Cover Image


View/Hide Left Panel
Click for next page ( 143

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 142
Self-Hardening Slurries and Stable Grouts from Cement-Bentonite to IMPERMIX~ Gilbert Tallard, President Env~roTrench Co., Pelham N.Y. ABSTRACT Conventional concretes and grouts, with which most engineers are familiar, are materials with a water/cement ratio generally below 0.5. The strength of these materials is In the tens of MPa (thousands of psi). The structural strength of these materials is their primary quality. A less familiar group of materials, particularly in the United States, consists of very watery slurries that set to form a solid of relatively low strength ranging from 0.1 to 3.5 MPa (15-500 psi). The water content of such materials may vary from 250 to over 500 percent, with a water/cement ratio as high as 10 or more. At these water content levels, in order to have a homogenous matenal, it is necessary to suspend the cement with a viscosifier showing thixotropic properties. The viscosifier is traditionally some kind of clay, generally bentonite. The notion of stabilizing a cement suspension with a tixothropic clay has made stable grouts possible. The degree of stability of such grouts is indicated by the amount of bleed or free water at set time. The essential purpose of these materials is to provide an economical means of controlling ground-water migration in the ground, be it soil or rock. Whether such materials are called a slurry or a stable grout depends on the application. A _ ~ ~ grout is used when the application is localized; a slurry is used when the application is in an open excavation, although the term slurry grout, meaning watery stable grout, is often used. The application field is below the ground water table and slightly above. Given the high water content, the dehydration and destruction of the hardened slurry material would occur if exposed to the elements, without any kind of protection. Conserved in water, these materials have an infinite life. Originating in natural ground-water seepage control in the dam construction and repair technology, the control of contaminated ground-water migration called for similar technology. The chemistry of the environment and the service under which these materials are to perform has brought forth certain shortcomings. The strongest limitation has been the lack of watertightness of mixes using Ordinary Portland Cement with respect to the regulatory conductivity threshold of 10-7 cm/s. Bentonite clay loses some of its properties in the presence of calcium in the cement, and as a highly dispersive clay, it is subject to shrinkage in the presence of certain organic chemicals. in view of these limitations, a different pair of constituents has been found and perfonns surprisingly well, from both chemical and permeability standpoints. Named ~ERMIX~, this flexible formulation is based on attapulgite or sepiolite clays and finely ground blast furnace slag cement. Slac cement's long setting time is taken advantage ot In some applications like slurry trenching in difficult ground, or shortened when strength buildup is necessary, as in sludge solidification. Given the low viscosity and low-solids content of these suspensions, a very penetrating environmental grout for sealing contaminated fractured rock can be found in IMPERMIX~ Whereas cement-bentonite slurries have found limited applications in environmental remediation, due to excessive permeability, the permeability threshold of self-hardening slurries has been lowered with IMPERMIX'~' by one and two orders of magnitude and occasionally more. With the fact that an I~ER~X'i' formulation can be tailored to satisfy a specific chemical condition, it is now possible to engineer and control, with a high degree of quality assurance, the self-hardening slurry application in site-specif~c conditions. Contaminated ground-water barriers have been created by D-124

OCR for page 142
APPENDIX~PAPERS PRESElITED D-125 the one-phase slurry trenching technique using backhoe and clamshell tools, the injection vibrating beam method, and jet grouting. Future use for ~n-situ fixation by soil mixing, pipe line abandonment, contained sludge in-situ solidification, and horizontal jetted barriers are a few fixture obvious applications. Solidifying low-level radioactive aqueous waste is a more engaging prospect. Definitions necessarily stable. Slurp: suspension of solids in a large amount of a liquid generally water-and not Thixotropy: a physical phenomenon occuning when a liquid builds up rigidity at rest, while reversing to a low-viscosity fluid in a dye namic condition; such liquids are also known as shear sensitive. Mineral grout: suspension of mineral powders in water, not necessarily stable, and containing at least one cementitious ingredient to create an eventual set after injection into . recelvmg grounc . Stable grout: suspension of at least one clay mineral and one cementitious powder capable of developing sufficient viscosity to maintain all particles in suspension with a minimum of bleed water. Self-hardening slurry: a cliluted, stable grout used in a one-step construction process in which the slurry in liquid phase is the slurry stabilizing the walls of a trench and in a set-~n-place solid phase providing the desired end product. Bentonite: a highly dispersive montmorillonite clay used extensively for viscosifying water-based drilling or trenching fluids, and for the preparation of common stable grouts; viscosities by adsorbing water between its structural platelets; high ion-exchange capability. Attapu1fgite: a clay mineral of the palygorskite family with a needle-like crystalline structure; viscosification of water occurs by mechanically stacking the needles and trapping water between them; very little ion-exchange capability by comparison to bentonite. Cement: common term for Ordinary Portland Cement, the most common cementitious binder used in the world; sets by forming at least three distinct chemical compounds and some noncomb~ned salts. Slag cement: finely ground blast furnace slag cement is a hydraulic binder manufactured from a residue of modern steel making; molten slag is blown into a cold-water spray, with the slag cooling into glassy pellets; once finely ground In a cement mill, an industrial pozzolanic material is obtained. Cement-bentonite (CBJ: a generic mixture of Portland cement and bentonite, with a solids content of generally less than 35 percent, mixed into a stable grout and used for alluvium and fractured rock grouting as well as a self-hardening slurry for cutoff wall construction.

OCR for page 142
D-126 BARRIER TECHNOLOGIES FOR ENVIRONMENTAL MANAGEMENT IMPERMI/: a proprietary mixture of attapulgite clay and blast furnace slag cement and other ingredients used as a self-hardening slurry, typically at a maximum solids content of 22 percent and exhibiting very low permeability, high strength, and very good stability under chemical attack. INTRODUCTION Chronologically, self-hardening slurries have their origin in stable grouts. Barrier technology, as understood in pure geotechnical construction, consists of controlling ground-water flow under or through a man-made structure. The technology started with grout curtains and cutoff walls having application to embankment dam construction and repair. In the waterproofing of porous granular soils, stable grouts do a very good job in sealing these formations, by comparison with neat cement grouts. Cement-bentonite (CB) grouts have been used extensively in alluvium grouting around the world, except in the United States. The best permeability to be expected for the grouted soil is in the 10-5 cm/s range. However, with grout cutoffs up to 25-feet thick, a rather low gradient is at play, and the safety of the system can be seen as higher than in the case of the thin cement-bentonite cutoff wall subject to a much higher gradient. This construction method was developed in the 1970's, when tools and evolution in the rheology of the self-hardening slurries permitted the construction of cutoff walls 2-feet thick and over 100-feet deep. LIMITATIONS OF CEMENT-BENTONITE For better understanding, economy, and quality assurance, design engineers generally prefer trenched cutoff walls to grout curtains. When designing water reservoir embankments or contemplating the repair of a leaky earth dam, a conductivity (K) in the range of 10-6 cm/s is quite c~ti~f~t~rv in rerlllcin~ the residual seepage to an acceptable value concurrently achieving a total ~ ~ . ~ ~ ~ 1 ~ _ _ A_ __1~ 1__ 1~ 1,.~ ~ Ill ;~11~1;- thy cessation of Internal erosion. c;emem-oemomle selI-n=~eIluly my Am wale ~l~"ll~" ". ills United States since the mid-1970's for the purpose of impeding the migration of a contaminated Knifer had to get a variance from regulatory agencies with respect to the coefficient of ~ . ~ . ~ 1 _ 1 ~ 1 _ ~ ~ permeability above the norm as established by the sol1-0entomte slurry Irencn oenon mark. oo111t; more stringent agencies, such as the California Water Board, require both compatibility and a permeability coefficient of less than 10-7 cm/s, with no exception. In this context, 10 years ago the Los Angeles County Sanitary District was looking to use both self-hardening slurry technology and to meet the K < 10-/ cm/s criterion for their numerous barriers to be installed across canyons and gullies confining its landfills. This writer was hired to formulate a cement-bentonite slurry that could perform reliably in the field and provide a low permeability. After a few months of testing, a cement-bentonite slurry prepared with 4.5 percent bentonite, 25 percent Portland cement, and 15 percent fly ash fluidified and retarded with an amount of lignosulfonate satisfied the criteria. Permeabilities at 60 days in the low 10-8 cm/s range were achieved. Practically, nearly a doubling of ingredient solids over a normal formulation were called for to reduce the permeability by one order of magnitude. The added cost for materials and preparation on-site was quite significant. At these solid levels, the cost for fluidifier/retarder is equal to that of the bentonite. The recourse to a high dosage of fluidifier for viscosity control can lead (and did) to defects by excess dosage, causing

OCR for page 142
APPENDIX~PAPERS PRESENTED D-127 suspended soil particles to settle in discrete bedding patterns of higher permeability. This solution could not be generalized to the uncontrolled hazardous waste remediation industry at large. MATERIALS As a combination, Wyoming bentonite, which is highly dispersive and prone to ion exchange, and Portland cement, which exhibits an early false set and some leachability, are not the best partners for creating a Tow-permeability, high-water-content, self-hardening slurry matenal. Hydrated bentonite will readily flocculate at the first encounter of a little amount of Portland cement. The flocculation causes a drastic change in the filtration characteristics of bentonite slurry, going from as low as ~ 3 cm3 on an American Petroleum Institute (APT) standard test ( 100 psi for 30 mini to over 80 cm3. Once the cement is mixed into the bentonite slurry, the filtrate, tinder API standard, becomes total (250-300 cow. It is clear that a head differential of 210 feet of water is unrealistic for most projects and the API standard test should be seen only as a Qualitative procedure. Concurrently with the flocculation and filtrate loss, a s~gn~cant increase of viscosity occurs. The Europeans, who are not blessed with our resources in Wyoming and South Dakota, have to permutate natural calcium bentonites into sodium bentonite. This industrial process allows the production of all kinds of bentonites and, in particular, clays much less sensitive to calcium ion exchange. Also, the number of cement types available in Europe is far greater than here, and certain blended cements offer a better match to bentonitic clays, as well as intrinsically higher chemical stability. CHEMISTRY When addressing hazardous waste barriers, the chemist of the contaminant can become a major factor in the barrier's durability. Particularly with certain volatile organic compounds, such as benzene, xylene, or methanol, the water adsorbed between the bentonite mineral platelets suffer a change in dielectric constant (whether in a soil bentonite or cement-bentonite configuration) and causes the bentonitic gel to contract by loss of dispersive energy. This causes an increase of conductivity that can be over an order of magnitude. The purpose of compatibility permeability testing under a higher than service gradient is to determine, in a few months, the degradation potential over the life of the barrier. When permanent enclosures are involved, a 50-year service life is far from overly ambitious. QUALITY CONTROL/QUALITY ASSURANCE This writer has always been adverse to the use of soil bentonite barriers for permanent enclosures. A coarse technology and a crude construction method cannot provide the level of quality control (QC) and quality assurance (QA) necessary to satisfy all parties having a vested interest in the enclosure. On the other hand, the self-harden~ng slurry approach to the construction of a barrier allows the sophistication of an engineered project: at the design stage, during construction QC stage, and at the post-installation QA stage. With cement-bentonite self-hardening slurries, only a number of sanitary landfill barriers could be satisfied with a K above ~ 0-7 cm/s with

OCR for page 142
D-128 BARRIER TECHNOLOGIES FOR ENVIRONMENTALMANAGEMENT no adverse consequences, which meant practically the end of the road for cement-bentonite In hazardous waste remediation in the 1980's, a profound disappointment to this wnter. ~ERMIX~ The history of IMPERMIX Starts with this frustration and builds on a multiplicity of past experiences and a taste for pluridiscipl~narity. As the steed mills of this country were closing In the 1970's, so were the sources of blended cement such as Portiand/ground blast-fumace slag-cement blends. Today, very few sources are producing a pure blast-furnace stag-cement product that is sold to the ready-mix concrete industry. A remarkable characteristic of hardened blast-furnace sTag- cement In very high water/cement ratio slurry is the strength, which is many fumes that of Portland cement mixed at the same ratio. In conventional concretes, the gain in strength is only 5-1S percent, with the permeability and resistance to salts improved noticeably. When, 10 years ago, this writer constructed his first soil attapulg~te slurry trench at a chemical facility on the Gulf of Mexico, actually using seawater to prepare the slurry, a good clay mate for the slag cement was recognized. A clay that viscosities mechanically instead of swelling and doesn't blink at the sight of electrolytes appeared to be a good candidate indeed. The concept of a controlled initial viscosity based on mixing energy was extremely appearing. Further research determined the range of interesting proportions. Surprisingly, a total of 15 percent solids by weight of water could produce a hardened slurry with a coefficient of Permeability less than ~o-7 COWS and a strength substantially above 100 psi. From this research, it- ~ ~ tin) . ~ . ~ ~ ~ 1 , , ~ 1 ~ ~ __ _ ~ _ _ ~ ~ 1 ~ ~ ~'ERMIX~ was born, and the base tormulahon was set at one part attapulglte to ~ pans OI Slag. At the present time, it is possible to target 10~9 cm/s. Any increase in stag content is matched to a certain increase in clay to assure stability and chemical balance. Any stag proportioning above 15 percent involves some structural design consideration. Once permeability to water was assured, compatibility with common pollutants was addressed. Organic chemicals whether aqueous or nonaqueous, heavy metals, strong alkalis and ~. . ~ ~ . . T T ~ ~ 1% ~ 1 1 _ _ 1 1 ~ ~ _ _ _ 1 ~ 1 ~ ~ ~ ~ ~ 4- 1^ ~ ~ acid solutions to a pti level or a, all caused a ~ong-~em1 aet;rt;~ 1[~ By we ~ degradation. Addition of fly ash, silica fume had no positive effect on the permeability. Addition of ground lime appeared to provide a positive decrease in permeability on well-cured samples, although initially retarding the set. Addition of filtrate reduction polymers proved very effective from a technical standpoint, although not always cost effective. Applications With a natural setting time of 70 hours (at room temperature, shorter in warm weather), the working hme of an IMPERMIX~ slurry is unusually long, allowing for the construction of practically jointless barriers. The long setting time coupled with high strength (for a self-hardening slurry, 150-250 psi) has allowed the construction of support of excavations where structural elements are placed and suspended inside the fluid trench until the slurry is hardened. A number of shoring systems can take advantage of this technique. The high fluidity of ~'ERMIX~ mixes, which have proven to be non-sh~ink, opens a market for backfilling abandoned utility ducts or sewers where only cellular concrete could do the job at a much higher cost. A means of accelerating the setting of I~'ERMIX~ has opened a new field of application for light-weight backfill material. IMI'ER=X~ is now being used to pretrench

OCR for page 142
APPENDIX~PAPERS PRESENTED D-129 along the alignments of the slurry walls supporting the excavation for the Central Artery In Boston. In this application, the IMPERMIX~ slurry is used as the trenching support slurry to minimize the size of the excavation required to remove obstacles such as granite blocks or Amber cribbings, inherited from the rich history of this New England capital city. The elimination of dewatering and shoring is a major simplification and a much safer procedure. In lieu of backfilling with a conventional lean mix, the modified IMPERMIX@~ sets in place overnight and hardens enough in 2 days to permit trenching of the slots for the guide wall construction ahead of the slurry wall panels excavation proper. ~ERMIX~'s fluidity is also an asset when using jet-grouting techniques to install a barrier under utilities and connecting to conventional trenched cutoffs or soil-displacement narrow walls. The potential for soil mixing in situ exists and has been demonstrated in the laboratory. The potential for sludge fixation also exists with the smallest increase in volume. One must bear in mind that THERMS may represent only 13 percent of solids; 87 percent is water (water content > 600 percent). It is perfectly conceivable to mix the contents of a leaky tank almost hill of contaminated water (e.g., low-radiation water) into an IMPERMIX~ grout that will not leach and to consider the problem solved satisfactorily over the long term, as long as the tank remains buned. IMPERM1X~ will only release water by drying. In a submerged condition, despite the already high water content, the water content of IMPERMIX~ tends to increase. Research Needs There is still much that needs to be learned about ~ERMIX~ before we start to understand why the performance of this combination of materials is so good. No electron microscope observation has been made, and no mercury permeation has been attempted to determine the system's porosity structure. An answer needs to be found to why IMPERMLX~ is capable of combining, after set, more water than its already extremely high water content. A low-gradient permeability test had water going in for a month with no water coming out at all and no change In volume. From a practical standpoint, when performing a compatibility test, the practice with materials having a substantial solids content is to percolate a multiple of the pore volume, the latter being defined by the ratio of dry weight and the initial sample weight. In the case of self-hardening slurries, this definition results in the pore volume being equal to the water content. With the very low permeability exhibited, it is clear that most of the water content is part of the structure and not of the pore space; therefore, the need arises to redefine "the pore volume" for realistic compatibility testing. A review of compatibility testing procedures in a flexible wall biaxial cell permeameter should be made to compare the typical method using unrealistically high gradients (50-200) through fairly thick samples (4 inch) and methods using low gradients through a large diameter and thin sample in an oversized biaxial permeameter cell (Haug, 19801. A new test for very low- pe~meability materials should be developed to establish threshold gradients below which the matenal can be considered truly impermeable. Permeability is not the only concern when dealing with hazardous waste. Sorbtive characteristics and electric diffusion are areas that deserve investigation. Sorbtive properties of attapulgite clays are well known, and slag cement/lime combinations have been tested successfully for chromium. No evaluation of ~ER~X~ sorbtive properties has been undertaken yet, nor has diffusion been studied.

OCR for page 142
D-130 BARRIER TECHNOLOGIES FOR ENVIRONMENTAL MANAGEMENT A better understanding of the material will permit the introduction of specific ingredients to improve chemical retention. This writer has started investigating the benefits of adding activated carbon to the mix. The chemical activation of the slag itself has caused a reduction in permeability for low solids IMPER~X~ formulations, an important improvement when trying to minimize the volume increase in fixation projects. Despite the unglamorous sulfur odor of fresh IMPERMIX~ samples, sulfur contained in slag may be a factor in explaining the exceptional strength and should be investigated. CONCLUSIONS Cement-bentonite slurries and grouts have opened many avenues in geotechnical engineering and construction. Lack of choice in better local ingredients and chemical and regulatory thresholds have limited the full use of these products in the hazardous waste remediation business despite their "eng~neenng" values. With contamination eventually migrating from soils into the rock basement, barriers must address both media, and self-hardening slurries are a good tool for dealing with both. A new pair of clay-cement ingredients, bearing the flag of IMPER~X~, is a good candidate to take over the self-hardening slurry technology to carry it to the hazardous waste environment. In times when remediation funding is getting tighter, the return to permanent vertical and horizontal enclosures is in order, where possible. The laxity in slurry trench battier technology (soil bentonite) that has prevailed up to now because of the planned temporary nature of the battier or because of the creation of negative gradient baIriers should give way to baITiers that are engineered and built to be barriers that act like baIriers. The bureaucratic observance of performance criteria is often limited to a coefficient of permeability threshold without even defining the laboratory-test~ng criteria. This attitude does not seem to factor in the constructability versus end-product quality elements that are essential to achieving one's objectives and has led to risky situations (not to mention law suits) that would not be tolerated in a permanent enclosure project. Some engineering and construction ngor will have to be reintroduced in the process and, if this is possible from an overall bus~ness/legal consideration, then long-term in-situ bioremediation (Nature's work) will take place safely at a fraction of the cost of the present approaches in areas where land can be furrowed for an extensive penod oftime. BIBLlOGRAPlIY D'Appolonia, D. I. 1980. Soil bentonite cutoffs. ASCE Journal. 106:(GT41339-417. Dupeuble, E. P., and P. Habib. 1978. Coupures Stanches en beton plastique, 13th ICOLD congress, Q. 4S, New Delhi. Evans, I. C., H.-Y. Fang, and T. I. Kugelman. 1985. Containment of hazardous materials with soil bentonite slurry walls. Paper presented at HMCRI, Washington, D.C. November 4-6. Haug, M. D. 1980. Optimization of slurry trench cutoff design. PhD. disertation. University of California, Berkeley. Jefffies, S. A. 1985. Clay slag cement grouts for pollution control. Workshop on Blast Furnace Slag Cements, Kings College, London. Johnson, A. I., R. K. Frobel, and N. S. Cavalli eds. 1985. Hydraulic barriers in soil and rock. Proceedings of the ASTM/ASCE Symposium, Denver, Colo., June 1984. ASTM STP 874.

OCR for page 142
APPENDIX~PAPERS PRESENTED D-131 Kahl, T. W. I. L. Kauschinger, and E. B. Perry, 1991. Plastic Concrete Cutoff Waits for Earth Dams, U.S. Army Corps of Engineers report REMG-GT-! 5. Millet, R. A., and I. Y. Perez. 1981. Current USA practice, slurry wall specifications. ASCE Journal. 107(GT81:1041-1056. Noguera, G. 1985. Diaphragm wall for Colbun main dam. Paper presented at the 15th [COLD Congress, Q.5S, Lausanne. Pare, I. I., P. I. GIover, and G. Dussault. 1985. Construction control of remedial work at LG 3 south dikes. Paper presented at the ~ 5th {COLD Congress, Q.5S, Lausanne. Ryan, C. R. 1977. Slurry Cutoff Wails. Interim report, technical course at Resource Management Tnc., February. Schmednecht, F. 1976. Slurry injection. Construction Digest. January 8. Tallard, G. R. 1984. Slurry trenches for containing hazardous wastes. ASCE C.E. Magazine, February. Tallard, G. R. 1990. New trenching method using synthetic big-polymers. ASTM STP 1129. June. TalIard, G. R. 1992. Hazwaste hydraulic barriers. Paper presented at Update for the 90's 45th CGS conference, Toronto, October. U.S. Environmental Protection Agency (USEPA). 1984. Slurry trench construction for pollution migration control. EPA.540/2-84-00l, Cincinnati, Ohio.