ARTICLE IN PRESS

0964-8305/$ - se

doi:10.1016/j.ib

�Correspondfax: +420235 5

E-mail addr

International Biodeterioration & Biodegradation 58 (2006) 224–230

www.elsevier.com/locate/ibiod

The biofiltration permeable reactive barrier:Practical experience from Synthesia

Lenka Veselaa,�, Jan Nemecekb, Martina Siglovac, Martin Kubald

aDEKONTA, a.s., Prague Office, Volutova 2523, 158 00 Prague 5, Czech RepublicbENACON, s.r.o., Marie Cibulkove 34/356, 140 21 Prague 4, Czech Republic

cDepartment of Fermentation Chemistry and Bioengineering, Institute of Chemical Technology in Prague, Technicka 5, 166 28 Prague 6, Czech RepublicdDepartment of Environmental Chemistry, Institute of Chemical Technology in Prague, Technicka 5, 166 28 Prague 6, Czech Republic

Received 30 June 2005; received in revised form 11 March 2006; accepted 13 June 2006

Available online 24 August 2006

Abstract

The paper refers to utilization of biological elements within permeable reactive barriers. The concept of a biofiltration permeable

barrier has been tested in the laboratory and in pilot-scale. Oxyhumolite (oxidized young lignite) was examined as an absorption material

and a biofilm carrier. Laboratory tests performed before the pilot verification confirmed that oxyhumolite adsorbs organic pollutants at a

minimum value, but that it can be used for biofilm attachment. An experimental barrier was built on premises of a chemical factory

contaminated mainly by various organic pollutants [benzene, toluene, ethylbenzene, and xylenes (BTEX), chlorobenzenes, naphthalene,

nitro-derivatives, phenols, trichloroethylene (TCE), and total petroleum hydrocarbon (TPH)]. Before the barrier was installed, a

preliminary survey of the unsaturated zone, hydrogeological investigation, and a microbiological survey had been performed. The barrier

was designed as a trench-and-gate system with an in situ bioreactor. During the year 2004, measurements of groundwater flux and

retention time under current hydrological conditions, together with chemical and microbiological monitoring, were carried out on the

site. The results showed high effectiveness of organic contamination removal. Average elimination varied from 57.3% (naphthalene) to

99.9% (nitro-derivatives, BTEX); microbial density in the bioreactor was approx. 105CFUmL�1.

r 2006 Elsevier Ltd. All rights reserved.

Keywords: Biobarrier; Biofiltration; Oxyhumolite; Bioreactor; Groundwater

1. Introduction

Permeable reactive barriers (PRBs) can provide cost-effective and long-term solutions for many groundwatercontamination problems. They are constructed belowground to intercept groundwater flow. As contaminatedgroundwater moves through the reactive filling, contami-nants are treated and transformed into harmless productsand/or their by-products.

Up to today, more than 60 PRBs have been installedthroughout the world. Most of them have been designedto treat halogenated aliphatic hydrocarbons (mainly

e front matter r 2006 Elsevier Ltd. All rights reserved.

iod.2006.06.013

ing author. Tel.: +420 235 522 252 5;

22 254.

ess: vesela@dekonta.cz (L. Vesela).

chlorinated) or heavy metals (e.g., hexavalent chromium).These barriers generally work on the principle of chemicalreduction on a reactive medium, which is mainly zero-valent iron (e.g., Puls et al., 1995; Reardon, 1995; Orthand Gillham, 1996; Gavaskar et al., 1997). Only asmall number of barriers has been installed for otherpollution treatments, such as petroleum substances, aro-matics, benzene, toluene, ethylbenzene, and xylenes(BTEX), or some explosives [e.g., cyclotrimethylenetrini-tramine (RDX)]. These pollutants are usually treatedin PRB by sorption on granulated activated carbon,or by aerobic degradation enhanced by intensive aerationin a reactor (USEPA, 2002; Alvarez et al., 2004; Drieset al., 2004; Birke et al., 2005). Special categories ofthese barriers are so-called air-sparging barriers, whichare designed as single or more lines of injecting wells

ARTICLE IN PRESSL. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230 225

built across groundwater flow. These wells are used forinjection of atmospheric air as an electron acceptor. In thecase of extensive injection this method is also calledbiosparging.

Some of the first biosparging barriers were built on sitesin East Garrington, Alberta, Canada, in 1995 (Bowles,1997) and in Alameda, California, USA, in 1997 (USEPA,2002). Biological barrier technology was also successfullyapplied in Europe, e.g., on sites in Italy, the Netherlands,and the United Kingdom (Bakker et al., 2000; Shields etal., 2000). During a pilot-scale application of a biologicalPRB on a site in Italy contaminated by various aromatichydrocarbons, a decrease of total hydrocarbon concentra-tion from 8000 mgL�1 to less than 200 mgL�1 wasobserved. It was also confirmed, by measurement ofhydrocarbon concentrations in soil air, that the majorprocess of contaminant removal was aerobic biologicalmineralization and not stripping of volatile organicsubstances. Further, a contamination level below 10 mgL�1

was achieved by a full-scale application (Bakker et al.,2000).

At the same time as the first biosparging barriers werebeing developed, PRBs, working on the biofiltrationprinciple, were also designed and tested. These biobarriersare generally designed as in situ reactor biofilter units,which contain suitable media with a bacterial biofilm.Unlike in a ‘classical’ PRB, pollutants are treated inbiobarriers by two ways: first they are absorbed on reactivemedia and then specific microorganisms decompose them.As biological elements are used in the biofiltration PRB, achoice of ‘bioreactive’ media is considered one of the mostimportant aspects of its design. A suitable ‘bioreactive’medium has to not only fulfil all requirements of the‘classical’ PRB (e.g., compatibility with undergroundenvironment, grain size, permeability, stability, low pur-chase price), but it must also ensure optimal growthconditions for microorganisms (i.e., it must not inhibit ornegatively influence metabolic activity of microbes, and itmust ensure satisfactory medium colonization and creationof an active biofilm). An important factor is also ensuringoptimal conditions for growth (suitable pH, nutrientcontent, and, in the case of aerobic barriers, sufficientdissolved oxygen).

Technological design of biofiltration barriers (e.g.,continual, funnel-and-gate or trench-and-gate) and alsothe type of a reactive segment (e.g., reactive gate, keson, insitu reactor) are similar to the ‘classical’ PRB. However, itis necessary to realize that implementation of biologicalelements into this technology and the necessity of ensuringoptimal growth conditions for microorganisms requirecertain technological amendments of PRB design and cancause some difficulties, for example, while ensuringsufficient oxygen level and optimal pH. So-called ‘biofoul-ing’ is caused by biofilm overgrowth and can lead todecrease of biobarrier permeability. This can furtherchange groundwater flow and cause biobarrier overflowing(i.e., bypassing).

Practical applications of biological PRB working on thebiofiltration principle have been relatively few so far.Successful realization of this PRB type can be shown, forexample, in a so-called ‘denitrification’ biobarrier, whichwas composed of a reactive medium of organic carbon anda bacterial culture of Pseudomonas sp. The ‘denitrification’biobarrier was built on a pilot-scale in Canada in 1995. Theresults showed that it is possible to achieve fast denitrifica-tion in this system and decrease nitrate content fromconcentrations typical for wastewater below limits fordrinking waters (Robertson and Cherry, 1995).This paper refers to pilot-scale testing results of a

biofiltration permeable barrier. Oxyhumolite (oxidizedyoung lignite) was used as an absorption material and abiofilm carrier.

2. Materials and methods

2.1. Tested reactive medium: oxyhumolite

Oxidized young lignite (oxyhumolite) contains up to 90% biochemi-

cally active humic substances. In the Czech Republic, one can find some of

the highest-quality oxyhumolites in the world. Some oxyhumolites from

North Bohemia, for example, contain more than 80% humic acids and

have a low content of bitumen (o0.1%) and inorganic compounds

(Novak et al., 2001; Vesela et al., 2005b). At present, oxyhumolites are

turned mainly into sodium and potassium humates, and partially also into

humic acids and humates of other metals. Owing to the low price of

oxyhumolites (they are waste products from coal mining) and their ion

exchange properties, it is believed that another use for them could be as

sorbents for wastewater treatment or removal of heavy metals (Novak et

al., 2001; Cezikova et al., 2001; Madronova et al., 2001). The fact that

humic acids extracted from oxyhumolites have similar or better ion

exchange properties in comparison to humic acids extracted from peat and

brown coal speaks for their utilization in this field.

During laboratory work performed before the pilot testing, it was

found that oxyhumolite adsorbs organic pollutants at a minimum value

(Vesela, 2005; Vesela et al., 2005a). Results of the batch sorption test

confirmed the generally expected assumption of a large surface area of the

tested oxyhumolites (up to 22.344m2 g�1). It was further found that all

tested samples adsorbed model organic substances (chlorobenzenes,

aniline, and nitrobenzene) to a small extent (in the range of

0.84–22.68mg g�1) and that experimentally determined maximal adsorp-

tion capacities did not correspond to the surface area of the tested

materials. Also, based on literature research, it is believed that the binding

strength of oxyhumolite to organic substances is very low (mainly

hydrogen bonds, Van der Waals forces) and therefore this material is not

suitable as an organic contaminant sorbent (Senesi, 1992; Piccolo, 1994;

Senesi et al., 1995; Vesela et al., 2005b). Thus it was concluded that

sorption of organic substances on oxyhumolites can be considered as low,

compared to oxyhumolite sorption ability for heavy metals (Cezikova

et al., 2001; Madronova et al., 2001), and will be neglected during further

testing.

The results of laboratory biofiltration experiments showed the

possibility of using oxyhumolite as the biofilm carrier (Siglova et al.,

2004; Vesela, 2005; Vesela et al., 2005a). It was also confirmed that

oxyhumolite is not suitable as the carrier itself, mainly because of its

ability to decrease pH, and therefore it is necessary to combine this

material with limestone.

2.2. Pilot-scale testing

The biofiltration PRB was built on the premises of a chemical plant

called Synthesia. The company was established in 1920 and is situated in a

ARTICLE IN PRESS

Fig. 1. Designed system trench-and-gate used on the Synthesia Pardubice site.

Fig. 2. Configuration of the in situ bioreactor.

L. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230226

suburb of Pardubice, in the East Bohemia part of the Czech Republic.

Throughout the more than 80 years of its development the company

experienced a number of changes of direction; at present it is focused on

the production of cellulose derivatives, pigments, and dyes, as well as

advanced organic intermediates. Former manufacturing technologies used

on the company premises caused massive pollution of groundwater and

led to production of many waste by-products.

During laboratory biofiltration experiments performed before this pilot

testing, it was verified that naturally occurring (autochthonous) microflora

derived from contaminated water of the model pilot site have become

adapted to local conditions and are capable of reproduction and

degradation in the temperature range of 8–30 1C, and that it is possible

to increase the activity of these microflora by the addition of N, P

nutrients (approx. 1 gL�1 of an NP fertilizer). Laboratory column tests

used for development of design parameters of the PRB pilot unit

confirmed that a retention time of 15 h is long enough for a 97% reduction

of all contaminants present in contaminated groundwater (with the

exception of poorly degradable substances such as nitrobenzene or N,N-

diethylaniline), providing that other conditions (mainly oxygen and

nutrient concentrations) are optimized.

Two bacterial species, Achromobacter xylosoxidans and Rahnella sp.,

were isolated from this contaminated groundwater. They were identified at

the Czech Collection of Microorganisms of the Masaryk University in

Brno. Rahnella sp. is known to be a nitrate reducer, and some strains are

able to fix nitrogen. A. xylosoxidans is also able to reduce nitrates and

some literature sources refer to its ability to degrade chlorobenzoates and

chlorocatechols (Jencova et al., 2004). Both bacterial species are gram

negative.

Before biobarrier installation on the pilot site, preliminary investiga-

tions of the unsaturated zone (boring up to 2.0m, sampling) and saturated

zone (drilling of five wells, pumping tests including sampling) were

performed. A geological cross-section of the site consists of sand and

sandy gravel-fluvial deposits of the Elbe River up to 2.5m thick. The

aquifer is underlined by impermeable upper-cretaceous claystones

weathered to clays.

On the basis of the results of hydrogeological investigation, the

biobarrier for pilot testing was designed as the trench-and-gate type

(Bowles, 1997). The system contained one drainage trench, which collected

contaminated groundwater to an underground bioreactor (gate). The

drainage trench was installed perpendicular to the groundwater flow. It was

approx. 13m long and 1.0–1.5m wide, and its depth interfered with the

underlying impermeable clay (2.6–3.0m below ground). The bottom side of

the trench was sealed with a plastic liner; above it was filled with permeable

gravel and native soil. Treated groundwater flowed by gravity to the

recharge gallery formed by two trenches filled with gravel approx. 10m long

and 1.0m wide, to the depth of impermeable clay. This helped to minimize

pressure losses of groundwater flowing by gravity through the system and

thus maximized the capture zone of the drainage trench (Fig. 1).

A reactive segment was constructed as an in situ bioreactor. The

reactor was a box of outside dimensions 2.0� 1.2� 4.8m. The first

chamber (marked as Chamber I) was equipped with an aeration segment

at its base. This chamber had an internal size of 0.9� 1m; the water

column fluctuated between 3.24 and 3.56m (depending on hydrological

conditions). The effective volume of Chamber I varied from 2.98 to

3.28m3 (average 3.13m3). Treated water flowed by gravity from the first

chamber to the second and third chambers (marked as Chambers II and

III). These chambers were connected in parallel and both were equipped

with a biofilter unit of 0.5m3. The filter of the second chamber was filled

with ‘keramzit’ (ceramic granulate of LIAPOR, Lias Vyntirov, Czech

Republic); the filter of the third chamber contained oxyhumolite (derived

from the Vaclav mine near Duchcov, Czech Republic), with limestone

(Vapenka Certovy schody, Czech Republic) as a pH buffer, and with

gravel. Bullet valves regulated the water inflow into Chamber I and also its

discharge to Chambers II and III. Piezometers were used to monitor

groundwater level in the drainage trench and the individual chambers, as

well as in both the arms of the recharge gallery. They were also used for

water sampling in all the chambers (Fig. 2).

The barrier installation was completed in November 2003. Its

construction on the site was carried out in two phases: (i) excavation

work and bioreactor bedding, and (ii) drainage trench and recharge gallery

installation.

The pilot testing included regular measurements of groundwater flow

and retention time (trace tests with sodium chloride) and groundwater

levels (the G-20 indicator, Geospol, Czech Republic), and analyses of

various inorganic [dissolved solids, NH4+, NO3

�, Cl�, HCO3�, SO4

2�, and

chemical oxygen demand (COD)] and organic parameters [BTEX,

chlorinated benzenes, naphthalene, nitro-derivatives, phenols, trichlor-

oethylene (TCE), and total petroleum hydrocarbons (TPH)]. Inorganic

ARTICLE IN PRESS

-651710 -651700 -651690 -651680 -651670 -651660 -651650 -651640

-1058730

-1058720

-1058710

-1058700

-1058690

-1058680

-1058670

-1058660

-1058650

V-801

V-802

V-803

V-804

V-805

214.64

214.69

214.62

214.48

214.71

Fig. 3. Groundwater table situations on the pilot site—the situation in

November 2003 (before PRB building).

-651710 -651700 -651690 -651680 -651670 -651660 -651650 -651640

-1058730

-1058720

-1058710

-1058700

-1058690

-1058680

-1058670

-1058660

-1058650

V-801

V-802

V-803

V-804

V-805

V-806 V-807 V-808

V-809V-810

214.74

214.81

214.74

214.64

214.94

214.88214.88 214.89

214.89214.89

Fig. 4. Groundwater table situations on the pilot site—the situation in

January 2004 (after PRB building).

L. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230 227

parameters were determined by the following methods: computation

(dissolved solids, HCO3�), spectrometry (NH4

+, NO3�), argentometry

(Cl�), chelatometry (SO42�), and permanganate titration (COD). Most

organic parameters (BTEX, chlorinated benzenes, naphthalene, nitro-

derivatives, TCE) were measured by solid phase microextraction (SPME)

with gas chromatography/mass spectroscopy (GC/MS); phenols were

analysed on GC/MS after acetylation. TPH were detected by Fourier

transform infrared spectroscopy (FTIR). All the above-mentioned

chemical analyses were done in a commercially accredited laboratory

(Monitoring, Czech Republic) by recognized analytical methods, follow-

ing standard operational procedures. Temperature, pH, conductivity, and

oxygen concentration were measured by the Multi 340i device (WTW,

Germany). Content of mineral nutrients (N, P) was determined by

ammonia, nitrate, and phosphate analytical sets (Marcherey-Nagel,

Germany). Total number of aerobic culturable psychrophylic bacteria

[number of colony forming units (CFU) per mL of groundwater] was

determined by plating on Petri dishes containing meat peptone agar

(MPA) No. 2 (Imuna Pharm, Slovakia) and incubating at 20 1C for 72 h.

3. Results and discussion

The pilot testing started in January 2004 and continuedfor a year. During the pilot testing, trace tests wereperformed in order to measure groundwater flux throughthe bioreactor segment under current hydrological condi-tions and to determine retention time in the individualchambers of the in situ bioreactor; chemical and micro-biological monitoring of decontamination process effec-tiveness was also carried out.

3.1. Groundwater flow on the PRB site

Groundwater levels on the model site were measuredduring the pilot testing. The level of groundwater beforebiobarrier installation (the situation in November 16, 2003)is shown in Fig. 3. Hydroizohypses were calculated by theSURFER programme by the kriging method. Obviousdeformation occurred near the well-marked V-804; generalgroundwater flow north–south was obvious. It could notbe ruled out that in the line of wells V-802 and V-805 alocal groundwater divide existed. Groundwater to thenorth of this divide (the V-801 surroundings) could thenflow north and be drained into a sewerage pipe that runsparallel to an internal factory road. The groundwater tablevaried from 214.40 to 214.70m asl on the site during thisperiod. The average hydraulic gradient was 0.01. Based oninformation obtained during this measurement, PRB wassituated in the area allocated by wells V-803, V-804, andV-805.

The situation of groundwater flow after installation ofPRB in January 27, 2004, is shown in Fig. 4. Despite thedeformation near wells V-803 and V-804, caused by thecalculation method and well position, it was clear that thedrainage trench preferentially captured groundwater flow-ing from the area near well V-805. The direction ofgroundwater flow was then veered southwest to thedrainage trench. After passing the in situ bioreactor, waterwas infiltrated via the recharge gallery into an aquifer inthe south part of the model site. From maps it is evidentthat, owing to local heterogeneity in permeability, the flow

captured by the drainage trench was rather narrow and didnot interfere with the surroundings of the V-801, V-802,and V-803 wells.Further, the results of groundwater table measurements

from the period October 31, 2003, to February 11, 2004,were compared. The groundwater table significantly

ARTICLE IN PRESSL. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230228

increased (up to 0.44m in the area of the V-805 well) in thesoutheast of the model site during this period. On the otherhand, groundwater table changes were minimal in thesouthwest part. The total increase of groundwater tablewas caused by higher precipitation and melting snow.Minimal groundwater table changes near the V-801 wellcan be explained by the drainage effect of the sewerage.The maximal groundwater table in the V-805 area showedits hydrogeological location on the divide. The aboveanalyses of groundwater flow ensured that groundwaterflowed to the in situ bioreactor preferentially from the areaof the V-805 well. This was confirmed by results ofchemical analyses; water flowing to the bioreactor hadhydrochemistry similar to water in the V-805 well.

3.2. Groundwater flux and retention time in the bioreactor

The first trace test was carried out in January 2004. Theaim of this test was to determine flow quantity in thebioreactor under current hydrogeological conditions and tomeasure retention time of groundwater in individualreactor chambers. Results of this trace test showed thatthe average flow rate through the bioreactor was0.47 Lmin�1; the resulting retention time in Chamber Iwas thus 114 h. In the case of optimal homogeneousdistribution of groundwater flow from Chamber I toChambers II and III, the retention time in both biofilters ofthese chambers was 14 h. Results also showed that waterdid not flow only through the biofilter units; it also flowedpartly through an interspace between the biofilter walls andboth chambers. Sealing of the space eliminated this adverseflow of groundwater in the bioreactor.

The second trace measurement was performed in June2004, when a decrease of the groundwater level on the sitewas observed. The aim of the test was to determinegroundwater flow rate under low hydrogeological condi-tions. For the trace test, the same methodology as in thefirst case was used. Over 78 h, the flow rate throughChamber I was 0.13 Lmin�1; it decreased further to0.08 Lmin�1. The retention time in Chamber I, resultingfrom the flow rate and water volume, was then 628 h; theretention time of the biofilters of Chambers II and III was83 h. Results of this trace test showed that there is minimalgroundwater flow through the bioreactor under lowhydrogeological conditions.

Such a long retention time for groundwater in theindividual chambers did not enable us to test theeffectiveness of two biofilter fillings in common conditions,and so an amendment of the PRB system was planned anda system of forced groundwater circulation was installed.This system allowed groundwater flow rate in the in situbioreactor to be controlled. Follow-up measurements ofgroundwater flow and its progression with time confirmedthat massive increases and decreases of groundwater werecommon on the site throughout the pilot testing. Thereforeit is clear that the technical amendment of flow through thebioreactor was absolutely essential on this site and it

enabled water flow rate to be maintained within the rangeof 0.3–0.5 Lmin�1. Average retention time at averagewater volume of 3.18m3 was approx. 129 h in Chamber Iand 16 h in Chambers II and III.

3.3. Monitoring of groundwater decontamination

The effectiveness of the decontamination process inevery chamber of the PRB was regularly monitored bychemical analyses of inorganic (dissolved solids, NH4

+,NO3�, Cl�, HCO3

�, SO42�, and COD) and organic

parameters (BTEX, chlorinated benzenes, naphthalene,nitro-derivatives, phenols, TCE, and TPH). Samples weretaken from the piezometers of the bioreactor, which werelocated at the inputs and outputs of all chambers; sampleswere also taken from monitoring wells situated at theend of each arm of the recharge gallery (wells V-809 andV-810).Results of inorganic parameter analyses showed a rather

small effectiveness of the biofiltration PRB for reduction ofinorganic contaminants. An average decrease of 21.6%inorganic contamination was detected in Chamber I (thechamber with an aeration hose). The lowest effectivenesswas measured for NO3

�, SO42�, and COD and conversely

the highest decrease, approx. 22%, was detected for NH4+

ions. In Chamber II (the chamber filled with ‘keramzit’), adecrease of inorganic parameters of only 10% wasachieved. Minimal effectiveness was found, once again,for SO4

2� and COD; the highest decrease was for Cl�. Inthe case of Chamber III (the chamber filled withoxyhumolite), inorganic parameters decreased up to 28%.Minimal effectiveness was achieved for SO4

2� and NH4+

ions; efficiency higher than 20% was detected for thefollowing three parameters; 28% for NO3

�, 25% for COD,and 21% for Cl�.Unlike the inorganic pollutants, organic contaminants

were removed with very high efficiency by the PRBbiofiltration system. The average degree of organiccontaminant removal varied from 20.5% to 97.5% inChamber I. The lowest efficiency, 20.5%, was achieved fornaphthalene; the highest, above 90%, was observed forBTEX (97.5%), TPH (96.2%), and nitro-derivatives(90.8%). A high decrease was also detected for otherorganic contaminants; chlorinated benzenes (86.6%), TCE(78.6%), and phenols (73.3%). In the case of Chamber II, adecrease of 9–93% was observed. The lowest efficiency wasachieved for removal of TPH (8.7%), naphthalene(30.9%), and phenols (43.5%); other parameters showeda decrease above 50%; TCE of 56.7%, chlorinatedbenzenes of 71.7%, nitro-derivates of 76.8%, and BTEXof 92.9%. Chamber III showed good efficiency in the rangeof 35–98%. The lowest efficiency was observed fornaphthalene (35.4%) and phenols (48.5%); other organicparameters showed decreases higher than 50%; chlorinatedbenzenes 61.4%, TPH 56.8%, TCE 52.5%, BTEX 98.2%,and nitro-derivatives 94.7%. The effectiveness of organic

ARTICLE IN PRESS

0

20

40

60

80

100

Naphthalene TCE BTEX CL

benzenes

N derivates Phenols TPH

Rem

ova

l eff

ecti

ven

ess

(%)

Chamber I Chamber II

Chamber III Total PRB

Fig. 5. Removal effectiveness of organic substances in the individual chambers and the whole PRB unit.

1.00E+03

1.00E+04

1.00E+05

1.00E+06

8.4. 3.5. 21.5. 26.7. 11.8. 26.8. 8.9. 21.9. 19.10. 26.10. 2.11. 9.11.

Aer

ob

ic b

acte

ria

con

ten

t (C

FU

/ml)

Chamber I

Chamber II

Chamber III

Fig. 6. Aerobic bacterial content in the individual chambers.

L. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230 229

contamination removal in individual chambers and in thewhole PRB pilot unit is shown in Fig. 5.

Together with chemical analyses of inorganic andorganic contaminants, the total number of aerobic cultur-able psychrophylic bacteria in groundwater (CFU mL�1)and mineral nutrient content (N, P) was measured at thesame sampling points. Owing to the technical design of thebioreactor and the weight of each biofilter unit, it was notpossible to sample directly the tested carrier materials. Themicrobial density of all three bioreactor chambers wasaround 105CFUmL�1 during the pilot testing. If wecompare this value with the input value of 103CFUmL�1

detected before start of the pilot, it can be said that theautochthonous microflora concentration increased by twoorders. Aerobic microorganism content trends in indivi-dual chambers are shown in Fig. 6. The concentration ofnutrients (N, P) in the bioreactor unit was sufficient

throughout the pilot testing. The ammonia and nitrateconcentrations ranged between 50 and 250 gL�1; phos-phates were approx. 3–5 gL�1.It is necessary to mention here that, based on the results

of laboratory work (mentioned in Section 2.2. or in Siglovaet al., 2004), it was decided to use autochthonousmicroflora (mainly A. xylosoxidans and Rahnella sp.) forthe groundwater decontamination process. The biofilterunits of the pilot PRB were not directly inoculated;autochthonous microflora present in groundwater wasenhanced by aeration and nutrient additions.

4. Conclusions

It was confirmed that incorporation of biologicalelements into PRB technology could contribute to treat-ment of heterogeneous and complex organic pollution.

ARTICLE IN PRESSL. Vesela et al. / International Biodeterioration & Biodegradation 58 (2006) 224–230230

Laboratory and further pilot testing on the model site ofSynthesia Pardubice showed high effectiveness of removalof various organic contaminants originating from formerchemical production.

Regarding the testing of oxyhumolite as the innovativereactive medium, it can be concluded that this materialabsorbs organic contaminants at a minimum value; itsmaximal sorption capacity is low and binding strength toorganic substances very small. Nevertheless, this materialcan be used for biofilm attachment, preferentially com-bined with limestone and gravel.

Acknowledgement

The authors would like to thank the Czech Ministry ofIndustry and Trade for financial support (Grant no. KD-K2/42).

References

Alvarez, P.J.J., Oh, B.T., Gregory, K., Scherer, M., Parkin, G., 2004.

Iron-based bioremediation of RDX and other redox-sensitive ground-

water pollutants. International Biodeterioration & Biodegradation 53,

199.

Bakker, L.M.M., Tonnaer, H., Michelberger, H.S., Pijls, C.G.J.M.,

Filipponi, P., 2000. Full-scale operation of aerobic biobarriers. In:

Telford, T. (Ed.), Proceedings of ConSoil Conference 2000, vol. 2,

FZK/TNO, Leipzig, pp. 932–938.

Birke, V., Burmeier, H., Dahmke, A., Ebert, M., 2005. The German

permeable reactive barrier (PRB) network ‘‘Rubin’’: temporary overall

results and lessons learned after five years of work. In: Uhlmann, O.,

Annokkee, G.J., Arendt, F. (Eds.), Proceeding of the ConSoil

Conference 2005, FZK/TNO, Bordeaux, pp. 2865–2869.

Bowles, M.W., 1997. The trench and gate groundwater remediation

system. M.Sc. Dissertation, University of Calgary, Faculty of

Graduate Studies, Calgary, Alberta, Canada.

Cezikova, J., Kozler, J., Madronova, L., Novak, J., Janos, P., 2001.

Humic acids from coals of the North-Bohemian coal field: II. Metal-

binding capacity under static conditions. Reactive & Functional

Polymers 47, 111–118.

Dries, J., Hendrick, H., Bastiaens, L., Springael, D., Agathos, S.N., Diels,

L., 2004. An anaerobic sequential reactive barrier concept for

remediation of groundwater containing mixed pollution. International

Biodeterioration & Biodegradation 53, 203–204.

Gavaskar, A.R., Gupta, N., Sass, B.M., Fox, T., Janosy, R.J., Cantrell,

K., Olfenbuttel, R., 1997. Design guidance for application of perme-

able barriers to remediate dissolved chlorinated solvents. Battelle,

Columbus, Ohio.

Jencova, V., Strnad, H., Chodora, Z., Ulbrich, P., Hickey, W.J., Paces, V.,

2004. Chlorocatechol catabolic enzymes from Achromobacter xylosox-

idans A8. International Biodeterioration & Biodegradation 54,

175–181.

Madronova, L., Kozler, J., Cezikova, J., Novak, J., Janos, P., 2001.

Humic acids from coals of the North-Bohemian coal field: III. Metal-

binding properties of humic acids—measurements in a column

arrangement. Reactive & Functional Polymers 47, 119–123.

Novak, J., Kozler, J., Janos, P., Cezikova, J., Tokarova, V., Madronova,

L., 2001. Humic acids from coals of the North-Bohemian coal field: I.

Preparation and characterisation. Reactive & Functional Polymers 47,

101–109.

Orth, W.S., Gillham, R.W., 1996. Dechlorination of trichloroethene in

aqueous solution using Fe0. Environmental Science & Technology 30,

66–71.

Piccolo, A., 1994. Interactions between organic pollutants and humic

substances in the environment. In: Senesi, M., Miano, T.M. (Eds.),

Humic Substances in the Global Environment and Implications on

Human Health, Proceeding of the 6th IHSS Symposium, Elsevier,

Amsterdam, pp. 961–977.

Puls, R.W., Powell, R.M., Paul, C.J., 1995. In-situ remediation of

groundwater contaminated with chromate and chlorinated solvents

using zero-valent iron: a field study. 209th ACS National Meeting,

American Chemical Society, Anaheim, California.

Reardon, E.J., 1995. Anaerobic corrosion of granular iron: measurement

and interpretation of hydrogen evolution rates. Environmental Science

& Technology 29 (12), 2936–2945.

Robertson, W.D., Cherry, J.A., 1995. In-situ denitrification of septic-

system nitrate using porous media barriers: field trials. Ground Water

33, 99–111.

Senesi, N., 1992. Binding mechanisms of pesticides to soil humus

substances. Science of the Total Environment 123-124, 63–68.

Senesi, N., D’Orazio, V., Miano, T.M., 1995. Adsorption mechanisms of

S-triazine and bipyridylium herbicides on humic acids. Geoderma 66,

273–283.

Shields, A.R.G., Hardisty, P.E., Butler, A., Johnes, A., Wallace, S., 2000.

Remediation of a former gas manufacturing plant using a research-

supported in-situ air sparging programme. In: Telford, T. (Ed.),

Proceedings of the ConSoil Conference 2000, vol. 2, FZK/TNO,

Leipzig, pp. 1005–1012.

Siglova, M., Vesela, L., Masak, J., Zachar, P., Cejkova, A., Mikes, J.,

Jirku, V., 2004. Bioremediation of the contaminated groundwater

based on the biofilm of the autochthonous microflora. In: Popov, V.,

Itoh, H., Brebbia, C.A., Kungolos, S. (Eds.), Waste management and

the environment II. WIT Press, Ashurst, pp. 571–579.

USEPA, 2002. Field applications of in-situ remediation technologies:

permeable reactive barriers. USEPA, online at /http://clu-in.org/

download/rtdf/fieldapp_prb.pdfS.

Vesela, L., 2005. The utilisation of oxyhumolite in remediation of

contaminated groundwater. Ph.D. Thesis, Institute of Chemical

Technology Prague, Faculty of Environmental Technology, Prague,

Czech Republic.

Vesela, L., Siglova, M., Nemecek, J., Kubal, M., 2005a. Biological

elements in permeable reactive barriers—practical experience from

Synthesia. In: Uhlmann, O., Annokkee, G.J., Arendt, F. (Eds.),

Proceedings of the ConSoil Conference 2005, FZK/TNO, Bordeaux,

pp. 1497–1506.

Vesela, L., Kubal, M., Kozler, J., Innemanova, P., 2005b. Structure and

properties of natural humic substances of the oxyhumolite type.

Chemicke Listy 99, 711–717.