Bioelectrochemical nitrogen fixation towards sustainable hydroponics

Abstract

Ammonia plays a crucial role as a nitrogen source in agriculture, but its conventional production through the Haber–Bosch process is highly energy- and carbon-intensive. Bioelectrochemical nitrogen fixation offers a promising route by coupling renewable electricity with the selectivity of biological nitrogenases. Here, we developed a bioelectrochemical–hydroponic system that integrates a nickel–molybdenum (NiMo) cathode for hydrogen evolution with a hydrogen-oxidizing diazotroph Xanthobacter autotrophicus (X. autotrophicus) for nitrogen fixation under ambient conditions. The NiMo@carbon rod electrode exhibited excellent electrocatalytic activity, reaching a current density of 10 mA cm⁻² at an overpotential of −97 mV. It maintained stable operation at 10 mA for 7 days in a microbial medium. Extracellular NH₄⁺ reached 13 mg L⁻¹ after inhibition of glutamine synthetase, indicating that ammonium accumulation was strongly limited by cellular assimilation under normal growth conditions. Acetylene reduction further confirmed nitrogenase activity in the biohybrid system. The resulting bioelectrolyte was directly employed in mug bean hydroponic cultivation, where it promoted shoot-oriented seedling development and produced a significantly lower root-to-stem length ratio than the mineral medium control. The approach was further evaluated in 3 L and 27 L hydroponic modules, demonstrating the compatibility of bioelectrolyte with larger hydroponic cultivation formats. This work establishes new opportunities to develop sustainable, scalable, and decentralized bioelectrochemical technologies for nitrogen fixation in agriculture for hydroponic systems.

---

1. Introduction

Ammonia (NH₃) serves as the primary building block for modern nitrogen fertilizer production, with more than 180 million tons produced annually and over 80% used for fertilizer manufacturing.^1–3 Ammonia synthesis remains dominated by the industrial Haber–Bosch process, which requires extreme conditions (400–500 °C, 150–300 atm) and relies on large quantities of H₂ from fossil fuels. This energy-intensive process accounts for ∼1.5% of global CO₂ emissions and is increasingly inconsistent with the transition toward carbon-neutral agriculture.^4,5 Therefore, developing decentralized and low-carbon ammonia production routes under ambient conditions is essential to support food and fertilizer production.

Green ammonia production has attracted broad attention as a low-carbon alternative to the Haber–Bosch process, given its sustainable and environmentally friendly advantages. In principle, photocatalytic and electrocatalytic nitrogen reduction convert N₂ and H₂O to NH₃ under ambient conditions,⁶ while their practical applications remain limited by poor selectivity, low faradaic efficiency, high catalyst cost, and competition with hydrogen evolution.^7–9 Recently, bioelectrochemical nitrogen fixation (e-BNF) has emerged as a different strategy that couples electrochemical energy input with the high selectivity of biological nitrogenases. The H₂-mediated bioelectrochemical system is particularly promising because cathodically generated H₂ could serve as a clean electron donor for hydrogen-oxidizing diazotrophs, enabling N₂ fixation under biologically compatible conditions.^10–12

The cathode material plays a critical role in regulating HER activity, which determines the availability of electron donors for microbial metabolism. Carbon-based electrodes (e.g., carbon rods and carbon felt) are biocompatible and inexpensive, but they are not ideal for hydrogen evolution due to their high overpotentials. Noble metal catalysts such as Pt exhibit excellent HER performance, yet their high cost and frequent reliance on acidic conditions limit their usage in microbial systems. Earth-abundant transition metal-based materials offer a more practical alternative. Among them, the NiMo alloy is attractive because of its favorable hydrogen adsorption properties, good electrochemical stability, and HER activity under neutral conditions.^13–15 These features make NiMo a suitable cathode material for linking electrochemical H₂ generation with microbial nitrogen fixation. However, how the Ni/Mo ratio shapes HER activity, and the catalyst morphology, crystalline structure, and surface chemistry under biologically compatible conditions, remain unclear.

Recent studies demonstrated that diazotrophs, including Azotobacter vinelandii, Rhodopseudomonas palustris, and Clostridium pasteurianum, can be coupled with electrochemical systems for nitrogen fixation.^10,16–18 These studies have advanced the understanding of electron transfer efficiency, hydrogenase activity, and nitrogenase-mediated N₂ reduction efficiency.^11,17,19 However, most work focused on ammonium production or microbial mechanisms, while the downstream agricultural function of the produced ammonium remains largely unexplored.¹² This gap limits the practical relevance of e-BNF, as agricultural use requires the bioelectrolyte to be not only ammonium-containing but also directly compatible with plant growth.²⁰ Hydroponic farming provides a suitable platform because plant growth is directly governed by the composition and bioavailability of the nutrient solution.²¹ It reveals significant advantages, including high nutrient uptake efficiency, reduced land and water consumption, and good adaptability to an indoor farming system.²² Yet current hydroponic cultivation still relies heavily on synthetic nitrogen fertilizers derived from Haber–Bosch ammonia, leaving an opportunity to connect green ammonia production with controlled-environment agriculture.

Here, we develop an integrated bioelectrochemical–hydroponic platform that couples electrochemical H₂ generation, microbial N₂ fixation, and plant cultivation (Fig. 1). A dual-chamber microbial electrochemical system was constructed, with an earth-abundant NiMo alloy cathode, to generate H₂ under biologically compatible conditions. X. autotrophicus, as a microbial catalyst, then used the electrochemically generated H₂ to support N₂ fixation and CO₂ assimilation through nitrogenase activity and the Calvin–Benson cycle. The resulting NH₄⁺-abundant bioelectrolyte was directly applied as the nitrogen source for hydroponic seedling cultivation. Plant responses, including root and stem elongation and overall growth performance, were evaluated to determine whether bioelectrochemically produced ammonium can function as a practical fertilizer input. The system was further examined in 3 L and 27 L hydroponic units to assess its scalability. This integrated platform connects nitrogen fixation under ambient conditions with direct plant nutrition, extending green ammonia toward practical agricultural applications. More broadly, this work provides a scalable framework for decentralized ammonia production and future controlled-environment crop cultivation.

Fig. 1 Schematic illustration of the integrated bioelectrochemical nitrogen fixation system coupled with hydroponic seedling cultivation. The microbial cell uses hydrogen generated via water electrolysis as an electron donor to reduce N₂ to NH₄⁺ via nitrogenase in Xanthobacter autotrophicus. The resulting NH₄⁺-rich bioelectrolyte is then applied to a hydroponic system to support plant growth.

2. Materials & methods

2.1 Electrode preparation

All the chemicals were of analytical grade and purchased from Sigma-Aldrich. All solutions were prepared with ultrapure water from a Milli-Q water purification system (Milli-Q Advantage A10, Germany).

NiMo-coated carbon rods (NiMo@CR) with different Ni/Mo ratios were prepared by electrodeposition.¹³ A three-electrode single cell was set up at a cathodic current density of 50 mA cm⁻² (Bio-Logic VMP300, France), in which a carbon rod (diameter: 6 mm) served as the working electrode, a piece of platinum foil as the counter electrode, and an Ag/AgCl electrode as the reference electrode. The electrolyte solution for electrodeposition comprised NiSO₄·6H₂O (0.15 mol L⁻¹), Na₂MoO₄·2H₂O (0.05, 0.1, 0.15 mol L⁻¹), and sodium citrate (0.3 mol L⁻¹). The pH value of the solution was adjusted to 9.0 by adding an appropriate amount of ammonia. The working electrode was then washed with DI water three times and dried at 60 °C for 1 hour to obtain the NiMo@CR electrodes.

2.2 Bacterial cultivation

Xanthobacter autotrophicus (ATCC 35674) was purchased from the American Type Culture Collection (ATCC). The lyophilized powder was rehydrated entirely with ∼0.5 mL of Luria–Bertani media, then transferred to 5–6 mL secondary tubes and cultured at 30 °C for 2–3 days. The succinate nutrient broth was prepared with 5 g per L nutrient broth, 4 g per L yeast extract, 3 g per L NaCl, and 5 g per L sodium succinate. After activation, the bacteria were inoculated onto a succinate nutrient agar plate overnight. A single colony was then suspended in succinate nutrient broth for overnight growth in a shaking incubator at 200 rpm and 30 °C. Stock strains were stored in 25% glycerol at −80 °C for future use.

X. autotrophicus was initially grown in succinate nutrient broth for 24 hours, and then harvested by centrifugation at 6000 rpm for 5 min (Centrifuge 5804, Germany). The cells were washed twice with phosphate-buffered saline (PBS) and resuspended in the minimal medium (K₂HPO₄ 1 g L⁻¹, KH₂PO₄ 0.5 g L⁻¹, NaHCO₃ 2 g L⁻¹, MgSO₄·7H₂O 0.1 g L⁻¹, CaSO₄·2H₂O 0.04 g L⁻¹, FeSO₄·5H₂O 0.01 g L⁻¹, Trace Mineral Mix SL-6 1 mL L⁻¹) under a headspace of 10% H₂, 10% CO₂, 2% O₂. The initial OD₆₀₀ is ∼0.20. After adaptation, X. autotrophicus was collected for experiments.

2.3 Bioelectrochemical setup and operation

The microbial electrochemical nitrogen-fixation system was operated in a dual-chamber reactor constructed from two equal-sized borosilicate glass bottles separated by a bipolar membrane (Fuel Cell Store Inc., the United States). NiMo@CR was used as the cathode, with a submerged working area of 4 cm², while Pt-coated titanium (2 cm length and 1 cm width) was applied as the anode because of its stability under neutral conditions. An Ag/AgCl reference electrode (3.5 M KCl) was inserted into the cathode chamber through a side port in the rubber stopper. The electrode was inserted through gastight rubber stoppers. Each chamber was filled with 200 mL of nitrogen-free inorganic culture medium and magnetically stirred at 200 rpm. For bioelectrochemical experiments, washed microbial cultures were resuspended to an OD₆₀₀ of ∼0.2 and inoculated into the cathode chamber. Prior to use, the catholyte was degassed with nitrogen gas for 20 minutes to remove dissolved oxygen. A gas mixture of N₂, CO₂, and O₂ (78/20/2% v/v/v) (Air liquid, Singapore) was then supplied into the cathode headspace at a flow rate of 20 sccm using a mass flow controller (Alicat Scientific, the United States). All experiments were conducted at room temperature. Data are presented as mean ± standard deviation from three biological replicates where applicable.

2.4 Hydroponic cultivation

Mung bean (Vigna radiata) seeds were commercially purchased and sterilized with 2% sodium hypochlorite for 5 min, followed by three rinses with sterile DI water. The sterilized seeds were pre-germinated on wet filter papers in a 30 °C incubator in the dark for 2 days and then transferred to a planting tray.

The bioelectrolyte collected from the e-BNF system was used as the cultivation solution to evaluate its ability to support plant growth in the 300 mL hydroponic assay. The solution was replaced with DI water and mineral medium in control treatments. Germinated seedlings were grown directly in the solution, and plant development was monitored at regular intervals. All seedlings were cultivated under controlled light (8 h/16 h day/night cycle) and at ambient temperature (20–25 °C) for 5 days. The root length and overall growth performance were recorded as the seedling-stage indicators to evaluate early plant growth and development allocation.²³ The statistical significance of the difference among treatments was assessed using a one-way analysis of variance (ANOVA), with significance set at p < 0.05. To further evaluate the scalability of this bioelectrochemical–hydroponic strategy, flow-through hydroponic cultivation systems with working volumes of 3 L and 27 L were constructed (SI Fig. S1). X. autotrophicus-containing bioelectrolyte was used because the e-BNF reactor did not yet generate sufficient sterile ammonium-containing electrolyte for liter-scale cultivation. The plant growth was evaluated based on root elongation and overall morphology.

2.5 Analytical methods

Ammonia concentration was determined by ion chromatography (IC, Metrohm, Switzerland). The mobile phase contains a mixed solution of acetonitrile, nitric acid/dipicolinic acid, and DI water (v/v/v: 100 : 147 : 753). The flow rate, injection volume, and column temperature were 0.6 mL min⁻¹, 100 µL, and 50 °C, respectively. Electrolyte samples were filtered through a 0.22 µm filter, diluted, and quantified using calibration curves (0–2 mg L⁻¹; Fig. S2).

The morphology of the electrodes was observed using Field-Emission Scanning Electron Microscopy (FESEM, JEOL JSM-7610F, Japan). The elemental distribution was measured using Energy Dispersive Spectroscopy (EDS) mapping. Ni and Mo valences were characterized by X-ray photoelectron spectroscopy (XPS, Kratos Analytical Ltd, Kratos AXIS Ultra, the United States).

2.6 Faradaic efficiency calculation

Faradaic efficiency (FE) for extracellular ammonium production was calculated according to , where F is the Faraday constant (96 485 C mol⁻¹), C is the produced ammonia concentration (g L⁻¹), V is the catholyte volume (L), M_NH4⁺ is the mole mass of NH₄⁺, I is the applied current (A), and t is the reaction time (s).

2.7 Microbial acetylene reduction assay

The microbe acetylene reduction reaction was performed following a modified literature procedure.¹² A 2 mL liquid sample from the cathode compartment of the biohybrid system was collected and then transferred into a 20 mL serum vial sealed with a butyl rubber stopper and aluminum cap. The vial was pre-filled with C₂H₂, and the reaction was conducted at room temperature. A headspace gas sample of 1 mL was collected using a gas tight syringe and analyzed with a gas chromatograph with a flame ionization detector (Agilent GC 8600, the United States).

3. Results and discussion

3.1 NiMo@CR cathode for biocompatible H₂ evolution

In the biohybrid system shown in Fig. 1, the electrode is not required to directly activate N₂; instead, it converts electrons into H₂, which is subsequently used by X. autotrophicus as a reductant for nitrogenase. H₂ evolution, normally considered a competing reaction in electrochemical N₂ reduction, was intentionally used in this study as the electron delivery process.

NiMo was electrodeposited onto carbon rods (NiMo@CR) at a constant cathodic current density of 50 mA cm⁻² for 600 s, forming evenly distributed micrometer-sized NiMo particles on the carbon surface (Fig. S3). The SEM image revealed the uniform coating of bimetallic nanoparticles, and EDS mapping confirmed the coexistence of Ni and Mo (Fig. 2a–c). XPS spectra further indicated the chemical states of the deposited metals. In the Ni 2p spectrum, peaks at binding energies of 856.4 eV and 873.7 eV correspond to Ni 2p_3/2 and Ni 2p_1/2, respectively, due to surface oxidation. Additional satellite peaks at 861.3 eV and 879.9 eV suggest the formation of the NiMo alloy. In the Mo 3d spectrum, peaks at 232.1 and 235.2 eV indicate the presence of Mo⁶⁺, attributed to air oxidation (Fig. 2d and e). The coexistence of surface oxides and the NiMo alloy was beneficial, as the alloy structure enhances HER activity through synergistic electronic interactions between Ni and Mo, while the surface oxides could promote interfacial wettability and charge transfer.^24,25

Fig. 2 Characterization and performance of NiMo@CR as a cathode. (a) SEM images and (b and c) EDS elemental mapping of the Mo and Ni catalyst, respectively. (d and e) XPS spectra of the NiMo catalyst. (f) iR-Corrected HER linear sweep voltammetry (LSV) curves of electrodes prepared by using different Mo and Ni ion ratios in 1 M KOH. (g) iR-Corrected LSV curves in a bioelectrochemical system. (h) Nyquist curves of the system, with the equivalent circuit inserted; Z′ and −Z″ are the real and imaginary parts of the impedance, respectively. (i) 7 day chronopotentiometry (CP) test of the bioelectrochemical system.

Electrochemical performance was probed using linear sweep voltammetry (LSV) with iR compensation. The HER activity was dependent on the Ni/Mo precursor ratio in KOH solution, from which the optimized NiMo@CR at a mole ratio of 0.05/0.15 revealed a current density of 10 mA cm⁻² at a low overpotential of −97 mV (Fig. 2f). This performance was comparable to those of efficient non-precious metal HER catalysts,^26,27 indicating superior catalytic activity. The optimized NiMo@CR electrode was selected and then operated continuously at a constant cathodic current of 10 mA for 7 days in the bioelectrochemical medium. The LSV curve showed an onset potential of approximately −96 mV at a current density of 0.1 mA cm⁻² (Fig. 2g). The results reveal that the NiMo cathode efficiently catalyzed hydrogen evolution in bacterial culture media without significant energy losses, emphasizing its suitability for sustained bioelectrosynthesis. Electrochemical impedance spectroscopy (EIS) results, shown in the Nyquist plots (Fig. 2h), presented a charge transfer resistance of ∼10 Ω, much lower than that for the pristine carbon rod treatment (∼50 Ω, Fig. S4), indicating fast interfacial kinetics and efficient electron transfer. The cathodic potential remained stable at −1.14 V vs. Ag/AgCl during operation (Fig. 2i), suggesting a long-term electrochemical stability and minimal electrode degradation. These results highlight the suitability of the NiMo alloy as a robust and cost-effective catalyst in microbial electrosynthesis systems.

3.2 Nitrogenase-mediated N₂ fixation to ammonium in a biohybrid system

The NiMo@CR cathode was then coupled with X. autotrophicus, a hydrogen-oxidizing diazotroph, to convert N₂ and cathodically generated H₂ into NH₄⁺. Previous microbial electrolysis studies showed that autotrophic cathodic communities enriched with Xanthobacter and Hydrogenophaga could produce NH₄⁺ from H₂, with CO₂ fixation supported, and relied heavily on synthetic microbial interactions.²⁸ This study, however, combines the nitrogen-fixing capability of Xanthobacter with in situ-generated H₂ from efficient HER to build a green biohybrid system.

The abiotic NiMo@CR control experiment revealed only a minor increase in ammonia concentration (∼1 mg L⁻¹ over 5 days) (Fig. S5), which was commonly observed in nitrogen reduction studies and may be due to the medium components or gas feed.²⁹ After adding X. autotrophicus, OD₆₀₀ was increased from 0.05 to ∼0.1 in the first 48 hours, and the extracellular NH₄⁺ concentration reached ∼2.3 mg L⁻¹ after 7 days (Fig. S6). The slight increase in NH₄⁺ concentration in biotic treatments doesn't confirm poor N₂ fixation; instead, it suggests a trade-off between the nitrogen fixation activity and assimilation into biomass.

In microbial nitrogen metabolism, fixed nitrogen is rapidly assimilated into amino acids and biomass, mainly through the glutamine synthetase/glutamate synthase pathway.³⁰ To probe the dynamic partitioning of these two processes, phosphinothricin (PPT), a glutamine synthetase inhibitor, was used after 48 hours of incubation. PPT blocks a key enzyme in the assimilation of ammonium into amino acids, and the disruption in nitrogen metabolism leads to extracellular ammonia accumulation.³¹ After PPT addition, extracellular NH₄⁺ rapidly accumulated and reached 13 mg L⁻¹ on day 7 (Fig. 3a). The improvement of NH₄⁺ yield was accompanied by a decline in OD₆₀₀ (Fig. 3b), consistent with growth inhibition caused by impaired nitrogen assimilation.³² The much higher NH₄⁺ level in the PPT-treated biohybrid than in the abiotic treatment also revealed that extracellular ammonium accumulation was mainly associated with microbial activity (Fig. S5 and 3a).

Fig. 3 N₂ reduction performance on the NiMo–X. autotrophicus hybrid system. (a) Time-dependent extracellular ammonia production. (b) OD₆₀₀ in 7 day experiments.

An acetylene reduction assay was conducted to further reveal the underlying microbial mechanism.³³ Acetylene serves as a well-established alternative substrate for nitrogenase and is reduced to ethylene by the same enzyme involved in N₂ reduction. The observed time-dependent accumulation of ethylene over 36 hours confirmed that X. autotrophicus retained nitrogenase activity in the bioelectrochemical system (Fig. S7). In all, these results revealed a nitrogenase-mediated synthesis of NH₄⁺ from N₂ reduction in the NiMo–X. autotrophicus biohybrid. The faradaic efficiency for extracellular NH₄⁺ production was ∼2.6%, which indicates soluble ammonium recovery rather than total nitrogen-fixation efficiency, as fixed nitrogen was partly retained in cellular biomass and cathodic electrons were also consumed by H₂-mediated microbial metabolism. In comparison with the reported microbial electrolysis cells, this relatively modest faradaic efficiency in the NiMo–X. autotrophicus biohybrid demonstrates the need for regulation strategies such as strain selection, metabolic regulation, and H₂ transfer regulation.

3.3 Hydroponic application of bioelectrochemical ammonium

The fertilization effect of bioelectrochemically produced NH₄⁺ was evaluated using mung bean (Vigna radiata) in a 300 mL hydroponic system. After 7 days of bioelectrosynthesis, the catholyte was filtered through a 0.22 µm membrane to remove cells and then used as the nutrient solution. Fig. 4 presents the seedling growth, including the root length and stem length, over a 5 day incubation. One parameter, the root-to-stem length ratio, was introduced to determine the elongation balance between belowground and aboveground investment.³⁴

Fig. 4 Growth performance of mung bean seedlings cultivated under 300 mL hydroponic conditions. (a) Root length changes; (b) stem length changes. (c) Ratio of the root length to stem length. (d) Root-to-stem length ratio on day 5.

Seedlings exposed to the mineral medium exhibited the highest root elongation (∼85–90 mm), surpassing both the DI water and bioelectrolyte groups (Fig. 4a). This pattern was consistent with plant nitrogen physiology, that is, the mineral medium provided a conducive ionic environment for root elongation; the presence of ammonium in the bioelectrolyte may alter the root-shoot balance.³⁵ External ammonium affects root and shoot growth disproportionally:³⁶ ammonium acted not only as a nitrogen source but also a physiological signal that can reshape root architecture, affect the pH microenvironment, alter hormone balance, and regulate the nitrogen-assimilation pathway.³⁵ At high or imbalanced levels, ammonium may inhibit root elongation and impose metabolic stress.³⁷ The lower root elongation in the bioelectrolyte group was therefore not treated as growth inhibition; instead, it reflected a shift in early seedling allocation under increased reduced-nitrogen availability. Stem growth showed different trends among treatments. The bioelectrolyte group reached 157 mm stem length on day 5, higher than that in the mineral medium (145 mm) and the DI group (130 mm) (Fig. 4b). The ammonium generated by the NiMo–X. autotrophicus system was biologically available to plants and supported early shoot development. The root-to-stem length ratio supported the dynamic pattern for treatment-dependent allocation (Fig. 4c). All groups showed a rapid decrease after transplantation, reflecting faster shoot elongation than root extension during early growth. The mineral medium primarily promoted root elongation, whereas bioelectrolyte produced comparable shoot development but a significantly lower root-to-stem length ratio, indicating a more shoot-oriented growth pattern. By day 5, the bioelectrolyte showed the lowest root-to-stem length ratio (p < 0.05) (Fig. 4d). The bioelectrochemical NH₄⁺ shifted early seedling development toward relatively greater aboveground growth.

3.4 Scale-up application in 3 L and 27 L hydroponic modules

The hydroponic system was further extended to scale-up cultivation modules to demonstrate the scalability of the system (Fig. 5). The bioelectrolyte containing live X. autotrophicus at an initial OD₆₀₀ of 0.05, rather than filtered ammonium-only catholyte, was used. At the same time, introducing a live bacterial suspension provided a more practical fertilization mode, in which microbial cells continuously served as in situ sources of ammonium within the hydroponic environment.^12,38

Fig. 5 Hydroponic planting for mung bean seedlings in scale-up reactors. (a) Configuration of 3 L deep-water culture. (b–d) Root length, stem length, and root-to-stem length ratio on day 5 in 3 L deep-water culture, respectively. (e) Configuration of 27 L flow-through culture. (f–h) Root length, stem length, and root-to-stem length ratio on day 5 in 27 L flow-through culture, respectively.

In the 3 L culture system, seedlings cultivated with the live X. autotrophicus-containing bioelectrolyte showed growth performance comparable to those grown in a mineral medium (Fig. 5a). The average root length was slightly lower in the bioelectrolyte groups, whereas the average stem length was slightly higher; as a result, the root-to-stem length ratio was marginally lower for the bioelectrolyte treatment (Fig. 5b–d). Although these differences were not statistically significant, the directional change was in accordance with the quantitative allocation pattern observed in the 300 mL bench assay (Fig. 4), where the bioelectrolyte tended to favor aboveground development relative to root extension.

We also constructed a prototype hydroponic unit with a working volume of 27 L. The system was designed as a closed-loop recirculating module, in which the nutrient solution was continuously cycled through a planting channel (Fig. S8). The channel contained evenly spaced planting holes (8 cm diameter, 100 cm length) connected to a central reservoir. By the final cultivation stage, plants in the bioelectrolyte-fed unit displayed visibly stronger growth compared with those in the control (Fig. 5e). They also showed a higher mean aboveground growth and a slightly lower root-to-stem length ratio than those in the mineral medium control (Fig. 5f–h), presenting a consistently shoot-oriented allocation pattern in the smaller hydroponic assays. These results reveal the operational compatibility of the biohybrid system for plant growth, while its mechanism for growth enhancement needs further investigation.³⁹ The live bioelectrolyte may offer a useful route for future e-BNF-integrated hydroponics, where diazotrophic cells could participate in continuous in situ nitrogen cycling during cultivation. Future work should extend the cultivation period and quantify biomass accumulation, plant nitrogen uptake, nutrient balance, and harvest yield to evaluate the agronomic potential of this bioelectrochemical fertilizer system.⁴⁰

4. Conclusion

This work presents a sustainable strategy for decentralized ammonia synthesis by coupling electrocatalytic hydrogen evolution with microbial nitrogen fixation under ambient conditions. Using an earth-abundant NiMo@CR cathode and the hydrogen-oxidizing diazotroph X. autotrophicus, the system converted cathodically generated H₂ into biologically fixed nitrogen. Inhibition of glutamine synthetase increased extracellular NH₄⁺ accumulation to 13 mg L⁻¹, revealing that ammonium release was strongly limited by cellular assimilation under normal growth conditions. The bioelectrolyte was further applied to hydroponic cultivation, where it promoted an early shoot-oriented seedling response, as reflected by a lower root-to-stem length ratio. Stepwise testing in 300 mL, 3 L, and 27 L hydroponic systems demonstrated the compatibility of the bioelectrolyte across increasing cultivation volumes. This validation highlights the potential of bioelectrosynthesis to bridge sustainable nitrogen fixation with real-world hydroponic systems. This effort opens new possibilities for future development of decentralized, carbon-neutral fertilizer systems. Further research is needed to focus on the development of new adapted strains, efficiency improvements, further system scale-up, technoeconomic analysis, and life-cycle impacts. Moreover, further steps to improve the efficiency of H₂ uptake by the microorganisms in this scheme may lead to improved energy efficiency. Together, these future research directions open up opportunities to further advance this approach toward real-world deployment.

Conflicts of interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data supporting this study are available at https://doi.org/10.5281/zenodo.17462189.

Supplementary information (SI) is available. See DOI: https://doi.org/10.1039/d5ta08745g.

Acknowledgements

This work was supported by the Singapore Ministry of Education (Tier 1) (WBS Numbers: A-0009304-00-00, A-0009304-01-00, and R-302-000-274-114). In addition, Panpan Zhang acknowledges the support from the Competitive Research Programme of the National Research Foundation Singapore (NRF-CRP27-2021-0004). We would like to acknowledge the assistance of Phei Shien in building the hydroponic device.

References

1. K. Rouwenhorst and G. Castellanos, Innovation Outlook: Renewable Ammonia, International Renewable Energy Agency, Ammonia Energy Association, Abu Dhabi, Brooklyn, 2022.
2. Ammonia Technology Roadmap – Analysis, https://www.iea.org/reports/ammonia-technology-roadmap, accessed 19 June 2025.
3. J. W. Erisman, M. A. Sutton, J. Galloway, Z. Klimont and W. Winiwarter, Nat. Geosci., 2008, 1, 636–639.
4. C. Smith, A. K. Hill and L. Torrente-Murciano, Energy Environ. Sci., 2020, 13, 331–344.
5. Z. Zhang, H. Zhang, H. Jiang and L. Li, J. Mater. Chem. A, 2024, 12, 33334–33361.
6. D. R. MacFarlane, P. V. Cherepanov, J. Choi, B. H. R. Suryanto, R. Y. Hodgetts, J. M. Bakker, F. M. F. Vallana and A. N. Simonov, Joule, 2020, 4, 1186–1205.
7. H. Jin, S. S. Kim, S. Venkateshalu, J. Lee, K. Lee and K. Jin, Adv. Sci., 2023, 10, 2300951.
8. X. Fu, J. Zhang and Y. Kang, Chem Catal., 2022, 2, 2590–2613.
9. M. Jiang, X. Chen, F. Chen, M. Wang, X. Luo, Y. He, C. Wu, L. Zhang, X. Li, X. Liao, Z. Jiang and Z. Jin, Chem, 2025, 11, 102441.
10. B. Wang, Y. Zhang and S. D. Minteer, Energy Environ. Sci., 2023, 16, 404–420.
11. S. Lu, R. M. Rodrigues, S. Huang, D. A. Estabrook, J. O. Chapman, X. Guan, E. M. Sletten and C. Liu, Chem Catal., 2021, 1, 704–720.
12. C. Liu, K. K. Sakimoto, B. C. Colón, P. A. Silver and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 6450–6455.
13. Q. Zhang, W. Xiao, W. H. Guo, Y. X. Yang, J. L. Lei, H. Q. Luo and N. B. Li, Adv. Funct. Mater., 2021, 31, 2102117.
14. F. Kracke, A. B. Wong, K. Maegaard, J. S. Deutzmann, M. A. Hubert, C. Hahn, T. F. Jaramillo and A. M. Spormann, Commun. Chem., 2019, 2, 45.
15. D. Song, J. Roh, J. Choi, H. Lee, G. Koh, Y. Kwon, H. Kim, H. M. Lee, M. Kim and E. Cho, Adv. Sci., 2024, 11, 2403752.
16. H. Chen, F. Dong and S. D. Minteer, Nat. Catal., 2020, 3, 225–244.
17. F. Li, F. Li, Y. Lin, L. Guo, L. Zhang, R. Li, Q. Tian, Y. Wang, Y. Wang, X. Zhang, J. Liu and C. Fan, J. Chem. Technol. Biotechnol., 2022, 97, 2163–2170.
18. L. Rago, S. Zecchin, F. Villa, A. Goglio, A. Corsini, L. Cavalca and A. Schievano, Bioelectrochemistry, 2019, 125, 105–115.
19. L. Zhang, L. Zeng, J. Wang, H. Wang, D. Zheng, X. Wang, D. Li and G. Zhan, ChemPlusChem, 2024, 89, e202400072.
20. R. Rad, T. Gehring, K. Pellumbi, D. Siegmund, E. Nettmann, M. Wichern and U.-P. Apfel, Cell Rep. Phys. Sci., 2023, 4(8), 101526.
21. S. Sankhalkar and V. Jamuni, in Hydroponics, ed. N. Kumar, Springer US, New York, NY, 2024, pp. 3–13.
22. E. C. Hann, S. Overa, M. Harland-Dunaway, A. F. Narvaez, D. N. Le, M. L. Orozco-Cárdenas, F. Jiao and R. E. Jinkerson, Nat. Food, 2022, 3, 461–471.
23. T. Mašková and T. Herben, Ecol. Evol., 2018, 8, 7143–7150.
24. M. Luo, J. Yang, X. Li, M. Eguchi, Y. Yamauchi and Z.-L. Wang, Chem. Sci., 2023, 14, 3400–3414.
25. L. Meng, S. Shang, S. Liu, L. Zhang, Q. Tang, H. Wang, F. Wang, C. Li, Y. Sun and H. Wu, Appl. Catal. B Environ. Energy, 2024, 358, 124388.
26. E. J. Popczun, J. R. McKone, C. G. Read, A. J. Biacchi, A. M. Wiltrout, N. S. Lewis and R. E. Schaak, J. Am. Chem. Soc., 2013, 135, 9267–9270.
27. M. Luo, J. Yang, X. Li, M. Eguchi, Y. Yamauchi and Z.-L. Wang, Chem. Sci., 2023, 14, 3400–3414.
28. L. Zhang, C. Tian, H. Wang, W. Gu, D. Zheng, M. Cui, X. Wang, X. He, G. Zhan and D. Li, Bioelectrochemistry, 2022, 144, 108044.
29. M. Kolen, D. Ripepi, W. A. Smith, T. Burdyny and F. M. Mulder, ACS Catal., 2022, 12, 5726–5735.
30. A. Suzuki and D. B. Knaff, Photosynth. Res., 2005, 83, 191–217.
31. P. J. Lea and B. J. Miflin, Plant Physiol. Biochem., 2003, 41, 555–564.
32. R. Dixon and D. Kahn, Nat. Rev. Microbiol., 2004, 2, 621–631.
33. M. J. Dilworth, Biochim. Biophys. Acta, Gen. Subj., 1966, 127, 285–294.
34. T. Ericsson, Plant Soil, 1995, 168–169, 205–214.
35. Y. Liu and N. von Wirén, J. Exp. Bot., 2017, 68, 2581–2592.
36. W. L. Pan, I. J. Madsen, R. P. Bolton, L. Graves and T. Sistrunk, Agron. J., 2016, 108, 2485–2492.
37. D. T. Britto and H. J. Kronzucker, J. Plant Physiol., 2002, 159, 567–584.
38. J. M. Sanchez-Yañez, in Current Trends in Wheat Research, ed. M.-R. Ansari, IntechOpen, London, 2022.
39. G. Rajaseger, K. L. Chan, K. Yee Tan, S. Ramasamy, M. C. Khin, A. Amaladoss and P. Kadamb Haribhai, Bioinformation, 2023, 19, 925–938.
40. W. E. Finch-Savage and G. W. Bassel, J. Exp. Bot., 2016, 67, 567–591.

---

Footnote

† These authors made equal contributions.

---