary HRT from 2 to 10 d in continuous flow mode; measure NO₃⁻ removal rate, N₂O accumulation, and power density.
Medium Cassava wastewater was diluted to 10 g COD·L⁻¹ (starch industry level) but not compared to flour industry wastewater (56.7 g COD·L⁻¹). Substrate strength effect on denitrification and power is unknown.
Introduction notes CPWW COD ranges from 11.6 to 56.7 g·L⁻¹ depending on source, but only the lower end was tested. High-strength wastewater may inhibit denitrifiers or shift to heterotrophic pathways.
Try. Test MFC-A with CPWW at 20, 40, and 56.7 g COD·L⁻¹ (diluted to same OLR by adjusting feed rate); measure NO₃⁻ removal, N₂O, and power across substrate strengths.
Medium Cyanide content of cassava wastewater was not quantified. Authors cite cyanide tolerance but do not measure its concentration or effect on denitrifying biofilm.
Introduction states CPWW contains cyanogenic glycoside (linamarin) that converts to cyanide, and authors note 'increments in feed loading rate...did not affect the current generation' despite increased cyanide input. But cyanide concentration was never measured.
Try. Measure free and total cyanide in influent CPWW and cathode effluent across the three loading phases using colorimetric or HPLC methods; correlate to denitrification rate and N₂O production.
Generated by MESSAI extraction pipeline · review against source PDF
Extracted values
Every value pulled from the paper, with its source, so each number can be checked against the text.
n = 73 extracted values
Select a citation to see the quote, page and confidence behind a value.
35 of 72 extracted rows modelable
- 18 not canonicalised (no canonical slug)
- 19 no SI value
| Parameter | Value | Unit | Confidence | Source | Notes | Modelable |
|---|---|---|---|---|---|---|
| Substrate | wastewater | — | 80% | llm | ||
| Organic Loading Rate | 2.67 | g COD·L⁻¹·d⁻¹ | 95% | extracted | "Phase III 40 3.75 2.67 0.26" · pH(an)=7, substrate=cassava processing wastewater | Modelable |
| Anolyte Flow Rate | 20 | mL·d⁻¹ | 95% | extracted | "Phase I 20 7.50 1.33 0.13" · pH(an)=7, substrate=cassava processing wastewater | Excluded: no SI value |
| Anolyte Flow Rate | 27 | mL·d⁻¹ | 95% | extracted | "Phase II 27 5.56 1.80 0.18" · pH(an)=7, substrate=cassava processing wastewater | Excluded: no SI value |
| Anolyte Flow Rate | 40 | mL·d⁻¹ | 95% | extracted | "Phase III 40 3.75 2.67 0.26" · pH(an)=7, substrate=cassava processing wastewater | Excluded: no SI value |
| Electrode Surface Area | 5 | cm | 95% | extracted | "Each individual chamber was fitted with a set of three interconnected stainless-…" · condition_hint=design parameter, length, electrode_material=stainless-steel mesh with activated carbon | Excluded: no SI value |
| Electrode Surface Area | 4.5 | cm | 95% | extracted | "Each individual chamber was fitted with a set of three interconnected stainless-…" · condition_hint=design parameter, width, electrode_material=stainless-steel mesh with activated carbon | Excluded: no SI value |
| Organic Loading Rate | 1.33 | g COD·L⁻¹·d⁻¹ | 95% | extracted | "Phase I 20 7.50 1.33 0.13" · pH(an)=7, substrate=cassava processing wastewater | Modelable |
| Hydraulic Retention Time Phase III | 3.75 | d | 95% | extracted | "Phase III 40 3.75 2.67 0.26" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Hydraulic Retention Time Phase II | 5.56 | d | 95% | extracted | "Phase II 27 5.56 1.80 0.18" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Nitrate Effluent | 4 | mg NO₃⁻-N·L⁻¹ | 95% | salicylic acid method | "Effluent 7.47 (±0.17) 132 (±47) 7.56 (±0.25) 4 (±2.4)" · pH(an)=7.56, substrate=cassava processing wastewater, steady_state | Excluded: not canonicalised (no canonical slug) |
| Nitrate Loading Rate Phase III | 0.26 | g NO₃⁻-N·L⁻¹·d⁻¹ | 95% | extracted | "Phase III 40 3.75 2.67 0.26" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Activated Carbon Mass | 3 | g | 95% | extracted | "Each individual chamber was fitted with a set of three interconnected stainless-…" · condition_hint=design parameter, electrode_material=stainless-steel mesh with activated carbon | Excluded: not canonicalised (no canonical slug) |
| Nitrate Loading Rate Phase II | 0.18 | g NO₃⁻-N·L⁻¹·d⁻¹ | 95% | extracted | "Phase II 27 5.56 1.80 0.18" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Hydraulic Retention Time Phase I | 7.5 | d | 95% | extracted | "Phase I 20 7.50 1.33 0.13" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Nitrate Loading Rate Phase I | 0.13 | g NO₃⁻-N·L⁻¹·d⁻¹ | 95% | extracted | "Phase I 20 7.50 1.33 0.13" · pH(an)=7, substrate=cassava processing wastewater | Excluded: not canonicalised (no canonical slug) |
| Nitrate Effluent | 21 | mg NO₃⁻-N·L⁻¹ | 95% | salicylic acid method | "Effluent 7.40 (±0.16) 192 (±26) 8.34 (±0.14) 21 (±1.6)" · pH(an)=8.34, substrate=cassava processing wastewater, steady_state | Excluded: not canonicalised (no canonical slug) |
| Chamber Volume | 0.15 | L | 95% | extracted | "Two dual-chamber MFCs with 0.15 L working volume in each chamber were used" · condition_hint=design parameter | Modelable |
| Chemical Oxygen Demand (COD) Concentration | 132 | mg COD·L⁻¹ | 95% | colorimetric method 5220-d | "Effluent 7.47 (±0.17) 132 (±47) 7.56 (±0.25) 4 (±2.4)" · pH(an)=7.47, substrate=cassava processing wastewater, steady_state | Excluded: no SI value |
| Chemical Oxygen Demand (COD) Concentration | 192 | mg COD·L⁻¹ | 95% | colorimetric method 5220-d | "Effluent 7.40 (±0.16) 192 (±26) 8.34 (±0.14) 21 (±1.6)" · pH(an)=7.4, substrate=cassava processing wastewater, steady_state | Excluded: no SI value |
| Anodic pH | 7.47 | — | 95% | portable ph meter | "Effluent 7.47 (±0.17) 132 (±47) 7.56 (±0.25) 4 (±2.4)" · pH(an)=7.47, substrate=cassava processing wastewater, steady_state | Modelable |
| Anodic pH | 7.4 | — | 95% | portable ph meter | "Effluent 7.40 (±0.16) 192 (±26) 8.34 (±0.14) 21 (±1.6)" · pH(an)=7.4, substrate=cassava processing wastewater, steady_state | Modelable |
| Cathodic pH | 7.56 | — | 95% | portable ph meter | "Effluent 7.47 (±0.17) 132 (±47) 7.56 (±0.25) 4 (±2.4)" · pH(an)=7.56, substrate=cassava processing wastewater, steady_state | Modelable |
| Cathodic pH | 8.34 | — | 95% | portable ph meter | "Effluent 7.40 (±0.16) 192 (±26) 8.34 (±0.14) 21 (±1.6)" · pH(an)=8.34, substrate=cassava processing wastewater, steady_state | Modelable |
| Organic Loading Rate | 1.8 | g COD·L⁻¹·d⁻¹ | 95% | extracted | "Phase II 27 5.56 1.80 0.18" · pH(an)=7, substrate=cassava processing wastewater | Modelable |
Methodology
What was the system and how was it run
Methodology
Methodology
Two dual-chamber MFCs (0.15 L each) were constructed with stainless-steel mesh electrodes filled with activated carbon and separated by either an anion exchange membrane (MFC-A) or cation exchange membrane (MFC-C). The anode was inoculated with anaerobic sludge and fed daily with cassava wastewater diluted to 10 g COD·L⁻¹ in sequential batch mode (exchange volumes 20–40 mL·d⁻¹, HRT 7.5–3.75 d). The cathode was inoculated with heterotrophic denitrifying sludge and fed with pre-nitrified CPWW (973 mg NO₃⁻-N·L⁻¹). Voltage was measured continuously via Arduino-based differential voltmeter at 5-min intervals. COD, NO₃⁻, and NO₂⁻ were measured daily using colorimetric and HACH methods. Polarization curves were generated at end-of-operation using 10 external resistances (52–1,000 Ω).
System configuration
| Field | Value |
|---|---|
| reactor type | MFC |
| chamber config | 2-chamber |
Performance metrics
| Field | Value |
|---|---|
| citation count | 2 |
Operational parameters
| Field | Value |
|---|---|
| substrate | ["wastewater"] |
Operating conditions
14 runs extracted| Run | T°C | pHan | pHcat | Substrate | [S]mg/L | HRTh | Mode | R_extΩ | V_appV | Inoculum |
|---|---|---|---|---|---|---|---|---|---|---|
| v2-012545f9e5187925 | — | — | — | cassava processing wastewater | 10000 | — | — | — | — | — |
| v2-0936dc6e007bcca8 | — | 7.56 | — | cassava_processing_wastewater | — | — | steady_state | 1000 | — | — |
| v2-1015a1458c073ec6 | — | 7 | — | cassava processing wastewater | — | — | — | 1000 | — | — |
| v2-123dea91b7989a0d | — | 7.4 | — | cassava processing wastewater | 10000 | — | steady_state | 1000 | — | — |
| v2-15ba370761db8312 | — | — | — | cassava processing wastewater | 10000 | — | steady_state | 1000 | — | — |
| v2-319f1ed1006c7597 | — | 7.47 | — | cassava processing wastewater | 10000 | — | startup | 1000 | — | — |
| v2-3abda78873dbd643 | — | 8.34 | — | cassava processing wastewater | 10000 | — | steady_state | 1000 | — | — |
| v2-6820b23a41f06e9e | — | 7 | — | cassava processing wastewater | 10000 | — | — | 1000 | — | — |
| v2-752e7d803115bbbf | — | 7.56 | — | cassava processing wastewater | — | — | steady_state | 1000 | — | — |
| v2-89cbcab198036ccc | — | 8.34 | — | cassava processing wastewater | — | — | steady_state | 1000 | — | — |
| v2-9982347bf5630182 | — | 7.47 | — | cassava processing wastewater | 10000 | — | steady_state | 1000 | — | — |
| v2-c81ff497f53c1817 | — | — | — | cassava processing wastewater | 10000 | — | — | 52 | — | — |
| v2-cebbb413134a140c | — | 7 | — | cassava processing wastewater | 10000 | — | startup | 1000 | — | — |
| v2-f586e9af22985397 | — | — | — | — | — | — | — | — | — | — |
Outcomes
What was measured and what worked
Electrochemistry
2 constants| Electrode | Reaction | Parameter | Value | Unit | Conf. |
|---|---|---|---|---|---|
| anode | — | internal_resistance | 22.3 | ohms | 100% |
| cathode | — | internal_resistance | 41 | ohms | 100% |
Pipeline
Raw extractions, artifacts, quality signal
Extraction quality
honesty signalCalibrate your trust in the extracted content against these signals — they're how the pipeline rated its own output for this paper.
Parameter completeness
38%
Fraction of expected parameters extracted
Data completeness
60%
Overall extraction coverage across fields
AI confidence
72%
Model self-reported confidence
Mean parameter confidence
86%
Average confidence across PaperParameter rows surfaced above
review against source PDF before citing · scores 0–1
Chart explorer
Move the inputs this paper used and see how the predicted outcome shifts.
Explore model
The reactor in 3D, with flow you can switch on and off.
AI insights
Contradictions, gaps and opportunities found by comparing this paper with the corpus.
Compare similar papers
The nearest papers by system, materials and organisms, with their outcomes side by side.
Source PDF
The paper as published.
Representative model, generic MFC
Tap to interactRepresentative model, generic MFC
Representative MFC — matched on the paper’s system type only, not its reactor or geometry.
Keywords
- Denitrifying bacteria
- Microbial fuel cell
- Wastewater
- Waste management
- Sewage treatment
- Pulp and paper industry
Cite
Aline Oliveira da Silva, Simone Perazzoli, Hugo Moreira Soares et al. (2023). Dual-chamber microbial fuel cell with denitrifying biocathode for the treatment of cassava processing wastewater. Engenharia Sanitaria e Ambiental. https://doi.org/10.1590/s1413-415220230116
Open in lab for full controls, parameter editing, and template overlays.
Open in lab