WOVEN GEOSYNTHETIC FABRIC WITH DIFFERENTIAL WICKING CAPABILITY

CROSS-REFERENCE TO RELATED APPLICATION

This application claims benefit of U.S. Patent Application Serial No. 12/359,876 filed January 26, 2009, which is incorporated herein by reference in its entirety.

TECHNICAL FIELD OFTHE INVENTION

The present invention is related generally to woven fabrics. More specifically, the present invention is related to g∞synthetic wicking fabrics and pavement structures employing same. BACKGROUND OF THE INVENTION

Frost heave and thaw weakening can cause damage to pavement structures, such as parking areas, roadways, airfields, etc., in northern regions. The formation of ice lenses in the pavement structure is a significant contributor to such damage, as illustrated in FIGURE 1. Three elements are necessary for ice lenses, and thus frost heave, to form. These arc: (1) frost susceptible soil, (2) subfreezing temperatures, and (3) water. Often, water is available from the groundwater table, infiltration, an aquifer, or held within the voids of fine-grained soil. By removing any of the three elements above, frost heave and thaw weakening can be at least minimized or eliminated altogether.

Techniques have been developed to mitigate the damage to pavement structures caused by frost heave and thaw weakening. One such method involves removing the frost susceptible soils and replacing them with non-frost susceptible soils. The non-frost susceptible soil is placed at an adequate thickness to reduce the strain in the frost-susceptible soil layers below to an acceptable level. Other methods include use of insulation to reduce the freeze and thaw depth. In areas where removal of frost susceptible soils and reduction of subfreezing temperature are difficult and expensive, removal of water can lead to savings in construction costs by reducing the formation of ice lenses. By breaking the capillary flow path, frost action can be less severe.

A capillary barrier is a layer of coarse-grained soils or gcosynthetic in a fine grained soil that (i) reduces upward capillary flow of soil water due to suction gradient generated by evaporation or freezing, and (or) (ii) reduces or prevents water from infiltrating from the overlying fine-pored unsaturated soil into the soil below the capillary barrier. In the latter case, if the capillary barrier is sloped, the infiltrating water flows in the fine soil downwards along the interface with the capillary barrier. Geosynthetic drainage nets (goonets) have been found to serve as capillary barriers because of their large pore sizes. The performance of nonwoven gcotextiles as a capillary barrier appears to be compromised by soil intrusion into their interiors, decreasing the pore size and increasing the affinity of the material to water. Further, as reported by Henry (1998), "The use of geosynthetics to mitigate frost heave in soils." Ph.D. dissertation, Civil Engineering Department, University of Washington, Seattle, hydrophobic geotextiles have been more effective in reducing frost heave than hydrophilic geotextiles. The above mentioned capillary barriers attempt to cut off the capillary water flow by generating a horizontal layer with very low unsaturated permeability under suction. The whole structure is permeable for downward rainfall infiltration. This type of capillary barrier requires that the barrier thickness exceed the height of the capillary rise of water in them. In addition, it provides conditions suitable for water vapor flow because of their high porosity and comparatively low equilibrium degrees of saturation.

Thus, there remains a need for a woven geosynthetic fabric with differential wicking capability that reduces or eliminates frost heave in soils. Accordingly, it is to solving this and other needs that the present invention is directed.

SUMMARY OFTHE INVENTION The present invention is directed to a woven geotextile wicking fabric. The wicking fabric comprises a polymeric yam disposed in one axis of the fabric and a plurality of wicking fibers disposed substantially parallel to one another and woven with the polymeric yam in another axis of the fabric. The wicking fiber comprises a non-round or non-oval cross-section and has a surface factor of about 100 cc/g/hr to about 2SO cc/g/hr. In one aspect of the present invention, the cross-sectional shape of the wicking fiber is multichannel, trilobal, or pillow.

In another aspect of the present invention, a wicking drainage system is disclosed. The wicking drainage system comprises a wicking fabric layer disposed on a layer of frost susceptible soil. A layer of non-frost susceptible soil is disposed on the wicking fabric. Optionally, a base layer for supporting asphalt and/or concrete is disposed on the non-frost susceptible soil. The wicking drainage system can further comprise an impermeable hydrophobic geomembrane disposed below the wicking fabric. Further, the wicking fabric can be tilted with respect to the water tabic and/or the asphalt and/or concrete layer being supported by the wicking drainage system.

Yet, in another aspect of the present invention, a wicking drainage system comprises a wicking fabric layer disposed on a first layer of frost susceptible soil. A second layer of frost susceptible soil is disposed on the wicking fabric layer. Disposed on the second layer of frost susceptible soil is a geotextile layer. A layer of non-frost susceptible soil is disposed on the geotextile layer. Optionally, a base layer for supporting asphalt or concrete is disposed on the non-frost susceptible soil. The geotextile layer can be another wicking fabric layer. It is to be understood that the phraseology and terminology employed herein are for the purpose of description and should not be regarded as limiting. As such, those skilled in the art will appreciate that the conception, upon which this disclosure is based, may readily be utilized as a basis for the designing of other structures, methods, and systems for carrying out the present invention. It is important, therefore, that the claims be regarded as including such equivalent constructions insofar as they do not depart from the spirit and scope of the present invention.

Other advantages and capabilities of the invention will become apparent from the following description taken in conjunction with the accompanying drawings showing the embodiments and aspects of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The invention will be better understood and the above objects as well as objects other than those set forth above will become apparent when consideration is given to the following detailed description thereof. Such description makes reference to the annexed drawings wherein:

FIGURE I is an illustration of the formation of ice lenses in a pavement structure;

FIGURE 2 is an illustration of wicking fiber cross-sections employed in the present invention;

FIGURE 3 is an illustration of a wicking drainage system in accordance with the present invention;

FIGURE 4 is an illustration of another aspect of the wicking drainage system in accordance with the present invention; FIGURE S is an illustration of yet another aspect of the wicking drainage system in accordance with the present invention;

FIGURE 6 is an illustration of still another aspect of the wicking drainage system in accordance with the present invention; FIGURE 7 is a graph illustrating sieve analysis of silt taken from the CREEL permafrost tunnel;

FIGURE 8 is a graph illustrating sieve analysis of Dl material in Fairbanks;

FIGURE 9 is a graph illustrating compaction test results for silts from CREEL permafrost tunnel; FIGURE IO is a graph illustrating compaction test results for Fairbanks Dl material with

10% fines; and

FIGURE 11 is comparison of gravimetric water content to metric suction for Fairbanks Dl material.

DETAILED DESCRIPTION OFTHE INVENTION The present invention is directed to a woven, wicking fabric that optimizes capillary tension substantially in a single axis to enhance dewatering around the fabric protected area versus conventional fabrics. For example U.S. Patent No. 6,132,633, which is incorporated herein by reference in its entirety, describes a geocomposite capillary barrier drain (GCBD) for displacing water from beneath pavement. The GCBD system employs a transport layer, a capillary barrier and a separator layer. Specifically, the GCBD transport layer utilizes the capillary properties of a fiberglass fabric to displace water away from the paved surface, hi accordance with the present invention, the novel woven fabric described below can be incorporated into the GCBD system by replacing the fiberglass fabric. Further, the novel woven fabric of the present invention can be employed to replace the GCBD system altogether. In accordance with the present invention, a geotextile woven, wicking fabric comprises a conventional yarn or a filament in one axis and a wicking fiber woven with the yarn or filament in another axis to form the fabric. For example, the wicking fiber can be woven into the wicking fabric in either the warp or the weft directions. The wicking fiber has a non-round or non-oval cross-section with a surface factor between about l.S and about 3.3. In another aspect the wicking fiber has a flux range of about 100 cc/g/hr to about 250 cc/g/hr. Yet, in another aspect the wicking fiber maintains at least about 80 % flux up to 60,000 ft-lb/fr*. Still, in another aspect the wicking fiber maintains unsaturated hydraulic conductivity in environments having saturations between 100 % and 17 %. As indicated above the fabric of the present invention finds utility in civil engineering applications. The polymers described below can be employed to make the conventional yam or filament.

In one aspect of the present invention, the wicking fabric has a specific surface area of 3,650 cm ² /g and a permeability of 0.55 cm/s, which is equivalent to a flow rate of 1,385 l/min/m ² . Further, the wicking fabric of the present invention can maintain saturation in a water infiltration test after being exposed to evaporation for three days.

The wicking fabric of the present invention can both drain the water out of the soil from the beneath and from the top of the soil when there is water ponding. This aspect of the invention provides for quick drainage of water in spring thaws. Further, the wicking fabric can be employed to reduce the moisture content in the soil and improve soil shear strength. Wicking Fibers

In one aspect of the present invention, wicking fibers are woven into a wicking fabric substantially parallel to one another. As a result, a fluid, such as water, is transported along the wicking fibers to the periphery of the woven fabric of the present invention. That is, the wicking fibers move the fluid substantially along a single axis. Wicking fibers employed in the present invention have a high surface factor of less than 1.5 as compared to a round cross-sectional fiber of the same denier having a high surface factor of 1.0. Such wicking fibers generate increased capillary action over round cross-sectional fibers of the same denier. Several types if fibers can be employed in the present invention and are described below.

U.S. Patent No. 5,200,248, which is incorporated herein by reference in its entirety, describes capillary channel polymeric fibers that can be employed in the present invention. Such fibers store and transport liquid and have non-round, cross-section shapes which include relatively long thin portions. The cross-section shapes are substantially the same along the length of the fiber. Further, these capillary channel fibers can be coated with materials that provide an adhesion tension with water of at least 25 dynes/cm. U.S. Patent No. 5,268,229, which is incorporated herein by reference in its entirety, describes fibers that can be employed in the present invention. These fibers have non-round cross-sectional shapes, specifically "u" and "E" shaped cross-sections with stabilizing legs.

Further, these fibers are spontaneously wcttable fibers and have cross-sections that are substantially the same along the length of the fiber.

U.S. Patent No. 5,977,429, which is incorporated herein by reference in its entirety, describes fibers having distorted "H" shape, a distorted "Y" shape, a distorted "+" shape, a distorted "U" shape, and a distorted shape of a spun fiber that is referred to as "4DG". Such fibers can be employed in the present invention. U.S. Patent No. 6,103,376, which is incorporated herein by reference in its entirety, describes a bundle of synthetic fibers for transporting fluids which can be employed in the present invention. The bundle comprises at least two fibers that when acting as individual fibers are poor transporters of fluids, yet when in a bundle the fibers provide a bundle that is an effective transporter of fluids. As described, the bundle has a Specific Volume greater than 4.0 cubic centimeters per gram (cc/gm), an average inter-fiber capillary width of from 25 to 400 microns, and a length greater than one centimeter (cm). At least one of the two fibers has a non- round cross-section, a Single Fiber Bulk Factor greater than 4.0, a Specific Capillary Volume less than 2.0 cc/gm or a Specific Capillary Surface Area less than 2000 cc/gm, and more than 70% of intra-fiber channels having a capillary channel width greater than 300 microns. Wicking fibers employed in the present invention are made from the major melt spinnablc groups. These groups include polyesters, nylons, polyolefins, and cellulose esters. Fibers from polyethylene terephthalate) and polypropylene are useful in the present invention at least because of their manufacturability and wide range of applications. The denier of each fiber is between about 15 and about 250, or between about 30 and about 170. In addition, wicking fibers can be formed from other polymers that shrink significantly when heated, such as polystyrene or foamed polystyrene. The step of shrinking introduces the distortion in the fiber that increases long-range distortion factor (LRDF) and short range distortion factor (SRDF). The relatively large values of LRDF and/or SRDF of the fibers described in U.S. Patent No. 5,977.429 provide their utility in absorbent products. Shrinking occurs for oriented amorphous polymeric fibers when the fibers are heated above their glass transition temperature. The shrinking occurs either prior to or in the absence of substantial crystallization.

As indicated above, the wicking fibers of the present invention can be made of any polymeric material that is insoluble in the fluid which is to be contacted with the capillary channel structures. For example, the polymer utilized can be a thermo-plastic polymer, which can be extruded and drawn via an extrusion process to form the final product. Examples of suitable polymeric materials, in addition to polyester, polystyrene and polyolcfins such as polyethylene and polypropylene, include polyamides, chemical cellulose-based polymers such as viscose and di- or tri-acc-. Co-, ter-, etc. polymers and grafted polymers can also be used. One type of thermoplastic polymer can be employed in the present invention arc polyesters and copolymers of dicarboxylic acids or esters thereof and glycols. The dicarboxylic acid and ester compounds used in the production of polyester copolymers are well known to those of ordinary skill in the art They include terephthalic acid, isophthalic acid, p.p'-diphenyldicarboxylic acid, p,p'-dicarboxydiphcny! ethane, p,p'-dicarboxydiphenyl hexanc, p,ρ'-dicarboxydiρhcnyl ether, p.p'-dicarboxyphenoxy ethane, and the like, and the dialkylesters thereof that contain from I to about S carbon atoms in the alkyl groups thereof.

Aliphatic glycols useful for the production of polyesters and copolymers are the acrylic and alicyclic aliphatic glycols having from 2 to 10 carbon atoms, such as ethylene glycol, trimethylcne glycol, tetramethylene glycol, pentamethylene glycol, and decamethylene glycol. It is additionally contemplated to utilize copolymers or graft copolymers, terpolymers, chemically modified polymers, and the like, which permanently exhibit high surface hydrophilicity and do not require the use of wetting agents, which may wash away from the structure surface upon contact with fluids. Modified polymers which can exhibit permanent hydrophilicity include chemical cellulose polymers such as cellulose acetates. In addition, one can also include pigments, deluslerants or optical brighteners by the known procedures and in the known amounts.

A type of polyester which can be employed in the present invention is glycol modified polyethylene tcrephthalnelate) (pETG) copolyester. Suitable PETG is available from Eastman Chemical Products, Inc. (Kingsport, Term., USA), under the name KODAR TM 6763, with a glass transition temperature of about 81 ºC. Another factor affecting polymer choice is amenability to chemical modification of its surface for increasing, for example, hydrophilicity. Thus, for capillary channel structures intended for absorbing and/or transporting aqueous based solutions, it can be advantageous to use a polyester-based polymer rather than, for example, a polypropylene. However, this selection option is not meant to thereby limit the scope of the invention. Also, depending upon the intended use of the structures, it can be desirable that the polymer material utilized be flexible at the temperatures at which the structures are intended to be used. Due to the relatively thin walls and bases of the structures hereof, even relatively high modulus polymers can be used to make structures that are both flexible and soft, yet which retain surprisingly high resistance to collapse. Flexibility will depend upon such factors as the thickness and dimensions of the capillary channel walls and base, as well as the modulus of elasticity. Thus, choice of polymer in this regard will be highly subject to the intended use and temperature conditions. Choice of such suitable polymer material is well within the ability of one of ordinary skill in the art.

Depending upon the intended use, the capillary channel structures can be made from polymers that are either hydrophilic or oleophilic, or can be treated to be hydrophilic or oleophilic.

The surface hydrophilicity of polymers used to make the capillary channel structures of the present invention can be increased to make the capillary channel walls more wettable to water or aqueous solutions by treatment with surfactants or other hydrophilic compounds (hereafter, collectively referred to as "hydrophilizing agents") known to those skilled in the art. Hydrophilizing agents include wetting agents such as polyethylene glycol monolaurates (e.g., PEGOSPERSE.TM. 200ML, a polyethylene glycol 200 monolaurate available from Lonza, Inc., Williamsport, Pa., USA), and ethoxylated oleyl alcohols (e.g., VOLPOTM.-3 available from Croda, Inc., New York, N.Y., USA). Other types of hydrophilizing agents and techniques can also be used, including those well known to those skilled in the fiber and textile arts tor increasing wicking performance, improving soil release properties, etc. These include, for example, surface grafting of polyacrylic acid. Suitable commercially available hydrophilizing agents include ZELCON TM. soil release agent, a nonionic hydrophile available from DuPont Co., Wilmington, Del. (USA) and Milease T.TM., comfort finish available from ICl Americas, Inc., Wilmington, Del., USA. In addition, ERGASURF, ceramic microbcads and vinyl pyrrolidone can be employed as hydrophilic or hygroscopic additives. The capillary channel structures of the wicking fibers have an axial base and at least two walls extending from the base, whereby the base and walls define at least one capillary channel. Certain of such fibers have at least five walls and at least four capillary channels. Others can have at least six walls and at least five capillary channels. There is no theoretical maximum number of capillary channels that the structure hereof can have, such maximum number of capillary channels being governed more by need for such structures and practicability of making them. In one aspect of the present invention, the capillary channels arc substantially parallel with one another and an open cross-section along at least about 20% of their length, along at least about 50% of their length or and along from at least 90% to 100% of their length. Wicking fibers of the present invention provide flexible, collapse-resistant, capillary channel structures comprising a polymer composition and having at least one intrastructure capillary channel, wherein the structures have an axial base and at least two walls extending from the base, typically (but not necessarily) along substantially the entire length of the base element, whereby the base element and walls define said capillary channd(s). In general, the walls should extend from the base for a distance in the axial direction of the base for at least about 0.2 cm. In another aspect of the present invention, the walls extend from the base for a distance in the axial direction of the base for at least about 1.0 cm. The actual length of the structure is limited only by practical concerns. Although the capillary channel structures hereof can have one capillary channel or a plurality of capillary channels, for convenience the plural form "channels" is used with the intent that it shall refer to a singular "channel" in structures having only one such channel or a plurality of channels in structures having more than one channel. The structures are further characterized in that the capillary channels are open along a substantial length such that fluid can be received from outside of the channel as a result of such open construction. In general, the structures will typically have Specific Capillary Volume (SCV) of at least about 2.0 cc/g, at least about 2.S cc/g or at least about 4.0 cc/g, and a Specific Capillary Surface Area (SCSA) of at least about 2000 cm ^2/g at least about 3000 cm ² /g or at least about 4000 cm ² /g. The procedures to be used for measuring SCV and SCSA are provided in at least one of the patents incorporated above.

The wicking fibers of the present invention have a surface composition that is hydrophilic, which may be inherent due the nature of the material used to make the fibers or may be fabricated by application of surface finishes. Hydrophilic surface finishes provide structures the surfaces of which have large adhesion tension (i.e., that strongly attract) with aqueous liquids and are therefore preferred for applications involving aqueous liquids such as those discussed below for temporary acquisition/distribution structures and permanent storage structures. In one aspect, the hydrophilic surface has an adhesion tension with distilled water greater than 25 dynes/cm as measured on a flat surface having the same composition and finish as the surface of the fiber. Some of the finishes/lubricants useful to provide large adhesion tensions to aqueous liquids are described or referenced in U.S. Pat. No. 5,611,981, which is incorporated by reference herein in its entirety. Surface finishes are well known in the art.

As discussed above, the wicking fibers have channels on their surface which may be useful in distributing or storing liquids when the proper surface energetics exist on the surface of the fibers, such as when the fibers satisfy the above equation relating to specific surface forces.

The surface energetics determine the adhesion tension between the surface and whatever liquid is in contact with the surface. The larger the adhesion tension, the stronger the force of attraction between the liquid and the surface. The adhesion tension is one factor in the capillary forces acting on the liquid in a channel. Another factor affecting the capillary forces acting on a liquid in a channel is the length of the perimeter of the channel. When the widths of the channels are small, the capillary forces arc relatively strong compared to the force of gravity on the liquid, since the force of gravity on the liquid in a channel is proportional to the area of the channel.

FIGURE 2 illustrates wicking fiber cross-sections of multichannel, trilobal, and pillow that can be employed in the present invention. However, as indicated in patent discussed above, other shapes can be employed in the present invention. The multichannel is also referred to as the "4DG" shape.

In one aspect of the present invention, a wicking fabric made from nylon has high wettability similar to fiberglass. The wicking fabric has a high specific surface area of 3650 cnr/g and high permeability of O.55cm/s (equivalent to a flow rate of 1385 1/min/m ² ).

Weaves

Weaves which can be employed in the present invention include, but are not limited to, plain, twills, specialty weaves, 3-D's. satins, sateens, honeycombs, lenos, baskets, oxfords, or Panamas. Figure 2 is a photomicrograph of a geosynthetic fabric of the present invention. Wicking Drainage System

Referring to FIGURE 3, in accordance with the present invention, a wicking drainage system 10 comprises a wicking fabric 20. a non-frost susceptible soil layer 30 disposed over the wicking fabric, and a base layer 40, such as an asphalt treated base, disposed on the soil layer 30. Asphalt and/or concrete SO are disposed on the base layer 40. The wicking fabric 20 is disposed on frost susceptible soil bed 60. The frost susceptible soil bed 60 is raised above the water table to form side drains 70 which facilitate water drainage. The thickness of the frost susceptible soil bed 60 is conventional. For example, soil bed 60 can be 40 inches above the water table. Non- frost susceptible soil layer 30, such as the Dl material with 10% fines content described below, should be of a sufficient thickness as to allow water drainage from the base layer 40 to the wicking fabric 20. In one aspect of the present invention, the thickness of the non-frost susceptible soil layer 30 is about 13 inches. However, the thickness can be varied as necessary depending upon soil conditions.

In another aspect of the present invention, the wicking drainage system comprises an impermeable hydrophobic geomcmbrane (not shown) disposed below the wicking fabric 20. The wicking fabric 20 allows water from the overlying soil to pass through the wicking fabric 20 when the overlying soil is saturated and transport water laterally to side drains 70. When the overlying soil is unsaturated, the wicking fabric can absorb water from the overlying unsaturated soil and transport it in the lateral directions. The impermeable hydrophobic geomcmbrane can repel water and completely cut off the capillary rise of ground water from beneath. In another aspect of the present invention, the geomembrane can be a one-way-valve geotcxtile.

In an alternate design, the wicking drainage system comprises the arrangement as shown in FIGURES 4-6. When installed in the pavement structure, the wicking fabric 20 is tilted at a slope from 5-10% so that infiltrating water will flow downdip. Furthermore, there should not be wrinkles of any significance that would cause water to pond on top of impermeable layer. Figure 4 illustrates the wicking drainage system 10 of FIGURE 3 with the tilted arrangement.

As illustrated in FIGURE 5, a second layer of wicking fabric 20 is employed in the wicking drainage system 10. Disposed between the respective layers of wicking fabrics 20 is a layer of frost susceptible soil. In another aspect of the present invention, as illustrated in Figure 6, the wicking fabric 20 is disposed on a layer of frost susceptible soil 60. Further, another layer of frost susceptible soil 60 is disposed on the wicking fabric 20. A geotextile separation layer 80 is disposed on the second layer of frost susceptible soil 60, and a layer of non-frost susceptible soil 30 is disposed on the geotextile separation layer 80.

The overall effect of the wicking drainage system is to cut off upward capillary water flow and drain most of the infiltrated water out of the pavement structure through the tilted

S drainage net by the wicking fabric. The diving force for the water flow in the drainage net is gravity and the driving forces for the water flow in the wicking fabric are gravity and suction generated by evaporation and freezing.

EXAMPLES

Example 1 : Sieve Analysis and Gradation Curves for Two Typical Soils in Alaska 0 Two typical soils employed in Alaskan pavements were collected. These soils were

Fairbanks silt obtained from the CREEI. permafrost tunnel and Dl material obtained from University Ready Mix Company. Silt is a frost-susceptible soil and typically used as subgrade for Alaska pavements. The silt from the CREEL permafrost tunnel was sieved to remove organic material. A sieve analysis was performed on the silt and is shown in FIGURE 7. S The Dl material was a typical non-frost susceptible material which is typically employed as base courses in Alaska pavements. To be qualified as a Dl material, the fines content has to be less than 4%. In this example, sieve analysis was made for the Fairbanks Dl material and fines with grain size less than 0.075mm was added to make a new frost susceptible material with 10% fines content. The gradation curves for the original and fabricated Dl materials are shown0 in FIGURE 8.

Example 2: Modified Proctor Compaction Tests

The Fairbanks silt and the Dl material with 10% fines content were compacted in accordance with ASTM D1SS7 in order to simulate the compaction process in the field. The compaction test results are as shown in FIGURES 9 and 10. 5 Example 3: Soil Water Characteristic Curve

Pressure plate tests in accordance with ASTM D232S-68 were used to obtain the water retention characteristic curve in the range from 0 to ISOOkPa. The salt concentration tests were used to measure the soil water characteristic curve for suction values are greater than I ₁ SOOkPa. FIGURE 11 shows the test results for Fairbanks Dl Material. Example 4: Soil Column Tests

Using the D-I material with 10% fines and at the optimum moisture content, cylinders were constructed. The cylinders were compacted in five layers, 52 blows to each layer. Geosynthetic materials were placed above the second layer. 13 different cylinders were made testing S different geosynthetic materials ((Nylon Wicking Fabric. Glass Fabric, HPS70, FW402, and HIPS board). S cylinders were made with the geosynthetic material being the same size as the cylinder and S cylinders were made with the appropriate geosynthetic material protruding outwards in order to understand the effects and advantages of drainage capabilities for each geosynthetic material. A membrane was placed around each cylinder in order to retain the moisture within the cylinder. Baths were setup to allow for water infiltration from the bottom of the cylinder. The evaporation within the room that the water baths were put in was measured by filling a glass full of water and measuring the weight of the glass of water each day for one week. Water was added to the water baths throughout the week.

Example S: Laboratory Capillary Rise Tests and Soil Water Characteristic Curves for Different Geosynthetks

The performance of six different geosynthetics at three different locations of layered pavement systems were tested through two groups of laboratory capillary rise tests. The three locations are in the base course, between the base course and the subgrade, and in the subgradc. The Dl material with 10% of fines content and Fairbanks sill was used to represent the base course layer and the subgrade of the pavement structure, respectively. In the first group of tests, all the geosynthetics were wrapped in the membrane, which is referred to as "no drainage" in the later discussion. In the second group of tests, only the top and bottom halves of the soil specimens were wrapped in the membrane while geosynthetics specimens had larger size (about 6 inches in diameter) and exposed partially in the air to increase the evaporation, which is referred to as "with drainage" in the later discussion. Six different geosynthetics were tested and total 36 tests were performed for three different locations. For each location, it included one reference soil tests, six soil columns with geosynthetics inside but no drainage, and six soil columns with geosynthetics inside and with drainage. The purposes of the two groups of tests were (1) to investigate if the geosynthetics can cut off the capillary rise, and (2) to investigate the influence of evaporation on the water content distribution of the pavement structure. The first group of tests was used to simulate the geosynthetic in the center of the pavement structure, while the second group of tests is used to simulate the performance at the shoulder of the pavement structure. For each group of tests, there was also a reference soil column with no geosynthetic inside. The gcosynthetic specimens used in the tests, where specimens I through 6 were MirafiΦ FW402, Mirafi® G-Serics Drainage Composites, Glass fabric, Mirafi® HP570, Mirafi Nylon Wicking Fabric, and Imp, respectively.

Specimens were compacted in three layers, 25 blows to each layer. A total of twenty six specimens were compacted. Each was 4.5 inches in height. After the specimens were made, a capillary barrier was placed on top of a specimen. Another specimen was placed on top of the capillary barrier. A plastic membrane was placed around each specimen for moisture control. The top of the silt specimens that were placed on top of the capillary barriers were sealed to eliminate evaporation. A total of 13 soil columns were made. The soil columns were then placed in a pan and water was periodically poured into the pan to maintain a height of about 0.5 inch to wet the soil from the bottom. After two weeks the specimens were taken out of their water baths in order to measure the moisture content at various heights. The specimens were taken apart and the capillary barrier was removed. A ruler was used to measure the appropriate width of each section. Each section was 1.5 inches in width. Both the top and bottom specimens were cut into three equal sections. A knife was used to cut each section. Once each section was removed, its weight was weighed on a scale the type of capillary barrier and its section height was recorded and the section was placed in a pan that would correspond to that particular specimen. This was done for each specimen. Afterwards, the pans were put in the oven and weighed again 24 hours later in order to obtain the dry weights.

Example 6: Salt Concentration Test and Pressure Plate Test

The salt concentration tests were used to measure the soil water characteristic curve for suction values are greater than 1,500 kPa. Specimen 2 and 3 show reasonable curves as shown in Figs 44 and 45, but the curve for specimen 5 seems a little strange as shown in Fig.46. For this reason, the results are currently being redone. The results may have been construed by a number of things. The first of which is the handling of the materials. Although gloves were used and precautions were taken to prevent moisture from escaping from the capillary barrier, this may have been a source of error. This may account for the extremely low moisture content levels that were found. Another reason may be that the salt concentration levels within the test containers arc off. A reason for this may be because the duck tape that was used is not adhering to the glass container as well as one might expect. The results from the next test should prove helpful in determining where the error is coming from.

The pressure plate tests in accordance with ASTM D2325-68 were used to obtain the water retention characteristic curve in the range from 0 to I SOO kPa. Data is currently being collected for the pressure plate test. After the data is collected, the specimens need to be dried in order to determine their dry weight which is used to determine the moisture content. Once the moisture contents are determined, the specimens will be saturated and put back into the pressure plate apparatus at a different suction.

Example 7: Configuration of pavement section Preliminary numerical simulations of performance of wicking fabric in expansive soils were performed by assuming material properties of the wicking fabric. Fig.47 shows an example of a typical configuration of the pavement section studied, and the mechanical boundary conditions are also shown.

In the example, the concrete slab was 0.2S meter ( 10-in) thick. Those concretes were made with gravel aggregates from Victoria, Texas, 0.4S of water-cementitious ratio (w/cm). The concrete has a Young's Modulus of E - 2* 10? kPa. Poisson's ratio v - 0.15. and hydraulic conductivity of K - 1 x 10.12 m/s. Due to the symmetry of the pavement structure, a 5-meter ( 16.4-ft) of width was chosen. The suction at a depth of 6.0 m was constant and assumed to be equal to 10 kPa, which is just above the ground water table. The suction at the ground surface was assumed to be 1000 kPa for the first approximation. For the left and right sides of the structure, only vertical displacements were allowed due to symmetry.

Example 8: Simulation of Soil-Structure Interaction

Coupled thermal-mechanical jointed (contact) elements in ABAQUS/Standard (2002) are used to simulate the interaction at the soil-concrete slab interface. The upper side of the contact clement is the bottom surface of the concrete slab and the lower side is the ground surface where the concrete slab is resting. The bottom face of the concrete slab is assigned to be the master surface and the ground surface is assigned to be the slave surface. Namely, the concrete can penetrate into the soil while the soil can not penetrate into the concrete (ABAQUS/Standard 2002). The "hard" contact relationship in ABAQUS is used to simulate the normal behavior at the soil-slab interface. During the simulation, the program will compute the thickness of the contact elements in the direction normal to the soil-structure interface. When the soil and the slab foundation are in contact (the thickness of the contact element is zero), any compressive load can be transferred from the slab to the soil. When the soil and the foundation are not in contact (the thickness of the contact element is greater than zero), no load can be transferred from the slab to the soil.

The basic Coulomb friction model is used to simulate the tangential behavior in the soil structure interaction in which the two contacting surfaces can carry shear stresses up to a certain magnitude across their interfaces before they start sliding relative to one another.

It is also assumed that no water is allowed to flow through the soil-slab interface. This condition is realized by defining a very low "gap conductance" to the jointed elements. The gap conductance of the contact elements is assumed to be 10 ^-30 S ^-1 when the slab and the soil arc in contact with each other. The gap conductance of the contact elements is assumed to be 0 when the slab and the soil are separated.

Example 9: Discussion of Simulation Results

The wicking fabric was installed at a depth of 1.0 m below the concrete slab. The wicking fabric was assumed to be under high compression with a bulk factor of I. It had an ability to transport water at a rate of 1.48 gal/hour/yard. This corresponds to an ability of horizontal permeability of 2x10 ^-3 m/s (for a wick fabric with a thickness of Imm, transmissivity is 2x10 ^-6 m ^: /s). Three different wicking fabrics were considered as follows:

I . The ability of the wicking fabric to transport water is limited so that the wicking fabric works as reinforcement only just like goo-textile. This case is referred to as "reinforcement only" in the following discussions; 2. The wicking fabric is highly permeable in all directions. This case is referred to as

"single layer wicking fabric" in the following discussions; and

3. The wicking fabric is highly permeable in the direction towards outside of the pavement only and impermeable in the other two directions. This case is referred to as "wicking fabric with impermeable layer" in the following discussions. It was used to simulate the wicking drainage board proposed in the previous progress report. Two different conditions were considered. One is that the concrete slab is integrated and there is no leakage form the slab to the subgrade. and the other is that there was a leakage at the center of the slab, which caused the suction in the range of 1.0 meter below the centerline were equal to 10 kPa (field capacity).

In order to investigate the influence of the wicking fabric on the performance of the pavement structure, conditions when there is no inclusion of wicking fabric were also considered. A total of eight simulations were performed as shown in Table 3.

The simulations were performed under steady state conditions. Two parameters were used to evaluate the performance of the pavement structure. The first one was the "length of unsupported slab", which is length of the slab which was not supported by the subgrade soils. This parameter is related to the differential settlements caused by the expansive soils under certain weather conditions. The second parameter was the Von Miscs stresses. A Von Mises stress is a stress- invariant used in yield criteria. It is calculated independently of the coordinate reference system, does not carry directional stress information such as normal and shear stresses, but carries enough information to identify hot-spots where failure might occur. The larger the Von Mises stresses, the higher possibility of damage there is. In the simulation of a pavement structure built on expansive soils with no wicking fabric and no leakage, the expansive soils underneath the concrete pavements are covered by the concrete slab so that there is no evaporation of water while the soils outside the concrete slab are subjected to evaporation. As a result, the soils underneath the concrete slab have lower suction values, which correspond to higher moisture contents. While the soils outside of the slab have high suctions and lower water contents (drying). The difference in moisture contents due to the coverage of the concrete slab can cause large differential settlements. The soils at the shoulder of the concrete pavements shrink more than the soils underneath the slab, which cause a phenomenon called "shoulder rotation" or "edge-drop" case. The differential settlement can be so large that part of the concrete slab loses support from the subgrade soils and make the concrete slab a cantilever. This will cause very large bending moments in the concrete slab, which can result in damage to the slab. The maximum Von Mises Stress for this case is 2399kPa, which is occurring in the center of the slab. The slab and the soils separated at the edge of the slab and the length of the separation is 1.1 m for a 5.0 m concrete slab as shown in Table 3.

Case 2: No Wicking Fabric, With Leakage

In the simulation of pavement structure built on expansive soils with leakage and no wicking fabric, there is a leakage underneath the center of the slab, which makes the soil wetter than the previous case. Outside of the slab, the soils were still dry due to evaporation. As a result, the differential movements are larger than the previous case. The length of unsupported slab is about 1.4m and the maximum Von Mises stress is 3597 kPa, about 50% higher than the previous case. In conclusion, leakage in the pavement structure will make the differential settlements much severe and more likely to result in damage to the pavement structure. Cases 1 and 2 were used as references to demonstrate the influence of wicking fabric on performance of the pavement structure.

Case 3: With Geotextile Reinforcement, No Leakage

In this case, a geotextile was included in the pavement structure at a depth of 1.0 m below the concrete pavement. The geotextile was assumed to have the same permeability as that for the soils because it is relatively thin. Its Young's modulus was assumed to be 200,00OkPa, which is much stronger than the expansive soils. In the simulation of pavement structure built on expansive soils with geotextile reinforcement and no leakage, the inclusion of the geotextile reinforcement had no influence on suction distribution. Although the length of the unsupported slab was l.tm (the same as that for case I), the maximum Von Mises Stress was 2668 kPa, 11% higher than that when there is no reinforcement. This case indicates the inclusion of a reinforcement does not cause any benefit for the pavement structure for the differential settlement caused by expansive soils.

Case 4: With Geotextile Reinforcement, With Leakage

In this case, there is a leakage underneath the center of the concrete slab. As a result, the suction was 10 kPa in the range of 1.0 m below the concrete slab. Just like case 2, the leakage significantly increases the differential settlements in the subgrade soils. As a result, the length of unsupported slab is 1.4m and the maximum Von Mises stress is 3600 kPa, which are basically the same as those in case 2. Again, this case indicates that inclusion of a geotextile reinforcement will not reduce the differential settlements caused by expansive soils. Case S: With a single layer of wicking fabric, no leakage This case is used to simulate the case when the wicking fabric is installed in a pavement structure. In the simulation of pavement structure built on expansive soils with a single layer of wicking fabric and no leakage, due to the high ability of the wicking fabric to transport water, the wicking fabric significantly increase the suction under the concrete slab and suction distributions in the pavement structure is more uniformly distributed with depth. As a result, the differential settlement in the pavement structure is very small.

The length of unsupported slab is only 0.162 m, which is mainly limited at a very small range close to the edge of the slab. Due to the fact that most of the slab is rest on the subgrade soils and suction difference underneath the slab is small, the stress in the slab is small (if the differential settlements arc zero, the stress in the slab will be the smallest). The maximum Von Mises stress is only S17.S kPa, less than 22% of the maximum Von

Mises stress for case 1 when there is no wick fabric. This case indicates that inclusion of the wicking fabric can significantly improve the pavement performance and the pavement is much less likely to damage compared with case I . Case 6: With a single layer of wicking fabric, with leakage The difference between cases 6 and S is that there is a leakage underneath the centerlinc of the concrete slab. Due to the leakage, the soil underneath the centerline of the slab is very wet with a suction of 10 kPa, while the outside still remains 100OkPa. The difference in suction is large. As a result, the differential settlements are very big. The leakage not only causes swelling for soil above the wicking fabric it also causes swelling of the soil beneath the wicking fabric. The final length of unsupported slab is 1.26 m, and the maximum Von Mises stress is 3S27 kPa. Compared with cases 2 and 4, inclusion of the wicking fabric only slightly improves the performance of the pavement structure when there is leakage. It is worth noting that case 6 is a steady state simulation in which the leakage is assumed to be lasting for a significant period of time. Under a real situation, a rainfall event only lasts for a short period of time. Therefore, the actual improvement made by including a wicking fabric might be greater than the simulation. This case was performed for comparison purposes only.

Case 7: Wicking Fabric with Impermeable Layer, no leakage

This case simulates the situation in which the wicking drainage board discussed above is installed in a pavement structure. In this simulation of a pavement structure built on expansive soils with the installation of the wicking drainage board and no leakage, the wicking drainage board significantly increases the suction under the concrete slab and suction distributions in the pavement structure is more uniformly distributed with depth as in case S. The differential settlement in the pavement structure is very small. The length of unsupported slab is only 0.162 m and the maximum Von Mises stress is only 517.5 kPa. The results obtained are like those obtained in case S. This case indicates that inclusion of wicking drainage board can significantly improve the pavement performance.

Different from case 7, in case 8 there is a leakage underneath the centerline of the slab. The leakage causes suction increase underneath the slab, resulting significant difference between the centerline and outside of the slab. However, due to the wicking drainage board is impermeable in the vertical direction, the wetting of the soil is limited between the concrete slab and the wicking drainage board. Also, because the drainage board is permeable on both sides, the bottom side can still drain water out of the pavement structure even when there is leakage on the top. As a result, the soil at the centerline is still drying below the wicking drainage board.

The wetting of the soil above the wicking drainage board causes the soil to swell, while the drying of the soil below the wicking drainage board causes the soil to shrink. These two effects counterbalance and reduce the differential settlement even when there is leakage at the center of the slab. In case 8, die slab and the soils are in good contact with a length of unsupported slab of 0.079m. Consequently, the maximum Von Mises stress is 1425 kPa, about 60% and 40% of the maximum Von Mises stresses in cases I and 6, respectively. The maximum Von Mises stresses and length of unsupported slab in cases 7 and 8 are much smaller than those under similar situations. It is concluded that inclusion of wicking drainage board can significantly improve the performance of pavement structure.

Example 10: Performance Under Raining Weather Conditions

The study of the performance of different geosynthetics under raining weather conditions was investigated by means of two different sets of tests. First, the performance of different geosynthetics placed between fully saturated Dl material with 10% fines and Fairbanks silt. Second, the performance of different geosynthetics placed between fully saturated Dl material with 10% fines and Dl material. The geosynthetics tested were Mirafi® FW402, Mirafi® G- Scries Drainage Composites, Mirafi® HPS70, and a nylon wicking fabric made in accordance with the present invention.

The interface is where each geosynthetic was placed. The material above the interface needed to be fully saturated as well as the geosynthetic itself in order to accurately understand the effects of rainfall on the performance of the geosynthetics. Membranes were wrapped around the outside of the compacted materials in order to control moisture loss due to exposure. The tests in which the geosynthetics were wrapped in a membrane arc referred to as having "no drainage." The tests in which the geosynthetics were partially exposed to the air arc referred to as having "drainage."

The Dl material was prepared and allowed to sit without exposure to the air. This allowed the moisture to be able to distribute itself throughout the sample. The prepared materials were then compacted in a plastic cylinder mold in 3 layers at 25 blows per layer. After the material was compacted, the surface was made smooth and was removed from the mold. Holes were cut into the bottom of plastic molds in order to allow for water infiltration. The plastic molds were also raised using spacers to serve this same purpose. To avoid loss of material during extraction, a cut was made along the length of the mold that allowed us to carefully fit the mold around the compacted Dl material. Once the mold was in place, duct tape was used to seal the cut that was made and hold the compacted Dl material firmly in place. Filter paper was placed in between the compacted Dl material and the holes that were cut into the plastic molds in order to prevent the loss of material. As the level of the water within the water bath increased, the water level within each cylinder would rise as well.

The performance of the nylon wicking fabric under raining weather conditions was evaluated. For both the D-l/D-l and D- I/Silt water infiltration test with drainage, the nylon wicking fabric out performed the others by having lowest moisture content distributions both above and below the interface. For the D-l/D-l water infiltration test with no drainage, the nylon wicking fabric out performed the others by having the lowest moisture content distribution both above and below the interface. The nylon wicking fabric demonstrated that it is effective in both drainage and non drainage applications. Example 11 : Performance Under Raining Weather Conditions

Example 11 was conducted as Example 10 except:

I . Instead of using D I material, uniform sand was employed to represent the coarse base material. This can significantly increase the uniformities of the soil specimens while the hydraulic properties of the two soils arc similar. 2. Instead of using compacted soil specimens, soil slurry was employed to make soil specimens which can approximate the uniformity of soil specimens. Other reasons why soil slurry was used instead of compacted soils are as follows:

(i) soils compacted at the optimum moisture contents usually have good mechanical properties and have less chance to cause problems for pavement structure, while soil with high moisture contents will. After frost heave, when the soil thaws, the moisture contents in the soils are as high as or even higher than soil slurry. If use of the wicking fabric can reduce the moisture content of high moisture content soils in the pavement structure, it will be highly beneficiary to the performance of the pavement structure.

(ii) high moisture contents in the soil specimens means high unsaturated permeabilities, which can lead to reduced experimental time for the tests.

3. Instead of putting geosynthetics in the center of the soil columns, the geolhnthetics were placed at the bottom of the soil specimens (8 inches in height). Two different soils were used: Fairbanks Silts and uniform medium sand. The experiment can be used to investigate the impact of geosynthetics on the soils above it after rainfall infiltrations. 4. In order to investigate the influence of geosynthetics on the soil below it, a series of tests were also performed. To facilitate the discussion, this group of tests is referred to as "Rainfall Infiltration/Top Test" in the following sections. Tests for the modified laboratory rainfall infiltration soil column tests were accomplished using silt and medium sand. In order to measure the effectiveness of the geosynthetic under saturated soil conditions due to weather, slurries were prepared from Fairbanks silt. After the slurry had been prepared, it was placed into a cylindrical plastic mold. The plastic mold was filled with the slurry and the top was leveled off. To densify the soil, the side of the plastic mold was tapped. An initially saturated geosynthetic and impermeable membrane were placed below the slurry. The impermeable membrane was placed directly under the geosynthetic and the geosynthetic was placed directly under the soil slurry. Holes were hammered into the top of each plastic mold using a hammer and a sharp metal object. The reason for the holes was to decrease the suction caused by drainage of water that would otherwise be inhibiting the flow of moisture through the soil slurry. The holes were made after the experiment was completely setup. The water was allowed to drain for 3 days. Initially, all of the geosynthetic materials were saturated and remained saturated. Some excessive water was drained out due to gravity and the amount of water flow reduced quickly with time in the first several minutes. There were 2 tests performed for each geosynthetic. The geosynthetic materials used in the experiment were periodically checked to sec if the geosynthetic was still saturated.

It was found that in the range where the soil slurry columns were sitting, in the direction of the nylon wicking fabric, the nylon wicking fabric remained wet after more than three days of testing, while outside of the range where the soil slurry columns were sitting, the nylon wicking fabric quickly dried out in less than one day.

Outside of the range where the soil slurry columns are sitting, the Mirafi® G Scries Drainage Composites remained relatively wet after three days, while the Mirafi® FW402 and the MirafiΦ HP570 quickly dry out in less than one day.

After 3 days, the molds were removed. The soil was then cut into 6 equal layers. The initial weight of each layer was recorded and each layer was put into the oven to dry for 24 hours. After 24 hours, each layer was taken out of the oven and the final weight was obtained. Using the initial and final weights, the moisture contents were found for each layer. The Rainfall Infiltration/Top Tests were performed at a moisture content of 28V* using sand. There were 2 tests setup for each geosynthetic. First, the sand and water were mixed together to obtain the correct moisture content. The sand slurry was poured into a plastic mold to a height that would leave 1.33 in. on top. A geosynthetic was placed in the mold at this height. The ends of each geosynthetic were cut so that they remained below the 1.33 in. mark along outside of the plastic mold for the entire test. Each geosynthetic that was used in the test was initially saturated. After the geosynthetic was firmly in place, the rest of the sand slurry was placed on top of the geosynthetic and filled to the top of the plastic mold. Aluminum foil was used to cover the soil slurry above the geosynthetic in order to not allow the moisture within the slurry to evaporate. After the tops of the molds had been covered, the experiment was allowed to sit for 5 days. After S days, the moisture distribution for each test was recorded. The same observations were made during the Rainfall Infiltration/Top Test as the

Modified Laboratory Rainfall Infiltration Soil Column Tests. The Mirafi® G-Serics Drainage Composites and Mirafi Nylon Wicking Fabric remained saturated for a longer period of time than the Mirafi® FW402 and the Mirafi® HPS70.

There were two series of tests performed for each geosynthetic. The average moisture distribution was found between the series of tests for each geosynthetic. The average moisture content throughout each averaged moisture distribution was also found. The data that was recorded in preliminary tests was not used in data analysis due to different original moisture contents.

For the sill water infiltration test using Mirafi Nylon Wicking Fabric, the moisture contents near the top of the specimen shifted to the left and were slightly lower than the moisture contents near the bottom. This observation may be due to the influence of gravity. The moisture distribution for each series of tests is relatively stable in that the moisture distribution trend docs not drastically change. In three days, the average moisture content reduced from 53% to about 40%. For the moisture distributions for the silt water infiltration test using Mirafi® HP570, the moisture contents near the lop of the specimen shifted to the left and were slightly lower than the moisture contents near the bottom. There was also a slight bend in the moisture distribution for each series of tests. In three days, the average moisture content reduced from 53% to about 43.35%. The moisture distributions for the silt water infiltration test using Mirafi® G-Series

Drainage Composites shows a lower moisture content on top, a slight bend in the middle, and a dipping moisture content at the bottom. In three days, the average moisture content reduced from 53% to about 43.54%.

For the moisture distributions for the silt water infiltration test using MirafiΦ FW402, more moisture was allocated in the central area of the cylinder. The lowest moisture contents arc at the bottom rather than the top. In three days, the average moisture content reduced from 53% to about 42.09%.

Of the average moisture distributions for the Mirafi Nylon Wicking Fabric, the MirafiΦ HP570, the Mirafi® FW402, and Mirafi® 0-Serics Drainage Composites, the Mirafi Nylon Wicking Fabric has the lowest moisture distribution. The average moisture distribution was found by taking the average of the results for the first and second series of water infiltration tests. The average moisture content throughout each averaged moisture distribution shows that the Mirafi Nylon Wicking Fabric has removed the most moisture from the silt slurry and the difference in the moisture contents varied from 2% to 3.5%. It is known that the undrained shear strength of fine grained soils can increase about 20% for 1% reduction in the moisture content. This means by using the wicking fabric, the undrained shear strength of soil slurry can be 45% to 90% higher compared with soil treated with other gcosynthetics.

The above test results have some important implications on the use of the wicking fabric of the present invention in a pavement structure. Usually, after a pavement structure is constructed, the moisture content in the pavement structure will crease due to the following reasons:

1. the evaporation is prevented in the vertical direction by the asphalt pavement;

2. accumulation of rainfall infiltration from cracks in the pavement, and

3. capillary rise of water induced by frost heave and other reasons.

As a result, there could be excess water in the pavement structure which would be much higher than the optimum moisture content when the soil was originally compacted.

Consequently, there will be increasing differential settlement and reduced shear strength of the soil. From the above test results, it can be concluded that inclusion of the wicking fabric in a pavement structure can lead to reduced moisture content, increased shear strength of the soil, and reduced different settlement. All of these are expected to significantly improve the performance of the pavement structure and service life. The rainfall infiltration tests for medium sand actually simulated the situations when there is a traditional gravel drainage layer in the pavement structure. In Alaska, A 4 inch thick Dl material is usually used for drainage purpose as swell as to prevent frost heave and thaw weakening. Its characteristic is similar to the sand used in the rainfall infiltration tests. When there is rainfall infiltration, a considerable amount of water can be trapped in this layer and cannot be drained out of the pavement structure. The sand is near saturation, when it is exposed to air, the water cannot be drained out under due to even small negative suction. The moisture content is about 25*/«. Under this situation, inclusion of the Mirafi® FW402 and Mirafi®HP570 may not help for draining the water, while the wicking fabric can help to reduce the moisture content by transporting the water out of the pavement structure.

When there is rainfall infiltration from the top of a pavement structure, a considerable amount of water will be trapped in the both the drainage layer and the silty soil layer. Under moisture situation, the relative humidity in the air is less than 90%, which corresponding to suction value of 10 MPa. As a result, soil exposed to air will dry out quickly and become nearly impermeable under negative pore water pressure (suction). The soils at the shoulder work like a large plastic mold. With a layer of geosynthetic reinforcement such as the Miraft® FW402 and Mirafi® HPS70. the differential settlement would still be large, as none of the Mirati® FW402 and Mirati® HPS70 can transport water under negative pore water pressure (suction). The Mirafi® G-Scries Drainage Composites cannot work very well to since it is design to transport water under positive pore water pressure conditions. There is high stress concentration in the pavement structure.

By contrast, when there is a layer of the wicking fabric made in accordance with the present invention in the pavement structure which has a high ability to transport water under negative pore water pressure in the transverse direction, the water content will be more uniformly distributed in the pavement structure along the wicking fabric as any suction difference can lead to water flow. As the suction value at the shoulder is higher, it can (1 ) reduce the moisture in the pavement structure, and (2) make the moisture content to be more uniformly distributed in the top soil layer in the transverse direction. Both effects are beneficiary to improvement of pavement performance and service life. When there is less water in the pavement structure, it is also expected that there is less chance of frost heave during winter. The wicking fabric made in accordance with the present invention aides in reducing the moisture content in the soil. In summer seasons when the soils are completely unfrozen, the suction in the center of the pavement structure is low, which corresponds to a high relative humidity (usually greater than 99.9%). The relative humidity in the air under most situations is less than 90%, which corresponds to very high suction. Once installed in the pavement structure, the wicking fabric can provide good water transportation channel under unsaturated conditions, the soil on both sides of the wicking fabric tend to be as dry as the soil near the shoulders of the pavement structure in order to maintain an equilibrium in the metric suction (or relative humidity). In this way, it can generate a zone with low water content and consequently low unsaturated permeability. This zone can work as a capillary barrier when winter comes because of the reduced unsaturated permeability of the soil. In addition, since there is less in-situ water, there is less frost heave, too.

During winter, the wicking fabric in the pavement structure can aide in preventing frost heave. The freezing process starts from the outside towards inside of the pavement structure. When the soil along the shoulder freezes, free water in the soil becomes ice, which reduces the unfrozen water content in the soil and increases the suction in the soil at the shoulder. The soil in the core of the pavement structure normally has higher moisture content and low suction value.

As a result, water flows from the core to the shoulder of the pavement structure, which also generates a zone with lower moisture content than it should have if there is no wicking fabric. As the freezing front approaches the wicking fabric from the top, there will be less frost heave.

From the Rainfall infiltration/lop test results, it is concluded that inclusion of the wicking fabric in the pavement structure also helps improve performance of the pavement structure during thawing seasons. When thawing occurs, it starts from the outside toward inside. The thawing process can be not uniform and cause ponding of water in the pavement structure as the frozen soils arc usually impermeable. From the Rainfall infiltration/top test results, it is found that the wicking fabric drains water from the top, which is very helpful for water reduction. The following conclusions can be obtained from the above analysis: 1. Traditional granular drainage layer cannot drain water our of the pavement structure under low suction value. The granular material can hold a considerable amount of water at the field capacity. 2. The Mirafi® HP570, the Mirafi® FW402, and the Mirafi® G-Scries Drainage Composites cannot drain water our of the soil under unsaturated situations. It was found that these geosynthetics dried out quickly when exposed to air. When these gcosynthetics are dry, they are impermeable to unsaturated water flow. 3. The wicking fabric of the present invention can maintain wet and work as a very good channel for water transportation under high suction values. All the test results indicated that the wicking fabric helps effectively reduce the water in the soil under negative pore water pressure.

4. Analysis indicates that if properly designed, inclusion of the wicking fFabric in a pavement structure can effectively reduce the moisture content in the pavement structure at all seasons.

Example 12: Moisture Migration

Two tests were performed to investigate the performance of different geosynthetics during the frost heave process: exterior and interior moisture migration tests. In the exterior moisture migration tests, the top halves of each soil specimen were surrounded by a geosynthetic and an impermeable layer which was wrapped around the geosynthetic. In the interior moisture migration tests, each geosynthetic was placed vertically within each soil specimen. The material that was used was a silty soil taken from the CREEL Fairbanks permafrost tunnel. The geosynthetic materials that were used in the tests included Mirafi® FW402. Mirafi® G-Series Drainage Composites, Mirafi® HP57O, and Mirafi Nylon Wicking Fabric. At least one reference was also made for each set of tests. The main purpose of the two groups of tests was to evaluate the moisture migration performance of each geosynthetic.

In the preliminary tests, water was allowed to infiltrate the soil from a water bath within the frost heave apparatus. A hammer was used in the preliminary tests as well. The soil was prepared at a moisture content of 25%. The soil specimens are then installed in the frost heave apparatus for frost heave tests. During the frost heave tests, the soils were frozen downward with the temperature maintained at -7 ⁰ C at the top of the specimen and 1°C at the bottom. The soil specimens were surrounded by insulation materials at the lateral in order to make sure the frozen process is one dimensional. During the frozen process, the temperatures at five different locations of the specimens were measured to monitor the freezing process. The frost heave is measured at the top of soil specimens using LVDTs. Usually, the tests lasted for at least three days until the soil specimens were fully frozen.

The tests that followed the preliminary tests were performed differently in the following ways. First, in the interior vertical moisture migration tests, water was not allowed to infiltrate the soil. Second, the soil was not compacted using a hammer. Instead, the sides of the plastic molds were tapped to eliminate voids and/or air bubbles within the soil. Third, the moisture content was increased to 40Vo. The amount of frost heave with respect to time has also been shown in this set of tests.

After frost heave test, each specimen was taken out and cut into six approximately equal portions along the height. AU six portions were then put into oven and the moisture contents were determined. From 0 to 4 inches, the Mirafi® FW402 moisture distribution remains relatively constant and centered on the original moisture content of 25%. From 4 to 8 inches, moisture migrated towards the top. Instead of a constant moisture distribution as seen in the 0 to 4 inch portion, the moisture distribution from 4 to 8 inches shows higher moisture contents at the top and lower moisture contents at the bottom.

There is a slight increase in moisture content just before the 4 inch mark. This is due to the impermeable layer impeding water migration, which resulted in a slight buildup of ice lenses. Although this is an open system, the moisture distribution below the 4 inch mark reach a constant moisture distribution that is slightly higher than the original moisture content, indicating there is little water migration into the soil.

From 0 to 4 inches, the Mirafi® G-Series Drainage Composites moisture distribution increased from 25% to 28%. From 4 to 8 inches, moisture migrated towards the top. The moisture contents at the top of the soil specimens reached 32.8%, 7.8% higher than the original moisture content of 25%. At 6 inch mark, the moisture content was 21.9%, 3.1% lower than the original moisture content. The lower part had a moisture content of 23%, 2% lower than the original moisture content. It was also found that at the upper part of the interface, the soil is relatively dry. All these indicated that there was a migration of water in the from 4 to 8 inches.

For the Mirafi® HP570, from 0 to 2 inches, moisture distribution remains relatively constant and centered on the original moisture content of 25%. From 2 to 4 inches, the moisture content increased from 25 % to 28.1 % due to water intake in this open system. The impermeable layer impedes water migration, and a buildup in moisture content at the 4 inch mark below the interface is expected. From 4 to 8 inches, moisture has migrated towards the top. Instead of a relatively constant moisture distribution as seen in the 0 to 4 inch portion, the moisture distribution from 4 to 8 inches shows higher moisture contents at the top and lower moisture contents at the bottom. The moisture content at the top was as high as 30.7%, 5.7% higher than the original moisture content while at the bottom it was 18.8%, 6.2% lower than the original moisture content. These results as well as the previous result clearly showed that the moisture was migrating during the freezing process. Since the upper part was a closed system, the overall moisture content should be maintained the same.

The Mirafi Nylon Wicking Fabric moisture distribution was determined by averaging the results of two different tests performed with Mirafi Nylon Wicking Fabric under the same conditions. From 0 to 4 inches, the Mirafi Nylon Wicking Fabric moisture distribution slightly increased. From 4 to 8 inches, moisture migrated towards the top. Instead of a relatively constant moisture distribution as seen in the 0 to 4 inch portion, the moisture distribution from 4 to 8 inches shows higher moisture contents at the top and lower moisture contents at the bottom. The moisture content at the top was as high as 30.6%, 5.6% higher than the original moisture content while at the bottom it was 15.8%. 9.2% lower than the original moisture content. These results as well as the previous result clearly showed that the moisture was migrating during the freezing process. Since the upper part was a closed system, the overall moisture content should be maintained the same. With respect to a reference soil specimen , the difference between the reference soil specimen and the previous one are that there is no geosynthetics in the reference soil specimen. The tendency of the moisture distributions are similar to those with geosynthetics. From 0 to 4 inches, there is a slight increase of moisture content due to the water intake and the average moisture content increased from 25% to about 26% with a relatively constant distribution. From 4 to 8 inches, moisture distribution linearly increased with height with 15.2% at the bottom. 22.9% in the middle and 29.6% at the top. This distribution indicated that there was water migration from the bottom to the top during the frost heave.

In the exterior moisture migration test, the original moisture content of the soil was 40% in order to simulate the situation when there is excess water in the pavement structure. For soil specimens with Mirafi® FW402, the lower part (0 to 4 inches) has water migration during the freezing process. Above the height of 3 inches, the moisture content of the soil specimen is greater than the original moisture content of 40%, indicating there is water intake. Below the height of 3 inches, the moisture content is smaller than 40%, indicating that the soil is drying. The moisture content at the bottom of the soil specimen was considered to be the initial moisture content due to free access to water. This test results indicated that for this specific soil, when the initial moisture content in the soil is high, the migration of water in the soil specimen is sufficient to match freezing process while the water intake from the water bath is very small.

This can be verified by results obtained from the upper part (4 to 8 inches), the middle of the soil specimen at 6 inches basically had the same moisture content as the initial moisture content. Above 6 inches, the moisture content is higher than 40%, indicating there was water intake. Below 6 inches, the moisture content is lower than 40%, indicating there was water loss. The top half is a closed system and the total moisture maintained constant.

For soil specimens with inclusion of Mirafi® G-Series Drainage Composites, the lower part (0 to 4 inches) has water migration during the freezing process. Above the height of 3 inches, the moisture content of the soil specimen is greater than the original moisture content of 40%, indicating there is water intake. Below the height of 3 inches, the moisture content is smaller than 40%, indicating that the soil is drying. This test results indicated that for this specific setup, when the initial moisture content in the soil is high, the migration of water in the soil specimen is sufficient to match freezing process while the water intake from the water bath is very small.

For the upper part (4 to 8 inches), the middle of the soil specimen at 6 inches basically had the same moisture content as the initial moisture content. Above 6 inches, the moisture content is higher than 40%, indicating there was water intake. Below 6 inches, the moisture content is lower than 40%, indicating there was water loss. The top half is a dosed system and the total moisture maintained constant.

The test results for soil specimens with inclusion of Mirafi® HPS70 differed. For the lower part (0 to 4 inches), the moisture content was lower than the initial moisture content. Yet, the moisture content was relatively uniformly distributed, which was not consistent with the results from the other tests in this group. However, the result was similar to the results in the preliminary exterior moisture migration tests. That is, when the initial moisture content was low, the moisture contents in the low halves of the soil specimens were relatively uniformly distributed.

For the upper part (4 to 8 inches), the middle of the soil specimen at 6 inches basically had a moisture content of 35.1 %, lower than the targeted 40%. The moisture content above the 7 inch mark is 45.1%, higher than the targeted 40%. The moisture content distribution shows the moisture migration during the freezing process.

The test results for soil specimens with inclusion of Mirafi Nylon Wicking Fabric and the soil specimen with no inclusion of geosynthetics, respectively, were similar to those with inclusions of Mirafi® FW402 and Mirafi® G-Series Drainage Composites. Both the upper and lower parts indicated water migration from bottom to top.

The moisture content for soils in the upper halves of the specimens of this group basically had similar sloped distributions, indicating that there is moisture migration during the freezing process. The differences in the moisture distributions among different geosynthetics were insignificant except the Mirafi® G-Serics Drainage Composites. The excess moisture content in the soil specimen with inclusion of Mirafi® G-Series Drainage Composites was due to the fact that Mirafi® G-Series Drainage Composites was initially wetted and contained more water. For the lower parts, there should have been no differences in the moisture content distributions since all the soil specimens had the same setup and access to water intake.

The interior vertical moisture migration test employed soil having an original moisture content of the soil was 40%. Except for the reference soil specimens, different geosynthetics were included vertically in the center of the soil specimens. Both ends of all soil specimens were sealed with no access to water. All soil specimens were frozen from to bottom with a constant temperature of -7 ºC at the top and 1 ºC at the bottom.

For Mirafi® FW402, the moisture content distributions below the 6 inch mark was basically uniform with a moisture content varying from 37.0% to 37.5%. Above the 6 inch mark, the moisture content increased to 48.4% at an approximate height of 7 inches. The results indicated that there was significant water migration at the very beginning of the freezing process. As the freezing front moved downward, the speed of water migration slowed down. This was demonstrated by the decrease in moisture content with height. The water migration gradually matched the moving freezing front and resulted in an approximately uniform moisture distribution below the 6 inch mark. For Mirafi® G-Series Drainage Composites, the moisture content distributions below the 3 inch mark was basically uniform with a moisture content varying from 36.6% to 36.8%. Above the 6 inch mark, the moisture content increased to 36.7% at 3.5 in mark to 52.9% at about 7 inch mark. Compared with previous soil specimen with inclusion of Mirafi® FW402, the sloped moisture distribution occurred at a larger depth of 3.7 inch. The moisture content below 3.7 inch mark was also lower than that in the previous soil specimen with inclusion of Mirafi® FW402.

The moisture distribution for Mirafi® HPS70 had the same pattern as that of the soil specimen with inclusion of Mirafi® G-Scries Drainage Composites. The difference was that the sloped moisture distribution was only in the range above S inch mark and the moisture content below S inch mark was basically uniform.

For Mirafi Nylon Wicking Fabric, the moisture content was the highest at the top (46.5%). It decreased to 35.9% till the mark of 3 inch and then slightly increased to 36.3 % and maintained relatively uniform below it. This was a closed system and the initial moisture content was 40% everywhere, it is evident that moisture migration was frost-induced.

The experimental results for the reference soil specimen were similar to those above, except that the reference soil transported more water to the top.

It can be concluded that when there is free water access, the Mirafi Nylon Wicking Fabric can transport water best as compared with the other gcosynthetics. Through the above groups of tests, several conclusions can be made: During the freezing process, capillary forces (suction) can be generated due to the freezing of tree water. When the soil is under the influence of suction, the soil is unsaturated and there is still water migration. Water migration due to the freezing process is the reason for the frost heave. In these group tests, it is shown that the Mirafi

Nylon Wicking Fabric, a wicking fabric made in accordance with the present invention, has high transmissivity under unsaturated conditions. Such property, when property used, can be used to prevent the frost heave for pavement in the cold regions.

With respect to the above description then, it is to be realized that the optimum dimensional relationships for the parts of the invention, to include variations in size, materials, shape, form, function and manner of operation, assembly and use, are deemed readily apparent and obvious to one skilled in the art, and all equivalent relationships to those illustrated in the drawings and described in the specification are intended to be encompassed by the present invention.

Therefore, the foregoing is considered as illustrative only of the principles of the invention. Further, various modifications may be made of the invention without departing from the scope thereof and it is desired, therefore, that only such limitations shall be placed thereon as arc imposed by the prior art and which are set forth in the appended claims.