Benjamin T. Adams1, S.M. ASCE, and Ming Xiao2, M. ASCE 1 Graduate student, Department of Civil and Geomatics Engineering, California State University, Fresno. 2 Assistant Professor, Department of Civil and Geomatics Engineering, M/S EE94. California State University, Fresno, California 93740. USA. Phone: (559) 278-7588. Email: [email protected]. ABSTRACT This study aims to explore a remediation method in which organic soil is mixed with sand to increase the sand’s resistance to piping erosion. Our preliminary experiments on internal erosion (piping and suffusion) of fibrous and amorphous peats (with organic contents of 23% and 32%, respectively) using rigid wall and flexible wall setups show that peat does not erode internally even when subjected to hydraulic gradients greater than those typically seen in the field. These initial investigations also suggest that mixing organic soils with erodible sand may reduce the potential for piping erosion. In this study, a mixture of green and manure compost, referred to as co-compost, is used as the source of organic soil. Hole-erosion tests are performed to quantify the erosion of a silty sand, the co-compost, and various ratios of sand–co-compost mixtures. The potential increase in consolidation settlement and reduction in shear strength and permeability due to the addition of organic matter are also investigated. Greater proportions of organic soil mixed with typical construction sand result in increased resistances to piping erosion. With the addition of organic matter to sand, consolidation settlement increases, undrained compression strength decreases, and permeability was reduced by two orders of magnitude. INTRODUCTION
Case histories indicate some earthen levees and dams built using granular soils are susceptible to piping, and piping may progress into catastrophic failures of the earthen embankments. Common physical preventive measures include installing filters at the seepage exit point to retain the soils or installing slurry cut-off walls to reduce the concentrated seepage and hydraulic gradient. This paper presents a preliminary investigation into a bio-remediation method in which the piping potential of an erodible granular soil is reduced through the addition of organic soil. The motivation for this research stems from our recent studies (Adams and Xiao, 2010; Xiao et al., 2010) using flexible-wall triaxial erosion and rigid-wall hole-erosion tests on various soil types, including fibrous and amorphous peats (with organic contents of 23% and 32%, respectively) and a sand-peat mixture. The experiments provided the following observations:
(1) When peat is consolidated under simulated embankment loading, even at a
small effective stress of 34 kN/m2, internal erosion (piping or suffusion) does not occur even at a high hydraulic gradient of 55.
(2) The addition of non-erodible peat to highly erodible mineral soil results in a drastic reduction in piping erosion.
The first objective of this study is to further test the hypothesis that introducing
organic soil into granular mineral soils can reduce the potential for piping. Also requiring attention is the potential degradation of the mechanical properties considered necessary to constitute a suitable building material. Organic soils such as fibrous peats exhibit significant compressibility, particularly secondary consolidation (Mesri and Ajlouni, 2007). The low specific gravity (1.5–2.0) of peat may also make it unsuitable as an embankment backfill. Accordingly, the second objective of this study is to quantify the changes in the soil’s geo-mechanical properties resulting from the introduction of organic soil. This will help aid in evaluating the feasibility of applying this bioremediation method to earthen embankments in the field. EXPERIMENTAL METHODOLOGY 1. Materials and Characterization
Two materials are used in this testing program. One is a sandy soil that was sampled from a construction site stockpile. The other, commonly referred to as co-compost, is a commercially produced compost containing equal proportion by mass of green waste and bio-solids. It contains large, relatively fresh organic particles and resembles very closely the Kerman Peat used in our previous testing (Adams and Xiao, 2010; Xiao et al., 2010). The similarity in organic content and particle shape is therefore one motive for choosing the compost over the peat that was studied in our previous tests. The other motivation stems from practical considerations. Compost is an end product of an aerobic process during which microorganisms decompose organic matter into a stable material, i.e., the compost will not further decompose and is free of compounds such as ammonia and organic acids that can be toxic to plant growth. Therefore, compost may serve as an artificial alternative to the naturally occurring peat and may promote environmental sustainability by transforming waste products into a potentially usable construction material.
In addition to these two original materials, four composite soils are created in the
laboratory by adding the co-compost to the sand in proportions of 5%, 10%, 15%, and 20% by mass. Each mixture is tested under identical conditions to study the changes in erosion resistance resulting from varying ratios of organic soil content. The basic characteristics of the sand, the co-compost, and the mixture with 20% co-compost are shown in Table 1.
The variations in particle sizes of the sand, the 20% mixture, and the co-compost
are shown in Figure 1. For the sand and the 20% mixture, the results from both sieve and hydrometer analyses are reported. Due to the high organic content and the subsequent floating of a large portion of the co-compost, analysis of the co-compost by hydrometer sedimentation is not possible. For the co-compost, only the results of the sieve analysis are reported. While a drastic difference in grain size distribution of the course fraction exists between the co-compost and each of the other two soils, the precise role that this parameter plays in piping erosion progression or resistance is still not fully understood. Previous research by the authors indicates that for a sandy soil and a fibrous peat that have nearly identical grain size distributions, a drastic difference in piping potential still exists (Adams and Xiao, 2010; Xiao et al., 2010).
Figure 1. Grain size distributions of the sand, the 20% mixture, and the co-compost
2. Hole-Erosion Experimental Setup
The improvised hole-erosion test setup shown in Figure 2 is used to conduct the erosion testing. The device is modeled after the apparatus developed by researchers at the University of New South Wales (Wan and Fell, 2004a, 2004b). The test is conducted on specimens measuring 7.0 cm in diameter and 14.0 cm in length. Clear acrylic tubing and end caps, which allow for easy observation of the specimen during all phases of testing, are used for the specimen mold. The end caps are designed in such a way that eroded soil particles are able to exit from the specimen unhindered. The specimen is created by compacting soil directly within the mold to its maximum
dry density in thin, uniform layers at optimum moisture content. Maximum dry density and optimum moisture content are obtained using the Harvard miniature compaction apparatus. Each of the three uniform layers of the specimen receives 25 tamps from the 20 lb (89 N) spring. A straight hole is pre-formed during the specimen compaction using a metal rod and has an initial diameter of 0.65 cm. This hole simulates an initial piping channel. A specially designed tamper covering nearly the entire cross sectional area of the specimen is used to ensure consistent densification within each layer. This ensures that the soils adjacent to the hole are not over- or under- compacted. De-ionized water is introduced at the top of the cylinder via a constant-head reservoir and then passes through a 2.0 cm thick layer of glass beads before reaching the specimen. The glass beads, measuring 0.6 cm in diameter, are used to evenly distribute influent before it enters the simulated piping channel. Effluent with eroded soil particles is incrementally collected in buckets directly beneath the specimen as the effluent exits the cylinder. With the downstream side of the specimen open to atmosphere, a constant hydraulic gradient of four is established for conducting the erosion tests, simulating possible field conditions.
Figure 2. Constant-head hole-erosion test apparatus and test specimen 3. Hole-Erosion Testing Program
Each specimen is tested for a duration of 60 min or until failure, whichever occurs first. Failure is defined as enlargement of the initial piping hole to the perimeter of the mold. Table 2 summarizes the testing parameters used for the erosion tests. During each time interval, the mass of water collected in each bucket is used to calculate the incremental seepage. Also calculated in each time increment is the total dry mass of eroded soil particles. This value is determined through decanting and drying of the collected effluent with eroded material. At the end of each test, a silicon rubber fluid (OOMOO® 25 Silicone Rubber, Smooth-On Inc.,
Easton, PA) is slowly injected into the piping hole. The fluid occupies all voids in the piping channel and solidifies in 75min at room temperature. It has negligible shrinkage and good tear strength. The solidified silicon rubber castings can be easily detached from the soil and accurately represent the shape of the piping hole in each test. Table 2. Hole-Erosion Test Parameters
Test Parameter Sand5% Mix
10% Mix
15% Mix
20% Mix
Co-compost
Duration of test (min) 6 15 60 60 60 60Duration of each increment (min) 0.75 3.00 5.00 6.00 10.00 10.00
4. Compressibility, Strength, and Permeability Testing
The sand and the 20% mixture are compacted to their maximum dry densities at optimum moisture content and tested for consolidation settlement, compression strength, and permeability. By comparing the results from the sand with the 20% mixture, the potential for degradation of these important engineering index properties is investigated. Consolidation using standard 24-hour incremental loading is used to study both the total and time-rate of consolidation. A pressure range of 15.4 kN/m2 to 247.0 kN/m2 is implemented, with time – deformation measurements taken at the pressures of 30.9 kN/m2 and 247.0 kN/m2. Standard triaxial consolidated-undrained compression and flexible-wall permeability tests are performed at an effective confining pressure of 55.2 kN/m2. This stress is achieved by providing a total confining pressure of 110.4 kN/m2 and a pore pressure of 55.2 kN/m2. RESULTS AND DISCUSSION 1. Hole-Erosion Test Results
Figure 3 shows a side-by-side comparison of the post-erosion piping hole castings from the sand, the 20% mixture, and the co-compost. The vertically oriented shaft extending the entire length of the photograph is the metal rod used to form the initial piping hole during the specimen compaction. To the left of the rod the cylindrical erosion test mold with specimen is shown to provide a physical reference with respect to the erosion hole castings. The piping hole in the sand without organic soil enlarged to the perimeter of the cylinder at the top portion of the specimen and the test was terminated at 6min. The co-compost and the 20% mixture sustained 60min of concentrated seepage without failure and showed only a slight enlargement in hole size.
Figure 3. Physical observation and comparison of post-erosion hole conditions for the sand, the 20% mixture, and the co-compost
The seepage and erosion rates plotted in Figures 4 and 5, respectively, are obtained using the incremental effluent volume and dry mass of eroded soils collected during each test. Seepage rate is the volume of effluent collected during each time increment divided by the duration of that increment. The erosion rate represents the percent erosion per liter of effluent collected during each increment, where the percent erosion is calculated as grams of dry eroded material collected during each increment divided by the initial dry mass of the specimen. Each of these values is plotted against the corresponding cumulative seepage volume for that increment. Under the constant-head condition established in each test, as erosion progresses and the size of the piping hole increases, a larger seepage volume is able to pass through the specimen per unit of time. The sand and the 5% mixture failed at 6min and 15min, respectively, where the piping channels enlarged to the perimeter of the mold at the top portion of the specimens. The other three mixtures and the compost were tested for 60min and the piping channels did not enlarged to the perimeter of the specimen. Figure 4 quantitatively reveals the effect of varying organic soil concentrations on the erosion potential. While each soil exhibits a relatively constant erosion rate throughout the duration of the test, the erosion rates for the co-compost and the 20% mixture are many orders of magnitude lower than those exhibited by the pure sand and the 5% mixture, suggesting a positive correlation between organic content in the soil and the reduction in erosion.
It is noted that the point representing the first increment of erosion for each
test in Figure 4 may not be reliable ⎯ during the formation of the piping channel at specimen compaction, the removal of the metal rod used to form the piping hole can cause some soil particles to dislodge from the wall of the hole. The assembly of the cylinder and the end caps can also cause a slight disturbance of the specimen. These
loose particles are easily washed from the specimen upon the initiation of the test (due to water hammer) and are collected in the first effluent bucket, causing the eroded soil mass to be high. This is supported by the fact that for every soil type, the first point represents the highest recorded erosion rate, despite the apparently arbitrary erosion progression exhibited by each soil.
Figure 4. Erosion progression in each test
Figure 5. Average seepage rate during each increment of elapsed time 2. Compressibility, Strength, and Permeability Results
Table 3 presents a summary of the consolidation settlement, compression strength, and permeability characteristics for the sand and the 20% mixture. Although the 20% mixture provided for a large reduction in erosion, the detriment to
settlement and strength properties would likely invalidate this material for use in embankment construction.
The compression indices that are calculated using Figure 6 increased by
approximately 233% after the addition of 20% co-compost to the sand. An increase in total settlement of approximately 140% is also observed. In addition to the total settlement, the rate and magnitude of time-dependent settlement appear to experience an increase as well, as shown by the time – deformation curves in Figure 7 for normal stresses of 30.9 kN/m2 and 247.0 kN/m2. Secondary consolidation of the 20% mixture is not revealed in Figure 7, even after seven days of consolidation at a pressure of 247.0 kN/m2. Longer time may be needed in order to observe the possibly large secondary consolidation, as is normally observed in organic soils. Table 3. Settlement, Strength, and Permeability Characteristics Soil Characteristics Sand 20% MixCompression index 0.030 0.100 Undrained peak shear stress at 55.2 kN/m2 effective stress (kN/m2) 400.2 229.9Coefficient of permeability, flexible-wall method (cm/s) 2.31�10-4 1.99�10-6
(a) Sand (b) 20% Mixture
Figure 6. Consolidation curves
The maximum deviator stress sustained by the 20% mixture is only about half that
of the pure sand, as shown in Figure 8. Much larger deformation is needed to mobilize the shear strength of the 20% mixture than the sand, indicating higher plasticity. Permeability reduction by adding the compost into the sand is significant; seepage velocity approaches the low values typically exhibited by clays and other low-permeability soils (Table 3).
(a) Normal stress = 30.9 kN/m2 (b) Normal stress = 247.0 kN/m2
Figure 7. Time – deformation curves for the sand and the 20% mixture under normal stresses of (a) 30.9 kN/m2 and (b) 247.0 kN/m2
Figure 8. Triaxial consolidated-undrained (CU) compression for the sand and the 20% mixture
CONCLUSIONS
This paper presents the preliminary findings of an investigation into a bioremediation method against piping erosion in sand. Constant-head hole-erosion tests on various ratios of sand-compost mixtures are performed in order to determine the role that the presence of organic matter may play in resisting piping erosion. The potential degradation of important engineering index properties due to the addition of organic material to sand is also investigated. Analysis of these findings suggests the following conclusions:
1) Greater proportions (by mass) of organic soil mixed with typical construction
sand result in increased resistances to piping erosion.
2) With the addition of organic matter to sand, consolidation settlement increases, undrained compression strength decreases, permeability was reduced by two orders of magnitude.
This research verified the hypothesis that adding organic matter to mineral soils
can increase the soil’s resistance to piping. The negative impact of the organic matter on the soils geomechanical properties is expected and remains a concern. We are continuing the research on the following three aspects: (1) the study of the fundamental mechanisms controlling erosion reduction including what types of organic matter can effectively reduce internal erosion and why; (2) with the knowledge gained in (1), identification of an effective and economical source of organic soil and its appropriate mixing ratio in an erodible soil in order to constitute a suitable embankment material; and (3) investigation into the long-term effects of deliberately introducing a potentially biochemically reactive material into earthen structures.
REFERENCES Adams, B.T., and Xiao, M. (2010). “Piping Mechanisms of a Fibrous Peat.”
Proceedings of the 5th International Conference on Scour and Erosion, Nov 7-10, 2010, San Francisco, CA.
Mesri, G., and Ajlouni, M. (2007). “Engineering properties of fibrous peats.” J. Geotechnical and Geoenvironmental Engrg. 133 (7):850-866.
Wan, C. F., and Fell, R. (2004a). “Investigation of Rate of Erosion of Soils in Embankment Dams” Journal of Geotechnical and Geoenvironmental Engineering, 130 (4): 373-380.
Wan, C. F., and Fell, R. (2004b). “Laboratory Tests on the Rate of Piping Erosion of Soils in Embankment Dams” Geotechnical Testing Journal, 27 (3): 295-303. ASTM International, West Conshohocken, PA, DOI: 10.1520/GTJ11903.
Xiao, M., Gomez, J., Adams, B., Shwiyhat, N., and E. Sinco (2010). “Experimental study of subsurface erosion of peats.” Proceedings of GeoFlorida 2010, Annual Geo-Congress of the Geo-Institute of ASCE. Feb 20-24, West Palm Beach, FL.
![Influence of fine particles on internal erosion of sand-clay mixtures · 2017-10-17 · 2004a, b] defined the variation of erosion rate assuming the erosion curves (i.e. erosion rate](https://static.documents.pub/doc/80x56/5e68b54713946b2f765b247a/influence-of-fine-particles-on-internal-erosion-of-sand-clay-mixtures-2017-10-17.jpg)
