Effect of Hydrated Lime on Dynamic Modulus of Asphalt-Aggregate ...

KSCE Journal of Civil Engineering (2010) 14(6):829-837 DOI 10.1007/s12205-010-0944-4

Highway Engineering

www.springer.com/12205

Effect of Hydrated Lime on Dynamic Modulus of Asphalt-Aggregate Mixtures in the State of North Carolina Sangyum Lee*, Youngguk Seo**, and Y. Richard Kim*** Received August 3, 2009/Revised December 4, 2009/Accepted February 9, 2010

···································································································································································································································

Abstract A comprehensive experimental study has been conducted primarily to characterize the effect of hydrated lime on a dynamic modulus of Hot Mix Asphalt (HMA). In this paper, five Superpave HMAs typically used in the State of North Carolina were modified with hydrated lime and evaluated with a series of uniaxial compressive complex modulus tests. And a new design method was proposed for lime modified HMAs based on their volumetric optimums. For a 9.5 mm surface mixture, three levels of lime addition methods were tested and their impacts on a dynamic modulus and phase angle were statistically analyzed. Finally, a resistance of lime modified HMA to moisture damage was demonstrated by comparing with unmodified HMA. Findings suggested that the inclusion of hydrated lime should make HMAs stiff especially at high frequencies and lower temperatures. The reduction of dynamic modulus due to moisture was smaller in lime modified HMAs than in un-modified HMA. Keywords: hot mix asphalt, hydrated lime, dynamic modulus, superpave ···································································································································································································································

1. Introduction It has been shown that hydrated lime does increase the stiffness of Hot Mix Asphalts (HMAs). When added to HMAs, hydrated lime produces the beneficial effects of filler as it also improves the resistance to moisture damage. Even with the great deal of literature and experience extolling the benefits of hydrated lime, most of research efforts devoted so far have been confined either to examining engineering properties, such as the resilient modulus, or to nonlinear regions of the material behavior (MeCann et al., 2003; Mohammad et al., 2000, 2006; Kim et al., 2003). However, little research exists regarding the impact of hydrated lime on the fundamental behavior and performance of asphalt concrete mixtures (Little et al., 2005; Petersen et al., 1987a, 1987b; Sebaaly et al., 2003; Anderson et al., 1973). Dynamic modulus of HMAs is a fundamental material property that relates stresses to strains induced under different time and temperature conditions. Many studies have dealt with dynamic modulus of HMAs (Shook et al., 1969; Kallas, 1970; Bohn et al., 1970; Cragg et al., 1971; Yeager et al., 1975; Majidzadeh et al., 1979; Akhter et al., 1985; Witczak et al., 1996), but its practical use in analysis and design of HMAs by state highway agencies was quite limited. Since the advent of AASHTO 2002 pavement design guide, public awareness and interest in dynamic modulus has increased because mechanistic-empirical analysis

procedures require dynamic modulus to determine the responses of pavement structures and thus, to predict the performance of asphalt concrete pavements (Seo et al., 2007). NCHRP 9-19 (Superpave Support and Performance Models Management) suggested that dynamic modulus should be the best indicators for distresses in asphalt concrete pavements (Pellinen, 2001). Therefore, complex modulus test for dynamic modulus calculation has been considered as simple performance test. The fact that the same material property can be used in both pavement response and performance prediction implies a significant advantage to state highway agencies because only one equipment and test method is necessary for both pavement thickness design and asphalt mixture design. Also, it will allow both thickness design and mixture design to be performed on the same theoretical framework, which has been regarded as one of the major goals in developing these new design methods. Therefore, states began to establish data base for the dynamic modulus. The primary objectives of this research are: (a) to understand the effect of lime modification on a dynamic modulus of HMAs typically used for flexible pavements in the State of North Carolina; (b) to investigate the impact of lime addition methods on the characteristics of both dynamic modulus and phase angle; and (c) to demonstrate the resistance of lime modified HMA to moisture damage.

*Member, Senior Researcher, Road Management Division, Seoul Metropolitan Government, Seoul 100-744, Korea (E-mail: [email protected]) **Member, Senior Researcher, Test Road Center, Korea Expressway Corporation, Hwaseong, Kyonggi 445-812, Korea (Corresponding Author, E-mail: seoyg89@ hotmail.com) ***Professor, North Carolina State University, Raleigh, NC 27696-7908, USA (E-mail: [email protected]) − 829 −

Sangyum Lee, Youngguk Seo, and Y. Richard Kim

2. Time-temperature Superposition

Table 1. Aggregate Gradation of Selected Mixtures Mix Types

For dynamic modulus characterization of HMAs, it is important to understand the principle of time-temperature superposition (Chehab et al., 2002; Kim et al., 2004). Simply stated, the same modulus value of a material can be obtained both at low test temperatures and long times (slow frequencies) or at high test temperatures but short times (fast frequencies). In general, the behavior of a material at high temperatures is the same as that under long loading times or slow loading rates/frequencies, and the material behavior at low temperatures is the same as that under short loading times or fast loading rates/frequencies. Materials that exhibit this type of behavior are called thermorheologically simple. The time-temperature superposition of a material can be checked by performing complex modulus tests at varying temperatures and frequencies, as is done in the AASHTO TP62 protocol (AASHTO, 2003). The dynamic modulus increases as the loading frequency increases and the temperature decreases. The simplifying feature afforded by time-temperature superposition is that all of these curves can be superposed to form a single continuous curve by means of horizontal translations only. The amount of horizontal shift varies by temperature and is quantified with the time-temperature shift factor. After the horizontal shift, the frequency at the reference temperature is called reduced frequency.

Fine, Coarse, Fine, Fine, Fine, S9.5A S9.5C S12.5C I19.0B B25.0B

#5

20.0

#67

41.4

22

#78

36.0

17.0

55.0

12.0

38.0

30.0

Dry screenings

0.0

33.0

1.0

28.0

22.0

Washed screenings

83.0

10.0

0.5

Sand/Baghouse fines Sum

2.0 100.0

100.0

Sieves (mm)

3. Materials and Specimen Fabrication 3.1 Mixture Selection In order to understand the effect of lime modification on a dynamic modulus, five HMAs typically used by the North Carolina Department of Transportation (NCDOT) were selected and the volumetric mixture design was performed. These same mixtures were then modified with hydrated lime and the optimum asphalt content was reevaluated. These modified and unmodified mixtures were then tested for the dynamic modulus to evaluate the effect of hydrated lime.

% Stock Pile

12.0 12.0

100.0

100.0

100.0

% Passing Blend

37.5

100

100

100.0

100

100

25.0

100

100

99.2

100

98

19.0

100

100

95.3

99

82

12.5

100

100

77.6

88

66

9.5

99

97

59.2

76

63

4.75

86

62

45.1

52

42

2.36

63

40

32.5

38

30

1.18

44

34

19.9

2

23

0.600

32

25

12.2

21

16

0.300

21

14

6.3

13

10

0.150

12

8

6.3

6

6

0.075

5.2

5.6

6.3

3.8

4

The NCDOT has categorized Superpave mixtures used in North Carolina based on the layer location, aggregate gradation, and traffic volume. In this study, all of S9.5A, S9.5C, S12.5C, I19.0B, and B25.0B mixture types have been evaluated. In each case two mix designs are created, modified and unmodified, for a total of 10 different mixture types. These mixes are the most frequently used Superpave mixes in North Carolina according to the NCDOT Pavement Management Unit. This selection in-

Table 2. Material Information of Selected Mixtures Mix Design

Fine, S9.5A

Fine, S12.5C

Fine, I19.0B

Aggregate Type

Fine, B25.0B

Coarse, S9.5C

Granite

Aggregate Source

Pineville/Charlotte

Concord/Cabarrus

Garner

Garner

Holly Springs

Aggregate Specific Gravity

2.932

2.757

2.633

2.639

2.652

Binder Grade

PG 64-22

PG 70-22

PG 64-22

PG 64-22

PG 70-22

0.5% of Binder

0.5% of Binder

0.5% of Binder

0.25% of Binder

0.5% of Binder

Binder Source Anti-strip Additive %

Citgo Wilmington

Anti-Strip Supplier

Arr-Maz (Ad-Here 6500 LOF)

Chemical Lime %

1% of mix

Lime Specific Gravity

2.300

Nini/Ndes/Nmax

6/50/75

8/100/160 − 830 −

7/75/115

7/75/115

8/100/160

KSCE Journal of Civil Engineering

Effect of Hydrated Lime on Dynamic Modulus of Asphalt-Aggregate Mixtures in the State of North Carolina

cludes three mixtures for surface layer (S), one for intermediate layer (I), and one for base layer (B). Aggregate gradations of the mixes are presented in Table 1. In addition to the mix types shown in Table 1, factors that are important, but not explicitly shown, are aggregate type and specifications of materials. These factors are presented in Table 2. Only granite aggregates are considered because they constitute approximately 70% of all pavement construction in North Carolina. To investigate the lime addition methods and their effect on a dynamic modulus and phase angle, three lime addition methods (two dry methods and one wet method) were tested based on S9.5C mixture type. Between two dry methods, one being modified with 1.0% hydrated lime without any modification to the aggregate structure (Leco), and the other being modified with hydrated lime that is substituted for a portion of the baghouse fines (Lsub). In wet method, hydrated lime is mixed with wet aggregate at a moisture content of 2-3% over Saturated-SurfaceDry (SSD) conditions (Lmod). Granite aggregate from the Pineville/Charlotte quarry, Reclaimed Asphalt Pavement (RAP) and PG 58-28 binder obtained from the Citgo refinery in Salisbury, North Carolina, were used to produce all the mixtures denoted as RS9.5C. It should be noted that the presence of RAP in the mixture changes the true PG of the asphalt binder in the mixture. According to the NCDOT, the true PG for this mixture is PG 64-22. The aggregate gradation of each mixture is presented in Table 3. It is seen that, in general, the coarse aggregate Table 3. Gradation Information for the Lime Modification Methods % Stock Pile

Mix Types Control

Leco

Lsub

Lmod

78-M Stone

20.0

19.8

20.0

22.5

Washed screens

41.4

41.0

41.4

32.5

Asphalt sand

12.0

11.9

12.0

17.5

Lime

0.0

1.0

1.0

1.0

Baghouse fines

1.5

1.5

0.5

1.4

RAP

25.1

24.9

25.1

25.1

Sum

100.0

100.0

100.0

100.0

Sieves (mm) 19.0

% Passing Blend 100.0

100.0

100.0

100.0

12.5

99.2

99.2

99.2

99.2

9.5

95.3

95.4

95.3

95.1

4.75

77.6

77.8

77.6

76.2

2.36

59.2

59.6

59.2

58.7

1.18

45.1

45.7

45.1

46.0

0.600

32.5

33.1

32.5

32.8

0.300

19.9

20.7

19.9

18.8

0.150

12.2

13.0

12.2

11.6

0.075

6.3

7.1

6.3

6.5

Vol. 14, No. 6 / November 2010

gradation does not change across the mixture types. Stockpile percentages of mixtures could be found elsewhere (Kim et al., 2007). 3.2 Specimen Fabrication All specimens were compacted by the Superpave gyratory compactor, to a height of 178 mm and a diameter of 150 mm. To obtain specimens of uniform quality and air void distribution, these samples were cored and cut to a height of 150 mm and a diameter of 100 mm for testing. After obtaining specimens of the appropriate dimensions, air void measurements were taken via the core-lock method, and specimens were stored until testing. During storage, specimens were sealed in bags and placed in an unlit cabinet to reduce aging effects. In addition, no test specimens were stored for longer than two weeks before testing.

4. Experimental Program 4.1 Test Setup and Complex Modulus Tests The closed-loop servo-hydraulic testing machine was a UTM25 machine manufactured by IPC Global of Australia. This machine was equipped with a 25 kN load cell. The environmental chamber for the UTM-25 was refrigerator driven and also utilized a feedback system to maintain consistent temperature during the testing. Measurements of axial deformations, load and crosshead movement were taken for all tests. In both machines the data acquisition system was identical and consisted of a National Instruments 16-bit data acquisition card and LabView software. Axial measurements were taken at 90o intervals over the middle 100 mm of the specimen with loose-core LVDTs from IPC Global. To obtain dynamic modulus, a series of complex modulus tests was conducted in load-controlled mode in axial compression (zero maximum stress) mode following the protocol given in AASHTO TP62-03 (AASHTO, 2003). Tests were completed for all mixtures in this study at -10o, 10o, 35o and 54.4oC and at frequencies of 25, 10, 5, 1, 0.5, 0.1, 0.05 and 0.01 Hz. Load levels were determined by a trial and error process so that the resulting strain amplitudes were between 50 and 70 micro strains to ensure an accurate viscoelastic characterization. 4.2 Moisture Conditioning The purpose of the moisture conditioning is to introduce a certain amount of moisture damage in the specimen prior to dynamic modulus characterization testing. Quing et al. (2007) performed the sensitivity test to determine the moisture conditioning process using various combinations of parameters. As a result, three parameters were determined for the preconditioning process (i.e., the moisture saturation before conditioning): moisture content (or saturation level), conditioning temperature, and conditioning duration. Of these three parameters, the conditioning temperature had the most significant effect on the moisture resistance of HMAs, followed by the conditioning duration and the moisture content.

− 831 −


Overall multiple freeze-thaw (F-T) cycling conditioning was conducted in accordance with AASHTO T-283. Because dynamic modulus characterization of lime modified HMAs suggested in this study is different from those specified in the AASHTO T283, the conditioning procedure was adjusted for appropriate saturation levels and post-conditioning temperatures. It is generally believed that the higher the air voids, the higher the permeability and, thus, the more water can penetrate into an asphalt pavement. That is, the permeability of asphalt mixtures is proportional to air void content when the properties and the structure of the aggregate are similar. In this study, all the mixtures consist of the same aggregate and structure, and have lower air voids (4%) than required for tensile strength ratio testing requires (7%). Due to the low air voids, it took three times longer than the maximum of 10 minutes specified in the AASHTO T283. Then, the saturated specimens (preconditioned specimens) were subjected to three multiple F-T cycles that consisted of freezing at -17.7oC (0oF) for 24 hours followed by 24 hours thawing at 60oC (140oF). The conditioned specimens were subjected to stabilization for two hours at 26oC (77oF) in a water bath. This procedure was considered to be effective in minimizing the creep by the weight of the specimens. In order to minimize the effects of saturation, all conditioned specimens were dried to a certain level of saturation by a Core Dryer instrument. Because Table 4. Level of Saturation for Moisture Conditioning Saturation Level (%) Mix Temp. o ( C) Stage

-17.7, 60, 26

Control

Lsub

PrePostPrePostconditioning conditioning conditioning conditioning

1

56.0

33.3

56.7

38.4

2

60.2

34.8

57.6

38.0

3

58.3

37.3

54.2

37.9

Avg.

58.2

35.1

56.2

38.1

the setup for dynamic modulus test takes 18 hours, the specimens were wrapped with parafilm to avoid the evaporation of the moisture until the test setup was completed. The test saturation level (post-conditioning) was decided based on the testing setup time. The saturation levels for pre-conditioning and post-conditioning are shown in Table 4 Level of saturation for moisture conditioning.

5. Implementation of Mixture Designs 5.1 Mixture Design for Five Lime-modified HMAs The NCDOT does not currently require the use of hydrated lime in asphalt concrete mixtures. So, for this study lime modified mix designs were developed based on the standard NCDOT mixtures. These mix designs were performed following the standard Superpave procedure. There are several proven and effective methods for adding hydrated lime to asphalt. However, for economic reasons, the dry method with a lime content of 1% by weight of mixture was chosen to evaluate the effects of lime on volumetric mix design. Results from the mix design process for both unmodified and modified mixtures are shown in Table 5. Fine and coarse mixtures are denoted with the final letter of the mixture identification. It is seen from Table 5 that the optimum asphalt content for all of the mixtures modified with hydrated lime decreased 0.35% to 0.90% (saving an average of approximately 12% of the untreated optimum asphalt content). Care should be taken in utilizing hydrated lime in existing mix designs without first reassessing the asphalt content. Because hydrated lime acts as an extender in asphalt concrete, failure to readjust the asphalt content could lead to excessively rich mixtures or very poor mixtures. Rich mixtures may be more prone to permanent deformation failures or vice versa. Conversely, it is possible that the benefits of hydrated lime could offset the higher than optimum asphalt content. It should be noted that some of the VMA values for the lime modified mixtures and one of the unmodified mixtures are lower

Table 5. Results of Mix Design Analysis Design Asphalt Content Mix type

Fine, S9.5A Fine, S12.5C Fine, I19.0B Fine, B25.0B Coarse, S9.5C

Opt. AC (%)

Unmod.

5.90

Lime

5.55

Unmod.

4.70

Lime

4.10

Unmod.

4.30

Lime

3.70

Unmod.

3.90

Lime

3.50

Unmod.

4.90

Lime

4.00

AC (%) Difference 0.35 0.60 0.60 0.40 0.90

VMA Gmm

(%)

2.659

17.9

2.620

18.5

2.591

14.1

2.592

13.2

2.492

13.1

2.495

12.0

2.506

12.5

2.488

12.4

2.492

14.2

2.497

13.0

− 832 −

VFA Min. 15 14 13 12 15

(%) 79.0 80.0 71.0 70.0 69.3 69.0 67.5 68.0 72.0 66.0

Min.-Max. 70-80 65-75 65-78 65-78 65-76



Table 6. Design Results from Lime Addition Study Mix Type Fine RS9.5C

Design Asphalt Content (%) Binder Grade

Control Leco Lsub

PG58-28

Lmod

%Gmm@Nini, Nmax

VMA

Opt. AC (%)

90.5% max.

98% max.

(%)

5.60

89.4

97.0

16.2

5.15

89.4

96.8

15.8

5.30

89.5

96.1

15.7

5.60

89.8

96.1

15.9

than the minimum criteria. Under normal design situations such mixtures would be rejected. However, the aim of this study is to determine the effect of lime addition on the optimum asphalt content; therefore, the VMA violation does not adversely affect the conclusions. It is of interest to note that the lime modification method utilized in this study causes a reduction in the VMA for all mixtures, except for Fine S9.5A.

VFA Min.

(%)

Min.-Max.

75.0 15

74.0 74.0

65-80

75.8

of the original mix design and that the optimum asphalt content remains the same. Results from the volumetric mix design are summarized in Table 6. It is observed that the Superpave volumetric criteria are met for all addition methods. Also the Leco mixture shows the most drastic reduction in optimal asphalt content, followed next by the Lsub mixture.

6. Results and Discussion 5.2 Mixture Design for RS9.5C Modified with Three Addition Methods To further investigate the lime addition methods, three different lime modified mixtures (Leco, Lmod and Lsub) and one unmodified (control) mixture were developed based on RS9.5C. For each of the mixtures, Superpave volumetric mix design is performed to obtain the optimal asphalt content. For this design 1%, by aggregate weight, hydrated lime is first added to the original unmodified mixture and then the proportions of the various stockpiles in this original mix design, particularly the sand stockpile, are slightly altered. The alterations are performed to ensure that the blended gradation is the same as that

6.1 Effect of Lime Modification on Dynamic Modulus Two replicates for each of five Superpave mixtures shown in Tables 1 and 2 were tested for unmodified and lime modified mixes. Their dynamic modulus master curves are displayed in Figs. 1 through 5. The reference temperature that was used as the basis for shifting the data is 10oC. The filled symbols are used to present the dynamic modulus values of the lime modified mixes, and the unfilled symbols are used for unmodified mixes. In general, the lime modified mixtures do show higher stiffness at a higher reduced frequency and lower stiffness at a lower reduced frequency. However, this difference could be related to specimen

Fig. 1. Dynamic Modulus Comparison of S9.5AF Mixture for Unmodified (UM) and Modified (M) with Two Replicates

Fig. 2. Dynamic Modulus Comparison of S12.5CF Mixture for Unmodified (UM) and Modified (M) with Two Replicates


− 833 −


Fig. 3. Dynamic Modulus Comparison of I19.0BF Mixture for Unmodified (UM) and Modified (M) with Two Replicates

Fig. 5. Dynamic Modulus Comparison of S9.5CC Mixture for Unmodified (UM) and Modified (M) with Two Replicates

Fig. 6. Effects of Lime Modification on Dynamic Modulus in: (a) Semi-log, (b) Log-log Scale Fig. 4. Dynamic Modulus Comparison of B25.0BF Mixture for Unmodified (UM) and Modified (M) with Two Replicates

variability, as can be seen in the spread of the replicate master curves. Further, results from dynamic modulus testing are summarized for all of the mixes with the line-of-equality plots shown in Fig. 6. The data are presented in both semi-log and log-log space to examine both the low temperature (semi-log) and high temper-

ature (log-log) results. It can be observed that the lime modified mixtures tend to be stiffer at the highest modulus conditions (i.e., under the lowest test temperature conditions), but show little difference at the low modulus (high temperature) conditions. It is noted that these results somewhat conflict those presented elsewhere (Witczak et al., 2004) where it is shown that hydrated lime increases the stiffness by a constant factor under all conditions.

− 834 −



Table 7. Statistical Analysis of Effect of Lime Addition Methods on Dynamic Modulus

Fig. 7. Effect of Lime Addition Methods on |E*| in: (a) Semi-log and (b) Log-log Space, and on Phase Angle (c)

6.2 Statistical Significance of Lime Addition Methods The impact of three lime addition methods on the dynamic modulus and phase angle was summarized in Fig. 7. Visual inspection of this figure indicates that no significant effect from the lime addition method is detected in the dynamic modulus test. However, for completeness a statistical analysis has been carried out. The first step in such an analysis involves a single factor analysis of variance (ANOVA) under the null hypothesis that the |E*| value for each mixture is equivalent. This analysis is performed with the significance level a=0.05 at the various reduced frequencies. The results of this analysis are presented in Table 7. It is observed that the null hypothesis is rejected at the lowest reduced frequencies, conditions marked with darken cells. To gain further insight into the cause of rejection, the Bonferroni multiple pairwise comparison method using Least Significant Difference (a=0.05) is used to make individual mixture comparisons. The results are also summarized in Table 7. The specimen that produced the highest value of dynamic modulus Vol. 14, No. 6 / November 2010

Reduced Frequency (Hz)

p-value

2.40E+04 1.40E+03

Least Significant Difference Ranking Control

Lsub

Leco

Lmod

0.3182

A

A

A

A

0.2315

A

A

A

A

2.00E+00

0.2330

A

A

A

A

3.00E-02

0.0053

B

A

A-B

B

1.00E-04

0.0027

B

A

A-B

B

1.00E-05

0.0000

C

A

B

C

1.30E-07

0.0092

B

A

A

A

(|E*|) at a given frequency was ranked “A.” From this point, the E* value for each of the other specimens was compared to this value to determine if they were significantly lower. If the specimen with the second highest value of E* is found to be significantly less it will be ranked “B”, if not, it also would be ranked “A.” The process is continued for the third and forth specimens for each of the frequencies to produce the results shown in Table 7. It is seen from this analysis that in general, at conditions rejected by the ANOVA analysis, that the Lsub, Leco, and to a lesser extent Lmod mixtures show a consistently higher modulus than the Control mixture. It is somewhat surprising to see these differences in the modulus, particularly considering that the results indicate no statistical difference in the data at the lowest reduced frequency. However, since the variability in this study was small, then it is possible that the higher data quality makes comparisons at low reduced frequencies clear. Regardless of the outcome of the statistical analysis, it is clear from Fig. 7 that the lime does not increase the stiffness of the material by a great amount. This mixture contains aggregate from the same source as the S9.5CF mixture used as the basis for the Control, Lmod, Leco, and Lsub mixtures. It is unclear from these limited results if similar results regarding the influence of lime on dynamic modulus would hold for other types of mixtures using this same aggregate source, however, the results do support the claim that the specific physico-chemical properties of the constituent materials strongly affect the impact of lime on the material behavior. 6.3 Effect of Lime Modification on Moisture-conditioned HMAs Control and Lsub mixtures at the optimum asphalt contents were moisture conditioned using the procedure described in the previous section. The moisture-conditioned specimens were then subjected to the dynamic modulus test used for the dry specimens. Test results are shown in Figs. 8 and 9 for the Control and Lsub mixtures, respectively. In general, moisture conditioning lowers the dynamic modulus, whereas the phase angle remains relatively the same. The reduction of the dynamic modulus due to moisture conditioning is smaller in the Lsub mixture than in

− 835 −


Fig. 8. Dynamic Modulus Test Results for Control (C) and Moisture-conditioned Control (CM) Mixtures: (a) Dynamic Modulus in Semi-log Scale, (b) Dynamic Modulus in Log-log Scale, (c) Phase Angle

Fig. 9. Dynamic Modulus Test Results for Control (C) and Moisture-conditioned Control (CM) Mixtures: (a) Dynamic Modulus in Semi-log Scale, (b) Dynamic Modulus in Log-log Scale, (c) Phase Angle

the Control mixture.

(one being modified with hydrated lime that is substituted for a portion of the baghouse fines). • Lime modification can cause a reduction in the VMA for most of HMAs except for fine 9.5 mm surface mixture • It was found that lime modification could be effective in reducing moisture damage. Also, the reduction of the dynamic modulus due to moisture conditioning is smaller in the Lsub mixture than in the Control mixture. The differences in optimum asphalt contents can be significant with only moderate changes to the dynamic modulus. However, dynamic modulus is only a single performance parameter which, based upon recent studies, may not be the best predictor of rutting and fatigue cracking. Additional experiments in further study are intended to provide a more comprehensive look at mix design and mixture performance. For instance, the addition of lime may extend the binder without adverse effects to the mix and still provide significant moisture sensitivity mitigation, crack

7. Conclusions New mix designs for Superpave HMAs commonly used in North Carolina have been developed with lime modification. Characterization of these mixtures was investigated with the newly developed dynamic modulus test protocol. Based on the results, the following conclusions were drawn: • An additional 1% lime by weight of mixture reduces the optimum asphalt content by 0.35 to 0.9%. • Lime modification of HMAs increased the dynamic modulus, especially at high frequencies and lower temperatures. • Among three lime addition methods, Leco mixture (one being modified with 1.0% hydrated lime without any modification to the aggregate structure) revealed the most dramatic decrease in optimal asphalt content, followed next by the Lsub mixture

− 836 −



pinning, and other performance improvements.

References Akhter, G. F. and Witczak, M. W. (1985). “Sensitivity of flexible pavement performance to bituminous mix properties.” Transportation Research Record 1034, Transportation Research Board, National Research Council, Washington, D.C., pp. 70-79 American Association of State Highway and Transportation. Officials (2003). “TP-62 standard method of test for determining dynamic modulus of hot-mix asphalt concrete mixtures.” Washington, D.C. Anderson, D. A. and Goetz, W. H. (1973). “Mechanical behavior and reinforcement of mineral filler-asphalt mixtures.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 42. pp. 3766. Bohn, A. O., Ullidtz, P., and Stubstad, R. (1970). “The dynamic modulus of asphalt concrete surfaces.” Dansk Vejtidsskrift, Vol. 47. Chehab, G., Kim, Y. R., Schapery, R. A., Witczack, M., and Bonaquist, R. (2002). “Time-temperature superposition principle for asphalt concrete mixtures with growing damage in tension state.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 71, pp. 559-593. Cragg, R. and Pell, P. S. (1971). “The dynamic stiffness of bituminous road materials.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 40, pp. 126-147. Kallas, B. F. (1970). “Dynamic modulus of asphalt concrete in tension and tension-compression.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 39, pp. 1-23. Kim, Y. R., Little, D. N., and Song, I. (2003). Mechanistic evaluation of mineral fillers on fatigue resistance and fundamental material characteristics. Proc., 82nd Transportation Research Board Meeting, Transportation Research Board, Washington, D.C. Kim, Y. R. and Muthady N. (2007). Implementation plan for the new mechanistic-empirical pavement design guide, Final Report to the North Carolina Department of Transportation, Report No. FHWA/ NC/2006-23. Kim, Y. R., Seo, Y., King, M., and Momen, M. (2004). “Dynamic modulus testing of asphalt concrete indirect tension mode.” Transportation Research Record 1891, in Transportation Research Board, National Research Council, Washington, D.C., pp. 163-173. Little, D. N. and Petersen, J. C. (2005). “Unique effects of hydrated lime filler on the performance-related properties of asphalt cements: physical and chemical interactions revisited.” Journal of Materials in Civil Engineering. ASCE. March/April 2005. pp. 207-218. Majidzadeh, K. S., Khedr, S., and El-Mojarrush, M. (1979). “Evaluation of permanent deformation in asphalt concrete pavements.” Transportation Research Record 715, Transportation Research Board, National Research Council, Washington, D.C., pp. 21-31. McCann, M. and Sebaaly, P. E. (2003) “Evaluation of moisture sensitivity and performance of lime in hot-mix asphalt: Resilient modulus,


tensile strength and simple shear tests.” Transportation Research Record 1832. Transportation Research Board, National Research Council, Washington, D.C., pp. 9-16. Mohammad, L. N., Abadie, C. D., Dranda, C., Wu, Z., and Zhongie, Z. (2006). “Permanent deformation analysis of hot-mix asphalt mixtures using simple performance tests and 2002 mechanisticempirical pavement design software.” Proc., 85th Transportation Research Board Meeting, Transportation Research Board, Washington, D.C. Mohammad, L., Abadie, C., Gokmen, R., and Puppala, A. (2000). “Mechanistic evaluation of hydrated lime in hot-mix asphalt mixtures.” Transportation Research Record 1723, Transportation Research Board, National Research Council, Washington, D.C., pp 26-36. Pellinen, T. K. (2001). “Investigation of the use of dynamic modulus as an indicator of hot-mix asphalt performance.” PhD Dissertation, Department of Civil and Environmental Engineering, Arizona State University. Petersen, J. C., Plancher, H., and Harnsberger, P. M. (1987a). Lime treatment of asphalt, Final Report Prepared for the National lime Association, Western Research Institute, Laramie, Wyoming. Petersen, J. C., Plancher, H., and Harnsberger, P. M. (1987b). “Lime treatment of asphalt to reduce age hardening and improve flow properties.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 56, pp. 632-653. Quing, L. and Harvey, J. (2007). “Inclusion of moisture effect in fatigue test for asphalt pavement.” Proc., 86th Transportation Research Board Meeting, Transportation Research Board, Washington, D.C. Sebaaly, P. E., Hitti, E., and Weitzel, D. (2003). “Effectiveness of lime in hot-mix asphalt pavements.” Transportation Research Record 1832. Transportation Research Board, National Research Council, Washington, D.C., pp. 34-41. Seo, Y. and Kim, Y. R. (2007). “Using acoustic emission to monitor fatigue damage and healing in asphalt concrete.” Journal of Civil Engineering, KSCE, Vol. 26, No. 4, pp. 237-243. Shook, J. F., Kallas, B. F., and McLeod, N. W. (1969). “Factors influencing dynamic modulus of asphalt concrete.” Journal of Association of Asphalt Paving Technologists, AAPT, Vol. 38, pp. 140-178. Witczak, M. W. and Bari, J. (2004). “Development of a master curve (E*) database for lime modified asphalt mixtures.” Report of National Lime Association. Witczak, M. W. and Fonseca, O. A. (1996). “Revised predictive model for dynamic (complex) modulus of asphalt mixtures.” Transportation Research Record 1540, Transportation Research Board, National Research Council, Washington, D.C., pp. 15-23. Yeager, L. L. and Wood, L. E. (1975). “Recommended procedure for determining the dynamic modulus of asphalt mixtures.” Transportation Research Record 549, Transportation Research Board, National Research Council, Washington, D.C., pp. 26-38.

− 837 −