Index > 4 Environmental remediation of radioactively contaminated sites >

4.5.1 Barriers (CT)

Contents
4.5.1.1 Subsurface barriers
4.5.1.2 Bored piles
4.5.1.3 Slurry walls or trenches
4.5.1.4 Keyed rammed piles
4.5.1.5 Sheet piles
4.5.1.6 Injection walls
4.5.1.7 Injection curtains
4.5.1.8 Ground freezing
4.5.1.9 Permeable reactive barriers

4.5.1.1 Subsurface barriers

Underground containment barriers are an important method for limiting or preventing the movement of radiological and non-radiological contaminants into the surrounding geological media and groundwater. In the past, containment has been used primarily at sites where there was no other efficient and cost effective option. However, subsurface barriers can be used in any number of situations where it is necessary to prevent the migration of contamination. Barriers are currently used, for instance as an interim step, while final remediation alternatives are being developed (or considered) in conjunction with other treatment techniques, e.g., reactive barriers. In many instances subsurface barriers are capable of effectively confining the contaminant for extended time periods in a cost effective way.

There are many subsurface barrier technologies commercially available and others are at various stages of development. The purpose and function of the containment system must be determined prior to designing and constructing the barrier. Site characterization is an essential part of choosing an appropriate barrier. Some of the factors that may need to be considered when designing a subsurface barrier are [IAEA-2006b]:

  • It is important to establish the barrier placement criteria, including location, depth and thickness.
  • A stress-deformation analysis needs to be performed on the surrounding area in order to assess the potential impacts of barrier construction.
  • A compatibility test needs to be performed to select the most effective barrier materials and, when necessary, appropriate mixture combinations.
  • It is necessary to determine the most effective and feasible construction methods.
  • Construction quality assurance/quality control is a crucial component of subsurface barrier emplacement.

Different types of subsurface barrier have different construction quality assurance criteria; however, there are two primary concerns. First, the installed barrier must have a hydraulic conductivity equal to or less than that specified in the design. The second concern is barrier continuity, which is difficult to assess; the methods available have had varying degrees of success. There is currently no method of guaranteeing the continuity of a subsurface barrier [IAEA-2006b]. Discontinuities may occur during grout application/installation and joint formation. Cracking due to curing, settling or wet/dry cycling may occur over time. Proper emplacement of a subsurface barrier is critical in ensuring the overall effectiveness of the containment system. Once a barrier has been installed, verification and monitoring are crucial. At this time, there is no uniform method for monitoring the emplacement, long term performance or integrity of the barrier.
The construction of subsurface barriers can be grouped into three basic technologies:

  1. Replacement of excavated materials with materials of lower permeability;
  2. Displacement with materials of lower permeability;
  3. Reduction of the permeability of the soil (Figure 4.5).

Impermeable liners made with clays or cement and clay mixtures are widely used in the construction of new landfills. Clay is subject to chemical attack by leachates from the waste material that can degrade the barrier and lead to increased infiltration and contaminant dispersal. Proper moisture content must be maintained to prevent shrinkage cracks in the clay. The development of new barrier concepts, materials and construction techniques is in the process of overcoming these deficiencies, however. The long term stability and effectiveness of new synthetic binders and polymers as sealants is being evaluated. Inorganic grouts are also being studied for use with or without clays.

Figure 4.5 Various containment constructions (top view)
Figure 4.5 Various containment constructions (top view)

4.5.1.2 Bored piles

Bored piles are a series of overlapping large size boreholes displayed in Figure 4.5 part (a). Rotary drilling equipment, soil mixers or line shaft excavators may be used. The boreholes are backfilled with a cement grout or concrete before the next hole in the row is drilled. Depending on the cement and aggregate used, nearly complete sealing can be achieved. Depths of several tens of metres can be reached. In principle the technique can be applied to many types of soil and rock, but the cost increases with the hardness of the material. Very inhomogeneous soils, containing boulders for instance, may prevent successful application. The technique may be combined with that of slurry walls.

4.5.1.3 Slurry walls or trenches

Slurry walls or trenches as displayed in Figure 4.5 part (b) are constructed by excavating a vertical trench around waste areas to a depth that is at or below the bottom elevations of contaminated soil or waste materials. Trench stability is maintained by placing a liquid slurry of bentonite and water in the trench as excavation progresses. When the trench reaches the proposed maximum depth, the slurry is displaced from the bottom upwards with a dense barrier material consisting of soil bentonite, cement grout, polymers, plastic concrete or other low permeability materials. Using a continuous trenching construction method (see also Figure 4.8), cavities for slurry walls can be continuously excavated with a backhoe or excavator, filled with slurry, and backfilled with low permeability material until the waste disposal areas are completely encircled. Slurry walls can be excavated to depths of more than 30 m and can have permeabilities as low as 10–8 to 10–9 m.s–1.
This technique is easiest to apply in sand and gravel formations and to a certain extent to cohesive materials, such as clays. It is more difficult to implement in hard rocks. Amendments can be added to the injected grouts that will act as additional sorbents for contaminants such as heavy metals and radionuclides. Slurry walls may also be combined with a plastic membrane to form combination walls displayed in Figure 4.5 part ( c ).

4.5.1.4 Keyed rammed piles

Prefabricated concrete piles may be rammed into the ground using a pile driver. In order to ensure water tightness, they are interlocked with slots and keys, see Figure 4.5 part (d). The applicability of this technique is largely restricted to unconsolidated or weakly consolidated sediments without large boulders.

4.5.1.5 Sheet piles

Sheet piling consists of vertical cut-off walls constructed by driving strips of steel, precast concrete, aluminium or wood into the soil. Sheet metal piling, which are corrugated sheets of iron that are shaped in such a way that they interlock (Figure 4.5 part (e)) with sealable joints, is commonly used. Interlocking sheets are assembled before installation and driven or vibrated into the ground by about a metre at a time until the desired depth is achieved. Sheets are sealed by injecting grout into the joints between the metal sheet piles. Continuous sheet piling walls can potentially be driven to depths of some 90 m in unconsolidated deposits lacking boulders. Bulk hydraulic conductivities of 10–8 to 10–10 cm.s–1 have been achieved in test cells constructed of joint sealed sheet piles.

4.5.1.6 Injection walls

An I-shaped pile is driven into the ground and upon extraction the remaining hollow space is backfilled with a bentonite or cement-bentonite slurry, see Figure 4.5 part (f). Each section overlaps with the preceding one to provide good keying in and water tightness.

4.5.1.7 Injection curtains

Injection curtains are constructed by pushing hollow injection tubes into the ground (unconsolidated materials) or by drilling injection boreholes (rocks), see Figure 4.5 part (g). A variety of inorganic and organic grouts may be injected to fill the pore space of the soils or rocks. Typical inorganic grouts are ordinary Portland cement (OPC), bentonite and water glass. The organic grouts used in civil engineering applications include polymers of methacrylate and epoxy resin. The possible interaction of organic grouts with organic contaminants has to be carefully studied before application, as the contaminants may lead to a dissolution or breakdown of the sealing components, or may prevent polymerization. The technique, in principle, is applicable to all types of soils and rocks. The sealing success depends very much on the homogeneity of the permeability distribution. Preferential pathways may lead to incomplete sealing. Some geological formations may have a too low permeability for injection, but still provide long term migration pathways. In such a case hydro-fracturing allows successful creation of injection curtains. To this end, sand, zirconia or other high strength spherical materials are injected under very high pressure to ‘fracture’ the rocks. Spherical materials stabilize the open fracture while providing a high permeability infill that allows injection of the actual grout. In addition to providing a hydraulic sealant, injected grouts can also act as sorbents for contaminants. This effect may be less effective for organic contaminants than for metals, including radionuclides.

A variant of injection curtains is the injection of non-miscible fluids with the intention to reduce water permeability. In recent times the effect on water permeability of injecting bio-degradable oils has been explored [IAEA-2006b].

4.5.1.8 Ground freezing

Temporary containment can be achieved by a variety of measures, including grouting and ground freezing. Either an impermeable screen around a contamination can be established or the contaminated material itself can be frozen in order to facilitate its handling or excavation. Artificial ground freezing (AGF) has been used for over 100 years to form impermeable barriers and temporary support for excavations, shafts and tunnels [IAEA-2006b]. Techniques such as grouting and artificial ground freezing are standard in civil engineering and mining for stabilizing, for instance, highly saturated soils or creating impermeable walls for tunnelling purposes. They are also used when constructing foundations below the groundwater table.

Laboratory studies have shown that frozen soil barriers with very low hydraulic conductivities (< 4 × 10–12 m.s–1) can be formed under saturated soil conditions. The formation of a frozen soil barrier in arid conditions will require a suitable method for homogeneously adding moisture to the soils to achieve saturated conditions. Formation of frozen soil barriers in areas where plumes of low freezing point contaminants (tri-chloro-ethylene, etc.) exist may require low temperature and more expensive cryogenics (e.g., liquid nitrogen and CO2) [IAEA-2006b].

Freezing is effected by a system of pipes that are inserted into or around the contaminated zone (Figure 4.6). A cooling liquid (brine) is circulated (a one phase system) in this pipe system. Another option is an open two phase process whereby liquid nitrogen is pumped into the ground. The N2 vaporizes and thereby extracts the heat from the soil. Thermosyphons forming a closed two phase system are an alternative. The working fluid is contained in a closed sealed vessel (a thermopile or thermoprobe) that is partially buried. Thermosyphons can function passively in cold climates during the winter months, at which time the above ground portion is subjected to cold ambient air that cools and condenses the working fluid. The condensed fluid gravitates to the below ground portion. Below ground, subjected to warmer temperatures, the working fluid warms, vaporizes and rises upwards to repeat the cycle. A closed two phase system can also be used in an active mode and is applicable when the ambient air temperature is above freezing [IAEA-2006b]. Such systems utilize ‘hybrid thermosyphons’. A typical system consists of multiple thermoprobes, an active (powered) compressor and condenser, an interconnecting supply and return piping network, and a control system. Thermoprobes consist of an evaporator and a passive condenser section. The hybrid system can function simultaneously in both passive and active modes when the ambient temperatures are sufficiently low, thereby reducing energy costs. Hybrid thermosyphons may operate in northern climates (locations that experience air temperatures below the target soil temperature) without external power. The temperature of the barrier can be adjusted to ensure the necessary liquid-solid phase change even though contaminants may lower the phase change temperature.

Figure 4.6 The principle of ground freezing as a barrier and to immobilize contaminants
Figure 4.6 The principle of ground freezing as a barrier and to immobilize contaminants

4.5.1.9 Permeable reactive barriers

The use of permeable reactive barriers or walls is distinguished from outright containment by the fact that the contaminant carrier as such (i.e., the groundwater) is not prevented from movement [IAEA-2004b]. The objective is rather to remove the contaminants from the mobile phase. Permeable reactive barriers are installed by excavating a portion of the aquifer, disposing of the excavated material and replacing it with a permeable material designed to react with the contaminant and remove it from the flowing water (Figure 4.7). The advantages over pump and treat systems are that no active pumping or process operation and maintenance is required, thus reducing energy and operation and maintenance costs, no treatment sludge’s are produced, thus reducing waste disposal costs, and no surface facility is required, which allows the land to be returned to productive use. The systems typically rely on the natural gradient of the groundwater table as the driving force. The barrier material must be designed to remain reactive for periods of many years to decades. Furthermore, the barrier permeability must be sustained throughout the duration of the groundwater treatment. The performance of permeable reactive barrier systems must therefore be monitored so that corrective action can be taken when required.

Permeable reactive barriers have been designed and implemented for the treatment of dissolved metals, acid mine drainage, radionuclides and dissolved nutrients. Contaminant removal can be effected in a variety of ways. Treatment processes include adsorption, simple precipitation, adsorptive precipitation, reductive precipitation and biologically mediated transformations [IAEA-2004b].

Changing the redox state can be a very effective method of immobilizing certain radionuclides (e.g., uranium and technetium). These radionuclides have two or more oxidation states, and the more reduced oxidation states are less mobile; for example, reduction of the hexavalent uranyl ion UO22+ to the tetravalent U(IV) state results in the precipitation of sparingly soluble precipitates, including UO2(s) or mixed U(VI)–U(IV). Zero valent iron is an abundant and inexpensive reducing agent that has been observed to reduce and precipitate uranium and technetium in laboratory studies [IAEA-2004b].

The oxidation products generated (e.g., ferric hydroxides) can provide a high capacity sorption substrate also for non-redox sensitive species, but their long term stability in relation to changes in redox conditions has to be carefully evaluated [IAEA-2004b].

Permeable reactive barrier systems containing zero valent iron have been installed for the treatment of uranium, technetium and other metals; these barriers demonstrate excellent removal of uranium and technetium. Examination of the reaction products has been conducted at a series of sites of permeable reactive barriers. Although the results of these characterization studies are inconsistent, all the reports indicate that a portion of the uranium entering the barrier system is reduced to U(IV), whereas some portion may remain in the U(VI) oxidation state. Other metals commonly associated with uranium mine waste, including arsenic, molybdenum, selenium, vanadium and zinc, are also removed from the groundwater, possibly as reduced phases (e.g., V2O3) or as sulphide minerals (As2S3, ZnS) [IAEA-2004b].

Figure 4.7 Sketch of a permeable reactive wall in combination with a Funnel-and-Gate system
Figure 4.7 Sketch of a permeable reactive wall in combination with a Funnel-and-Gate system

Organic reductants, such as sawdust, have also been used to promote the reduction and precipitation of uranium. Passive treatment systems containing organic carbon have been used to remove both uranium and nitrate from groundwater at sites where these two constituents coexist as a result of releases from nuclear weapon production facilities [IAEA-2004b].
Sorption can remove contaminants from groundwater and can maintain low concentrations of radionuclides. Sorptive materials that have been evaluated or deployed in permeable reactive barrier systems for treating radionuclides include zeolites (e.g., clinoptilolite), phosphate based adsorbents (e.g., bone char apatite and Apatite II) and hydrous ferric oxides (e.g., amorphous ferric oxy-hydroxide (AFO)) [IAEA-2004b].

The majority of the reactive barriers installed to date have been continuous barriers installed across the entire width of the plume. Contaminant fluxes also can be focused on the reactive barrier by an array of non-reactive barriers, such as slit or slurry walls, to form a Funnel-and-Gate system [IAEA-2004b].

Funnel-and-Gate systems reduce the physical length of the treatment portion of the barrier and prevent contaminants from circumflowing the treatment zone. The volume of reactive material required to treat contaminated groundwater is determined by the contaminant concentrations, groundwater geochemistry and flow rate. For many contaminant plumes, the volume of reactive material will be similar, whether a continuous barrier or Funnel-and-Gate configuration is employed. Since the installation of continuous barriers is typically less expensive than that of Funnel-and-Gate systems, this installation technique has been preferred. Furthermore, because Funnel-and-Gate installations focus the flow to across a small cross-sectional area, there is greater potential for clogging by the formation of secondary precipitates.

Depending on the reactive material to be used, deployment techniques may include injection wells (for grouts, gels and soluble reactants) or trenches cut by a suitable excavator (for grouts and particulate material such as granular iron, sawdust, etc.), see Figure 4.8.

Figure 4.8 (a) Continuous trenching machine used to install a 46 m long, 7.3 m deep and 0.6 m wide granular iron permeable reactive barrier  [IAEA-2004b]
Figure 4.8 (a) Continuous trenching machine used to install a 46 m long, 7.3 m deep and 0.6 m wide granular iron permeable reactive barrier [IAEA-2004b]
Figure 4.8 (b) Simultaneous excavation and replacement of aquifer material with granular iron as the horizontal trencher advances [IAEA-2004b]
Figure 4.8 (b) Simultaneous excavation and replacement of aquifer material with granular iron as the horizontal trencher advances [IAEA-2004b]

Development work on efficient methods to emplace reactive barriers with minimal disturbance, even in awkward places, is ongoing. Adaptation of more novel civil engineering techniques, such as directional or horizontal drilling, the use of guar gum slurries for barrier installation, hydraulic fracturing and jet grouting techniques, can be used for the emplacement of barriers at depths beyond the capabilities of the conventional excavation techniques displayed in Figure 4.9 [IAEA-2004b].

Computer simulations conducted using reactive solute transport models can be used to determine design parameters for barrier installation, to predict the potential for barrier clogging and to assess the potential benefits of barrier performance. The performance of a reactive barrier installed at the Elizabeth City US Coast Guard Support Centre was simulated using the reactive solute transport model MIN3P. Comparison of the simulation results with subsequent measurements showed good agreement in Figure 4.10. The performance of the permeable reactive barrier installed at Monticello Canyon, Utah, USA, was simulated using the PHREEQC model [IAEA-2004b].

Figure 4.9 The principle of directional drilling and grouting
Figure 4.9 The principle of directional drilling and grouting
Figure 4.10a
Figure 4.10a
Figure 4.10b
Figure 4.10b

Figures 10a and 10b are showing simulated heterogeneous reactions at a permeable reactive barrier at the Elizabeth City US Coast Guard Support Centre [IAEA-2004].

The limitations on permeable reactive barrier performance and lifespan include constraints on the reactive material longevity and the barrier permeability. Of these concerns, the potential for barrier clogging and the permeable reactive barrier evolving into an impermeable reactive barrier is the most significant. Since the total mass of contaminant that accumulates in the barrier is modest, the principal precipitates resulting in clogging are the products of reactions between the barrier material and the major ions present in the water, or between the barrier material and the water itself. The use of zero valent iron (Fe0), the most commonly used reactive material, results in the reduction of water and an increase in the pH to between pH10 and pH11. This increase in pH favours the precipitation of carbonate minerals, principally calcite (CaCO3) and siderite (FeCO3). Over periods of several years to decades, the accumulation of these precipitates potentially may be sufficient to reduce the pore space of the reactive material and limit barrier permeability. Reactive barrier technology has evolved recently, and the oldest barriers are now approaching ten years of operation. Clogging to a degree that is sufficient to impair barrier performance has yet to be observed, although long term monitoring programmes are required to assess this concern.

The long term fate of the reactive barrier after remediation is complete or after the barrier becomes ineffective depends on the nature of the contaminant and on the characteristics of the barrier. Concerns include the potential for remobilization of contaminants retained in the barrier and the potential for clogging in the barrier to alter natural groundwater flow conditions. In many barrier systems, the contaminant is converted to a form that is stable in the geochemical environment that prevails in the aquifer. Furthermore, because the mass of contaminant is small relative to the mass of the barrier material, the residual barrier material may be classified as non-hazardous. In these systems, it may be acceptable for the barrier to remain in place. In other cases, the mass of contaminant may exceed soil guidelines, the contaminant may have the potential for remobilization or the contaminant may be sufficiently hazardous to warrant excavation of the reactive material and placement in a secure waste disposal facility. In these cases, excavation of the barrier, or a portion of the barrier, may be required.

Although considerable research on the performance of reactive walls is continuing worldwide, some techniques have reached commercial maturity [IAEA-2004b].