ENVIRONMENTAL ENGINEERING SCIENCEVolume 20, Number 5, 2003© Mary Ann Liebert, Inc.

Microbial Degradation of Perchlorate: Principles and Applications

Jianlin Xu, Yanguang Song, Booki Min, Lisa Steinberg, and Bruce E. Logan*

Department of Civil and Environmental EngineeringThe Pennsylvania State University

University Park, PA 16802

ABSTRACT

Perchlorate (ClO42) release into the environment has occurred primarily in association with its manufac-

ture and use in solid rocket propellant. When released into groundwater, perchlorate can spread over largedistances because it is highly soluble in water and adsorbs poorly to soil. Two proven techniques to re-move perchlorate from drinking water are anaerobic biological reactors and ion exchange. In this review,we focus on the application of microbiological systems for degrading perchlorate. Some bacteria can useperchlorate as an electron acceptor while oxidizing a large range of substrates. Perchlorate-respiring bac-teria (PRB) are widely distributed in the environment, and are enriched at perchlorate-contaminated sites.We review the pathways by which PRB degrade perchlorate, and the different biological treatment pro-cesses that have been developed to remove perchlorate from water sources. We also discuss the effectsof alternate electron acceptors in the water, such as oxygen and nitrate, on perchlorate removal. Althoughmany different biological treatment systems using PRB have so far only been proven at the bench scale,all pilot scale tests performed to date with a few of these systems have been successful. The success ofthese perchlorate bioreactor tests indicates that biological treatment is a suitable method for soil remedi-ation and water treatment of perchlorate-contaminated water.

Key words: perchlorate; application; percolate-respiring bacteria; bioremediation

405

*Corresponding author: Department of Civil and Environmental Engineering, 212 Sackett Bldg, Penn State University, Uni-versity Park, PA 16802. Phone: 814-863-7908; Fax: 814-863-7304; E-mail: blogan@psu.edu

INTRODUCTION

PERCHLORATE (ClO42) is a highly oxidized (17) chlo-

rine oxy-anion manufactured for use as the oxidizerin solid propellants for rockets, missiles, explosives, andpyrotechnics (Urbansky, 1998; Gullick et al., 2001; Lo-gan, 2001a). Approximately 90% of all perchlorate saltsare manufactured as ammonium perchlorate for use in

rocket and missile propellants. The periodic replacementand use of solid propellant has resulted in the dischargeof more than 15.9 million kg of perchlorate salts into theenvironment since the 1950s (Motzer, 2001). Perchloratesalts are highly soluble in water. Sodium perchlorate hasa solubility of about 2 kg/L, allowing large amounts tobe readily transported through surface and ground wa-ters. The U.S. EPA has identified perchlorate users and

manufacturers in 44 states, and perchlorate releases in atleast 20 states (U.S. EPA, 2002). Such perchlorate re-leases are estimated to have affected the drinking waterof 15 million people.

Perchlorate can be detected by many methods includ-ing ion-selective electrodes, ion chromatography, capil-lary electrophoresis, HPLC, and spectrophotometry (Ur-bansky, 1998). In 1997, the California Department ofHealth first reported an ion chromatographic method thatwas capable of detecting perchlorate concentrations of 4mg/L. Since then, ion chromatography has been the mostcommonly used detection method for perchlorate. Per-chlorate is only known to occur in the natural environ-ment in Chilean caliche, a material that is used for somefertilizers (Ericksen, 1983). The EPA developed amethod for measuring perchlorate concentrations in fer-tilizers (Collette et al., 2001), and has used this methodto survey large numbers of fertilizers and related materi-als. It was concluded that most fertilizers did not containperchlorate, and therefore, that fertilizers did not con-tribute to the extensive environmental perchlorate con-tamination that has been observed (Urbansky et al.,2001).

Although there is currently no federal drinking waterstandard for perchlorate, perchlorate has been included onthe federal Contaminant Candidate List (U.S. EPA, 1998).High concentrations of perchlorate are known to affect thefunction of the thyroid gland in humans by inhibiting theuptake of iodide (Wolff, 1998), but direct epidemiologi-cal evidence that indicates that perchlorate is toxic to ex-posed humans is lacking (Crump et al., 2000; Li et al.,2001; U.S. EPA, 2002). Recent studies have indicated thatlow concentrations of perchlorate significantly inhibit io-dide uptake in humans and animals (Lawrence et al., 2000;OEHHA, 2002; U.S. EPA, 2002). Perchlorate contami-nation of the environment poses a threat to indigenouswildlife as well as human health. Smith et al. (2001) havefound perchlorate contamination of wildlife and vegeta-tion at concentrations that can pose a threat to the normalgrowth and development of amphibian populations (Gole-man et al., 2002a, 2002b). The Office of EnvironmentalHealth Hazard Assessment in California EPA has pro-posed a Public Health Goal of 6 mg/L for perchlorate indrinking water (OEHHA, 2002). In a recent perchloraterisk assessment draft report, the U.S. EPA (2002) pro-posed a draft reference dose of 0.03 mg/kg of body weightper day, which could produce a drinking water equivalentlevel of 1 mg/L to protect human health. Based on this in-formation, the California Department of Health Service inCalifornia decreased the action level for perchlorate indrinking water from 18 to 4 mg/L (DHS, 2002b). In NewMexico, the action level was set at 1 mg/L (www.clu-in.org/studio/perchlorate-060402.

In the last 5 years, several reviews have been publishedon various perchlorate issues that include: bacterialdegradation (Herman and Frankenberger, 1998; Logan,1998); chemistry and analytical chemistry (Urbansky,1998, 2000a, 2000b; Urbansky and Schock, 1999) toxi-cological studies and drinking water standards (Wolff,1998; Urbansky, 2000a; Soldin et al., 2001; OEHHA,2002; U.S. EPA, 2002); and contamination sources andoccurrence data (Urbansky, 2000a; Cheremisnnoff, 2001;Gullick et al., 2001; Logan, 2001a). However, there havebeen important advances made in the treatment of per-chlorate-contaminated water since the microbiology andexisting treatment technologies were reviewed in 1998(Herman and Frankenberger, 1998; Logan, 1998). Oneof the most important developments since these reviewshave been reports of microbial treatment processes ca-pable of removing perchlorate down to levels expectedto be suitable for drinking water (,4 mg/L). These pro-cesses are the basis of several recent patents on biologi-cal treatment processes for perchlorate treatment (see At-taway et al., 1999; Coppola and McDonald, 2000;Frankenberger, 2000; Gaudre-Longerinas and Tauzia,2001; Logan, 2001b; Van Ginkel et al., 1999).

In the current review, we focus on recent developmentsthat have improved our understanding of the bacteria re-sponsible for perchlorate degradation, the pathways usedfor perchlorate degradation, and the treatment systemsdeveloped to use perchlorate respiring bacteria (PRB).We provide only a brief background on the chemistry,occurrence, health issues, and drinking water issues forperchlorate, and refer the reader to other more compre-hensive studies and reviews on these subjects. We con-centrate on topics related to microbial degradation of per-chlorate because biological treatment is a particularlypromising option for remediation of perchlorate-contam-inated drinking water. Although these biological per-chlorate treatment technologies are quite new, there aresigns that such treatment systems will be accepted for usein water treatment plants. The California Department ofHealth Services recently approved the use of fluidizedbed reactors for the treatment of perchlorate contaminatedgroundwater (DHS, 2002a), and biological denitrificationhas been used at one site in Oklahoma for the treatmentof nitrate-contaminated groundwater (www.pall.com).

PECHLORATE RESPIRING BACTERIA

Ubiquity and diversity of perchlorate reducing bacteria

It has been known for several years that the ability toreduce perchlorate is not limited to a single bacterialspecies, although some of this earlier evidence of per-

406 XU ET AL.

chlorate degradation was inferred from information onbiological chlorate degradation (Logan, 1998). Bacteriacapable of chlorate reduction were found to inhabit a va-riety of environments including rivers, sediments, soils,and wastewater treatment plants (van Ginkel et al., 1995).Utilizing media with chlorate as the only electron ac-ceptor, Coates et al. (1999) similarly found that the ac-etate-oxidizing chlorate reducing bacteria (CRB) repre-sented a significant population whose abundance rangedfrom 2.31 3 103 to 2.4 3 106 cells per gram of samplesobtained from a variety of sources, including pristine andhydrocarbon-contaminated soils, aquatic sediments, pa-per mill waste sludges, and farm animal waste lagoons.All 13 isolates obtained in this study were also capableof growth on acetate using perchlorate, leading to earlyspeculation, via the abbreviation of (per)chlorate, that allCRB were also capable of perchlorate reduction (Coateset al., 1999).

More recent studies have provided evidence that notall CRB are PRB, although the converse is true, that allPRB are CRB. Wu et al. (2001) used perchlorate or chlo-rate in anaerobic growth medium to compare the abun-dance of PRB and CRB in different environmental sam-ples. They found that when perchlorate was the soleterminal electron acceptor in the medium, the number ofPRB in a pristine soil was up to 1,000-fold lower thanthat found using chlorate in the medium to enumerateCRB for the same samples (Wu et al., 2001). This pro-vided indirect evidence that not all CRB were PRB, andthat PRB were less abundant than CRB in these envi-

ronments. Confirmation that not all CRB were PRB wasprovided by three different studies. Only 8 of 10 CRBisolated from wastewater were found to be capable ofrespiring perchlorate (Logan et al., 2001c). In anotherstudy, it was mentioned that a microorganism was iso-lated that was capable of using chlorate, but not per-chlorate (Coates et al., 2000). Wolterink et al. (2002) iso-lated a CRB strain Pseudomonas chloritidismutansAW-1from an anaerobic bioreactor treating wastewater con-taining chlorate, and found that this strain did not reduceperchlorate. These studies indicate that while most CRBare able to degrade perchlorate, there is a subset of CRBthat cannot use perchlorate as an electron acceptor forcell respiration.

Perchlorate reducing isolates

Many PRB have now been isolated (Table 1). ThesePRB are all Gram-negative, facultative anaerobes. Anal-ysis of the 16S rRNA sequences of tested strains indi-cated that all isolates were members of the class Pro-teobacteria. The majority of these PRB, including GR-1,perc1ace, KJ, PDX, HZ, JDS5, and 15 other strains(Achenbach et al., 2001), are located in the b-subclassof Proteobacteria. Achenbach et al. (2001) proposed twonew genera, Dechloromonas and Dechlorosoma, forthese b-subclass lineages, which represent the predomi-nant PRB in the environment. The phylogenetic tree inFig. 1 shows the general positions of several PRB in com-parison to other bacteria.

MICROBIAL DEGRADATION OF PERCHLORATE 407

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

Table 1. Perchlorate-respiring bacterial isolates.

Isolate References

Vibrio dechloraticans Cuznesove B-1168 Korenkov et al. (1976)Wolinella succinogenes HAP-1 Wallace et al. (1996)Dechloromonas agitata CKB Bruce et al. (1999)Dechloromonas sp. SIUL Coates et al. (1999); Coates et al. (2000)Dechloromonas sp. MissR Coates et al. (1999); Coates et al. (2000)Dechloromonas sp. CL Coates et al. (1999); Coates et al. (2000)Dechloromonas sp. NM Coates et al. (1999); Coates et al. (2000)Dechlorosoma sp. SDGM Coates et al. (1999); Coates et al. (2000)Dechlorosoma sp. PS Coates et al. (1999); Coates et al. (2000)Dechloromonas sp. JM Miller and Logan (2000)Dechloromonas sp. HZ Zhang et al. (2002)Dechlorospirilium anomalous WD Michaelidou et al. (2000)Dechlorosoma sp. KJ Logan et al. (2001c)Dechlorosoma sp. PDX Logan et al. (2001c)Dechlorosoma sp. GR-1 Rikken et al. (1996)Dechlorosoma sp. perc lace Herman and Frankenberger (1999)Citrobacter sp. IsoCock1 Okeke et al. (2002)Dechloromonas sp. JDS5 Shrout and Parkin (2002)

Electron donors and acceptors used by PRB for growth

All PRB are capable of dissimilatory reduction of chlo-rate to chloride for energy and growth, and many PRBcan also reduce nitrate. For both perchlorate and chlo-rate, reduction does not occur in the presence of a highconcentration of dissolved oxygen. Most isolates can useoxygen and many can respire using nitrate as a terminalelectron acceptor. HAP-1 was initially reported to be anobligate anaerobe (Wallace et al., 1996), but in a laterstudy it was reported that it was a microaerobic organ-ism (Wallace et al., 1998), although no further details onoxygen tolerance were provided. It seems reasonable thatPRB should be able to use, or at least tolerate, dissolvedoxygen as it is produced during the decomposition ofchlorate and perchlorate. Although many PRB are capa-ble of complete denitrification, CKB does not grow onnitrate (Bruce et al., 1999), and Cuznesove B-1168 (Ko-renkov et al., 1976) and HAP-1 (Wallace et al., 1996)reduce nitrate only to nitrite and do not produce ammo-nia or nitrogen gas.

In some cases PRB have been tested for their ability touse other electron acceptors such as metals and sulfate.GR-1 can utilize Mn (IV) as an electron acceptor. Most

PRB and CRB cannot reduce sulfate (Riken et al., 1996;Wallace et al., 1996; Bruce et al., 1999; Coates et al., 1999;Herman and Frankenberger, 1999; Michaelidou et al.,2000; Wolterink et al., 2002). The only report of CRB(Acinetobacter spp.) capable of growth using sulfate wasprovided by Stepanyuk et al. (1992), but the ability torespire perchlorate was not tested and these bacteria couldnot use nitrate. PRB have not been found to be capable ofusing other electron acceptors, including: Fe (III), selenate,malate, and fumarate (Coates et al., 1999).

Both heterotrophic and autotrophic PRB have been iso-lated. Acetate has been most frequently used as a singlesubstrate for heterotrophic perchlorate reduction (Ko-renkov et al., 1976; Wallace et al., 1996; Bruce et al.,1999; Coates et al., 1999; Herman and Frankenberger,1999; Logan et al., 2001c), but hydrogen or formate wasrequired as an electron donor for the growth of HAP-1on acetate (Wallace et al., 1996). Perchlorate reductionby Citrobacter sp. IsoCock1 (Okeke et al., 2002) wassustained on acetate, but yeast extract was found to im-prove growth. A wide variety of organic substrates, in-cluding alcohols and carboxylic acids, can be used asgrowth substrates by PRB although the use of these sub-strates is strain-dependent. Dechloromonas sp. JM(Miller and Logan, 2000), a strain isolated from the bac-terial consortium in an autotrophic packed-bed biofilmreactor, reduced perchlorate using dissolved hydrogen,but could not grow using hydrogen as the sole electrondonor. Two autotrophic isolates, Dechloromonas sp. HZ(Zhang et al., 2002) and Dechloromonas sp. JDS5(Shrout and Parkin, 2002), have recently been isolatedthat can grow in a minimal inorganic medium using per-chlorate, hydrogen, CO2, and nutrients.

Nutritional requirements for PRB

There is no detailed information on the best mediumto use for PRB or what trace nutrients or metals areneeded for growth. Several research groups have used aphosphate buffer system (Attaway and Smith, 1993;Rikken et al., 1996; Wallace et al., 1996; Herman andFrankenberger, 1999; Logan et al., 2001c; Wu et al.,2001), while others (Bruce et al., 1999; Coates et al.,1999) have used a bicarbonate-buffered freshwatermedium amended with a complex vitamin solution. Thereis no evidence that these additional vitamins are neces-sary for PRB growth, but in one study it was shown thatseveral PRB strains could not grow without the tracemetal solution (Michaelidou et al., 2000). Iron, molyb-denum, and selenium appear to be important for PRBgrowth and perchlorate degradation. Perchlorate reduc-tase purified from GR-1 was found to contain 11 mol ofiron, 1 mol of molybdenum, and 1 mol of selenium permol of heterodimer (Kengen et al., 1999). Bender et al.

408 XU ET AL.

Figure 1. Neighbor-joining phylogenetic tree based on the 16SrRNA sequences of PRB (Dechloromonas and Dechlorosoma)and others. Bootstrap values (100 replicates) are indicated on thenodes. The prealigned 16S rRNA sequences in TREECON for-mat (Van de Peer and De Wachter, 1994) were downloaded fromthe rRNA WWW Server (HTTP://RRNA.UIA.AC.BE/)

(2002) also reported that molybdenum was important forPRB growth. Wallace et al. (1996) used yeast extract andpeptone in their medium for HAP-1. Zhang et al. (2002)found that yeast extract improved the growth of the au-totrophic PRB strain HZ, but that yeast extract was notneeded for growth.

It is likely that the different trace nutrients used forlaboratory media are not needed for bioremediation oflow perchlorate concentrations in natural systems. As ex-plained below, several field studies have achieved per-chlorate degradation only through the addition of an ox-idizable substrate (acetate, ethanol, etc.), nitrogen, andphosphorus (Green and Pitre, 2000; Hatzinger et al.,2000; Logan et al., 2001a; Evan et al., 2002; Min et al.,2003).

THE PERCHLORATE DEGRADATION PATHWAY

Much of what is known about bacterial degradation ofperchlorate has resulted from earlier studies on chloratereduction. Quastel et al. (1925) found that chlorite wasproduced from chlorate by one strain of Escherichia coliwithout further reduction of chlorite (ClO2

2). The fail-ure of chlorate respiration in this strain was likely due tothe toxic effects of chlorite and an absence of the enzymechlorite dismutase. Later, it was found that some bacte-ria could reduce chlorate to chloride (Aslander, 1928;Bryan and Rohlich, 1954). The reduction of perchlorateto chloride by several species of heterotrophic bacteriawas first demonstrated with the use of 36Cl-labeled per-chlorate (Hackenthal et al., 1964). Hackenthal (1965)concluded that chlorate was the product of perchloratereduction by cell free extracts obtained from nitrate-

adapted cells of Bacillus cererus. They also found thatchlorate could be reduced by the same cell free prepara-tion, and that it competitively inhibited perchlorate reduction. The first proposed perchlorate reduction path-way was ClO4

2 ClO32 ClO2

2 Cl2O2 Cl2 (Hack-enthal et al., 1964; Hackenthal, 1965). The reduction ofperchlorate or chlorate to chloride by bacteria was sub-sequently confirmed by other researchers (Korenkov etal., 1976; Malmqvist et al., 1991; Rikken et al., 1996;Coates et al., 1999; Wu et al., 2001).

Research on the perchlorate reduction pathway did notmake further progress until a new enzyme, chlorite dis-mutase, was purified from the PRB strain GR-1 and foundto produce oxygen from chlorite (Rikken et al., 1996; vanGinkel et al., 1996). Rikken et al. (1996) proposed athree-step mechanism of perchlorate reduction (Fig. 2) inwhich chlorate, chlorite, and dissolved oxygen were se-quentially produced. This pathway, which is now widelyaccepted for bacterial respiration using perchlorate andchlorate, is: ClO4

2 ClO32 ClO2

2 O2 1 Cl2. Chlo-rite dismutase has been isolated from two other bacteria,Ideonella dechloratans, and strain CKB. Stenklo et al.(2001) purified chlorite dismutase from I. dechloratansthat had catalytic properties that were similar, but notidentical, to those found for the chlorite dismutase en-zyme obtained from GR-1. Stenklo et al. (2001) deter-mined the 22-residue N-terminal amino acid sequence forthis enzyme and found no homologue in the protein se-quence of the purified enzyme from I. dechloratans.Coates et al. (1999) also purified a chlorite dismutasefrom the strain CKB, which had characteristics similar tothe enzyme purified from strain GR-1. The finding byCoates et al. (1999) that all 13 CRB isolates obtained intheir laboratory could disproportionate chlorite into chlo-ride and oxygen makes it likely that chlorite dismutase

MICROBIAL DEGRADATION OF PERCHLORATE 409

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

Figure 2. Perchlorate reduction pathway (adapted from Rikken et al., 1996)

is the central enzyme for the dissimilatory reduction ofperchlorate to chloride in PRB.

The impact of nitrate on perchlorate reduction is im-portant because nitrate is a common contaminant of per-chlorate-contaminated waters. Many PRB are capable ofpartial or complete denitrification, and the presence ofnitrate usually decreases the rate of perchlorate reduction.The similarities between the perchlorate reduction anddenitrification pathways led to an early suggestion thatthe perchlorate and nitrate pathways were catalyzed bythe same nitrate reductase (Hackenthal et al., 1964;Stouthamer, 1967). In addition, it was found that per-chlorate reductase obtained from GR-1 reduced nitrate aswell as chlorate and perchlorate (Kengen et al., 1999).

The concept of a shared enzyme for nitrate and per-chlorate reduction is unlikely given more recent findings.It has been found that some PRB, for exampleDechloromonas agitata CKB (Bruce et al., 1999), canrespire using perchlorate but not nitrate. In a PRB capa-ble of denitrification (strain perclace), it was found thatnitrate and perchlorate reductases were located in differ-ent cell fractions (membrane and periplasmic fractions,respectively), indicating that these two enzymes wereseparate (Giblin and Frankenberger, 2001). Further evi-dence of completely separate denitrifying and perchlo-rate pathways was provided by Xu et al. (2002). Theseresearchers found that perchlorate degradation and deni-trification in Dechlorosoma sp. KJ and PDX were sepa-rately induced. When cells were grown on only perchlo-rate, there was minimal nitrate reduction. Also, there wasno perchlorate or chlorate reduction when cells were ini-tially grown on only nitrate. However, the presence ofboth perchlorate and nitrate in the medium stimulated thedegradation of both electron acceptors. Their kinetic re-sults were supported by whole cell protein profiles usingSDS-PAGE that showed the appearance of bands in thegels at locations expected for perchlorate reductase, chlo-rite dismutase, and nitrate reductase under conditions thatagreed with the kinetic studies (Xu et al., 2002).

It is still not clear if only a single enzyme is used byPRB for chlorate and perchlorate reduction, or if thereare separate enzymes used for perchlorate and chloratereduction. Research by Kengen et al. (1999) suggestsonly one enzyme is needed for chlorate and perchloraterespiration. They found that a single enzyme could cat-alyze both chlorate and perchlorate reduction, and thatthis oxygen-sensitive enzyme was located in theperiplasm. However, the maximum reaction rates (mea-sured with methyl viologen) for perchlorate (3.8 U/mg)were actually less than those with either nitrate (6.2U/mg) or chlorate (11.3 U/mg) (Kengen et al., 1999). Be-cause it appears likely that separate enzymes are used fornitrate and perchlorate (as explained above), it may also

be that there are separate chlorate and perchlorate en-zymes. Additional evidence for separate enzymes is pro-vided indirectly by the fact that not all CRB are capableof respiration with perchlorate, although this questionwill require further research to resolve.

MICROBIAL TREATMENT PROCESSES—BENCH SCALE PROCESSES

The first bench-scale treatment system reported to de-grade perchlorate was a suspended growth system de-veloped to treat high concentrations of perchlorate (Att-away and Smith, 1993); this system has been adequatelydescribed in previous reviews (Herman and Franken-berger, 1998; Logan, 1998). More relevant for drinkingwater treatment and bioremediation studies are reports onfixed and fluidized bed bioreactors primarily designed totreat low concentrations of perchlorate, and laboratorybatch tests to evaluate the feasibility of in situ bioreme-diation. These different bench scale studies are summa-rized in Table 2.

Heterotrophic fixed-bed reactor studies

The first reported fixed-bed bioreactor was an up-flowanaerobic reactor inoculated with a mixed culture con-taining Wollinella succinogenes HAP-1 (Wallace et al.,1998). This system was designed to treat high perchlo-rate concentrations (about 10%) found in rocket washoutwastewaters. The reactor was made of acrylic tubing(1.17 m in length; 7.6 cm inside diameter) filled with di-atomaceous earth pellets (mean pore diameter of 20 mm).The substrate for the mixed culture (BYF-100) consistedof 54% naturally occurring protein, peptides, free aminonitrogen, vitamins, and trace elements. Perchlorate wasreduced to ,300 mg/L over the entire 43-day period ofoperation at hydraulic retention times (HRT) and influ-ent perchlorate concentrations of 1.17 h and 1500 mg/L,and 0.46 h and 500 mg/L, respectively. For both influentconcentrations, the effluent concentration was below 100mg/L (the detection limit in this system) for 95% of thesamples taken during the reactor operation.

Perchlorate was found to be degraded to lower con-centrations in a sand media bioreactor inoculated withperc1ace, which was isolated from wastewater biosolids(Herman and Frankenberger, 1999). The reactor was aglass column (2.8 cm diameter; 14 cm long) packed withsterilized, oven-dried sand (40 to 70 mesh size). Follow-ing a start up period, it was found that perchlorate couldbe degraded from a concentration of 130 mg/L to belowthe detection limit (4 mg/L) at an HRT of 3 h. It was alsoreported that a celite-packed bioreactor inoculated withthe same isolate completely removed perchlorate from

410 XU ET AL.

738 mg/L at a flow rate of 1 ml/min (Giblin et al., 2000a).When the flow was increased to 2 mL/min, 92 to 95%of the perchlorate was removed.

Perchlorate concentrations of 20 mg/L were found tobe completely removed in a sand-packed bioreactor (2.5cm diameter; 28 cm long) inoculated with a perchlorate-degrading enrichment (Kim and Logan, 2000). The en-richment was developed with acetate and perchlorate us-ing primary digestor solids from a wastewater treatmentplant. Perchlorate was removed to nondetectable levels(,4 mg/L) over 35 days, after a 10-day startup period, atHRTs ranging 18 to 51 min. The same sand column sys-tem was also used to examine perchlorate degradation

rates achieved when the reactor was inoculated with apure culture of Dechlorosoma sp. KJ vs. that obtainedusing a perchlorate-degrading enrichment (Kim and Lo-gan, 2001). The reactor inoculated with the pure cultureremoved perchlorate from 20 mg/L to less than 4 mg/Lat a much shorter HRT of ,1 min than the mixed cul-ture reactor (12 min). Acetate was used as the substratefor the culture, and the molar ratio (6.6, n 5 156) of ac-etate-to-perchlorate for the pure culture was more thantwice the ratio (2.9, n 5 6) of the mixed culture.

Perchlorate degradation using granular-activated car-bon (GAC) in a bioreactor was compared to reactor per-formance with sand media in fixed-bed reactors (Kim and

MICROBIAL DEGRADATION OF PERCHLORATE 411

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

Table 2. Bench-scale bioreactors for treating perchlorate in water.

PerchlorateReactor type (media) Inoculum Substrates (mg/L) References

Suspended growth Wolinella succinogenes Protein nutrientsa 7750 Attaway and SmithHAP-1 in mixed culture (1993)

Fixed bed Wolinella succinogenes BYF-100b 500, 1500 Wallace et al. (1998)(diatomacous HAP-1 in mixed cultureearth pellets)

Fixed bed (sand Perclace Acetate 0.13, 0.738 Herman andor Celite) Frankenberger (1999);

Giblin et al. (2000a)Fixed bed (sand Mixed culture Acetate 20 Kim and Logan (2000)

GAC)Fixed bed (GAC) Tapwater A mixture of acetate, 0.051–0.55 Brown et al. (2000)

lactate, and pyruvateFixed bed (sand) Dechlorosomas sp. KJ Acetate 20 Kim and Logan (2001)Fixed bed Primary sludge and Acetate 0.1, 1 Burns et al. (2001)

(cylindrical effluent from waste-pall rings) water treatment plant

Fixed bed Perclace Acetate 0.8 Losi et al. (2002)(Celite pellets)

Fixed bed Mixed culture Hydrogen 0.740 Miller and Logan(2000)

Fixed bed (Celite Biosolids from the Hydrogen 0.740 Giblin et al. (2000b)R-635) municipal water

treatment plantFixed bed (glass Mixed culture Hydrogen 0.073 Logan and LaPoint,

beads) (2002)Fluidized bed (sand Biological solids from Ethanol, methanol, 25 Greene and Pitre,

or GAC) an anaerobic digester or a mixture of the (2000); Hatzingertwo alcohols et al. (2000)

Fludized bed (GAC) Mixed culture Acetic acid and ethanol 11–23 Togna et al. (2001)Hollow-fiber Ralstonia eutropha Hydrogen 0.006–0.100 Nerenberg et al.

membrane (2002)bioreactor

aAged brewer’s yeast, cottonseed protein, or whey power; bNaturally occurring protein (54%), peptides, free amino nitrogen, vitamins, and trace elements.

Logan, 2000). The GAC media (12 3 40 mesh-sieved)had a bulk density of 0.44 g/cm3 (NORIT Americas Inc.,Atlanta, GA), while the sand had an average size of 0.425mm (density of 2.67 g/cm3). Both reactors were fed thesame perchlorate-degrading enrichment. The GAC-packed reactor completely removed perchlorate for thefirst 4 days of operation, but after backwashing the me-dia on day 6, the reactor performed erratically with com-plete perchlorate removal on some days and detectableperchlorate concentrations on others. The poor operationof the GAC reactor was attributed to desorption of per-chlorate from the GAC in the lower parts of the reactorwhere biological activity was insufficient to degrade theperchlorate. Based on these results, the researchers rec-ommended that GAC would not be suitable for fixed-bedbioreactors that would need backwashing to controlbiofilm formation and prevent clogging.

A slightly different GAC system was examined byBrown et al. (2000) for perchlorate degradation. They ex-amined the use of biologically active carbon (BAC) fil-ters for perchlorate removal at low influent perchlorateconcentrations (Brown et al., 2000). The BAC filterswere constructed of glass pipes 2.54 cm in diameter andTeflon™ endcaps. The filters were rendered biologicallyactive by feeding 600 bed volumes (BV) of dechlorinatedtapwater containing 50 mg/L perchlorate and 20 mg/Lbromate. Deionized water containing 50–55 mg/L per-chlorate was then fed to the column for 5,800 empty BV,with perchlorate removal occurring mainly by sorptionto the carbon. Perchlorate removal was then stimulatedby adding a mixture of acetate, lactate, and pyruvate (2mg/L as carbon) to the water that also contained 2.5 mg/Lof dissolved oxygen and nitrate. Perchlorate removal wasfound to be a function of the concentration of nitrate inthe water. Perchlorate was removed from 52 mg/L to be-low the detection limit, for an influent nitrate concentra-tion of 1.4 mg/L (0.07 mg/L in the effluent). However,perchlorate was not appreciably removed (from 51 to 50mg/L) at a higher influent nitrate concentration of 4.5mg/L (2.0 mg/L in the effluent). The effects of nitrate onperchlorate reduction are further discussed below.

Burns et al. (2001) reported that two fixed-bed biore-actors placed in series could remove perchlorate from aninfluent concentration in the range of 0.1 to 1 mg/L, tobelow the detection limit even in the presence of highconcentrations of nitrate (50 mg/L). Both reactors (70mm diameter; 460 mm in a height) initially inoculatedwith primary sludge were fed a synthetic solution con-taining acetate (250 mg/L) and sodium sulfite (to removedissolved oxygen in the reactor influent). HRTs of 1 h(0.1 mg/L) and 10 h (1 mg/L) were needed for completeremoval of perchlorate.

Another bench-scale packed-bed biological reactor

was examined in a 30-day study for its efficiency at re-moving low concentrations of perchlorate (,1 mg/L)from groundwater (Losi et al., 2002). The Plexiglas col-umn reactor (total volume of 3,062 mL) was filled withCelite pellets and inoculated with the perchlorate-reduc-ing bacterium perc1ace (Herman and Frankenberger,1999). Groundwater from a perchlorate-contaminated sitein California was pumped (upflow mode) into the reac-tor along with nutrients (N and P) and acetate (approxi-mately 500 mg/L). Influent perchlorate concentrations ofabout 800 mg/L were degraded to nondetectable levels(,4 mg/L) at residence times as low as 0.3 h. Acetateconcentrations were less than 50 mg/L in the effluent. Ni-trate (20 mg/L NO3

2-N) was also completely removedwhile sulfate reduction was not observed.

Heterotrophic fluidized bed bench scale reactors

Fluidized bed reactors (FBRs) have also been exam-ined for perchlorate removal in bench scale studies(Greene and Pitre, 2000; Hatzinger et al., 2000). TheFBRs contained either sand or GAC with particle sizesof 0.2 to 0.6 mm and 0.9 to 1.4 mm, respectively, as themedia. The reactors were fed ethanol, methanol, or a mix-ture of the two alcohols as substrates for perchloratebiodegradation. Ethanol or the ethanol–methanol mixtureproduced better performance than just methanol. TheFBR packed with GAC showed better performance thanthe sand reactor, and achieved consistent perchlorate re-moval from 25 mg/L to less than the detection limit of 4mg/L (Greene and Pitre, 2000; Hatzinger et al., 2000).

A laboratory-scale FBR was used to treat perchlorate-contaminated groundwater from the Longhorn Army Am-munition Plant (Togna et al., 2001). The glass column re-actor containing GAC as the fluidization medium was fedboth acetic acid and ethanol as the electron donors. Ef-fluent perchlorate concentrations were less than the de-tection limit except during a low substrate study con-ducted to determine the point of treatment failure.

Autotrophic H2 reactors

Nitrate has been degraded in hydrogen-fed bioreactorsfor many years (Gayle et al., 1989; Kapoor and Vi-raraghavan 1997), but only very recently there have beenreports of using autotrophic bioreactors for the degrada-tion of perchlorate. Miller and Logan (2000) developedan unsaturated-flow, packed-bed bioreactor for thedegradation of perchlorate in drinking water. The biore-actor was inoculated with a hydrogen-oxidizing, au-totrophic, and perchlorate degrading enrichment culturedeveloped in a chemostat. The packed-bed reactor wasfed with a gas mixture of hydrogen (5%) and carbon diox-ide and a feed stream containing only perchlorate and in-

412 XU ET AL.

organic nutrients. The reactor was initially operated fora 10-day start up period using a high concentration ofperchlorate (50 mg/L) to build up the biofilm, and wasthen switched to a lower level of perchlorate (740 6 110mg/L) for the next 145 days of operation. During this op-eration period, perchlorate was reduced to 460 6 80 mg/Lat a constant hydraulic loading rate of 0.45 cm/min(1.1–1.3 min detention time).

A slightly modified version of the reactor used byMiller and Logan (2000) was tested to examine the re-moval of perchlorate from a perchlorate contaminatedgroundwater (Logan and LaPoint, 2002). The reactor wasalso fed a gas mixture of hydrogen (5%) and carbon diox-ide as the electron donor and carbon source, respectively.An average of 25 6 5% of perchlorate was removed at aHRT of 1.5 min from a groundwater containing 73 6 2mg/L of perchlorate and 21 6 5 mg/L of nitrate. It wasestimated that a reactor with a detention time of 8 to 42min (95% C.I.) a length of 0.65 to 3.4 m would be nec-essary to completely remove perchlorate in this type ofunsaturated-flow system.

A 120-mL bioreactor packed with Celite R-635 (CeliteCorporation, Lompoc, CA) was used to degrade per-chlorate with hydrogen (Giblin et al., 2000b). The au-totrophic consortium of bacteria was fed a pressurizedwater stream containing dissolved hydrogen and bicar-bonate. Perchlorate (740 mg/L) was removed to belowthe detection limit at a 2-h detention time. Perchlorateremoval was incomplete at shorter detention times, withperchlorate concentrations of 100 and 200 mg/L and de-tention times of 1 and 0.67 h, respectively. It was spec-ulated that the perchlorate breakthroughs at the shorterdetention times resulted from nonuniform distribution ofbiomass, pH variations of the groundwater, and limitedhydrogen transport to the bacteria.

The degradation of perchlorate in a groundwater hasalso been demonstrated using a hollow-fiber membranebioreactor (Nerenberg et al., 2002). The membrane mod-ule in the reactor contained 98 hollow fibers (93 cm long;280 mm outside diameter) in a PVC pipe shell. Hydro-gen gas fed through the fiber pores was used to supporta biofilm growing on the membrane and in the liquid inthe reactor. The groundwater contained both perchlorate(6–100 mg/L) and nitrate (2.5–3 mg/L). It was found thatperchlorate could be removed to below the detection limitwhen the nitrate concentration from the effluent was lessthan 30 mg/L in the presence of at least 300 mg/L of dis-solved hydrogen in the bulk solution.

Effects of nitrate on perchlorate removal

Both perchlorate and nitrate are common contaminantsin surface and ground waters in the United States. Asmost of the PRB are also denitrifiers, some papers have

discussed the simultaneous removal of perchlorate andnitrate from contaminated waters. Nitrate has been foundto have different effects on perchlorate removal rates. Ni-trate has inhibited the perchlorate reduction rate in somestudies (Herman and Frankenberger, 1999; Brown et al.,2002) but not in others (Burns et al., 2001; Logan andLaPoint, 2002). A PRB isolate, perc1ace, was able tocompletely remove perchlorate (130 mg/L) and to simul-taneously remove more than 95% of the nitrate (20 mg/L)in a sand-packed bioreactor (Herman and Frankenberger,1999). Coppola and McDonald (2000) reported that amixed culture of W. succinogenes could completely de-grade perchlorate (1.2–1.5 g/L), chlorate (3–3.5 g/L), andnitrate (0.2 g/L) in a 7-L continuously stirred tank reac-tor at a HRT of 16 h. Nerenberg et al. (2002) showedthat perchlorate and nitrate reduction to nondetectablelevels could be concurrently achieved using a hydrogen-fed, autotrophic membrane bioreactor. In pilot-scale tests,Min et al. (2003) found that perchlorate and nitrate wereboth completely removed in a fixed-bed bioreactor. Thus,it appears likely that perchlorate and nitrate can be si-multaneously removed in bioreactors. However, the con-centration of nitrate will increase the total amount of elec-tron donor needed by the system for complete perchlorateremoval.

Soil bioremediation feasibility tests

Important questions for soil bioremediation arewhether the added substrate will be used for perchloratedegradation or lost to other biochemical routes (such asmethanogenesis), and whether there are sufficient micro-organisms in the soil to achieve perchlorate degradationusing added substrate. The first indication that perchlo-rate bioremediation could be easily accomplished werestudies that demonstrated perchlorate reduction could beachieved solely by adding substrate and nutrients towastewater and soil samples. For example, it was shownthat by adding acetate to headspace-free BOD bottles50% of 20 mg/L perchlorate was degraded within 4 days,and 74% was removed within 9 days (Kim and Logan,2000). The cultures obtained after 9 days of acclimationwere transferred (5% by volume) to freshly prepared me-dia containing acetate (200 mg/L) and either perchlorate(50 mg/L) or chlorate (50 mg/L), or same concentrationsof both anions (100 mg/L). The perchlorate and chloratedegradation rates were similar (75 and 78%, respectively)within 5 days using the acclimated cultures (Kim and Lo-gan, 2000).

Other researchers have shown that CRB are ubiquitousin soil samples, providing indirect evidence that biore-mediation to remove perchlorate would not requirebioaugmentation. For example, van Ginkel et al. (1995)found CRB in rivers, sediments, soils, and wastewater

MICROBIAL DEGRADATION OF PERCHLORATE 413

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

treatment plants (van Ginkel et al., 1995) and Coates etal. (1999) found CRB at concentrations of 2.31 3 103 to2.4 3 106 cells per gram in pristine and hydrocarbon-contaminated soils, aquatic sediments, paper mill wastesludges, and farm animal waste lagoons.

Direct evidence for the feasibility of perchlorate biore-mediation with different substrates was provided by Wuet al. (2001). They examined perchlorate degradation us-ing samples from soils, natural waters, and wastewaters.Complete degradation of 210 mg/L perchlorate wasachieved within 4 to 7 days and 8 to 29 days for rawwastewater and creek water, respectively, using four dif-ferent substrates (acetate, lactate, citric acid, or molasses).PRB were nonmeasurable using a five-tube MPN methodin some soil samples, but were present at high concen-trations at perchlorate-contaminated sites. For example,perchlorate reduction was not achieved with 10 g of “pris-tine” soil, but it was observed with 100 g of soil. Micro-organisms in 10 g of a perchlorate-contaminated soil(from a site in Texas) completely removed perchloratefrom 100 mg/L within only 7 days using polylactate orlactate.

Hunter (2001) demonstrated that injecting sandcolumns with a slowly degradable substrate (vegetableoil) could remove perchlorate without the need forbioaugmentation with perchlorate-acclimated bacteria.Columns were seeded with bacteria from soil, and per-chlorate (20 mg/L) was added to water and pumped intothe columns at a rate of ,25 mL/day. After 14 days ofoperation, 0.47 mg of soybean oil was injected onto thetreatment columns. Although perchlorate levels remainedhigh in the control columns (no soybean oil), perchloratein the effluent of the treatment columns decreased by,99% over a 17-week operation in the columns con-taining the vegetable oil.

MICROBIAL TREATMENT PROCESSES—PILOT AND FULL-SCALE PROCESSES

Fixed-bed reactors

A pilot-scale bioreactor was operated as an ex situtreatment process to treat perchlorate-contaminatedgroundwater at the Naval Weapons Industrial ReservePlant in McGregor, TX (Perlmutter et al., 2000). The re-actor was constructed of steel (5-ft diameter, 18 ft inheight) and fed water containing 7 to 20 mg/L of per-chlorate. It was found that adding only acetate (acetate:ClO4

2 ratio of 5:1) and nutrients (nitrogen and phos-phorus) at a flow rate of 20 gpm, the perchlorate con-centration could be completely removed (,20 mg/L, thedetection limit at that site).

Two pilot-scale bioreactors were operated for about 7

months to treat perchlorate-contaminated groundwater(average 76 mg/L) at the Texas Street Well Facility inRedlands, CA (Logan et al., 2001a; Evans et al., 2002;Min et al., 2003). The 7-ft tall reactors contained two rec-tangular beds each 2 ft 3 1 ft. One reactor contained sandmedia (1 mm diameter) and the other one plastic media(3.76 cm diameter). The reactors were inoculated with apure culture of Dechlorosoma sp. KJ. Acetic acid (47 6

9 mg/L for the sand media reactor; 53 6 10 mg/L for theplastic media reactor) and an ammonium phosphate so-lution (C:N ratio of 5:1) were added to the groundwater.The groundwater contained nitrate (4 mg/L), oxygen (7mg/L), and sulfate (33 mg/L) as potential terminal elec-tron acceptors in addition to perchlorate. Perchlorate wascompletely and consistently removed to nondetectablelevels (4 mg/L) at flow rates of 1 gpm in the plastic me-dia reactor, and 2 gpm in the sand reactor. Backwashing(weekly) was critical to maintain consistent perchlorateremoval rates in the sand reactor. In the absence of reg-ular backwashing, there was short circuiting in the sandreactor, resulting in insufficient groundwater detentiontime for consistent perchlorate removal.

Fluidized bed bioreactor

A pilot-scale FBR was tested in 1996 at the Aerojectfacility in Sacramento, CA, using water containing 7,000to 8,000 mg/L of perchlorate and 1.5 mg/L of nitrate-ni-trogen (Catts and McCullough, 1998). There are few de-tails available on this system, but it was reported that ef-fluent concentrations of ,400 mg/L were achieved forperchlorate, and ,50 mg/L for nitrate-nitrogen. Ethanolwas added in proportion to perchlorate at a molar ratioof 4:1.

In 1998, a second FB bioreactor was tested at the Bald-win Park site in the San Gabriel Basin in California. Thereactor (20 in high, 15 ft long) contained activated car-bon (10 3 30 mesh), and was similar to one used at theAerojet site, except this reactor was inoculated withsludge from a food processing industry (vs. a wastewatertreatment plant). The reactor was fed groundwater (30gpm, 114 L/min) amended with ethanol (40 to 70 mg/L)and nutrients. Effluent ethanol concentrations rangedfrom ,10 mg/L to nondetectable concentrations. Per-chlorate (50 to 100 mg/L) removal was generally greaterthan 90% when dissolved oxygen was low (near 1 mg/L).At other times, removal varied from 23 to 45%. Nitrate(5 to 6 mg/L NO3

2-N) removal was generally greaterthan 99%. The denatured ethanol used for the study con-tained low concentrations of methanol, methyl isobutylketone and isopropyl alcohol (Catts and McCullough,1998). These chemicals can be toxic to humans, andtherefore the addition of these chemicals to drinking wa-ter sources should be avoided. Nitrosodimethylamine

414 XU ET AL.

present in the water at 70 to 80 mg/L was not affected bythe bioreactor treatment.

A full-scale FBR was operated to treat perchlorate ingroundwater at a site in Rancho Cordova, CA (www.en-virogen.com/perchlorate.htm; Green and Pitre, 2000;Hatzinger et al., 2000). The system consisted of fourFBRs each 14-ft in diameter treating 4,000 gpm ofgroundwater. The reactors were filled with GAC, inocu-lated with a perchlorate-degrading enrichment, and fedethanol as the electron donor. Following a startup periodof 3 weeks, the effluent concentration of perchlorate wasremoved from 6 to 8 mg/L to less than the detection limit(4 mg/L) for 8 weeks of operation during periods of ex-cess ethanol addition. Periods of incomplete perchlorateremoval occurred at lower ethanol doses (Green and Pitre,2000; Hatzinger et al., 2000).

FBRs have recently been approved for treatment ofperchlorate contaminated drinking water by the Califor-nia Department of Health Service (DHS, 2002a). How-ever, the presence of methanol and other impurities inethanol may prohibit the use of ethanol as a substrate inthese bioreactors if they are to be used for drinking wa-ter treatment.

In situ bioremediation

Pilot-scale tests for perchlorate remediation haveshown that perchlorate reduction can be achieved byadding various electron donors to create anoxic condi-tions in the subsurface. An excess of an electron donorcan be used for the degradation of perchlorate, nitrate,and other pollutants. In situ bioremediation field tests ofperchlorate were conducted at the Naval Weapons In-dustrial Reserve Plant in McGregor, TX (Perlmutter etal., 2000). Trenches were dug at the site to collect ground-water and direct the flow of this water through moundsof material containing various organic sources such ascompost (from an edible mushroom production facility)and cottonseed meal. Perchlorate in the water was re-moved from 16–27 mg/L to less than 100 mg/L (detec-tion limit in this study) after only 2 weeks of operation.Nitrate was also reduced from 15 mg/L to nondetectableconcentrations.

Cox et al. (2001) reported that perchlorate in situbiodegradation was successfully achieved in a deepaquifer at the Aerojet Superfund site in California. Ac-etate was added into the aquifer to support perchloratereduction. Perchlorate concentrations in the groundwaterdecreased from 12 mg/L to less than the detection limit(4 mg/L) within 15 feet of the electron donor deliverywell. In addition to perchlorate, trichloroethylene, an-other common contaminant of groundwater, was com-pletely dechlorinated to ethane.

MICROBIAL TREATMENT PROCESSES—SYSTEMS USED IN CONJUNCTION WITH

OTHER PROCESSES

Processes such as anion exchange (Venkatesh et al.,2000; Gu et al., 2001), membrane filtration (Urbanskyand Schock, 1999), and activated carbon (Brown et al.,2002) have been studied for the removal of perchlorate,especially for treating waters contaminated with low(,100 mg/L) concentrations of perchlorate. For example,anion exchange resins have been shown to be highly ef-fective for perchlorate removal (Urbansky, 2000a). Afield experiment has shown that one bed volume of aunique bifunctional resin can treat greater than 100,000BV of groundwater with an initial perchlorate concen-tration of 50 mg/L (Gu et al., 2000). Because these phys-ical/chemical techniques do not degrade perchlorate, fur-ther treatment using chemical or biological processes isnecessary. Perchlorate in the brine wastes can be reducedto chloride by chemical methods using catalytic air-sen-sitive metal cations such as titanium or ruthenium (Abu-Omar et al., 2000; Espenson, 2000), or by electrochem-ical reduction (Urbansky, 1998). One company (CalgonCarbon) has patented a treatment process (Venkatesh etal., 2000) for perchlorate removal that uses ion exchangeto remove perchlorate and a bioreactor or catalytic reac-tor for complete perchlorate destruction. This treatmentprocess has been approved for treatment of perchloratecontaminated water by the California Department ofHealth Service (http://www.calgoncarbon.com/indus-try/productdata.php?id511).

Biological treatment of concentrated waste streamsfrom ion exchange processes can be difficult to treat us-ing biological systems due the high salt content of thebrine. In the case of an anion exchange process, the re-generation of the resin typically generates a 7–12% NaClbrine solution enriched in perchlorate. Gingras andBatista (2002) were unable to adapt a PRB culture to de-grade perchlorate in an ion exchange brine. As little as1% NaCl reduced perchlorate reduction rates by their per-chlorate-degrading culture by half (Gingras and Batista,2002). High NaCl systems may not be required for cer-tain types of specialized ion exchange resins. In one sys-tem recently developed by Gu et al. (2001), the resin wasregenerated using a tetrachloroferrate displacement tech-nique. Using this method, only 5 BV of regenerant solu-tion were needed to achieve nearly 100% recovery of theresin. Large-scale field application of this technique isreported to be under way (Gu et al., 2001).

Although biological degradation of perchlorate in ionexchange brines can be difficult, some bacteria can growand degrade perchlorate even in highly saline solutions.A culture containing primarily W. succinogenes HAP-1

MICROBIAL DEGRADATION OF PERCHLORATE 415

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

was reported to degrade perchlorate in brines derivedfrom the regeneration of ion exchange resins (7% NaCl,180 mg/L perchlorate) (Coppola and McDonald, 2000).However, the 7% brine solution had to be diluted to 3%with water to allow biodegradation of the perchlorate.This dilution is undesirable, as it increases the total vol-ume of salt solution requiring disposal. Logan et al.(2001b) demonstrated that the biological reduction ofperchlorate was possible at salinities of up to 11% NaCl.The perchlorate-degrading enrichment was developed us-ing samples from The Great Salt Lake. Although per-chlorate was degraded in these highly saline solutions,the degradation rates were slow. Okeke et al. (2002) iso-lated a salt tolerant PRB strain, Citrobacter sp. IsoCock1,that was able to achieve a 32% reduction of perchloratein a 7.5% solids solution in 1 week at an initial perchlo-rate concentration of 0.5 g/L. A 46% reduction of per-chlorate was achieved by a mixture of this bacterium andthe strain perc1ace (Okeke et al., 2002).

Reverse osmosis (RO) can also be used to remove per-chlorate from drinking water. The waste stream contain-ing perchlorate that was produced from an RO processcontained much lower amount of salts (,1%) than theNaCl brine generated using an ion exchange process. Apacked bed bioreactor, inoculated with the pure cultureperc1ace, was tested for its ability to remove perchloratefrom a simulated RO rejectate (Giblin et al., 2002). Itwas found that this system removed 98% of perchloratefrom a twice-concentrated rejectate (total dissolved solidsof 0.4%) with an influent perchlorate concentration of 8mg/L and a residence time of 2 h. Nitrate was removedsimultaneously with perchlorate from an initial concen-tration as high as 900 mg/L to below 4 mg/L. Despite theefficiency of perchlorate removal, the system sufferedfrom clogging due to the high total dissolved solids ofthe twice-concentrated rejectate.

It is possible to simultaneously remove and degradeperchlorate using BAC. BAC has been used in drinkingwater treatment to remove trace amounts of halo-aceticacids in drinking water (Xie and Zhou, 2002). Kim andLogan (2000) reported that GAC could be used to treatwater containing 10 mg/L of perchlorate. They found thatperchlorate was completely removed by the reactor at thebeginning of operation, but that after backwashing reac-tor efficiency was substantially reduced due to perchlo-rate desorption from the carbon in regions of low bio-logical activity near the end of the column. At much lowerinfluent perchlorate concentrations of 50 mg/L, Brown etal. (2002) found, over a 103-day study, that it was pos-sible to remove perchlorate to below the detection limitin a BAC operated at a 25-min empty bed contact time.Although the BAC was not bioaugmented with PRB, theyfound that Dechloromonas and Dechlorosoma were pres-

ent in the carbon bed during the period when perchlorateremoval was successful (Lin et al., 2002).

PHYTOREMEDIATION

Perchlorate contamination of the environment may af-fect agricultural plants as well as naturally occurring flora(U.S. EPA, 2002). Chlorate has been used as a defoliant(van Wijk and Hutchinson, 1995), and therefore, it is notsurprising that perchlorate can also be taken up by plants.The accumulation of perchlorate in plants is of concernfor several reasons. Perchlorate can be toxic to someplants, if the perchlorate accumulates and is not degraded,and the death of the plant may release perchlorate backinto the environment that could be toxic to other plantsor wildlife (Urbansky et al., 2000). Perchlorate accumu-lation in food plants could present another route of hu-man exposure to perchlorate. Perchlorate contaminatedwater, such as Lake Mead or the Colorado River, ispresently used for irrigating food crops (Susarla et al.,1999; Urbansky et al., 2000). Hutchinson et al. (2000)are currently studying the accumulation of perchlorate inlettuce irrigated with perchlorate-tainted water.

Because perchlorate can accumulate in plants, phy-toremediation has been suggested as a potential mecha-nism for degrading perchlorate in soil systems. Phytore-mediation may occur by phytoextraction (accumulationin the branches and leaves), phytodegradation, or rhizo-transformation (degradation in the root sphere primarilydue to microbial activity). Although many plants haveshown the ability to accumulate perchlorate, some plantscan drive perchlorate degradation completely to chloride(Nzengung et al., 1999; Susarla et al., 1999; Hutchinsonet al., 2000; Nzengung and Wang, 2000). Nzengung andWang (2000) found that willow trees could degrade 100mg/L of perchlorate in 53 days, and that minced spinachand tarragon leaves could degrade 7 mg/L of perchloratein 30 days. There were no lag times for perchlorate degra-dation in either experiment. Perchlorate degradation byplants was found to occur in two stages (Nzengung andWang, 2000). The first stage consisted of an initial up-take of perchlorate proportional to the water uptake bythe plant, and a slow transformation of perchlorate tochloride in the plant tissues. The second stage was char-acterized by a rapid removal of perchlorate by degrada-tion in the root zone with little perchlorate taken up bythe plant. This second stage was assumed to arise fromthe stimulation of growth of PRB in the rhizosphere(Nzengung et al., 1999; Nzengung and Wang, 2000).

Several phytoremediation studies have been conductedto study the rate of perchlorate uptake and degradation,and the effects of other chemicals on degradation rates.

416 XU ET AL.

Nzengung et al. (1999) examined removal and degrada-tion of perchlorate in water amended with acetate anddiffering amounts of nitrate using sand or hydroponic sys-tems cultivated with willow (Salix nigra). Both systemtypes were capable of removing perchlorate from the wa-ter. Degradation was found to occur due to transforma-tion in the leaves and stems, and degradation by the mi-crobial population in the plant rhizosphere. Multipledosing of sand systems with 100 mg/L perchlorate, andless than 100 mg/L of nitrate, achieved a maximumdegradation rate of perchlorate of 2.35 mg/L-h (Nzen-gung et al., 1999). Krauter (2001) constructed a wetlandsystem to degrade both nitrate and perchlorate in ground-water. The system consisted of four tanks containinggravel and planted with a variety of indigenous wetlandplants. Influent water contained nitrate (68 mg/L), per-chlorate (4.5 mg/L), and trichloroethene (54 mg/L), but itwas further amended with sodium perchlorate (0.1 mg/L).It was discovered that although the young wetland sys-tem was capable of supplying sufficient carbon to sup-port denitrification, perchlorate reduction was carbon-limited. When acetate (260 mg/L) was added to theinfluent, both nitrate and perchlorate were completely de-graded within 52 h. The maturity of the plants in thebioreactor and the time of year were found to greatly in-fluence the nitrate and perchlorate degradation rates inthe bioreactor (Krauter, 2001).

CONCLUSIONS

Many heterotrophic biological treatment systems havebeen tested to degrade perchlorate including suspended,fixed-bed, and fluidized-bed reactors (Attaway andSmith, 1993; Wallace et al., 1998; Attaway et al., 1999;Green and Pitre, 2000; Coppola and McDonald, 2000;Hatzinger et al., 2000; Perlmutter et al., 2000; Logan etal., 2001a; Evans et al., 2002). Organic electron donorsthat have been used include simple compounds such asacetate and ethanol, as well as more complex organic sub-strates such as those found in compost piles. Perchloratedegradation has also been achieved in bioreactors usingonly inorganic amendments. These reactors are sustainedby hydrogen gas delivered by pressurization, gas trans-fer across liquid films, or synthetic membranes (Giblinet al., 2000b; Miller and Logan, 2000; Nerenberg et al.,2002). These hydrogen-based technologies are promisingtechnologies for water treatment because less biomass isproduced by autotrophic processes than heterotrophicprocesses. Large-scale tests are needed to evaluate pro-cess efficiency and the economics of these different hy-drogen-based systems.

Although at least one biological treatment process has

been approved for use in the state of California for drink-ing water treatment (DHS, 2002a), little has been doneto study the removal of PRB from the treated water.Membrane bioreactors can be used to keep the bacteriaseparated from the contaminated water (Batista and Liu,2001), but these systems are at a less advanced stage ofdevelopment than other biological perchlorate treatmentsystems. It has been suggested that reactors based on en-zymes to reduce perchlorate could avoid the potentialhealth problems associated with biological treatment(Betts, 1999). Perchlorate reductase, which can reduceboth perchlorate and chlorate (Kengen et al., 1999), andchlorite dismutase (van Ginkel et al., 1996; Coates et al.,1999; Stenklo et al., 2001) have both been purified. How-ever, no such enzyme-based systems have been reportedin the literature for treatment of perchlorate contaminatedwater.

The presence of alternate electron acceptors in per-chlorate contaminated water will be an issue for all typesof biological reactors. Oxygen is an important interme-diate in the perchlorate degradation pathway (Rikken etal., 1996). It is well known that for PRB oxygen is a pref-erential electron acceptor to perchlorate, and that highconcentrations of dissolved oxygen inhibit perchlorate re-duction (Kengen et al., 1999; Song and Logan, 2002). Itis not clear what concentration of dissolved oxygen willcompletely inhibit perchlorate reduction, how long bac-teria can withstand exposure to high concentrations ofoxygen before losing the ability to reduce perchlorate, orhow long it would take oxygen-exposed bacteria to re-gain the ability to reduce perchlorate. However, the pres-ence of oxygen, nitrate, or sulfate in bioreactor feedstreams does not appear to be a problem for the steadyoperation of such systems. In a pilot-scale test for ex situgroundwater treatment, it was found that oxygen, nitrate,and perchlorate were all completely reduced but that sul-fate was not measurably degraded (Logan et al., 2001a;Evans et al., 2002). Thus, it is likely that the main im-pact of oxygen and nitrate on a treatment system will beto increase the requirement of substrate (such as acetateor hydrogen) that is oxidized by the bacteria.

One of the most important issues for designing a per-chlorate treatment system will be the regulatory require-ment for perchlorate removal. It now appears likely thatthe removal of perchlorate to very low levels will be nec-essary. The U.S. EPA is expected in the near future to fi-nalize a draft of its final assessment of its toxicologicaleffects of perchlorate, which could lead to recommenda-tions for perchlorate removal to ,1 mg/L for drinkingwater (Renner, 2002). The action level of perchlorate fordrinking water in several states has been set at levels of1–4 mg/L based on the release of the draft report(www.clu-in.org/studio/perchlorate-060402). If enacted,

MICROBIAL DEGRADATION OF PERCHLORATE 417

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

these regulatory levels could require perchlorate treat-ment systems for users of large water sources such asLake Mead and the Colorado River that serve major citiesin California, Utah, and Arizona (Motzer, 2001). Both bi-ological and chemical treatment systems can be used totreat perchlorate contaminated water to low levels. Themost appropriate system will likely be site and case-spe-cific, with economic, social, and political factors playinga role in the selection of each treatment system.

ACKNOWLEDGMENTS

The authors thank Husen Zhang for constructing thephylogenetic tree, and support from the National ScienceFoundation (Grants BES9714575 and BES0001900), andsupport via a grant from the United States Environmen-tal Protection Agency, the East Valley Water District, andthe American Water Works Association Research Foun-dation (AwwaRF Grant No. 2557).

REFERENCES

ABU-OMAR, M.M., MCPHERSON, L.D., ARIAS, J., andBÉREAU, V.M. (2000). Clean and efficient catalytic reduc-tion of perchlorate. Angew. Chem. Int. Ed. 39, 4310–4313.

ACHENBACH, L.A., MICHAELIDOU, U., BRUCE, R.A.,FRYMAN, J., and COATES, J.D. (2001). Dechloromonasagitata gen. nov., sp nov and Dechlorosoma suillum gen.nov., sp nov., two novel environmentally dominant (per)chlo-rate-reducing bacteria and their phylogenetic position. Int. J.Syst. Evol. Microbiol. 51, 527–533.

ASLANDER, A. (1928). Experiments on the eradication ofCanada Thistle, Cirsium arvense, with chlorates and otherherbicides. J. Agric. Res. 36, 915–928.

ATTAWAY, H., and SMITH, M. (1993). Reduction of per-chlorate by an anaerobic enrichment culture. J. Indust. Mi-crobiol. 12, 408–412.

ATTAWAY, H.H., SHANAHAN, J.F., and HURLEY, J.A.(1999). Process for bioreduction of propelleant wastewater.U.S. patent 5,948,260.

BATISTA, J., and LIU, J. (2001). Biological perchlorate re-moval from drinking waters incorporating aucroporousmembranes. The Sixth International Symposium, In Situand On-Situ Bioremediation, June 4–7, 2001 San Diego,California.

BENDER, K.S., CHAKRABORTY, R., LACK, J.G.,COATES, J.D., and ACHENBACH, L.A. (2002). Isolationand characterization of genes involved in (per)chlorate re-duction and the utility of these genes as metabolic probes. InThe 102nd general meeting of American Society for Micro-biology, Salt Lake City, Utah, USA.

BETTS, K.S. (1999). A wide variety of bugs can break downperchlorate. Environ. Sci. Technol. 33, 515A.

BROWN, J.C., SNOEYINK, V.L., and KIRISITS, M.J. (2002).Abiotic and biotic perchlorate removal in an activated filter.J. Am. Water Work Assoc. 94, 70–79.

BROWN, J.C., SNOEYINK, V.L., LIANG, S., RASKIN, L.M.,CHEE-SANFORD, J.C., and LIN, R. (2000). Removal ofperchlorate in biologically active carbon adsorption systems.The AWWA Inorganic Contaminants Workshop, February27–29, Albuquerque, New Mexico.

BRUCE, R.A., ACHENBACH, L.A., and COATES, J.D.(1999). Reduction of (per)chlorate by a novel organism iso-lated from paper mill waste. Environ. Microbiol. 1, 319–329.

BRYAN, E.H., and ROHLICH, G.A. (1954). Biological re-duction of sodium chlorite as applied to measurement ofsewage B.O.D. Sewage Ind. Wastes 26, 1315–1324.

BURNS, N.L., SCHULTE, D.D., and DAHAB, M.F. (2001).Dual removal of perchlorate and nitrate from groundwaterthrough biological reduction. In WQTC 2001 proceedings,American Water Works Association.

CATTS, J.G., and MCCULLOUGH, P.E. (1998). Draft Final:Phase 1 Treatment Study Report Perchlorate in Groundwa-ter Baldwin Park Operable Unit San Gabriel Basin. Techni-cal Report. Irvine, CA: Harding Lawson Associates Engi-neering and Environmental Services.

CHEREMISINOFF, N.P. (2001). National defense programslead to groundwater contamination: The perchlorate story.Pollut. Eng. August 2001, 38–43.

COATES, J.D., MICHAELIDOU, U., BRUCE, R.A., O’CON-NOR, S.M., CRESPI, J.N., and ACHENBACH, L.A. (1999).Ubiquity and diversity of dissimilatory (per)chlorate-reduc-ing bacteria. Appl. Environ. Microbiol. 65, 5234–5241.

COATES, J.D., MICHAELIDOU, U., O’CONNOR, S.M.,BRUCE, R.A., and ACHENBACH, L.A. (2000). The diversemicrobiology of (per)chlorate reduction. In E.T. Urbansky,Ed., Perchlorate in the Environment. New York: KluwerAcademic/Plenum Publishers, pp. 257–270.

COLLETTE, T.W., ROBARGE, W.P., and URBANSKY, E.T.(2001). Ion Chromatographic Determination of Perchlorate:Analysis of Fertilizers and Related Materials. EPA/600/R-01/026. Cincinnati, OH: U.S.EPA.

COPPOLA, E.N., and MCDONALD, G.R. (2000). Bio-degra-dation of amonium perchlorate, nitrate, hydrolysates andother energenic materials. U.S. patent 6,077,432.

COX, E.E., MCMASTER, M.L., NEVILLE, S.L., and BON-SACK, L.T. (2001). Successful field demonstration of in situbioremediation of perchlorate in groundwater. Abstract forIn situ and on-situ bioremediation in the sixth internationalsymposium, San Diego, June 4–7, 2001.

CRUMP, C., MICHAUD, P., TELLEZ, R., REYES, C., GON-ZALEZ, G., MONTGOMERY, E.L., CRUMP, K.S., LOBO,

418 XU ET AL.

G., BECERRA, C., and GIBBS, J.P. (2000). Does perchlo-rate in drinking water affect thyroid function in newborns orschool-age children? J. Occup. Environ. Med. 42, 603–612.

DHS. (2002a). Biological treatment to remove perchlorategiven conditional California DHS approval. Department of Health Services, Sacramento, California. http://www.safedrinkingwater.com/archive/sdwn051502.htm.

DHS. (2002b). Perchlorate drinking water action level and regu-lation. Department of Health Services, Sacramento, California.http://www.dhs.ca.gov/ps/ddwem/chemicals/perchl/action-level.htm.

ERICKSEN, G.E. (1983). The Chilean nitrate deposits. Am. Sci.71, 366–374.

ESPENSON, J.H. (2000). The problem and perversity of per-chlorate. In E.T. Urbansky, Ed., Perchlorate in the Environ-ment. New York: Kluwer Academic/Plenum Publishers, pp.1–8.

EVANS, P., CHU, A., LIAO, S., PRICE, S., MIN, B., and LO-GAN, B.E. (2002). Pilot testing of a reactor for perchlorate-contaminated groundwater treatment. In Proceedings ThirdInternational Conference on Remediation of Chlorinated andRecalcitrant Compounds, May 20–23, 2002.

FRANKENBERGER, W.T. (2000). Bacterial removal of per-chlorate and nitrate. U.S. patent 6,077,429.

GAUDRE-LONGERINAS, M., and TAUZIA, J.M. (2001).Process for the biological purification of a water containingammonium perchlorate. U.S. patent 6,328,891.

GAYLE, B.P., BOARDMAN, G.D., SHERRARD, J.H., andBENOIT, R.E. (1989). Biological denitrification of water. J.Environ. Eng. 115, 930–943.

GIBLIN, T., and FRANKENBERGER, W.T. (2001). Perchlo-rate and nitrate reductase activity in the perchlorate-respir-ing bacterium perclace. Microbiol. Res. 156, 311–315.

GIBLIN, T., HERMAN, D., DESHUSSES, M.A., andFRANKENBERGER, W.T. (2000a). Removal of perchloratein ground water with a flow-through bioreactor. J. Environ.Qual. 29, 578–583.

GIBLIN, T.L., HERMAN, D.C., and FRANKENBERGER,W.T. (2000b). Removal of perchlorate from ground water byhydrogen-utilizing bacteria. J. Environ. Qual. 29, 1057–1062.

GIBLIN, T., LOSI, M.E., HOSANGADI, V., and FRANKEN-BERGER, W.T. (2002). Bacterial perchlorate reduction insimulated reserve osmosis rejectate. Bioremed. J. 6, 105–111.

GINGRAS, T.M., and BATISTA, J.R. (2002). Biological re-duction of perchlorate in ion exchange regenerant solutionscontaining high salinity and ammonium levels. J. Environ.Monit. 4, 96–101.

GOLEMAN, W.L., CARR, J.A., and ANDERSON, T.A.(2002a). Environmentally relevant concentrations of ammo-nium perchlorate inhibit thyroid function and alter sex ratios

in developing Xenopus laevis. Environ. Toxicol. Chem. 21,590–597.

GOLEMAN, W.L., URQUIDI, L.J., ANDERSON, T.A.,SMITH, E.E., KENDALL, R.J., and CARR, J.A. (2002b).Environmentally relevant concentrations of ammonium per-chlorate inhibit development and metamorphosis in Xenopuslaevis. Environ. Toxicol. Chem. 21, 424–430.

GREENE, M.R., and PITRE, M.P. (2000). Treatment ofgroundwater containing perchlorate using biological flu-idized bed reactors with GAC or sand media. In E.T. Ur-bansky, Ed., Perchlorate in the Environment. New York:Kluwer Academic/Plenum Publishers, pp. 241–256.

GU, B.H., BROWN, G.M., ALEXANDRATOS, S.D., OBER,R., DALE, J.A., and PLANT, S. (2000). Efficient treatmentof perchlorate-contaminated groundwater with bifunctionalanion exchange resins. In E.T. Urbansky, Ed., Perchloratein the Environment. New York: Kluwer Academic/PlenumPublishers, pp. 165–176.

GU, B.H., BROWN, G.M., MAYA, L., LANCE, M.J., andMOYER, B.A.A. (2001). Regeneration of perchlorate(ClO4

2)-loaded anion exchange resins by a novel tetrachlo-roferrate (FeCl42) displacement technique. Environ. Sci.Technol. 35, 3363–3368.

GULLICK, R.Q., LECHVALLIER, M.W., and BARHORST,T.A.S. (2001). Occurrence of perchlorate in drinking watersources. J. Am. Water Work Assoc. 93, 66–77.

HACKENTHAL, E. (1965). Die reduktion von perchlorat durchbakterien 2. Die identitat der nitratreduktase und des per-chlorat reduzierenden enzyms aus B. cereus. Biochem. Par-macol. 14, 1313–1324.

HACKENTHAL, E., MANNHEIM, W., HACKENTHAL, R.,and BECHER, R. (1964). Die reduktion von perchlorat durchbakterien 1. Untersuchungen an intakten zellen. Biochem.Parmacol. 14, 195–206.

HATZINGER, P.B., GREENE, M.R., FRISCH, S., TOGNA,A.P., MANNING, J., and GUARINI, W.J. (2000). Biologi-cal treatment of perchlorate-contaminated groundwater usingfluidized bed reactors. In The Second International Confer-ence on Remediation of Chlorinated and Recalcitrant Com-pounds, Monterey, CA.

HERMAN, D.C., and FRANKENBERGER, W.T. (1998). Mi-crobial-mediated reduction of perchlorate in groundwater. J.Environ. Qual. 27, 750–754.

HERMAN, D.C., and FRANKENBERGER, W.T. (1999). Bac-terial reduction of perchlorate and nitrate in water. J. Envi-ron. Qual. 28, 1018–1024.

HUNTER, W.L. (2001). In situ removal of perchlorate fromgroundwater. In The Sixth International Symposium on InSitu and On Situ Bioremediation. San Diego, CA.

HUTCHINSON, S.L., SUSARLA, S., WOLFE, N.L., and MC-CUTCHEON, S.C. (2000). Perchlorate accumulation fromcontaminated irrigation water and fertilizer in leafy vegeta-

MICROBIAL DEGRADATION OF PERCHLORATE 419

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

bles. In Second International Conference on Remediation ofChlorinated and Recalcitrant Compounds, May 2000, Mon-terey, CA.

KAPOOR, A., and VIRARAGHAVAN, T. (1997). Nitrate re-moval from drinking water—Review. J. Environ. Eng. 123,371–380.

KENGEN, S.W.M., RIKKEN, G.B., HAGEN, W.R., VANGINKEL, C.G., and STAMS, A.J.M. (1999). Purification andcharacterization of (per)chlorate reductase from the chlorate-respiring strain GR-1. J. Bacteriol. 181, 6706–6711.

KIM, K., and LOGAN, B.E. (2000). Fixed-bed bioreactor treat-ing perchlorate-contaminated waters. Environ. Eng. Sci. 17,257–265.

KIM, K., and LOGAN, B.E. (2001). Microbial reduction of per-chlorate in pure and mixed culture packed-bed bioreactors.Water Res. 35, 3071–3076.

KORENKOV, V.N., ROMANENKO, V.I., KUZNETSOV,S.I., and VORONNOV, J.V. (1976). Process for purificationof industrial waste waters from perchlorates and chlorates.U.S. patent 3,943,055.

KRAUTER, P.W. (2001). Using a wetland bioreactor to reme-diate ground water contaminated with nitrate (mg/L) and per-chlorate (ug/L). Int. J. Phytoremed. 3, 415–433.

LAWRENCE, J.E., LAMM, S.H., PINO, S., RICHMAN, K.,and BRAVERMAN, L.E. (2000). The effect of short-termlow-dose perchlorate on various aspects of thyroid function.Thyroid 10, 659–663.

LI, F.X., SQUARTSOFF, L., and LAMM, S.H. (2001). Preva-lence of thyroid diseases in Nevada counties with respect toperchlorate in drinking water. J. Occup. Environ. Med. 43,630–634.

LIN, R., CHOI, Y.C., BROWN, J., SNOEYINK, V., MOR-GENROTH, E., and RASKIN, L. (2002). Monitoring a per-chlorate-removing biofilm reactor using terminal restrictionfragment length polymorphism and mathematical modeling.In International Specialised Conference on Biofilm Moni-toring, Porto, Portugal.

LOGAN, B.E. (1998). A review of chlorate and perchloraterespiring microorganisms. Bioremed. J. 2, 69–79.

LOGAN, B.E. (2001a). Assessing the outlook for perchlorateremediation. Environ. Sci. Technol. 35, 482A–487A.

LOGAN, B.E. (2001b). Method and apparatus for treating perchlorate-contaminated drinking water. U.S. patent6,214,607.

LOGAN, B.E., and LAPOINT, D. (2002). Treatment of per-chlorate- and nitrate-contaminated groundwater in an au-totrophic, gas phase, packed-bed bioreactor. Water Res. 36,3647–3653.

LOGAN, B.E., KIM, K., and PRICE, S. (2001a). Perchloratedegradation in bench- and pilot-scale ex-situ bioreactors. InA. Leeson, B.M. Peyton, J.L. Means, and V.S. Magar, Eds.,

Bioremediation of Inorganic Compounds. Columbus, OH:Battelle Press, pp. 303–308.

LOGAN, B.E., WU, J., and UNZ, R.F. (2001b). Biological per-chlorate reduction in high-salinity solutions. Water Res. 35,3034–3038.

LOGAN, B.E., ZHANG, H.S., MULVANEY, P., MILNER,M.G., HEAD, I.M., and UNZ, R.F. (2001c). Kinetics of per-chlorate- and chlorate-respiring bacteria. Appl. Environ. Mi-crobiol. 67, 2499–2506.

LOSI, M.E., GIBLIN, T., HOSANGADI, V., and FRANKEN-BERGER, JR., W.T. (2002). Bioremediation of perchlorate-contaminated groundwater using a packed bed biological re-actor. Bioremed. J. 6, 97–103.

MALMQVIST, A., WELANDER, T., and GUNNARSSON, L.(1991). Anaerobic growth of microorganisms with chlorateas an electron-acceptor. Appl. Environ. Microbiol. 57,2229–2232.

MICHAELIDOU, U., ACHENBACH, L.A., and COATES,J.D. (2000). Isolation and characterisation of two novel(per)chlorate-reducing bacteria from swine waste lagoons. InE.T. Urbansky, Ed., Perchlorate in the Environment. NewYork: Kluwer Academic/Plenum Publishers, pp. 271–283.

MILLER, J.P., and LOGAN, B.E. (2000). Sustained perchlo-rate degradation in an autotrophic, gas-phase, packed-bedbioreactor. Environ. Sci. Technol. 34, 3018–3022.

MIN, B., EVANS, P., CHU, A., and LOGAN, B.E. (2003). Per-chlorate removal in pilot plant scale packed-bed bioreactors.Water Res., submitted.

MOTZER, W.E. (2001). Perchlorate: Problems, detection, andsolutions. Environ. Forensics 2, 301–311.

NERENBERG, R., RITTMANN, B.E., and NAJM, I. (2002).Perchlorate reduction in a hydrogen-based membrane-biofilm reactor. J. Am. Water Work Assoc. 94, 103–114.

NZENGUNG, V.A., and WANG, C.H. (2000). Influences onphytoremediation of perchlorate-contaminated water. In E.T.Urbansky, Ed., Perchlorate in the Environment. New York:Kluwer Academic/Plenum Publishers, pp. 219–230.

NZENGUNG, V.A., WANG, C.H., and HARVEY, G. (1999).Plant-mediated transformation of perchlorate into chloride.Environ. Sci. Technol. 33, 1470–1478.

OEHHA. (2002). Public health goal for perchlorate in drinkingwater. Office of Environmental Health Hazard Assessment(OEHHA) in California EPA, Sacramento, CA http://www.oehha.org/public_info/facts/perchloratefacts.html#down-load.

OKEKE, B.C., GIBLIN, T., and FRANKENBERGER, W.T.(2002). Reduction of perchlorate and nitrate by salt tolerantbacteria. Environ. Pollut. 118, 357–363.

PERLMUTTER, M.W., BRITTO, R., COWAN, J.D., PATEL,M., JACOBS, A., LOGAN, B., and CRAIG, M. (2000).Bioremediation of perchlorate-contaminated groundwater at

420 XU ET AL.

naval weapons reserve plant McGregor, Texas. Proc. In Na-tional Defense Industrial Association Proceedings, 26th En-vironmental Symposium and Exhibition, Long Beach, CA,27–30, 2000.

QUASTEL, J.H., STEPHENSON, M., and WHETHAM, M.D.(1925). Some reactions of resting bacteria in relation to anaer-obic growth. Biochem. J. 14, 304–317.

RENNER, R. (2002). Perchlorate drinking water recommenda-tion drops. Environ. Sci. Technol. (Online News. http://pubs.acs.org/subscribe/journals/esthag-w/2002/jan/policy/rr_per-chlorate.html).

RIKKEN, G.B., KROON, A.G.M., and VANGINKEL, C.G.(1996). Transformation of (per)chlorate into chloride by anewly isolated bacterium: Reduction and dismutation. Appl.Microbiol. Biotechnol. 45, 420–426.

SHROUT, J.D., and PARKIN, G.F. (2002). Isolation of hy-drogen-utilizing, autotrophic, perchlorate-degrading bacteria.In The 102nd general meeting of American Society for Mi-crobiology. Salt Lake City, UT.

SMITH, P.N., THEODORAKIS, C.W., ANDERSON, T.A.,and KENDALL, R.J. (2001). Preliminary assessment of per-chlorate in ecological receptors at the Longhorn Army Am-munition Plant (LHAAP), Karnack, Texas. Ecotoxicology10,305–313.

SOLDIN, O.P., BRAVERMAN, L.E., and LAMM, S.H.(2001). Perchlorate clinical pharmacology and human health:A review. Ther. Drug Monit. 23, 316–331.

SONG, Y., and LOGAN, B.E. (2002). Effect of O2 on perchlorate reduction and recovery of perchlorate degra-dation following O2 exposure. In The 102nd general meet-ing of American Society for Microbiology. Salt Lake City,UT.

STENKLO, K., THORELL, H.D., BERGIUS, H., AASA, R.,and NILSSON, T. (2001). Chlorite dismutase from Ideonelladechloratans. J. Biol. Inorg. Chem. 6, 601–607.

STEPANYUK, V.V., SMIRNOVA, G.F., KLYUSHNIKOVA,T.M., KANYUK, N.I., PANCHENKO, L.P., NOGINA,T.M., and PRIMA, V.I. (1992). New species of the Acineto-bacter Genus—Acinetobacter hermotoleranticus sp. Nov.Mikrobiologiya 61, 490–500.

STOUTHAMER, A.H. (1967). Nitrate reductase in Aerobacteraerogens 1. Isolation and properties of mutants strainsblocked in nitrate assimilation and resistant against chlorate.Arch. Microbiol. 56, 68–75.

SUSARLA, S., BACCHUS, S.T., WOLFE, N.L., and MC-CUTCHEON, S.C. (1999). Phytotransformation of perchlo-rate and identification of metabolic products in Myriophyl-lum aquaticum. Int. J. Phytoremed. 1, 97–107.

TOGNA, A.P., GUARINI, W.J., FRISCH, S., VECCHIO,M.D., CLIFFMURRAY, J.P., and TOLBERT, D.E. (2001).Case study of ex-situ biological treatment of perchlorate-con-taminated groundwater. In The Sixth International Sympo-

sium on In Situ and On-site Bioremediation, June 4–7, SanDiego, CA.

U.S. EPA. (1998). Federal Register notice about the Contami-nant Candidate List (March, 1998, 63 FR 10273).http://www.epa.gov/OGWDW/ccl/ccl_fr.pdf

U.S. EPA. (2002). Perchlorate Environmental Contamination:Toxicological Review and Risk Characterization (ExternalReview Draft). NCEA-1-0503. Washington, DC: U.S. Envi-ronmental Protection Agency, Office of Research and De-velopment.

URBANSKY, E.T. (1998). Perchlorate chemistry: Implicationsfor analysis and remediation. Bioremed. J. 2, 81–95.

URBANSKY, E.T. (2000a). Perchlorate in the Environment.New York: Kluwer Academic/Plenum Publishers.

URBANSKY, E.T. (2000b). Quantitation of perchlorate ion:Practices and advances applied to the analysis of common.Crit. Rev. Anal. Chem. 30, 311–343.

URBANSKY, E.T., and SCHOCK, M.R. (1999). Issues in man-aging the risks associated with perchlorate in drinking wa-ter. J. Environ. Manage. 56, 79–95.

URBANSKY, E.T., COLLETTE, T.W., ROBARGE, W.P.,HALL, W.P., and SKILLEN, J.M. (2001). Survey of Fertil-izers and Related Materials for Perchlorate: Final Report.EPA/600/R-01/tba. Cincinnati, OH: U.S. EPA.

URBANSKY, E.T., MAGNUSON, M.L., KELTY, C.A., andBROWN, S.K. (2000). Perchlorate uptake by salt cedar(Tamarix ramosissima) in the Las Vegas Wash riparianecosystem. Sci. Total Environ. 256, 227–232.

VAN DE PEER, Y., and DE WACHTER, R. (1994).TREECON for Windows: A software package for the con-struction and drawing of evolutionary trees for the Mi-crosoft Windows environment. Comput. Applic. Biosci. 10,569–570.

VAN GINKEL, C.G., KROON, A.G.M., and VAN VIJK, R.J.(1999). Process for the degradation of chlorite. U.S. patent5,891,339.

VAN GINKEL, C.G., PLUGGE, C.M., and STROO, C.A.(1995). Reduction of chlorate with various energy substratesand inocula under anaerobic conditions. Chemosphere 31,4057–4066.

VAN GINKEL, C.G., RIKKEN, G.B., KROON, A.G.M., andKENGEN, S.W.M. (1996). Purification and characterizationof chlorite dismutase: A novel oxygen-generating enzyme.Arch. Microbiol. 166, 321–326.

VAN WIJK, D.J., and HUTCHINSON, T.H. (1995). The eco-toxicity of chlorate to aquatic organisms: A critical review.Ecotoxic. Environ. Safety 32, 244–253.

VENKATESH, K.R., COBE, E.R., JENNINGS, D.L., andWAGNER, N.J. (2000). Process for the removal and de-struction of perchlorate and nitrate from aqueous streams.U.S. patent 6,066,257.

MICROBIAL DEGRADATION OF PERCHLORATE 421

ENVIRON ENG SCI, VOL. 20, NO. 5, 2003

WALLACE, W., BESHEAR, S., WILLIAMS, D., HOSPA-DAR, S., and OWENS, M. (1998). Perchlorate reduction bya mixed culture in an up-flow anaerobic fixed bed reactor. J.Ind. Microbiol. Biotechnol. 20, 126–131.

WALLACE, W., WARD, T., BREEN, A., and ATTAWAY, H.(1996). Identification of an anaerobic bacterium which re-duces perchlorate and chlorate as Wolinella succinogenes. J.Indust. Microbiol. 16, 68–72.

WOLFF, J. (1998). Perchlorate and the thyroid gland. Phar-macol. Rev. 50, 89–105.

WOLTERINK, A.F.W.M., JONKER, A.B., KENGEN,S.W.M., and STAMS, A.J.M. (2002). Pseudomonas chlori-tidismutans sp. nov., a non-denitrifying chlorate-reducingbacterium. Int. J. Syst. Evol. Microbiol, 52, 2183–2190.

WU, J., UNZ, R.F., ZHANG, H.S., and LOGAN, B.E. (2001).Persistence of perchlorate and the relative numbers of per-chlorate- and chlorate-respiring microorganisms in naturalwaters, soils, and wastewater. Bioremed. J. 5, 119–130.

XIE, Y.F., and ZHOU, H.J. (2002). Use of BAC for HAA re-moval-Part 2, column study. J. Am. Water Work Assoc. 94,126–134.

XU, J., TRIMBLE, J.J., and LOGAN, B.E. (2002). Perchloratereduction and denitrification pathways are separate in per-chlorate reducing bacteria. In The 102nd general meeting ofAmerican Society for Microbiology, Salt Lake City, Utah.

ZHANG, H.S., BRUNS, M.A., and LOGAN, B.E. (2002).Chemolithoautotrophic perchlorate reduction by a novel hy-drogen-oxidizing bacterium. Environ. Microbiol. 4, 570–576.

422 XU ET AL.