Phosphorus Sequestration in Biosolids, Nuisance Struvite Control via PAD and Chemical Addition to TH-AD Digestate, and Downstream Effects

Purpose: Struvite and other common precipitates have long posed persistent maintenance and operational challenges at wastewater treatment plants. These precipitates can be enhanced at plants that utilize biological phosphorus removal and can precipitate within anaerobic digesters and on final dewatering equipment which causes damage and decreases treatment efficiency. Struvite can be sequestered as nutrient rich biosolids if properly controlled to remain in the cake. In a digested solids storage tank (DSST) that is in between anaerobic digestion and final dewatering, precipitation can be controlled by manipulating variables such as mixing, aeration of digestate to achieve a shorter solids retention time (SRT) version of post aerobic digestion (PAD) and by chemical addition. Optimizing this via batch and pilot testing can control nuisance struvite and other crystal formation, enhance downstream effects such as increasing phosphorus in the cake, improve biosolids odor profile and cake dryness, reducing polymer demand, and achieve additional volatile solids reduction (VSR).
This research will review findings from batch and pilot scale tests and discuss the nature of these precipitates. Objectives include determining the mechanism that provides optimal phosphorus removal, odor control, cake characteristics, and analyzing the precipitated solids of interest. The presentation will discuss results across batch, pilot, and full-scales and examine the lessons learned of each.

Motivation and Background: The Atlantic Treatment Plant (ATP) in Virginia Beach, VA, operated by the Hampton Roads Sanitation District's (HRSD) is a high-rate facility with an A/O process configuration. In solids handling, a Cambi Thermal Hydrolysis Process (THP) is utilized prior to anaerobic digestion, leading to the formation of scaling precipitates and nuisance struvite. ATP is actively engaged in efforts to control these issues for optimal operational efficiency.
Precipitate formation is determined predominantly by the solids solubility product (Ksp) and the saturation in solution and can be controlled by pH [1]. Manipulation of pH via CO2 stripping from agitation and aeration has successfully been shown to control solubility reactions [2, 3]. Increasing the saturation index via chemical addition can also control precipitation reactions, decrease polymer demand, and improve cake dryness with an SRT of a few hours [4]. Likewise, PAD has been shown to achieve ammonia removal, additional VSR, and break down odor compounds formed in anaerobic digestion at a 5-6 day SRT [5]. Based on these findings, a pilot scale set up (Figure 1) will run at an SRT of 3 days to provide benefits from PAD, but with a longer reaction time that may be more beneficial for phosphorus sequestration. Various mixing, aeration rates, and chemical addition will be manipulated to determine the optimal operation conditions on phosphorus sequestration, cake characteristics, VSR and odors.

Pilot Design and Preliminary Operation: The pilot set up at ATP consists of four tanks. Each tank is 11 feet tall with a volume of about 63 gallons. Simulating the DSST, the tanks are operated as daily batch fed continuously stirred tank reactors (CSTRs) maintained by pump recirculation with a 3-day SRT. The tanks are aerated with fine bubble diffuser membranes which can be operated at constant or intermittent air flow rates, or via Dissolved Oxygen (DO) or pH set points. Preliminary operation has included mixing and aerating tank 1 at a constant airflow rate of 5 L/min, while recording online measurements of temperature, DO, and pH (Figure 2). At this operation, about 50% OP removal across the pilot has been measured with minimal NH3 removal due to lack of pH increase at this airflow rate. Future plans include tracking COD across the pilot, increasing aeration, and the addition of Ca(OH)2 or Mg(OH)2 at various dosages. Batch Testing Results: Prior to running pilot scale tests, some objectives can be mechanistically answered by preliminary batch testing. Chemical Addition: The following test was conducted on a Phipps and Bird PB-900 series Jar Tester at a continuous mixing speed of 300 rpm. Five jars of digestate consisting of a control with no chemical addition, two jars dosed with a carbide lime slurry at Ca2+ + Mg2+/P ratios of 0.5:1 and 1:1, and two jars dosed with thioguard, similarly, at Ca2+ + Mg2+/P ratios of 0.5:1 and 1:1. Chemical dose was injected as a spike at time zero of the test. Analytical measurement of OP-P and pH were determined daily via spectrophotometry (Phosphorus HACH TNTplus Vial Test UHR 2-20 mg-PO4-P/L) and pH bench meter, respectively) (Figure 3).
The pH of the digestate sample was about 7.4 and increased up to a maximum of 8.7 within 24 hours likely due to CO2 stripping from the mixing intensity. OP removal in the control condition with no chemical addition was measured as 68%. The addition of chemical helped to increase OP removal as high as 95%. The digestate pH increasing towards alkaline within the first test day showed the greatest impact on OP removal rate. The higher Ca2+ + Mg2+/P molar ratio tested of 1:1 for both the carbide lime slurry and thioguard had higher OP removal rates than those dosed at lower ratios of 0.5:1. Thioguard compared to the carbide lime slurry overall had a slightly more significant effect on OP removal. The pH decrease that is measured after 24 hours may be due to struvite's production of hydrogen ions in the formation reaction, however, future plans of repeat experiments as well as X-Ray Diffraction for precipitate analysis will further determine this.

Cake Characteristics: A separate test was completed to look at aeration effects on cake characteristics. A sample from the DSST was conditioned, dewatered, and analyzed to evaluate polymer demand and cake solids. Polymer was tested on two buckets of digestate - a gently mixed control bucket and a bucket aerated at about 5 L/min. After 72 hours, polymer was used to condition each digestate. After polymer was added to the solids, it was mixed, and a capillary suction time (CST) instrument was used to find the optimal dose of polymer. The solutions with the lowest CSTs were identified to be the optimal dose of polymer needed and were then dewatered by the centrifuge cup method and measured for cake solids and solids capture, per the Higgins method [6]. In Table 1, the Aerated bucket had a higher polymer optimal dose, TS simulated cake value, and TSS simulated filtrate value compared to the Control bucket while the Control bucket had a higher VS simulated cake value. From these results, the aerated digestate had poorer solids capture when dewatered with the polymer compared to the Control bucket. In addition to cake dryness, improving cake odors will be researched in the future by the characterization of odor profiles and gas sampling the reactor head space using Chromatography Mass Spectrometry solid phase microextraction and Gastec Tubes.

Conclusion: This paper will present the challenges that ATP faces as well as batch and pilot research conducted to combat these issues. Optimization learned on this scale will advance the discussion on similar issues shared by wastewater treatment plants across the country. Further advancements have the possibility to improve plant efficiency and save plant capital costs.