:1; . I M I .r.“ ' a: a thHk-a' MATERTAL mom “"5- (.59 PLACE IN RETURN Box to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE DATE DUE DATE DUE AAA .2‘5 2007 9’ .7 6/01 cJCIRC/DateOutpfi-pw ' INEERN DC MOD ln Depan” ENGINEERING CHARACTERISTICS OF POLYMER MODIFIED ASPHALT MIXTURES By Mohammad Jamal Khattak A DISSERTATION Submitted to Michigan State University In partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Civil and Environmental Engineering 1999 ENGINEERN . \lllllll | ' I v... . 1‘ " ‘Jmfln' ' n-f‘I \\ut\tlll A) " :‘k‘ ”\- ‘ 7-.” ‘av‘on. ‘ '4 fi ! . Tm“ v'HwCE ‘\" _a ' . . K. ”g“ \~ . “ ‘ I .. \‘W ~ I t i '1‘ t‘ ‘r. l Nun“ r‘\h'h‘L‘ - T ' - H: Li}; AAA-:" 5" it"':‘~a|’"i.' -:y I xv. 4.. '4‘10u5 k:v~_\1\ U. u V‘K 4“! “a ‘ ‘rll ‘ ""m. ‘nf 'm 1... A '-l.§| l““.‘u ‘t ..‘ . Itf "- 33‘.“\0. ~r'.\l‘\"‘b¥“ Q. . - M 0. vs. ‘i-l "‘ ". 5.; ' \‘Iha. _~ ._. ‘ .5 men. P . I a: 3. «.- 51‘: \. ~ . “‘AKEEL E f \ ’5. m.,.,‘.‘ A“? ' ‘ . “1:010...“ _‘ t-Mi. . [ T {5:}: . . 5‘:n a“, t l\‘ ‘ ‘x‘s W a \V‘ ‘c- \ \s 3E "'Tv a I . W‘ v -.‘"‘9 ~ .4 ‘ T" e «a -, ‘ "1 "A: \[H ‘ ‘ uk . l ‘p .q‘ s ’3. “-.\i‘. . -_“‘l‘ a -,l .- “: T‘H ABSTRACT ENGINEERING CHARACTERISTICS OF POLYMER MODIFIED ASPHALT MIXTURES by Mohammad Jamal Khattak Asphalt cement is viscoelastic material, which is time and temperature dependent. At high temperatures asphalt cement behaves as a viscous material. At below freezing temperatures it acts as a brittle solid and is more susceptible to temperature cracking. In order to improve both the high and low temperature properties, asphalt binders were modified with various kinds of polymers. In this study, structural and engineering properties of polymer modified asphalt (PMA) mixtures were investigated in lieu of the rheological properties, morphology and adhesive properties of binders. Experimental program matrices were established for two types of structural tests; indirect tensile strength test and indirect tensile cyclic load test to evaluate the engineering properties of straight and PMA mixtures. The tests were conducted at low (23°F), moderate (77°F) and high temperatures (140°F) for aged and unaged Marshall size samples. Asphalt- polymers mixing procedures were developed based upon the improvement of rheological properties. All asphalt mixtures were designed according to the Michigan Department of Transportation (MDOT) specifications. The SBS polymer system was selected as a typical polymer to discuss the improvements in the structural and engineering properties due to polymer modification. Comparison of the structural and engineering properties of asphalt mixtures modified with . - ' ~5“QI)I i" .y. lu~ w ' t" a! I l {A ’7: . ..,. he... \9 _ ‘|H ‘F‘ ‘ R .Luu.-.;u‘.‘i. \'g A} ‘IAZA on“: . ... u -- zx‘ln ~n‘ania ‘Ts “L\i \ \A v 1 ”.I an! .L "all SIJNI- -! . ... ....,. u ‘ l 7‘ V. 1" l N . fiLn—Lu U1 30: AL\. .lu. ‘ -\1.\ I . \ "‘ 1r- - ‘ . ‘ \. \ ‘ ‘ fl . &~n ”£2 {a ..L\11 fr! ‘Ni O:C\ I'" 9 . " "Q’x \ ~-¢>. various polymers is presented and discussed. A new fatigue life criterion based on the rate of accumulation of horizontal plastic deformation was developed and compared to the rate of crack growth criteria (based on fracture mechanics). It is shown that the rate of accumulation of horizontal plastic deformation criterion produced consistent and repeatable results for straight and PMA mixtures. Fatigue life and rut depth models were developed based on rheological properties of binders and physical and engineering properties of mixtures. To My | To My Parents, Brothers and Sister TAM to A: \lzrht 2‘"- ~"t':‘: .1355 5a.) x. r {\ ‘q_--o‘ u~lQ‘-n~ on“ w‘ ---\u‘.~ ‘uaudnA\e and. e:-\\ .a h '1. ‘ fi~~l~ mpn‘J‘g" Ah.._\"" Ce ;.I\..«x \ :) \lgiaci Home 5 I l o 130‘ " .. ilKC i0- C\'.L’.".. an- 13;...3132 and the Palm \ P‘)‘~9m a: "v-“,.\‘ ' .‘ . o ‘9 4.2:. 3.2mm lacmouiedge xx 21?. l .LL“ .".‘..'v 9' ‘ ‘ T .‘H ‘ ! "’ A T ‘5 JAE film?“ 5::LL s T‘a-u- l' "N. \ 5 9L ‘ Thar. 295a: A" .‘H 'i :.::'é|. 'q .' I o .‘-. Arie ‘aOCTa: A“ ‘ ‘-- I Tr ~- . . ‘9 won 01M; ( 2‘: ‘TcC ‘L I . "£4 33510 "~\ ,. ‘- r r-., ‘i , " u. . T‘W‘d is ?l.4 x..." ACKNOLEDGEMENT Thanks to All Mighty Allah for providing me with an opportunity to undertake and complete this task. Special thanks to my advisor Dr. Gilbert Y. Baladi for his constant guidance and encouragement throughout this research study. Many thanks to my doctoral committee members Dr. Thomas Wolf, Dr. Neeraj Buch, Dr. Lawrence Drzal and Dr. Michael Thorpe. I would like to extend special thanks to the Michigan Department of Transportation and the Pakistani Ministry of Education for sponsoring my study and the Department of Civil and Environmental Engineering at Michigan State University for financial support. I acknowledge with gratitude the efforts and guidance provided by Mr. Jamal Dass during the initial stage of this study. Thanks to all graduate and undergraduate students Heath, Zaffar, Attullah, Maqbool, Owais Rehman, Nazaral Wazir and F azail for their help in the laboratory investigations. The support of Mr. Chang and other friends is greatly appreciated. Last but not least, special thanks to my parents, brothers and sister for their patients, understanding and continuos guidance. INTER l NilODl’CTION " REMESIS RESEARCH ( lBJl (' k RESERRCH PLAN 792515 LW llT MIR: ENTIRE RENE“ CSEPAL . ASPPA'LT (EVER I mu -‘I~\..1L_'\LT BIXDI :. I, I. n ""1 LU“ Tcmrw' 1 ~ ‘ ..:.l.l Pent" " w Ux‘l. ' J J 1’ 0/ f l. I ”'1 -a ._ f 1"! 'v- ‘ 4 a .414 If}! - TABLE OF CONTENTS CHAPTER 1 INTRODUCTION 1 . 1 GENERAL ............................................................................... 1 1 -2 HYPOTHESIS ....................................................................... 3 1 .3 RESEARCH OBJECTIVES .................................................................. 3 1 .4 RESEARCH PLAN ........................................................................ 3 1 .5 THESIS LAYOUT ........................................................................ 5 CHAPTER 2 2. 1 GENERAL ................................................................................... 7 2.2 ASPHALT CEMENT COMPOSITION .......................................................... 7 2.3 ASPHALT BINDER PROPERTIES .................................................................. .8 2.3.1 Low Temperature Properties ..................................................... 8 2.3.1.1 Penetration (ASTM D5) ........................................................ 10 2.3.1.2 Ductility ............................................................. 11 2.3.1.3 Fraass Breaking Point .......................................... 11 2.3.1.4 Limiting Binder Stiffness .......................................... 12 2.3.1.5 Critical Stress (Strain) Method ................................. 12 2.3.2 High Temperature Properties .......................................... 13 2.3.2.1 Viscosity ............................................................ 13 2.3.2.2 Temperature Susceptibility .......................................... 13 2.3.2.3 Viscoelastic Properties .......................................... 15 2.3.2.4 Ring and Ball Softening Point ................................. 17 2.4 EFFECT OF POLYMERS ON ASPHALT BINDER PROPERTIES ...... 18 2.4.1 Effect of polymer on Low Temperature Properties ........................ 18 2.4.1.1 Effect on Penetration ................................................... 18 2.4.1.2 Effect on Ductility ................................................... 18 2.4.1.3 Effect on Stifiness and Tensile Properties ........................ 21 2.4.1.4 Effect on Fraass Point .......................................... 22 2.4.2 Effect of polymer on High Temperature Properties ........................ 25 2.4.2.1 Effect on Viscosity ................................................... 25 vi Ilvlll ‘ V I ‘ I.— o \ vat v I va. .1. . . .\ .0. uuuuuu iii. . .1..\ I.“ .‘ ,‘\r.. \ a 0'1 ‘ a Q‘. 4 0‘. h. 9| ‘11 fl. — \\.. " ‘A‘A .u “11‘ x V P L.- n 11. 11 -1 .1 C 1. a . a . x . _..: . l L \1 b at u- 5A liq ‘ll. . . t . Ark . . . . 9-... firm . .fl. . t \‘ 1 vJ .: \s x» ‘41 .1 .\. .\. .\. .\. .\w .\... xnl. .\. .\. .‘.. \... .E in R \I \ mu ‘4; A; P hr» A1 A. A. ~\ 1..\.«.1..~.a.r~ \.fl.1. «. .t.r.r. R DR [1 . \L . M1. 0 ‘5 0 hulk N P \4 n1” -1. a. R T. .\\. MM .1 a . .3 .nfi. 1.. we .0 V- -1. . e 1‘- \ut h (H —-V ‘1 “in; ‘.~ \. s ‘.s ‘ ~ ri\ \\. fi‘. §\. DNA l‘. 1‘ t‘ 1‘ F. a. c N. o T R C 1" Q! I«. §\ I11 \\ Qt. .1. D \. \ .. \. \ mu 80 Truly. . ‘ ..~ .~ 2.4.2.2 Effect on Temperature Susceptibility ........................ 26 2.4.2.3 Effect on Viscoelastic properties ................................. 28 2.4.2.4 Effect on Softening Point .......................................... 29 2.5 ENGINEERING CHARCTERISTICS OF ASPHALT MIXTURES ...... 31 2.5.1 Asphalt Mixture Properties ................................................... 31 2.5.1.1 Stress-Strain Characteristics .......................................... 31 2.5.1.2 Resilient Modulus ................................................... 31 2.5.1.3 Dynamic Complex Modulus ............................................ 33 2.5.1.4 Fatigue and Tensile Characteristics ................................. 36 2.5.1.5 Permanent Deformation and Plastic Characteristics ............... 36 2.5.1.6 Thermal Cracking Potential of Asphalt mixtures ............... 38 2.5.2 Effect of Polymer on Asphalt Mixtures Properties ........................ 39 2.5.2.1 Effect of Polymers on Modulus ................................. 39 2.5.2.2 Effect on Fatigue and Tensile Characteristics ............... 41 2.5.2.3 Effect of Polymers on Permanent Deformation and Plastic , Characteristics ................................................... 42 2.5.2.4 Effect of Polymers on Thermal Cracking Potential of Asphalt Mixtures ............................................................ 46 2.6 FATIGUE AND RUT MODELS .................................................................... 49 2.6.1 Fatigue Models ............................................................ 49 2.6.2 Rut Models ..................................................................... 59 CHAPTER 3 LABORATORY INVESTIGATION .................................................................... 64 3.1 GENERAL ................................................................................. 64 3.2 MATERIALS USED ................................................................................ 65 3.2.1 Aggregates 65 3.2.2 Asphalt Cement ............................................................................................... 65 3.2.3 Polymers 65 3.3 AGGREGATE HANDLING, SIEVING AND WASHING 66 3.3.1 Aggregate Collection and Handling .................................................... .66 3.3.2 Aggregate Sieving 67 3.3.3 Aggregate Washing . . .67 3.4 PHYSICAL PROPERTIES ...................................................... g: 3.4.] Specific Gravity ................................................................................ 63 3.4.2 Absorption Capacity .................................................................... 70 3.4.3 Durability ............................................................................................ 70 3.4.4 Viscosity ............................................................................................ vii 35 R‘IXIXG PRtlt I 35.1 liming} 3.52 \lixzng‘: «vq :.:~ Mm; : 3.54 MM; FF 3: MANN .\‘.l‘ n 1.. r... . . .beI . .‘Ak‘vt 5. n , -‘ ‘r‘ \rn-n‘... "~‘D \‘Was [.5 Q .- a 4" Rs‘.."\ - y.- | 5‘ A” Fg—‘F . ' :.Lr’:?l.‘\11-\ - I é.‘i'xl’IlR 4 EHIOLOGK'AL Mt FROPERIIIS OF P\ i: 379.0. W], ' $310616. 1 Visa-w- 4;: g. .. mg: 4‘ g s 9‘ \ ("m' ' .w r- I“ “ 1 a ‘ ‘3‘] 4“ « fl . .~.‘\ 4‘ '§ . '”‘4 E:.‘{A », “\ILLI . ‘1...” ‘ ‘h ‘ it\ '1. N1 .. “l \—C: 's (f) "f - v I Z {7" 0‘1. «'41 i) .._.. 1.1 p...- ———4 f v .I' '1' (It '1‘ '1' .IJ‘IJIJIIJIJ- -_.a H u._‘ 3 .5 MIXING PROCEDURES OF PMA BIN DERS ............................................ 72 3.5.1 Mixing Procedure of SBS/SEBS Polymer ............................................ 72 3.5.2 Mixing Procedure of SBR latex Polymer ............................................ 72 3.5.3 Mixing Procedure of EAM Polymer .................................................... .73 3.5.4 Mixing Procedure of Crumb Rubber Modifier (CRM) .................... 74 3 .6 MARSHAL MIX DESIGN ................................................................................ 74 3.6.1 Aggregate Gradation ............................................................................. .74 3.6.2 Sample Preparation of Mixtures .......................................... 86 3.6.3 Results of Mix Design .. .......................................................................... .87 3 .7 TEST PROGRAM ............................................................................................ 97 3.7.1 Indirect Tensile Cyclic Load Test .......................................... 97 3.7.2 Indirect Tensile Strength Test ............................................................... .102 3.8 EXPERIMENT PROGRAM MATRIX ...................................................... 105 CHAPTER 4 RHEOLOGICAL, MORPHOLOGICAL AND MICROSTRUCTURAL PROPERTIES OF PMA BINDERS .................................................. 111 4.1 INTRODUCTION .................................................................... 1 1 l 4.2 RHEOLOGICAL PROPERTIES .................................................. 113 4.2.1 Viscosity .................................................................... 113 4.2.2 Storage and Loss Moduli .................................................. 117 4.2.3 Morphology Analysis .................................................. 126 4.2.3.1 Network Morphology .................................................. 126 4.2.3.2 Polymer-Phase Morphology ......................................... 143 4.2.4 Effect of Processing on Short-Term Aging ....................... 154 CHAPTER 5 DATA ANALYSIS AND DISCUSSION .................................................. 157 5.1 INTRODUCTION .................................................................... 157 5.2 THE STRUCTURAL AND ENGINEERING PROPERTIES .............. 159 5.2.1 Styrene Butadiene Styrene (SBS) PMA Mixtures ....................... 160 5.2.1.1 Load-Deformation Characteristics ................................ 160 5.2.1.2 Tensile and Compressive Strengths ................................ 162 5.2.1.3 Resilient and Equivalent Moduli ................................ 175 5.2.1.5 Plastic Deformation Characteristics ................................ 181 5.2.1.6 Binder-Aggregate Adhesion Properties ....................... 185 viii ' C L , '1 I a ‘ "7"" n‘.-.‘ (Olinf... .\1:\’.‘.‘r: ~» ' 1 ~ I .‘.....‘..1 ‘Y‘ nu‘ '\ x .11 ---1. D‘ ‘ MI ---.I.- .“\ “' \l "—-|_ \_ Mm Mr .rmm a GENERI'L " 5:315 "HUN (1 i 6‘} r1.; :- IBHM‘L R“ ' \ AV, \ . " €71“: 5.2.2 Other Polymer Systems .................................................. 196 5.2.3 Comparison-Structural and Engineering Properties of PMA Mixtures .......................................................................................... 196 5.2.3.1 Comparison-Tensile and Compressive Strengths .................. 196 5.2.3.2 Comparison-Resilient and Equivalent Moduli .................. 208 5.2.3.3 Comparison-Plastic Deformation Characteristics .................. 213 5.2.3.4 Comparison-Binder—Aggregate Adhesion Properties ...............216 5.3 FATIGUE LIFE ANALYSIS .......................................................... 227 5.3.1 Fatigue Life Criteria ........................................................... 227 ' 5.3.1.1 The Rate of Accumulation of HPD Criterion .............. 228 5.3.1.2 The Rate of Crack Growth Criterion ................................ 230 5.3.1.3 Comparison of the two Fatigue Life Criteria ....................... 234 5.3.2 Fatigue Life .................................................................... 244 5.3.3 Rut Potential .................................................................... 254 5.3.4 Temperature Cracking .................................................. 260 CHAPTER 6 TENSILE (FATIGUE) AND COMPRESSIVE (RUT) STRAINS MODELS .....273 6.1 GENERAL ............................................................................. 273 6.2 SELECTION OF VARIABLES .................................................. 274 6.3 CORRELATION ANALYSIS .................................................. 274 6.4 REGRESSION ANALYSIS ........................................................... 276 6.4.1 Fatigue Life .................................................................... 281 6.4.2 Vertical Plastic Deformation (VPD) ......................................... 291 CHAPTER 7 SUMMARY, CONCLUSIONS AND RECOMMENDATIONS .............................. 297 7.1 SUMMARY ............................................................................ 297 7.2 CONCLUSIONS ......................................................................................... 299 7.3 RECOMMENDATIONS .............................................................................. 300 LIST OF REFERENCES ix E31, .11 uka v EuJ .0 1s\\\\ \. EOO"..I I Aib\ 3 LIMJC" 11.. NH ,.. \. Table 2.1 Table 2.2 Table 2.3 Table 2.4 Table 2.5 Table 2.6 Table 2.7 Table 2.8 Table 2.9 Table 2.10 Table 2.1 1 Table 2.12 Table 2.13 Table 3.1 Table 3.2 Table 3.3 Table 3.4 Table 3.5 Table 3.6 Table 3.7 Table 3.8 Table 3.9 LIST OF TABLES Effect of polymers on binder penetration. ................................. 19 Effect of polymers on binder ductility. .................................. 20 Effect of polymers on cracking temperature. .................................. 23 Viscosity measurements of polymer modified asphalt. ................ 27 Effect of polymer on the ring and ball softening ball. ................ 30 Effect of polymers on resilient modulus and tensile strength. . ...40 Effect of polymers on fatigue lives of mixture. ......................... 43 Summary of Wheel Tracking test results. ......................... 45 Limiting stifliress values for the asphalt mixtures. ......................... 47 Stiffness values in (psi) for a (20,000 sec.) loading time. ................ 48 Prioritized list of asphalt modifiers and related issues. ................ 50 Summary of the literature review. ........................................... 51 Summery of results for field performance. .................................. 54 Specific gravity and absorption capacity of the aggregates. ................ 69 Summary of mixing time and temperature for PMA binders. ...75 Aggregate gradation; G1 and asphalt contents. ......................... 76 Aggregate gradation; G2 and asphalt contents. ......................... 77 Aggregate gradation; G3 and asphalt contents. ......................... 78 Aggregate gradation; G4 and asphalt contents. ......................... 79 Aggregate gradation; GS and asphalt contents. ......................... 80 Aggregate gradation; G6 and asphalt contents. ......................... 81 ............... 82 Aggregate gradation; G7 and asphalt contents. I 'r, I.» v .le.“ --:.u.. I I .w-.‘.- L..obut s or a T u. hr?“ ' at _v \ as... :1 ll\\.L- q 1. n u‘ -\ 1‘4 «v \- nu“. I‘m. A. CUM-n.“ 1 m“! N‘.-" 5 o .5 I I o 1‘ " a .. . . . Ab . SLL‘YJ‘L N In ‘ ~| \ inhir\ ' c ‘- ~ A .‘ \l'n‘"\ ‘- U1... .3. Ct'anL’i chfiz-st ‘ ..l . :;‘ “ah ‘\ . .. . ‘_.‘-\~ '~ I ‘ ;u Table 3.10 Table 3.11 Table 3.12 Table 4.1 Table 4.2 Table 5 .1 Table 5.2 Table 5.3 Table 5.4 Table 5.5 Table 5.6 Table 5.7 Table 5.8 Table 5.9 Table 5.10 Asphalt mix design results for AC5 straight and the aggregate gradation G7. ...................................................................... 89 Theoretical maximum specific gravity of AC5 and aggregate gradation G7. ............................................................................... 90 Summary of the results of the asphalt mix design for the straight and polymer modified asphalt mixtures. ............................................ 96 Viscosity of the straight, processed and PMA binders (centipoise; cP). ............................................................ 1 14 Storage modulus (G’), loss modulus (G”) and tand values of three PMA binders at temperatures of 95 and 1400F. ................................ 119 Summary of the indirect tensile strength test results of asphalt concrete mixtures at test temperature of 77oF. ................................ 164 Summary of the indirect tensile strength test results of aged asphalt concrete mixtures at temperature of 77°F. ................................ 172 Resilient and equivalent moduli of SBS PMA mixtures at 77°F . 1 78 Summary of the tensile and compressive strengths of straight, processed and PMA mixtures tested at their respect optimum polymer contents and at 77°F. ..................................................................... 1 19 Summary of the tensile and compressive stress ratios of straight, processed and PMA mixtures tested at their respect optimum at 77°F . . .204 Summary of the tensile and compressive strengths of oven aged PMgA mixtures tested at their respect optimum polymer contents and at 77 F .207 Summary of the resileint and equivalent moduli of straight, processed and PMA mixtures tested at their respective optimum polymer contents andzag9 77°F . .............................................................................. Quantitative summary of fracture mophology of PMA mixtures. .....223 Summary of lap-shear strengths and fracture toughness of PMA mixtures at various temperatures. ................................................... 224 Sample characteristics and sample geometry for the analysis of the fatigtge5 life criteria. .................................................................... 2 xi 3‘; 0‘, ‘ It ‘- r," '10 A ..,..... . .8... mans. A Stiff 1‘"? 7‘5 '~ .1 51.1": I'Cif‘t‘t', The 311.: FI\. .' 11.1““. . Tr!) ‘F ‘55 u. \ t1. \(. -l P" ‘ ~ 1.01“ . 5"? 1'3: 5, Table 5.1 1 Table 5.12 Table 5.13 Table 5.14 Table 5.15 Table 5.16 Table 5.17 Table 5.18 Table 6.1 Table 6.2 Table 6.3 Table 6.4 Table 6.5 Table 6.6 Table 6.7 Table 6.8 A summary of the indirect tensile cyclic load test results of SBS PMA mixtures at test temperature of 77°F . ................................ 246 A Summary of the average fatigue lives of PMA mixtures for various polymer contents at 77°F . .................................................. 251 A Summary of the increases in the fatigue lives of PMA mixtures with respect to straight mixtures for various polymer contents at 77°F . ....... 253 The number of load cycles required to develop 0.01 inch of VPD of PMA mixtures for various polymer contents at 77oF. ............................. 258 The increases in the number of load cycles required to develop 0.01 inch of VPD in PMA mixtures of various polymer contents at 77oF relative to the straight mixtures. ........................................................... 259 A summary of the indirect tensile strength test results of asphalt concrete mixtures at test temperature of 23°F . .......................................... 263 A summary of the tensile strength and fracture toughness of straight, processed and PMA mixtures (at optimum polymer contents) at 23°F. .269 A summary of the equivalent moduli and vertical deformation at failure of straight, processed and PMA mixtures at their respective optimum polymer contents at 23oF. ................................................... 270 The most influential variables that affect the fatigue life (FL) and the vertical plastic deformation ( VPD). ........................................... 275 Pearson correlation matrix for SBS polymer system. ....................... 277 Pearson correlation matrix for SEBS polymer system. .............. 278 Pearson correlation matrix for SBR polymer system. .............. 279 Pearson correlation matrix for the variables of VPD model for three polymer systems. ........................................................... 280 A summary of the transformation functions for each independent variable. .................................................................... 286 A summary of nonlinear regression analysis of the fatigue model for three polymer systems. .................................................. 289 A summary of the transformation function for each independent variable. .................................................................... 292 xii r‘uk e 688 O 1 9 1 t . ... I‘ ... g r ..... Ion analySIS of the V PD models for three l’(, Vlllel E; y S 81118 ............................................... 29 I xiii . “ .wo; min-b .9 U ’ 11 t’!‘ ' 3. l'fidb . . - ‘ 1 I"‘.: 4 ”one .. . . - ~.. . ..‘2 ‘ \ 5“ u.' ‘ .-.k 10“ L Tun-y“. V iSk‘Ol‘. \TSCOC. R6>uh. Figure 1.1 Figure 2.1 Figure 2.2 Figure 2.3 Figure 2.4 Figure 2.5 Figure 2.6 Figure 3.1 Figure 3.2 Figure 3.3 Figure 3.4 Figure 3.5 Figure 3.6 Figure 3.7 Figure 3.8 Figure 3.9 Figure 3.10 Figure 3.11 Figure 3.12 LIST OF FIGURES Flow chart for the research plan. ............................................ 4 Two-phase asphalt model representing asphaltenes, resins and oils. ........ 9 Viscoelastic model for asphalt behavior. .................................. 16 Effect of polymers on F raass point. ........................................... 24 Stress-strain behavior of conventional asphalt mixes. ................ 32 Viscoelastic behavior of asphalt. ........................................... 35 A summary of fatigue test characteristics. .................................. 37 Viscosity-temperature chart for straight and polymer modified asphalt mixtures. ...................................................................... 71 Aggregate gradation curves (G1 to G3) for mix design along with MDOT maximum and minimum specifications. .................................. 83 Aggregate gradation curves (G4 to G6) for mix design along with MDOT maximum and minimum specifications. .................................. 84 Aggregate gradation (G7) curve. ........................................... 85 Results of Marshall mix design for AC5 mixtures. ......................... 91 Results of Marshall mix design for AC10 mixtures. ......................... 92 Results of Marshall mix design for AC20 mixtures. ......................... 93 Results of Marshall mix design for AC5 mixtures with 5 percent SBS polymer. ...................................................................... 94 Results of Marshall mix design for AC10 mixtures with 2 percent SBS polymer. ...................................................................... 95 Typical load deformation cycles with 0.1 second loading time and 0.4 seconds relaxation period. .................................................... 98 Stress-strain behavior of conventional asphalt mixtures. ................ 99 Typical load-deformation curve for straight and modified asphalt xiv a. S ‘ ”Ti‘ wink - . . 'Fn' ' w afi. . mutur. Than? 1651. The ex" Vin-K.“ WSW". I" "v tht\1. 113°.“ 1 1“ "bth. Disk” E1EX1 <.P.. ~>~1 \ ; a ~““-4|. ~ E3131 1.5851 ESEX: «1".11 M 111 ..1( ESLLI ~11L171. E51); 5 l i I ., ‘ w .l.,,‘\., Figure 3.13 Figure 3.14 Figure 4.1 Figure 4.2 Figure 4.3 Figure 4.4 Figure 4.5 Figure 4.6 Figure 4.7 Figure 4.8 Figure 4.9 Figure 4.10 Figure 4.11 Figure 4.12 Figure 4.13 mixtures. ...... q .............................................................. 104 The experimental program matrix for the indirect tensile cyclic load test. ............................................................................. 106 The experimental program matrix for the indirect tensile strength test... 107 Viscosity at 2750F of SBS, SBR and EAM PMA binders as a function of polymer content and mixing conditions. ................................ 1 14 The storage modulus (G’) and loss modulus (G”) of AC5-SBS PMA binders as a function of the SBS polymer content at 95 and 140°F. ..... 124 Effect of polymer content on the storage modulus (G’), loss modulus (G”) and tanfi of AC10 modified with SBR polymer at temperature of 140°F. ............................................................................. 125 ESEM micrographs showing network morphology of AC5-straight binder and AC5 modified with five-percent SBS polymer content. .............. 128 ESEM micrographs showing the kinetics effect on the development of the network morphology of the AC5-straight and AC5 modified with 5 percent SBS polymer content. ......................................... 129 ESEM micrographs showing morphology and network structure of AC10 straight binder and the AC10P/AC10A mix. ................................ 131 ESEM micrographs showing morphology and network structure of the ISBSAC10/AC10A and 6SBSAC10/AC10A mixes. .............. 133 ESEM micrographs showing network morphology of the AC10-processed and AC10 modified with 5 percent SBR polymer content. .............. 134 ESEM micrographs showing network morphology of the AC10-processed and AC10 modified with 3 percent EAM polymer content. .............. 136 ESEM micrographs showing network morphology of the aged AC5 & AC10-EAM processed binders. .......................................... 138 ESEM micrographs showing network morphology of aged AC5 and AC10 binders modified with 3%EAM polymer. ................................ 139 ESEM micrographs showing network morphology of the AC5-CRM processed binders and stored at 350°F for 0, 5 and 24 hours ............... 140 ESEM micrographs showing network morphology of the AC5 modified XV . ‘1 . ‘P’J‘ :4 .‘fi' ‘ . “me; e.“ ( g. "-~‘._1 . .7- , v; I .g‘.* ' | "L, I {J . 4", ’i II (I. 5.. 15V. 1‘. cm"r III! 03.. W, 1.u1‘.t\ 1‘. \L' Isn't. 5 TN. 7’ .".L 'I‘, 3 tkll r; ‘ .i ‘ ‘1',n\ lau‘..\ ‘ 11"“ " \nu‘. ‘ {A L‘\\ 1., r . _. .“i\1 Figure 4.14 Figure 4.15 Figure 4.16 Figure 4.17 Figure 4.18 Figure 4.19 Figure 4.20 Figure 5.1 Figure 5.2 F igue 5.3 Figure 5.4 with 10 percent CRM content and stored for times of 0, 5 and 24 hours at 350°F. ............................................................................ 142 LSM micrographs showing the surface of AC10 straight binders, a), b) & c) with reflected light at 100x, 200x and 1000x, respectively; and d) with fluorescent light at 1000x. The sample was imaged using blue (488nm) light. ................................................................................ 144 LSM micrographs showing the surface of AC10-5%SBS binders, a), & c) with reflected light at 200x and 1000x respectively; and b) & d) with fluorescent light at 200x and 1000x respectively. The sample was imaged using blue (488nm) light. .................................................... 145 ESEM micrographs showing polymer-phase morphology of the AC10 modified with 5 percent SBS content before and after exposure to the electron-beam. ........................................................... 147 LSM micrographs showing the surface of a), & b) AC10-straight binder c) & d) AC10-5%SBR binder. a) & b) with reflected light at 333x; and b) & d) with fluorescent light at 333x. .......................................... 149 LSM micrographs showing the surface and polymer distribution of an AC10-3%SBR a) & c) with reflected light at 200x; and b) & d) with fluorescent light at 200x. .................................................... 150 LSM micrographs showing the surface and polymer distribution of AC10- 1%SBR a) & c) with reflected light at 200x; and b) & d) with fluorescent light at 200x. ....................................................................... 151 ESEM micrographs showing polymer-phase morphology of the AC10 modified with 5 percent SBR content before and after exposure to the electron-beam. ........................................................... 1 53 Typical load-deformation curves of straight and SBS PMA mixtures under static loading at 77°F. ........................................................... 161 Tensile and compressive strengths of AC5 and AC10 mixtures modified with SBS polymer as a function of polymer content at 77°F. ............ 163 Stress-Ratio of AC5 and AC10 mixtures modified with SBS polymer as a function of polymer content at 77°F. ......................................... 169 Normalized tensile strength of oven aged SBS PMA mixtures as a function of polymer content at 77°F. ................................ 171 xvi , n l ..-.,\‘ .‘an ow I I .“,.‘ ..JC-U 0! r"3‘ ”Adar b . I ”'11 i~nb .. .. l '7'9‘9 ‘i‘b... u v>.}." .‘t."‘:l'l .‘~‘. '1 't-Cfii ”H. 1- ‘4, I1 ' "i‘ ~‘4 I n '\ ‘ .‘\ i A' . 5. ‘ - i Figure 5.5 Figure 5.6 Figure 5.7 Figure 5.8 Figure 5.9 Figure 5.10 Figure 5.11 Figure 5.12 Figure 5.13 Figure 5.14 Figure 5.15 Figure 5.16 Figure 5.17 Figure 5.18 Figure 5.19 Normalized resilient and equivalent moduli of SBS PMA mixtures as a function of polymer content at 77°F. ................................ 177 Effect of the oven aging of equivalent moduli of straight, processed and SBS PMA mixtures at 77°F. .................................................. 180 Accumulation of horzontal and vertical plastic deformations (HPD&VPD) of AC5 SBS PMA mixtures as a function of the number of load cycles at 77°F . ............................................................................. 182 Horizontal and vertical plastic deformation (HPD & VPD) rates of AC5- SBS PMA mixtures as a function of polymer content at 77°F. 184 The lap-shear strength and fracture toughness of AC5-SBS PMA binders as a function of polymer content at 68°F. ................................ 187 ESEM micrographs showing strands in AC10-2%SBS PMA binders for various displacements at 68°F. ......................................... 188 Normalized Lap-shear stiffness and fracture toughness of aged AC10-SBS PMA binders as function of polymer content at 68°F . .............. 190 Typical fracture surfaces of the failed straight and PMA mixtures at 23°F. ............................................................................. 193 The normalized lap-shear strength and fracture toughness of AC10-SBS PMA mixtures as a function of polymer content at 32 and 14°F. 195 ESEM micrographs showing failure processes and fibrils of AC10 mixture modified with 2 percent SBS polymer content at 32°F. 197 The normalized tensile strength of straight and processed mixtures as a function of the mixing conditions (temperature & time) of various polymer systems. ........................................................... 200 The tensile strength of PMA mixtures at the optimum polymer contents normalized relative to the straight mixtures. ................................ 202 The tensile strength of oven aged AC 10 PMA mixtures at the optimum polymer contents normalized relative to the straight ones. .............. 206 The normalized resilient moduli of straight and processed mixtures as a function of the mixing conditions (temperature & time) of the various polymer systems. ........................................................... 210 Resileint and equivalent moduli of processed and PMA mixtures xvii 3m i 1ll ...a. ' v.4 p 1"": N ‘11 1“. ..- u '3‘ . “1°51. >>I" . ..- ~ ° ";1 ‘ i-i . u . a F; \ ah. 0“ - O ...3‘1 q. \ norm..- Curr. m, 9]. .. 1111\Iu. \ RIC iix n- E1 ESE\1 pCIC'CF' 915:: ESL\f AC5? ESEXT an ‘ '1- l a.\ r.. f‘.’ r-- ‘-\. P0713? 'h 1~. 11L :1 ‘w 4 r .... ‘\l,|‘I:‘V‘. Figure 5.20 Figure 5.21 Figure 5.22 Figure 5.23 Figure 5.24 Figure 5.25 Figure 5.26 Figure 5.27 Figure 5.28 Figure 5.29 Figure 5.30 Figure 5.31 Figure 5.32 Figure 5.33 normalized with respect to the processed mixtures at 77°F . .............. 212 Cumulative of the horizontal plastic deformation (HPD) of AC 1 0-PMA mixtures as function of the number of load cycles at 77°F . .............. 214 The horizontal plastic deformation rates of AC10 PMA mixtures at 77°F. ............................................................................. 215 ESEM micrographs showing fibrils in AC 10 mixtures modified with 2 percent SEBS polymer content at 68°F at two magnification rates. 217 ESEM micrographs showing fibrils of AC5 mixtures modified with SBR polymer and subjected to two displacement levels. ....................... 219 ESEM micrographs showing crack and fibril morphology in processed AC5 binder and AC5 modified with 1 and 2 percent EAM content. 220 ESEM micrographs showing (a, b) the interaction between rubber particles and the asphalt network for two exposure times and mgnification levels and (c, d) the typical fracture morphology in AC5 modified with 10 percent CRM at two magnification levels. ................................ 221 Shape factor (b)for a circular plate with crack at the center. .............. 232 Typical plots of the total modulus, normalized compliance and half the crack length as a function of the number of load repetitions at 68°F . ....237 Typical plots of cummulative HPD, normalized rate of the crack growth and cumulative HPD as a function of the number of load repetitions at 68°F. ............................................................................. 238 A typical K-da/dN curve for indirect tensile sample of asphalt concrete mixtures. .................................................................... 240 Comparison of the fatigue lives calculated using between the rate of cumulative HPD and the rate of crack growth criteria. .............. 242 Three-dimensional crack propogation model for centrally cracked cylindrical sample subject to indirect tensile cyclic load. .............. 243 Effect of polymer content on the fatigue lives of SBS PMA mixtures at 77°F. ............................................................................. 245 The number of load cycles required to develop 0.3 inch of vertical plastic deformation in the SBS PMA mixtures tested at 77°F. .............. 255 xviii fine 3:4 x033”... itiw v I‘. - ,Po;\“ \‘F'V‘ ..Lau..-. . L...... . ‘fi ‘3‘ 1 . u..\ s.' .~....(, - .‘. . 9 \ ‘ Th.) '.‘. :“" A£\ n. . 1L " ' I v. F“ . '. 5" ‘~ “b\,.1 FJ.A‘, ‘ \ 7 ,\‘.. It “\ .-.._~‘ ‘~:*.. F.1" “~»-‘~ s -o 5 ’r‘?‘ I' \..§ \ .1 "“ 0‘ ' N "s ‘h‘ ' .‘§‘_. )0,“ ““hu~ \ Ii . ‘- hi . t‘l‘ ‘1' FE" v— . N, . t‘:§§ P'AS 1LH.‘\' "793;, r “ ~.Q P151114 \ Figure 5.34 Figure 5.35 Figure 5.36 Figure 5.37 Figure 6.1 Figure 6.2 Figure 6.3 Figure 6.4 Figure 6.5 Figure 6.6 Normalized tensile strength, fracture toughness and equivalent moduli of AC10-SBS PMA mixtures at temperature of 23°F. ....................... 262 Normalized rates of HPD of AC5 and AC10 SBS PMA mixtures as a function of polymer content at 23°F. ......................................... 265 The stiffness of AC5-SBS PMA as a function of the polymer content at temperature of -l 1°F. ........................................................... 266 The rates of accumulation of horzontal plaastic deformation (HPD) of AC10 PMA mixtures at 23°F ................................................... 271 Fatigue life of AC5-SBS PMA mixtures as a function of the horizontal plastic deformation rate at failure. ......................................... 282 Fatigue life of AC5-SBS PMA mixtures as a function of indirect tensile strength (ITS). ........................................................... 283 Fatigue life of AC5-SBS PMA mixtures as a function of tanB. 284 Fatigue life of AC5-SBS PMA mixtures as a function of polymer content. .................................................................... 285 Predicted versus observed fatigue lives of the three polymer systems... 290 Predicted versus observed VPD for three polymer systems. .............. 296 xix ll GENERtL ~v~ 'rw- gangmc‘fl' " i . .- l‘ - .La:.§....‘p \i . i. 0“,. q. ‘21,”... I .h'....\..-: .3 a“ 1\ u "‘.|‘ ~0 e. 4 — \ ta»»&...‘h‘§.‘ L . I 9"» ' ..\ .. - ~ . \ :"' n'V‘M'w‘d” . 1...-.. ‘I \ . .u 13th.... ~ i. ‘ i c siL-i “.1 j : \'.. "s. ~i¥nu£§¢4 Lat-\- u. o:‘ i. . x M. _"'~1‘.““\T I'c‘ v‘ o .e , “3:"; . 1. “ti-b! ~~., ..‘TNF". -. 1 Nu». lit. V. 13* .. ‘- "\ ';. ‘., .¥\.a ‘x. . \’. ..1-. D CHAPTER 1 INTRODUCTION 1.1 GENERAL The characteristics of the asphalt concrete play an important role in the short and long-term pavement performance. The performance of asphalt concrete surfaced pavements is a fiinction of traffic loads and volume, the physical and engineering characteristics of the asphalt concrete, the properties of the aggregate base and sand subbase and the environment. Typically, pavement deteriorates over time due to environmental cycles and increase in the number of traffic load repetitions. This deterioration manifests itself in several types of distress modes such as rutting, fatigue cracking, thermal cracking, stripping, and ravelling. An asphalt mixture made with AC20 asphalt grade, generally performs well relative to rutting at high temperatures. It is because AC20 asphalt grade maintains a high stiffness at high temperatures, which makes the mix more resistance to plastic deformation. However, this AC20 mix will be extremely susceptible to thermal cracking because of the brittleness and lack of flexibility at low temperatures. The opposite is true for asphalt mixtures made with AC5 asphalt grade. They perform well at low temperatures but are highly susceptible to rutting at high temperatures. Thus, the ability 0f attaining an asphalt pavement that will perform well at both low and high temperature iS the key to improving performance. Various researchers (1, through 7) have shown that polymers can successfully improve the performance of asphalt pavements at low and high temperatures. The .. - I. )'- j u““" 1’. ‘:\"‘.- \ dab-.43.; L‘s 311' .u\.- u ‘ .;-\-9‘“',)J VF ",}". . b‘..hdb\bU LA: in\\‘ :3: “11:.‘4'83 01 h . l-‘IJ‘“ i.” 'P ‘ : _L...DU“ X“ ~'."""\' ‘Wfi‘n -v- . 1 u D .. :..~\. 5L1...w\\-I 1. i ' 'h. “9" r1 - an» .: .‘ e1.\ 5 ;L:;::T Z: VhC pr‘ )1.“ q. .... \.u4‘:\ undc ‘r‘l-J. , 5. iii; ' '3 D) 1.; ~m. 5Flruv‘t T1 - rjc T‘" . n, k . “3 L3?” K‘k m")? :‘ixkiPa-i k“\) .9 .~ i“ )i‘ iCiU'fi . . g“ 31.1" 1‘8, s “ . 3, . - N 401", inc “if: “,2.” ‘3‘ dc“ addition of polymers also improves the fatigue and tensile strength properties. Shuler et al (5) conducted indirect tensile tests using AC5 binder modified with SBS polymer system at temperatures of 6, 77, and 106°F. They reported that the tensile strength of AC5 modified with 6 percent SBS polymer increased significantly over that of the straight AC5. Based on a stress-controlled fatigue testing and a fracture mechanics evaluation, Button et a1 (6) reported that AC5 modified with SBS polymer exhibited superior fatigue properties compared to the straight asphalt at 68 and 32°F. A three phase research study was conducted at Michigan State University (MSU) to characterize the properties and the possible benefits of polymer modified asphalt (PMA) mixtures under low, intermediate and high temperatures. The study was funded by the Michigan Department of Transportation (MDOT). The three phases are: 1. The fundamental physical, chemical, and thermodynamic properties of polymer- asphalt binder. 2. The basic morphology and microstructure of polymer-fiber-asphalt-aggregate mixtures. 3. The structural and engineering properties of the polymer modified asphalt mixtures. This dissertation focuses on the structural and engineering properties of PMA mixtures and their interaction with rheological properties, morphology and microstructures of PMA mixtures. 12 mmmrs W’s ofplls‘x '* ,.......b. A h‘ ‘u; 9“,; ‘H . in5\ ntl\l_i\".:\ ‘ T u - Pg "|rI-\h . Ah m‘. i! ..I A‘ ' "‘ "v '1- LILUA. ~ 3"," '1"\ '~.. _.¢-.\. ..l. ‘ ~..\\ 'I h-- ' 'v‘ L‘_ t "f‘ 3.4-J “‘"‘r A..5\u‘ I ’ ‘ i 33 RESEARt'H' ~16 Gian c ' ? r ' 3:,f'fi'm, o5. "“HQL u \. mi ' ‘ .1 ..Ifi \ hug ‘t.‘._",""‘"' s““\x L)“. w ”Hui \ 'v J‘QFO; ‘ H\.‘i‘r“ 9L.‘ . Mk \ *4 '~ ‘ 1“ 1\ 3i" l,‘,l' ’1‘ x x . ‘k. 1“. .I t‘ . <_ _ ‘a‘.-O““ ‘ p. ,' > .. 1.2 HYPOTHESIS It is hypothesized that the improvement in the structural and engineering properties of PMA mixtures can be related to: The rheological properties of the polymer-asphalt binder The morphology of the polymer-asphalt network structure and binder-aggregate adhesion. It is further hypothesized that a relationship can be made between rheological and the structural properties of PMA mixtures. 1.3 1.4 RESEARCH OBJECTIVES The objective of this study can be summarized as follows: Determine the effect of different polymers and polymer contents on the structural and engineering properties of asphalt mixtures at low, moderate and high temperatures. Identify the controlling fundamental properties that affect the engineering properties. Develop a model relating the rheological properties of the polymer-aspahlt binder to the engineering properties of these mixtures. RESEARCH PLAN To accomplish the above-mentioned objectives, a proposed research plan was designed and is shown in Figure1.1. A comprehensive literature search was conducted to identify different type of polymers and the improvement in engineering properties due to finial. ( hemlw. Tltrmrindmmx Prop: £831!th P‘l \ Bl TN. R a. P ‘ mu. 2 . .‘05. , 'h..1\ V‘. 4z':~‘ Q" RESEARCH PLAN tructural ropertiesof. Straight/PMA ..., ‘" '* Mixtures ‘ Materials (AC, Agg, Polymers) <————> V Experiment Program Matrices Literature Review Marshall Mix Design/MDOT Spec } Indirect Indirect Tensile Tensile Strength Cyclic Load Test Test Morphology and. . Microstructure of" ‘I Straight/PMA Mixtures V l ' l. Tensile strength 1. Fatigue life 2. Fracture toughness 2. Resilient modulus 3. Compressive strength 3. Horizontal plastic 4. Equivalent modulus defamation 5. Vertical deformation 4. Vertical plastic (Failure). defamation. 1. Glass transition I. Networ< temperature. morphology 2. Dynamic _ -,~: ' ' analysis. mechanical ' - A 2. Fracture tou ness. analysis (G’&G”). ' DataAnalysls ‘ 3. Void analysigsl.‘ 3. Viscosity. 1 . . » 4. Binder/Aggregate interface analysis. t1gue&Rut Models a V"Concltisibns - p and ‘ "Recommendations Figure 1.1 Flow chart for the research plan. C;.. “'; 3.3.4». .AJOUliixdut "‘ .1 ‘ - n. i. :€.§ ‘ :A'S \\'..~,§ . ‘ ' -.'_.._. ii . L 3;...)4‘3. A1; 3&3. 7.1: 1'; 'i‘a l" ""- .~‘¢.“~5 A i "‘ q. u ' I . yva 0);.» i w . y“ ‘e' 0-“>*‘. \\.t:. Adlhhl v ..... _. ‘ 1 »_; :e v .. K. v."- A.‘\. a. AK 5‘ |_ \ i“ w l . : . '2 0". c. :S-nes “‘. k‘ i I _._\,.. . \‘ ‘ ”1"“, _ .1‘ . \. “\“‘~~\ 1. PM . ‘ct “\1~. wk, 32;) ~‘ \0 \ . \ ~§ ' .‘ y- n l\ _ v ‘\ Us. ...i'u ‘uk .0, ‘ polymer modification. Different materials such as asphalt cements, aggregates and polymers were collected for preparation of mixtures. Experiment design matrices were established. All asphalt mixtures were designed using the Marshall mix design method and the specifications of the Michigan Department of Transportation (MDOT). In order to evaluate the engineering properties of straight and PMA mixtures, two types of structural tests; indirect tensile strength test and indirect tensile cyclic load test were conducted at low (23°F), moderate (77°F) and high temperatures (140°F) using Marshall size samples. In general, the structural and engineering properties and failure mechanisms of asphalt mixtures are functions of the binder chemical, physical, and rheological properties, aggregate gradation, fine content and angularity, binder-aggregate adhesion, and voids (size and distribution). For polymer modified asphalt mixtures, the failure mechanism is also a function of the polymer-asphalt chemistry, the morphology of the polymer-asphalt network structure, and the amount and distribution of the polymer. To assess the interaction of these properties on mixture performance, test data from the other two phases of the study was incorporated into this phase as shown in Figure 1.1. The data was analyzed in conjunction with the structural and engineering properties. Finally, fatigue and rut models were developed using the statistical methods. 1.5 THESIS LAYOUT This thesis is organized into seven chapters as follows. Chapter 2- Literature Review Chapter 3- Laboratory Investigation Chapter 4- Rheological, Morphological and Microstructural Properties of PMA Binders Chapter 5- Data Analysis and Discussion Chapter 6- Tensile (Fatigue) and Compressive Strains (Rut) Models Chapter 7- Summary, Conclusions and Recommendations :1 GENERAL ‘9 m? \ “" Ill-‘1 5‘ .\_ “‘1‘ ' I “A r 7' mm: «.»‘s- all“ E ‘gai \r. . 'fi. ‘ I' '. I.“ . §‘.‘o3‘ll‘.ntln\\ dgi'uv. I ~‘"r" \. '7 .- - . ’3‘ Y';- I .i ~:\..~‘ A ‘ Is A (I, 1.,",~,JI;T 1'. . ...\ ..JKMJA'STM EL\ n \ a» ”‘2" , ,1 ' ~~|k.-u-I: chKALr-s ‘u M. ‘-. 2%.. “‘5,“ ..J\ v-J . ' ‘nxa 2 - ._ ‘Il 5. \\.v ‘1 ..s ' ”v,‘ 4 "\e ”I i A511. .- , ' \\.l ‘. .- I r~\“.:‘. h h“ ,i k v K ‘\.. ‘ x 'L \1 2.. ‘- “N. ‘\ hr'\ - CHAPTER 2 LITERATURE REVIEW 2.1 GENERAL In order to understand the effect of polymer modifiers on the pavement performance, it is necessary to understand the basic properties of asphalt binder and mixtures that control the performance. Pavement performance is directly related to the way asphalt performs under the different traffic and environmental conditions. This chapter presents a review of the basic structural and engineering properties of the unmodified and polymer modified asphalt binders and mixtures related to fatigue cracking, resistance to low temperature cracking and rut potential of asphalt concrete pavements. 2.2 ASPHALT CEMENT COMPOSITION Asphalt is a complex mixture of many different hydrocarbons consisting primarily of molecules that contains mainly carbon and hydrogen atoms. A generic asphalt elemental analysis shows that approximately 84 percent of the sample is carbon, 10 percent hydrogen, 1 percent is oxygen and the remainder consists of several trace elements including nitrogen, vanadium, nickel and iron (8). For convenience, asphalt constituents are classified into three categories: oils, resins, and asphaltenes. Oils are the light compounds in the asphalt, which have lower molecular weights (25-800). Resins are intermediate molecular weight compounds (800-2000) and asphaltenes are the highest molecular weight compounds (> 2000). As mentioned earlier, asphaltenes contains trace elements, and these elements may react with polymer (9). An average asphalt sample has ~ . v _ v Q a” -..-..~>-.v "' 5“] \ t . ‘5‘ “.W‘ '\ “_ AAI-‘ -..n-h ~ . .w v- - xv- .Q“ ,. 3 .9\ q h.._b§tab- I“ V ‘:I ‘--‘ I gn.‘ I h .|.' 5'.'...\'...'u\-ls.\- ..\ t «. w “H--,o,.'..‘e ...' b‘. .QL.».4\....\ . hi ‘F‘;'\w - \\ I 4 \ ‘ 4 ‘ ..3 F a - ~‘.—-—.~\.z\) \ VS“~.\ ” 1 » 5‘ § “0.; 5.... e | .1 .1 ‘ an M3 ....'s\..‘\ tk‘rTTl \ 3",". -1 )n a .- ui~l\‘\'\l‘ .‘ u t . g i ‘ w . a-._\._r§“\ "g-w‘ r... “ ‘tthn..- \ 5 . . . ya . \ ‘51L‘u :‘qe n‘o s “"'-._‘_ '-., 4 \ 5., ‘0‘ u.‘::\ {)rn‘q‘r‘ ' C S '3'. ‘ . h i the.‘ ‘3"an " an asphaltene/resin/oil ratio of approximately 23/27/50 and asphaltene content for AC20 asphalt is higher than AC2.5 (10). Using the asphaltene/resin/oil classifications, a two-phase asphalt model was developed (10) as shown in Figure 2.1. Phase-I is an assembly phase consisting of asphaltenes and resins that are dispersed in phase-II, the solvent phase consisting of the oily constituents. The resins behave as peptizing agents that stabilizes the asphaltenes in the oily constituents. Moreover on oxidation, they yield asphaltene type of molecules. Asphaltenes can exist both in a randomly oriented particle aggregate form and or in an ordered micelle form. They are the most complex components with the highest polarity and have a very high tendency to interact and associate. Asphaltenes play a major role as the viscosity building component of asphalt cements. The engineering properties of asphalts are directly related to the quantity of asphaltenes, the size of the micelle structure, and the nature of the dispersion medium, oils and resins. Asphalt molecules contains many organic functional groups such as ketones, phenols, sulfoxides, acid anhydrides, pyroles, and quinines. These functional groups can react with a reactive polymer modifier, such as an epoxy. They are also important in age hardening, which is essentially oxidation of these functional groups, which cause increased stiffness (10). 2.3 ASPHALT BINDER PROPERTIES 2.3.1 Low Temperature Properties It is well known that thermal cracking of asphalt concrete pavement is directly related to the stiffness and tensile strength of asphalt binder at low service temperature. /O Resms Asphaltenes O ._ “—0 @ /V Oils 4’ Phase I = Asphaltenes + Resins (Assemblies) Phase II = Oils (Solvent) Figure 2.1 Two-phase asphalt model representing asphaltenes, resins and oils. isr to cor; .. 1p- 3‘? ”IQ. J: L‘“ . .. , ~H'xfi“; {fro-1 "r- n Lek-hes”; -t—5in:...: 0.1-p ~\_7v*- f, I. ._\ ia‘ -.AL.’ 9 s I ”J." ”Iv-.3 k‘n‘wt .7 m...€..l..l.'s c ”J‘n’. 1 n ‘F" "'. ‘91-“) ‘ ‘ 0.x.-. “mini Ln; ‘1‘ .34‘1‘4‘ I m" x .uuukcu xii rs ”"wr-n - f‘ » “,3 ’ ~ 3"" ,. I ‘svoos... 3t:\\ . e .4: . 'b ”‘7'“. k‘_‘ ..J"!~ ._ .-..~....L “Nut". ‘ra ...g ”1‘- " v' . “\3 #4..) \t 2;... , \'.. '~ i I 'T“ ‘ “Cuff 3.1.1 'r~ “O I¢h«| \‘5 ‘ Ip - “In Lk ‘6)i\' \r". ‘ ¢Q\ Ts ‘I‘~ 99 r ‘lker' \-. «.“l‘.“ “1.» ‘t.-. “ma ‘1 \ r- it“. . P .r.‘~e “‘ ifl. A! “Oil". .. ~. n}. a- “I. ‘k I ' ”0.52%. "‘k.., . .‘ t'I .3 “u x In:- I ..a \'\. s‘ ‘3 "“ 16‘..-» 'L.‘1" ' ‘4. \‘ u -< ,- Thus, in order to control the thermal cracking of asphalt concrete pavement, the consistency and tensile strength properties of the asphalt binders have to be determined. It is recognized that one of the most significant factors to be considered in low temperature behavior of asphalt mixtures is that the consistency and stiffness characteristics of asphalt cement used in the mix (11). It can be stated that the transverse crack phenomena may be reduced or retarded by the use of softer asphalt (asphalt exhibiting lower temperature susceptibility and/or better flow properties at lower temperatures). This means that asphalt binder stiffness should be low at low temperatures. Another important factor to be considered is aging. Aging is the phenomena of hardening with time due to environmental factors. This phenomenon is important because the hardening effect due to aging increases the stiffness of the asphalt cement (11). Also, the tensile strength of the asphalt binder and the asphalt mixtures at low service temperature reflects the ability of the material to resist cracking, and therefore, has to be evaluated. The following section provides a review of the properties of asphalt cement that measures the consistency, the stiffness and tensile strength of the asphalt cement. 2.3.1.1 Penetration The penetration test measures the consistency of asphalt binder. It is defined as the thickness that a standard needle vertically penetrates a sample of the material under known conditions of loading, time and temperature. Heukelom (12) related asphalt stiffness to its penetration index (PI), which is a measurement of temperature susceptibility and is used to determine asphalt stiffness at low temperature. McLoad (I 3) pointed out that the improved values of penetration-viscosity number (PVN) indicates reduced chances of transverse 10 l ..q. I . *mc TCX'J: , '1‘ Ln .5; .. .. .1. NV, ”are. . e lo“ 1..., 13.1.2 Ductility iii: intuit} u: 1.... ??.’.:i..fli. ‘.\ 3LT. 5 r1 '3“! o l‘ .r .....m.:e. inc v I. .0”... F . ‘ Ha."' ‘\ _I.‘ t \ it“sub In-.. ' ‘ \~ I . J'- ‘O n . NA‘EA.»‘LI£ 51:." -‘ LL11“.- '1’7-.- ‘ . i f .. ’ ‘AMAEH‘L “h u flus‘l r; n‘ I . I f‘a‘hi ‘6: WM. if) Oliiilc a TC’W’V‘M 9 x‘HK'iL hfre are ‘ 4.14. lmss Hm I». MN} ’ t‘fi.’ . ub‘s‘“. 0" . u“. ‘ I ' AA \ n ‘ Li [DC 4 .. , I eI-“.. “Gigi :4 '. ‘ s 0i 1‘( ‘7‘ ". , u. i “ xt :\ I, e i r MC..- it L. I“ * -.‘:‘ u. cracking. Some researchers recommended the penetration test at temperatures below 70°F to control the low temperature performance (14). 2.3.1.2 Ductility The ductility of asphalt is determined by the distance to which it will elongate before breaking, when two ends of the specimen are pulled apart at a certain rate (speed) and temperature. The significance of the ductility test as a means of asphalt quality control has been debated because of its empirical nature and poor re-produceability (15). According to Hastead (16), the ability of asphalt to undergo elongation is not the primary Characteristic affecting durability, but, rather, that the ductility test results are an indication Of an internal phase relationship of the asphalt constituents. Sieggmann (I 7) pointed out that When ductility of the asphalt derived from different crudes are compared at equal Penetration, there are great differences, which can be correlated with P1. 23.1.3 Fraass Breaking Point The F raass test consists of flexing a thin steel plate coated with a thin film of asphalt ‘0 introduce a tensile stress in the asphalt film (18). The test gives the temperature at which a CTack forms in the asphalt film, which has been subjected to tensile stresses, while being cOOled at a rate of 1°C per minute. It was stated that binders with lower F raass points Would be expected to resist cold weather abrasion better than binders with high F raass Points because they are more brittle (I 9). 11 13.1.4 Limiting Him I I I “ ' ~.'.“ 9 V " ' ,1.“.‘.'st. u'lfi \Jitl’.’ t'i l {~l" . 1 . ;.v' ~ ~umb:.\ rugra‘h\a\ I'th. 'C‘ "II! \IP- v i - ' ... -.v. . . .1: .it 10.1.13 ‘0 . I g g . \ j ' I m. P}. 9' In“ uk.t.} Autuo'\ J; 31%" ' ' ‘ .. ntical Stress Tn: concert u? 23% ' . _ -.. iii other u t in} ..f.'.. “ ~ based on I?" IN. 4‘... _.\1‘tn.1$ ~ . ‘\s O . ' ~ Irh‘\ LT ‘40 .‘i x “r - as: JCTCC: [U ht r. .2 "rs KER ‘u'iii' 2.3.1.4 Limiting Binder Stiffness The idea is to estimate the temperature at which the asphalt reaches a critical value of "limiting stiffness" where it becomes susceptible to cracking. The Canadian researchers preposed the value of limiting stiffness as 145000 psi at loading time of 30 minutes (20). Strategic highway research program (SHRP) proposed a limiting stiffness value of 29,000 psi (200 MPa) at loading time of 60 second in the bending beam stiffness test. Thus, the critical temperature at which asphalt stiffness reaches the value of "limiting stiffness" is considered to be the predicted cracking temperature (21). 2.3.1.5 Critical Stress (Strain) Method The concept of this procedure is to control the stresses (strains) developed in the asphalt. In other words, the developed stress (strain) should not exceed a critical value. Hills (22) based on the estimation of the thermal stresses developed in asphalt and observations of mix cracking, established a value of the critical stress of 73 psi. This means that the temperature at which the developed tensile stress in the binder reaches 73 psi is considered to be the cracking temperature. SHRP binder specification established a critical strain value of 1 percent when the sample is pulled apart at a rate of 0.04 inch/min (1 rum/min). Thus, the temperature at which the fracture strain reaches this critical value iS the predicted cracking temperature (21). 12 m Hlflh Temp?” Hi ._1 \"Lscosil." \‘iicofi‘éll l5 ‘1 .. - a": ‘ . , ,..;-;uf,1‘ 1. ‘ I LI n’U‘UuhlfiU “L - - .. 1 I \ISCOSIll at H ct: temperature. 1 . , s. w...” 5"; oMOY‘ \ his. uib «nulk loadi.’ - Hoxicxer. th. .!‘. .h 3, t "t1” {we gym '“Arl— .mnl.» IIIK ‘ x "-1-. h. 31v ix ‘ ' luff...) piedmi ll" ' ‘3, 4.1 Temperature 2.3.2 High Temperature Properties 2.3.2.1 Viscosity Viscosity is a measure of internal friction or resistance to flow. High viscosities indicate stiffer asphalt (low penetration and high fluidity). Traditionally, viscosity test is conducted at 275°F to determine the suitable processing viscosity and temperature. In order to obtain a good processability, reasonably low viscosity is required. In addition, the ability of the binder to adhere to the aggregates is another key issue; in this respect, low enough viscosity is needed. Viscosity at 140°F was traditionally used as a tool to evaluate consistency at high service temperature. The ability of this test to predict how any given asphalt will perform under the traffic loading is questionable especially in the case of polymer modified aSphalts. However, this test could be an indication of the relative performance of different asphalt binders. King (23) stated that the absolute viscosity (at 140°F) alone is not able to accurately predict how any binder will behave. 2.3.2.2 Temperature Susceptibility Temperature susceptibility as defined by Mcload (24) is the change in consistency (Penetration or viscosity) of the asphalt cement for a given change in temperature. Mcload (24) recommended the use of the penetration viscosity number (PVN) to characterize temperature susceptibility of asphalt cement. PVN value of 0.0 indicates low temperature susceptibility, while value of -1.5 indicates high temperature susceptibility. l3 I ’~ ‘ 9 ' )f, l5» 2m. .ers in . . I I s. Pznemtzon :cr \K'u'g .l\iL, ‘ P3T262”211».‘\r. it * Pfirtmzz..n \- \\‘~ .ifire‘ D'I‘H‘ . ‘ “Ni” t. . ““1Lll..ln R H. Many parameters were used to characterize temperature susceptibility. These include (25, 26): l. Penetration temperature susceptibility (PTS): PTS=[L0g(P' 7P”) (2.1) T2 " T1 Where, P 1 = Penetration at temperature T], and P 2 = Penetration at temperature T2. 2. Penetration Index (PI): 1 = 20 — 500PTS (2.2) 1 + SOPTS 3 . Penetration Viscosity Number (PVN): PVN=1.5[L'L03(Y”5)] (2.3) L — M Where, L = 4.2580 - 0.79674 (Log Penetration at 25°C) M = 3.4628 - 0.61094 (Log Penetration at 25°C) 4~ Penetration Ratio (PR): 1’. PR=—x100 (2.4) 25 Where, P4 = Penetration at 4°C, 200g, 60 sec. P 2 5 = Penetration at 25°C, 100g, 60 sec. 14 ~ . \ seeming 1‘ Vx' *rViSCOSii} ‘31 " ~ 2 . . it“ rxxekazmls‘u k. 5'05‘ "‘-“"“_ in has stud} ‘ at I- "I h ' 2..*="’tcf“.llurt 5.1.». a..- b5..-: 1313 Viscoelastic l’ a... .. ‘. .. w {3.11;qu w 5‘ . ‘ I ‘ ‘wx. . ’jfi .‘«..’~ 1' _.. ...iu 8231:0221. la ' 3'” t -. "“-‘§\.\ In '8 ‘ 1m 3"? f‘d‘tnigut \ Rita'filt‘ 1260“ IS . According to Muncy et al (2 7), high temperature consistency can be evaluated by absolute viscosity at 140°F, or softening point while the low temperature consistency is better be evaluated by the glass transition temperature, F raass point, or low temperature ductility. In this study, PVN values of polymer modified asphalts showed that they are less temperature-susceptible than the control asphalts. 2.3.2.3 Viscoelastic Properties The balance between the viscous and the elastic nature of asphalt cement affects the way particular asphalt behaves in an asphalt mixture. This behavior under varying traffic and environmental conditions is important when considering the use of asphalt cement in pavement construction. Regarding asphalt cement properties, the linear Viscoelastic theory is applicable. To better understand the Viscoelastic nature of asphalt, Goodrich (28) suggested a mechanical model of shock absorber and spring as shown in Figure 2.2. At high temperature asphalt is easily deformed with small elastic component and large viscous flow component (i.e., the model becomes mostly shock absorber). When a load is applied to this binder at high temperature, a permanent deformation is developed, and this causes "rutting". In this report, Goodrich pointed out that the fundamental purpose of using polymers in asphalt is to gain better balance between the viscous and the elastic properties over a wide range of temperature and loading conditions. This means that the change in the asphalt properties with changing temperature (i.e., temperature susceptibility) should be minimized using polymers. Laboratory testing methods allow us to measure the spring (elastic) and the shock 15 Hot Viscoelastic Cold Brittle Figure 2.2 Viscoelastic model for asphalt behavior. 16 abs-fie: mscousv mi N y - . ‘ - N gilt}. in this lC>L .5 5.1:; response Is r - v v‘p-c‘ v- 11‘n" s ‘ ‘ ...LQL .L‘iboil 4IL hi all -.- i. e. ...cIO.’ Ol .ne TC>I>L u W's-HE . .Asx.-.ri:. a} 3.1:“.‘u I‘MQRV a. ' 3V_..Q‘ .. = ~ in it sfiuqutg 1.4..) ~.~ 5' (J '. - H ‘ e’“ : '1 -. _ KN”: hd h '} ‘ .. .1”- 1::\. ”‘7‘. A T 5c (h): [L’i 9 14; H :~~ . "“if"1 4 I. "»o‘ “ 50.7““ x 1"- ‘9‘.- "-59. ‘:.‘ 4H... 2‘43: 1‘ a .3, absorber (viscous) nature of asphalt cement by means of dynamic mechanical analysis (DMA). In this test, a sinusoidal strain (or stress) is applied to a specimen and the resulting response is monitored as a fianction of frequency. The primary response of interest in dynamic mechanical analysis is the complex modulus (G*), which is an indicator of the resistance of an asphalt binder to flow under a given loading conditions. The complex dynamic modulus is computed in strain controlled testing using the following equation: G * .. = :1 (2. 5) 7 w Where, G*w = The complex dynamic shear modulus at frequency w. rw = The absolute magnitude of the applied dynamic shear stress response. yw = The absolute magnitude of the applied dynamic shear strain. 2.3.2.4 Ring and Ball Softening Point Ring and ball-softening point is a meaningful indicator for the high temperature performance of the binder at high temperatures. European specifications have recommended softening point as a fundamental indicator of high temperature performance (23). 17 ll EFFECT OF I 14.1 Effect of Pol} r Lil] Effect on Pent: it Ms reports: 32:31:02 shelvu (>8 ‘ _s- .. £441.31 \ 3‘ ‘s- ' “.4151“ t V 'Q -. -..- :0... gn9~ 1... -~- ‘L¢\.\\Ile.\ Sk-‘I‘ 4,3 . ; sum. PI) and PCT :_:‘:“-‘: x.“ "« . s... ‘ (l . “k.” “:3§.I ,‘ “\AL {Affine 1...?" x; . "Hi. a. k“ 4I w \‘L"‘ :::: ‘\\‘l“\ I h . . c}; ‘i. ‘1‘, T‘\_ :‘\-_ ~353.\~, . ““1311 L . I. ' . .L_ “:36 a, PG. N .‘ N I. .. ‘ ‘\. ‘\.‘\ WIN h I 'H - "- 35-. ‘K: T‘h ‘4.‘ 2.4 EFFECT OF POLYMERS ON ASPHALT BINDER PROPERTIES 2.4.1 Effect of Polymers on Low Temperature Properties 2.4.1.1 Effect on Penetration It was reported by various researchers (25, 29, 30, 31) that the low temperature penetration (below 68°F) values of polymer modified asphalt are substantially higher than those of straight asphalt which indicates a softer binder and less thermal cracking potential. However, some references (29, 32) reported less (stiffer binder) or equal penetration values in the case of polymer modified asphalt as compared to straight one. Table 2.] presents some of these results. Haiping Zhou etal. (33) used the penetration index and (PI) and penetration viscosity number (PVN) approach and found out that control AC20 is more susceptible to temperature effects then the polymer modified AC20. They used three polymers for their study, a) styrelf, a polymerized binder with a thermoplastic styrene-butadiene block copolymer, b) AC20R, a polymerized binder with a thermosetting styrene-butadiene latex anion polymer, and c) CA(p)l, a polymerized binder with a thermoplastic ethylene-vinyl-acetate random copolymer. 2.4.1.2 Effect on Ductility Various studies (5, 33) reported that polymer modified asphalt binders are more ductile than straight asphalt binders (see, Table 2.2). Shuller (5) measured five parameters from the ductility test that is; maximum engineering stress, maximum true stress, area under stress-strain curve, true "asphalt modulus", and true "asphalt polymer modulus". All of these parameters were found to be higher in modified binders compared to straight binders. The area under the stress strain curve which represents fracture toughness (the 18 u .—\ Reference 3 Table 2.1 Effect of polymers on binder penetration. Penetration at 39.2”F (4°C) Reference AC Control Polymer Used AC Control AC Modified 3% SBS 25 16 AR1000 3% SEBS 27 19 3% EVA 35 3% Kraton 4O 3% Styrelf 30 17 120/150 4% Elvax 27 32 6% Kraton 44 6% Styrelf 35 5% SBS 27 18 80/100 5% EVA 24 29 5% LDPE 28 19 A020 Ground tire 15 21 Rubber 2.8% Neoprene 15 20 A010 2.8% SBS 18 17 2.8% SBR 15 l9 Table 2.2 Effect of polymers on binder ductility. Ductility at 39.2°F (4°C) Reference AC Control Polymer Used AC Control AC Modified 3% SBS 74 16 AR1000 3% SEBS 23 45 3% EVA 48 20 AC-S 3% Kraton 10 48 6% Kraton 63 23 40/50 5% SBR l 2.4 5% EVA 2.0 20 C” “ffCiCd' i0 C33? .‘ .1 .. $935; 9.,"" ' M79,av ‘25..” SHILJIC> L . is; W 1*? \‘if‘ 0,4 l'fi'". ,..'..-S‘ mk “IIASU \ Ill mum} and 6.353: L9,. \‘i's'rm 5i '» ..._i..... .r.‘ {3&1 L ..: L hi 1 ;‘-JP.)"‘\ ~- . » . _. '.o n—b¥ml\ UaiI\A\-e\£ .. 7341M...) - L... ...auxtcflblltN 'G"~1.zn 9' _ h n». ht" “CI-l CI‘n‘s .h} .0. .tl3 Effect on Stiff fr -» . .“L‘ \ ~1- m. .r . f\\3 bd5i‘d I‘ ‘;3“ "w .‘ll’ 3?; 4;. . ‘ “‘ew'.‘ d ”A“ -' .C ”5111M :‘§,.- “fill-E.) wa‘ : T ‘1 m {his 33; ‘~ Elfin-3 2 r. ‘53? H. ‘_ ‘ s V" . -‘ be p :-.~ . .';1 i‘.. :‘u.\}hl9 -y ‘3‘..." t in \\‘ . v I -“ ‘_>" 7.; s >\.l‘s ‘Fg' ! '-Iv-. “Lu-[C \ s "u I ‘ \_ h‘k‘v‘_ ‘1: r. 'c A! K‘ :" ‘ "all (g . ~ ‘I' .j.‘ .‘ ..n >kéxh l h.‘ .QiO ' ‘k‘t t" energy needed to cause fracture) of the binder, was found to increase by polymer addition which indicates better low temperature fracture resistance and tensile properties in the polymer modified binders. Haiping Zhou etal. (33) used ASTMDl ll testing procedure for ductility and elastic recovery tests and determined that at 39.2°F ductilities for polymer modified binders were higher than the conventional AC20. There was no noticeable difference for ductility tested at 77°F . Elastic recovery test results indicates similar characteristics; binder with polymer additives had considerably higher elastic recovery then conventional AC20. 2.4.1.3 Effect on Stiffness and Tensile Properties Quite few studies have been carried out to predict the low temperature fracture of asphalt mixes based on binder properties for polymer modified asphalt. An extensive review of the asphalt additive (4) included an evaluation of thermal cracking potential of asphalt binders with different types of modifiers. In this study, the cracking temperature was determined using the concept of limiting stiffness and critical stress. The limiting stiffness used in this study was 145,000 psi at a loading time of 30 min, and the critical stress (73psi) was used according to Hills method that was mentioned earlier. Results showed that some polymers (Microfil 8, Elvax, and Novophalt) did not cause a significant change in the cracking temperature. Dow Latex was found to increase the cracking temperature while crumb tire rubber was very effective in lowering cracking temperature. King etal. (34) determined the cracking temperature of SBS modified asphalt and straight asphalt based on the limiting stiffness method and the critical strain method. In 21 I 11+ r 5. - / Er’ (b (I: Hr-s p—A-q 2.2 V II 1 ...p‘wna'gm )P‘ "'.‘\ o In 5.5-mu..- sbt.t. nut—1: \ . O '9': ' 9 ‘h . I ) . IL §5A\.‘J.' 7L . A "~“‘ ‘MQ , ,J L‘.e .‘.Q: 51.1.1.” 31"“ :~.\. K- . ...n, -..J‘n *u.~-.AL..Ig >h.iu.x . C35 1 >7...“ . ' ~~I~~.L. abfihil a” ‘ .I . f min 11‘. . k. I‘TPW‘JRJI'.‘ a. ‘ -._...i. rhyming .7. ‘. a F‘c._‘\:‘ne‘§ \ I "‘ 31;. “31:. - this study the SHRP specifications were used (i. e the limiting stiffness is 29,000 psi at 0.35 min in the direct tension test, and the critical strain is 1 percent in the direct tension test). Results (see, Table 2.3) showed that SBS modified asphalt has substantially lower cracking temperatures than straight asphalt. It should be recognized that the cracking temperatures predicted from the critical stress (strain) and the limiting stiffness methods are not the same. Also, the values used as a limiting stiffness and as a critical strain (stress) were developed as a criteria for straight asphalt, and therefore, might not hold for polymer modified asphalt. Moreover, when the predicted cracking temperatures found using these methods were compared to the cracking temperatures obtained from the cooling tensile test for polymer modified asphalt, they were different (33). The conclusion of this study is that the cracking temperature obtained from the bending beam test is not the same temperature at which the pavement is expected to fail. Haiping Zhou etal (33) determined that binders modified with polymer have higher toughness values than the conventional AC20 asphalt. The polymer modified residue asphalt exhibits similar characteristics except styrelf. These test results imply that binders with polymer have a higher tensile strength than the binder without. 2.4.1.4 Effect on Fraass Point The available literature shows that Fraass points of polymer modified asphalt binders are lower than straight binders (25, 32, 35). Serfass (33) reported that Fraass point decreased linearly as SBS content increased (see Figure 2.3). Haipiing Zhou etal (21) modified AC20 asphalt with styrelf, AC20R, and CA(P)1 and found that polymer 22 Table 2.3 Effect of polymers on cracking temperature. Direct Tensile Bending Beam AC Grade Fraction of SHRP B 006 SHRP B 002 Polymer Temperature at 1% Temperature at 200 failure strain; °F MP3 Stiffness; °F 40/50 0 17.06 7.88 40/50 x l3.10 4.82 40/50 1.5x 1.76 40/50 2.0x 60/70 0 12.02 6.22 60/70 x 17.34 3.02 60/70 1.5x ! -2.02 0.50 60/70 2.0x -4.90 -1.48 80/ 100 0 8.42 2.84 80/ 100 x —l .48 -0.94 80/100 1.5x -8.32 -1.84 80/100 2.0x -7.96 -3.82 180/200 0 -0.22 -3 .46 180/200 x -5.98 -7.42 180/200 1.5x -7.42 -10.48 180/200 2.0x -l 6.60 -12.46 23 P 0 _| A. V.» .dL 3 u 3.. ban-nu... -. Temperature (’C) b) C N O h C 1, SBS Content Figure 2.3 Effect of polymers on F raass point. 24 ,. _ . 5." 3.13161 asphalt blue 22;: suggests that -. 2: :5 2'26 com enzxnr In conclusion} | 732135 and 10“ 1:1 #33? £315.13 id; EU I l ' "‘ ‘ "‘ 4' . V“! : 3.35:1. l“ rm. 1.; 5: 2.22:0. : l ‘. L '10“: ”-g- (4.. .5 o . ~ .L’aild be m 1": Effect Of Pair 142 “‘:nl\" e, ~ H"‘a‘t(1h H “rt “Q“. I£~ F . ”ed... \ 8“: ti» ,. w. .. . guttl‘h‘} s' modified asphalt binders had lower Fraass points than the conventional AC20 asphalt, which suggests that the polymer modified asphalt are more flexible at cold temperature than is the conventional AC20 binders. In conclusion, there is a clear trend that shows improvement in the tensile properties and low temperature fracture resistance due to polymer modification. The other important factor in the resistance of thermal cracking is the binder stiffness (which is expressed by many parameters such as penetration and bending beam test) was found to be improved (lowered) by polymer addition in most of the literature. However, some reports found the stiffness of polymer modified asphalt to be higher or comparable to the straight binder. 2.4.2 Effect of Polymer on High Temperature Properties 2.4.2.1 Effect on Viscosity All the available literature showed that polymer modification increased binder viscosity at all temperatures. Table 2.4 presents some viscosity values for polymer modified asphalt compared to control asphalt as obtained from the literature. This table shows that polymer modified asphalt have higher viscosities at high temperatures. From processing (mixing and compaction) stand point, in order to obtain reasonable spraying viscosities of the polymer modified asphalt excessive heating might be needed, which causes deterioration of both polymer and asphalt. Serfass etal. (36) stated that it is necessary to reduce the viscosities of polymer modified asphalt and this can be achieved either by cutting the modified asphalt with a fluxing agent or by emulsifying it. However, 25 . -L. swans: shomd ”its: Oflhc unmoc The higter \1: m "" .f :“ .h‘h ‘,| _ :'~:::. aflm u C nix“ I _, ”v.0 <9 "1~ \‘ r1 .Ir..:..0.tcdu . q - > '13“: ’- " ysLA. ;";1~ ‘ ' I Q, - 430. rsnrlncc x; V AP ' l ,_ :0 ‘e . w h _ 1. ~ ‘. “ha kit“ Neil \\ ‘ "O ‘ ‘11 ; 3|. ‘ pt 0 \h Ii. ‘>“- ~ . ~,.1{'f'r-._{W~‘A* ' 14. u a- . .‘y . \1.\.“ ‘1 J: ”1 5 "~- . dxc-p: ~. em, VL“‘ ." I ‘NC‘S. p . _ ‘3‘ m; 4.2:?”F‘; ' ‘ ‘ndex ‘ 1 . ~_ “(5 C1311 I J \V.. :~:V._Y . «Nga‘ 51:..4“ r ‘ N: v \-l\ '7‘ experience showed the polymer modified asphalt can be applied at temperatures similar to those of the unmodified asphalt (23). The higher viscosities gained by polymer modification compared to the control asphalt, indicate higher stability and stiffness at high temperatures. Thus, polymer modified asphalt are expected to be stiffer than the control asphlat at high temperatures as evident from the measurements of the absolute viscosity results (see, Table 2.4). This is expected to reduce rutting potential of asphalt. 2.4.2.2 Effect on Temperature Susceptibility Jain (25) studied penetration properties of unmodified and polymer modified asphalt, and reported that temperature susceptibility was significantly reduced upon polymer modification. Serfass etal (3 6) evaluated the high temperature consistency using the ring and ball softening ball and the low temperature consistency using F raass brittle point, and reported that the SBS addition increased stability at high temperature (i.e higher ring and ball softening ball), and reduced consistency at low temperature (i.e lower Fraass point). He concluded that the overall temperature susceptibility was reduced in the case of polymer modified asphalt. Andeson etal (32) pointed out that for the modified asphalt, temperature susceptibility was reduced for temperatures above 77°F, while at low temperatures, polymers have little effect on temperature susceptibility as it was indicated in the penetration index values (PI). Bouldin and collins (29) pointed out that temperature susceptibility can be evaluated either by the change in the viscosity or the dynamic mechanical analysis parameters (storage modulus; 0’, loss modulus; G”) with the change in 26 Cay—PP; —(_ Z—.-..I:..-,_. A-lB» an.» ..—:th 2 ..r.—.—J.Av. via.» ...ztw- 3 biblayvl?’ 2.1—:1:...J.....—X=Z ..U...>_.:_..C z...........u:.z=.v.: >-.t:.e_r...> Zhut.’ :3. n. N ..e~.~= F .yf .8588 a. 8 awe 02 a £885 .. 235 58 9.6 8% 830. $2 25< 2.5< o? m: 23 82 5555 $8 2592 5 mafia 2: 5mm .5 a. mzou own one 5mm .5 age: 4 m5 _. 555.5 .5 own . 2855 $0 $29 ”em 02 2 5 mm... 55.5 0% 5:02 25< mom a? "555.5 exam an 8: 205.5. .5 m SN ova <>m com 0:8 58 oz. 42 8: as 55 .5 825.. 2.5< 25¢. 2m 88 5mm .5 m E can .55 $3 at 8 ES 0% 92m 88 32 5mm $2 23. _ 25< 25¢. E... E: 55.55wa 5:95.. 55.50 5:55.. .5528 age a .5555 E§® +3. 0... +3. 3. :5: .5528 525.55 5552 £3 E 5:95.. 3. E55 .925 5.555 2.0855 E555 .2933 wommuofi 3838 .5 35883308 $585 vd 038. 27 2:, .me. The is»: mam suscepti?‘ In continuos. 2:213“ illipfl“ CIT.- ~ 2.5 21:: that pol) O _; immatures t ab 31372:: mien-ed as.” If; :3 expected to . \ gh‘fiéf . ‘ \:\( £73156”. ‘1"; 1.1;- G "or (i \ K." 33‘4“ LE3. .x ‘ ..h'5 I n. v a TA h ‘4'; .." - “Q RI“. \ «‘51 p “\‘\.“I?T‘ ‘9 ‘l . .4. .1' . temperature. The last study also reported that polymer modified asphalts are less temperature susceptible at temperatures above 68°F . In conclusion, even though some studies showed that polymers did not yield significant improvement in temperature susceptibility at temperatures below 68°F, all these studies agree that polymers provide substantial decrease in temperature susceptibility at high temperatures (above 68°F). However, it appears from most of the literature that polymer modified asphalt are less dependent on the changes in temperature and, thereby, they are expected to show enhanced performance over the entire range of service temperatures. 2.4.2.3 Effect on Viscoelastic properties Dynamic mechanical analysis (DMA) permits the finger printing of asphalt over a variety of temperatures and frequencies. Bouldin and Collins (16) used the DMA parameters (G’, G*) to evaluate temperature susceptibility of the binder using: B/ : L0g(Ml —M2) Tl ’Tz (2. 6) Where, M is G’ or G* They found that polymer modified asphalts are significantly less temperature susceptible at temperature above 68°F, which indicates that the change in G’ or 6* values with temperature has been minimized upon polymer modification. Anderson etal (32) conducted the DMA over a temperature range of -22°F to 140°F using the torsion bar and the parallel plate geometry, and reported that the dynamic modulus of modified binders was higher than that of the straight binder. The conclusion of this study 28 , o 125E 003me ”‘1‘ ‘~ l :13 3';le Willi]! TL“ ‘. r‘n'.‘. O . \ MM.- kimono . I y' t ‘ “‘\‘.’ st- Ln“ ’lfi.‘ v- 34.... ...iU\i§. bulk: . ' 1 I ‘H'd‘h .h s. 1.. .16 az‘tteurar‘ .‘v " ' \Lt'fii“ Cl. l} . 'z-q‘onr. . - ...c.;...'e> .0 331;. twisted l“ l“: I 2“; .1?“ 4? \.‘~.IA.:§‘ LO ‘EY h~'-"T- ‘k. § ‘1... ’1'; mg .. vi ‘X'nf‘ “L: 1d... I.‘.\ ' 54“”)! ' «z. "in L.;“ ’ I 8-3 s ‘ . 1 er- 4;. vs.» “gm sh.) , L‘ \ \\h ‘.I AL\ was that polymer modified asphalt have more stability at high temperature and thereby they have more rutting resistance. Bouldin et a1 (3 7) used the parallel plates geometry in the RMS8OO rheometrics to measure the DMA parameters for the asphalt modified and straight asphalt binders, and pointed out that the presence of a polymeric network is manifested through the appearance of plateau modulus at high temperature. Bouldin and Collins (29) found that polymer modification substantially increased both the storage modulus (G’) and the dynamic modulus (G*). Moreover, the desired effect of increasing better elastic behavior at elevated temperatures to achieve better balance between the elastic and the viscous behavior, was accomplished by polymer modification (38). In this study the loss tangent (tan 6 = G”/G’) was found to be higher in the case of polymer modified asphalt compared to the conventional asphalt, which indicates better balance between the elastic and the viscous behavior. King et a1 (23) found a good correlation between DMA parameters and the rut depth from the wheel- tracking test. 2.4.2.4 Effect on Softening Point Most of the available literature shows that polymer modification increases the softening point of the asphalt binders. Table 2.5 presents some of the experimental results as obtained from the literature. 29 I III I. lit b lll lll‘ll’ I'IIII. llt' .uaul—Cgupf J- ilqulm 1. I 111 1 ._ trI. . Ivr I: I t. .\.%H.INHNN.J~K4NII 22:2». v . wax 73—. A..—: v 2...;— u....:..::.l. 52.95:.— .22;—1...:.4:..I:25..:= 1...: 3.: :3 5...... ».~...~.~.. _.4.4~.~.\ VIN UT~2~ <2 $2 mmm so 3.: m: mmm 9% egg: as .3555 so 28$? :3 <2 <2 <2 x<>qm 9% om SE 25 SE5 :2. 20.25. can «a: 83.2 $2 <2 :2 2: 42,553.28 8°92 63 ”.82 <>m :2 8.3 a 3”: e2 mmmm cam 82%.. 6: 2Hm< 3: mmm 3m 3.: ~25wa 8.3 a 3: 2: mmm :3 o5< as EE< 3: 2 8582ng 62:02 hofibom+u< .2280 U< can: 2:20 3:20qu .89 a. V E...— wsfiecm 552.5 .8280 u< :3 wcfiofig =3 98 mat 05 no 58b2— mo Spam Wm 033. 30 15 ENGINEERI 15.1 Asphaltfilitt 15.1.] Stress-Strain l‘negeneral 51 retail: load is shun ramstreSs b} : Elastic strain tr‘mediatel} a J .. Viscoelastic s tecox'etable ui Plastic strain, 1 l l . 31"“! v ‘ v has be noted hen ifiltfuible for the tat; ”*3 . «em. A perfecth 2.5 ENGINEERING CHARACTERISTICS OF ASPHALT MIXTURES 2.5.1 Asphalt Mixtures Properties 2.5.1.1 Stress-Strain Characteristics The general stress-strain behavior of the conventional asphalt mixtures subjected to cyclic load is shown in Figure 2.4 (39). This figure shows that, the material responds to the applied stress by a total strain, which consists of three components. These are: 1. Elastic strain. It is stress and temperature dependent and is recovered immediately after releasing the load (stress). 2. Viscoelastic strain. It is time, stress and temperature dependent and is completely recoverable with time. 3. Plastic strain. It is permanent and stress & temperature dependent. It should be noted here that the tensile plastic strain or permanent deformation is responsible for the fatigue damage and consequently results in fatigue failure of the pavement. A perfectly elastic material will never fail in fatigue regardless of the number of load applications. Similarly, compressive plastic strain or deformation causes pavement rutting. 2.5.1.2 Resilient Modulus Resilient modulus is defined as the applied axial stress divided by the elastic portion of the strain. In flexible pavement design, the resilient modulus has long been used in lieu of the modulus of elasticity. In general, high modulus indicates high deformation resistance. Also, a high modulus asphaltic surface layer will protect the 31 Stress \N D A ~ . ilzbllC Strain Figure 3 , Unload Stress Strain Plastic Strain Elastic Strain Viscoelastic Strain Figure 2.4 Stress-strain behavior of conventional asphalt mixes. 32 em from being m . .. e. 2*: ...lnb: AS mentioned C. 527a: mzttures 15 Use.‘ Div." ‘ a )‘ mutt: it. not .111 um rain at L e strain is r Ibr— ._.. I .’I ‘.I «pelt/I Resilient film pinion of the x! $.13 Dtnamic Com Dynamic mod; at: describes the t l\ in pax'eniezx til-match. The tilt}. .t as is in the — ”sperm. while Ck ' “in mlildUiUS Cu" .1] Tun J subgrade from being over stressed and therefore it will reduce the possibility of subgrade failure. As mentioned earlier that in the theory of elasticity the resilient modulus of asphalt mixtures is used as the major characteristic of the strength of the material. Asphalt mixtures are not an elastic material, but rather a Viscoelastic material. However, large portion of the strain is recoverable, and thus it can be considered elastic for the analysis purpose (40). Resilient modulus is defined as the applied axial stress divided by the elastic portion of the strain. 2.5.1.3 Dynamic Complex Modulus Dynamic modulus of the material is one of the essential engineering properties, which describes the Viscoelastic nature of asphalt mixtures. This modulus was also used sometimes in pavement analysis as indicator of material strength and resistance to deformation. The difference between the complex dynamic modulus and the resilient modulus test is in the shape of loading wave. Resilient modulus test uses a wave that has a rest period, while complex modulus is tested in a sinusoidal or haversian loading. Complex modulus consists of two components: M*2 =\/M’2 +M”2 (2.7) Where, M * = The complex modulus, M = The storage modulus (elastic component) M” = The loss modulus, and Tan A = M”/M’ 33 Frame complex mo. ‘9‘.“ ‘v‘r~fl“ , ‘1) 343.2 lodulu‘: mUuk a- males. ll nun; :a-les re trimmings: extentttres hem e giant dztte enee i ;t ,t ' mums the l m. ...es 0: me Find: \i- i u; C" ' the hinders L at; sclid and MW An\.l “so. ‘ t «since-elastic n‘ 1C: “‘fifsa CU ' ‘Ka Qc' ‘ Dynamic complex modulus is measured by applying a cyclic load, in tension, compression, or shear loading mode. The value of the dynamic modulus is not the same for all of these loading modes. Huang (40) cited Kallas, and reported that the differences in the loading modes are insignificant at temperatures between 40 to 70°F and frequency of 1 to 16 HZ. At temperatures between 70 to 100°F and frequency of 1 HZ, however, there was a significant difference in complex modulus values. Similarly the DMA is conducted on asphalt binders to investigate the rheological properties of the binders. The dynamic shear rheometer (DSR) is used to measure the shear modulus (G*) and phase angle (8) (time lag between the applied shear stress and the shear strain) of the binders using the sinusoidal loading wave. If 6 = 0, the binder behaves as elastic solid and when 5 = 90" it acts as viscous fluid. When 0<5<90° the asphalt binder acts as visco-elastic material. From 6* and 8 the storage modulus (G’) and loss modulus (G”) can be calculated using the following equations: G/ =G*Cos(5) (2.8) G” = G * Sin(6) (2. 9) The values of rheological properties are highly dependent on temperature. At high temperatures, asphalt behaves as a completely viscous fluid. In this case, G’ would be small and (I would be 90 degree. At low temperature asphalt behaves like an elastic solid with a large G’, small G” and 6 equal to 0 degree. Under normal operating temperatures asphalt behaves as a combination of the two. This behavior is called Viscoelastic. Figure 2.5 shows same value of G*, Asphalt 2 is more elastic than Asphalt 1 because its 6 is smaller. Asphalt 34 Viscous Behavior I OH G1,? G*2 G2” 51 52 G1, G2, Elastic Behavior Figure 2.5 Viscoelastic behavior of asphalt. 35 It‘ll rec-over much it (A I “ -‘ : wither: dexnmt: 15.1.4 Fatigue and ' The fatigue lit .‘x‘i Figure 3.6 Shim as a C‘3‘Cile‘ load is fixed. In the past “3”} COmpJete.‘ ~ fatigue life i .‘ ' . V, a “ hills -~-~.u.'rtec in the It ‘ n fachi’k “ v- ‘0 in]' ‘l .‘t,.'[ ,‘ autJe lift Til“: In“ Ald11()n , . (Onsitiidmor Aéiéft‘gdte CL";- ’ ~ 1. ‘kpilail 1\ DC ' . Efl‘yirQnment “l . \. \‘P ”Datum d In .- .. general‘ (it: 3:91“ 2 will recover much more deformation than Asphalt 1. It is important to consider both G* and 5 when describing asphalt rheology (10). 2.5.1.4 Fatigue and Tensile characteristics The fatigue life of laboratory samples is a function of test type and the criteria used. Figure 2.6 shows various types of fatigue testing of asphalt mixtures. In all of these tests, a cyclic load is applied to the sample, and the number of cycles to failure is measured. In the past, many efforts have been made to estimate the fatigue life of laboratory compacted samples. Such estimates are highly dependent on the criterion used to define fatigue life of indirect tensile test samples. Various criteria were developed and are reported in the literature and are discussed in section 4.4. l. The following factors effect the fatigue life of flexible pavements: 1. Tire Inflation and tire-pavement contact pressure. 2. Consolidation and field compaction. 3. Aggregate angularity sand and mineral filler. 4. Asphalt type content. 5. Environmental Factors. 2.5.1.5 Permanent deformation and Plastic Characteristics In general, the plastic characteristics of any material is divided into two categories: 1. Permanent deformation is the cumulative plastic deformation under cyclic load. 36 Loading lat Type Configurst liinl Pam Prim (ester i PM! : Plum 4 x;\ l Cums“ i \l\ billing inning, l l Ital ‘. infill i <5 Failure in Loading Stress Loading Loading Performace State bending Test Type Configuration distribution Wave Form Frequency deformation of moment Allowed Stress or Tensile stress Third : : Heversine l-l.67 No a Yes Point Load Rest g Flexure i :73 Sine, Center 3 Same as Triangular, Point | Above Rectangular l : 100 No 3 No Flexure 1 Load Rest- E 1:100 5 Sine, 25 Triangular, Cantilever Rectangular l : 100 No 3 No Load Rest- g 1:100 5 T Rotating l6.67 No 3 Yes Cantilever C E D T Axial 8.33-25 No .73 Yes C S n Diametral l .0 Yes 3 No V' 5 Va“ V 5 Supported F‘Iexure Heversine 0.75 Yes 3 No (Beam) .g D Figure 2.6 A summary of fatigue test characteristics. 37 1 Creep ' aeonn Passes: let 23.56 ll Cil'...‘ ‘vr‘i- V : Nautn ("‘4 (L “ —. \W‘og' ' u‘. us; ¢ . ‘ YV _, .‘ ,l' ‘ kn»; \ 2. Creep is a measure of the total deformation (elastic, Viscoelastic and plastic) under a constant static load. Permanent deformation is a basic concern in the structural design of flexible pavement because it causes two different distress modes in the pavement, rutting and fatigue cracking. The following factors affect permanent deformation: 1. Lower percent air voids results in lower permanent deformation (4], 42). 2. Softer asphalt binder causes higher permanent deformation (4]). 3. Higher asphalt contents cause higher permanent deformation. 2.5.1.6 Thermal Cracking Potential of Asphalt Mixtures The previous methods tried to predict the fracture at low temperature based on the binder properties. The aggregate properties were accounted for in an empirical manner. Low temperature cracking is a result of many factors related to environment, and asphalt concrete properties. Material properties such as aggregates (type and gradation), asphalt (grade and content), mixture density (air voids), and thermal conductivity, are the determining factors in thermal cracking resistance of the pavement. Therefore the thermal cracking potential of the asphalt mixture should be evaluated. Many researchers attempted to characterize the low temperature fracture properties based on the mixture properties. The available literature suggests that the critical condition for fiacture in asphalt concrete occur at low temperature and/or rapid loading rates. This resulted from the fact that asphalt behaves in brittle manner at these conditions. In other words, the failure occurs as soon as the developed stress reaches the tensile strength of the material without any ductile flow. King (34) pointed out that the low temperature fracture occurs at low strain as a brittle 38 fire rather than du. I Y N.‘ ’\ ' ' 1533111 tUileTC‘li‘ Wall; 15.1 Effect of P01; 252.1 Effect of Pol‘ .ICCOtiiag n. *1}; .11” f t . “an M Ute" 35% b ‘l ‘\ V skewed mitt :esi'jezt modulus at ills result it e 105111110 (44.. ti» 37:13: Betti reported ”‘I‘s'ildpb \)_ :1 . 1 ' not" ...i‘pnrted that 'h" k \ 0“ ". 9 (“3:1 , ‘ ““11 at h1.~h ‘1 e c' t‘th ‘ “rd-FLI‘Q \ t“ 01 0.. I \_ . .‘ F ‘1 «‘3'» ‘5 le 10 Per .3"! knihy ' ‘ L 'g a. & K S M). P I lk'.‘ 3“». ”TI failure rather than ductile failure. The following is a study of the basic properties of the asphalt concrete related to low temperature cracking. 2.5.2 Effect of Polymer on Asphalt Mixture Properties 2.5.2.1 Effect of Polymers on Modulus According to many reports, polymer modification resulted in higher resilient modulus for the asphalt mixture as compared to straight asphalt. A group of researchers (43) showed that polyolefene/latex modified asphalt mixtures substantially increased resilient modulus at 104°F over straight mixtures (see Table 2.6). This result was supported by many other studies for different polymers. Mayama and Yoshino (44) used ethylene-ethyl-acrylate (EEA) and styrene-butadiene rubber with asphalt and reported a substantial improvement in the modulus at low and high temperatures. Another study (45) used polyolefene and Kraton D4406 modified asphalt and reported that the resilient modulus was improved upon polymer modification especially at high temperatures. Shuller etal (5 ) used the indirect tensile test to demonstrate the resilient modulus at high temperatures, and reported that the modulus of SB and SBS modified mixtures are significantly higher than straight mixes. King (38) conducted dynamic modulus testing over a range of temperatures, 14°F to 104°F, and frequencies, 0.1 Hz to 20 Hz. The moduli were 10 percent less at low temperatures and 17 percent more at high temperatures compared to straight mixes. This indicates better resistance to embrittlement at low and to mix tenderness at high temperatures. Little (45 ) carried out 39 latle 3.6 171‘: \ \liiturc T} pc- Table 2.6 Effect of polymers on resilient modulus and tensile strength (31). Tensile Percent Mixture Type Resilient Modulus (ksi) Strength Air Voids (psi) 10017 34°F 77°F 104°F 10°F AC5 3725 2070 242 22 476 4.3 AC5, 5%PI 4301 2319 278 57 534 3.8 AC5,5%PI, 3%L 4415 1960 288 61 581 3.3 AC10 3823 2497 266 43 430 3.9 AC10, 5%PI 4173 2447 298 64 482 4.4 AC10, 5%PI, 3%L 3593 2648 441 96 503 3.7 AC20 4765 2286 343 66 483 4.6 AC20, 5%PI 3709 4121 352 113 535 3.3 AC20, 5%PI, 3%L 4980 3079 353 119 559 2.9 40 icmdard (ASTM . 1res are able to reta‘ 1 mi: mixtures. K}: mares modified it ~ lid-ii \thich were .1 Zimmes. their 11. litttied AC10 shot-I An filmsh c - “rt“ 31 in“ 11‘; d Itgc f '5 .1: 1, I 42.313 3 hm“ then: 3:111: mffigth at A ll 113313110 50:11th 3%.“ f, it .\C. ‘ t - a .‘1; ”k 913311 . . - an}, Siren. .3: I I ' I‘ I A “ a £351.43 the standard (ASTM D4123) at high temperature (104°F), and found that Novophalt mixes are able to retain a resilient modulus that is substantially greater than the that of straight mixtures. Khosla (46) reported that at low temperatures, AC5 and AC1 0 mixtures modified with reacted SBS (Styrelf), and with carbon black (Microfil 8) showed moduli which were almost identical to those of straight AC5 and AC 1 0, whereas at high temperatures, their moduli were substantially higher than straight AC5 and AC10. Modified AC20 showed higher moduli at all temperatures. 2.5.2.2 Effect on Tensile and Fatigue Characteristics An extensive investigation of asphalt additives (4) included studying tensile properties at low and high temperatures, and reported that at low temperatures and high loading rates, polymers increased mixture tensile strengths over the straight mixtures which indicates better thermal cracking resistance. Another study (43 ) showed that the indirect tensile strength at —53.6°F was higher in the case of polyolefene/ latex modified asphalt as compared to straight AC5 and AC10 asphalt mixes. Shuler etal (4) conducted the indirect tensile test using AC5 with SB and SBS as modifiers, and reported that the tensile stress at failure (tensile strength) for AC5 with 6% SBS is significantly higher than straight AC5 at temperatures of -5, 77, and 105°F. Little (28) performed the indirect tensile test at temperatures of 32, 77, and 104°F at a strain rate of 2 in/min. the results of this test at 77°F and shows that Novophalt mixes have smaller deformation rates than straight AC20 mixes. Results of the indirect tensile test and resilient modulus presented above indicate that polymer modification improves the overall tensile properties and fracture resistance, 41 ‘1 ‘4' ,.I' 73701139 ..ltla A v :..._. 1; - 5 2.461116: 01 . >85 30: mer is l t Bljfficct 0 P01} 171:? 31.517.311.111} fro 1135 1135 1* '8 i '73.)“ v - is! “as Q ‘~ JO ““‘:”"a~“~ . LI 1 s, ”5.11.“. 91) improving fatigue cracking resistance and extending fatigue life. Jain (25) reported that fatigue lives of asphalt mixes were increased 9 times upon the addition of 2 percent of SBS polymer (see Table 2.7). 2.5.2.3Effect of Polymers on Permanent Deformation and Plastic Characteristics Polymer modification was reported to be very effective in reducing permanent deformation from binder testing. Goodrich (28) reported that the creep behavior of AC mixes was well correlated with the dynamic rheological properties of the binder. The creep test was conducted on polymer modified mixes (4). The load was applied for duration of 0.1, 1, 10, 100, and 1000 see. Total permanent deformation was measured at 2, 4, 8, and 12 min after unloading. The 1000 sec-load was used to measure creep compliance. Creep compliance characteristics showed that the addition of Novophalt to AC5 transforms the compliance characteristics of the blend to those which are statistically the same as straight AC20, and that the compliance of AC5 and SBR or EVA at 100°F are significantly higher than those of straight AC5. Further, the permanent deformation of modified mixes was lower than straight mixes. Khosla (46) conducted the incremental static creep test at temperatures of -20, 0, 20, 40,70, 90, and 120°F, under a 20 psi creep stress. It was found reveals that straight asphalt mixes exhibited higher deformation than carbon black (Microfil 8) modified mixtures. The dynamic tensile test is another way to obtain permanent deformation of mixes. Tayebali etal (4 7) studied axial dynamic loading of modified and straight mixes and reported that after a number of loading cycles at 104°F, the plastic strains of modified 42 ldhie 1329585 / .‘1‘59 [SBS / Table 2.7 Effect of polymers on fatigue lives of mixture (19). Sample No. of Cycles Sample No. of Cycles A80/100 4,500 A+2%LDPE 48,000 B60/70 1 1,500 A+5%LDPE 4,800 A+2%SBS 48,800 A+10%LDPE 750 A+5%SBS 91,000 A+2%PE+2%EVA 10,840 A+10%SBS 63,800 A+2%PE+5%EVA 14,750 A+2%EVA 4,900 B+2%PPW 8,800 A+5%EVA 8,900 B+1%HTPB 46,500 A+10%EVA 10,100 B+2%HTPB 16,500 43 rites “ere qurte >11 frherflmer .10 20.713116le the cur little '45' 51 ":1". 1“ . ests: uni:1\i.‘tl . Armenia} static e1 right AC20 mix 1 serial) at high 11‘ actuated pew .1 p n J- «mo .‘L.\ .-.is shou that .\ titration 1th st 13.5.3.8 111 ‘ .his stud The six ‘VC 1' mime rutrin: ...peratttre and he 31135 “t .1 “WMUCICQ V‘j.‘ .. ‘> ~~E\.=l‘~'mer m- ' v 11 > ‘ p. 1 1“ . My ‘41::r l ‘ ‘ Twill)" Pu . e 1‘01 .g_-. ‘=;:1\Vra“1'\ I “~th ‘. r,“ Aul \ \ .‘ Ff; ‘ « $.‘1Q13: PSI v “ diaI“ 4113'?“ Raw-J . r‘\1l|c‘ «\Q dtt‘“-. 51.1 mixes were quite smaller than straight mixes. The practical significance of this result is that the polymer modification provides mixtures with a superior rutting resistance as compared to the control mixes. Little (45) studied the deformation properties of different mixes. He conducted four tests: uniaxial compressive, creep compliance, uniaxial cyclic loading, and uniaxial incremental static creep. Findings of the uniaxial compressive creep testing showed that straight AC20 mix exhibited a very high compliance compared to Novophalt mixtures, especially at high loading times (10,000 see). From cyclic loading test results, the accumulated permanent strain was measured versus the number of loading applications. Results show that Novophalt mix is substantially more resistant to permanent deformation than straight mix which fails at 20,000 cycles. Air voids were an important variable in this study that affected both the modified and straight mixes. The above results along with binder testing results indicate clearly that polymers do improve rutting resistance of asphalt mixtures under extreme conditions of high temperature and heavy loading. To fitrther verify this improvement, the wheel-tracking test was conducted. Bouldins and Collins (21) reported a significant decrease in rut depth upon polymer modification. Table 2.8 presents the relative ranking of mixtures according to their rutting resistance. This ranking was consistent with the results of the dynamic mechanical analysis results. King et a1 (38) conducted the French wheel-tracking test. Results of rut depth versus number of loading cycles showed (see Table 2.8) that polymer modified asphalt was able to withstand four to ten times more loading cycles before ruts of specified depth. 44 Table 2.8 Wheel Tracking (16). Rut Rate No. of Cycles to a Asphalt Cement Polymer (W t.%) (mm/1000 cycles) rut depth of 10mm AR1000 SEBS (3%) 0.075 133,300 AR1000 SBS (3%) 0.11 90,900 AR1000 EVA (3%) 0.185 54,000 AR1000 Control (0%) 0.32 31,250 45 Lilliiicct of P011 The concept railing. Mcload 1: its x tines. khosl. Exes. lle tend. 4.:._-' .1'. '. ...3311 10mm: 11.? \ E‘n‘i, ~-., & 1331 at 414. as; 3. . tiling :11 5. CS, “-4. ‘21 “I It 2.5.2.4Effect of Polymers on Thermal Cracking Potential of Asphalt Mixture The concept of limiting stiffness was applied for AC mixtures. To eliminate thermal cracking, Mcload (48) has suggested a limiting stiffness value given in Table 2.9. Based on these values, Khosla (48) evaluated the cracking potential of polymer modified asphalt mixtures. He conducted the static creep test at temperatures of -20, 0, 20, and 40°F at different loading times up to 1000 see, and then used time temperature superposition to evaluate mixture stiffness at 20,000 see. Results are listed in Table 2.10. Comparing these values to the limiting stiffness values in Table 2.4. He concluded that, at -30°F, only mixtures made with AC5, AC5 with SBS, AC5 with Microfil, and AC10 with SBS, are able to mitigate low temperature cracking. At -20°F, in addition to the aforementioned mixtures, AC10 and AC10 with Microfil will also perform well. At 0°F however, only AC20 would not eliminate transverse cracking under these criteria. Saal (49) used the same concept and determined the limiting stiffness (750 ksi) at loading time of 10000 sec and a drop of temperature from 32°F tol4°F. The critical stress method was also used for the AC mix. The postulated mechanism for cracking is based on the concept of induced thermal stress, which exceeds the tensile strength of the mixture. Further, cracking temperature and cracking potential were evaluated using the indirect tensile test and the cooling tensile test. Shuller etal (5 ) conducted the indirect tensile test at -O.4°F, and reported that AC5 with 6% SBS displays significantly higher tensile strains at failure than straight AC5, which indicates that it is more ductile and softer at low temperatures than straight AC 5. The low temperature fracture toughness of 46 Tfmpcrature l Table 2.9 Limiting stiffness values for the asphalt mixtures. Minimum Pavement Stiffness at which transverse , Stiffness at which Temperature (°F) cracking are expected (2,0000 cracking can be seconds) eliminated -30 800,000 400,000 -20 600,000 300,000 -10 400,000 200,000 0 250,000 125,000 47 Table 2.? \ \lixtu re \ AC5 AC5~SBS ACS‘ .\llCR(.)T-‘ll AC10 .lfltl‘SBS AC ZILI‘XTICROH {2.0 ACCNSBS .ttilti‘xtlg‘R()I_~: \, Table 2.10 Stiffness values in (psi) for a (20,000 sec.) loading time. Temperature (°F) Mixture -30 -20 -10 0 AC5 350,000 270,000 180,000 34,000 AC5+SBS 240,000 100,000 72,000 16,000 AC5+ MICROFTL 290,000 210,000 78,000 31,000 AC 1 0 732,000 290,000 190,000 61,000 AC 1 0+SBS 410,000 1 40,000 93,000 34,000 AC 1 0+MICROF IL 670,000 240,000 100,000 57,000 AC20 1,600,000 870,000 650,000 170,000 AC20+SBS 910,000 510,000 3 30,000 80,000 AC20+MICROF IL 1 50,000 810,000 630,000 160,000 48 Xmophah mixe: :1 compare- It 33111'1‘1616 the .1. 531T per hour t1: 3m tempera ln sunxm AAL ‘C‘R'V‘Anr ,. .1...“ 1116 .117. «9.5 x ' . agreement In Novophalt mixes was found to be higher than straight AC20. Newcomb (43) reported that the indirect tensile strength at 10.4°F was higher in the case of polyolefene/latex modified mix compared to straight AC5 and AC10 mixes. King (34) conducted the cooling tensile test, where the asphalt mix sample is held at constant length while cooling the sample at 50°F per hour until fracture occurs. Polymer modification was found to lower the fracture temperature. In summary, improving the material properties yields substantial increase in pavement life and quality. Polymer modified asphalt is expected to yield the desired improvement in pavement performance. Finally, the summary of different types of polymers used as asphalt modifiers and literature reviewed is shown in Tables 2.11 through 2.13. 2.6 FATIGUE AND RUT MODELS 2.6.1 Fatigue Models Miner’s cumulative damage concept has been widely used to predict fatigue cracking. It is generally agreed that the allowable number of load repetitions is related to the tensile strain at the bottom of the asphalt layer. The amount of damage is expressed as a damage ratio, which is the ratio between predicted and allowable number of load repetitions. Damage occurs when the sum of damage ratios reaches a value of 1.0. Various mechanistic-empirical fatigue models, which have been reported in literature (40) such as Asphalt Institute and Shell models, relate the number of load 49 Types 1 \e’.‘.\ 011‘. .V m,‘ "- ‘ ' \ ‘ $13....Ut‘mllt5 \ .4 4;- R¥dk k114i \v pohmers Table 2.11 Prioritized list of asphalt modifiers and related issues. Modifier Representative Modifiers Issues Types and Main Results Network SBS/SEBS - improved rutting and solubility, content, glass thermoplastics thermal cracking resistance. transition temperature, SBR and SBR latex - improved rutting, processing temperature, fatigue cracking, and thermal cracking effect on aggregate- resistance. asphalt adhesion Epoxy - improved rutting, thermal reactivity as function of Reacting cracking resistance. temperature, reacting polymers Furfural, MAH - improved rutting, functional group, thermal cracking, better adhesion, but content, compatibility lower cohesion. with aggregate, processing temperature EA - improved thermal cracking phase separation, glass Dispersed resistance. transition temperature, thermoplastics HTPB, PPW, and PE - improved source, content, effect on thermal cracking and fatigue cracking aggregate-asphalt resistance. adhesion, processing EVA - improved viscosity in temperature laboratory tests, tendency of rutting, fatigue cracking, and stripping in field. Fibers Cellulose fiber - better strength and content, gradation, elasticity, less bleeding and better strength and modulus, stabilization. adhesion to asphalt, Mineral, glass, and polyester fiber - processing temperature better strength and elasticity. Particles Microfil 8 - improved rutting, fatigue type, content, gradation, cracking, and thermal cracking resistance. CRM - improved thermal cracking resistance, severe aging, lower tensile and shear strength, ravelling, reflective cracking, and stripping. adhesion to asphalt, processing temperature 50 ii 1 1111.1 1 1 1 I 1. l I 7.... {3:1 2. .1; 2.3:... 1:25.: 2.513;? .1; $ 7:.._:..:?... :.:.....:.:...Z. ..:.< ::...: <12 1 4:925:1122 3.— 9:12.... 7:2. 7...: <4) ._ .2... ’25:.3 52:: 1:... 22:22.4 . . «f; \» 2: $ 2.4.5.259 r3).— u:1...: 3.62.2.2: .62.... r._.¢.._,:_..; 7.42.122. ...i.._-£.:._- lu»..22m::t.. N ..z....~.¢..d.. \ :‘_\.P..w~.>< Z: .\ II 1 111 .ITlll... t.|.|T|lll|ll ill-till] 11 I ITil' 1 111 . ill 1 t 1.. . . 1 I. a. 3.1. a. - .d = 3. v .595 .1. -= =1- 3.‘ .0; = 3 =95 =.\.~..v. h . 1.--- i ll, - - I11- .1- I - L.v.~.\.-~=~< ..I’.a,..a.4- .J...-.:.:: 1.1: :v 91:2:::..f. ..s .1 .d~.s-..~ .0835 80505 20500 03500000 0880:0580: 0000— 05 :0 E5. 05 00 :0: 003 8080558 .:0>0>5I 58:00 :05 0080000580: 80:00:00 558: 0030— 00: 020500 005008 08.305- 808305 505 000:»:— 0803 0:80: 000: 500500008305 .8080 0:0 00080. 808808 60:3: 05 00: 800:5 00500880555 025:0 $00: 5:800 00:0: 550 .5005 65:0..— Swiot0009 80830: moms ® 00:80::0 0: £02: @ 000:8 02:008 505 00558 0.: 00:80:00.. 5003:. 0088808 0: mo: 00 moom-© 0005.50 8030— 0 0:55 2:? 000055 0:0 5200.0 .0855 0: 00:80:00: 0000208 £55.50 0:800 555- 0880:0580: >5— :0 0:0 mo~0© 00000.08 08:0,: :0 8080 0:80...- moan ® 00000.88 £50.00 0:809. @3080 88:05. 58:00 :05 80:00 08000: .205 :06,» .moom :05 000—- 0850:0588 Pa .0 00 50:05:08 :8: 930008 82:00: 56380.36 0:0 0058 02:008 0082—05- 083:: .0530: 2:880:04 €05.03 :803— 03:00:01 8:802 0w:0:0 00: 0:0 :5 >5: 000008000 8:00 gem :0 52037 Pm ._0 :0 00.0% .05: 5.5. 000380 .203: 0:0 30:3 8 0:. 0520.: 530—0: 0050.050 003 0:030:00 58853:- 05—50530: 8080558 0.:— 0zwz0m :00: 00: <>m 002:000 802:0 0:0 20.800 A<>mv 505 0205800 00>: 05:0: 00558 05: 800:3 005008 008305- b90503 0000000 38> 0:035...“ 0:38:0an 000080.:0m 00:08:0‘0h 00500—2 .3030: 0:30:05 05 :0 E0885 S .m 030,—. 51 1 _ .2: 7.1.051; _ rr 5. :...:.:;:.:I :31. .; __ :. i; .I. ,;.;.....: _ :: ;. ._ If; ...I .. 4 tau=mz=-U=:.V LQ-Q-ISL-uan .v.:-~.-L.C--.va~ Lanz—Iit .0..J~.:-:A-. v ...u~ .Mc .J~.~:.\ A883: NV 388 w83£ 80:3: 33 Gem _ -v 3:880:88 8032 3 33808 5803 0:88 800.35- .00: 3:: 30. 3:880:88 3 0830 o: 33 805 603808 b_80:0w 30:8 3538 9:880 EB mo “:02: 8 o 80.: w8w:8 3:880:88 8805: © 2:338 80:63:. 22833 882% com: Ea ”rain-8 mmm 833:5 33083 M80538 («0 35:3 0:: :53 33800: 3 moR® 33803? 55:30085 88:02 £080 -_>:_> 0:035: 3,33: £530 52 .__w a 8:3 H 223 .Uow : - 3 3::3 3538 808:0: X: 08 :8 :&:0:u 80:3: £83053? 08 .3 38388 :53 803:: 3538:: 3a 3538 .0830: 3030: 580 838m 88 5303 0:88 “08:08 8 30:80.5: 38m- .8033 3538 :85 8 33:83 3 8035 3538:: :8 >688 038 :38 330.58 028 033 «:0: 3H- .Uomn 8 mm 0mg 0:380:88 05 8:53 3:93 3538 .0830: 3.: 0:380:88 8:30:08 :53 >363 038 33803 8m8 2:338 83:»: 89: SM. 8:830 8083:: 330% :53 30:30:08 038 0:03 80050 0>m:<- .UooE© >538? 8283 3:8: 3: 333 3538 80520:- 3:30 m=80> 83:8 33 mm0::w:8 80:8: 3: 3:93 3538 80830:- .0588 800508 58 .om U< .3 _0>0_ 05 8 £2238 8053: 3a Far—a :0 38:05 3: .0330: .mohnmuv 33.33: 8032 3: 38:3 3538 80830:- 8:338 288—>0 Ammmv 3:85— 888330 3.0885: 0038.8t0m 35.32 .8288 N8 038 52 II I I . .w‘.‘--1.!Id.|- .~.-un.ual.l. .l. <5::.::2. » m; .m. .dQ: \. $880388“. 28808 Sta fimcobm 0:88 8868 .o 3:608 80:68 9588 8: Eu 80582- 68% 05 203 >05 Mot. 2 .888 82,558 .8 kiwi 29$ moo“ © fiwcobm 0:88 3265- .888 3:608 ..8 82m:— 382888 803 moi© 3:608 8233*- 828308 886830. 0:0 8:33 88 550 some wo $3 853» 203 warm“: 88.608 80:60.. ”—03 80¢ 3:58 mo ECOFE- 959 5:3 .=So>o 8089898 80:3: 3238 A3158 5:3 3m 83, _ Emma Rh © 88 m 0:38 gm 535$ move ® 8:832 20c ozv 3m .120 5:3: 3:608 82:3. 8 83.58 “onwE toga b88888 $5.58 25 02. 83808 .wEV—oflo 88.55 cams—m Accflobomv docmwamoa x080 9 8:883. 5826. 6088:8800 gonzo 8889 go 8:3 32:58 Emission 5088 :8 .moov E .6888 82: 38888883 Q85 8832 @8355 584 308.8 32:25 :8 £02: 3? .Uoom 88 Doc ® 8:888 338% 1838. m- 3 Egan—coma 23808 :0 “out“. 235- @3238 3 83888 bfiuocow 33 835 8 cosmscomom- 822% fism 8&8 CEfioSZXSREobog 333.230 hflwoaflhah ouflflahchhoh .5882 .3888 N3 2an 53 ..au__.:::.,.._uL .49: .r: 1:21.85. ...».,..:.:::..J. ”g .m. 972.; m0_o> 80 0. 080 800000 w808800800 28300 - 80.8800 05 00 .80 m: 3 000800. 0.03 .0000 88m - 808800 0. 00.00800 Qm 3 000800. 003 w8§080 00.038881- $88080 080588.. $8.8m mm ._0 .0 $8880an E282 x005 808.00 0008.88.05 w8800.08 08.80. 88%880 80888—080 0. 030 003 “m0. 80>O-.=._H 0.0.800 .00 805 0088 02.8008 .00 880 0088 02.8008 .0.“ .030. 083 0250 00.80008 08 w88008£ 0w<- .w8wm 0.0... .080 not. .0 .0583 .0838 mm: .0088 005002- .w8wm 0.0... .080 moo: @ 8008; 08800.80 .030. 008 .0088 005002- .w8wm 80¢ .080 monnmw 88.8.0800 .0888 088 .0088 005002 w8w< am .033: $88 .328 08808—0800 80.088.25 00:88.0t0m .5885: 008088850 20¢ .00 08800. («0 b0888m m _ .N 030% 54 . . ‘ 7., CIIOFIS 10 11 tom: "1 ..)"0 “ .ius. ERROR for 2H 7'13 9', . - ‘ “.3116 5th i§v¥kv "lull. \, ‘ N tflnw - MCI. ‘L 'T- $.40- “Ht. repetitions to the tensile strain and stiffness of the mixtures. These are of the following form: N, = f. (a, )"’ (5)-13 (2.10) Where, 8; = Tensile strain, E Stiffness, and f 1, f2 and f 3 are regression constants. For the standard mix used in design, the regression parameters of Asphalt Institute equation for 20% of area cracked are: f1 = 0.0796, f2 = -3.291 f3 = -0.854 and for the Shell equation these are: f1 = 0.0685, f2 = -5.671 f3 = -2.363 The regression parameters f2 and f 3 are determined from laboratory testing, while f 1 must shift from laboratory to field values by calibration. Pell and cooper (50) correlated the laboratory fatigue life with the mix factors of asphalt concrete mixtures and established the following regression equation: Log (NI) = 4.13 Long) + 6.95 L0g(TR,B)-11.13 (2.11) Where, Nf = Fatigue life. Vb = Volume of binder (percent of the total volume); and TR, B Ring and ball temperature. 55 Tie; con: i Uch . 1 33331153. bum [.Ug [\f We. V.» ‘1 tn “Wrs l They concluded that asphalt content is the most important variable affecting the fatigue life of asphalt mixtures. Baladi (51) conducted stress-controlled indirect tensile tests on Marshall samples to develop a fatigue equation using the statistical methods. The regression model is presented below: Log (Nfl = 36.631 - 0.1402 (TT) — 2.30 L0g(CL) — 0.5095 (A V) —0.001306 (KW + 0.06403 (ANG) (2.12) Where, Nf = Fatigue life. TT = Test Temperature (°F). CL = Cyclic load (pounds). A V = Percent air voids. K V = Kinematic viscosity. ANG = Aggregate angularity. Based on his filed study of flexible pavements in Michigan and Indiana, Baladi reported that the predicted fatigue life of asphalt samples were 20 times the fatigue life of the observed pavements. Mukhtar H. (52) studied the effect of asphalt mix properties and asphalt mix design factors on the fatigue cracking of asphalt pavements. The statistical model developed by Mukhtar H. is given below. Log(FL) = 2.1261 + 0.0068 (ANG)3 — 2.4266 (Log(AC)) ~— 0.0183 (SL) + 0. 7520 (SPGTY)2 — 1. 484 (Log(FINE)) (2.13) 56 there. Where, FL = ANG = AC = SL = SPGYT= Laboratory fatigue life (thousands of load cycles). Angularity of coarse aggregate particles in the AC mix (1=rounded, 5 = crushed on all sides). Asphalt content (percent by weight of the total AC mix). Pavement service life from construction or last rehabilitation to coring (years). Sample bulk specific gravity. Based on his study, Mukhtar concluded that the coarse aggregate angularity is the most influential factor affecting the fatigue life. Moreover, a fine content of less than 5 percent and a sand content of less than 20 percent provide a better fatigue cracking resistance. Majidzadeh etal. (53) utilized the concept of fracture mechanics, stress intensity factor and rate of crack propagation to predict the fatigue life of asphalt concrete pavements. According to their study, the fatigue life can be expressed as the number of load applications, which causes an initial flaw termed as “starter flaw”, co, to grow according to the crack propagation law; Paris law until it reaches a “critical size”, Cf, corresponding to the failure state. The Paris law is expressed by the following equation: dc ——=AK" 2.14 dN ( ) 57 Where. .4 and n are mm «5. . A ; mil 73331321.". 1... . - .. «is tx‘vstulatc vi- “-4!“ \\ 3‘ I 0% Where, d . . j = Rate of crack propagatlon w1th respect to number of load applicaton. K = Stress intensity factor. A and n are material constants and are determined experimentally. After rearranging and integrating the equation 2.14, the following equation is obtained. C f dc N = 2.15 f (“I AK" ( ) It was postulated that “starter flaw”, CO, is a material constant which is effected by the material characteristics, such as, void content and distribution, surface characteristics, aggregate angularities etc. However, c]: determines the critical crack size at which the material reaches the state of instability. The stress intensity factor can be determined either using the compliance approach or by notched beam specimen. In their study, the tests were conducted on both notched and un-notched beam specimens for sand asphalt and asphaltic concrete at temperatures of 23, 41 and 78.5°F . It was found that the average value of “starter flaw” co, is 0.05 for sand asphalt and 0.025 for asphaltic concrete. However, the values of n were 3.05, 3.35 and 4 at temperatures of 78.5, 41 and 23°F respectively. The value of n=4 is same as theoretical value reported by Paris for metals. This indicates that at low temperature of 23°F the asphalt concrete behaves as elastic solid. 58 lfil Rut \lud Rut prcdi and mechanistic- :inicis. them Barksdal meted of predi. 2:05 using pc: ‘Jgefiier with .11‘ sadism cleszi. 5.011515 not ' T0 prcd ;‘ I f ,. f3- 2.6.2 Rut Models Rut prediction models can be divided, in general, into two groups: mechanistic and mechanistic-empirical. The mechanistic models are based either on the theory of elasticity, theory of plasticity or the Viscoelastic theory. Barksdale (54) and Romain (55) first proposed the general concept of layer-strain method of predicting the rut depths. The layer-strain method consists of predicting the rut depths using permanent-deformation characteristics determined from laboratory tests together with an analysis procedure for the pavement structure using either linear or nonlinear elastic theory. The non-linear elastic theory should provide more accurate result (56), it has not been used very extensively because of its added complexity. To predict the amount of the permanent deformation due to load repetitions, each layer of the pavement is divided into several sublayers, and the stress state is calculated at the center of each sublayer directly beneath the load using the elastic theory. With the average stress state at the center of each sublayer, the corresponding axial plastic strain can be ascertained from the results of laboratory tests. The total rut depth for each layer is than calculated using the following equation. Ap = 2K6.” )(Ah. )1 (2.16) i=l Where, Ap = Total rut depth. 8,.” = The average plastic strain in the ith sublayer. Ah; = The thickness of I'm sublayer. n = The total number of sublayers. 59 Ihciayer-stra 0': rut depth. ' and; Kmfi spraiVES) A ‘ 1,) , firx‘nx- Ssn\ “i “Kb 0.1 I“ the \ @3330th 51H: 0“,»: A -‘ ‘ ILL. Cr 18 “K 0553.”- 'K‘Ar.lv The layer-strain method is considered a simplified engineering approach for predicting the rut depth, which permits the flexibility of using either linear or non-linear elastic analysis. Kenis (5 7, 58) has presented an overall framework for a Viscoelastic analysis system (VESYS) which consists of four major interactive models, termed primary response, general response, damage, and performance models. The primary response model is a probabilistic linear Viscoelastic solution for the mean and variance of the time dependent stress, strain, and deflection at prescribed positions in a layered Viscoelastic system. The rut depth predictive submodel is based upon the fundamental assumption that rut depth is proportional to the log of load repetitions. The permanent strain behavior of all materials in the pavement system is modeled by an equation of the form: a, (N) = £,11N"“ (2. 17) Where, a and )1 are material constants and a is the peak haversine load strain for a load pulse with 0.1 sec load duration at the 200th repetition. In the VESYS methodology, the relationship between total, resilient, and permanent strains is: 8p(N)=£—a,(N) (2.18) Where, 8r is the resilient strain. Using this assumption, expressions for the resilient compliance, Dr and resilient modulus, E, is as follows: D,(N)= D(1 —#N‘“) (2.19) 60 [ms formuliti (331,155. In 1th p b \ ‘ '0‘ I . .11 11.3. anous In the .-\ 1i r ' . my .Cpflllloil.‘ if)” f‘ . 1., 0. the subz'r viz“. . « ‘71 e“ i‘“ CU L€&h\ d' ‘8 ml); Farah . ..‘C the: . mflde a 8'” - v. 1021+ In this formulation, the two main material characterization parameters are or and p. variables. In this case: EzEN 2.20 , N, w 1 1 a > 0 signifies a stress hardening materials. a = 0 signifies a constant (elastic) material. a < 0 signifies a stress softening material. For bituminous materials, the a and )1 parameters are functions of temperature. In the Asphalt Institute and Shell design methods (40), the allowable number of load repetitions Nd to limit rutting is related to the vertical compressive strain ac on the top of the subgrade by the following equation. N, = f, (a, )‘f 2 (2.21) This equation is also used by other agencies with the values of f 1 and f2 shown in Table 2.14. As can be seen from the Table 2.14, the exponent f 2 falls within a narrow range, but the coefficient f 1 varies a great deal. Both these coefficients should be calibrated by comparing the predicted performance with field observations. Leahy and Witczak (59) assessed the influence of repeated triaxial test conditions and mix parameters on the permanent deformation characteristics of asphalt concrete and presented a statistical model of the form: Log (5],) = -15.83 + 7.132 Logm + 1.105 Log (S) — 0.118 Log (V) + 2,155 Log(EAC) + 1.11 7 Log (VOL) + 0.986 (TIEMP'O- 102 VMA —0-158) LOW”) (2. 22) 61 10'.in- 1111‘ H11. 771.11 LV Balad] (011/ Ct “sent depth in the A Liz-".1101 = ~10 * U [I if: ' Hi [1 1f: 1' ( ‘5 RD :1 l' M .H Where, Permanent deformation. Test temperature (°F). Deviator stress (lb/inz). Viscosity at 70°F (106 poise). effective asphalt content (percent by volume). Percent air voids. Test temperature (°F). Load repetitions. Baladi (60) correlated the AC mix and other pavement layer parameters to predict the rut depth in the AC layer. He presented an equation of the form. Log (RD) = -1.6 + 0.067 (A10 — 1.4 L0g(TAC) + 0.07 (AAT) — 0. 000434 (KV) + 0.15 L0g(EASL) —0.4 Log(MRRB) — 0.50 L0g(MRB) +0.1 L0g(SD) + 0. 01 Where, L0g(CS) —0. 7 L0g(TBEQ) +0. 09 Log (50-(TAC + TBEQ)) (2.23) RD = Rut depth (inch). A V = The percent air voids. T AC = Thickness of AC layer (inch). AAT = Average annual temperature (°F). K V = Kinematic viscosity at 275°F. 62 Baladi. conclude acids lead 10 hi; ESAL = The number of 18-kip ESALs at which the rut depth is being calculated. MR R B = Resilient modulus of road bed (psi). SD = Pavement surface deflection. TB EQ = Equivalent thickness of base material, which is the actual thickness of the base layer plus the equivalent thickness of the subbase layer reduced by the ratio of the modulus of the subbase to that of the base material. Baladi, conclude that the air voids is the most important factor affecting rut and higher voids lead to higher rut potential. 63 31 GENERA lhc p11 ma primers on the s reiterate and hi; clerical and rhet mature and mic armor cohesio Ellmnies of PM in pa ement p :3 (l . "lmi‘mer mod | lndireett! 0 Fracture! ' Theresili ' lhechar: rutandlo l0 accomplish t‘ EYE? 0f tempera] Risen“ CHAPTER 3 LABORATORY INVESTIGATION 3.1 GENERAL The primary objective of this study was to determine the effect of different polymers on the structural and engineering properties of asphalt mixtures at low, moderate and high pavement service temperatures. Moreover, by relating the physical, chemical and rheological properties of the binders, morphology of polymer network structure and microstructure analysis (voids, binder-aggregate interface, adhesion/cohesion) of polymer modified asphalt (PMA) mixtures with the engineering properties of PMA mixtures, the controlling fundamental properties responsible for long term pavement performance can be identified. The structural and engineering properties for polymer modified asphalt mixtures are as follows: 0 Indirect tensile and compressive strengths. 0 Fracture toughness. o The resilient and equivalent moduli. o The characteristics of the permanent deformation, which effect fatigue cracking, rut and low-temperature cracking potential. To accomplish the objectives of this study, several materials and tests were selected for a range of temperatures and experiment design matrices were established. These are presented in the following section. 64 3; MATERN The matenrii ill Aggregates The coarse .: 5mm Asphalt. an 1148'!) in tour dil‘l’e' trsferred to the .\1 mil quantities \\ l‘. Lie Michigan Depar iii fiphalt mix 008. 31.2 Asphalt C e Three \‘isew 81.0. These 35pm will t ' . ‘8 a COUlalnef 335.: . 323 Polymers Fm t.Vpes 0 3.2 MATERIALS USED The materials used throughout this study are presented below: 3.2.1 Aggregates The coarse and fine aggregates used in the asphalt mixtures were obtained from Spartan Asphalt, and were stored at the storage facility of Michigan State University (MSU) in four different piles according to the aggregates sizes. The aggregates were then transferred to the MSU Pavement Research Center of Excellence (PRCE) laboratories in small quantities where they were sieved, washed, dried, and stored. Fly ash, supplied by the Michigan Department of Transportation (MDOT), was used as the mineral filler for the asphalt mix design. 3.2.2 Asphalt Cement Three viscosity graded asphalt cements used in this study were AC5, AC 1 0 and AC20. These asphalt cements were supplied by Amoco in 5 gallon containers. Upon opening a container, the asphalt cement was stored in a freezer to prevent the aging, which could occur at room temperature. 3.2.3 Polymers Five types of polymers were selected for modifying the asphalt cements. These are: - Styrene- Butadiene- Styrene (SBS) 65 . Styrene- lit} . Sti'rene- BL' I . Ehaloy :\.\i O Crumb RUl‘" The SBS. SE tertiorl thermoplas lhe EAM acts as a inelastic network \ rot form a netttorl. Ly) try 1.. aggregate and .1 33 AGGREG The pieced. 1M 1‘ washout this sin 33] Aggregate Aggremtex 4‘. xiii-‘0'}? \. J 0 0‘ \t“‘ 65 “We 51. <0» 0 \v 1‘ \ tut film o Styrene- Etylene- Butylene- Styrene (SEBS) Kraton o Styrene- Butadiene- Rubber (SBR) latex o Elvaloy AM (EAM) o Crumb Rubber (CRM) The SBS, SEBS and SBR latex are classified according to their function as network thermoplastics, as they behave like resins and form a network inside the asphalt. The EAM acts as a reacting polymer, as it bound molecularly with the asphalt and forms an elastic network within the asphalt. However, CRM is a particular modifier and does not form a network of its own. The rubber particles disperse in the binder and adhere to the aggregate and asphalt surfaces. 3.3 AGGREGATE HANDLING, SIEVING AND WASHING The procedures adopted for the aggregate handling, sieving and washing throughout this study are presented below. 3.3.1 Aggregate Collection and Handling Aggregates were brought from a close by MSU facility located on Jolly Road. The aggregates were stored there in four different piles according to their sizes. These aggregates were brought in brown color 5-gallon containers. Standard sampling techniques were used in order to ensure that the material brought be the true representative of the piles. 66 332 Aggregate S llie aggrega 2010230“ F and th here sietecl. Tao 5r pantand - ‘~' 4 siete placed in the nest til retr‘ .ed on each sic l 10th prominent lih 33.3 Aggregate J The aggrcg. Samples. Specified so that the C}'llltd€li fret ttater from a rest rw OhOUrs ori Mm“? aQEremt; 055710 avoid an\ 3.3.2 Aggregate Sieving The aggregates brought from the facility were dried in an oven at temperature of 220 to 230°F and then allowed to cool down at room temperature. The dried materials were sieved. Two sets of sieves were used i.e., + # 4 sieve (3/4”, 1/2”, 3/8”, 5/ 16”, #4 and pan) and - # 4 sieve (#4, #8, #16, #30, #50, #100, #200 and pan). The aggregates were placed in the nest of sieves on the sieve shaker and shook for 8 minutes. The material retained on each sieve was carefully poured in separate blue color 5-gallon containers with prominent labeling of respective aggregate sizes on them. 3.3.3 Aggregate Washing The aggregates were washed thoroughly before they were used for the Marshall samples. Specified quantity of aggregate was poured into the aggregate washing machine so that the cylinder of the washing machine was filled one third of its total volume. The fi'esh water from a tap was turned on into the machine. Washing was carried out for at least two hours or until clean water can be seen came out of the machine. During washing, aggregate size and starting time was noted down and pasted on the machine in order to avoid any confusion about the aggregate size and washing time. After the washing was completed, material was oven dried at the temperature of 230° F. The aggregates were allowed to cool down at room temperature and then poured in red colored 5-gallon containers with prominent labeling of respective aggregate size. It should be noted here that different colored containers were used in order to avoid any confusion between unsieved & sieved and unwashed & washed aggregates. 67 3,; Pllt'SlCAl Various Phi his and} and 31 e l" 3.l.l Specific (i r The speciti. sizes it ere tested if gti‘ities “ere deter triples 1881)) we: intenal has tested 3.4 3.4 PHYSICAL PROPERTIES Various physical properties of the aggregates and asphalt binder were evaluated in this study and are presented below: 3.4.1 Specific Gravity The specific gravity of the aggregates was determined using the ASTM D 127 for the coarse aggregate and ASTM D 128 for the fine aggregate. The individual aggregate sizes were tested in triplicate samples. As a result, the bulk and the apparent specific gravities were determined. The bulk specific gravities of the saturated surface dried samples (SSD) were also determined in this test. The bulk specific gravity of the fly ash material was tested by MDOT using the standard AASHTO procedures. Table 3.1 shows the specific gravity of the individual aggregate, which represents the average of triplicate samples. The specific gravities of the coarse aggregates (-1/2” to + #4 sieve) and the fine aggregates (- #4 sieve to +#200 sieve) were averaged based on the proportions of each sieve size in the proposed aggregate gradation. 3.4.2 Absorption Capacity The absorption capacity of the aggregates is defined, as the ratio of the weight of water is required to saturate all the permeable voids in the aggregate while keeping their surface dry to their oven-dry weight. The absorption capacity of the aggregates was determined from the specific gravity test data (ASTM D 127 for the coarse aggregate and ASTM D 128 for the fine aggregate). The absorption capacity of individual aggregates is listed in Table 3.1. It can be seen that the absorption capacity varies from 1.78 percent for 68 TX.» >...:2_= » >..>=...» 0......u....uu< 1.32.... 3.9.1.1.”: . . ....... .3828...» >833...» i _- I i i i | . . . . . >232... » 338....» . :.:.:.Cv.£< 0:32;. 23:8 2.0.3»: . . . . 023.....0. attaiv. .055. . . :c..€..=v.:< 9:30.31. . . OMGLO>< .32.? .33.}. uvv. 0:2: .Sumog 2.09.0..— . 4.1.: 2.9.3:va 42:0 20.31.14.100: 05.-.: ..A....J:Q3.J 9.3.}... .25: .12.. >...>: .m U.....J.U.~..J. » .n- 072%». 306% 2.000% n Om .0 820.8 8.83% u 0% 083 m ........................ a. 80.. - .8... .5... N a: Z S: 2.. 8%.. N N... S. mm: a... 8;... N 42 we. $2 040.. on... 05.25% + 2 3.- :00 m l E. .0... m: 30.. on... 808008. 2.. $0.. 0 80.. Z w... .8. 0:... 8.. 2 m3 2 we: .00.. 00+ 8 3.0 :..N 2... 00.0 a. + 0.68% + 2 ..S- M: No.0 2 m: 000 .08? 2000.00,. 3.0. M: .3 :..~ .2 8.. .0... + 8.80 _ ll 00 4: E..- m: .8. fix; 500880 30:20 00.0w0.ww< 0080 000w0.ww< 3.00800 mfi’wwu 538.0 330.0 8038.030~ 0500.5 130,—. 8000 00 . . 0500.5 0500.5 08m 00.203. .8; 28.8 00 0.8... 2003.50.08 3.3.8.8. 880.....me 8.20.3 8:8 .0000m0.wm0 05 ..0 0.00000 8050.030 05. 330% 0Eo00m ..m 0300. 69 the {me aggrc gut me 12"5'126 510 me. I. 3.43 Durabi‘ If)”: 556d On ‘ x was: av the fine aggregates of plus size #200 to 0.60 percent for the coarse aggregate passing on the 1/2" size sieve. The average absorption capacity of the aggregates is found to be 1.10 percent. 3.4.3 Durability Durability of the coarse aggregate was evaluated using the Los Angeles Abrasion Test based on Michigan Test Procedure MTM 102-78. A blend of about 5000gm of the coarse aggregates was made. The blend was tested for 500 revolutions in the Los Angeles Abrasion machine. It was found that the percent loss by weight of the aggregates was 24.20 percent, which satisfied the MDOT specifications. 3.4.4 Viscosity The viscosity of the straight and the polymer modified asphalt (PMA) binders were measured using the Brookfield viscometer. The results for the AC5 and AC10 are presented in Figure 3.1. It can be seen that the SBS modified binders have higher viscosity than the straight binders. For the mixtures with straight asphalt binders, the mixing and the compaction temperatures were selected at 290°F based on MDOT recommendations. For the mixes with the SBS modified binders, the mixing and compaction temperatures were selected at 350°F. The viscosity of the modified binders at this temperature is within the range used for the straight asphalt and within the range of temperatures recommended by the industry. It was noticed that the SBS modified mixtures were easy to handle at this temperature level. 70 Viscosity (cps ) WW! i— lflll F “ill b. 10,000 _______________________________________ _I_____..I_.-_-- ::::::j:::::::::::::::::::t:::::: —o—AC-5 ::::::I:_:::::::::::::::C::::::+AC-10 . ---------- \---—--——--—-------++Ac5+5%SBs ~ L ““““““ {\\ “““““““““ i +AC-10+5% SBS' 1 ,000 Viscosity (cps) 100 10 200 220 240 260 280 300 320 340 360 Temperature (F) Figure 3.1 Viscosity—temperature chart for straight and polymer modified asphalt mixtures. 71 35 MIXING The mixi ghtscd upon 1h“ ' Q.‘ grade 01 trozen healer “'21:" than gzsbcated to 27 rtquzred propor L , ‘ Lind]. 512mlp16. g”? ‘; CU JibmrSCd ‘ k w . - 13: \thre 11 u 3350 for I“ 0 } median. At' Sill“ i ‘ ‘ed m ”"361; 3.5 MIXING PROCEDURES OF PMA BINDERS The mixing procedures for polymer modified asphalt binders were developed (based upon the improvement of rheological properties) by the Department of Chemical Engineering at MSU are presented below: 3.5.1 Mixing Procedure of SBS/SEBS Polymer A dry 400ml beaker was filled with approximately 300 grams of the required grade of frozen asphalt. The beaker was weighed before and after filling with asphalt. The beaker was then covered with a tin foil and heated for one hour in the oven that was preheated to 275°F . An oil bath was heated to an equilibrium temperature of 350° F. The required proportion of SBS/SEBS was also weighed in a weighing boat. After heating the asphalt sample, the SBS/SEBS was mixed in it with stirring rod until the polymer was well dispersed throughout the asphalt. The beaker was then immersed in the preheated oil bath where it was mixed using a “Fischer Scientific” 115 volt, low shear mixer at full speed for two hours. Upon completion, the sample should appear homogeneous to visual inspection. At the end of mixing, the sample was removed and used appropriately or stored in freezer at the temperature of -10° F. 3.5.2 Mixing Procedure of SBR latex Polymer It has been the industry standard to mix asphalt and SBR latex on-site using a spray flash technique. However, in laboratory, the following procedure was adopted. Approximately 300 grams of asphalt were heated in 400ml beaker at 275°F for one hour in a preheated oven (in order to obtain a good melt). The beaker of asphalt was 72 hen placed in an C m3 m1 incremenb percent water 1. A!“ sin: a stirring rm 81301131ng water inthe hot oil bath shear mixer tapprt tattered into the h tap ’he pol} mer DEM The sampl Beth. tight!) cm cr one hour. The scar“ 353 Mining Pr About ISL a the temperature ”fitted the com 331'? The :sphn ‘h‘ trad \' 3W6sired am I 1" Scre~~ d 310.“ h . ..w . m: if “Peril ““idlr‘fl' . I then placed in an oil bath that was preheated to 350°F. SBR latex was added to the asphalt in 2 ml increments (SBR latex solution consists of seventy percent polymer and thirty percent water). After each increment of polymer was added, the blend was mixed by using a stirring rod for one minute (being careful of the bubbling asphalt due to evaporating water). After all of the polymer had been added, the blend was allowed to sit in the hot oil bath undisturbed in order to allow any residual water to evaporate. A low shear mixer (approximately 1600 rpm; 2-inch diameter, 4-blade impeller) was then lowered into the hot blend. The speed of mixing was slowly increased (being careful to keep the polymer from climbing the stem of the mixer) to its maximum over a ten minute period. The sample was mixed for 30 minutes. The sample was removed from the oil bath, tightly covered with an aluminum foil, and allowed to cool at room temperature for one hour. The sample was then stored in a freezer. 3.5.3 Mixing Procedure of EAM Polymer About 150 grams of the original asphalt was heated in a 400ml beaker in an oven at the temperature of 275° F for 40 to 60 minutes. After a good melt of the asphalt was obtained, the container with asphalt was placed in a hot oil bath, which was preheated to 380°F. The asphalt was stirred slowly (~50%) for 10 minutes before adding the EAM. The desired amount of EAM was added slowly to the asphalt and the stirring speed was increased slowly to ~100 percent. The blend was mixed for additional two hours at the same temperature. It was necessary to check the blend periodically and the speed of the blending might be adjusted as the viscosity changed. At the end of mixing, the sample was removed and used appropriately or stored in freezer. 73 3.5.4 Mixing Approx hour no. obtain hath that llJS h. hensloul} ad- dimeter. J-hin 530M} increasc ninetes at 351; minimum tint]. hen stored in ; Finaii} ltissmd} is pr 3'6 MARS 3“ Aggrq “i 331nm} a.-.. 3.5.4 Mixing Procedure of Crumb Rubber Modifier (CRM) Approximately 300gm of asphalt are heated in a 400ml beaker at 275°F for one hour (to obtain a good melt) in a preheated oven. The beaker of asphalt is placed in an oil bath that has been preheated to 350°F. The required quantity of the CRM was weight and then slowly added into the beaker containing asphalt. A low shear mixer (2-inch diameter, 4-blade impeller) is then lowered into the hot blend. The speed of mixing is slowly increased to ~1200 rpm over a five minute period. The sample is mixed for 30 minutes at 350°F. The sample was removed from the oil bath, tightly covered with an aluminum foil, and allowed to cool at room temperature for one hour. The sample was then stored in a freezer. Finally a summary of mixing time and temperatures for all the polymers used in this study is provided in Table 3.2. 3.6 MARSHAL MIX DESIGN 3.6.1 Aggregate Gradation The aggregate gradation used throughout this study was selected based on a trial and error procedure. Seven aggregate gradations G1 through G7 were made and the resulting asphalt mixtures were tested using the Marshall mix design procedure. These gradations are shown in Tables 3.3 through 3.9 and Figures 3.2 through 3.4. The G7 gradation shown in Figure 3.4 and Table 3.8 satisfies the MDOT specifications of voids in mineral aggregate (VMA) and the MDOT criteria for 4C asphalt mix. 74 Table 3.2 1 \ Pob'mer T3] Table 3.2 Summary of mixing time and temperature for PMA binders. Mixing Mixing Time Polymer Type Temperature (°F) (hours) Remarks SBS 350 2 Network Thermoplastics SEBS 350 2 Network Thermoplastics SBR 350 1/2 Network Thermoplastics EAM 380 2 Reacting Polymer CRM 350 1/ 2 Particle Modifier 75 Table 3.3 Aggregate gradation and asphalt contents. a) G1 gradation data. Sieve Sieve Gradation weight Cumulative size opening G1 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0 0.0 (To 0.0 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 65.0 34.0 1836.0 1890.0 5/16 7.700 50.0 15.0 810.0 2700.0 No.4 4.750 35.0 15.0 810.0 3510.0 No. 8 2.370 22.0 13.0 702.0 4212.0 No. 16 1.180 15.0 7.0 378.0 4590.0 No. 30 0.600 10.0 5.0 270.0 4860.0 No. 50 0.300 8.0 2.0 108.0 4968.0 No. 100 0.150 6.0 2.0 108.0 5076.0 No. 200 0.075 5.0 1.0 54.0 5130.0 Pan Fly ash 0 5 270.0 5400.0 b) Asphalt contents. AC Content AC Weight Required Total weight of the mix (%) for 5400 gm batch 4.00 225.00 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 76 Table 3.4 Aggregate gradation and asphalt contents. a) G2 gradation data. Sieve Sieve Gradation weight Cumulative size opening G1 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0 0.0 0.0 0.0 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 81.0 18.0 972.0 1026.0 5/16 7.700 67.0 14.0 756.0 1782.0 No. 4 4.750 43.0 24.0 1296.0 3078.0 No. 8 2.370 31.0 12.0 648.0 3726.0 No. 16 1.180 22.0 9.0 486.0 4212.0 No. 30 0.600 15.0 7.0 378.0 4590.0 No. 50 0.300 10.0 5.0 270.0 4860.0 No. 100 0.150 6.0 4.0 216.0 5076.0 No. 200 0.075 4.0 2.0 108.0 5184.0 Pan Fly ash 0 4 216.0 5400.0 b) Asphalt contents. AC Content AC Weight Required Total weight of the mix (%) for 5400 gm batch 4.00 225.1% 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 77 Table 3.5 Aggregate gradation and asphalt contents. a) G3 gradation data. Sieve Sieve Gradation weight Cumulative size opening Gl retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 1000 0.0 0.0 0.0 1/2 inch 12.500 100.0 0.0 0.0 0.0 3/8 inch 9.500 88.2 11.8 637.2 637.2 5/16 7.700 70.0 18.2 982.8 1620.0 No. 4 4.750 51.9 18.1 977.4 2597.4 No. 8 2.370 33.6 18.3 988.2 3585.6 No. 16 1.180 24.1 9.5 513.0 4098.6 No. 30 0.600 16.0 8.1 437.4 4536.0 No. 50 0.300 10.5 5.5 297.0 4833.0 No. 100 0.150 6.7 3.8 205.2 5038.2 No. 200 0.075 5.2 1.5 81.0 5119.2 Pan Fly ash 0 5.2 280.8 5400.0 b) Asphalt contents. AC Content AC Weight Required Total weight of the mix (%) for 5400 gm batch 4.00 225.00 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 78 Table 3.6 Aggregate gradation and asphalt contents. a) G4 gradation data. Sieve Sieve Gradation weight Cumulative size opening G4 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0 ' 0.0 (To 0.0 , 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 73.0 26.0 1404.0 1458.0 5/16 7.700 56.0 17.0 918.0 2376.0 No. 4 4.750 35.0 21.0 1134.0 3510.0 No. 8 2.370 24.0 11.0 594.0 4104.0 No. 16 1.180 17.0 7.0 378.0 4482.0 No. 30 0.600 12.0 5.0 270.0 4752.0 No. 50 0.300 9.0 3.0 162.0 4914.0 No. 100 0.150 7.0 2.0 108.0 5022.0 No. 200 0.075 5.0 2.0 108.0 5130.0 Pan Fly ash 0 5 270.0 5400.0 b) Asphalt contents. AC Content AC weight required Total weight of the mix (%) for 5400 gr batch 4.00 225% 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 79 Table 3.7 Aggregate gradation and asphalt contents. a) G5 gradation data. Sieve Sieve Gradation weight Cumulative size opening G5 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0 0.0 070 0.0 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 77.0 22.0 1188.0 1242.0 5/16 7.700 59.0 18.0 972.0 2214.0 No. 4 4.750 35.0 24.0 1296.0 3510.0 No. 8 2.370 24.0 11.0 594.0 4104.0 No. 16 1.180 17.0 7.0 378.0 4482.0 No. 30 0.600 12.0 5.0 270.0 4752.0 No. 50 0.300 9.0 3.0 162.0 4914.0 No. 100 0.150 7.0 2.0 108.0 5022.0 No. 200 0.075 5.0 2.0 108.0 5130.0 Pan Fly ash 0 5 270.0 5400.0 b) Asphalt contents. AC Content AC Weight required Total weight of the mix (%) for 5400 gr batch 4.00 225.00 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 80 Table 3.8 Aggregate gradation and asphalt contents. a) G6 gradation data. Sieve Sieve Gradation weight Cumulative size opening G6 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0r 0.0 0.?) 0.0 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 81.0 18.0 972.0 1026.0 5/16 7.700 63.0 18.0 972.0 1998.0 No. 4 4.750 35.0 28.0 1512.0 3510.0 No. 8 2.370 24.0 11.0 594.0 4104.0 No. 16 1.180 17.0 7.0 378.0 4482.0 No. 30 0.600 12.0 5.0 270.0 4752.0 No. 50 0.300 9.0 3.0 162.0 4914.0 No. 100 0.150 7.0 2.0 108.0 5022.0 No. 200 0.075 5.0 2.0 108.0 5130.0 Pan Fly ash 0 5 270.0 5400.0 b) Asphalt contents. AC Content AC Weight for Total weight of the mix (%) for 5400 gr batch 4.00 225.00 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 81 Table 3.9 Aggregate gradation and asphalt contents. a) G7 gradation data. Sieve Sieve Gradation weight Cumulative size opening G7 retained weight (mm) Passing Retained (gr) retained 3/4 inch 19.000 100.0 0.0 0.0 0.0 1/2 inch 12.500 99.0 1.0 54.0 54.0 3/8 inch 9.500 81.0 18.0 972.0 1026.0 5/16 7.700 63.0 18.0 972.0 1998.0 No. 4 4.750 35.0 28.0 1512.0 3510.0 No. 8 2.370 22.0 13.0 702.0 4212.0 No. 16 1.180 16.0 6.0 324.0 4536.0 No. 30 0.600 11.0 5.0 270.0 4806.0 No. 50 0.300 9.0 2.0 108.0 4914.0 No. 100 0.150 7.0 2.0 108.0 5022.0 No. 200 0.075 5.0 2.0 108.0 5130.0 Pan Fly ash O 5 270.0 5400.0 b) Asphalt contents. AC Content AC weight required Total weight of the (%) for 5400 gr batch mix 4.00 225.00 5625.00 4.50 254.45 5654.45 5.00 284.21 5684.21 5.50 314.29 5714.29 6.00 344.68 5744.68 6.50 375.40 5775.40 7.00 406.45 5806.45 82 co;V we... 0030a ”AB—5 9.30:0 0.55 00000090000 808.000 000 8080.08 .009). 5.3 $8.0 $000 0...: 00.. GO 0.. .00 00300 000000& 000w00mw< NM 00%.... 0m.m oo.m omd cod om. co; and 00.0 _ 0.0 - 0.0. \ .x I I x I I I I I x I I I I I + \ . . 0 0m . .d t 0.0. n 0 \\ a - . w. . \ 0 0.. J 3 . . m. .. ..X o om 0w . .w. 00... 3.000.. 0 00 M 8380.0 0.. n m. . 2 . 0.0.. cm. / 300000.025 Hoe). a x. u 1 . \ m0 00.00005 IT 0.00 0 \ ‘4 N0 GOUQUMGIOII . .\ . \ 0.0. . \..x .0 80.005191 . i . L 0.00. 83 00+ .80..8¢.oo% 808.88 my... .3025 x88. 3.506 o>o.m .80 808.088 HOD: 8.3 mac? $.80 x8 8.. GO 8 38 838 00.83% 80383 m.m 20m... ohm oo.m OWN CON on. co. and cod I o x I I i I Ix .1 u \ \[I\ o. -- \\\A ON . \ x k \ \ a on n x 4 . F . 1 3 Q \ I \ .. \\\ ‘x 0m \ - \ \ 2.: €80 . \ \ \ u 8 828.082 .. x 0.. / . \ \ mcosaofiooqm PODS. n x. a \ t . x \ ‘ cocofiuaolfll 8 \ 80_0 a u. \\ m0 .005+l-00 N . _. .x 5 8835+ ‘ | _ . 00. aqfigam Kq Bugssud wanna 84 0... WM 800% 06 8 080 3.80 99 00.8080 Sumo; Wm 0.80.... 0.m 3.0 .8300 38.5 00.0 96% Wm 0N m.— III; j' ' ' ' '- ' I ' ' ' 4 0.0. 0.0N 0.0m 00.8080 20983. 0.0V 1P 0.0m r 0.00 / _ 5.0009 808% 0.0.. 0.0m 0.00 0.00. mfigam Kq Sugssnd manna 85 3.6.2 phsse says: 0h; d. n; . \. l): I“. \ 3.6.2 Sample Preparation of Mixtures The process of creating asphalt Marshall sample was comprised of two separate phases; the first phase was the combination of the aggregate and asphalt which was separated into three individual samples. The second phase was the compaction of the three separate samples by a test hammer. In the beginning of phase one, the aggregates were added together according to the design specifications. These specifications designate how many grams of material were needed from each sieve size. Once combined the aggregates were placed into the oven with the mix tools and frozen asphalt/polymer beaker then heated to the desired temperature (asphalt is heated to ranges of 280, 350 and 380 degrees, depending on the type of polymer being used). When the asphalt reached the appropriate temperature, the mixing pot was placed on the scale and the scale was zeroed. Then the aggregate was poured in and weighed to ensure the proper amount had been used, the scale was then zeroed again. Then the required amount of asphalt was poured into the container. The mixing pot was then placed in the mixer and stirred for 90 seconds and next transferred to the heated plate where the mix was deposited and ready to be quartered. With the use of a trowel, the mix was corralled into a circular pile and separated into four sections, one of the sections was then quartered again and distributed among the three remaining piles. Each pile was then placed into a container in required amounts. Three containers were created out of each mix, these containers were then placed into the oven again with the necessary molds and casts until the mix reached the desired temperature. The hot plate on which the test hammer rests was also turned on and pre- 86 brazed at remaining l Mfr? T3. . placed on 3.63 heated at this time. After this step, the tools and mixing pot should be cleaned of the remaining asphalt before using them again. When the samples reached the desired temperature the cast and Marshall mold were removed from the oven along with the first sample. A piece of wax paper was then placed on the base of the mold and the material was poured in. A heated spatula was then used to spade around the outside of the material/mold ring 15 times, then the inside was spaded an additional 10 times. The cast was placed on the base of the automated compactor and covered with a piece of wax paper before the hammer was placed on top. The sample was pounded 50 times, then the mold was reversed and pounded 50 times on this side. Once this was completed, the mold was then removed and labeled with the date, asphalt type and polymer type and allowed to cool for eight hours before the sample could be extracted from the mold. This process was continued consecutively for the remaining samples. 3.6.3 Results of Mix Design The mix design of the straight and the modified asphalt mixtures were conducted using the standard Marshall Mix design procedure. The aggregate gradation G7 was used for this mix design. All Marshall samples were compacted and tested in the following order: 1. Bulk specific gravity (ASTM D 2726) 2. Stability and flow test (ASTM D 1559) 3. Maximum theoretical specific gravity (ASTM D 2041). The optimum asphalt content was determined based on MDOT specifications. These include: 87 Rice test mm . flibzilz‘. 01mm :7 Cf) ._. . F1 W‘H-o ‘ l : ..‘4 , i ‘likd l ‘b‘stx 1. A target air voids of 3.5 percent based on the traffic volume. 2. A minimum of 16 percent VMA. 3. A minimum Marshall stability of 1200 pounds and Marshall flow in the range of 8 to 16. 4. An asphalt content between 5 and 8 percent by weight of the total mix. Tables 3.10 and 3.11 summarize the results of the Marshall mix design and the Rice tests for specific gravity, respectively, conducted on the triplicates made by using straight AC5 and aggregate gradation G7. Figure 3.5 depict, the specific gravity, the stability of the mix, flow, the percent VMA, the percent air voids and the percent voids filled with asphalt contents. Based upon the results and MDOT specifications, the optimum asphalt content of AC5 was determined at 5.7 percent by weight of the total mix. Similar results were obtained for AC10 and AC20 asphalt mixtures and the results plotted in Figures 3.6 and 3.7. Marshall mix design for polymer modified asphalt mixtures were also conducted. Figures 3.8 and 3.9 display the curves for the specific gravity, the stability of the mix, the flow, the percent VMA, the percent air voids and the percent voids filled with asphalt contents for AC5 modified with 5% SBS, and AC10 modified with 2% SBS. Based upon the results and MDOT specifications, the optimum asphalt content of AC5 and AC10 PMA mixtures was determined at 5.6 and 5.7 percent, respectively, by weight of the total mix. Finally a summary of optimum asphalt content for straight and polymer modified mixtures is shown in Table 3.12. 88 .hmv 3.1—.3211 3.31.9330: ..2: 7:: 21127:). V. v(\ .3.‘ 2:329. :vw.2.J~v .r:: :-:;.I.(\ C~ ». .JT~=.~. 0.0 002 22 3.0 002 00.2 02.2 00.2 2.2 00.00 00.00 20.0 3.0.0 03.0 .30 0.03 0.02 000.0 2.0.0 00.0 0.020 0.002 0.0:. 0.002 0 0.0 002 00.0 002 2.02 00.2 2.00 000.0 03.0 .30 0.02 000.0 00.0 0.000 0.002 0.0:. 3.02 N 0.0 002 0.0 002 00.2 00.2 00.00 .000 03.0 :60 0.0: 000.0 00.0 :00 0.32 0.000 0,002 2 0.0 0.0 002 00: 0.0 00: 00.: 2.2 00.: :2 0020 00.00 000.0 20.0 000.0 000.0 0.0: 0.03 0000 000.0 00.0 0.20 0.002 0.00... 0.002 0 0.0 22 00.0 2.2 2.2 00.2 8.00 000.0 000.0 000.0 0.0: .0000 00.0 0.000 0.002 0.5. 0.002 N 0.0 002 00.0 00: 00.2 00.2 020 000.0 000.0 02.0 0.0: 0000 00.0 .02 0.22 3.00 0.202 0 0.0 0.0 .02 002 S. 00: 00.2 00.: 00.2 2.3 :2. 3.20 20... 000.0 80.0 02.0 0.02 0.02 000.0 000.0 00.0. 2.30 .002 0.000 0.002 0 0.0 2: 00.0 00: 00.2 2.2 00.00 02... 200.0 02.0 0.0: 000.0 00.0 0.22 0.3.2 022. 2.2 0 0.0 002 0.0 002 2.: 2.2 00.: 000... 200.0 05.0. 0.02 2.0.0 00.0 0.000 0.002 0.02. 0.3.2 _ 0.0 00 002 .02 3.0 002 00.2 00.: 00.2 00.: 02.00 00.00 000... 000... 000.0 000.0 00: 0.0: 000.0 000.0 00.0 0.20 0.002 0.0:. .002 0 0.0 00: 00.0 002 2.2 00.: 0.0.00 000.0 000.0 000.0. 0.0: 000.0 20.0 0.20 0.002 0.02. 0002 0 0.0 22 02. 002 00.2 00.2 0000 0.00 000.0 000.0 3.: 000.0 000 0000 0.002 0.20 0.002 2 0.0 0,0 80. 002 2.0 002 00.2 00.2 00.2 00.2 2.20 00.00 000.0 00: 000.0 000.0 0.02 0.02 000.0 000.0 20.0 0.000 0.002 0.000 0.002 0 0... 800 2.0 0000 00.2 00.3 8.00 000.0 000.0 000.0 3.: 000.0 00.0 0.200 0.002 0.000 0.002 0 0... 002 03. 202 00.2 20.3 00.00 000.0 02.0 000.0 0.0: 200.0 00.0 0.000 0.202 0.02. 0002 _ 0.0 .5... A .35. .0350 80:0 80 .580 3‘ 0082 02.8.: 35 300.3802 300 03.8.6 380 a? 0.0.... 0.0 54.0%.:an 3.3 5. .0z 0 only: 8.02.89< ttlc_ to... .0035. <20 >< 50> .05 5;— 92 .922 00 8.8:: 0.5.05 43.2: 0.2 0000 .08.: 0230.95 30.50 3 5| 90.5300.“ ”FADE? $20 00 808980 20.2 0.0 0... 00.0 3.000 208.00 200 00< 0.0 0000 2.000.» 0.000 25 04. $013.00 .00 0000030 000000000 05 .80 2.0.0.00 00¢. 00.. 00.002 00.000 0.8 0.8.00.0. 0. .m 030... 89 Es a U I U 0% oo 0 C I oo 0 _ 02.0 nouaho>< _ q 000.0 :00 a 30.0 000.0 _ 0:0 _ 360 0060009. no .00 05:00th 202 020.2 020.2 020.2 0202 E50 2.00000 52000 00 m6 w m.m m mé 3.0 0.200.200 502000. on. 0.02 0.02 0.002 ~62 0.02 $00.03 E053 0023232502 2200 030 0:0 800 mmm.m 3:: ESE 02.2850 23532 0— 0.000 «.30. 0.000. 0.000 0.0.: Qiézgmo 0.95.50, .— 0002” o. _ 00m 0.0000 0.30m 0.30m 3-8 00:2 m>om< 002.35 00 .5905» v— v.ommv v.00? 0.0000 0.0000 0.003 2200000284000 0520 0B<3 00 E053 _. 0.9002 0.00: 0.02: 00.22 9002 3-”: 00:2 mo EOE? I Emowm mwowm 00000 0.0000 Emmwm 2005.0 no .5059 O 0.03..“ 00on 0. 5: 0.300 0.33 092 + 203.0 “20 E0295 m .023me 0.300 0.0000 0.500 0.02000 mE<3+§+2m§ "20 E059 m d... >m Gum—5mm: :U—ghm mU< Ecru .0ng 00620 ”$35 .00 con—600w oamwogwmm 0:0 mU< mo 306% $6000 805588 3202802. :0 030.0 90 592E... $3.237. 2:... 5:337. :253. . 8:: Stability (lbs) Flow (1/100)" Specific Gravity 2.54 2.52 2.50 2.48 2.46 2.44 2.42 2.40 2.38 2.36 2.34 2.32 4 4. 5 5 Asphalt Cement Content (%) 5.5 6 6.5 7 l 800 1 700 1600 1500 1400 1300 10.0 4.5 5 5.5 6 6.5 Asphalt Cement Content (%) 9.5 9.0 9° Vt 9° :3 >3 Vt >1 o 9‘ u- 4 4.5 5 5.5 6 6.5 Asth Cement Content (%) '/o VMA 9.x: VIC 91‘? cue °/o Air Voids 9’ .3 5‘ u. o v. .Nsos» DUO Vods Filled with Asphalt 0\ O‘ \l \l 00 00 \O O M O M O kl! O Ll. M 18.0 17.8 17.6 17.4 17.2 17.0 16.8 16.6 16.4 16.2 16.0 4 4.5 5 5.5 6 6.5 Asth Cement Content (%) 7 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) Figure 3.5 Results of Marshall mix design for AC5 mixtures. 91 Specific Gravity Stability (1b.) Flow (1/100)" 2.54 2.52 2.50 2.48 2.46 2.44 2.42 2.40 2.38 2.36 2.34 2.32 2000 1900 1800 1700 1600 1500 1400 1300 4 17.0 16.0 15.0 14.0 13.0 12.0 1 1.0 10.0 9.0 8.0 7.0 4 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) °/o VMA 17.6 17.4 17.2 17.0 16.8 16.6 16.4 16.2 16.0 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) “/0 Air Voids 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) Vods FilledwlthAsphalt 4 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) Figure 3.6 Results of Marshall mix design for AC10 mixtures. 92 a a .e... '2 t .- ..v.- N a lid}.- . a an]. 2.52 2.50 2.48 g“ 2.46 i 2.42 m 2.40 2.33 2.36 2.34 4.5 5.0 5.5 6.0 6.5 Asphalt Cement Content (%) 7.0 2100 2050 2000 1950 3‘8me (1h) '55 8 1850 \ i> 1800 1750 4.5 5 5.5 Asphalt Cement Content (%) 6 6.5 7 — H O 10.0 Flow (11100)" 9.0 8.0 7.0 4.5 5 5.5 6 6.5 Asphalt Cement Content (%) °/o WA 17.6 17.4 17.2 17.0 16.8 16.6 16.4 16.2 16.0 15.8 15.6 4.5 % Air Voids ”00000 AO‘MO Vods Filled with Asphalt \1 \l \l \l 00 00 N «b O\ 00 O N \l O 4.5 4.0 3.5 3.0 2.5 2.0 1.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) ‘\ < 4.5 5.0 5.5 6.0 6.5 7.0 Asphalt Cement Content (%) 4.5 5 5.5 6 6.5 Asphalt Cement Content (%) Figure 3.7 Results of Marshall mix design for AC20 mixtures. 7 41 u u t a. a. a. \- l-uba.-. .- v bin-56". A 1.: 6 3:27.21. :2»... a. 3...... 2.52 17.8 2.50 17.6 2.48 ”'4 17.2 g 2.46 17.0 5 2.44 E 16.8 > ° 2.42 °\° 16-6 2 40 16.4 m ' 16.2 2.38 16.0 2.36 15.8 2.34 ”'6 4.5 5 5.5 6 6.5 7 4'5 5 5‘5 6 6'5 7 Asphalt Cement Content (%) Asphalt Cement Content (%) 2500 4 5 2400 K\ 4.0 3 2300 i, a 3.5 c d>\ E g 2200 5 3.0 :a: \ .- w 2100 2.5 2000 T—j 2.0 1900 . 1.5 4.5 5 5.5 6 6.5 7 4-5 5 5-5 6 6.5 7 Asphalt Cement Content (%) Asphalt Cement Content (%) 13.0 90 12.0 W M 1?. a ' a :Eg‘IOD 0 // E 80 A E 9.0 / <5 :75 8.0 / 7.0 (r 70 4.5 5 5.5 6 6.5 7 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) Asphalt Cement Content (%) Figure 3.8 Results of Marshall mix design for AC5 mixtures with 5 percent SBS polymer. 94 2.52 2.50 2.48 2.46 2.44 2.42 2.40 2.38 2.36 2.34 Specific Gravity 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) Stability (lbs) § / <> \ 2200 \. 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) 15.0 140 1 ' ‘ / > 13.0 12.0 / Flow (1/100)" 11.0 t 10.0 ‘I’ l 4.5 5 5.5 6 6.5 Asphalt Cement Content (%) % VMA °/o Air Voids Vods Filled with Asphalt 17.2 17.0 16.8 16.6 T 16.4 16.2 ol/i / 16.0 15.8 4.5 5.0 5 5.5 6 6.5 7 Asphalt Cement Content (%) 4.5 4.0 3.5 3.0 2.5 & 2.0 \o 1.5 4.5 \IQQQOOOOOOOOOOO NAO~OOONAQOOO 4.5 5 5.5 6 6.5 7 Asphalt Cement Content (%) 5 5.5 6 6. 5 7 Asphalt Cement Content (%) Figure 3.9 Results of Marshall mix design for AC10 mixtures with 2 percent SBS polymer. .$.J.....X..—: 2.2—17.: 73.....332. ...J.:>_Ca. 32:. 22:23:! 5.: I: 21,—7.3: 7...: ...273LZ ...u.:.~: 2:279. .32.}. .»:..::::.r.. r...\ a .iaxa...‘ and man: 5.: ommd wand SYN we cmw 3.: $2 3.3 omnd mafia coed od 8.: com: 2.2 coma mafia pad Wm an U< cod: 5w: cod: ommé mmmd mg.” oh 8.: whom 3.: «mad mind Ema we and 9.0.: gem one: GEN bamd we: ed 3.: $3 5.3 coed coed SYN Wm mmmfieu 3 U< cod $2 3.0: omvé Nwmd oovd ed 3.: ma: 2.: oz..— wamd wQ‘u m6 cod: mm: No.2 o3: mmmfi weed o6 Ehm 9”,: «mo: eme— omm.m cwmd vwvd Wm 3 U< 9...: «we: 3.: oww.m gmm awed ow ofiw com: 3.: came mafia mend mé 3.: no: 8.: cmvd ahmd em: We on comm 2.: oncem :wmd weed od ooh end 32 :8: ova ~om.m am: On mum c\om m U< ocfi ocvm Nmé— ovmé wamd momd ow Omd we: was: GEN ohmd me: We omd com: NW: oamd mwmd need o6 ova ovfi $2 2.6: and bwmd whim Wm m U< 2.x wmm: no.2 aboé cth oom.~ ow cad 5w: 8.2 owed mmmd sand was a . Aka 88.80 5.3: Ame-:85 Aaxb 33.2.0 0 m 9:89: .3 58 8880 3:20 .533 E. b. _ a . . as; a? 280% 3:28.; .33 2...... i E: 2:... . a a... 8.85:6 E ...e am < E > ~==m 858882 88.80 04.. .3826.— _ a < .8838 2938 35:58 3830a 28 Emfibm 2: 8.: :wmmon x8 2883 05 .«o 83%: 08 go .9886 N _ .m 033. 96 3.7 In)”; 3.7 TEST PROGRAM Two types of tests were conducted to evaluate the structural properties of straight and polymer modified asphalt mixtures. These include; Indirect Tensile Cyclic Load Test (ITCT) and Indirect Tensile Strength Test (ITST). For all tests, Marshall size samples (4- inch diameter & about 2.5-inch thick) were used. The test procedures are presented below: 3.7.1 Indirect Tensile Cyclic Load Test The indirect tensile cyclic load tests were conducted to evaluate various properties of the asphalt mixtures. First, the sample was placed on the 05-inch wide strip on the loading device and a 30-pound sustained load was applied. When the sample came to rest, a cyclic load of 170-pounds was applied The deformation of the sample was measured in three directions using LVDT’S. Each loading cycle consists of O. 1 -sec load-unload period and 0.4 sec rest period. Figure 3.10 shows the Sinusoidal load cycles applied to the sample and a typical measured deformation. Figure 3.11 (which repeated for convenience) displays a typical behavior of an asphalt mixture subjected to cyclic loading. It can be seen that the total strain can be divided into 3 components, elastic, Viscoelastic and plastic. Part of the measured deformation, which was recovered immediately upon unloading is called the elastic deformation. Another portion that was recovered after a longer period of time is the Viscoelastic deformation. The third portion is plastic (permanent) in nature. This plastic deformation accumulates as the number of load cycles increases and it ultimately causes failure. Several trial tests were conducted to determine the magnitude of the cyclic, total peak and the sustained load. When a cyclic load of 300 pounds and a sustained load of 50 97 Defamation Load /\_l_/L Time J DP= Plastic Defonnation Figure 3.10 Typical load deformation cycles with 0.1 second loading time and 0.4 seconds relaxation period. 98 Load ——4 b Strain ‘ r Plague Strain kamc Strain Viscoelastic Strain Figure 3.11 Stress-strain behavior of conventional asphalt mixtures. 99 «1 guns at. n me: 0111. is l ‘ a ' »‘91 mt lm ~.- 4:: AuL‘Uill Resilil Kl}... Ll~‘ pounds were applied, the sample failed after few cycles in shear failure. After several trial tests, the magnitude of the cyclic and sustained load was selected as 170 and 30 pounds respectively. These make the total peak load 200 pounds and the minimum load of 30 pounds. The Indirect Tensile Cyclic Load test was used to determine the resilient modulus, the fatigue life, plastic deformation and the deformation rate of the straight and polymer modified asphalt mixtures. Resilient Modulus - The resilient modulus is an essential measure of the stiffness of a Where, particular asphalt mix under cyclic loading. The resilient modulus of the asphalt mixtures were calculated at test temperature of 23, 77 and 140°F using the following equation (51): MR = (0.253680*H + 3.9702876*V- 0.0142874 *A)/D (3.1) D = 1.105791(H2 + V2 + A2) - (h-0.0627461*V + 0.319145*A2),- H = DH*L/P; V = DV*L/P; A = DL/P; and M R: resilient modulus; D H: horizontal resilient or total deformation of the specimen along the horizontal diameter (inches); DV= vertical resilient or total deformation of the specimen along the vertical diameter (inches); 100 lat Fatigue Life - DL= longitudinal deformation along the longitudinal axis (thickness) of the specimen (inches); P = the magnitude of the applied load (pounds); L = sample thickness (inches). Fatigue cracking is one of the major load-related distress modes experienced in asphalt concrete pavements. In the laboratory, the fatigue response of the straight and the modified asphalt mixtures were studied under stress controlled tests using an indirect tensile test device. The test frequency was set at 2 cycles per second (2 Hz). Each loading cycle consisted of 0.1 sec of loading-unloading periods and 0.4 sec of relaxation time. All tests were conducted at a temperature of 77° F. For all tests, the fatigue life of the samples was defined as the number of load cycles that caused fracture in the sample. Due to time constraint, the test for some samples was terminated at 1,000,000 cycles and the fatigue life was considered as unlimited. Plastic Deformation- From the results of the indirect tensile cyclic test, the vertical plastic deformation versus the number of load cycles was obtained. This represents the resistance of the material to permanent deformation, which leads to rutting. The number of cycles that is required to develop certain levels of plastic deformation was considered to be the basis for the relative performance of the polymer modified and unmodified 101 Dela asphalt concrete mixtures. These four levels of plastic deformation are 0.01, 0.03, 0.05, and 0.07 inch. Deformation Rate - The deformation Rate (inch/ cycle) is defined as the magnitude of the horizontal plastic deformation per load cycle. This represents the rate at which the damage (permanent strain) is induced into the sample. This permanent strain is ultimately responsible for initiating fatigue cracks. It should be noticed that this deformation is caused by the tensile stress that is induced into the sample and it simulates the tensile strain that is developed at the bottom of the asphalt layer of the pavement due to traffic loading. This tensile strain causes the fatigue cracks in the asphalt layer when it exceeds the material capacity. Higher deformation rates, cause shorter and lower fatigue life. 3.7.2 Indirect Tensile Strength Test The indirect tensile strength test employs the indirect method of measuring the tensile properties. For this test, the same set up as the indirect tensile cyclic test was used, and a ramp loading was applied to the sample at a rate of 2-inch per minute until failure. During the test, the magnitude of the load and the sample deformation were measured. The data were used to calculate: Tensile strength - which is the maximum tensile stress applied to the sample during the test. Higher tensile strength indicates better resistance to cracking. Equivalent modulus - which is an indication of the stiffness of the material. Higher equivalent modulus indicates higher deformation resistance. The 102 Ira Co 7.1 equivalent modulus was obtained by calculating the slope of the first half of the curve. Since the load deformation curve remains almost linear up to half the maximum load at failure, equivalent modulus (EM) was calculated as follows (51): ' l P EM _ 543] (3.2) where, P is the maximum load at failure (pounds), and 5 is the deformation at half the maximum load at failure (inch). Fracture toughness - which is the total energy (lb -.inch) required to cause complete failure of the sample. This is equivalent to the area under the load- deformation curve at failure. Higher toughness indicates a material that is more resistant to cracking. Compressive strength - which is the maximum compressive stress applied to the sample during the test. Higher compressive strength indicates better resistance to rutting. The indirect tensile strength tests were conducted at test temperatures of 23, 77 and 140°F. The Marshall samples were placed in the temperature chamber for 4 to 5 hours and then tested within one minute. Typical load-deformation curves for straight and modified asphalt mixtures are shown in Figure 3.12. For each sample, the indirect tensile strength (ITS) and compressive strength (CS) were calculated for each sample using the following equations (51) : 103 ‘7‘.\IIIIIIII‘ Ill q a q 4 q u a fi 4 1‘1 d _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ a _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ 1..IrtsulurlrllTrun+IIIITIII+IIIITIII..IrrITrIILIItuT..uILIIII u _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ r1IILIIllr:InLllnlhuluLIIururulLnlnuulnrurrrnLuuntrunnLIIrI r _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ . _ _ I _ . _ _ _ _ _ _ _ _ _ _ II..I.IIII.I..1|.._II..|AuanuIIII_II:|_I..|:...uurvflurlJnIIIflIIu:_nli| u _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . In11:71:11uIIJIIIIananuIIIaIIII_I|u|4rllunnluJulnlfllu11711..1 _ _ _ _ _ _ _ _ _ . . . _ _ _ _ _ _ _ _ _ h _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ [Illinsaurllu|_nlll+IIIITIII+IIIITIIIill IT IIIIITIIIITIIII _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ -IIILuuln IIIL IIILrIItL _ r L Full: _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ l _ _ _ . I _ _ 1 111. IIIIIIIIIIIIIIII _ 111111111 _ lllllllllllllllllllll _..11..u _ _ _ _ _ _ . _ _ _ _ _ _ _ . [1113:1111 IIIIIIII _v||| _ululr _ . _ _ _ _ _ _ _ _ _ _ _ . _ 11:11:11: 11111111 + IIIIIIIIIIII Tutti! nIITIqu _ _ _ _ _ _ _ _ . _ _ _ . _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ P .— b b — p _ F P l? q — 4| 4 Id u « 4| Iql u a O 0 0 O 0 0 0 O O 0 O 0 O 0 O O 0 0 0 0 0 0 0 0 O O 0 6 4 2 O 00 6 4 2 0 8 6 4 2 2 2 2 2 11 .l. .l 1.. .I. 96855 1.31— — AC5-3%SBS 104 Deformation (inch) - - 'ACS-Straight 0.000 0.010 0.020 0.030 0.040 0.050 0.060 0.070 0.080 0.090 0.100 Figure 3.12 Typical load-deformation curve for straight and modified asphalt mixtures. liters 3.8 ‘1“ tr 3 Str Pol ITS=0.156x(—f—) (3.3) CS = 0.475 x [B (3. 4) Where, L = thickness of the sample (inch); and P = maximum load at failure (pounds). 3.8 EXPERIMENT PROGRAM MATRIX To accomplish the objectives of this study, several materials and test methods were used. Prior to commencement of the tests, an experiment design matrix was established. The experiment matrices for Indirect Tensile Strength Test (ITT) and Indirect Tensile Cyclic Test (ITCT) are shown in Figures 3.13 and 3.14 respectively. Each cell in the matrix represents a triplicate. There are three test temperatures, 23° F (-5°C), 77°F (25°C) and 140°F (60°C). Three types of asphalt grades (AC5, AC10 and AC20), five types of polymers (SBS, SEBS, SBR, EAM and CRM) with different polymer content. For each asphalt grade, three types are included in Figures 3.13 and 3.14. These are: Straight asphalt - virgin asphalt that was mixed with the aggregates at 290°F Polymer modified asphalt (PMA)- an asphalt that was mixed with polymers according to the established mixing protocol for two hours at 350° F or 380°F depending on the type of polymer used, after one hour of melting at temperature of 275°F. 105 Av.tv ‘..n.‘ ‘31.:‘\ “.Il‘ . o‘ul~t‘\‘v ‘ l.‘.clp.‘lll‘».t ‘p‘l‘tI‘I‘III. .33 88— 0:9? 0:88 “858 06 8.: 88.8 88»on 38088088 2:. m: .m 8:th commooam no\eo swag u: v83» 2d m=3 33m 8339.5 a 832%: :8 zoom Bang 2 m a: u 23 “swim co 35:2 228.. =8: 0383 05 8 woman: 08 33$ V8— 0880:: as 8:880 25 2%; 5. fig 3 2; H8056 a fig v 3.5.5:... 9: 815m 106 .41u::.1- fl...h.mullsfir....lu.h. -~N=§L~Tu .v~.~.l.:.v.~. s.v.v.-\~:~ .38 Emmet». 0:88 80qu 05 Ba “Fame 88on 358596 2F 3 .m 0.5mm 3388 0% SN n “:8 83% mo Easz 323.3. "as: 2.. 3 Etna ._ 32 95m :3 8:850 5:332 232 5 so... 3 E. .325... a as: v 35s.: 2. 8.95m 107 Pr: Processed asphalt - asphalt was processed in the exact conditions (temperature and time of mixing) as the polymer modified asphalt except that this was mixed without the addition of any polymers. Thus, during the mixing processes, this asphalt was subjected to the same aging conditions as that of the polymer modified asphalt. The reason for making this type of asphalt is to examine the effect of aging that occurs during the polymer mixing process on the engineering properties of the mixtures. The samples made from the above mentioned asphalt cements were cured for 12 hours or more at room temperature. For oven aging, the samples were placed in an oven for 7 days at 140°F and then cured for 9 to 12 hours at room temperature. For all the samples the following number designation procedure was adopted: 1. The first digit is the Test Type: Marshall Mix Design = 1 Indirect Cyclic Tensile Test = 2 Indirect Tensile Strength Test = 3 2. The second digit indicates Test Temperature: Temperature of 140° F= 1 Temperature of 77° F = 2 Temperature of 23°F = 4 3. The third digit indicates the Asphalt Type: AC5 = 2 AC10 = 3 AC20 = 4 108 4. 5. The fourth digit indicates the Asphalt Additive: Straight Asphalt = 0 ~ Asphalt + SBS polymer = Asphalt + SEBS polymer = 2 Asphalt + SBR polymer = b) Asphalt + EAM polymer = 4 Asphalt + CRM polymer = 5 The fifth digit indicates the Polymer Contents: Polymer content of 0%= O Polymer content of 1%: ' 1 Polymer content of 2%: 2 Polymer content of 3%: 3 Polymer content of 4%= 4 Polymer content of 5%= 5 Polymer content of 6%: 6 Polymer content of 7%: 7 Polymer content of 10%: 10 Polymer content of 15% = 15 The sixth digit indicates the Aging: Processed and PMA samples Oven aged samples (7 days, 140° F) Straight samples 109 7. The seventh digit indicates the Sample Number (SN) by order of test: SN: 1to3 For example, the designation number of third sample of AC 1 0 asphalt mixture containing 2 percent of EAM polymer tested under indirect cyclic load at 77°F is given as “2234213”. 110 CHAPTER 4 RHEOLOGICAL, MORPHOLOGICAL AND MICROSTRUCTURAL PROPERTIES OF PMA BINDERS 4.1 INRODUCTION As an integral part of this study, a series of experiments were conducted by the Department of Chemical Engineering at Michigan State University (MSU) to determine the physical, chemical and rheological properties of the straight, processed and polymer modified asphalt (PMA) binders. Details of the experiments can be found in references 1 through 3. The binder properties were measured by conducting the following four experiments: 0 Dynamic Mechanical Analysis. 0 Thermal Mechanical Analysis. 0 Rotational Viscometry. 0 Bending Beam Rheometer. The binder properties measured from the above experiments are: 0 The complex modulus (G*) and the time-lag between stress and strain. 0 The glass transition temperature (Tg). o The viscosity of the binders (n). o The binder stiffness (S) and its rate of change relative to loading time (the m-values). Recall that fiom the complex modulus and the time-lag data, the loss modulus (6”), the storage modulus (G/) and tan8 (the ratio of G”/G/) were calculated (see, chapter 2; Figure 111 binder m-Vah the the 4: 1411851 1 ‘ Mom EEIKTUI: 3‘1: . lL WW1 2.5). The G/ and G” represent the elastic and viscous component of 0*, respectively and the tan5 indicates the relative viscous to the elastic behavior of the binder. The T8 of the binder reflects the temperature at which the binder behaves like glass (brittle solid). The m-values indicate the rate of change of the binder stiffness with loading time. It reflects the thermal cracking potential of the asphalt concrete pavements. The Composite Materials and Structural Center at MSU also conducted various tests to characterize the basic morphology and microstructure properties of the binders, and binder-aggregate interface adhesion. Environmental Scanning Electron Microscope (ESEM) and Laser Scanning Microscope (LSM) were used to obtain the microstructural and morphological properties of the binders. At high magnification the ESEM allowed direct viewing of the thin film of asphalt binders and mixtures. This provided important information about the additional network structure due to the polymer modification and failure mechanisms, crack initiation and propagation under stresses. The binder-aggregate adhesion properties of straight, processed and PMA mixtures were studied by conducting Lap-shear test at different temperatures. The following properties were obtained from the Lap-shear test. 0 Lap-shear strength and stiffness. o Lap-shear fracture toughness. 0 Adhesive and cohesive failures. The analysis and discussion of rheological, morphological, microstructural and adhesive properties of binders are presented below. 112 4.2 RH E01 The rhe- investigated du the storage and and the moduli Engineering at f nemork and po the rheological ; engineering pro results ot‘the Th! 43.1 Viscosi Viscosit lied [0 CXpress mess to Shear 5 i‘7'0 L .:3 F‘ 3'5th shear strain is e 3r“? Table 4 b » ~ g imous p01 3 x1 . - ‘ twitield m {/1 BS. SBR and 4.2 RHEOLOGICAL PROPERTIES The rheological properties of the straight, processed and PMA binders investigated during the study include, the flow characteristic (Viscosity) of the binders, the storage and loss moduli and the network morphology of the binders. The viscosity and the moduli of PMA binders were studied by the Department of Chemical Engineering at MSU. The Composite Materials and Structural Center investigated the network and polymer-phase morphology of PMA binders. It was found that changes in the rheological properties of asphalt binder strongly influence the structural and engineering properties of the asphalt concrete mixtures. Therefore, for convenience, the results of the rheological properties are summarized below. 4.2.1 Viscosity Viscosity is a fundamental consistency measurement of asphalt cement binder used to express the resistance of the binder to flow. It is essentially the ratio of shear stress to shear strain at any given temperature and shear rate. At high temperatures such as 275°F, asphalt cements behave as simple Newtonian liquids (the ratio of shear stress to shear strain is constant) and have a totally viscous response. Since the mixing temperature of asphalt mixtures is 275°F or higher, the viscosity measurements provide sufficient representation of the workability of the binders (61, 62). Table 4.1 provides a list of the viscosity of straight, processed and PMA binders for various polymers and polymer contents at 275°F. The data were measured using the Brookfield viscometer. Figure 4.1 depicts the viscosity of the AC5 binder modified with SBS, SBR and EAM polymers as a function of the polymer contents. It can be seen that 113 Table - Poll C0 1 .13743u4\eww Table 4.1 Viscosity of the straight, processed and PMA binders (centipoise; cP). Polymer SBS Content (%) AC5 AC5 AC5 AC10 0 200 200 200 SBR EAM CRM 300 260 988 897 1,207 1,418 2,473 2,920 3,913 11,267 1 3 ,000 tests were 114 6M 5mm 1 , . in... l . . 3.0m 1l. 2.0m ll - - Viscosity ((‘ontipoisc; (P) 1.001) 1 . 51mm Figure 4.1 _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ r. _ _ _ _ mm , _ _ lllllll _ . h TilllllJlllIll llllllr _ um czmz _ . _ m m _ _ _ . Cm“ _ _ _ _ 8 _ _ _ _ .mwr _ _ _ lllllll _ _llll ILIIIIlIF Illllr _ ..MI ”(Cam _ I _ _ m333 _ . _ _ mm _ . . _ _ . . _ _ . . . r _ _ _ iiiiiii _ c _Iittii._lt- tint til: _ m1 SR _ _ _ _ .m BB _ _ _ _ P. SS _ _ _ _ _ _ _ . ”W. _ _ it itLti r. iiiiii L tttttt it: 1‘ _ m. ..‘ _ _ ..i _ S _ _ _ . _ . _ . _ _ _ _ . _ _ _ _ _ _ _ _ w iiiiii . tttttt AIAIII lllll J tttttt at it: _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ ttttttt _tIIItikIttIiLtiittiLitt trill it: _ _ _ _ _ / _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ . _ _ w iiiiii _ iiiiii A iiiiii _ tttttt i. tttttt 11‘ 1.. _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ lF » P _ P 0 0 0 0 0 0 0 0 0 O m 0 0 0, 0 0., .. 0., 0., 6 5 l Polymer Content (%) Straight Processed Figure 4.1 Viscosity at 275°F of SBS, SBR and EAM PMA binders as a function of polymer content and mixing conditions. 115 the Viscosity of There are three I hinders. These Lt l. The $110. I. The rent 3. The aim These mechanis and eausmg ine 11 Can he binder processe Weight unproer. 3.5 times hlght term aging duri 4.3.4). Higher n ‘he t‘iscositx' o: .- gher molecu}, asphall. On lht‘ ‘tl A Elbewlar “El; s balms. Slmilnl "3W that the I M = Cért' .lelpgise: CP the viscosity of the asphalt-polymer binder increases as the polymer content is increased. There are three main mechanisms responsible for the increase in the viscosity of PMA binders. These are: 1. The short-term aging of PMA binders during processing (see, section 4.2.4). 2. The reaction of the polymers with asphalt cement. 3. The absorption of the oils (light molecular weight materials) by the polymers. These mechanisms increase the higher molecular weight materials in the asphalt binder and causing increases in the viscosity. It can be seen from Table 4.1 and Figure 4.1 that the viscosity of straight asphalt binder processed under the SBS mixing conditions is about 1.5 times higher than the straight unprocessed asphalt. Binders processed under EAM mixing condition exhibits 3.5 times higher values. This higher viscosity of the processed binder is due to the short- term aging during processing and mixing of the polymer with the asphalt (see, section 4.2.4). Higher processing temperatures and longer time cause more aging. The increase in the viscosity of the EAM PMA binders with the increase in polymer content is due to the higher molecular weight material formed by the reaction of the EAM polymer with the asphalt. On the other hand, the SBS and SBR polymer systems absorb the lower molecular weight materials (oils) and hence result in higher content of the higher molecular weight material. This causes significant increase in the viscosity of the PMA binders. Similar results were obtained for the AC10 and AC20 PMA binders. It was also found that the ACS modified with 2 percent and higher EAM polymer contents did not pass the Strategic Highway Research Program (SHRP) viscosity specification (3000 centipoise; cP at 275°F). 116 Similarl were determine the (PM modit' higher CRM e0: seen from the d; ACZO with 15 p 375'? Various he etieet of 510 be“? about mor (the binder, n *he abSOl’ptign ( VlSCOSil)‘ in the the CRM COnte 4.2 2 ' StOragt The Ste»: ”11 fits f0r x'ari 3199mm). A it i ‘9, .. ‘ 36 Similarly the viscosity of the asphalt binder modified with various CRM contents were determined and are provided in Table 4.1. The data indicates that the viscosity of the CRM modified asphalt increases as the CRM content is increased. At 15 percent and higher CRM contents the mix becomes unworkable due to its high viscosity. It can be seen from the data in Table 4.1 that AC5 and AC20 modified with 20 percent CRM and AC20 with 15 percent CRM did not pass the SHRP viscosity specification of 3000 cP at 275°F. Various CRM modified asphalt binders were stored in oven at 350°F to examine the effect of storage time on the binders. It was excepted that longer storage times would bring about more interactions between the CRM and asphalt binders without severe aging of the binder. It was found that additional swelling of the CRM particles occurred due to the absorption of the oils from the asphalt phase. This resulted in an increase of the viscosity in the binders. It was also observed that the effect of storage time increases as the CRM content increases. The reason is that higher CRM contents produce higher surface area and hence more oil was absorbed. 4.2.2 Storage and Loss Moduli The storage modulus (G/) and loss modulus (0”) of straight, processed and PMA binders for various polymer contents were obtained using the Shear Rheometer (RMS-8OO apparatus). A frequency of 10 radians per second was used in compliance with SHRP specifications and strain levels of 7 to 10 percent were controlled to insure that the testing was conducted in the linear Viscoelastic range. Temperature sweeps were conducted from 86 to 167°F with measurements taken at 9 degrees intervals with an equilibrium period of 117 mo mmules m prondC‘d in Td The re> polymer conIL‘l reached. after ‘ The figure shot polymer col‘llt‘l‘ pohmer conten The $83 pol} n‘ pohmer conten moduli Values 0 concentrations; molecular itei g reduction in prc SBS polymer. Structure causir mene end bln issiiimateriai 9 Each other it P ‘i I that is were or. Mi? l’vi" mer S‘Sten W1 ”Ste 1 ‘ . m (at four 4‘— ‘1 . two minutes (63). A summary of the results (G’, G” and tan8) obtained at 95 and 140°F is provided in Table 4.2. The results indicated that for SBS, SEBS and EAM polymer systems, increasing polymer contents cause increases in G/ and G” until a certain “optimum” polymer content is reached, after which the moduli values decreases. This behavior is illustrated in Figure 4.2. The figure shows the G/ and G” of AC5-SBS PMA binders as a function of the SBS polymer content at 95 and 140°F. It can be seen that the values of G/ and G” increase as the polymer content is increased and they exhibit an optimum at five-percent polymer content. The SBS polymer system shows a sharp reduction in the GI and G” values at four-percent polymer content. Moreover, at polymer contents higher than five-percent, drop in the moduli values occurred. The SBS polymer is a network thermoplastic, at low polymer concentrations; it slowly destroy the natural structure of the asphalt by absorbing the light molecular weight oils. At about four-percent polymer content, the asphalt suffers a sharp reduction in properties due to the disruption of the polar resin-asphaltene network by the SBS polymer. At a slightly higher polymer content, the SBS polymer forms a network structure causing increases in the moduli values. The SBS polymer is composed of two styrene end blocks connected with each other to a thread like butadiene block. The styrene is stiff material while the butadiene is elastic in nature. The styrene end blocks are attracted to each other to form a network, which causes increases in the moduli values (63). Similar results were obtained for AC10-SBS binders and AC5 & AC10 modified with SEBS polymer system. However, no sharp reduction in the GI and G” values of SEBS polymer system (at four-percent or other intermediate polymer contents) was observed (see, Table 4.2). 118 hi... .i - l - t - _ H h a a : till. I:.:... r A F l_ H _l 4 Fl \...M.....I.~\ .I —i TXL 2.3-.09 Leah—cl Khmtzsg .Vf. ...2: {at «5.3?— 3...~ m3,_3 IU..3.3..9LC.0. .1 mLUUCF~ <2; 09:2..3 2.1.2:; .52.: 12,... a ..va 74:233.: 7.7.»: 4 ,- v.» 74:333.: blittf. his . 33:. h $063980 was» $8 on $8 3865 £06 33am U802 8.: RIM @151. 8.0 Sod :0 o: 8.0 Sod m3 3; mpg mo< mmmm 8.2 SE 30 mo mom 83 w 3w £62 95 03 £8 \b 9; mod Rod $.N £3 omo< : .8 5.2 no .. an 83 fine... . :23 29: e. . «anew who $3 9: 8.0 Bod a: 2% 20¢. mmm 3%. ~02 83 mo .. 3;. 08.0 8.2 8va 3.2m. of NE 2.0 08.0 9: . 42 3.0 5.0 80.0 2..” «.2 m2. 3; wo< 3.2 3:. 8.2 RS 3 3mm 9.: mg 83 flammouazh «Sum—«hum w v m N ~ c a «53. P. Rim. 2.; Aft 2.8.50 ..oEbom :35: 95,—. U< Leia—e.— .mooE 98 3 mo 85:23an «a E035 <35 085 mo 82$, 95 was 90v 9::on 32 ADV 3:608 omfloflm NV 033. 119 . h .. i an vi: .1: i. P Q: a. F TX; --u-:c.v ...u.->.—¢A- N ==-CQQ\ a§n-fih er. .73-3:0; .M.U--::~u. .v M1 8.- ..{QJ N. .wuaoswaoo was» 53 o: 88 3865 $8 392m “302 2%.... .. «m..- 33...”... 2; won ...: ...»..ovéfges 2.4.1..) $5.. ~m.m bmd 05.: was“ .n a”! (I..:/r... I: we. a seam” ...“... few. 8 fl Fwd 2nd ..S. .551 I/.>AVJ......¢/...nll. ...: 2.53.2.1 5. .. hunky,» ....3;w.#w»#wwa# sfiumwfifix ” a? .2. m m .o wed Sod o: o_U< 5..." «#3559 55.2.5... 25.5,. 3} ...«hfiua . u. a... ...»; 3 . . ... . xx» ... ”Kg 5 a: $3,“. ~N N mod meme 351$ 2...... . ...... ”5%. ”NM?! 111.7»:545 .. I . . ... I .. influx? £3 $4551.12”? 2 .. #2.. $5.... . r... ....) .5... . . . . .1 . 5.. u ” amen?» mv b— >zfi29uixry." Cd wofiw «9.3%. H” ... gxfie 0o. N. , . .. N ..«w. .....s. . we»? mmé N3; mmm . A... . “magma... mo N 15...}... ohgm mnvwm was 6...... any" ”annufifinurfii 2.0 owed 5.0 mood o3 Ewe 35w and up; 3.3. ...... «5...... ”gm .5. ...; <. ...5... ... $1.. «an Etc Homd mU< mm. 3.x www.mH end bhmd mod 30.0 o: vo.~ vmmé 5.5 52.2 5w mofifi ONU< ma 3.0 20.: who made 00.0 Sod mmMm o: aim 3&4. NW: wofiw 3;» commouogm c Nmog 20¢. ma lining a Agv €8.50 SEA—om F— :35: on»? Les—hon— ? 2:5 95,—. U< 38:80 3. 29¢ 120 _ U — 1‘ ~ 9 ~ A: av 2.9.:ch hush-Gan N mzl \mtzuez .15.; .~ Q.=.~.L. rV Q. .‘ab3~= 2: fiV Al. 173a 92.»..5 .-.v=~>.~°n~ 3~£~A F .3838". $3 $3 on «a5 8865 £8 @2995 6qu M, Emma“... 3:: was too :0 03 Rod 33 was o~o< 52.2 \O 3 82 O 20.: was SS :0 9; ..., 20.0 0 33. was 20¢. 2/-‘ Figure 4.7 ESEM micrographs showing morphology and network structure of the ISBSACIO/ACIOA and 6SBSAC10/AC10A mixes. 133 , ‘3‘." .1 l g ’u) :\(‘Ill—I’ruccsscd Binder 3-min. li\p0surc \ 251mm 1 I M, m [\(‘ms‘msnk Billdcl‘ fr“ ‘ Scwrul llllll. l’.\pusurc Figure 4.8 ESEM micrographs showing network morphology of the AC10-processed and AC10 modified with 5 percent SBR polymer content. 134 It was observed that the fibril diameter was approximately 5 to 10 pm, which was similar to the AC10 straight binders (68). The highly oriented fibril behavior of the AC10- 5%SBR binder was attributed to the strong texture of the SBR strands (1 to 3 mm diameter) present in the asphalt fibrils (discussed in section 4.2.3.2) which clearly affect the final network structure. c) EAM PMA Binders The network morphology of the EAM PMA binders was also investigated using the ESEM. Figure 4.9 shows the ESEM micrographs of the AC10-EAM processed and AC10-3%EAM PMA binders. The EAM-processed binder (see, Figure 4.9a) exhibit similar network structures as that of the straight binders with randomly arranged asphalt fibrils having a diameter of approximately 10pm and features resembling those of the aged binders (rougher or corrugated strand surfaces). This is because of the short term aging due to the processing and mixing of the binder at high temperature (discussed in section 4.2.4). The network structure of the AC10-3%EAM PMA binder (Figure 4.9b) appeared similar to that of the AC10-straight binders. However, the network formed in less than one minute, while the AC10-EAM processed binder required four minutes of continues exposure. This may be because of the less oil phase available due to the absorption of the oil by the EAM polymer. Another interesting observation was that the morphology of the EAM PMA binder did not showed any sign of aging in-spite of the processing and mixing at high temperature. This indicates that the EAM polymer modification prevent or retard the aging process (70). 135 .r’ r k/ " r \, k; , — ‘ ' I ACIU-ICAM Processed V" / [fl . \ ; I I J-Inm. l‘.\posurc h) \(‘IU- 3"hl‘ \\l Binder l- -IIIiII. E\posmc _ K “I... / hifllyt ‘ 1 'I: I . ‘ <5. 100 IIIII T, Figure 4.9 ESEM micrographs showing network morphology of the AC10- processed and AC10 modified with 3 percent EAM polymer content. 136 Figure 4.10 shows typical network structures of AC5 & AC10 EAM processed binders aged under TFO/PAV conditions. The network structure was not very well defined for the aged binders. After prolonged ESEM exposure (7-14 minutes), only a few traces of a highly coarsened network structure were seen in the EAM processed binders. The network structure of the AC5-EAM processed binder (Figure 4.10a) shows typical aged strands afier 7 minutes exposure. The fibrils were rough and coarsened with an average diameter of about 20 to 30 pm. In case of the AC10-EAM processed binders (Figure 4.10b) no network strands were observed even afier 10 minutes of exposure. When the binders were modified with three-percent EAM, the network morphology was smoother, less coarsened and finer, especially for AC10 binders as shown in Figure 4.11. The average diameters of the strands were 15 to 20pm for AC5 and 10 to 15 pm for AC10 binders. These results are consistent with those from all other PMA binders and further confirm that EAM PMA binders reduce the effect of aging (71). (1) CRM Asphalt Binders Thin films of crumb rubber modified (CRM) AC5 and AC10 asphalt binders were observed in the ESEM to determine the effects of rubber modification on the network structure. Three CRM concentrations of 0, 10 and 20 percent were used. Some samples were tested immediately after mixing. Others were stored for a period of 5 hours at 350°F. Still other samples were stored for 24 hours at 350°F. Figure 4.12 shows micrographs of the network structures in AC5-CRM processed binder for O, 5 and 24 hours of storage time at 350°F. There is no noticeable change in the network structure of those binders stored for O and 5 hours. However, the 24 hours stored 137 my ' . -.k II) Aged .~\('S-E.»\ M Processed 7—min. Exposure 4 .‘ fl ‘ h) Aged .-\(‘l()—l{.»\.\l Processed 7- min. li\posure Figure 4.10 ESEM micrographs showing network morphology of the aged AC5 &AC10-EAM processed binders. 138 I I f A. j; ‘. :1.1n.. ' -' a) Aged AC5-5%liAM Binder ,. -nIin. Exposure ' K Irv-1' f ,1) , ,, l))Aged.v\(‘ll)-3"/£.E.-\l\1 Binder ‘1 _-' 7-nIin. l‘:\[)()Slll‘C Figure 4.11 ESEM micrographs showing network morphology of aged AC5 and AC10 binders modified with 3%EAM polymer. 139 [T Pr”. ‘ IL A. n) A( 5- ( RH} PIIIIesseIl (0- hour) 4- min. Immune ‘9 \ \I < b) A( 5- ( RAI Processed (5— houls) 5- min. l‘\posure W )7: , g , ‘c x. 7 ltlll__LIIII c) A( 5- ( R\l Processed I24- houn) ( 5- min E\|H)\lllt‘. Figure 4.12 ESEM micrographs showing network morphology of the AC5- CRM processed binders and stored at 350°F for O, 5 and 24 hours. 140 binders show coarser and rougher network structures typical of aged asphalt. The mean diameter of the former network is about 10 to 12 um while that of the latter is 17 to 20 micron. These are consistent with the dimensions of all other binders investigated in this study. The CRM asphalt binders have a similar network structure, since the rubber is a particular modifier and does not form a network structure of its own. The rubber particles can be seen just below the surface and are visible after electron beam etching. Figure 4.13a shows the morphology of the AC5 modified with 10 percent CRM content for O-hour storage conditions. The rubber particles are evenly dispersed through the binder and the asphalt binder network fibrils are clearly seen to adhere to the rubber surface. Figure 4.13b shows the AC5 modified with 10 percent CRM content after 5 hours of storage time. The network morphology is similar to that seen in unmodified binder except for the presence of the rubber particles. Figure 4.13c shows the morphology of the AC5 modified with 10 percent CRM after 24 hours of storage time. The morphology is observed to be different than that of unmodified binders stored for 24 hours. The high temperature storage did not affect the coarseness or roughness of the network fibrils. The diameters of CRM modified binder networks ranged from 10 to 12 um regardless of storage time. These lead to the conclusion that the CRM modified asphalt binders have an increased resistance to high temperature aging. Similar results were observed for AC10 binders modified with CRM (72). 141 "-.. Y .»\... ' “ l V " fin/‘3’" 3 lab} \ ’_ "a),II's—III'LII‘RMIII—hour) - 4-nlin. li\posure H .1 . . h) .\(‘5-l0"/I.(‘R\l (5 hours) Z—min. Ewosure , . . .... , Figure 4.13 ESEM micrographs showing network morphology of the AC5 modified with 10 percent CRM content and stored for times of 0, 5 and 24 hours at 350°F. 142 4.2.3.2 Polymer-Phase Morphology Laser scanning microscopy (LSM) and ESEM were used to examine the polymer- phase morphology of the straight, processed and PMA binders. The LSM at Michigan State University (Zeiss 10) used in the study was a high performance light microscope which can be used to image samples using reflected, transmitted, or fluorescent light in either confocal or non-confocal modes. Examination of the asphalt binders with the LSM provided useful understanding of the size and distribution of the polymer-phase in asphalt modified with various polymers (68). A summary (based on polymer type) of the polymer-phase morphology results of PMA binders is presented below. a) SBS and SEBS PMA Binders Asphalt films of AC10-straight and AC10-5%SBS PMA binders were examined I at several magnifications using reflected light in the LSM. The low magnification images (100x-200x) appear featureless in all samples except for occasional voids. A typical LSM images of the AC10-straight asphalt binder are shown in Figures 4.14a and b using reflected light. The vertical lines in Figure 4.14a show the effect of laser damage to the specimen when filter was not used. At 1000x magnification, small black particles were observed in the AC10-straight binders using the reflected light as shown in Figure 4.14c. The particles were approximately 0.2 to 0.5 pm in diameter and well dispersed. The same area is shown in florescence mode in Figure 4.14d. No florescence from black particles was detected. Similarly, the AC10-5%SBS PMA binder was featureless when observed under reflected light at low magnification (200x) as shown in Figure 4.15a. However, when 143 “58. £2: a a»: vague SE a a. 3 a Q m a»: Aficwwvv 0:3 m5? woman: mm? 29:8 25. .503 8 3w: E8882.“ 53» Av 98 503838“ .xooofi USN @6qu 2355 30¢. mo Sufism 05 macaw 3&8?me 2mg 3 a. 2:5 144 Em: Gags 85 mam: comma: 83 29:3 2E. 503838." x83 98 xoom an Em: E8882.“ 515 G a. 3 23 53689.2 uncoo— Ea 38 a :3 cases as, a a. a a .8285 mmm$36< do 8&5 as wages Bandage 2mg 2 E: 2: v 05w; 145 examined under fluorescent conditions, large particles are clearly visible due to excitation with the blue 488 nm laser as shown in Figure 4.15b. The fluorescing particles were well dispersed and ranged in size from 0.5 to 2.0 pm in diameter. At higher magnification (1000x), the small black particles were again visible with reflected light as shown in Figure 4.15c. The same region is shown using the fluorescent light in Figure 4.15d. It should be noted here that the gray scale in the fluorescent image was inverted to show the black particles on a white background. Figure 4.14b and d clearly show that the large particles fluoresce strongly due to excitation with the green or blue laser; however, there is no detectable fluorescence from the smaller black particles observed in Figure 4.14c. These results indicate that the SBS polymer is the source of the fluorescent light observed in Figures 4.15b and 4.15d since the unmodified samples do not fluoresce. Thus, the use of fluorescence-LSM can provide valuable information on the size, shape, and distribution of the SBS polymer phase in the AC10 asphalt (68). Similar results were obtained for SEBS PMA binders. It was found that in addition to many small spherical particles (1 -3 microns in diameter), which fluoresced strongly, some larger block-like particles of SEBS polymer was also observed. These particles were approximately 5 pm in width and 20 pm in diameter (71). During ESEM examination of asphalt films at temperature of 50°F, it was observed that SBS polymers could easily be identified in the asphalt. It may be because the beam etches only the upper surface of the asphalt at temperatures of 50°F or below and therefore a network does not form during exposure. ESEM images of AC 1 0-5%SBS PMA binder are shown before exposure to the beam in Figure 4.16a. The ESEM was then used to etch the surface of the asphalt for several minutes and images are shown in Figure 146 u) \(‘ItJ—S'CrhSBS Before lileetron-Beuni li\posure h) .\('l0-5‘7'£.SBS After Iileetron-Benm I-Aposnl'e 50 IIIII Figure 4.16 ESEM micrographs showing polymer-phase morphology of the AC10 modified with 5 percent SBS content before and after exposure to the electron-beam. 147 4.16b. Particles on the order of 1 to 3 pm in diameter can be seen on the surface of the binder after exposure to the electron beam. In addition, a network does not form even after prolonged exposure since the temperature remained at 50°F. b) SBR PMA Binders Asphalt films containing 5, 3, and 1 percent SBR polymer were examined at several magnifications using reflected light and fluorescent light in the LSM. Typical LSM images of AC10 asphalt modified with 5 percent SBR are shown in Figure 4.17a and b using reflected and fluorescent light, respectively. The image formed with reflected light is featureless at 333x magnification except for occasional voids. When the same area is viewed with fluorescent light, a large number of SBR polymer strands (1 to 2 pm in diameter and 100 to 300 pm in length) were seen. It should be noted that in the fluorescent mode images the contrasts were inverted for clarity (71). LSM images of the AC10-3% SBR are shown in Figures 4.18 at 200x magnification for two separate areas. The fluorescent light images (Figure 4.18b & d) show a strongly textured SBR polymer-phase with fewer visible strands of SBR compared to the 5 percent SBR content (Figure 4.17). Discrete particles of SBR were also observed. LSM images of the AC10-1%SBR are shown in Figures 4.19 for two different areas at 200x magnifications. The fluorescent images (Figure 4.1% & d) indicate that the SBR polymer morphology consists almost entirely of discrete particles (3 to 5 pm in diameter). One possible explanation for the change in morphology is that the strand diameter in the one-percent SBR asphalt is now below the resolution of the LSM (i.e. less than 0.2 pm in diameter). However, higher magnification images of the one-percent SBR 148 .xmmm a 2m: Engage a? G a. 3 2a can a 3% Become a? 3 a. a .523 mmmsméa< a. a. a 323 “@8920... 3 a. .3 co Sea 2: Warsaw 292328 23 a A A. 2am 149 38:2 a? G a. a mmmgm , xoom 8 3w: “anemones—h 85 G a. 3 EB uhcom an Em: 30¢. 5 mo 89::me .6833 was 08.35. 05 wagon? wages? 2mg w“. a. 2&5 150 xoom 3 am: 608205 a? G a. 3 En £8 a am: 8882 .23 a a @ mmmxveoa. do 85:55 5&on Ea 8&5 2: mason... £385? 23 8:. 2:5 E: on: 8:. 2: .111 , . tummy . 5 . 151 asphalt also suggest that the SBR polymer consists of well-isolated particles. Finally, by comparing the AC10 asphalt modified with 5, 3, and 1 percent SBR polymer, one may observe a significant change in the SBR polymer phase morphology ranging from fine strands in the 5 percent sample to discrete particles in the 1 percent sample (71). An AC10 film modified with 5 percent SBR polymer was etched in the ESEM for several minutes using standard conditions to form a network. The sample was then frozen, and several areas examined in the LSM using reflected and fluorescent light. By comparing the network fibrils to the strands in the fluorescent image it was conclude that the fibrils show strong fluorescence, which is only possible if SBR is present. This result suggests that the network fibrils contain SBR polymer ( 71). This test method can be used to finger print the presence or absence of polymer, its content and its type. Similar results were also found when AC10 asphalt modified with 5 percent SBR polymer was exposed to electron-beam of ESEM. The un-etched asphalt film is shown in Figure 4.20a and the etched film is shown in Figures 4.20b. Strands of SBR polymer with approximately 2 pm in diameter are clearly visible on the surface of the binder. This image correlates well with the fluorescent-light LSM image of AC 1 0- 5%SBR PMA binder shown in Figure 4.17 (71). d) EAM PMA Binders The LSM technique was also applied to investigate the polymer phase morphology of the EAM PMA binders but it was found that the EAM polymer did not fluoresce using the blue or green laser and the polymer size and distribution could not be determined using the LSM images ( 72). 152 II) .\(‘|0-5'.’I.SBR Before Electron—Ben III 200 IIIII . .x I .0 .' 3- .. l .. h) A(‘|0-5"ruSBR After Electron-Benin ” ' ’8 4‘17“ Figure 4.20 ESEM micrographs showing polymer-phase morphology of the AC10 modified with 5 percent SBR content before and after exposure to the electron-beam. 153 e) CRM PMA Binders The LSM) was used to determine the distribution of CRM particles in modified AC5 and AC10 binders. Two types of lasers, blue (488 nm) and green (514 nm) were used for exciting fluorescence in the crumb rubber particles. Red-orange band pass filters were used to detect the fluorescence emitted from the samples. It was determined that crumb rubber did not fluoresce under these conditions. To confirm this, pure crumb rubber particles were observed directly under the LSM using both reflected and fluorescent modes. It was found that the crumb rubber particles did not fluoresce, however some fluorescence (white color) is observed from impurities and debris present in the rubber. Thus confocal LSM cannot be used to determine the phase distribution in CRM asphalt binders ( 72). 4.2.4 Effect of Processing on Short-Term Aging Recall that the mixing procedures for the PMA binders were developed by the Department of Chemical Engineering at Michigan State University (based upon the improvement of rheological properties) (see, chapter 3, section 3.5). In order to obtain improved rheological prOperties of the binder and complete formation of polymer network structure and or reaction of polymer with asphalt, it was recommended that PMA binders be heated to and mixed at specific processing conditions of temperature and time. The processing conditions used for various polymers investigated in this study are mentioned in section 3.5, chapter 3 and, for convenience, are tabulated below. 154 Polymer Type Mixing Mixing Time Remarks Temperature (°F) (hours) SBS 350 2 Network Thermoplastics SEBS 350 2 Network Thermoplastics SBR 350 1/2 Network Thermoplastics EAM 380 2 Reacting Polymer CRM 350 1/2 Particle Modifier The processing of the PMA binder at higher temperatures and the subsequent mixing with the aggregate resulted in aging of the binder. Aging occurs mainly due to two mechanisms: 1. Loss of volatile components or evaporation of lower molecular weight materials (oils and resins) during high temperature processing. 2. Oxidation/polymerization of the asphalt molecules at the surface. Both mechanisms result in an increase in the viscosity of the binders. Recall (section, 4.2.1) that the viscosity of processed binders were significantly higher than the straight ones. The viscosity of the EAM-processed binders exhibits the maximum increase among all other asphalt binders. This is because the EAM processed binders were mixed at higher temperature (3 80°F for a period of two hours) relative to the other processed binders as mentioned in the above table. The rate of evaporation of lower molecular weight materials and oxidation/polymerization of higher molecular weight materials increases with higher temperatures (73). Therefore the aging effect is significantly higher for EAM processed binders compared to the other processed binders. The increase in the viscosity due to short-term aging of the binders was responsible for the increase of various engineering properties (tensile and compressive strengths, equivalent and resilient moduli, fatigue life and rut potential) of the processed 155 mixtures (discussed in section 4.3). It should be noted here that the improvement in the engineering properties of the processed mixtures due to the short-term aging of the binder does not necessarily indicates better performance of the processed mixtures relative to the straight ones. These and other issues affecting the engineering characteristics of asphalt mixtures are discussed in the next chapter. 156 CHAPTER 5 DATA ANALYSIS AND DISCUSSION 5.1 INRODUCTION The structural and engineering properties and the failure mechanisms of polymer modified asphalt (PMA) mixtures are functions of: o The polymer-asphalt binder chemical, physical, rheological and adhesive properties. 0 The distribution of the polymer in the asphalt cement. o The morphology of the polymer-asphalt binder network structure. 0 The aggregate gradation and angularity which govern the void size and distribution. Two types of tests were conducted to determine the structural and engineering properties of straight, processed and PMA mixtures. o The Indirect Tensile Cyclic Load Test (ITCT). o The Indirect Tensile Strength Test (ITST). The structural and engineering properties obtained from the ITCT are: 0 The resilient modulus. o The characteristics of the vertical and horizontal plastic deformations that affect the fatigue life and rut and temperature cracking potentials of the asphalt mixtures. The structural and engineering properties obtained from the ITST are: 0 Indirect tensile and compressive strengths. 157 - Fracture toughness 0 Vertical deformation at failure. 0 Equivalent modulus. Recall that (see chapter 3) the equivalent modulus (which reflects the stiffness of the material) was obtained by calculating the slope of the linear part of the load-deformation curve (the curve is almost linear up to half the maximum load at failure). Higher equivalent modulus indicates higher deformation resistance. All tests were conducted on triplicate samples at low (23°F), moderate (77°F) and high (140°F) temperatures. In this chapter, the measured structural and engineering properties of straight, processed and PMA mixtures are analyzed and discussed in relation to: o The physical, chemical and rheological properties of the binders. o The morphology, microstructure and binder-aggregate interface adhesion properties of the PMA binders and mixtures. As stated in chapter 3, the polymers used in this study were divided into three groups as follows: 0 Network thermoplastic (SBS, SEBS and SBR). They behave like resins and form a network of themselves within the asphalt cement. o Reacting polymer (EAM). It chemically bounds to the asphalt cement particularly, the asphaltene and forms an elastic network. 0 Crumb rubber modifier (CRM). The crumb rubber does not. form a network of its own. The rubber particles disperse in the asphalt cement and peptizing agent such as resins is required to stabilize the modified system. 158 The structural and engineering properties of polymer modified asphalt (PMA) mixtures analyzed and discussed in lieu of the effects of the rheological properties, morphology and adhesive properties and are presented below. 5.2 THE STRUCTURAL AND ENGINEERING PROPERTIES The structural and engineering properties of straight, processed and PMA mixtures were obtained at test temperatures of 23, 77 and 140°F for various percentages of polymer contents. Samples aged in an oven at 140°F for seven days were also investigated in this study. It was found that for each polymer type, increasing the percent polymer from zero to an optimtun value causes an increase in the engineering properties of PMA mixtures. This behavior was due to the improvement of the rheological properties (storage modulus; GI, loss modulus; G” and tan5) and to increases in the adhesion properties of PMA binders. Increasing the polymer content above the optimum causes either no significant improvement or decrease in the properties. The polymer modification showed significant improvement in the tensile and compressive strengths, equivalent modulus, fracture toughness and resistance to plastic deformation. The improvement in the structural and engineering properties caused higher fatigue lives, high resistance to rutting and low temperature cracking potentials of the PMA mixtures. The SBS polymer system was selected as a typical polymer to illustrate the improvement in engineering properties due to polymer modification. The Analysis and discussion of the test data are presented below. A comparison between the affects of various polymers and the engineering characteristics of asphalt mixtures is presented in section 5.2.3. 159 5.2.1 Styrene Butadiene Styrene (SBS) PMA Mixtures 5.2.1.1 Load-Deformation Characteristics Figure 5.1 illustrates two typical load-deformation curves of the straight and SBS PMA mixtures subjected to static loading. It can be seen that as the load increases the total deformation of the mixtures also increases until a maximum load is reached where the sample fails. This value of the maximum load at failure is referred to as the peak load (Pf) andthe corresponding deformation is called the total deformation at failure (6]). It can be seen from Figure 5.1 that the SBS PMA mixture (curve-B) exhibits significantly higher peak load (PBf) than straight mixture (PA/r, Curve-A). However, the total deformation at failure of both straight (5,41) and SBS PMA mixtures (63;) remained almost the same. Moreover, for the same applied load (p), the SBS PMA mixture experienced substantially lower total deformation (63) than the straight one (6,.) Since the total deformation constitutes of elastic, Viscoelastic and plastic deformations, one can conclude that the straight asphalt mixture experiences higher plastic deformation than the SBS PMA mixture. These and other issues are presented in the next subsections. Examination of Figure 5.1 indicates that the load-deformation curves are almost linear in the range of zero to a load value equal to half the peak load (Pf). Therefore, an equivalent load-deformation ratio is defined here as: (P, ) NI— EM = 6 (5.1) (l/Zl’f) Where, EM = Equivalent Modulus (Pounds/Inch). 6 (”211!) = The measured deformation at half the peak load (l/2Pf). 160 2,400 PB! : : Curve-B (PMA Mixture) I I I I 2,000 ~ ——————— I ----- I. ------- I ——————— I —————— I I I I A 1,600 - ------- ' —————— I. ——————— I ——————— L —————— m I l l c I | I I = I I l I = | I l l O I l I l 5" 1,200 ‘ """" 1" —————— -: It ------ i ------- :- ------ 3 I I I . I 3 p 1 : Curve-A (Straight:Mixture) 800..” "In “I... ..... I ....... L ...... P ' I I I I I I 400 -- _- --. ....... a]... ..... , _______ r ______ O 53 5,1 6B 6Af 1 i 0.00 0.02 0.04 0.06 0.08 0.10 Deformation (inches) P A]. P Bf: Peak load of straight and SBS PMA mixtures, repectively. 6A , 63 = Total deformations of straight and SBS PMA mixtures, respectively. 6A,, 63, g Total deformation at failure of straight and SBS PMA mixtures respectively; and p = Applied load. Figure 5.1 Typical load-deformation curves of straight and SBS PMA mixtures under static loading at 77°F. 161 The EM values were then compared to the resilient modulus values. The benefit of the EM is that the test can be conducted in a short time period and it does not require sophisticated and expensive equipment as the resilient modulus test. 5.2.1.2 Tensile and Compressive Strengths Tensile and compressive strengths are the maximum tensile and compressive stresses, which a sample can sustain before failure. The tensile and compressive strengths are two fundamental engineering properties, which are used to evaluate the resistance to fatigue low temperature cracking and to the rut potential of asphalt mixtures. The load- deformation curve (see, section 5.2.1.1) and equations 3.2 and 3.3 (see, chapter 3) were utilized to calculate the tensile and compressive strengths of the straight, processed and PMA mixtures. The analysis and discussion of tensile and compressive strengths of SBS PMA rrrixtures are presented below. The improvements in the tensile and compressive strengths of SBS PMA mixtures are shown in Figure 5.2. The figure depicts the average tensile and compressive strengths of triplicate samples of AC5 and AC10 straight, processed and SBS PMA mixtures at 77°F for various polymer contents. The actual strength values of triplicate samples are listed in Table 5.1. It can be seen from Figure 5.2a & b that the average strength values (tensile and compressive) show significant improvements with the addition of SBS polymer, and exhibits an optimum value at the five-percent polymer content. The AC5 and AC10- processed mixtures exhibit increases in the average values of strengths over the straight mixtures of approximately 36 and 45 percent respectively. These increases in strength are mainly due to the short-term aging (see, section 4.2.4). It should be noted here that the 162 A AC10 PMA Mixtures $5 539% «=25 Straight Processed Polymer Content (%) a) Tensile Strength. I AC5 PMA Mixtures A AC10 PMA Mixtures I $5 $555 «22035.5 Processed Straight Polymer Content (%) b) Compressive Strength. Figure 5.2 Tensile and compressive strengths of AC5 and AC10 mixtures modified with SBS polymer as a firnction of polymer content at 77°F. 163 . . 5 . ... .s a”... . . 5.......<.......>.”, iii as ” :14:- < . .. 5. 5.. . ..v 5.. I. IS (...... a... .. ”infifififfi 2...“, .. . . ..r .. i” .n. . ”an”... #1. PM” aw? ..4. 155...? ..v.. . 5.5.53.” ...? :5 ehmd 5 ... . a... . I .3», flirfffiaf. Sod avod mmvéo m—ENNM :o.o $56 $432 Qum bbmd N—V—Nmm mad $6.9 $5.3 cod :Emmm a ,5 . a"... , ”In, . . ...... . . .n .....2 . . . 5....9 5, ,2. ., haughty, 1:1, .. 65$: ..ut. Ariana». 223m mbod Swdv 3M wmod v3.9: cud 3m.~ 322m mmmxmfiuaq. 2+ benm 2mg?“ Boommm mom.~ m Sommm mmm.~ Eoommm commoooi 3m .N m A oommm mmm-mo< mmod mawroc «m emmd Begum :ocwmm 03.3 on 2.4 092% am owed momd oncoamm 23 23 35.0 5.: mm E E as 82 £82 no.0 $3 89.? a as on :1. $2 4882 :3 $2 2% 4m ”S R can Rm.“ 88$ swims? 23 on; 296 4m 8N E mg 82 28$ sod ES 3.8 2 EN : 2w :2 382m :5 .53 E: 233. £3 €15 as 55 28> 5. 92:9 eoaasz 28:5 luck 2: hi:— ea iota—Fahua man—m5: magma—much. flaws—0.5m JEEP—«m . . . Homuaflwmmun hue-uh—On— fig E85.— 959on 3 nous—Echo: 13E?» Eofizsum 9:532..— ozmnoaaeu 955,—. 2..an 99C. :2 not. mo 838388 68 “a $538 880qu 2.398 .«o 3158 v.8 amoeba 0:28 82%.: 05 mo 5883 5m 2%... 164 228m ommfimmfi 22mg .. 5.5.} 1M.» , .234 $.24 H . 5.. 73".”... oZAmS at me”? .3? . X)“. .. 5': mad», , wag . “’59 u... . $5522 . .. 1. (3,2; .4” ”an Hg.” . 11’}:er . Ess. M 5”” , , ”3.... v :Nfimmm fins” ”m ”mu/,3 a,» mm» ...: it» BOOMNM omvdw mm :m I; EQN: 3w m—m 03 EOOMNM owN.mN_ 5 Sn m2 303% 35.8 ow :m N: m—OOmNm «#0.; mm Em m: Seemmm Nwodm :comumm , , am”. 08“ h macaw $3” macaw S ~8on webdm .3: .. 332 .6. Sign vcmfiw SSNNM wmw$b$U< $1 1...: 3953 :SNNM c8800.:— mmmA: U< 2 2 mam wood: 3 N? NE Emam 54. $2” 223m mmm$m$U< www.mfi 3. w? 3; awn woe mwmd :Bmmm :5 EB: :5 9.5—Rm €25 GE: 92: £25 9.5 2.5 .__< 515.5 baa—i «5:50 :«oh 2: :2. 2w nozafiheun 3.352 “mac—Esau. 539.5 5»:th was 59.5 2 .8.— EES—wmmon .5835— was «a Essa—Leon Ear—3w «no—«3.5”.— EEuEm ozuuoEEoU cinch gaun— a .— ... m 295m 99:. £2 .8838 3 23 165 5 omm_: 51mm a“ .. _ SmJS 29o Need an m: N 3% 0mg“ 22mg :99 owed >36: mm wmv EH main _m.m wmd m—Smmm awed Sod No mg 0: mmflm £6 me :Smwm mmvnm: mm 5mm 322m mad 3062 no 3% mme 223m Sod abods mm 5% med 22mg :od 52592 E. m3 memN Sm—mmm :od Eta—3 5 com ommd 228m om omm «mmw Eflmmm M205 ovwd: » @842 SN 218m Sod was 8%: S ”2 238m no.9 MK 0mm 53% .... 228m wmmfauo _ U< mmmgméfioxx mmmX?OHU< omod 3m 2: mm mmm EL wwo N am and fled Sod woof; F own 33 mvofi 85 omd NH imam mmmfimé5< Sod Sod www.mi on Em NS mowd co.»V Rd ZDMNM a5 ES: 3: 9.3.3— cm5 326 :2: A55 A3: . San—=2 2.3.50 . Eo> .__< 3395 xaum 2: =5— 3 53an.3: 2:352 amen—i=5. 5»:th :Enuaam ES: 595 u. 82— 535539: bah—om ...—s as 5%=—Lao: Ext?» “no—«25”— ouauahm 92895::5 9:32. swam . A ... m 2.3—am 2:“. £2 85.28 E 23 166 improvements in the strengths of processed mixtures due to short-term aging does not necessarily indicates better performance of processed mixtures relative to the straight ones. At the optimum polymer content of five-percent, the strengths of the SBS polymer modified AC5 and AC10 mixtures are approximately 1.45 and 1.32 times higher than the processed ones. Moreover, the strength values decrease after the five-percent polymer content. The improvements in the tensile and compressive strengths of SBS PMA mixtures are due to the following reasons: 1. The increase in the viscosity of the SBS PMA binder. 2. The increase in the GI and G” of the SBS PMA binder. 3. The formation of polymer network structure. 4. The improvement in the binder-aggregate adhesion properties (see, section 5.2.1.5). Recall (section 4.2.1) that the viscosity of SBS PMA binder increased as the polymer content increases. The higher viscosity of SBS PMA binder produces stiffer mixtures relative to the straight and processed ones. At seven-percent polymer content, it was observed that the viscosity of the binder exhibit considerable increase of about 4.5 times the processed binder, and the strengths of the mixtures decreased (see, Figure 5.2). It was also observed during mixing that the polymer-asphalt binder was very viscous and difficult to mix with the aggregate. First it was thought that the decrease in the tensile and compressive strengths are due to the high viscosity which may have adversely affected compaction. However, the densities and air voids of the samples were similar to those mixtures made with lower polymer contents (see, Table 5.1) which likely rule out the 167 compaction problem. Examination of the dynamic mechanical analysis (DMA) test results showed a decrease in both the G/ and G” values of SBS PMA binder with more than the five-percent polymer content (see, section 4.2.2), which resulted in the decrease in strength. The study conducted on the GI and G” of SBS PMA binders (see, section 4.2.2) showed similar trend as that of the strengths of the mixtures. The G/ and G” values of the binders exhibit substantial improvement with the SBS polymer modification. Recall that the SBS polymer composed of two styrene end blocks, which are connected with each other to a thread like butadiene block. The styrene is stiff material, and the butadiene is elastic in nature. The styrene end blocks are attracted to each other to form a network structure. The formation of the network structure of SBS polymer in the asphalt binder causes increases in the tensile and compressive strengths. Similar results were obtained for the mixtures tested at temperature of 140°F, and the aged mixtures that were tested at 77 and 140°F. However, no significant improvements were observed for both unaged and aged PMA mixtures at the test temperature of 23°F. Since all samples were tested at the same cyclic load (same stress), the higher values of strengths significantly decrease the stress-ratio of the PMA mixtures. Figure 5.3 shows the effect of SBS polymer content on the stress-ratio (stress/strength) of the AC5 and AC10 mixtures modified with the SBS polymer as function of polymer content at 77°F. The figure shows that the stress-ratio decreases as the polymer content is increased and it bottoms out at polymer content of five-percent. Moreover, the stress-ratio of AC10-SBS PMA mixtures is lower than the AC5-SBS PMA mixtures. This is because the AC10 mixtures have higher strengths than the AC5 mixtures. The decrease in stress- ratio indicates that the SBS PMA mixtures are subjected to lower levels of total strain 168 MM --WW ......s lam AA .A 3.5— 32% 0.00 Processed Straight Polymer Content (%) Figure 5.3 Stress-Ratio of AC5 and AC10 mixtures modified with SBS polymer as a function of polymer content at 77°F. 169 relative to the straight and processed ones. Recall that the total strain consists of elastic, Viscoelastic and plastic strains. Therefore one can conclude that the SBS PMA mixtures experience less plastic strain for the same applied load. The significant improvements in the tensile and compressive strengths and the decrease in the stress-ratio clearly indicate higher resistance to cracking and rut potential of SBS PMA mixtures. In order to examine the effect of oven aging on the tensile and compressive strengths of straight, processed and SBS PMA mixtures, the average tensile and compressive strengths of oven aged mixtures were normalized relative to the average tensile and compressive strengths of unaged mixtures. Since the normalized tensile and compressive strengths values of aged mixtures were similar, only the normalized tensile strength data was plotted as a fimction of polymer content and is shown in Figure 5.4. The actual values of triplicate samples of aged mixtures are listed in Table 5.2. Examination of Figure 5.4 and the data in Table 5 .2 reveals that: 1. The average tensile strengths of oven aged straight AC5 and AC 1 0 mixtures are approximately 25 percent higher than the unaged mixtures. 2. The average tensile strengths of the oven aged AC5 and AC10 mixtures which were processed under SBS mixing conditions show on average increase of about 20 percent with respect to the unaged mixtures. The oven aging of the mixtures (7 days at 140°F) caused the lower molecular weight materials (oils and resins) of the asphalt cement to evaporate. This resulted in higher viscosity of the binder and hence stiffer mixtures relative to the straight ones. Recall that the processed mixtures have already experienced short-term aging during processing and mixing at 350°F (see, section 4.2.4). Therefore, the oven aging of processed mixtures not 170 _ 5% ...m m... mm PS S 8% S0 -517 CC AA Er... _ n lllll P _ _ _ _ _ ii]. .............................. _ _ _ < _ n 5 0 7 5. 1 1 9:55.: wow< .8 nawuobm 2.2—oh. toga—Eel Processed Straight %) Polymer Content ( Figure 5.4 Normalized tensile strength of oven aged SBS PMA mixtures as a fimction of polymer content at 77°F. 171 .../w: ., ...H. g. .....-“ ...... .... “a... firm... up... ...”... .... w... ....“ ... mfiod owed oom Go mm Nov Nm— ~m_.N.. ~o.v. NON mNm~.NNm 23 Rod 3.? an :3 m2 $3 93 ”2 82% ~25 33 Row: 2. g E 35. N PM ”3 :28.” 82: :3. on” 282m 8qu a 2:. 2; OS. 52 N832 02.2. 5 SN 8 R... 92 mean ”8.8 8 ca 3 2:. a.” N852 a :w we fiuofiwmm h. u. .5...)$,..>$...> n. ?S..<.<...5.......... .<... . mgr. .. mmw$m60< Emmoooi mmm-m0< mmod 336 8va mm mww vo omm _ cod wmd owoommm :od mmod Omhfim Vm own mm 3;: ~m.v bmd mvoommm 23 $3. 2 $3 3 3m 2: $3 m3. mg 3582 33.33% :3 Rod 89? mm SN 8 S 3 v3. 52 258% Bod mmod 20.3 mm 03 S «2.; 3+ om.~ 3533 E: Ea . 38.80 .33 :3.— 233. a £5 E .5 $5 an: an: is» 5. £35 .282 55?: 2.1632 anon-$2; fiwuuhm flushm was . ,:o_§:u_nun 2: =3 8 Susan—89¢ no 3:... 9:52 35335.5 2.2:. :3 .59.“:— uEuoam 2956 can :2.anqu 185.5 > a _ . . H m . . m 25,—. 52 mo: mo 838388 on 3538 880:8 :2me vows mo £38 88 “#anch 0:38 686E 2: MO 3885 Nm 2%... 172 a» 1. 30¢ mm m :3 285 Eng ow Sod covod moods be mg ovo.m omfi mmd NVOOMNm 22. 8:; 08.8 S w: «2 08a 3; mm.” 308% mad ”So 8%: S 8? 5 83 3% mg 822m :3 :3 3.3: 3 MS a: $3 o3 o2 822m 23 Sod 232 3 8v 5 3% 3m mg 53% mmmfimmm :3 $3 232 3 ME. 5 :3 3+ 93 £3 23 93: mm E. E :3 8+ SN 828m :3 Sod 82: mm 8m 02 $3 3; mg 82% mac—Nam amazméo,‘ mmwghflulw mmm§m$U< Sod 35o etum: ov a? S; wo~.~ ohm 2mm Sod God 333.: mm “Q «E nmqm Ge. wwm mmimwm wmmc\$13< Sod 95.0 0:32 mm 9% mi 83 $4. emu Simmm :5 c5 35 9:. c as 33 9 c .5 E: 25:50 was an: 9.5—ma..— «a . . a . . a Eo> .3. 5395 a z hon—ho..— magoE amonawach 59.25 fiwnobm EBA , acts—awn:— 2: =3 8 zeta—Egan: 5 a 5:5— opEuEm ,u>_mmo.:_EcU 255... £35 335m 328% «ES-“w =5 5:55.8qu 32:3, . _ . . . . 25G. :2 .823qu 3 2a; 173 m, , . , .. .....k5z..." . mmodm— MN: mmm ooofinm 3 So com NN: mam «6 So now a: mmm meJéN 2&3: mmflmmm max”: on chm 52 mmmfimmm mvmdom :. emm NM: :mfimmm 3%: 332m 83: S. 9% 3: 8+ RN $5M . 3 8m 3: gm Rum Siam Go fimm mmw$>é~o< wmmfoméfioxx mmvaA: U< $93: 5 SN 228m was: mm 3m «2 EN mmflmwm :32 RN 822m 5M mmmgmé~0< Bod 366 5:3 2. mom m3 tum mmofimmm :od _mo.o cmvnwfi v» me om: 2 .m «mm 33 mam 888on Sod Sod 3.32 S G; _2 PM. mom :2 «NM mmm-20< :5 E 35 9:5 33 as is 2.52 22:8 .33..— xaom 95:5.— .a . . . . 2.5 L2 55.20 .55.??— ?3—52 may—Eng... :kuubw aauohw ES..— . dots—Human: 2: :2 «a grantee: 5135—. 9:522 3.39.5.5 a 25 3 383m #:9on u 95 can act—«EH89: 396.5» a . H m . . U a ... z .— _ w 2:“. £2 .uoséaoo S ”SS 174 only cause further loss of the lower molecular weight materials but also oxidize and polymerize the higher molecular weight materials (asphaltenes). This aging results in relatively more brittle mixtures as compared to the unaged mixtures. This can also be seen from the considerable decreases in the vertical deformation at failure (see, Table 5.2) and the increase in the modulus values (see, section 5.2.1.3) of the processed mixtures due to aging. Further examination of Figure 5.4 and the data in Table 5.2 indicates that: 1. At polymer contents lower than seven-percent, the oven aged AC5 and AC10 mixtures modified with the SBS polymer exhibit no significant increases in the average tensile strength relative to the unaged mixtures. 2. At seven-percent polymer content, the oven aged AC5 and AC 10 mixtures modified with the SBS polymer show an increase of about 31 percent in the average tensile strength as compared to the unaged mixtures. These observations indicate that the asphalt binders modified with SBS polymer system substantially reduce the aging of asphalt concrete mixtures. The five-percent polymer content shows the maximum improvement in the resistance to aging, however the seven- percent polymer content seems to increase the effect of aging. The improvement in the resistance to aging of SBS PMA mixtures was also confirmed by the binder morphology analysis (section 4.2.3) and binder-aggregate adhesion properties (section 5.2.1.6). 5.2.1.3 Resilient and Equivalent Moduli The resilient modulus is an essential measure of the elastic response stiffness of asphalt mixtures subjected to cyclic loading. It is an based on the instantaneous strain 175 under repeated load and is defined as the ratio of the cyclic stress to the instantaneous elastic strain (40, 74). On the other hand, the equivalent modulus (which reflects the stiffness of the asphalt mixtures) is obtained by calculating the slope of the linear part of the load-deformation curve (the curve remains almost linear up to half the maximum load at failure). Higher equivalent modulus indicates higher deformation resistance (51, 7). The average values of the resilient and equivalent moduli (at 77°F) of ACS and AC 1 0 mixtures modified with the SBS polymer were normalized relative to the average values of the resilient and equivalent moduli of AC5 and AC10 mixtures processed under SBS mixing conditions, respectively. The normalized data were than plotted as a fimction of polymer content as shown in Figure 5.5. The actual modulus values are listed in Table 4.5. Examination of Figure 5.5 and the data in Table 5.3 indicate that: 1. The normalized resilient moduli of processed mixtures show increases from 38 to 52 percent with respect to the straight ones. The increases are due to the short-term aging of the asphalt mixtures that occurred during processing of the binder at 350°F for a period of two hours and subsequently mixing with aggregates at the same temperature. 2. There is no significant difference between the normalized resilient modulus values of SBS PMA mixtures and the processed mixtures. 3. The normalized equivalent moduli of SBS-PMA mixtures show considerable improvement over the straight and processed mixtures. At the optimum 5 percent polymer content, the increases are 2.17 and 1.60 times the processed values for AC5 and AC10 mixtures modified with SBS polymer, respectively. 176 E AC5 SBS PMA Mixtures E AC10 SBS PMA Mixtures 1.50 125 . 0. 2:35: 23:33— beam—«ESZ Polymer Content (%) a) Normalized Resilient Modulus. E AC5 SBS PMA Mixtures AC10 SBS PMA Mi. 2.5 2.0 ~L< .5 ‘T’""""” 2:35: u=o_«>_=vfl toga—Eel Processed Straight ) % Polymer Content ( b) Normalized Equivalent Modulus. Figure 5.5 Normalized resilient and equivalent moduli of SBS PMA mixtures as a function of polymer content at 77°F. 177 Table 5.3 Resilient and equivalent moduli of SBS PMA mixtures at 77°F. Asphalt Modulus Cement 32 339 756 Resilient Modulus (Pounds/Inch”) A .= U = Q U} '1: = 5 O 9.4 V U) 3 = 1: g H = d.) '3 .2 = c- [:1 NA Note: Shaded cells indicate that not test was conducted. 178 Since the applied cyclic load is the same, the second observation implies that the elastic strain remains almost the same regardless of polymer addition. The third observation indicates that with the addition of the SBS polymer, the total strain of the SBS-PMA mixture decreases significantly. Since the total strain is the sum of the elastic, Viscoelastic and plastic strains, one can conclude that the plastic strain of the SBS PMA mixtures decreased considerably. Similar results were obtained for SBS PMA mixtures tested at temperature of 140°F, and for the aged mixtures tested at 77 and 140°F. It should be noted here that at temperature of 140°F, the deformations of the straight and processed mixtures were considerably high and therefore the samples failed before the 50th load cycle was reached. Hence, the resilient moduli were not obtained for these mixtures. Moreover, at test temperature of 23°F, no significant difference was observed in the moduli values of both unaged and aged SBS PMA mixtures. In order to examine the effect of aging on the equivalent moduli of straight and SBS PMA mixtures, the equivalent moduli of aged mixtures were normalized relative to the equivalent modulus of unaged mixtures. The normalized data was then plotted as a function of polymer content as shown in Figure 5.6. The actual modulus values of triplicate samples of aged mixtures are listed in Table 5.2. Examination of Figure 5.6 and the data in Table 5.2 reveals that: 1. The average equivalent moduli show increases of about 22 and 30 percent for aged AC5 and AC10 straight mixtures, respectively. 2. The average equivalent moduli .of aged AC5 & AC10 processed mixtures exhibit considerable increases of about 34 and 64 percent, respectively. 179 .w w _ _ _ _ _ TII L m, _ m. x .l. .1 _ M M_ _ WW tn: Dist IIIII P S_ S B_ m s" 5 w_ C C" A A_ 11.3 $4: > _ _ . _ A _ _ _ _ _ _ _ _ _ Tll|_lllll_ llllllllllllllllllllllllllll _ _ _ _ _ _ _ _ _ _ _ _ _ _ t---w---u ............................ _ ‘ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . . _ _ _ i— L! Ly _ H _ — _ a A A a ~ _ .— 5 0 5 0 5 O 5 0 5 0 2. 0. 7. 5. 2. 0. 7. 5. 2 O. 2 2 .l. 1 .I. .l. 0 0 0 0 3.352 .3925 8 955—3— mun—=52 .83.. ..e 2:352 2.9—«255m 103—«Ecol Processed Straight Polymer Content (%) ivalent moduli of straight, processed and f equ 77°F. aglng 0 Figure 5.6 Effect of the oven IantUI'eS at SBS PMA 180 3. The percent increases in the equivalent moduli of aged SBS PMA relative to the unaged mixtures decrease as the polymer content is increased and exhibit maximum decreases at five-percent polymer content. 4. At polymer content of seven-percent, the increases in the equivalent moduli due to aging are about 45 and 120 percent for AC5 and AC10 mixtures modified with SBS polymer, respectively. The above observations are similar to the observations obtained for tensile and compressive strengths, thus further confirming that the asphalt binders modified with SBS polymer system substantially reduce the aging of the asphalt concrete mixtures. The five-percent polymer content shows an optimum improvement in the resistance to aging, however the seven-percent polymer content increases the effect of aging. 5.2.1.4 Plastic Deformation Characteristics The improvement in the resistance to plastic deformation with polymer modification can further be confirmed from Figure 5.7. This figure depicts the accumulation of horizontal plastic deformation (HPD) and vertical plastic deformation (V PD) of AC5-SBS PMA mixtures as a function of the number of load cycles at 77°F. It can be seen that the required number of load cycles to accumulate given values of HPD and VPD in the SBS-PMA mixtures is much higher than the straight and processed mixtures. Further, the number of load cycles to accumulate any value of plastic deformation increases as the polymer content is increased until the 5 percent optimum polymer content is reached. Similarly for any number of load cycles the accumulation of plastic deformation is substantially lower for the SBS PMA mixtures than the straight 181 H M “HUN 3 .III WSS [33.11: nu qoD So. o/Vo/ 047 Illtl|llll '|lll llll'll. Illlll llll'll 'l'll + O‘VoStraight + 3%SBS —-— 5%SBS Tllllllllllll Yl'llllvllnlrall Illill"-| YIIIlIII'lIIIII vl“||vl|l|‘l vll'lll'llllul lvlllllviillll VIIIIIII'IIII IIIIIIIIIIIII' villlliulllll O. 55 Anna 5:355 came—m REST—em he acts—5:33.. 1,000,000 100,000 1 ,000 100 Number of Cycles a) Horizontal Plastic Deformation (HPD). :2: ADE acts—FEE: 9:35 .3558» he =33_=E=uo< u--,- -nuunuuuunuuuununuumuuuu rm HHHHHHHTHHHHHHHHHH e c S Sui: uuuuuuuuuuuuuuuuuuuu ._ m B B 5.. a “::: .......... mu M .../3}“ ..... r- I 14. fl u 1t h .g m. S S ..u B B d So So V V o/ 0 3 5 qui I In IIIIIUI? lllll jll'l Huuuunnuuumnuuuuuuuummuunniuwfl llllllllll rill-llllllllrll'llllrlll _ HHHHHHHHHHHHHHHHHHHHH ._ .......... T-------Li--!-- __ lllll filli'llllllilllrlllllll ‘H . .............. ..... 5, 2 O 8 6 4 2 J J O. O. 0. O. O 0 0 O 0 0 100,000 1,000,000 10,000 Number of Cycles (N) 1,000 a) Vertical Plastic Deformation (VPD). 100 Figure 5.7 Accumulation of horzontal and vertical plastic deformations (HPD&VPD) of AC5 SBS PMA mixtures as a fimction of the number of load cycles at 77°F. 182 ones. Since, the accumulation of the HPD and VPD are responsible for the fatigue cracking and rutting of asphalt mixtures, respectively; the increased resistance to plastic deformation considerably increases the fatigue lives and resistance to the rutting (see, section 4.4). Similar observation can also be made by simply examining the rate of accumulation of HPD and VPD shown in Figure 5.8. The deformation rates decrease as the polymer content is increased and exhibit significantly lower values at polymer content of five-percent as compared to the straight and processed mixtures. Stated differently, the magnitude of the plastic (permanent) deformation per load cycle, which represents the rate of permanent damage delivered to the sample decreases with increasing polymer content. This permanent deformation damage is ultimately responsible for rutting and fatigue cracking. Higher deformation rates cause shorter and lower fatigue life and higher rut depth. Similar results were also obtained for aged and unaged AC10 and AC20 mixtures modified with SBS polymer system at test temperature of 77°F and AC 1 O SBS PMA mixtures at 23°F. The higher number of load cycles to develop the same amount of plastic deformation and the almost constant resilient modulus (see, section 4.3.1.3) indicate that: 1. The SBS polymer system causes a decrease in the energy stored in the samples due to plastic deformation. 2. The resiliency (ability to expand and contract) of the asphalt mixtures is not affected by the addition of the SBS polymer. Recall that the deformations of straight and processed mixtures at 140°F were substantially high and the samples failed before the 50‘h load cycle was reached. Hence, the 183 -6 Defamation Rate (inch/cycle) x 10 100 p—a O .3 fl 0.0 l 0 0 1 2 3 4 5 6 7 Straight Processed Polymer Content (%) Figure 5.8 Horizontal and vertical plastic deformation (HPD & VPD) rates of AC5-SBS PMA mixtures as a function of polymer content at 77°F. 184 resilient modulus and plastic deformation data were not obtained for these mixtures. However, AC10 mixtures modified with five-percent SBS polymer content sustained an average of 175 load cycles. This result indicates that even at high temperature, the resistance of SBS PMA mixtures to plastic deformation is higher than the resistance of straight and processed mixtures. 5.2.1.5 Binder-Aggregate Adhesion Properties Failure between asphalt cement and aggregate may occur in two mechanisms, adhesion and cohesion. Adhesive failure occurs when the bond between the asphalt binder and the aggregate is broken. Hence, the aggregate surfaces along the failure surface exhibits either no or little asphalt cement. A cohesive failure occurs within the asphalt binder and the aggregates on both sides of the failure surface are covered with asphalt cement. The Composite Materials and Structural Center at MSU conducted lap- shear tests and ESEM tensile tests to investigate the adhesive properties of the straight, processed and PMA binders. The results of the lap-shear test give a measure of either the adhesive bond strength and stiffness or the adhesive fracture toughness between flat parallel plates having an overlap adhesive joint when pulled apart under tensile loading. The lap-shear fracture toughness was obtained by calculating the area under the load- deflection curve at failure, which is the indication of the strain energy required for complete failure of the sample. The binder-aggregate adhesion properties were investigated at moderate temperatures (68 and 77°F) and low temperatures (14, 23 and 32°F). The analysis and discussion of the results are presented below. 185 a) Moderate Temperatures (68 and 77°F) Figure 5.9 depicts lap-shear strength and the fracture toughness of AC5-SBS PMA binders as a function of polymer content at 68°F. It can be seen from the figure that the lap- shear strength decreases at low polymer contents and increases at high polymer contents. On the other hand, the lap shear-fracture toughness initially shows a slight decrease at the two-percent polymer content and then increases significantly with the increase in polymer content. At low polymer contents, the polymer absorbs the low molecular weight oil phase thus reducing the wettability (surface wetting ability) of the binder. At that time the effect of the reduction in surface wetting of the binder is more pronounced than the strength. However, when the polymer content is increased the strength of the binder eventually increases (66). The improvement in the lap-shear fracture toughness indicates that the strain energy for complete failure of the sample has increased. This suggests that higher amount of external work is required for complete failure of the sample. Hence, the SBS-PMA mixtures require more number of load cycles (external work) for failure of the sample as compared to the straight mixtures. Visual inspection of the failed Marshall size samples at 77°F revealed that the SBS PMA mixtures showed strands of asphalt-polymer blend bridging the crack in the sample. No such strands were observed for the straight and processed asphalt mixtures. A study conducted by E. Eugene Shin et a1. (75) at MSU on the SBS-PMA mixtures, using the ESEM showed that the number of fibrils formed in AC5 and AC10 mixtures modified with the SBS polymer are considerably higher than straight asphalt mixtures. Figure 5.10 depicts typical ESEM micrographs of AC10-2%SBS PMA mixtures at 68°F. It was found that the fibrils were longer and thinner for the SBS-PMA mixtures than straight ones and the 186 387%ch oc— x “859.3. 8522,.— .325-an O 0 0 0 0 0 0 0 0 0 0 1 8 6 4 2 0 h _ h _ d _ d _ . _ _ _ _ . _ _ . _ _ _ _ _ S _ _ . fl. _ _ _ _ . .m. _ L ....... - We. . _ _ _ _ n. _ _ . . _ Ma m. . . _ _ . n _ _ _ m m. _ _ S F. _ _ 1+ IT IIIIII _ IIIII All! I. lllllll r w w. _ _ h h. . _ S S. _ _ . . _ _ _ p p. _ _ m m. _ _ _ . _ o A. _ _ _ _ _ iiiiiii .IIIIIII_sIII....I_ ‘Jtllllti _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ . _ . II ... III_II 1‘ T IIIIII _ IIIIIII _II I ... II _ Illlfi _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ p _ _ a 4 «d A 0 0 0 0 0 0 5 4 3 2 1 3.555 saw—3.5m Lao—.mfiad Processed Polymer Content (%) Figure 5.9 The lap-shear strength and fracture toughness of AC5-SBS PMA binders as a function of polymer content at 68°F. 187 x ' n) .\(‘10-2"/uS|§S I’\l.\ Binder ‘1 '«E 'I'clnpenllure = 68'1“ Displacement = 0.001 inrh . 5 1?. a: ‘1 LR I00 rim "' T‘A‘“ F.‘-‘+.~’uEu'~LW’ 3. . ‘” » ‘4 1 I!) .\('l0-2"A.SBS l’\1.\ Binder 'l‘cnipcrnturc = 08"]? Displacement : 0.004 inch . 'fi 100 ll!" Figure 5.10 ESEM micrographs showing strands in AC10-2%SBS PMA binders for various displacements at 68°F. 188 locations at which they were formed were areas with good adhesion and favorable geometry relative to the direction of crack propagation. The elongation at break was 50pm for AC5 straight asphalt mixtures while AC5 and AC10 modified with SBS polymer showed fibrils which were stable even at elongation up to 130nm and 90um, respectively ( 75). Since the length and number of the fibrils of SBS PMA mixtures were higher than the straight one, these mixtures absorb more energy before failure. Moreover, it was observed that upon complete failure of the sample the fibrils stretch back in an elastic mode. The above observations further confirm that the mixtures modified with the SBS polymer system require more energy to failure relative to straight ones. The lap-shear results also showed improvement in the resistance to aging of SBS PMA binders. The samples for lap-shear test were aged using the rolling thin film oven (RTFO) and Pressure Aging Vessel (PAV). The RTFO simulates binder aging during asphalt mix production and construction of pavement. While, the PAV simulates the aging of the binder during the service life of the asphalt concrete pavement. Figure 5.11 depicts the normalized lap-shear stiffness and toughness of aged AC10-processed and AC10-SBS PMA binders as a function of the polymer content at 68°F (the stiffness and toughness data of PMA binders were normalized relative to the stiffness and toughness values of processed binders). Examination of Figure 5.11 indicates that: 1. At lower polymer content (less than four percent), the lap-shear toughness of the aged AC10-SBS PMA binder are lower than the aged processed binder. This is because of increases in the binder stiffness. The binder stiffness increases by approximately 50 percent at three-percent polymer content. This indicates that at 189 . q JI 4| . _ _ _ ...v 5555 ..5 5 u. 5 .5555 5.. 5.. .5 u. . 55.. 555 .w. 5 A5 . ...5 . . . . . . 5 5.. 5 . .. m. seas... “$5.5... ... .....a... are... 55“.}... 5 ..5. In“... 95%.5v555vvv.55 v. Mm.“ “.....5fi55555 durum“. 53.955559“. .. 5.55.55 5A5 . .n5...5.nm“m55555 5“...“ 555.W55555“ ”.... ... 55.52.55 5. .v ..."... 553.. .5...“ .55..5.. .55A5.. 555 5555... 55.3595 55. 5.. I; .Ial'l. E Lap-Shear Stiffness 2.0 .L _ _ _ _ _ ILap-Shear Fracture Toughness _ _ _ 2.5 mmocawghogfiaum 23 29:55 5.3.—mums..— teas—55oz O Processed Polymer Content (%) iffness and fracture toughness of aged AC10-SBS 190 PMA binders as fimction of polymer content at 68°F. Figure 5.11 Normalized Lap-shear st low polymer content (up to three-percent) the aged binder exhibited brittle behavior. 2. At polymer contents higher than four-percent, significant increases in the Lap- shear toughness and considerable decreases in the lap-shear stiffness relative to the processed binder can be seen. This implies that the aged AC10-SBS PMA binder can sustain higher deformation (at higher polymer content), exhibiting ductile mode of failure. Another observation was that the lap-shear stiffness of the aged binder increased approximately 30 times relative to the unaged binder at three-percent polymer content. However, at polymer contents of 5 and 6 percent the lap-shear stiffnesses of aged binders were only twice the unaged ones. This decrease in stiffness of the aged binders at polymer content higher than three-percent indicates that the effect of aging is reduced with the addition of SBS polymer. Moreover it was found that the lap-shear strength at five-percent polymer content was significantly higher than the processed one. Similar results were also obtained for aged AC5-SBS PMA mixtures. These observations imply that the asphalt mixtures modified with the SBS polymer system considerably improve the resistance to aging of the asphalt mixtures. b) Low Temperatures (14, 23 and 32°F) The resistance to low-temperature cracking of asphalt concrete mixtures is a function of the bond (adhesion) between the binder and the aggregate and the cohesion of the asphalt binder. The binder-aggregate bond plays an important role in failure and fracture of the asphalt concrete at low temperatures. Therefore a thorough analyses and 191 examination of the bond strength and fracture surface of the failed samples can provide an insight into the low temperature cracking potential of the asphalt concrete mixtures. Visual inspection of fractured surfaces of Marshall sample revealed that the failure surfaces exhibited both cohesive and adhesive failures and are temperature dependent. Adhesion failure is breaking of the bond between the binder and the aggregate surfaces. Cohesion failure occurs when the asphalt binder cracks. At room temperature (77°F), the straight and PMA mixtures exhibited both cohesive and adhesive failures. However, at low temperature (23°F) a third failure mechanism was observed, that is, the PMA mixtures showed that the cracks were propagated through the binder as well as some aggregates. Figure 5.12 illustrates typical fracture surfaces of the straight and PMA mixtures at 23°F. Sections AA and BB show typical fractured surfaces of a failed Marshall sample of a straight asphalt mixture. Sections CC and DD indicate typical fractured surfaces of a PMA mixture. The broken aggregates can be seen in light shades of gray and white color. The black shade represents either asphalt cement or aggregates coated with asphalt cement. Examination of Figure 5.12 reveals that: 1. Only few coarse aggregates are broken for the straight mixtures (sections AA and BB). It should be noted that the MDOT specification allows 10 to 15 percent of soft aggregates in the asphalt mixtures. By close examination of the failure surfaces, it was found that all broken aggregates were of the soft aggregate. 2. The straight asphalt mixtures exhibited very little asphalt cement on the surface of the unbroken aggregate indicating adhesive failure. 3. For PMA mixtures, a large quantity of hard and soft coarse aggregates were broken along with the binder (sections CC and DD). This implies that at low 192 .38 an 3ng S)?“ was Ewfibm Emma 05 .«o moo—«Em 0585.“ E29? .2 .m oSwE DU 5.88 Encouww< unamoewwxx can: a. tom 5on 9.552 :23... 858: bar...— 3 mm 55% << cocoow d ...... . 295m :25: 8:... waxwouwwxx tom 5on E0500 zap—Q35 =3 . m :2 193 temperature, the strength of the polymer-asphalt binder becomes either equal to or higher than the strength of most coarse aggregate. The above observations indicate that the adhesive strength of PMA binders is significantly higher than that of the straight binders. The improvement in the adhesive properties of SBS PMA mixtures at low temperature were further confirmed from the lap-shear test results. The lap-shear strength and fracture toughness of AC10-SBS PMA binders obtained at 14 and 32°F were normalized with respect to the lap-shear strength and toughness of the processed mixtures. The results are shown in Figure 5.13. Examination of the figure indicates that: 1. At 32°F (see figure 5.13a), the PMA binders show lower adhesive strength and higher toughness than the processed binder. This observation implies that PMA binders sustain higher strains (more ductile behavior) at failure than the processed binders. The reason is that at 32°F, the PMA binders have lower freezing temperatures than that of the processed binders. For example, the glass transition temperatures of straight AC5 and AC10 asphalt ranges from 5 to 14°F however, the SBS polymer shows —76 to —1 36°F. Therefore upon addition of the polymer in asphalt, the blend’s glass transition temperature becomes significantly less than the straight asphalt (76). Hence, the straight and processed binders act more like brittle solid than PMA binders. The microscopic analyses of fracture surfaces of the lap-shear test specimens indicated that the processed mixtures exhibited adhesive failure, however, the SBS PMA mixtures showed cohesive failure (66). This indicates that the failure 194 Z Lap-Shear Strength E Lap-Shear Toughness 0. 2 1.5.------———-- moi—ugh. ..5 5.58 325-93 .3152 Processed Polymer Content (%) a) Temperature = 32°F. {21 Lap-Shear Strength Lap-Shear Toughness 30 2.5. 2.0—~——~——~— 15+—~——~— 10»- 0.5—»~ -- 00. 829.5. a...“ nanny—«m 5:25-95 cog—«Eel 0 Processed Polymer Content (%) b) Temperature = 14°F. Figure 5.13 The normalized lap—shear strength and fracture toughness of AC10-SBS PMA mixtures as a function of polymer content at 32 and 14°F. 195 mode changed from adhesive to cohesive with the addition of SBS polymer in asphalt cement representing improved adhesive properties. Thus, the data in Figure 5.13a and observations of the fractured surfaces indicate that PMA mixtures are more ductile and less solid, and they have lower adhesive strength and higher toughness than straight mixtures. The implication of this significant observation on pavement performance is that PMA mixtures have cold temperature cracking potential than straight mixtures. Figure 5.13b shows that, at 14°F, the adhesive strength and the toughness of PMA mixtures are higher than those of processed mixtures are and they increase with increasing polymer. At the three-percent optimum polymer content the lap-shear strength and toughness are 2.0 and 2.7 times higher than those of the the processed mixtures. It can also be seen that as the polymer content increases above the 3 percent level, both strength and toughness decrease. The observation obtained from the lap-shear test results (strength, toughness and analysis of fractured surfaces) at low temperatures of 14 and 32°F further indicate that: 1. The optimum polymer content of AC10-SBS PMA mixtures decreases from five to three-percent as the temperature decreases from 140 to 14°F. The failure mode of SBS PMA mixture changes from cohesive to adhesive as the temperature decrease from 32°F to 14°F. This change causes an improvement in the low temperature properties. Figure 5.14 depicts two typical ESEM micrographs that were taken sequentially during the failure process of AC10 mixture modified with 2 percent SBS polymer content at tested at temperature of 32°F. Examination of the micrographs indicate that: 196 n) .\('Ill-2”i..\BS I’\l.\ .\li\luI'L‘ 'l cmpcrnturc = 32"1’ 500 11m - h) AFN—204.808 l’.\l \ .\li\turc "‘ ,. n ~ lcnlpcrnturc = 32 l‘ 200 um Figure 5.14 ESEM micrographs showing failure processes and fibrils of AC10 mixture modified with 2 percent SBS polymer content at 32°F. 197 l. The cracks were seen to propagate both along the interface (between the binder and the aggregate) and through the aggregates. The exposed aggregate surface, which was crazed by the crack, reveals some asphalt residues. 2. The formation of fibril along the failure surface. Although the number of fibrils was not as high as that at room temperature, it is higher than that of the straight binder. These two observations indicate better adhesion properties of the SBS PMA binders. 5.2.2 Other Polymer Systems The other polymer systems (SEBS, SBR, EAM and CRM) used in this study also showed significant improvement in the structural and engineering properties of PMA mixtures. The level of the improvement however varied from one polymer type to another. To avoid unnecessary repetitions, the structural and engineering properties of PMA mixtures at the optimum polymer content for each polymer type were selected and compared. The results of such comparisons are presented below. 5.2.3 Comparison of the Structural and Engineering Properties of PMA Mixtures 5.2.3.1 Comparison of the Tensile and Compressive Strengths Table 5.4 provides a list of the average tensile and compressive strengths of straight, processed and PMA mixtures at their respective optimum polymer contents. In order to investigate the effect of processing conditions (mixing temperature and time) on the strengths of the mixtures, the average tensile and compressive strengths of the processed mixtures were normalized relative to the straight ones. Figure 5.15 depicts the 198 6326.80 we? «m3 0: $5 8385 $8 33m 682 2.0 8m 0% N: E. E E. 9:. e8... 3 m S :5 2.. 5mm 2. w? Sm w? .2 So< M m m .l. J 1 em 9 8m 03 m2 me. o: E. 0:. RN 3.. u. m. 0: m2 .5 SN 5 03 5 § 33. N. L ) J EN 5 2: SN :1 n: E w... So< .m. m m a: mo a E 8. 5; NE 3 3.. E250 ENE—3. S tangent Eu§5<~ 50233.; m “5333.; m 5338.5 33395 Emihm h: e :0 :6 .5:— c at a e c m c ... Ca. . u ..— 520 25— E; mum—m mam 3.; taro.— , mo: 3 use 3528 5:53 Sago Bean: :2: a 3:9 8.53:8 <35 use 8368a .2325 we flawfibm 033958 was 0:82 06 mo baa—gm Ym 2%... 199 21 AC5 Mixtures AC 1 0 Mixtires Normalized Tensile Strength 0.0 - " Straight SBR/CRM SBS/SEBS EAM Processing temperature(°F) 275 350 350 380 Processing time (hour) 0 0.5 2 2 Mixing Conditions (Temperature; °F & Time; hour) Figure 5.15 The normalized tensile strength of straight and processed mixtures as a function of the mixing conditions (temperature & time) of various polymer systems. 200 effects of the processing conditions of the various polymer systems on the normalized tensile strength. It can be seen from the figure that the normalized tensile strength increases as the mixing temperature or time is increased. This increase is due to aging of the mixtures during processing of binders and subsequently mixing with aggregates. The compressive strengths of straight and processed mixtures showed exactly the same trend as that shown in Figure 5.15. The tensile strengths of PMA mixtures at the optimum polymer contents were normalized with respect to the tensile strengths of straight mixtures as shown in Figure 5.16. Examination of the figure reveals that: l. The EAM PMA mixtures exhibit 2.4 to 2.7 fold increases in the tensile strength values relative to the straight mixtures. 2. The SBS, SEBS and CRM PMA mixtures exhibit similar increases in the tensile strength values of about 1.90 to 2.20 times the straight mixtures. 3. The SBR PMA mixtures show approximately 30 to 40 percent higher tensile strengths than the straight ones. It should be noted here, a part of the increases in the tensile strengths is due to short-term aging (processing) of the mixtures. For the SEBS PMA mixtures the short-term aging causes about 38 to 44 percent increases relative to the straight mixtures. Processing of the EAM mixtures causes increases of about 2.1 to 2.4 times the straight ones (see, Figure 5.15). This indicates that the improvements due to polymer modification for the SBS and SEBS polymer systems are higher than the other polymer systems. The improvements are due to the increases in the binder-aggregate adhesion properties. The SBS and SEBS 201 saw—.25 9:238 BEEF—oz i sw mm fimm ..... pm. lam ..... _ _ ..... AA “ n ma U u . , u u _ 4. n f + W _ 0 5. O 5 O. 5 0. 3 2 2 .I. .l. 0 0 l-2%EAM 10%CRM 5%SEBS 3%SBR 5%SBS Straight Polymer Type Figure 5.16 The tensile strength of PMA mixtures at the optimum polymer contents normalized relative to the straight mixtures. 202 PMA binders show better adhesive behavior relative to the other polymer systems (see, section 5.2.3.4). The improvement in the tensile strength of the SBR-PMA mixtures is significantly less than the improvement in the other polymer systems. This can be explained by the difficulties encountered during the processing and mixing of SBR PMA mixtures. The mixing procedure of SBR polymer with asphalt binder was established during the study of the chemical and thermodynamic properties of polymer modified asphalt (65). The procedure calls for optimum mixing time and temperature of l/2-hour and 350°F, respectively. During mixing the SBR-asphalt binders with aggregates, it was visually noticed that the SBR-asphalt blend exhibited a strainy structure, and it was not quite homogenous. Moreover, a difficulty of mixing with aggregates was experienced at four and five percent SBR polymer contents. Although the mixing temperature was 350°F, the SBR- asphalt blend was very sticky. The mixing of the SBR polymer modified binder with aggregate was somewhat difficult at the SBR polymer content of four-percent. Moreover, during mixing the SBR modified binders tended to stick to the mixing impeller along with the fine aggregates, which caused some segregation in the mix. The strainy structure of the binder was also confirmed by the polymer-phase morphology analyses conducted by LSM and ESEM (see, section 4.2.3). Similar results were obtained for all mixtures tested at temperature of 140°F, and for the aged mixtures tested at 77 and 140°F. However, no significant difference was observed for both unaged and aged PMA mixtures at test temperature of 23°F. Table 5.5 shows the effect of various polymer systems on the tensile and compressive stress-ratio (stress/strength) of AC5, AC10 and AC20 PMA mixtures at 203 68328 mm? 63 on 35 8865 38 395m ”802 204 ..fi Sod Sod Sod Rod «86 owed N86 owed good 33. a .0 v8.0 awed Sod 39o Sod awed wood 33 RS Su< M 8m ... ... Sod ES 23 om; ES 23 £3 23 $3 mu< o m m. 95o Sod v8.0 306 N85 83 $3 Sod 98¢ 83. m 36¢ Sod Rod 83 owed owed mood god RS 23. W m Sod ES RS 83 Rod 23 Sod 23 83 H3. m... a 2.250 :53: e 98$.me wolnqcufim n .49anch m .flmnlwam m 33”....“ L .e .m ax; «5:50 .5539— Eu :2”.— mmm mmmm mam 2.? .3528 mot. 8 Sancho 638k :05 8 3.39 8528 <33 was $3685 .Ewfibm mo 8.58 868 368588 can £me“ 2: mo gm On 033. 77°F. The data indicates that the stress-ratios of EAM mixtures are the smallest compared to the other polymer systems. The decreases in the stress ratios of the EAM mixtures can mainly be attributed to the increases in strengths due to short-term aging during processing. Such improvements in the stress-ratio are not desirable as it may effect the low temperature cracking potential of the mixtures. The decreases in stress-ratios of the SBS and SEBS polymer systems, on the other hand, are mostly due to polymer modification. The SBR polymer system also show decreases in the stress-ratio relative to the straight ones, however the level of improvement is not as significant as that of SBS and other polymer systems. Since all samples were tested at the same cyclic load, the higher values of strengths significantly decrease the stress-ratio of PMA mixtures. The significant improvements in the tensile and compressive strengths and considerable decreases in the stress-ratios of SBS and SEBS PMA mixtures clearly indicate higher resistance to fatigue cracking and rut potential relative to the straight and other PMA systems. In order to examine the effect of aging on the tensile and compressive strengths of the PMA mixtures, the average tensile and compressive strengths of oven aged mixtures were normalized relative to the average tensile and compressive strengths of the unaged mixtures. For illustration, the normalized data for ACS and AC10 PMA mixtures are shown in Figure 5.17. The average tensile and compressive strengths of triplicate samples of aged mixtures are listed in Table 5.6. Examination of Figure 5.17 and the data in Table 5.6 indicates that: 1. The average tensile strengths of oven aged AC5 and AC10 mixtures are approximately 20 and 25 percent higher than the unaged mixtures. 205 meal—z Ewan—D 8 9533— mun—5&2 cow< he naweobm 9:258 tea—«Ecol 5%SBS 5%SEBS 3%SBR 2%EAM Straight Polymer Type the ght ones mixtur es at Figure 5.17 The tensile strength of oven aged AC10 PMA tl'aJ optimum polymer contents normalized relative to the s 206 .8358 2: no @38qu was 63 o: :2: 38%E m=oo 393m ”802 mmw moo 20¢. 3 S 0 ) n w 80 N2 So< .m m m 0 am. a.” u. 5. am 02 mo< m Ira :N 304.. ) m. u w: NS 23. W m m. w. w c5 mom mm mm_ mm mU< EoEoU «:::—3a «:89. N 382: 838.. 888.. a ME... A: u n— ests—u... U h w n— 6 m .— m Ae\ev :::—BU heath—om e not a c e a SEQ wmmm 99C. ..oEbom mo: 8 98 $838 Babom Sancho 83.0.8 :2: “a 838 85pr <35 coma :26 mo 3&5ch omeanoo can 0:38 2t mo 5885 cm 03.3. 207 2. The SBS, SEBS and EAM PMA mixtures exhibit no significant difference in the tensile strength values after oven aging. However, the SBR PMA mixtures show an average increase of about 15 percent. These observations indicate that the PMA mixtures significantly reduce the effect of aging. The improvement in the resistance to aging of PMA binders were also confirmed by the binder morphology analysis (see, section 4.2.3) and binder-aggregate adhesion properties (see, sections 5.2.1.5 and 5.2.3.4). 5.2.3.2 Comparison of the Resilient and Equivalent Moduli Table 5.7 provides a list of the average resilient and equivalent moduli of straight, processed and PMA mixtures tested at their respective optimum polymer contents and at temperature of 77°F. Figure 5.18 depicts the normalized resilient moduli of the processed mixtures as a function of processing conditions (mixing temperature and time) of the various polymer systems. It can be seen that all mixtures processed under EAM mixing conditions exhibit significant increases in the normalized resilient moduli (65 & 73 percent for AC10 & AC5 mixtures, respectively) as compared to the straight ones. However, the mixtures processed under SBS/SEBS and SBR/CRM mixing conditions show average increases of about 41 and 29 percent increases, respectively. Once again these are mainly due to the short-term aging. Since EAM mixtures were processed for two hours and at higher temperatures than the other polymer systems, the effect of aging is significantly higher. In order to examine the effect of polymer modification on the resilient and equivalent moduli of PMA mixtures, data were normalized with respect to the resilient 208 .8388 05 no 325:8 33 82 o: 35 88qu £3 33am ”302 235m :32 $32 33.2 238 832 Wadi 832 33.: m 3%: 83m. $32 $32 83: 83m 832 9%.; 33: 9.30 Eda m. Ed: 832 2&2: tame 5.5 «8.8 8a.? «2.8 ”3.8 m .. ,. 23mm 839. £39. 023:. 5&9. onus... 83:. e~o< W8 $38 23% 2.3:. 313. £3.02. Stem. $5? $5.53 23% £33 238 Eu< mm 83$ 83% Steam S052 83% Exam 83% mm}; 9.38 mg. M w EoEoU “Egg—mt. S 338:...— 3o§5<~ 6332:.— M 82395 m 333?:— m 323?:— Ewmabm Afiev 25:80 :::—am e a: e e c e c Emu 23. «mm mmmm mam 25. .25.: mo: 3 was $588 Earn—on EEO 958%»: :05 an 888 8528 <33 98 commoooa .Emfibm mo €608 E03233 Ea “£262 ofimo gm 5m 2an 209 Z AC5 Mixtures AC10 Mixtures .N O O I I l I I I I | I I I I I l l I I I I I I | I l I | ___.____|______c Normalized Resilient Modulus 0.00 Straight SBR/CRM SBS/SEBS EAM Processing temperature (”F) 275 350 350 380 Processing time (hour) 0 0.5 2 2 Mixing Conditions (Temperature; °F & Time; hour) Figure 5.18 The normalized resilient moduli of straight and processed mixtures as a function of the mixing conditions (temperature & time) of the various polymer systems. 210 and equivalent moduli of processed mixtures. The normalized data were then plotted as shown in Figure 5.19. Examination of the data in Table 5.7 and Figure 5.19 indicates that: 1. There are no significant increases in the resilient moduli of all PMA mixtures relative to the processed ones. This indicates that the elastic strain remains almost the same for all polymer systems. 2. The equivalent moduli of AC5-SBS systems exhibits the maximum improvement (about 220 percent higher than the processed mixtures) followed by the SEBS (185 percent), SBR (170 percent) and EAM 110 percent) polymer systems 3. The equivalent moduli of AC10 mixtures modified with SBS, SEBS, SBR and CRM polymer systems show similar improvement of about 155 percent. 4. The asphalt mixtures modified with EAM polymer system exhibits no significant increase in the equivalent moduli. The observations indicate that the SBS polymer system shows the maximum and the EAM system shows the minimum improvement in the equivalent moduli amongst all polymer systems used in this study. This improvement trend is similar to that of the G” of PMA mixtures (see, Table 4.2, section 4.2.2). The G” values of SBS PMA binders exhibit the maximum improvement relative to the other polymer systems. This implies that the SBS polymer system has significantly higher resistance to plastic deformation as compared to other polymer systems. 211 Z AC5 Mixtures CS 1.75 1.so~~—-—--—-—4' I l 125——————————+— l l | 33...: 22:3 e§=§¢z 5%SBS 5%SEBS 3%SBR l-2%EAM 10%CRM Processed Polymer Type a) Normaliized Resilient Modulus. i Z AC5 Mixtures AC 1 0 Mixtures 2.50 3.332 2.9—«2:3— uni—.532 5%SBS 5%SEBS 3 %SBR 1-2%EAM 10%CRM Processed Polymer Type b) Normaliized Equivalent Modulus. 212 Figure 5.19 Resileint and equivalent moduli of processed and PMA mixtures normalized with respect to the processed mixtures at 77°F. 5.2.3.3 Comparison of the Plastic Deformation Characteristics Figure 5.20 depicts the accumulation of the horizontal plastic deformation (HPD) of AC 1 O-PMA mixtures as a function of the number of load cycles at 77°F. It can be seen that the PMA mixtures exhibit substantial improvement in the resistance to HPD. For any HPD value, the EAM polymer system is subjected to higher number of load cycles followed by the SBS and SEBS polymer systems. The CRM exhibit the least number of load cycles to any vale of HPD. The HPD rates of PMA mixtures at 77°F exhibit similar trend as shown in Figure 5 .21. It can be seen that the HPD rates of SBS, SEBS and EAM polymer systems are almost the same. However, the SBR and CRM exhibit higher HPD rates. Similar results were obtained at 23 and 140°F. Recall that the HPD rate represents the rate at which the damage (tensile plastic strain) is induced into the sample. Since fatigue cracking is due to the accumulation of the tensile plastic strain, the significantly lower HPD rates of SBS, SEBS and EAM PMA mixtures indicate considerably higher resistance to fatigue and perhaps temperature cracking. These polymer systems show significant improvement in the resistance to the vertical plastic deformation (V PD) indicating higher resistance to rut potential. The improvements in the plastic properties of PMA mixtures are also reflected in the tan?) values (G”/G/) of PMA binders. The tan5 values (at optimum polymer contents) of SBS, SEBS and EAM PMA binders are significantly lower than the SBR and CRM binders (see, Table 4.2, section 4.2.2). For example, the tan5 value of CRM is 2.82 at its optimum content of 10 percent, however it is 1.10 for SBS at polymer content of five-percent. Since lower values of tanfi indicate higher resiliency and lower plasticity, the SBS PMA mixtures are more elastic than the CRM mixtures and have lower fatigue cracking and rut potential. 213 f- Iii... IIIIflWMIImIIMIIJfliwIHIIIlIlIIII IIIqu-iIW-III ' *Yfiwd "T‘— '“fi fir‘fiflTYTT'T—‘T'V‘T‘T’HT I TV __—— T'J "— _T 0.04 -2 - , .22-? m- 8.5 nets—Eamon 382m .3553: 95255.0 100,000 1 ,000,000 000 9 Number of Cycles (N) 10 0 1,00 100 l _' " —2§_I65/:CRKII I ' +~ Straight A—u— 5%SBS A~5%SEBS 4+ 3%SBR —-n—- 2%EAM | L , Figure 5.20 Cumulative of the horizontal plastic deformation (HPD) of AC10-PMA mixtures as function of the number of load cycles at 77°F. 214 Horizontal Plastic Deformation Rate (inch/cycle) x 10" .° 9 O .— 0.001 Straight . 5%SEBS -74 --4 --4 lllllll illliill 3%SBR Polymer Type _____ 2%EAM l 0%CRM Figure 5.21 The horizontal plastic deformation rates of AC10 PMA mixtures at 77°F. 215 Further, the SBS and SEBS polymer systems also exhibit significant improvement in the binder-aggregate adhesive properties as discussed below. 5.2.3.4 Comparison of the Binder-Aggregate Adhesion Properties Several samples of AC10 modified with 2 and 5 percent SEBS polymer contents were examined using the ESEM. The results showed that SEBS modification had a less marked effect on the fracture mechanism of PMA mixtures than SBS modification. The number of fibrils formed was not as numerous and strong compared to the AC10 modified SBS, as shown in Figure 5.22 ( 75). Moreover, the length of the fibrils at break was smaller than SBS PMA mixtures. The length of fibrils at failure of SEBS modified asphalt binders was approximately 200 microns, while SBS modified asphalt binders showed fibril length of 300 microns (77). In comparison with asphalt binder modified with 5 percent SEBS polymer content, the 2 percent SEBS PMA showed a marked reduction in the number of fibrils. The aggregate surface however did not show a predominant adhesive failure at room temperature and the asphalt residue on the aggregate surface showed clear sign of cohesive failure ( 75). This indicates that the adhesion between the asphalt and aggregate increases as the polymer content is increased. The ESEM tensile test micrographs of SBR PMA mixtures showed that the binder morphology underwent a vast change (66). It was observed that adhesive failure between the aggregate and the binder was reduced. On the other hand, cohesive failure along with crack bridging mechanism like high amounts of fibril formation of the binder was dominant. The binder remained soft and could sustain high deformation. The binder’s fibrils were longer and thinner compared to the straight asphalt and firmly connected to 216 RES" 1. Temperature = 68"F h) AC5-2‘VIISEBS PM. Indu Temperature = 68"l9 . Displacement = 0.01 inch Figure 5.22 ESEM micrographs showing fibrils in AC10 mixtures modified with 2 percent SEBS polymer content at 68°F at two magnification rates. 217 two faces of the aggregates as shown in Figure 5.23. Moreover, the entire fibril elongates instead of necking as observed in other PMA binders. Another observation was that the fibril density increased as the polymer content was increased. It was found that the fibrils were more densely populated and are clearly attached to the aggregates at their ends for asphalt binder modified with 5 percent SBR polymer content relative to 1 percent. This indicates enhanced adhesive behavior of SBR PMA binder relative to the straight one. The ESEM micrographs of AC5-processed mixtures and AC5 mixtures modified with 1 and 2 percent EAM polymer contents and subjected to tension are shown in Figure 5.24. The micrographs show that these mixtures have almost the same network and crack morphology. The cracks were observed to form predominantly near the binder-aggregate interface rather than in the binder phase. Moderate fibril formation (crack bridging) was observed across the crack face and the effect of the high binder stiffness can be seen in the coarseness of the asphalt fibril structures. The EAM polymer concentration seems to have no visible effect on the fracture morphology. The fibril density and general crack morphology of EAM PMA mixtures indicate very stiff binder with reduced Viscoelastic behavior. Similar to the other polymer systems the CRM shows increased adhesive behavior. The ESEM micrographs of AC5-CRM PMA are shown in Figure 5.25. It can be seen that the fibrils are clearly attached to the fractured surface. It was observed that the fibrils have considerably high elongation at break (approximately 600 pm) with a diameter of 2 pm. Moreover the fibril density was higher than the straight asphalt EAM PMA binder. The fracture surface appeared to be covered with asphalt, thereby exhibiting cohesive failure and increased adhesive properties. Recall that the ESEM network morphology showed that 218 Figure 5.23 ESEM micrographs showing fibrils of AC5 mixtures modified with SBR polymer and subjected to two displacement levels. 219 c) ..\(‘5-I".,I:. M'Bindci- Displacement: 0.01"” inch Figure 5.24 ESEM micrographs showing crack and fibril morphology in processed AC5 binder and AC5 modified with 1 and 2 percent EAM content. 220 —....~,-. .. .. ., i r .. 3,. 4 :I).\ It)‘!v..(’l{\l ... . l)|\( 5-10‘5‘..('R\l Storage time 2 0 hr I z . , Storage time : 0 hr liumsure : 2min. ' r‘ . 3 . . K ‘ l".\|)IIsIIre :4 min. n "'1". " 'I'"'VJ> .. tl).\('5-l0"m( l{\l Binder 201mm I'LISLQII.‘ Ill” Figure 5.25 ESEM micrographs showing (a, b) the interaction between rubber particles and the asphalt network for two exposure times and mgnification levels and (c, d) the typical fracture morphology in AC5 modified with 10 percent CRM at two magnification levels. 221 the CRM in the asphalt cement is in dispersed form and does not form any network of its own. The rubber particles could be seen adhering to the asphalt surface, which caused on in increase in the adhesion between the binder and the aggregates. Finally a quantitative summary of the changes observed in fracture morphology during ESEM tensile tests of PMA binders is presented in Table 5.8. Table 5.9 shows the lap-shear strengths and fracture toughness of PMA mixtures at 68, 50, 32, 14 and —4°F. Each cell in the table represents the average of two measurements. For each modifier, an optimum polymer content (shown in bold figures values) was determined based on the average lap-shear performance in terms of strength and toughness of PMA binders, especially, emphasizing low temperature performance on fracture toughness. It was found that the effects of polymer modification on the adhesion behavior of asphalt binders were strongly temperature dependent, i.e., a negative effect at temperatures of 32°F and higher and positive effect at temperatures lower than 32°F. However, the degree of improvement varied depending upon modifier types and concentration. In general, both polymer concentration and test temperature dependency of EAM modified binders were similar to those of other PMA binders (SBS, SEBS, and SBR). However, even though the strength of EAM binders are equivalent to or slightly higher than those of others at temperatures from 50°F to 14°F, the value drops down to the lowest at —4°F. This may be partly due to the high glass transition temperature (Tg) and associated brittleness of the therrnoset epoxy on the adhesive strength and to the high temperature processing resulting in short-term aging. The effect of CRM modification on 222 Table 5.8 Quantitative summary of fracture mophology of PMA mixtures. . . Average F'ihn'l F'IEn'l Densily ' . Average Flbnl , . . Mixture Type Diameter at Break Relative to Straight Length (um) 1 (pm) AC5 Straight 50-100 2 + AC5-SBS 70-130 2 ++ AC5-SEBS 70-130 2 ++ AC5-SBR 1000-2000 0.7 +++++ AC5-EAM 70-100 2 + AC5-CRM 600 2 ++ I Higher number of positive (+) sign indicates increasing fibril density. 223 .30138 33 38 on $5 886:— E8 Bumsm .202 TE. 2 New: No.2 .....fi... .. .. 5. EN.” E EN .: ON ON N: N. : aN_.NN on...“ . . cm was New SN . 2 m— doof Sf NNNN . . 2m 33 m9. 3N . 2 2 $35.34.. [$12 on; 2:: . .H. .. N3 :3 SN NmN m RNfi N25 N52 ,. . a... eNN Name we. a ...... 0 v8.2 $33 82: $3 sewage... 8% 8... SN N RNi Non: N52 SE .. ......m..... a. N$d o? N: N .3.»— meleaN 3&2 2.5 5...»... «a? NImm ..INm _ 2 Ahoy c 52.8.2 895—2.; .5855: EDGE .._< 2.8.8.— ouauaaoafioh 8.532 ._ . . «3.83.. anon. .385 um: oswsfl 2: mo $9423 2: com bHoEoow 0388 28 35:38.85 295m 2 .m 2an 235 A typical normalized compliance (C/Co) and total modulus (E) of indirect asphalt samples tested at 68°F are plotted in Figure 5.27a as a function of the number of load repetitions (N). A best-fit equation was obtained expressing the modulus as a function of the number of load repetition. The normalized compliance of the sample was then calculated by simply taking the inverse of the modulus (calculated from the best-fit equation) and dividing the result by the compliance value at the first data point. It can be seen from Figure 5.27a that the modulus value decreases and the normalized compliance increases as the number of load cycles increases. Since the normalized compliance and the crack length are uniquely related, the crack length was calculated using equation 5.13 and plotted as a function of the number of load repetitions as shown in Figure 5.27b. Similarly the measured cumulative HPD was plotted against the number of load repetition as shown in Figure 5.28a. The normalized rate of cumulative HPD was calculated from equation 5.3. Similarly the normalized rate of crack growth was obtained by calculating the incremental slope of the crack with respect to the number of load repetition and then dividing it with the first calculated slope value (2“ data point, i=2). The normalized data was then plotted against the logarithmic value of the number of load repetitions as shown in Figure 5.28b. It can be seen that the normalized rate of cumulative HPD decreases first forms a valley and then starts to increase. Similarly, the normalized rate of crack grth decreases first reaches a valley and then increases. The decrease in the rate of crack growth may be because the crack is being abstracted or shielded by large aggregates in the asphalt concrete mixtures. The crack had to deflect and change its path and propagate around the boundary of the aggregates. This crack deflection is referred to as crack 236 3.8 85.335 3535.52 5 0 5. 0 5. 2 2 1 .l. 0 _ if P . .. _ . . _. . _ l. . .0 . F _ . . . . C _ . _ C . . _ O . M _ _ _ _ . .1 1111.1111.111 61.111171. .ID..11 . _ ,. _ _ m _ _ _ _ O . _ .0 _ . m C _ . . _ . .0 . . A_ mm 1111111.? 11.1111.1 1414.11.11 _ . _ . . . _ . _ M _ . . m _ . _ _ . O o . . _ . T N 1111.11 11111.1111_1111 1. 11 _ _ . _ _ O A . . . _ . _ . . . _ _ _ . . . . . _ _ . _ _ . 1111.1 .1T111_1111_1111.A 11.11111 . _ . . . _ . _ . . . . _ . . _ _ . . . . . . . _ . . _ . . 1111 11.1111.1111.1111_ 11.11111 . . . _ . . _ O . _ _ .A . . . _ _ _ _ . . _ _ . . .0 . _ . . _ 1111.1 11T111.1111.1111.1 1.11111 . . _ _ _ _ . . _ . _ _ . O . . . . . . _ _ . . _ O. _ . . . 1111.1 11.1111.1111_1111.1 11.11111 + . . _ . . _ . . _ . _ . . _ . . O. _ . _ . .0 _ . _ . . .?T t. + . . O O O 0 0 0 0 0 0 O 0 0 O 0 O O: 0., 0.. O: 0.. 0: 0.. 0 0 0 0 0 0 O 4 2 0 00 6 4 2 1 1 1 .2.me 3.352 .35. 50,000 60,000 70,000 80,000 30,000 40,000 10,000 20,000 0 Number of Cycles (N) a) Total modulus and normalized compliance. :0... .3 i=3 V.25 ...“: 30,000 40,000 50,000 60,000 70,000 30,000 Number of Cycles (N) 20,000 10,000 b) Half the crack length. Figure 5.27 Typical plots of the total modulus, normalized compliance and half the crack length as a function of the number of load repetitions at 68°F. 237 T 2 O 0 figmnamov .....ueaawoeo: cums—m 1355.8: o>t£=E=U 0.00 20,000 30,000 40,000 50,000 60,000 70,000 80,000 Number of Cycles (N) 10,000 0 a) Cumulative horizontal plastic deformation (HPD). £55 U.25 ... as. .3152 25258.6 he 83— team—«5.8 Z 1 00,000 10,000 1 ,000 100 Number of Cycles (N) b) Normalized rate of cummulative HPD and crack growth. Figure 5.28 Typical plots of cummulative HPD, normalized rate of the crack growth and cumulative HPD as a function of the number of load repetitions at 68°F. 238 shielding mechanism, which decreases the rate of crack growth. The crack tip shielding mechanisms have also been reported by Suresh (83) in polycrystalline alumina under cyclic loading. The crack growth rates were found to decrease with increasing crack length before arresting. When the crack reaches a critical dimension, it starts to accelerate in a similar manner as that shown in Figure 5.28b. Relative to the asphalt samples, the number of load applications at which the crack starts to accelerate is considered the fatigue life of the asphalt concrete mixture. The stress intensity factor (K) was also calculated and the rate of crack propagation was plotted against the stress intensity factor as shown in Figure 5.29. Similar results were also obtained for PMA mixtures (4-inch diameter and 2.5-inch thick samples) without slot and straight asphalt concrete mixtures (,6-inch diameter and 1.375- inch thick samples) with and without a slot. The rate of crack growth decreases first and then starts to increase, hence violating the Paris’ law, which states that the rate of crack growth increases with the increase in the stress intensity factor and obeys the power law as given below: gag/Mm dN Where A and m are material constants. The decrease in the rate of crack growth is likely due to the many secondary cracks surrounding the main crack. Hence, the crack tip contains many micro-cracks embedded in the plastic zone. The rate of crack propagation and the path it follows depends mainly on the energy balance at the crack tips. Therefore, the energy due to the applied load is utilized by the main crack as well as the micro-cracks. Further, if there is 239 IILL1+IILIILL1rIItIL+L1T III+++1TIm Tillll+114314111111224411 1114141711 LIILL1+11$IILL1TIIIIL+L1T III+++1T11 LECLLIkIILCELL1rI1CCLFLIF ICEFrk1F11 :2... . :_2.. . .:..... :2... . JQJJJIH11fl43441fl11j34441fi 1334fi41114 LECLLIPI1EFCLL1F1ICCLFL1FIICCFHFIF1 :2... . :2... . :2... . :2... . g :2... . :.2.. . .:._... :2... . JD3JJIAI14§34J1fl1133444Ifi 1334fl41W11 :2... . :.2._ . .:._... :2... . :2... . :.2.. _ .:._... :2... . J.41.JI.I4I14.4.J4J11112.21.4417 Ian—1.4041711. :2... . :2.... .:_.... .2_2 .. :2... . :_2.. . .:._... :2... _ .22.. . :_2_. . :2... . :2.... :_2_. . _:._._. :2... . :.2.. . _FprrL.L1 _ ».cu p b ... . _ . he.._. L . LIILL1+11+IILL1T1III +L1T III+++1T14 lIIiil+114IILIIT1IiI +¢1T III+++ITIL LIILL1+IIE+ILLIT1I12 +¢1T 1II+++1T11 LCCLL1k11LFCLL1r11tC kLIF ICCFrF1F1i .:._... :2... . :_.._ . .:..... JJDJJ14114QJJJIfl11fl3 JIfl 1334fiq1711 LCCLLIHIILCCLLIP11tCI LIFIICCFHF1F1L .:._... :2. . . :.2 _ .2_._.. :2... . :2. _ . :2. . :2... . JJDJJIA11433 11fl11 I 41711 :2... . 2.2 . .. . .:..... :2. . . .... J33JJ1411 44411 1334141711 :2... . .... .2_2 .. :2... . ... . :2... _ :2... . .... .2_2 .. :2... . ... . :2... . wF»F_»_ _ ___ p p ._E»h— LIILL1+11 +L1T III+r41T11 LIIli1+II +L1T III+++1T11 LIILL1+11 t +41T III+++1T11 LECLLIF11 L FL1F 1CCFFFIF11 :2... . _. ... .:.2 .. 4334J1fl11 4441111 441fi 1334fl41711 LCCLLIPN\.HCLLIr11ECLHL1FI1CCFPF1F11 :2... . 2.2.. . .:._... :2... . :2... . 2.2.. . .:._... :2... . 433JJ1H11433441fl11fl34441fl11334fi41711 :2... _ :.2.. _ .:._... :2... . :2... . :2... _ :2... _ :2... . ; J33JJ1411.434J11112244411 133414171 :2... . :2.... 2.____ _:.2 _. :2... . :.2.. . .:.2 .. .2..... :2... . :.2.. . .:._... :2... . :2._. . :_2_. . .:._... :2... . 2.... . . . ..rpp . p1 P .rp1r. _ . _..1»L.L, . 2 3 4 5 6 ...... m ...... m m 0. 0. 0. 0. 0. l l. I. l. 1 293...»... .23.... 1,000 100 10 Stress Intensity Factor (K); psi-finch)“2 le sample of asphalt concrete irect tensi C Figure 5.29 A typical K-da/dN curve for ind mixtures. 240 a large amount of plastic deformation, the crack tips become blunt. Hence, the presence of micro-cracks and blunting mechanism at the crack tip reduces the effective stress intensity factor and the rate of crack growth (53, 83). The test data was analyzed and the fatigue lives of all samples listed in Table 5.10 were obtained using the two criteria. Figure 5.30 shows the fatigue life that was calculated using both criteria. It was observed that the fatigue life based on the rate of crack growth criterion yielded an average of 45 percent higher values than the rate of accumulation of HPD criterion. The difference in the fatigue life obtained from the two criteria can be explained by examination of the crack growth criterion. It is based on the theory of elasticity with the assumptions that the material is homogeneous, isotropic elastic solid and there are negligible changes in the material properties during testing. This is not the case for the asphalt samples, which are heterogeneous mix of asphalt cement and fine and coarse aggregates. Further, the criterion is based on two- dimensional crack growth. For cylindrical sample tested in indirect cyclic tensile loading, the crack propagates in three dimensions. Given the crack initiates at the center of the sample, the crack propagates along both the vertical and horizontal diameters as well as along the thickness of the sample as schematically shown in Figure 5.31. The crack length equation only calculates the crack length along the vertical diameter of the sample. This may result in higher fatigue life relative to the cumulative HPD criterion. On the other hand the rate of accumulation of HPD criterion is based on direct measurement of HPD from the test sample. Since the fatigue of cylindrical sample is mainly due to the accumulation of HPD, the rate of accumulation of HPD is a true representation of fatigue crack growth. This criterion produced consistent and repeatable results for the straight, 241 1,000,000 «1444 #14 14<40.mN.N.N NvodNN 03.de 000.0N 000.NwN meN mNmN mNN bo.mN¢m.N m0.mwv.N 00No.o oNv.NvN om0.NNm o00.ov 000.9NN vomN NN mN mNN <2 m0.mNNb. N <2 oww .mvN owN. EM 000. SN mumN NNmN mNN wowwmuw. .N :.m.0.:me.N m 5.3. .N...m..N.. obw .N.,mm ooo.N.N .000 03 mNonNN o0.meo.N m0.mNN.w.m 300.0 omN.mNN womfimm ooo.0N 000.8 36 wme NNoN .mNN o0.vaN m0.mNNN.N. NmNoo 0:.moN www.mmm ooo.mN 000.5. vmé NomN NNoNMNN om00MNN N33 oNN .vN.N nwoomNN N N 50m oMN . r... $5.; vmoomNN tommoUONm mmm0N U< swim M... 3-3.... m. 5.; N93: ooo ....2 NENNN No.5: 2353 :86 cacao NmNd: ooo; : NENNN mmmeau... 8&me 23:? NS: NNNé. NSNNN 89% 832 :NNNNN so? 225 a: 38: 3.2 8% a 295 a one £5 92%.: N83? 3.5.2 .535 , 5.32—:89: 5 3315 35> Egg—U HEN-«Echofl flotaaheon— oNN—=00: 2:300: ob..— . fieflaflwmmofl Noah—0h acme—m 9533 N35. ..5—Nana “Eu—Nah 3 86.6 ..N< Havana uEuoam oNN—Bum NE.“ no.3. ES Nat—cute! owns—.— . . . . .Ne Ban—=2 . 13:3th Nat-outca— egeaoo :« 2.3 247 w0.mNmm.w mo-mNom.v oNoo.0 M304: mmodmv ooo.0NN ooo..00w m0-mNoN.o 8+ SENNN 8.93 8-8: 88.8 :45: 33% 88.88 _ 88.83 8+ 82 NENNN . S 8.8 58.82 83.: 88.8: 83: com ONNN :DNE 8.8.3 383 E88 <2 8S9. 88.8 88.3: N2 NNNN SENNN 8.83 8.33 88.8 9......8N 833. 88.8 832 ..N... 82 NSNNNN 258..” 8-83 :88 “8.3; 55.3... 88.8_ 88.8NN 8. m NNNN :2 NNN mmmxmoNUAN 0088on mmmoNU< oo-mmo.N mo-mNNN.m $23 03.3 N353. oooom ooo.mN QM mNmN mmooNNN oo-mNmo.N mo.meN.m owNoo mafia $0de ooo?u oooJR m2. momN NmooNNN Ewfibm0NU< No-MNoN.m mo-mNmN.m 28.0 $503 3:..me ooo.oN ooo..NN. NNN. womN NmooNNN wodmmw 3.9.: 323 Nmo.NoN oNvfimm ooo.ooN ooo.mN.m 0N4. NcmN onNmNN No-mNNw.N vo-mN.m.o ooooo wdeoN wadmm ooodmN ooodoo mN.m NNmN mNmNmNN mmmgmoNUN‘ No-mNoc.N wo-mNmo.w 295 #3.? NEde 25.3 ooofiom Sam ome EmNmNN 293 £25 a: , . £25 33. 588 2. unwas— 2. :2: as 0293 02950 . .855 5280 5.5-“.5550 9.5—Na..— mN=o> 3.5.20 ..5—.5589: :oNNaESNon 2:352 2:352 85 . eotunflmun ..oENNNeAN 2855 . 3 8.9.5 ..5. E350 9:5on 283m 3820 55,—. 33.89% gut—Nu 2.5—am Nix 2.3. 52 NSEESNN . 5 BANE—:2 Nag—58.50 5258.5: .8288 :w 2.3 248 all mixtures, the stress—ratio decreases considerably. Recall that lower values of stress-ratio indicate lower strains. This implies that the resistance to total strain of SBS PMA mixtures is higher than the straight ones. The resilient and total moduli. The resilient and total moduli of SBS PMA mixtures show similar trend as that of fatigue lives. These two moduli of processed mixtures exhibit substantial increases relative to the straight mixtures and that there are no significant differences in the moduli of processed and SBS PMA mixtures. Since the resilient and total moduli were calculated on the basis of the instantaneous recoverable strain (elastic) and total recoverable strain, respectively, one can conclude that elastic plus Viscoelastic strains remain almost the same regardless of polymer addition, while the total strain of SBS-PMA mixture decreases significantly. Given that the total strain is the sum of elastic, Viscoelastic and plastic strains, one can conclude that the plastic strain of the SBS PMA mixture decreases considerably. Such decreases result in higher fatigue life. Improvements in the resistance to the HPD. Recall that for any load cycle the accumulation of horizontal plastic deformation is substantially lower for the SBS PMA mixtures than the straight ones. Moreover the plastic deformation rates decrease as the polymer content is increased and exhibit significantly lower values at the optimum polymer content. Since the plastic deformation rate represents the rate at which the permanent deformation or damage is induced into the sample the number of load cycles to fatigue cracking is higher and hence the fatigue life. 249 Improvements in the binder-aggregate adhesion properties. At the optimum polymer content, the lap-shear strengths and fracture toughness exhibit considerable improvements. Similarly the ESEM micrographs show significant increases in the number of fibrils firmly attached to the aggregate boundaries of SBS PMA binders as compared to the straight ones. Moreover the fibril can sustain more deformation before failure. This behavior suggests SBS PMA mixtures can absorb higher energy and therefore higher amount of external work is required before failure is reached. Improvements in the rheological properties. At optimmn polymer content the G” values of SBS PMA binder are significantly higher than the straight binders. Since the G” reflects the amount of plastic strain induced into the asphalt binder due to load, the significant decrease in G” value indicates that the plastic strain is considerably lower for SBS PMA binder. Similarly the tan5 values (G”/G’) of SBS- PMA binder decrease substantially with the addition of polymer. Recall that the lower values of the tan5 indicate either more elastic behavior and/or lower plastic behaviour relative to higher values of tan8. The decreases in the tan5 values clearly indicate that the SBS PMA binder develops more resistance to plastic deformation. Similarly, the SBS polymer system, the other polymer systems (SEBS, SBR and EAM) exhibit significant improvements in the fatigue lives of PMA mixtures. However the level of improvement varies for each polymer type. A summary of the average fatigue lives of PMA mixtures for various polymer contents at 77°F is shown in Table 5.12. The data in the table show that the fatigue lives for all polymers increase with the increase in the polymer content and reaches an optimum value, after which the fatigue lives 250 doNNNNE a 5:”— 058 a om: 2&3 55 mSaoNNEN ..D: 325:8 as ea 8 as 888. £8 83m 582 0x0 38:50 BEN—om 8%.“..058... . .. 88.8 8No< «8.2 E... 88w 20... 2.20 am. .ewémmw N8.N 3... we... .....e; 8.8 8% 22.. 23. 88.2 8N B< . . 88.8 80.... 83... . . .. H. N88 8% Eu... mow . 8.3: . .. N. NE... N8.N 3... WE... 82E 88.8 83. 88.82 88.2. 8.8 8ch $3: 83 S u< mmmm 88.8 88.2. N88 Neeem 58w N8.N mo< ”HE , e... 8.8 88.8 80.... N882 o8 mm 88.2 New}: 8% 20¢. mum Seem N88 88.? NeeéN 88w N8.N 82 00885.5 Ewfihm 2: o o EoEWU 09C. BEN—om . :2...... .NNeNN. Na 3558 5830a £553 50 8538 <33 .8 m2:— oswofl 092on 05 .No .N.—«885m < NN .m 033. 251 decrease. One important point should be noted is that asphalt concrete mixtures modified with the EAM polymer did not experience fatigue failure even at one million load cycles (unlimited fatigue life, shown as “U” in Table 5.12). Therefore, to analyze the effect of EAM polymer modification on the fatigue life, the number of cycles to develop 0.0025 inch of HPD for straight, processed and EAM PMA mixtures were obtained and compared. In order to investigate the relative increases in the fatigue lives of PMA mixtures due to polymer modification, the average values of the fatigue lives of the PMA mixtures were normalized with respect to the fatigue lives of the straight mixtures. The normalized data are listed in Table 5.13. It should be noted here that for the EAM PMA mixtures, since the fatigue life is unlimited, the number of cycles to develop 0.0025 inch of HPD is normalized relative to the number of cycles to develop 0.0025 inch of HPD of straight mixtures. For each polymer system the optimum polymer contents is selected to compare the improvements in the fatigue lives of PMA mixtures. Examination of data in Table 5.13 indicates that: 1. The EAM polymer system shows the maximum improvements in the fatigue lives followed by SEBS and SBS polymer systems. 2. The SBR polymer system exhibits the least increases in the fatigue lives of PMA mixtures. 3. The CRM shows no improvement in the fatigue lives. These three observations are similar to those made regarding to the tensile strengths of PMA mixtures (see Table 5.4). For example, at optimum polymer contents, the tensile strength of AC5-EAM PMA mixtures are twice the tensile strength of SBR PMA 252 SU< :2”.— o._ mU< «firm o; ONU< . S 29.. new o.~ o; 84.. Va 3 30¢. Em o; 2 U< mam—m m.~ o._ mU< in A: o~U< 5 o." o5< mam m.~ A: wu< cyanogen Ewihw 2.5. 2.3. c o 25:50 ..oEbom :2.—3. met. an 893:8 cog—om £8.53 8m $538 Emir—um 8 Somme. £39 85an «93 mo 8% Qumran 05 E 33803 2: me 3885 < 2 .m 03$. 253 mixture. Similarly the AC5 mixtures modified with SEBS and SBS polymers exhibit approximately 65 and 50 percent higher strengths, respectively than the AC5-SBR PMA mixtures. Once again, since all PMA mixtures were subjected to the same cyclic load, the substantial increases in the tensile strengths of EAM PMA mixtures cause decreases in the stress-ratios. Hence, the EAM PMA mixtures experience lower plastic strains relative to the other polymer systems. 5.3.3 Rut Potential Pavement rutting is manifested by the accumulation of plastic strain due to repeated traffic loading. Progressive movement of materials occurs either in the asphalt concrete layer or in the underlying layers (base, subbase and roadbed). Rutting of the asphalt layer is due mainly to of further compaction of the asphalt concrete due to traffic and lateral plastic flow of the asphalt cement at high temperature (52, 61). In the laboratory, rut is simulated by tracking the number of load cycles that is required to develop certain levels of vertical plastic deformations (VPD) are obtained from ITCT and plotted as a function of the polymer content. This result provided one of the bases for the relative performance of the straight, processed and the PMA mixtures with respect to rutting. Figure 5.33a depicts the number of load cycles required to develop 0.03-inch of VPD as a function of SBS polymer content at 77°F. It can be seen that the number of load cycles required to develop 0.03 inch of VPD increases considerably as the polymer content is increased until the optimum polymer content is reached. The VPD data of SBS PMA mixtures were normalized with respect to the processed mixtures and plotted in Figure 5.33b. It can be seen that the AC5 and AC10 254 —_—_4 V///.////A I'--|-'|l"'|||lll|ll|l'|l|l‘ Tullllllll'llll'l" Illl'l lllll 0 _ 1 _ u....n.u....u.” C _ “WEE A _ “unfinuuaw _ a _ _ l - l m 5 ......... _ ........... C _ _ _ A _ . _ a . _ _ _ . _ _ k _ . _ . _ _ . _ . . _ _ _ L i b _ 4 4 1 A _ 0 0 O 0 0 0 0 O 0 0 0 0 O 0 O 0 5 0 5 0 5 2 2 l l gun—anew: 25:.— _3_to> he :8. 9295: S 296 he 3:85: Processed Straight Polymer Content (%) a) Number of cycles to develop 0.03 inch of VPD. 15 r r 0 _ _ 1 . _ C . _ mm”. A . _ a — _ .......... _ . _ _ 5 _ . u 4 C lllll _ ::::: L ..... p nnnnn A _ . . a _ _ _ . _ _ . _ _ _ . . _ _ . _ _ . _ _ _ . . _ _ _ _l h h P T 4 A a d 2 9 6 3 0 1 :::-3:80: ogr— Rumto> .8 :85 9:86: 8 2&0 he 338:2 3335.82 Processed Straight Polymer Content (%) b) Normalized number of cycles to develop 0.03 inch of VPD. Figure 5.33 The number of load cycles required to develop 0.3 inch of vertical plastic deformation in the SBS PMA mixtures tested at 77°F. 255 processed mixtures show about 7 and 6 times higher number of load cycles to reach 0.03- inch plastic deformation relative to the straight mixtures. These increases are due to the short-term aging during mixing and processing of the mixtures. Optimum polymer content of five-percent, the improvements due to SBS polymer modification are about 13 and 7 times the processed mixtures, respectively. Moreover a decrease in the resistance to VPD can be seen at seven-percent polymer content. The improvement in the resistance to VPD indicates that the work done by the external load to deve10p the same amount of VPD is significantly higher for the SBS-PMA mixtures than the processed ones. Since the accumulation of VPD is a measure of the rut potential of the asphalt mixtures, the SBS polymer system exhibits substantial improvements in the resistance to rutting at moderate temperatures of 77°F. The improvements in the resistance to rut potential of SBS PMA mixtures are due to: 1. Increases in the compressive strengths. The compressive strengths of the PMA mixtures increase significantly with polymer addition at 77 and 140°F. The compressive strength is one of the fundamental engineering properties, which is used to evaluate the resistance to the rutting of asphalt mixtures. Higher compressive strength means lower compressibility and higher shear resistance, which indicates that the mixture is less susceptible to lateral flow. Hence, the two mechanisms (compression and shear flow) that cause pavement rutting are favorably affected by the addition of SBS polymer. 2. Improvements in the resistance to accumulation of VPD. Recall that for any load cycle the rate of accumulation of VPD is substantially lower for the SBS PMA mixtures than the straight ones. This indicates higher resistance to rut. 256 3. Improvements in the binder-aggregate adhesion properties. The lap-shear strengths and fracture toughness exhibit considerable improvements. Moreover the ESEM micrographs show that fibrils are firmly attached to the aggregate boundaries. This suggests better bond between the aggregates and the binders, resulting in higher resistance to shear deformation and lateral flow of asphalt and improved resistance to rutting potential. 4. Improvements in the rheological properties. At optimum polymer content, the G” values of SBS PMA binder are significantly higher than the straight binder at 77 and 140°F. Moreover the tan5 values of SBS-PMA binder decrease substantially with the addition of polymer. This implies that the SBS-PMA mixtures can recover more deformation than the straight ones and hence exhibit low level of VPD. Another observation is that the viscosity of the binder also increases. Since viscosity represents the resistance to flow, the higher viscosity of the binder indicates high resistance to rutting. Similarly, the other polymer systems (SEBS, SBR, EAM and CRM) exhibit significant improvements in the resistance to rut potential of PMA mixtures. However the level of improvement varies for each polymer type. In order to investigate the increases in the resistance to rut potential of PMA mixtures due to the processing and polymer modification, the number of load cycles required to develop 0.01 inch of VPD were obtained form the test results and the data are listed in Table 5.14. The data were then normalized with respect to the straight mixtures and is shown in Table 5.15. The data in the tables indicate that at optimum polymer contents, the EAM PMA system shows that: 257 oofim coed I». vam . one? a (r! .. I I ... I I7 I m I II ..I I I III I I I I I I II I» I «.I «I I ((511! . I I I 3x.) I.» II I n v Ifiw 35......" I I I I...» II ... I II . as» I I 42......” I I I I I II I II . II I I «... II ”III I I I I I335. I I I II I I ..5 I. .n. III Mm; 853?.5. I ”IInmmméfiwIn/I Wmmwwmm II.» I II. ...».III 5.. II” ImeIIfixw» IIIIIII2. mU< wmflmw o4 o~o< 3 20¢. Eta— og m0< 3 OS... mfiwé 3 23. gm . 2 S 8< ca 3 omu< fin...“ 3. 3 20¢. mum—m fime 04. o._ mU< ed A: omu< 3. 3 on? mam c... o.» R2 338.. was . . . .. . . a“... SEA—cm . . AR; 2.3.50 beam—om “_asnmaw .3538 “swag 05 8 0358 no: 3 $888 hoax—om 32.8.» mo 8538 EA E 95 mo :05 8.0 9:26“. 8 8:5on 83.8 32 mo $983 on“ E 82882 BE. 2 .m 2an 259 1. The maximum increase in the resistance to rut potential followed by the SEBS and SBS polymer systems. However for EAM PMA mixtures, most of the increases are due to the short-term aging during processing (11 to 21 times the straight mixtures). 2. The SBR and CRM polymer systems exhibit the minimum improvement with respect to the resistance to rut potential. This trend is very similar to the compressive strengths of PMA mixtures. For example, at optimum polymer contents, the compressive strength of the AC10-EAM PMA mixture is 75' percent higher than the AC 1 O-SBR PMA mixtures. This indicates that the vertical strain induced by cyclic loading is higher for SBR as compared to EAM polymer system. Another observation is that although the compressive strength of CRM PMA mixtures (at optimum polymer content) is approximately 47 percent higher than the SBR PMA mixtures, but still, the improvement in the resistance to the VPD is higher for SBR polymer system. This may be due to the better binder-aggregate adhesion behavior of SBR polymer system. It was found that the lap-shear strengths and fracture toughness of SBR PMA binders were significantly higher than the CRM PMA binders at room temperature of 68°F (see, Table 5.4), indicating improved adhesion properties 5.3.4 Temperature Cracking Temperature cracking in asphalt concrete pavement occurs due to the following two mechanisms (40, 61). As the pavement temperature decreases, the tensile stress increases due to shrinkage. When the temperature induced tensile stress exceeds the 260 tensile strength of the asphalt concrete, the asphalt concrete pavement cracks. This mechanism is normally referred to as low-temperature cracking. 1. If the tensile stress due to low temperature is less than the tensile strength, the pavement will crack after a large number of temperature cycles (due to the accumulation of plastic strain). This mechanism is called thermal fatigue cracking. The SBS polymer system is selected to illustrate the improvement in the resistance to temperature cracking potential of PMA mixtures. The results and discussion are presented below. The average values of the tensile strength, fracture toughness and equivalent moduli of AC10 straight, processed and SBS PMA mixtures at 23°F are normalized with respect to the tensile strengths, fracture toughness and the equivalent moduli of processed mixtures. The normalized data is shown in Figure 5.34. The actual values of triplicates are listed in Table 5.16. Examination of Figure 5.34 and Table 5.16 indicates that: 1. The values of the equivalent moduli of processed mixtures increase by 21 percent relative to the straight mixtures. This is due to the short-term aging during processing of the binder and subsequently mixing with the aggregates at 350°F. This causes which resulted in a stiffer mix relative to the straight ones. 2. There is no significant difference between the equivalent moduli of processed and the SBS-PMA mixtures. 3. The tensile strengths and fracture toughness exhibit no significant improvement with the addition of SBS polymer. 261 1.4 8. 6. 0 0 5a: tea—«Euoz 0.4 1H , Processed Straight %) Polymer Content ( lEl Tensile Strength m Fracture Toughness Z Equivalent Modulus I fracture toughness and equivalent moduli le strength, i of AC10-SBS PMA mixtures at temperature of 23°F. Figure 5.34 Normalized tens 262 ,., 2.3. ix. . ...k J ...P 5.1“” #4. ”I: . ... ...... . z “5%" new. , ,bxfitmw.” 3,,» a. ,. ...?é. . .. ”Hung, $5., a a” was 83.8 3.5% 38.0 mbm€am $5va ::mem 22va 223m COOKOM :2va com mum 23c Mama :2va 83 Rod 2;: 223m 23 . ::mvm 20.va was 33 832 2 2 2% £3 83 83% 223m :3 om; 85: m E :2qu mmmgm-o_ U< mmmficYS U< wmmgmé—nZQ Sod mmod o2 223m vummoooi wad emod :_.om~ 2; 223m mmm65< mmod $0.3m m3 229% 29o God 331mm 9: 2b; 3m Sod mmé omd mmooflsm wad _mod mmm.mm~ o2 _Nm; 0% coed and mm.~ mmooflsm Ewfibméfifi‘ de owed NNN.~N~ 5: ~34 n3. coed £5 38 Hmoomvm I .N.—c 2.3.5 .53 is. 25 22;..— .s are EB: 53 £9.25 :2: an: 23» e? .355 .252 :::...” 5:25:39 2:35: 895%ch . :fifihm ES..— gun-ammun— 2: :2. an Ear—8w Eo_§_=am 9:522..— ozmmEanU 2.2—uh. ion 33.5“— uEuoam San—mm .55“ gag—.539 . . . 95,—. £2 .momm mo 238%an 58 3 8.528 8888 3:33 we £88 68 fiwqobm 0:38 6865 05 mo @5888 < 2 .m ...—nah 263 The HPD rates of AC5 and AC10 mixtures modified with SBS polymer are normalized with respect to the HPD rates of processed mixtures and plotted as a function of polymer content at 23°F in Figure 5.35. It can be seen that the normalized HPD rate of the AC5-SBS PMA mixtures decrease considerably as the polymer content is increased. Recall that the HPD rate indicates the rate at which the horizontal tensile plastic deformation is accumulated with the application of cyclic load. Lower HPD rate means less accumulation of plastic deformation. Therefore a decrease in the HPD rate at low temperatures implies higher resistance to thermal fatigue cracking due to temperatures cycles. The HPD rates of AC5-SBS PMA mixtures decrease substantially with the increase in the polymer content and show maximum improvement at optimum polymer content of five-percent. On the other hand, the HPD rates of AC10 SBS PMA mixtures exhibit significant decrease at three-percent polymer content and then start to increase indicating a shift in the optimum polymer content at low temperature. Recall that at high and room temperatures the optimum polymer content is five percent. This improvements in the HPD rates of SBS PMA mixtures suggest that higher amount of the external work in the form of temperature cycles is required for the accumulation of HPD and hence increases the resistance to thermal fatigue cracking of asphalt mixtures. The stiffness of aged ACS and AC5 modified with SBS polymer were obtained using the Bending Beam Rheometer (BBR) (63). Recall that the RTFO (which simulates the asphalt binder aging during the manufacturing and construction of asphalt concrete pavement (61)) was used to age the straight and PMA samples. Figure 5.36 depicts the stiffness of the aged AC5 and AC5 modified with SBS polymer as a function of the polymer content at -11°F. It can be seen that at lower polymer contents there is no 264 1.5 = 1 .3 El AC5 SBS PMA Mixtures E a AC10 SBS PMA Mixtures 8 I 0 l G .2 1.0 4 § 3 33 a? 3 a g a. .5 a 3 5 m 0. ‘1 1: 0 .5} E 6 Z 0.0 0 3 4 5 Processed Polymer Content (%) Figure 5.35 Normalized rates of HPD of AC5 and AC10 SBS PMA mixtures as a fimction of polymer content at 23°F. 265 . _ . . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ fl _ _ _ . . _ _ _ . _ . T _ _ _ L _ _ _ _ _ _ _ _ _ _ _ _ . _ _ IIIII _IIIIL rllllrllll.lllli _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ lllll _llllL .Illfllllrl|l|.lllll _ . _ . _ _ _ _ . _ _ _ . _ _ _ . _ lllll 7.31.... 4 wuunuTnlnluluir _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ . . _ _ _ . _ IIIII _IIII IIIA fl IllqllllJllllt _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ _ rnouIIIu unuuquuiiflunnJ lllll _ uuuuu _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . I. _ _ _ _ L _ _ _ _ . _ _ _ _ _ _ _ . _ . _ _ _ _ . _ . _ _ _ _ _ _ . _ n 5 n r l l 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0, 0, 0., 0, 0., 0, 0, 0 0 O 0 0 0 0 7 6 5 4 3 2 1 ...... x9 saga Polymer Content (%) Figure 5.36 The stiffness of AC5-SBS PMA as a function of the polymer content at temperature of -1 1°F. 266 significant difference in the stiffness values. However, at polymer contents higher than four-percent, a substantial decrease in the stiffness can be observed. The decrease in the stiffness values indicates that the AC5-SBS PMA has sustained higher strains level relative to the straight ones. As stated before low temperatures cause the asphalt concrete to contract causing high tensile stresses to develop. The high stresses produce high strains. If the ability of the mixtures to undergo higher strains is small, a crack will occur in the asphalt mixture. Asphalt cements, which have high stiffness at low temperatures, are highly susceptible to temperature cracking because they are less flexible and cannot sustain higher strains. On the other hand, asphalt cements with low stiffness can sustain higher strains at low temperatures and hence perform better with respect to the low temperature cracking (61). This implies that the AC5-SBS PMA mixtures have higher resistance to temperature cracking. The resistance of asphalt concrete mixtures to low-temperature cracking is considerably effected due to the bond between binder and aggregate. The binder- aggregate bond plays an important role in failure and fracture of the asphalt concrete at low temperatures. As stated in section 5.2.1.5 the binder-aggregate adhesion properties of SBS PMA binder exhibit considerable improvement over the straight binders at low temperatures. This implies that the SBS PMA binders could sustain higher deformation before failure as compared to the straight mixtures. Further, microscopic analyses of the fracture surfaces of the lap-shear test specimens indicated that the processed mixtures exhibited adhesive failure, while, the SBS PMA mixtures showed cohesive failure (66). These indicate that the failure mechanism changes from adhesive failure to cohesive failure. Hence the potential for low temperature cracking decreases. 267 Similarly, the other polymer systems (SEBS, SBR, EAM and CRM) exhibit significant improvements in the resistance to temperature cracking potential. However the level of improvement varies for each polymer type. Tables 5.17 through 5.18 provide a list of the tensile strengths, fracture toughness, equivalent modulus and vertical deformation at failure of straight, processed and PMA mixtures at the optimum polymer contents and at 23°F. Examination of the data reveals that there is no significant difference in the tensile strengths and fracture toughness of processed and PMA mixtures at 23°F. However, the AC10-10% CRM and AC20-3%SBR exhibit increases of about 31 percent in the fracture toughness values over the processed mixtures. This is due to slight increases in the vertical deformation at failure and a decrease in the equivalent moduli of mixtures. The other polymer systems show no significant differences in the vertical deformation at failure or in the equivalent moduli. Figure 5.37 illustrates the HPD rates (inch/cycle) of AC10 straight PMA mixtures at test temperatures of 23° F. It can be seen that except for SEBS polymer system, the HPD rates are considerably lower than the straight ones and the AC10 mixtures modified with 2 percent EAM show the lowest HPD rates, followed by SBR and SBS systems. These indicates that polymer modified asphalt mixtures have higher resistance to horizontal plastic deformation at low temperature of 23°F. Further, the AC10 mixtures modified with EAM and SBR polymer systems exhibit significant increases in the lap-shear strengths and fracture toughness at 14 and 32°F (see, Table 5.9). This indicates that the binder-aggregate adhesion properties improve with the polymer modification at low temperatures. These observations implies that the EAM and SBR polymer systems are the most effective polymers in increasing the resistance to low temperature and thermal fatigue cracking of 268 In? .3333 33 $8 on 85 33:2: £8 33m ”20 Z SA 2: mg m: c3 52 c2 o2 of _D e~0< W m M 8 an n .. u- m m3 m2 m2 o3 m3 3; o2 o3 o2 o3 SU< m m M s o5. mow men Sn one Rm wan mmm mmm Nnm 30¢ ) W .4 mmm o3 vmn vmm mmn mun NR Gm N3 2m EU< m N 26:60 ..5—.74 E 328:...— 9812o0, tan5 cannot 281 1,000,000 1 \ z . . -—Power (AC5-SBS) 3 \ Ln v 10,000 9.; \“~ A \ o O :3 O .50 1,000 :2 y = 2.9m"-56 R2 = 0.86 100 10 1.0E-08 1.05-07 1.0E-06 1.0E-05 HPDR at FL (inch/cycle) Figure 6.1 Fatigue life of AC5-SBS PMA mixtures as a fimction of the horizontal plastic deformation rate at failure. 282 1,000,000 1 o AC5-SBS -—- Power (AC5-SBS) 100,000 4 e 3 r”. 53 v 10,000 5.4:» 3 ..J g , 3 c ’ .31? 1,000 :8 Li. Y = 5E-06x4'6433 R2 = 0.65 100 10 60 80 100 120 140 160 Indirect Tensile Strength (ITS); psi Figure 6.2 Fatigue life of AC5-SBS PMA mixtures as a fiinction of indirect tensile strength (ITS). 283 1,000,000 1 I O AC5-SBS —— Power (AC5-SBS) 100,000 4 A . : A E: 10,000 \ é \ “a“ ’ \t .31” 1,000 a _ 4.86 F‘- y—100167.07x R2 = 0.85 100 10 0 2 4 6 8 Tanfi Figure 6.3 Fatigue life of AC5-SBS PMA mixtures as a function of tanfi. 284 100,000 4 4 3 e O 3 10,000 EL 9 AC5-SBS Q 5 ——Poly. (AC5-SBS) Q 5 .29 ‘5 an 1,000 y = -l629.3x2+17157x + 3000 R2 = 0.65 100 Polymer Content (%) Figure 6.4 Fatigue life of AC5-SBS PMA mixtures as a function of polymer content. 285 Table 6.6 A summary of the transformation functions for each independent variable. I$:g::?::t Traiisfrsztion Equation 1:365:33]? function HPDR Power N f: a(HPDR) ” a and b ITS Power Nf= C(ITS)d c andd Tan 6 Power Nf= e(Tan 5)f e andf PC Polynomial N f: g+hPC+iPC 2 g, h andi VS Power N f = ijS) k j and k 286 be zero for any asphalt binder; PC = Polymer content (percent of the binder); [TS = Indirect tensile strength (pounds/inchz); and VS = Viscosity of the binder (Pouis); Similar nonlinear regression analyses of the fatigue life data were conducted for the SEBS and SBR polymer systems. The resulting models are presented below: SEBS polymer system. _ _ 2 1.4409 FLW =23.363(HIDDR)‘0550"[1 00953” O'OOIZPC J [(ITSXVS)]°'8363 (4.15) T an6 R2 = 0.88 F = 50.96 p-value = 1.868 x 10'17 SBR polymer system 1 0 1620PC 0 0462PC2 "8768 FL ,3, = 13.574(HPDR)‘°-'°°° + ' ' [(ITSXVS) ”33 (4.16) ‘ T and R"- = 0.86 F = 31.48 p-value = 4.299 x 10'12 It can be seen that the final forms of the fatigue models are same for all the three polymer systems, except that the regression coefficients are different. Therefore, the fatigue model can be generalized as follows: 2 (1 FL = k(HPDR)"[1 + ”P TC +561) C ] [(ITSXVS)]" (4.1 7) an Where k, a, b, c, d and e are regression constants. 287 Results of the statistical analyses of the three polymer systems are summarized in Table 6.7. It can be seen that the R2 of models are 92, 88 and 86 percent for SBS, SEBS and SBR polymer systems, respectively. These indicate that the prediction of fatigue life using the above nonlinear regression equations is useful. In order to test whether there is relation between the dependent variable and the set of independent variables in a regression model F-statistics is performed. The null hypothesis (H0) is established as there is no relationship and the alternate hypothesis (H 1) is the relationship exits between the dependent and independent variables in a model. The results of F -test can be presented using the p-values, which is a probability number that measures the extent to which the data are consistent with the conclusion H0 (84). For example, a value of 0.01 based on F-test for a model indicates that the chance of finding that the dependent variable has no relationship with any of the independent variables is one out of one hundred. It also means 99 percent of confidence level for the appropriateness of the regression model. It can be seen from table that p-values of all the PMA systems (For example, 2.034 x 10'17 for SBS polymer system) are considerable small, using the F-statistics. This implies that the null hypothesis is rejected and hence the relationship of fatigue life with the independent variables is significant. The observed and the predicted values of fatigue life for SBS, SEBS and SBR polymer systems were plotted on the logarithmic scale in Figure 6.5. The straight solid line indicates the locus of points where the observed and predicted fatigue life is same. It was found that approximately 82 percent of the predicted fatigue life lie with in the 55 percent of observed values. This indicates that a reasonable fit exits between the data and the nonlinear regression model. 288 Table 6.7 A summary of nonlinear regression analysis of the fatigue model for three polymer systems. Dependent Variable FL fi J Source Polymer Regression Residual Uncorrected Corrected Type Total Total Degree of SBS 6 34 40 39 Freedom SEBS 6 41 47 46 (DF) SBR 6 32 38 37 Sum of SBS 3.11~:+11 2.6E+10 3.4E+11 2.2B+11 Squares SEBS 1.3E+11 l.7E+10 1.4E+1 l 5.1E+11 (SS) SBR 5.6E+10 9.5E+09 6.5E+10 3.4E+1 Mean SBS 5254.10 7.8E+08 “4.3353425: ”“3355? “’5. 25;. Square SEBS 2.1E+10 4.1E+08 (MS) SBR 9.3E+09 3.0E+08 _______ . As m totic As m totic 95% Polymer Regressron Values St’aniiard Confidepnce Interval Type Constants Error Lower Upper k 6.4326 25.4751 -45.34 g 58.20 a 03059 0.0455 -0.398 0213 SBS b -0.1736 0.0567 -0.289 -0.058 C 0.0142 0.0103 -0.007 0.035 d 1.5807 0.5937 0.374 2.787 e 0.9108 0.3290 0.241 1.580 R 23.3627 0.0609 23.240 23.486 a -0.5500 80.443 -l62.64 162.53 SEBS b 00953 0.0437 -0.184 .0007 C -0.0012 0.3104 -0.629 0.626 d 1.4409 0.2913 0.852 2.030 e 0.8363 0.0062 0.824 0.849 11: 18.574 0.2417 18.081 19.067 a 0.1060 72.557 -l48.09 147.87 SBR 1) 0.1620 0.3705 -0.594 0.918 c 00462 0.1177 -0.286 0.194 (1 1.8768 1.2459 -0.664 4.418 e 0.8933 0.0458 0.800 0.987 R2= (l-Residual SS/Uncorrected total SS) F—Statistics p-value 0.92 67.22 2.03E-l7 SEBS 0.88 50.96 1.87E-l7 SBR 0.86 31.48 4.30E-12 289 1,000,000 E" -—l--P-:1-|-I:H:|:l---I:-i-P E :SBS Polymer System H :: ‘NumberofDataPoints=40 rmn‘ _"_T”l7l’l"] 100,000 ~— ___________ _ i 1T2 10,000 2' _ a'é'ei _, 3:-.:: 3:8 7 1"”1‘ ' ‘ITI‘I _1 1T 1,000 3' 21'; 100 ' 1,000,000 E: ::T:‘t—l:l:l—13EI:—::I——1:i: a E iSEBS Polymer Syatem :5 JNumber of Data Points = 47 0 100,000 -__ a :3 3 5 T— j .29 :: 4 H y—— —— <1 G 1 3' 10,000 :: :1 '6 EE 3 8 :: r—tvi-l—H -J1 5 ~——— r-—1—1—1—1-+ g 1,000 -3 a _C: U E:__3 100 I l l l |||| I I I I I II | l I I III] 1,000,000 _:::|::tj:':':';'t1:::i::':t§§§ E SBR Polymer System 33; F Number of Data Points =37 11f 100,000 : i 5. . Z I 321 ::: n ‘21 :13: 1 ¢‘.‘ 10,000 — _ ' __ 1 :13; 3 5E 1,000 t ___________ 100 ‘ ‘1 100 10,000 100, Measured Fatigue Life Figure 6.5 Calculated versus measured fatigue lives of the three polymer systems. 290 6.4.2 Vertical Plastic Deformation (V PD) Similar to the fatigue models the nonlinear regression analysis was conducted on the SBS polymer system using the SPSS statistical computer program for the development of VPD models. The transformation function, which provided the highest R2 was incorporated in the final rut model. Finally, a summary of the transformation functions for each independent variable is presented in Table 6.8. After examining each of the transformation function, they were incorporated into one equation for the estimation of the VPD using the SPSS statistical computer program. The final equation j for SBS polymer system is presented below. , 00736 0.3567 VPD”, = 0.0029[(N)5-6”5[ 1731?] [ML] )(A V)1 -4799 (4.18) R2 = 0.83 F = 496.78 p-value = 2.18 x 10'194 Where, VPD = vertical plastic deformation (x106; inch); N = number of load cycles; tan6 = ratio of storage to loss modulus of the binder, tan8 >0, tanS cannot be zero for any asphalt binder; PC = polymer content (percent of the binder); M, = resilient Modulus (ksi); and A V = percent air voids. Similar nonlinear regression analyses of the VPD data were also conducted for the SEBS and SBR polymer systems. The resulting models are presented below: 291 Table 6. 8 A summary of the transformation function for each independent variable. Type of . Independent Transformation Equation Regressnon Variables . Constants function Nr Power VPD: a(Nr) b a and b Tan 6 Power VPD: C(Tan 5) d c and d PC Polynomial/Power VPD: e(1+PC)f e and f Mr Power VPD= gWr) h g and [7 AV Power VPD: 11,4101 i and j 292 SEBS polymer syatem. Tami (”223 1 0.2918 VPDWS = 0.0590[(N)2-8"“°l m] [7] ](A V)0'4306 (4.19) R2 = 0.78 F = 602.63 p-value = 1.95 x 10'274 SBR polymer syatem. I 'I'ums‘ 00472 03948 I a VPDW = 0.0110[(N)6'8047[i+m') (47,] )(A 1009165 (4.20) i. ~ '5 R2 = 0.83 F = 584.56 p-value = 2.43 x 10'223 Since the forms of the all three VPD models are same, therefore the VPD model can be generalized as follows: VPD = k(N)”' (A V)" (4.21) Where (Tané' j], l c m=a — 1+PC M, k, a, b, c and n are regression constants. The results of nonlinear regression analysis of the VPD models for SBS, SEBS and SBR polymer systems are summarized in Table 6.9. It can be seen that the R2 of the models are 83, 78 and 83 percent for SBS, SEBS and SBR polymer systems, respectively. This indicates that the prediction of rut depth using the above nonlinear regression equation is useful. Moreover, the considerable smaller p-values (For example 2.22 x 10'194 for SBS 293 Table 6.9 A summary of nonlinear regression analysis of the VPD models for three polymer systems. Dependent Variable WD _ _ Source H.332“ Regression Residual Un'crtzg‘elcted €3.33]th Degree of SBS 5 515 520 519 Freedom SEBS 5 837 842 841 (DF) SBR 5 578 583 582 Sum of SBS 8.8E+05 1.8E+05 l.1E+06 5.4E+05 Squares SEBS 9.2E+05 2.6E+05 1.2E+06 6.4E+05 (SS) SBR 1.4E+06 2.7E+05 Mean SBS 1.8E+05 3.5E+02 Square SEBS 1.8E+05 3.1E+02 (MS) SBR 2.8E+05 4.7E+02 A m i As an totic 95% Polymer Regression Values Sinai-tic Coni'idepnce Interval Type Constants Error Lower Upper 1( 0.0029 0.0015 -0.0001 0.0059 a 5.6275 0.4074 4.8271 6.4279 SBS b 0.0786 0.0024 0.0739 0.0834 C 0.3567 0.0111 0.3350 0.3785 (1 1.4799 0.1861 1.1143 1.8455 R 0.0590 0.0217 0.0165 0.1016 a 2.8609 0.235 2.3995 3.3224 SEBS b 0.1223 0.0039 0.1147 0.1298 C 0.2918 0.0124 0.2674 0.3161 (1 0.4306 0.1318 0.1720 0.6893 1( 0.011 0.0050 0.0008 0.0206 a 6.8047 0.572 5.6805 7.9289 SBR b 0.0472 0.0033 0.0407 0.0536 0 0.3948 0.0134 0.3684 0.4211 (1 0.9165 0.1422 0.6372 1.1959 :99; - - L R2= (l-Residual SS/Uncorrected total SS) F—Statistics p-value SBS 0.83 496.78 2.22E-l94 SEBS 0.78 602.63 1955-274 SBR 0.83 584.56 2.43E-223 294 o - fl I j! Em " .; polymer) obtained using the F-statistics suggest that the relationship of rut with the independent variables is significant. The observed and the predicted values of VPD for SBS, SEBS and SBR polymer systems were plotted on the logarithmic scale in Figure 6.6. The straight solid line indicates the locus of points where the observed and predicted rut depths are same. It was found that approximately 90 percent of the predicted rut lie with in the 50 percent of observed values. This indicates that a reasonably accurate fit exits between the data and the nonlinear regression model. The sensitivity analyses of the fatigue and rut models were not performed because various variables in the models are dependent on each other. For example, if the polymer content PC in the fatigue model is changed it will not only effect the viscosity of the binder but the tam? values of the binder will also change. 295 . Ian: 53~ L 3542; . 1000 I 11.414ll 44 1 LI _E_L_I_Ipl|:_._._k_v._.l.:.CCLLI n _DUuLIlBDHHHI 1| ++I:.I.._1_I_I I EELLIIEErrrI r 4| 14'432417 I r+| .. 2.: h. a- 43:31.: rr: r ..I 14|432qu iiiiii _l.I| _ ..I ...,ILEILLII __ _ _ __ ______._ rrl r ..1 aaudj13|_li __ _ . __ : . .1: _I I rkltttLLil _ _ .0 __ _:._._. _ _ _ __ _ _ _ _ __ _ __ m ...i ¢+| 1 1H 4 IIIHHH IIIIHI 31. tel _I _:_|_ ..h: A. Jr. _. +| 1.. .5: _ ___ __ _._.__I.aal._.l _I_... _lfll __ __a __ Try If «4| ,_ ___ _, __ ___ __ ._ ._ ___ __ 0 I Pvt... it‘ll 1 IIIHHI Du rruaErCLIJ _:._.u [L r+uttrIL|1 rrl [L "+1 :+I.II_I1 rrl [L + :tILLll ._ __ _:._.u I rrl i 5 ,_ 7 + 00 1 3 _ 7 . 5 I- 8 T_ 5 _. = __ = u = "h m ,2 m S _ S __ __ m.m __ w m _m.m __ HR 0 u S 0 H6 0 r» .I. +_ P +1 yp 1 W. S J “up. 14 j fly 5 w m 4- a m s m I H" CD HI mD 4 flD 44 f I f .1 ?+ __ NO _I 1W0 _mrm : __2 dd. I a. P 41 I. 2.144% :PM 2 ._92_ _Doiuw _:_____ ES m ttEELLI FIL“ « 1.. v Etkrk :B u __:____ ::.:_ _R m __ SN . __ _B u ___ __ _ v _ nsN ___ __ :::__ _ _ HELL? :___a_ _a 0 0 0 1 1 0 0 0 1 1 0 0 l 0 0 0 1 0 W W m 0 ..I. 0 1 0 1 1 1.1 l 32: ”4.: he 95 8.25.5 inch) . 9 Observed VPD (x 10“ Figure 6.6 Calculated versus measured VPD for three polymer systems. 296 CHAPTER 7 SUMMARY, CONCLUSIONS AND RECOMMENDATAIONS 7.1 SUMMARY The performance of asphalt concrete surfaced pavements is a function of traffic loads and volume, the physical and engineering characteristics of the asphalt concrete, the properties of the aggregate base and sand subbase, the environment and construction practices. Asphalt pavement deteriorates over time due to environmental cycles and increase in the number of traffic load repetitions, which manifest themselves in several types of distress such as rutting, fatigue cracking, thermal cracking, stripping, and raveling. Asphalt cement is Viscoelastic material, which is time and temperature dependent. At high temperatures asphalt cement behaves as a viscous material. At below freezing temperatures it acts as a brittle solid and is more susceptible to temperature cracking. In this study, the structural and engineering properties of polymer modified asphalt (PMA) mixtures were investigated at low, moderate and high temperatures and in lieu of the effects of the rheological properties, morphology and adhesive properties. Three viscosity-graded asphalt cements (AC5, AC10 and AC20) were used in this study. Five types of polymers Styrene- Butadiene- Styrene (SBS), Styrene- Etylene- Butylene- Styrene (SEBS) Kraton, Styrene- Butadiene- Rubber (SBR) latex, Elvaloy AM (EAM) and Crumb Rubber (CRM) were selected for modifying the asphalt cements. The procedures for mixing polymers with asphalt cement were developed (based upon the improvement in the rheological properties) by the Department of Chemical Engineering at Michigan State University (MSU). All asphalt mixtures were designed using the Marshall 297 mix design method and the specifications of the Michigan Department of Transportation (MDOT). Experimental program matrices were established for two types of structural tests; indirect tensile strength test and indirect tensile cyclic load test to evaluate the engineering properties of straight and PMA mixtures. The tests were conducted at low (23°F), moderate (77°F) and high temperatures (140°F) for aged and unaged Marshall size samples. The SBS polymer system was selected as a typical polymer to discuss the improvement in the structural and engineering properties due to polymer modification. The polymer modified asphalt mixtures showed significant improvement in the tensile and compressive strengths, equivalent modulus, fracture toughness and resistance to plastic deformation. The improvement in the structural and engineering properties of the PMA mixtures caused higher fatigue lives and higher resistance to rutting and low temperature cracking potentials. Further comparison of the structural and engineering properties of asphalt mixtures modified with polymers were also discussed. During the course of the study, a new fatigue life criterion was developed based on the rate of accumulation of horizontal plastic deformation and was compared to the rate of crack growth criteria (based on fracture mechanics) (85). The rate of accumulation of horizontal plastic deformation criterion produced consistent and repeatable results for straight and PMA mixtures. Finally statistically based models relating the laboratory fatigue life and rut depth to the rheological properties of binders and engineering properties of PMA mixtures were developed using the nonlinear regression analysis. 298 7.2 CONCLUSIONS Based on the laboratory test results and analysis and comprehensive comparison of the engineering properties among various PMA mixtures, the following conclusions were drawn: 1. The engineering properties (fatigue life, rut and temperature cracking potentials, and tensile and compressive strengths) of PMA mixture increase as the polymer content increases from zero to an optimum value above which, the properties remain constants or decrease. The improvements in the engineering properties of PMA mixtures at low, moderate and high temperatures are due to the improvements in the rheological properties and morphology of the binders and the binder-aggregate adhesion. 2. The rates of accumulation of tensile (horizontal) and compressive (vertical) plastic strains decrease as the polymer content increases to an optimum value. 3. There is no significant difference in the resilient moduli between the processed and PMA mixtures. Since, the applied cyclic load is the same, this implies that elastic strain remained almost the same. 4. The higher number of load cycles to develop the same plastic deformations (horizontal and vertical) and the almost constant resilient modulus indicate that the polymer systems cause a decrease in the energy stored in the sample due to plastic deformation. Since the applied cyclic load was the same for all mixtures, it implies that the plastic properties of PMA mixtures are improved. The improvements in the plastic properties result in high resistance to: 299 7.3 0 Fatigue cracking o Rutting 0 Low temperature cracking The EAM PMA system showed the maximum improvement in the resistance to fatigue cracking and rut potentials followed by SBS and SEBS PMA systems, while the SBS and SBR PMA systems increase the resistance to low temperature cracking of mixtures as compared to other PMA systems. The new fatigue life criterion based on the rate of accumulation of horizontal plastic deformation produced consistent and repeatable results. This criterion showed slightly lower fatigue life than that obtained from the crack growth criterion (based on fracture mechanics). For a cylindrical sample subjected to indirect tensile cyclic loading, the crack propagates in three dimensions. The statistically based laboratory fatigue and rut models for SBS, SEBS and SBR have the similar equation form. These models provide good fit between the laboratory fatigue and rut data and the rheological properties of binders and engineering properties of the PMA mixtures. RECOMMENDATIONS Based on the data analysis and conclusions of this study the following recommendations are made. 1. Field tests be conducted to monitor the pavement performance of straight and PMA mixtures and to assess the benefits and costs of polymer modification. 300 The data from the field tests should be used for the development of fatigue and rut models and be compared with the laboratory models. The effect of polymer modification to the resistance on stripping and raveling should be investigated. Based on the improvements in the engineering properties at high, moderate and low temperatures, it is strongly recommended that the SBS polymer system be used for asphalt modification to improve the pavement performance. The processing conditions of temperature and time for PMA mixtures (as discussed in chapter 3) should be followed while producing the PMA mixtures in the asphalt mixing plant. 301 10. 11. LIST OF REFERENCES Stuart K. D., Asphalt mixtures Containing Chemically Modified Binders, Report, Federal Highway Administration, F HWA-RD__92-101, 1991. Goodrich, J .L., Asphalt and Polymer Modified Asphalt Properties Related to the Performance of the Asphalt Concrete Mixes, Proceedings of the Association of the Asphalt Paving Technologists, Vol. 57, 1988. Studies of the Polyolefin and Latex Modifiers for Asphalt Mixtures, Polymer Modified Asphalt Binders, ASTM STP 1108, Wardlaw, and Shuler, Eds. Philadelphoa, 1992. Newcomb, D. E., Stroup-Gardiner, M., and Epps, J. A., Laboratory and Field ' Little, D. N., Button, J. W., White, R. M., Ensley, E. K., Kim Y., Ahmed S. J., ..., Investigation of Asphalt Additives, Report, Federal Highway Administration, 1987. Shuler, T.S., Collins, H.J., and Kirkpatrick, J. P., Polymer Modified Asphalt Concrete Properties Related to Asphalt Concrete Performance, ASTM STP 941, CE. Brisco, Ed. 1987. Button, J .W., Little, D.N., Kim, Y., and Ahmed, J., Mechanistic Evaluation of Selected Asphalt Additives, Proceedings of the Association of the Asphalt Paving Technologists, Vol. 56, 1987. Khattak, M. J ., and Baladi, G. Y., Engineering Properties of Polymer Modified Asphalt mixtures, Transportation Research Record 1638, TRB National Research Council, Washington, DC, 1998. Material Reference Library Asphalt Data, in Strategic Highway Research 9 Program, National Research Council, Washington, DC. 1992. Highway Materials Engineering- Asphalt Materials and Paving Mixtures. Vol. F HWA-I-90-008, Washington, DC. 1990, U.S.D.O.T, Federal Highway Administration, National Highway Institute. Lawrence, M. F ., Chemical, Physical, and Thermodynamic Properties of Neat and Polymer Modified Asphalt Binders, M. S. thesis, Michigan State University (MSU), 1997. Yoder, Eldon Joseph and Witczak, M. W., Principles of Pavement Design, Wiley Interscience Publication, 2“d edition, 1975. 302 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Heukelom, W., Observations on the Rheology of Bitumens and Asphalt Mixes, Proceedings, Association of Asphalt Paving Technologists (AAPT), Vol. 35, 1966. McLoad, N.W., Asphalt Cement: Pen-visc Number and its Application to Moduli of Stiffness, Journal of Testing and Evaluation, July 4m, 1976. Gaw, W.J., Measurement and Prediction of Asphalt Stiflness. Technical Publication, ASTM, Vol. 628. 1977, Philadelphia, PA. p. 57-67. Kandhal, P.S., Low Temperature Properties of Paving Asphalt Cements, Report, Transportation Research Board, Washington, DC 1988. Halstead, W.J., The Relation of Asphalt Ductility to Pavement Performance, Proceedings, Association of Asphalt Paving Technologists (AAPT), Vol. 32, 1963. Siegmann, M.C., Pfieffer, J., The Properties of Asphalt Ductility to Pavement Performance, Elsevier Publishing Corporation, N.Y., 1950. Anderson, D.A., Rheological Properties of Polymer Modified Polymer Modified Asphalt Binders, ASTM STP 1108, 1992, American Society for Testing Materials: Philadelphia. Rogge, D.F., Terrel ,R.L., and George, A.J., Polymer Modified Hot Mix Asphalt - Oregon Experience, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor., American Society for Testing and Materials: Philadelphia. 1992, p. 151-172. Kandhal, P.S., Low Temperature Shrinkage Cracking 0f Pavements In Pennsylvania, Proceedings, Association of Asphalt Paving Technologists Technical Sessions Lake Buena Vista, Florida February 13, 14, 15,1978): p. 73-114. Performance Graded Asphalt Binders Specifications and Testing, Asphalt Institute, Superpave Series SP-l. Hills, J .F., Predicting the Fracture of Asphalt Mixes by Thermal Stresses, Journal of Institute Petroleum, 1974. King, G., Influence of Asphalt Grade and Polymer Concentration on the High Temperature Performance of Polymer Modified Asphalt, Proceedings, Association of Asphalt Paving Technologists Technical Sessions, Materials: Philadelphia, 1992 p. 29-66. 303 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. McLoad, N.W., Using Asphalt Rheology to Impair or Improve Asphalt Pavement Design and Performance, in Asphalt Rheology: Relationship to Mixture, O.E. Brisco, Editor. 1987, American Society for Testing and Materials: Philadelphia. p. 5 1 -75. Jain, P.K., Sangita, SB, and Arya, I.R., Characterization of Polymer Modified Asphalt Binders for Roads and Airfields, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing. Heather, W.P.F ., Comparative Performance of Polymer Modified Hot Mix and the Practical Problems of Ensuring Performance, in Polymer Modified Asphalt Binders, ASTM STP 1108, American Society for Testing Materials: Philadelphia. Muncy, H.W., King, G.N., and Prudhomme, J .B., Improved rheological properties of polymer modified asphalt, in Asphalt Rheology: Relationship to Mixture, ASTM STP 941, O.E. Brisco, Editor. 1987, American Society For Testing Materials: Philadelphia. Goodrich, J .L., Asphaltic Binder Rheology, Asphalt Concrete Rheology and Asphalt Concrete Mix Properties, Asphalt Paving Technology, 1990. Bouldin, MG. and Collins, J .H., Influence of Binder Rheology on Rut Resistance of Polymer Modified and Unmodified Hot Mix Asphalt, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing and Materials: Philadelphia. p. 50—60. Shuler, T.S., D.L. Hanson, and McKeen, R.G., Design and Construction of Asphalt Concrete Using Polymer Modified Asphalt Binders, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing Materials: Philadelphia. Vallerga, BA. and Bagley, J .R., Design of Asphalt Rubber Single Surface Treatment with Multilayered Aggregate Structure. Asphalt Pavement Construction: New Materials and Techniques, editor, J .A. Schrocman. ASTM STP 724. 1980, American Society for Testing and Materials: Philadelphia. p.22-38. Anderson, D.A., Rheological Properties 0f Polymer Modified Emulsion Residue, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing Materials: Philadelphia. Haiping Zhou, Scott E. Nodes, and Nichols, J .E., Evaluation of three polymer modified Asphalt Concretes, Transportation Research Record 1454, 1994. 304 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. King, G.N., Influence of Asphalt Grade And Polymer Concentration On The Low Temperature Performance Of Polymer Modified Asphalt, Prepared for Presentation at the Annual Meeting of the Association of Asphalt Paving Technologists, 1993. Stroup-Gardiner, M., Newcomb, DE, and Tanquist, B., Asphalt-Rubber Interactions, presented in Transportation Research Board 72nd Annual Meeting January 10-14, Washington, DC, 1992. Serfass, J .P., Joly, A., and Samanos, J., SBS©Modified Asphalt for Surface Dressing © A comparison Between Hot©Applied and Emulsified Binders, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society For Testing Materials: Philadelphia. Bouldin, M.G., Collins, J .H., and Berker, A., Rheology and Microstructure of Polymer / Asphalt Blends, Rubber Chemistry And Technology, 1991. 64(No. 4, September-October 1991): p. 577-600. King, G.N., Muncy, H.W., and Prudhomme, J .B., Polymer Modification: Binder's Effect 0n Mix Properties Asphalt Paving Technology, 1986. Vol. 55, Proceedings Association of Asphalt Paving Technologists Technical Sessions Clearwater Beach Florida February 17, 18 and 19, 1986, p. 519-537. Chatti, K, Characterization of Asphalt Paving Mixtures Under Static and Cyclic Loading, M.S. thesis Michigan State University, 1987. Haung, Yang H., Pavement Analysis and Design, Prentice Hall, Englewood Cliffs, New Jersey, 1993. Haas, R. and Meyer, F., Cyclic Creep of Bituminous Materials under Transient High- Volume Loads, Transportation Research Record 549, TRB National Research Council, Washington, DC, 1975. p. 1-14. Brown, SF. and Cooper, KB, A Fundamental Study of the Stress-Strain Characteristics of Bituminous Materials, Proceedings of Asphalt Paving Technologists, 1980. 149: p. 476-499. Newcomb, D.E., Stroup-Gardiner, M., and J. Epps, A., Laboratory and Field Studies of Polyolefin and Latex Modifiers for Asphalt Mixtures, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society For Testing Materials: Philadelphia Mayama, M., Yshino, M., and Hasegawa, K., An Evaluation Of Heavy Duty Binders, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing Materials: Philadelphia. 305 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. Little, D., Performance Assessment of Binder-Rich Polyethylene-Modified Asphalt Concrete Mixtures (No Vophalt), Transportation Research Record, TRB 1317, 1991. Khosla, P.N., Eflect of the Use of Modifiers on Performance of Asphaltic Pavements, Transportation Research Record 1317, 1991.p. 10-22. Tayebali, A.A., Influence of Rheological Properties of Modified Asphalt Binders on the load Deformation Characteristics of The Binder©Aggregate Mixtures, in Polymer Modified Asphalt Binders, ASTM STP 1108, K.R. Wardlaw and S. Shuler, Editor. 1992, American Society for Testing Materials: Philadelphia. McLoad, N.W., A Four Year Survey of Low Temperature Cracking on Three Onterio Test Roads, Proceedings, Association of Asphalt Paving Technology vol. 41, 1972. Saal, R.N.J., Physical Properties OfAsphaltic Betumin, Surface Phenomena, Thermal and Electrical Properties, Properties Of Asphalt Bitumen, Elsevier, 1950. Pell, P. S., The eflcet of testing and mix variables on the fatigue performance of bituminous materials, Proceedings of Association Asphalt Paving Technology Vol.44, 1975. Baladi G. Y., Integrated material and structural design method for flexible pavement, Vol. 1: Technical Report, FHEA, 1988. Mukhtar, Hamid, Reduction of pavement rutting and fatigue cracking, Ph.D. thesis, Michigan State University (MSU), 1993. Majidzadeh, K., Kauffrnan, E. M., and Ramsamooj, D. V., Application of fracture mechanics in the analysis of pavement fatigue, Proceedings of Association Asphalt Paving Technology Vol. 40, 1971. Barksdale, R.D., Laboratory evaluation of rutting in base course materials Proceedings, third International Conference on the structural design of Asphalt Pavements, Vol. 1, London, 161-174, 1972. Romain, J .E., Rut depth prediction in asphalt pavements Proceedings, third International Conference on the structural design of Asphalt Pavements, Vol. 1, London, 205-210, 1972. Brown, SF, and Bell, C.A.., The validity of design procedure for the permanent deformation in the asphalt pavements, Proceedings, The Asphalt Paving Technologists, Vol. 48, 1977. 306 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. Kenis, W.J., Predictive design procedures- A design methods of asphalt pavements, Proceedings, Fourth International Conference on the structural design of Asphalt Pavements, Vol. 1, Ann Arbor, 205—210, 1977. Kenis, W.J., Sharwood, and McMahon, Verification and Application of VESYS Subsystems, Proceedings, Fifth International Conference on the structural design of Asphalt Pavements, Vol. 1, Delft, The Netherlands, 333-346, 1982. Rita B. Leahy, and Mathew W. Witczak, The influence of test conditions and asphalt concrete mix parameters on permanent deformation coeflicients Alpha and MU, APT Vol. 60, 1991. Baladi G. Y., Fatigue life and permanent deformation characteristics of asphalt concrete mixes, Transportation Research Record, TRB 1227, 1973. Robert, Freddy L., Kandhal, Prithvi S., Brown, E. R, Lee, Dah-Yinn., and Kennedy, Thomas W., Hot Mix Asphalt Materials Design, and Construction, NAPA Education Foundation Lanham, Maryland, Second Edition, 1996. Asphalt Institute Performance Graded Asphalt Binder Specification and Testing, Super Series 1 (SP-l), 1994. ‘ Shull, Jeffery C., The Fundamental Chemical, Physical, and Thermodynamic Properties of Polymer Modified Asphalt Cements, M.S thesis, Michigan State University (MSU), 1995. Brule’, B., Y. Brion, and A. Tanguy., Paving Asphalt Polymer Blends: Relationships Between Composition, Structure, and Properties, Association of Asphalt Paving Technologist, Vol. 57, 1988, pp. 41-60. Lawrence M. France, Chemical, Physical, and Thermodynamic Properties of Neat and Polymer Modified Asphalt Binders, M.S thesis, Michigan State University (MSU), 1997. Hawley M., L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Final report, Michigan Department of Transportation (MDOT), 1997. Hawley M., L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Quaterly Report, Michigan Department of Transportation (MDOT), June 16 1995 to September 15, 1995. Hawley M., L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Quarterly Report Michigan Department of Transportation (MDOT), September 16 1995 to December 15, 1995. 307 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Quaterly Report, Michigan Department of Transportation (MDOT), December 16 1995 to March 15, 1996. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase [1, Quarterly Report Michigan Department of Transportation (MDOT), March 16 1996 to June 15, 1996. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Quarterly Report Michigan Department of Transportation (MDOT), December 16 1995 to March 15, 1996. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase II, Quarterly Report Michigan Department of Transportation (MDOT), September 16, 1996 to December 15, 1996. Traxler, R.N., Durability of Asphalt Cements, Association of Asphalt Paving Teshnologys, Vol. 32, 1963. ASTM Standards, Standard Test Method for Indirect Tension Test for Resilient Modulus of Bituminous Mixtures: D 4123-82, Road and Paving Materials; Paving Management Technologies, Volume 04.03, 1994. E. Eugene Shin, Alekh Bhurke, Edward Scott, Steve Rozeveld, and Lawrence T. Drzal, Microstrucrure, Morphology, and Failure Modes of Polymer-Modified Asphalts, Transportation Research Record 1535, TRB, 1996. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase 1], Quarterly Report Michigan Department of Transportation (MDOT), June 16 1994 to September 15, 1994. Hawley M, L. T. Drzal, and G. Y. Baladi, Polymer in Bituminous Mixtures, Phase 1], Quarterly Report Michigan Department of Transportation (MDOT), March 16 1995 to June 15, 1995. Baladi G.Y., Fatigue life and permanent deformation Characteristics of Asphalt Concrete mixes, Transportation Research Record 1227, TRB, National Research Council, Washington, DC, 1989. Kim Y.R., Effect of temperature and mixture variables on fatigue life predicted fatigue testing, Transportation Research Record 1317, TRB, National Research Council, Washington, DC, 1991. Goodier, J. N., and Hoff, N. J ., Structural Mechanics, Proceedings of first symposium on Naval strcutural mechanics, Pergarnon Press New York, 1960. 308 81. 82. 83. 84. 85. Brock, David, The Practical Use of Fracture Mechanics, pages 52 to 54, Kluwer Academic Publishers, Boston, London, 1988. Rooke, D. P., and Cartwright, D. J ., Compendium of Stress Intensity Factor, Londan Her majesty’s Stationary Office, 1976. Herzberg, R. W., Deformation and Fracture Mechanics of Engineering Materials, pages 588 to 595, John Wiley & Sons Inc., New York, 1989. Neter, J., Wasserrnan, and GA. Whitrnore, Applied Statistics. 4th Edition. Allyn and Bacon, Boston, 1983. 3" Khattak, M. J., and Baladi, G. Y., Analysis of the Fatigue Life of the Asphalt Concrete Samples, presented and accepted for publication in symposium on Asphalt, Number One Thermoplastic Polymer. Proceedings sponsored by , Industrial and Engineering Chemistry Co-sponsored by Petroleum Chemistry 1L Division 217‘h American Chemical Society (ACS). National meeting held at Anaheim, California (March 21-25, 1999). 309 \ M