Investigation of Warm-Mix Asphalt Using Iowa Aggregates Final Report, April 2011

The implementation of warm-mix asphalt (WMA) is becoming more widespread with a growing number of contractors utilizing various WMA technologies. Early research suggests WMA may be more susceptible to moisture damage than traditional hot-mix asphalt (HMA) mixes. The objectives of this study are to test the binder and mix properties of WMA technologies for both field- and laboratory-produced mixes to determine the performance of WMA compared to traditional HMA. Field- and laboratory-produced mixes were studied. The laboratory-produced mixes compared HMA control mixes with WMA mixes that had the same mix design. The WMA technologies used for the laboratory study were Advera, Sasobit, and Evotherm. The field study tested four WMA field-produced mixes. Each of the four mixes had a corresponding control HMA mix. The WMA technologies used in the field study included: Evotherm 3G/Revix, Sasobit, and Double Barrel Green Foaming. The three main factors for this study were WMA/HMA, moisture-conditioned/not moisture-conditioned, and reheated/not reheated. Mixes were evaluated based on performance tests. Binder testing was performed to determine the rheological differences between HMA and WMA binders to determine if binder grade requirements change with the addition of WMA additives. The conclusions of this study are as follows:  Reduced mixing and compaction temperatures were achieved.  Statistical differences were found when comparing tensile strength ratio (TSR) values for both laboratory- and field-produced mixes. In the laboratory, none of the WMA additives performed as well as the HMA. For the field mixes, all TSR values passed Iowa’s minimum specification of 0.8 but, on average, WMA is lower compared to HMA TSR values.  Dynamic modulus results show that, on average, HMA will have higher dynamic modulus values. This means the HMA exhibits stiffer material properties compared to WMA; this may not necessarily mean superior performance in all cases.  Flow number results show that WMA has reduced flow number values compared to HMA. The only exception was the fourth field mix and weather delayed production of the control mix by nine days. The laboratory mixes showed that flow number values increased significantly with the addition of recycled asphalt pavement (RAP).  In the laboratory study, Advera reduced TSR values. Given that Advera is a foaming agent, the increase in moisture susceptibility is likely attributed to the release of water necessary for the improvement of the workability of the asphalt mixture.

Improving the Accuracy and Usability of Iowa Falling Weight Deflectometer Data, May 2013

This study aims to improve the accuracy and usability of Iowa Falling Weight Deflectometer (FWD) data by incorporating significant enhancements into the fully-automated software system for rapid processing of the FWD data. These enhancements include: (1) refined prediction of backcalculated pavement layer modulus through deflection basin matching/optimization, (2) temperature correction of backcalculated Hot-Mix Asphalt (HMA) layer modulus, (3) computation of 1993 AASHTO design guide related effective SN (SNeff) and effective k-value (keff ), (4) computation of Iowa DOT asphalt concrete (AC) overlay design related Structural Rating (SR) and kvalue (k), and (5) enhancement of user-friendliness of input and output from the software tool. A high-quality, easy-to-use backcalculation software package, referred to as, I-BACK: the Iowa Pavement Backcalculation Software, was developed to achieve the project goals and requirements. This report presents theoretical background behind the incorporated enhancements as well as guidance on the use of I-BACK developed in this study. The developed tool, I-BACK, provides more fine-tuned ANN pavement backcalculation results by implementation of deflection basin matching optimizer for conventional flexible, full-depth, rigid, and composite pavements. Implementation of this tool within Iowa DOT will facilitate accurate pavement structural evaluation and rehabilitation designs for pavement/asset management purposes. This research has also set the framework for the development of a simplified FWD deflection based HMA overlay design procedure which is one of the recommended areas for future research.

Improving the Accuracy and Usability of Iowa Falling Weight Deflectometer Data: Tech Transfer Summary, May 2013

This study aims to improve the accuracy and usability of Iowa Falling Weight Deflectometer (FWD) data by incorporating significant enhancements into the fully-automated software system for rapid processing of the FWD data. These enhancements include: (1) refined prediction of backcalculated pavement layer modulus through deflection basin matching/optimization, (2) temperature correction of backcalculated Hot-Mix Asphalt (HMA) layer modulus, (3) computation of 1993 AASHTO design guide related effective SN (SNeff) and effective k-value (keff ), (4) computation of Iowa DOT asphalt concrete (AC) overlay design related Structural Rating (SR) and kvalue (k), and (5) enhancement of user-friendliness of input and output from the software tool. A high-quality, easy-to-use backcalculation software package, referred to as, I-BACK: the Iowa Pavement Backcalculation Software, was developed to achieve the project goals and requirements. This report presents theoretical background behind the incorporated enhancements as well as guidance on the use of I-BACK developed in this study. The developed tool, I-BACK, provides more fine-tuned ANN pavement backcalculation results by implementation of deflection basin matching optimizer for conventional flexible, full-depth, rigid, and composite pavements. Implementation of this tool within Iowa DOT will facilitate accurate pavement structural evaluation and rehabilitation designs for pavement/asset management purposes. This research has also set the framework for the development of a simplified FWD deflection based HMA overlay design procedure which is one of the recommended areas for future research.

MEPDG Work Plan Task No. 5: Characterization of Unbound Materials (Soils/Aggregates) for Mechanistic-Empirical Pavement Design Guide, 2009

The resilient modulus (MR) input parameters in the Mechanistic-Empirical Pavement Design Guide (MEPDG) program have a significant effect on the projected pavement performance. The MEPDG program uses three different levels of inputs depending on the desired level of accuracy. The primary objective of this research was to develop a laboratory testing program utilizing the Iowa DOT servo-hydraulic machine system for evaluating typical Iowa unbound materials and to establish a database of input values for MEPDG analysis. This was achieved by carrying out a detailed laboratory testing program designed in accordance with the AASHTO T307 resilient modulus test protocol using common Iowa unbound materials. The program included laboratory tests to characterize basic physical properties of the unbound materials, specimen preparation and repeated load triaxial tests to determine the resilient modulus. The MEPDG resilient modulus input parameter library for Iowa typical unbound pavement materials was established from the repeated load triaxial MR test results. This library includes the non-linear, stress-dependent resilient modulus model coefficients values for level 1 analysis, the unbound material properties values correlated to resilient modulus for level 2 analysis, and the typical resilient modulus values for level 3 analysis. The resilient modulus input parameters library can be utilized when designing low volume roads in the absence of any basic soil testing. Based on the results of this study, the use of level 2 analysis for MEPDG resilient modulus input is recommended since the repeated load triaxial test for level 1 analysis is complicated, time consuming, expensive, and requires sophisticated equipment and skilled operators.

Investigation into Shrinkage of High-Performance Concrete Used for Iowa Bridge Decks and Overlays, TR-633, 2013

High-performance concrete (HPC) overlays have been used increasingly as an effective and economical method for bridge decks in Iowa and other states. However, due to its high cementitious material content, HPC often displays high shrinkage cracking potential. This study investigated the shrinkage behavior and cracking potential of the HPC overlay mixes commonly used in Iowa. In the study, 11 HPC overlay mixes were studied. These mixes consisted of three types of cements (Type I, I/II, and IP) and various supplementary cementitious materials (Class C fly ash, slag and metakaolin). Limestone with two different gradations was used as coarse aggregates in 10 mixes and quartzite was used in one mix. Chemical shrinkage of pastes, free drying shrinkage, autogenous shrinkage of mortar and concrete, and restrained ring shrinkage of concrete were monitored over time. Mechanical properties (such as elastic modulus and compressive and splitting tensile strength) of these concrete mixes were measured at different ages. Creep coefficients of these concrete mixes were estimated using the RILEM B3 and NCHRP Report 496 models. Cracking potential of the concrete mixes was assessed based on both ASTM C 1581 and simple stress-to-strength ratio methods. The results indicate that among the 11 mixes studied, three mixes (4, 5, and 6) cracked at the age of 15, 11, and 17 days, respectively. Autogenous shrinkage of the HPC mixes ranges from 150 to 250 microstrain and free dying shrinkage of the concrete ranges from 700 to 1,200 microstrain at 56 days. Different concrete materials (cementitious type and admixtures) and mix proportions (cementitious material content) affect concrete shrinkage in different ways. Not all mixes having a high shrinkage value cracked first. The stresses in the concrete are associated primarily with the concrete shrinkage, elastic modulus, tensile strength, and creep. However, a good relationship is found between cementitious material content and total (autogenous and free drying) shrinkage of concrete.

Ranking of HMA Moisture Sensitivity Tests in Iowa, RB00-012, 2012

Several agencies specify AASHTO T283 as the primary test for field acceptance of moisture susceptibility in hot mix asphalt. When used in this application, logistical difficulties challenge its practicality, while repeatability is routinely scrutinized by contractors. An alternative test is needed which can effectively demonstrate the ability to screen mixtures based on expected performance. The ideal replacement can be validated with field performance, is repeatable, and allows for prompt reporting of results. Dynamic modulus, flow number, AASHTO T283, Hamburg wheel tracking device (HWTD), and the moisture induced sensitivity test (MIST) were performed on plant produced surface mixes in Iowa. Follow-up distress surveys were used to rank the mixes by their performance. The rankings indicate both the quantity of swelling from MIST conditioning and submersed flow number matched the performance ranking of all but one mixture. Hamburg testing parameters also appear effective, namely the stripping inflection point and the ratio between stripping slope and the creep slope. Dynamic modulus testing was ineffective, followed by AASHTO T283 and ratios produced from flow number results of conditioned samples.

Beneficial Effects of Selected Additives on Asphalt Cement Mixes, HR-278, 1987

Effects of polyolefins, neoprene, styrene-butadiene-styrene (SBS) block copolymers, styrene-butadiene rubber (SBR) latex, and hydrated lime on two asphalt cements were evaluated. Physical and chemical tests were performed on a total of 16 binder blends. Asphalt concrete mixes were prepared and tested with these modified binders and two aggregates (crushed limestone and gravel), each at three asphalt content levels. Properties evaluated on the modified binders (original and thin-film oven aged) included: viscosity at 25 deg C, 60 deg C and 135 deg C with capillary tube and cone-plate viscometer, penetration at 5 deg C and 25 deg C, softening point, force ductility, and elastic recovery at 10 deg C, dropping ball test, tensile strength, and toughness and tenacity tests at 25 deg C. From these the penetration index, the viscosity-temperature susceptibility, the penetration-viscosity number, the critical low-temperature, long loading-time stiffness, and the cracking temperature were calculated. In addition, the binders were studied with x-ray diffraction, reflected fluorescence microscopy, and high-performance liquid chromatography techniques. Engineering properties evaluated on the 72 asphalt concrete mixes containing additives included: Marshall stability and flow, Marshall stiffness, voids properties, resilient modulus, indirect tensile strength, permanent deformation (creep), and effects of moisture by vacuum-saturation and Lottman treatments. Pavement sections of varied asphalt concrete thicknesses and containing different additives were compared to control mixes in terms of structural responses and pavement lives for different subgrades. Although all of the additives tested improved at least one aspect of the binder/mixture properties, no additive was found to improve all the relevant binder/mixture properties at the same time. On the basis of overall considerations, the optimum beneficial effects can be expected when the additives are used in conjunction with softer grade asphalts.

Field Evaluation of Class A Subbase Using Fly Ash, MLR-87-3, 1988

This project consisted of slipforming a 4-inch thick econocrete subbase on a 6-mile section of US 63. The project location extends south from one mile south of Denver, Iowa to Black Hawk County Road C-66 and consisted of the reconstruction and new construction of a divided four-lane facility. The econocrete was placed 27.3 feet wide in a single pass. Fly ash was used in this field study to replace 30, 45 and 60 percent of the portland cement in three portland cement econocrete base paving mixes. The three mixes contained 300, 350 and 400 pounds of cementitious material per cubic yard. Two Class "C" ashes from Iowa approved sources were used. The ash was substituted on the basis of one pound of ash for each pound of cement removed. The work was done October 6-29, 1987 and May 25-June 9, 1988. The twelve subbase mixes were placed in sections 2500 to 3000 feet in length on both the north and southbound roadways. Compressive strengths of all mixes were determined at 3 and 28 days of age. Flexural strengths of all mixes were determined at 7 and 14 days. In all cases strengths were adequate. The freeze/thaw durability of the econocrete mixes used was reduced by increased fly ash levels but remained above acceptable limits. The test results demonstrate the feasibility of producing econocrete with satisfactory properties even using fly ash at substitution rates up to 45 percent.

Laboratory Evaluation of Asphalt-Rubber in Asphalt Paving Mixture, MLR-89-15, 1990

Interest in the use of ground rubber from used tires as a hot asphalt mix binder has been increasing due to the magnitude of the disposal problem posed by the annual addition of millions of waste tires to the refuse stream. This study evaluates, through laboratory means, the performance of asphalt-rubber as a hot mix binder as compared to conventional asphalt. The results indicate that asphalt-rubber outperforms its base asphalt in mixes of identical gradation and comparable void content on tests that are heavily dependent on binder characteristics (resilient modulus and indirect tension). An appreciable increase in rut resistance due to the use of asphalt-rubber is not indicated.

Evaluation of Asphalt Mix Permeability, MLR-90-2, 1992

Efforts to eliminate rutting on the Interstate system have resulted in 3/4 in. aggregate mixes, with 75 blow Marshall, 85% crushed aggregate mix designs. On a few of these projects paved in 1988-1989, water has appeared on the surfaces. Some conclusions have been reached by visual on-sight investigations that the water is coming from surface water, rain and melting snow gaining entry into the surface asphalt mixture, then coming back out in selected areas. Cores were taken from several Interstate projects and tested for permeability to investigate the surface water theory that supposedly happens with only the 3/4 in. mixtures. All cores were of asphalt overlays over portland cement concrete, except for the Clarke County project which is full depth AC. The testing consisted of densities, permeabilities, voids by high pressure airmeter (HPAM), extraction, gradations, AC content, and film thicknesses. Resilient modulus, indirect tensile and retained strengths after freeze/thaw were also done. All of the test results are about as expected. Permeabilities, the main reason for testing, ranged from 0.00 to 2.67 ft per day and averages less than 1/2 ft per day if the following two tests are disregarded. One test on each binder course came out to 15.24 ft/day, and a surface course at 13.78 ft/day but these are not out of supposedly problem projects.

Characteristics of Iowa Fine Aggregate, MLR-92-6, 1996

The objective of this research was to evaluate the quality (angularity, mortar strengths and alkali-silica reactivity) of fine aggregate for Iowa portland cement concrete (PCC) pavements. Sands were obtained from 30 sources representative of fine aggregate across Iowa. The gradation, fineness modulus and mortar strengths were determined for all sands. Angularity was evaluated using a new National Aggregate Association (NAA) flow test. The NAA uncompacted void values are significantly affected by the percent of crushed particles and are a good measure of fine aggregate angularity. The alkali-silica reactivity of Iowa sands was measured by the ASTM P214 test. By P214 many Iowa sands were identified as being reactive while only two were innocuous. More research is needed on P214 because pavement performance history has shown very little alkali-silica reactivity deterioration of pavement. Six of the sands tested by P214 were evaluated using the Canadian Prism Test. None were identified as being reactive by the Canadian Prism Test.

Laboratory Evaluation of Polymer and Multi-Grade Asphalt Binders, MLR-90-5, 1992

A number of claims have been made that polymer modified asphalt cements, multi-grade asphalt cements, and other modifications of the liquid asphalt will prevent rutting and other deterioration of asphalt mixes, thereby, extending the service life of asphalt pavements. This laboratory study evaluates regular AC-20 asphalt cement, PAC-30 polymer modified asphalt cement and AC-10-30 multi-grade asphalt cement. PAC-30 was also evaluated with 15% Gilsonite and 15% Witcurb in a 75% crushed stone - 25% sand mix. These mixtures were evaluated for all Marshall properties along with indirect tensile, resilient modulus, and creep resistance.

Strengthening of an Existing Continuous-Span, Steel-Beam, Concrete-Deck Bridge by Post-Tensioning, HR-308, 1990

The need to upgrade a large number of understrength and obsolete bridges in the U.S. has been well documented in the literature. Through several Iowa DOT projects, the concept of strengthening simple-span bridges by post-tensioning has been developed. The purpose of the project described in this report was to investigate the use of post-tensioning for strengthening continuous composite bridges. In a previous, successfully completed investigation, the feasibility of strengthening continuous, composite bridges by post-tensioning was demonstrated on a laboratory 1/3-scale-model bridge (3 spans: 41 ft 11 in. x 8 ft 8 in.). This project can thus be considered the implementation phase. The bridge selected for strengthening was in Pocahontas County near Fonda, Iowa, on County Road N28. With finite element analysis, a post-tensioning system was developed that required post-tensioning of the positive moment regions of both the interior and exterior beams. During the summer of 1988, the strengthening system was installed along with instrumentation to determine the bridge's response and behavior. Before and after post-tensioning, the bridge was subjected to truck loading (1 or 2 trucks at various predetermined critical locations) to determine the effectiveness of the strengthening system. The bridge, with the strengthening system in place, was inspected approximately every three months to determine any changes in its appearance or behavior. In 1989, approximately one year after the initial strengthening, the bridge was retested to identify any changes in its behavior. Post-tensioning forces were removed to reveal any losses over the one-year period. Post-tensioning was reapplied to the bridge, and the bridge was tested using the same loading program used in 1988. Except for at a few locations, stresses were reduced in the bridge the desired amount. At a few locations flexural stresses in the steel beams are still above 18 ksi, the allowable inventory stress for A7 steel. Although maximum stresses are above the inventory stress by about 2 ksi, they are about 5 ksi below the allowable operating stress; therefore, the bridge no longer needs to be load-posted.

Strengthening of Existing Single Span Steel Beam and Concrete Deck Bridges, HR-238, Part II, 1985

The unifying objective of Phases I and II of this study was to determine the feasibility of the post-tensioning strengthening method and to implement the technique on two composite bridges in Iowa. Following completion of these two phases, Phase III was undertaken and is documented in this report. The basic objectives of Phase III were further monitoring bridge behavior (both during and after post-tensioning) and developing a practical design methodology for designing the strengthening system under investigation. Specific objectives were: to develop strain and force transducers to facilitate the collection of field data; to investigate further the existence and effects of the end restraint on the post-tensioning process; to determine the amount of post-tensioning force loss that occurred during the time between the initial testing and the retesting of the existing bridges; to determine the significance of any temporary temperature-induced post-tensioning force change; and to develop a simplified design methodology that would incorporate various variables such as span length, angle-of-skew, beam spacing, and concrete strength. Experimental field results obtained during Phases II and III were compared to the theoretical results and to each other. Conclusions from this research are as follows: (1) Strengthening single-span composite bridges by post-tensioning is a viable, economical strengthening technique. (2) Behavior of both bridges was similar to the behavior observed from the bridges during field tests conducted under Phase II. (3) The strain transducers were very accurate at measuring mid-span strain. (4) The force transducers gave excellent results under laboratory conditions, but were found to be less effective when used in actual bridge tests. (5) Loss of post-tensioning force due to temperature effects in any particular steel beam post-tensioning tendon system were found to be small. (6) Loss of post-tensioning force over a two-year period was minimal. (7) Significant end restraint was measured in both bridges, caused primarily by reinforcing steel being continuous from the deck into the abutments. This end restraint reduced the effectiveness of the post-tensioning but also reduced midspan strains due to truck loadings. (8) The SAP IV finite element model is capable of accurately modeling the behavior of a post-tensioned bridge, if guardrails and end restraints are included in the model. (9) Post-tensioning distribution should be separated into distributions for the axial force and moment components of an eccentric post-tensioning force. (10) Skews of 45 deg or less have a minor influence on post-tensioning distribution. (11) For typical Iowa three-beam and four-beam composite bridges, simple regression-derived formulas for force and moment fractions can be used to estimate post-tensioning distribution at midspan. At other locations, a simple linear interpolation gives approximately correct results. (12) A simple analytical model can accurately estimate the flexural strength of an isolated post-tensioned composite beam.

Field Demonstration and Laboratory Evaluation of Foamed Asphalts - Muscatine County, HR-233, 1985

In view of the energy, environmental, and economic advantages of the foamed asphalt process using local aggregates in cold mixes and the promising results from Research Project HR-212, a 4.2-mile section of county road in Muscatine County was built with foamed asphalt and local aggregates during August-September 1983. Extensive laboratory evaluation was carried out on five plant mixes representing foamed mixes used in the nine test sections, a laboratory prepared foamed mix, and a laboratory prepared hot mix similar to Plant Mix 1. The foamed mixes were compacted, cured under 15 curing conditions and tested for bulk specific gravity, Marshall stability at 77° F and at 140° F, cured moisture content, resilient modulus and effects of moisture damage due to freeze-thaw cycles, water soaking, and vacuum saturation. In addition, four sets of 83 core samples were taken at 1 to 15 months and tested for moisture content, specific gravity, Marshall stability, and resilient modulus. In summary, the test road has performed satisfactorily for almost two years. The few early construction problems encountered were to be expected for experimental projects dealing with new materials and technologies. Overall results to date are encouraging and foamed asphalt mixes have proved to have the potential as a viable base material in areas where marginal aggregates are available. It is hoped and expected that performance evaluation of the test sections will be continued and that more foamed asphalt trial projects will be constructed and monitored so that experiences and findings from this project can be verified and mix design criteria can be gradually established. For future foamed asphalt projects it is recommended that anti-stripping additives, such as hydrated lime, be added in view of the potential moisture susceptibility of foamed mixes observed in the laboratory evaluation.