Lessons Learned from the Start-up of the TPAD Process at San Jose-Santa Clara RWF

Temperature Phased anaerobic digestion, thermophilic digestion (55°C, 131°F) followed by mesophilic anaerobic digestion (38°C, 98°F), was the process selected by the San Jose-Santa Clara RWF (SJ-SC) as the stabilization process for the overall facilities expansion and modernization projects, with the purpose of renewing facilities, increasing operating capacity and starting the transition of SJ-SC away from the use of sludge lagoons to manage biosolids. TPAD was identified as the preferred digestion alternative at SJ-SC as part of the 2010 Biosolids MTPAD was identified as the preferred digestion alternative at SJ-SC as part of the 2010 Biosolids Master Plan. The TPAD process was formed based on converting four existing anaerobic mesophilic digesters to serve as the thermophilic anaerobic digesters and connecting that system to the existing 12 mesophilic anaerobic digesters through a sludge cooling heat exchanger system.

One of the first questions the start-up team had to answer was how to seed the digesters. The regionally relevant operating thermophilic digesters within proximity to San Jose are at the EBMUD Facility in Oakland, CA and the Hyperion WWTP in Los Angeles, CA. Given the large volumes of sludge and haul distances and schedule impacts it was elected to start the digesters from a mesophilic seed sludge.
The implications of starting with a mesophilic seed sludge are important to understand as they impact the startup schedule and duration to process maturity. The primary challenge is that the biomass operating in the system are not representative of a thermophilic operation. Following seeding, the temperature was increased rapidly to the target operating temperature of 55°C (131°F), with no sludge feed to the system. With the temperature increase there was a significant increase in the volatile acids concentration in the thin seed sludge, increasing from 100 mg/L as acetate to approximately 417 mg/L as acetate, over the first 12 days of operation, after which the VFA levels were rapidly depleted, decreasing from 403 mg/L as acetate to 7.92 mg/L as acetate between days 14 and 17. This was indicative of the reestablishment of methanogenic activity in the digester, indicating that thermophilic tolerant methanogens are present in mesophilic sludge and feed sludge as well, suggesting using thermophilic seed sludge is unnecessary. Loading commenced around day 20 of operations, which coincided with a rapid increase in volatile acid levels within the digester, resulting a cessation of feeding on day 32. After a decrease on VFA levels loading recommenced, which lead to a further spike in VFA. However, rather than ceasing loading the loading rate was held relatively constant, until there was a sharp decrease in VFA levels around day 50. The rationale for this strategy was to provide sufficient substrate to generate the needed VFAs to maximize methanogenic growth rate, while not souring the digester. This strategy proved successful as, loading increase steadily to a maximum of 50,000 lb-VS/d for the single digester at start up, without further uncontrolled imbalance between VFA production and loading rate (Figure 1). Concurrent with onset of stable loading, biogas production stabilized, producing measurable levels within the digester (Figure 2). Based on these observations, it took approximately 40 days from first heating the digester until full thermophilic digestion was established and in balance (Figure 2).

Similar trends were observed in the subsequent 3 digesters started up. Additional evaluation of the data found that daily change in volatile acids was means understanding digester progress towards as stabilized system. Figure 3, plots the daily change in volatile acids the start-up for Digester 8, concurrent with the digester observed VFA concentration, as measured by titration. What is apparent is that during the population stabilization period days 1-40, daily VFA generation remained positive, with the exception of the initial drop prior to any loading. However, once loading commenced the VFA generation (change VFA mass per day) remained above zero. However, with the onset of methanogenesis (~ day 50) increasing negative generation days were observed along with zero days. With the maturation of the biomass both structurally and in concentration it is expected that the VFA generation rate will converge around zero. Utilizing the daily change in VFA maybe an effective indicator of process maturation, both from an onset of methanogenesis and longer-term stabilization of the process.

Since commissioning the digestion process has provided significant additional operational design related lessons learned. Overall, since conversion to TPAD the volatile solids destruction has increased from a typical value of 30-40% to 55-60%, resulting in less solids and more biogas (Figure 4). Another critical observation was in the cooling heat exchangers, where there was significant accumulation of struvite, that impacted the ability of the system to cool the sludge in the mesophilic phase (Figure 5). Interestingly, mesophilic digester performance was not negatively impacted, even while operating at ~115°F, suggesting for future TPAD systems cooling all the way down to conventional mesophilic temperatures may not be necessary.
This paper benefits operators and design engineers alike will gain valuable insight into the TPAD process operations and benefits.