Chapter 5 Federal Highway Administration Development of Asphalt Bound Surfaces with
5.4 Materials and Mix Design
Test Track Section W7 – Micro-Surfacing. The selected friction aggregates were a calcined bauxite and Texas sandstone. Ashapura provided the quantity of calcined bauxite needed for the Test Track. This material had an 84% Al2O3 (aluminum oxide) content. There were no laboratory or field studies in the United States with the 84% Al2O3 calcined bauxite prior to this
study, so the friction performance was unknown. Great Lakes Minerals maintains a supply of calcined bauxite for HFST meeting AASHTO PP 79-14 (87% minimum Al2O3) but was not able to donate the required quantity. Photos of both samples are shown in Figure 1. The photo of the 84% Al2O3 calcined bauxite shows both crushed particles and rounded particles, which is atypical of the manufactured calcined bauxite used for HFST shown in the right photo.
Figure 1 Calcined Bauxite Samples
As shown in Figure 2, micro-deval tests on the 84% Al2O3 calcined bauxite had 8% loss
compared to 2% loss for the 87% Al2O3 calcined bauxite used in the previous FHWA HFST study.
The mass loss results for the 84% Al2O3 calcined bauxite sample were similar to the best regional friction aggregate (taconite) from the earlier FHWA HFST study (1). The graph also shows two other aggregates with lower mass loss, flint, and slag. These aggregates had lower friction performance compared to taconite in the field, therefore, the 8% mass loss for the 84%
Al2O3 calcined bauxite stockpile may or may not correlate to lower friction performance in the field. Accelerated laboratory friction testing using the NCAT Three Wheel Polishing Device (TWPD) and DFT on a polymer resin-bound 84% Al2O3 calcined bauxite slab specimen was determined to be invalid due to the DFT malfunctioning.
A previous NCAT laboratory evaluation of a sandstone source from Oklahoma showed very good friction performance, so the Texas sandstone was expected to perform well (2). The Texas sandstone source was a preferred source used by Vance Brothers who worked in conjunction with NCAT for the pavement preservation test sections.
The binder used in the micro-surfacing was a highly polymer modified asphalt (HiMA) processed into a CSS-1HP emulsion for use in micro-surfacing applications and was supplied by Ergon Asphalt & Emulsions in Vicksburg, Mississippi. The HiMA base asphalt was selected to improve surface durability and aggregate particle retention compared to conventional emulsion.
Figure 2 Comparison of Micro-Deval Test Results for Two Calcined Bauxite Samples (1) The mix design of the calcined bauxite micro-surfacing and sandstone micro-surfacing were performed by Paragon Technical Services, Inc. The gradations for both are shown in Table 1.
The calcined bauxite gradation was predominately 26% retained on the #8 sieve and 70%
retained on the #16 sieve. This was blended with a Calera, Alabama limestone sand with 51%
passing the #16 sieve. The mix was a 50:50 aggregate blend to satisfy micro-surfacing gradation targets. The coarse fraction of the blend (aggregate retained above the #16 sieve) was 67%
calcined bauxite. The emulsion content was 12.5% by weight of dry aggregate (8.1% residual asphalt by aggregate weight).
Table 1 Gradations by Percent Passing of Section W7A&B Micro-surfacing
Sieve Size Bauxite (50%) Limestone (50%) JMF Blend W7A Sandstone W7B
3/8” 100 100 100 100
#4 100 99 100 90
#8 74 76 75 53
#16 4 51 28 32
#30 1 33 18 21
#50 1 20 11 15
#100 1 13 8 12
#200 0.7 10.8 6.7 9.7
The mix design gradation for the sandstone micro-surfacing had 47% retained above the #8 sieve compared to only 25% retained above the #8 sieve for the calcined bauxite gradation. The sandstone gradation had a coarse fraction of 68% retained above the #16 sieve, which was similar to the calcined bauxite gradation. The CSS-1HP emulsion content for the sandstone micro-surfacing was 12% by weight of dry aggregate (7.6% residual asphalt by aggregate weight).
Test Track Section W3 – SMA. To accomplish a surface with a dominant exposure of calcined bauxite, the SMA was designed as a 4.75 mm nominal maximum aggregate size (NMAS) mixture. The mix design process started with a previous 2003 Test Track Section (N7) granite SMA job mix formula (JMF) using a Marshall 50-blow design procedure. The initial W3 calcined
bauxite SMA mix design used the Great Lakes Minerals calcined bauxite stockpile and was later adjusted (increased binder content) for the Ashapura calcined bauxite stockpile used for field production and placement. A PG 76-22 binder was selected for the W3 mix design and the final binder content was 8.3%. The final combined gradation was 40% calcined bauxite, 59% granite, and 1% filler as shown in Table 2. The coarse fraction of the gradation (retained on the #16 sieve) was 67% calcined bauxite. The Hamburg Wheel-Tracking Device (HWTD) performance test identified a potential for stripping in the laboratory mixture, so the field mixture was produced with an anti-strip agent. Figure 3 shows the summary plot of the HWTD test and Figure 4 is a photo of one of the failed test specimens. Based on the excessive amount of clean sand around the failed HWTD specimens, it appears that the stripping was caused by
incompatibility between the granite and asphalt.
Table 2 Gradation by Percent Passing of Section W3 SMA with Calcined Bauxite Aggregate
Sieve Size JMF Blend Bauxite (40%) Granite (59%) Fly Ash (1%)
ẵ” 100 100 100 100
3/8” 100 100 100 100
#4 100 100 100 100
#8 69 32 85 100
#16 47 1 65 100
#30 32 1 50 100
#50 20 1 35 100
#100 12 1 23 100
#200 8.1 0.5 14.1 100.0
Test Track Sections W8 and W9 – HFST. As a point of reference, three HFSTs from the earlier FHWA study remained in place and were tested quarterly (1). W8A used granite aggregate, W8B used calcined bauxite aggregate, and W9 used flint aggregate. W8B complies with the AASHTO PP 79-14 HFST standard with the calcined bauxite meeting the minimum 87%
aluminum-oxide specification requirement and is considered the “gold standard” for high friction.
Figure 3 Calcined Bauxite SMA Laboratory Mixture Hamburg Wheel-Tracking Test Results
Figure 4 Calcined Bauxite SMA Mixture Stripping Failure