Ecotoxicology and Environmental Safety 53, 323d329 (2002)

Environmental Research, Section B

doi:10.1006/eesa.2002.2225

Microbial Toxicity Tests and Chemical Analysis as Monitoring Parametersat Composting of Creosote-Contaminated Soil

J. Ahtiainen,*�1 R. Valo,- M. JaK rvinen,* and A. Joutti**Finnish Environment Institute, Research Laboratory, Hakuninmaantie 4-6, FIN-00430 Helsinki, Finland; and -Soil and Water Ltd.,

P.O. Box 50, FIN-01620 Vantaa, Finland

Received November 1, 2001; published online July 25, 2002

Traditionally, chemical analyses are used in the assessment ofcontaminated soil and in monitoring the e7ciency of soil remedi-ation processes. We investigated if chemical analysis could besupported and even partly replaced by biological toxicity tests. Intwo case studies creosote-contaminated soil was composted out-doors in 5- and 100-m3 windrows. Degradation of polyaromatichydrocarbons (PAHs) was followed by chemical analysis andtoxicity tests. Polyaromatic hydrocarbons were quanti5ed andiden5tifed by HPLC. Because the soil was also contaminated bycopper-, chromium-, and arsenic-containing fungicides, theseelements were analyzed by atomic absorption spectrometry. Thetoxicity of soil samples was assessed by a soil-contact modi5ca-tion of the luminescent bacteria (Vibrio fischeri) test and in theother case also by enzyme synthesis inhibition (Toxi-ChromoPadtest, Escherichia coli). The toxicity of soil water extracts wasmeasured by the standard luminescent bacteria (V. fischeri) testand bacterial (Pseudomonas putida) growth inhibition test. Afterthe 5rst 4 months of the composting period the total amount ofPAHs was reduced in all windrows, and in particular, the loss oftwo- and some three-ring compounds was high, almost 90%.Toxicity decreased concurrently with the decrease in PAH con-centration during composting, but after 4 months, one of the pilesinoculated with mycobacteria and containing more three- andfour-ring compounds was found to be more toxic than at thebeginning. After the next summer, total PAH content was fur-ther reduced but some four-ring or heavier compounds weredemonstrated to be poorly degraded. The toxicity was alsoreduced to the same level as in the control pile. The total PAHcontent and the toxicity were both reduced signi5cantly during5 months of composting � 2002 Elsevier Science (USA)

Key Words: creosote; polyaromatic hydrocarbons; soil; com-posting; bacteria; toxicity.

INTRODUCTION

Polyaromatic hydrocarbons (PAHs) have been the majorissue of environmental concern at creosote-contaminated

1To whom correspondence should be addressed. Fax: #358-9-40300890. E-mail: jukka.ahtiainen@ymparisto.".

32

sites. The PAH fraction of creosote represents some 85% ofthe hydrocarbon content, compared with 5% fraction ofother aromatic heterocyclic compounds (AHCs) (Muelleret al., 1989). Compounds in both groups and their metab-olites are known to be toxic and some of the componentsare mutagenic. However, the PAHs have low aqueous solu-bility, which limits bioavailability. In comparison, many ofthe more recalcitrant AHCs are more polar compounds andhave a higher water solubility. Hence the AHCs may also bea risk to groundwater and biota, although these chemicalsare not routinely analyzed.Composting has been applied to remediate creosote-

contaminated soils. The degradation of PAHs is well char-acterized (Cerniglia, 1992), but knowledge of the degrada-tion of AHCs is limited (Bundy et al., 2001). In most caseswhen soil has been contaminated for decades, the high-molecular-weight PAHs are detected in the soil while thesmaller PAHs have degraded.Creosote has been used for wood preservation at several

places in Finland. Other wood impregnants, such as metalcompounds (arsene, chrome, and copper), may have beenused at the same site. Therefore, the soil at such sites maycontain both creosote and heavy metals. A high metalconcentration potentially inhibits PAHs from degrading, aswell as inhibiting AHC-degrading microbes. Because of theabove-mentioned complexity of the pollution, and the po-tential formation of biotransformation products fromPAHs, bioremediation and monitoring of creosote-con-taminated soil should be carefully planned, assessed, andevaluated. Toxicity assessments during the remediation andespecially at the end can provide valuable and complement-ary information to the chemical analysis. Only a limitednumber of compounds can be analyzed by chemical analy-sis. One major advantage of biological toxicity tests, overchemical analysis, is direct assessment of the potential haz-ard to the soil ecosystem caused by both the original PAHsand their potential biotransformation products.Exposure of the terrestrial biota to harmful environ-

mental compounds may result in direct or indirect e!ects,

3

0147-6513/02 $35.00� 2002 Elsevier Science (USA)

All rights reserved.

324 AHTIAINEN ET AL.

which cannot realistically be assessed even in very complexmodel test systems by using only chemical analytics. Often,various single-species screening tests are carried out to de-tect the possible harmful e!ects of chemicals on biotic sys-tems. In the environment di!erent species and complex foodwebs are exposed. However, we have already learned fromaquatic toxicology that the actual impacts often concern fewrelevant targets on the biomolecular level. The functionalunits (respiration, enzyme functions, membrane transport)are often similar in di!erent living organisms (Wenzel et al.,1997) which makes it possible to monitor toxicity by select-ing only a few key parameters of simple organisms. Ofcourse, too simplistic use of results from few parameters oftoo few organisms can mislead the risk assessment. Di!erentorganisms and species have di!erent exposures to com-pounds, di!erent metabolisms, and di!erent tolerance toenvironmental stress factors, harmsful chemicals being onlyone of these.In recent years, various bacterial bioassays have been

developed for screening of chemicals, e%uent, sediment, andsoil toxicity. Most are based on measurements of growthinhibition, respiration, and viability of bacterial cells.Microorganisms, especially bacteria, have several advant-ages for use in toxicity testing. Microbial tests aresimple, rapid, sensitive, and inexpensive (Bitton and Dutka,1986).The inhibition of light production by <ibrio ,scheri indi-

cates disturbance of the energy metabolism of this hetero-trophic bacterium. The luminescence pathway is a directbranch of the electron transport chain, and from theluminescence measurement the metabolic status of this bac-terium can be assessed (Hastings, 1978). The change inbacterial luminescence when these bacteria are exposed towastewater samples can be used as an indicator of potentialtoxicity (Bulich et al., 1981). Tests can be performed withsoil water elutriates or with fresh homogenized soil sampleswith solid-phase modi"cation of the standard test. TheToxi-ChromoPad test is a biotest based on the inhibition of�-galactosidase enzyme synthesis in Escherichia coli whenthe bacterial cells are directly exposed to contaminatedsamples (Kwan, 1995). The bacterium Pseudomonas putidarepresents a common aquatic heterotrophic microorganism.WhenP. putida cells are cultured under speci"ed conditions,in a de"ned medium with di!erent concentrations of sam-ples over several generations, toxic substances present in thesample can inhibit the cell multiplication of the bacteria(Bringmann and KuK hn, 1977). Hence these three bacterialtoxicity tests cover three di!erent and important endpoints:respiration inhibition, enzyme synthesis, and cell multiplica-tion.In this study, the bioremediation of two soils with heavy

creosote and metal contamination was monitored in out-door pilot composts by chemical analysis and by bacterialtoxicity tests.

MATERIALS AND METHODS

Experimental Setup of Pilot-Scale Composting

In the smaller-scale case the soil from a sawmill area nearVilppula in southern Finland was very heavily con-taminated with wood impregnants: creosote oil and metals(As, Cr, Cu). Two pilot composts (5 m3) were constructed inthe "eld. Spruce bark chips were added as a bulking mater-ial to both composts. One of the piles was inoculated witha Mycobacterium and the other pile was left uninoculatedand thus contained only indigenous bacteria. For chemicalanalysis and toxicity tests, the pooled soil samples (at thestart, after the "rst summer, after the second summer) werehomogenized and sieved through an 8-mm sieve to removethe added bark chips and other coarse particles.The other site was located at Ilmajoki, in western Fin-

land. The soil from an old wood-preserving facility washeavily contaminated with creosote oil and metals (As, Cr2,Cu). A larger-scale pilot composting pile (100 m3) was con-structed in the "eld. The soil was prescreened to removecoarse particles (more than 50 mm). The soil was pretreatedwith 50% hydrogen peroxide to speed up the breakdown ofthe recalcitrant four- to six-ring PAH compounds. After thehydrogen perioxide had completely degraded (tested bya "eld-strip test) a microbial inoculum of PAH degraders,nutrients (N, P, and micronutrients), and bark chips wasadded. The pile was sampled for toxicity and chemicalcomposition at the beginning, after the "rst week, and thenevery 4 weeks during the 163-day composting period. Thecompost was thoroughly mixed before sampling. Thepooled soil samples (during one summer) were homogenizedand sieved through an 8-mm sieve to remove the added barkchips and other coarse particles.

Chemical Analysis

At sampling the pooled soil samples were homogenizedand the PAHs were quanti"ed and identi"ed by HPLC.PAHs were extracted by 6 h cyclohexane/dichloromethaneSoxhlet extraction. The extract was evaporated and dilutedto acetonitrile and the PAH quanti"cation was based on themethod of external standard. In the larger-scale case, thePAHs were analyzed by gas chromatography with massselective detection after toluene extraction. Also, the totalhydrocarbon content of the soil was analyzed from thepetrole ether extract by gravimetry. The metals (copper,chromium, and arsene) were analyzed by atomic absorptionspectrometry.

Soil Toxicity Assessments

Toxicity of the soil samples was tested by solid-phasemodi"cation of the luminescent bacteria test and with the

TABLE 1Chemical Characteristics of the Soils in the Two Cases

before Composting

mg/kg fresh wt soil

Small scale, Vilppula Larger scale, Ilmajoki

Total PAHs 23,500 10,960Total hydrocarbons Not analyzed 27,000

Arsene 2,500 870Chromium 7,000 520Copper 3,000 400

325MONITORING COMPOSTING OF CREOSOTE-CONTAMINATED SOIL

Toxi-Chromo enzyme synthesis test, and the toxicity of soilelutriates with the standard luminescent bacteria test andwith the bacterial (P. putida) growth inhibition test.

TABLConcentrations and Reduction of Analyzed PAH Compounds inPile at Start (June), after One Summer (October), and at the

Compost PAHs At start

Inoculated pile Naphthalene (2)a 500Asenaphtene (3) 7,100Fluorene (3) 1,900Phenanthrene (3) 6,400Anthracene (3) 2,500Fluoranthene (4) 2,200Pyrene (4) 1,000Benzo[a]anthracene (4) 300Chrycene (4) 1,000Benzo[b]#uorene (5) 500Benzo[k]#uorene (5) 60Benzo[a]pyrene (5) 60Dibenzo[a]anthracene (5) (30Benzo[ghi]perylene (6) (30

Total identi"ed 23,600

Control pile Naphthalene (2)a 500Asenaphtene (3) 7,100Fluorene (3) 1,900Phenanthrene (3) 6,400Anthracene (3) 2,500Fluoranthene (4) 2,200Pyrene (4) 1,000Benzo[a]anthracene (4) 300Chrycene (4) 1,000Benzo[b]#uorene (5) 500Benzo[k]#uorene (5) 60Benzo[a]pyrene (5) 60Dibenzo[a]anthracene (5) (30Benzo[ghi]perylene (6) (30

Total identi"ed 23,600

aNumber of rings in the molecule.

For the solid-phase luminescent bacteria test homogen-ized soil subsamples of 5 g were weighed to 50-ml centrifugetubes and diluted in 16 ml deionized water. This suspensionwas amended with 2 ml of 20% NaCl solution and with aninoculum of 2 ml of overnight-grown <ibrio ,scheri (strainNRRL B 1117) luminescent bacteria. Tubes were incubatedfor 15 min at 153C in a water bath. After incubation thetubes were quickly centrifuged (5 min, 1660 rpm, 300g) toseparate solids from the exposed bacteria. The luminescenceof a 1-ml sample of this supernatant was measured witha luminometer (BioOrbit 1253 Model, Bio Orbit, Turku,Finland). The sample luminescence was then compared withluminescence in deionized water.The Toxi-Chromo Pad test with soil samples was

carried out according to the manufacturer's instructions(Environmental Biodetection Products, Inc., Ontario,Canada). Freeze-dried E. coli were rehydrated and added to

E 2the Mycobacterium-Inoculated Pile and Noninoculated Controlend of the Next Summer (October) at Small-Scale Composts

Concentration, mg/kg (% of reduction from start)

After "rst period After second period

(10 (98) 0 (100)(10 (99) 70 (99)3,400 170 (91)14,300 430 (93)3,300 1,400 (44)5,200 1,520 (31)2,700 900 (10)600 3002,000 400 (60)220 (56) 100 (80)90 100110 70(10 20(10 20

31,980 5,500 (77)

(10 (98) 0 (100)(10 (99) 0 (100)330 (83) 50 (97)1,700 (73) 100 (98)2,700 1,000 (60)2,300 1,500 (32)1,400 900 (10)330 3001,100 400 (60)120 (76) 100 (80)50 (17) 10050 (17) 70(10 20(10 0,02

10,120 (57) 4,560 (81)

TABLE 3Soil Toxicity Assessed by Solid-Phase Luminescent Bacteria

Test, Standard Luminescent Test, and P. putida Growth Inhibi-tion Test, Total PAH content, and Reduction of Inoculated Pileand Noninoculated Control Pile (at small-scale Composts) at theBeginning, after One Summer, and at the End of the NextSummer

At startAfter "rstsummer

After secondsummer

Inoculated pileLUM Solid 1,540a 714 (54%)b 285 (81%)LUM Elutr. 500 154 (69%) 100 (80%)PGI Elutr. 3,330 222 (93%) 91 (97%)

Total PAHs 23,600 mg/kg 31,800 mg/kg (*) 5500 mg/kg (77%)

Control pileLUM Solid 1,540 286 (81%) 250 (84%)LUM Elutr. 500 33 (93%) 100 (80%)PGI Elutr. 3,330 222 (93%) 100 (80%)

Total PAHs 23,600 mg/kg 102,10 mg/kg (57%)45,600 mg/kg (81%)

aSTU"1/EC50�1000.bPercentage reduction.

326 AHTIAINEN ET AL.

the reaction mixture. After 20 min preincubation 0.5 ml ofthis solution was mixed with 0.5 g of homogenized fresh soil.The mixture was incubated at 373C for 2 h. The mixture washomogenized again and a small drop of incubated soil}bac-teria mixture was pipetted onto a chromogenic pad (a glassmicro"ber "lter soaked with blue chromogenic substrate for�-galactosidase). The pads were incubated at 373C over-night for color development. The intensity of enzyme syn-thesis was indicated by the color blue; possible toxicityreduced enzyme synthesis and color development.Freshly homogenized soil samples were eluted to de-

ionized water by weighing 5 g of soil to 50 ml of deionizedwater and shaking overnight at room temperature. The pHof the elutriates was neutralized to 7 to obtain comparableresults. These soil elutriates were tested according to thestandardized luminescent bacteria test (ISO 11348, Part 3,1998). The luminescence inhibition test was accomplishedby combining 500 �l elutriate with 500 �l luminescent<. ,scheri NRRL B-11177 suspension. The test tubes werethen incubated in a 15$13C water bath. Luminescence wasmeasured after 30 min incubation with a luminometer(BioOrbit 1253). Inhibition of luminescence was calculatedby comparing sample luminescence with that of the controlcontaining deionized water.The elutriates were also tested by automated modi"ca-

tion of the standard P. putida test (ISO 10712, 1995) witha Bioscreen C analyzer (Labsystems, Helsinki, Finland). Inthis test P. putida MIGULA (DSM 50026) were grown for16 h in a liquid medium in special cuvettes and turbidity due

to bacterial growth was measured by vertial photometry.Toxicity was assessed as growth inhibition (%) in di!erentdilutions of the soil extracts compared with deionized water.The pH of the samples was checked and they were "rstcentrifuged for 10 min/500 g and then sterile-"ltered (0)2�m) to eliminate background bacterial growth. In this test,the nutrients in the tested extract can also stimulate bacter-ial growth and mask possible toxic e!ects.

Expression of Toxicity

The toxicity of the soil samples or soil elutriates wasestimated as EC50 values on the basis of dose}responsecurves and expressed as soil toxicity units (STU"1/EC50�1000) when possible, except for the qualitative Toxi-ChromoPad test.

RESULTS

Chemical characteristics of the soils in both cases beforecomposting are summarized in Table 1. The concentrationsof arsenic, chromium, and copper were very high in bothcases. Also, both soils were heavily contaminated with thecreosote oil.In the small case study, the concentrations and reduction

of analyzed and identi"ed PAHs in the Mycobacterium-inoculated pile and in the noninoculated control pile at thestart and during the test are listed in Table 2. At the startmost of the PAHs (about 78%) were of low molecular size(two and three rings) and hence rather quickly biodegrad-able. However, the total amount of PAHs was high enoughto cause potential inhibition of biological activity.Total PAH content was reduced by 57% in the control

pile during the "rst summer of composting (Table 2). Naph-thalene content was reduced by 98% and asenaphthene by99%, and in the noninoculated control pile, #uorene by83% and phenanthrene by 76%. Chemical analysis in-dicated an increase in PAHs in the inoculated pile. How-ever, after the second summer the amounts of PAHs werereduced to the same low levels as in the control pile, with80% reduction.The toxicity and total PAH content of soil and their

reduction at the beginning and during composting are pre-sented in Table 3. The solid-phase luminescent bacteria test,standard luminescence test, and P. putida growth inhibitiontest revealed a similar increase in toxicity following anincrease in PAH content in the inoculated pile. Also, thereduction of toxicity during composting followed PAH con-centrations.In the larger case, the concentrations of PAHs in the

compost during composting after the peroxide pretreat-ment, starting from 22 May, are presented in Table 4. TotalPAH concentration decreased 76% during the 163 days ofcomposting. Degradation and reduction were 96% for

TABLE 4Concentrations of PAHs during 163 Days of Composting at Larger-Scale Compost

Time of sampling (days from start)Reduction

Compound 0 42 71 91 133 163 from start

Concentration (mg/kg dry wt)Naphthalene (2)a 66.6 22.2 7.7 2.5 0.3 1.0 98%1-Methylnaphthalene (2) 97.9 63.8 27.1 10.1 1.3 0.9 99%2-Methylnaphthalene (2) 115.8 58.7 22.8 5.5 0.8 0.7 99%2,6-Dimethylnaphthalene (2) 95.7 80.8 39.9 23.2 2.8 2.5 97%2,3,5-Trimethylnaphthalene (2) 31.9 30.9 22.6 15.5 2.4 3.3 88%Biphenyl (2) 57.6 20.1 7.8 2.4 0.3 0.4 99%

Acenaphthene (3) 979.1 825.7 512.4 279.0 53.2 51.8 95%Acenaphthylene (3) 23.2 16.0 21.5 16.2 3.5 6.8 71%Fluorene (3) 810.6 581.6 311.0 182.5 29.9 20.7 97%Phenanthrene (3) 2005.4 956.2 508.1 299.8 51.0 31.0 98%1-Methylphenanthrene (3) 56.2 68.2 58.8 36.7 5.5 9.1 84%Anthracene (3) 311.5 243.4 197.9 130.6 26.3 40.7 86%

Fluoranthene (4) 2038.2 1761.8 1967.0 1411.0 1232.0 743.1 64%Pyrene (4) 1085.8 1115.8 1315.6 1031.3 1036.0 569.7 48%Benzo[a]anthracene (4) 360.2 346.9 321.7 324.3 109.8 194.2 46%Chrysene (4) 423.3 397.4 403.7 390.0 337.0 255.7 40%

Benzo[b#k]#uoranthene (5) 329.6 284.0 288.0 340.0 83.2 262.0 21%Benzo[e]pyrene (5) 155.5 89.2 128.6 142.7 32.1 76.0 51%Benzo[a]pyrene (5) 194.3 117.6 165.1 185.1 41.4 86.0 55%Perylene (5) 132.2 50.1 114.3 119.0 10.1 24.1 82%Dibenzo[a,h]anthracene (5) 137.4 50.6 103.0 110.7 6.6 15.5 89%

Indeno[1,2,3-cd]pyrene (6) 160.3 60.7 129.5 140.9 13.9 30.0 81%Benzo[ghi]perylene (6) 120.3 38.5 103.7 111.7 12.5 23.8 80%

aNumber of rings in molecule.

327MONITORING COMPOSTING OF CREOSOTE-CONTAMINATED SOIL

low-molecular-weight PAHs (two and three rings). Also thecompounds with four to six rings were degraded e$ciently,with 56% reduction from the start.Total PAH content of the soil during 5 months of com-

posting and the toxicity of soil elutriates in the luminescent

FIG. 1. Total PAH content of the soil (mg/liter) during the entire5-month composting period, and toxicity of soil elutriates in the lumines-cent bacteria test as soil toxicity units (STU50"1/EC50�1000).

bacteria test are presented in Fig. 1 and Table 5. There wasa continuous decrease in PAH concentration with someplateaus. These phases seemed to re#ect the temporarilyincreased concentrations of four- and "ve-ringed com-pounds and soil toxicity. At the end the reduction of PAHswas almost 80% and a reduction in toxicity was observedwith both modi"cations of luminescent bacteria tests butnot with the P. putida growth inhibition test. The toxicity ofthe soil to P. putida was rather low throughout composting.However, the observed oscillation in toxicity based on STUunits calculated on the basis of EC10 values correlated wellwith the other tests (Fig. 2, Table 4). The Toxi-ChromoPadtest indicated a similar reduction in toxicity to enzymesynthesis in bacteria. After 2�

�weeks of composting no

inhibition could be observed.

DISCUSSION

Heavy contamination of soil with high concentrations ofmetal salts (As, Cr, Cu), as well as creosote, did not totallyinhibit microbial activity and the composting processes. Theadded wood bark and nutrients and mixing of the compost

TABLE 5Total Polyaromatic Hydrocarbon and Total Hydrocarbon Concentrations, and Toxicity of the Soil Elutriate in the Luminescent

Bacteria Test and P. putida Growth Inhibition Test, Reduction (%), and Qualitative Toxicity of Soil in Toxi-Chromo Tests duringLarge-Scale Composting

Time of sampling (days from start of composting 22 May)

0 42 71 91 133 163

Total PAHsa 10,106 7,545.3 (25) 6,777.7 (33) 5,311.0 (47) 3,091.8 (69) 2,450.0 (76)Total HCsa 25,000 19,000 (24) 18,000 (28) 17,000 (32) 14,000 (44) 11,000 (56)LUM elutr. (as STU50) 200 77 (62) 167 (17) 67 (67) 71 (65) ndb

PGI elutr. (as STU10) 62.5 20.8 (67) 50 (20) 20 (68) 91 ndToxi-ChromoPadc ! ! ! # # nd

amg/kg dry wt.bnd, not done.c#when enzyme is synthesized.

328 AHTIAINEN ET AL.

piles provided favorable conditions for biodegradation ofPAHs and other hydrocarbons.The increase in certain analyzed and identi"ed PAHs in

the inoculated pile in the small-scale composting is note-worthy. A probable explanation for this is the heterogeneityof the compost. The tarry creosote was unevenly distributedin the pile. That made representative sampling of the piledi$cult. If the PAH content of the inoculated pile (Table 2)is compared at the start and after the "rst summer, it can beseen that concentrations of all the PAH components areabout two times higher. This supports the heterogeneity ofthe samples. Even if the homogenized sample was composedof several subsamples after mixing of the whole pile, it is stillpossible that it contained more creosote than the pile onaverage. Even one &&hot spot'' could be responsible for this.Analysis of several composite samples could have revealed

FIG. 2. Toxicity of soil elutriates as EC50-based soil toxicity units(STU50) with luminescent bacteria test (LUMelutr.) and as extrapolated soiltoxicity units with Pseudomonas growth test (PGIelutr.), and inhibition ofluminescence in solid-phase test (LUMsoil inh%) during composting.

this. Increased toxicity was not observed in toxicity testsafter the "rst summer (Table 3), although the PAH concen-tration was increased.After the second summer almost complete degradation of

PAHs and a similar reduction in toxicity were noted in bothpilot piles. Also, in the larger-scale study, a #uctuating butactual reduction of toxicity was recorded during the5 months of composting. Plateaus in degradation were alsonoted in all toxicity tests as temporarily increased toxicity.The explanation could be an increase in individual PAHswith higher toxicity like benzo[a]pyrene. Also, certain un-known transformation products and other hydrocarbonslike aromatic hydrocarbons of high toxicity or degradationto more toxic individual compounds could cause toxicity.The microbial toxicity test, especially the solid-phase

luminescent bacteria test, proved to be a practical tool forassessing toxicity during bioremediation of creosote-con-taminated soil. Even if the high metal content causes its ownbasic level of toxicity during composting it did not mask thetoxicity caused by changing concentrations of PAHs.The semiquantitative or more qualitative Toxi-Chromo-

Pad test also gave an indication of reduced toxicity duringcomposting. The method can be used for toxicity quanti"ca-tion using a liquid modi"cation of the test on soil elutriates(Dutka et al., 1996).A novel possibility is a kinetic modi"ed method of the

luminescent bacteria test that can be used to assess soilsamples (Lappalainen et al., 1999). In this method a kineticmeasurement of the luminescence signal is started at thesame time as the bacteria are added to the sediment sample.Luminescence is measured throughout the 30-s exposureand the possible luminescence inhibition is compared withthe initial peak luminescence at the start. This should helpto avoid some of the color or adsorption interferences.Juvonen et al. (2000) found the luminescent bacteria test,especially with the above-mentioned #ash modi"cation,the most suitable test for screening during composting of

329MONITORING COMPOSTING OF CREOSOTE-CONTAMINATED SOIL

various hydrocarbons containing soil. Obviously, the sensi-tivity of the luminescent bacteria tests for soil toxicity as-sessment depends on the soil contaminants, theirbioavailability, and their modes of toxic action. The test andits di!erent modi"cations seem to detect the toxicity causedby PAHs, as also observed in this study, nitroaromaticcompounds (Maxam et al., 2000), and chlorophenols(Laine et al., 1997) in studies of soil contamination andremediation.Luminescent bacteria assays measure the e!ect on energy

metabolism, the Toxi-Chromo-Pad test indicates inhibitionof enzyme synthesis and the P. putida growth inhibition testmeasures the e!ects on cell multiplication. These three bac-terial tests form a relevant triad of simple tests measuringtwo important and separate impacts of contaminants. How-ever, in the assessment of the eventual environmental risk ofsoil contaminationwith hazardous substances or evaluationof soil remediation processes a set of di!erent trophic levelbiotests should be used, and also other environmental con-cerns like genotoxicity (Belkin et al., 1994) should be evalu-ated. This has proven to be essential, because variousorganisms and hence toxicity tests have di!erent sensitivi-ties to various contaminants (Maxam et al., 2000).

CONCLUSIONS

Inherent microbes of soil were able to degrade PAHsunder composting conditions, when the environmental con-ditions were optimized. Inoculation of the compost withknown PAH degraders did not speed up the process mark-edly. Toxicity tests on elutriates and soil seemed to re#ectrather a reduction in PAHs in both cases. However, theelutriate tests were less sensitive at the end of composting.This could possibly be explained by the limited extractabil-ity of certain compounds to the elutriate.Direct soil-contact modi"cation of the <. ,scheri

luminescent bacteria test appears to be a cheap and suitabletool for assessing toxicity changes during bioremediationprocesses.

ACKNOWLEDGMENTS

The authors thank Miia Aalto and Riitta Mero for skillful technicalassistance.

REFERENCES

Belkin, S., Stieber, M., Frimmel, F. H., Abeliovich, A., Werner, P., andUlitzur, S. (1994). Toxicity and genotoxicity enhancement during poly-

cyclic aromatic hydrocarbon biodegradation. Environ. ¹oxicol. =aterQual. 9, 303}309.

Bitton, G., and Dutka, B. J. (Eds.) (1986). Introduction and review ofmicrobial and biochemical toxicity screening problems. In ¹oxicity¹esting ;sing Micro-organisms, Vol. 1, pp. 1}8. CRC Press, BocaRaton, FL.

Bringmann, G., and KuK hn, R. (1977). Limiting values for the damagingaction of water pollutants to bacteria Pseudomonas putida and greenalgae Scenedesmus quadricauda in cell multiplication inhibition test. Z.=asser Abwasser-Forsch. 10, 87}98.

Brower, H., Murphy, T., and McArdle, L. (1990). A sediment-contactbioassay with Photobacterium phosphoreum. Environ. ¹oxicol. Chem.9, 1353}1358.

Bulich, A. A., Greene, M. W., and Isenberg, D. L. (1981). The reliability ofthe bacterial luminescense assay for the determination of toxicity of purecompounds and complex e%uents. In Aquatic ¹oxicology and HazardAssessment (D. R. Branson and K. L. Dickson, Eds.), STP 737,pp. 338}347. Am. Soc. for Testing and Materials.

Bundy, J. G., Morris, A. W. J., Durham, D. G., Campbell, C. D., and Paton,G. I. (2001). Development of QSARs to investigate the bacterial toxicityand biotransformation potential of aromatic heterocyclic compounds.Chemosphere 42, 885}892.

Cerniglia, C. (1992). Biodegradation of polycyclic aromatic hydrocarbons.Biodegradation 3, 351}368.

Dutka, B. J., McInnis, R., Jurkovic, A., Liu, D., and Castillo, G. (1996).Water and sediment ecotoxicity studies in Temuco and Rapel RiverBasin, Chile. Environ. ¹oxicol.=ater Qual. 11, 237}247.

Hastings, J. W. (1978). Bacterial luminescence: An overview. In Methodsof Enzymology (M. Deluca, Ed.), Vol. 57, pp. 123}153. Academic Press,New York.

ISO 10712 (1995). =ater Quality: Pseudomonas putida Growth Inhibition¹est (Pseudomonas Cell Multiplication Inhibition ¹est).

ISO 11348, Part 3 (1998). =ater Quality: Determination of the InhibitoryE+ect of =ater Samples on the ¸ight Emission of <ibrio ,scheri(¸uminescent Bacteria ¹est).

Juvonen, R., Martikainen, E., Schultz, E., Joutti, A., Ahtiainen, J., andLehtokari M. (2000). A battery of toxicity tests as indicators of decontami-nation in composting oily waste. Ecotoxicol. Environ. Saf. 47, 156}166.

Kwan, K. K. (1995). Direct sediment toxicity testing procedure usingsediment-chromotest kit. Environ. ¹oxicol.=ater Qual. 9, 193}196.

Laine, M. M., Ahtiainen, J., Was gman, N., OG berg, L. G., and JoK rgensen,K. S. (1997). Fate and toxicity of chlorophenols, polychlorinateddibenzo-p-dioxins, and dibenzofurans during composting ofcontaminated sawmill soil. Environ. Sci. ¹echnol. 31, 3244}3250.

Lappalainen, J., Juvonen, R., Vaajasaari, K., and Karp, M. (1999). A new#ash method for measuring the toxicity of solid and coloured samples.Chemosphere 38, 1069}1083.

Maxam, G., Rila, J.-P., Dott, W., and Eisentraeger, A. (2000). Use ofbioassays for the assessment of water-extractable ecotoxic potentialof soils. Ecotoxicol. Environ. Saf. 45, 240}246.

Mueller, J. G., Chapman, P. J., and Prichard, P. H. (1989). Creosotecontaminated sites: Their protential for bioremediation. Environ. Sci.¹echnol. 23, 1197}1201.

Wenzel, A., Nendza, M., Hartman, P., and Kanne, R. (1997). Test batteryfor the assessment of aquatic toxicity. Chemosphere 35, 307}322.