ENGINEERING RESEARCH INSTITUTE THE UNIVERSITY OF MICHIGAN ANN ARBOR Final Report BITUMINOUS MIXTURES CONTAINING FLY ASH'* ~ - DETROT,,,,. MICGAN Julyt 2519 DERI v.M.C.HI'GN v~~Jl 19.56..

5-Z <v ll-. Il- R i'-D —, -q

The University of Michigan * Engineering Research Institute TABLE OF CONTENTS Page ABSTRACT i OBJECTIVE iii INTRODUCTION 1 SCOPE 1 MATERIALS 2 TESTS EMPLOYED 2 SELECTION OF MIXTURES FOR COMPARISON 3 DISCUSSION OF TEST DATA 4 Stability 4 Compressive Strength 5 Flow Unit Weight Percent Voids in Aggregate Filled with Asphalt 5 Resistance to Water Action 6 Interrelationships 6 CONCLUSIONS 7 APPENDIX A. MATERIALS 13 APPENDIX B. TEST METHODS AND PROCEDURES 18 APPENDIX C. TEST DATA AND CHARACTERISTIC CURVES 23 APPENDIX D. DATA FROM IMMERSION-COMPRESSION TESTS I- -ii

The University of Michigan ~ Engineering Research Institute ABSTRACT Mineral fillers of pulverized limestone and fly ash, high and low carbon, were compared. Specimens of asphaltic concrete, similar in all respects except for the fillers, were tested for (1) stability and flow by the Marshall method and (2) compressive strength, dry and after periods of immersion in water. On the average, mixtures containing limestone dust gave the highest stabilities, followed in order by those containing high-carbon fly ash and low-carbon fly ash, although some characteristic other than carbon content seems to be responsible for differences in behavior of the fly-ash fillers. Mixtures within the range of Michigan State Highway Department criteria and current practice can be prepared with any of the fillers studied, although those containing fly ash from some sources seem to be more critical with respect to mixture control requirements. The compressive strengths of the mixtures were not materially affected by any of the fillers and all mixtures containing the several fillers were highly resistant to water action. OBJECTIVE This project has as its objective the determination of relative efficiencies of fly-ash fillers and asphaltic concrete paving mixtures., - iii

The University of Michigan ~ Engineering Research Institute INTRODUCTION The specifications for dense-graded asphaltic concrete of the Michigan State Highway Department require the use of a fine mineral filler. Pulverized limestone is the most widely used material for this purpose, but fly ash is available in some areas and is acceptable if it meets certain requirements, including a carbon content between 7 and 12% by weight. These limits of carbon content were established some years ago and a large amount of fly ash conforming to them has been used from plants of The Detroit Edison Company. Two plants of the Company now produce fly ash containing considerably less than 7% carbon. It is the objective of this investigation to compare the characteristics of asphaltic concrete mixtures in which are used as fillers: (1) limestr dust, (2) low-carbon fly ash, and (3) high-carbon fly ash. The sources of fillers were as follows: National Lime and Stone Co. limestone Waukesha Products Co. limestone The Detroit Edison Co. fly ash Marysville (10.3% carbon) The Detroit Edison Co. fly ash Conners Creek (9.9% carbon) The Detroit Edison Co. fly ash Trenton Channel (2.9% carbon) The Detroit Edison Co. fly ash St. Clair (3.05% carbon) SCOPE Since the principle use of filler is in the dense-graded asphaltic concrete employed in surfaces constructed under Michigan State Highway Depart= ment specifications, the mixtures used in this study were prepared in accordance with these requirements. The typical composition is as follows: Asphalt 5.5% by weight Filler 6.0% by weight Fine aggregate 3350.5% by weight Coarse aggregate 55.0% by weight 1' ~"

The University of Michigan * Engineering Research Institute The current design is based on density-stability studies made a number of years ago and modified by experience on past construction of this type. This design conforms closely to the following criteria when tested by the Corps of Engineers Marshall stability method:* Stability (Marshall) 1500 lb or above Voids 3 to 5% Flow (Marshall) 20 (max) Percentage of voids of the aggregate filled with asphalt 75-85% MATERIALS With the exception of the two low-carbon fly ashes, all the materials used conform to Michigan State Highway specifications for use in dense-graded asphaltic concrete. In order to preserve uniformity, the coarse aggregate was divided into sizes and recombined for each batch of mixture to a typical gradation within the State specifications. Tests of the aggregates, fillers, and asphalt are given in Appendix A. TESTS EMPLOYED In addition to tests on the constituent materials, compacted mixtures were tested for (1) strength or "stability," (2) flow or deformation, and (3) strength retention after immersion in water. Details of specimen preparation and testing procedures are described in Appendix B. For determinations (1) and (2), the Marshall method of specimen preparation and testing was used. In this method mixtures are prepared of identical composition except for the asphalt content which is varied in 0.5% steps from very lean to very rich. Cylindrical specimens are compacted by the prescribed procedure and the density of each is determined. From the known specific gravities and percentages of the constituent materials and the specific gravity of the compacted specimen, there are computed (1) the voids in the specimen, (2) the percentage of voids of the aggregate filled with asphalt, and (3) the weight per cubic foot of the compacted mixture. Each of these computed characteristics is plotted against asphalt content. * Described by Gayle McFadden and Walter C. Rickets, Proceedings Assn. of Asphalt Paving Technologists, 17:92-113 (1948). 2u

The University of Michigan * Engineering Research Institute The compacted specimens are tested for "stability" or strength and "flow" or deformation in the Marshall apparatus. The specimen on edge is pressed between upper and lower testing heads conforming to the curvature of the specimen but spaced 3/4 in. apart, thus permitting deformation. An attachment permits measurement of deformation, recorded as "flow," as the load is apt plied at a standard rate. The "stability" of the specimen is the maximum load which the specimen will sustain, and the "flow" value is read as the amount of deformation at the instant the maximum load value is obtained. Curves of both stability and flow are plotted against asphalt content. These constitute two additional characteristic curves. Mixtures were prepared and tested with three filler contents (4%, 6%, and 8%) by weight for each of the six fillers of this investigation. The five sets of characteristic curves and the data from which they were plotted for each of the six fillers are presented in Appendix C, Tables I through V and Figs. 1 through 30. The relative resistance to water of the six fillers was determined by the immersion-compression test developed by the U. S. Bureau of Public Roads. In this test, test specimens of the mixtures are prepared by doubleplunger direct compression without tamping. The specimens are tested in unconfined compression at 77cF, three without exposure to water, three after 4 days' immersion in water, and three after 14 days' immersion in water. The compressive strength of the specimens after water immersion divided by the compressive strength of the specimens which are not exposed to water gives the percentage of retained strength or index of resistance. SELECTION OF MIXTURES FOR COMPARISON Asphaltic paving mixtures are so designed that when laid the pavement will contain some air voids. One in which the voids of the aggregates are entirely filled with asphalt will deform under heavy traffic and prove unsatisfactory. Thus, the Michigan State Highway Department designs its mixtures to contain 3 to 5% voids when compacted in the laboratory by the methods used in this study. For the purpose of selecting comparable mixtures conforming to State Highway Department practice, compositions, compacting to 4% voids were chosen. From the characteristic void-asphalt (Appendix C) curves for mixtures containing 4, 6, and 8% of filler from each source, the asphalt content resulting in 4% voids (average of the 35-5 limits of the State Highway Department) was de"Standard Method of Test for Effect of Water on Cohesion of Compacted Bituminous Mixtures," ASTM, D1075-54. 3

The University of Michigan ~ Engineering Research Institute termined. At this asphalt content the remaining properties (stability, flow, weight per cu ft, and percent aggregate voids filled with asphalt) were picked from the corresponding characteristic curves. The required amounts of asphalt to give 4% voids did not vary a great deal for the several fillers as will be noted in Table I. The water-immersion tests were made on mixtures typical of Michigan State Highway Department practice with 6% filler and with the asphalt contents which would give 4% voids when compacted by the Marshall method. The asphalt contents were those of Table I, for 6% filler. DISCUSSION OF TEST DATA The following comments are based on the test characteristics of mixtures containing 4% voids and are identical in composition except for the variation in asphalt content needed to produce 4% voids and, also, for the sources of filler. The test characteristics of those mixtures containing 6% filler are considered most significant, since most Michigan State Highway Department mixtures carry close to this amount. STABILITY The Marshall test for stability is one of a number of arbitrary strength tests. It has been developed for dense-graded mixtures containing coarse aggregates and its use is rather widespread. Test results are not expressed in terms of shear or compressive stress, but are merely the maximum total load in pounds which a specimen is able to resist under the conditions of the specified test procedure. The criteria of satisfactory behavior of mixtures in this test are established by the observations on test tracks and highways. Exact duplication of test results is not possible. As shown by Table I and Fig. 1, all mixtures containing the six fillers at 4% void content possess stabilities over 1500 lb, with the exception of that one with 8% of the low-carbon fly ash from the St. Clair plant. The stability of the mixture containing high-carbon fly ash from the Marysville plant is similar to those of the two limestone mixtures. The mixtures containing Conners Creek fly ash, also high carbon, is similar in stability to the two low-carbon fly-ash mixtures. The average of the stabilities of the highcarbon fly-ash mixtures is a little below the average of those containing limestone dust, and the stability of mixtures in which low carbon is used is lowest 4~~~~~~~~lws

The University of Michigan ~ Engineering Research Institute COMPRESSIVE STRENGTH In discussing the strength or stability of the mixtures, attention should be directed to the unconfined compressive strengths of dry specimens which were determined in connection with the immersion-compression test series. The specimens were identical with those containing 6% filler in Table I, except that they contained somewhat higher percentages of voids. These test results are given in the first column of Table III. Whereas in the Marshall stability test the specimen is partially confined, the specimens whose strengths are reported in Table III were entirely without lateral support. The unconfined compressive strength is often regarded as a valid criterion for design. In this study, the unconfined compressive strengths of all mixtures, regardless of the source and nature of the filler, were not significantly different, indicating that perhaps the mineral filler in the amounts used in such mixtures is less significant with respect to strength than the other components. FLOW Flow measurements are made as deformation progresses in the Marshall stability test. The deformation at maximum load is recorded. Plastic mixtures deform more and rigid mixtures deform less. A certain degree of plasticity is desired although no correlation exists between the laboratory results obtained on this test and the service behavior of a mixture. The Michigan State Highway Department practice is to design mixtures which have flow values of 20 or below. All the mixtures at 4% voids possessed flow values well under this limit as shown in Table II and Fig. 2. UNIT WEIGHT Figure 3 shows the unit weight of mixtures containing the six fillers at 4% voids. These weights do not vary greatly and are not significant in this study. They should be closely similar since the mixtures are compacted to the same percentage of solid density. The only variables are the minor variations in asphalt content and the specific gravities of the several fillers. PERCENT VOIDS IN AGGREGATE FILLED WITH ASPHALT The set of values, shown in Fig. 4, is of interest only in indicating that compaction to 4% voids, and with the typical 6% filler used by the Michiga State Highway Department, the aggregate voids are likely to be a trifle less than 75% filled with asphalt. 5

The University of Michigan ~ Engineering Research Institute RESISTANCE TO WATER ACTION When mixed with asphalt and stirred in water under the conditions of the water-asphalt preferential test, all the six fillers proved to be satisfactory. For the water-immersion compression tests, mixtures were chosen which contain 6% filler and the appropriate asphalt content to compact to 4% voids by the Marshall method. These mixtures are identical with those whose Marshall stabilities and asphalt contents are shown in the middle columns of Table I. However, under the double-plunger compression method of compaction prescribed for the preparation of immersion-compression specimens, the voids could not be reduced to 4%. Rather, the average voids of the specimens were as follows: Average Voids of Immersion-Compres sion Filler Nature Specimens, % National Lime and Stone Limestone 6.70 St. Clair Low-carbon fly ash 6.02 Trenton Channel Low-carbon fly ash 5.64 Conners Creek High-Carbon fly ash 6.11 Marysville High-carbon fly ash 5.39 Waukesha Products Limestone 6.41 The higher percentage of voids above 4% is not of consequence. The two strergth tests (Marshall stability and direct compression) are so different that they cannot be expected to produce comparable results,. The higher percentage of voids might be expected to facilitate the penetration of water to the interior of the specimen and in this manner somewhat increase the severity of the test. Kctadily the results of thle compresfive-strength tests ceach series were remarkably uniform. They are shown in Table III and Figs. 5a and 5b. There i, no significant decrease in strength in mixtures containing any one of the six fillers. The retention, after 14 days' immersion, of 75% of the dry strength has been suggested as indicating satisfactory resistance to water action. All mixtures were above this limit. INTERRELATIONSHIPS With increases in asphalt content beyond a certain point, the stability and voids of a mixture decrease and the flow increases. There are indications in this study that the balance between the minimum Marshall stability of 1500 lb, the asphalt content, minimum desired voids, and percentage of fili ler as expressed by the current Michigan State Highway Department practice may be somewhat critical when fly ash from some of the sources is used as filler~ 6....

The University of Michigan * Engineering Research Institute For some reason, perhaps shape of the particles, mixtures containing 6% fly ash from St. Clair, Trenton Channel, and Conners Creek gave 1500-lb stability with 5.5% asphalt, and, at this point, contained 351, 3.5, and 3.1% voids, respectively. An increase in asphalt would reduce the stability below 1500 lb and also would reduce the voids below the design limit. The two limestone film lers and the Marysville fly ash tolerate 6%, or a little more, asphalt while still maintaining 1500-lb stability, although at this asphalt content the voids also would be critical. The following data, picked from the charts of Appendix C, indicate these relationships: Approximate Maximum Asphalt Content, with Corresponding Voids to Give Marshall Stability of 1500 lb Filler Asph. Voids Asph. Voids Asph. Voids National Lime and Stone 5.8 3.2 6 O+ 2.6 6.4 2.9 St. Clair 5.8 3.0 5.5 3.1 4.6 4.2 Trenton Channel 6.0 3.0 5.5 5.5 5-5 3.2 Conners Creek 6.1 2.6 5.5 3.1 5-1 357 Marysville 6.5+ 2.6 6.2 2.3 6.0 2.7 Waukesha Products 6.2 2.9 6.1 2.5 5.9 2.5 CONCLUSIONS The following conclusions are based on this rather limited study den signed to investigate the relative effects of certain fillers when used in a single, typical, dense-graded asphaltic concrete mixture. 1. The Marshall stabilities are somewhat affected by the fillers from the several sources. The limestone dusts gave highest stability values, followed in order by the high-carbon fly ash and the low-carbon fly ash. However, all the fillers do produce mixtures under current Michigan State Highway Department practices which possess stabilities of above the minimum design limit of 1500 lb. 2. In the unconfined compressive-strength test the source of filler was not a significant factor. Mixtures containing high-carbon fly ash and lowcarbon fly ash possessed strength at least equal to those containing limestone dust 5. Flow values of the mixtures determined by the Marshall test procedure show no significant difference attributable to the source of filler when other design criteria were satisfied~ 7

The University of Michigan * Engineering Research Institute 4. All the mixtures tested, regardless of the source of filler, were satisfactory with respect to resistance to water action as determined by the immersion-compression test. 5. There were some indications that mixtures containing fly ash from three sources are more critical with respect to design relationships between asphalt content, voids, stability, and flow than are those containing the limestone dusts and the fourth (Marysville) fly ash.

The University of Michigan ~ Engineering Research Institute 2500 WaMukesha Products > I -- Marysville I — 1 ~Nat'l Lime a Stone u 2000 - Q Conners Creek:E e - Trenton Channel 1500 l -J 5O' — Minimum value: g I Michigan State Highway Dept. practice St. Clair Cn I 000 I l 4.0 6.0 8.0 PERCENT FILLER Fig. 1. Stabilities of mixtures at 4% voids. TABLE I. STABILITIES OF MIXTURES AT 4% VOIDS, lb Filler Filler Asphalt Filler Asphalt Filler Asphalt National Lime and Stone 2060 5.4 2260 5.1 2080 5.25 St. Clair 1840 5.2 1820 4.9 1360 4.75 Trenton Channel 1740 5.3 1760 5.2 1660 5.15 Conners Creek 1660 5.2 1880 4.8 1840 4.9 Marysville 2300 5.25 2200 5.1 2300 5.1 Waukesha Products 2160 5.25 2500 5.0 2490 4.8 Average Stabilities of Mixtures at 4% Voids, lb 2 limestones 2110 2380 2285 2 high-carbon fly ash 1980 2040 2070 2 low-carbon fly ash 1790 1790 1510

The University of Michigan * Engineering Research Institute 20 Michigan State Highway Dept. practice (maximum flow) 0 8 I Treenton Channel Woukesha Products _J St. Clair Not'l Lime a Stone 10 l 4.0 6.0 8.0 PERCENT FILLER Fig. 2. Flow values of mixtures at 4% voids. TABLE II. FLOW VALUES OF MIXTURES AT 4% VOIDS (1/100 in.) Filler 4% 6% 8% National Lime and Stone 11.6 10.8 10.8 St. Clair 12.5 12.9 11.5 Trenton Channel 12.0 12.8 1355 Conners Creek 11.0 12.2 12.4 Marysville 11.4 11.3 12.5 Waukesha Products 10.8 11.2 12.7 10

The University of Michigan ~ Engineering Research Institute 0 o 152- Woukesha Products Not'I Lime 8 Stone <{:'- St. Clair J Conners Creek 150 3J ___ Trenton Channel I-1 |~~~~~~~~~~~~~~~~~ Marysville Z 148 4.0 6.0 8.0 PERCENT FILLER Fig. 3. Unit weight of mixtures at 4% voids. 0 Michigan State Highway Dept. 78 practice: 75-85% CO Nat'l Lime 8 Stone w I -J C") 76 o Z5 Marysville ITrenton Channel Fg Waukesh4. P ~aConners Creek aProducts [Waukesha Products 74 Creek ICJ W 72 4.0 6.0 8.0 PERCENT FILTER Fig. 4. Percent aggregate voids filled with asphalt at mixture voids of 4%. 11

The University of Michigan * Engineering Research Institute Nat'l St. Trenton Conners Waukesha 400 Lime B Stone Clair Channel Creek Marysville Products In 300 z 200 ct Oo 0 0 4 4 10 4 14 0 4 14 0 4 14 0 4 14 0.4 14 DAYS OF IMMERSION Fig. 5a. Strength of mixtures tested dry and after immersion. o 1.20 1.002,, a.80 o.60 o. 0 0 4 14 0 4 14 0 4 14 0 4 14 0 4 14 0 4 14 DAYS OF IMMERSION Fig. 5b. Ratio of strength of mixtures tested dry and after immersion. TABLE III. COMPRESSIVE TESTS OF DRY AND IMMERSED SPECIMENS Filler Dry 4-Day 14-Day D 4-Day 14-Day immersion Immersion y Immersion Immersion. National Lime and Stone 349 365 336 1.00 1.05 0.96 St. Clair 328 323 327 1.00 0.98 1.00 Trenton Channel 355 379 367 1.00 1.07 1.03

APPENDIX A MATERIALS

The University of Michigan ~ Engineering Research Institute SOURCES OF MATERIALS FILLER The fly-ash fillers used in this investigation were the by-products of the combustion of powdered coal from The Detroit Edison Company plants at St. Clair, Trenton Channel, Conners Creek, and Marysville, Michigan. The limestone fillers used were the standard products of the National Lime and Stone Company, Carey, Ohio, and the Waukesha Products Company,Waukesha, Wisconsin. FINE AGGREGATE The fine aggregates used were obtained from the Whittaker and Gooding pit in the vicinity of Ann Arbor, Michigan, and were being used as a fine aggregate in the production of bituminous concrete by the Ann Arbor Construction Company. COARSE AGGREGATE The coarse aggregate was crushed natural gravel from American Aggregates Corporation, Green Oaks, Michigan. It was furnished as coarse aggregates meeting Michigan State Highway Department requirements for 25A coarse aggregate ASPHALT The asphalt was a standard 85-100 penetration asphalt from the Lion Oil Company, Eldorado, Arkansas, and obtained from the Wayne County Road Commission from their supply on hand......~ ~14

SIEVE SIZES 270 200 140 80 5 DIAMETER IN MILLIMETERSo 0 0 0 0 0 A 4 3 00 100 0 0 000 0 ( O3 National Lime EN Stone Co. a St. Clair Plant 90 0 Trenton Channel Plant - N~~ Canners Creek Plant A Marysville Plant * Waukesha Products Co. 80__ ~70 z a.60 05 ww 40, 30 20w 10 FILLER MATERIALS. HYDROMETER ANALYSES OF MATERIAL PASSING NO. 40 SIEV Fig. 1NA. Size distribution of filler materials (only that portion passing no. 200siv used. in preparation of test mixtures).

The University of Michigan * Engineering Research Institute TABLE I-A. TESTS OF FILLER MATERIALS Fly Ash Trenton Conners Chemical Properties St. Clair Channel Creek Marysville. (percent by weight, water-free basis) Silicon Dioxide, Si02 39.1 46.0 40.7 44.5 Aluminum Oxide, A1203 25.0 27.7 21.9 Ferric Oxide, Fe203 23.1 17.0 21.9 Calcium Oxide, CaO -- 2.6 1.7 -a Magnesium Oxide, MgO 1.1 0.9 o0.9 1.5 Sulfur Trioxide, SO3 0.7 0.7 0.7 0.5 Loss on Ignition 3.05 2.9 9.9 10.3 Moisture 0.1 0.2 0.3 Specific Gravity 2.634 2.42 2.47 2.28 Water-Asphalt Preferential Test Satisfactory Satisfactory Satisfactory Satisfactory Limestone Dusts Chemical Properties National Lime and Stone Co. Waukesha Products Co. Specific Gravity 2.83 2.83 Water -Asphalt Preferential Test Satisfactory Satisfactory 16

The University of Michigan * Engineering Research Institute TABLE II-A. GRADATIONS OF AGGREGATES Coarse Aggregate Gradation Usead 1Michigan State Highway Department Sieve assing Specifications Retai.ne.d % Passing.% Passing 5/8 in. 0 100.0 100 1/2 in. 5.0 95.0 90-100 3/8 in. 41.2 53.8 -- No. 4 41.3 12.5 10-25 No. 10 12. e 0 0-10 Fine Aggregate Michigan State Highway Gradat ion Sieve Department Used Specifications Pass no..4 0 100 Pass no. 4, ret'd no. 10 2.6 0-5 Pass no. 10, ret'd no. 40 25.6 5-35 Pass no. 40, ret'd no. 80 46.2 30-60 Pass no. 80, ret'd no. 200 25.6 15-35 Pass no. 200 0 0-5 TABLE III-A. ANALYSIS OF ASPHALT CEMENT General Characteristics Semi-solid Specific Gravity at 25~C/250C 1.018 Penetration at 25~C, 100 g, 5 sec 84 Flash Point, Cleveland Open Cup, C. 352 Loss on Heating at 163~C, 5 hr, 50 g, % 0.184 Penetration of Residue from Loss on Heating, 25~C, 100 g, 5 sec 73 Ductility at 25~C, cm 110+ Solubility in CC14, g 99.93 Oliensis Spot Test Negative 17

APPENDIX B TEST METHODS AND PROCEDURES

The University of Michigan ~ Engineering Research Institute MARSHALL PROCEDURES FOR STABILITY AND FLOW TESTS PREPARATION OF -AGGREGATES A sufficient quantity of aggregates, to make the required number of specimens, was air-dried to substantially constant weight and then separated into the required size ranges by means of mechanical sieving. To maintain uniformity of gradation of the test specimens, the separate size fractions were recombined by weight into individual batches of sufficient volume to mold three specimens, 4 inches in diameter by 2-1/2 inches compacted height. PREPARATION OF MIXTURE AND COMPACTION OF SPECIMENS The weighed aggregate fractions of each batch were heated to 300'F and thoroughly mixed dry by stirring with a spatula. The 12-quart-capacity mixing bowl of a Hobart mechanical mixer was then charged with the batch of heated aggregate and a crater formed in the dry blended mixture; the required amount of bitumen, at a temperature of 3000F, was then weighed into the mixture. Mixing of the aggregate and bitumen, using a wire, or bird-cage-type mixing paddle, was immediately started, and, after a mixing period of 1-1/2 minutes, the mixer was stopped and the mixture of fines and bitumen adhering to the sides of the bowl and to the paddle was scraped back into the mixing area, after which mixing was continued until a total mixing time of 3 minutes was completed. The mixture was then placed in a shallow pan, covered, and placed in the oven until the compaction temperature was attained. Compaction was accomplished, with the mix at a temperature of 280 ~ 5~F, using standard Marshall test apparatus, with 50 blows applied with the compaction hammer on one side of the specimen, removing the base plate and collar, reversing the mold, reassembling, and applying the same number of blows to the opposite side of the specimen. After compaction, the mold containing the specimen was placed in cool water for a minimum of 2 minutes after which the sample was removed, by means of the sample extractor, from the mold and identified as to mix and specimen number. The specimens were allowed to cool to room temperature in air for approximately 24 hours, after which they were weighed in air and in water for specific-gravity determinations. In order to maintain close control, a maximum tolerance of 0.03, between the maximum and minimum range of specific gravities of the three specimens of each mix, was used and mixes not complying to this tolerance were repeated. STABILITY AND FLOW DETERMINATION The test specimens were brought to the test temperature of 1400F by immersing them in a water bath at 140~F for a period of one hour. Using the 19

The University of Michigan * Engineering Research Institute Marshall-type specimen mold holder, breaking head, loading jack, proving ring assembly, and flow meter, specimens were subjected to load applications at a constant rate to produce a uniform vertical movement of 2 inches per minute until failure of the specimen occurred. The maximum reading of the dial in the proving ring assembly, converted to pounds, is the stability value for the individual specimen, and the flow meter reading, expressed in hundredths of an inch, is the flow, or deformation under stability loading, value of the specimen. The stability value varies directly with the thickness, or height, of the specimen. Therefore, it was necessary to correct the stability values for specimens of a thickness varying from the standard 2-1/2 inches. Flow values do not vary appreciably with change in thickness of the specimen and therefore no corrections were made for the flow value. SELECTION OF MIXES From the test results obtained, and the values computed the average results of the three specimens for each filler, at each bitumen and filler content, were plotted and shown in Figs. 1 through 30, Appendix C. On the basis of these curves, the median of the standard void range, 3 to 5%, with 6% filler content was selected as the basis for determining the bitumen content for each of the six fillers. WATER-IMMERSION AND UNCONFINED COMPRESSION-TEST PROCEDURES Mixtures with bitumen contents as determined from the 4% void curves with 6% filler were used for the water-immersion test series. PREPARATION OF AGGREGATES A sufficient quantity of aggregates, to make the required number of specimens, was air-dried to substantially constant weight and then separated into the required size ranges by means of mechanical sieving. To maintain uniformity of gradation of the aggregates, the separate size fractions were recombined by weight into individual batches of sufficient size to mold one specimen, 4-1/2 inches in diameter by 4-1/2 inches compact height. PREPARATION OF MIXTURE AND COMPACTION OF SPECIMENS The weighed aggregate fractions of each batch were heated to 3000F and thoroughly mixed dry by stirring with a spatula. The 12-quart-capacity mixing bowl of a Hobart mechanical mixer was then charged with the batch of 20

The University of Michigan * Engineering Research Institute heated aggregate and a crater formed in the dry blended mixture; the required amount of bitumen, at a temperature of 3000F, was then weighed into the mixture. Mixing of the aggregate and bitumen, using a wire, or bird-cage-type mix ing paddle, was immediately started, and, after a mixing period of 1-1/2 minutes, the mixer was stopped and the mixture of fines and bitumen adhering to the sides of the bowl and to the paddle was scraped back into the mixing area, after which mixing was continued until a total mixing time of 3 minutes was. completed. The mixture was then placed in a shallow pan, covered, and placed in the oven until the compaction temperature was attained. Compaction was accomplished, with the mixture at a temperature of 280 + 50F, by placing it in a molding cylinder which, together with the top and bottom plunger, had been preheated in the oven maintained at the molding temperature. With the bottom plunger in place and the molding cylinder supported temporarily on brass bars, the mixture was spaded around the inside of the mold with a spatula to reduce surface "honeycomb." It was then compressed between the top and bottom plungers under an initial load of approximately 150 psi to set the mixture against the sides of the mold. The support bars were then removed to permit full doubles plunger action and the full molding load of 3000 psi, or 48,000 lb, was applied and maintained for two minutes. After removal from the mold, specimens were oven cured 24 hours at 1400F and thereafter brought to test temperature, 770F, by storing in air at this temperature for approximately 18 hours. COMPRESSION AND WATER-IMMERSION TESTS Nine specimens for each of the six fillers were prepared, and, after determining the oven-dry weight, the surface-dry weight, and the weight in water for each specimen, the nine specimens were divided, at random, into groups of three specimens each. One group was immediately tested in axial compression without lateral support at a uniform rate of vertical deformation of 0005 inch per minute per inch of height. The compressive strength for each specimen was computed in pounds per square inch by dividing the maximum vertical load obtained during deformation at the rate specified by the original cross-sectional area of the test specimen. The second group of specimens was immersed in dis= tilled water for four days at 120~F, then transferred to a second water bath, maintained at 770F, for a period of two hours. The compressive strength was then determined. The third group of specimens was immersed in distilled water for 14 days at 1200F, then transferred to a second water bath, maintained at 77F,| for a period of two hours. The compressive strength of the third group was then determined RETAINED STRENGTH The numerical index of resistance of the bituminous mixtures to the 21

The University of Michigan ~ Engineering Research Institute possible detrimental effect of water was determined by dividing the strength of the specimens after immersion by the strength of the specimens which have not been immersed. WATER-ASPHALT PREFERENTIAL TEST* This test is used to determine the relative affinity for water or bin tumen of a mineral filler which is to be used in a bituminous mixture. Fifty millileters af asphaltic oil (SC-3A), heated to 140~F, was placed in an eight-ounce sample bottle with wide mouth and screw cap. Ten grams of material passing no. 200 sieve was added to the bituminous material. The bottle was placed in a hot-water bath maintained at a temperature of 140'F and the mixture stirred for five minutes with a mechanical mixer, which revolved at a controlled speed of 1500 rpm. The mixer was equipped with a stirring apparatus containing paddles and a specially designed dispersion cup equipped with baffles. After five minutes, 100 ml of distilled water at 140~F was added and the mixture stirred for a second five-minute period. After the sample was allowed to settle for approximately 24 hours the amount of clean dust in the bottom of the bottle was observed. If the mineral filler is of satisfactory quality it will remain in the bituminous material and the water will remain clear. If the dust is of a poor quality, showing a strong affinity for water, the dust will be concentrated in the bottom of the bottle having separated from the oil. If the dust is entirely satisfactory there will be 0% separation; if entirely unsatisfactory there will be 100% separation. A separation of up to an estimated 25%, based on the total sample is considered satisfactory for this test. ~ST-35 method of test for water-asphalt preferential of mineral filler. W. S. housel, Applied Soil Mecehaiics, Laboratory Manual of Soil Testing Proced:J~aer 1948... 22

APPENDIX C TEST DATA AND CHARACTERISTIC CURVES

The University of Michigan * Engineering Research Institute I0 ml - 4% Filler i - 6% Filler ) - 8% Filler 8 06 cn R9 |\,Michigan State Highway Dept. l.4 4design limits z 2 I i l I 4.0 4.5 5.0 5.5 60 6.5 PERCENT ASPHALT Fig. 1-C. Characteristic curves: % asphalt vs void in mix. Filler: National Lime and Stone Co. -. 24

The University of Michigan * Engineering Research Institute I0 ['l - 4% Filler ~ -6- %Filler X - 8%Fil ler X 8 IC) 0 Fille Michiganr: State ighway Dept. limClair Plant a ~~.... 25.....

The University of Michigan * Engineering Research Institute I0 m -4% Filler A -6% Filler 8 - 8% Filler x m I0o i> 4b Michigan State Highway Dept. limits L 4 w 04.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 3-C. Characteristic curves: % asphalt vs voids in mix. Filler: Trenton Channel Plant 26

The University of Michigan * Engineering Research Institute 10 ml - 4% Filler A -- 6% Filler @)- 8% Filler 8 06' C.f.''Michigan State Highway Dept. limits z - I I I I I 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 4-C. Characteristic curves: % asphalt vs voids in mix. Filler: Conners Creek Plant 27

The University of Michigan * Engineering Research Institute I0 l -4% Filler j -6% Filler ( - 8% Filler 8 4 -J(/ ro 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 5-C. Characteristic curves: % asphalt vs voids in mix. Filler: Marysville Plant 28

The University of Michigan * Engineering Research Institute i' - 4% Filler.~ - 6% Filler a ()- 8% Filler 10 8 PERCENT ASPHALT Fig. 6-C. Characteristic curves': asphalt vs voids in mix. Filler: Waukesha Products Co..............~ O29

The University of Michigan * Engineering Research Institute 3000 E -4% Filler l\ -6% Filler / ( — 8% Filler 2500 Q-J 2000' 1500 Michigan State Highway Dept. minimum 1000 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 7-C. Characteristic curves: % asphalt vs stability. Filler: National Lime and Stone Co. 30

The University of Michigan * Engineering Research Institute 3000 m1 -4% Filler j - 6% Filler 2500 -8% Filler 2500 m 2000 2000 ~ Filler:Michigan State ig Dept. Cair Plant I000 ~ 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 8-C. Characteristic curves:' asphalt vs stability. Filler: St. Clair Plant..... ~~~31

The University of Michigan * Engineering Research Institute 3000 -. 4%o Filler -- 6% Filler - 8% Filler 2500 2000 |u _ D - 1500 -' IDo TMichigan State Highway Dept. minimum 1000 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 9-C. Characteristic curves: % asphalt vs stability. Filler: Trenton Channel Plant.... 32

The University of Michigan ~ Engineering Research Institute 3000. [m - 4% Filled A.- 6% Filled |() - 8% Filled 2500 2000 ~m \ \a - 1500C Michigan State Highway Dept. minimum 1000 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 10-C. Characteristic curves: % asphalt vs stability. Filler: Conners Creek Plant 33

The University of Michigan ~ Engineering Research Institute 3000 m - 4% Filler k~ A- Ad N - 6% Filler ()- 8% Filler 2500'2000 1500 Michigan State Highway Dept. minimum 1000 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 11-C. Characteristic curves: % asphalt vs stability. Filler: Marysville Plant 34

The University of Michigan * Engineering Research Institute 3000 (I" El -4% Filler A -6%/OFiIler ) -8% Filler 2500 - -' 2000, I'C) 1500 Michigan State Highway Dept. minimum I 000I I 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 12-C. Characteristic curves: % asphalt vs stability. Filler: Waukesha Products Co.

The University of Michigan ~ Engineering Research Institute 30 0 - 4% Filler 28 h/ - 6 % Filler - 8 % Filler 26 24 22 / O. 20 / z Michigan State Highway Dept. maximum limit 8 18 m -0 14./ 12 / 8 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 13-C. Characteristic curves: % asphalt vs flow. Filler: National Lime and Stone Co. 36

The University of Michigan * Engineering Research Institute 44 ~~~~~~44 t —r~~~~~~~ I 42 40 l- 4%/ Filler j 36. 1I 34 32 I 30 I 28 I z I 8 26 24L / LL 22 20 Michigan State Highway Dept. I / maximum limit 18 14 / / 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 14-C. Characteristic curves: % asphalt vs flow. Fillers St. Clair Plant 37

The University of Michigan * Engineering Research Institute 34 - 4/4% Filler (i 32-, 6% Filler 0) - 8% Filler 30 28, 26 24 22 0 20 Michigan State Highway Dept. maximum limi/ LL. 18 16 // v 14 12 / 10 l I I I I 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 15-C. Characteristic curves: o asphalt vs flow. Filler: Trenton Channel Plant 38 -

The University of Michigan * Engineering Research Institute 34 32 P 30 30 -- 4% Filler / 28- 6% Filler 28 -- 8% Filler 26 24 022 o / 20 Michigan State Highway Dept. 0 maximum limit. 18 16 6 10 8 A /\4.0 4.5 5.0 55 6.0 6.5 PERCENT ASPHALT Fig. 16-C. Characteristic curves: % asphalt vs flow. Filler: Conners Creek Plant 739

The University of Michigan * Engineering Research Institute 36 3 -4% Filler 34 & -6% filler 32 A-8% Filler 30 28 26 24 O / 20 2 22 — / 18 / / 16 — 14// / 124.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 17-C. Characteristic curves:, asphalt vs flow. Filler: Marysville Plant 40

The University of Michigan * Engineering Research Institute 28 [] - 4% Filler 26 A ^ - 6% Filler (i)- 8% Filler 2422 z20 a Michigan State Highway Dept. maximum limit 0 8 1 3 / m 18 = 16 0 14 12 41 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 18-C. Characteristic curves: ~ asphalt vs flow. Filler: Waukesha Products Co................ 41..

The University of Michigan * Engineering Research Institute m1- 4 % Filler,A - 6 / Filler 152 — 8 % Filler 151 =15ol 0 /,l z 150 147 0 i / / 4.0 4.5 5.0.5.5 6.0 6.5 PERCENT ASPHALT Fig. 19-C. Characteristic curves: % asphalt vs unit weight. Filler: National Lime and Stone Co. "149J 1 ~ ~ ~42 42..

The University of Michigan * Engineering Research Institute El- 4 ~/ Filler i- 6/0 Filler 152 (- 8% Filler 151 150 3 149 z 148 147 lI I! I 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 20-C. Characteristic curves: % asphalt vs unit weight. Filler: St. Clair Plant 43

The University of Michigan * Engineering Research Institute [ -- 4% Filled 152 - 6% Filled 152 - 8% Filled 151 ILL:: 150 1: 9 149 F I,, /.' z O0 148 I 147 I I. I II I I 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 21-C. Characteristic curves: * asphalt vs unit weight. Filler: Trenton Channel Plant.44

The University of Michigan * Engineering Research Institute -4 % Filler j-6 % Filler j-8%/ Filler 152 1 51 _ i./ J /l 2 149 1z / 148 147 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 22-C. Characteristic curves: % asphalt vs unit weight. Filler: Conners Creek Plant

The University of Michigan * Engineering Research Institute mn - 4% Filler 152 - 6% Filler (i) - 8% Filler 151 lL. o 150 a. 149.I Iop ID z 0~J 147 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 23-C. Characteristic curves: % asphalt vs unit weight. Filler: Marysville Plant 46 -...

The University of Michigan * Engineering Research Institute 152 151 / Ir' A/ / LL _ / CID w 149 z:D / rT~ -- 4%' Filler 148 -6% Filler | m - - 8% Filler 147 40 45 50 55 60 65 PERCENT ASPHALT Fig. 24-C. Characteristic curves: % asphalt vs unit weight. Filler: Waukesha Products Co. 47

The University of Michigan * Engineering Research Institute l -- 4% Filler A - 6% Filler () - 8% Filler 90 I-,s - Michigan State Highway Dept. limits 80 _J E70 w 60 z 50 z w X0 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 25-C. Characteristic curves: % asphalt vs %A aggregate voids filled. Filler: National Lime and Stone Co. 48

The University of Michigan ~ Engineering Research Institute F~ - 4% Filler - 6% Filler - 8% Filler 90 I< I A 0. Michigr:::t ate Hi:Thway Iept. limits U) 80 0t40 ufJ 0 0L A, I I I I I 4.0 4.5 5.0 5.5 6. 0 6.5 PERCENT ASPHALT Fig. 26-C. Characteristic curves: % asphalt vs % aggregate voids filled. Filler: St. Clair Plant 49

The University of Michigan * Engineering Research Institute El - 4% Filler A - 6 % Filler ( - 8/% Filler 90 C) Michigan State Highway Dept. limits Y 80 w 70 -J I: (. z 50 az I I 4:0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 27-C. Characteristic curves: % asphalt vs % aggregate voids filled. Filler: Trenton Channel Plant

The University of Michigan ~ Engineering Research Institute E - 4% Filler A - 6% Filler 0 - 8% Filler o90 2: Qa Michigan State Highway Dept. limits 0z 50 I 70 z 50 60 cr - 551 a. 4.0 4.5 5.0 5.5 6.0 6.5 PERCENT ASPHALT Fig. 28-C. Characteristic curves: % asphalt vs % aggregate voids filled. Filler: Conners Creek Plant 51l

The University of Michigan ~ Engineering Research Institute |E - 4% Filled - 6%/o Filled - 8%/o Filled 90 a.I t Michigan State Highway Dept. limits C) < 80 JLL 70 56 0P60N / z_ 0 Fig. 29-C. Characteristic curves: % asphalt vs lo aggregate voids filled. Filler: Marysville Plant 52 52....

The University of Michigan ~ Engineering Research Institute [m - 4% Filler A - 6% Filler () - 8% Filler 90 II 0 Michigan State Highway Dept. limits U) 60 _J CD 0 (35 z 0 a. 4.0 4.5 5.0 5.5 6,0 6.5 PERCENT ASPHALT Fig. 30-C. Characteristic curves: l asphalt vs k aggregate voids filled. Filler: Waukesha Products Co. 55

TABLE I-C DATA FOR CHARACTERISTIC CURVES Filler: National Lime and Stone Company Mix Number 101 102 105 104 105 106 107 108 109 110 111 112 115 114 15 16 17 18 Thickness of Specimen, in. 2.521 2.521 2.50 2.50 2.479 2.458 2.541 2.50 2.50 2.50 2.50 2.50 2.50 2.521 25.0 249 25 Specific Gravity 2.567 2.589 2.401 2.416 2.409 2.406 2.575 2.409 2.422 2.428 2.420 2.409 2.404 2.596 246 240 248 248Unit Weight, lb/cu ft 147.72 149.07 149.82 150.76 150.52 150.15 148.22 150.54 151.12 151.55 150.99 150.50 150.00 149.51 15.8 5.6 1087 5.2 Stone, % by weight 57.2 56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.5 55.0 54.7 54.4 54.7 54.4 540 5. 554 51 Sand, % by weight 34.8 34.7 5.4.5 54.5 54.1 55.9 54.1 55.9 55.7 55. 5 55.5 55.1 55.5 55.1 5.0 28 5.6 54 Filler, % by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0o 6.0 6.0o 6.0 8.0 8.0 8. 80 80 00 pAsphalt, % by weight 4.0 4.5 5.0 55 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 50 55 6. 65 Specific Gravity (apparent) Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 276 27 6 7 Sand 2.694 2.690, 2.694 2.69L) 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2. 694 264 26 2 o Filler.852.S-3 -:'-833 2.855 -2. 833 2.855 2.855 2.855 2.855 2.855 2.855 2.855 2.855 2.855 2.855 285 2.535 Asphalt 1.018 1.00-8 1.0-18 1.0-18 1.018 1.018 1.018 1.018 1.018 1.0o18 1.018 1.018 1.018 1.018 108 1.8 Cl Specific Gravity Compacted Aggregate 2.27i2 2.282 2.281 2.285 2.264 2.250 2.280 2.501 2.501 2.295 2.275 2.252 2.508 2.288 2.9 227 Theoretical Maximum Density of Mix 2.567 2.508 2.:528 2.509 2.489 2.470 2.569 2.550 2.530 2.510 2.491 2.471 2.571 2.551 2.5 252 27 Theoretical Maximuir Density of Total Mineral Aggregate, Void Free 2.741 2.742 2.742 2.742 2.742 2.745 2.744 2.744 2.744 2.744 2.744 2.744 2.746 2.746 274 76 7 — Voids in Mineral Aggregate 17.10 16.78 16.82 16.74 17.42 17.99 16.90 16. 14 16.16 16.56 17.11 07.92 15.95 16.67 1.4 1. Voids, * - Total Mix 7.8 6.55 5. 05 5.72 5.25 2.60 7.54 5.52 4.28 5.29 2.86 2.51 6.49 6.08 45 h Co Total Voids Filled with Asphalt, * 54.50 62.96 70.08 77.83 81.49 85.51 55.35 65.88 75.52 80.16 85.54 85.99 59.28 65.62 722 801 8 Marshall Stability (corrected), lb 1685 2420 2148 2050 1698 1741 2020 2769 2569 1819 1606 1592 2410 2587 940 16 146 1 Flow: 0.01 in. 8.7 9.1 10.7 11.8 12.8 18.0 11.5 8.0 10.2 12.7 17.7 25.2 9.0 8.2 102 1. 170.

TABLE II-C DATA FOR CHARACTERISTIC CURVES Filler: St. Clair Plant - Mix Number 201 202 203 204 205 206 207 208 209 210 2'11 212 215 214 21 26 27 28 Thickneas of Specimen, in. 2.541 2.541 2.50 2.50 2.50 2.50 2.541 2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.0.5 250.0 Specific Gravity 2.593 2.399 2.411 2.415 2.41.6 2.403 2.599 2.425 2.419 2.420 2.415 2.595 2.400 2.424 245 241 245 255 C Unit Weight, lb/c-u ft 149.54 149.72 150.45 150.70 150.76 149.97 149.70 151.54 150.95 151.05 150.57 149.45 149.78 151.24 11.4 5.0 1072 495 Stnne, % by weight 57.2 _56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.5 55.0 54.7 54.4 54.7 54.4 540 57 5.4 51 W Sand, % by weight 54.8 54.7 54.5 54.5 54.1 55.9 54.1 55.9 55.7 55.5 55.5 55.1 55.5 55.1 5.0 28 526 24 VI Filler, % by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 6.0 6.0 8.0 8.0 8. 80 80 80 * Asphalt, % by weight 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5. 5. 60 65 Specific Gravity (apparent) Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 276 276 276 276 Sand 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.9 264 264 264 Filler 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.654 2.5 264 264 264 Asphalt 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 108 108 108 108 f Specific Gravity Compacted Aggregate 2.297 2.291 2.291 2.282 2.271 2.247 2.505 2.516 2.298 2.287 2.268 2.240 2.504 2.515 2.0 228 227 225 Theoretical Maximum Density of Mix 2.561 2.541 2.521 2.502 2.485 2.464 2.559 2.559 2.520 2.500 2.481 2.461 2.557 2.557 258 2.9 247 249 Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.754 2.755 2.754 2.754 2.755 2.754 2.751 2.752 2.752 2.751 2.751 2.751 2.729 2.729 279 279 279 278 ( Voids in Mineral Aggregate 15.97 16.16 16.20 16.53 16.90 17.80 15.66 15.25 15.90 16.26 16.94 17.99 15.56 15.18 1.6 1.5 1.1 1.1 ( Voids, % - Total Mix 6.53 5.5 4.31 3.48 2.70 2.46 6.28 4.48 4.02 5.19 2.74 2.68 6.15 4.47 5.8.0 290.6 Total Voids Filled with Asphalt, % 58.95 65.59 75.15 78.94 84.08 86.19 60.15 70.57 74.71 80.45 85.86 85.15 60.62 70.56 764 810 8.5 8.7 Marshall Stability (corrected), lb 1801 1825 1914 1686 1422 1550 1950 1867 1811 1462 974 858 2055 1586 115 95 02 97 Flow: 0.01 in. 8.7 9.2 9.8 14.8 15.5 25.0 8.8 12.7 12.8 14.2 21.7 44.8 8.4 10.5 128 9. 2.7 40

.... T~L~ III-C ~" ~b DATA FOR CHARACTERISTIC CURVES f'' Filler: Trenton Channel Plant...'I m. Mix Number 301 302 303 304 305 306 507 308 309 310 311 512 313 514 315 316 317 318 O Thickness of Specimen, in. 2.552 2.531 2.541 2.511 2.50 2.511 2.~42 2.552 2.511 2.50 2.521 2.542 2.563 2.521 2.521 2.521 2.521 2.532 Spec if ic Gravity 2. 370 2. 393 2. 398 2.4O2 2. 399 2. 387 2. 565 2. 591 2. 598 2. 399 2.39O 2. 379 2. 562 2. 577 2. 589 2.404 2. 592 2. 582 Unit Weight, lb/cu ft 147.88 149.31 149.65 149.86 149.74 148.92 147.55 149.18 149.60 149.72 149.15 148.44 147.40 148.35 149.08 150.00 149.23 148.65 Stone, % by weight 57.2 56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.5 55.0 54.7 54.4 54.7 54.4 54.0 53.7 53.4 53.1 Sana, % by weight 5~.8 34.7 34.5 3&.3 54.1 35.9 34.1 35.9 35.7 33.5 33.3 33.1 53.3 35.1 35.0 32.8 32.6 32.4 Filler, ~ by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 6.0 6.0 8.0 8.0 8.0 8.0 8.0 8.0 (~ Asphalt, ~ by weight 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.C 6.5 Specific Gravity (apparent) r~1 Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766:~ Sand 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 ~(~ Filler 2.42 2.42 2.42 2.42 2.42 2.42 2.42 2.42 ~.42 2.42 2.42 2.42 2.42 2.42 2.42 2.42 2.42 2.42:~ Asphalt 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1.018 1.O18 1.018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 Specific Gravity Compacted Aggregate 2.275 2.285 2.279 2.270 2.256 2.232 2.270 2.283 2.278 2.267 2.247 2.226 2.268 2.271 2.270 2.272 2.248 2.228'1 Theoretical Maximum Density of Mix 2.552 2.552 2.513 2.494 2.475 2.456 2.~46 2.527 2.507 2.488 9.469 2.449 2.539 2.520 2.50i 2.481 2.462 2.443:~ Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.723 2.723 2.724 2.723 2.725 2.724 2.716 2.716 2.716 2.716 2.715 2.715 2.708 2.708 2.708 2.708 2.7Q7 2.707 Voids in Mineral Aggregate, ~ 16.46 16.08 16.36 16.66 17.16 18.07 16.42 15.94 16.13 16.54 17.25 18.07 16.26 16.16 16.20 16.11 16.96 17.72 f$ Voids, % -- Total Mix 7.14 5.50 4.57 3.71 3.05 2.85 7.12 5.59 4.37 3.59 3.20 2.87 6.97 5.66 4.48 3.12 2.87 2.49'1 f) Total Voids Filled with Asphalt, ~ 56.68 65.81 72.07 77.85 82.31 84.28 56.65 66.30 75.02 78.35 81.54 84.16 57.15 65.04 72.42 80.70 83.12 85.96:~' Marshall Stability (corrected), lb 2039 1906 1789 1695 1474 1258 2287 2127 1986 1510 1313 1068 2409 2223 1738 1513 1171 1007 Flow: 0.01 in. 8.5 8.8 10.4 13.3 17.8 21.6 8.0 9.2 ll.9 14.7 23.7 28.4 8.9 9.5 12.0 17.5 22.2 33.2 f"~ C F-P,,,,,,,,,..................... 1 1 1...,,,1 l..

TABLE IV-C ~." DATA FOR CHARACTERISTIC CURVES f" Filler: Conners Creek Plant -— e'I Mix Number 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 Thic!uless of Specimen, in. 2.563 2.521 2.521 2.90 2.50 2.50 2.50 2.521 2.50 2.438 2.521 2.50 2.563 2.541 2.50 2.50 2.50 2.50 Specific Gravity 2.383 2'.393 2.394 2.423 2.411 2.396 2.396 2.405 2.420 2.411 2.404 2.586 2.358 2.400 2.409 2.415 2.403 2.383 Unit Weight, lb/cu ft 148.70 149.34 149.37 151.17 150.47 149.49 149.53 150.07 151.03 150.45 150.03 148.87 147.16 149.76 150.30 150.67 149.95 148.68 ~' Stone) ~ by weight 57.2 56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.3 55.0 54.7 54.4 54.7 54.4 54.0 53.7 53.4 53.1 Sand, ~ by weight 34.8 34.7 34.5 34.3 34.1 33.9 34.1 33.9 33.7 33.5 33.3 33.1 33.3 33.1 33.0 32.8 32.6 32.4 k.J] Filler, ~ by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 6.0 6.0 8.0 8.0 8.0 8.0 8.0 8.0 Asphalt, % by weight 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 Specific Gravity (apparent) r~1 Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 Sand 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 Filler 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 2.47 — " Asphalt 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 028 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1. 018 1.018 Specific Gravity Compacted Aggregate 2.288 2.286 2.274 2.289 2.267 2.240 2.301 2.297 2.300 2.278 2.260 2.231 2.264 2.292 2.288 2.282 2.259 2.228'1 Theoretical Maximum Density of Mix 2.554 2.534 2.515 2.496 2.477 2.458 2.549 2.~30 2.510 2.491 2.472 2.452 2.544 2.524 2.505 2.486 2.467 2.447 Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.726 2.726 2.726 2.726 2.725 2.726 2.720 2.720 2.720 2.719 2.719 2.719 2.713 2.713 2.714 2.713 2.713 2.713 Voids in Mineral Aggregate, ~ 16.07 16.15 16.58 16.02 16.82 17.83 15.42 15.53 15.45 16.21 16.87 17.96 16.55 15.53 15.69 15.88 16.73 17.88 Voids, ~ -- Total Mix 6.73 5.55 4.82 2.94 2.65 2.54 5.99 4.94 3.57 3.21 2.74 2.70 7.30 4.91 3.84 2.87 2.59 2.63'1 Total Voids Filled with Asphalt, ~ 58.43 65.62 70.94 81.67 84.28 85.78 61.16 68.18 76.94 80.32 83.83 84.99 55.99 68.40 75.54 81.95 84.54 85.01 ~" Marshall Stability (corrected), lb 2208 2106 1996 1978 1567 1231 2380 2108 1736 1527 1191 914 2368 2366 1731 1561 1175 970 Flow: 0.01 in. 7.2 8.2 10.0 13.4 16.6 24.5 8.2 9.6 14.3 15.8 23.2 32.5 9.2 10.5 12.8 15.5 22.3 99.5 r-~ C

TABLE V-C DATA FOR CHARACTERISTIC CURVES Filler: Marysville Plant Mix Number 501 502 503 504 505 506 507 508 509 510 511 512 515 514 51 56 57 56 0 Thickness of Specimen, in. 2.563 2.563 2.50 2.50 2.50 2.521 2.563 2.563 2.541 2. 521 2.50 2.521 2.563.h.6 2.5412.1 256 Specific Gravity 2.363 2.572 2.393 2.599 2.598 2.585 2.572 2.585 2.588 2.598 2.595 2.588 2.549 2.5853.8.9 255 259~ Unit Weight, lb/cu ft 147.45 148:05 149.52 149.70 149.66 148.84 148.04 148.82 148.99 149.61 149.45 149.05 146.58 148.72 188 4.4 188 4.5 - Stone,% by weight 57.2 56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.5 55.0 54.7 54.4 54.7 54.4 540 5. 5.4 51 Sand, %by weight 34.8 54.7 34.5 54.5 54.1 55.9 54.1 55.9 55.7 55.5 55.5 553.1 55.5 55.1 5.0 28 5.6 24 Filler, *by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 6.0o 6.0o 8.0o 8.0... o co Asphalt,% by weight 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5. 0 5.5 6.0 6.5 4.0 4.5 5. 55 60 65 Specif ic Gravity (apparent) Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 276 2.6 276 276 Sand 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2. 694 2.694 2.694 2.9 264 264 264 Filler 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.28 2.8.8 228.8 Asphalt 1.0 18 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 i.o18 io8 1.8 108 108 Specific Gravity Compacted Aggregate 2.269 2.266 2.275 2.267 2.255 2.250 2.277 2.278 2.269 2.266 2.251 2.255 2.255 2.276 2.6 225 224 221 Theoretical Maximum Density of Mix 2.546 2.526 2.506 2.488 2.469 2.450 2.556 2.517 2.498 2.478 2.459 2.440 2.527 2.507 2.8 246 245 241 Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.716 2.715 2.715 2.716 2.715 -2.716 2.704 2.705 2.704 2.7o4 2.704 2.705 2.695 2.692 265 262 262 261 f Voids in Mineral Aggregate, *16.47 16.55 16.27 16.55 16.96 17.88 15.78 15.79 16.10 16.30 16.74 17.58 16.26 15.44 158 161 1.7 1.9 Voids, * - Total Mix 7.19 6.08 4.51 3.58 2.86 2.64 6.44 5.24 4.41 5.24 2.60 2.12 7.05 4.94 4.4 52 264.5 Total Voids Filled with Asphalt, % 56.58 65.51 72.28 78.54 85.16 85.22 59.15 66.81 72. 66 79.99 84.46 87.84 56.72 68.09 759 801 848 8.9 Marshall Stability (corrected), lb 2077 2250 2469 2055 1759 1657 2801 2874 2265 1955 1478 1505 2224 27153 49 19 64 16 Flow: 0.01 in. 9.7 9.0 11.2 11.8 16.5 19.7 9.5 8.8 11.2 14.7 19.7 24.5 7.8 10.5 127 5. 2.2 55

TABLE VI-C DATA FOR CHARACTERISTIC CURVES Filler: Waukesha Products Company Mix Rumber 601 602 605 604 605 606 607 608 609 610 611 612 613 614 61 66 61 68 Thickness of Specimen, in. 2.563 2.541 2.50 2.50 2.50 2.458 2.521 2.50 2.50 2.50 2.458 2.50 2.50 2.50 2.0.5 247 2.8 Specific Gravity 2.566 2.574 2.415 2.421 2.413 2.405 2.596 2.410 2.450 2.451 2.427 2.407 2.419 2.426 2.5 2.5 2.6 2.1 Unit Weight, lb/cu ft 147.66 148.16 150.68 151.07 150.59 150.07 149.51 150.40 151.65 151. 69 151.46 150.22 150. 95 151.58 15.7 5.8 1158 506 Stone, % by weight 57.2 56.8 56.5 56.2 55.9 55.6 55.9 55.6 55.5 55.0 54.7 54.4 54.7 54.4 540 5. 5.4 51 Sand, % by weight 54.8 54.7 34.5 54.5 54.1 55.9 54.1 55.9 55.7 53.5 55.5 55.1 55.5 553.1 5.0 28 5.6 24 Filler, % by weight 4.0 4.0 4.0 4.0 4.0 4.0 6.0 6.0 6.0 6.0 6.0 6.0 8.0 8.0 8. 80 80 80 Asphalt, % by weight 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5.0 5.5 6.0 6.5 4.0 4.5 5. 55 60 6. Specific Gravity (apparent)m Stone 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.766 2.6 276 276 276 Sand 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2.694 2. 694 2.694 2.694.2.694 2.694 2.694 2.9 264 264 264 Filler 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.85 2.853.5 28 28.5 - Asphalt 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.018 1.8 108 108 108 Specific Gravity Compacted Aggregate 2.272 2.267 2.294 2.288 2.269 2.249 2.500 2.5302 2.508 2.297 2.282 2.251 2.522 2.517 2.1 250 226 228 Theoretical Maximum Density of Mix 2.568 2.548 2.528 2.509 2.489 2.470 2.569 2.550 2.550 2.510 2.491 2.471 2.571 2.550 2.5 251 249 247 Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.742 2.742 2.742 2.742 2.742 2.745 2.744 2.744 2.744 2.744 2.744 2.744 2.745 2.745 2.4 276 275 275 Voids in Mineral Aggregate, * 17.15 17.51 16.54 16.56. 17.26 18.02 16.18 16.11 15.88 16.28 16.85 17.98 15.42 15.59 156 1.2 1.9 176 Voids, % - Total Mix 7.85 6.82 4.48 5.51 5.04 2.65 6.75 5.48 5.95 5.15 2.56 2.58 5.92 4.86 5.4.1 265.5 Total Voids Filled with Asphalt, * 54.25 60.63 72.57 78.85 82.42 85.40 58.52 66.0o6 75.17 80.71 84.87 85.67 61.60 68.79 767 809 845 8.6 Marshall Stability (corrected), lb 1651 1949 2297 1859 1651 1558 2807 2516 2497 1758 1674 1550 2955 2766 252 13 104 05Flow: 0.01-in. 8.2 10.0 9.0 11.5 15.2 18.7 7.8 7.9 11.5 15.7 17.4 20.6 9.4 10.6 155 5. 175 29

APPENDIX D DATA FROM IMMERSION-COMPRESSION TESTS

TABLE I-D "~:2' MIX PROPORTIONS AND TEST RESULTS OF WATER-IMMERSION TESTS f...,,.,,, f% -1 FiLler Spe c ific at ions Nat ionale"~ St. Clair Plant Trenton Channel Plant ConneTs Creek Plant Marysville Plant Waukesha Products Co. Li~e and StQne Co, O Mix Number A B C A B C A B C A B C A B C A B C -"~ Thickness of Specimen, in. 4.50 4.479 4.479 4.479 4.458 4.575 4.458 4.479 4.479 4.50 4.50 4.50 4.50 4.50 4.50 4.479 4.50 4.50 m, Specific Gravity 2. 359 2. 354 2. 354 2. 368 2. 372 2. 374 2. 360 2. 358 2. 356 2. 363 2. 363 2. 364 2. 355 2. 364 2. 357 2. 370 2. 369 2. 362 zr Unit Weight, lb/cu ft lb7.16 146.89 146.77 147.74 147.99 148.12 147.28 147.14 147.01 147.47 147.43 147.50 146.95 147.51 146.60 147.89 147.85 147.53 (~ Stone, ~ by weight 55.2 55.2 55.2 55.3 55.3 55.3 55.2 55.2 55.2 55.4 55.4 55.4 55.2 55.2 55.2 55.3 55.3 55.3 Sand, ~ by weight 35.7 33.7 33.7 33.8 33.8 33.8 35.6 55.6 33.6 53.8 35.8 35.8 35.7 33.7 33.7 33.7 33.7 35.7 Filler, ~ by weight 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0 6.0' O% }-J Asphalt, ~ by weight 5.1 5.1 5.1 4.9 4.9 4.9 5.2 5.2 5.2 4.8 4.8 4.8 5.1 5.1 5.1 5.0 5.0 5.0 r1~ Specific Gravity (apparent) Stone 2. 766 2. 766 2. 766 2. 766 2. 766 2. 766 2.766 2. 766 2. 766 2. 766 2.766 2. 766 2. 766 2. 766 2.766 2.766 2. 766 2. 766 Sand 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2. 694 2.69L 2. 694 2. 694'" Filler 2.833 2.833 2.833 2.634 2.634 2.634 2.42 2.42 2.42 2.47 2.47 2.47 2.28 2.28 2.28 2.83 2.83 2.83 Asphalt 1.018 1.Ol8 1.018 1.C18 1.018 1.018 1.O18 1.018 1.O18 1.018 1.018 1.O18 1.018 1.O18 1.018 1.018 1.018 1.018 Specific Gravity Compacted Aggregate 2.319 2.23~ 2.234 2.252 2.255 2.258 2.258 2.235 2.234 2.250 2.249 2.250 2.235 2.243 2.237 2.252 2.251 2.246'I Theoretical Maximum Density of Mix 2.525 2.525 2.525 2.523 2.523 2.523 2.499 2.499 2.499 2.517 2.517 2.517 2.493 2.493 2.493 2.530 2.530 2.530 Theoretical Maximum Density of Total Mineral Aggregate, Void Free 2.744 2.744 2.744 2.731 2.731 2.731 2.716 2.716 2.716 2.719 2.719 2.719 2.704 2.704 2.704 2.744 2.744 2.744 Voids in Mineral Aggregate, ~ 15.48 18.60 18.60 17.55 17.42 17.35 17.61 17.70 17.76 17.25 17.28 17.24 17.56 17.04 17.27 17.94 17.95 18.14'I Voids, ~ -- Total Mix 6.56 6.74 6.79 6.15 6.00 5.92 5.55 5.64 5.72 6.11 6.13 6.09 5.54 5.18 5.45 6.32 6.35 6.55 Total Voids Filled with Asphalt, ~ 64.33 63.52 63.48 64.94 65.91 65.89 68.48 68.11 67.79 64.61 64.50 64.65 68.06 69.59 68.46 64.80 64.71 63.95 F Dry Strength, psi 349 328 355 339 375 292 Water / 4-Day Immersion, psi 365 325 579 342 566 299 e-~ Immersion k14-Day Immersion, psi 335.7 327 567 300 359 277 ~' e' Retained Strength, ~ 104.6 96.2 98.4 99.7 106.7 103.4 100.8 88.5 97.7 95.8 102.4 94.9 e"~ 1

UNIVERSITY OF MICHIGAN 3 9015 02826 8277