Degree

Doctor of Philosophy (PhD)

Department

The Gordon A. and Mary Cain Department of Chemical Engineering

Document Type

Dissertation

Abstract

Biological denitrification is an effective and cost-efficient approach for nitrate removal; however, conventional processes often rely on liquid carbon sources such as methanol or acetate that require precise dosing and monitoring. Solid-phase denitrification (SPD) using polyhydroxybutyrate (PHB) provides a self-regulating alternative, but its high cost limits widespread application. This dissertation investigated strategies to develop lower-cost solid carbon media and simplify SPD operation for recirculating aquaculture systems (RAS). PHB/cellulose (PHBC) blends containing 20–40% cellulose were first evaluated against pure PHB based on nitrate conversion, carbon release, substrate consumption, and denitrification cost. PHBC60:40 and PHBC70:30 achieved the highest apparent peak nitrate conversion rates (3.9 ± 0.80 kg NO₃⁻-N m⁻³ d⁻¹), while PHBC60:40 reduced the cost per kilogram of NO₃⁻-N removed by approximately 18% compared with pure PHB. PHBC60:40 was subsequently evaluated in flow-through biofilters with and without nitrogen-driven influent degassing. Although degassing increased the maximum denitrification rate by approximately 9%, dissolved oxygen in the non-degassed system was depleted within the first 5–10% of the media bed, while other performance parameters remained comparable, demonstrating that oxygen-stripping pretreatment was unnecessary. To achieve further reductions in denitrification cost, soy wax (SW) and stearic acid (SA) were evaluated as PHB-free solid carbon alternatives. SW and SA achieved higher maximum denitrification rates (1.13 ± 0.10 and 1.05 ± 0.10 kg N m⁻³ d⁻¹, respectively) than PHB (0.76 ± 0.03 kg N m⁻³ d⁻¹) and reduced the cost per kilogram of NOx⁻-N removed by approximately 48% and 51%, respectively. Microbial analysis revealed substrate-specific communities, with Comamonadaceae dominating PHB biofilms and Pseudomonas pseudoalcaligenes predominating in SW and SA systems, demonstrating that substrate composition influenced denitrifying community structure. Although nitrite accumulation occurred with lipid-based substrates during laboratory operation, pilot-scale evaluation of SW as the sole carbon source in a commercial RAS demonstrated operational feasibility with negligible nitrite accumulation when operated alongside an aerobic biofilter. These findings demonstrate that engineered PHBC blends and lipid-based substrates can substantially reduce denitrification costs while maintaining effective nitrate removal, providing practical pathways toward simpler, more economical, and passive self-regulating SPD systems for RAS applications.

Date

10-12-2026

Committee Chair

Benton, Mike

LSU Acknowledgement

1

LSU Accessibility Acknowledgment

1

Download

Share

COinS