|
|
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
|
Glynn Holleran, Vice President, VSS Asphalt Technologies, USA Jack Van Kirk, Director of Asphalt Technology, Basic Resources Incorporated, USA Back to VSS Technology Papers Library 1. SUMMARY The purpose of a road surfacing is Roads are in essence mechanisms by which stress applied from moving wheel loads are transferred to the earth. To operate effectively they must not crack, rut or wash away. The distress of road surfaces generally take one of these forms. However in instances where the road surface or structure has failed then the rehabilitation method must address and correct the failure mode. Techniques to achieve these ends are many. They range from hot mix overlays, polymer modified membranes as stress absorbing seals or interlayers, recycling of distressed pavement layers with emulsions or foamed asphalt, chip seals and slurry seal and microsurfacing. (1) Asphalt (bitumen) is a naturally occurring material and has limitations based on its chemistry, the method that was used for its refining and the materials with which it is blended, for example cutter or other diluents. (2,3). Addition of polymeric materials has been widely used to extend the application of asphalt and improve its properties, particularly in the area of thermal susceptibility and flexibility. (4,5,6). A significant problem in western countries has been the development of a vast used tyre stockpile. Tyre rubber is a mixture of a number of compounds including SBR of medium to high molecular weight, natural rubber, carbon black, processing oils and fillers.(7) Several methods have been used to dispose of these tyres, amongst them have been burning, addition to asphalt for road building (8) and pyrolysis to carbon black (9). As the main constituent of tyre rubber is SBR then the addition of this material to asphalt to extend its performance would seem logical, economic and environmentally advantageous. This paper examines the technical and performance aspects of asphalt rubber and compares it to the use of polymers. It examines some of the performance information available that indicates the use of such materials. 2. COMPATIBILITY - THE CREATION OF A SUITABLE ASPHALT RUBBER BINDER 2.1 Compatibility and its effects on Morphology: The compatibility of asphalt / polymer systems may be defined in several ways. (4,5). It may be in terms of the achievement of a particular morphology, i.e. the structural arrangement of the polymer particles, chains or groups within the asphalt matrix. It may be in terms of thermodynamic stability, i.e. whether the conformation of the polymer particles or chains are in a low energy state, i.e. whether there is a driving force to increase entropy. It may be in terms of a practical storage stability, i.e. will it separate on standing. Or it may be based on whether a given property or set of properties are achieved and can be maintained for a suitable period of time (that is until the material has been applied). In asphalt rubber compatibility is an issue that has received little attention (with a few exceptions (12,13). Asphalt Rubber blends of 18-20% can increase viscosity by an order of magnitude overnight. It has long been acknowledged that extender oils do improve general storage stability and stabilize the product, even if only briefly (12-24hrs). A reaction is claimed to occur (10) in which the asphalt and the rubber particle interact to form a gel coated particle. This is analogous to the process of swelling that occurs in polymer asphalt systems. (5).
How well this model reflects the actual situation and the relative effect of particle sizing is not sure but, based on polymer and asphalt chemistry it seems adequate. It also explains why there is a significant change in properties over time as such a system is not thermodynamically stable. That is the large increase in viscosity over its early life is due to the continuation of this solvation process.(14). This can be shown in examination of micrographs of asphalt rubber digested with and without extender oil and in low shear mixing compared to material put through a colloid mill. As shown in figure 2 and figure 3. (relative sizing is important ;all are to same scale).
2.2 Morphology and Physical behaviour This has clear implications for the application and properties.
However it is claimed that materials that have been digested for longer periods or those which have finer morphologies exhibit poorer fatigue properties (15). This is explained by a mechanism similar to that which is employed in materials such as high impact polystyrene to make them more impact and crack resistant. In such materials a rubber (in this case butyl rubber or butadiene rubber) is dispersed in relatively large (above micron) sizing particles. The fatigue fracture begins as a crazing in the dispersing (asphalt) phase and propagates by the increase of strain at the root of the craze ; this leads to cracking and the crack similarly propagates. If this crack encounters a large mass of flexible material that can dissipate the strain, then the crack will be blunted. It cannot simply go around the mass because of the magnitude of its size. In polymer systems or finely dispersed systems the crack blunting must be achieved by the properties of the binder itself. Thus the polymer asphalt binder must be isotropic on a micron or even molecular scale. This is possible in polymer systems (4,6) but may not be in asphalt rubber systems due to the presence of fillers and the crosslinked nature of the rubber. See figure 4.
This may be a consideration for applications of asphalt rubber where crack resistance is a key reason for using such materials such as in fatigue, reflection cracking or low temperature cracking. Low temperature cracking may be expected to be still improved as each morphology will improve tensile strength to a high degree. Figure 4 is for hot mix but it might be expected that the binder effect will be similar in other straining situations. In seals the magnitude of the differences will be changed as film thickness is an order of magnitude greater and stiffness plays a more important role. 3. EMULSIFICATION Polymer emulsions have been used to a great extent in chip sealing and slurry/microsurfacing (16,17). The polymers generally are pre-blended with the asphalt, co-milled as latex either by direct injection into the soap or pre mixed with the soap, or post added. Ground tyre rubber usually can not be added so simply. Pre-blended crumb, must be very finely dispersed, free of any metal (to avoid damage to the emulsion mill) to be able to be emulsified. Other methods by which tyre rubber may be incorporated in emulsion are post addition approaches. Ground solid crumb rubber has been added into slurry mixes as a dry ingredient, similar to the method mentioned above. In such a case,the rubber becomes a part of the aggregate phase and acts mainly as a filler. Such processes are in general use in USA. However, to create much of a change in elasticity, increase in cohesion, and other desirable properties,the rubber needs to be fully or partially digested so that it may coat particles. This is the basis of the process to be discussed. The increase in cohesion should improve properties such as deformation resistance (in rut filling),surface abrasion resistance, crack resistance and allow increased binder films without flushing. 3.1 Solvent Dispersion The process referred to as the wet process may be adapted for use with emulsions. If a suitable solvent can be used that allows dispersion in the emulsion water phase, then a material capable of being post added may be produced. The solvent type is of obvious importance as it must not create an environmental hazard, nor degrade either the asphalt properties nor those of the rubber. On the other hand, if the solvent is able to swell or soften the rubber, then it may improve wetting and adhesion. A range of solvents has been used to optimize the dispersion and other additives such as wetting agents and carbon black included. In general terms, an oil additive is preferable with a high aliphatic content and a boiling range that meets emission requirements but also allows swelling of the rubber. The material will be referred to by its designation of RG-1. 3.1.1 Emulsion Properties and RG-1 RG-1 is a semi swelled dispersion of crumb rubber (40-50%) in a petroleum solvent. It is supplied as a free running high viscosity material that can be readily poured and pumped. RG-1 is used by post addition into the emulsion with simple mixing. (Figure 5 shows some properties). As may be seen the emulsion is not greatly affected by addition of RG-1 except that the sedimentation rate is high. This is not surprising as the RG-1 is a separate phase and so the emulsion must be thoroughly mixed before use. There is no obvious breaking caused by the presence of the RG-1 and this is true to concentrations in excess of 20%.
3.1.2 Direct Emulsification This work is relatively new and involves preblending the asphalt rubber via a colloid mill system to create a very fine distribution as in figure 3. The material is not very stable and is a very coarse emulsion. The particles as shown in figure 6 are up to 20 microns in size. This explains the high settlement. However break and adhesion are very good. More work is needed.
4. APPLICATION AND PERFORMANCE Application methods for these materials are standard. However, for hot sealing, the temperature needs to be increased to approximately 180-190C. 4.1 Hot Applications and Summary of States use Crumb rubber has been used by many states over the years with varying degrees of success. The most successful and extensive use in the USA has been in Arizona, Florida and California. 4.1.1 Crumb Rubber Use in Arizona Arizona's early applications of crumb rubber were in chip seals as stress absorbing membranes (SAM's) and stress absorbing membrane interlayers (SAMI's) and in sealants. This experience with SAM's and SAMI's was extensive in the early years and these applications proved to be extremely cost-effective. The Arizona Department of Transportation (ADOT) and the City of Phoenix began using crumb rubber in hot mix in the 1970's, but more extensive use began in 1985 using open and gap graded mixes. Currently Arizona uses only the wet process. Their experience with crumb rubber in the dry process has been unsuccessful. The open graded mixes have been placed in thicknesses of 1.5 inches (40mm) or less over both flexible and rigid pavements. The gap graded mixes have been placed in thicknesses of 2.5 inches (60mm) or less. Most of the projects utilizing crumb rubber are with open graded mixes. The performance of open and gap graded mixes in Arizona has been very successful and continues to be extremely cost-effective. 4.1.2 Crumb Rubber Use in Florida The Florida Department of Transportation (FDOT) began their use of crumb rubber in hot mix in 1988-1989. Their decision to begin use of crumb rubber was prompted by State legislative interest. However, FDOT took a very structured approach. Their main use is 5 % crumb rubber by total weight of binder in dense graded friction courses. These are placed in a 1 inch (25mm) thickness to improve the resistance to rutting, particularly at intersections. On their freeways FDOT uses a ½ inch (15mm) thin layer of open graded friction coarse containing 12 % crumb rubber (by weight of total binder). The crumb rubber is used to improve the durability of the hot mix. FDOT also has developed a wet process which uses about 5-10 % ultra fine crumb rubber in the hot mix (by weight of total binder). The crumb rubber is introduced into the asphalt just prior to the binder being introduced to the hot plant. There is no lengthy reaction time. FDOT does not use the dry process. 4.1.3 Crumb Rubber Use in California California's experience with crumb rubber began in the 1970's in chip seals or SAM's. The California Department of Transportation (Caltrans) began experimenting with crumb rubber in hot mix applications in 1978 and since then Caltrans has been a leader with hot mix usage and has had considerable experience with dense and gap graded mixes. Caltrans experience with the dry process has not been successful. Some field trials using the dry process continue, but the wet process has proved to be the most cost effective and it is predominant in California. Because of the wide range of experience with crumb rubber in hot mix and seals the California experience will be looked at in more detail in this paper. a) Laboratory and Field Testing Laboratory research by Caltrans (18) indicated that crumb rubber modified asphalt concrete (RAC) mixes were more abrasion resistant when compared to conventional dense graded asphalt concrete (DGAC). Field permeability testing also showed that dense graded RAC mixes had extremely low permeability. It was felt that these low permeability would reduce the infiltration of water into the mat and, therefore, cut down on the freeze-thaw damage. The low permeability should also reduce oxidation and thereby lower the aging rate. b) Caltrans RAC Usage On the RAC projects constructed by Caltrans prior to 1983, the RAC was compared to equal thicknesses of conventional DGAC. However, in 1983 a project was constructed (on RT. 395 in northeastern California) using various overlay strategies including three test sections of reduced thickness RAC (when compared to the conventional DGAC overlay design thickness) (19). Also placed on the project were various thicknesses of conventional DGAC. This project, though not realized at the time, later became the turning point for Caltrans rehabilitation strategies involving RAC mixes. For awhile after 1983, Caltrans continued to construct and compare equal thicknesses of RAC and conventional DGAC on other projects, while reviewing and accumulating data on the RT 395 project. By 1987, it became evident that substantially thinner overlays of RAC, when compared to conventional DGAC, could provide a longer service life at a reduced cost. At this point in time, Caltrans strategy for RAC test section overlays changed. It was decided that all subsequent projects if appropriate would involve RAC overlays that were thinner than those required if conventional DGAC were used. Projects utilizing reduced thicknesses continued until 1992. At that time the use of reduced thickness RAC became a routine strategy in the Caltrans rehabilitation program. Caltrans rehabilitation projects using RAC overlays have included rubber modified: dense graded asphalt concrete (DGAC) placed in thicknesses ranging from 1 ¼ inches to 3 inches (30mm-75mm), open graded asphalt concrete (OGAC) placed in thicknesses of 1 inch or 1 ¼ inch (25mm or 30mm), gap graded asphalt concrete (GGAC) placed in thicknesses of 1 ¼ inch to 2 ½ inches (30mm to 60mm), and an Arizona-type, three-layer system, which is a 1 ¼ inch (30mm) leveling coarse of DGAC, a SAMI and finally a 1/1/4 inch (30mm) layer of RAC. These overlays have been placed over flexible pavement (AC) as well as rigid pavement (PCC). On these projects, six different types of RAC have been used: devulcanized reclaimed rubber has been added using a dry process to conventional DGAC mix; vulcanized reclaimed tyre rubber has been added using a dry process to a gap-graded aggregate and conventional asphalt; and vulcanized reclaimed tyre rubber has been preblended with a conventional asphalt to form asphalt-rubber binder (wet process) which was then added to a dense graded, open graded or gap graded aggregate; or the binder was used with a pre-coated chip as a stress absorbing membrane interlayer (SAMI) in an Arizona-type, three-layer system. SAMI's also have been used with other RAC mixes. Currently, Caltrans uses predominantly gap graded RAC mixes in their rehabilitation program. Caltrans uses crumb rubber modified hot applied SAM's in their maintenance program. They have proven to be very cost-effective. The asphalt rubber binder is sprayed at a temperature of 385 F to 415 F (196 C to 212 C) and then hot precoated chips are spread at a temperature of 260 F to 325 F (127 C to 163 C). The hot precoated chips have significantly improved the rock retention on their chip seals. The asphalt rubber SAM's have shown a superior resistance to reflective cracking. These types of SAM's are placed on the more heavily cracked pavements. c) Caltrans Overlay Design Caltrans uses a deflection-based design procedure for rehabilitation of flexible pavements. This procedure is also used for RAC overlays. On the early RAC projects, the RAC overlay design thickness was the same as that provided from the deflection study for conventional DGAC. The design life normally used is ten years, and during this time only minor maintenance is expected. In 1992, Caltrans presented a proposal to the Federal Highway Administration (FHWA) to allow the use of reduced thickness RAC overlays as an approved strategy on federally funded rehabilitation projects. This proposal was approved based primarily on the successful field experience of reduced thickness RAC projects. At that time an Interim Thickness Determination Guide for asphalt-rubber hot mix-gap graded was developed and also approved by the FHWA. This interim guide has not been changed and is currently being used on a wide spread basis. The reduced thickness theory has been validated through a Caltrans research effort in South Africa using the Heavy Vehicle Simulator (HVS) (20) and by research at the University of Alaska, Fairbanks (21). This research has concluded that the reduced thickness guidelines are conservative. A different approach is used for portland cement concrete (PCC) pavement rehabilitation involving DGAC overlays. In order to try and obtain the desired ten year design life, the early approach (prior to 1982) was to place a DGAC overlay 3 inches to 6 inches (75 mm to 150 mm) in thickness depending on the condition of the existing PCC pavement. However, the 10-year design life was not being achieved. After 1983, this strategy changed. PCC pavements are now cracked and seated, a leveling course of DGAC is placed, a pavement reinforcing fabric (PRF) is placed, and finally the pavement is overlaid with DGAC 3 inches (75 mm) thick placed in two lifts. This same approach is used for RAC overlays over PCC pavements. However, the PRF is replaced with a SAMI and the 3 inch (75 mm) thick DGAC is replaced with a 1 ¼ inch (45 mm) thick RAC overlay. This is referred to as an Arizona-type, three layer system. The RAC is usually a gap-graded or open graded mix. d) Mix Design Mix design methods used for crumb rubber modified mixes in the USA are basically modified Marshall and Hveem mix designs. The basis for these designs comes from field experience. Most open graded binder contents are determined using a formula. This usually involves increasing the unmodified binder content by some predetermined factor. Different approaches are used around the USA, but, most dense and gap graded binder contents are based on void content in the laboratory. A major difference in the laboratory design is that the mix and compaction temperature is increased. The required aggregate temperature is between 300 F and 325 F (149 C and 163 C). The required compaction temperature for the combined mix is between 290 F and 300 F (144 C and 149 C). In California the Hveem stability requirement is lowered to a minimum value of 23. Also, in California the binder content for open and dense graded RAC mixes is normally about 20% higher than that for the conventional DGAC mixes and the binder content for gap-graded mixes is normally about 40% higher than that for the conventional DGAC mixes. In Arizona the binder contents for open graded mixes is about 50% to 60% higher than that for conventional open graded mixes and the same as California for gap graded mixes. Florida's values are about 12% higher for open graded mixes and about 5% higher for dense graded mixes. Florida requires a Marshall stability and flow for their dense graded mixes, however Arizona does not require either. With the advent of performance based (SHRP "Superpave") and performance related (Australia / New Zealand) specifications and test procedures, asphalt rubber mixes can be shown,unequivocally, to produce superior results to conventional mixes. Modifications to laboratory compaction (gyratory) are required; higher compaction temperatures and more revolutions are needed (mix specific). This insures that the laboratory void contents are similar to the in-place voids in the field. However, more work is required in this area. Laboratory testing of asphalt rubber binder properties is a matter of controversy in the USA and more work is required to develop a rheological specification that truly reflects field performance. 4.1.4 Cost analysis The cost of asphalt-rubber binder is little over twice the cost of conventional asphalt. This increases the cost per ton of crumb rubber modified hot mix by about 30%-40% over that of conventional DGAC. The cost increase for open graded mixes ranges from about 30% to 100% depending on the binder content. However, if the design thickness of the mix is reduced compared to the conventional design thickness, as is done in California, the overall project cost can be less for the crumb rubber modified project. 4.1.5 Construction The construction of the crumb rubber modified overlays has been very similar to that of conventional hot mix overlays, although there are a few important differences. First, the mix must be placed at a higher temperature, preferably in a range between 300 F to 325 F (149 C and 163 C). Raking should be completed before the mix drops below 290 F (144 C). As the mix cools, it becomes stiffer and raking becomes very difficult. Breakdown rolling should be completed before the mix drops below 290 F(144 C). If the mix is compacted at 290 F (144 C) or above, excellent relative compaction has usually been achieved quite easily. Values of 96% to 98% relative compaction (compared to a laboratory compacted briquette) are quite common for dense graded crumb rubber modified hot mix. However, compaction of gap-graded crumb rubber modified hot mix has been more difficult because the mix cools faster than conventional DGAC. If proper temperatures are used, adequate compaction can be achieved. 4.1.6 Crumb Rubber Hot Applied Performance Crumb rubber hot mix has been used in many parts of the USA and in different climate regions. Many of the early projects were placed to resolve specific problems such as abrasion resistance, OGAC night placement, thin flexible bridge overlays, and desert hot mix pavement rehabilitation. Generally, control sections containing conventional DGAC were placed on the early projects so that direct comparisons could be made. Crumb rubber modified hot mix has been used on bridge decks, roadside rests, parking lots, and low, medium and high volume roadways. Over the years crumb rubber modified hot mix has proved to provide cost effective performance in all climatic regions and in many parts of the USA. 4.1.7 Summary: The USA has had considerable experience with crumb rubber modified hot mix. From this experience, it has been learned that when compared to conventional DGAC, these mixes can tolerate higher deflections. These mixes have shown that they can outperform substantially greater thicknesses of conventional DGAC. They can exhibit lower permeability which in turn decreases oxidation and aging. They have also proven to be more abrasion resistant in the snow regions and when distress does develop it progresses at a much slower rate. All these desirable qualities that crumb rubber modified hot mix possess lead to decreased maintenance costs and ultimately lower annual equivalent costs. At this time, it is believed that rubber modified asphalt hot mix will probably play a major role in the rehabilitation of rigid and flexible pavements in the future. The role that these mixes will play depends on their performance on recently completed projects and projects to be constructed in the next few years. If reduced thickness crumb rubber modified hot mix continues to show the success that has been demonstrated on earlier projects there will be a definite increase in the usage of the product. 4.2 Slurry and Microsurfacing 4.2.1 Laboratory Method International Slurry Surfacing Association (ISSA) guidelines were followed exactly with the RG-1 mixed in the emulsion with hand stirring. Mixing times were different for the RG-1 modified systems, and retarder levels needed to be increased significantly compared to asphalt emulsion from 0.25% to 1.0%. Water levels too were increased by around 25-30%. For flex testing,samples were prepared according to ISSA 146. However, as the ISSA device was not available, a method was devised where the strips of metal with the slurry covering were bent in a continuous rate of about 5cm per minute and the on set of visible cracks observed. The lateral distance that the strip was bent from the horizontal was measured and the difference noted. This is obviously very operator dependent but appeared to be as reproducible as the ISSA 146 test. The results may only be taken as an indication however and more work is required. All mixes met both the requirements for microsurfacing and slurry seal. 4.2.2 Results a) Wet Track testing. (figure 7) Wet Track showed that the rubber additive gave improved resistance to stone loss under this test compared to asphalt. This may give the opportunity to reduce binder levels. In combination with latex the results were further improved. This showed less improvement than with latex alone. This is probably due to the presence of particles of rubber in the surface that are taken out by the abrasion head. This is not an unexpected result. It should not be interpreted in terms of insufficient binder however. The general increase in resistance to abrasion does indicate that binder levels may be increased for these systems by at least 1%. This will have the effect of increasing durability,as film thickness is the determining factor for this property (Holleran 1996).
b) Loaded Wheel Test (Figure 8) Mixes with RG-1 showed improvement in resistance to deformation relative to asphalt. The improvements were significant but not as large as those for polymer modification. When combined with latex, the results for the RG-1 mixes were improved significantly. These results show the effect of increased viscosity and modulus. Shear resistance cannot be gauged by this test but the results are consistent with improvements in rut resistance and indicate an application for this type of material.
The setting of the slurry /microsurfacing was not compromised by the addition of the rubber, in fact it was improved marginally. The polymer materials were better still.
Figure 10 illustrates this for Vialit plate test with these binders.
Standard boiling tests for adhesion showed some improvement and Schulze Breuer testing showed higher compatibility, so this effect is significant to the mix performance. (All rubber modified mixes had AAA ratings and qualified as microsurfacing mixes). d) Flexural Strength- Crack Resistance Figure 11 shows the test results on these mixes in flexure. Within the limitations previously indicated, the rubber modified material does appear to have better crack resistance in flexure. It should, therefore, improve resistance to traffic deformation in rut filling and allow an improvement in performance of the slurry coat of a cape seal.
Field trials have been carried out using RG-1 and the long term effects are being quantified. The main trial carried out was on a residential street with moderate surface raveling, shrinkage cracking, and alligator cracking. This was divided into three sections, alligator cracking was worst in section III. Cracks were sealed with cold crack pour sealant in section I and this section was covered with a type II polymer modified (latex at 3.5%) slurry. Section II was overlaid with the same type II slurry as above. Section III was done with a 5% residual addition of RG-1. All applications were at about 20lb/yd2. After 24 months section I had intermittent hairline reflection cracks. This was less than for comparable streets done with unmodified slurry seal in the same trial. Moderate raveling occurred. Section II looked similar to section I. Section III has been performing extremely well, even in the worst alligator cracked areas,little, if any cracking has been reflected through. Stone retention was excellent with little or no raveling. 4.2.4 Summary The general conclusions on slurry and microsurfacing application to date are: 1. RG-1 was easy to blend into the emulsion using an in line blending system and ensuring that the emulsion was well mixed before use. 2. The emulsion mixed easily with the mixes and,although higher levels of retarder and water were required, laid as normal. 3. The slurry set and cured normally. 4. Long term effects of traffic and aging cannot be concluded from results to date. 5. The use of the crumb rubber product appeared to retard reflective cracking. 6. Stone retention appears to have been improved 4.3 Use In Chip Sealing applications 4.3.1 Emulsion Properties/ Stone Retention In Figures 5 and 12 some properties of CRS-2 type emulsion are shown. It is clear that the rubber has some effect on the residual properties but the recovery method is likely to play a part in this. The effect of the binder on chip performance was examined using a modified Vialit plate test. Some of these tests were carried out by US Oil and Refining and some by VSS.
Extra testing was carried out to compare the effect of other polymers on stone retention using the same method. Figures 13 and 14 show the results. It is evident that the RG-1 has a significant effect on stone retention and that this is comparable and even superior to emulsion modified with latex. In combination with latex,the RG-1 performs even better. Also the emulsified rubber performs well.
This may be caused by an increase in cohesion produced by the rubber but may also be associated with improved wetting due to residual solvent. This may also be the case with the CRS-AR2 as extender oil is a part of the formulation. 4.3.2 Seal Design and application The rubber loading rates are significantly lower in the final binder than is usual for the hot applied systems for RG-1. No trials have been conducted with CRS-AR2 as yet. Design was carried out, as per Austroads application rates, for both aggregate and binder. No allowance was made for the fact of rubber being present. The seals made across a range of jobs in residential streets and secondary roads over the last 5 years have shown good results. There was some indication that reflective cracking was retarded in these trials. 5. CONCLUSIONS
6.0 BIBLIOGRAPHY
Back to VSS Technology Papers Library Copyright © 2000 Valley Slurry Seal Co All Rights Reserved. |
