Effect of O2 exposure on perchlorate reduction by Dechlorosoma sp. KJ

Anaerobic bioreactors have been developed to remove perchlorate from water, but backwashing and operational interruptions can expose biofilms to oxygen. While it is well known that oxygen is a preferential electron acceptor to perchlorate for perchlorate-respiring bacteria, little is known about the effect of oxygen exposure or redox potentials on perchlorate reduction. Four different dissolved oxygen scavengers were tested for their ability to quickly restore anaerobic conditions and allow perchlorate reduction by a facultative, perchlorate respiring bacterium Dechlorosoma sp. KJ. Of the four different oxygen scavengers tested (Oxyrase trade mark, L-cysteine, Na2S and FeS), only Oxyrase trade mark was able to rapidly (<30 min) scavenge dissolved oxygen and allow cell growth. There was no cell growth after addition of Na2S and FeS, and l-cysteine produced a long lag in cell growth. To investigate the effect of dissolved oxygen on perchlorate reduction, anaerobically grown cultures Dechlorosoma sp. KJ, were exposed to dissolved oxygen for various periods ranging from 1 to 32 h. Perchlorate reduction and redox potential were then measured for cells returned to an anaerobic environment containing an oxygen scavenger. It was determined that cells exposed to dissolved oxygen for more than 12h were incapable of reducing perchlorate. Cells exposed to dissolved oxygen for less than 12h quickly reduced the redox potential to negative values (-127 mV to -337 mV) and were able to reduce perchlorate or chlorite. Our results suggest that aeration during backwashing of biofilm reactors, or exposure of perchlorate-degrading cell suspensions to dissolve oxygen for less than 12h, will not be detrimental to the ability of perchlorate-degrading bacteria to use perchlorate as an electron acceptor.     

Kinetics of a hydrogen-oxidizing, perchlorate-reducing bacterium

This paper provides the first kinetic parameters for a hydrogen-oxidizing perchlorate-reducing bacterium (PCRB), Dechloromonas sp. PC1. The qmax for perchlorate and chlorate were 3.1 and 6.3 mg/mgDW-day, respectively. The K for perchlorate was 0.14 mg/L, an order of magnitude lower than reported for other PCRB. The yields Y on perchlorate and chlorate were 0.23 and 0.22 mgDW/mg, respectively, and the decay constant b was 0.055/day. The growth-threshold, Smin, for perchlorate was 14 microg/L, suggesting that perchlorate cannot be reduced below this level when perchlorate is the primary electron-acceptor, although it may be possible when oxygen or nitrate is the primary acceptor. Chlorate accumulated at maximum concentrations of 0.6-4.3 mg/L in batch tests with initial perchlorate concentrations ranging from 100 to 600 mg/L. Furthermore, 50 mg/L chlorate inhibited perchlorate reduction with perchlorate at 100 mg/L. This is the first report of chlorate accumulation and inhibition for a pure culture of PCRB. These Chlorate effects are consistent with competitive inhibition between perchlorate and chlorate for the (per)chlorate reductase enzyme.     

Reduction of bromate to bromide coupled to acetate oxidation by anaerobic mixed microbial cultures

Bromate, a weakly mutagenic oxidizing agent, exists in surface waters. The biodegradation of bromate was investigated by assessing the ability of mixed cultures of micro-organisms for utilization of bromate as electron acceptor and acetate as electron donor. Reduction of bromate was only observed at relatively low concentrations (<3.0 mM) in the absence of molecular oxygen. Under these conditions bromate was reduced stoichiometrically to bromide. Unadapted sludge from an activated sludge treatment plant and a digester reduced bromate without lag period at a constant rate. Using an enrichment culture adapted to bromate, it was demonstrated that bromate was a terminal electron acceptor for anaerobic growth. Approximately 50% of the acetate was utilized for growth with bromate by the enrichment culture. A doubling of 20 h was estimated from a logarithmic growth curve. Other electron acceptors, like perchlorate, chlorate and nitrate, were not reduced or at negligible rates by bromate-utilizing microorganisms.     

Kinetics of a chlorate-accumulating, perchlorate-reducing bacterium

Kinetics parameters for perchlorate and chlorate reduction were determined for Dechlorosoma sp. HCAP-C, also known as Dechlorosoma sp. PCC, a novel perchlorate-reducing bacterium (PCRB) that accumulates significant amounts of chlorate during perchlorate reduction. This is the first report of such behavior, and we hypothesized the perchlorate reduction kinetics would be markedly different from other PCRB. In batch tests with initial perchlorate concentrations ranging from 200 to around 1400 mg/L, maximum chlorate accumulation ranged from 41 to 279 mg/L, and were consistently around 20% of the initial perchlorate concentration. For perchlorate, parameters were determined using a competitive inhibition model. The maximum specific substrate degradation rate qmaxP was 11.5mgClO4-/mgdry weight (DW)-d, and the half-maximum rate constant KP was 193 mgClO4-/L. For chlorate, the qmaxC was 8.3 mgClO3-/mgDW-d and the KC was 58.3 mgClO3-/L. The high KP values relative to conventional PCRB, values suggests that HCAP-C does not play a significant role at low perchlorate concentrations. However, the relatively high qmaxP, and the potential for syntrophic relationships with chlorate-reducing bacteria that relieve the effects of chlorate inhibition, suggest that HCAP-C could play a significant role at high perchlorate concentrations.     

Feasibility and kinetics study on the direct bio-regeneration of perchlorate laden anion-exchange resin

Anion exchange is one of the most promising treatment technologies for the removal of low levels of perchlorate. The spent anion-exchange resins, however, need to be disposed of or regenerated because they contain high contents of perchlorate. This study investigated the feasibility and kinetics of a direct bio-regeneration method. The method accomplished resin regeneration and biological perchlorate destruction concurrently, by directly contacting the spent resin with the perchlorate-reducing bacteria (PRB). The results indicated that the method was effective in regeneration of perchlorate and nitrate loaded resin and the resin could be repeatedly regenerated with the method. The regenerated resin was effective, stable, and durable in the filtration treatment of perchlorate in well water from the Saddle River area, NJ. Moreover, the method was also effective in regeneration of the spent A-530E resin, which had high perchlorate affinity and was yet very difficult for regeneration with the conventional brine desorption technique. Besides, the results further suggested that the perchlorate and nitrate desorption from the loaded resin coupling with their subsequent biological reduction could be the direct bio-regeneration mechanism. No biofilm was formed on the regenerated resin surface according to a scanning electron microscopy (SEM) analysis.     

Characterization of bromate-reducing bacterial isolates and their potential for drinking water treatment

The objective of the current study was to isolate and characterize several bromate-reducing bacteria and to examine their potential for bioaugmentation to a drinking water treatment process. Fifteen bromate-reducing bacteria were isolated from three sources. According to 16S rRNA gene sequencing, the bromate-reducing bacteria are phylogenetically diverse, representing the Actinobacteria, Bacteroidetes, Firmicutes, and α-, β-, and γ-Proteobacteria. The broad diversity of bromate-reducing bacteria suggests the widespread capability for microbial bromate reduction. While the cometabolism of bromate via nitrate reductase and (per)chlorate reductase has been postulated, five of our bromate-reducing isolates were unable to reduce nitrate or perchlorate. This suggests that a bromate-specific reduction pathway might exist in some microorganisms. Bioaugmentation of activated carbon filters with eight of the bromate-reducing isolates did not significantly decrease start-up time or increase bromate removal as compared to control filters. To optimize bromate reduction in a biological drinking water treatment process, the predominant mechanism of bromate reduction (i.e., cometabolic or respiratory) needs to be assessed so that appropriate measures can be taken to improve bromate removal.     

The synthesis of the dissimilatory nitrate reductase under aerobic conditions in a number of denitrifying bacteria, isolated from activated sludge and drinking water

A number of denitrifying bacteria were isolated from activated sludge and drinking water. These bacteria were tested for the synthesis of the dissimilatory nitrate reductase under aerobic conditions (dissolved oxygen concentration above 4 mg · l⁻¹). The synthesis of this enzyme varied from total repression by oxygen in some bacteria, especially those isolated from drinking water, until a nearly non oxygen-repressed synthesis in other bacteria (strains 15 and N4). The effect of the dissolved oxygen concentration during growth of the bacteria on the synthesis of the dissimilatory nitrate reductase in cells of strain 15 was studied more extensively. A considerable repression of the enzyme synthesis was obtained when the dissolved oxygen concentration was relatively high (approx 15 mg·l⁻¹). Addition of chlorate to the growth medium of strain 15 (using NH⁺₄-N as nitrogen source) also resulted in a serious repression of the nitrate reductase synthesis during aerobic growth (dissolved oxygen above 4 mg·l⁻¹). The dissimilatory nitrate reductase of aerobically grown cells of strains 15 and N4 was found to be mainly localized in the membrane fraction.     

An electrokinetic/Fe0 permeable reactive barrier system for the treatment of nitrate-contaminated subsurface soils

Effective nitrate removal by Fe⁰ permeable reactive barriers (Fe⁰ PRB) has been recognized as a challenging task because the iron corrosion product foamed on Fe⁰ hinders effective electron transfer from Fe⁰ to surface-bound nitrate. The objectives of this study were (i) to demonstrate the effectiveness of an electrokinetic/Fe⁰ PRB system for remediating nitrate-contaminated low permeability soils using a bench-scale system and (ii) to deepen the understanding of the behavior and fate of nitrate in the system. Bench-scale laboratory experiments were designed to investigate the influence of the Fe⁰ content in the permeable reactive barrier, the pH in the anode well, and the applied voltage on remediation efficiency. The experimental results showed that the major reaction product of nitrate reduction by Fe⁰ was ammonium and that nitrate reduction efficiency was significantly influenced by the variables investigated in this study. Nitrate reduction efficiency was enhanced by either increasing the Fe⁰ content in the Fe⁰ reactive barrier or decreasing the initial anode pH. However, nitrate reduction efficiency was reduced by increasing the applied voltage from 10 V to 40 V due to the insufficient reaction time during nitrate migration through the Fe⁰ PRB. For all experimental conditions, nearly all nitrate nitrogen was recovered in either anode or cathode wells as nitrate or ammonium within 100 h, demonstrating the effectiveness of the system for remediating nitrate-contaminated subsurface soils.     

Kinetics of nitrate and perchlorate removal and biofilm stratification in an ion exchange membrane bioreactor

The biological degradation of nitrate and perchlorate was investigated in an ion exchange membrane bioreactor (IEMB) using a mixed anoxic microbial culture and ethanol as the carbon source. In this process, a membrane-supported biofilm reduces nitrate and perchlorate delivered through an anion exchange membrane from a polluted water stream, containing 60 mg/L of NO₃⁻ and 100 μg/L of ClO₄⁻. Under ammonia limiting conditions, the perchlorate reduction rate decreased by 10%, whereas the nitrate reduction rate was unaffected. Though nitrate and perchlorate accumulated in the bioreactor, their concentrations in the treated water (2.8 ± 0.5 mg/L of NO₃⁻ and 7.0 ± 0.8 μg/L of ClO₄⁻, respectively) were always below the drinking water regulatory levels, due to Donnan dialysis control of the ionic transport in the system.Kinetic parameters determined for the mixed microbial culture in suspension showed that the nitrate reduction rate was 35 times higher than the maximum perchlorate reduction rate. It was found that perchlorate reduction was inhibited by nitrate, since after nitrate depletion perchlorate reduction rate increased by 77%. The biofilm developed in the IEMB was cryosectioned and the microbial population was analyzed by fluorescence in situ hybridization (FISH). The results obtained seem to indicate that the kinetic advantage of nitrate reduction favored accumulation of denitrifiers near the membrane, whereas per(chlorate) reducing bacteria were mainly positioned at the biofilm outer surface, contacting the biomedium. As a consequence of the biofilm stratification, the reduction of perchlorate and nitrate occur sequentially in space allowing for the removal of both ions in the IEMB.     

Microbial ecology of a perchlorate-reducing, hydrogen-based membrane biofilm reactor

The hydrogen-based membrane biofilm reactor (MBfR) has been shown to reduce perchlorate to below 4 microg/L, but little is known about the microbial ecology of this or other hydrogen-based reactors, especially when influent perchlorate concentrations are much lower than the influent oxygen and nitrate concentrations. Dissimilatory (per)chlorate-reducing bacteria (PCRB) can use oxygen as an electron acceptor, and most can also use nitrate. Since oxygen and nitrate can be reduced concurrently with perchlorate, they may serve as primary electron acceptors, sustaining PCRB when the perchlorate concentrations are very low. We studied five identical MBfRs, all seeded with the same inoculum and initially supplied with oxygen, or oxygen plus nitrate, in the influent. After 20 days, perchlorate was added to four MBfRs at influent concentrations of 100-10,000 microg/L, while the fifth was maintained as a control. One day after perchlorate addition, the MBfRs displayed limited perchlorate reduction, suggesting a low initial abundance of PCRB. However, perchlorate reduction improved significantly over time, and denaturing gradient gel electrophoresis (DGGE) analyses suggested an increasing abundance of a single Dechloromonas species. Fluorescence in-situ hybridization (FISH) tests showed that the Dechloromonas species accounted for 14% of the bacterial count in the control MBfR, and 22%, 31%, and 49% in the MBfRs receiving nitrate plus 100, 1000, and 10,000 microg/L perchlorate, respectively. The abundance was 34% in the MBfR receiving oxygen plus 1000 microg/L perchlorate. These results suggest that oxygen is more favorable than nitrate as a primary electron acceptor for PCRB, that PCRB are present at low levels even without perchlorate, and that the presence of perchlorate, even at low levels relative to nitrate or oxygen, significantly enhances selection for PCRB.     

Sulfolane biodegradation potential in aquifer sediments at sour natural gas plant sites

Sulfolane is used in the treatment of sour natural gas. It is a highly water soluble compound that has been introduced into soils and groundwaters at a number of sour gas processing plant sites. Aquifer sediments from contaminated locations at three sites in western Canada were assessed for microbial activity and their ability to degrade sulfolane under aerobic and five anaerobic (nitrate-, Mn(IV)-, Fe(III)-, sulfate- and CO₂-reducing) conditions. The microcosms were supplemented with 200 mg/L sulfolane and adequate supplies of N, P, and the appropriate terminal electron acceptor. Microcosms containing contaminated aquifer sediments from each of the three sites were able to degrade sulfolane aerobically at 8°C and 28°C, and the biodegradation followed zero-order kinetics. The lag times before the onset of sulfolane biodegradation were shorter when sulfolane-contaminated sediments were used as inocula than when uncontaminated soils were used. No anaerobic sulfolane biodegradation was observed at 28°C, nor was sulfolane biodegradation observed at 8°C under Fe(III)-, sulfate- and CO₂-reducing conditions. At 8°C, anaerobic degradation of sulfolane coupled to Mn(IV) reduction was observed in microcosms from two sites, and degradation coupled to nitrate reduction was seen in a microcosm from one of the contaminated sites.     

Aerobic and anoxic biodegradation of benzoate: stability of biodegradative capability under endogenous conditions

Aromatic organic compounds are degraded by different enzyme systems under aerobic and anoxic conditions. This raises the question of how bacteria in biological nitrogen removal processes, which cycle bacteria between aerobic and anoxic environments, regulate their enzyme systems for degrading aromatic compounds. As a first step in answering that question, mixed microbial communities were grown on benzoate as sole carbon source in chemostats under fully aerobic and fully anoxic (nitrate as the electron acceptor) conditions and tested for their ability to degrade benzoate in batch reactors after exposure to aerobic or anoxic conditions in the absence of substrate. Aerobically grown biomass retained its ability to degrade benzoate without loss of activity after endogenous exposure to aerobic conditions for up to 8 h. However, when exposed to anoxic conditions, the biomass rapidly lost its aerobic benzoate degrading activity, retaining less than 20% of the initial activity after 8 h. Similarly, anoxically grown biomass retained its ability to degrade benzoate without loss of activity after endogenous exposure to anoxic conditions for up to 8 h. However, when anoxically grown biomass was exposed to aerobic conditions, only 20% of its initial activity was lost in the first 2 h, after which the remaining activity was retained for up to 8 h. Similar experiments with pyruvate showed that the 20% loss of activity was not due to loss of denitrifying enzymes, suggesting that it was due to loss of catabolic enzymes.     

Microbial reduction of perchlorate with zero-valent iron

Microbial reduction of perchlorate in the presence of zero-valent iron was examined in both batch and column reactors to assess the potential of iron as the electron donor for biological perchlorate reduction process. Iron-supported mixed cultures completely removed 65 mg/L of perchlorate in batch reactors in 8 days. The removal rate was similar to that observed with hydrogen gas (5%) and acetate (173 mg/L) as electron donors. Repeated spiking of perchlorate to batch reactors containing iron granules and microorganisms showed that complete perchlorate reduction by the iron-supported culture was sustained over a long period. Complete removal of perchlorate by iron-supported anaerobic culture was also achieved in a bench-scale iron column with a hydraulic residence time of 2 days. This study demonstrated the potential applicability of zero-valent iron as a source of electrons for biological perchlorate reduction. Use of zero-valent iron may eliminate the need to continually supply electron donors such as organic substrates or explosive hydrogen gas. In addition, iron is inexpensive, safe to handle, and does not leave organic residuals in the treated water.     

Microbial reduction of perchlorate in pure and mixed culture packed-bed bioreactors

Perchlorate (ClO4-) has been detected in a large number of surface and ground waters in the US. Due to health concerns of perchlorate in drinking water, the California Department of Health Services has established a provisional action level of 18 microg/L. Several microbial isolates have been obtained capable of microbiological perchlorate reduction through cell respiration, but few of these have been tested for perchlorate removals to these low levels. The feasibility of using one isolate (KJ) for water treatment was tested in a packed-bed bioreactor by comparing minimum detention times necessary to achieve complete removal of perchlorate. Perchlorate was reduced approximately from 20 mg/L to non-detectable (< 4 microg/L) levels in acetate-fed columns inoculated with KJ or mixed cultures. The complete conversion of perchlorate to chloride was demonstrated by a stoichiometric ratio of perchlorate to chloride of 1.0 +/- 0.14. Perchlorate removal to non-detectable levels required a minimum empty bed contact time (EBCT) of only 2.1 min for the column inoculated with KJ, vs. 31 min for the mixed culture column. Acetate was used at a molar ratio of C2H3O2-/ClO4- of 2.9 (n = 6) for the mixed culture, while more than twice as much acetate was consumed on average (6.6 +/- 2.0, n = 156) by the pure culture. These results demonstrate that detention times of packed-bed bioreactors can be substantially reduced using isolate KJ, but that larger concentrations of acetate will be necessary to reduce perchlorate to low levels necessary for drinking water.     

Assessment of the formation of inorganic oxidation by-products during the electrocatalytic treatment of ammonium from landfill leachates

This work investigates the formation of oxidation by-products during the electrochemical removal of ammonium using BDD electrodes from wastewaters containing chlorides. The influence of the initial chloride concentration has been experimentally analyzed first, working with model solutions with variable ammonium concentration and second, with municipal landfill leachates. Two different levels of chloride concentration were studied, i) low chloride concentrations ranging between 0 and 2000 mg/L and, ii) high chloride concentrations ranging between 5000 and 20,000 mg/L. Ammonium removal took place mainly via indirect oxidation leading to the formation of nitrogen gas and nitrate as the main oxidation products; at high chloride concentration the formation of nitrogen gas and the rate of ammonium removal were both favored. However, chloride was also oxidized during the electrochemical treatment leading to the formation of free chlorine responsible of the ammonium oxidation, together with undesirable products such as chloramines, chlorate and perchlorate. Chloramines appeared during the treatment but they reached a maximum and then started decreasing, being totally removed when high chloride concentrations were used. With regard to the formation of chlorate and perchlorate once again the concentration of chloride exerted a strong influence on the formation kinetics of the oxidation by-products and whereas at low chloride concentrations, chlorate appeared like an intermediate compound leading to the formation of perchlorate, at high chloride concentrations chlorate formation was delayed significantly and perchlorate was not detected during the experimental time. Thus this work contributes first to the knowledge of the potential hazards of applying the electro-oxidation technology as an environmental technology to deal with ammonium oxidation under the presence of chloride and second it reports efficient conditions that minimize or even avoid the formation of undesirable by-products.     

Biodegradation of nitrilotriacetate (NTA) by bacteria—II. Cultivation of an NTA-degrading bacterium in anaerobic medium

Bacterial strains with the ability to utilize nitrilotriacetate (NTA) under aerobic and anaerobic conditions have been isolated from natural waters exposed to NTA. One of these strains (NTA-A2) is a facultative anaerobe which grows under anaerobic conditions on NTA if nitrate is available in the medium. (Under aerobic conditions this strain can utilize acetate, glucose and some other sugars as well as NTA but not EDTA (ethylenediaminotetraacetate) or lactose as the sole carbon source.) The properties and characteristics of the strain NTA-A2 are described.     

Chlorate reduction capacity and characterisation of chlorate reducing bacteria communities in sediments of the rio Cruces wetland in southern Chile

This study investigated chlorate reduction kinetics in multiple samples of sediments from a longitudinal profile of a wetland located downstream of the effluent discharge of a cellulose plant, including characterisation of the bacterial communities involved. The sediments were exposed to different initial chlorate concentrations in microcosm tests, with and without the addition of acetate as an external electron donor, and in a matrix of natural water or a defined medium. At a high initial chlorate concentration of 100 mg/L, in the absence of an external electron source, the degradation curves presented first-order kinetics, influenced by electron donor availability. The first-order kinetic constant varied between 0.05 and 0.17 day⁻¹. Subsequently, when the initial chlorate concentration was reduced to 7 mg/L, a zero-order kinetic was obtained, with the kinetic constant presenting values between 1.1 and 1.3 mg/L-day. No correlation was observed between chlorate degradation kinetics and the location of the sampling points or the previous history of exposure to chlorate. Other factors evaluated, such as the availability of organic matter or the chlorate reducing bacteria count, also proved not to have any incidence on the results. The richness of chlorate reducing bacteria species in the different samples analysed were also similar, with the greatest similarity being found between cld genes in the samples from the upstream or downstream sampling points. Additionally, cld genes most similar to those present in PCRB like Dechlorospirillum sp., Alicycliphilus denitrificans, Dechloromonas agitata, Dechloromonas sp. LT1 and Ideonella dechloratans were detected. This study showed that the anaerobic sediments of the Cruces river wetland present a high potential for chlorate natural attenuation, regardless of the previous history of exposure to chlorate. This capacity is associated with the presence of a diverse community of chlorate reducing bacteria.     

The impact of temperature on the performance of anaerobic biological treatment of perchlorate in drinking water

A 20-month pilot-scale study was conducted to examine the impact of temperature on the performance of an anaerobic biological contactor used to treat perchlorate-contaminated water. The contactor was successfully acclimated with indigenous microorganisms. Influent temperatures varied from 1.4 to 30 °C. The objectives of the study were to investigate the effects of temperature on perchlorate removal, nitrate removal, nitrite formation, dissolved oxygen consumption, sulfide production, and nutrient acetate consumption. The results confirmed that consistent biological perchlorate removal to 2 μg/L is feasible at temperatures above 10 °C. Effluent concentrations of perchlorate, nitrate, and dissolved oxygen varied inversely with temperature, while sulfide varied positively with temperature. Under the conditions that prevailed during this study, 10 °C was a threshold temperature below which microbial activity, including perchlorate reduction, decreased dramatically.     

Stability of chlorine dioxide in aqueous solution

Chlorine dioxide in unbuffered aqueous solution at pH 9 and 25°C decomposes to chlorate, chlorite, chloride and oxygen by coupled slow and rapid reactions. The initiation of the second rapid decomposition depends on the initial chlorine dioxide concentration and ionic strength of the solution. Chloride ion has a catalytic and inhibiting effect, changing the product distribution to a 1:1 molar ratio of chlorate and chlorite, thus increasing the oxidizing potential of chlorine dioxide. In buffered solutions at pH 8.8 the reaction is pseudo-second order, yielding chlorate and chlorite as products.     

Fluidized bed reactor for the biological treatment of ion-exchange brine containing perchlorate and nitrate

The removal of perchlorate and nitrate from contaminated drinking water using regenerable ion-exchange processes produces a high salt brine (3-10% NaCl) laden with high concentrations of perchlorate and nitrate. This bench-scale research describes the operation of acetate-fed granular activated carbon (GAC) based fluidized bed reactors (FBR) for perchlorate-only, and combined nitrate and perchlorate removal from synthetic brine (6% NaCl). The GAC was inoculated with a salt-tolerant culture developed by the authors and used previously in batch systems. An FBR was an effective design for perchlorate reduction and exhibited first-order degradation kinetics with respect to perchlorate concentrations. Nitrate was also removed by the organisms in the column and had no negative effects on the removal of perchlorate using the FBR design. However, at higher concentrations of nitrate the FBR was more difficult to operate due to loss of carbon and biomass from the formation of nitrogen bubbles and the high recycle flow rates needed.