All that Jazz. $$
Field Description
The field experiments were adopted at the Rifle Integrated Field Research Challenge (RIFRC) located at Rifle, Colorado. The groundwater flows 7-8 m/day through the alluvial aquifer which have approximately 3 m below the ground surface. Four monitoring wells exist with distance from approximately 5 meters each others. Also 32 wells exist that each well has 6 hose with different depth that can be used to injection.
Injection Scheme
Injection were performed while 14 days. Molasses were used for culturing indigenous bacteria in the experimental site. Fertilizer were also used to provide abundant carbon source to the system, by dissolved by the urea hydrolysis bacteria.
Detailed injection amount were presented in table X.
Experimental section
Materials
Two soils were chosen for all experiments. Ottawa 20/30 sand (emax = 0.742, emin = 0.502) and Silcosil 125 (US Silica, 95.1% smaller than 75 μm) were selected to represent sand and silt, respectively. Reagent grade chemicals including calcium chloride (CaCl2), sodium chloride (Na2CO3), urea, yeast extract, agar, and ammonium sulfate were obtained from Fisher Scientific and were used as received. Sporosarcina Pasteurii (s. Pasteurii) was obtained from ATCC 6453. Materials for the bender elements were purchased from Piezo Systems and were constructed before testing began.
Methods
Control experiments: Shear wave velocity was measured for soil samples that consisted of: 100% sand, 90% sand/10% silt, and 80% sand/ 20% silt. Each soil was tested at both loose and dense densities. For the sand/silt mixtures, 10% and 20% of volumetric fraction of silt were thoroughly mixed with sand using a rotating soil mixer for 24 hours to ensure homogeneity. The homogenized soil samples were then wet pluviated into a rigid wall test cylinder (10.2 cm diameter by 10.2 cm height). The specimen was loaded into a lever arm 1-D consolidometer (ELE International), with specially instrumented top and bottom platens to measure shear wave velocity (after Lee and Santamarina, 2005). Bender elements were attached to both the bottom platen and top cap, with an initial spacing of approximately 10.2 cm, which varied with each loading stage. Signals were measured with an oscilloscope (Agilent, DSO5014A), with the frequency of the input signal was equal to 20 Hz and the voltage of the input signal was 10 Volts (Agilent, 33210A). Output signals were filtered with a low-pass filter at 20 kHz and a high-pass filter at 500 Hz to eliminate utility frequency and other noise (Krohn-hite 3364 filter). Shear wave velocity and specimen height were measured at every loading step, ranging from stresses of 13.5 kPa to 98.2 kPa. Control tests were performed in both dry and saturated conditions, with loose and dense samples at each volumetric fraction of sand – silty sand mixtures (12 total control tests, Table 1).
Table 1. Experimental Matrix
Sample | Fluid conditions | Vol. fraction of sand (%) | Vol. fraction of silt (%) | Relative density | Treatment type |
RD0D | Dry | 100 | 0 | Dense | Control |
RD0L | Dry | 100 | 0 | Loose | Control |
RD10D | Dry | 90 | 10 | Dense | Control |
RD10L | Dry | 90 | 10 | Loose | Control |
RD20D | Dry | 80 | 20 | Dense | Control |
RD20L | Dry | 80 | 20 | Loose | Control |
RS0D | Saturated | 100 | 0 | Dense | Control |
RS0L | Saturated | 100 | 0 | Loose | Control |
RS10D | Saturated | 90 | 10 | Dense | Control |
RS10L | Saturated | 90 | 10 | Loose | Control |
RS20D | Saturated | 80 | 20 | Dense | Control |
RS20L | Saturated | 80 | 20 | Loose | Control |
C0D | Saturated | 100 | 0 | Dense | Chemical (abiotic) |
C0L | Saturated | 100 | 0 | Loose | Chemical (abiotic) |
C10D | Saturated | 90 | 10 | Dense | Chemical (abiotic) |
C10L | Saturated | 90 | 10 | Loose | Chemical (abiotic) |
C20D | Saturated | 80 | 20 | Dense | Chemical (abiotic) |
C20L | Saturated | 80 | 20 | Loose | Chemical (abiotic) |
B0D | Saturated | 100 | 0 | Dense | Biological |
B0L | Saturated | 100 | 0 | Loose | Biological |
B10D | Saturated | 90 | 10 | Dense | Biological |
B10L | Saturated | 90 | 10 | Loose | Biological |
B20D | Saturated | 80 | 20 | Dense | Biological |
B20L | Saturated | 80 | 20 | Loose | Biological |
Abiotic precipitation experiments. Chemical treatment tests were conducted for same three soil conditions (100%, 90% sand / 10% silt, and 80% sand / 10% silt), in order to measure the impact of precipitation in abiotic conditions (six tests total, Table 1). The sand samples and the sand/silt mixture samples were mixed with between 42 - 64 grams of solid phase CaCl2. The soil/CaCl2 mixture was then air pluviated into the test chamber, after which one pore volume of Na2CO3 solution was injected from the bottom platen. The sample was allowed to equilibrate for 24 hours during which time the CaCl2 dissolved. The final target concentration of the CaCl2 and Na2CO3 were 1.31 M and 1.45 M, respectively. After the sample equilibrated, loading was applied in increments of approximately double the previous load, and shear wave velocity was measured at each load increment.
Biomediated experiments: Sporosarcina Pasteurii from frozen stock were cultivated in a nutrient broth plate containing 20 g/L of yeast extract, 10 g/L of ammonium sulfate, 20 g/L of agar, and 0.13M of Tris buffer (ATCC 1376). Bacteria were grown in a liquid medium containing 20 g/L of yeast extract, 10 g/L of ammonium sulfate, and 0.13 M of Tris buffer at 36 ̊C for approximately 30 hours. Sample absorbance was measured using a spectrophotomer (Make and model), and it was determined that optical density (OD600) reached than 0.8 after 30 hours of growth. After 30 hours of growth, one pore volume of the bacterial suspension was injected into the test chamber, which contained soil samples prepared with the same methodology as the control experiments (six tests total, Table 1). The soil sample was first permeated with tap water injected from the bottom platen to ensure saturation, and then one pore volume of bacterial suspension was injected from the bottom platen. Next, the reagent solution with 3 g/L of nutrient broth, 0.1 M of CaCl2, and 0.1 M of urea was then continuously injected at a flow rate 0.24 mL/min for ten days. Shear wave velocity measurements were performed once a day for ten days at a constant vertical stress of 13.5 kPa. After ten days, the vertical stress on the sample was increased in increments of two times the previous load, and shear wave velocity was measured at ach increment up to a vertical stress of 98.2 kPa. After the experiment concluded, the specimens were extruded and sliced into three layers (approximately 3 cm thick), and total carbon content was quantified (Make and model) as a function of distance from injection point.
Results
Because shear wave velocity of granular soils depends on the inter-particle contacts and the state of fabric (Santamarina et al. 2001), it is predictable that the shear wave velocity should increase with increased applied stress and relative density. Additionally, it is well known that the shear wave velocity in dry conditions and saturated conditions does not show significant differences due to the propagating properties of shear waves, i.e., the shear modulus of water is zero.
As anticipated, the control soil specimens prepared at high relative density had higher shear wave velocities at all levels of confining stress when compared with samples prepared lower relative density (Figure 1). In the initial state (i.e., applied stress = 13 kPa) for all tests performed, the shear wave velocity ranged from 100 - 150 m/sec for the untreated soil samples. Increases in applied vertical stress resulted in predictable increases in shear wave velocity (Vs = ~260 m/sec at 100 kPa). Increasing the silt content in the control specimens (no chemical or biological treatment) resulted in measured shear wave velocities that increased approximately 10% at a sand/silt ratio = 90%/10% compared to 100% sand, and increased 5% at a sand/silt ratio of 80% / 20% compared to 100% sand, which is consistent with results obtained by XX (references). Shear wave velocity measurements measured in saturated and dry conditions were essentially identical.
Shear wave velocities measured in soils that were subjected to chemical treatment showed similar trends, but different magnitudes when compared to untreated specimens. Measured velocities for the chemically treated specimens ranged from 250 – 400 kPa, depending on density and vertical stress (Figure 2). In all cases, the shear wave velocities for the chemically treated soils were higher than the untreated controls, and in a similar trend to the untreated specimens, the chemically treated dense specimens had higher shear wave velocity when compared to chemically treated loose specimens. Increases in applied vertical stress resulted in an increase in measured shear wave velocity for all chemically treated specimens of approximately 100 m/sec, which was comparable to the results for the control specimens.
Shear wave velocities for specimens that were subjected to biological treatment demonstrated the highest measured values, ranging from initial Vs = 350 – 480 m/sec (vertical stress = 13.5 kPa), to maximum values Vs =500 – 550 m/sec (vertical stress = 100 kPa) (Figure 3). Similar to other tests, densely packed specimens had higher shear wave velocity when compared to loosely packed specimens.
At values of low vertical stress, comparison of the measured shear wave results demonstrated that biological and chemical treatment of the specimens resulted in a shear wave velocities that was more than two times higher than the control specimens. At high vertical stress (100 kPa), the measured shear wave velocity increase approximately two times for the biologically treated specimen and 1.5 times for the chemically treated specimen. Assuming uniform precipitation of calcium carbonate in the specimen, the increased Vs for biologically treated specimens implies a larger mass of CaCO3 precipitated in the specimen. Additionally, the increments of shear wave velocity increase with increasing applied vertical load were similar for all three methods, which means the effect of applied load on both the bio-treated and chemical-treated samples were relatively small compared to the control experiments. It is important to note that precipitation tends to occur more extensively in loose samples, which lessens the difference between measured shear wave velocities in dense and loose samples (XX% difference in treated samples compared to XX% difference in control samples)..
Replacement of silt on bio-treated samples show significant increase than control samples. Shear wave velocities of both control samples and chemically treated samples increased approximately 10% and 5% when 10% and 20% of sand were replaced by silt, respectively. On the other hand, those increased more than 10% when 20% of sand were replace by silt at bio-treated sample.
Discussion and Analysis (WHY??)
Precipitation of calcium carbonate on the surface of granular particles like the sand matrix will act to increase the effective particle size, while precipitation of calcium carbonate on silt particles or bacteria that reside in the pore space can be regarded as addition of new granular particles that are filling the pore space and reducing voids. Both deposition mechanisms will contribute to an increase in soil density, will strengthen the inter particle connectivity which will decrease deformability.
In summary, the measured shear wave velocity was a function of the following, in decreasing order of importance: treatment method, applied vertical stress, relative density, and silt content.
(a)
(b)
(c)
(d)
Shear wave velocity according to applied vertical stress. (a) Dried condition sample (b) saturated condition sample (c) Chemical treated sample (d) bio-treated sample