MODIFIED BITUMINOUS CONCRETE

Transcription

1 , '. ef-80 MODIFIED BITUMINOUS CONCRETE ROUTE H, F BY. CHARLES K. i4cinroe MARESH R. PATEL NJDOT BUREAU OF MATER! ALe OICTOBER.1991

2 PURPOSE. THE PURPOSE OF THIS LABORATORY INVES.TIGATION IS TO CONDUCT A TECHNICAL ANALYSIS OF A RUBBER MODIFIED BITU MINOUS CONCRETE DESIGN TO DETERMINE HOW FEASIBLE IT IS AS A HIGHWAY PAVEMENT. TESTXNas I" FIELD SAMPLES GF-80 RUBBER GRADATIONS - PRODUCER BIN GRADATIONS - VERIFICATION MARSHALL SPECIMENS - COMPOSITION ANALYSIS LAB SAMPLES - ACTUAL BIN MARSHALL SPECIMENS - THEORETICAL DESIGN MARSHALL SPECIMENS - SPECIFIC GRAVITY AND AIR VOIDS - VISCOSITIES, PENETRATIONN, AND SPECIFIC GRAVITY OF VIRGIN AC-20 - VISCOSITIES, PENETRATION, AND SPECIFIC GRAVITY OF GF-80 MODIFIED AC-20

3 '~.:. ROUSE RUBBER INDUSTRIES GF-80 STANDARD SIEVE ANALYSIS PERCENT PASSING SIEVE SIZE TRIALS 1 & 2 TRIALS 3 & 4 TRIALS 5 & 6 SPECIFICATION (100 g) (50 9) (25 g) I '.. MODIFIED SIEVE ANALYSIS PERCENT PASSING SIEVE SIZE TRIAL 1 TRIAL 2 TRIAL 3 SPECIFICATION (50g/WET SIEVED/ (50gI-DRY SIEVED/ 50g/DRY SIEVED/ 1 HR) RUBBER STOPPERS/ RUBBER STOPPERS 1 HR) TALC/l HR) NOTE: 1 THE STANDARD SIEVE ANALYSIS IS NOT POSSIBLE WITH THE GF-80. WE TRIED THREE.HODIFIED SIEVE ANALYSIS. TRIALS 1 AND 3 YIELDED THE BEST RE SULTS. TRIAL 3 RESULTS ~ERE HORE ACCURATE THAN TRIAL 1. NOTE: 2 FOR THE NEXT PHASE WE WILL USE 1" RUBBER BALLS INSTEAD OF RUBBER STOP PERS. THIS PROCEDURE IS RECOKHENDED BY THE PRODUCER.,,

4 ROUSE RUBBER INDUSTRIES af-acl MODIFIED SIEVE ANALYSIS PERCENT PASSING SIEVE SIZE TRIAL 1 fl00g TRIAL 2 (50g TRIAL 3 (50g SPECIFICATION GF-SO & TALC)/ GF-80 & TALC)/ GF-80/25g TALC) 1" BALLS/ 1 HR 1" BALLS I 1 HR 1" BALLS/ 1 HR " S i MODIFIED SIEVE ANALYSIS PERCENT PASSING SIEVE SIZE TRIAL 4 (50g 50g TALCUM SPECIFICATION GF-SO ONLY) 1". POWDERI 1" BALLS/ 1 HR BALL/O.S HR ci c! S5.S NOTE: 1 TRIAL 3 YIELDED THE BEST RESULTS. TRIALS 1 AND 2 ALSO PASSED. SOME OF THE SIEVES WERE JUST SLIGHTLY OUT OF SPECIFICATION BUT THIS IS DUE.TO THE SMALL AMOUNT OF TALC THAT ADHERED TO THE RUBBER. NOTE: 2 BASED ON THESE RESULTS. TRIALS AND 3 WERE ACCEPTABLE. WE RECOM- MEND USING TRIAL 3 BEC~USE THE SMALLER AMOUNT OF MATERIAL IS EASIER TO WORK WITH. NOTE: 3 THE. GRADATION OF TALCUM POWDER IS FOR REFERENCE PURPOSES ONLY..

5 ElRADATlONSa SIEVE SIZE ACTUAL BIN COMB. THEORETICAL COMB. SPECIFICATIONS 1"' /4"' /2"' /8"' # # # # # i #100 # RaHALT CEMENT TEST RESULiSI TESTS VIRGIN AC-20 MODIFIED AC=20 ABSOLUTE 140 F. POISES KINEMATIC 275 F, CENTISTOKES F, 100g, 5 sec SPECIFIC GRAVITY

6 -- - MARSHALL TEST RESULTS SERIAL NUMBER STABILITY FLOW BULK SPa GRAVITY MAX. SPa GRAVITY " AIR VOIDS ; " ; 3.4 -AVERRGE AVERAGE : AVERAGE AVERAGE NOTE: VERIFICATION SET VERIFICATION SET ACTUAL FROM BIN SAMPLES THEORETICAL FROM PRODUCER'S DESIG~

7 COMPOSITION ANALYSIS RESULTS FIELD SAMPLES SERIAL NO SPECIFI.cATIONS TRUCK LOAD 12TH 20TH 8TH 32ND 54TH 41ST MIN MAX SIEVE SIZE PERCENT PASSING 2" 1 1/2". I 1" /4" /2" /8" # " # # # # #100 # % AC

8 COMPOSITION ANALYSIS RESULTS VERIFICATION SAMPLES - SERIAL NO. PLUG NO. 1 SIEVE SIZE SPECIFICATIONS MIN MAX PERCENT PASSING 2" 1 1/2" 1" 100 3/4" " 93 3/8" 83 #4 60 # #16 31 #30 22 # , #100 # :.: AC 7.44R NOTE: IT HAS NOT BEEN DETERMINED WHY THESE TWO VALUES ARE HIGHER THAN THE OTHERS. THE FILTER RING USED ON THESE SAMPLES WERE THICKER THAN THE ONES USED ON THE OTHERS. THERE WAS ALSO SOME OF THE RUBBER FOUND IN THE CORRECTION FINES.

9 CONCLUSION, SEVERAL METHODS FOR GRADING THE GF-80 RUBBER WERE ATTEMPTED. IT WAS DETERMINED THAT 50g OF RUBBER, 25g OF TALCUM POWDER, TWO 1" RUBBER BALLS, SHAKEN FOR 1 HR. YIELDS THE BEST RESULTS. THIS PROCEDURE IS SIMILAR TO THE ONE RECOMMENDED BY THE PRODUCER. THE MARSHALL TEST RESULTS FOR THE PLUGS MADE FROM, I" THE ACTUAL BIN SAMPLES YIELDED HIGH "AIR VOIDS. THIS IS' USUALLY ASSOCIATED WITH HIGH PERMEABILITY. THEREFORE, EVEN IF STABILITIES ARE GOOD THE VOIDS SHOULD BE ADJUS TED. THIS COULD BE ACCOMPLISHED BY INCREASING THE AMOUNT OF MINERAL FILLER IN THE MIX. WE LATER DISCOVERED THAT THE BIN SAMPLES THAT WE RECIEVED WERE IN FACT LOW ON FINES. THE COMPOSITION ANALYSIS REVEALED THAT SOME OF THE RUBBER IS FOUND IN THE CORRECTION FINES AND POSSIBLY DIS SOLVED IN THE SOLVENT. THE ASPHALT CEMENT TESTS ON GF-80 MODIFIED ASPHALT CEMENT SHOWED A SIGNIFICANT INCREASE IN THE VISCOSITIES, A SIGNIFICANT DECREASE IN THE PENETRATION, AND A SMALL INCREASE IN THE SPECIFIC GRAVITIES. IT HAS BEEN DETERMINED THAT THE RUBBER MODIFIED MIX IS FEASIBLE AND THAT IT WILL BE USED ON A SECTION OF IN TERSTATE 95.

10 GLASS (CRCG) AND RUBBER (GF-8(1) IN ADDITION TO THE TESTING FOR THE ROUTE 1-95 AND PROJECT WE TRIED A MIXTURE OF CRUSHED GLASS WITH GROUND RUB BER. THE MIX DESIGN USED AN 1-4 TOP COURSE FROM THE WALTER EARLE CORPORATION. THIS MIX DESIGN WAS MODIFIED BY ADDING 10 PERCENT CRUSHED RECYCLED CONTAI NER GLASS <'CRCG) AND 10 PERCENT GROUND TIRE RUBBER (GF-BO). THESE ADDITIVES WERE COMBINED WITH #67, #B, AND #10 STONE AND WASHED SAND TO OB TAIN THE FOLLOWING PERCENTAGES PASSING: SIEVE SIZE THEORETICAL COMBINATION SPECIFICATION MIN. - MAX. 1 " / "00 1/ B /8" #4 5' b5 # # # # # THE MI X WAS TESTED AT VA RI DUS AS PHAL T CONTENTS AND AN " OPTIMUM CONTENT OF 5.1 PERCENT WAS OBTAINED. THESE RESULTS ARE SHOWN BELOW: ~ A. C STABILITY FLOW ~ AIR VOIDS VMA VFA (I C> '=' "j ~.~ 17.5 B7.4 NOTE: ALL VALUES ARE AVERAGE OF 3 SPECIMENS

11 , " SPECIFICATIONS DESIGN RESULTS MIN. MAX. STABILITY (1 bs) (> FLOW (0.01 i y,) AIR VOIDS ()(. ) VMA ()(. ) VFA (1- ) 82.3 THEORETICAL SP. GR LBS/SQ. YO/IN THE OPTIMUM ASPHALT CONTENT OBTAINED FROM THIS DATA WAS 6.1 PERCENT. THE FOLLOWING RESULTS ARE AT THIS AS- PHALT CONTENT: THESE VALUES ALL COMPLY WITH THE NJDOT SPECIFICATIONS AND WE ENCOUNTERED NO DIFFICULTIES IN THE LABORATORY. AF- TER A VISUAL INSPECTION FOR'STRIPPING WE FOUND 5 TO 10 PER- CENT OF THE TOTAL SURFACE AREA WAS UNCOATED. THIS TEST DOES NOT PRESENT ANY CONCLUSI VE DATA. IT HAS BEEN De.TERM I NED, THEREFOR, THAT THIS PARTICULAR MIX DESIGN CAN BE USED WITH 10 PERCENT CRCG AND 10 PERCENT GF-80.

12 MAY 27, 1992 ASPHALT CEMENT AND GF-80 HUBIlER MIXTURES OF 15~ AND 20% RUBBER WERE MIXED WITH DIF FERENT GRADES OF ASPHALT CEMENTS. THE 20~ RUBBER MIXED WITH AC-20 WAS TOO VISCOUS SO NO VISCOSITY VALUES WERE OBTAINED. FOR THE 15% MIXTURE WITH AC-20, ONLY AN ABSOLUTE VISCOSITY VALUE WAS OBTAINED. THIS VALUE, HOWEVER, WAS TOO FAR OUT OF SPECIFICATION TO MAKE AN ANALYSIS. THE 20% RUBBER MIXED WITH AC-5 WAS ALSO TOO VISCOUS TO OBTAIN ANY VALUES. THE 15% MIXTURE WAS VISCOUS ENOUGH TO TEST BETWEEN 330 AND 350 DEGREES FARENHEIT BUT NOT AT TEST TEMPERATURES. BASED ON THE RESULTS OF THESE TESTS WE RECOMMEND THAT NO MORE THAN 10% RUBBER BE USED WITH ANY GRADE OF ASPHALT CEMENT.