Ecology

8 WATER AS A RESOURCE | 1.0

An accessible version of the article is available athttp://waterresources.fona.de/reports/bmbf/annual/2010/nb/English/30/1_-ecology.html

9

Water is the source and foundation of all life – a precious commodity that is increasingly endangeredin many regions of the planet as a result of man’s influence. This makes it all the more vital to ensurethe sustainable handling and protection of water resources. The BMBF is working to achieve this goalthrough numerous research projects covering areas such as the restoration of domestic river basins,the biological clean-up of soil and groundwater, and flood management. The aim of integrated waterresource management is to co-ordinate the social, ecological and economic aspects of this work.

ECOLOGY

10 WATER AS A RESOURCE | 1.1.0

Underground application – efficient in-situ ground water remediation methods

Toxic substances that escape from under disused,unsealed landfill sites, extensive contamination ofthe ground and soil from long-since closed industrialsites, widespread contamination from fuel and theaftermath of weapons: these are just some exam-ples of hazardous abandoned waste in the soil thatoccur in great numbers and pose a threat to thequality of groundwater. Often no one knows pre-cisely where this abandoned waste is locatedbecause toxic substances have been disposed ofincorrectly or there are no longer any witnesses.Despite these difficulties, scientists are increasinglysucceeding in tracking down abandoned waste,analysing it and using new methods – directlyunderground (“in situ”) – to render it harmlessbefore it reaches the groundwater. This is essential,as around 70 percent of Germany’s drinking water isobtained from groundwater.

Germany is one of the world’s most densely populatedcountries. Combined with the unchecked dumping of haz-ardous substances in the past and the intensive use ofgroundwater and soil resources, this all amounts to a largeamount of abandoned waste and many suspected areas ofcontamination. With around 296,500 suspected aban-doned waste sites recorded in Germany in 2009, there is aconsiderable need to treat this waste. Abandoned wastecan be found underground at many disused and stillactive industrial sites. Its detection and efficient remedia-tion presents a real challenge.

The Federal Ministry of Education and Research (BMBF)tackled this issue with new vigour as part of its “Researchfor sustainable development” programme, supportingselected research and development projects to solve theproblem of abandoned waste. One of the aims of this is todevelop pilot, transferrable solution examples.

Cleaning directly in the groundwater

Treating the groundwater and soil “in situ”, i.e. makingthe hazardous substances harmless while still under-ground, is a promising approach. Be it the addition ofnutrients, gas injection (project 1.1.08) or the introductionof micro-organisms – the range of in-situ proceduresbeing (further) developed and tested by researchers andengineers in Germany is technically challenging anddiverse. One advantage is the generally lower costsinvolved in on-site treatment. Unlike conventional meth-ods, the contaminated groundwater and soil material

does not need to be raised for treatment in plants aboveground.

Groundwater remediation: a priority task

Germany places great value in developing efficient reme-diation procedures. There are numerous projects under-way seeking to optimise these forms of technology interms of practicality and transferability. Here are someexamples of joint projects funded by the BMBF: ● Under the KORA principle (retention and degradation

processes to reduce contaminations in groundwaterand soil), tests were undertaken to determine how andwhether remediation of abandoned waste could beimproved through natural cleaning processes, (pro-jects 1.1.01 and 1.1.02).

● The BMBF founded the RUBIN research programme(use of permeable treatment walls for site remedia-tion) in May 2000 with the specific aim of developingin-situ treatment walls (projects 1.1.03, 1.1.05 and1.1.06).

● SAFIRA (remediation research in regionally contami-nated aquifers). Also the name of the Bitterfeld-basedmajor research facility conducting the work, SAFIRAbegan researching new technologies and methods in1999 for the in-situ remediation of groundwater con-taminated with complex hazardous compounds (pro-ject 1.1.07).

● The BMBF and the state of Baden-Württemberg alsoprovided funding to set up VEGAS (Versuchseinrich-tung zur Grundwasser- und Altlastensanierung/research facility for the remediation of groundwaterand abandoned waste), the purpose of which is todevelop innovative remediation methods that enableexperiments to be performed under near-natural con-ditions (projects 1.1.09, 1.1.10, 1.1.11).

ECOLOGY | GROUNDWATER 11

Experience has shown that conventional remedia-tion procedures are often hit by technical or finan-cial limitations, meaning that impurities cannot befully removed from soil and groundwater. Naturalprocesses for reducing hazardous substances couldbe of great support here. Research was undertakenunder the BMBF KORA principle (retention anddegradation processes to reduce contaminations ingroundwater and soil) between 2002 and 2008 todetermine how and whether these natural process-es could be used in the remediation of abandonedwaste.

In the past, more and more sites were being uncoveredthat were polluted with substances hazardous both tohealth and the environment as a result of previous pro-duction and storage activities. Before now, such aban-doned waste was sanitised by digging up and removingthe polluted soil or pumping out then treating the con-taminated groundwater. Alternatively, safeguardingmeasures can prevent further spread of harmful sub-stances or treatment (decontamination) with chemical orbiological agents can be carried out. However, these con-ventional remediation methods are often subject to tech-nical and economic limitations, e.g. due to deepaquifers◄, heterogeneous underground conditions, built-up areas, the type of toxic substances present and theiroften uneven distribution.

Building on natural processes

It has been determined at a number of sites that naturallyoccurring processes such as biological degradation, chem-ical precipitation◄, decomposition, sorption◄, dilutionand volatilisation can render harmless or withhold haz-ardous substances in groundwater and soil under theright conditions. As such, contaminant plumes◄ expandonly to a limited degree in the groundwater and diminishonce the source of the toxin runs dry. Great emphasis hasbeen placed on researching, evaluating and finally utilis-ing these processes in a targeted manner for a sustainableway to handle our resources.

In-situ remediation of abandoned waste – making good use of biological cleaning processes

Systematic investigations at24 reference sites

The natural retention and degradation of hazardous sub-stances depends on the type of contamination and theconditions in the soil and groundwater. A few key ques-tions must be answered to be able to assess whether natu-ral processes for reducing contamination (referred to as“natural attenuation” (NA) processes) can be used or stim-ulated at a given location:● Are the hazardous substances being effectively broken

down or withheld by natural processes under the pre-vailing conditions underground?

● Are any (undesirable) intermediary products accumu-lating through biological degradation?

● Could degradation be stimulated through the target-ed modification of on-site conditions?

The diverse range of issues related to NA processes waswhat prompted the BMBF to provide approximatelyEUR 20 million of support to the “retention and degra-dation processes to reduce contaminations in ground-water and soil” (KORA) principle between 2002 and2008. In total, 74 projects were carried out at 24 typicallycontaminated sites representing a wide range of compa-rable cases in Germany, testing whether and under what

WATER AS A RESOURCE | 1.1.0112

Overview of the thematic networks (TN) and working aids from the KORA funding principle

TN

1

2

3

4

5

6

7

8

Name of thematic network(industry-specific pollutants)

Refineries, fuel tanks, fuels/mineral oil(TPH, BTEX, MTBE)

Gas works, coking plants, coal tar process-ing (PAH, coal tar, heterocyclics)

Chemical industry(VCHC, BTEX)

Landfills, abandoned waste disposal sites(landfill pollutants)

Former munitions works(Compounds typically found in explosives)

Mining, sediments(Trace metals, acidity/sulphate, pesticide)

Modelling and prognosis

Derivation of MNA concepts, legal and eco-nomic issues, acceptance by the officialbodies and the public

Investigatedsites

5

4

6

4 (+ 2)+

3

2 (+ 1)+

–

–

Short name forworking aid

TN 1 guidelines(ISBN 978-3-89746-093-9)

TN 2 guidelines(ISBN 978-3-934253-50-6)

TN 3 guidelines(ISBN 978-3-00-026094-0)

TN 4 guidelines(ISSN 1611-5627, 04/2008)

TN 5 guidelines(ISBN 978-3-00-025181-8)

TN 6 guidelines(ISBN 978-3-89746-098-X)

TN 7 synopsis(ISSN 1611-5627, 05/2008)

Handling recommendationswith method collection(ISBN 978-3-89746-092-0)

+ Additional sites were investigated as part of associated projects

ECOLOGY | GROUNDWATER | IN-SITU REMEDIATION OF ABANDONED WASTE 13

conditions natural degradation and retention processescould be put to effective use – in particular:● Assessing the risks associated with contaminated

groundwater and soil ● Calculating and performing hazard control measures

(specific remediation)● Calculating and performing follow-up measures.

In order to assess whether natural toxin reductionprocesses can be used and to monitor their effectiveness,regular soil and groundwater samples must be taken(monitored natural attenuation, MNA). The use of NAprocesses is therefore not a “do-nothing option”. Quite theopposite: only through clear and thorough data collection(monitoring) and assessment (forecasting) can the antici-pated processes be implemented effectively and naturalmethods of reducing contamination be a viable alterna-tive or supplement to conventional remediation methods.

Toxin-related research

The aim of the research performed under KORA was toform the basis for considering natural processes for reduc-ing contamination and to derive plausible opportunitiesand parameters for implementation from an ecological,economic and administrative perspective. This not onlyrequired the development of appropriate methods toprove the efficacy of the NA processes, but also the valida-tion of tools for their assessment. Universities, engineer-ing bodies and authorities worked together to developMNA concepts for the 24 investigated sites with tailoredsolutions for each one, and in many cases these processeswere implemented too. In doing so, they created refer-ence sites to be used as examples of how NA processes canbe recorded, assessed and considered in a graduated pro-cedure. The experts further developed innovative in-situremediation procedures based on stimulating naturalprocesses for reducing contamination. They also investi-gated various measures that are to encourage acceptanceof NA processes for abandoned waste remediationthrough (risk) communication.

Industry guidelines and handling recommendations for practical application

The results of this funding principle were documented inthe KORA handling recommendations (with integratedmethod collection), in six sets of industry guidelines andin the KORA synopsis “Systemanalyse, Modellierung undPrognose der Wirkungen natürlicher Schadstoff min-derungsprozesse” (systems analysis, modelling and fore-casting effects of natural processes for reducing contami-nation) (see table). There is thus a variety of supplemen-tary material available to assist authority representatives,engineering planners and those in charge of remediation.They can be used to check the potential ways in which NAprocesses and MNA concepts can be applied to treat aban-doned waste. The working aids provide recommendationsand assistance in using monitored or stimulated naturalcontamination reduction processes to treat abandonedwaste in Germany and cover NA processes for groundwa-ter that is already contaminated. Relevant existing(inter)national working aids, concepts and guidelineswere considered during the compilation of the handlingrecommendations and industry guidelines. The workingaids can be referenced at the website for this principle(www.natural-attenuation.de/bestellung) or downloadedin PDF format.

Project website ►www.natural-attenuation.de

DECHEMA e. V.Dr. Jochen Michels, Christopher FreyTheodor-Heuss-Allee 2560486 Frankfurt, GermanyTel.: +49 (0) 69/75 64-157, -440Fax: +49 (0) 69/75 64-117E-mail: michels@dechema.de, frey@dechema.deFunding reference: 02WN0335

University of StuttgartInstitut für Wasserbau (water engineering institute), VEGASDr.-Ing. Hans-Peter KoschitzkyPfaffenwaldring 6170650 Stuttgart, GermanyTel.: +49 (0)7 11/6 85-647 17Fax: +49 (0)7 11/6 85-670 20E-mail: koschitzky@iws.uni-stuttgart.deFunding reference: 02WN0336

The former premises of a dry cleaner’s for leatherwork clothing in the Harburg region of Lower Saxony was contaminated with perchloroethyleneover a period of several decades. As the groundwa-ter is extremely low down and the contamination isextremely difficult to access, conventional surveyingand remediation procedures are not suitable for thisabandoned waste. One of the projects within theKORA principle (retention and degradation process-es to reduce contaminations in groundwater andsoil) therefore aimed to use suitable monitoring andforecasting procedures to estimate the degree towhich natural degradation and retention processesare sufficient in ruling out any risk to a neighbour-ing water conservation area.

Perchloroethylene (PCE)◄was and still is used in tradeand industry to remove paint, as a solvent and to degreasematerials – as it was in a former specialist dry cleaner’s forleather work clothing in Rosengarten-Ehestorf in Har-burg. After use, the chemical along with the improperlycleaned wastewater filtered underground over thedecades below the company’s premises measuringapprox. 3000 sqm.

An inaccessible aquifer

The surface of the affected aquifer◄ is 30 to 40 metresunderground – unusually low down. With one stretch upto 230 metres deep, it is also extremely powerful. Thesefactors – combined with the heavily built-up area – wouldmake it extremely difficult and expensive to carry outremediation using conventional methods (pump-and-treat procedure◄). The Harburg authorities thereforelooked at alternatives and opted to take part in the BMBFKORA principle. Run from 2003 to 2006, the project enti-tled “Field-scale quantification of the potential of NAin deep large-scale aquifers – example: VOC contami-nation from a dry cleaners in Rosengarten-Ehestorf”involved project planners determining the extent of natu-ral attenuation (NA) in the soil since the introduction ofcontamination and estimating its subsequent course. Theproject partners included the Harburg authorities, theInstitut für Gewässerschutz und Umge-bungsüberwachung (institute for the protection of natu-ral waters and environmental monitoring) in Kiel, theState Authority for Mining, Energy and Geology inHanover and the Tübinger Grundwasser-Forschungsinsti-tut (Tübingen groundwater research institute).

Keeping tabs on pollutants – recording and assessing natural degradation and retention processes

In order to quantify the nature of the contaminantplume◄ and the natural degradation and retentionprocess underground, the PCE introduction and its distri-bution had to be recorded first. Existing measuring pointswere upgraded and new ones set up in order to locate theprecise location of the contamination source and theensuing discharge.

Innovative systems for taking soil gas andgroundwater samples

The site conditions made it extremely expensive to samplethe groundwater using conventional wells, which is whyonly five were installed. The experts secured additionalsamples, but above all site surveys, using what is known asthe direct-push procedure◄. This innovative drillingmethod is a faster, more flexible and considerably morecost-effective alternative to the conventional procedure.The direct-push procedure was further developed duringthe project so that the required underground depths

WATER AS A RESOURCE | 1.1.0214

A direct-push sounding device at work

ECOLOGY | GROUNDWATER | KEEPING TABS ON POLLUTANTS 15

could also be reached. The scientists also expanded thewells to create multilevel measuring points to enable per-manent sampling of soil gas and groundwater at variousdepths. This innovative procedure enabled a preciseanalysis of the vertical distribution of the contamination.Conversely, it proved difficult to determine the horizontalspread of the contamination by attempting to pump at thewells. The impenetrable soil did not permit any conclu-sions to be drawn regarding the overall breadth of thecontaminant plume.

Steady introduction of PCE

Key results of the survey: the scientists measured the high-est concentrations of PCE at a soil depth of five to tenmetres below the surface. They determined that the soilcontinuously feeds PCE into the groundwater. However,the quantity continually drops over time until the dis-charge finally comes to a halt. Because the contaminant isprimarily accessing the groundwater with the drainagewater, the current and future amount of contaminantdelivered and thus the lifetime of the pollution source issignificantly affected by the amount of precipitation. Onaverage, the total delivery over the non-sealed surfacearea of the company premises amounts to nine grams aday, of which over seven grams can get into the ground-water. The rest evaporates into the atmosphere, where it isbroken down into harmless substances.

Pollution transport models

The project team made use of what are known as transportmodels to obtain a quantitative estimate of the futureintroduction of pollution into the groundwater. Withtheir help the team ran through various versions of pollu-tion spread since the assumed introduction of the contam-ination some 40 years ago. According to the most likelyscenarios, which were determined by a comparison withthe results measured on site, the PCE contaminationplume in the aquifer has been stable and virtuallyunchanged for around 20 to 25 years. Calculations sug-gest that it is between 400 and 500 metres long. Theresults from the measuring points, which are extremelyheterogeneous in part, lead to the conclusion that theremay not be just one contaminant plume but it could havesplit into two or more branches.

Pollution source will be “clean” in around40 years

Scientists expect that the source of the pollution – i.e. thePCE still in the soil above the groundwater – will havecompletely disappeared in around 40 years’ time. The PCEplume in the groundwater will completely break downover time due to the existing natural degradation processes.

These statements are indispensable for a legal assessmentof using natural attenuation. They enable the authoritiesto dispense with the extremely expensive and frequentlyonly partly efficient measures of active remediation with-out any risk. Overall, the procedure developed at theRosengarten site for surveying and assessing the contami-nation situation has shown that decisions on whether andto what extent natural pollution reduction is taking placecan also be made with sufficient certainty for deepaquifers.

Landkreis Harburg (Harburg County)(Head of soil/air/water department)Gunnar PeterPostfach 144021414 Winsen/Luhe, GermanyTel.: +49 (0) 41 71/6 93-4 02Fax: +49 (0) 41 71/6 93-1 75E-mail: g.peter@lkharburg.deInternet: www.landkreis-harburg.deFunding reference: 02WN0437

Groundwater is an especially important and also vul-nerable resource for our supply of drinking water.Often only the slightest amount of contamination isenough to render thousands of litres of waterundrinkable. In order to dispense with the need topump contaminated groundwater to the surface forcleaning, researchers in Germany and elsewherehave been intensively focussing on “in-situ” proce-dures for some years – technologies that can beapplied directly in the aquifer◄. To drive forwardnew and further development of permeable treat-ment walls in Germany, the BMBF has been support-ing the RUBIN research programme since 2000/2001.

Cleaning polluted groundwater is a complex task. Mostremediation procedures involve an active pump-and-treatmethod, active meaning that the water is pumped to thesurface and treated there in downstream plants. This is anextremely expensive process, and as the water is then fre-quently channelled into the sewer system this results inyet more cost.

Remediation within the aquifer

Recent years have seen intensive work on the develop-ment and testing of passive forms of in-situ remediationtechnology using permeable treatment walls. These areinstalled in the groundwater flow from the source of con-tamination to eliminate the pollutants in the aquifer itself.A unique new procedure is thus saving costs as there is nolonger a need to pump or discharge the groundwater.Two types of technology have been primarily introducedto practical applications: the all-over permeable treatment wall and the funnel-and-gate system, wherebythe groundwater is fed via guiding walls (funnel) to a per-meable reaction zone (gate).

An all-over permeable wall involves a ditch reachingdown to the layers where the groundwater flows beingfilled with reactive material. Elemental iron or activatedcarbon◄ is primarily used for this purpose. The contami-nated groundwater flows through the permeable barrier,and the reactive material operates like a filter to removeor break down the pollutants.

The technology involved in permeable treatment wallswas primarily driven forward in North America. Germanyhad little practical experience with treatment walls.

Permeable treatment walls – underground structures for successful remediation

RUBIN – the project

The BMBF-funded research programme entitled Use ofpermeable treatment walls for site remediation(RUBIN) saw industrial companies and research facilitiesworking closely together in an interdisciplinary fashion atvarious locations across Germany. The main task was toconduct a detailed, far-reaching and co-ordinated investi-gation into the potential and limitations as well as theenvironmental impact and economic viability of perme-able treatment walls at a large number of locations in Ger-many. One key goal of the research programme was todevelop generalised criteria for the application of perme-able treatment walls, such as● Layout, design, construction and operation● Performance and durability (degradation of pollution,

as well as for mixed contamination/reactor systemsand reactive reactor filling materials)

● Framework conditions and limitations for use● Economy (procedural costs) and ecology (environmen-

tal impact).

Results for practical application

Projects were conducted at six sites to investigate durabili-ty, degradation levels, changes in material, material con-version processes, treatment wall systems, reactor fillingmaterials and monitoring.

WATER AS A RESOURCE | 1.1.0316

Schematic representation of a fully permeable treatment wall (Mull u.Partner Ing. Ges. mbH, Hanover)

ECOLOGY | GROUNDWATER | PERMEABLE TREATMENT WALLS 17

Another focus of the work saw employees from the Univer-sity of Kiel set up and test quality management rules aspart of the RUBIN project, with the aim of enabling reli-able planning, execution and monitoring of standardisedtreatment walls in future. Experts from the University ofTübingen also compared the economic efficiency of themethods with conventional remediation procedures.These studies put those using this new technology in aposition to carry out an in-depth cost comparison infuture.

The quality management rules and economic efficiencyobservations are, along with the overall results of the pro-gramme, integral components of the handbook on the useof permeable treatment walls for site remediation, whichpresents the key results and findings of the project tointerested parties. Created under the guidance of OstfaliaUniversity, the co-ordination site of the RUBIN pro-gramme, this handbook is to serve primarily as a generalorientation aid for users – e.g. authorities, those in chargeof remediation, planners and environmental technologyproviders.

Scientists conducted further investigations to examine theresults and findings at technical pilot facilities, with exten-sive planning, construction and trials involved. As a result,the technology was able to be rendered marketable andintroduced to practical groundwater remediation withinGermany. The BMBF funding therefore made it possible todevelop a new, more cost-effective and more environmen-tally friendly method of removing pollution directly fromthe groundwater – one of the country’s most significantsources of drinking water – ready to put into practice andthus enhance Germany’s status within environmentaltechnology.

Project website ►www.rubin-online.de

Ostfalia – University of Applied SciencesFaculty of Civil and Environmental EngineeringProf. Dipl.-Ing. Harald BurmeierProf. Dr. Volker BirkeHerbert-Meyer-Straße 729556 Suderburg, Germany Tel.: +49 (0) 58 26/9 88-6 11 40, -6 15 60E-mail: h.burmeier@ostfalia.dev.birke@ostfalia.deFunding reference: FZK 0271241 and 02WR0828

Permeable treatment wall

Contaminant plume

Clean groundwater

Contamination

Soil/non-saturated zone

Impervious layer

Direction of groundwater flow

Aquifer

Ground water surface

Schematic representation of a fully permeable treatment wall (source: Mull u. Partner Ing. Ges. mbH, Hanover)

Treatment (reactive) walls are an extremely promis-ing approach to sanitising or safeguarding contami-nated aquifers◄. Germany’s first all-over permeablereaction wall was installed as a pilot scheme atRheine in North Rhine-Westphalia in 1998, in a qua-ternary aquifer polluted with chlorinated hydrocar-bons (DBU project). In a subsequent, further-reach-ing research and development project funded bythe BMBF, partners from science and economicsworked together to investigate the durability of thetreatment wall and the degree to which iron can beused as a reactive material at that particular site.

In June 1998, Hanover-based Mull und Partner Inge-nieurgesellschaft mbH used funding from the DeutscheBundesstiftung Umwelt (DBU, German environmentalfoundation) to erect a groundwater treatment wall as afield-scale project in order to remove volatile halogenatedhydrocarbons (VHHC)◄ such as tetrachloroethylene(PCE)◄ or trichloroethylene (TCE)◄ from contaminatedgroundwater. This involved using a new porous zerovalent iron (Fe0), “iron sponge”, as the reactive material.

Pollutants from a laundry

The reaction wall was installed approximately 700 metresdownstream of a massive source of underground PCE con-tamination. The cause of this pollution is a laundry thatonce operated on the premises. The treatment wall is astructure enabling all-over permeability of contaminatedgroundwater and measures around 6 metres deep,22.5 metres long and 88 centimetres thick. It is filled withtwo reactive materials up to a height of around 3.5 metres,i.e. above the maximum anticipated level of the ground-water. The materials used are an iron sponge supplied byMITTAL Steel Hamburg GmbH (formerly ISPAT Hamburg-er Stahlwerke GmbH) and a 70% pea gravel and 30% greycast iron granulate blend from Gotthart Meier AG in Rhe-infelden. This twin-layered effect allows the performanceof the materials to be compared. This procedure has seenthe amount of PCE reliably drop from its initial concentra-tion of several thousand micrograms per litre by over 99%since 1998.

Reliable long-term effectiveness – in-situ treatment wall at the Rheine site

Studies on long-term effectiveness

As well as the activity mentioned, the plant was also usedfor various long-term investigations. The BMBF funded thefollowing projects: Pre-investigation, monitoring andquality management regarding Reactive Walls(Christian Albrecht University of Kiel), Evaluation of thedurability of an iron-reactive wall with the example ofthe site at Rheine (Mull und Partner Ingenieurge-sellschaft mbH) and Biological processes in a reactiveiron wall (TU Berlin). The objective of the research activi-ties was to observe the long-term behaviour of the treat-ment wall and to develop a monitoring programme toinvestigate the geochemical, hydro-geological and bio-logical processes in and around the Fe0 treatment wall.

Reliable for many years now

The reaction wall at Rheine has been providing assuredconstant remediation of the groundwater for over tenyears now, with the two different materials producing dif-ferent cleaning performances. The groundwater investi-gations have shown that the iron sponge achieved a

WATER AS A RESOURCE | 1.1.0418

GardenStreet

FilmHDPE 2 mm

Gas drain pipes

Calcite gravel

Protective layer of sand

Drainage gravel

Filled with non-contaminated dug-out material

Iron granulate filled borehole

Max. groundwater level

Quaternary base

Vertical structure of the treatment wall at the Rheine site

ECOLOGY | GROUNDWATER | IN-SITU TREATMENT WALL AT RHEINE 19

cleaning performance of around 70 to 80% in the first 6 to12 months following installation, a figure that then rose tomore than 99%. For several years now the concentrationsof VHHC measured downstream have been under10 micrograms per litre.

The concentrations of VHHC downstream of the section ofwall filled with a mixture of grey cast iron granulate andpea gravel were somewhat different: an excellent clean-ing performance of 99% was recorded at the start, but oncethe reaction wall had been in operation for around 8 to12 months only around 80% of the inflowing VHHC con-tent was still being broken down. This level of cleaningperformance has remained virtually constant up to thepresent day. The scientists were able to determine fromcore drilling that a partial separation of gravel and ironwas the cause of the reduced degradation in this section ofthe treatment wall.

Both flow modelling and pump-and-trace attempts clearlydemonstrated that the permeability of the treatment wallis assured and nothing is flowing over or around it. Theexperts were also able to substantiate that hydraulicchanges had taken place during the time of operationthrough precipitates◄ or gas formation.

Biological aspects

The scientists at TU Berlin also demonstrated during theproject the appearance of bacteria in the two iron materi-als used after a few years. The description of the biologicalactivities of all relevant physiological bacteria groupsoffered fundamental insights into the microbial colonisa-

tion. As it was discovered, there are no anticipated nega-tive effects within a foreseeable timeframe on the long-term stability of the dechlorination performance of irontreatment walls as a result of the micro-organisms present.

Applying the results

The synopsis of all the research work clearly demonstratesthat the pilot and demonstration treatment wall at Rheinecan be successfully applied for long-term use. The projectresults also included Mull und Partner Ingenieurge-sellschaft mbH registering a trade name for the ironsponge. It is marketed under the name “ReSponge” andhas been registered with the patent and trademark officesin Europe (07/2005) and the USA (12/2005). The companyalso entered a contractual agreement with MITTAL SteelGmbH in Hamburg in 2003 for the marketing of ironsponge for remediation purposes.

Project website ►www.rubin-online.de

Mull und Partner Ingenieurgesellschaft mbHDr. Martin WegnerJoachimstraße 130159 Hanover, GermanyTel.: +49 (0)5 11/12 35 59-59Fax: +49 (0)5 11/12 35 59-55E-mail: wegner@mullundpartner.deFunding reference: 0281238

Christian Albrecht University of KielInstitute of GeosciencesProf. Dr. Andreas DahmkeDr. habil. Markus EbertLudwig-Meyn-Straße 1024118 Kiel, GermanyTel.: +49 (0)4 31/8 80-28 58, -46 09Fax: +49 (0)4 31/8 80-76 06E-mail: ad@gpi.uni-kiel.deme@gpi.uni-kiel.deFunding reference: 02WR0208

Technische Universität BerlinDepartment of Environmental Technology (ITU)Environmental hygiene working groupDr. Martin SteiofAmrumer Straße 3213353 Berlin, GermanyTel.: +49 (0) 30/31 42 75-32Fax: +49 (0) 30/31 42 75-75E-mail martin.stelof@tu-berlin.deFunding reference: 0271262

Reactive wall Downstream

Upstream

ReSponge®

Reducing the concentration of pollution by treatment wall permeation

The Lang tar works in Offenbach prepared andprocessed tar from 1915 to 1929. After most of thebuildings were torn down in 1930 and a period ofvarious intermittent uses, the premises are now pre-dominantly wasteland. However, the soil andgroundwater are just as polluted with tar oil andtar-oil-related substances as they were before. Whatis therefore required is a straightforward, economi-cally feasible and safe remediation procedure thatalso affects the surrounding office locations as littleas possible. Scientists working under the RUBINresearch programme used an innovative “funnel-and-gate system” with three built-in bioreactors◄to develop a procedure to meet these requirements.

The impact of 14 years of tar production at the Offenbachsite is still clearly measurable today: the contaminating taroil and tar-oil-related substances have penetrated rightthrough to the base of the quaternary, sandy-gravellyaquifer◄. Underneath this lies tertiary◄ clay (Rupelianclay), which has a blocking effect to prevent further pene-tration of the contaminants. Several measuring pointsindicate this pollution to be a 20 to 80 centimetre tar oilphase◄ at the base of the quaternary aquifer. Polycyclicaromatic hydrocarbons (PAH) are the dominant pollutantspresent; analyses indicated over 150 milligrams per kilo-gram of soil, and up to several grams in places. Theresearchers found BTEX◄ concentrations of up to 30 mil-ligrams per litre at the source of the contamination.

Low maintenance and control requirements

After a detailed inspection of the site, experts estimatedthat it would cost around EUR 18 million to perform a con-ventional clean-up by means of digging up the soil. Thatled to a search for more cost-effective alternatives. In astudy of the various options, scientists weighed upwhether to clean up just some areas or to seal off the con-taminated soil with a barrier and a surface seal. A funnel-and-gate system with bioreactor was discussed as an addi-tional option. This solution should be simple to imple-ment with low maintenance and control requirements,and would have minimal impact on the existing landusage. The estimated costs were comparatively low, com-ing in at EUR 1.5 million. However, there was a problem:such a reactor had never been built before; the feasibilityof this proposal needed to be proven first.

Funding from the BMBF smoothed the way from the initialidea through to the now obtained proof of functionality

Funnel and gate – an innovative reactor conceptsuccessfully combating contamination

and effectiveness. Working under the RUBIN researchprogramme (see project 1.1.03), a team of scientists per-formed the required experiments (in the lab and on site),produced the necessary models and implemented theconstruction and test run of the reactor on a pilot scale.Other key requirements for the success of the project werethe approval of Darmstadt council, Department for theenvironment in Frankfurt, and the equipment providedby the state of Hessen.

Planning, construction and operation

The concept implemented on a pilot scale involved a heav-ily structured reactor: it comprised a baffle plate thickenerto remove iron and other solids from the water, three in-series bioreactors and an activated carbon phase.Upstream of the baffle plate thickener and before each ofthe three bioreactors was an open water area (free waterarea) to distribute the groundwater over the entire flowcross-section of the bioreactors. Oxygen (as H2O2) andnutrients were also fed into the groundwater at severalpoints within these free water areas to stimulate the bio-logical degradation of the pollution. The reactor concepttherefore took the general development away from pas-sive, difficult-to-control systems to ones that enable inter-vention and control.

The pilot funnel-and-gate system was constructedbetween October 2006 and March 2007. The 30 metrelong guiding walls (funnel) connect to the permeablereaction zone (gate) on the east and west flanks. They weredesigned as 550 millimetre thick walls in the mixed-in-place (MIP) procedure◄ and extend at least one metre intothe Rupelian clay. The actual reaction areas between thefree water areas were filled with a gravel with a grain sizeof 2 to 8 millimetres to serve as a growth body (carrier) forthe pollution-degrading micro-organisms.

The reactor is operating during the pilot with a flow rateof 230 to 500 litres per hour, which cannot be achieved bythe natural gradient of the groundwater alone. A pump istherefore necessary to achieve this. This active interven-tion ensures a constant throughflow and keeps feedingquantities and degradation conditions as constant as pos-sible, leading to a clear reduction in operating and moni-toring requirements compared with a passive approach.The flow rate control also enables modification to thehydraulic framework conditions at any time (e.g. ground-water sampling from the area).

WATER AS A RESOURCE | 1.1.0520

ECOLOGY | GROUNDWATER | FUNNEL AND GATE 21

Oxygen and nitrate are added to the water in order tostimulate aerobic◄ and aerobic-denitrifying degradationof the pollutants. O-phosphate◄ also ensures that there issufficient phosphorus present in the water. Nitrogen –another essential nutrient – is available to the micro-organisms in the form of naturally occurring ammoniumwithin the groundwater. Hydrogen peroxide (H2O2) solu-tions, sodium nitrate and a mixture of monopotassiumphosphate (KH2PO4) and buffer solution Na2HPO4 areburied to serve as working chemicals.

The microbial colonisation of the four gate modules wascontrolled by feeding in additives and monitored over800 days across the entire bioreactor system using amicrobiological monitoring program.

Effectiveness proven

In the baffle plate thickener, the addition of H2O2 convertsthe iron, which is then deposited as sludge through sedi-mentation. A large proportion of the pollutants is alreadybroken down at this stage through the aerobic stimulationin the baffle plate thickener. The main reduction is that ofPAH and BTEX aromatics (benzene, toluene, ethyl ben-zene, xylenes), to around 70%. The other contaminantsrelating to the tar oil (tar-oil-based pollutants) such asNSO-HET (NSO heterocyclic compounds)◄ and the otheraromatic hydrocarbons◄ are reduced by around half.

The remaining aromatic hydrocarbons and PAH 2-16 arereduced to around 40% each in bioreactor 1 by the aerobic-denitrifying processes; the BTEX aromatics and NSO-HETto above 20%. The remaining pollutants are broken downin bioreactors 2 and 3.

The first time that no traces of tar-oil-based pollutantswere recorded in the bioreactor process was in September2009. The trialled funnel-and-gate system is removing thepollution from the groundwater by means of aerobic-den-itrifying degradation alone. Other potential eliminationprocesses such as retardation◄ and volatilisation are play-ing little or no role.

System expansion

The positive experiences with the test reactor are nowleading the scientists to investigate the suitability of thesystem for treating the entire downstream contamination.They are currently comparing various versions with one ortwo gates and also passive and active components.

HIM GmbH, Remediation of Contaminated SitesDivisionDipl.-Ing. Christian Weingran64584 Biebesheim, GermanyWaldstraße 11Tel.: +49 (0) 64 28/92 35 11Fax: +49 (0) 64 28/92 35 35E-mail: asg.weingran@t-online.deFunding reference: 02WR0293

CDM Consult GmbHDipl.-Ing. Jörn Müller64665 Alsbach, GermanyNeue Bergstraße 13Tel.: +49 (0) 62 57/50 43 15Fax: +49 (0) 62 57/50 43 60E-mail: joern.mueller@cdm-ag.de

I.M.E.S. GmbHDr. Hermann Schad88279 Amtzell, GermanyMartinstraße 1Tel.: +49 (0) 75 20/92 36 00Fax: +49 (0) 75 20/92 36 04E-mail: hermann.schad@imes-gmbh.net

DVGW Water Technology Center (TZW)Dr. Andreas Tiehm76139 Karlsruhe, GermanyKarlsruher Straße 84Tel.: +49 (0)7 21/9 67 81 37Fax: +49 (0)7 21/9 67 81 01E-mail: andreas.tiehm@tzw.de

Operating station

Sedimentation area

Fill area for activated carbon

Free water area 2

Dosing station

Connecting structures

Piping (culvert)

Free water area 1 Free water area 3 Free water area 4

Bioreactor B1 Bioreactor B2 Bioreactor B3

Longitudinal cross-section of the innovative funnel-and-gate system

Iron granulate reaction walls have proven theircleaning effectiveness in several field studies todate. However, the effect of modified pH values◄onthe long-term stability of the reactive material wasstill unclear. Scientists from the use of permeabletreatment walls (RUBIN) programme used the site ofa former barracks in Berlin to test a correspondingpilot facility on the groundwater there, which washighly contaminated with trichloroethylene (TCE)◄.The result was overwhelming: the iron granulatewas also highly efficient and economical in the longterm and it was possible to remove all the TCE.

There are 11,000 known abandoned waste sites in Ger-many, all representing a significant risk to the groundwa-ter. Quick remediation, e.g. through excavation or pump-ing out the polluted groundwater, has proven technicallyunfeasible or disproportionately expensive in many cases.This is especially true when the pollutants are organic sub-stances such as volatile halogenated hydrocarbons(VHHC)◄ or tar oils, and when this pollution takes theform of a separate flow phase◄under the ground. Assuch, technologies aiming to secure the contaminantplumes◄ being emitted from the sources of contamina-tion are increasingly gaining favour. One example of suchtechnology is permeable treatment walls. Development ofthis technology began in North America and it has sincebeen funded in Germany via the BMBF’s RUBIN R&D project (see project 1.1.03).

Testing long-term stability

A modular pilot facility was set up in 2001 on the site of aformer Soviet barracks in Bernau (approximately 30 kmnorth-east of Berlin) to treat the groundwater there, whichwas highly contaminated with trichloroethylene (TCE).The main aim of the project was to test the durability anddegradation performance of an iron granulate treatmentwall under quasi in-situ conditions. The procedure utilisedthe reduction potential of metallic iron coming into con-tact with water and halogenated hydrocarbons. The corro-sion reaction of the iron served to dehalogenate◄ the par-tially carcinogenic substances. However, the oxygenreduction process increased the pH value, which trig-gered a string of secondary reactions restricting the long-term stability of the reactive material.

Application of treatment walls – combating high TCE concentrations

Project aims achieved

The scientists began by erecting a funnel-shaped barrieraround the source of the contamination. This barrier bor-ders the reactor, which is placed beneath the naturalgroundwater table and thus permits passive horizontalpermeation. The reactor comprises 18 individual cylindri-cal modules measuring 2.8 metres in diameter andaround 2.2 metres high; each module can be consideredto be a standalone reactor. By coupling several modules inseries or in parallel, it is possible to control flow lengthsand thus the amount of time the water remains in thereactor according to requirements. The research projectoperated the modules in series.

The most important aims of the project were achieved infull:● Efficient elimination of the high TCE concentrations

remaining stable in the long term.● Manageable volume flows and pollutant concentra-

tion/quantity over a long flow course in the iron reac-tor.

● Access to the reactive material to enable countermea-sures to be taken against any unforeseen negativeinfluences.

WATER AS A RESOURCE | 1.1.0622

Average retention time (hrs)

TCE degradation in lab with iron granulate

ECOLOGY | GROUNDWATER | TREATMENT WALLS COMBATING TCE 23

The grey cast iron granulate used as the reaction materialachieved a cleaning performance of over 99.5%. It is both ahighly efficient and economically beneficial prospect forfield-scale projects where pollution concentrations are atleast 25 milligrams per litre, as found at the Bernau site.Follow-up cleaning of the slightly contaminated reactoroutlet with water activated carbon has proven to be highlyeffective over the three and a half years the reactor hasbeen in operation and the most economic measure forremediation available. Running over several years, theresearch project saw a total of 450 kilograms of TCEremoved from 15,000 cubic metres of water overall, andthe treated groundwater cleaned up to the TCE detectionlimit.

Ready for practical application

The results of the research project have been implement-ed in site remediation since February 2007. The “eastVCHC plume” is being cleaned using the iron granulatetreatment wall. The requirements for this were:● Upgrade/retrofit of the equipment. The iron granulate

from 10 reactor modules had to be removed, mechani-cally processed and reinstalled.

● Installation of new piping plus measuring and controltechnology as a pump system to and from the reactormodules.

● Implementation of six pumping wells plus the neces-sary pumping technology.

As the cleaning performance of the reaction wall is insuffi-cient for the cDCE degradation product◄, follow-upcleaning with water activated carbon is also required.

The contaminated groundwater is pumped at a speed of2.4 cubic metres an hour from six wells. Ten parallel mod-ules are used to clean the water contaminated withvolatile chlorinated hydrocarbons◄ (VCHC, predominant-ly TCE). The treated water then flows through the follow-up cleaning system and is then fed back into the soil. Theequipment is operated and monitored centrally.

The scientists have already modified the equipment sever-al times during operation in order to optimise perform-ance. The water now flows up through the modules in theA section. A mixture of iron and filter sand was placed inthe feed area of the B section modules and a gas drainagesystem made from perforated polyethylene tubes hasbeen installed.

In the last three and a half years, the cleaning installationat the Bernau site has removed approximately 3,200 kilo-grams of VCHC from around 55,000 cubic metres ofgroundwater underground. It has removed virtually allthe TCE from this location, part of which was not fullydechlorinated but converted to cDCE instead.

Brandenburgische Boden Gesellschaft für Grundstücksverwaltung und -verwertung mbHMartina FreygangWaldstadt Hauptallee 116/615806 Zossen, GermanyTel.: +49 (0) 33 77/3 88-157Fax: +49 (0) 33 77/3 88-180Martina.Freygang@bbg-immo.dewww.bbg-immo.de Funding reference: 0251231

IMES Gesellschaft für innovative Mess-, Erkundungs- und Sanierungstechnologien mbHDr. Hermann SchadMartinstraße 188279 Amtzell, GermanyTel.: +49 (0) 75 20/92-36 00Fax: +49 (0) 75 20/92-36 04info@imes-gmbh.net www.imes-gmbh.net

ISTEV GmbHPeter HeinBismarckstraße 114109 Berlin, GermanyTel. +49 (0) 30/80 94-15 76Fax +49 (0) 30/80 94-15 78hein@istev.comwww.istev.com

Iron granulate

Output

Gas valve

SamplingWater and gas area

SandGeotextile

Filter gravelDrainage

Input

View of the reactor ditch with partly filled modules (left) and side viewof a filled module

Coal mining and chemical industry have heavily con-taminated the soil and groundwater in the Bitter-feld area. Experiences from the last 20 years indi-cate that hydraulic soil and groundwater remedia-tion◄ is often ineffective – particularly on largeabandoned waste sites if the source of the contami-nation either cannot be precisely located or is verydifficult to remove. The remediation research inregionally contaminated aquifers◄ (SAFIRA)research project therefore used the sample modelsite at Bitterfeld-Wolfen to develop new technolo-gies and methods for in-situ remediation of ground-water contaminated with complex pollutant com-pounds.

The Bitterfeld-Wolfen area is still suffering the effects ofabandoned waste. The ground beneath the former indus-trial and landfill sites is polluted, the groundwater heavilycontaminated in places with organic compounds, prima-rily chlorinated hydrocarbons (CHC◄), over an area ofaround 25 square kilometres. In Bitterfeld this contamina-tion reaches a depth of 30 to 40 metres and affects an esti-mated 250 million cubic metres of groundwater. Conven-tional remediation procedures demand mainly protractedand expensive pump and treat measures. However, forsuch widespread contamination and such a complex mixof pollutants, in-situ cleaning is an appealing approach asit does not involve digging up and moving the soil.

Active and passive methods

In-situ remediation methods are divided into active tech-nologies (e.g. soil gas suction) and passive methods, thelatter requiring little or no energy feed during the remedi-ation process. The most developed passive version is reac-tion walls. While this method is already successfullyapplied to simple combinations of pollutants, develop-ment is still needed to handle complex mixtures.

The SAFIRA research project investigated the hydro-geo-logical and geochemical framework conditions for cost-effective in-situ procedures and tested these at the Bitter-feld site. Researchers at the UFZ centre for environmentalresearch at Leipzig-Halle and the universities of Dresden,Halle, Kiel, Leipzig and Tübingen were able to use areasformerly home to the Bitterfeld chemical industry todevelop and test the suitability of technologies for passivedecontamination in a real-life situation.

SAFIRA joint research project – remediation research using the Bitterfeld site as a model

The main objectives of the project were: ● Development and gradual implementation of effi-

cient passive water treatment technologies for organ-ic pollutant mixtures from the lab stage to a pilotphase.

● Technical-economic optimisation of the new tech-nologies plus their combination.

● Demonstration of their long-term stability under fieldconditions.

● In addition, the actual operating costs and the envi-ronmental law and planning aspects of in-situ reac-tion zones were to be evaluated.

Pilot facility for various procedures

At the heart of this project is a pilot facility installed in Bit-terfeld’s groundwater, 23 metres below the surface of thesite. The scientists used this to investigate seven proce-dures that had been successfully tested in the lab and insmall-scale field trials: ● Adsorption◄ and microbial degradation of pollutants

with activated carbon◄● Zeolite-supported palladium (PD) catalysts◄● Oxidation all-metal catalysts◄● Oxygen-reducing combination reactors◄● Membrane-supported palladium (PD) catalysts◄● Adsorption with activated carbon ● Anaerobic◄microbial degradation

WATER AS A RESOURCE | 1.1.0724

Water flow 30 ... 1000 l/h

Dual-phase separator (separation of substances heavier and lighter than water)

Precipitation/flocculation (4-chamber agitator vessel)

Separation (stainless steel lamella baffle plate thickener, lamella surface area: approx. 1 m²)

Filtration (backwashing possible), V = 185 l, sand filling (approx. 140 l coarse, medium and fine sand), p up to 10 bar

Activated carbon air filter 2 Activated carbon filters in series, 1-2 and 2-1 poss., with V = 216 l, carbon quantity each: 160 l

Gas washer for washing out HCl (pH neutralisation)

Stripping (compact stripper)

Compact stripper with droplet separator

Stripping air fan with volume flow rate of up to 135 m³/h

Catalytic oxidation exhaust gas cleaning2 reactors, each approx. 30 l

(electric heating 8 kW (approx. 350°C))

Catalytic reduction exhaust gas cleaning2 reactors, of approx. 30 l and 6 l

(electric heating 2 x 2 kW (approx. 100...300°C))

Stripping via 2 hollow-fibre membrane modules

Gas flow: 2...50 l/min N2, 0.1...5 l/min H2

Water flow: 100...1000 l/min water

Incl. vacuum pump: 4 m³/h at normal pressure

Vacuum up to 100 mbar absolute

Procedural steps carried out in the pilot scheme

ECOLOGY | GROUNDWATER | SAFIRA RESEARCH PROJECT 25

Transferring these procedures to a larger pilot scaleproved problematic in some cases. Although the micro-organisms at the site are in a position to break downchlorobenzene◄under anaerobic conditions, the rates ofdegradation were insufficient in practice so proceduresfor adding oxygen also had to be developed. The scientistsalso ascertained that although palladium catalysts weresuitable for quick reduction dechlorination◄ they neededbetter protection in groundwater containing sulphatesagainst the products used for the microbiological reduc-tion of sulphate such as hydrogen sulphide. Conventionaladsorbents (such as activated carbon) help to remove pol-lutants. The operating life of treatment walls can be signif-icantly extended through microbiological colonisation.The project also showed that oxidation catalytic proce-dures can also be used to treat complex pollutant mix-tures. Methods for reactivating the surfaces of the cata-lysts is an area where optimisation was required.

Implementation in specific remediationconcepts

In additional investigations, the project partners adaptedtheir research to the extremely diverse composition of pol-lution at each site, in particular the many different sub-stances and the high concentrations of individual sub-stances at Bitterfeld. The new projects focused on threeareas promising huge benefits for future remediation con-cepts at Bitterfeld-Wolfen: ● Innovative remediation technologies (catalysis◄) ● Coupled systems (hydrochemistry/microbiology) ● Spatial effect (digital area model, visualisation,

models)

Within innovative remediation technologies, the projectteam developed a new method that uses vacuum strips◄through a hollow-fibre membrane to convert pollutantsfrom the liquid phase◄ to the gaseous phase, whichenables a highly efficient level of destruction throughcatalysis. This technology deals with a broad range of con-taminants in high concentrations. Once the procedurehad been successfully used in a pilot scheme, it was devel-oped further at various locations. The focus of this devel-opment was on removing and destroying crucial sub-stances from a procedural perspective. A new form oftechnology will now make it much easier to treat the spe-cific contaminated groundwater in the Bitterfeld-Wolfenregion. It was tested in a pilot scheme that included a pro-cedural stage for stripping◄ out pollutants with hollow-fibre membrane modules and devices for applying inno-

vative procedural steps for removing the sulphur from thestripped-out gases (UFZ patent). The facility aims todemonstrate alternatives for complex groundwater con-tamination where there has previously been no economi-cally viable cleaning concept. The “treatment train”approach was followed to achieve a sufficient overallcleaning performance, in other words the intelligentinterlinking of modular standard and/or innovative indi-vidual processes.

Another focus was on the combination of different micro-biological degradation paths in corresponding condi-tioned aerobic◄/anaerobic zones permitting the succes-sive degradation of certain pollutants or groups of pollu-tants. The third area of focus looked at the creation of adigital database for Bitterfeld, which included a geologi-cal structure model, a description of regional groundwa-ter qualities at various periods and remediation scenariosbased on land usage.

Helmholtz Centre for Environmental Research –UFZProf. Dr. Holder WeißProf. Dr. Frank-Dieter KopinkePermoserstraße 1504318 Leipzig, GermanyTel.: +49 (0)3 41/2 35-12 53Fax: +49 (0)3 41/2 35-18 37E-mail: holger.weiss@ufz.defrank-dieter.kopinke@ufz.derobert.koehler@ufz.deInternet: www.ufz.deFunding reference: 02WT9911/9

Many organic pollutants in groundwater arebiodegradable under aerobic◄, or oxygenated, con-ditions. Adding oxygen can accelerate the degrada-tion process; a relatively cost-effective way of doingthis is to use direct gas injection. A project teaminvestigated the processes determining the efficien-cy of such direct gas injections in porous media(aquifers◄), and developed forecast models for usewithin remediation. Several demonstration projectshave already been successfully completed on thebasis of the research results.

One way to accelerate the degradation of pollutantsunderground is to use direct gas injections – an in-situremediation procedure. Injection lances moving both hor-izontally and vertically are used for the targeted introduc-tion of oxygen into the stream of contaminated ground-water, where it establishes itself in the aquifer in the formof small, finely distributed bubbles. Low-permeability lay-ers of sediment prevent the gas from dissipating upwards;sideways-expanding oxygen-flow capillary networks◄form instead. The immobile gas phase◄ acts hydraulicallyand biologically like a reactive oxygen wall that cleans thegroundwater flowing through it. This means the bubblesdissolve slowly and add oxygen to the groundwater pass-ing by, supporting the process of degrading the pollutantsit contains, while new oxygen is constantly fed in fromabove.

Controlling the spatial effect of the gasaccumulator

Between 2000 and 2010, the BMBF financed the SAFIRA,PROINNO 1 and 2 and ZIM demonstration projects inorder to research this procedure. The research sitesincluded the site of a former chemical plant in Leuna andthe Auensee recreation area in Leipzig, which is underthreat from the groundwater being contaminated withPCE-TCE◄. Conventional technologies for direct gas injec-tion are based on homogeneous gas distribution throughlances. The distribution itself is not measured, meaningthat these technologies operate “blind”. Basically, the gasdistribution at the injection lances behaves like a dynamicgas accumulator comprising branch-like and inter-related(coherent) channels of gas and non-related (incoherent)gas clusters. The injection or release processes cause theseeither to expand or draw closer together.

Direct oxygen injection – measuring and modelling dynamic gas accumulators

The heart of the innovation and the scientific challenge ofthe projects lay in the small-scale metrological recordingof the gas dispersion and accumulation processes in theheterogeneous conditions underground and the interpre-tation and control of these using computer models. Theaim was to control the spatial effect of the gas accumula-tion. As well as conventional low-pressure injection (LPI),the scientists also experimented with high-pressure injec-tion (HPI) for the first time.

Improved monitoring

Sensatec in Kiel worked together with the UFZ centre forenvironmental research at Leipzig-Halle to develop a newsystems technology for coupled LPI-HPI direct gas injec-tion. This involved both a reliable in-situ gas measure-ment system and a dynamic gas dispersion model on thebasis of which the new injection process can be controlledand optimised. The measuring system features an innova-tive set of sensors (sensor array), which enables it to meas-ure and store large amounts of data extremely quickly.This makes it suitable for recording changes in the gastransport and accumulation processes in the otherwiseheterogeneous conditions underground. As this is muchfaster than typical groundwater transport processes, stan-dard systems are unsuitable for this type of monitoring.With regard to the measurement data, the scientists usedsuitable geostatistical processes to calculate intermediatevalues (interpolation) and visualised all the data in 3D.These 3D data fields formed the basis for developing thegas dispersion model.

WATER AS A RESOURCE | 1.1.0826

Low-pressure injection High-pressure injection

3D representation of gas distribution underground with LPI andHPI direct gas injection

ECOLOGY | GROUNDWATER | DIRECT OXYGEN INJECTION 27

Technologically relevant results

The research projects produced the following results:● Independent measuring of horizontal and vertical

hydraulic conductivity is essential for model-support-ed forecasting of dynamic gas accumulation under-ground. Thin clay layers in particular can significantlyinfluence the dynamics of the gas flows reaching verti-cally upwards. In existing groundwater models, thevertical conductivity was calculated from the horizon-tal conductivity using an empirical factor (< 1). Howev-er, this basic approximation breaks down when itcomes to predicting the dispersion behaviour ofdynamic gas accumulators.

● A fine-scale site survey is required to ensure the opti-mum vertical and horizontal positioning of injectionlances and sensors. As a minimum, monitoring mustinvolve dynamic drilling◄ and injection logs◄ inorder to determine hydraulic conductivity and geo-electric profile recordings◄.

● The gas must be measured in a sealed in-situ sensornet adapted to the local conditions (approx. 60 meas-uring points per injection lance; 1 sensor per m3),which is essential for successful predictions of gas dis-persion in heterogeneous sediments.

● The experts derived from the laboratory experiments aworking hypothesis that low-pressure injections leadto incoherent gas transport and high-pressure injec-tions lead to coherent gas transport. This finding isimportant when it comes to ascertaining the correctdimensions for the gas barriers.

● A sensor system must be able to measure and store lotsof data in a short period of time and to differentiatebetween coherent and incoherent gas transport.

● LPI scenarios and combined LPI/HPI scenarios with anextremely low injection rate of 0.18 cubic metres anhour lead to an incoherent accumulation of gas. Con-versely, LPI scenarios with an injection rate ten timeshigher than this and pulsed HPI scenarios lead to acoherent accumulation of gas. This means that it isclearly the injection rate and not the injection pres-sure that predominantly affects the various gas distri-bution patterns. The only significant differencebetween the two injection methods is that high-pres-sure injection achieved a higher level of gas saturationin the lower range.

● A 3D gas dispersion model for optimising the coupledLPI/HPI gas supply procedure must:a) be a multi-phase model,b) factor in heterogeneous horizontal and vertical

permeability◄ and capillary pressure fields◄, c) use parameter fields conditioned for the measure-

ment data.● The experiments regarding the gas dispersion process-

es conducted in the lab and the field are also of greatinterest to CCS technology, i.e. the underground accu-mulation of greenhouse gases.

Project website ►http://safira.ufz.de

Helmholtz Centre for Environmental Research –UFZHydrogeology departmentProf. Dr. Helmut GeistlingerTheodor-Lieser-Straße 406120 Halle, GermanyTel.: +49 (0)3 45/5 58-52 20Fax: +49 (0)3 45/5 58-55 59E-mail: helmut.geistlinger@ufz.deFunding reference: 02WT9947/8

Example 3D computer simulation of gas distribution with LPI and HPI direct gas injection

Pollution is contaminating soil and groundwater allover the world. Land affected by this is difficult touse and hard to sell because remediation of the soiland groundwater is extremely complex and expen-sive. It therefore often remains unused. At the sametime, an area of around 90 hectares a day in Ger-many alone is being redeveloped as green-field land.The specialists at the Versuchseinrichtung zurGrundwasser- und Altlastensanierung (research facil-ity for the remediation of groundwater and aban-doned waste, VEGAS) are therefore developing tech-nologies that enable efficient surveying and remedi-ation of contaminated land to render it marketableonce more.

The soil and groundwater at many former industrial sitesand also in numerous urban areas has been contaminatedthrough improper disposal methods, accidents, the effectsof war and careless handling of hazardous substances.Consumers have also worsened the problem in the past:household chemicals, paint and other toxins that are nowdealt with as specialist waste were disposed of at unse-cured landfill sites with permeable ground underneath.For example, exhausted gravel pits were filled with allkinds of refuse and then planted over.

Conventional in-situ remediation techniques are oftenprotracted and expensive. The physical properties of thepollutants means that they collect in the soil pores and aredifficult to remove with traditional flushing methods.However, if contaminated material is dug out and dis-posed of at a dump, this only serves to delay the problem.Built-up land or land with deep-lying, hard-to-pinpointsources of pollution cannot be dug up in any case. One ofthe most important tasks we face today therefore is todevelop new technologies offering economical and eco-logical remediation before the abandoned waste poses athreat to both people and protected natural resources.

The VEGAS concept

With support from the BMBF and the Ministry of the Envi-ronment in Baden-Württemberg, the major research facil-ity VEGAS was established by the University of Stuttgart inSeptember 1995 (size: 670 m2; large-scale test rig surfacearea 18 x 9 m, height 4.5 m, divisible into three compart-ments). The engineers and scientists there develop survey-ing and remediation technologies and operate field appli-cations and technology transfers. The large-scale test rigsare used to perform experiments under near-natural con-

VEGAS research facility – ecological remediation without digging up and shipping out

ditions, their dimensions being somewhere between con-ventional lab equipment and actual field conditions.There are good reasons for such middle ground: conven-tional lab tests are not directly transferrable to actual con-ditions “in the field”, and the results from time and cost-intensive field studies at existing pollution sites are onlymeaningful to a limited extent. This is because more oftenthan not the remediation specialists are unaware of boththe overall amount of pollution and its precise spatial dis-tribution. The few distributed measurement points do notproduce a sufficiently detailed overview. Furthermore,environmental protection laws currently in force prohibitthe injection of remediation chemicals into the aquifer◄ ifit cannot be guaranteed that they are harmless. However,such a guarantee cannot be given when testing new tech-nologies under development.

Innovations in technology

Abandoned waste in the soil can either be cleaned up orsafeguarded. The remediation approach involves remov-ing the source of the contamination or the plume throughchemical, biological or hydraulic methods. Safeguardingprevents the further spread of the pollution, e.g. by con-taining the source. In the last few years, VEGAS has con-centrated primarily on source surveying and remediation.Appropriate procedures, some of which have already beenput into practice in the field in collaboration with compa-nies include:

WATER AS A RESOURCE | 1.1.0928

Society

• Acceptance

• Social compatibility

Environmental policy

• Remediation goals

• Threshold values and

standards

Ecology

• Impact

• Sustainability

Laboratory experiments

and principles

Simulation technology and

hydroinformatics

Practice

• Field investigations

• Remediation

experiments

Teaching and training

(interdisciplinary)

VEGAS

Large-scale technological

experiments

Technical feasibility

Efficiency

VEGAS – Contributing to research, practice and teaching

ECOLOGY | GROUNDWATER | VEGAS RESEARCH FACILITY 29

● Thermal remediation technologies: Energy feeds –either in the form of vapour or a vapour-air mix or as afixed source of heat – increase the temperature of thecontaminated soil and groundwater areas. Thisreduces the surface tension◄, viscosity◄ and densityof the pollution while simultaneously increasing itsvapour pressure. This boosts its transformation intogas and enables it to be sucked out via the soil gas.

● In-situ reduction of pollution through minute piecesof iron (nanoscale iron) in treatment walls: this reme-diation procedure is applied to CHC plumes. Expertsare currently examining how this technology couldalso be applied within built-up areas. This involvesinjecting nanoscale iron directly into the source of thecontamination by means of suspension◄. The issues oftransportation, reactivity, and long-term stability andalso economics still need to be clarified; these too mustbe answered through the pilot sites within variousresearch consortia.

Other types of technology (further) developed by universi-ties, companies, local authorities and institutes withinVEGAS involve procedures such as the injection of ten-sides, alcohol cocktails or microemulsions. Other newoptions included special remediation wells, in-situ chemi-cal oxidation (ISCO◄) and reduction (ISCR◄), immobilisa-tion of heavy metals and improvement of the naturaldegradation processes within the aquifer (enhanced natu-ral attenuation, ENA) through the addition of nutrients.

Focus on measurement technology

The field of measurement technology has been consistent-ly expanded within VEGAS:1. Surveying: The position and concentration of a pollu-

tion source must be precisely identified in order toimplement remediation. The VEGAS researchers havetherefore developed new methods, e.g. based on sen-sors and fibre optics, to enable a quick and cost-effec-tive way of pinpointing sources of pollution.

2. Monitoring: New instruments for on-site measure-ment technology enable a prompt return of informa-tion on the distribution of and decrease in pollution,thus lowering the cost of remediation.

3. Long-term observation: Once remediation is complete,the experts must use automated long-term observa-tion to prove that the concentration of pollution is notrising again and that the threat is therefore effectivelyneutralised.

4. Geothermal energy: To date there has been littleresearch on the influence of geothermal systems ongroundwater. The monitoring of water temperatureand quality in the vicinity of geothermal probes isintended to contribute to their safe use in the longterm.

Transfer of technology and knowledge

In order to ensure sustainable protection for soil andgroundwater, it is not enough simply to develop technolo-gy. That is why the research facility holds regular trainingevents for specialists working at authorities and engineer-ing bodies. The technologies developed are conveyed to abroad specialist audience through pilot schemes conduct-ed at actual pollution sites.

Project website ►www.vegasinfo.de

University of StuttgartInstitut für Wasserbau (water engineering institute), VEGASPfaffenwaldring 6170550 Stuttgart, GermanyTel.: +49 (0)7 11/6 85-6 47 17Fax: +49 (0)7 11/6 85-670 20Funding reference: 02606861

VEGAS scientific managerDr. Jürgen BraunTel.: +49 (0)7 11/6 85-6 70 18E-mail: juergen.braun@iws.uni-stuttgart.de

VEGAS technical managerDr.-Ing. Hans-Peter KoschitzkyTel.: +49 (0)7 11/6 85-6 47 16E-mail: hans-peter.koschitzky@iws.uni-stuttgart.de

Using methods from the crude oil industry as a mod-el for protecting the environment? A research teamfrom the Versuchseinrichtung zur Grundwasser- undAltlastensanierung (research facility for the remedi-ation of groundwater and abandoned waste, VEGAS)at the University of Stuttgart has shown that this isindeed possible. Experts at the institute for waterengineering (IWS) there and the institute for hydro-mechanics at the University of Karlsruhe (IfH) devel-oped an alcohol-based form of remediation technol-ogy for treating contaminated aquifers◄. They con-centrated primarily on contamination throughhydrocarbons of varying density and medium to lowsolubility (LNAPL/DNAPL◄). Their work has resultedin a procedure that enables aquifers contaminatedin this way to be cleaned “in situ” – i.e. in its existinglocation.

The crude oil industry first of all managed to use tradition-al pumping methods to extract around 40% of the accu-mulated crude oil from the discovered source. The mainreasons for the low yield were the surface tension◄ andthe different viscosities◄ and densities of water and oil.The specialists at the oil company then injected alcoholand tensides into the oil fields as solubilisers to test theireffect. These substances reduced the surface tension andsignificantly increased pumping efficiency.

However, certain types of alcohol prevent the CHCs frommoving freely. This put alcohol flushing back in the pic-ture for use in in-situ groundwater remediation. Thismethod is only economically viable though if alcohols canbe found that enable a quick and controlled discharge ofcontaminants in dissolved form or as a free phase◄ andthat can be recovered and reused several times duringremediation.

Large-scale tests

The BMBF-funded project entitled “Development of anadvanced groundwater remediation technique forthe removal of anthropogenic chlorinated hydrocar-bons with high density (CHC) by alcohol injections”saw scientists from IfH and VEGAS testing whether alcoholflushing was suitable for remediation and investigatinghow to determine the relevant dimensions. The researchfocused on efficiency, stability of the cocktail (separabili-ty), production costs and above all hydraulic control. On

Remediation using alcohol – using methods from the crude oil industry as a model

the one hand, this means that a cocktail of alcohol with itsspecific physical properties must be transported in a tar-geted manner to the source of the contamination; on theother, that uncontrolled mobilisation of the contamina-tion must be avoided. The IfH investigated what happenswhen alcohol cocktails are injected in a spatially targetedmanner into a contaminated aquifer and whether theeffects can remain under control. To clarify these ques-tions, the experts from both institutes performed twolarge-scale tests among others in realistic conditions.

The researchers produced a mixture of hydrocarbons(BTEX◄) of low density (LNAPL) in a 6 x 3 x 4 metre tank sothat the pollution lay in residual saturation – i.e. capturedthrough capillary forces – beneath the surface of the waterand as a floating phase. They then used a horizontal wellto target injection of a mixture of alcohol and water (iso-propanol and water at a ratio of 60:40) into the artificialaquifer. The alcohol permeated the contaminated regionand dissolved the pollution, which could then be pumpedout via two vertical wells. The result: almost 90% wasremoved.

In further tests, the scientists introduced a source of CHCcontamination (TCE) to a heterogeneous artificial aquiferin a large tank (9 x 6 x 4.5 m). They used a groundwater cir-culation well to inject a cocktail of alcohol into the lowerregions of the aquifer. At the same time, the same well was

WATER AS A RESOURCE | 1.1.1030

Water - Alcohol - Contaminant

Vertical wells

LNAPL

Horizontal well Alcohol injection

Alcohol flushing of aquifers contaminated with LNAPL

ECOLOGY | GROUNDWATER | REMEDIATION USING ALCOHOL 31

used to draw out the mixture of alcohol, water and pollu-tion from the upper regions. Scientists managed again toremove over 90% of the contamination – in dissolved butalso mobilised form. It must be emphasised that no uncon-trolled downwards vertical displacement of the pollutiontook place.

The alcohol used was recycled in order to reduce costs andthe amount of wastewater. The project team designed andbuilt a wastewater treatment facility for this purpose.

Alcohol cocktails for tough cleaning

These successful tests meant that the researchers couldnow make more accurate statements on which alcoholcocktails are suitable for which types of remediation. Forexample, a cocktail of 2-propanol (54% volume), 1-hexanoland water (both 23% volume) is recommended for CHCcontamination. This mixture enabled the soil material tobe cleaned safely and efficiently in all tests. The requiredinitial concentration of the mixture depends on the flowconditions and the heterogeneity of the soil. It must benoted that the higher the proportion of alcohol, the moreexpensive the remediation process. The lower the propor-tion, the greater the risk that the cocktail will separate.

Complex use cases

Finally, the project partners used the test data to producemathematical equations for the dependency of the densi-ty, viscosity and surface tension on temperature and themixture ratio. They are currently being used to expand theMUFTE-UG (Multiphase Flow, Transport and Energy Model– Unstructured Grid) numeric model at the department ofhydromechanics and modelling of hydrosystem at IWSwith a module that can simulate a complex multi-phase/multi-component flow.

Project website ►www.vegasinfo.de

Karlsruhe Institute of Technology (KIT) Department of Civil Engineering, Geo and Environ-mental Sciences Institute of HydromechanicsProf. Gerhard H. JirkaKaiserstr.1276128 Karlsruhe, GermanyTel.: +49 (0)721/608-2200Fax: +49 (0)721/608-2202Funding reference: 02WT0065

University of StuttgartInstitut für Wasserbau (water engineering insti-tute), VEGASPfaffenwaldring 6170550 Stuttgart, GermanyTel.: +49 (0)7 11/6 85-6 47 17Fax: +49(0)7 11/6 85-6 70 20E-mail: vegas@iws.uni-stuttgart.deFunding reference: 02WT0064

VEGAS scientific managerDr. Jürgen BraunTel.: +49 (0)7 11/6 85-6 70 18E-mail: juergen.braun@iws.uni-stuttgart.de

VEGAS technical managerDr.-Ing. Hans-Peter KoschitzkyTel.: +49 (0)7 11/6 85-6 47 16E-mail: hans-peter.koschitzky@iws.uni-stuttgart.de

Large VEGAS well (alcohol flushing for DNAPL contamination)

High temperatures are a good way to extract pollu-tion from contaminated soil. This has been provenby researchers at the Versuchseinrichtung zurGrundwasser- und Altlastensanierung (research facil-ity for the remediation of groundwater and aban-doned waste, VEGAS) within the University ofStuttgart’s water engineering institute. Their ther-mal in-situ remediation technology (THERIS) – justone of the procedures used for thermal remediation– uses a strong infeed of heat to clean out contami-nation from underground, which is faster than theconventional means of “cold” soil gas suction.Experts at the VEGAS research facility (see project1.1.09) have compared both processes with eachother as part of a research project – with very clearresults.

Medium to low solubility liquid hydrocarbons(LNAPL/DNAPL◄) contaminate soil and pose a risk togroundwater. Landowners frequently turn to a processcalled soil gas suction (SGS) for remediation. This involvesthe pollutants, which convert into gas at a sufficientlyhigh natural temperature, being sucked out of the con-taminated ground together with the air within throughpipes (soil gas level). A discharged air treatment facility fil-ters out and removes the pollution from the contaminatedsoil air on site and the system then releases the cleaned airinto the atmosphere. However, this method quickly reach-es its limits: the organic pollutants only convert from liq-uid to gas◄ in very small amounts at the usual soil temper-ature of about 10°C, which makes the remediation processmuch slower and much more expensive. Furthermore,fine-grain soils – in which pollutants most often accumu-late – are not permeable enough for this process. The soilgas is then very difficult to suck out. After several years, itis still uncertain whether many of these facilities areachieving their remediation objective.

Heat infeed boosts contaminant discharge

These problems can be overcome by applying heat, e.g. viafixed sources of heat such as the electrical heat lances onwhich the THERIS procedure is based. The soil heats up,pollutants convert more quickly to gas and are also dis-charged from smaller areas of permeable soil. SGS sucksout gaseous contaminants directly from under the groundalong with the soil gas. The suctioned, contaminated soilgas is then cleaned on site. THERIS can be used with vari-ous soil types, particularly those with loose stones (sand,

Heat as an accelerator – heat lances turning soil pollutants into vapours

silt, clay) in the soil zone above the groundwater, i.e. thenon-saturated soil zone. Unlike thermal in-situ remediation procedures – vapouror vapour-air injection – no heat medium is injected dur-ing the THERIS process. This makes THERIS also suitablefor heating and cleaning large, low-permeability soil lay-ers. Comparative experiments in the large tank at theVEGAS research facility and at a field site should now clari-fy the differences between SGS and THERIS in terms ofremediation time, degree of effectiveness and energy consumption.

Experiment in the large VEGAS tank

The researchers began testing the THERIS procedure in a150 cubic metre tank at the VEGAS research facility, whichwas filled with a layered soil structure and equipped withmeasuring instruments. They added 30 kilograms oftrimethylbenzene (TMB, flash point 169°C) to a localisedarea of a one-metre thick layer of fine low-permeabilitysand.

To begin with, they applied cold soil gas suction for twomonths. They then put four quadratically arranged, fixedheat sources (THERIS) into operation in this fine-struc-tured layer, which heated up to 500°C. The SGS continuedto deliver a constant rate of suction. As the TMB was fullyremoved after just 20 days, the soil temperature betweenthe heating elements had not even hit 100°C.

WATER AS A RESOURCE | 1.1.1132

Zone U. Hiester , H.-P. Koschitzky T. Theurer, A. Winkler,

Ground to air (gas) extraction

Surface

Fixed heat sources

e.g. silt or clay layer

Non-saturated zone

Groundwater

Contamination

Principle of the THERIS procedure

ECOLOGY | GROUNDWATER | HEAT AS AN ACCELERATOR 33

Remediation in the field test

The researchers also began a field test: an approximately3.5 metre thick clay-marl layer over an area of around 80square metres still contained high concentrations ofCHC◄ after several years of cold SGS, primarilyperchloroethylene◄ (PCE, flash point 121°C). To beginwith, the SGS was left to continue without any change.Then 22 heat lances from the THERIS system were imple-mented, equipped with the latest measuring, control,data recording and transfer technology. They heated thesoil evenly and in a way relatively uninfluenced by differ-ences in geological structure. After just one month thetemperature was around 80°C, and largely over 90°C aftertwo. The researchers sucked out the gaseous pollutantsmobilised by THERIS using SGS.

THERIS has the edge

The evaluation of the tests showed that THERIS cut theremediation time by about 90% compared with cold SGSand reduced the amount of energy required by two-thirds.Differences in the absolute values between the tank andthe field test were traced to location factors such as geolo-gy, the type of contamination and its distribution plus sys-tem-specific details (e.g. the well arrangement).

THERIS takes just a few weeks to clean even low-perme-ability soil quickly, reliably and with sustainable results.The significant increase in performance is predominantlyattributable to: ● the increased gas permeability and more effective dif-

fusion as a result of the dried soil ● the accelerated conversion of liquid hydrocarbons to

gas as a result of the higher temperatures.

The THERIS facilities are robust, enabling them to beinstalled quickly and operated safely whatever the weath-er. Due to the cost of putting the heating elements andmonitoring systems in place, the installation costs arehigher than for cold soil gas suction. However, these addi-tional costs are compensated for by the shorter remedia-tion time, lower operating costs (energy, equipmentrental etc.) and lower personnel costs.

Project website ►www.vegasinfo.de

University of StuttgartInstitut für Wasserbau (water engineering institute), VEGASPfaffenwaldring 6170550 Stuttgart, GermanyTel.: +49 (0)7 11/6 85-6 47 17Fax: +49 (0)7 11/6 85-6 70 20E-mail: vegas@iws.uni-stuttgart.deFunding reference: 02WT0266

VEGAS scientific managerDr. Jürgen BraunTel.: +49 (0)7 11/6 85-6 70 18E-mail: juergen.braun@iws.uni-stuttgart.de

VEGAS technical managerDr.-Ing. Hans-Peter KoschitzkyTel.: +49 (0)7 11/6 85-6 47 16E-mail: hans-peter.koschitzky@iws.uni-stuttgart.de

THERIS procedure: areas of application Heat lances and soil gas suction level being installed in the large tank

Clay Silt Sand Gravel

“Cold” SGS

Vapour-air injection

THERIS