Technical Manual. Advanced Crack Mitigation Technology for Asphalt Pavement Overlays. Stronger ideas for a sustainable world

Transcription

1 Technical Manual Advanced Crack Mitigation Technology for Asphalt Pavement Overlays Stronger ideas for a sustainable world

2

3 I. INTRODUCTION AND HISTORY OF CRACKING II. III. IV. ADFORS INTERLAYER SOLUTIONS GlasGrid RESEARCH GlasGrid GP RESEARCH V. INSTALLATION VI. VIi. ENGINEERING RESOURCES REFERENCE LITERATURE - 1 -

4 I. Introduction and History of Cracking Many pavements, which are considered to be structurally sound after the construction of an overlay, prematurely exhibit a cracking pattern similar to that which existed in the underlying pavement. This propagation of an existing crack pattern, can form discontinuities in the old pavement, into and through a new overlay and is known as reflective cracking. Cracking and deformation that typically deteriorates flexible pavements are attributed to excessive traffic loading, age hardening of asphalt binder, or temperature cycling. Moisture can readily enter a cracked pavement, weakening the underlying layers, further accelerating damage to the pavement. Research has shown that cracks are likely to reappear at a rate of 2 cm (1")/year after rehabilitation construction such as in unreinforced asphalt overlays. Many new pavement overlays are subject to reflective cracking from the underlying pavement into the new overlay. These premature reflective cracks significantly decrease structural strength and allow water to penetrate into sub-layers, resulting in severe damage. The cracking in the new overlay surface is due to the inability of the overlay to withstand shear and tensile stresses created by movements of the underlying pavement. These movements are mostly caused by either traffic loading or by thermal cycling (expansion and contraction). Traffic Loading Load associated cracking occurs when shear and bending forces due to heavy traffic loading create stresses that exceed the fracture strength of the asphalt overlay. This is a pavement stability issue. Pavement instability is generally due to the presence of one or more of the following: poor soil conditions, improper drainage, increased traffic load, and age. Unstable Portland cement concrete (PCC) slabs are often identified by excessive movement or deflection during loading accompanied by the presence of water and fines pumping upward at the joint. Once these fines exit the pavement structure they leave behind a void. Voids negatively affect support for the PCC slab, resulting in the slab becoming even less stable

5 Instability in asphalt cement concrete (ACC) pavement is typically characterized by a series of closely spaced, multidirectional fatigue cracks. The distinctive pattern is often referred to as alligator cracking because of its similar appearance. I. Introduction and History of Cracking Alligator Cracking Pavement rehabilitation strategies for fatigue include drainage improvements such as edge drains and surface sealing. Structural improvements for full depth asphalt include subgrade reinforcement and sufficient structural overlay thickness to adequately support the traffic load. Thermal Stress Temperature associated cracking occurs when horizontal movement, due to thermal expansion, contraction and curling of base pavement layers, creates tensile stresses in the overlay that exceed the strength of the asphalt. Overlays placed on both ACC and PCC pavements are subject to thermal cracking. Thermal cracks usually appear in transverse and longitudinal directions. Temperature cycling occurs over an extended period of time. The resultant horizontal stress loading happens at a very slow rate, as compared to traffic loading stress rates. Under these very slow loading rates, the stiffness or fracture resiliency of the asphalt material is quite low, perhaps 1,000 to 10,000 times lower than the modulus exhibited by these same materials under traffic induced loading rates. Crack growth due to thermal stress in an overlay Overlay Old Pavement Contraction & Curling Flexible overlays placed on PCC pavements are particularly susceptible to thermal cracking at the slab joints. Thermal rates of expansion and contraction vary between materials such that any slab joint spacing almost always assures premature joint reflection. PCC Joint Cracking Longitudinal Cracking Transverse Cracking - 3 -

6 II. ADFORS Interlayer Solutions Future Life Cycle Costs The Economical Solution When asphalt pavement cracks, water permeates the base and reduces the life of a road. Improving the asphalt s natural ability to resist cracks improves the drainage capabilities of a road, and the Federal Highway Administration (FHWA) suggests that by doing this a road s life can be extended. GlasGrid improves asphalt s natural ability to reduce cracking by up to 3 times by providing a tensile element to the asphalt. This helps keep water out of the base and improves the drainage capabilities of a road by 10%. Based on 25+ years of successful installations around the world, GlasGrid typically provides a 50% reduction in future investment cost (e.g. maintenance, rehabilitation and use. costs) over the life of The GlasGrid Product Line With Glasgrid* Without Glasgrid *With added life benefit of 1.1 due to improved drainage FHWA GlasGrid is composed of high strength glass filament yarns knitted into a grid and coated with an elastomeric polymer. ADFORS advanced coating improves the toughness and durability of glass filaments, and enhances the bond between layers of asphalt. It is also backed with a pressure sensitive adhesive to improve the ease of installation. GlasGrid TF (Tack Film) provides all the benefits of our conventional GlasGrid with the added benefit of a highly engineered film designed to replace the need for spray tack coats on new leveling/binder course. The tack film enhances the performance of the GlasGrid, resulting in extension of life up to 5 times better than unreinforced asphalt. Route Type Traffic [ESALS] Waterproofing Required or Application to Milled Surface Alligator Cracking (Aging) Block Cracking Cracks < 6 mm (1/4 in) Product Is Applied: Full Width Detail Repair Full Width GlasGrid GP & GlasGrid CG GlasGrid GG, GlasGrid TF 2 & GlasGrid GP Residential Streets, Minor Country Roads, Golf Cart Paths, Parking Lots, Tennis Courts, Bike Trails < 300K < 1 % Heavies GP25 GP50 CG50 GGPG GGPM GP25 GP50 GG50 GP25 GP50 GG50 County, Regional or Municipal Connector Roads 300,000 1,000, % Heavies GP25 GP50 CG50 CG100 CG200 GP25 GP50 GG50 GP25 GP50 TF100 GG50 Interstate Highways or Inter-Urban Roads > 1M > 5% Heavies GP25 GP50 CG50 CG100 CG200 TF100 GG50 TF100 GG50 Airports Private / Municipal General Aviation Traffic GP25 GP50 CG50 CG100 CG200 GP25 GP50 TF100 GG50 GP25 GP50 TF100 GG50 Airports Regional / International Commercial Traffic CG50 CG100 CG200 TF100 GG50 TF Industrial Ports or Intermodal Facilities Axle Loads > 20kip CG50 CG100 CG200 TF100 TF100

7 GlasGrid CG (CompoGrid) is a composite material consisting of our conventional GlasGrid bonded to a nonwoven paving geotextile that provides the added benefit of waterproofing and the ability to be installed on milled surfaces. The nonwoven geotextile is an AASHTO approved paving fabric. II. ADFORS Interlayer Solutions GlasGrid CGL (CompoGrid Lite) is a composite material consisting of our conventional GlasGrid bonded to a lightweight nonwoven polyester geotextile that provides the added benefit of waterproofing, using a fraction of the tack coat required by standard paving fabrics. GlasGrid PG (Patch) is our CGL composite material pre-saturated with bitumen to create a rapid repair system for asphalt reinforcement of small areas. The product can be installed directly on a milled surface and provides a waterproofing benefit as well as a mitigation for reflective cracking without the need of an external bond coat. GlasGrid PM (Manhole Patch) is our CGL composite material pre-saturated with bitumen and die cut to create an asphalt reinforcement repair system for around manholes. The product can be installed directly on a milled surface and provides a waterproofing benefit as well as a mitigation for reflective cracking without the need of an external bond coat. GlasGrid GP (GlasPave TM ) is a hybrid geosynthetic paving mat consisting of a high strength, fiberglass grid, coated with an elastomeric polymer, and embedded between two spun bond polyester textiles. GlasGrid GP provides the waterproofing benefit of a paving fabric, while only requiring half the amount of bitumen to provide the same level of waterproofing. Additionally, GlasGrid GP provides a tensile element at low strain that conventional paving fabrics do not. Block Cracking Cracks > 6 mm (1/4") PCC Joint Reflective Cracks Thermal Cracking Warm Region Thermal Cracking Cold Region Lane Widening Cracks (Sand Subgrade) Lane Widening Cracks (Clay Subgrade) Full Width or Detail Repair GlasGrid GG, GlasGrid TF 2 & GlasGrid GP Detail Repair GlasGrid GG & GlasGrid GP GP25 GP50 TF100 TF100 GG50 GP25 GP50 TF100 GG50 GP25 GP50 GG50 GG100 GP25 GP50 GG100 TF100 TF100 GG50 GG100 GG100 TF100 GG50 GG100 TF100 TF100 GG50 GG100 TF100 TF100 2 GlasGrid TF (Tack Film), a film tack coat is bonded to the GlasGrid and can be used for applications with fresh leveling course. Available in TF100 for full width coverage

8 II. ADFORS Interlayer Solutions High Potential Crack Reflection Activity Low Non-Woven Fabric Low Material Composition As with any engineered product, it is essential to begin with quality raw materials. Asphalt reinforcement materials must provide increased tensile strength at low deformation. ADFORS uses E glass with an elastomeric polymer coating that has been developed over a 25 year period. This process improves GlasGrid s resistance against installation damage and enhances the bond within the pavement composite. It also provides a fundamental increase to tensile strength at a very low deformation. Our coating is fully compatible with asphalt, thermally stable, resistant to creep and will not chemically breakdown over time. Ease of Installation Environmental and site specific conditions vary greatly. Practical application of any reinforcement requires the ability to conform to a wide range of paving operations. The use of pressure sensitive adhesive permits an efficient and uniform placement with the product remaining secure during the entire paving operation. Interlayer Strain Resistance High High Tensile Strength High modulus E -fiberglass (73,000 MPa/10.6 million psi) exhibits a tremendous strength-to-weight ratio and is pound for pound stronger than steel. With glass having a real modulus advantage over asphalt concrete, GlasGrid clearly provides the stiffness required to redirect crack energy. Grids made with C-glass have a lower tensile strength and modulus compared to E-glass. C-glass is less water and alkali-resistant than E-glass and is not recommended for asphalt reinforcement. Thermal, Lane Widening or Block Cracking Block or Flexural Cracking Paving Mats Glass Based Composite Grids GlasGrid & Steel Grids Polyester Based Composite Grids Low Elongation Asphalt concrete pavements possess poor elastic qualities and once oxidization has taken place, they become extremely brittle. Micro-cracking can be easily observed in pavements, even with minimal deformation. As pavements crack with less than 3% elongation, materials with elongation greater than this cannot utilize their full tensile potential. Polymer coated glass filaments realize their full tensile strength with less than 3% elongation, a necessary characteristic for a pavement reinforcement. No Long-Term Creep Many reinforcement materials that appear to be initially stable exhibit creep deformation due to constant stress over long periods of time. Fiberglass is ideal to withstand long-term tensile stress without creep deformation. These types of stresses are commonly found in asphalt pavements due to thermal movement and are a leading cause of reduced pavement performance. Since fiberglass exhibits no creep, it assures long-term performance

9 Asphalt Compatibility The specially formulated elastomeric polymer coating was designed to deliver superior asphalt compatibility. The benefit of an engineered coating results in the stress energy within the pavement to be directly channeled to the glass filaments with minimal transfer loss. The coating is penetrated to the center of the glass filaments; all filaments are thoroughly protected, they are well bonded together, which further increases the mechanical strength of the material overlay. The elastomeric coating has been designed to soften and bond extremely well to hot mix asphalt, unlike bitumen-coated grids where the bitumen coating can melt and flow away from the grid when contacted with hot asphalt. Thermal Stability GlasGrid is thermally stable well above 232 C (450 F), ensuring stability throughout the paving process, including the elevated temperatures commonly found with polymer modified asphalt mixes. Fiberglass itself possesses a melting point above 1000 C (1800 F). Therefore, GlasGrid does not shrink when contacted with hot mix asphalt like grids or fabrics made from polypropylene. Grid shrinkage under hot asphalt can cause grid movement and premature cracking. II. ADFORS Interlayer Solutions Polymer Coated GlasGrid Filaments Glass Fiber Polymer Coated GlasGrid Filaments Chemical Stability The elastomeric polymer coating has been designed to protect our GlasGrid products from chemicals typically found in asphalt pavements. This ensures that the reinforcement can perform effectively for many years. Millability and Recyclability With the exception of steel grids in thin asphalt overlays (steel grids are not millable), most geosynthetic interlayer systems are millable using conventional reclaiming equipment. However, when it comes to recycling reinforced asphalt, pavements reinforced with the GlasGrid GG/GlasGrid GP Systems have been proven to be reusable in other road projects as a recycled asphalt pavement or RAP, confirmed by RWTH Aachen University (Germany) and the National Center for Asphalt Technology (NCAT-US). Physical Durability The specially formulated elastomeric polymer coating provides protection from mechanical damage during construction including asphalt compaction, safeguarding its ability to perform over the long term. Some competitive grids are sold without any protective coating. Though these uncoated grids meet the requested nominal strength, they demonstrate a lack of compatibility with the asphalt resulting in poor bonding and increased risk of glass filament damage during installation. Width Due to the desirable low elasticity of GlasGrid, the standard widths of 1.5 m (59") and 2.0 m (78.75") are ideal on road curves. Relocating the interlayer joint farther away from centerline of the road significantly improves the ability to reinforce the pavement construction joint. For detailed crack repair applications using our GlasGrid 8502/8512 or GlasGrid PG (Patch) our standard width of 1.5 m (59") or 1.2 m (47.25") ensures a minimum of 0.75 m (2.5') or 0.6 m (2') of reinforcement respectively on either side of a crack to control crack energy. GlasGrid GP (GlasPave) is supplied in widths up to 3.8 m (150") to cover lane widths with minimal overlaps. The dimensions of the GlasGrid makes it easier to handle and install leading to a well performing project

10 II. ADFORS Interlayer Solutions Interlock and Confinement Asphalt concrete gains its compressive strength through compaction. The aggregate mixture is specifically selected to provide strong interlock and confinement within the load bearing stone structure, and asphalt cement (AC) is the glue that holds the particles together. The quality of both the aggregate and the AC will determine the quality of the final pavement structure. As particles strike through the GlasGrid apertures, they become mechanically interlocked within the composite system. This confinement zone impedes particle movement. Asphalt mixtures can achieve better compaction, greater bearing capacity and increased load transfer with less deformation. The gradation of the asphalt mix determines the aperture size. For asphalt mixes with a maximum particle size 12.5 mm (½") or less, a product with a standard sized aperture of 12.5 mm by 12.5 mm (½" by ½") is appropriate. For coarser mixes, a larger 25 mm by 25 mm (1" by 1") aperture grid should be used. The aperture size may also be selected based on local environmental conditions, mix stability, past performance or user preference. Overlay Test Results: Three distinct modes of failure are observed in overlay tests. Mode I: Cracks that propagate from the bottom up straight through the interlayer Mode II: Cracks that propagate from the bottom up and are then redirected/ dissipated horizontally along the interlayer Mode III: Cracks that propagate from the bottom up to the interlayer and then start again from the top down Modes I and III occur when the interlayer acts as a strain relieving material and Mode II occurs when it acts as a reinforcing material. Bottom-up cracks reach the interlayer under two scenarios, depending on the geosynthetic type and interlayer conditions. In strain relieving mode, the crack stalls at the interlayer level and then starts moving upward into the overlay. In the reinforcing mode, the crack turns horizontally and propagates to just below the grid at a distance where it has no more energy to propagate any further. Tested in the lab, GlasGrid produces Mode II failure with conventional tack coats

11 GlasGrid may be the most tested interlayer product on the market. Since its' introduction in the early 80 s, GlasGrid has been tested by numerous facilities including: III. GlasGrid Research Texas Transportation Institute (TTI) and Texas A&M University (US) Fatigue Testing Thermal Cracking Testing Delft University (Netherlands) Thermal Cracking Testing present - National Center for Asphalt Technology at Auburn University (US) Full Scale Accelerated Load Test Track The University of Nottingham (UK) Fatigue Testing Reflective and Thermal Cracking Testing University of PARMA (Italy) Flexural Beam and Slab Testing EMPA (Swiss) Fatigue Testing MMLS IFSTTAR (France) Full Scale accelerated load test track The use of interlayers for reflective cracking has been extensively researched over the last 30 years. In particular, seven key research projects have quantified the benefits of using GlasGrid and help define its areas of application. This testing and research has continually supported the initial proposition by TTI/A&M, that GlasGrid improves the life of an asphalt pavement by % times through reflective crack retardation

12 III. GlasGrid Research Texas Transportation Institute (TTI) & Texas A & M University For many years, engineers have investigated the use of interlayers within the overlay to reduce the effects of reflective cracking. Interlayers can dampen stress, relieve strain, and provide tensile reinforcement to the asphalt. TTI is a higher education-affiliated research agency, seeking solutions to these problems and challenges facing all modes of transportation. The Overlay Tester at TTI has become the standard for looking at fracture performance of asphalt and testing the effects of load cycling. The effects of many interlayer materials of varied strengths, configurations, tack coats and embedment quantities have been evaluated at TTI. Schematic Diagram of the Overlay Test Schematic Diagram of the Beam Fatigue Test 380 mm (15") Grid 19 mm (3/4") P P Grid Fixed Plate Movable Plate R R Studies using the Overlay Test and the Beam Fatigue Test on reinforced asphalt beams demonstrated a two-fold to three-fold improvement in the life of a GlasGrid reinforced overlay compared with an overlay constructed using the same thickness of unreinforced asphalt. These two test methods are still used widely to evaluate the performance of asphalt mixes and interlayer systems. In addition to the main laboratory testing undertaken at Texas A&M University, a reflective cracking performance prediction model was developed using the test data. Using traffic, temperature and pavement geometry variables, comparisons were made of the predicted performance for unreinforced and GlasGrid reinforced overlays. For the example shown below, the predicted performance benefit of the 100kN GlasGrid reinforced overlay is 1.5 to 2 times that for the unreinforced overlay. Number of Days in the Life of an Overlay 1,500 days 7.6 cm (3") thick overlay 12.7 cm (5") thick overlay 17.8 (7") cm thick overlay 2x 1000 days Unreinforced 2x 500 days GlasGrid 2x GlasGrid GlasGrid Unreinforced Unreinforced Unreinforced 30 F 50 F 70 F 30 F 50 F 70 F 30 F 50 F 70 F 17 C 28 C 39 C 17 C 28 C 39 C 17 C 28 C 39 C Temperature Differential

13 Delft University (Netherlands) Following an extensive field performance evaluation study and comprehensive laboratory testing program, a performance prediction model was developed for pavement overlays subject to reflective cracking caused by thermally induced stresses. This model was subsequently used to develop ARCDESO (Anti-Reflective Cracking Design Software.) For 10 years, ARCDESO has demonstrated that for a range of input parameters, the reflective cracking design life of an asphalt overlay reinforced with GlasGrid can be more than double that of an unreinforced overlay. The graph below shows a simulation of the predicted percentage of reflected cracks for two rehabilitated pavements; one with a 20mm (0.79") thick leveling course with a PG64-22 CRS tack coat applied at 0.16 L/m2 (0.035 gal/yd2) followed by 60 mm (2.36") thick wearing course and two, a 20 mm (0.79") thick leveling course using GlasGrid 8501 interlayer, with a PG64-22 CRS tack coat applied at 0.16 L/m2 (0.035 gal/yd2) followed by 50 mm (1.97") thick wearing course. III. GlasGrid Research Percentage reflected cracks mm GG mm PG64 CRS mm No GG 60 mm PG64 CRS Years The University of Nottingham (UK) An interface bond test was used to measure the quality of the bond between various interlayers and the asphalt. The test results concluded that the presence of a non-woven fabric and not grids result in a serious reduction in the interface shear stiffness. This explained why the grid systems without a fabric performed better than the systems with fabric. Semi-continuously supported beam tests were also conducted at the University of Nottingham to determine the ability of interlayer materials to resist crack propagation in notched asphalt beams. This test simulates a stress distribution similar to that found in pavements under normal traffic conditions. A software program "OLCRACK" was developed that includes fatigue factor to the other sections in order to represent the reduced rate or increased rate of crack propagation; i.e., the greater the fatigue factor, the lower the crack propagation rate. The fatigue factor and bond strength associated with a particular interlayer are key input parameters for the performance model developed. Based on the data shown, the model predicts that a GlasGrid interlayer will enhance the crack propagation resistance for an asphalt overlay by a factor of 2 to 3 times. Direction of Load Application Interface Bond Test Apparatus Rollers Epoxy Stiffener h Interface Normal Load O Specimen Load Cell Tie Bar Epoxy (continued...)

14 III. GlasGrid Research EMPA EMPA is the Swiss federal laboratory for materials science and technology, oriented to meeting the requirements of industry and the needs of our society. GlasGrid was tested using an Model Mobile Load Simulator (MMLS3) device that induces a unidirectional load to the pavement to simulate traffic loading. The MMLS3 is comprised of four pneumatic tires mechanically linked together in a loop as shown in the diagram below. Each wheel simulates a scaled load of a single tire. These tires are 300 mm (11.8") in diameter, approximately one-third the diameter of standard truck tires. This device can be used both in the laboratory and in the field on actual pavement. Crank for Height Setting 300 mm (11.8") Pneumatic Tire Drive Motor Model Mobile Load Simulator From this test it was determined that with GlasGrid about 3 times longer resistance against reflective crack induced failure at constant room temperature was found than for the unreinforced control slab. Visual inspection of the reinforced slabs showed that the cracks were interrupted by propagating along the reinforcing grids. Model Mobile Load Simulator (EMPA, Swiss) Cycles Control GlasGrid MMLS3 Test temperature 20C (68F) 7200 passing per hour Tire pressure: 6bar (87 psi) Load: 2.1 kn (472 lbf) Test Setup grid 30 mm (1.2") 25 mm (1") 25 mm (1") 30 mm (1.2") 12 mm (0.47") L=1405 mm (55.3") Cuts, depth 25 mm (1") Passing direction mm 95 mm 95 mm 150 mm 150 mm mm 25 mm 245 mm (9.6") 1-6 deformation sensors 175 mm (6.9") 325 mm (12.8")

15 University of Parma The objective of this research was to quantify the effectiveness of the GlasGrid interlayer system to affect fatigue cracking and extend pavement service life. Center point bending tests were conducted according to ASTM E399, using beam and slab samples designed to simulate bottom-up cracking. Both GlasGrid and GlasGrid TF were tested to evaluate their benefits to tensile stress/strain reduction and reflection crack prevention. III. GlasGrid Research Test specimens were 60 mm (2.36") thick, composed of leveling 20 mm (0.78") and wearing 40 mm (1.57") layers to simulate field cross-sections. Beam specimens were 400 mm x 100 mm (15.75" x 3.94") and slab specimens were of 500 mm x 500 mm (19.7" x 19.7"). Testing was conducted at 20 C (68 F) for the control and 100 kn/m GlasGrid reinforced specimens under a central loading metal dowel. This configuration allowed for the evaluation of the interlayer's ability to enhance the overlay system behavior in terms of post-cracking ductility and energy absorption. Beam Samples GlasGrid reinforced specimens showed about 1.5 times of a higher maximum load than unreinforced specimens. Significant differences were observed in fracture energy, to crack a specimen. Reinforced specimens responded to higher energy dissipation at the crack tip, requiring higher fracture energy to crack. As shown in the figure below, GlasGrid TF reinforced beams showed at least twice the fracture energy than that of GlasGrid reinforced beams. Three Point Bending Test Flexural Test Results Fracture Energy Control GlasGrid GlasGrid TF 0 Flexural Beam Test Flexural Slab Test Slab Samples Three point bending slab tests were performed to reduce the scale effects of beam samples and to evaluate the mesh (3D) effect. GlasGrid reinforced specimens provided enhanced ductile behavior, improving the tensile toughness of the asphalt. GlasGrid reinforced slabs absorbed more than 1.5 times the fracture energy compared to the unreinforced control specimens. Furthermore due to the enhanced bond, GlasGrid TF allowed asphalt concrete specimens to deform much more before the damage became great enough for a crack to initiate. To that end, GlasGrid TF reinforced specimens required at least three times higher strain to failure compared to the other reinforced specimens

16 III. GlasGrid Research University of Parma, continued Strain localization and damage distribution were recorded and measured by using a Digital Image Correlation (DIC) System that achieves a highly accurate strain map of the specimens during loading on a 60 mm x 60 mm (2.36" x 2.36") central area. The figures below show progressive tensile strain maps from the DIC. As can be seen in the images, the difference between reinforced and unreinforced specimens is dramatic. Strain maps on the reinforced specimen confirm GlasGrid s ability to act as a barrier against propagation of cracks to the upper layer. Conversely, the strain maps on the unreinforced specimen show a big crack developing in the central region which extends from the bottom edge to the surface layer along the vertical plane. No Grid Strain Contour Crack Initiation Crack Initiation GlasGrid GlasGrid TF 2 mm (0.08") 4 mm (0.16") 10 mm (0.39") 14 mm (0.55") Flexural Deflection

17 IFSTTAR The fatigue benefits of 100 kn/m GlasGrid were evaluated at the IFSTTAR facility in France. The IFSTTAR Accelerated Pavement Testing (APT) facility, in Nantes, is an outdoor circular accelerated pavement test track dedicated to full-scale pavement experiments. The test track consists of a central tower and four arms each 20 m long (65.6') equipped with wheels, running on a circular test track. The experimental circular pavement has a mean radius of 17.5 m (57.4') and a width of 6 m (19.7'), and thus a total length of approximately 110 m (360'). The position of the loading module can be adjusted for different radii on each arm. During loading, a lateral wandering of the loads can be applied to simulate the lateral distribution of loads in real traffic. III. GlasGrid Research Testing was carried out on low traffic pavement crosssections consisting of an 80 mm (3") thick bituminous wearing course, over a granular sub-base (300 mm or IFSTTAR Full Scale Testing Carrousel (Nantes, France) 11" thick), and a sandy subgrade soil, with a subgrade modulus of about 95 MPa. GlasGrid was placed in the lower part of the bituminous layer, 2 cm (0.8") above the interface with the subbase and covered by 5 cm (2") bituminous mix. The bituminous mix was a standard French 0/10 mm wearing course bituminous material. The four arms of the fatigue carrousel are equipped with standard dual wheels running at a constant speed of 6 rounds/minute (43 km/h). A load of 65 kn (standard French equivalent axle load) was applied to the test section up to 1,000,000 cycles

18 III. GlasGrid Research Crack Monitoring - Results Crack percentage was determined by the ratio between the length of pavement with cracks and the initial length. During the experiment, loading was stopped approximately every 100,000 cycles to perform various distress measurements. The first cracks were observed in the unreinforced section after 800,000 cycles. Due to no cracking in the GlasGrid section at 1,000,000 cycles, the load was raised to 70 kn for the remainder of the test. It should be noted that the surface cracks observed were based on crack initiation and not true crack reflection. Unreinforced GlasGrid Crack Maps at 1.2 Million Loading Cycles The test was then run until the extent of cracking increased such that the control section was too damaged to continue at 1.2 million cycles. At this point the control section exhibited a cracked area of 70%, while the GlasGrid reinforced section exhibited insignificant cracking. The figure below presents the evolution of the extent of cracking, as a function of the cycles of traffic. To compare the results of the GlasGrid section and the control we normalized the applied traffic loading to equivalent standard axles by applying the fourth power law. (AASHO Road Test AASHO 1961). The graph corresponds to the percentage of the length of the pavement affected by cracks (for a transverse crack, the affected length is considered, arbitrarily, to be 500 mm). GlasGrid delayed the cracks by a 2.4 times versus the control. 65 kn % Cracking Area GlasGrid Control 2.4x , , ,000 m 800,000 1,000,000 1,200,000 1,400,000 1,600,000 Cycles

19 North Carolina State University (NCSU) The rutting performance of a control slab (without reinforcement) was compared to a reinforced slab made using GlasGrid 8511 and was evaluated at North Carolina State University using the MMLS3 test at 50 C (122F). The slab configuration is shown below. The following conclusions were drawn based on 400,000 load cycles: III. GlasGrid Research The results from the reinforced slab confirmed that the reinforcement reduced the downward rut depth by 26% and also showed improved shear flow resistance compared to the control slab by reducing the height of the shear humps in the shear flow area and therefore, the total downward rut depth, by 25%. The increased load bearing capacity that is due to the tension resistance of the GlasGrid and the confinement and increased friction due to the aggregate particles locked in the GlasGrid openings seemed to be the major factors for this rutting performance improvement. Cross-Section View of Pavement Structure 81 mm (3.2") wheel 300 mm (11.8") Thermocouple Reinforcement + Tack Coat HMA Slab Tack Coat Street Base 559 mm (22") Verticle Displacement (mm) Transverse Distance (mm) Verticle Displacement (mm) Transverse Distance (mm)

20 III. GlasGrid Research National Center for Asphalt Technology (NCAT) The National Center for Asphalt Technology (NCAT) completed construction of an accelerated loading test track in 2000 at their Auburn Facility. The 46 section, 2760 m (1.7 mile) long test track is where rapid testing of a large number of test sections can be carried out simultaneously, with the focus on the upper 10 cm (4") of asphalt cement concrete. GlasGrid was placed full width in 30 m (100') of 60 lm (200') long section identified as W1 along with 0.03 gal/yd² (0.14 L/m2) of CQS-1h tack and paved with a 12.5 mm (0.5") SMA asphalt mix. The entire section W1 was paved as one section with the NCAT officials reporting that the installation of the GlasGrid went very smoothly, including the construction of the 5 cm (2") overlay and the achievement of the targeted 95% compaction was verified. Over the next 12 years the pavement section was subjected to 40 million ESALs (Equivalent Standard Axel Loads) in the four phases of loading. Extracted GlasGrid piece In 2006 after 20 million ESALS the subsection not having the GlasGrid, started to show some signs of distress specifically at the pavement joint located at the center line of the road, while the GlasGrid subsection continued to show no signs of distress. Further detail inspections that followed the 30 million and 40 million ESAL s loading phases has reconfirmed the distresses in the control subsection were further deteriorating, while the GlasGrid subsection is still not exhibiting any signs of distress cracks. The results clearly show that GlasGrid is reinforcing the pavement joint and preventing any cracks at the joint from reflecting to the surface even after 40 million ESALS. Exploratory samples have been removed from the GlasGrid section and the findings were that the GlasGrid is well bonded and the grid is intact

21 - 19 -

22 IV. GlasGrid GP (GlasPave TM ) Research Water Proofing Claim Whether an asphalt pavement interlayer is a paving fabric, paving mat or composite grid, water proofing of the pavement can be a desired feature and benefit of the product. Preventing water ingress into the lower layers of the pavement structure is key to long-term performance of the pavement. GlasGrid GP (GlasPave) with its design of a fiberglass laid scrim laminated between two thin layers of high performance polyester mat, results in a thinner, lower porosity interlayer compared to paving fabric or paving mat. GlasGrid GP is an engineering paving mat that delivers the highest tensile strength in the market when compared to other paving mats or fabric. GlasGrid GP only requires 0.45 L/m2 (0.10 gal/yd2) of hot asphalt cement binder (AC), as measured in the ASTM D6140 Asphalt Retention Test, for saturation and 0.68 L/m2 (0.15 gal/yd2) is recommended for application. Traditional paving fabrics or mats require at least L/m2 ( gal/yd2) of hot AC to fully saturate the interlayer. Typically, paving fabric or mat suppliers then recommend an application rate of between L/m2 ( gal/yd2), depending on the existing surface conditions and amount of AC binder likely to be absorbed by the pavement. This translates to between 25% and 40% less hot AC required for GlasGrid GP compared to paving mat or paving fabric, which means significant petroleum conservation to the contractor and asset owner and an environmentally friendly smaller carbon footprint on the job site. Though GlasGrid GP requires significantly less hot AC, it still provides equivalent waterproofing compared to a paving fabric or paving mat. Once the interlayer is fully saturated, it can become an impermeable layer and act as a waterproof membrane. GlasGrid GP was 3rd-party tested against both paving fabric and paving mat using ASTM D5084, Measurement of Hydraulic Conductivity of Saturated Porous Materials Using a Flexible Wall Permeameter. Results of the testing are shown in the table below. Interlayer Tape Asphalt Application Thickness mm (inches) Coefficient of Rate L/m2 (gal/yd2) Permeability (cm/sec) GlasGrid GP (0.15) 1.0 (0.04) 2.8 x Paving Mat 1.04 (0.23) 1.3 (0.05) 2.6 x Paving Fabric 1.44 (0.32) 2.0 (0.08) 2.6 x These results showed that all three of the AC saturated interlayers tested had extremely low water permeability rates and are essentially impermeable membranes. Based on literature cited, a membrane, tested in this manner, will greatly enhance the waterproofing of a pavement if the measured water permeability is less than approx 1 x 10-3 cm/sec. (1) What is significant in this finding is that the GlasGrid GP 25 saturated membrane exhibits the benefit of being a moisture barrier even though it only has half of the AC binder content and half of the thickness compared to paving fabric. References: (1) Baker, T.L., Results of Melt Through Asphalt Absorption and Permeability Tests, Internal Communication, Amoco Fabrics and Fibers Co.,

23 The Effect of GlasGrid GP in Reclaimed Asphalt Pavement (RAP) on Asphalt Mixture Performance A study by the National Center for Asphalt Technology (NCAT) was conducted to evaluate the recycleability of GlasGrid GP in Reclaimed Asphalt Pavement (RAP) by comparing the performance properties of two asphalt mixtures: one containing control RAP and the other containing GlasGrid GP RAP materials. IV. GlasGrid GP (GlasPave TM ) Research A two-layer test section was constructed. The asphalt material used in both layers was a 9.5 mm (.37") nominal maximum aggregate (NMAS) mix with a PG binder. GlasGrid GP fabric was installed at the interface of the two layers. The control asphalt mixture was obtained by milling the upper portion of the test section that did not include GlasGrid GP fabric, and the GlasGrid GP RAP was obtained by milling the middle portion of the test section, where the GlasGrid GP was installed. The two RAP materials were brought back to the NCAT laboratory, dried out and characterized. They were then mixed with virgin aggregate and binder to produce two 12.5 mm (0.49") NMAS Superpave mix designs. One design contained 30% of the control RAP by weight of aggregate, and the other contained 30% of the GlasGrid GP RAP by weight of aggregate. Both mixes used the same virgin aggregates a mixture of limestone, granite, and sand and a PG virgin binder. The two mix designs were then tested to evaluate mixture performance characteristics such as moisture susceptibility, stiffness, and resistance to rutting and low-temperature cracking. Control RAP mixture (left) and RAP mixture containing GlasGrid GP (right)

24 IV. GlasGrid GP Research The Effect of GlasGrid GP in RAP on Asphalt Mixture Performance Conclusion 1) Characterization of the two 12.5 mm (0.49") NMAS mix designs (with and without RAP) showed that they were very similar in terms of aggregate gradation and volumetric properties, making them ideal for comparative laboratory performance testing. 2) The extracted binder content of the GlasGrid GP RAP was slightly higher than that of the control RAP, most likely due to sampling and testing variability. There was no significant difference between the two mix designs in terms of the recovered binder content and the gradations of the recovered aggregates. 3) The TSR results for the control RAP and GlasGrid GP RAP mix designs were not significantly different. This supports the similarity of the two mixes with respect to the recovered binder and gradations. 4) Results from the Hamburg wheel-tracking device showed that the control mix design had a lower average rut depth and that the GlasGrid GP RAP design had a higher Stripping Inflection Point (SIP) value. The slightly higher rut depth for the GlasPave RAP was due to the slightly higher recovered binder content. However, the differences were not statistically significant, and the two mix designs should have similar resistances to rutting and moisture damage. 5) The dynamic modulus test showed that while the control RAP mix design was slightly stiffer than the GlasGrid GP RAP mix design, the higher stiffness of the control RAP mix was likely attributed to the slightly lower binder content. 6) The critical temperature analysis indicated that there was no significant difference in the critical low temperature at which the two mix designs would experience thermal cracking. In summary, the mix design containing the GlasGrid GP RAP had statistically the same properties compared to the control mix and therefore proves that GlasGrid GP is in fact recyclable

25 Boundary Conditions On flexible pavements, structural deficiencies must be addressed prior to designing for reflective cracking V. Installation Requirements PCC Slab Lengths must be <6 m (20 ft) without load transfer efficiency data Load Transfer Efficiency must be >70% measured by FWD (Falling Weight Deflectometer) or other device. An appropriate rehabilitated pavement design is required in heavily stressed areas, (mix, structure, thickness etc.) Crack width max. 25 mm ( 1") wide. Cracks greater than 6.35 mm (0.25") directly beneath and in contact with the GlasGrid GP, should be crack filled Loading Thermal conditions - should not exceed a 131 F (55 C) difference between average monthly maximum and minimum temperatures General Installation Requirements No installation of self-adhesive grids on milled surfaces unless a micro milled surface (6mm (0.25") grind or less) is specified. Milled surfaces are not recommended for PCC joint or thermal reflective cracking Placing A leveling course is strongly recommended for aggressive reflective cracking (minimum 20 mm (0.75") thick) The pavement surface must be clean and dry If a tack coat is specified without grid, one should also be used with the grid Minimum overlay thickness of 4 cm (1.5") Adhesion Requirements - Refer to specific product installation guides Unwinding Easy installation procedure for GlasGrid The engineered adhesive, which is activated by pressure is an essential component of GlasGrid and allows faster installation. A much as 21,000 m2 (25,000 yd2) of grid can be placed in a day using a standard placing unit. The installation of the GlasGrid System can be adaptable to overcome various weather or construction conditions. For GlasGrid CG and GP, please refer to individual product installation guides. Activating Adhesive Environmental compliance: a. Glass filaments: 1) APFE report 2) REACH compliance certificate b. Dust analysis during milling (EUROFIN) c. Reduction in CO2 and VOC emissions d. Recyclable milling-able (NCAT-GP, RWTH Aachen-GG, IFSTTAR) Overlaying

26 V. Installation Requirements Tack Coats (Bond Coats) A tack coat (bond coat) is a light coating of liquid asphalt applied either to an existing pavement surface or on top of the installed GlasGrid material. It is used to improve the bond of a new asphalt concrete course to the existing pavement surface. When the GlasGrid System was first introduced, tack coats (bond coats) were not universally used on new leveling courses. More recently however, the pavement industry has been implementing changes to asphalt mixes in order to make them leaner, stiffer and more rut-resistant. Consequently, these changes and the need to maximize the bond between lifts have resulted in most authorities mandating the use of a tack coat between all lifts of asphalt. The GlasGrid System uses a pressure sensitive adhesive to adhere to the pavement surface, unlike other paving and geocomposite grids that require a tack coat. When a tack coat has been specified, it should be used in accordance with the following guidelines*: Emulsion Tack Coats High quality polymer modified tack coat containing a minimum of 60% solids Examples include: North America: Europe: Hot spray AC NTSS-1H or equivalent low pen trackless tack coats Cationic, rapid set, CRS-2P. In general, cationic emulsions can break and set more quickly to reduce cure time, and help expedite the start of the paving operation C60BP1-S, polymer modified bitumen emulsion with 60% bitumen, and a short breaking time C69BP4-OB, special polymer modified emulsion with 69% bitumen and mid breaking time, specially for surface dressing AC20-5TR, PG 64-XX, UK's PEN In general, hot spray AC tacks work well in cooler weather, when surface temperatures are at or below 80 F (27 C). When surface temperatures exceed 80 F (27 C), the manufacturer recommends that a bituminous tack that is stable in warm conditions be applied Emulsions used with the GlasGrid System must break and then "cure" before any additional asphalt is placed. Breaking is defined as the point at which the brown colored fluid turns black. Curing occurs when the residual asphalt cement contains no solvents (water or any volatiles). Reference should be made to the GlasGrid Installation Guide for additional information. *Use of a tack coat type other than those specified above is not recommended and will likely influence the application and curing time of the tack coat. In this case, on-site supervision by the specifying engineer would be required. Fluxed bitumen or cutback asphalt must not be used as a tack coat with GlasGrid, as the solvent will attack the adhesive on the GlasGrid reducing or eliminating the bond to the surface

27 VI. Engineering Resources GlasGrid 8501/8511 Complete Road System Mill and Fill Rehabilitation GlasGrid 8501/8511 Complete Road System Asphalt Cement Concrete (ACC) Pavement

28 VI. Engineering Resources GlasGrid x 8501/8511 Complete Road System Portland Cement Concrete (PCC) Pavement GlasGrid 8502/8512 Detailed Repair System Portland Cement Concrete (PCC) Pavement Overlay to Address Joint Reflective Cracking

29 VI. Engineering Resources GlasGrid 8502/8512 Repair System Asphalt Cement Concrete (ACC) Pavement to address Thermal and Isolated Reflective Cracking GlasGrid 8502/8512 Repair System Lane Widening (Example Section)

30 VII. Reference Literature Brown, S.F., Thom, N.H. and Sanders, P.J. A Study of Grid Reinforced Asphalt To Combat Reflective Cracking, AAPT Annual Meeting, De Bondt, A. H. Anti-Reflective Cracking Design of (Reinforced) Asphalt Overlays, Ph.D thesis, Delft University of Technology, 1999 C.M. Aldea and J.R. Darling, Effect of Coating on Fiberglass Geogrid Performance, 5th International RILEM Conference, Limoges, France, May 2004 Hook, K, GlasPave TM Technical Note Water Proofing Claim, Saint-Gobain Technical Fabrics, 2010 Lytton, R. L. Reinforcing Fiberglass Grids for Asphalt Overlays, Texas A&M University, 1988 Mai Lan Nguyen, Juliette Blanc, Jean Pierre Kerzrého, and Pierre Hornych Review of Glass Fiber Grid Use for Pavement Reinforcement and APT Experiments at IFSTTAR, LUNAM Université, IFSTTAR Infrastructures and Mobility Department, 2013 Meyer,A and Schulze, C, Investigation on millability and recycling of glass fibre reinforcement asphalt layers, from RWTH University Aachen Germany, 2013 Partl,M and Raab,C, Asphalt-Reinforcement Testing with MMLS3, EMPA Materials Science & Technology (CH), Test Report No , Penman, J and Hook, KD, The use of Geogrids to Retard Reflective Cracking on Airport Runways, Taxiways and Aprons, RILEM Conference, Chicago, IL, 2008 Powell, R. Buzz, PhD, P.E., Installation and Performance of a Fiberglass Geogrid Interlayer on the NCAT Pavement Test Track, National Center for Asphalt Technology, 2008 Tebaldi, G and Romero, E, Characterization of Reinforced Asphalt Pavement Cracking Behavior Using Flexural Analysis, University of Parma, SIIV 5th International Congress Sustainability of road Infrastructures, 2012 Thom, N. H. Grid Reinforced Overlays: Predicting the Unpredictable, School of Civil Engineering, University of Nottingham, UK, 3rd INTERNATIONAL CONFERENCE MAINTENANCE AND REHABILITATION OF PAVEMENTS AND TECHNOLOGICAL CONTROL, 2003 The Effect of GlasPave TM in RAP on Asphalt Mixture Performance, Research Synopsis NCAT Report, National Center for Asphalt Technology at Auburn University (US),

31 - 29 -

32 Learn more about how GlasGrid Pavement Reinforcement System products can increase the life of your paving projects. SAINT-GOBAIN ADFORS America 1795 Baseline Road Grand Island, NY USA Tel: (+1) Fax: (+1) GlasGrid is manufactured at an ISO 14001:2004 registered facility of Saint-Gobain ADFORS. GlasGrid is a registered trademark and GlasPave is a trademark of SAINT-GOBAIN ADFORS. U.S. Patent 8,038,364 and 8,349,431. Additional patents pending SAINT-GOBAIN ADFORS /14