National Academies Press: OpenBook
« Previous: 1 Background
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 10
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 11
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 12
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 13
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 14
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 15
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 16
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 17
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 18
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 19
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 20
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 21
Suggested Citation:"2 The PCAPP Biotreatment System." National Research Council. 2013. Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant. Washington, DC: The National Academies Press. doi: 10.17226/13494.
×
Page 22

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

2 The PCAPP Biotreatment System GENERAL DESCRIPTION OF THE BIOREACTOR SYSTEM TDG mineralized) can lower the pH in the bioreactor and inhibit the bacteria. Therefore, caustic addition (NaOH) Immobilized cell bioreactors (ICBs) will be used to biode- may be required, especially during start-up, to maintain the grade the thiodiglycol (TDG) and other organic compounds pH between 7 to 8 in the bioreactor. During normal opera- produced by the hydrolysis of the mustard agent (HD or HT) tions with an acclimated biomass, both the acid addition to in the munitions currently stored at Pueblo Chemical Depot. the feed tank and caustic addition to the ICB units may not Honeywell International, Inc., is the technology provider for be needed. The inorganic nutrients urea and diammonium the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP) phosphate (DAP) are also added to meet the stoichiometric ICB biotreatment system, which includes equipment for pro- nitrogen and phosphorus requirements for biodegradation. cessing the feed, off-gas, and effluent from the bioreactors as As mentioned above, a total of 16 ICB units will be well as recycling pumps and storage for the feed and effluent, installed in 4 parallel modules with 4 units per module. as shown in Figure 2-1. Each unit, holding approximately 42,000 gal, contains three The ICBs at PCAPP are aerobic attached-growth bio­ chambers in series (21,000 gal, 10,500 gal, and 10,500 gal). reactors in which bacteria grow and are immobilized on a Most of the degradation and biomass growth is expected support medium made of plastic and an elastomer foam. to occur in the first chamber, which hydraulically operates The overall bioreactor system consists of 4 parallel mod- similarly to a continuous-flow stirred tank reactor (CSTR) ules, each of which contains 4 parallel units, for a total of from the high internal recycle flow rate and aeration. The 16 parallel units. Figure 2-2 shows two of the four modules; remaining two sequential chambers also operate in the mode Figure 2-3 identifies major components of a module. The of CSTRs. They provide additional contact time to improve principal target compound for treatment in the ICBs is the effluent quality and biotreatment capacity and to mitigate Schedule 2 compound TDG. Each ICB unit is expected to against potential overload or temporary upsets. The ICBs are handle 485 lb TDG/day (57.9 lb TDG/1,000 gal diluted expected to remove at least 95 percent of the influent TDG hydrolysate/day) and achieve TDG removal efficiencies and 85 percent of the chemical oxygen demand (COD), and greater than or equal to either 95 percent at the maximum each unit is designed for a maximum flow rate of diluted design flow rate or 98 percent at 50 to 90 percent of the hydrolysate of 9,700 gpd (BPT, 2006a). The flow rate to each maximum design flow rate.1 ICB unit is expected to vary between approximately 4,800 The pH of the incoming hydrolysate feed is expected to and 9,700 gpd. At these flow rates, the average hydraulic fall between 10 and 13, so the process design includes the retention time in each unit is approximately 8.6 to 4.3 days, ability to add acid to the feed tank to provide neutraliza- respectively, to allow sufficient contact time with bacteria. tion if the high pH proves inhibitory to the biomass. TDG The ICB feed requires dilution (normally 1 part hydro- is expected to be metabolized in the ICBs, mainly to CO2 lysate to 7 parts process water) to achieve desired, sub-­ and sulfuric acid (H2SO4), by aerobic bacteria, resulting in inhibitory influent concentrations of TDG. Even when biomass growth (sludge production) of about 150-200 lb/ diluted, the hydrolysate has a relatively high concentration day per unit.2 The generated sulfuric acid (0.8 g-H2SO4/g- of (1) TDG (about 7,000 mg/L, or 2,730 mg/L as TOC), (2) other organics (estimated at 570 mg/L as TOC in addi- 1George Lecakes, Chief Scientist, PCAPP, “PCAPP’s Water Recovery tion to the TDG), and (3) reduced inorganic compounds. The System and Brine Reduction System Briefing,” presentation to the com- combination of all three categories of compounds exerts a mittee, May 1, 2012. 2Ibid. significant oxygen demand (15,000 mg/L as COD) (BPT, 10

THE PCAPP BIOTREATMENT SYSTEM 11 FIGURE 2-1  Conceptual diagram for the 16 immobilized cell bioreactor (ICB) units during planned normal operation. NOTE: GAC, granular activated carbon. SOURCE: Adapted from George Lecakes, Chief Scientist, PCAPP, “PCAPP’s Water Recovery System and Brine Reduction Figure 2-1 and S-1 System Briefing,” presentation to the committee, May 1, 2012. Bitmapped FIGURE 2-2  Two of the four biotreatment modules at PCAPP, each of which has four immobilized cell bioreactor units. SOURCE: Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design and Operating Conditions for PCAPP’s Biotreatment Process,” presenta- tion to the committee, November 28, 2012. 2006a). Plans are for this oxygen demand to be met by aera- or if the blower air stops flowing to the bioreactors. The tion from coarse-bubble air diffusers (powered by blowers) system was designed according to applicable standards.3 at the bottom of the reactors. 3Among The system includes local programmable logic controllers the standards involved are those from the following organiza- that are data-linked to the facility control system for remote tions: The American National Standards Institute/ Instrumentation, Systems, and Automation Society; the American Petroleum Institute; the Air Condi- monitoring and status and alarm indication. The alarms are tioning and Refrigeration Institute; the American Society of Mechanical set to alert if various parameters, such as pressure, pH, dis- Engineers; the American Society for Testing Materials; the American Weld- solved oxygen, and temperature, fall outside normal ranges ing Society; the Hydraulic Institute; the National Electrical Manufacturers

12 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP FIGURE 2-3  Major components of one of four biotreatment modules at PCAPP. SOURCE: Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design and Operating Conditions for PCAPP’s Biotreatment Process,” presentation to the committee, November 28, 2012. The effluent from the ICB will be sent directly to the water of the biomass without the need for separate biomass recov- recovery and brine reduction system (WRS-BRS), which is ery devices and biomass recycle (Rittmann and McCarty, expected to recover and recycle 80 percent of the ICB efflu- 2001; Metcalf & Eddy, Inc., et al., 2003). A relatively high ent and other streams as process water. The bioreactors will concentration of active biomass can be maintained within also potentially generate vapors containing toxic volatile the ICB for longer solids retention times in comparison to organic compounds (VOCs), which will be collected and suspended-growth biotreatment processes. The longer solids routed to an off-gas treatment system containing a granular retention times also facilitate greater biomass decay with activated carbon (GAC) adsorber (see Figure 2-1). resulting lower amounts of sludge production. The immo- bilized biomass provides greater resistance to toxic shocks and fluctuations in organic loading than suspended growth COMMITTEE’S REVIEW OF THE SYSTEM processes such as sequencing batch reactors (SBRs). One The committee’s understanding of the principal operating final noteworthy advantage of the ICB system that has been and hydrolysate characteristics pertinent to the ICB units is designed for use at PCAPP is the oxygen supply system that given in Table 2-1. These values are used in the calculations utilizes a duckbill coarse bubble diffuser and is therefore presented in this chapter. All calculations are based on a unlikely to clog. single ICB unit unless stated otherwise. In early studies in the mid- to late 1990s, some testing had Operation of the ICBs relies on biomass that grows on been performed using SBRs instead of ICBs. Some of these a solid layer or support medium and forms a biofilm. This studies will be discussed later in this chapter. These studies immobilization permits excellent retention and accumulation were performed to address the planned use of biotreatment to process the mustard agent hydrolysate produced from the bulk stocks of agent stored at the Aberdeen, Maryland, site. Association; the National Fire Protection Association; and the Tubular Exchanger Manufacturers Association (BPT, 2006a). The Aberdeen Chemical Agent Disposal Facility (ABCDF)

THE PCAPP BIOTREATMENT SYSTEM 13 TABLE 2-1  Key Operating and Feed Characteristics for the Immobilized Cell Bioreactor Units Characteristic Hydrolysate (total)a Process Water (total)a ICB Influent (total)a ICB Influent (per unit) Flow (gpd) 16,766 117,362 134,128 8,383 Hydraulic retention time (days) 4.98b Volume (gal) 41,783 Concentration of TDG (mg/L) 56,000 0 7,000 7,000 Concentration of TSS (mg/L) 8,000 0 1,000 1,000 Concentration of COD (mg/L) 120,000 0 15,000 15,000 Concentration of TOC (mg/L) 26,400 0 3,300 3,300 Concentration of iron (mg/L) 2,160 0 270 270 Concentration of NaCl (mg/L) 57,600 0 7,200 7,200 pH 10-13 7-8c aThis is the total to all 16 units. bThis was calculated from the flow and volume shown. c This will be maintained via acid production within the unit and caustic or acid addition as needed. SOURCE: Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design and Operating Conditions for PCAPP’s Biotreatment Process,” presentation to the committee, November 28, 2012; personal communication from Paul Usinowicz to the committee, November 29, 2012; and BPT (2006a). did subsequently process the HD mustard agent into hydro- TDG. SBR laboratory studies demonstrated that TDG was lysate. This hydrolysate, however, was ultimately shipped for inhibitory at concentrations greater than 2,000 mg/L (or processing at Dupont’s million-gallon biotreatment Secure approximately 2,000 ppm).4 These studies also indicated Environmental Treatment facility in Deepwater, New Jersey, that hydrolyzed heel material could be inhibitory.5 Based rather than treated biologically onsite at the ABCDF. Onsite on these studies, recommendations had been that operators treatment is being required by the State of Colorado for the should “maintain cell operations so that TDG concentrations hydrolysate that will be produced at PCAPP. As part of the in all cells are maintained at less than 2,000 ppm TDG during ACWA program planning for PCAPP, studies conducted operations”(BPT, 2005). beginning in the early 2000s switched from the use of SBRs Once at steady state, biodegradation and dilution of to ICBs. Initially, it was planned that the agent hydrolysate the influent hydrolysate should maintain concentrations would be mixed with energetics hydrolysate produced from of TDG below 700 mg/L in the first chamber of each unit the tetrytol and M8 propellant associated with the projectiles (the design influent TDG concentration in the hydrolysate and mortars stored at Pueblo Chemical Depot. Following is 7,000 mg/L; see Table 2-1). Therefore, under normal cost-cutting initiatives and regulatory and public acceptance, operating conditions, the TDG concentration in the units the energetic materials will now be shipped off-site for pro- will not exceed 2,000 mg/L. During periods of off-normal cessing, and only the agent will be hydrolyzed and subse- operation and start-up, however, care should be taken to oper- quently biodegraded at PCAPP. ate below the recommended maximum TDG concentration (i.e., 2,000 mg/L). Indeed, the PCAPP design influent TDG Finding 2-1. The committee supports the selection of a bio- concentration compares favorably to TDG concentrations film process for the biotreatment system at PCAPP. In com- used in SBR experiments of approximately 8,500 mg/L, parison to a suspended-growth process, ICBs have greater 12,700 mg/L, and 13,750 mg/L in which no apparent toxic- resiliency during fluctuating loading, faster recovery from ity or inhibition of the biomass was observed (Harvey et al., off-normal conditions (e.g., toxic inhibition and excursions 1997; SBR Technologies, Inc., 1996; Harvey et al., 1996). in pH and temperature), and longer biomass residence time. 4James P. Earley, Principal Engineer and PCAPP Task Manager, SAIC, Complexity of the Influent Stream and Related Toxicity/ “Review of Biotreatment Testing for the Pueblo Chemical Agent Destruction Inhibition Pilot Plant (PCAPP) Project,” presentation to the Committee on Chemical Demilitarization, September 15, 2011, National Research Council, Wash- Toxicity Impacts on Biodegradation ington, D.C. 5Ibid. Also, “heel material” refers to mustard agent that over time in stor- Satisfactory operation of the ICBs at PCAPP will hinge age has degraded to a semi-solid or solid material in the agent cavities of on the ability of the bacteria to adapt to and tolerate initially munitions. This material generally is removed from the munitions by high pressure water jets. Sulfonium ion species comprise a large percentage of high, and potentially inhibitory, influent concentrations of this material.

14 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP With respect to toxicity testing, an SBR study observed inhibitory if they are not removed in the WRS-BRS system toxicity at loadings of 0.2 g TOC/g of mixed-liquor sus- (via activated carbon or the evaporator/crystallizer). Indeed, pended solids (MLSS)/day (SBR Technologies, Inc., 1996). of the incoming TOC, only approximately 83 percent is TDG The committee compared this value to the expected ICB (Table 2-1)8; not all of the remaining TOC in the influent is design loading at the PCAPP site using the equation below. characterized. The organic mass in units of “g TOC” (SBR Technologies, If urea and DAP are added as nutrients, as currently Inc., 1996) were converted to “g TDG” using two methods: planned, nitrification (the microbial oxidation of ammonia to (1) the theoretical TOC:TDG ratio of 0.39 and (2) averag- nitrate) may occur in the second and third chambers where ing the TOC:TDG ratios provided in the report6 to generate the biodegradable organic concentration may be sufficiently a TOC:TDG ratio of 0.45. After conversion, the threshold low. Because nitrifiers are more susceptible to inhibition by for inhibition or toxicity was calculated to be 0.44 to 0.51 g toxicants than heterotrophs (the organic carbon-degrading TDG/g MLSS/day. organisms expected to dominate in the ICBs), an interruption To determine how the expected design loading (EDL) for in nitrification may provide an early warning sign for the the ICBs at PCAPP compares to the range at which toxicity presence of toxic compounds. Nitrification was reported dur- or inhibition might be expected, the committee performed ing several phases of pilot-scale operation in a 75-gal SBR the following calculation: (SBR Technologies, Inc., 1998). By monitoring ammonium, nitrite, and nitrate leaving the ICB units, PCAPP staff may (g TDG/L) × (gal influent/day) × (L/gal) × be able to respond more quickly to toxicity. (0.7 g VSS/g TSS) EDL = (g TDG degraded/L) × (Yobs) × (gal influent/day) × Inhibition by Heavy Metals (L/gal) × (average SRT) Table 2-2 shows the characterization of HD mustard where: agent hydrolysate used in ACWA biodegradation testing in • VSS is volatile suspended solids and represents the 2003 (Guelta and Fazekas-Carey, 2003). The hydrolysate suspended organic matter; produced came from the hydrolysis of drained liquid agent • TSS is total suspended solids and includes both and the solid heel material in 4.2-inch HD mortar rounds organic and inorganic matter that is suspended; stored at Pueblo Chemical Depot. Consequently, the heavy • The unit flow and TDG concentration given in Table metals concentrations given in Table 2-2 can be considered 2-1 was assumed; representative9 of what will be encountered when PCAPP • A 50-day solids retention time (SRT) was assumed begins processing. Bacteriostatic effects exerted by high in the ICBs; TDG concentrations could be exacerbated by the presence • An observed biomass growth yield (Yobs) of 0.036 g of some heavy metals with antimicrobial properties (e.g., Ag VSS/g TDG was used7; and and Cu). Whereas such metals can be inhibitory and even • 90 percent of the influent TDG was assumed to be bacteriocidal in the low milligrams/liter concentration range, biodegraded. there are several factors that would mitigate their toxicity to bacteria. First, sorption to precipitated iron is a likely fate This yielded an expected design loading in the PCAPP for many of the heavy metals. The presence of salt (NaCl) ICBs of 0.43 g TDG/g TSS/day. As discussed earlier, this at high concentrations (e.g., 7,200 mg/L NaCl in the diluted is slightly below the range of values expected to be toxic/ influent; see Table 2-1) would promote coagulation of iron inhibitory (0.44 to 0.51 g TDG/g MLSS/day) from a previ- precipitates and sorbed metals, thus removing the metals ous study (SBR Technologies, Inc., 1996). Again, with the from solution and thereby reducing their bioavailability expected dilution in the units operated as CSTRs, this may and toxicity. Common ligands, such as chloride, phosphate, not be a problem, but during off-normal or start-up periods, sulfide, and organic matter, may also associate with bacte- the biomass in the ICBs may be inhibited or killed by the riocidal metals (e.g., Ag) to form dissolved complexes that influent TDG. have lower toxicity than the free metal ions (Xiu et al., 2011). An additional toxicity issue of concern is the presence of uncharacterized compounds in the hydrolysate feed that may 8One gram of TDG is equivalent to 0.39 g TOC. buildup in the recycled process water. At hydraulic residence 9Variations in heavy metal concentrations may result from differences in times in the range of 4.3 to 8.6 days, COD degradation inter- lots during manufacture, long-term storage conditions, degree of agent deg- mediates may also build up in the recycled water and become radation, and interactions with metal surfaces in the agent storage cavity of munitions. Since no comprehensive sampling of heavy metals constituents 6See SBR Technologies, Inc. (1996), pp. 11, 27, and 28. for all of the approximately 780,000 munitions stored at Pueblo Chemical 7James P. Earley, Principal Engineer and PCAPP Task Manager, SAIC, Depot has been done, the committee had to rely on the information avail- “PCAPP Biotreatment System Update,” presentation to the Committee on able from prior test samples. The values given in Table 2-2 were the most Chemical Demilitarization, September 15, 2011, National Research Coun- complete analysis seen over the course of the committee’s research in terms cil, Washington, D.C. of heavy metals composition.

THE PCAPP BIOTREATMENT SYSTEM 15 TABLE 2-2  HD Hydrolysate Characterization from 2003 would rapidly oxidize any sulfide formed and released from Biotreatment Testing the biofilm back to sulfate in the bioreactor fluid; therefore, Concentration (mg/L) it will be difficult to determine if sulfide toxicity is occurring, Constituent (unless otherwise indicated) because it would occur within the biofilm itself. Thiodiglycol (TDG) 17,537 One option for mitigating potential sulfidogenic con- Dithane 2,093 ditions in the biofilm is to add nitrate as a supplemental Thiox 47.9 electron acceptor. If there is insufficient dissolved oxygen Chemical oxygen demand COD) 43,100 to keep the biofilm entirely aerobic, nitrate can readily dif- Total organic carbon (TOC) 8,120 fuse into the biofilm and, as a preferential electron acceptor Percent TOC as TDG 84.9% over sulfate, prevent the development of sulfidogenic con- COD:TOC ratio 5.31 (ratio) Sulfate 84 ditions. For this operational strategy, the ammonium and/ Sulfur 6,010 or urea in the nutrient supply should be replaced with an Total dissolved solids (TSS) 28,000 adequate quantity of a nitrate salt to provide nitrogen both Total suspended solids (TSS) 1,000 for biomass synthesis and as a favorable electron acceptor. pH 13 (pH) No information was found in the literature on the biodegrad- Specific gravity 1.03 g/mL ability of TDG under nitrate-reducing conditions, although it Aluminum 1.99 Arsenic 0.579 seems likely; other glycols (ethylene and propylene) readily Barium 0.033 undergo mineralization under nitrate-reducing conditions Cadmium 3.2 (Klotzbücher et al., 2007). Calcium 10.9 Chloride 10,800 Finding 2-2. Toxicity of TDG is unlikely to cause system Copper 0.281 upset during normal ICB operations at PCAPP. Under Iron 520 Lead 3.69 p ­ eriods of off-normal operation and start-up, however, TDG Magnesium 5.74 could be inhibitory or toxic to the biomass in the ICBs. Mercury 0.013 Molybdenum 0.065 Finding 2-3. Uncharacterized components in the wastewater Nickel 0.330 feed to the ICBs at PCAPP could be inhibitory, and these Phosphorus 0.456 Potassium 15.2 compounds may build up over time if not removed in the water Silver 5.73 recovery and brine reduction systems as the water is recycled. Sodium 10,630 Zinc 3.59 Finding 2-4. Although there are heavy metals present in the SOURCE: Guelta and Fazekas-Carey (2003). feed to the ICBs, the committee believes that heavy metals toxicity will not be a major concern for the biotreatment process at PCAPP. However, among factors that could be explored regarding poor ICB performance, especially dur- Furthermore, some dissolved metal removal by biosorption ing off-normal conditions, would be heavy metals toxicity. is likely to occur. Finding 2-5. Sulfide formation can be minimized by ensur- Inhibition by Sulfide ing sufficient aeration and dissolved oxygen penetration throughout the biofilm. This can hinder the development of Another inhibitory inorganic species is sulfide (likely anaerobic niches. present as H2S and HS–), which can be toxic to bacteria at concentrations at or above 100 mg/L. Unlike the heavy Finding 2-6. Inclusion of nitrate as a supplemental electron m ­ etals that are present in the influent to the ICBs, sulfide acceptor in the feed to the PCAPP ICBs likely will mitigate could be produced in the bioreactor by the reduction of sul- the development of sulfidogenic conditions in areas of the fate if anaerobic zones within the biofilm develop. Up to 0.79 biofilm that are deficient in oxygen. g of sulfate can be produced from the biological oxidation of 1 g of TDG, and with 98 percent removal of 7,000 mg/L of Recommendation 2-1. Care should be taken at PCAPP TDG in the influent, up to 5,400 mg/L sulfate would form. during ICB start-up and off-normal periods to avoid toxic- Given the high oxygen demand exerted by the hydrolysate, ity/inhibition by reducing mass hydrolysate loading (e.g., anaerobic micro-niches may develop within the biofilms increasing the hydraulic retention time or further diluting the where sulfate would be reduced to sulfide (referred to as hydrolysate) or operating in batch mode by discontinuing sulfidogenic conditions). Indeed, the potential formation of the feed until the ICB unit has recovered. sulfide is acknowledged in the reports reviewed by the com- mittee (BPT, 2006a). Dissolved oxygen in the bulk liquid

16 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP Recommendation 2-2. Potentially inhibitory organic com- at warmer temperatures that enhance microbial growth, pH pounds in the process water recycle should be monitored values outside of the 7 to 8 range may be better tolerated. at PCAPP. Operators should be aware that if such products In bench-scale tests with ICBs designed and operated to buildup in the WRS-BRS system, toxicity/inhibition of the simulate the full-scale units, the Bechtel Pueblo Team (2005) biomass in the ICBs may result. found that pH levels above 10 led to significant foaming, and performance was hampered for a longer period. Recovery Recommendation 2-3. If TDG removal or oxygen consump- of the ICB units was possible after 1 to 2 days following tion begins to decline, the concentration of metals should be short-term (i.e., 1- to 2-hour) excursions in pH above 10. The analyzed in the influent to the ICBs at PCAPP and within Bechtel Pueblo Team recommended that influent pH should the ICBs themselves to determine if the metals might be be lowered below 10 prior to the first ICB cell to prevent contributing to inhibition. unacceptably high pH conditions from occurring. The milligrams of oxygen consumed per gram of MLSS • Oxygen consumption should be monitored by mea- per hour was measured with MLSS from an SBR operated suring oxygen in the off-gas as the ICBs are sealed with HT hydrolysate over a pH range of 6.5 to 9.0 (Harvey and the off-gas will be sent through granular activated et al., 1997). This test, called the specific oxygen uptake carbon. rate (SOUR), is an indicator of the activity of the biomass • If dissolved metals reach inhibitory concentrations, and oxygen demand rates. Over several tests, the SOUR measures to remove them from the influent (e.g., increased when the pH increased from 6.5 to 7, but it precipitation through pH adjustment and addition decreased above pH 8. These results are consistent with the of coagulants in the holding tanks immediately intent of operating the ICB units at pH 7 to 8. upstream of the ICBs) should be considered. Caustic is a low-cost and convenient method for deliv- ering alkalinity. However, sodium bicarbonate gave more Recommendation 2-4. With the planned use of urea and precise pH control than caustic in a laboratory-scale ICB diammonium phosphate to supply nutrients, nitrification study (Guelta and DeFrank, 1998). activity should be monitored at PCAPP based on ammonia, nitrite, and nitrate levels in the ICB influent and effluent. Finding 2-7. To optimize TDG removal, pH control in the Inhibition of nitrification could alert operators to the pres- range of 7 to 8 is essential for the PCAPP ICBs. ence of increasing concentrations of toxic compounds. Recommendation 2-6. PCAPP should consider using Recommendation 2-5. Careful attention should be given to b ­ icarbonate rather than caustic (NaOH) for pH control in the control sulfide formation. ICBs, especially during start-up, because this strategy was shown to provide more precise pH control. This may be accomplished by maintaining aerobic condi- tions within the biofilm or by adding nitrate to the feed to Temperature Impacts on Biodegradation the PCAPP ICBs. Temperatures in Pueblo, Colorado, vary throughout the year. Biological processes are significantly impacted by pH Impacts on Biodegradation temperature, which could lead to decreased TDG removal Control of pH will be critical to the successful operation efficiency or even reactor failure during periods of extreme of the ICB units. The generation of sulfuric acid during the temperature in summer and winter. This was noted previ- biodegradation of TDG makes this especially challenging. ously in the NRC report Interim Design Assessment for The target range for proper operation is pH = 7 to 8.10 During the Pueblo Chemical Agent Destruction Pilot Plant (NRC, pilot testing with the 1,000-gallon ACWA unit, a significant 2005, p. 21): decrease in performance (i.e., less than 90 percent removal of TDG) occurred from insufficient pH control and unsea- The ICBs can operate in temperatures ranging from 95°F to sonably low temperatures (Guelta et al., 2002). During this 41°F, but ambient temperatures in Pueblo, Colorado, range period, the pH was 5.0 to 6.5, indicating the need for pH from 115°F to −20°F. Therefore, the design must provide neutralization. Nevertheless, the interactive impact of low appropriate cooling and heating of air and water fed to the ICBs to ensure optimum operating temperatures. pH and low temperatures (55°F and 65°F) was not evalu- ated in these studies (Guelta et al., 2002); it is possible that An SBR study was specifically performed to investigate the stability of performance (TDG and TOC removal) over a range of operating temperatures (SBR Technologies, Inc., 10Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design 1996). Stable performance was observed at temperatures and Operat­ing Conditions for PCAPP’s Biotreatment Process,” presentation from 8 to 35°C (46 to 95°F); the reactors were said to “fail,” to the committee, November 28, 2012.

THE PCAPP BIOTREATMENT SYSTEM 17 however, when the temperatures reached 35 to 40°C (95 to Carey, 2003; Guelta, 2006). These studies also reported aver- 104°F) (SBR Technologies, Inc., 1996). age feed VSS concentrations ranging from 30 to 200 mg/L. Design of the ICBs takes into account the potential for Thus, approximately 50 to 75 percent of the suspended low temperatures during winter operation (Golder Associates solids is expected to be inorganic. Indications are that most Inc., 2010). To maintain the reactors at 80°F, steam will be of the inorganic suspended solids contain iron. For example, injected at up to 580 lb/hr/reactor. However, no information analysis of unfiltered feed to pilot-scale ICBs measured iron was found in the calculations for the possibility of the influ- concentrations of 130 to 400 mg/L, while filtered samples ent becoming too warm. Consequently, there appears to be contained 1 to 1.6 mg/L. Effluent iron concentrations from no provision for cooling the influent. Since the intent is to these systems ranged from 2 to 28 mg/L, indicating that iron operate the reactors in the mesophilic range, sustained influ- solids were building up in these systems. ent temperatures above the range 95 to 104°F may result in Inorganic solids can also form during treatment, espe- a deterioration in TDG removal. cially considering the high concentration of iron in the feed. Bench-scale tests indicated that iron oxide precipitated Finding 2-8. Ambient temperatures in Pueblo, Colorado, during treatment (BPT, 2005). While it may be difficult to have ranged from −31° to 109°F. Biological degradation determine whether iron oxides precipitate (e.g., FeOOH, is sensitive to temperature, and, based on data provided to Fe2O3, Fe(OH)3) during treatment, it is clear that these the committee, performance may suffer or the process may oxides were present in the pilot ICBs, probably entering fail completely if the PCAPP ICBs reach temperatures over with the feed. In addition, given the relatively high bio- approximately 100°F for sustained periods. chemical oxygen demand of the feed and the likely devel- opment of anaerobic conditions in the deeper zones of the Recommendation 2-7. The following activities are recom- biofilms, the committee cannot rule out the possibility that mended for the PCAPP biotreatment system to address local sulfate would be biologically reduced to sulfide, which in annual ambient temperature extremes: turn could precipitate with iron (e.g., FeS). It is also likely • A heat balance calculation should be performed to that iron phosphate (FePO4) will form from the addition anticipate operating temperatures in the ICB units. of DAP as a nutrient. One pilot study reported addition of • During systemization, the temperature ranges during 44 mg/L of phosphorus (from DAP) resulting in an ICB summer and winter periods that might be routinely effluent concentration of about 4 mg/L phosphorus (SBR expected in the ICB units should be determined by Technologies, Inc., 1998). The concentration of phosphorus testing. required for the biomass growth on TDG is approximately • If needed, PCAPP staff should make provisions for 5 mg/L. This amount of phosphorus is derived from using cooling the ICB units or feed flow. a bacterial yield of 0.036 mg VSS/mg TDG and 98 per- cent removal of 7,000 mg/L TDG (see Table 2-1) and by assuming that the biomass contains 2 percent phosphorus Solids Buildup Concerns by weight (Metcalf & Eddy, Inc., et al., 2003). Thus, Solids buildup in the ICBs could have an adverse impact approximately 35 mg/L phosphorus would be available for on long-term performance. Indeed, clogging and impedance precipitation as FePO4. This could form approximately 11.9 of wastewater flow through the biofilm-supporting media lb/day per unit of FePO4 solids (assuming a flow of 8,383 could be conducive to channeling, reduced contact with the gpd; see Table 2-1). Based on the estimates provided above, biofilm, and decreased treatment efficiency. In addition, if the production of solids within the ICB units likely ranges large quantities of biomass build up in the system, periodic from about 35 to 60 lb/day (approximate concentration in large sloughing events could cause excess biomass loss, 8,383 gpd is 500 to 860 mg/L). thereby causing reduced treatment efficiency. Three sources PCAPP plans to add DAP to the feed tank, which could of solids are (1) influent suspended solids; (2) precipitates result in more FePO4 precipitation due to the combined fac- from the high iron content of the diluted hydrolysate; and tors of feed tank residence time (10 to 30 days) and increased (3) biomass formed during treatment. Excess solids build- iron availability. If, however, DAP was added in the ICB up has been reported as a potential problem in several of the recycle line, less contact between iron and phosphorus might pilot studies conducted (Earley et al., 2003; Guelta, 2006). occur, decreasing the amount of FePO4 precipitation. The design feed TSS concentration to the ICBs is Since the ICBs use biofilm processes, excessive biomass 1,000 mg/L with a design flow of 8,383 gpd (see Table 2-1); could build up on the elastomer foam. As mentioned above, therefore, the solids load to each unit of 70 lb TSS/day could such accumulation could cause anaerobiosis; it could also cause significant clogging. Data from pilot studies indicate cause clogging, channeling, and biomass sloughing. Bio- that feed TSS varied widely, from 100 mg/L to 1,000 mg/L, mass production rates can be estimated from the observed and average values fed to pilot ICB reactors were between bacterial yield (Yobs). Several values for Yobs using different 100 and 400 mg/L (Earley et al., 2003; Guelta and Fazekas- units were reported from the various pilot-scale studies.

18 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP These include 0.036 g VSS/g TDG removed,11 0.16 g VSS/g Finding 2-10. Sloughing and solids release events are likely TOC removed, and 0.293 g TSS/g TOC removed (Guelta et to be intermittent for the PCAPP ICBs. Thus, large quantities al., 2001); 0.06 to 0.15 g VSS/g biological oxygen demand of solids might be released occasionally. (BOD) removed (Golder Associates Inc., 2009); and 0.07 to 0.15 g VSS/g BOD5 removed (Nurdogan et al., 2012). Using Recommendation 2-8. Once operational, PCAPP should Yobs values based on TDG and TOC, along with the data perform ongoing solids balance calculations for the ICBs, given in Table 2-1, and assuming 3,300 mg/L TOC in the using influent and effluent solids and influent solution feed,12 98 percent removal of TDG, and 80 percent removal chemistry to estimate the precipitates that are likely to form of TOC,13 estimates of biomass production (as VSS) and and the amount of biomass produced. This should be used total suspended solids (TSS) production range from 17 to to anticipate issues related to solids buildup and to assist 30 lb VSS/day and 26 to 54 TSS lb/day, respectively, for with preventive maintenance scheduling, such as flushing/­ each ICB unit. scouring or even system shut-down and solids removal. While it is not possible to directly combine the three types of solids described above, it is clear that the amount Recommendation 2-9. Composite samples should be taken of all three (influent solids, solids precipitated during treat- because the release of solids from the ICBs at PCAPP can ment, and biomass production) have the same order of mag- be intermittent. These composite samples should be used to nitude, and their total production is about 100 to 150 lb TSS/ characterize the effluent and perform mass balances of solids. day per unit. These values are close to the design estimated total biomass production of 150 to 200 lb/day per unit.14 In Recommendation 2-10. The preliminary quantification by a flow of 8,383 gpd, this represents concentrations of 1,500 the committee points to likely problems with excess solids. to 2,200 mg/L TSS. These levels could cause clogging and PCAPP staff should consider using a settling tank to remove associated performance deterioration. This issue must be total suspended solids from the influent to the ICB units. addressed through discussions with the technology provider and potential trial-and-error operation to determine the Recommendation 2-11. If solids accumulation and phos- optimal biomass sloughing schedule, rotation between units phorus nutrient availability become a problem, PCAPP staff to avoid too much mass being sent to the evaporator/crystal- should consider moving the diammonium phosphate (DAP) lizer, and intensity of scour needed. If necessary, packing feed point from the feed tank to the recycle line within each media could be removed from the units to enhance fluidiza- unit to avoid excessive FePO4 precipitation and DAP use. tion and sloughing. There is also a plan to remove solids from the bottom of the reactor if they accumulate; nevertheless, Recommendation 2-12. If the ICB technology provider this may be difficult to perform, considering the flat floor of cannot provide adequate information regarding biomass the reactors and the relatively small access port.15 accumulation and sloughing expectations with the waste that will be processed through the ICBs at PCAPP, either Finding 2-9. The hydrolysate produced at PCAPP will have the technology provider or the PCAPP staff should consider high influent total suspended solids concentrations (up to testing an ICB unit during systemization for 3 to 6 months 1,000 mg/L); therefore, solids may build up in the ICBs. with a benign and common substrate, such as molasses (fed Precipitation during treatment may also produce inorganic at the same chemical oxygen demand concentration as the solids. Furthermore, excessive biomass may build up within hydrolysate), to determine likely biomass accumulation the packing of the ICB units. Solids accumulation may cause and sloughing issues. Time should be provided to allow the changes in the hydraulic characteristics of the units, poten- biomass to dry fully and be flushed from the system before tially causing clogging and reduced contact with the biofilm start-up with hydrolysate. that decreases TDG removal efficiency. Oxygen Demand and Flux Issues Air will be supplied to the ICBs via two 3,600-scfm 11James P. Earley, Principal Engineer and PCAPP Task Manager, SAIC, b ­ lowers per module, equivalent to 1,800 scfm per ICB tank. “PCAPP Biotreatment System Update,” presentation to the Committee on The air will be distributed via 40 coarse air diffusers in the Chemical Demilitarization, September 15, 2011, National Research Coun- first chamber and 20 each in the second and third chambers.16 cil, Washington, D.C. The basis for this design is a set of calculations performed by 12Ibid. 13Ibid. Tideflex Technologies using a Diffused Aeration Modeling 14George Lecakes, Chief Scientist, PCAPP, “PCAPP’s Water Recovery Program (version 02-20-2009; Tideflex Technologies, 2010). System and Brine Reduction System Briefing,” presentation to the com- mittee, May 1, 2012. 15Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design 16Paul Usinowicz, Water/Wastewater Research Leader, Battelle, “Design and Operat­ing Conditions for PCAPP’s Biotreatment Process,” presentation and Operat­ing Conditions for PCAPP’s Biotreatment Process,” presentation to the committee, November 28, 2012. to the committee, November 28, 2012.

THE PCAPP BIOTREATMENT SYSTEM 19 Several assumptions used in the model agree reasonably ibility of operation and improve the ability to better well with the assumptions used by the committee regard- match the oxygen demand and supply. ing feed flow rate and composition (see Table 2-1). For 3. Addition of nitrate to the feed to serve as a sup- example, Tideflex assumed a peak flow rate of 9,648 gpd and plemental electron acceptor and nitrogen source an influent COD of 14,631 mg/L, 85 percent of which was (instead of urea). Nitrate is more soluble than oxygen assumed to be BOD. Based on an assumed process oxygen and can, therefore, be fed at a higher concentration, stoichiometric coefficient of 1.1 kg O2/kg of BOD (which is facilitating better penetration into the interior of the reasonable for an aerobic system with an SRT over 40 days), foam media; this can mitigate sulfidogenic condi- the predicted oxygen demand is 498 kg/day per ICB unit. The tions. As mentioned above, biodegradation of TDG mass rate of oxygen required was converted to an air flow under anoxic conditions has not yet been demon- rate of 1,880 scfm based on several assumptions, the most strated but seems likely, given its biodegradability important of which was a diffuser efficiency of 7 percent, under aerobic conditions and the likelihood that which is the fraction of oxygen delivered by the submerged catabolism occurs via a pathway in which oxygen- diffuser that is transferred to the water phase. This is likely ases are not involved. to be conservative, since the coarse bubbles will be broken 4. Pure oxygen could be used instead of air. This would up as they pass through the elastomer foam media, thereby also mitigate the potential for developing flux-limit- improving the oxygen transfer efficiency. ing conditions within the biofilm. The design of the ICB aeration system has the entire hydrolysate feed flow entering the first chamber (one half the The oxygen demand predicted by Tideflex Technologies volume of the total unit), yet the air supply will be uniformly assumes that nitrification will not occur at any point in the distributed throughout the 42,000-gallon tank volume (based ICB units. This should be monitored closely, since nitrifi- on uniform distribution of the diffusers). The ICB system as cation would add considerably to the oxygen demand. If currently configured for operation does not permit recircula- most of the BOD is removed in the first and/or second tank tion between chambers or step feeding of the influent to the and there is an excess of ammonium in the feed (e.g., cre- second and third chambers. Consequently, it is possible that ated by urea addition and a high demand for DAP from the anaerobic conditions will develop within the pore spaces of precipitation of phosphorus with iron),17 nitrification could the foam media, possibly leading to development of sulfido- become established in the third chamber. If an excessive genic conditions in the first chamber. oxygen demand due to nitrification does develop, it may be The calculations by Tideflex Technologies did not take advisable to switch from DAP and urea to a different form into account oxygen flux-limited conditions within the bio- of phosphorus (e.g., potassium phosphate) and nitrogen (e.g., film, which could lead to decreased TDG degradation and nitrate) to have better control over nutrient addition. The use excessive sloughing. Providing sufficient oxygen to meet of nitrate instead of ammonium for the source of nitrogen the stoichiometric oxygen demand of the waste does not provides three major benefits. The first is the elimination guarantee that oxygen will diffuse into the biofilm as fast as of nitrification, which will lower the oxygen demand in the its demand is exerted. In biofilm reactors, the oxygen sup- ICBs that may already have a marginal oxygen supply. Sec- ply must often be higher than the stoichiometric demand to ond, the nitrification of ammonia to nitrate generates protons increase the concentration gradient, and thus the diffusive (acid), so using nitrate will lessen the need for neutralization flux into the biofilm, to ensure that the rate of degradation is with caustic addition. Finally, nitrate will eliminate growth not limited by the rate of oxygen diffusion. of nitrifying bacteria, which will help to reduce the amount Several strategies may be used to mitigate oxygen defi- of biomass solids produced. cient conditions in the first chamber, including the following: Finding 2-11. The mass rate of oxygen demand predicted 1. Modification of the influent piping to allow for by Tideflex Technologies is in good agreement with assump- introduction of feed to the second and third cham- tions used by the committee for the PCAPP ICB influent. bers. Step feeding would distribute the loading more evenly, to ­ etter match the oxygen supply with b Finding 2-12. The air supply rate was calculated by Tideflex demand. Operation in this manner could be reserved Technologies based on the total mass rate of BOD and on for periods when the oxygen level in the first tank an evenly distributed air supply in proportion to the volume falls below the target bulk concentration of 2 mg/L. of each chamber. However, the BOD will not be distributed 2. Modification of the recirculation piping to allow for evenly for operation as currently planned; all of the load- recirculation from the second and third tanks to the ing would be added to the first chamber of each ICB unit. first. As currently configured for operation, recircu- lation is only to take place within the first chamber. 17Note that phosphorus precipitation was not taken into account in the These recirculation options would improve the flex- nutrient demand calculations performed by Golder Associates sent to the committee (Golder Associates Inc., 2010).

20 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP Consequently, oxygen-deficient conditions may develop in issues, potential for solids buildup, and challenges with the the first chamber. This is likely to result in development of oxygen supply, pH buffer, and nutrient addition discussed anaerobic conditions within the foam medium. above. Therefore, acclimation may require a ­ onger duration l than planned, with TDG feed concentrations increasing very Recommendation 2-13. PCAPP should consider performing slowly to the desired 7,000 mg/L. trials with a benign and common substrate, such as molasses, ­ to determine if sufficient oxygen can be delivered to main- Finding 2-13. The mixture of mustard agent hydrolysate tain aerobic conditions within the biofilm in the current ICB and recycled water to be produced at PCAPP constitutes a configuration. unique and complex feed to be processed through the ICBs. The committee believes that start-up will be the most critical Recommendation 2-14. In an aerobic biofilm process, meet- phase in establishing stable treatment in the ICBs. ing the stoichiometric oxygen requirement is not sufficient to ensure that oxygen flux (by diffusion into the biofilm) is Recommendation 2-19. Extensive monitoring of the hydro- not rate limiting. If it has not already been performed, the lysate as it is produced and of the PCAPP ICBs should occur, biofilm should be modeled to indicate if the biofilm will be particularly in the start-up phase, to ensure that the system oxygen flux-limited under best-case conditions. is robust and that the PCAPP operators have an accurate understanding of oxygen use, pH control, TDG degrada- Recommendation 2-15. PCAPP should consider modify- tion, biomass acclimation and development, and the solids ing both the influent and recirculation piping to permit step mass balance. feeding and chamber-to-chamber recirculation that would provide more flexibility for matching the air supply with the Recommendation 2-20. The committee understands that the oxygen demand. current plans are to start up the ICBs with the hydrolysate feed and achieve full design strength within 30 days. PCAPP Recommendation 2-16. PCAPP should consider add- should consider a longer ramp-up period to avoid system ing nitrate to the feed to serve as a supplemental electron failure and provide more time for the organisms to acclimate acceptor and nitrogen source; it should be provided at a suf- to the hydrolysate feed. Approaches include the following: ficiently high dose to prevent sulfidogenic conditions within the biofilm. • Starting up some units in parallel with different ­hydrolysate loading or sequentially to use the knowl- Recommendation 2-17. In the event that nitrification is estab- edge from one operating period to help optimize the lished and contributes excessively to the oxygen demand, conditions for the next operating period. This would PCAPP should consider changing from ­ iammonium phos- d allow time to observe the response of a subset of the phate to another form of phosphorus and urea to another form PCAPP ICB units with hydrolysate feed and fully test of nitrogen (e.g., nitrate) to have better control of nitrogen out potential operating parameters before attempting and phosphorus addition. to bring all 16 units online at the maximum design feed flow and TDG concentration. Recommendation 2-18. If the oxygen supply from delivered • Testing the ICB operation in advance with a benign air becomes inadequate, pure oxygen should be considered substrate during the systemization phase (as noted as a replacement for air in the PCAPP ICBs. in the section above, “Solids Buildup Concerns”) to provide valuable performance and operating data that will aid the start-up phase of the ICB units with the Start-up Issues hydrolysate containing TDG. An aggressive start-up phase is proposed for the ICB • Using surrogates that are easy to analyze, such as units at PCAPP. For example, the goal is to first treat the dissolved or total organic carbon, to predict/monitor h ­ ydrolysate from the 155-mm projectiles; this results in the TDG degradation in near-real time. highest TDG mass loadings and feed flow to the ICB units. In the current plans, the biomass seed will be collected from TDG accounts for about 85 percent of DOC or TOC the Colorado Springs publicly owned treatment works. The v ­ alues. So if 95 percent removal of DOC or TOC occurs, PCAPP staff has suggested that acclimation would require one can be confident that a substantial amount of the TDG is approximately 30 days: that is, the amount of ­ ydrolysate in h being removed. The committee is not recommending that a the feed will be ramped up from high dilution to the design surrogate only be used from Day 1, rather, that DOC or TOC feed (1 part hydrolysate to 7 parts recycled process water) data be collected along with TDG data to establish if there within 30 days. The committee believes that start-up and is a good correlation between their behaviors. After having acclimation will be critical to the success of the system, a track record of monitoring both parameters, eventually the particularly in light of the potential toxicity and inhibition DOC or TOC might then be used for routine monitoring,

THE PCAPP BIOTREATMENT SYSTEM 21 and then TDG could be sampled only as needed, e.g., for feed TSS accumulates in the bioreactor and how much is regulatory compliance. formed in the bioreactor. The amount of biomass (typically measured as VSS) in the effluent will be dictated by periodic sloughing and the planned scour and backwash. Spikes in Analysis and Composition of the Effluent and Off-Gas from effluent biomass concentration could be large during these the Bioreactors episodic events. The PCAPP staff has indicated that they will monitor The type and concentration of constituents in the ICB effluent ammonium to determine nutrient uptake and urea effluent may affect the performance of downstream pro- addition requirements for the bioreactors. There are two pos- cesses. For example, inorganic constituents and bacteria sible competing strategies for adding nitrogen. The first is as may contribute to fouling of the WRS-BRS system (e.g., a nutrient for biomass synthesis, and the amount of nitrogen hinder heat transfer and evaporation). Residual VOCs such as supplied should be just enough to meet the stoichiometric oxathiane or dithiane may condense upon cooling in the vent need of the biomass. Thus, the nitrogen concentration in the condenser, potentially blocking it. Some dissolved organic effluent should be low. The second need for nitrogen might compounds, such as biosurfactants and proteins, could cause be addition of nitrate to suppress sulfate reduction, and this foaming and damage compressors. will likely require a considerably higher nitrate concentra- Little is known about the composition of the bio­eactor r tion. Consequently, the addition of nitrogen may end up off-gas. Although one study with a SBR (Harvey et al., being a complex operational issue. Furthermore, if ammo- 1997) that was fed approximately 8,500 mg/L TDG with nium is supplied in excess, nitrification could take place in an hydraulic residence time of approximately 13 days did the second and third chambers of a unit, oxidizing the ammo- not find detectable levels of vinyl chloride and 1,2-dichloro- nium to nitrate. This possibility is discussed above. Indeed, ethane in the effluent (detection limits of 0.2 and 0.1 mg/L, this was observed in earlier pilot testing (SBR Technologies, respectively), the hydrolysate is known to possibly contain Inc., 1998). Simply monitoring for the residual ammonium trace volatile contaminants that could end up in the off-gas of concentration in the effluent and/or bioreactor units under- the bioreactor. The treatment train includes passing the off- estimates the presence of other nitrogen species (nitrate and gas through a granular activated carbon adsorber to remove nitrite), which could lead to excessive urea addition to the any odorous and toxic VOCs (BPT, 2006b). unit, increasing the oxygen demand and cost. Finding 2-15. The precise composition and characteristics of Finding 2-14. Nitrification could take place in the second the effluent and off-gas from the PCAPP ICBs are unknown, and third chambers of a given PCAPP ICB unit, transform- variable, and might affect downstream processing. ing ammonium to nitrate. If nitrogen dosing is based only on effluent ammonium, excessive urea addition to the unit Recommendation 2-22. The composition and characteris- could result. tics of the effluent and off-gas from the PCAPP ICBs should be closely monitored, especially during start-up, to anticipate Recommendation 2-21. There are complications involved potential long-term concerns for downstream processing. with the addition of nitrogen, so a more comprehensive mon- itoring strategy should be employed. Ammonium, nitrate, REFERENCES and nitrite should be monitored for the PCAPP ICBs and an effluent target of 1 mg/L of residual ammonium as nitrogen Bechtel Pueblo Team (BPT). 2005. Test Report for Bench-Scale Evalua- tion of HT, HD, and Energetics Hydrolysis and Biotreatment. Pueblo, should instead be changed to 1 mg/L of total residual nitro- Colo.: Bechtel. gen (ammonium, nitrate, and nitrite, as nitrogen). Bechtel Pueblo Team (BPT). 2006a. System Design Description (SDD) for Biotreatment System No. B09. Pueblo, Colo.: Bechtel. The amounts of solids, soluble microbial products Bechtel Pueblo Team. 2006b. System Design Description (SDD) for (organic compounds produced by the biomass), TDG, recal- Bioreactor Off-Gas Treatment System No. B11. 24852-RD-3YD-B11- BOO01. Pueblo, Colo.: Bechtel. citrant organics from the hydrolysate feed, and inorganics Earley, J.P., M.A. Guelta, and J.R. Mashinski. 2003. Biological Treatment in the effluent from the ICBs are likely to be variable and of Agent and Energetic Hydrolysates Generated from the Washout of are currently unknown. The ICB performance requirement Mustard (HD) Munitions. Aberdeen Proving Ground, Md.: U.S. Army is to remove at least 95 percent of feed TDG. The treatment Soldier and Biological Chemical Command. goal for COD is to remove at least 85 percent of the feed Golder Associates, Inc. 2009. Golder’s Immobilized Cell Bioreactor (ICBTM) General Technology Description. Atlanta, Ga.: Golder Asso­ concentration. Beyond these two parameters, there is uncer- ciates, Inc. tainty with respect to the concentration and composition of Golder Associates, Inc. 2010. MEB Calculation 101, Rev. 5. Atlanta, Ga.: the ICB effluent. The amount of dissolved solids will be Golder Associates, Inc. controlled in part by the salt content of the hydrolysate and in part by the amount of caustic and nutrients that must be added. The effluent TSS will depend on how much of the

22 REVIEW OF BIOTREATMENT, WATER RECOVERY, AND BRINE REDUCTION SYSTEMS FOR PCAPP Guelta, M.A. 2006. Biodegradation of HT Agent from an Assembled Klotzbücher, T., A. Kappler, K.L. Straub, and S.B. Haderlein. ������������ 2007. Biode- Chemical Weapons Assessment (ACWA) Projectile Washout Study. gradability and groundwater pollutant potential of organic anti-freeze ECBC-TR-486. Aberdeen Proving Ground, Md.: Edgewood Chemical liquids used in borehole heat exchangers. Geothermics 36(4): 348-361. Biological Center. National Research Council (NRC). 2005. Interim Design Assessment for Guelta, M.A., and J.J. DeFrank. 1998. Performance of Immobilized Cell the Pueblo Chemical Agent Destruction Pilot Plant. Washington D.C.: Bioreactors for Treatment of HD and VX Hydrolysates. ERDEC- The National Academies Press. TR-437. Aberdeen Proving Ground, Md.: Edgewood Research, Devel- Metcalf & Eddy, Inc., G. Tchobanoglous, F. Burton, and H.D. Stensel. 2003. opment and Engineering Center. Wastewater Engineering Treatment and Reuse, 4th Edition. Boston, Guelta, M.A., and L. Fazekas-Carey. 2003. Biodegradation of Hydrolyzed Mass.: McGraw-Hill. Mustard from an Assembled Chemical Weapons Assessment (ACWA) Nurdogan, Y., C.A. Myler, G.D. Lecakes, P.J. Usinowicz, and E.P. Blumen- Projectile Washout Study. ECBC-TR-291. Aberdeen Proving Ground, stein. 2012. Biological Treatment of Chemical Agent Hydrolysate by Md.: Edgewood Chemical Biological Center. Immobilized Cell Bioreactor Technology. New Orleans, La.: Water Envi­ Guelta, M.A., N.A. Chester, S. Lupton, M. Koch, I.J. Fry, and M.H. Kim. ronment Federation Technical Exhibition and Conference. Available at 2001. Biodegradation of HD and Tetrytol Hydrolysates in a Pilot http://www.golder.com/global/us/modules.php?name=Publication&sp_ Scale Immobilized Cell Bioreactor. ECBC-TR-192. Aberdeen Proving id=267&page_id=212#/!ts=1358887602837!. Accessed January 22, Ground, Md.: Program Manager for Assembled Chemical Weapons 2013. Assessment. Rittmann, B.E., and P.L. McCarty. 2001. Environmental Biotechnology: Guelta, M.A., N.A. Chester, C.W. Kurnas, M.V. Haley, F.S. Lupton, and M. Principles and Applications. Boston, Mass.: McGraw-Hill. Koch. 2002. Performance of the ACWA Pilot Scale Immobilized Cell SBR Technologies, Inc. 1996. Biodegradation of HD Hydrolysate in Bioreactor in Degradation of HD and Tetrytol Payloads of the M60 S ­ equencing Batch Reactors. South Bend, Ind.: SBR Technologies, Inc. Chemical Round. Aberdeen Proving Ground, Md.: Edgewood Chemical SBR Technologies, Inc. 1998. Final Report Evaluation of the Impact of Biological Center. Waste Water from the HD Neutralization/Biodegradation Treatment Harvey, S.P., L.L. Szafraniec, W.T. Beaudry, M.V. Haley, T.E. Rosso, G.P. Process on the Aberdeen Proving Ground Edgewood Area Federally Young, and J.P. Earley. 1996. HD Hydrolysis/Biodegradation Toxicol- Owned Treatment Works, Test 7: Organic Loading Study. South Bend, ogy and Kinetics. ERDEC-TR-382. Aberdeen Proving Ground, Md.: Ind.: SBR Technologies, Inc. Edgewood Research Development and Engineering Center. Tideflex Technologies. 2010. Diffused Aeration System Modeling: PCAPP, Harvey, S.P., L.L. Szafraniec, W.T. Beaudry, J.T. Earley, and R.L. Irvine. Pueblo, Colorado—ICB Bioreactors Coarse Bubble Aeration and Mix- 1997. Neutralization and Biodegradation of Sulfur Mustard. ERDEC- ing System. Carnegie, Pa.: Tideflex Technologies. TR-388. Aberdeen Proving Ground, Md.: Edgewood Research Develop- Xiu, Z.-M., J. Ma, and P.J.J. Alvarez. 2011. Differential effect of com- ment and Engineering Center. mon ligands and molecular oxygen on antimicrobial activity of silver nanoparticles versus silver ions. Environmental Science and Technology 45(20): 9003-9008.

Next: 3 The Water Recovery and Brine Reduction Systems »
Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant Get This Book
×
Buy Paperback | $39.00 Buy Ebook | $31.99
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The Pueblo Chemical Depot (PCD) in Colorado is one of two sites that features U.S. stockpile of chemical weapons that need to be destroyed. The PCD features about 2,600 tons of mustard-including agent. The PCD also features a pilot plant, the Pueblo Chemical Agent Destruction Pilot Plant (PCAPP), which has been set up to destroy the agent and munition bodies using novel processes. The chemical neutralization or hydrolysis of the mustard agent produces a Schedule 2 compound called thiodiglycol (TDG) that must be destroyed. The PCAPP uses a combined water recovery system (WRS) and brine reduction system (BRS) to destroy TDG and make the water used in the chemical neutralization well water again.

Since the PCAPP is using a novel process, the program executive officer for the Assembled Chemical Weapons Alternatives (ACWA) program asked the National Research Council (NRC) to initiate a study to review the PCAPP WRS-BRS that was already installed at PCAPP. 5 months into the study in October, 2012, the NRC was asked to also review the Biotreatment area (BTA). The Committee on Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant was thus tasked with evaluating the operability, life-expectancy, working quality, results of Biotreatment studies carried out prior to 1999 and 1999-2004, and the current design, systemization approached, and planned operation conditions for the Biotreatment process.
Review of Biotreatment, Water Recovery, and Brine Reduction Systems for the Pueblo Chemical Agent Destruction Pilot Plant is the result of the committee's investigation. The report includes diagrams of the Biotreatment area, the BRS, and WRS; a table of materials of construction, the various recommendations made by the committee; and more.
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!