'Bleed Water Testing Program for Controlled Low Strength Material

Bleed water measurements for two Controlled Low Strength Material (CLSM) mixes were conducted to provide engineering data for the Tank 20F closure activities. CLSM Mix 1 contained 150 pounds of cement per cubic yard whereas CLSM Mix 2 contained 50 pounds per cub yard. SRS currently used CLSM Mix 2 for various applications. Bleed water percentages and generation rates were measured along with flow and compressive strength. This information will be used to select a mix design for the Tank 20F closure activities and to establish the engineering requirements, such as, lift height, time required between lifts and quantity of bleed water to be removed from the tank during the placement activities. Mix 1 is recommended for placement within Tank 20F because it has better flow characteristics, less segregation, lower percentage of bleed water and slightly higher strength. Optimization of Mix 1 was beyond the scope of this study. However, further testing of thickening additives, such as clays (bentonite), sodium silicate or fine silicas maybe useful for decreasing or eliminating bleed water.


BLEED WATER TESTING PROGRAM FOR CONTROLLED LOW STRENGTH MATERIAL (CLSM)
.      ., Bleed water percentages and generation rates were measured along with flow and compressive strength. This information will be used to select a mix design for the Tank 20F closure activities and to establish the engineering requirements, such as, lift height, time required between lifts and quantity of bleed water to be removed from the tank during the placement activities.
Mix 1 is recommended for placement within Tank 20F because it has better flow characteristics, less segregation, lower percentage of bleed water and slightly higher strength. Optimization of Mix 1 washyond the scope of this study. However, further testing of thickening additives, such as clays (bentonite), sodium silicate or fine silicas may be useful for decreasing or eliminating bleed water.

EXPERIMENTAL PROCEDURE
Two mixes were prepared by RaytheonlEBASCO personnel at the concrete testing facility in N-Area. The ingredients and proportions of the mixes used are given in Table 1. A three cubic foot paddle mixer was used for the batching of the CLSM mix designs. In addition, 0.25 cubic foot batches were prepared under laboratory conditions to provide better control of the mixing proportions and curing. Preparation of the CLSM mixes was in accordance with ASTM D4832. Flow (evaluated per ACI 229) and temperature was measured on each of the batches prepared for the bleed water experiments.
After mixing, the CLSM mixes were poured into one of three 55 gallon drums. Pour heights in the drums were 1, 2, and 3 feet, respectively. The drums were covered after filling to prevent evaporation. Bleed water was measured as a function of elapsed time after pouring. The measurements were made by siphoning the water off and measuring the volume. The water was returned to the drum after each measurement to monitor changes versus time. The testing was done in accordance with the testing procedures identified in Attachment "A, . Laboratory samples were cast at the Savannah River Technology Center (SRTC) in sealed 12 inch by 41/2 inch seethrough plastic cylinder molds. Bleed water was pipetted off of the molds for measurements.
The water was returned to the molds after each measurement to monitor changes versus time. ,

Laboratory Flow Experiments
Flow was evaluated on a small scale. Samples were prepared under laboratory conditions. The flow results for Mix 1 and Mix 2 containing the maximum water content were about 11 and 9-314 inches respectively. (See Table 2  with minimum water content ranged from 10-1/2 to I 1 inches. The maximum water content showed a range from 11 to 12-1/4 inches. The spread diameter for Mix 2 ranged from 9-318 to 10-112 inches for the minimum water content and 11 to 11-114 inches for the maximum water content. A Slight amount of segregation was observed from Mix 2 but not from Mix 1.

Bleed Water Laboratory Experiment
Bleed water is the migration of water to the top surface of freshly placed CLSM caused by the settlement of the solid materials (cement, sand and fly ash) within the mass. Essentially, the water within the placed CLSM 'bleeds" to the surface of the placement. Settlement is a consequence of the combined effect of vibration and gravity or just gravity alone.
Bleed water was measured for Mix 1 and Mix 2 prepared with the maximum water contents under laboratory conditions. After 24 hours of curing, 11.1% of the initial mixing water segregated as bleed water for Mix 1 and 11.8% for Mix 2. After 4 days, a small amount of the bleed water was reabsorbed into Mix 1 (1 0.5% bleed water) and Mix 2 (1 I .6% bleed water). These indications mean that Mix 2 absorbs less bleed water than Mix 1.

55 Gallon Drum Bleed Water Experiment
Bleed water measurements for Mix 1 with minimum and maximum water content are presented in Tables 4, 5 and 6 for heights of 1, 2 and 3 foot placements respectively. The results for Mix 2 with a maximum water content and a 3 foot placement are provided in Table 7.
After 24 hours of curing, the percentage of bleed water generated from Mix 1 was 9.89% (average) with minimum water content and 13.0%. (average) for maximum water content. Some bleed water was reabsorbed for all pour heights evaluated between the initial measurement and the day 4 measurement. After 4 days, the percentage of bleed water did not significantly change. The average values derived from the placements for Mix 1 (1,2 and 3 foot placements) ranged from 8.7% (minimum water content) to 12.1% (maximum water content). (See Figures I , 2 and 3

.)
After 24 hours for the test performed for Mix 2,14.2% bleed water was measured. After 4 days, 13.4% was measured.

Compressive Strength
The compressive strength results for Mix 1 and Mix 2 are presented in Table 3 and Figures 8 and 9. The strength was measured after 7 and 28 days of curing. Mix 1 showed higher strength at 7 and 28 days than Mix 2 for both the minimum and maximum water contents.
All mixes tested met the SRS recommended requirements for CLSM of 30 pounds per square inch to 150 pound per square inch after a 28 day curing period. However, Mix 2 with the maximum water content showed an average strength that just met the requirements.

Compressive Strength, Flow, And Bleed Wafer Properties
The strengths of Mixes 1 and 2 are directly related to the cement and inversely related to the water content. More cement results in higher strengths. Consequently, Mi x 1 with 150 pounds of cement per cubic yard has higher strengths at 7 and 28-days than does Mix 2 (50 pounds of cement per cubic yard). Mix 2, made with the minimum water content of 62 gallons per cubic yard, has higher strength than the same mix design made with a maximum water content of 66 gallons per cubic yard.
Flow was found to be directly dependent on the cement and water content. Greater flow was achieved by increasing both the cement and water content. Mix 1 had more flow than Mix 2 for equivalent water contents. This trend was observed on both the taboratory and field prepared mixes (55 gallon drums) (See Table 2.). The lower flow measured in the laboratory samples is attributed due to a lower moisture content of the sand.
The equations for predicting flow between the 62 to 66 gallons per cubic yard m g e are shown in Figure 11 for Mix I and Mix 2. For the same amount of water, Mix 1 has 10 percent more flow than Mix 2. Additional water must be added to Mix 2 in order to increase the flow of Mix 2 by 10 percent to make it equivalent to Mix I . For batch plant proportioning, the amount of additional water is approximately 5 gallons per cubic yard.
Bleed water will be apparent since the addition of more water will cause segregation. Therefore, adding extra water to Mix 2 to achieve better flow characteristics comparable to that of Mix 1 is not recommended.
The percentage of bleed water is also a function of the amount of mixing water and the amount of cement. Mixes with higher water contents will generate a higher percentage of bleed water. CLSM mixes with higher cement content will generate a lower percentage of bleed water.
Bleed water results were confirmed in the laboratory experiments. The lower percentages of bleed water measured in the laboratory samples is attributed to a slightly lower moisture content of the sand used compared to that used in the experiment with the 55 gallon drums.

Tank 20F Closure
The amount of bleed water is proportional to the water in the design mix and the volume of CLSM poured into Tank 20F. Therefore, the following conversion should be used: Gallons of Bleed Water = % Bleed Water (gals of Water, / cyd of CLSM) * (cyd of

CLSM)
For Mix 1 with maximum water, the amount of bleed water expected after 24 hours for a one foot lift is approximately 1800 gallons. If this water is distributed evenly within Tank 20F, the depth of the bleed water would be about one half inch. Since it is difficult to remove such a small layer, much thicker lifts are needed to be practical for removal (See Section 6.0, Recommendations). This results in a more homogeneous mix and also reduces the percentage of bleed water.
The bleed water, as a percent of the total mix water generated in a closed tank, such as Tank 20F, using Mix 1 fill material, is independent of the lift height per CLSM volume. It is, however, dependent on the amount of water in the mix. Mix 1 has an allowable water content of 62 to 66 gallons per cubic yard. After curing for various times, the relationship between the percent bleed water and the amount of water in the original mix are given by the following equations: where: x = amount of water in the mix, gallons per cubic yard, y = flow, inches.

I .
Use Mix 1 because of the better flow characteristics, less bleed water and segregation and higher strength.

2.
Use as little water as possible to achieve the desired flow. This should be done and supervised by a knowledgeable person familiar with CLSM or an experienced Construction Engineer.

3.
Pour 3000 to 3500 cubic yards of Mix 1 before attempting to remove bleed water. This will generate at least six inches of bleed water to make it easier for removal.
Optimize Mix 1 to further minimize bleed water. Inorganic sluny thickeners such as clay (bentonite, etc.), sodium silicates or fine silica, in addition to air entraining and superplastizer admixtures, are recommended for future testing.

5.
Prior to the start of filling Tank 20F. a trial mix should be made at the batch plant and evaluated at SRS for drop height, segregation, flow, pumpability and the minimum water requirement. If pumpability is questionable, the mix proportions must be adjusted accordingly.  1  Table 2  Table 3  Table 4   Table 5   Table 6 Table 7 Figures:                 All work shall be charged to WAD No. WE1 14925, Activity Code No. W11492500.

BLEED WATER TEST MATERIALS:
1. 18 Containers With Lids a) The containers are to be a minimum 2 feet diameter in width and the height shall be tall enough to accommodate 1,2 and 3 foot depths of the material plus excess bleed water. They are to be made of steel or plastic. 55 gallon drums are an example of an acceptable container. It is important that lids are included with the containers.

CLSM Mix Design
a) The test CLSM mix design shall consist of the following: ii) * one test for minimum (62 gal) and another for maximum (66 gal) of water content, 251 5 #Icyd of sand, 600 Ncyd of fly ash, Type F * The minimum water content as indicated here means optimum water needed to make it flowable. Therefore, the quantity may be adjusted.
b) The material can be obtained from an on site delivery, if available or may be batched. c) Several batches (maximum 4) may be prepared provided that there is less than 15 minutes lapse time between batches. A single batch is preferred.

Miscellaneous Equipment
. a) Tape, ruler, measuring jar, etc. for measuring the height of the initial CLSM placement and the height and volume of the bleed water.