PRB Introduction

1
An Introduction To Permeable Reactive Barriers
(PRB) Volker Birke Ernst Karl
Roehl
University of Applied Sciences
University of Karlsruhe Applied
Geosciences Karlsruhe
Fachhochschule Nordostniedersachsen
2
Definition
Permeable Reactive Barriers are "passive in situ
treatment zones of reactive material that
degrades or immobilizes contaminants as ground
water flows through it. PRBs are installed as
permanent, semi-permanent, or replaceable units
across the flow path of a contaminant plume.
Natural gradients transport cont-aminants through
strategically placed treatment media. The media
degrade, sorb, precipitate, or remove
chlo-rinated solvents, metals, radionuclides, and
other pollutants."
EPA (1999), Remedial Technology Fact Sheet,
542-R-99-002
3
Source http//www.eti.ca/eti.html
4
contamination source
GW
clean groundwater
plume
Aquifer
Aquitard
reactive barrier
LNAPL light non-aqueous phase liquids
DNAPL dense non-aqueous phase liquids
5
PRB Concept
"Emission oriented remediation approach" ? Deconta
mination of the plume (vs. removal of the
contaminant source) Passive system ? No active
pumping of groundwater ? Low maintenance
following installation
6
Basic Concept "Emission oriented remediation
approach" ? Clean-up of the plume, not the
source Passive system ? No pumping required
Application ? Unclear location of source(s) ?
Slow contaminant release from source ? Low
solubility of contaminants ? Large volumes of
contaminated soil ? Built-up areas
7
Site Characteristics ? Flow field
(hydraulics) ? Contaminant concentrations ? Total
contaminant mass expected ? Groundwater
characteristics
Treatability Study ? Choice of attenuation
mechanism and reactive material ? Column
tests ? Determination of required residence
time ? Calculation of barrier thickness
8
Types of reactive walls
Degradation Chemical and/or biological
reac-tions converting the contaminants to
harmless by-products. Sorption Contaminant
removal from ground-water through adsorption or
complexation. Precipitation Fixation of
contaminants in insoluble compounds and minerals.
9
Types of reactive walls
a) Continuous Barrier (CRB)
b) Funnel-and-gate (FG) system
10
Source Gavaskar et al. 1998
11
Reactive Material Requirements

• High contaminant attenuation
• Good selectivity for target contaminants
• Fast reaction rates
• High hydraulic permeability
• Long-term stability
• Environmental compatibility
• Sufficient availability in homogenous quality
• Cost-effectiveness

12
Reactive Materials targeting Organic Contaminants
Main source Dahmke et al. (1996) own additions
13
Reactive Materials targeting Inorganic
Contaminants
Main source Dahmke et al. (1996) own additions
14
PRB Operating Requirements
Hydraulic conductivity A minimum permeability
must be guaranteed during barrier operation to
avoid that contaminated groundwater by-passes the
system.
Homogeneity In areas of favoured flow-paths
there is the danger of a fast consumption of the
reactive material's contaminant attenuation
capability.
15
PRB Operating Requirements
Barrier life-time
Period during which the reactive material keeps
its ability to remove the target contaminants
from the groundwater.
Period during which the PRB keeps its hydraulic
performance.
16
Long-term Performance Aspects
The barrier life-time is governed by

• Type and concentration of contaminants
• Type and kinetics of sorption and/or degradation
  processes.
• Type and mass of reactive material
• Hydraulic characteristics of the site (flow
  velocity)
• Geochemical characteristics of the ground-water
  (Eh, pH, composition)

17
Long-term Performance Aspects
Considerations on mass flux
Hydraulic model of the former gas works site in
Portadown, Northern Ireland.
Source Kalin, R., presentation at PRB-net
Workshop, April 2001, Belfast, Northern Ireland
18
Long-term Performance Aspects
Processes that might impair the long-term
performance of PRBs

• Coatings on the particle surface of the reactive
  material by
• ? precipitation of secondary minerals
• ? corrosion ("rust")

• Clogging of the pore space between the particles
  by
• ? precipitation of secondary minerals
• ? gas formation (H2)
• ? Biomass production

• Consumption of the reactivity by
• ? arriving at the material's sorption capacity
• ? dissolution of the reactive material

19
Zero-valent Iron (Fe0) Walls
granular Fe0
foamed Fe0 aggregates
Organic contaminants abiotic reductive
degradation of chlorinated hydrocarbons (e.g.,
PCE, TCE, VC) Inorganic contaminants abiotic
reductive immobilisation of heavy metals and
others (e.g., Cr, U, Mo, Tc, As, NO3). Costs 200
- 400 /t
20
Zero-valent Iron (Fe0) Walls
Results of column tests conducted using
commercial iron and groundwater from a
contaminant plume at an industrial site. PCE
dechlorination, formation of cDCE, and subsequent
cDCE degradation.
Source Gillham O'Hannesin, 1994
21
Zero-valent Iron (Fe0) Walls
Degradation of chlorinated hydrocarbons
Electron transfer from Fe0 surface (oxidation) to
the chlorinated hydrocarbon (reduction,
dehalogenation) 2Fe0 ? 2Fe2 4e- 3H2O ? 3H
3OH- 2H 2e- ? H2 X-Cl H 2e- ? X-H
Cl- 2Fe0 3H2O X-Cl ? 2Fe2 3OH- H2
X-H Cl-
22
Zero-valent Iron (Fe0) Walls
Uranium
Molybdenum
Removal of uranium and molybdenum from
contaminated groundwater in porous Fe0 aggregates
of a PRB system (Durango uranium mill tailings,
Colorado, USA).
Source http//www.doegjpo.com/perm-barr/index.htm
23
Zero-valent Iron (Fe0) Walls
Reductive immobilisation of heavy metals
Reduction of mobile and oxidised metal compounds
followed by mineral precipitation
Chromium Fe0 ? Fe2 2e- 2H2O ? 2H
2OH- 2H 2e- ? H2 Fe0 ? Fe3
3e- Cr(VI)O42- 4H2O 3e- ? Cr(III)(OH)3
5OH- Fe0 Cr(VI)O42- 4H2O ? Fe(III)Cr(III)(OH
)6 2OH-
24
Zero-valent Iron (Fe0) Walls
Coatings
Coatings might block access to the reactive
surfaces. Further precipitation blocks the pore
spaces between some iron particles increa-sing
flow velocity and decrea-sing the residence time.
Source Powell Associates Science Services
http//www.powellassociates.com/
25
Zero-valent Iron (Fe0) Walls
Iron corrosion
Anoxic Fe0 ? Fe2 2e- 2H2O ? 2H 2OH- 2H
2e- ? H2 Fe0 2H2O ? Fe2 H2
2OH- Oxic Fe0 ? Fe2 2e- H2O ? H OH- ½O2
2e- ? O2- Fe0 H2O ½O2 ? Fe2 2OH-
26
Zero-valent Iron (Fe0) Walls
Precipitation of secondary minerals
Carbonates HCO3- OH- ? CO32- H2O Fe2
CO32- ? FeCO3 (s) Ca2 CO32- ? CaCO3
(s) Iron minerals Fe2 2OH- ? Fe(OH)2
(s) 3Fe(OH)2 (s) ? Fe3O4 (s) 2H2O H2
Siderite
Calcite
Magnetite
27
Zero-valent Iron (Fe0) Walls
Iron geochemistry

• Stability fields for the system Fe-CO2-H2O with
  the following solid phases
• Am. iron hydroxide Fe(OH)3
• Siderite FeCO3
• Iron hydroxide Fe(OH)2
• Zero-valent iron Fe
• (25C, Fetotal 10-5 M, Ctotal 10-3 M, from
  Stumm Morgan 1996).

28
Zero-valent Iron (Fe0) Walls
Clogging
Carbonate, Ca and Fe concentration in
ground-water passing through a Fe0 wall. Obvious
precipitation of calcite and siderite, especially
in the upstream pea gravel (Denver Federal
Center, Denver, USA).
Source McMahon, P.B., Dennehy, K.F. Sandstrom,
M.W. (1999), Ground Water, 37, 396-404.
29
Zero-valent Iron (Fe0) Walls
Carbonate precipitation
Carbonate concentrations in the zero-valent iron
filling of a Fe0 wall (industrial site
contaminated by chlorinated hydrocarbons, New
York, USA).
Source Vogan, J.L. et al. (2000), J. Haz. Mat.,
68, 97-108.
30
Zero-valent Iron (Fe0) Walls
Silicon dioxide
Distribution of dissolved silicon dioxide in a
Fe0 wall (Moffett Naval Station, Mountain View,
CA).
Source Gavaskar et al. (2000)
31
Zero-valent Iron (Fe0) Walls
Consumption
Dissolved iron with pH in Fe0 column experiments
(ZVI) Clear dissolution of iron, but only
relevant at pH values lt 7.
Source U.S. Department of Energy Grand Junction
Office (GJO) http//www.doegjpo.com/perm-barr/
32
Zero-valent Iron (Fe0) Walls
Groundwater constituents

• Decrease of concentration in the wall
• Ca, Mg, Si, bicarbonate, sulphate, H
• Showing some influence on the reaction kinetics
  (corrosion, dehalogenation)
• Bicarbonate, sulphate, nitrate, phosphate,
  chloride, dissolved oxygen

33
Zero-valent Iron (Fe0) Walls
Mass balancing
Precipitation in a Fe0 wall, Copenhagen, Denmark
(Kiilerich et al., 2000) 13,3 kg iron
hydroxides, 2,7 kg CaCO3, 2,7 kg FeCO3 and 0,8 kg
FeS per 1000 kg iron filling per year
Loss of porosity in a Fe0 wall, Denver Federal
Center, Denver, USA (McMahon et al.,
1999) 0,35  of total porosity per year
(calculated only for the assumed precipitation of
calcite and siderite)
34
Activated Carbon

• Activated carbon
• Adsorption of organic contaminants
• Specific surface approx. 1000 m2/g
• Granular
• Reaction kinetics Diffusion controlled
• ? Critical parameter contact time!

35
Activated Carbon
Retardation
Retardation factor f(c) adsorption isotherm
(linear, Freundlich, Langmuir) va groundwater
flow velocity vS contaminant transport velocity
PAH R gt 3000 (Schad Grathwohl,
1998) Trichloroethene R ? 5000 -
20000 Chlorobenzene R ? 10000 - 20000 (Köber et
al., 2001)
36
Activated Carbon
Maximum barrier life-time estimation
d reactive wall thickness va groundwater
flow velocity R retardation factor
Horizontal flow through an activated carbon
reactor of 1,8 m diameter with a flow velocity of
0,5 m/d and a retardation factor of R 3000
maximum life-time 30 years
37
Activated Carbon
Factors influencing barrier life-time

• Groundwater composition
• Competition effects Natural groundwater
  constituents and contaminants compete for the
  adsorption sites
• Precipitation of secondary minerals Coatings
  block the access to the particle surfaces and
  alter the reaction kinetics
• Formation of biomass
• Negative effect clogging of the free pore space
• Positive effect biological degradation of sorbed
  contaminants possible

38
PRB Construction
39
Karlsruhe, Germany
40
Monitoring
Targets Validation of Performance Longevity
41
Monitoring
Longevity

• Checking of hydraulics
• Checking groundwater chemistry
• Hydrochemical parameters pH, electr.
  conductivity
• cations Ca2, Mg2, Fet,
• anions HCO3-, SO42-, Cl-, PO42-, NO3-
• Investigation of the reactive material
• Coring carbonate, XRD, REM

42
Current Research
Focus of current RD

• Selection of appropriate materials and processes
  for selective and efficient removal of
  groundwater pollutants.

• Evaluation of longevity and long-term
  performance development of models.

• Upscaling applicability and transfer of
  lab-scale results into the field

• Hydraulics of PRBs.

43
Current Research Tri-Agency-Initiative
Tri-Agency Initiative, USA
44
Current RD
Reaktionswände und -barrieren im
Netz-werkverbund (RUBIN), BMBF, Germany

• PRB projects co-operating in a network (RUBIN)
• Launched May 2000, 3 years
• Financial means ca. 4 Mill. Euro.
• Coordination University of Applied Sciences
  (Prof. H. Burmeier, Dr. V. Birke, Dipl.-Ing. D.
  Rosenau)
• 11 projects
• 8 projects dealing with design, erection and
  operation of pilot- or full-scale PRBs in Germany
  and/or important general preparatory RD work
• 3 projects addressing general issues and missions.

45
Conclusions
PRB long-term behaviour is a function of the
deployed reactive material.
PRB longevity is influenced by the pollutants to
be treated and the groundwater ingredients, i.e.,
groundwater chemistry.
The main groundwater components reveal a
specific, important influence predominantly due
to their higher concentrations compared to the
pollutants concentrations.
Surface reactions at the reactive material cause
significant changes in geochemical conditions
(pH, Eh) regarding pore space that is passed by
groundwater and therefore hydrochemical changes
in the composition of the groundwater.
46
Conclusions
Mineral formation (coatings), alteration of
surfaces, gas evolution and biomass can influence
reactivity and permeability of a PRB.
Alteration of surfaces and mineral formation can
be mostly observed directly upgradient of a PRB.
However, only pertaining to a few cases,
detrimental effects regarding efficiency of the
PRB have been observed so far.
Geochemical processes are predominantly
well-known and well understood. However,
quantitative approaches for long-term
behaviour/performance are still lacking. Current
RD projects address these issues.