BRENNEN

Industries / Energy, water and waste / Environmental services / Environmental remediation

Treating in place

Treating in place cleans polluted soil and groundwater where they lie, by pulling out vapors, heating the ground, or injecting chemicals, microbes or carbon, without digging the soil up or pumping the water out.

6

Milestones achieved, from 1989 to 2016.

1

To watch for: announced, not yet confirmed outside.

2

Open deliverables. Nobody has shipped them.

The road

AchievedWatch forOpen

Achieved

  1. 1989

    Vacuum pulls solvents from soil

    Soil vapor extraction pumps air out of wells set in the dry soil above the water table, so solvents that evaporate easily leave as vapor. In a demonstration for EPA's SITE program, which tests vendors' cleanup methods, at the Groveland Wells Superfund site in Massachusetts, Terra Vac's vacuum extraction system ran for 56 days in soil polluted with trichloroethylene (TCE), a degreasing solvent, and removed 1,300 lb of volatile organic compounds in eight weeks, by EPA's account.

    LessonEPA found the method also worked in low-permeability soils such as clays, as long as the soil held enough air-filled pore space.

    osti.govosti.gov

    Credit

    Terra Vac, Inc.; evaluated for EPA by M. K. Stinson

  2. 1992

    Heating ground to strip fuel

    Lawrence Livermore National Laboratory (LLNL) and the University of California, Berkeley, combined steam injection, electric heating of clay layers and underground electrical imaging, a method they called dynamic underground stripping, to clean a gasoline spill at the laboratory that lay partly below the water table. Over several months of field operations it recovered more than 26,500 liters (7,000 gallons) of gasoline, the laboratory reported.

    LessonHeat only moves the pollutant into vapor that must still be captured and destroyed above ground, so LLNL went on to develop hydrous pyrolysis oxidation, which adds oxygen so the chemicals break down underground.

    osti.govosti.govosti.gov

    Credit

    R. D. Aines, R. L. Newmark and colleagues, LLNL; University of California, Berkeley

  3. 1997

    Bacterium turns solvent into ethene

    Tetrachloroethene (PCE), the main dry-cleaning solvent, can be broken down by microbes that strip off its chlorine atoms one at a time, ending in ethene, which is harmless. A Cornell University team isolated strain 195, a bacterium that carries PCE all the way to ethene while living on hydrogen; its genetic sequence placed it in no known group of bacteria. Later papers call its relatives Dehalococcoides.

    LessonThe strain grew only when fed extracts from mixed microbial cultures, an early sign that it depends on other microbes.

    ebi.ac.uk

    Credit

    X. Maymó-Gatell, Y. Chien, J. M. Gossett and S. H. Zinder, Cornell University

  4. 2002

    Bacteria added to an aquifer

    Bioaugmentation means injecting cultured microbes into the ground. At Kelly Air Force Base in Texas, where groundwater held about 1 mg/L of PCE, feeding the native microbes methanol and acetate broke the PCE down only as far as cis-dichloroethene (cDCE), a partly dechlorinated product. After 13 liters of KB-1, a culture containing relatives of Dehalococcoides ethenogenes, were injected into a test area holding about 64,000 liters of groundwater, PCE, TCE and cDCE all fell below 5 µg/L within 200 days, and ethene accounted for the loss.

    LessonThe method is now routine: by EPA's count, 9 of the 28 Superfund groundwater decisions that chose bioremediation in fiscal years 2021 to 2023 included bioaugmentation.

    ebi.ac.ukepa.gov

    Credit

    D. W. Major and co-authors, GeoSyntec Consultants

  5. 2006

    Source treatment rarely meets standards

    A source zone is the patch of ground holding undissolved solvent that feeds a groundwater plume. Reviewing 59 chlorinated solvent sites whose source zones were treated in place by chemical oxidation, enhanced bioremediation, heating, or flushing with surfactants and cosolvents, the authors found median drops of 88% or more in the main solvent, yet none of the 59 projects met drinking-water limits (maximum contaminant levels) throughout the treated zone.

    LessonConcentrations can rebound after treatment: at seven chemical-oxidation sites with at least a year of follow-up, the median drop shrank from 90% right after treatment to 78% one to five years later, while at ten bioremediation sites it grew from 77% to 96%.

    api.crossref.org

    Credit

    Travis M. McGuire, James M. McDade and Charles J. Newell

  6. 2016

    Results from 235 projects pooled

    GSI Environmental, for the Defense Department's Environmental Security Technology Certification Program (ESTCP), gathered before-and-after groundwater data from 235 in-place treatment projects at chlorinated-solvent sites, begun between 1991 and 2012. Typical projects cut concentrations by 0.5 to 2 orders of magnitude, that is, to between about a third and a hundredth of the starting level. Rebound was most common after chemical oxidation, in about 30% of monitoring wells, then after bioremediation, in about 25%, and least common after heating or chemical reduction. Typical unit costs ran from $100 to $300 per cubic yard, and cost did not track results.

    LessonNo site feature the team tested, such as geology, size or depth, strongly predicted how well a treatment would work.

    serdp-estcp.milsepub-prod-0001-124733793621-us-gov-west-1.s3.us-gov-west-1.amazonaws.com

    Credit

    Travis McGuire, David Adamson, Charles Newell and Poonam Kulkarni, GSI Environmental; ESTCP project ER-201120

Watch for

  1. Carbon barriers for decades

    Colloidal activated carbon is charcoal ground fine enough to inject through wells; it coats the aquifer's grains so that PFAS, the long-lasting "forever chemicals," stick to them instead of flowing on. An early published field case, at a site in Canada, kept PFOA and PFOS below 30 ng/L for 18 months (McGregor, 2018). In May 2026 Jeremy Birnstingl, a vice president of REGENESIS, which sells such carbon, reviewed field data from 37 barriers, reported drops of up to five orders of magnitude, and wrote that modeling shows well-designed barriers can stay protective for "multidecadal to century-scale times." In June 2026 a team from Carnegie Mellon University, the US Navy's engineering center and others reported in lab columns that PFAS already held on the carbon can wash off again: with diesel-range organics in the water, about 70% of the PFOA mass came back out. EPA's January 2025 remedy report counted injected activated carbon among Superfund groundwater remedies for the first time, with eight selections in fiscal 2021 to 2023.

    Achieved whenindependent field data, starting with the Defense Department's evaluation at Naval Air Station Jacksonville (ESTCP project ER20-5182, due in 2026), show barriers holding PFAS for a decade or more, or a regulator's review of field data confirms the forecast.

    api.crossref.orgregenesis.comapi.crossref.orgapi.openalex.orgepa.govserdp-estcp.mil

    Claimed by

    Jeremy Birnstingl, REGENESIS, May 2026

Open

  1. 01

    Independent tally of carbon barriers

    A public table, independent of any supplier, of every colloidal activated carbon barrier injected to hold PFAS for which monitoring data are public: site, date, carbon dose, PFAS upstream and downstream by compound, years monitored, and whether any compound has broken through. Each site gets an open transport model that forecasts when breakthrough would come.

    The brief
    Credit · unclaimed
  2. 02

    Update the in-place results

    An open, project-by-project table of in-place treatments at chlorinated-solvent sites begun after 2012, coded with the same fields as the 2016 Defense Department database (technology, scale, geology, treatment dates, and before-and-after concentrations of the main solvent), so that its findings on typical drops and on rebound can be rechecked on newer projects and newer methods, such as injected carbon.

    The brief
    Credit · unclaimed

Open 01 · the brief

Independent tally of carbon barriers

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

A public table, independent of any supplier, of every colloidal activated carbon barrier injected to hold PFAS for which monitoring data are public: site, date, carbon dose, PFAS upstream and downstream by compound, years monitored, and whether any compound has broken through. Each site gets an open transport model that forecasts when breakthrough would come.

Why it is still open

The largest multi-site review, of 37 barriers, comes from the carbon's maker. A 2024 conference paper by Grant Carey of Porewater Solutions, with co-authors from the Navy, the Air Force and the firm APTIM, summarized 20 case studies but gives no site-level results on its public page, and the Defense Department's independent evaluation covers one site, Naval Air Station Jacksonville, with results due in 2026. Other independent field records cover single sites over short periods: Sweden's government pilot barrier at Örnsköldsvik airport showed no notable breakthrough within two years, and its researchers say long-term performance "still remains uncertain, as PFAS breakthrough under field conditions can take decades." Lab work in 2026 found conditions under which held PFAS is released again.

api.crossref.orgmeetingorganizer.copernicus.orgapi.openalex.orgproceedings.battelle.orgserdp-estcp.mil

Built with what exists

EPA's downloadable remedy component data and five-year review reports (EPA counted eight Superfund decisions selecting injected carbon in fiscal 2021 to 2023) and state regulator files; MODFLOW 6 and MT3D-USGS run through the FloPy Python library, with sorption settings from published column tests; a model built for extracting from long documents, such as Muse Ember (see the Note "The jobs agents repeat"); an agent that cites the page behind each value (see the Note "Findings that cite notes").

  1. First step

    Download EPA's remedy component data, list the decisions that selected in-place carbon, and collect each site's latest monitoring report.

  2. Done looks like

    At least 15 barriers with at least two years of public data after injection, every value linked to its source page; two reviewers agree on the breakthrough flag for at least 90% of barriers; forecasts are published before each site's next sampling round and scored against it.

Open 02 · the brief

Update the in-place results

Credit · unclaimed

The first to publish is credited on this line, and the deliverable moves from Open to Achieved.

What to build

An open, project-by-project table of in-place treatments at chlorinated-solvent sites begun after 2012, coded with the same fields as the 2016 Defense Department database (technology, scale, geology, treatment dates, and before-and-after concentrations of the main solvent), so that its findings on typical drops and on rebound can be rechecked on newer projects and newer methods, such as injected carbon.

Why it is still open

The 2016 database is the largest pooled set found, and ESTCP posts its project data in a free spreadsheet, but the newest of its 235 active projects began in 2012. A search found no later pooled update.

serdp-estcp.milsepub-prod-0001-124733793621-us-gov-west-1.s3.us-gov-west-1.amazonaws.com

Built with what exists

The Data sheet of the 2016 spreadsheet as the template; newer project data from EPA five-year reviews, ESTCP final reports and state databases such as California's GeoTracker; a model built for extracting from long documents, such as Muse Ember (see the Note "The jobs agents repeat"); an agent that cites the page behind each number (see the Note "Findings that cite notes"); Python with pandas.

  1. First step

    Download the 2016 spreadsheet, list the fields on its Data sheet, and code five projects from ESTCP final reports published since 2016 into the same fields.

  2. Done looks like

    At least 100 projects begun after 2012, each value linked to its source page; two people agreeing on at least 95% of fields in a 10% sample; and the 2016 measures recomputed with confidence intervals, set beside the 2016 figures of about 30% of wells rebounding after chemical oxidation and about 25% after bioremediation.