icdf stabilization

7/30/2019 ICDF Stabilization

1/78

BNL-60977INFORMALmR!rIN-SITU CONTAINMENTAND STABILIZATIONOF BURIED WASI'E - ANNUAL REPORT Fy 1994

M.L.Allan and L.E.Kukacka

October 1994

Energy Efficency and Conservation DivisionDepartment of Applied ScienceBrookhaven National LaboratoryUpton, New York 11973

This work was performed under the auspices of the U.S. epartment of EnergyWashington, D.C. Under Contract No. DE-ACO2-76CHOOO16


2/78

DISCLAIMERThis report was prepared as an account of work sponsored by anagency of the United States Government. Neither the United StatesGovernment nor any agency thereof, nor any of their employees, norany of their contractors, subcontractors, or their employees makesany warranty, express or implied, or assumes any legal liability orresponsibility for the accuracy, completeness, or usefulness of anyinformation, apparatus, product or process disclosed, or representsthat its use would not infringe privately owned rights. Reference

herein to any specific commercial product, process, or service bytrade name, trademark, manufacturer, or otherwise, does not neces-sarily constitute or imply its endorsement, recommendation, orfavoring by the United States Government or any agencythereof. Theviews and opinions of authors expressed herein do not necessarilystate or reflect those of the United States Government or anyagency, contractor, or subcontractor thereof.ii


3/78

DISCLAIMERPortions o f this do cum ent m ay be illegiblein electronic image products. Ima ges areproduced from th e best availabte originaldocument.


4/78

CONTENTSPage

LIST OF T .............................................LIST OF FIGURES ...........................................S ....................................................3 .0 Introduction ..........................................2 . 0 In-Situ Stabilization of Chromium Contaminated Soil ...2.1 Introduction ..........................................2.2 Experimental Work......................................

ivviviii1223

2.2.12.2.22.2.32.2.42.2.52.2.62.2.72.2.82.2.92.2.10

Materials...................................... 3Specimen Preparation .......................... 6Curing ........................................ 6Viscosity...................................... 7Specific Gravity................................ 7Flow Time ..................................... 7Leach Tests ................................... 7Permeability Coefficient ...................... 8Unconfined Compressive Strength ............... 8Wet-Dry Durability Tests ...................... 9

2.3 Results and Discussion................................. g2.3.1 Unhardened Grout Properties ................... 92.3.2 TCLP Tests .................................... 102.3.3 Deionized Water Leach Tests ................... 212.3.4 Permeability Coefficient ...................... 222.3.5 Compressive Strength .......................... 242.3.6 Wet-Dry Cycling................................ 262.3.7 General Discussion ............................ 282.3.8 Recommendations for Future Work on In-SituStabilization ................................. 28

3.0 In-Situ Containment of Stabilized Chromium Plume ...... 303.1 Introduction .......................................... 303.2 Experimental Work...................................... 303.2.1 Materials ..................................... 30 3.2.2 Specimen Preparation .......................... 313.2.3 Curing......................................... 323.2.4 Unhardened Grout Properties ................... 323.2.5 Permeability Coefficient ...................... 323.2.6 Unconfined Compressive Strength ............... 333.2.7 Chromium Diffusion Tests ...................... 33

iii


5/78

Page3.2.8 Permeation Grouting ............................ 35

3.3 Results and Discussion ................................. 353.3.1 Permeability Coefficient of Slag Modified SoilCements ........................................ 353.3.2 Permeability Coefficient of Soil Cured GroutBefore and After Drying ........................ 403.3.3 Permeability Coefficient of Soil Cured SoilCements Before and After Drying ................ 423.3.4 Effect of Confining Pressure................... 443.3.6 Chromium Diffusion ............................. 443.3.7 Permeation Grouting Cores ...................... 473.3.8 General Discussion of In-Situ Containment ofChromium Plume................................. 473.3.9 Recommendations for Future Work on In-Situ

Containment of Chromium Plumes ................. 48

3.3.5 Compressive Strength ........................... 4

4.0 Jet Groutina Field Trials .............................. 495.0 In-Situ Immobilization of Tritium Plums................ 536.0 In-Situ Containment of Tritium Plume ................... 546.1 Introduction ........................................... 546.2 Experimental Work ...................................... 54

6.2.16.2.26.2.36.2.46.2.56.2.66.2.76.2.86.2.9

Materials ...................................... 54Specimen Preparation ........................... 55Curing ......................................... 6Permeability Coefficient....................... 56Unconfined Compressive Strengt.h ............... 56Wet-Dry Cycle Tests ............................ 56Tritium Diffusion Tests ........................ 57Unhardened Grout Properties .................... 56Flexural Strength .............................. 56

6.3 Results and Discussion ................................. 576.3.1 Unhardened Grout Properties.................... 576.3.2 Permeability Coefficient ....................... 586.3.3 Compressive Strength ........................... 606.3.4 Flexural Strength ..............................606.3.5 Tritium Diffusion .............................. 63

7.0 Site Visit ............................................. 668.0 Technoloav Transfer .................................... 678.1 Patents ................................................ 67

iv


6/78

8 . 28 .38 .4

8 . 5

PUblications ..........................................Publications in Press..................................Publications under Review ..............................Conferences............................................9.0 Conclusions............................................

10.0 References............................................

Page676767

676768


7/78

L I B T OF TABLESPage

Table 1. Mix Proportions of Grouts......................... 4Table 2. Mix Proportions of Treated Soils.................. 5Table 3. Rheological Properties of Unhardened Grouts....... 9Table 4. Specific Gravities of Unhardened Grouts........... 10Table 5. TCLP Results for Landfill Soils Treated with 40%Slag/60% Cement Mix............................... 19Table 6. Results for Deionized Water Leach Tests........... 22Table 7. 28 Day Wet Cured Compressive Strength of StabilizedSoil........ ...................................... 26Table 8. 84 Day Wet Cured Compressive Strength of Stabilized

S o i l . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26Table 9. Residual Compressive Strength of Stabilized Soil.. 28Table 10. Mix Proportions of CWL Barrier Grouts............ 31Table 11. Mix Proportions of CWL Soil Cement Barriers...... 31Table 12. Mix Proportions of Microfine Cement Grout........ 35Table 13. Properties of Unhardened Microfine Cement Grout.. 35Table 14. Permeability Coefficients of Slag Modified SoilCements.......................................... 36 Table 15. Permeability Coefficients of Wet Cured SlagModified Soil Cements After Wet-Dry Cycles....... 36Table 16. Permeability Coefficients of Plain and FibreReinforced Grout Before and After Wet-Dry Cycles. 40Table 17. Permeability Coefficients of Plain and FibreReinforced Soil Cements Before and After Wet-DryCycles ........................................... 42Table 18. Compressive Strength of Slag Modified SoilCements.......................................... 44Table 19. Mix Proportions of Grouts for Jet Grouting FieldTrials........................................... 51Table 20. Properties of Unhardened Grouts Prepared inLaboratory ....................................... 51

vi


8/78

PageTable 21. Jet Grouting Parameters and Grout Properties..,.. 52Table 2 2 , Mix Proportions of Grouts for MWL Barriers....... 55Table 23. Mix Proportions of MWL Soil Cement Barriers...... 55Table 24. Unhardened Grout Properties for MWL Barriers..,,. 58Table 25. Permeability Coefficients Before and AfterWet-Dry Cycles for MWL Barrier Grouts............ 58Table 26. Compressive Strength of MWL Barrier Grouts...,.., 60Table 27. Flexural Strength of MWL Barrier Grouts.......... 63Table 28. Intrinsic Diffusion Coefficients for Tritium... .. 64

vii


9/78

LIST O F FIGURESPage

Figure 1.Figure 2.Figure 3.

Figure 4.

Figure 5.

Figure 6.Figure 7.

Figure 8.

Figure 9.

Figure 10.

Figure 11.Figure 12.Figure13.Figure 14.

TCLP results for 200 ppm Cr(II1) soil treatedwith slag modified grouts ......................... 1TCLP leachate pH for 200 ppm Cr(II1) soil treatedwith slag modified grouts ......................... 13TCLP results for 200 ppm Cr(V1) soil treatedwith slag modified grouts at soilt/cementitiousmaterial ratio = I............................... 15TCLP results for 200 ppm Cr(V1) soil treatedwith slag modified grouts at soilt/cementitiousmaterial ratio = 2............................... 15TCLP results for 200 ppm Cr(V1) soil treatedwith slag modified grouts at soilt/cementitiousmaterial ratio = 5............................... 16TCLP leachate pH for 200 ppm Cr(V1) soil treatedwith slag modified grouts ......................... 17TCLP results for 200 ppm Cr(V1) soil treatedwith 40% slag/60% cement grout before and afterexposure to air for 7 months..................... 18TCLP results for 200 ppm Cr(V1) soil treatedwith 80% slag/20% cement grout before and afterexposure to air for 7 months..................... 20Permeability coefficient versussoil/cementitious material ratio for 200 ppmCr(II1) soil treated with slag modified grouts ... 23Compressive strength versus soil/cementitiousmaterial ratio for 200 ppm Cr(IX1) soil treatedSchematic diagram of diffusion cell used forchromium and tritium diffusion experiments ...... 34Permeability coefficient of slag modified soilcements after 28 days wet curing ................ 37Permeability coefficient of slag modified soilcements after 84 days soil curing ............... 38Permeability coefficient of wet cured slagmodified soil cements before and after 12

with slag modified grouts ....................... 25

wet-dry cycles..................................39v i i i


10/78

PageFigure 15. Effect of 0.2% Fibrillated (0.2F) and 0.2%Monofilament Fibres (0.2MF) on PermeabilityCoefficient of Soil Cured Grout Before and AfterDrying and Re-Saturation........................ 41

Monofilament Fibres (0.2MF) on PermeabilityCoefficient of Soil Cured Soil Cement Before andAfter Drying and Re-Saturation .................. 43Figure 17. Relationship Between Permeability Coefficientand Confining Pressure for Cracked Soil Cementwith 0.2% Monofilament Fibres (47S2 0.2MF). ..... 45

Figure 16. Effect of 0.2% Fibrillated (0.2F) and 0.2%

Figure 18. Compressive strength versus soil/cementitiousmaterial ratio for slag modified soil cements... 46Figure 19. Permeability coefficients of wet cured groutsbefore and after 12 wet-dry cycles.............. 59Figure 20. Compressive strength of grouts after 28 and 84days wet curing ................................. 61Figure 21. Flexural strength of grouts after 28 days wetcuring.. ........................................ 62Figure 22. Tritium diffusion versus time for groutsand soil cements................................ 65

ViX


11/78

SUMMARYResearch on advanced grouting materials for in-situstabilization and containment of buried hazardous waste that wasinitiated in FY 92 continued in FY 94. The three main areas offocus for FY 94 were in-situ stabilization of chromium contaminatedsoil, subsurface containment barriers for chromium plumes andsubsurface containment barriers for tritiated water plumes.Grouts containing blast furnace slag as a partial replacementfor cement were studied for stabilization of trivalent andhexavalent chromium contaminated soil, The levels of slag,soil/grout ratios and water/cementitious material ratios werevaried in order to determine the best combination of properties,Soil stabilized with grout was subjected to a variety of tests,including EPA Toxicity Characteristic Leaching Procedure, ANS 16.1deionized water leaching, strength and permeability. Treatment ofcontaminated soil with slag modified grout resulted in a strong,low permeability product that passed EPA leaching tests. Bothtrivalent or hexavalent chromium could be stabilized. Landfillsoil samples with 6110 ppm Cr were also successfully treated. Itwas determined that addition of slag to grout was necessary forreduction of hexavalent to trivalent chromium,' followed byimmobilization as chromium hydroxide. A slag content greater than20% by weight of total cementitious materials was necessary forgreatest leach resistance. This is compatible with Type IS cementwhich is preblended with 25 to 75% slag. The great advantage ofusing a slag modified grout for in-situ stabilization of chromiumcontaminated soil is that either Cr(II1) or the more mobile andtoxic Cr(V1) can be treated since no pre-reduction of Cr(V1) isneeded. This simplifies and reduces the time and costs associated

with Cr(V1) remediation immensely.Investigation of resistance to permeable cracking ofcementitious subsurface containment barriers continued, Additionof slag to grouts for containment barriers was initiated and thegrout properties were characterized. Soil cements containing slagwere a l s o developed and characterized. Jet grouting field trialsusing plain and slag modified grouts were conducted at anuncontaminated site at SNL.Grout and soil cements were investigated as potential barriersfor soil contaminated with tritiated water. Diffusion coefficientswere measured and determined to be of the order of 10"' to lo-" m2/s.

These values are similar to those previously reported for grout andconcrete. It is evident that grout and soil cement barriers do notafford permanent containment of tritium. At the relatively lowlevel of tritium activity in the MWL plume it is marginal whetherplacement of a cementitious barrier for tritium containment wouldbe sufficiently beneficial.

X


12/78

1.0 INTRODUCTIONThe ongoing objective of this work is to develop, demonstrateand implement advanced grouting materials for in-situ installationof subsurface containment barriers and for in-situ stabilization of

chromium contaminated soils. The work is conducted as part of theMixed Waste Landfill Integrated Demonstration (MWLID) headquarteredat Sandia National Laboratory (SNL) The two landfills of specificinterest are the Chemical Waste Landfill (CWL) and the Mixed WasteLandfill (MWL) , both located at SNL. The work on groutingmaterials was initiated in FY 92 and the previous accomplishmentsare documented in the FY 92 and FY 93 annual reports.The work is comprised of two subtasks: (1) In-Situ Barriersand (2) In-Situ Stabilization of Contaminated Soils. The mainenvironmental concern at the CWL is a chromium plume resulting fromdisposal of chromic acid and chromic sulphuric acid into unlinedpits. This program has investigated means of in-situ stabilization

of chromium contaminated soil and placement of containment barriersaround the CWL. In-situ immobilization of tritiated water with cementitious grouts wasnot considered to be a method with a high probability of successand was not pursued. This is discussed further in Section 5.0.Containment barriers for the tritium plume were investigated.

The MWL contains a plume of tritiated water.

FY 92 work examined stabilization of chromium contaminatedsoil and determined that reduction of hexavalent chromium wasnecessary before treatment with either Portland cement grout orcalcium hydroxide, Polymer, cementitious and polymer modifiedcementitious barrier grouts for the CWL were investigated in FY 92.The grouts were formulated for use as either monolithic (i.e. groutonly) barriers or for in-situ mixing with soil to produce soilcement barriers. Additives for significantly improving the curedproperties of cementitious grouts and grout treated soils overconventional materials were identified. Polymer grouts wereconsidered to not be cost effective for the site studied since theproperties of interest were not significantly improved overcementitious grouts, yet the raw material cost was high. Inaddition, cementitious grouts had other practical benefits such aseasy clean up, handling and environmental safety, reaction controland worker familiarity.

In FY 93 work concentrated on optimization andcharacterization of CWL barrier grout and soil cement formulations.Mixes were tested for properties such as permeability, strength,wet-dry durability, shrinkage, resistance to shrinkage cracking,leach resistance, freeze-thaw durability and flow behaviour.Fibres were investigated as a means of improving crack resistance.While fibres improved resistance to restrained shrinkage cracking,the effects on permeability and wet-dry durability were variable.Fibre reinforcement studies were continued in FY 9 4 and will bediscussed in this report. Some problems with obtaining uniformmixing of fibres were encountered, and field tests would be

1


13/78

necessary to completely evaluate the practicality of their use ingrouts.FY 94 work focused on stabilization of chromium contaminatedsoil with blast furnace slag modified grouts to bypass the stage ofpre-reduction of Cr(VI), barriers for tritiated water containment

at the MWL, continued study of barriers for the CWL, and jetgrouting field trials for CWL barriers at an uncontaminated site atSNL. Cores from the FY 93 permeation grouting field trails werealso tested in FY 94.200 IN-SITU BTABILIZATION OF CHROMIUM CONTAMINATED BOIL2.1 Introduction

Between 1962 and 1985 solutions of chromic and chromicsulphuric acid were dumped into unlined pits at the CWL. Sincethen, migration of chromium to a depth of 23 m has been detected.The total chromium concentrations in the soil vary with depth andthe typical value is approximately 200 ppnn. Concentrations up to6000 ppm exist in some locations. The chromium is present in boththe trivalent and hexavalent oxidation states.

Trivalent chromium can be stabilized by addition of basicreagents, such as cement or calcium hydroxide, to precipitatechromium hydroxide. In addition to formation of Cr(OH),, Mollah &-l. (1992) and Ivey et al. (1990) suggested that Cr(II1) maysubstitute for Si in the calcium silicate hydrate phase of cement.Ivey et al. (1990) also postulated that Cr(II1) substituted for A1and Fe in calcium aluminate hydrates and calcium aluminoferritehydrates. Kindess et al. (1994) did not find evidence ofincorporation of Cr(II1) into calcium silicate hydrate, but didfind that Cr(II1) could substitute for A1 in calcium aluminatehydrates.The more mobile and toxic hexavalent state of chromiumrequires pretreatment by reduction to Cr(II1) before stabilizationas the hydroxide (Connor, 1990). Leaching experiments performed byZamorani et al. (1988) have demonstrated that Cr(V1) is notpermanently immobilized by addition of Portland cement. Reductionagents for Cr(V1) include ferrous sulphate, bisulphites andsulphides (Connor, 1990). Potential drawbacks of these agents forthe landfill remediation include the necessity for pH adjustmentprior to mixing with ferrous sulphate and generation of SO2 with

bisulphites. Other problems with soil pretreatment by solutionsare that reduction may be slow and incomplete due to previousdiffusion of chromium into soil particles (Connor, 1990). Thepotential safety and environmental hazards of sulphide orbisulphite treatment and the EPA regulations regarding H 2Sproduction should also be regarded. Any unreacted Cr(V1) willremain mobile and this is cause for concern.In FY 92, treatability studies of soil contaminated with

2


14/78

either trivalent or hexavalent chromium were conducted (Allan,Kukacka and Heiser, 1992). Soil spiked with 1000 ppm Cr(II1) wassuccessfully stabilized by a pumpable, superplasticized Portlandcement grout. Reduction of Cr(V1) with Fez+ solutions andstabilization with lime was also investigated and found to besuccessful.Work in FY 94 examined means of overcoming problems and costsassociated with a two stage treatment process for Cr(V1). Theobjective was to combine the reduction and stabilization steps.Reduction of Cr(V1) to Cr(II1) and subsequent stabilization hasbeen observed for cements containing ground granulated blastfurnace slag (Kindness et al., 1994; Langton, 1988, Langton andWong, 1991). This is attributed to the low redox potential ofsulphur bearing slag modified cements as compared to ordinaryPortland cements (Macphee and Glasser, 1993). Langton (1988)demonstrated significantly lower total chromium concentrations inEPA Toxicity Characteristic Leaching Procedure (TCLP) tests whenblast furnace slag was added to saltstone formulations. Analysis

of pore fluid expressed from cement pastes spiked with 5000 ppmCr(V1) showed that ordinary Portland cement had 71 ppm total Cr ofwhich 70 ppm was Cr(V1) whereas a blend with 10% ordinary Portlandcement and 90% slag had


15/78

slag) ratio (w/c). Superplasticizer was added at a rate of 20ml/kg cementitious material. Silica fume modified grout was alsoinvestigated. The silica fume was suppllied by Norcem ConcreteProducts. The grout was added to soil at soil/cementitiousmaterial ratios (s/c) by weight of 1, 2 and 5. Additional mixingwater was required at the higher soil/cementitious material ratios.Soil from an uncontaminated site near the CWL was used. The soilwas dried prior to use so that moisture content was constant.

The mix proportions of the grouts and treated soils are givenin Tables 1 and 2 respectively. Slag was used at cementreplacement levels of 0, 20, 40, 60 and 80% by weight of totalcementitious materials (cement plus slag) and the corresponding MixNumbers are 47, 80, 74, 75 and 78. Silica fume was used at areplacement level of 5% by weight of total cementitious material.The last number in the mix code refers to the s/c value (1-5).The silica fume modified mix (58-5) was used at a s/c value of 5.The proportions may require alteration for field applicationdepending on the type of placement equipment used and the soilconditions.T a b l e 1. Mix Proportions of Grouts

80-180-280-574-1

971 243 0 583 11.0 24.3751 188 0 675 13.2 18.8453 113 0 794 15.3 11.3722 481 0 577 10.9 24.1

4


16/78

N , B . SP = superplasticizerTable 2. Mix Proportions of Treated soils

Mix No . Cement Slag Water Bentonite SP Soil47S1 816 0 391 6.5 16.3 8164782 537 0 388 7.7 10.7 107547s5 268 0 376 7.4 5.4 1342

(ks/m3) (kg/m3) (ks/m3) (kg/m3 ( (ks/m3)

In the first series of tests, chromium was added touncontaminated soil from near the actual landfill site. The soilwas alluvial with silty to gravelly sand type texture and a nativepH of 8.4. Studies on trivalent chromium were performed on soilspiked with CrCl3.6H,O to give 200 ppm Cr(II1) , Hexavalent chromiumcontaminated soils were produced by spiking with CrO, to a level of200 ppm Cr(V1). The dilution of the soil with grout resulted inthe chromium concentrations in the treated soils being 80, 106 and135 ppm for s/c = 1, 2 and 5 respectively, The second series oftests was performed on soil samples from the landfill, The samplesused were retrieved from depths of 9.2, 10.8 and 12.3 m. Thereported amounts of Cr(V1) and Cr (total) were 291 and 6110 pprnrespectively at 9.2 m, 112 and 138 ppm, at 10.8 m and 0.11 and 11.2ppm at 12.3 m (9). Total Cr of the landfill samples was measuredby inductively coupled plasma spectroscopy following acidextraction in accordance with EPA SW 846 Method 3050A and Cr(V1)was determined colorimetrically after water extraction.

5


17/78

2-2.2 Specimen PreparationThe laboratory spiked soils were prepared by adding therequired amount of predissolved CrC1,.6H20 or CrO, to theuncontaminated soil. The amount of distilled water used to

dissolve the chromium compounds was taken into account in the mixproportions. The spiked soils were either mixed by hand or with aplanetary mixer, depending on the quantity. Spiked soil was usedimmediately for preparation of grout stabilized specimens.Grout was mixed in a blender for leach and permeabilityspecimens or in a planetary mixer for compressive strength testcylinders. The order of addition was water, superplasticizer,bentonite, cement and any slag. For small. leach and permeabilityspecimens, the grout was mixed with the chromium contaminated soilby hand until a uniform product was achieved. Grout stabilizedsoil for strength cylinders was mixed in a planetary mixer.Bentonite hydration was limited to that which occurred duringmixing, Trials showed that the bentonite hydration time had littleeffect on grout viscosity, but did increase the gel strength. Forfield applications the amount of time allowed or bentonitehydration prior to mixing will depend on practicality.It is realized that the laboratory mixing equipment differsfrom that which would be used in the field for treating soils.Hence, some variations in mix uniformity and final propertiesbetween laboratory prepared specimens and jet grouted or augermixed soil can be expected.Six cylinders 65 mm high and 30 mm diameter were cast for TCLPtests. Two of these were used and the other four were archived.Permeability and deionized water leach tests were performed oncylinders 36 mm high and 26 mm diameter. Three specimens were usedfor each batch. The cylinders used for strength tests were 152 mmhigh and 7 6 mm diameter and six specimens per batch were tested.

2.2.3 CurinqAll materials were covered with plastic sheet after casting toprevent evaporation and plastic shrinkage cracking. Groutstabilized soil specimens were demoulded 24 hours after casting andcured for 28 days for leach, permeability, compressive strength,and wet-dry cycle tests. Two types of curing were used fordifferent tests. Conventional type curing in the form ofsubmersion in water at room temperature was conducted in order todetermine the properties when adequate cement hydration occurs.Simulation of the in-situ curing conditions for subsurface barrierswas achieved by burying 24 hour old specimens in 200 1 containersof site soil. 28 day compressive strength and wet-dry cylinderswere cured in water. Other test specimens, and 8 4 day compressivestrength cylinders were cured in soil.

6


18/78

2.2.4 ViscositvViscosity of grouts was measured using a Fann 35A coaxialcylinder viscometer at 300 rpm. Measurements were takenimmediately after mixing in the blender. All viscosity

measurements were conducted at room temperature.2.2.5 SDecific Gravity

The specific gravity of grouts and soil cements was measuredusing a Baroid mud balance following ASTM D 4380-84.2.2.6 Flow Time

The flow, or efflux, time of the stabilization grouts wasmeasured using a Marsh funnel. The flow time indicates therelative flowability of grouts by measuring the time of efflux ofa specific volume of grout through a standardized cone. Theprocedure for measuring flow time was to fill the Marsh funnel tothe 1500 ml mark and measure the efflux time for 1000 ml of groutinto a measuring cylinder. For water, the flow time of 1000 mlinitially filled to the 1500 ml mark is 28 seconds..2.2.7 Leach Tests

Two types of leach tests were performed. These were the EPATCLP test using Extraction Fluid #2 (5.7 ml glacial acetic aciddiluted to 1 1, pH = 2.88) and the American Nuclear Society ( A N S )16.1 test using deionized water. The grout treated soils werecured for 28 days by burying in dry site soil prior to leachtesting. In accordance with the TCLP test procedure, the curedsoil cements were crushed to pass through a 9.5 mm sieve beforetesting. In addition, crushed specimens prepared from grout with40% slag/60% cement and 80% slag/20% cement were exposed to theatmosphere for seven months and re-tested to determine the effectof oxygen and carbon dioxide on the stability of the treated soil.TCLP tests were also performed on untreated spiked and landfills o i l s .

At the conclusion of the TCLP tests, the pH of the leachatewas measured. The leachate was then acidified with a measuredquantity of concentrated HNO, and analyzed for hexavalent and totalchromium. Atomic absorption spectrophotometry was used for totalchromium measurements. Hexavalent chromium concentration wasdetermined by reacting the acidified leachate with 1,5diphenylcarbazide and measuring the absorbance at 540 nm with a Wspectrophotometer.

Preliminary TCLP tests were performed on uncontaminated soilstreated with grout to determine whether detectable chromium wouldleach from cement or soil, thereby necessitating backgroundcorrection. Analysis for iron was also conducted on the blank

7


19/78

specimens to determine if sufficient iron would leach from blastfurnace slag, cement or soil to interfere with measurement of totalchromium by atomic absorption spectrophotometry and hexavalentchromium by colorimetry. It was found that no chromium was leachedfrom the uncontaminated specimens and that: iron concentration wasless than 1 ppm, and therefore unlikely t o interfere with chromiumanalysis. The absence of matrix effects was confirmed by analyzinguncontaminated and contaminated leachates with standard additionsof chromium.

Monolithic grout treated cylinders were subjected t o ANS 16.1leach tests in deionized water for 24 weeks. Deionized water leachtests were performed on trivalent chromium soils stabilized withgrouts containing 0 and 40% slag replacement and on hexavalentchromium soils stabilized with 40 and 80% slag grouts.Soil/cementitious material ratio ranged from 1 to 5. Hexavalentchromium soils treated with ordinary Portland cement grout were nottested since previous studies show that chemical immobilization isnot achieved (Kindness et al., 1994; Zamorani et al., 1988).The leachates from the ANS 16.1 tests were analyzed andreplaced daily for the first five days and subsequently every twoweeks. The sampling period in the initial stages of the test waschanged from the shorter intervals given in the standard becausenegligible release occurred in periods of 2 and 5 hours. Theleachates were tested for pH, calcium concentration and totalchromium concentration. Calcium concentration was used , s anindicator of leach resistance of the cementitious material sincedeionized water is aggressive to hydrated cement. Atomicabsorption spectrophotometry was used to determine calciumconcentration.

2.2.8 Permeability CoefficientThe water permeability coefficient was measured using a rigidwall uniaxial flow permeameter. Specimens were cast in glass tubes26 mm diameter and 50 mm long. The soil used was spiked with 200ppm trivalent chromium. After curing in soil for 28 days thespecimens were trimmed to a length of 36 imm. The diameter of thespecimens was slightly oversized to achieve a leakproof seal withthe rubber gasket. The seal was further ensured by coating thegasket walls with silicone grease. Specimens were vacuum saturatedwith water prior to testing. The applied pressure for thepermeability tests was 2 atm. Preliminary experiments conducted atdifferent pressure gradients confirmed that flow through t he grout

treated soils obeyed Darcy's Law. Measurements did not commenceuntil constant flow rate was established. Three specimens perbatch were tested.2.2.9 Unconfined ComDressive Strensth

Grout treated soil cylinders 76 mm diameter and 152 mm longwere measured for unconfined compressive strength measurementsfollowing ASTM C 39-86. Measurements were conducted on six

8


20/78

specimens per batch for curing times of 28 and 84 days. The 28 daycure tests were performed on soil cements cured in water and thespecimens for 84 day test were cured in dry site soil. Allspecimens were prepared from soils spiked with 200 ppm trivalentchromium.

Mix No. Viscosity (cp)74-1 4674-2 1174-5 4

2.2.10 Wet-Drv Durability TestsCylinders of treated soil 76 mm diameter and 152 mm long weresubjected to 12 wet-dry cycles at 25OC. The drying temperature islower than that of 71OC used in ASTM D 559-89 and was chosen as amore realistic simulation of subsurface temperatures to which thematerial would be exposed in the landfill. The cycles consisted of5 hours submersion in water followed by 43 hours drying in air atrelative humidity 40- 50%. Compressive strength was measured after24 hours soaking at the completion of the 12 wet-dry cycles. Thesoaking was performed to give a specimen moisture content similarto the original 28 day wet cured specimens for strength comparisonpurposes. Mass changes were not monitored since tests on grout

treated soil conducted in FY 93 showed that variations were due touptake and loss of moisture rather than loss of integrity.

Marsh Flow Time (s)52.831.228.2

2.3 Results and Discussion2.3.1 Unhardened Grout Properties

The viscosities and Marsh flow times of the 74 series ofgrouts (40% slag/60% cement) are given in Table 3. The specificgravities of the unhardened grouts are presented in Table 4.T a b l e 30 Rheological Properties of Unhardened Grouts

The 74 series of grouts had relatively low viscosities and lowflow times. The mix proportions of these grouts could be alteredto suit the flow requirements of the chosen placement technique ifnecessary. The specific gravity data showed the expected decreasein values as the amount of slag replacement for cement increased,due to the lower specific gravity of slag itself.

9


21/78

Table 4 . s p e c i f i c G r av it ie s of Unhardened Grouts

58-580-180-280-574-1

Mix No. Specific GravityII1.381.841.651.391.82

47-1 1.8547-2 1.6647-5 1.40

74-274-575-175-2

1.811.62

78-178-2

75-5 j 1.381.791.61

II 78-5 I 1.372.3.2 TCLP Tests

The results of the TCLP tests on 200 ppm Cr(II1) spiked s o i ltreated with plain and slag modified grout are shown in Figure 1.The effect of slag content and soil/cementi.tious material ratio areevident. The grout without any slag has the greatest amount ofleached Cr, but the level remains below the EPA limit of 5 ppm. Atthe 80% slag replacement level no chromium was detected in theleachate. Hence, it is apparent that addition of slag improves theTCLP leach resistance of grout stabilized trivalent chromium andthat high amounts of soil can be added to the grout withoutdetriment to TCLP leachability. The concentration of chromiumleached from spiked unstabilized soil was 6.3 ppm.The silica fume modified mix was used to stabilize 200 ppmCr(II1) spiked soil at a s/c = 5 (58S5). The total leached Cr was0.3 ppm and the leachate pH was 11.03. The amount of leachedchromium was the same as that for the plain grout (4785), andhigher than the slag modified mixes.The leachate pH values of the grouted soils with 200 ppm

10


22/78

3.00n2.50EQaY5 .00-

1.50w0Z00 1.00

0.50

0.00 0 40 60SLAG REPLACEMENT LEVEL (%)

1 S/C=l s/c=2ggg$js/c=5I80

F i g u r e 1 .modified grouts.TCLP results for 20 0 ppm C r ( I I 1 ) soil treated with slag


23/78

Cr(II1) are shown in Figure 2. The corresponding leachate pH ofthe unstabilized soil was 5 . 8 2 . As indicated, the pH decreaseswith increasing soil/cementitious material ratio and withincreasing slag content. The higher pH of the Portland cementgrout may result in increased solubility of chromium.The concentrations of hexavalent and total chromium in theleachate from Cr(V1) spiked specimens are plotted against amount ofslag for s/c = 1 to 5 in Figures 3to 5. In all cases, thegreatest amount of chromium is leached from the specimens withoutslag, and leachate concentrations decrease with increasing slagreplacement of cement. The leached chromium remains below EPAlimits for all mixes. However, the ordinary Portland cement groutis not expected to give permanent chemical stabilization of Cr(V1)(Kindness et al., 1994; Zamorani et al., 1988). The leachresistance of the soil treated with ordinary Portland cement isprobably due to physical containment, rather than chemicalstabilization.A significant feature of Figures 3to 5 is that the proportionof chromium that is in the hexavalent state is reduced withincreasing slag replacement level. This indicates reduction to Cr(111). When slag is absent the majority of leached chromium is inthe undesirable hexavalent state. The results show that slagreplacement levels equal or greater than 40% give the best leachresistance, even at high soil loadings. The untreated soil spikedwith 200 ppm Cr(V1) had leachate concentrations of 8 . 7 5 ppm.Cr(V1)and 9.5 ppm Cr total. It is possible that ferrous ions within thesoil caused some conversion of Cr(V1) to Cr(II1).The pH values of the leachates from the treated 200 ppm Cr(V1)soils are shown in Figures 6. The values are similar to those forthe 200 ppm Cr(II1) soils treated with grout with the same mixproportions. The untreated soil had a leachate pH of 5 . 4 9 . Someneutralization of the soil by the added CrO, probably occurred.Another possible Cr(II1) stabilization mechanism is substitutionfor A13+ ions in bentonite used in the grouts. It is also possiblethat Cr3+is adsorbed on bentonite (Connor, 1990). However, thesemechanisms would only account for a small proportion of the totalstabilization since a low amount of bentonite was used.Specimens with s/c = 5 prepared from grout with 80% slagreplacement exhibited surface cracks after the 2 8 day curing periodin dry soil. The cracks are attributed to excessive drying

shrinkage and this deterioration is of concern.Leach tests were performed on 200 ppm Cr(V1) spiked soilstabilized with grout containing 40 % sl.ag/60% cement that wascrushed after curing and exposed to air for seven months. Theresults are shown in Figure 7. The initial results for the samemixes are also shown. Comparison indicates that the concentrationsof hexavalent and total chromium are similar for the two series oftests when s/c = 1 and 2. This indicates stability when the grout

12


24/78

7 I I I I I I I1 I I I I I I0 20 40 60SLAG REPLACEMENT LEVEL (%) 80

0 /c=1 s/c=281 s/c=5

Figure2.slag modified grouts.TCLP leachate pH for 200 ppm Cr(II1) soil treated with

13


25/78

3-00nIF i g u r e 4.modified grouts at soil/cementitious material ratio = 2.TCLP results for 200 ppm Cr(V1) s o i l treated w i t h slag

15


26/78

nEZ0

aQY

!i-zW0z00

3.00

2.50

2.00

1.50

1oo

0.50

0.00 0 20 40 60SLAG REPLACEMENT LEVEL (%)

80

Figure 5. TCLP results f o r 200 ppm Cr(V1) soil treated with slagmodified grouts at soil/cementitious material ratio = 5.16


27/78

12

11

IW 102aY 9I0

8

7

\J I

0

0

_ .

20 40 60SLAG REPLACEMENT LEVEL (%) 80

0 s / c=1 A s / c = 2 s / c = 5

Figure 6. TCLP leachate pH for 200 ppm Cr(V1) soil treated withslag modified grouts.17


28/78

0.90Z 0.80

0.70s0.60

c1Q

, -

I-0$ 0.50

0.4006 0.300.200.100.00 1 2 5

SOIUCEMENTITIOUS MATERIAL RATIOI Cr(V1) Initial Cr(Tota1) Initialu r(Vl) Final Cr(Total) Final 1

Figure 7. TCLP results for 200 ppm Cr(V1) soil treated with 409slag/600 cement grout before and a f t e r exposure t o air for 7months.18


29/78

treated soil is exposed to oxygen and carbon dioxide. However,when s/c = 5, the amount of leached chromium is significantlyhigher after seven months exposure. It i s also significant thatmost of the leached chromium from the exposed mix with s/c = 5 isin the hexavalent state. The results for crushed and leachedspecimens prepared from 80% slag/20% cement grout are depicted inFigure 8, along with th e initial 28 day results. It is evidentthat the leachate chromium concentrations are similar before andafter exposure to air. At s/c = 5, the total Cr after a ir exposurewas 0.15 ppm compared with 0.10 ppm at 28 days and Cr(V1) levelswere similar (0.04 ppm at 28 days and 0.03 ppm after seven monthsexposure). Thus, the large increase i.n Cr(V1) concentrationobserved with the 40% slag/60% cement grout was not repeated forthat with 80% slag/20% cement. It is possible that the higherCr(V1) level with 40% slag was due to an anomaly such as non-uniform mixing.

LeachatePH11.0710.465.2811.2810.624.3411.189.975.74

The pH of the leachates from the seven month old crushedspecimens wa s lower than the original values. For 40% slag/60%cement grout the pH values were 6.87, 6.90 and 6.73 for s/c = 1, 2and 5 respectively, compared with original values of 11.28, 10.38and 7.63. For 80% slag/20% cement grout the leachate pH valueswere 6.91, 7.27 and 6.35 for s/c = 1, 2 and 5*respectively,compared with initial values of 9.74, 8.52 and 7.43. Th e reductionin leachate pH is due to carbonation of the cementitious materialcaused by reaction with atmospheric carbon dioxide.

Leachate Cr Leachate Cr(VI) (PPm) Total (ppm)0.04 0.370.55 0.8311.5 14.30.01 0.170.03 0.141.00 1.100.00 0.080.00 0.080.05 0.09

Slag replacement levels of 40% and s/c values of 1 and 2 wereselected for TCLP studies on contaminated soil from t he landfill.The results for the treated and untreated soils are given in Table5.

10.8 m12.3 m12.3 m12.3 m

Table 5.

Untreated12

Untreated

TCLP Results for Landfill Soils Treated with 40% Slag/60% Cement Mix (74)ty-+9.2 m Untreated10.8 m10.8 m

19


30/78

nEQQZ0FW

2I-ZW0Z00

0.50

0.40

0.30

0.20

0.10

0.00

..... ... . .. ..

1 2 5SOIUCEMENTITIOUS MATERIAL RATIO

1 Cr(VI) Initia l Cr(Tota1) Initia l0 r(VI) Final Cr(Tota1) Final 1

Figure 8 , TCLP r e s u l t s for 200 ppm C r ( V 1 ) soil treated w i t h 80%slag/20% cement grout befo re and a f t e r exposure t o a i r for 7months20


31/78

The majority of TCLP leachable chromium from the untreatedlandfill soils was in the hexavalent state. The unleached 10.8 msample was the closest in original chromium analysis to that of thespiked soils with 112 ppm Cr(V1) and 138 ppm total Cr. The spikedsoils had significantly higher leached concentrations than thislandfill soil ( 8 . 7 5 ppm Cr(V1) for spiked soil compared with 1.00ppm Cr(V1) for landfill soil). This suggests that the landfillsoil was less leachable than the spiked soil which could be due toformation of low solubility chromium complexes with soil compoundsand diffusion of chromium into rocks over time.

The slag modified grout successfully immobilized Cr (VI) in thelandfill soil. It is planned to repeat the TCLP tests on crushedsamples exposed to air to determine the loingevity of stabilizationby these grouts.Factors affecting the TCLP results imclude crushed particlesize. As the s/c value and slag replacement level increased, the

treated soils were weaker and tended to fragment into smallerpieces when crushed. Since the TCLP sample preparation procedurerequires that material passes through a 9.5 mm sieve and does notgive any lower bounds on particle size, finer pieces. are accepted.The result is that the weaker specimens tended to have greatersurface area exposed to leachate. This is expected to increase theleachate pH and increase the leachate chromium concentrationcompared with the same material tested with a greater particlesize.2.3.3 Deionized Water Leach TestsThe mean final leachate pH and cumulative leached calcium andchromium for the stabilized spiked soils are presented in Table 6.The oxidation state of the chromium added to the soil is indicatedafter the mix number.The leachate pH remained alkaline for all the testedmaterials. No detectable chromium was leached from the specimensprepared with 200 ppm Cr(II1) spiked soil. Chromium leached in theearly stages of the tests from the Cr(V1) spiked soils stabilizedwith 40% slag/60% cement grout at s/c = 5 and with the 80% slag/20%cement grout at s/c = 2 and 5. The cumul-ative leached fractionsfor chromium were 0.025, 5.9~10-~nd 5.1~10-~or mixes 74S5 (VI),7882 (VI) and 7885 (VI) respectively. Leaching of chromium ceasedafter 4 weeks.The deionized water leach test is probably more representativeof the exposure conditions for the landfill under study than theTCLP test since exposure to organic acid leachants is not expected.Hence, the low leachability of chromium observed in the deionizedwater tests is indicative of acceptable immobilization for existingfield conditions.

21


32/78

Table 6 . Results for Deionized Water Leach TestsMix Final Leachate Mean MaximumCumulative Cumulative

Leached Creached Ca(ms) (mg)4 7 6 5 (111) 1 0 . 4 9 f 0 . 0 1 3 52 22 0.00

PH

~~ ___

74S1 (111) 1 0 . 4 9 f 0 . 0 5 3 4 5f 4 0.007452 (111) 1 0 . 3 0 f 0 . 3 0 1 6 9 f 2 0.007 4 s 5 (111) 1 0 . 4 1 k 0 . 0 5 2 8 2 + 2 0.00

0.000.00

74S1 (VI) 1 0 . 6 0 + 0 . 1 1 3 7 3 2 27482 (VI1 1 0 . 2 9 2 0 . 1 3 2 8 1 2 2

~~ __ _ ~

7 4 6 5 ( V I ) 1 0 . 4 0 2 0 . 0 2 3 1 32 3 0 . 1 37 8 6 1 (VI1 1 0 . 2 8 f 0 . 0 3 2 1 5 f 3 0.00

~~~~ ~ ~

* 0 . 0 2 48 5 2 (V I ) 1 0 . 2 2 2 0 . 0 2 2 2 2 2 27 8 8 5 (VI) 1 0 . 1 3 f 0 . 0 6 2 05 22 0.026

i

The total amount of calcium leached from the stabilized soilsdecreased as slag content increased, reflecting the lower originalinventory of calcium. The cumulative leached fraction could not becalculated accurately because the original calcium concentrationsof the specimens were unknown. However, estimations of cumulativeleached fractions for comparative purposes were made using thecalcium content of the cement and slag used in the mixes andassuming a value of soil calcium content. The estimations from theresults for the Cr(V1) spiked soils showed that the approximatecumulative leached fraction of calcium was higher f o r the 4 0 %slag/60% cement than the 80% slag/20% cement mixes for the samesoil calcium content. Cumulative leached fraction of calcium wasalso estimated to increase with increasing soil content, the degreedepending on the assumed value of calcium in the soil. Theincreased leached fraction of calcium as soil content increased isdue to the corresponding higher water/cementitious material ratiosand associated greater leachability of cement rather than the soilcontent itself since estimations assuming calcium free soil showthe same trend.

2 . 3 . 4 Permeability CoefficientThe permeability coefficients of the grout treated Cr (111)soils after 28 days curing in dry soil are shown in Figure 9 . Theerror bars represent one standard deviation. Permeability gives ameasure of durability and resistance to penetration by mostaggressive species will improve as permeability decreases. Theresults show the expected trend of permeability increasing as soil


33/78

1o - 61o-

1o-81o-

1o - O

l o - 10 - l 2 1 2

SOI L/C EMENTITI

= 0% SLAG0 0% SLAGr g 60% SLAG[XI 80% SLAG

OUS MATERIAL5

RATIO

Figure 9. Permeability coefficient versus soil/cementitiousmaterial ratio for 200 ppm Cr(II1) soil treated with slag modifiedgrouts.23


34/78

and water content increases. At high soil contents th epermeability is probably controlled by flow through uncemented soilparticles. Any connected microcracks will also increasepermeability. Surface microcracking was not observed on the 80%slag mix with s/c = 5, although cracks were present on the largerTCLP specimens. The effect of slag replacement level onpermeability of the treated soil is variable, and is notsignificant at s/c = 5.

The permeabilities were reasonably low, thereby indicatingresistance to liquid penetration. Further work is necessary todetermine whether t he same low values can be achieved under fieldconditions.2.3.5 ComDressive StrengthThe mean 28 day wet cured compressive strengths and standard

deviations for the 200 ppm Cr(II1) soils stabilized with grout aregiven in Table 7. Figure 10shows the effect of slag replacementlevel and soil/cementitious material ratio on compressive strengthgraphically. The error bars represent one standard deviation.Strength decreases with increasing slag content when s/c = 1 and 2.At s/c = 5, strength tends to increase as slag content increases.The strengths of the stabilized soils are adequate for withstandingoverburden loads and occasional vehicular traffic.Compressive strengths were also measured on specimens preparedfrom the 40% slag replacement grout (74) after 84 days curing indry soil. Slight increases instrength occurred with time for the buried specimens, although dry

soil curing conditions were not ideal for cementitious materials.The results are given in Table 8.


35/78

nUaI

II-0zWII-v,Wv,0WlxaI00

W

L

4035

30

25

20151050

T

T

1 2SO IL/C EM ENT IT

0% SLAG0 0% SLAG40% SLAG6 0 % SLAG80% SLAG

5IOUS MATERIAL RATIO

Figure lo. Compressive strength versus soil/cementitious materialr a t i o f o r 200 ppm Cr (II1 ) so i l t r e a te d wi th slag modified grouts.25


36/78

Table 7 , 28 Day Wet Cured Compressive Strength of Stabilized Soil

~~ ~

47s580S1

Mix No.R

6.320.530.8k4.0

28 Day Compressive StrengthI (MPa)

805280S574S174S27465

20.7f2.16.3f0.435.4k3.720.523.26.0k0.5

47S1 32.922.8II

75S1758275s5

II 4762 I 23.2k2.2

29.2k3.115.6k0.97.6kO. 4

Mix No.74S174527455

84 Day Compressive Strength36.1k1.626.2fl. 28.9k0.4

(MPa)

I I8S2 13,550.6II 7865 I 8.0f0.3Table 8, 84 Day Soil Cured Compressive Strength of Stabilized Soil

2.3.6 Wet-Drv CvclinqThe grouted treated soils developed surface cracks after thesecond wet-dry cycle. Cracking is due to stresses induced bycycles of drying shrinkage and followed by swelling of the surfacelayer. Crack widths were greatest on the specimens with s/c = 5,

26


37/78

particularly those with 80% slag replacement for cement (7885).All of the specimens remained intact. Table 9 presents theresidual compressive strengths of the stabilized soils after 12wet-dry cycles.

478574S17452748575S1

Table 9. Residual Compressive Strength o f Stabilized Soil

1O.lfO. 533.3f3.022.3f3.97.220.534.7+1.*3

Residual Compressive Strength37.8k5.533.1kO. 3

7882 12.9fl. 67885 6 . 4 k 0 . 5

P

17.9f2.58.3f0.125.2k0.7

75S2758578S1

27


38/78

2-3.7 General Discussion of In-Situ StabilizationStabilization of hexavalent and trivalent chromiumcontaminated soils with slag modified grouts appears a viable one-

stage remedial option. The results indicate that leach resistance,impermeability and mechanical strength of treated soil can beachieved. While high soil loadings can be used, superiordurability and stability are expected with soil/cementitiousmaterial ratios less than 5. Water/cementitious material ratio ofthe grout is also a significant factor in controlling theperformance of the stabilized soil and lower values are favoured.A grout with 40% slag/60% cement and s/c = 1-2 gives the bestcombination of properties. Higher slag contents have lower leachedCr in TCLP tests and lower leached Ca in deionized water tests, butobservations of shrinkage cracking in some specimens suggests thathigh slag contents may be undesirable, Field testing withappropriate mixing and placement equipment is necessary to furtheroptimize grout mix proportions and waste loadings.

The ability of the soil stabilized with slag modified mixes topass TCLP tests even at high soil and water contents .also signifiesthat the grout treatment process has a forgiving nature. Temporaryaberrations in process control, within limits, may not necessarilyobstruct successful stabilization.Use of slag in treatment grouts may also contribute 'benefitssuch as resistance to sulphate attack. For treatment of thechromium plume the temperature and heat generated by grout must beconsidered since this may result in thermal cracking. In theproposed remediation, grout would be mixed in-situ with

contaminated soil to give an array of interconnected columns, eachapproximately 1 to 2 m diameter and up to 25 m deep, depending onplume depth at a given location. If heat dissipation is inhibitedby surrounding soil, excessive thermal stresses may eventuate,particularly if conditions approach adiabatic. Addition of slagand optimization of soil/grout ratio to reduce heat generation, butretain adequate toughness, will reduce the risk of thermalcracking.2.3.8 Recommendations for Future Work on In-SituSeveral aspects of in-situ stabilization of chromium

contaminated soil with slag modified grouts require investigationbefore full-scale remediation can be implemented. The data give afoundation that justifies further fine tuning so that the processcan be applied with the greatest degree of success.

Stabilization

The aspect of greatest importance in transferring the slagmodified grout treatment to full-scale use is field trials todetermine the optimum grout flow properties (and hence,water/cementitious material ratio) and equipment parame%ers for the


39/78

chosen placement method. This would best be achieved throughconducting field trials on grouts with different mix proportionsand varying the placement equipment parameters in a systematicmanner. An uncontaminated site with soil conditions similar tothose of the actual landfill is required for these field trials.The dimensions and quality of the stabilized soil would then bedetermined and correlated with the materials and equipmentparameters used. Mixes 74-1 and 74-2 are recommended for fieldtrials. Variations of these should also be tested, with theobjective of minimizing the water/cementitious material ratio thatis compatible with the placement technique and final productquality.

The field trials must be conducted in conjunction with anexperienced, professional grouting contractor with appropriatemixing and placement equipment. Successful remediation and optimumvalue from field trials will require full cooperation from asuitably qualified contractor, since success not only depends onthe grout material but also the care and c:ompetency with which itis placed. It is also recommended that a contractor from privateindustry be used so that technology transfer i s enhanced. Jetgrouting and deep s o i l mixing are the recommended placementtechniques. It is emphasized here, and in a future section of thisreport, that permeation grouting with the grouts used in this workis inappropriate forthe site conditions and totally unsuitable forstabilization of the chromium plume.Simplification of the grout batching process should also beinvestigated. Type IS cement is a blend of ground granulated blastfurnace slag and Portland cement in which the slag comprises 25 to75% of the total mass. These amounts are within the range of themixes studied. While 20% slag replacement: did not convert all ofthe hexavalent chromium, a value of 25% nay give better results.If Type IS cement is successful in stabilizing Cr(V1) contaminatedsoil it would not be necessary to add cement and slag in separatestages, and this would simplify grout production. A concern withType IS cement is that the variable slag content may give variablestabilization capacity. By performing TCLP tests on soilsstabilized by Type IS cement with different slag contents, thesuitability of this cement can be evaluated.Further durability tests of the stabilized soil should be

conducted. In particular, resistance to deterioration caused bysoluble sulphates and sulphur oxidizing bacteria that may bepresent in the soil requires investigation to ensure that longevityis maximized.Specimens of landfill soil stabilized with Mix 74 grout havebeen archived for future study. Some have been crushed and arecurrently being exposed to air for repeat TCLP tests.Stabilization of soils contaminated with hlCgher Cr levels should beperformed to ensure that the grouts are capable of treating any

29


40/78

high localized concentrations in the landfill.3.0 IN-SITU CONTAINMENT OF STABILIZED CHROMIUM PLUME3.1 Introduction

Grout based subsurface containment barriers continued to beinvestigated in FY 94. Thefirst is for placement around the CWL chromium plume and assumesthat the plume is stabilized either before or after barrierinstallation. Thus, it is assumed that the barrier will not beexposed to chromium leachants. The barriers will be placed inuncontaminated soil and therefore, do not require TCLP testing.The CWL barriers under consideration are either soil cementproduced by in-situ mixing of soil and grout and monolithic groutproduced by pumping grout into a cavity in the soil. The secondtype of barrier is for containment of tritiated water at the MWLand this is dealt with in Section 6.0.

Two types of barriers are considered.

Several aspects of subsurface barriers for the CWL wereexamined. Work on saturated permeability of cracked barriers andthe effect of fibre reinforcement initiated in FY 93 was continuedand completed. Further details of this work are available in theFY 93 Annual Report. Core samples from th e FY 93 permeationgrouting field trials were tested for strength and permeability.Jet grouting field trials were conducted at SNL. Addition of slagto grouts f o r subsurface barriers was also investigated. Worstcase scenario tests were conducted to evaluate t he consequences ifthe chromium is not stabilized.3.2 Experimental

3.2.1 MaterialsThe materials used for the CWL barriers are th e same as thosedescribed in Section 2.2.1, except for Mix 70 which contained sand.Mix 70 is intended for a monolithic grout barrier produced byreplacing soil with grout. The other grouts are intended f o r in-situ mixing with soil. Bentonite was added to prevent settling ofsand and cement. In practice, if the grouts are mixed in a highshear mixer and pumped directly after mixing bentonite may bereduced. The mix proportions of the grouts are presented in Table10. The letters IIf11 and "c" refer to fine and coarse grades of

sand respectively and are described further in the FY 93 AnnualReport.

30


41/78

Table 10. Mix Proportions of CWL Barrier GroutsMix Cement Slag47-1 1221 047-2 948 074-1 722 481

No. (ks/m3) (k9/m3)Sand Water Bentonite SP (l/m3)0 586 11.1 24.40 679 13.3 19.00 577 10.9 24.1

(k9/m3) (k9/m3) (k9/m3

The mix proportions of the soil cements produced by mixinggrouts with dry, uncontaminated site soil are presented in Table11.T a b l e 11. Mix Proportions of CWL Soil Cement Barriers

Fibres were added to Mixes 47S1, 47S2 and 70 in a continuationof work from FY 93. Polypropylene collated fibrillated andmonofilament fibres supplied by Forta Corporation were used. Bothtypes of fibres were 19 mm in length. The collated fibrillatedbundles open during mixing to form a three-dimensional network.The fibre network has improved mechanical bonding under tension ascompared with monofilament fibres due to stress transfer throughoutthe fibrils, rather than solely at the fibre/matrix interface. Thevolume fraction of fibres was 0.2%. Fibrillated fibre reinforcedmixes are denoted by the suffix I10.2F1l nd monofilament fibres aredenoted by 'IO. 2MF".3.2.2 SDecimen PreDarationNeat cement grouts and soil cements were mixed in a planetary

3 1


42/78

mixer as described in Section 2.2.2. The sanded grout (70) wasalso mixed in a planetary mixer with sand being added last. Fibreswere added to grouts after cement and any sand were mixed. Thefibres were added in increments in an attempt to prevent clumpingand mixed for 2 minutes. Fibrillated fibres tended to clump andtangle during mixing. Less mixing problems were encountered withthe monofilament fibres.

Fibre reinforced soil cements were prepared by adding thefibre reinforced grout to the soil and mixing. The volume fractionof fibres was based on the soil cement volume. Monofilament fibreswere used with 47S2, and both fibre types were used with 47S1 and70.3.2.3 CurinqThe two types of curing, wet and soil, as described in Section

2.2.3, were used for the barrier materials. Strength andpermeability tests were performed on the 74 series of soil cementscured for 28 days in water. The same soil cements were tested forpermeability after three months soil curing. Permeability wasmeasured on plain and fibre reinforced 47S1-5 and 70 after fourmonths soil curing. The results of wet cured properties for 47S1-5and 70 are reported in the FY 9 3 Annual Report.3.2.4 Unhardened Grout ProDertiesThe rheological properties and specific gravities of the 47series and 70 grouts are described in the FY 93 Annual Report andthe properties of the 74 series are described in Section 2.3.1

above.3.2.5 Permeability CoefficientThe water permeability coefficient (hydraulic conductivity) ofthe 47 and 74 series of soil cements and Mix 70 under saturatedconditions was measured in a flexible wall triaxial cellpermeameter on specimens 75 mm diameter and 105 mm long. Thepermeant was de-aired tap water. The applied pressure gradient was207 kPa and the confining pressure was 414 kPa. The chosen valueof confining pressure was based on the maximum bearing pressure ofinterest. Three specimens per batch were tested and all specimenswere saturated prior to measurement. Measurements conducted over

pressure gradients from 69 to 207 kPa on cracked and uncrackedspecimens indicated that Darcy's Law of flow through porousmaterials was applicable.The effect of confining pressure on measured permeability forcracked specimens was also determined at 310, 345 and 414 kPa.In the first series of tests on plain and fibre reinforcedgrouts and soil cements conducted in FY 93, the specimens were


43/78

cured in water for 28 days. Permeability coefficient was measuredand the specimens were then subjected to 23 wet-dry cycles beforere-saturation and re-measurement, The cycles consisted ofsubmersion in water at 25OC for 5 hours followed by drying in airat 25OC and relative humidity of 40-50% for 43 hours.

In FY 94 permeability coefficient before and after drying inair at 25OC and 40-50% relative humidity for three months wasmeasured on the plain and fibre reinforced 47 series of soilcements and the 70 grout. Field curing conditions were simulatedby burying the specimens in site soil for four months. Saturationwas performed before measurement of permeability coefficient. Thepurpose of the drying tests was t o eliminate the crack healing thatarises during wet-dry cycling when calcium hydroxide leaches outof the grout or soil cement and reacts with carbon dioxide to formcalcium carbonate. This enabled comparison of material performanceunder conditions that do not favour crack healing.The permeability coefficient of 7481-5 was measured after 28days wet curing and then after 12 wet-dry cycles and re-saturation.Permeability after 84 days soil curing and saturation was alsomeasured on 7461-5.3.2-6 Unconfined Cornm-essive StrenathCompressive strength measurements were performed on the CWLbarrier materials using the same method described in Section 2.2-9.The strength properties of 4781-5 and 70 are described in the FY 93Annual Report. Strength of 7481-5 was measured after 28 days wet

curing,3.2.7 Chromium Diffusion TestsThe diffusion of chromium through discs of grout and soilcement was investigated, The purpose of the experiment was tosimulate a worst case scenario where unstabilized chromium withinthe plume is mobile and contacts the barrier. The response of thebarrier material to direct contact with chromium was studied.Chromic acid solution with a concentration of 3 . 8 ~ 1 0 - ~ (200 ppmCr(V1) with a pH of 2.51 was used to accelerate the aggressivenessof the test. Chromic acid is known to be deleterious tocementitious materials (Biczok, 1972). Two soil cements (74S1 and

74S2) were tested.The diffusion cell used is shown in Figure 11. Two chambersconstructed from uPVC pipe were separated by a disc of grout orsoil cement. The chambers had polymethylmethacrylate caps withscrew in plugs to enable filling and sampling, The discs had adiameter of 56 mm and a thickness of approximately 10-11 mm.Specimens were cast in 76 mm diameter cylinders and trimmed to fitthe cell. The actual dimensions of the disc were measured before

33


44/78

Grout/Soil Cement

Non-Active

Figure11.and tritium diffusion experiments.Schematic diagram of diffusion cell used for chromium34


45/78

placement in the cell. The joint between the discs and cell wallswere sealed with a silicone rubber compound. One side (active) ofthe diffusion cell contained 375 ml of the chromic acid solutionand the other side (non active) contained 375 ml of distilledwater. Samples of 1 ml were withdrawn from the non-active side atperiodic intervals for Cr analysis by atomic absorptionspectrophotometry. The cell contents were stirred before sampling.

Bentonite SP ( i / m 3 )icrof ine Water (kg/m3)Cement (kg/m3)

3.2.8 Permeation GroutinqThe details of the permeation grouting field trials at S N L inFY 93 are presumably available from S N L . As discussed in the FY 93Annual Report, BN L believed that permeation grouting would have alow probability of forming a continuous barrier in a controlled andpredictable manner due to the fine particle size of the site soil.However, a microfine cement grout was used in field trials. Themix proportions of this grout are given in Table 12 and theunhardened properties are given in Table 13. The flow time wasmeasured using an ASTM C 939 flow cone, and should not be confusedwith a Marsh funnel. A pulverized grout was also used in the fieldtrials, but was not part of the BNL program, nor were othermaterials such as sodium silicate and wax.

717 I 717

T a b l e 1 2 . Mix Proportions of Microfine Cement Grout

14 4 1 14.3Viscosity (cp) ASTM Flow Time (s)

9 9 . 4Specific Gravity

1.47Core samples of the grouted soil were shipped to BNL inNovember, 1993. The age of the specimens was approximately threemonths and the core diameters were 84 mm. The cores were soaked inwater for compressive strength testing and saturated in water for

permeability measurements.3.3 Results an8 Discussion

3.3.1 Permeabilitv Coefficient of Slaq Modified Soil CementsThe permeability coefficients for the 28 day wet cured and the8 4 day soil cured soil cements prepared from Mix 74 are given inTable 14. The mean value and one standard deviation are indicated.

35


46/78

The results are also shown graphically in Figures 12and 13.Table 14. Permeability Coefficients of Slag Modified 8011 Cements

Mix74S174827485

28 Day Wet Cured 84 Day Soil CuredPermeability (xlO-" cm/s) Permeability (xlO-" cm/s)1.7 k 0.1 5.7 f 0.83.0 k 0.9 21 f 512 f 1.0 26 f 8

The results can be compared with those for the soil cementswithout slag (47 series) measured with the same instrument in FY93. The permeability coefficients of 28 day wet cured 47S1 and47352 were 6 . 0 ~ 1 0 - ~ ~0.5x10-'' cm/s and 7.5x10-" f 0.8~10-'' cm/s,respectively. This indicates that slag improved the impermeabilityat s/c ratios of 1 and 2 for wet curing conditions. When the soilcements were cured by burial in soil for 84 days the permeabilitieswere higher, but remained below the cm/s limit. Thepermeabilities were similar for s/c = 2 and 5. The increase inpermeability when the materials are cured in soil is due to reduced

Mix745174827485

hydration of cement.

Permeability After 12 Wet-Dry Cycles(cm/s)

3.3~10-~ 1 ~ 1 0 - ~i.o~io-~ 2 ~ 1 0 - ~2 . 8 ~ 1 0 ~ ~4 ~ 1 0 - ~

The permeability coefficients of12 wet-dry cycles are presented ininitial values graphically in FigureTable 15. Permeability Coefficients

Soil Cements After Wet-Dry

the wet cured specimens afterTable 15 and compared with14.of Wet Cured Slag ModifiedCycles

The standard deviation of the permeability coefficients afterthe specimens were subjected to wet-dry cycles is high and this isdue to heterogeneity of cracking. The soil cements with s/c = 1and 2 remained below the E P A permeability limit of cm/s.Although the laboratory testing conditions do not replicate exactlythe field conditions to which the barriers will be exposed, thesusceptibility of Mix 7485 to formation of permeable cracks limitsi t s usefulness as a barrier material. Furthermore, after testingthe specimens of 7485 were maintained in air and developed surfacedeterioration in the form of flaking after 4 months. Therefore,

3 6


47/78

nv)\EvIZW0LLW00>-J

W

--

tma-wtEWa

1o-8

1 o-

1 o-O1 2 5SO1L/C EM ENTlT lO US MATERIAL RATIO

Figure12.after 28 days wet curing.Permeability coefficient of slag modified s o i l cements37


48/78

1o -8

1 o - O

T

T

1 2 5S0 IL /C E M ENT ITI0US MATERIA L RAT O

Figure 13.after 84 days soil curing.Permeability coefficient of slag modified soil cements38


49/78

this mix with its high s/c and w/c values is not recommended forthe CWL barriers.

Mix 112 Day Soil CuredPermeability (x10-locm/

70 2.4 f 0.170 0.2F 5 . 8 f 0.37 0 0.2MF 2.5 f 0.6

3.3.2 Permeability Coefficient of Soil Cured Grout Before andFfter Drving

Permeability AfterDrying and Re-saturation (.x10-~' m/s)3.6 f 1.28 . 8 f 0 . 88.5 k 2.0

The permeability coefficients of the plain and fibrereinforced versions of Mix 70 after four months curing in soil andsaturation are shown in Table 16and Figure 15. The codes 0.2F and0.2MF refer to 0.2% fibrillated fibres and 0.2% monofilament fibresrespectively. The permeabilities for the same specimens afterdrying in air for three months and re-saturation are also given.T a b l e 16. Permeability Coefficients of Plain and Fibre ReinforcedGrout Before and After Drying

The extent of microcracking on the dried specimens appearedless than that for the wet-dry specimens tested in FY 93 (seeAnnual Report) and crack healing did not occur. After drying andre-saturation, the permeability coefficients of the fibrereinforced grouts are greater than those for the plain grouts. Forthe wet cured and wet-dry cycle specimens reported in FY 93, thepermeability coefficient of the monofilament fibre reinforced grout(70 0.2MF) was significantly higher than the plain grout (9.2xlO-'cm/s compared with 1.4x10-' cm/s) .The initial permeability coefficient for the grout with 0.2%fibrillated fibres is higher than the plain and monofilament fibrereinforced materials and this is possibly due to batch variation.The higher initial permeability coefficient of the fibrillatedfibre reinforced grout may obscure the crack resistance if

permeabilities are compared on absolute terms. The relativeincreases were SO%, 52% and 240% for the plain, 0.2% fibrillatedfibre and 0.2% monofilament fibre reinforced mixes, respectively.Therefore, the addition of monofilament fibres to the grout resultsin reduced performance for the mix and conditions studied.

40


50/78

20

15

10

5

0

- I-INITIALFINAL

T

PLAIN 0.2F 0. IMF

Figure 15. Effect of 0.29 Fibrillated (0.2F) and 0.2% MonofilamentFibres (0.2MF) on Permeability Coefficient of S o i l Cured GroutBefore and After Drying and Re-Saturation.41


51/78

3.3.3 Permeability Coefficient of Soil Cured Soil CementsBefore and After Drvinq

Mix

47S147S1 0.2F47S10.2MF

Table 17 and Figure 16 shows the permeability coefficientbefore and after drying and re-saturation for the soil cured soilcements. In Figure 16 S1 and S2 refer to 47S1 and 4782,respectively.

112 Day Soil CuredPermeability (x10-lo

2.4 f 0.22.1 f 0.72.8 2 0.3

cm/ s

Table 17. Permeability Coefficient of Plain and Fibre ReinforcedS o i l Cements Before and After Drying

476247820.2MF

16 2 1.123 f 1.0

Permeability AfterDrying and Re-saturation ( x ~ o - ~ O cm/s)3.5 f 1.03.5 f 1.018 f.3.050 f 1.0150 f 30

The four month soil cured permeabilities were all low,indicating suitability for barrier materials if the same propertiescan be achieved in the field. The initial values of permeabilityshow that 47S2 has greater sensitivity to curing conditions than47S1. After drying and re-saturation the higher permeability of4782 signifies reduced crack resistance compared with 47S1. It isproposed that soil curing followed by wet-dry cycling would resultin a higher f nal permeability coefficient than 5. Oxlo-' cm/s sincewet-dry cycles are more aggressive than continued drying. Thus,superior impermeability under such exposure conditions wouldrequire use of a tough material which is less sensitive toinadequate curing conditions. The CWL conditions are not expectedto be as aggressive as those encountered in repeated wetting anddrying used in the laboratory studies.

For 47S1, the final permeability coefficient after soil curingfollowed by drying was not altered significantly when 0.2%fibrillated fibres were added and the mean value for both the plainand fibrillated fibre mixes was 3.5~10-~' m/s. Addition ofmonofilament fibres to 47S1 resulted in a steep increase in finalmean permeability coefficient (1.8x10-' cm/s) as compared with theplain and fibrillated fibre reinforced mixes. Since all of themixes with s/c = 1 had similar magnitudes of initial permeability,

42


52/78

IO-

c

1 o- *

T T

s1 s1PLAIN 0.2F

I

11 I20.2MF PLAIN 0.2MFFigure 16. Effect of 0.2% Fibrillated (0.2F) and 0.2% MonofilamentFibres (0.2MF) on Permeability Coefficient of Soil Cured SoilCement Before and After Drying and Re-Saturation.

4 3


53/78

the increased final value of the fibre reinforced material cannotbe attributed to batch variation.

Mix No. 28 Day Wet Cured Compressive Strength(MPa)

74S1 35.0 f 4.27482 22.1 2 2.5

. 74s5 6.3 f 0.4-

Addition of 0.2% monofilament fibres to 4782 resulted in afinal mean permeability coefficient of 1.5~10-~m/s compared with5.0x10-' cm/s for the plain version. The final permeabilitycoefficients of the monofilament reinforced soil cements weresignificantly higher than those for the unreinforced or fibrillatedreinforced versions.

3.3.4 Effect of Confinina PressureThe relationship between confining pressure and permeabilitycoefficient shown in Figure 17 indicates that the confiningpressure applied in the permeameter tends to close microcracks.This phenomenon has also been reported by Daniel et al. (1985).The implication is that the saturated flow permeabilitycoefficients of the microcracked materials would be higher forunconfined conditions. Thus, the overburden pressure to which thebarrier is exposed will influence the permeability when microcracksare present. Since the confining pressure was kept constant forall of the test results presented in Tables 16and 17, the relativeeffect of crack closing should be similar for all materials.3.3.5 Compressive StrensthThe compressive strengths of plain and fibre reinforcedversions of 47S1, 4782 and 70 were presented in the FY 93 AnnualReport. The 28 day wet cured strengths of the slag modified soilcements are presented in Table 18 and Figure 18. Comparisonbetween Table 18and Table 7 shows that the strengths are similarfor the 40% slag/60% cement treated soils with or without chromium.Therefore, 200 ppm Cr(II1) does not appear to have a significanteffect on compressive strength.

T a b l e 18. Compressive Strength of Slag Modified Soil Cements

3.3.6 Chromium DiffusionAt the time of writing, no chromium has been detected in thenon-active side of the soil cement diffusion cells after 90 days.

44


54/78


55/78

403530

2520

151050

1T

1 2 5S0 IL/C EMENT IT IOUS MATER IA 1 RAT10

Figure 18.ratio for slag modified soil cementsafter 28 days wet curing.Compressive strength versus soil/cementitious material

4 6


56/78

3.3.7 Permeation Groutina CoresSix core samples from the permeation grouting field trials atSNL were received. Five of the cores were soil grouted with the

pulverized cement and the sixth was of the soil grouted withmicrofine cement. Examination of the cores showed that the groutsuccessfully penetrated in areas where the soil had a gravellytexture. The soil was significantly more coarse than samplesshipped to BNL. Areas of finer soil in the cores were notpenetrated by either grout. The uncemented soil crumbled duringcutting of the samples. The cores exhibited coarse air voids andfiner voids within the grout phase. For the small core ofmicrofine cement treated soil, it was evident that the grout hadcaused fracture of the soil, and that the grout penetrated thefractures in preference to the fine soil. Thus, the groutpermeation appears to be inhomogeneous.Two of the pulverized grout cores were long enough forcompressive strength tests. The measured strengths were 28 and 33MPa, which is relatively high and more than adequate for thepurpose. The density of the samples was around 2300 kg/m3. Thehigh strength was attributed to the high grout content of thecores. Areas where grout does not penetrate the soil are expectedto have lower strengths.

The permeabilityof two cores containing the pulverized cement grout were 5.3x10-'and 2.7x10-' cm/s. The one core of microfine cement grout treatedsoil had a permeability of 8.4x10-' cm/s. While it is not possibleto draw definitive conclusions from the small numbers of samplesobtained from a grouted cobble layer, it is evident that themicrofine cement grout does not meet the impermeabilityrequirements for subsurface barriers. This concurs with the visualobservations of incompletely grouted soil within the sample. Thepermeability of the soil grouted with pulverized cement wasreasonably low but is meaningless if the grout does not penetratein a uniform and predictable manner. The results also suggest thatalternative placement techniques such as jet grouting and soilmixing using the materials developed in this program must be fieldtested.

Permeability tests on cores were performed.

3.3.8 General Discussion of In-Situ Containment of ChromiumPlumeIt is recognized that mixing and exposure conditions in thefield will vary somewhat from those used in these laboratoryexperiments. Grouts with partial replacement of cement with slagappear suitable for producing soil cement barriers.Soil/cementitious ratio and water/cementitious material ratio mustbe kept as low as practical to enhance performance. Addition of


57/78

slag may also provide benefits of reduced heat generation andsulphate resistance as discussed in the section on stabilization,but this requires confirmation.The practicality of placing fibre material reinforced barriersrequires further investigation before use of fibres can berecommended. For simulated in-situ curing followed by drying inair, fibrillated fibres do not result i.n improvement of crackresistance and this may be due to reduced interfacial bonddevelopment associated with less extensive hydration. Monofilamentfibres were detrimental to permeability when microcracks formed insoil cured materials. It is proposed that the observed increasedfinal permeability coefficient associated with monofilament fibrereinforced materials is due to high conductivity paths developedwhen interconnected microcracks form between fibres and alongfibre/matrix interfaces. The failed interfacial surface length islikely to be greater for monofilament than fibrillated fibrereinforced materials. This is because tensile stresses will tendto result in failure along the entire length of a monofilamentstrand, whereas stress transfer across fibrils will occur infibrillated fibres and reduce the length of fibre disbondment.For field conditions additional factors, such as thepermeability of the soil surrounding the barriers and thesaturation level of the barrier itself, also influence thecriticality of cracks. These factors have been considered byWalton and Seitz (1992). Cracks may be less catastrophic for anarid, unsaturated environment than for saturated conditionsdepending on water tension within cracks. Thus, flow through acracked barrier in unsaturated conditions may be controlled by thepermeability of the uncracked matrix. Conversely, flow through asaturated cracked barrier will be controlled by crack permeability.Hence, the results obtained for saturated flow on uncoveredmaterials require some qualification before application to anunsaturated, backfilled environment. The presence ofpolypropylene/matrix fracture surfaces may complicate unsaturatedflow due to altered surface tension as compared with matrixfracture flow paths.3.3.9 Recommendations for Future Work on In-Situ Containmentof Chromium Plumes

As was the case for the stabilized chromium contaminated soil,the CWL barrier materials should be tested for durability tosulphate attack and sulphur oxidizing bacteria. The chromiumdiffusion experiments will continue to be monitored until adiffusion coefficient can be calculated.It is proposed to investigate the hydraulic behaviour of theCWL type subsurface barriers for unsaturated conditions. Thepermeability studies conducted to date on cracked and uncrackedgrout and soil cement barriers have been performed under saturated

48


58/78

conditions. Inboth instances sufficiently low permeabilities haveusually been measured. In practice, permeability coefficients aregreater under saturated conditions than unsaturated, hence theobtained results provide a "worst-case scenario1' or the typicallyarid conditions at SNL. However, if the barriers contain cracksflow will depend more significantly on saturation levels. Forsaturated conditions flow will occur through cracks in preferenceto the uncracked matrix, whereas the opposite will probably occurin unsaturated conditions and the impact of cracks on performancemay be insignificant. Hence, the results for saturatedpermeability of microcracked barriers also represent an upperlimit. By measuring the effect of cracking on flow for unsaturatedconditions a better indication of the exact impact of cracks onbarrier performance under the typical site conditions will beachieved.

Future work for CWL barrier field trials is discussed inPermeation grouting with cementitious materials is notection 4.recommended.

4 . 0 J E T G RO UT IN G F I E L D T R I A L SThe BNL component of the FY 94 jet grouting field trials at

SNL was to provide a test plan, grout formulations and conductmeasurements on core samples. The trials were initiated on 30July, 1994 at an uncontaminated site close to the CWL.In the original test plan dated 19 April, 1994 a total of fourgrouts were identified for test. Both columns and laminar panelswere to be produced. The objective of investigating panels was to

determine whether reduced wall thickness could be achieved inpractice. The grouts consisted of an ordinary Portland cementbased material and a slag modified material, at two different watercement ratios. The mix proportions were similar to those for 47-1,47-2, 74-1 and 74-2, but had lower w/c ratios.It was stipulated in the original test plan that grout shouldbe used as a drilling fluid in preference to water since drillingwith water would effectively dilute the grout and change the w/c inan uncontrolled manner.The target column diameter was 1.0 m and tests were necessaryto determine what diameters could be achieved in the site

conditions. The original desired column length was 5 . 0 m. It wasalso planned to place adjoining columns so that the barriercontinuity could be studied.The extent of the field trials and the column heights werereduced from the original plan. The jet grouting was performed bythe Westinghouse Hanford Company using Casagrande Jet 5 equipment.Other grouts were evaluated at the same site, but these were notpart of the BNL component.

4 9


59/78

A major obstacle to the jet grouting trials was the absence ofa suitable grout mixer. BNL had assumed that either a high shearor paddle grout mixer of suitable capacity would be available,since this is a standard and fundamental piece of equipment forgrouting operations. Production of the type of high quality groutnecessary for environmental remediation or other high performanceapplications requires an efficient mixer. Also, since theobjective was t o test the grouts under field conditions, a groutmixer of the type routinely used by the grouting industry wasdesired.

It was suggested to BN L by S N L and Westinghouse Hanfordpersonnel conduction the trials that a concrete ready mix truckcould be used for grout mixing. This suggestion was rejected fora variety of logistical reasons (e.g. change in grout fluidproperties during transit, possible need for retarder) andpractical reasons. Ready mix trucks are designed for mixingconcrete containing coarse aggregate. It was believed that theneat cement grouts would separate into two components: a pastestuck to the barrel walls and a liquid comprised predominantly ofwater. On 28 and 29 July, 1994, work mot related to the BN Lcomponent was conducted by Westinghouse Hanford using a groutcomprising of 1 part cement to 1 part water by weight, mixed andtransported to the site in a ready mix truck. It was observed thatthe grout was non-uniform, with lumps of cement paste and waterygrout being ejected at different times. 'I'his type of ineffectivemixing and poor quality grout is inappropriate for CWL barriers.As a compromise, small scale mixing was improvised using adrill mixer in two 200 1 containers. The mixed grout was thentransferred to another 200 1 drum and recirculated by a centrifugalpump. Houlsby (1990) describes this type of mixing as Itrelativelycrude and is only tolerated on very low standard, cheap groutingt8.This should be kept in mind when evaluating the results of the jetgroutingtrials. Proper mixing would improve the properties of t h egrouts and grouted soil.The mix proportions of the grouts used in the jet groutingfield trial

icdf stabilization

Documents