1. Report No.

FHWA/TX-91+468-2

4. Title and Subtitle

2. Government Accession No.

EVALUATION OF METHODS OF DETERMINING THE THEORETICAL MAXIMUM SPECIFIC GRAVITY OF ASPHALT CONCRETE PAVING MIXTURES

Technical Report Documentation Page 3. Recipients Catalog No.

5. Report Date April 1989 6. Performing Organization Code

1-------------------------------1 8. Performing Organization Report No. 7. Author(s)

Research Report 468-2 Mansour Solaimanian and Thomas W. Kennedy

9. Performing Organization Name and Address

Center for Transportation Research The University of Texas at Austin Austin, Texas 78712-1075

10. Work Unit No. (TRAIS}

11. Contract or Grant No. Research Study 3-9-85/8-468

1-------------------------------1 13. Type of Report and Period Covered 12. Sponsoring Agency Name and Address Interim Texas State Department of Highways and Public

Transportation; Transportation Planning Division P. 0. Box 5051 Austin, Texas 78763-5051 15. Supplementory Notes

14. Sponsoring Agency Code

Study conducted in cooperation with the U. S. Department of Transportation, Federal Highway Administration.

Research S tudv Title: "Field Evaluation to Obtain Densitv in Asohal t Mixtures 11

16. Abstract

A comparative study was performed to compare the maximum theoretical specific gravities of asphalt-aggregate paving mixtures obtained using several methods. The study included experimental work and analysis of the resulting data. The agreement between results obtained using the Texas C-14 method and the Rice method were excellent. Results obtained by backcalculat theoretical maximum densities from a single Rice test were also found to be satisfactory. Theoretical approach based on bulk specific gravity of aggregate is not recommended because of s nificantly low theoretical maximum specific gravities and high relative densities.

The C-14 method results in lower des asphalt content than the Rice method if the latter is not corrected for water absorption. This case occurs if the design asphalt content is selected at 97% relative density for both methods. However, the design asphalt contents from the two methods will be much closer if the asphalt content is selected at 97% relative density based on the C-14 method, and 96% based on the Rice method.

17. Key Words

specific gravities, maximum, theoretical, asphalt-aggregate paving mixtures, densities, design asphalt content, water absorption

18. Distribution Statement

No restrictions. This document is available to the public through the National Technical Information Service, Springfield, Virginia 22161.

19 Security Classif. (of this report)

Unclassified

20. Security Classif. (of this page)

Unclassified

21. No. of Pages

44

22. Price

Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

EVALUATION OF METHODS OF DETERMINING THE THEORETICAL MAXIMUM SPECIFIC GRAVITY OF

ASPHALT CONCRETE PAVING MIXTURES

by

Mansour Solaimanian Thomas W. Kennedy

Research Report Number 468-2

Research Project 3-9-85/8-468

Field Evaluation to Obtain Density in Asphalt Mixtures

conducted for

Texas State Department of Highways and Public Transportation

in cooperation with the

U.S. Department of Transportation Federal Highway Administration

by the

CENTER FOR TRANSPORTATION RESEARCH

Bureau of Engineering Research

THE UNIVERSITY OF TEXAS AT AUSTIN

April 1989

The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily represent the official views or policies of the Federal High­way Administration. This report does not constitute a standard, specification, or regulation.

ii

PREFACE

This is the second report in a series of reports deal­ing with the findings of a research project concerned with density of asphalt mixtures. This report summarizes the findings of a study comparing methods of determining the theoretical maximum specific gravity of asphalt mix­tures.

The work required to develop this report was provided by many people. Special appreciation is extended to Messrs. James N. Anagnos and Eugene Betts for their assistance in the testing program. Also, the assistance of personnel from the various districts is acknowledged. In addition, the authors would like to express their appreciation to Messrs. Billy R. Neeley and Paul Krugler of the Texas State Department of Highways

and Public Transportation, to Peter A. Kopac of the pavement division, HNR-20, and to William E. Elmore of the Center for Transportation Research for their suggestions, encouragement, and assistance, and to other district personnel who provided data and information. Appreciation is also extended to the Center for Transportation Research staff who assisted in the preparation of the report. The support of the Federal Highway Administration, Department of Transportation, is acknowledged.

April1989

Mansour Solaimanian Thomas W. Kennedy

LIST OF REPORTS

Report No. 468-1, "Evaluation of the Troxler Model 4640 Thin Lift Nuclear Density Gauge," by Mansour Solaimanian, Richard J. Holmgreen, Jr., and Thomas W. Kennedy, is an evaluation of the Troxler Model 4640 Thin Lift Nuclear Gauge's ability to predict core densi­ties. July 1990.

Report No. 468-2, "Evaluation of Methods of Determining the Theoretical Maximum Specific Gravity of Asphalt Concrete Paving Mixtures," by Mansour Solaimanian and Thomas W. Kennedy, summarizes a study comparing methods of estimating the theoretical maximum specific gravity of asphalt mixtures. April 1989.

ABSTRACT

A comparative study was performed to compare the maximum theoretical specific gravities of asphalt-aggre­gate paving mixtures obtained using several methods. The study included experimental work and analysis of the resulting data. The agreement was excellent between re­sults obtained using the Texas C-14 method and the Rice method. Results obtained by backcalculating theoretical maximum densities from a single Rice test were also found to be satisfactory. A theoretical approach based on bulk specific gravity of aggregate is not recommended because of significantly low theoretical maximum spe­cific gravities and high relative densities.

iii

The C-14 method results in lower design asphalt content than the Rice method if the latter is not corrected for water absorption. This case occurs if the design as­phalt content is selected at 97 percent relative density for both methods. However, the design asphalt contents from the two methods will be much closer if the asphalt content is selected at 97 percent relative density based on the C-14 method and 96 percent based on the Rice method.

SUMMARY

This report summarizes the results of a comparative study of different methods for determining the theoretical maximum specific gravity of asphalt-aggregate mixtures. Six different methods are compared: (1) Rice method with results not corrected for water absorption, (2) Rice method with results corrected for water absorption, (3) SDHPT method as outlined in Construction Bulletin C-14, hereinafter designated as the C-14 method, (4) theo­retical method which uses the bulk specific gravity of ag­gregates, (5) backcalculated theoretical maximum specific gravity at different asphalt contents from one Rice test at a certain asphalt content with results not cor­rected for water absorption, and (6) backcalculated theo­retical maximum specific gravity at different asphalt con­tents from one Rice test at a certain asphalt content with

results corrected for water absorption. The effect of dif­ferent design asphalt content is also investigated.

Thirteen asphalt-aggregate mixtures from construc­tion projects were selected. Experiments were performed and the data were analyzed. Results indicated excellent agreement between the theoretical maximum densities obtained using the C-14 and Rice methods. The backcalculation method also gives promising results. The result from the theoretical approach is not reliable and generally yields a significantly higher relative density and lower calculated air voids. Design asphalt content from the C-14 method is slightly lower than that from the uncorrected Rice method. Values were found to be closer for design asphalt content when the corrected Rice method is compared with the C-14 method.

IMPLEMENTATION STATEMENT

The Rice method of determining the maximum theo­retical specific gravity of an asphalt mixture is generally accepted as the best method currently available. This study indicated that excellent agreement exists between values obtained using the Rice method and the Texas-C-14 method Nevertheless, the C-14 method generally will produce values indicating higher relative densities and lower air voids in the compacted mixture. Thus, the Rice method should definitely be used for compaction control.

In terms of mixture design, using the current Texas design procedure, which selects the asphalt content corre­sponding to 3 percent air voids, the Rice method will

iv

generally yield a higher asphalt content than the C-14 method. In order to obtain the same asphalt content, an air void content of about 4 percent should be used with the Rice method

Backcalculation of maximum specific gravities at other asphalt contents based on a measured Rice specific gravity at a given asphalt content produces acceptable re­sults. The calculation of maximum specific gravity using the specific gravity of the asphalt cement and the bulk specific gravity of the aggregate is not acceptable and definitely should not be done.

CONCLUSIONS AND RECOMMENDATIONS

The following conclusions are made based on the data analyzed in this study:

(I) The theoretical maximum specific gravities from the theoretical approach (OJ are, in general, signifi­cantly lower than those obtained from other meth­ods, and therefore yield the highest relative densi­ties.

(2) The theoretical maximum specific gravities from the wtcorrected Rice approach (Gru,) are the highest and result in the lowest relative densities compared to other methods.

(3) Relatively good agreement exists between the theo­retical maximum specific gravities from the C-14 method (Gc14) and those from the uncorrected Rice method (Gru,).

(4) The best agreement is observed between Gc14 val­ues and the corrected Rice method (Ore) values.

(5) Backcalculating theoretical maximum specific grav­ity (Gmax) values from one single Rice test is also promising, and backcalculated Gmax values com­pare well with those directly from the Rice tests.

v

(6) Design asphalt content at 97 percent relative density based on Gc14 is smaller than that at 97 percent relative density based on Gru.

(7) Relatively good agreement exists between design asphalt contents when 96 percent relative density is used based on Gru and 97 percent relative density is used based on Gct4·

(8) In this study it was found that the largest differ­ences between the specific gravities measured by the two methods generally do not exceed 2 percent. This difference causes the relative density based on the corrected Rice method to be about 1 to 2 percent less than that based on the uncorrected Rice method. The Rice method is gaining widespread acceptance both at national and at state levels.

(9) An additional consideration should be stated in the C-14 method. If saturation for a specimen is going to be considered complete following the peak, the relative density of that specimen based on the theo­retical approach must exceed 100 percent; otherwise erroneous results will occur.

TABLE OF CONTENTS

PREFACE---------·······-····-··---·---···--·-·-·-···---·---------··--·--·-·-... -·--····--···-·-···----················--···-······ iii

LIST OF REPORTS .......................... - .................................................................................................................................................................... iii

ABSTRACT ................................................................................................................................................................................................................. iii

SUN1MARY ................................................................................................... - ....................... -.............................................................................. iv

CONCLUSIONS AND RECOMMENDATIONS......................................................................................................................................... v

CHAPTER 1. INTRODUCTION·-·-----------------------------·---------······ .............................. .

CHAPTER 2. METHODS OF DETERMINING THE THEORETICAL MAXIMUM SPECIFIC GRAVITY

Specific Gravities of Aggregates··----------------------·---·----·--·-·........................................................ 2 1be Rice Medlod... ......................................................................................................................................... -....................................... 2 Texas SDHPT Method, C-14 ............................................................................................................................................................. 2

Theoretical Method .................................................... -------·----·--·-·---·-·-·-···-·---·-........................................... 3 Rice Backcalculation Method ..................................................................................................... _ _,............................................... 3

CHAPTER 3. EXPERIMENTAL PROGRAM

Laboratory Testing .................................................................................. -...................................................................................................... 4

Aggregate Specific Gravity .................... ---·-·--·----·--------·---·· ................................................ -.......................... 4

Mixture-Specimen Specific Gravities··--·--.. -------------·-·-............................................... _............................. 4 Calculations ________ , ................................................................................. - ........................................... _ .. , ____ , .................. -... 4

Analysis................................................................................................................................................................................................................. 4

CHAPTER 4. DISCUSSION OF RESULTS ............. - ................... -....................................................................................................... 5 Effective Specific Gravity............................................................................................................................................................................ 5 1beoretical Maximum Specific Gravity ................. ._............................................................................................................................ 5 Relative Densities............................................................................................................................................................................................ 5

Design Asphalt Conrent................................................................................................................................................................................. 6 Asphalt Absorption........................................................................................................................................................................................... 6

A Word of Caution Concerning the C-14 Method·-----------------·--· .............................................................. 6

CHAPTER 5. CONCLUSIONS AND RECOMMENDATIONS ....................................................................................................... 13

APPENDIX. FIGURES AND TABLES ... _ ................................................................................................................................................. 15

vi

CHAPTER 1. INTRODUCTION

Past experience and studies have shown that density is considered to be among the most important factors af­fecting engineering characteristics, service life, and per­formance of hot mix asphalt concrete pavement Proper density of the pavement is decided based on the amount of air voids. The amount of air voids is calculated from the relative density of the pavement structure and is re­ported as a percent of the total volume. Relative density is defined as the ratio of the bu1k specific gravity of the compacted mixture to the zero-air-void specific gravity (i.e., specific gravity of the mixture when no voids are present). Texas specifications require the air voids to be between 8 and 3 percent (92 and 97 percent relative den­sity when based upon the Rice method). The whole crite­rion for evaluating density in the mix depends on the val­ues obtained for the zero-air-void specific gravity or the theoretical maximum specific gravity. The Texas method of mixture design (Item 340) selects the design asphalt content which produces a specimen at 3 percent air voids

1

(97 percent relative density) provided that Hveem stabil­ity of the specimen satisfies the minimum requirements based on laboratory tests. It is the purpose of this report to evaluate and compare different methods of determining

the theoretical maximum specific gravity (Gmax), and to analyze their effects on selecting the design asphalt con­tent First, the methods will be briefly described; a pre­sentation of experimental data then follows. The results obtained from the tests are discussed and analyzed, and a conclusion is made.

This comparative study is performed on the maxi­mum specific gravities, the effective specific gravities, and the relative densities obtained from different meth­ods. The effect of varying the design asphalt content is investigated. The relationship between the differences in the maximum specific gravity and percent asphalt absorp­tion is also studied

CHAPTER 2. METHODS OF DETERMINING THE THEORETICAL MAXIMUM SPECIFIC GRAVITY

Currently, there are three methods which have been or are being used in Texas to estimate the theoretical

maximum density (zero air void density), Gmax· These methods are:

(1) Rice method (Test Method Tex-227-F, Ref 1), (2) SDHPT method as outlined in Construction Bulletin

C-14 (Ref 2), and (3) theoretical method based on the bulk specific grav­

ity of the aggregate.

These methods, along with a fourth method (backcalculation method), will be briefly discussed later.

In the C-14 and backcalculation methods, the effec­tive specific gravity of aggregates is used to determine the theoretical maximum specific gravity. Therefore, be­fore discussing the above methods, the effective specific gravity and its difference from other specific gravities of aggregates are briefly explained.

SPECIFIC GRAVITIES OF AGGREGATES The specific gravities of aggregates are required to

calculate the theoretical maximum specific gravity of the mixture. However, there are three different specific gravities, all of which will result in different values of the theoretical maximum specific gravity of the mixture. Thus, it is very important to distinguish clearly between different specific gravities of aggregates and their effects on the estimated theoretical maximum specific gravity of the mixture. Figure I illustrates the different types of specific gravities which can be calculated for the aggre­gate. The bulk and apparent specific gravities are ob­tained directly from tests on the aggregates. However, the effective specific gravity is obtained only through tests on asphalt-aggregate mixtures. The effective spe­cific gravity is greater than the bulk specific gravity. The difference between the bulk and effective specific gravi­ties increases as the amount of absorbed asphalt in­creases.

THE RICE METHOD

This method (Texas Test Method Tex-227-F and its equivalent,ASTM D-2041) was developed by James Rice (Ref 3), and thus is commonly known as the Rice method. In this method, a measured amount of asphalt­aggregate mixture is placed in a pycnometer and submerged in water. A vacuum is then applied to remove entrapped air at room temperature. The pycnometer is then filled with water and weighed. The weight of the pycnometer filled with water (without presence of the aggregate) is also determined. The volume of the mixture is calculated based on these weight

2

measurements. The theoretical maximum specific gravity is calculated using the dry weight and the volume of the mixture.

For mixtures containing porous aggregates or aggre­gates which are not completely coated, it is recommended that the saturated surface dry weight be used to determine the volume rather than the dry weight, to correct for er­rors introduced during testing by water absorption. The higher the water absorption, the greater the difference be­tween the corrected and uncorrected values. The follow­ing factors contribute to water absorption into the mixture during the Rice test (Ref 1): (a) insufficient or very thin coating of aggregates by asphalt, (b) very porous aggre­gate, and (c) excessive vacuum applied and excessive time used for vacuum.

In this report, the Rice specific gravity based on dry weight is referred to as the "uncorrected Rice Specific

Gravity," Gru. If saturated-surface-dry mixture is used to calculate volume, the gravity is termed "corrected Rice

Specific Gravity," Grc· The corrected gravity is always smaller than the uncorrected gravity and thus will result in a higher relative density. The uncorrected Rice method is generally used to determine the theoretical maximum specific gravity.

TEXAS SDHPT METHOD, C-14

This procedure is outlined in the Construction Bulle­tin C-14 (Ref 2). In this method, the effective specific gravity of the aggregate is first determined. Then, the ef­fective specific gravity is used to calculate the theoretical maximum specific gravity using the following formula:

where

G = 100 C-14 %Agg %AC --+--

Ge Gac

Gc.14 = theoretical maximum specific gravity, %Agg = percentage of aggregate in mix by

weight, %AC = percentage of asphalt in the mix by

weight, Ge = effective specific gravity of the

combined aggregates, and Gac = specific gravity of asphalt.

Thus, the effect of asphalt absorption is taken into account through the use of the effective specific gravity.

This method is unique in the way it determines the effective specific gravity of the aggregate. A series of specimens are compacted at different asphalt contents and

their bulk specific gravities are measured. Generally, the bulk specific gravity of the specimens increases as the as­phalt content increases until all voids are filled with as­phalt (saturated) and then decreases with additional as­phalt Specific gravity of the compacted specimens with one increment more asphalt cement than the compacted

specimens with the highest specific gravity (GcJ is used to calculate the effective specific gravity of the combined aggregate using the following formula:

where

100- %AC 100 %AC --+--Gcs Gac

Gcs = specific gravity of the compacted specimens with one increment more asphalt cement than the compacted specimens with the highest specific gravity.

THEORETICAL METHOD In this method, the bulk specific gravity of the com­

bined aggregates (which is measured according to Test Method Tex-201-F) is used to calculate the theoretical maximum specific gravity through the following equa­tion:

100 Gt= %Agg %AC

--+--~ Gac

where ~ = bulk specific gravity of the combined

aggregates.

The aggregate potential for asphalt absorption is ignored when bulk specific gravity is used. Therefore, this method yields a lower theoretical maximum specific gravity, higher relative density, and lower optimum as­phalt content compared to the first two methods.

RICE BACKCALCULATION METHOD

Another approach, in addition to the three methods mentioned above, is the Rice backcalculation approach, which is similar to the method used in C-14. In this method, only one Rice test is performed on one mixture. Generally, a mixture with a high asphalt content is selected to ensure complete coating of aggregates. The theoretical maximum specific gravity from the Rice test is used to calculate the effective specific gravity of the

aggregate, which is then used to calculate Gmax at other asphalt contents. This method takes significantly less time to find the maximum specific gravities at various asphalt contents than actually performing the Rice test at those asphalt contents. It is reasonable to believe that the effective specific gravity is independent of the amount of asphalt content as long as coating of aggregates with

3

asphalt is complete and the same type of asphalt is used. However, the chances of making errors in performing the Rice test increase at very high asphalt contents, mainly because part of the asphalt can easily be lost by its sticking to the pan and other implements.

Apparent specific gravity: G _ Ws a- (Vs + Vv)

Bulk specific gravity:

Gb= Ws (Vs + Vv + Vw)

Effective specific gravity: G _ Ws e- (Vs + Vv + Vw- Va)

Gb<Ge<Ga

V s = Volume of solids Vw= Volume of voids permeable to water Va = Volume of voids permeable to asphalt cement Vv = Volume of voids never filled with water or

asphalt cement Ws =Weight of solids yw = Unit weight of water

1 . -'Yw

1 . -'Yw

1 . -'Yw

Fig 1. Graphics representation to indicate different types of specific gravities of aggregates.

CHAPTER3. EXPERIMENTALPROGRAM

Thirteen projects were selected for this study. The aggregates included limestone and silicious materials. The asphalt cements were either AC-20 or AC-10 for all mixtures. The asphalts and aggregates, their specific gravities, and the asphalt sources are summarized in Ap­pendix Table A.l. Data were obtained from tests con­ducted by project and district personnel.

All data for projects one, two, and three were from the corresponding districts, and all data for projects six through thirteen were from tests conducted by project personnel. For projects four and five, the data concern­ing Rice specific gravities at different asphalt contents, and the specific gravities of compacted specimens, were obtained from the corresponding districts; however, the Rice specific gravities at design asphalt content or at a high asphalt content-and the bulk specific gravities of aggregates-were determined by project personnel.

For projects six through eleven, Rice specific gravi­ties were determined at design asphalt content and at a high asphalt contenL However, for projects twelve and thirteen, both corrected and uncorrected Rice specific gravities were determined at six to eight different asphalt contents.

LABORATORY TESTING The laboratory testing included determination of

specific gravities of the aggregate, - Rice specific gravities of mixtures, and - specific gravities of compacted specimens.

AGGREGATE SPECIFIC GRAVITY The bulk specific gravities of the coarse portion of

the aggregates (retained on No. 80 sieve) were deter­mined according to Texas Test Method Tex-201-F. The apparent specific gravities of the fine portion (passing No. 80 sieve) were determined according to Texas Test Method Tex-202-F.

MIXTURE-SPECIMEN SPECIFIC GRAVITIES

Two groups of asphalt-aggregate mixtures were pre­pared at various asphalt contents for each individual project for which asphalt and aggregate were available. The mixtures were divided into two batches. The mix­tures from batch one were used to determine the cor­rected and uncorrected Rice specific gravities of the mix­ture according to Texas Test Method Tex-227-F.

Mixtures from batch two were compacted into speci­mens using the Texas gyratory compactor according to Texas Test Method Tex-206-F. The specific gravities of the compacted specimens were then determined in accor­dance with Texas Test Method Tex-207-F. The specific

4

gravities of these compacted specimens were used in the C-14 method.

CALCULATIONS The results of the laboratory tests and the data ob­

tained from the districts were used to calculate the differ­ent parameters as follows:

(1) Corrected and uncorrected Rice specific gravities obtained at various asphalt contents were used to calculate the effective specific gravities of the ag­gregates.

(2) The combined bulk specific gravities of aggregates were used to calculate the maximum theoretical spe­cific gravities of the mixtures at various asphalt contents.

(3) The specific gravities of compacted specimens and the maximum theoretical specific gravities calcu­lated in step (2) above were used in the C-14 method to determine the effective specific gravities of the aggregates and the maximum specific gravi­ties of mixtures.

(4) The effective specific gravities from the corrected and uncorrected Rice specific gravity at optimum asphalt content or at a high asphalt content were used to backcalculate the maximum specific gravi­ties at other asphalt contents.

(5) Four types of relative densities at various asphalt contents were calculated. The relative densities were calculated based on four types of maximum specific gravities mentioned before.

(6) The amount of asphalt absorbed for different as­phalt aggregate mixtures was calculated with the aid of the effective specific gravities of aggregates.

ANALYSIS The following parameters were compared and evalu­

ated:

(1) The effective specific gravities obtained from cor­rected and uncorrected Rice specific gravities at various asphalt contents.

(2) The effective specific gravities at optimum asphalt content obtained from different maximum specific gravities.

(3) The maximum specific gravities at optimum asphalt content from different procedures.

(4) The maximum specific gravities at various asphalt contents for each individual projecL

(5) The relative densities based on different maximum specific gravities.

(6) The amount of asphalt absorptions determined from different procedures.

CHAPTER 4. DISCUSSION OF RESULTS

EFFECTIVE SPECIFIC GRAVITY

The effective specific gravity (Ge) of the combined aggregates for each project was backcalculated from the Rice specific gravities at different asphalt contents. The results, on a distorted scale, are shown in Fig 2. There is no consistent trend between the backcalculated specific gravity and asphalt content at which the Rice gravity was determined. In addition, the variations are small. The maximum difference between the backcalculated effec­tive specific gravities for the same project does not ex­

ceed 2 percent of the average Ge for all cases. Table A-2 gives numerical values for Ge

backcalculated using different methods for different projects, and Table A-3 presents a statistical summary on

differences in Ge values from various approaches. It can

be seen that average difference between Ge values from uncorrected Rice and C-14 methods is 0.021 (less than 1 percent). Much closer agreement exists between Ge val­ues from the corrected Rice method and those from the C-14 method. Table A-3 indicates that, in this case, aver­age difference and standard deviation are about 0.004 and 0.013, respectively, when algebraic values are used The difference, in no case, exceeds 0.023.

Effective specific gravities from different methods

are also compared in the scatter plots of Fig 3. Ge values from the Rice tests are obtained from tests at design as­phalt content. Each point is representative of one project

In almost all cases, Ge values from the C-14 method are smaller than those from the uncorrected Rice method. However, in the case of the corrected Rice method, al­most half of the values from the Rice method are larger than those from the C-14 method.

It is also worth noting that the Ge value calculated directly from the Rice test at design asphalt content is very close to the one backcalculated from another Rice test at a different asphalt content (Fig 3 and Table A-2).

THEORETICAL MAXIMUM SPECIFIC GRAVITY

Table A4 presents numerical values obtained for the

theoretical maximum specific gravity (Gmax) for different projects at different asphalt contents, and Table A-5 gives

a statistical summary on differences between Gmax val­ues from different projects. In almost all cases, the

uncorrected Rice method yields a higher Gmax than the C-14 method, as can be seen in Fig 4. Average difference

in Gmax from the C-14 and the uncorrected Rice method

is 0.015 (less than 1 percent of typical Gmax values).

5

However, the agreement is significantly better in the case

of the corrected Rice method. The difference in Gmax obtained from the C-14 and from the corrected Rice method has an average value of 0.003 when algebraic values are used and of 0.011 when absolute values of dif­ferences are used (Table A-5). The standard deviation is 0.013 in the former case and 0.007 in the latter case. Fig­ure 5 indicates that differences are both positive and negative. Notice that both C-14 and Rice methods give

significantly higher values for Gmax than the theoretical approach, which uses the bulk specific gravity of aggre­gate and does not take into account the effect of asphalt absorption (refer to Fig 4 and Tables A-4 and A-5). A

summary of Gmax values at design asphalt content for different projects, and differences between them, are given in Tables A-6 and A-7, respectively. Good agree­ment exists between the C-14 and uncorrected Rice meth­ods at design asphalt content. However, the agreement is better when the corrected Rice method is compared with the C-14 method. Figure 4 supports this conclusion. The theoretical approach, as expected, yields significantly

lower values for Gmax than other methods (Fig 4). Figures A-1 through A-6 indicate differences be­

tween Gmax values from various methods in the case of projects for which Rice specific gravities at different as­phalt contents are available. In all cases except for

project twelve, Gmax from uncorrected Rice approach is higher than that from C-14 method. There are only two projects (twelve and thirteen) for which both corrected and uncorrected Rice specific gravities are available, and these two suggest that very close agreement exists be­tween the corrected Rice method and the C-14 method.

In one case, Gmax from the C-14 method is higher than that from the corrected Rice, and in the other case, the in­verse is true. The figures also indicate that

backcalculation values of Gmax from one Rice test com­pare very well with those obtained from the Rice test di­rectly.

RELATIVE DENSITIES Figures A-7 through A-12 and Tables A-8 and A-9

were developed from the data available on Gmax values and specific gravities of compacted specimens. Relative densities based on different approaches are given in these figures. It can be seen that, because maximum theoreti­cal specific gravities from the C-14 method are smaller than those from the uncorrected Rice method, the former results in higher relative densities than the latter.

Notice that backcalculation methods yield good re­sults compared to the C-14 and Rice methods. Relative

6

Project 1 -"0 ;::.~ ·- (.) > (I) e :::: 2.62o (,!:)0

.!:::? g 2.610 :t:::::::;) (.)-

~ u.J 2.600 (/.)~ g; cc 2.590 ·- c:: -o ~ -o 2.580 ~(I)

u.J ~ 2.570 c:a

2.560 4.0 5.0 5.5 6.0 7.0 8.0

Asphalt Content(%)

2.680 Project3 Project 4 ::0 ::0

;::. ~ 2.670 ;::.~ ·- u ·- u ~ ~ 2.660

> Q) e:::: (,!:)0 (,!:)0

.!:::? g 2.650 u (.) 2.660 ·- c:: :t:::::::;) :t:::::::;)

(.)- u-~ u.J 2.640 <Uw (I)~ c..u

(I)-g; a: 2.630 (I) a: ·- c:: .~ c:: -o 'QO ~ "0 2.620 <U-o ~(I) ~Q)

u.J ~ 2.610 w rn C'CI c:a

2.600 5.0 6.0 7.0 8.0 4.0 5.0 6.0 7.0

Asphalt Content(%) Asphalt Content(%)

Fig 2. Effective specific gravities of aggregates obtained from Rice test at different asphalt contents.

densities based on the corrected Rice method are given only for projects twelve and thirteen. In most cases, ex­cept one, the difference between relative densities from the Rice and C-14 methods does not exceed 1 percent, even when the uncorrected Rice specific gravity is com­pared with that from the C-14 method.

DESIGN ASPHALT CONTENT Texas specifications require the design asphalt con­

tent to be selected at 97 percent relative density. Because the relative density from the C-14 method is higher than that from the uncorrected Rice method, the former is ex­pected to yield a lower design asphalt content than the latter (Fig 6a). For this reason, in Fig 6b the asphalt con­tent at 96 percent relative density based on the uncorrected Rice method is compared with the asphalt content at 97 percent relative density based on the C-14 method. In this case, better agreement is observed. Since the C-14 method generally yields a drier mix than the uncorrected Rice method at 97 percent relative den­sity, equivalent mixes are expected if design asphalt con­tent is selected at 97 percent based on the former and at 96 percent based on the latter. Unfortunately, sufficient

data are not available for the corrected Rice method to compare with data from the C-14 method. However, the design asphalt contents from these two methods are gen­erally expected to be very close.

ASPHALT ABSORPTION To investigate the existence of a meaningful relation­

ship between the amount of asphalt absorption and the difference in maximum specific gravities, Fig 7 was pro­vided. The abscissa in this figure is the average percent absorption for each project, which is found based on per­cent absorptions at different asphalt contents. The aver­age difference in maximum specific gravities also pre­sents the average of differences at different asphalt contents. The data are scattered and do not suggest any meaningful correlation.

A WORD OF CAUTION CONCERNING THE C-14 METHOD

As mentioned before, in the C-14 method, the assumption is that no air voids exist in the compacted specimen when the saturation point is reached, and

Project 12 "'0

.2:- ~ 2.650 ·- (.) > Q)

~ ::: 2.640 (.!)0

(.) :E :5 2.630 (.)-

:g_ UJ 2.620 enS:? ~a: 2.610 ·- c: -o ~ "'0 2.600 :j::Q) UJ en iil 2.590

4.0 5.0 5.3 6.0

Asphalt Content(%)

2.640 Project 13

.2:-~ 2.630 ·- (.)

~ ~ 2.620 (.!)0

-~ g 2.610 ~:::::l (.)-

:g_ UJ 2.600 enS:? Q) a: 2.590 .2: c: -o ~ "'0 2.580 :j::Q)

UJ ~ 2.570

4.0 5.0 5.5 6.0 6.5

Asphalt Content(%)

7.0 8.0

2.640

.2:- ::0 2.630 ·:;: s ~ ~ 2.620 (.!):::

-~ 8 2.610 --'(3 UJ :g_ S:2 2.600 en a: ~ c: 2.590 ~0

~ ~ 2.580 :~=en UJ~ co

7.0 7.5

Project 12

5] 5~ 6] 7]

Asphalt Content(%)

Project 13

5B ~0 6B 7]

Asphalt Content(%)

7

Fig 2 (continued). Effective specific gravities of aggregates obtained from Rice test at different asphalt contents.

increasing the asphalt content beyond this point causes reduction in the specific gravity of the specimen. The specific gravity of the specimen following the highest specific gravity is used to find the effective specific gravity of the combined aggregates, because experience indicates that the specimen with the highest specific gravity may have not reached saturation yet and that complete saturation is probably achieved at the asphalt content of the specimen following the peak. It was found that in almost all cases, the relative densities of the specimens with specific gravities following the peak are

equal to or exceed 100 percent when Gmax is calculated based on the bulk specific gravity of the aggregates (i.e., when the theoretical approach is used). No problem is imposed as long as such is the case. However, a review of a number of projects in districts where the C-14 method was used revealed that in some projects, the set of specimens with specific gravity following the peak yielded a relative density less than 100 percent. Assuming that the bulk specific gravity of the aggregates was correct, the only way that the relative density could

be less than 100 percent was that saturation was not complete and there were still some voids left in the specimen. There is no way that, on one hand, the specimen could have a lower density than 100 percent based on the theoretical approach, and, on the other, be saturated. Because of this problem, these projects reported effective specific gravities less than the bulk specific gravity, which is impossible, and obtained negative asphalt absorptions.

If theoretical maximum specific gravities are calcu­lated based on the bulk specific gravity, the mix design will yield a mix with less asphalt than could be obtained by any other method. However, erroneous values for the effective specific gravity from the C-14 method will cause this method to yield a mix with even less asphalt content than what should be obtained from the theoretical approach. During the field density control stage, relative densities calculated based on the theoretical approach are generally not conservative. Erroneous results from the C-14 method may indicate significantly higher relative den­sities than actually exist

8

2.660 2.640

2.640 2.620

2.620 2.600

"<t • "<t ..-2.600

..-I t1 u E E 2.580 • e 0

2.580 ..... - • -CD CD C'.!:J • C'.!:J

2.560 2.560 • 2.540 2.540

2.520 2.520 2.520 2.540 2.560 2.580 2.600 2.620 2.640 2.660 2.520 2.540 2.560 2.580 2.600 2.620 2.640

Ge from RICE (Uncorrected) Ge from RICE (Corrected)

2.660 2.640

- 2.630 2.640 'C • CD -(.) 2.620

£" e -n 2.620

.....

• 0 2.610 (.)

e c: .... ::::::> 0 ~ 2.600 !2. 2.600 0 UJ :;::;

2.590 !::::! "' a: "3 E 2.580

(.)

«i 2.580 • ~ g (.)

"' 2.570 cff 2.560 a:. E e - 2.560

2.540 (1:1

C'.!:J 2.550

2.520 2.540 2.520 2.540 2.560 2.580 2.600 2.620 2.640 2.660 2.540 2.560 2.580 2.600 2.620 2.640

Ge from RICE (Uncorrected) Ge from RICE (Uncorrected)

Fig 3. Comparing effective specific gravities from different methods at design asphalt content.

9

2.460 2.440

2.440 2.420 • 2.420

""" """ 2.400 ..... ..... <:,)

2.400 <:,) • E

2.380 e I -~ 2.380 )( Cl:l

E E ~ ~ 2.360

2.360

2.340 2.340

2.320 2.320 2.320 2.340 2.360 2.380 2.400 2.420 2.440 2.460 2.320 2.340 2.360 2.380 2.400 2.420 2.440

Gmax from RICE (Uncorrected) Gmax from RICE (Corrected)

2.460 2.440

'6' 2.440 s

(.) 2.420 2:! '6' ....

(1) 0 t) 2.420

(.) c::

2:! ::I 2.400 .... -0 c:: 8 2.400 0

~ UJ .!5! (.) ::I 2.380 cc (.)

E 2.380 ~ e (.)

Cl:l - CD 2.360 ~ 2.360 E E 0 .... ~ -

2.340 ~ 2.340 E ~

2.320 2.320 2.320 2.340 2.360 2.380 2.400 2.420 2.440 2.460 2.320 2.340 2.360 2.380 2.400 2.420 2.440

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

Fig 4. Comparing theoretical maximum specific gravities from different methods at design asphalt contents.

10

2.460

2.440

2.420 '<t

).: ._, 2.400 E E -~ 2.380 E

C!)

2.360

2.340

• •

2.320 L---'----L--'--...I--.L.............I---'--...1 2.320 2.340 2.360 2.380 2.400 2.420 2.440 2.460 2.480

2.480-

:::::- 2.440 ~ ;;; u

;;::l

t'!: 2.400 0 (1.)

..c:: I-

E ,g 2.360 X <tS E

C!) 2.320

2.280 2.280

2.480

- 2.440 -~ ;;; .g (1.) 2.400 0 (1.)

..c:: 1-

E 2.360 0 .::.

X <tS

E C!)

2.320

2.280 2.280

Gmax from Backcalculation (Uncorrected)

• •

•

2.330 2.380 2.430 2.480

Gmax from C-14

, • •

2.320 2.360 2.400 2.440 2.480

Gmax from RICE (Uncorrected)

Fig 4 (continued). Comparing theoretical maximum specific gravities from different methods at design

asphalt contents.

11

0.030

en 0.025 <»en :::::1-o-(ijcv

0.020 >.c.-xQi J., E:; I 0.015

C!)VLLJ

ciS:2 0.010 .uua:: ;-c-o _c.u

0.005 .urcs-a::lw~ Q) c..J ~

0.000 u- c c a: u ~ec <» c :::> -{) 005 :c:: ..... - . a-

-{).010

-{).015 (a)

0.020

en <D(I) :::::1-o "iijc->.c:V 0.010 -...->< <» I res:; u e..,. I

C!l.,...w clu <»U-<»-oa:: 0.000 .!c-o .un:l<» a::lwtS CD (...) <» u- .._ c a: 0 ; E~ -{).010 ;:!: a-

-{).020

FigS. Differences between values or the theoretical maximum specific gravities from (a) C-14 and uncorrected Rice methods and (b) C-14 and corrected Rice methods.

12

• •

•

5S 6] 6S 7D Optimum AC Based on RICE

(Uncorrected at 97% Relative Density)

5S 6] 6S 7]

Optimum AC Based on RICE (Uncorrected at 96% Relative Density)

Fig 6. Comparing design asphalt contents selected based on C-14 and Rice methods.

(/)

~ 0

0.030

.c 0.025 -s ::iii E ,g "#' 0.020

~~ ::~c..>

~ "g O.D15 x"' "'w EU

<!'cc c:: 0.010

0.005

0.000 0.23 0.40 0.61 0.66 0.67 0.79 0.92 1.08 1.72

Absorption (%)

Fig 7. Difference in Gmax values from Rice and C-14 methods versus percent of asphalt absorption.

CHAPTER 5. CONCLUSIONS AND RECOMMENDATIONS

The following conclusions are made based on the data analyzed in this study:

(1) The theoretical maximum specific gravities from the theoretical approach (Gt) are, in general, signifi­cantly lower than those obtained from other meth­ods, and therefore, yield the highest relative densi­ties.

(2) The theoretical maximum specific gravities from the uncorrected Rice method (Gru) are the highest and result in the lowest relative densities compared to other methods.

(3) Relatively good agreement exists between the theo­retical maximum specific gravities from the C-14 method (Gct4) with those from the uncorrected Rice method (Gru).

(4) The best agreement is observed between Gct4 val­ues and Grc values.

(5) Backcalculating theoretical maximum specific grav­ity (Gmax) values from one single Rice test is also promising, and backcalculated Gmax values com­pare well with those directly from the Rice tests.

13

(6) Design asphalt content at 97 percent relative density based on Gct4 is lower than that at 97 percent rela­tive density based on Gru.

(7) Relatively good agreement exists between design asphalt contents when 96 percent relative density is used based on Gru and 97 percent relative density is used based on Gct4·

(8) In this study it was found that the largest differ­ences between the specific gravities measured by the two methods generally do not exceed 2 percent. This difference causes the relative density based on the corrected Rice method to be about 1 to 2 percent less than that based on the uncorrected Rice method. The Rice method is gaining widespread acceptance both at national and at state levels.

(9) An additional consideration should be stated in the C-14 method. If saturation for a specimen is going to be considered complete following the peak, the relative density of that specimen based on the theo­retical approach must exceed 100 percent; otherwise erroneous results will occur.

REFERENCES

1. Manual of Testing Procedures, Volume 2, Texas State Department of Highways and Public Transportation, Austin, Texas, 1982.

2. Construction Bulletin C-14, "Hot Mix Asphaltic Concrete Pavement," Design of Laboratory Mixes, Texas State Department of Highways and Public Transportation, Austin, Texas, April1984.

14

3. Rice, J. M., "Maximwn Specific Gravity of Bitumi­nous Mixtures by Vacuum Saturation Proce­dure," ASTM Special Technical Publication No. 191, June 1956.

APPENDIX. FIGURES AND TABLES

2.500 Project1 2.500 Project1

2.450 2.450

v • "<t • I ~ c.J 2.400 2.400 E E 0 0 .... .... - -X X

"' 2.350 C"CI 2.350 E E (.!:) (.!:)

2.300 2.300

2.250 2.250 2.250 2.300 2.350 2.400 2.450 2.500 2.250 2.300 2.350 2.400 2.450 2.500

Gmax from RICE (Uncorrected) Gmax from Backcalculation (Uncorrected)

2.460 Project1 2.500 Project1 '6' .s 2.440 u <U 2.450 .... • .... -0 2.420 @. u c

::::::l <a 2.400 u 2.400 c = • .g <U .... <1:1 2.380 0

'3 <U ..c::

2.350 u I-<a 2.360 E u .::.::. 0 u ..... <1:1 2.340

...... ro X 2.300 co E E e 2.320 (.!:) -X 2.250 C"CI E 2.300

(.!:)

2.280 2.200 2.280 2.320 2.360 2.400 2.440 2.480 2.250 2.300 2.300 2.350 2.400 2.450 2.500

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

Fig A-1. Theoretical maximum specific gravity from different methods for Project 1.

15

16

2.480 Projecl2 2.480 Project 2

2.420 2.420

-.:r • -.:r ...... ...... I I u u

E • E e 2.360 e 2.360 - -X X C'<:l C'<:l E E

C.!') (.!:)

2.300 2.300

2.240 2.240 2.240 2.300 2.360 2.420 2.480 2.240 2.300 2.360 2.420 2.480

Gmax from RICE (Uncorrected) Gmax from Backcalculation (Uncorrected)

2.500 Project 2 2.500 Projecl2 15' <U n !!: 2.450 2.450 ..... -0 @ (.) c

:::::l a; -c • (.) :;::;

.Q 2.400 !!: 2.400 1:;3 0 • "5 <U

.c (.) ..... ~ E 2.350 • .:.::. 2.350 (.) 0 ..... C'<:l -co X

E C'<:l E e

2.300 (.!:)

2.300 -X C'<:l E

(.!:)

2.250 2.250 2.250 2.300 2.350 2.400 2.450 2.500 2.250 2.300 2.350 2.400 2.450 2.500

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

Fig A-2. Theoretical maximum specific gravity from different methods for Project 2.

17

2.460 Project 3 2.460 Project 3 6' ~ Q) • :::

2.420 • 0 2.420 u c:

'<t ::l -.,.... I c:

(,.) 0

E ~ e 2.380 3 2.380 u - ~ X

<':! .:..: E u (!) • <':!

CD

2.340 E 2.340 0 .... -)( CIS E

(!)

2.300 2.300 2.300 2.340 2.380 2.420 2.460 2.300 2.340 2.380 2.420 2.460

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

2.460 Project3

'<t .,.... I

u 2.420 ~ .> ~ (!) u

;;::: 2.380 '(3

a:> c.

C/)

E e -X 2.340 <':! E

(!)

2.300 ___ __,_ ___ ......_ ___ ...._ __ __.

2.300 2.340 2.380 2.420 2.460

Gmax from Backcalculation (Uncorrected)

Fig A-3. Theoretical maximum specific gravity from different methods for Project 3.

18

2.520 2.520 Project 5

2.480 2.480 ...,. ...,. ...- ...-I 6 u E E 0 2.440 0 2.440 ... ... - -X X

<'tl co E E

C!) C!)

2.400 2.400

2.360 2.360 2.360 2.400 2.440 2.480 2.520 2.360 2.400 2.440 2.480 2.520

Gmax from RICE (Uncorrected) Gmax from Backcalculation (Uncorrected)

2.520 Project 5 2.520 Project 5 S' Q)

u l!: ..... -0

2.480 - 2.470 (..) ~ c ::::l a:i c

(..) :;::!

0 e :;::! • "' 0 '3 Q)

2.420 2.440 .s:::. (..) 1-;;:;

E (..) .;:,:: 0 • (..) ... <'tl -co X

E co

e 2.400 c5 2.370 • -X

"' E C!) • 2.360 2.320

2.360 2.400 2.440 . 2.480 2.520 2.320 2.360 2.400 2.440 2.480 2.520

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

Fig A-4. Theoretical maximum specific gravity from different methods for Project 4.

2.480 Project12

2.440

~ .,... c.1 E 0 2.400 .... -)( "' E

C!'

2.360

2.320 !&.....---'-----'-------''-----' 2.320 2.360 2.400 2.440 2.480

Gmax from RICE (Uncorrected)

2.500 Project12

2.460

~

~ 2.420 E e -)( E 2.38o C!'

2.340

2.300 w:.------"----1..---'----'-----J 2.300 2.340 2.380 2.420 2.460 2.500

Gmax from Backcalculation (Uncorrected)

19

2.500 Project12

2.460

~ .,... c.1 2.420

E e -)( E 2.38o C!'

2.300 __ ___.. __ __... __ __._ __ __._ __ __,

2.300 2.340 2.380 2.420 2.460

2.460

~ 2.420

~ E e - 2.380 ~ E

C!'

2.340

Gmax from RICE (Corrected)

Project 12

2.500

2.300 ----'---...1..--...1----'-----J 2.300 2.340 2.380 2.420 2.460 2.500

Gmax from Backcalculation (Corrected)

Fig A-S. Theoretical maximum specific gravity from different methods for Project 12.

20

2.500 Project12 2.460

s Project 12 <I)

t.)

s 2.460 ~ <I) 0 2.420 t.)

(..) c::

(I) ::J ... ... -0 2.420 c:: <:....:) 0 - :;::::; UJ <'tl

~ '3 2.380 0: ~

<'tl

E 2.380 (..)

-"'= 0 (..) ... <'tl - co X

"' E E e 2.340 (!) 2.340 -X

<'tl

E C!l

2.300 2.300 2.300 2.340 2.380 2.420 2.460 2.500 2.300 2.340 2.380 2.420 2.460

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

2.460 Project 12 2.460 Project 12

s .$! (..)

-=- ~ 2.420 .... 2.420 8 0

c.:> cti -~ • c::

0 (I) t; 0 '3 <I)

..c:: 2.380 (..) 2.380 ..... ~

E (..)

-"'= 0 (..) ... <'tl - cc

X E "' C!JE 2.340 e 2.340 -X <'tl

E C!l

2.300 2.300 2.300 2.340 2.380 2.420 2.460 2.300 2.340 2.380 2.420 2.460

Gmax from RICE (Uncorrected) Gmax from RICE (Corrected)

Fig A·S (continued). Theoretical maximum specific gravity from different methods for Project 12.

21

2.460 Project 13 2.420 Project 13

2.420 • '<I" '<I" 2.380 I

..... '-' ~ E E 0 2.380 e .... - -X X

"" • "' E E (!:) (!:) 2.340

2.340

• 2.300 lol:---....l..---..1------1---....1

2.300 2.340 2.380 2.420 2.460 2.300 1&...----L----....1--------J

2.300 2.340 2.380 2.420

Gmax from RICE (Uncorrected) Gmax from RICE (Corrected)

2.460 Project 13 Project 13

2.420 2.420 '<I" • '<I" .,.... I

.,.... '-' • ~ E • E 0 2.380 2.380 .... • e -X -"" • ~ E E (!:) (!:)

2.340 • 2.340

• •

2.300 "------1.----'-----'-------J 2.300 ~---'----.1...----L.--_1 2.300 2.340 2.380 2.420 2.460 2.300 2.340 2.380 2.420 2.460

Gmax from Backcalculation (Uncorrected) Gmax from Backcalculation

Fig A-6. Theoretical maximum specific gravity from different methods for Project 13.

22

2.460 Project 13

2.460 ::0

Project 13 (1)

0 2.460 (1)

::0 ..... .....

(1) 2.420 0 2.420 0 0 c: E ::> ..... -0 c: 2.420 8 0

L.LJ ;; S:? 2.380 "3 2.380 a: 0

E 13 ~ 0 (..) 2.380 ..... - C'C

X cc C'C E E 2.340 2.340

<:..::) e .... X e 2.34o

<:..::)

2.300 2.300 2.300 2.340 2.380 2.420 2.460 2.300 2.340 2.380 2.420 2.460

Gmax from RICE (Uncorrected) Gmax from RICE (Uncorrected)

2.440 Project 13 2.460

Project 13 '0 .s 0 (1) .....

~ .....

2.400 0 2.420 u - -~ • c: 0

~ ~ ..... • "3 0 (1)

2.360 • 0 2.380 .c 13 1- • E ~

(..) e C'C - • cc X C'C c5 2.320 • 2.340

><

• C'C E

0

2.280 2.300 2.280 2.320 2.360 2.400 2.440 2.300 2.340 2.380 2.420

Gmax from RICE (Uncorrected) Gmax from RICE (Corrected)

Fig A-6 (continued). Theoretical maximum specific gravity from different methods for Project 13.

-102

100 ~ e.... 98 .&!' ·~ 96 <1)

0 94

.~ ;;; 92 (j) ex:

Project 1

-C-14 - - RICE (Uncorrected)

90

884~----~~----~------~------~8 5 6 7

-102

100 ~ e.... 98 Z' ·~ 96 <1)

0 94 ~ iii 92 (j) ex:

Asphalt Content(%)

90

88~------~----~------~------~ 4 5 6

Asphalt Content(%)

-102

100

~ 98 .&!' ·~ 96 Q)

0 94 ~ iii 92 ~

Project 1

-C-14 • •••• Backcalculation Method

(Uncorrected) •• ••••• ••• ••• •• •••

23

... •···········

90

88~----~~----~------~------~ 4 5 6

Asphalt Content(%)

Project 1

All Methods Combined

88 4~----~5~----~6------~------~

Asphalt Content (%)

Fig A·7. Relative densities based on different methods for Project l.

24

101 Project2 101

100 ,- 100

;:i 99 ,,

- 99 0

,., eft -98 ; -98 ~ ~ ·;;; 97 ; ·~ 97 t:: ; ~ 96 ; ~ 96 Q) 95

; Q) 95 .~ ; -~ .....

94 ; -C-14 - 94 ••••• Backcalculatioo Method (Uncorrected) .£Sl .£Sl Q) ; - - RICE (Uncorrected) Q) -C-14

ex: 93 ; ex: 93

92 ; 92

91 4 5 6 7 8 914 5 6 7 a

Asphalt Content(%) Asphalt Content (%)

101 101

100 100

~ 99 ~ 99

~ 98 ~ 9a ·;;;

97 ·;;; 97 t:: t:: Q)

96 Q)

96 Cl - C-14 Cl Q)

95 - - Theoretical Method Q) 95 > >

N 94 ~ 94 ..... Backcalculation Method (Uncorrected)

Q) Q) ·- Theoretical Method ex: 93 ex: 93 - - RICE (Uncorrected)

92 92 -c-14

91 4 5 6 7 8 6 7 a

Asphalt Content(%) Asphalt Content(%)

Fig A-8. Relative densities based on ditTerent methods for Project 2.

100

-99 ~ ~ .;::;. 98 ·c;; c: cg 97

~ ~ 96 Q)

a:: 95

Project3

-- RICE (Uncorrected) -C-14

94~--~~--_.----~----~----._--~

100

-99 ~ ~ .;::;. 98 ·c;; c: Q) 97 CJ Q)

> E 96 Q)

a: 95

94

100

;;? 99 ~ .;::;. 98 ·c;; c: Q)

97 CJ

~ :;:::: 96 "" cu a::

95

94

5 5.5 6 6.5 7

5

5

Asphalt Content(%)

Project 3

.,_..,-. ;

.,_..,-. .,_..,-. """"""

"""""" """"""

~;;; ••••• Backcalculation Method (Uncorrected) ;' - G-14

5.5

5.5

6 6.5 7 7.5 8 Asphalt Content(%)

Project 3

..... Backcalculated Method (Uncorrected) - - RICE (Uncorrected) -C-14

6 6.5 7 7.5 8

Asphalt Content(%)

Fig A-9. Relative densities based on different methods for Project 3.

25

26

100

99 ;g 98 :::._. ~ 97

~ 96 CD

c 95 CD > ~ 94

~ 93

92

914

100

99 ;g 98 :::._.

•

Project 5

- - RICE (Uncorrected) - C-14 ----

,..

--.,..,. ....,.,. .,..,.

5 5.5 6 Asphalt Content(%)

,.. ,.. ,..

ProjectS ...... -~· ..... · ,..--

,.. ,.. ,..

,._ RICE (Uncorrected) - C-14

4.5 5 5.5 6 6.5 7

Asphalt Content(%)

100

99 ;g 98 :::._. .z:. 97 -~ 96 CD c CD 95 > ~94 ~ 93

92

Projects

• • • • • Backcalculatlon Method (Uncorrected) - C-14

All Methods

,.. ..

,.. ,..

Project 5

,. , . ,. ,. ...... -

~· ..... · ,..--

91~---L----L----L----L----L--~ 4 4.5 5 5.5 6 6.5 7

Asphalt Content(%)

Fig A-10. Relative densities based on different methods for ProjectS.

27

101 Project 12 101 Project 12 100 100

~ 99 -99 0 'if!. -98 -98 2:- 2:-·~ 97 ·~ 97

~ 96 ~96 (1)

95 -C-14 ~ 95 > ••• •• Backcalculation Method (Uncorrected) :;:::; -- RICE (Uncorrected)

·+::;

"' 94 * 94 -C-14 <i3

a: 93 a:ga

92 92

6 7 8 91 4 5 6 7 8 Asphalt Content(%) Asphalt Content (%)

101 Project12 ,-•""" 101 Project12 100 ~·""";-- 100

~ 99 ·-- ~99 0 r -98 .~ -gs ~ ...... .....:..- 2:-·~ 97 ...... ., ·~ 97 ,... --~ 96 . ..,.- ~96 /; ~ 95 ~ •- RICE (Corrected) ~ 95

:;:::; .~ - - RICE (Uncorrected) ·+::;

~ 94 * 94 (1) v a: 93 a: 93

92 92 91 91 4 4 6 7 8 5 6 7 8

Asphalt Content(%) Asphalt Content(%)

Fig A-11. Relative densities based on different methods for Project 12.

28

101 Project 12 101 Projecl12 100 100

;e 99 -99 <;/!. "' -98 -98 ?;- :;::;. ·u; 97 -~ 97 c: ~ 96 ~96

""' 95 ~ 95 .2!: ·.;:::: ~ 94 ~ 94 ""' a: 93 a: 93

92

6 6

Asphalt Content{%) Asphalt Content(%)

101 Project12 101 100 100

-99 ;e 99 <;/!. -98 !:_. 98

~ 97 -~ 97 c: a; 96 ~ 96 Q

~ 95 ""' 95 - C-14 .2!:

1S! 94 -•- Theoretical Method <G 94 ""' iii a: 93 a: 93

92 92

91 4 5 6 7 8 6 7 8

Asphalt Content{%) Asphalt Content{%)

Fig A-ll (continued). Relative densities based on different methods for Project 12.

29

102 Project13 102 Project13 101 101

l100 -cloo .2:- 99 ~ 99 ····· .... -~ 98 0 •••

98 ... .,.,. c ...

QJ QJ •••······ 0 97

,.,. 0 97 ••

Q) QJ •• .2: _,-' - - RICE (Corrected) > •• ... •• Backcalculalion Method .Ei 96 ~ 96

.. •• .,...,. -c-14 •• -C-14 QJ Q) •• a: 95 a: 95 •• •• ••

94 94 •••••• 934 934_;··

.5 5 5.5 6.5 7 7.5 5 5.5 6 6.5 7 7.5

Asphalt Content (%) Asphalt Content (%)

102 Project13 102 Project13 101 101 --·--l100 - .,-. .,-. ~ c1oo ..,.. .

.2:- 99 ~ 99 . ~· ... ~ ~···· ·v;

98 ·v;

98 ,. .... -.:1'-c c ,. ... Q) Q) . ~· ~tr······ 0

97 0

97 Q) ·- Theoretocal Method Q) • .; ;" ••••• Backcalculalion Method > >

:;:; 96 -C-14 ·.;::::; 96 .,. • .; .,..:tt' - - RICE (Corrected) «! t:C

o.; o.; 95 • ~··· ·- Theoretocal Method a: 95 a:

••••• - G-14 94 94 •••••

934.5 •••

5 5.5 6 6.5 7 7.5 934.5

Asphalt Content(%)

Fig A-12. Relative densities based on difFerent methods for Project 13.

30

101

100 -rfl. 99 -~ 98 "(';; c: Q)

c Q)

. 2: (; (i) a:

97

96

95

94

93

924.5

101

100

~99 ~ 98 •<;; ~ 97 r; 96 > .fii 95

~ 94

93

924.5

5

5

Project 13

- - RICE (Corrected) - C--14

5.5 6 6.5 7

Asphalt Content(%)

Project 13

- - RICE (Corrected) ··- RICE (Uncorrected)

5.5 6 6.5

Asphalt Content(%)

7.5

101

100

~ 99 -~ 98 "(';; c: 97 (1.)

Cl 96 (1.)

> :;::: 95 ('Q

(i) a: 94

93

924.5

101

100

~99 -.2:- 98 "(';;

~ 97

~ 96 > f! 95 ~ 94

93

•• •• •• •• •• ••

Project 13

•• •• .. ... ·······•·

.... ·········· ... ••• •• Backcalculation Method

(Uncorrected) -C-14

····· •••

Project13

All Methods ; ·' ... -...;:;;-·tt:··· ., ..-:~:· ., ttt···· ~ ~··· Backcalculation Method r;::t=f:!:.' (Uncorrected)

.;,tl"• ••• •• •- RICE (Corrected) ·""' •••• - - RICE (Uncorrected) ••• •• - C-14 ••

Fig A-12 (continued). Relative densities based on different methods for Project 13.

31

TABLEA-1. PROJECTS USED IN THE PRESENT STUDY

Aggregate Aggregate Aggregate Mix Asphalt Asphalt Asphalt Bulk Design

Project District Coun!,! Source Type Type Type Source S.G. S.G. A. C.

1 1 Lamar B.H., F.P. Limestone D AC-20 Ardmore 0.992 2.542 5.8 2 1 Hunt B.H., T.F. Limestone D AC-20 Texaco 1.029 2.529 5.6 3 13 Jackson R.W.,S.F. Limestone D AC-20 Gulf St 1.024 N/A 4.5 4 6 Midland T.P. Rhyolite D AC-20 Fina 1.029 2.501 6.2 5 13 Calhoun B.I., R.D., W.F. GRVlJL.S.* D AC-20 TFA 1.022 2.578 5.0 6 B.H.,T.X. Limestone D AC-10 TFA 1.021 2.500 5.3 7 14 Travis C.M. Limestone D AC-20 TFA 1.022 2.534 6.0 8 16 San Pat. R.W., P.B., H.F. Limestone D AC-10 TFA 1.020 2.584 4.2 9 16 San Pat. R.F.,P.B. Limestone B AC-10 TFA 1.020 2.571

10 17 Robertson G.H., D.F. Silicious D AC-20 TXGulf 1.027 2.597 4.9 11 16 San Pat. G.H.,O.P. Limestone D AC-20 Gulf St 1.024 2.573 4.3 12 19 Panola G.H.,S.P. Silicious c AC-20 Lion 1.022 2.587 5.3 13 21 Hidalgo FD.,BG.,M.C. D AC-20 TFA 1.022 2.542 5.2

* Crushed Gravel and Limestone

BG. = Ballenger O.P. = Owens Prop. Filed Sand B.H. = Boorhem Fields P.B. = Parker Brothers B.I. = Bay Incorporation R.F. = Rodriguez Filed Sand C.M. = Colorado Materials R.W. = Redland Worth D.F. = Downing Field Sands S.F. = Scholemen Field Sand FD. = Fordyce S.P. = Simms Pit Field Sand F.P. = Farris Pit Field Sand T.F. = Tyne Pit Field Sand G.H. = Gifford Hill T.P. = Transpeco H.F. = Hinze Field T.X. = TXI Field Sand M.C. = Monte Cristo Filed Sand W.F. = Wederrtier Field Sand

32

TABLE A-2. EFFECTIVE SPECIFlC GRAVITY OF THE COMBINED AGGREGATES AT VARIOUS ASPHALT CONTENTS FOR DIFFERENT PROJECTS

Ge Ge Ge Ge Asphalt From From From From Ge HighAC Content RICE RICE Backcalculation Backcalculation From Used For

Project (%) (Uncorrected) (Corrected) (Uncorrected) (Corrected) C-14 Backcalculation

1 4.0 2.617 N!A 2.581 N!A 2.567 8.0 5.0 2.596 N!A 5.5 2.597 N/A 5.8 2.576 N!A 6.0 2.603 N!A 7.0 2.589 N/A 8.0 2.581 N/A

2 4.0 2.602 N/A 2.547 N/A 2.536 8.0 5.0 2.577 N/A 6.0 2.562 N/A 7.0 2.549 N/A 8.0 2.547 N/A

3 5.0 2.645 N/A 2.618 N/A 2.604 8.0 6.0 2.621 7.0 2.634 8.0 2.618

4 6.2 2.545 2.534 2.562 2.551 2.540 9.0 9.0 2.562 2.551

5 4.0 2.640 5.0 2.641 N/A 2.659 N!A 2.636 7.0 6.0 2.662 7.0 2.659

6 5.3 2.583 2.563 2.580 2.561 2.554 9.0 7 6.0 N/A 2.595 N/A N/A 2.613 N/A g 6.0 2.603 2.598 N/A N/A 2.594 N/A 9 6.0 2.640 2.619 N/A N/A 2.606 N/A

10 4.9 2.624 2.609 2.611 N/A 2.616 6.5 11 4.3 2.618 2.593 N!A N/A 2.603 N/A

12 4.0 2.599 2.596 2.625 2.611 2.619 7.0 5.0 2.611 2.596 5.3 2.605 2.595 6.0 2.599 2.590 7.0 2.625 2.611 8.0 2.642 2.589

13 4.5 2.595 2.583 2.614 2.598 2.578 7.0 5.0 2.588 2.577 5.2 2.605 2.599 5.5 2.601 2.593 6.0 2.615 2.597 6.5 2.601 2.585 7.0 2.614 2.598 7.5 2.592 2.566

TABLE A-3. STATISTICAL DATA ON DIFFERENCES BETWEEN EFFECTIVE SPECIFIC GRAVITY OF AGGREGATES FROM DIFFERENT METHODS

33

When Algebraic Values Are Used: When Absolute Values Are Used:

(Uncorrected) (Corrected) (Uncorrected) (Uncorrected) (Corrected) (Uncorrected) RICE RICE RICE RICE RICE RICE C-14 C-14 B.C.• C-14 C-14 B.C.•

Count 12 9 9 Count 12 9 9 Standard Dev 0.025 0.013 0.030 Standard Dev 0.020 0.007 0.016 Average 0.021 -0.004 0.006 Average 0.025 0.011 0.024 Minimum -0.020 -0.023 -0.025 Minimum 0.004 0.004 0.003 Maximum 0.066 0.013 0.055 Maximum 0.066 0.023 0.055

*B.C. :The Backcalculation Method

34

TABLE A-4. THEORETICAL MAXIMUM SPECIFIC GRAVITIES THROUGH DIFFERENT METHODS AT VARIOUS ASPHALT CONTENTS

Asphalt Gt Back- Back- HigbAC Content RICE RICE Theory cak:ulatlon calculation Used for

Project (%) (Uncorrected) (Corrected) C-14 (Uncorrected) (Uncorrected) (Corrected) Backcalculation

4.0 2.456 N/A 2.414 2.392 2.426 N/A 8.0 5.0 2.402 N/A 2.378 2.358 2.390 N/A 5.5 2.385 N/A 2.361 2.341 2.372 N/A 6.0 2.372 N/A 2.344 2.324 2.355 NIA 7.0 2.327 N/A 2.310 2.291 2.321 N/A 8.0 2.288 N/A 2.278 2.260 2.288 N/A

2 4.0 2.452 N/A 2.396 2.390 2.405 N/A 8.0 5.0 2.397 N!A 2.363 2.357 2.372 N/A 6.0 2.352 N/A 2.331 2.326 2.340 NIA 7.0 2.310 N!A 2.300 2.295 2.308 N/A 8.0 2.278 N/A 2.270 2.265 2.278 N/A

3 5.0 2.451 N/A 2.417 N/A 2.429 N/A 8.0 6.0 2.397 N/A 2.383 N!A 2.394 N/A 7.0 2.373 N/A 2.350 N/A 2.361 N/A 8.0 2.328 N/A 2.318 N/A 2.328 N/A

4 6.2 2.332 2.323 2.328 2.297 2.345 2.336 9.0 9.0 2.259 2.251 2.244 2.216 2.259 2.251

5 4.0 2.483 NIA 2.479 2.430 2.499 N/A 5.0 2.447 N/A 2.443 2.396 2.462 N/A 7.0 6.0 2.428 N/A 2.408 2.362 2.426 N/A 7.0 2.391 N/A 2.374 2.330 2.391 N/A

6 5.3 2.389 2.373 2.366 2.322 2.387 2.372 9.0

7 6.0 N/A 2.376 2.390 2.327 N!A N/A N/A

8 6.0 2.381 2.377 2.374 2.366 NIA N/A NIA

9 6.0 2.410 2.394 2.384 2.356 N/A N/A N/A

10 4.9 2.438 2.426 2.432 2.416 2.427 N/A 6.5

11 4.3 2.450 2.429 2.438 2.412 N/A N!A N!A

12 4.0 2.448 2.445 2.465 2.438 2.470 2.458 7.0 5.0 2.423 2.410 2.429 2.403 2.434 2.423 5.3 2.407 2.399 2.419 2.393 2.423 2.413 6.0 2.379 2.372 2.394 2.369 2.399 2.388 7.0 2.365 2.355 2.361 2.337 2.365 2.355 8.0 2.345 2.306 2.328 2.305 2.332 2.322

13 4.5 2.427 2.417 2.413 2.383 2.443 2.430 7.0 5.0 2.404 2.395 2.396 2.366 2.425 2.412 5.2 2.411 2.406 2.389 2.360 2.418 2.405 7.0 5.5 2.397 2.391 2.379 2.350 2.408 2.395 6.0 2.391 2.377 2.362 2.334 2.391 2.378 6.5 2.364 2.351 2.346 2.318 2.374 2.361 7.0 2.357 2.345 2.330 2.302 2.357 2.345 7.5 2.324 2.305 2.314 2.287 2.341 2.329 9.0

35

TABLE A-5. STATISTICAL DATA ON DIFFERENCES BETWEEN THEORETICAL MAXIMUM SPECIFIC GRAVITIES FROM DIFFERENT METHODS

Wben Algebraic Values Are Used: (Uncorrected)

(Uncorrected) (Corrected) (Uncorrected) (Corrected) (Uncorrected) (Corrected) RICE RICE RICE RICE RICE RICE RICE C-14

(Corrected) C-14 C-14 Gt* Gt* B.C.** B.C.** Gt* RICE -Count 40 22 36 21 30 13 37 21

Stand. Dev O.ol5 0.013 0.017 0.017 0.018 0.008 0.013 0.008 Average O.ol5 -0.003 0.040 0.026 0.001 -0.011 0.026 0.012 Minimum -0.017 -0.022 0.010 0.001 -0.022 -0.024 0.005 0.003 Maximum 0.056 0.017 0.067 0.051 0.047 0.001 0.062 0.039

When Absolute Values Are Used:

(Uncorrected) (Uncorrected) (Corrected) (Uncorrected) (Corrected) (Uncorrected) (Corrected) RICE

RICE RICE RICE RICE RICE RICE C-14 (Corrected)

C-14 C-14 Gt* Gt* B.C.** B.C.** Gt* RICE

Count 40 22 36 21 30 13 37 21 Stand. Dev 0.011 0.007 0.017 O.Dl7 0.010 0.007 0.013 0.008 Average O.Dl8 0.011 0.040 0.026 0.014 0.011 0.026 0.012 Minimum 0.004 0.001 0.010 0.001 0.000 0.001 0.005 0.003 Maximum 0.056 0.056 0.067 2.451 2.451 0.051 0.062 0.062

* G 1: Theoretical maximum specific gravity from the theoretical approach based on the bulk specific gravity of the combined aggregates

**B.C.: Backcalculation maximum specific gravity from a RICE test performed at a high asphalt content

TABLE A·6. THEORETICAL MAXIMUM SPECIFIC GRAVITIES FOR DIFFERENT PROJECTS AT DESIGN ASPAHLT CONTENT

OR AT AN ASPHALT CONTENT CLOSE TO THE DESIGN VALUE

(Uncorrected) (Corrected) (Uncorrected) Gt Project RICE RICE Backcalculation C-14 Theory -

1 2.372 2.355 2.344 2.324 2 2.352 2.340 2.331 2.326 3 2.397 2.394 2.383 4 2.332 2.323 2.345 2.328 2.247 5 2.447 2.462 2.443 2.396 6 2.389 2.373 2.387 2.366 2.322 7 2.376 2.390 2.327 8 2.381 2.377 2.374 2.366 9 2.410 2.394 2.384 2.356

10 2.438 2.426 2.427 2.432 2.416 11 2.450 2.429 2.438 2.412 12 2.407 2.399 2.423 2.419 2.393 13 2.411 2.406 2.418 2.389 2.360

36

TABLE A-7. DIFFERENCES BETWEEN THEORETICAL MAXIMUM SPECIFIC GRAVITIES AT DESIGN ASPHALT OR AT AN ASPHALT CONTENT CLOSE TO THE DESIGN VALUE

(Uncorrected) (Uncorrected) (Corrected) (Uncorrected) (Corrected) (Uncorrected) RICE

RICE RICE RICE RICE RICE C-14 (Corrected)

Project C-14 C-14 Gt* Gt* B.c.•• Gt• RICE

1 0.028 0.048 0.020 2 0.021 0.026 0.005 3 0.014 4 0.004 ....{).005 0.085 0.076 ....{).022 0.081 0.009 5 0.004 0.051 0.047 6 0.023 0.007 0.067 0.051 ....{).014 0.044 0.016 7 -D.014 0.049 0.063 8 0.007 0.003 O.Q15 O.Oll 0.008 0.004 9 0.026 0.010 0.054 0.038 0.028 0.016

10 0.006 ....{).006 0.022 0.010 ....{).001 0.016 0.012 11 0.012 ....{).009 0.038 0.017 0.026 0.021 12 ....{).012 ....{).020 0.014 0.006 ....{).024 0.026 0.008 13 0.022 0.017 0.051 0.046 ....{).012 0.029 0.005

Count 12 9 11 10 5 12 8 Stand. Dev 0.012 0.013 0.023 0.026 0.010 0.023 0.006 Average 0.013 -D.002 0.043 0.030 ....{).015 0.033 0.011 Minimun ....{).012 ....{).02 0.014 0.006 ....{).024 0.005 0.004 Maxinum 0.028 O.Gl7 0.085 0.076 ....{).001 0.081 0.021

* Gt: Theoretical maximum specific gravity from the theoretical approach based on the bulk specific gravity of the combined aggregates

**B.C.: Backcalculation maximum specific gravity from a RICE test performed at a high asphalt content

37

TABLE A-8. RELATIVE DENSITIES THROUGH DIFFERENT METHODS FOR PROJECTS FOR WffiCH RICE TESTS AT DIFFERENT ASPHALT CONTENTS ARE AVAILABLE

Asphalt Relative Density (%) Based On

Content (Uncorrected) (Corrected) (Uncorrected) (Corrected) Project (%) RICE RICE B.C.* B.C.* C-14 Gt**

4.0 89.5 90.6 91.1 91.9 5.0 93.0 93.5 93.9 94.7 5.5 94.0 94.5 95.0 95.8 6.0 95.3 96.0 96.4 973 7.0 98.3 98.6 99.0 99.8 8.0 99.6 99.6 100.1 100.9

2 4.0 91.6 93.4 93.8 94.0 5.0 94.8 95.8 96.2 96.4 6.0 98.3 98.8 99.2 99.4 7.0 99.6 99.7 100.0 100.3 8.0 100.4 100.4 100.8 101.0

3 5.0 94.1 95.0 95.4 6.0 97.2 97.3 97.8 7.0 98.2 98.7 99.2 8.0 99.6 99.6 100.0

5 4.0 91.6 91.0 91.7 93.6 5.0 94.3 93.7 94.5 96.3 6.0 96.0 96.1 96.8 98.7 7.0 97.2 97.2 97.9 99.8

12 4.0 92.1 92.2 913 91.7 91.5 92.5 5.0 96.1 96.6 95.7 96.1 95.8 96.9 6.0 97.2 97.5 96.4 96.8 96.6 97.6 7.0 99.0 99.4 99.0 99.4 99.2 100.2 8.0 99.3 101.0 99.8 100.2 100.0 101.0

13 4.5 93.4 93.8 92.8 93.3 93.9 95.1 5.0 94.6 95.0 93.8 94.3 95.0 96.2 5.5 95.6 95.9 95.2 95.7 96.4 97.5 6.0 96.7 97.3 96.7 97.2 97.9 99.1 6.5 97.8 98.4 97.4 97.9 98.6 99.8 7.0 98.2 98.7 98.2 98.7 99.4 100.6 7.5 99.5 100.3 98.8 99.3 100.0 101.1

* G t= Theoretical maximmn specific gravity from the theoretical approach based on the bulk specific gravity of the combined aggregates

**B.C.: Backcalculation maximmn specific gravity from a RICE test performed at a high asphalt content

38

TABLE A-9. DIFFERENCES IN RELATIVE DENSITIES THROUGH DIFFERENT METHODS

Corrected C-14 C-14 RICE Gt Gt

(Uncorrected) (Corrected) (Uncorrected) (Uncorrected) (Corrected) Project RICE RICE RICE RICE RICE

1 1.6 2.4 0.9 1.7 1.0 1.8 1.1 2.0 0.7 1.5 0.5 1.3

2 2.2 2.4 1.4 1.6 0.9 1.1 0.4 0.7 0.4 0.6

3 1.3 1.0 0.6 0.4

5 0.1 2.0 0.2 2.0 0.8 2.7 0.7 2.6

12 -0.6 -0.7 0.1 0.4 0.3 -0.2 -0.8 0.5 0.8 0.3 -0.6 -0.9 0.3 0.4 0.1

0.2 -0.2 0.4 1.2 0.8 0.7 -1.0 1.7 1.7 0.1

13 0.6 0.2 0.4 1.7 1.4 0.3 -0.0 0.4 1.5 1.2 0.7 0.5 0.2 1.9 1.7 1.2 0.6 1.6 2.4 1.8 0.8 0.2 0.5 1.9 1.4 1.2 0.6 1.5 2.3 1.8 0.4 -0.4 0.8 1.6 0.8