YEAR 2006

INFLUENCE OF MODIFIED MARSHALL COMPACTION TECHNIQUE ON ENGINEERING PROPERTIES OF POLYMER MODIFIED AND NEAT BITUMINOUS CONCRETE MIXES.
By S.S. Awanti1, M.S. Amarnath2 & A. Veeraragavan3

ABSTRACT

This paper reports about the various laboratory investigations carried out on two types of polymer modified bituminous concrete mixes with Styrene-Butadiene-Styrene (SBS) copolymer and Styrene-Butadiene-Rubber (SBR) and a neat bituminous concrete mix using 80/100 bitumen to study the effect of adopting modified Marshall compaction technique on engineering properties of bituminous mixes. Marshall mix design was carried out to determine optimum binder content and Marshall parameters, Static indirect tensile strength at different temperatures, tensile strength ratio and resilient modulus ratio were determined on specimens prepared with modified and standard Marshall compaction techniques. Indirect tensile fatigue tests were carried out on cylindrical specimens prepared using standard and modified Marshall compaction technique. Marshall test results indicate higher stability, higher flow, higher unit weight, lower air voids and lower optimum binder content for all the mixes compacted with modified Marshall hammer when compared to standard Marshall compaction. Static indirect tensile strength at different temperatures, tensile strength ratio, resilient modulus, resilient modulus ratio and fatigue life were found to be higher for mixes compacted with modified compaction when compared to standard Marshall compaction. Modified compaction shows lower permanent deformation as well as deformation rate. The performance of polymer modified bituminous concrete mix with SBS copolymer is found to be superior when compared to the other two mixes prepared with polymer-modified bitumen with SBR as well as neat bitumen.

1. INTRODUCTION

In India, Marshall mix design is the sole method used for the design of bituminous concrete mixes. This is a standard laboratory method used by and large in different parts of the world for determining the strength and flow characteristics of bituminous paving mixes. It is also recognized that the impact compaction used in standard Marshall method does not simulate the state of compaction and orientation of aggregate particles attained in the field compaction by road rollers and subsequently by the pneumatic wheels of vehicles during service due to absence of kneading and shearing action during the compaction blows. Accordingly, there has been a growing concern among the highway engineers that the Marshall method should be replaced with a better method, which will measure the fundamental properties of bituminous mixtures. Such a method should be able to predict pavement behaviour in terms of fundamental properties such as fatigue, rutting and low temperature cracking. This led to the development of Superpave (Superior Performing Asphalt Pavement) Asphalt Mixture Design Method in USA developed by Strategic Highway Research Program (SHRP) (Pandey 2002). This method includes test equipment, test methods and criteria. The key features in the Superpave mix design method are laboratory compaction and performance testing. Superpave Gyratory Compactor (SGC) accomplishes laboratory compaction. As it is possible to obtain similar degree of reorientation of aggregate particles caused by the shearing action imparted to the bituminous mixture in the field. The Superpave mix design procedure is not yet popular and the cost of equipment for Superpave mix design method is quite high and equipment is not easily available in India. Therefore, the other simpler and a low cost option to gyratory compaction was first reported in South Africa (Sabita, 1993; Rust et al 1992). Subsequently for evaluation of bituminous mixtures this modified Marshall hammer was used in India at

IIT Kharagpur by Palit et al (2001). The modified Marshall hammer or Hugo hammer method introduces number of differences from standard Marshall method, such as modification for the face of the Marshall hammer by providing indents on the compaction face. The compaction process consists of turning the hammer face by 300 after every five blows for the first sixty-five blows. The final ten blows were applied with a standard Marshall hammer so as to provide an even surface on the Marshall specimen. The process was repeated on the other face also.
1.1. Practical Application of Modified Compaction

Modified Marshall method has the potential of replacing gyratory compaction device. With performance data from the field, appropriate mix design criteria can be developed to suit the prevalent traffic level. Densities comparable to those obtained immediately after the construction of the pavement and those after years of traffic can be duplicated in the laboratory by varying the number of blows. Like in the case of Superpave compaction, ultimate density that can be achieved by the in-service bituminous layer can also be established and those mixes whose air voids approach zero can be discarded. It is desirable that laboratory tests are conducted on the specimens of Bituminous Concrete and Dense Bituminous Macadam whose densities are closer to those attained after a few months of traffic by which time the pavement is stabilized.

1.2. Objectives

· To compare Marshall parameters obtained by modified Marshall or Hugo hammer method of mix design and standard Marshall mix design method for polymer modified and neat bituminous concrete mixes.
· To compare indirect tensile strength values obtained at different temperatures for Marshall specimens prepared by using modified Marshall and standard Marshall mix design method for polymer modified and neat bituminous concrete mixes.
· To compare the moisture susceptibility in terms of Tensile strength ratio and ratio of resilient modulus for specimens prepared using modified Marshall and standard Marshall mix design method for polymer modified and neat bituminous concrete mixes.
· To compare fatigue lives and rut resistance obtained by indirect tensile fatigue test for specimens prepared using modified Marshall and standard Marshall mix design method for polymer modified and neat bituminous concrete mixes.

2. LITERATURE REVIEW
Limited work has been reported on effect of Modified Marshall mix design technique on engineering properties of polymer modified and neat bituminous concrete mixes. Following two researchers worked on Hugo hammer technique.

Palit et al (2001) carried out a laboratory study on the effect of adopting modified Marshall compaction on the engineering properties of neat and crumb rubber modified bituminous mixtures. They reported that the densities attained using modified Marshall compaction were similar to those obtained by gyratory compaction. Indirect tensile strength, elastic modulus, fatigue life as well as resistance against rutting were found to improve with modified compaction. It was also observed that the performance of mixes prepared with bitumen modified using crumb rubber is much superior when compared to that of normal mixes prepared with neat bitumen for both standard and modified Marshall compaction. Kohli et al (2003) carried out laboratory study on the use of Hugo hammer method in design of dense bituminous macadam. They reported that by using Hugo hammer method, there is clear-cut saving in bitumen in the range between 6per cent and 18per cent for the three gradations namely gradation suggested by Kandhal (USA), FHWA (0.45 power gradation) or maximum density gradation and MoRT&H 2001 for DBM using large size aggregate. Stability values obtained are little higher than the recommended value given by MoRT&H 2001 for the three gradings using Hugo hammer method. The stability and bulk density values for maximum density gradation using Hugo hammer was as high as 2500 kg and 2.4 gm/cm3 respectively indicating that the compactive effort plays a major role in achieving the strength of mix.

3. LABORATORY INVESTIGATIONS

3.1. Materials

· Neat Bitumen: 80/100 grade.
· Polymer modified bitumen of 70 grade (PMB-70): PMB-70 with Styrene-Butadiene-Styrene (SBS) and PMB-70 with Styrene-Butadiene-Rubber (SBR) are of readily blended type (in commercial form) using neat 80/100 as base bitumen. As per supplier’s (M/s Hindustan-Colas Limited, Tamil Nadu) report for the modification of base bitumen 3.5per cent of SBS in solid form and 5per cent of SBR in latex form was mixed.
· Aggregates and filler: Crushed granite coarse and fine aggregates. Ordinary Portland cement and stone dust were used as the mineral filler.
3.2. Physical Properties of Aggregates
Crushed granite coarse and fine aggregates, mineral filler viz, cement and stone dust used in the investigations had specific gravity of 2.65, 2.70, 3.10 and 2.61, respectively.

Crushing value, Impact value and Los Angeles Abrasion value are 39.78 per cent, 26.90 per cent, and 12.20 per cent respectively for crushed granite aggregates. These are fulfilling MoRT&H-2001 requirements.
3.3. Gradation of Aggregate

Grade-2 midpoint aggregate gradation for BC layer as per MoRT&H-2001 was adopted and is shown in Table 1.
3.4. Properties of Neat and Polymer Modified Bitumen

Various physical properties of neat and polymer modified bitumen were determined and the results are shown in Table 2. Bituminous binders are composed of organic molecules; they react with oxygen from environment. This reaction is called oxidation and it changes the structure and composition of bituminous molecules. Oxidation causes the bituminous binder to become more brittle, generating the


term oxidative hardening or age hardening. In practice, a considerable amount of oxidative hardening occurs before the bitumen is placed. At the hot mixing facility, bitumen is added to the hot aggregate and the mixture is maintained at elevated temperatures for a period of time. Because the bitumen exists in thin film covering the aggregate, the oxidation reaction occurs at a much faster rate. To simulate aging of binder during mixing and rolling or compaction in the field the Polymer modified binders such as SBS and SBR were subjected to Thin Film Oven Test (TFOT). The results of these tests are also shown in Table 2.

3.5. Mixing and Compaction Temperatures for PMB (SBS, SBR)

According to Hensley (1998), mixing temperature can be defined as that temperature, which produces a uniform and sufficient coating of the coarse aggregates, which is to be estimated on the basis of experience. Based on several trails, mixing temperature for modified binder was found to be 1800C and the compaction temperature was taken as 1700C. In order to establish further the above findings, viscosity temperature relationship is established for PMB with SBS and SBR and also for neat 80/100 bitumen.
As per IRC:SP:53-2002, for PMB mixing and compaction temperatures corresponds to viscosity values of 4 poise (max) and 5 poise, (max) respectively. As per ASTM D 1559-89, for neat bitumen mixing and compaction temperatures corresponds to viscosity values of 1.56 ± 0.18 poise and 2.57 ± 0.28 poise respectively. From viscosity temperature relationship as shown in Fig. 1 it is observed that, for PMB with SBS and SBR the desirable mixing and compaction temperatures are 1800C and 1700C, respectively. The above values for neat bitumen are 1500C and 1400C, respectively.

To investigate further the validity of this selection for PMB-SBS and PMB-SBR Marshall specimens were prepared with arbitrary binder content of 4.7 per cent for SBS and 5.0 per cent for SBR at different mixing temperatures of 160, 170, 180 and 1900C. Compacting temperatures were taken as 100C less than the corresponding mixing temperature. Grade-2 middle limit aggregate gradation as per MoRT&H-2001 was used and is shown in Table 1. From investigations it has been observed that maximum stability is obtained corresponding to a mixing temperature of 1800C for SBS and also for SBR. Other Marshall properties corresponding to this mixing temperature also satisfy the requirements as per IRC:SP:53-2002. Hence for further investigation mixing and compaction temperatures were taken as 1800C and 1700C respectively. Marshall properties at different mixing and compaction temperatures for SBS (PMBC1) and SBR (PMBC2) are shown in Table 3 and Table 4, respectively. Requirements of PMB mix as per IRC: SP: 53-2002 is shown in Table 5.

3.6. Marshall Test

The Marshall test is basically an unconfined test. The purpose of the test is to measure the strength of an asphalt mixture that is compacted to a standard laboratory compactive effort. This test is also used as part of the Marshall mix design procedure for selecting design bitumen content and also for the quality control of bituminous mixtures. The cylindrical specimen of 10 cm diameter and 6.35 cm height is loaded vertically with a compressive load along a diagonal at a rate of 50 mm per minute at 600C temperature through semicircular testing heads. The temperature of 600C was selected since this approximates the maximum pavement temperature in the summer, thereby providing the weakest condition for the hot bituminous mixtures. The Marshall stability is the maximum load, which the specimen can withstand. The flow value is the total vertical deformation of the specimen at maximum load. The Marshall quotient is the property that is used to characterize bituminous mixtures, which is a ratio of Marshall stability and flow. This is an empirical stiffness value used by some engineers, to evaluate the quality of bituminous mixtures. A higher value of Marshall quotient indicates a stiffer mixture and, hence, indicates the mixture is likely to offer more resistance to permanent deformation. This method has, by and large, stood the test of time because of limits on air void, VMA, voids filled with bitumen and flow coupled with feed back from field performance data. Marshall test can not predict fatigue or permanent deformation behaviour of in service pavement. In India, this method has not given satisfactory mixtures for heavy traffic roads and bleeding, cracking and permanent deformation soon after construction using this mix is not uncommon.

In the present investigation, a modified Marshall hammer with 3.2 mm indents (Fig. 2) on circular tamping face equipped with a 4.5 kg weight constructed to provide a free fall of 45.7 cm as in the case of standard Marshall hammer was fabricated. The compaction process consists of turning the hammer face by 300 after every five blows for the first sixty-five blows. The final ten blows were applied with a standard Marshall hammer so as to provide an even surface on the Marshall specimen. The process was repeated on the other face also. Marshall tests were conducted according to ASTM-D 1559-82. Optimum binder content was calculated as the average of binder content corresponding to maximum stability, maximum unit weight, and 4per cent air voids.

3.6.1 Mixes Investigated: Six types of mixes were considered for this investigation namely:
(i) BC/M, BC/S: Bituminous concrete mixes prepared with neat bitumen.
(ii) PMBC1/M, PMBC1/S: Bituminous concrete mixes prepared with polymer-modified bitumen PMB-70 with SBS.
(iii) PMBC2/M, PMBC2/S: Bituminous concrete mixes prepared with polymer-modified bitumen PMB-70 with SBR
(iv) M and S indicates modified and standard Marshall compactions, respectively.
(v) Marshall properties at optimum binder content for all six types of mixes are shown in Table 6.

Variations of Marshall stability, Flow, Unit weight and per cent Air voids with binder content for different mixes are shown in Figs. 3 and 4, respectively.
From Marshall test results it is observed that, the trend of stability versus binder content for PMBC1 and PMBC2 mixes are similar to BC mix. The maximum stability values are 56 per cent and 29 per cent higher for PMBC1 and PMBC2 when compared to BC using modified Marshall method, whereas, these values are 44 per cent and 31per cent with standard Marshall method. It is also observed that modified compaction gives 11per cent, 3per cent and 6per cent higher stability than standard compaction for all three mixes viz. PMBC1, PMBC2 and BC. In general, it can be seen that, the flow values increase with increase in binder content. However, higher flow values is recorded for modified compaction than standard compaction in the case of both neat and modified mixes. Higher unit weights are achieved for PMBC as well as

BC mixes compacted by modified Marshall compaction when compared to standard Marshall compaction. In general, air voids decreases with increase in binder content in case of modified as well as plain BC mixes. In modified compaction, air voids are low as compared to standard compaction at particular binder content for plain as well as modified mixes. This shows that modified compaction imparts better kneading action to bring proper orientation of aggregate particles. It is observed that the optimum binder content is reduced by 8 per cent, 9 per cent and 1.50 per cent for PMBC1, PMBC2 and BC mixes compacted by modified Marshall compaction when compared to standard Marshall compaction. This reduction brings substantial economy in PMBC mixes and a marginal economy in BC mixes.

3.7. Indirect Tensile Test
The static indirect tensile test was carried out as per ASTM: D-4123-82 (1995) to study the behaviour of paving mixes at different temperatures and also to fix up stress levels in indirect tensile fatigue test. The split tensile strength of bituminous mixes was determined by applying a compressive load to Marshall specimens along the vertical diametral plane, through two-curved steel strips 12.5 mm wide with the same inside curvature as that of the Marshall specimens. The load was applied at a rate of 5.08 cm/min until failure occurred. A nearly uniform tensile stress is developed normal to the direction of the applied load and along the same vertical plane causing the specimen to fail by splitting along the vertical

diameter. The ultimate load was obtained to calculate maximum indirect tensile strength. Specimens at each test condition were made in triplicate, and the test parameters reported as the numerical average of the test data. Specimens were prepared at the optimum binder content for PMBC1, PMBC2 and BC mixes obtained by modified as well as standard Marshall method. Tests were carried out at temperatures of 15, 20, 25, 30, 35 and 400C. Indirect tensile strength of specimen is calculated by using the equation (1) as per (ASTM: D-4123-82-1995).

Indirect Tensile Strength, s = 2P/pd t ......... Equn. 1.
Where, P= Load at failure in N; d = diameter of specimen in mm; t = thickness of specimen in mm
The indirect tensile strength at different temperatures for PMBC1, PMBC2 and BC specimens prepared with modified and standard Marshall compaction methods are shown in Table 7 and Fig. 5.

From static indirect tensile test results at test temperatures of 15, 20, 25, 30,3 5 and 40OC, the indirect tensile strength values for PMBC1, PMBC2 and BC specimens prepared using modified Marshall compaction are found to be higher in the order of 11per cent, 6 per cent, 6 per cent, 6 per cent, 8 per cent and 6 per cent; 4 per cent, 11per cent, 17 per cent, 18 per cent, 7 per cent and 15 per cent; 18 per cent, 25 per cent, 9 per cent, 24 per cent, 22 per cent and 12 per cent respectively when compared to above mixes prepared with standard Marshall compaction. From this it can be seen that modified compaction improved the performance of the mixes compared to standard compaction for all types of mixes. It is also observed that the polymer modified bituminous mixes such as PMBC1 and PMBC2 shows higher indirect tensile strength at all temperature levels when compared to BC mixes.

3.8. Tensile Strength Ratio (TSR)
Moisture susceptibility of bituminous mixes may also be determined in terms of TSR, which is expressed as the percentage of average static indirect tensile strength of the conditioned specimens to the average static indirect tensile strength of the unconditioned specimens. Conditioning consists of soaking the specimens in water maintained at 600C for 24 hours and allowing them to cure at 250C for one hour. The test was conducted at 250C. Specimens were compacted to about 6 per cent to 8 per cent air voids at design binder content. One subset of three specimens was considered as unconditioned or control specimens. The other subset of three specimens was considered as conditional subset (AASHTO T 283). The results of tensile strength ratio (TSR) for PMBC1, PMBC2 and BC specimens prepared with modified and standard Marshall compaction are presented in Table 8.

From the results it is found that TSR values are 3 to 4 per cent higher for all the mixes compacted with modified Marshall method when compared to standard Marshall method.

3.9. Resilient Modulus Ratio (MRR)

Resilient Modulus, MR is a ratio of applied stress to recoverable resultant (horizontal) strain. The resilient modulus test is basically a repetitive load test using the stress distribution principles of the indirect tensile test. Resilient modulus is close to conditions that prevail in the field since a loading pulse is followed by a rest period. The loading time and rest period can be varied to represent different vehicle speed and traffic intensity. The values of resilient modulus can be used to evaluate the relative quality of materials as well as to generate input for pavement design or pavement evaluation and analysis. The repeated load indirect tensile test for determining resilient modulus of bituminous mixtures was conducted by applying compressive loads with a haversine waveform. The load was applied vertically on the vertical diametral plane of cylindrical specimen of bituminous concrete. The resulting horizontal deformation of the specimen was measured. The loading frequency of 2 Hz inducing 0.25 s loading period and the rest period of 0.45 s were used. From the instantaneous resilient horizontal deformations the resilient moduli of PMBC and BC mixes were calculated at test temperature of 250C using the equation (2) as per ASTM: D-4123 (1995).

 

The resilient modulus ratio is computed for each mixture, which is expressed as the percentage of average resilient modulus value of the conditioned specimens to the average resilient modulus value of the unconditioned specimens. Conditioning consists of soaking the specimens in water maintained at 600C for 24 hours and allowing them to cure at 250C for one hour. The test was conducted at 250C. A repeated load of 2 kN was applied for all the mixes. Resilient horizontal deformations were measured after 50-100 cycles of load repetitions. The results of resilient modulus ratio (MRR) for PMBC1, PMBC2 and BC specimens prepared with modified and standard Marshall compaction are presented in Table 9

From the results it is found that MRR values are 2 to 4 per cent higher for all the mixes compacted with modified Marshall method when compared to standard Marshall method. From TSR and MRR values it is apparent that the moisture susceptibility reduces by using modified Marshall compaction when compared to standard Marshall compaction due to improved compactive effort, which results in lower moisture penetration and that leads to higher TSR and MRR

3.10 Repeated Load Indirect Tensile Fatigue Test

This test is similar to static indirect tensile test, but in this case repeated loads were applied and horizontal and vertical deformation measurements were made. This test method was selected because of its simplicity and the ease with which the samples can be prepared. This test can also be performed on field cores. A number of investigators in the past (Kennedy 1978; Salter and Rafati-Afshar 1987; Mohammad and Paul 1993; Palit et al 2004) employed this test for evaluating the fatigue performance of asphalt mixes. For this purpose, a repeated load test setup was designed and fabricated at the Civil Engineering Department, Bangalore University, Bangalore for evaluating the characteristics of pavement material under repeated loading condition. The test facility is complete with a computer interfaced data acquisition system used to control the nature, magnitude, and frequency of loading and to collect data from different sensors. This equipment can be used for conducting repeated load tests, indirect tensile fatigue tests, and fatigue tests on beam specimens. The setup consists of loading frame with a capacity of 2 tonnes and a double-ended shaft hydraulic cylinder with a servo valve to apply the load and hydraulic power pack consists of motor pump, relief valve, cooling arrangement, etc. The input to the servo valve is from servo amplifier, which receives its input either from built in function generator or from computer. A load cell of 1 tonnes capacity was used as feed back element. Vertical and horizontal deformations of the specimens were measured by using linear variable differential transducers (LVDT). PMBC1, PMBC2 and BC specimens were prepared at optimum binder content by mixing the aggregates and binder at desired mixing temperatures, and compacted by modified Marshall hammer or standard Marshall hammer at desired compacting temperatures to achieve the required bulk densities as per modified or standard Marshall mix design method. The specimen after conditioning for two hours at required temperature of 30OC was placed between two steel strips so that the central axes of the strips, specimen and piston were in the same vertical plane. A constant repeated load of 1.8 kN was applied at a loading frequency of 2 Hz with a loading period of 0.25 s and rest period of 0.45 s. An approximate haversine type of loading waveform was used. The load repetitions were continued till the specimen failed. In this study the specimen is treated as failed when the permanent horizontal deformation resulted is 5-mm. Fatigue lives is the number of load repetitions to cause failure of specimen as above. The specimens were tested at a temperature of 30OC. Load, vertical deformation and horizontal deformation patterns were accurately traced by taking 100 readings per cycle. From these instantaneous resilient horizontal deformations the resilient moduli of PMBC and BC mixes were calculated at test temperature using the equation (2) as per ASTM: D-4123 (1995).

3.10.1 Initial Tensile Strain: The initial tensile strain is a recoverable tensile strain determined after 50 to 200 load cycles (ASTM: D-4123). This is an indicator of the performance of a bituminous mix under repeated load. The initial tensile strains of PMBC1, PMBC2 and BC mixes were calculated using Equn.3.

Indirect tensile fatigue results obtained for PMBC1, PMBC2 and BC mixes prepared with modified and standard Marshall compaction are presented in Table 10.

From the results it is observed that, deformation rate of PMBC1, PMBC2 and BC mixes compacted with modified Marshall hammer are nearly 2.46, 1.43 and 1.68 times lower than the corresponding mixes compacted with standard Marshall hammer hence this leads to development of lower tensile strain in mixes compacted with modified Marshall hammer and hence leads to higher fatigue life.

From the fatigue test results it is observed that for PMBC1, PMBC2 and BC mixes, the modified compaction offers higher fatigue lives in the order of 42 per cent, 22 per cent and 44 per cent respectively when compared to mixes prepared with standard compaction method. It is also seen that higher resilient modulus and lower initial tensile strains are induced for modified compaction when compared to standard compaction for all the three mixes.

3.11 Deformation Rate

The deformation rate (mm/cycle) is defined as the magnitude of the horizontal plastic deformation per load cycle. This represents the rate at which damage (permanent strain) is induced into the sample. This permanent strain is ultimately responsible for initiating fatigue cracks. It should be noticed that this deformation is caused by the tensile

From the results it is observed that, deformation rate of PMBC1, PMBC2 and BC mixes compacted with modified Marshall hammer are nearly 2.46, 1.43 and 1.68 times lower than the corresponding mixes compacted with standard Marshall hammer hence this leads to development of lower tensile strain in mixes compacted with modified Marshall hammer and hence leads to higher fatigue life.

3.12 Repeated Indirect Tensile Tests

Repeated indirect tensile test was also found (Kennedy 1978; Qi and Witczak 1998; Palit et al 2004) to be a convenient test for determining the permanent deformation characteristics of paving mixtures under repeated loading condition though the effect of aggregate quality is not fully reflected due to lack of confinement in this test. Repeated indirect tensile tests were conducted in this investigation to evaluate the permanent deformation characteristics of neat and polymer modified bituminous concrete mixtures compacted using stress that is induced into the sample, and it simulates the tensile strain developed at the bottom of the asphalt layer of the pavement due to traffic loading. Cumulative plastic tensile strain ultimately causes fatigue cracks when it exceeds the tensile capacity of the material.

Horizontal deformation values obtained from repeated load indirect tensile test conducted on PMBC1, PMBC2 and BC mixes compacted with modified and standard Marshall hammer were used to determine deformation rate in mm/cycle. These results are shown in Table 11.

standard as well as modified Marshall hammer under repeated loading condition at 30°C temperature. A repeated load of 1.80 kN amplitude was applied on the specimen. Loading time was maintained as 0.25 s and rest period as 0.45 s. Vertical compressive strain values at the end of different loading and unloading cycles were computed from the measured vertical deformation and the Poisson’s ratio (assumed as per TRL for 30°C temperature) using the Equation 4 as per Kennedy 1978.

As can be seen from the figure that, the mixes compacted with modified Marshall hammer displayed slower buildup of irrecoverable deformation when compared to mixes prepared with standard Marshall method. It is only the early part of the curves that is relevant to describe the deformation behavior of road mixes since the deformations in the latter part are large due to tensile cracks (Palit et al 2004). In order to compare the deformation behaviour of different types of mixes, the accumulated compressive strain at an arbitrary level of 1000 load repetitions were compared from Fig. 6. It is observed that at 1000 load repetitions the strains in PMBC1, PMBC2 and BC mixes compacted with modified Marshall hammer are 7920, 7100 and 9979 micro strains. Whereas, the above values for mixes compacted with standard Marshall hammer are 9210, 10,500 and 16,260 micro strains, respectively. From this it can be inferred that modified Marshall compaction bring lower amount of permanent deformation because of better re-orientation of aggregates due to kneading action as well as higher densities attained during compaction.

4. CONCLUSIONS
1. The modified Marshall compaction results in 11 per cent, 3 per cent and 6 per cent higher stability than standard compaction for all three mixes viz PMBC1, PMBC2 and BC. Higher flow values are recorded for modified compaction than standard compaction in the cases of both neat and modified mixes. Higher unit weights are achieved for PMBC as well as BC mixes compacted by modified Marshall compaction when compared to standard Marshall compaction.

2. In modified compaction, air voids are low as compared to standard compaction at particular binder content for neat as well as modified mixes. This shows that modified compaction imparts better kneading action to bring proper orientation of aggregate particles.

3. The optimum binder content is reduced by 8 per cent, 9 per cent and 1.50 per cent for PMBC1, PMBC2 and BC mixes compacted by modified Marshall compaction when compared to standard Marshall compaction. This reduction brings substantial economy in PMBC mixes and a marginal economy in BC mixes.

4. The indirect tensile strength values for PMBC1, PMBC2 and BC specimens prepared using modified Marshall compaction are found to be higher in the order of 11per cent, 6 per cent, 6 per cent, 6 per cent, 8 per cent and 6 per cent; 4 per cent, 11per cent, 17 per cent, 18 per cent, 7 per cent and 15 per cent; 18 per cent, 9 per cent, 24 per cent, 22 per cent and 12 per cent at test temperatures of 15,20,25,30,35 and 40OC, respectively when compared to above mixes prepared with standard Marshall compaction.

5. Tensile Strength ratio (TSR) values are higher for all the mixes compacted with modified Marshall method when compared to standard Marshall method.

6. Resilient Modulus Ratio (MRR) values are higher for all the mixes compacted with modified Marshall method when compared to standard Marshall method.

7. The moisture usceptibility reduces by using modified Marshall compaction when compared to standard Marshall compaction due to improved compactive effort, which results in lower moisture penetration and that leads to higher TSR and MRR.

8. The modified compaction offers higher fatigue lives in the order of 42 per cent, 22 per cent and 44 per cent for PMBC1, PMBC2 and BC mixes respectively when compared to mixes, prepared with standard compaction method.

9. Higher resilient modulus values and lower initial tensile strains are induced for the mixes compacted with modified compaction when compared to standard compaction for all the three mixes.

10. Deformation rate of PMBC1, PMBC2 and BC mixes compacted with modified Marshall hammer are nearly, 2.46, 1.43 and 1.68 times lower than the same mixes compacted with standard Marshall hammer.

11. Modified Marshall compaction lowers the amount of permanent deformation in the order of 16 per cent, 48 per cent and 63 per cent for PMBC1, PMBC2 and BC mixes, respectively, because of better re-orientation of aggregates due to kneading action as well as higher densities are attained during compaction.

ACKNOWLEDGEMENTS

The authors wish to acknowledge with thanks M/s Hindustan Colas Limited, Tamil Nadu, for supplying polymer-modified bitumen for laboratory experiments to Bangalore University.

REFERENCES

1. ASTM: D-4123-1995, “Standard Test Method for Indirect Tension Test for Resilient Modulus of Bituminous Mixtures.” American Society for Testing and Materials. Vol. 04. (03, 1995).

2. ASTM: D-1559-82, “Resistance to Plastic Flow of Bituminous Mixtures Using Marshall Apparatus.” American Society for Testing and Materials. Vol. 04. (03, 1988)

3. Hensley, J.,‘‘Establishing Hot Mix Asphalt Mixing and Compaction Temperatures at the Project Level.’’ Asphalt, 12(2), 19–23, 1998.

4. IRC: SP: 53 “Tentative Guidelines on Use of Polymer and Rubber Modified Bitumen in Road Construction” Indian Road Congress, New Delhi 2002.

5. IS: 73. “Paving Bitumen- Specification (Second Revision).” Bureau of Indian Standards, New Delhi, Feb. 1992.

6. John Maddison Read “Fatigue Cracking of Bituminous Paving Mixtures.” Ph.D. Thesis, Unpublished, Dept. of Civil Engineering., University of Nottingham, May 1996.

7. Kennedy, T. W., ‘‘Practical Use of The Indirect Tensile Test for the Characterization of Pavement Material.’’ Proceedings of Australian Road Research Board, 1978, 9 (3), 36–45.

8. Kohli, M. Singh., Vashi, B.D., Damodriya, S.M. and Bhatt, R.D., “Case Study for Use of Hugo Hammer Method in Design of Dense Bituminous Macadam.” Proceedings of National Conference on Modern Cement Concrete and Bituminous Roads. College of Engineering, GITAM., Visakhapatnam, pp 58-64, Dec. 2003.

9. Mohammad, L. N., and Paul, H. R., ‘‘Evaluation of Indirect Tensile Test for Determining Structural Properties of Asphalt Mix.’’ Transportation Research Record 1417, 1993, Transportation Research Board, Washington, D.C., 58–63.

10. Palit, S.K., Sudhakar Reddy, K. and Pandey, B.B., “Evaluation of Bituminous Mixtures Using Modified Marshall Compaction Technique.” Highway Research Bulletin, No. 65, Indian Roads Congress, Dec. 2001

11. Palit S. K., Sudhakar Reddy, K and Pandey, B. B., “Laboratory Evaluation of Crumb Rubber Modified Asphalt Mixes” Journal of Materials in Civil Engineering, ASCE Vol. 16, No. 1, February 1, 2004.

12. Pandey, B.B., “Bituminous Mix Design.” Journal of New Building Materials and Construction World, Vol.8, No.5, pp 57-63, Nov., 2002.

13. Qi, X., and Witczak, M. W. (1998). ‘‘Time-Dependent Permanent Deformation Models for Asphaltic Mixtures.’’ Transportation Research Record 1639, Transportation Research Board, Washington, D.C., 83–93.

14. Rust, F.C., Grobler, J.E., Myburg, P.A., and Hugo, E., “Towards Analytical Mix Design for Large Stone Asphalt Mixes.” Proceedings of the 7th International Conference on Asphalt Pavements, Vol. 3, pp 311-325, 1992.

15. SABITA, “Large Aggregate Mixes for Bases (LAMBS), Manual 13, South Africa 1993.

16. Salter, R.J. and Rafati-Afshar, F., “Effect of Additives on Bituminous Highway Pavement Materials Evaluated by the Indirect Tensile Test”. Transportation Research Record 115, 1987, pp 183-195.

17. Specifications for Road and Bridge Works, Ministry of Road Transport and Highways (MoRT&H). Govt of India, Fourth edition, 2001.