Best Practices for Crack Treatments for Asphalt Pavements (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4C H A P T E R 2 Summary of the Literature Review Scope of Work A literature review was conducted on the state-of-the-art for crack sealing and crack filling. Approximately 115 techni- cal publications, state specifications, and test methods were reviewed. For ease of reference, the state-of-the-art summary and the state-of-the-practice survey are organized into the same categories. Those categories are: • General Issues/Project Selection; • Contracting Procedures; • Materials; • Construction; • Quality Control; and • Performance. General Issues/Project Selection Crack sealing and crack filling have been used as a mainte- nance procedure for asphalt pavements for many years. The technical literature is in general agreement with the following definitions. Definitions Crack sealing: Materials are placed into and/or above “working” cracks in order to prevent the intrusion of water and incom- pressibles into the cracks (“working” cracks refer to cracks that undergo significant amounts of movement). Crack sealing is commonly used as a transverse crack treatment (70). Crack filling: Materials are placed into “non-working” cracks to substantially reduce water infiltration and reinforce adjacent cracks. Crack filling is commonly used as a longitudinal crack treatment (70). Crack routing: Routing is used to open up the crack to accommo- date enough sealant to provide an effective seal, even after the pavement crack opens due to contraction at low temperature during the winter months (35). Adhesion: The binding force exerted by molecules of unlike sub- stances when brought into contact (76). Cohesion: That force by which molecules of the same kind or of the same body are held together so that the substance or body resists separation (76). Working crack: Identifying whether the crack is “working” (i.e., moving as a result of contraction and expansion) or not is a challenge. In the 1999 LTPP report, FHWA defined the amount of movement for “working” classification as 2.5 mm; however, currently the value most commonly referenced is 3 mm or approximately 1⁄8″ (23, 33, 35, 37, 70). Masson et al. (24) present Graphic 2-1 to illustrate poten- tial cracking conditions. Cracking illustrated in the top two line sketches is appropriate for crack treatment. The bot- tom sketch illustrates a branched crack condition that is not appropriate for crack treatment. Photographs 2-1 and 2-2 illustrate pavements where cracking is excessive and where crack treatments were inappropriately applied. CalTrans (22) uses the criteria shown in Table 2-1 for crack sealing/filling. These criteria fit within the parameters previ- ously described. Season for Sealing Masson et al. (24) demonstrate the effect of the time of year on sealing with Graphic 2-2. Graphic 2-2 can be interpreted as follows: • When sealing in the winter, the crack will be at its maxi- mum width, as shown in the first row of the graphic. In the other seasons, the crack reduces in size and squeezes the sealer out of the reservoir. State-of-the-Art in Crack Treatments

5 • The center image of the middle row illustrates sealing in the spring/autumn. The crack is at a “middle” size and will have less deformation of the sealant during cold and hot temperatures. • The bottom images demonstrate that if the crack is filled in the summer when the crack is at its smallest size, extreme stresses will be induced on the sealant during the winter, potentially leading to cohesive failure. Crack Development Cracks initiate in asphalt pavements for multiple reasons, the discussion of which is beyond the scope of this report. After development of the crack, expansion and contraction of the pavement during hot and cold weather, respectively, causes movement in the crack. In cold weather, the crack widens as the pavement contracts. This widening allows debris to enter the crack. In hot weather, the pavement expands, thereby clos- ing the crack. However, the debris collected in cold weather restricts closure of the crack in hot weather, resulting in dete- rioration of the cracked pavement. Cycles such as this cause continued deterioration of the pavement. Masson and Lacasse (31) provide a discussion of adhe- sive and cohesive failures. A cohesive failure occurs in seal- ant that is still adhered to the crack walls. Adhesive failures occur due to debonding at or near the sealant/asphalt mix- ture interface. Their discussion includes precautionary com- ments about the compatibility of sealants and aggregates at a specific location. Cracking Theory • Cracks Happen • Cracks Move • Cracks Grow • Cracks Get Worse • Cracks Accelerate Pavement Deterioration —Jim Chehovits, 2012 National Pavement Preservation Conference (79) Crack Types The Long-Term Pavement Performance Program (LTPP) Distress Identification Manual (80) identifies six primary types of cracking for asphalt pavements, namely: • Fatigue Cracking • Block Cracking • Edge Cracking Graphic 2-1. Cracking graphic (24). Photograph 2-1. Wrong application (24). Photograph 2-2. Excessive crack filling (photo by Dale Decker).

6• Longitudinal Cracking (both in the wheelpath and between wheelpaths) • Reflection Cracking at Joints • Transverse Cracking While treating any crack may ultimately provide some benefit to the underlying pavement structure through the reduction of moisture intrusion, the most advantageous applications for crack sealing and/or crack filling are block, longitudinal, reflection, and transverse. Unless the crack treatment is done in early-stage distress development, crack treatments for fatigue cracking do not substantially improve pavement performance; however, the treatment may reduce further deterioration of the pavement. Fatigue cracking is indicative of a structural failure in the pavement system and can only be remedied by removing and replacing the failed materials. Many references reviewed recommend not performing crack treatments on fatigue cracks (AKA, alligator or chicken wire cracks) or edge and slippage cracking. Examples include References 3, 4, 5, 17, 22, 23, 24, 25, 33, 35, 43, 49, 72, 78, and 82. Crack Shape Factor In the late 1950s and early 1960s, Tons (57) and Schutz (58) established that the shape of the crack sealing material was sig- nificant. Both concluded that the crack sealing material does not change volume, just shape (cross-sectional area), during expansion and contraction. Tons showed that more shallow seals developed lower strains in the sealer. Both demonstrated that the depth-to-width proportion (so-called shape factor) had a critical effect on the capacity of the sealer to withstand extension and compression. Subsequent work by Khuri and Tons (64) and Wang and Weisgerber (38) determined that a rectangular shape of the sealer was preferred. Khuri and Tons recommended wide and shallow seals with a width-to-depth ratio > 1.5 to minimize strains in the sealer. Schutz recom- mended a width-to-depth ratio of 2 based on evaluating the strain on the sealant. Subsequent work by Chong and Phang (35) in 1988 con- cluded that a 4 to 1 width-to-depth ratio performed well, particularly in cold regions, for the following reasons: 1. The strain developed in the sealant was decreased. 2. Cohesive failure in the sealant was decreased. 3. A 4:1 ratio provided greater bonding area horizontally in the crack compared to the vertical faces for square configurations. 4. Lower adhesive stress was developed on the sealant. 5. It was easier for the router operator to follow meandering cracks. 6. There was less stress on the router machine and router bits, resulting in higher productivity at lower cost. Chong (37) further recommended that a 12 mm × 12 mm (½″ × ½″) rout configuration provides good performance in warmer climates and particularly for urban expressways. Schutz noted that a bond breaker was necessary at the bot- tom of the crack to allow the sealer to expand and contract properly. Wang and Weisgerber further commented that bonding to the bottom of the reservoir does not have a signifi- cant effect on adhesion to the vertical walls but may lead to Crack Sealing Crack Filling Applicable Width 0.12"-1.00" 0.12"-1.00" Edge Deterioration <25% <50% Annual Horizontal Movement >0.12" Working <0.12" Non-Working Appropriate Type of Crack Transverse Thermal Transverse Reflective Longitudinal Reflective Longitudinal Cold Joint Longitudinal Reflective Longitudinal Cold Joint Longitudinal Edge Block, distantly spaced Table 2-1. CalTrans cracking criteria (22). Graphic 2-2. Seasonal impact on sealing operations (24).

7 cohesive failure in the sealant. Use of backer rod as a bond breaker is illustrated in Graphic 2-3. Seal Geometry Numerous crack seal configurations have been used. The following are the most common: • Recessed Crack Seal Configuration • Flush Fill Crack Seal Configuration with Routed Crack • Flush Fill Crack Seal Configuration with Non-Routed Crack • Overband Crack Seal Configuration with Routed Crack • Overband Crack Seal Configuration with Non-Routed Crack Graphics 2-4 to 2-6 illustrate the configurations. The recessed crack configuration in Graphic 2-4 is com- monly used when an overlay is to be placed. Flush fill, as shown in Graphic 2-5, is used in many applications and can be used prior to placement of a surface treatment. The overband is used in many applications but is commonly limited to low- speed roads. The reservoir applications where routing is performed have the advantages of being more aesthetically acceptable, not being exposed to traffic, better adhesion to the vertical faces of the crack, and reduced tensile strains in the sealant. The only disadvantage of the reservoir configuration is the additional work and cost to the project for the routing activity. Johnson et al. (30) report that routing transverse cracks improved seal- ant performance, but that routing of longitudinal cracks was not necessary. Graphic 2-3. Backer rod (shown as an ellipse) as a bond breaker (21). Graphic 2-4. Recessed crack seal configuration. Recess Depth Reservoir Width Reservoir Depth Graphic 2-5. Flush fill crack seal configuration, both routed (left) and non-routed (right). Reservoir Width Reservoir Depth Routed Crack Non-Routed Crack Graphic 2-6. Overband crack seal configuration, both routed (left) and non-routed (right). Overband Width Reservoir Depth Routed Non-Routed Overband Width Reservoir Width

8Chehovits and Manning (36) describe the advantages and disadvantages of overband (also known as band-aid) versus reservoir configurations. The main advantage of the overband is the ease and speed of application. Basically, the procedure is to apply sealant into the crack and level with a squeegee. How- ever, the disadvantages are aesthetics, exposure of the surface sealant to environmental and traffic deterioration (including snowplows), and the large and localized tensile strains that develop above the crack. Eaton and Ashcraft (23) caution that overband should not be used on city streets, parking lots, or sidewalks due to the potential for tracking. Based on a pooled fund study, Al-Qadi et al. (84) recommend the use of over- band for crack filling and crack sealing. CalTrans (22) advises against using an overband, preferring a squeegeed approach for any material left above the surface. The concerns expressed are that ride quality will suffer, with potential bumps and fat spots forming during subsequent overlays. Overbanding can be used on low-speed roads that are not slated for overlay within six months. Filice (72) recommends a 40 mm × 10 mm (1-½″ × 3⁄8″) rout for transverse cracks, a 40 mm × 15 mm (1-½″ × 5⁄8″) rout for transverse cracks where the pavement has a chip seal, and a 19 mm × 19 mm (¾″ × ¾″) rout for longitudinal routing. {Note: Throughout this report, conversions from metric to English units are rounded to the nearest practical unit.} Ponniah and Kennepohl (33) recommend that rout and crack sealing not be used if: • Crack openings are less than 3 mm (1⁄8″); • Cracks are fatigue type; • Crack density is high (80–100% of the pavement, or trans- verse cracks less than 10 m [30′] apart); • Pavement condition is poor; or • Overall pavement thickness is less than 50 mm (2″). Chong and Phang recommend that rout and seal treat- ment be accomplished within the first five years of service life of the pavement. Contracting Procedures Two significant schools of thought exist for the installa- tion of crack treatments. The first is that the agency will self- perform the crack treatment installation and the second is that the agency will contract for the crack treatment services. The decision is usually based on perceived cost-effectiveness. If done in-house, oversight of the process is often not well- defined. Employees are directed to do crack sealing, the dir- ective is followed, and little is done to verify installation qual- ity. If contractor services are employed, owners use a variety of techniques for the purchase of crack sealing services. These techniques include: • Unit Price—Low Bid • Lump Sum/Firm Fixed Price • Cost Plus • Indefinite Delivery/Indefinite Quantity • Warranty As one might expect, there are advantages and disadvan- tages to each of the contracting approaches. The decision on how to purchase crack treatment is both an economic and political choice. This report makes no attempt to address the procedure for that decision-making process. It is noteworthy that Michigan DOT has successfully made use of crack seal project warranties (SS-14). The war- ranty period chosen was two years. The warranty approach relieves the owner of future issues on sealant performance. The approach also heightens the attention-to-detail of the contractor to ensure the sealing is done properly. Materials The Nebraska Department of Roads Pavement Maintenance Manual (82): “A value engineering study concluded 66% of total cost of crack sealing operations was for labor, 22% for equipment, and 12% for materials. Because crack sealing takes a lot of time, workers are exposed to traffic and motorists encounter delays. Therefore, it is safer and usually more cost-effective to use a product that will last longer, even if it is more expensive.” The materials used for crack treatments have varied widely over the years, ranging from neat liquid asphalt to asphalt emul- sions to polymer and/or filler modified materials. This report does not address specific products by name but addresses material types and required properties. The products most commonly used currently can be broadly characterized as modified asphalt products. A wide variety of modification schemes are used to satisfy the specification requirements. Discussion of the specific types of modifiers used is beyond the scope of this report. American Society for Testing and Materials (ASTM) D6690 (TM-11) has been the reference standard for sealants for many years. Sealant manufacturers produce a variety of products that satisfy the ASTM requirements. ASTM D6690-12 identi- fies four different types of sealants as follows: Type I: Sealant for moderate climates, with low-temperature performance tested at -18°C with 50% extension.

9 Type II: Sealant for most climates, with low-temperature performance tested at -29°C with 50% extension. Type III: Sealant for most climates, with low-temperature performance tested at -29°C with 50% extension—special tests are also included {ASTM notes that these specifica- tion requirements were formerly Federal Specification SS-S-1401C}. Type IV: Sealant for very cold climates, with low-temperature performance tested at -29°C with 200% extension. Table 2-2 indicates the tests used for each type of material. The reader is referred to ASTM D6690 (TM-11) for details of the specific test requirements and procedures. While the ASTM procedures have been in use for many years, it is well known that fundamental engineering proper- ties of the materials are not developed from the procedures. In addition, there is poor correlation between field performance and lab tests. As noted in Table 2-2, aging of the material is not usually evaluated. Further complicating the evaluation from a producer’s perspective is the fact that many states modify the test values for local conditions (8). Recent research by Al-Qadi et al.1 (8) in the characteriza- tion of sealants has resulted in the development of a new grading system for sealants, loosely based on the same test methods as used for Superpave PG asphalts. The concept is to develop standard methods and procedures based on fun- damental material properties. This new approach is called the Performance-Based Grading System for Hot-Poured Crack Sealant. The materials are identified by a Sealant Grade (SG) designation. The sealant grading is identified as shown in the following example: SG 70-16 Where SG = Sealant Grade 70 = the high temperature performance based on track- ing resistance, °C -16 = the low temperature performance based on stiffness, adhesion, and cohesion properties, °C As with the Superpave PG grades, the SG grades can be tai- lored to meet the environmental requirements for the appli- cation. The grading system is based on both a high and low temperature requirement. Any combination of high and low temperature grades shown in Table 2-3 is theoretically possi- ble. However, it is unlikely that there will be availability of all grades in a given region. Sealant manufacturers undoubtedly produce a few products for a climatic area, but it is unlikely that all products will be available everywhere. At low in-service temperatures, the key issues for the seal- ant are to achieve proper adhesion for bonding and to have adequate flexibility and extendibility to tolerate the movement of the crack. The tests used to evaluate these low-temperature Test Procedure Type I Type II Type III Type IV Cone Penetration at 25oC Softening Point, oC Bond, non- immersed Bond, water- immersed Resilience, % Oven Aged Resilience, % Asphalt Compatibility Table 2-2. ASTM tests for each sealant type (TM-11). 1 It is noted that this referenced report is an executive summary of many years of research conducted as a pooled fund program administered by FHWA. Each of the test recommendations is thoroughly evaluated in separate reports. High Temperature Low Temperature 46 -46 52 -40 58 -34 64 -28 70 -22 76 -16 82 -10 Table 2-3. Sealant Grade high and low temperatures (8).

10 issues are the direct tension test, bending beam rheometer, and adhesion tests. At high in-service temperatures, the key issues are for the sealant to have sufficient elasticity against intrusion of debris and to resist flow and softening that could result in sealant tracking. The dynamic shear rheometer test is used to evaluate these properties. At installation temperatures, the rotational viscometer is used to evaluate the sealant properties for easy and proper installation. In the development of the SG system, some modifications to the PG test protocols were required. The following pro- vides a general overview of the protocol modifications to accommodate sealant products: • Rotational Viscometer: Used for measuring the flow prop- erties of the sealant; hence, upper and lower thresholds were identified as well as a change in the testing procedure. Instead of the binder hook used for conventional asphalts, a stiff metal rod replaces the wire hook and attaches to the spindle. Testing is conducted at the sealant application temperature. • Vacuum Oven Aging: Used to simulate aging of the sealant during service. A modification to the shelves in the oven is required to allow a uniform temperature profile in the oven. • Bending Beam Rheometer: Used to evaluate the flexibility of the sealant at low temperatures. The specimen is doubled in thickness, requiring a minor modification to the device to allow both binder and sealant testing. • Adhesion: Used to evaluate the bonding between sealant and aggregate. The tests are used to ensure the sealant adheres to the crack walls and that the bond will endure the applied thermal stresses on the sealant. • Direct Tension: Used to simulate field crack movements and to evaluate a sealant’s ability to withstand extension. The PG test protocol is modified and the equipment has slight modifications. • Dynamic Shear Rheometer: Used to evaluate tracking resis- tance at high temperatures. The specimen is doubled in thickness and the Multiple Stress Creep Recovery (MSCR) test is performed. The SG system provides a set of evaluation protocols that will assist users in selecting the proper grade of seal- ant for a specific application. The tests are new to sealant products but are familiar in the asphalt cement testing side of the industry, albeit with minor modifications. By evaluating the rheological properties of the sealant materials, this system provides an opportunity for sealant testing to be focused on performance-based criteria. The SG research reports provide recommendations for test criteria (8). As experience with the SG system expands, there may be modifications to the recommendations. This same “fine tuning” approach occurred with the implementation of the PG grading system for asphalt cements. Sampling Sealant Many specifications have sampling requirements. Examples of the requirements can be found in References TM-11, SS-4, SS-15, SS-16, and 83. Samples may be taken: (a) from the plant or warehouse prior to delivery, (b) at the time of delivery, (c) from the melter, or (d) from the applicator nozzle. As with any sample, documenting the sample project name, date, prod- uct, and location is critical for proper record keeping. Speci- fications generally define the lot/sublot size and the random sampling procedure required for the product. As an example, Wyoming (SS-4) defines a lot as no more than 90,000 pounds of sealant, with sublots of 30,000 pounds each. Construction Even with the best of materials, improper installation of the crack sealant compromises the performance of the appli- cation. It is therefore vital to have the sealant installed in a proper manner. This section discusses research activities that have helped to establish Best Practices. Discussion of the spe- cific Best Practices is presented in Chapter 4. Project Design It is critical that the condition of the existing pavement be evaluated prior to any preservation treatment. Crack treat- ments are no exception to that statement. It is imperative that a determination is made about potential crack movement, i.e., “working” versus “non-working” cracks, and that the pave- ment’s past and future rehabilitation activities are understood. Currently, there is not a universally accepted standard proto- col for this evaluation. Preparation for Crack Sealing/Filling In order for the sealant to bond, the crack must be clean and dry. Compressed air is commonly used to clean the crack. Routing of cracks is generally performed on transverse cracks that are “working” and greater than 3 mm (1⁄8″) in width prior to crack sealing. Ponniah and Kennepohl (33) recommend routing cracks between 3 mm (1⁄8″) and 19 mm (¾″) wide to a configuration of 40 mm × 10 mm (1-½″ × 3⁄8″). For milder cli- mates, Chong and Phang (35) indicate a rout of 19 mm × 19 mm (¾″ × ¾″) is also acceptable. Chong (37) subsequently indicates that a 12 mm × 12 mm (½″ × ½″) routing configuration works well for urban expressway applications. Eaton and Ashcraft

11 (23) caution that routing may be detrimental to pavements over 6 years old due to aging of the mixture. Smith and Romine (53) recommend the use of cutter wheel routers, as shown in Photographs 2-3(A) and 2-3(B). Sharp bits are required to achieve a clean cut. Crafco (81) further rec- ommends that the unit be capable of following random cracks and be designed to adjust cutting widths. The unit should be equipped with a cutterhead fitted with carbide-tipped cutting tools and have variable depth control. The machine must be capable of cutting approximately 1,000 to 1,200 linear feet per hour and provide a reservoir in the pavement that meets the design for the project. A hot air lance (HAL) (shown in Photograph 2-4) is also recommended by FHWA (53) to remove dust and moisture from the crack to ensure a better bond between the pave- ment and the sealant. Crafco (81) recommends that the HAL be capable of producing air temperatures up to 750°F and be constructed of suitable hardware. The equipment should be provided with separate valves to control propane, burner air, and lance air. The fuel and the burner air should be mixed only at the point of combustion before leaving the burner tube. A separate air lance tube should pass inside the burner chamber and have a maximum orifice of ¼″. At the fuel source, the propane should be controlled by a high-pressure regulator to control fuel pressure from 5 PSI to 30 PSI and to prevent flashback. Burner BTU should range from 20,000 to 500,000 BTU. A wheel kit constructed to keep the unit at the proper height and angle from the pavement, and to prevent debris from striking the operator, may also be used. Caution should be taken when using the HAL to not overheat and oxidize the pavement. A slight darkening of the pavement is acceptable. For crack filling, generally the only preparation recom- mended is to clean and dry the crack. Chong and Phang (35) recommend that the maximum distance between cleaning and sealing operations be 60–80 feet. Schutz (58) presented an argument that backer rod should be used to maintain proper shape factor for the sealant. By not having the sealant adhered to the bottom surface of the crack, the expansion and contraction of the sealant is not constrained on the horizontal surface. Installation of Crack Sealing For the installation to proceed, the sealant must be brought to application temperature. Crafco (81) recommends that the melter for hot-poured applications be a self-contained double boiler device with the transmittal of heat through heat transfer oil to the sealant vessel. It must be equipped with an on-board automatic heat-controlling device to per- mit the attainment of a predetermined sealant temperature and, then, maintain that temperature as long as required. The (B) Crack router (21).(A) Cutting wheel (photo by Dale Decker). Photographs 2-3. Cutting wheel router.

12 melter must be capable of safely heating product to 400°F. The temperature control should not allow the heat transfer oil to exceed 525°F. There should be temperature readings of the sealant within the melting vessel and within the discharge plumbing to provide monitoring of the sealant throughout the operation. The unit shall also have a means to vigorously and continuously agitate the sealant that meets requirements of ASTM D6690. The sealant should be applied to the pave- ment under pressure supplied by a gear pump with a direct connecting applicator tip. Chong (37) recommends overfilling the crack to just cover both edges of the crack and to allow for shrinkage during cool- ing. This approach minimizes snowplow damage for routed cracks. Quality Control Quality Control of a crack treatment operation consists of: (a) inspection of the operation, (b) sealant sampling and test- ing, (c) calibration of the equipment, and (d) inspection of the equipment. This section contains a brief discussion of each activity. Inspection Unfortunately, pavement preservation activities often do not command an adequate amount of attention for inspection ser- vices. With millions of dollars for a pavement reconstruction/ rehabilitation project, hundreds of thousands of dollars for a surface treatment project, and only tens of thousands of dollars for a crack treatment project, it is easy to understand how an agency will prioritize activities of limited inspection personnel with limited budget. Likewise, training is often not a high-priority activity for crack treatment operations. Personnel need to understand the importance of their activities and the proper method of application. Many organizations depend on on-the-job training. In some cases, this approach works well. However, it is all too easy for uniformity of on-the-job training to suffer when work needs to get done on a time schedule and manpower is limited. In addition, if bad habits are developed, generations of employees all learn the same bad habits. Training resources on crack treatments are available, for example, through FHWA’s NHI course #131110C, the National Center for Pavement Preservation, and References 3, 4, 5, 21, 22, 25, and 84. Many states require inspectors to be certified for construc- tion inspection. The development of an appropriately scoped certification program for crack treatment operations should be considered. As an example of one training approach, Nebraska Department of Roads (82) requires a one-hour training ses- sion prior to crack sealing activities. “Tailgate training” is an approach that has been used in a variety of situations ranging from safety training to materials handling and is an option that could be viable. Material Sampling and Testing When sampling any material for evaluation, it is critical that the sample truly represent the materials being evaluated. A bad sample provides bad information on the material. Calibration of the Equipment The key calibration component for crack sealing equip- ment is to ensure that the temperature control on the melter is working properly. Based on research by Masson et al. (56), overheating may cause damage to the sealant material. Inspection of the Equipment Equipment should be visually inspected for obvious defects prior to the start of each workday. Equipment manufacturers include maintenance recommendations with their specific equipment. These recommendations should be followed. Performance AASHTO’s National Transportation Product Evaluation Program (NTPEP) has performed evaluations for a variety of crack treatment products and for several state agencies. Photograph 2-4. Hot air lance (courtesy Crafco).

13 Details on these evaluations can be found at www.ntpep.org. Work done by the crack treatment pooled fund study (8, 84) also includes performance evaluations. Review of these eval- uations is recommended to the reader. Even as early as NCHRP Report 38 in 1967, it was recog- nized that cold-poured materials were not performing as well as hot-poured materials (29). Yildirim et al. (21) report crack sealing without routing using cold-poured materials has a typical life cycle of 1–2 years, while hot-poured materials have a typical life cycle of 3–5 years. CalTrans (22) reports that emulsion sealants in unrouted flush fill applications have a life expectancy of 2–4 years, whereas hot-poured applica- tions (either flush fill or overband) have a life expectancy of 6–8 years. Ponniah (34) reports that hot-poured crack treat- ments extend pavement life 2–5 years. Eaton and Ashcraft (23) report from their survey that emulsions for sealers (cold- poured applications) appear to only work where freeze/thaw cycles are not present for the pavement. Cost-Effectiveness To establish the cost-effectiveness of rout and seal mainte- nance treatments, Chong and Phang (35) suggest the following information is required: 1. The effectiveness of the treatment—(a) performance of sealant materials over time and (b) performance of various rout width and depth sizes over time to establish the most efficient rout configuration; 2. The extension of pavement service life—(a) retarding of additional crack development and (b) delaying the dete- rioration process of the existing distress; and 3. The influence of time—at which point of the pavement’s life cycle the treatment is applied most cost-effectively. While their focus was on rout and seal approaches, the sug- gestions for evaluation of cost-effectiveness are true for other crack treatment applications as well. Eaton and Ashcraft (23) report that chip seal applications cost 3–14 times as much as crack sealing and that overlays cost 8–26 times as much as crack sealing. With an overlay, cracks typically reappear 1–2 years after the application, depending on the thickness of the overlay. “No matter how expensive your sealant is, it is the least expensive part of the job.” —Eaton and Ashcraft, 1992 (23) Conclusion Significant research has been conducted over many years regarding proper crack treatment materials, processes, and procedures. This literature review has documented the state- of-the-art for the processes.