Identification of abiotic and biotic reductive dechlorination in a chlorinated ethene plume after

20 Pages·2017·4.05 MB·English

by  Alice Badin

Downloaded from orbit.dtu.dk on: Jan 28, 2023

Identification of abiotic and biotic reductive dechlorination in a chlorinated ethene
plume after thermal source remediation by means of isotopic and molecular biology
tools
Badin, Alice; Broholm, Mette Martina; Jacobsen, Carsten S.; Palau, Jordi; Dennis, Philip; Hunkeler,
Daniel
Published in:
Journal of Contaminant Hydrology
Link to article, DOI:
10.1016/j.jconhyd.2016.05.003
Publication date:
2016
Document Version
Publisher's PDF, also known as Version of record
Link back to DTU Orbit
Citation (APA):
Badin, A., Broholm, M. M., Jacobsen, C. S., Palau, J., Dennis, P., & Hunkeler, D. (2016). Identification of abiotic
and biotic reductive dechlorination in a chlorinated ethene plume after thermal source remediation by means of
isotopic and molecular biology tools. Journal of Contaminant Hydrology, 192, 1-19.
https://doi.org/10.1016/j.jconhyd.2016.05.003

General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright
owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

 Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
 You may not further distribute the material or use it for any profit-making activity or commercial gain
 You may freely distribute the URL identifying the publication in the public portal

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
JournalofContaminantHydrology192(2016)1–19
ContentslistsavailableatScienceDirect
Journal of Contaminant Hydrology
journal homepage: www.elsevier.com/locate/jconhyd
Identification of abiotic and biotic reductive dechlorination in a
chlorinated ethene plume after thermal source remediation by
means of isotopic and molecular biology tools
AliceBadina,MetteM.Broholmb,CarstenS.Jacobsenc,JordiPalaua,
PhilipDennisd,DanielHunkelera,⁎
aUniversityofNeuchâtel,CentreforHydrogeology&Geothermics(CHYN),RueEmileArgand11,CH2000Neuchâtel,Switzerland
bTechnicalUniversityofDenmark(DTU),DepartmentofEnvironmentalEngineering,Miljøvej,DTUB113,DK2800Kgs.Lyngby,Denmark
cGeologicalSurveyofDenmarkandGreenland(GEUS),DepartmentofGeochemistry,Ø.Voldgade10,1350KøbenhavnK,Denmark
dSiREM,130ResearchLane,Guelph,Ontario,N1G5G3,Canada
a r t i c l e i n f o a b s t r a c t
Articlehistory: Thermaltetrachloroethene(PCE)remediationbysteaminjectioninasandyaquiferledtothe
Received23December2015 releaseofdissolvedorganiccarbon(DOC)fromaquifersedimentsresultinginmorereduced
Receivedinrevisedform6May2016 redoxconditions,acceleratedPCEbiodegradation,andchangesinmicrobialpopulations.These
Accepted17May2016 changesweredocumentedbycomparingdatacollectedpriortotheremediationeventandeight
Availableonline26May2016
years later. Based on the premise that dual C-Cl isotope slopes reflect ongoing degradation
pathways,theslopesassociatedwithPCEandTCEsuggestthepredominanceofbioticreductive
Keywords: dechlorinationnearthesourcearea.PCEwasthepredominantchlorinatedethenenearthesource
Chloroethenes areapriortothermaltreatment.Afterthermaltreatment,cDCEbecamepredominant.Thebiotic
Stableisotopes
contributiontothesechangeswassupportedbythepresenceofDehalococcoidessp.DNA(Dhc)
Molecularbiology
andDhctargetedrRNAclosetothesourcearea.Incontrast,dualC-Clisotopeanalysistogether
Thermaltreatment
withthealmostabsentVC13CdepletionincomparisontocDCE13CdepletionsuggestedthatcDCE
wassubjecttoabioticdegradationduetothepresenceofpyrite,possiblesurface-boundiron(II)
orreducedironsulphidesinthedowngradientpartoftheplume.Thisinterpretationissupported
bytherelativelackofDhcinthedowngradientpartoftheplume.Theresultsofthisstudyshow
thatthermalremediationcanenhancethebiodegradationofchlorinatedethenes,andthatthis
effectcanbetracedtothemobilisationofDOCduetosteaminjection.This,inturn,resultsinmore
reducedredoxconditionswhichfavoractivereductivedechlorinationand/ormayleadtoaseries
ofredoxreactionswhichmayconsecutivelytriggerbioticallyinducedabioticdegradation.
Finally, this study illustrates the valuable complementary application of compound-specific
isotopic analysis combined with molecular biology tools to evaluate which biogeochemical
processesaretakingplaceinanaquifercontaminatedwithchlorinatedethenes.
©2016ElsevierB.V.Allrightsreserved.
1.Introduction particularly adapted for source treatment in subsurface
sediments of relatively high permeability such as sandy
Managementof sites contaminated withchlorinated eth- aquifers(vonSchnakenburg,2013).Thisremediationstrategy
enesisknowntobechallenging.Amongthevariousdeveloped isknowntoreleasedissolvedorganiccarbon(DOC)(Friisetal.,
remediationmethods,thermaltreatmentbysteaminjectionis 2005)theincreaseofwhichmaytriggerachainofmicrobially-
mediatedredoxprocesses.Whennaturalattenuationhasbeen
observed prior to active source remediation, steam injection
⁎ Correspondingauthor. mightthusinfluencethenaturallyoccurringdegradation.
http://dx.doi.org/10.1016/j.jconhyd.2016.05.003
0169-7722/©2016ElsevierB.V.Allrightsreserved.
2 A.Badinetal./JournalofContaminantHydrology192(2016)1–19
Natural degradation of chlorinated ethenes might occur turn affect the likelihood that biotically induced abiotic
bioticallyduetothepresenceofadequateactivemicroorgan- degradation will take place (Tobiszewski and Namieśnik,
ismsinspecificredoxconditions,aswellasabioticallyinthe 2012).
presence of reduced iron (Fe) minerals. Sequential biotic Inordertoexploretheoccurrenceofsuchprocessesinthe
reductivedechlorinationoftheubiquitousgroundwatercon- subsurfaceandthusevaluatetheeffectofremediationorsite
taminanttetrachloroethene(PCE)consecutivelyyieldstrichlo- management,varioustoolsmaybeemployed.
roethene(TCE),cis-dichloroethene(cDCE),vinylchloride(VC) In recent decades, compound specific isotopic analysis
and eventually non-toxic ethene. This process takes place in has been increasingly used to explore chlorinated ethene
strictlyanaerobicsystems(Wiedemeieretal.,1999;Bradley, degradation processes. It was demonstrated that the extent
2000) and is the most commonly encountered naturally of biodegradation could be determined based on isotopic
occurringbioticdegradationofchlorinatedethenes.According measurements (Hunkeler et al., 2010). Additionally, it was
to laboratory and field observations, PCE could undergo suggested that dual C\\Cl isotopic analysis may help to
reductive dechlorination in virtually all anaerobic conditions differentiate degradation pathways, for example biotic from
while reductive TCE, cDCE and VC dechlorination would abioticdegradation(Elsneretal.,2005;Abeetal.,2009;Audí-
generally occur in more reduced conditions, such as iron- Miróetal.,2013)despitesomelimitations(Badinetal.,2014;
reducingforTCEandideallysulfate-reducingtomethanogenic Renpenningetal.,2014).Therangeoflaboratorydetermined
forcDCEandVC(Vogeletal.,1987;Chapelle,1996;Bradley, dualC\\Clisotopeslopesassociatedwithvariouschlorinated
2000; Tiehm and Schmidt, 2011). The presence of and ethene degradation processes has increased in recent years,
competition for molecular hydrogen (H ), a key electron thusenrichingthedatabasetowhichdualC\\Clisotopeslopes
2
donor, can also be a determining factor (Ballapragada et al., measuredinthefieldcanbecomparedforprocessidentifica-
1997).Dependingonitsmineralogy,thepresenceofironmay tion(Fig.1)(Abeetal.,2009;Audí-Miróetal.,2013;Cretniket
alsoinducecompetitiveinhibitionofchlorinatedethenebiotic al.,2013;Kuderetal.,2013;Wiegertetal.,2013;Badinetal.,
reductive dechlorination (Paul et al., 2013). The occurrence 2014;Cretniketal.,2014;Renpenningetal.,2014).Moreover,
of reductivedechlorination furtherdepends on the presence rapidadvancesinmolecularbiologyopennewpossibilitiesfor
andactivityofspecificdechlorinatingmicroorganisms.Mem- thecharacterizationofmicrobialcommunitiespresentinthe
bers from various bacterial genera such as Sulfurospirillum, subsurfaceandtheassessmentoftheiractivitybasedonmRNA
Dehalobacter,Desulfitobacterium,DesulfuromonasorGeobacter analysis.Forexamplebacterial16SrRNAgenepoolcharacter-
have been reported to catalyze some steps of chlorinated ization via amplicon pyrosequencingcan be used to identify
ethenereductivedechlorination.However,whilesomeenrich- the present bacterial communities in high detail (Novais
mentculturesandconsortiaareabletodechlorinatetoethene andThorstenson,2011).Itwasmoreoverdemonstratedthat
(Flynnetal.,2000;Aulentaetal.,2002;Duhameletal.,2002; pyrotagsequencingisarobustandreproduciblemethodthat
HoelenandReinhard,2004),todate,theonlyorganismswhich can be used for reliable microbialcommunityexploration in
have been reported tocatalyzecompletereductive dechlori- naturalsystems(Pillonietal.,2012).Additionally,ascDCEand
nationtoethenearesomespeciesofthegenusDehalococcoides VCdegradationusuallyrepresentsthebottleneckofchlorinat-
(Dhc) (Löffler et al., 2013). cDCE is hence often found to ed ethene natural attenuation, it is essential to screen for
accumulateinthesubsurface.Microbialoxidationmightalso markers of their degradation. Dhc screening has thus been
takeplace,particularlyinthecaseofcDCEandVC(Hartmanset carried out in numerous studies since it is the only class of
al., 1985; Bradley and Chapelle, 1998, Bradley and Chapelle, microorganismsreportedtoperformcDCEandVCdechlorina-
2000).Despitetheirpresenceinthesubsurface,microorgan- tion.Furthermore,assessingthepresenceofgenesthatencode
ismsmaydisplayalowactivity,aresometimesinactiveorare forVCreductivedehalogenases(rdhA)knowntocatalyzeVC
even dormant (Meckenstock et al., 2015), which may addi- reductiontoethene,suchasvcrA,andmeasuringgenes'mRNA
tionally hinder reductive dechlorination. Abiotic reductive level constitutes a stronger line of evidence to support
dechlorinationcanalsotakeplacenaturally,providedthatthe complete reductive dechlorination. The vcrA and bvcA genes
adequate minerals and geochemical conditions are present. identified in Dhc are so far the only two functional genes
Iron sulphidessuchasmackinawite(FeIIS) or pyrite(FeIIS ), describedtoencodeVCrdhA(Krajmalnik-Brownetal.,2004;
2
iron oxides such as magnetite (FeIIO·FeIII O ), and iron Mülleretal.,2004)andtheirpresenceinfieldsamplesfrom
2 3
hydroxidessuchasgreenrusts,whicharecorrosionproducts sitescontaminatedwithchlorinatedetheneswassuccessfully
ofironorsteel([FeII FeIII(OH) )]X+[(A) yH O]X−whereAis relatedtocompletedechlorination(Scheutzetal.,2008;van
6-x x 12 x/n 2
ananion,typicallySO2−orCl−),havebeenreportedtocatalyze derZaanetal.,2010;Damgaardetal.,2013a).Itwasmoreover
4
abiotic reductive degradation yielding less chlorinated com- shownbasedonfieldsamplesthatrdhAgenesdirectlyinvolved
poundsandothernon-toxiccompoundssuchasacetylenein in dechlorination should be targeted in addition to Dhc, as
variousproportions(TobiszewskiandNamieśnik,2012).Pyrite differentmicrobialspeciesmightharbourvcrAandbvcAgenes
isknowntoreduceallchlorinatedethenes(LeeandBatchelor, duetohorizontalgenetransferandarethereforealsoableto
2002a)whilemackinawitewasshowntoreducePCEandTCE dechlorinateVCdowntoethene(vanderZaanetal.,2010).
butwasnon-reactivewithcDCE(ButlerandHayes,1999,Jeong Finally,targetingmRNAconstitutesanessentialcomplemen-
et al., 2011). Surface-bound Fe(II) is also known to catalyze taryanalysisasitwilladditionallyinformontheactivityofthe
abiotic degradation of reducible contaminants (Elsner et al., correspondingdegrader(Bælumetal.,2013).
2004; McCormick and Adriaens, 2004, Han et al., 2012). Here we combine such innovative methods to assess the
Furthermore,theactivityofvariousbacteriainthesubsurface impact of source remediation by steam injection on a
maychangethelocalredoxconditions.Thismightaffectthe chlorinated ethene plume that occurrs in a complex biogeo-
redoxpotentialofmetalscontainedinminerals,whichmightin chemical system where iron minerals are present. More
A.Badinetal./JournalofContaminantHydrology192(2016)1–19 3
Fig.1.LiteraturevaluesfordualC\\Clisotopeslopesassociatedwithvariousdegradationpathwaysofchlorinatedethenes(Abeetal.,2009;Audí-Miróetal.,2013;
Cretniketal.,2013;Kuderetal.,2013;Badinetal.,2014;Cretniketal.,2014;Renpenningetal.,2014).Slopesaregivenforeachofthesubstrates.Greenarrowsrepresent
dihaloelimination,i.e.themainpathwayforabioticreduction,whichtypicallyoccurrsinpresenceofzero-valentFe.Bluearrowsrepresentaerobicoxidationof
chlorinatedethenes.Redarrowscorrespondtohydrogenolysisoccurringduringreductivedechlorinationmediatedbybacteriaviatheirspecificcorrinoid-containing
reductivedehalogenaseenzymes.DualC\\Clisotopeslopesassociatedwithhydrogenolysisbytheseenzymesandcorrinoidsaswellasbychemicalmodelsmimicking
corrinoids(cobalamin,cobaloxime)werealsorecentlyreported(Renpenningetal.,2014).However,duetothelackofcorrelationsbetweentheseslopesandslopes
associatedwithbacteriallymediateddechlorination,thesevaluesarenotreportedhere.(Forinterpretationofthereferencestocolorinthisfigurelegend,thereaderis
referredtothewebversionofthisarticle.)
specifically, the aim was to evaluate if a plume detachment operated between 1964 and 2001. A sandy aquifer N50 m
occurredduetosteaminjectionasreportedbyapreviousstudy thickoccasionallycontaininglesspermeablesiltandclaylenses
(SleepandMa,1997),andinaddition,ifnaturaldegradation was characterized based on former site characterization
wasstimulatedbythethermaltreatment. campaigns.Achlorinatedetheneplumeof~2kmlengththat
Amajoradvantageofthisfieldsiteresidesinthefactthat followsthegroundwaterflow(Fig.2)southwardfromthesite
the plume was formerly well characterized and studied and turns southeast after 1 km was identified based on
(Hunkeleretal.,2011),whichallowsforcomparisonbetween extensivemonitoringwellcover(55multilevels).Thediffer-
before and after (7–8years later) source remediation. An enceinequipotentiallinesbetween2006and2014indicates
extensivecampaignwascarriedouttoevaluatetheimpactof thatthegradientislesssteepin2014thanin2006(Fig.2).Low
steaminjection where redox parameters,chlorinated ethene amounts of organic matteraswell ashigh levels of iron are
concentrationsandisotopiccompositionsthereof,454pyrotag expectedinaquifersinthispartofJutland(Postmaetal.,1991).
sequencing, Dhc DNA and rRNA, bvcA and vcrA functional An average groundwater velocity of 0.24 m·day-1 was
genes,andgenetranscripts(mRNA)wereanalysedinsamples previouslyestimated(Hunkeleretal.,2011),thoughthismay
taken from 20 wells at different screening depths along the varylocallyduetodifferenthydraulicconductivitiesresulting
plumeflowline. fromthelargegrainsizedistribution.
Thermalremediationbysteaminjectionwasappliedtothe
2.Materialsandmethods sourcezonebetweenOctoberandDecember2006.Theentire
plume was sampled in 2006 before remediation and some
2.1.Studysitedescription pointsfurtheroutintheplumethatwerenotyetimpactedby
remediation were sampled in 2007. The data previously
ThestudiedsitepreviouslydescribedbyHunkeleretal.is discussedbyHunkeleretal.(Hunkeleretal.,2011)consistof
locatedinsouthernJutland,Denmark,inthetownofRødekro datacollectedduringthesetwosamplingcampaignsandare
(Hunkeleretal.,2011).Briefly,theoriginalPCEgroundwater referredtoasdatafrom2006,forsimplificationpurposes.The
contamination comes from a dry-cleaning facility which mainconclusionsdrawnfromthepreviouscampaignwerethat
4 A.Badinetal./JournalofContaminantHydrology192(2016)1–19
Fig.2.Groundwaterequipotentiallinesfrom2006(blue)and2014(red),approximateplumeflowline(bluedashedarrow)andmonitoringwells(greyandorange
targets).Wellssampledin2014appearinorange.(Forinterpretationofthereferencestocolorinthisfigurelegend,thereaderisreferredtothewebversionofthis
article.)
(i)PCEandTCEwerelikelybioticallydegradedbyreductive Concentrationsofredoxspeciesandchlorinatedetheneswere
dechlorinationinthefirst400mdowngradientofthesource, regularly measured between 2004 and 2014 as part of the
(ii)cDCEwasnotaffectedbydegradation(neitherbioticnor monitoring process. The evolution of redox species between
biotically induced abiotic) in the first 1050 m downgradient 2004and2014isgiveninFig.3Bforsamplingpointsfromthe
from the source, but it was degraded between 1050 and plumecenterlinelocatedclosetothesource(B16-1;100m),in
1900 m downgradient, likely at least partially by biotic themiddleoftheplume(B34-4;1050m)andatthefrontofthe
reductivedechlorination,and(iii)VCwastransformedfurther plume(B61-2;1700m).
byundeterminedprocesses.TheprocessresponsibleforcDCE Basedontheassessedaveragegroundwatervelocity,itcan
degradationremaineduncertainduetothelackindualC\\Cl beestimatedthatspeciesmighthavebeentransportedabout
studies associated with abiotic reductive dechlorination of 600mdowngradientbetweentheendof2006andthenew
cDCEtowhichthedualC\\ClslopeobservedinRødekrocould campaign carried out in 2014. Such an estimation should
becompared. howeverbetreatedcautiouslyasheterogeneitiesmightchange
A.Badinetal./JournalofContaminantHydrology192(2016)1–19 5
Fig.3.A:Redoxsensitivespeciesconcentrationsin2006(Hunkeleretal.,2011)and2014(thisstudy).B:Evolutionofredoxspeciesconcentrationsbetween2004and
2014alongtheplumecentrelineat100m(wellB16–1),1050m(B34–4),and1700m(B61–2)downgradientfromthesource.Thesewellsare“circled”inA.Dissolved
oxygen(DO),nitrate(NO3−),Fe(II),dissolvedorganiccarbon(DOC)andmethane(CH4)concentrationsaregivenbythelefty-axiswhilesulfate(SO42−)concentrations
aregivenbytherighty-axis.DOCconcentrationsweremeasuredfromthetimeofsourceremediationinB16-1andonlyin2014inB34-4andB61-2.Thereddashedline
correspondstothesourceremediationevent.Notethattheleftgraphhasadifferentlefty-axisthantheothertwographs.(Forinterpretationofthereferencestocolorin
thisfigurelegend,thereaderisreferredtothewebversionofthisarticle.)
6 A.Badinetal./JournalofContaminantHydrology192(2016)1–19
the hydraulic conductivity by a few orders of magnitude in 1500) with detection limits of 0.43 μg·L−1, 0.94 μg·L−1,
areas where different subsurface materials such as gravel/ 0.47μg·L−1and0.38μg·L−1,respectively.
coarsesandorfinesand/silt/claylensesarepresent.Further- Isotopicanalysis:Cisotopicanalysiswasperformedatthe
more,withincreasingdistanceanddepthfromthesource,the University of Neuchâtel (CHYN, Switzerland) by the system
claylensesprogressivelydisappear.Thisresultsinadivergence previouslydescribedforsamplescontainingchlorinatedethene
oftheflowdirectionwhichcausestheplumetodivedeeper concentrationsexceeding5μg·L−1(Badinetal.,2014),except
intotheaquifer. that a QS-PLOT column was used instead of a DB-VRX to
improve VC separation. The compounds degassed by N
2
2.2.Groundwatersampling purgingwereretainedonaVocarb3000trap(VICI),transferred
toacryogenictrap(TekmarDohrmann)at−120°Ctoenable
Groundwaterin42screensfrom20differentlocationswas compoundconcentration,andsenttothegaschromatograph
sampledinMay2014afterpurging3timesthevolumeofwells (GC)column(QS-PLOT,30m,0.32mm,10μm)ofanAgilent™
andcheckingthestabilityofpH,temperature,dissolvedoxygen 7890aGCforcompoundseparation(35°Cfor6min,rampof
(O )andconductivity.Samplesfornitrate(NO−)andsulfate 15°C·min−1until130°Ckeptfor0.1minfollowedbyaramp
2 3
(SO2−)analysiswerecollectedinhardplasticbottles;samples of20°C·min−1until240°Ckeptfor5min).Aftercombustion
4
for DOC were collected in glass bottles after filtration and viaanIsoprimeGC5combustioninterface,theresultingCO gas
2
spikedwithH PO uponarrivalinthelaboratory;samplesfor wassenttoanIsoprime™100isotoperatiomassspectrometer
3 4
dissolved iron (Fe(II)) concentration were sampled in hard (IRMS)tomeasuretheCisotoperatio.Samplesweremeasured
plasticbottlesafterfiltration(sterilefilter,0.45μm)andspiked induplicate.Standarddeviationsσofthein-housereference
withHNO (topH2)uponarrivalinthelaboratory;samplesfor materialsmeasuredinthesamesequencesassamplesfromthe
3
chlorinatedetheneconcentrationwerecollectedin40mLglass fieldsitewere0.6‰(n=32),0.3‰(n=24),0.5‰(n=24),
vialsclosedwithoutheadspacewithaTeflon-coatedcapand 1.0‰(n=26)forPCE,TCE,cDCE,andVC,respectively.The
analyseduponarrivalinthelaboratory.Samplesforchlorinated standard uncertainty of duplicate measurements was deter-
etheneisotopeanalysiswerecollectedin40mLglassvialsor mined according to ISO guidelines (BIPM, 1993) asσ/√2, i.e.
1 L Schott bottles (for isotope analysis with low chlorinated 0.4‰,0.2‰,0.3‰,0.7‰forPCE,TCE,cDCE,andVC,respective-
etheneconcentrations)closedwithoutheadspacewithacap ly.Samplesthatcontainedreferencecompoundswithknown
containing a Teflon-coated septum. HNO was added previ- isotoperatios(ElementalAnalyser-IRMSmeasurement)were
3
ouslytothecontainersinordertoreachpH2whenfilledwith included in each sequence to verify the method accuracy
the sample and to stop any microbial activity. Samples for exceptforVCwhichwasnotcharacterizedbyEA.
methane,acetylene,etheneandethanewerecollectedin6mL Forthesamplesshowingchlorinatedetheneconcentrations
Exetainer®glassvials(LabCo,UK)with3mLheadspacewhich below 5 μg·L−1, 1 L bottles were manually connected to a
werepreviouslyevacuatedandfilledwith100μLconcentrated purgesystemthatconsistedofafritfromagas-washingbottle
H SO .Cautionwastakensothatnoairbubbleswereinjected asdescribedformerly(Hunkeleretal.,2012)insteadofpassing
2 4
with the groundwater samples. All aqueous samples were 40mLglassvialsbyautosampler.Thebottleswerepurgedfor
storediniceboxestoppedwithicepacksuntilarrivalatthe 30minatarateof150mL·min−1whichledtoaremovalof90,
laboratory where they were stored at 4 °C until analysis. 75, 50 and 100% of the dissolved PCE, TCE, cDCE and VC,
Samplesfor454pyrotagnextgenerationsequencinganalysis respectively,consideringHenrycoefficientsat20°Cof0.533,
werecollectedbypassing300–400mLofgroundwaterthrough 0.314, 0.14, 0.891 (gas/water) for PCE, TCE, cDCE and VC,
Sterivex™Filters(EMDMillapore,Billerica,MA,USA)andwere respectively.
then shipped to Guelph (SIREM, commercial laboratory, Cl isotopic analysis of PCE and TCE was performed as
Canada)withicepacks.SamplesforDhcDNAandrRNA,bvcA previously describedwithanAgilent7890GCcoupledto an
andvcrAfunctionalgenes(DNA),andgenetranscript(mRNA) Agilent5975Cquadrupolemassselectivedetector(SantaClara,
analysiswerecollectedasdescribedpreviouslyinBælumetal., CA,USA)(Badinetal.,2014).ADB-5column(30m,0.25mm,
2013snap-shotfrozeninliquidN andkeptat−80°Cuntil 0.25μm,Agilent)withaconstantheliumflowof1.2mL·min−1
2
analysisinCopenhagen(GEUS,Denmark). wasused to perform chromatographic separation.Molecular
ionsweretargetedandcalculationswerecarriedoutaccording
2.3.Analyses tothemethoddevelopedbyAepplietal.,2010.Calibrationwith
twoexternalstandards(δ37Cl =+0.3‰andδ37Cl =
EIL1 EIL2
Chemical analyses: NO−, Fe(II), SO2− and DOC were −2.5‰ for PCE and δ37Cl = +3.05‰ and δ37Cl =
3 4 EIL1 EIL2
measuredbyaccreditedmethodsbytheaccredited(DANAK, −2.70‰ forTCE)whichwereformerly characterizedbythe
International Standards Organization 17,025) laboratory ALS Holtmethod(Holtetal.,1997)attheUniversityofWaterloo
(ALS Denmark A/S, alsglobal.dk, alsglobal.com) in Denmark was completed for each sequence to obtain δ values on the
with quantification limits of 0.03 mg·L−1, 0.01 mg·L−1, Standard Mean Ocean Chloride (SMOC) scale. Cl isotopic
0.5mg·L−1and0.1mg·L−1respectively.Chlorinatedethene analysisofcDCEwasperformedatIsotopeTracerTechnologies
concentrations were analysed at ALS by Purge & Trap Gas Inc. (Waterloo) according to the method developed by
Chromatograph-MassSpectrometer(GC–MS)withadetection Shouakar-Stashetal.,2006,usingaContinuousFlow(CF)IRMS.
limitof0.02μg·L−1andstandarddeviations of10%.Ethene, Molecular biology analysis: DNA extraction for the 454
methane, ethane and acetylene were determined at the pyrotag analysis was carried out at SiREM (Guelph, ON,
TechnicalUniversityofDenmark(DTU,Kgs.Lyngby,Denmark) Canada) as follows: Sterivex™ filters were opened and the
byheadspaceGC-FlameIonisation Detector(FID)(Shimadzu filter membrane with attached biomass was removed and
GC.14Awithapackedcolumnwith80/120CarbopackB/3%SP- placedintotheBeadSolutionofaPowerMagDNAIsolationKit
A.Badinetal./JournalofContaminantHydrology192(2016)1–19 7
(MoBio,Carlsbad,CA,USA)andpulverizedusingasterilepipet clay particles. Extracted DNA and RNA was purified using
tip. Cell lyses were performed using a MiniBeadbeater-8 NucleoSpin RNA Clean-up XS kit (Macherey-Nagel, Duren,
(Biospec Products Bartlesville, OK, USA) at 50% of the Germany).RNA was converted to cDNAandDNA andcDNA
maximum setting for 30 s. DNA was purified using a PCRamplifiedusingstandardprotocolswithadetectionlimit
KingFisher™Duo (ThermoFisher Waltham, MA, USA) and of104copies·L−1.Thedetailedprotocolscanbefoundinthe
eluted in 150 μL. DNA was quantified using a NanoDrop SI.
spectrophotometer (NanoDrop Inc. Wilmington, DE) and
stored at −80 °C after extraction. 16S rRNA genes were 2.4.Calculationsforisotopicdatainterpretation
amplifiedfromDNAextractswithuniversalprimers926f(5′-
AAACTYAAAKGAATTGACGG-3′)and1392r(5′-ACGGGC 2.4.1.Cisotopebalance
GGTGTGTRC-3′)for454pyrotaganalysis.Thereverseprimer Inordertoevaluateisotopicdataandmoreparticularlyto
alsocontaineda10nucleotidebarcodeand454FLXTitanium determinewhetherdegradationreleasedasignificantamount
Lib-L ‘B’ adapter. PCR was performed under the following of compounds which were not detected, such as ethene or
conditions:94°Cfor3min;25cyclesof94°Cfor30s,52°Cfor ethane during complete sequential reductive dechlorination,
30 s, and 72 °C for 1 min; and finally 72 °C for 10 min. theCisotopebalancewasdeterminedforeachsamplingpoint
Amplicons were purified with GeneJET PCR Purification Kit accordingto:
(Life Technologies, Burlington, ON, Canada) and sequenced
with Roche GS-FLX Titanium series kits and system (Roche, δ13C ¼½PCE(cid:2)(cid:3)δ13CPCEþ½TCE(cid:2)(cid:3)δ13CTCEþ½cDCE(cid:2)(cid:3)δ13CcDCEþ½VC(cid:2)(cid:3)δ13CVC
Branford,CT,USA)atGenomeQuebecandMcGillUniversity sum ½PCE(cid:2)þ½TCE(cid:2)þ½cDCE(cid:2)þ½VC(cid:2):
InnovationCentre(Montreal,PQ,Canada).Finally,analysisof
where[PCE],[TCE],[cDCE]and[VC]arethemolarconcentra-
the reads was performed using QIIME v.1.8 (Caporaso et al.,
tions of PCE, TCE, cDCE and VC, respectively, and δ13C ,
2010).Initially,rawreadsweredemultiplexedandfilteredby PCE
quality (NQ20) and length (N250 nt) using the pick_otus.py δ13CTCE, δ13CcDCE, and δ13CVC their corresponding C isotopic
composition.Theuncertaintywasdeterminedbyerrorprop-
scriptwithusearch61(Edgaretal.,2011)optionandrepresen-
agation(Reddyetal.,2002).
tativesequenceswereselectedusingthepick_rep_set.py.The
sequences were aligned to the Greengenes Core reference
2.4.2.Extentofdegradation
alignmentbyPyNAST(Caporasoetal.,2010).Putativechimeric
In order to estimate the extent of degradation in certain
sequences were removed using ChimeraSlayer (Haas et al.,
partsoftheplume,thefollowingcoefficientwascalculated:
2011). Taxonomic assignment of the operational taxonomic
units(OTU)wasperformedbyassign_taxonomy.pyscriptwith !
Δδ13C
the Ribosomal Database Project (RDP) method (Wang et al., D¼1−exp
ε
2007;Martinsetal.,2013).Asequencewasdefinedasbelonging
toaparticularOTUwhenthesimilaritylevelwasatleast97%.
In orderto evaluatewhichandto whichextentmicroor- whereΔδ13Ccorrespondstothedifferencebetweentheinitial
ganisms relevant to redox processes potentially occurring in andfinalCisotopiccompositionoftheconsideredchlorinated
the subsurface were present, OTU reads per sample were ethene.Suchcalculationwasperformedonlyforcompoundsin
transformed to cells·L−1 based on the total bacteria count samplingpointswherenoprecursorwaspresent(e.g.forcDCE
(i.e. total DNA extracted from the samples). It was assumed whennoPCEorTCEwasdetected)toensurethattheisotopic
thatalltheextractedDNAwasprokaryotic,whichleadstoa compositionwasmerelyaffectedbythecompounddegrada-
slightoverestimation,andthattheaveragemicrobialgenome tionandnotbyitsproduction.InthecaseofPCE,suchcautionis
contains4⋅10−6ngDNA⋅cell−1(Paul,1996).Sincechlorinated unnecessary as it can only be degraded. Minimum and
ethenedegradationaswellasredoxprocessesoccurringinthe maximum enrichment factors reported in the literature as
subsurface were of interest, detected microorganisms were well as a field determined enrichment factor were used as
groupedintaxonomiccategoriessuchasgeneraofwhichsome summarisedinTableS3,SI.
strainsareknowntoperformcompletereductivedechlorina-
tion, partial reductive dechlorination of PCE and/or TCE, 3.Resultsanddiscussion
oxidationofcDCEandorVCunderaerobicconditions,bacteria
reportedtobefoundiniron-andsulfate-reducingconditionsas Inthissection,resultsfromchemical,isotopicandmolec-
well as during pyrite oxidation. Bacteria counts were then ularbiologyanalysesaregivenanddiscussedwiththeaimto
summed in each group and divided by the sum of bacteria understandtheeffectofthermalremediationonthechlorinat-
countsofall targetedgroupsin each sampleto evaluatethe edetheneplume.Onlythedatafrom27screensof13wells
proportionofeachbacteriagroupwithineachsamplerelative whicharelocatedalongtheplumecentrelinearediscussed.
tothegroupsofinterest.ThesedataaresummarisedinTableS
1andTableS2oftheSupportingInformation(SI). 3.1.Redoxconditions
Dehalococcoides DNA and rRNA, bvcA and vcrA functional
genes (DNA) and genes transcripts (mRNA) analysis was Concentrationsofredoxsensitivespeciesmeasuredin2006
performed on co-extracted DNA and RNA using a combined (Hunkeler et al., 2011) and 2014 are shown in Fig. 3A. In
phenol-chloroform and mechanical beadbeating method general,O and/orNO−concentrationslargerthan1mg·L−1
2 3
(Bælumetal.,2013).Inbrief:priortocelllysis,thesamples aredetectedintheshallowtoppartoftheaquifer(downto10–
were mixed with 0.5 mL liquid G2 DNA/RNA enhancer 15mdepth)whereasverylowconcentrations(b0.1mg·L−1)
(Ampliqon, Odense, Denmark) to cover binding sites of the arefoundindeeperparts.Here,thepresenceofFe(II)and/or
8 A.Badinetal./JournalofContaminantHydrology192(2016)1–19
methane indicates the occurrence of reducing conditions. 2014intheupperpartoftheaquiferfromthesourcezoneto
Below15mdepth,ConcentrationsofSO2−rangefrom23to 750mdowngradientcouldalsobeattributedtotheoxidation
4
59mg·L−1andareabove40mg·L−1,i.e.inthehigherpartof of organic matter released from remediation, which conse-
theconcentrationrange.Basedonthesulfurisotopiccompo- quentlyledtomorereducedconditionsandtemporallackof
sitionofSO2−,Hunkeleretal.suggestedthatthedisappearance pyrite oxidation due to lack of oxygen. Indeed, higher Fe(II)
4
ofO /NO−andtheincreaseofFe(II)/SO2−concentrationswith concentrations(N1mg⋅L−1),especiallyintheshallowpartand
2 3 4
depth could be associated with pyrite oxidation processes in B28, as well as slightly lower SO2− concentrations are
4
(Hunkeleretal.,2011),whichissupportedbythepresenceof observedinthispart,indicatingmorereducedconditions.Such
pyriteinsandyaquifersinJutland(Postmaetal.,1991). aDOCimpactleadingtomorereducedconditionsisparticu-
Unlike in 2006, in 2014, mixed redox conditions are larly reflected by B16-1 (Fig. 3). High DOC and SO2−
4
generallyobservedatthelocalscale.Indeed,in2014,markers concentrations (probably due to pyrite oxidation due to
fordifferentredoxconditionsaresimultaneouslyfoundwithin steam injection saturated with air) were indeed observed
somescreens(e.g.presenceofO andFe(II)inthesamewell). right after remediation in this sampling point, where less
2
Thismightbeexplainedbyacertainsubsurfaceheterogeneity concentrated NO− and O concentrations (due to DOC
3 2
thatresultsinvariousredoxzonesoverlayingeachotherwithin oxidation)wereconcomitantlyobserved.TheDOCconcentra-
screenintervalsrangingfrom1to4m.Suchaspatialchangein tion then gradually decreased (B16-1 in Fig. 3). Another
redoxconditionsmayresultfromthetemporalchangeinredox strikingchange between 2006and 2014 is the lack of Fe(II)
conditions affecting geologically different layers at different detected in 2014 between 1000 and 1500 m downgradient
speed.Thistemporalchangeissupportedbytheredoxspecies from the source (wells B34, B47 and B58). In these wells,
evolutionbetween2004and2014asdepictedinFig.3andis concentrations of methane increased whereas SO2− concen-
4
especiallyapparentinthefirst750mafterthesourcewhere trations showed up to 20% decrease in 5 out of 8 sampling
both Fe(II) and O are present. More particularly, the NO− points, which suggests the occurrence of SO2− reduction
2 3 4
concentrationsuddenlydrops100mdowngradientfromthe followed by the precipitation of metastable iron sulphide or
source where Fe(II) concentrations increase right after the Fe(II) binding to other minerals. This suggests that the DOC
remediationevent(Fig.3),whichindicatesashifttowardmore release affected this part of the aquifer too as the flowline
reducedconditionsin2014comparedto2006inthefirst750m descendswheretheclaylayerobservedunderthesourcearea
of the plume (Fig. 3). A striking change directly following disappears. The occurrence of sulfate-reducing conditions in
thermal remediation is the appearance of DOC in and this part of the plume would be additionally favorable for
downgradient of the source area that decreases over time chlorinated ethene microbial reductive dechlorination, espe-
(Fig.3AandB16-1inFig.3),presumablyduetoitstransport ciallyofcDCEandVC(Vogeletal.,1987;Chapelle,1996).Atthe
and consumption. Although aquifers in this part of Jutland fringeoftheplume(i.e.N1800mfromthesource),concentra-
generallycontainlowlevelsoforganicmatter,highconcentra- tions of Fe(II) increased in 2014 compared to 2006
tionsofDOCaremeasuredimmediatelydowngradientofthe (N1mg·L−1)andmethaneisdetectedinB64,whichindicates
sourceareaafterthethermalremediationtreatment,reaching morestronglyreducedconditionsin2014thanin2006.
valuesof6.1mg·L−1(F3)and3.1mg·L−1(B16,located100m
downgradient) in 2014. Such a release of sediment-bound 3.2.Evolutionofchlorinatedetheneconcentrations
organicmatterduetothermaltreatmenthasbeenpreviously
reportedneartreatedsourceareas(NewmarkandAines,1995, The distribution of individual chlorinated ethenes was
Friis et al., 2005). Friis et al. confirmed via experiments evaluated in a vertical section along the plume centerline
performed with field material that up to 8% of sediment- before (2006, Hunkeler et al., 2011) and eight years after
bound organic carbon could be released in temperature (2014)performingthethermalsourceremediation(Fig.4).In
conditionsusuallyachievedwiththermaltreatments(Friiset bothcases,inthefirst350mfromthesource,thecontaminant
al.,2005).Releasingorganicmattercanbeexpectedtoaffect plumeisconfinedintheupperpartoftheaquiferduetothe
redox conditions. This is indeed observed immediately presenceofaclayeylayerat20mdepth.Furtherdowngradient,
downgradient of the former source area (wells F2 and F3) the plume widens and dives toward deeper zones while
where O (b1 mg·L−1) and NO− (up to 12.1 mg·L−1) exhibitingmore reduced conditions, asdiscussed above.The
2 3
concentrationsaremuchlowerin2014thanthosemeasured remediationprimarilyresultedinadramaticdropinchlorinat-
in 2006 in nearby wells (B5, in the source and B11, 100 m edetheneconcentrationimmediatelydowngradient.
downgradient),whereconcentrationsofupto5.2mg·L−1and In2014,astrongdecreaseinPCE,TCEandcDCEconcentra-
28.0mg·L−1forO andNO−weremeasured,respectively.The tions of more than 85% is observed in wells immediately
2 3
lowervaluesmeasuredin2014couldberelatedtoO andNO− downgradient(i.e.wellsF2,F3andF4,Fig.2)between2006
2 3
consumptionduringoxidationoforganicmatter.Additionally, and 2014. Much lower concentrations are also generally
theDOCcontentinB16,B20,B22,andB23,whicharelocated measuredinwellssituatedwithin750mfromthesourcein
upto400mdowngradientofthesource,washigherin2010 2014comparedto2006,withtheexceptionofhighervalues
(Westergaardetal.,2011)thanin2014,supportingthegradual obtained for TCE at the bottom of B23 (e.g. 1.1μmol·L−1 in
consumption of DOC released by the thermal treatment. 2014 instead of 0.43μmol·L−1 in 2006). Further
Concomitantly,depletioninO wasgenerallyobservedinB16 downgradient, lower concentrations are generally found for
2
B20 B22 B23 and B28 in 2010 (Westergaard et al., 2011) PCEandTCE,whicheventuallydisappear1050m(PCE)and
compared to 2014, which coincides with DOC consumption 1450 m (TCE) downgradient. Similarly, cDCE concentrations
leadingtomorereducedconditions.Thesignificantchangein decreased 1050 m downgradient of the source in 2014
Fe(II) and SO2− concentrations observed between 2006 and comparedto1050mdowngradientofthesourcein2006.The
4
A
.B
a
d
in
e
t
a
l.
/
Jo
u
rn
a
l
o
f
C
o
n
ta
m
in
a
n
t
H
y
d
ro
lo
g
y
1
9
2
(2
0
1
6
)
1
–
1
9
Fig.4.Chlorinatedetheneconcentrationinthesubsurfacein2006(Hunkeleretal.,2011)and2014(thisstudy).
9