Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
TRB 2011 90th Annual Meeting Modified Micro-Deval Procedure for Evaluating the Polishing Tendency of Coarse Aggregates D. Stephen Lane (Corresponding Author) Associate Principal Research Scientist Virginia Transportation Research Council 530 Edgemont Road, Charlottesville, VA 22903 Phone: (434) 293-1953 Email: [email protected] Cristian Druta Senior Research Associate Virginia Tech Transportation Institute 3500 Research Transportation Plaza, Blacksburg, VA 24061 Phone: (540) 231-1056, Fax: (540) 231-1555 Email: [email protected] Linbing Wang Professor The Via Department of Civil and Environmental Engineering Virginia Polytechnic Institute and State University 301N Patton Hall, Blacksburg, VA 24061 Phone: (540) 231-5262, Fax: (540) 231-7532 Email: [email protected] Wenjing Xue Graduate Research Student The Via Department of Civil and Environmental Engineering Virginia Polytechnic Institute and State University 200 Patton Hall, Blacksburg, VA 24061 Phone: (540) 231-1069 Email: [email protected] Submission date: November 15, 2010 Word count: 3345 Words + 4 Tables (1000) + 6 Figures (1500) = 5845 TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 2 ABSTRACT Coarse aggregates used in asphalt concrete should be tough and durable to withstand HMA production, transportation, construction, traffic loads and environmental effects. Similarly, their morphology not only plays an important role in aggregates’ mechanical properties but also influences the asphalt pavement skid resistance, an essential characteristic of the pavement performance. As the predominant aggregate resources located in the western parts of Virginia are carbonate rocks, the mineral components of these rocks tend to be relatively soft and when subjected to abrasive wear under traffic the aggregate surface polishes fairly rapidly. This leads to a smoothing of the aggregate surface and thus potential safety issues are created due to loss of surface friction. Considering all these aspects, the increasingly popular Micro-Deval apparatus was employed to assess the abrasion/polishing (A/P) resistance of coarse aggregates suspected to show a range of polishing characteristics. While the conventional focus of the Micro-Deval test is on attrition or degradation (mass loss) reflecting the durability of particles, it also induces a polishing effect on susceptible particles and consequently was selected for use in this study. Hence, a suite of carbonate and non-carbonate coarse aggregates from ten Virginia sources were tested in the Micro-Deval using both the standard procedure and a cyclical A/P procedure. Also, 2-D digital aggregate images, before and after abrasion, were obtained and analyzed using specialized software to characterize their morphological parameters (i.e., shape, angularity, and texture). Testing results showed that wear-resistance aggregate groupings based on the A/P procedures were consistent with the polishing tendency based on lithologic characteristics. Although, the abrasion effects induced by both Micro-Deval procedures were subjectively discernable and trends could be perceived in the image analysis data, these differences were not statistically verified. INTRODUCTION In both asphalt concrete and hydraulic cement concrete, mineral aggregates are the main component to withstand the traffic and environmental loads. When used in asphalt pavement surface layers they need to be rough-surface-textured to provide skid resistance as well as resistant to polishing under wear by vehicle tires (1, 2). In order to maintain a high level of pavement surface friction (skid resistance), aggregates should wear non-uniformly; that is, they should not become rounded or polished smooth under traffic. Because the rate of aggregate polishing is closely related to the types of minerals they contain, a high percentage of hard mineral grains is desirable in the aggregates to provide polishing resistance. Also, asphalt pavement wearing surfaces often exhibit dissimilar friction characteristics under similar traffic conditions, whether they are constructed with aggregates of different rock types, or of the same rock type but from different sources. The purpose of this study was to develop objective and practical methods to assess the polishing/abrasion characteristics of carbonate and non-carbonate coarse aggregates for use in pavement surface layers. This would provide VDOT district materials engineers with needed tools to exercise sound judgment in the selection of aggregates for use in surface courses that would permit cost savings where appropriate and provide justification for the use of higher-cost aggregate when necessary. To achieve this goal, samples of coarse aggregate from 10 sources across VDOT’s western districts were evaluated using laboratory techniques to assess their A/P tendencies. The techniques included petrographic examination, acid-insoluble residue of the carbonate rocks, and two procedures using the Micro-Deval (MD) apparatus to subject the TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 3 particles to abrasive wear. Also, an image analysis system was employed to evaluate the changes in particle morphology resulting from the abrasion testing. BACKGROUND Virginia Department of Transportation (VDOT) specifications call for nonpolishing aggregate for use in most surface layers. However, VDOT materials engineers have few methods or tools with which to assess the polishing characteristics of aggregates to distinguish between polishing and nonpolishing aggregates. This is primarily a problem in the western part of Virginia where the predominant source materials for aggregates are carbonate rocks composed primarily of relatively soft minerals; they are thus considered to be polishing aggregates. Various tests are used to evaluate aggregates for abrasion resistance. Among these tests, the most common is the Los Angeles test that measures the ability of aggregates to withstand impact and abrasion. However, due to some criticism regarding its inability to render accurate aggregate degradation results (3, 4, 5) the MD apparatus is gradually gaining acceptance by agencies as an alternative abrasion resistance/durability test (6, 7, 8, 9, 10). Hossain et al. (5) present a detailed literature review and test results on this topic. They conclude that the MD test can accurately differentiate between good- and poor-performing aggregates, with much lower test variability, and recommend its use for fine and coarse aggregate. METHODS AND MATERIALS Aggregate Selection Coarse aggregates were selected from three VDOT districts: Bristol, Salem, and Staunton. The respective district materials engineers were asked to suggest aggregate sources currently considered to be polishing and nonpolishing, i.e., carbonate and non-carbonate aggregates, respectively. Based on their experience and familiarity with the use and performance of the aggregates, they suggested 10 sources for inclusion in the study. The selection of the carbonate aggregates was not directly related to the actual field performance in pavements, rather, it related to geographic distribution and variation in the mineralogical composition of the rock. The selected sources which represent the most predominant aggregate types available in the western part of Virginia were divided into two groups. The first group included two noncarbonate sources, currently meeting the nonpolishing definition by VDOT (11), representing better quality materials (Maymead and Salem) and one for which the general quality was more questionable (Nolichucky). The second (or the polishing) group included two limestones with low or very fine insoluble residue (IR) that would be expected to polish rapidly (Strasburg, Frazier) and several dolostone sources that early work indicated had higher and coarser IR that may be relatively less susceptible to polishing (Perry, Broadway, Rockydale, Castlewood, Staunton). Table 1 provides information on the aggregate types and sources. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue TABLE 1 Coarse Aggregate Sources and Sizes Maymead Granite Salem Quartzite Nolichucky Feldspathic quartzite, weathered Limestone, dolomitic Limestone Limestone, dolomitic Limestone, dolomitic Limestone Dolomite Limestone Broadway Strasburg Perry Frazier Castlewood Rockydale Staunton 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Aggregate Description 37 - 25 x NonCarbonate Aggregate Source Carbonate 1 4 Aggregate Sizes (mm) 25 - 19 19 - 12.5 12.5 - 9.5 x x x 9.5 - 4.75 x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x Petrographic Analysis Geologic information and other data pertaining to each aggregate source were obtained from the Virginia Division of Mineral Resources, industry sources and by searching the existing literature on Virginia geology. Samples from each source were examined using microscopic techniques including thin sections to determine the predominant rock types or minerals. The carbonate sources were evaluated for acid-insoluble total residue using Virginia Test Method (VTM) 37. In addition, particle mounts of the carbonate sources were prepared and a ground, polished surface was etched with dilute hydrochloric acid to distinguish between calcite and dolomite or acidinsoluble material. Brief lithologic descriptions are listed in Table 2. Based on the lithologic character of the aggregate sources they can be grouped according to their presumed susceptibility to polishing. The grouping is based on relative hardness of the minerals present (calcite being the relatively soft: 3 on the Mohs hardness scale, dolomite slightly harder, Mohs 3.5-4; and quartz, Mohs 7), grain size and uniformity (uniformly fine grained rocks are more susceptible to polishing). Using these criteria, a tentative grouping is shown in Table 3. 19 20 21 22 23 24 25 26 27 TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue TABLE 2 Lithologic Descriptions and Insoluble Residue Results Aggregate Source Granite Quartzite, medium to fine angular grains Nolichuckync Iron-cement sandstone, medium grained Broadwayc Dolomite -fine-grained microspar with some coarser areas; micritic intraclasts and peloidal zones; quartz sand seams Limestone organic rich micrite with microspar and bioclasts Dolomitic limestone - microsparite; micritepeloidal micrite Limestone - uniform fine-grained microsparin micrite Dolomitic limestone - microsparite, densely packed Dolomite - micrite, peloidal; micrite with dispersed microspar; coarsely crystalline Dolomite - microsparite with frequent coarser grains; microcrite with microspar Perryc Frazierc Castlewoodc Rockydalec Stauntonc 10 11 12 13 14 15 16 17 18 19 20 21 22 Insoluble Total Residue % N/A N/A Insoluble Residue N/A N/A N/A N/A 18.0 Clay/shale lumps, quartz grains 1.9 Carbon 15.6 Predominately clay with some clay/shale lumps and minor chert 10.0 Carbonaceous clay 24.8 Clay; clay/shale lumps 13.4 Clay, shale lumps 3.7 Clay Subscripts c and nc denote carbonate and non-carbonate, respectively. TABLE 3 Tentative Polishing Tendency Based on Principal Component Mineral Hardness Polishing Tendency Readily Readily to Moderate Moderate Nonpolishing 7 8 9 NonCarbonate Maymeadnc Salemnc Strasburgc 2 3 4 5 6 Lithologic Description Carbonate 1 5 Source Strasburg, Frazier Castlewood, Perry Broadway, Rockydale, Staunton Maymead, Salem, Nolichucky Mohs Hardness 3 3.5-4 3.5-4 6-7 Micro-Deval Testing The abrasion resistance of the coarse aggregates was determined in accordance with AASHTO T 327-05: Resistance of Coarse Aggregate to Degradation by Abrasion in the Micro-Deval Apparatus (the standard procedure). Details of the procedure have been provided by Hossain et al. (5) and others (12, 13, 14). Rezaei et al (15) used the method in a study of polishing characteristics of Texas aggregates, noting that some aggregates readily lost their surface texture during the course of the MD test while others retained a considerable amount of the original roughness. In the standard MD test the aggregate is tumbled upon itself in the presence of stainless steel balls. Thus, much of the abrasive action is between the aggregate particles themselves and is influenced by the nature and homogeneity of the source, such that for a limestone sample, the particles are abrading against other particles of equal hardness. Standard tests such as the British Polishing Wheel and the Aggregate Abrasion Value Machine (16) use an abrasive grit composed of a relatively hard material to induce wear on the particles. Consequently, we used the MD apparatus in a non-standard procedure that attempted to simulate TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 6 the A/P action of polishing tests which use grit. In the procedure devised, 300 g of coarse aggregate was placed in the jar mill together with 200 g of silica sand (abrasive grit) conforming to the ASTM C 778 (20-30) designation (17), 1250 g of stainless steel balls and 750 g of water. The standard sand was selected for this purpose since it is also used in the Aggregate Abrasion Value test and is a very uniform material. One potential downside to its use is that the particles are quite rounded and of uniform size, thus diminishing its abrasive effect. In this respect, a sharp (angular), graded sand might present an advantage in inducing wear. The jar was revolved at 100 rpm for three successive 15, 30, and 45 min periods. After each period (cycle), the aggregate particles were removed, dried, and weighed en mass, and the loss was determined for the sample. Thus, the same aggregate particles were abraded for 15, 45, and 90 min in total. Following each weighing, images were taken of each particle with backlighting for image analysis of the aggregate morphology. Aggregate Morphological Assessment Image analysis and processing were performed on all the aggregates described above to assess their morphological characteristics (i.e., shape, angularity, and surface texture). Image analysis was conducted on individual particles composing the sample, before and after abrasion cycles. The individual particle results are then aggregated to produce a single tri-parametric result for the sample as a whole. For this purpose, digital pictures were acquired and analyzed for approximately 20-25 aggregate particles tested using both the MD standard (AASHTO T 327) and non-standard procedures. The number of particles was selected in order to cover a broad range of shapes and was based on prior asymptotic analysis, when statistically stable values for morphological parameters were acquired from 15 particles (18). For the non-standard procedure, the same particles were imaged before and after each cycle, but individual particle orientation during imaging was random, except for a few sets, where an effort was made to image the particles in the same orientation after each cycle. For the standard MD procedure, individual particles were oriented in several ways for imaging, but the actual particles imaged before and after abrasion were from different sub-samples of the same source material. Ideally, particles from the test sample would be imaged before and then following testing. For the A/P MD tests, additional images were taken of selected particles with incident lighting for a visual record of the change induced on the particle by the non-standard A/P process. Figure 1 presents an example of such imaging as well as particle orientation. Morphological parameters were obtained using Matlab program code, based on a unified Fourier method, developed for processing any digital picture color format. Details with respect to image analysis and processing methodology were presented by Wang et al. (18). Hence, a total of 1,300 digital pictures (500 for the standard procedure and 800 for the A/P procedure) were analyzed to assess the wear induced by the MD apparatus on the aggregate particles. All the pictures were acquired at 640 x 480 pixel resolution. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 RESULTS AND DISCUSSION 7 Abrasion/Polishing Results 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 7 FIGURE 1 Aggregate particles before (1) and after Micro-Deval cycling (1a - 15 min; 1c - 90 min). The results of the standard MD tests (2 h) and the cycle A/P MD tests are presented in Figure 2. Detailed information is provided in the Virginia Transportation Research Council (VTRC) report (19). With two exceptions - Strasburg limestone and Nolichucky sandstone - the aggregates can be considered as “suitable” aggregates for surface courses since they exhibited loss values of less than 15% in the standard MD tests as suggested by White et al. (13) (Figure 2). Values ranged from a low of 5.4% (Broadway) to a high of 19.9% (Nolichucky). In the same light, Salem quartzite could be considered “questionable” as its loss value (13.2%) is closer to the 15% limit. Previous tests of Nolichucky and Salem (unpublished data) had been conducted in response to concerns about the quality of the material being supplied yielding values of 23.1% and 19.1%, respectively. In general, the Nolichucky material is somewhat friable with the component sand grains easily loosened from the cementing matrix. With the Salem material, the earlier sample was obtained when field personnel suspected that the supply was being produced from a more weathered area of the deposit where the matrix had become weakened. The Salem material included in this study was of better quality than that previously tested. Also, from Figure 2, it can be observed that the abrasion loss from the non-standard A/P procedure followed the same pattern as for the standard procedure. This indicates that the nonstandard procedure did not alter the basic wear relationships between aggregates in the MD apparatus. Furthermore, the new devised method can considerably reduce the testing time and still provide consistent abrasion results. In these figures, aggregates were grouped based on their lithologic characterization in Table 2. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 8 FIGURE 2 Micro-Deval test results (Standard (2h) and Non-standard procedures). For the carbonate rocks tested, the standard MD test results can be separated into two groups: those “less wear-resistant” (LWR) with losses in excess of 10% (i.e., Strasburg, Frazier, and Rockydale) and those “more wear-resistant” (MWR) (i.e., Broadway, Castlewood, Perry, and Staunton). Comparing these groupings with those of Table 3, the lithologic polishing tendency, it is noted that the Strasburg and Frazier were grouped together as readily polishing, with the others split between moderate-to-readily and moderate. The obvious non-conformant in these groupings is Rockydale, which was included in the moderate group in Table 3 but falls into the LWR group based on MD loss. An explanation for the apparent aberrant behavior of the Rockydale is that some particles showed signs of weathering, resulting in localized softening and splitting along fracture or bedding planes within the particles (see Figure 3). The MD procedure has been shown to be an effective tool in identifying reduction in material quality resulting from weathering (5). Additional samples of the material should be obtained and tested to verify this explanation and assess the implications regarding sampling, testing and acceptance of materials for specific applications. Examining the results of the cyclic A/P MD tests, the carbonate rocks separate into two groups based on loss that remain the same through the progression of wearing cycles. Using loss values of 1.0% at 15 min, 2.5% at 45 min, and 4.5% at 90 min, the carbonate aggregates can again be grouped into LWR (>x), i.e., Strasburg, Frazier, and Rockydale, and MWR (<x), i.e., Broadway, Castlewood, Perry, and Staunton. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 9 FIGURE 3 Rockydale aggregate particle showing oxidation along internal boundaries caused by weathering. For the non-carbonate rocks, both Nolichucky and Salem fell into the LWR group based on both the standard and cyclic A/P MD criteria. The Maymead falls with the MWR based on the standard test result and the LWR with the A/P results. At this juncture, it is important to draw a distinction between simple mass loss as determined in the standard and non-standard MD procedures, and polishing, where the surface of the particle is worn in a uniform manner that eventually results in a very smooth surface. The tendency of a rock to polish is a function of mineral composition, component grain size and fabric (arrangement of grains). A rock composed of a relative soft mineral such as calcite, when grain size and fabric are uniform can polish readily with minimal loss of material, whereas a rock of more heterogeneous character might exhibit a similar mass loss without polishing because differential wear of individual grains, loss of poorly bound grains, or loss of grains along fracture or cleavage planes, perpetuating a relatively rough surface. These differences can be observed in images of Strasburg or Frazier and Maymead particles, which had similar losses in the MD tests, but whose surfaces after polishing cycles are quite different (19). Figure 4 illustrates these differences showing the same particles of Salem, Strasburg, Maymead, and Staunton before the A/P MD tests and after the initial and final cycles. Here one can see the contrast between two carbonates, Strasburg (LWR) which readily smoothes and rounds, and Staunton (MWR), which smoothes but continues to maintain fairly sharp edges; and non-carbonates Salem and Maymead which maintain rougher surfaces and sharper edges despite material loss. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 10 FIGURE 4 Images of particles before (left), after initial (center), and final (right) A/P MD cycles. (Salemnc, Maymead nc, Staunton c (MWR), and Strasburg c (LWR) aggregates. Image Analysis Statistical Results In this study, statistical analysis was performed to quantify the distribution of aggregate morphologic characteristics (shape, angularity, texture) in the aggregate samples and to compare these characteristics among the samples. Sufficient aggregate profiles were selected for the analysis as suggested by Wang et al. (20) in an earlier study. Although all three parameters were included in this study, texture and, to a lesser degree angularity, are expected to reflect changes induced by abrasion. Figures 5 and 6 illustrate the mean values for the aggregates’ characteristics before and after MD abrasion testing for the standard and non-standard A/P procedures, respectively. Aggregate sources were grouped according to their lithologic description in Table 2. Examining the image analysis data for the non-standard A/P procedure in Figure 6, the two aggregates with the steepest progressive loss of angularity through two cycles fell into the LWR category based on simple mass loss and were those classified as readily polishing by lithologic characteristics. The third steepest loss in angularity was shared by two sources, one of which was TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 11 the third aggregate to fall into the LWR group. Similarly, the same four aggregates, Strasburg, Frazier, Rockydale, and Broadway, are four of the five showing the steepest progressive loss in texture through two cycles. Even though these subtle distinctions can be visually perceived (see Figure 4), Student’s t-tests were performed among all sets of aggregate parameters (15, 30, and 45 min testing) to determine if the aggregate mean characteristics values were statistically different. For this evaluation, a one-tail distribution and equal variances were selected, at a confidence limit of 95% (α = 0.05). They did not indicate significant differences among the same type of stone or source. Seventeen results (less than 10%) of 180 comparisons were statistically significant. P values less than α (0.05) are provided in Table 4 (shaded cells). Most of the differences (15 out of 17) were related to the shape and angularity (Strasburg, Frazier, Castlewood, and Rockydale), and only two related to the surface texture (Strasburg and Staunton). All of the shape-related differences as well as the majority of the angularity differences were tied to Strasburg or Frazier, both of which exhibited a tendency to rapidly round and lose their edges (see Strasburg in Figure 4). However, differences in the orientation of the particles when imaged (shape and angularity) may also have played a role. TABLE 4 Student’s t-test P Values Cycles (min) 0-15 0-45 0-90 15-45 15-90 45-90 Shape 0.1926 0.0943 0.0069 0.3655 0.0489 0.0610 0-15 0-45 0-90 15-45 45-90 0.0620 0.0002 0.0007 0.0384 0.1474 0-15 15-45 0.1164 0.0809 0-90 0.0518 0-15 0.2367 Strasburgc t-test P-value Angularity 0.1286 0.0213 0.0014 0.2077 0.0286 0.1116 Frazierc 0.0076 0.0001 0.0240 0.0552 0.0183 Castlewoodc 0.0483 0.0111 Rockydalec 0.0477 Stauntonc 0.4754 Texture 0.4768 0.0972 0.2700 0.1092 0.2554 0.0432 0.1072 0.0790 0.0903 0.1292 0.2585 0.1942 0.3271 0.4077 0.0119 As illustrated in Figure 6, the texture decreased quite uniformly or remained constant over the three testing periods as the same aggregates were pictured and analyzed in approximately the same orientations. Similarly, the standard MD procedure (AASHTO T 327) introduced some changes to the aggregates’ shape, angularity, and texture, as can be observed from Figure 5, but statistically significant differences were not observed among these parameters. In this case, the TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 12 differences related to the shape and angularity could have occurred because the aggregates analyzed before and after abrasion were not the same. Although visually discernable differences could be noted in particle angularity and texture between A/P cycles, these differences were not being satisfactorily identified through the image analysis processing. However, the Fourier analysis method developed for analyzing aggregate morphology works reasonably well for differentiating among shape, angularity, and texture of different aggregate source material (20). Furthermore, a new approach aimed at the way the acquired images are cropped for analysis as well as enhancement to the program code regarding the number of terms necessary for a better estimation of the morphological parameters are currently being developed. When the aggregate particles are subjected to the cyclic A/P testing, the observed changes are mainly in texture and to a lesser degree, angularity. Changes in shape are not a major factor except for materials that are very abrasion susceptible. 15 16 TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 13 FIGURE 5 Aggregate morphological parameters evaluation (AASHTO T 327 Procedure). 6 7 TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 14 FIGURE 6 Aggregate morphological parameters evaluation (Non-standard procedure). CONCLUSIONS This study introduced a new methodology to evaluate the abrasion/polishing resistance and morphology of coarse aggregates which was used to examine materials from western Virginia sources. The following are the main findings based on the test results: Based on petrographic examinations of the samples, the carbonate aggregates were subjectively assigned into three categories (Readily, Readily-Moderate, and Moderate) reflecting presumed polishing tendency. Using 10% loss in the standard (2 h) Micro-Deval (MD) procedure as the criterion, the carbonate aggregates were grouped into LWR and MWR categories. In the cyclical MD abrasion procedure devised for this study with successive 15-min, 30min, and 45-min wearing cycles, the carbonate sources fell into the same LWR and MWR grouping as with the standard MD procedure using respective losses of 1.0%, 2.5%, and 4.5% for the three cycles. Similar trends were observed for all aggregate sources based on the loss values of 1% (15 min), 2.5 (45 min), and 4.5 (90 min), respectively. Although the number of aggregate sources was limited, the new cyclic MD method could significantly shorten the testing time and still provide reliable results. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 15 The wear-resistance groupings based on the MD procedures were consistent with the polishing tendency based on lithologic (i.e., mineralogy, grain size, and fabric) characteristics except for one aggregate which appeared to have been affected by weathering. The wear and polishing effects induced by both the cyclical and standard MD abrasion procedures were subjectively discernable and trends could be perceived in the image analysis data, but these differences were not statistically verified. More differences were observed on the shape/angularity side rather than the texture side. However, potential problems that may be interfering with the ability of the image analysis program to clearly discern angularity and textural effects indicative of polishing have been identified and enhancements to the program are being developed. REFERENCES 1. Mahone, D.C. and C. Sherwood. The Effect of Aggregate Type and Mix Design on the Wet Skid Resistance of Bituminous Pavement: Recommendations for Virginia's Wet Accident Reduction Program. Research Report FHWA/VA-96-R10, Virginia Transportation Research Council, 1995. 2. Webb, J.W. The Wearing Characteristics of Mineral Aggregates in Highway Pavements. Research Report VHRC 70-R7, Virginia Highway Research Council, 1970. 3. Senior, S.A., and C.A. Rogers. Laboratory Tests for Predicting Coarse Aggregate Performance in Ontario. In Transportation Research Record: Journal of the Transportation Research Board, No. 1301. Transportation Research Board of the National Academies, Washington, D.C., 1991, pp. 97-106. 4. Cooley, A., Jr., and R.S. James. Micro-Deval Testing of Aggregates in the Southeast. In Transportation Research Record: Journal of the Transportation Research Board, No. 1837. Transportation Research Board of the National Academies, Washington, D.C., 2003, pp. 7379. 5. Hossain, M.S., D.S. Lane, and B.N. Schmidt. Use of the Micro-Deval Test for Assessing the Durability of Virginia Aggregates. Research Report VTRC 07-R29. Virginia Transportation Research Council, 2007. 6. Wu, Y., F. Parker, and P.S. Kandhal. Aggregate Toughness/Abrasion Resistance and Durability/Soundness Tests Related to Asphalt Concrete Performance in Pavements. In Transportation Research Record: Journal of the Transportation Research Board, No. 1638, Transportation Research Board of the National Academies, Washington, D.C., 1998, pp. 8593. 7. Gatchalian, D., E. Masad, A. Chowdhury, and D. Little. Characterization of Aggregate Resistance to Degradation in Stone Matrix Asphalt Mixtures. In Transportation Research Record: Journal of the Transportation Research Board, No. 1962, Transportation Research Board of the National Academies, Washington, D.C., 2006, pp. 55-63. 8. Barksdale, R.D. The Aggregate Handbook. National Stone, Sand and Gravel Association, Arlington, VA, 1991. 9. Atkins, N.H. Highway Materials, Soils, and Concretes. 4th Edition, Prentice Hall, NJ, 2003. 10. Alexander, M., and S. Mindess. Aggregate in Concrete. Taylor and Francis Group, London, New York, 2005. TRB 2011 Annual Meeting Paper revised from original submittal. Lane, Druta, Wang, and Xue 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 16 11. Virginia Department of Transportation. Road and Bridge Specifications, Richmond, 2007. 12. Saeed, A., J.W. Hall, and W. Barker. Performance-Related Tests of Aggregates for Use in Unbound Pavement Layers. Research Report NCHRP No. 453. Transportation Research Board, Washington, D.C., 2001. 13. White, T.D., J.E. Haddock, and E. Rismantojo. Aggregate Tests for Hot-Mix Asphalt Mixtures Used in Pavements. Research Report NCHRP No. 557. Transportation Research Board, Washington, D.C., 2006. 14. Lang, A.P., P. H. Range, D. W. Fowler, and J. J. Allen. Prediction of Coarse Aggregate Performance by Micro-Deval and Other Soundness, Strength, and Intrinsic Particle Property Tests. In Transportation Research Record: Journal of the Transportation Research Board, No. 2026, Transportation Research Board of the National Academies, Washington, D.C., 2007, pp. 3-8. 15. Rezaei, A., E. Masad, A. Chowdhury, and P. Harris. Predicting Asphalt Mixture Skid Resistance by Aggregate Characteristics and Gradation. In Transportation Research Record: Journal of the Transportation Research Board, No. 2104, Transportation Research Board of the National Academies, Washington, D.C., 2009, pp. 24-33. 16. Rogers, C., B. Gorman, and B. Lane. Skid-Resistant Aggregates in Ontario. Proc., 10th Annual Symposium, International Center for Aggregates Research, 2002. 17. ASTM C 778. Standard Specification for Standard Sand. Annual Book of ASTM Standards. Vol. 04.01. ASTM, West Conshohocken, Pa., 2009. 18. Wang, L., D. S. Lane, Y. Lu, and C. Druta. Portable Image Analysis System for Characterizing Aggregate Morphology. Research Report VTRC 08-CR11. Virginia Transportation Research Council, 2008. 19. Wang, L., C. Druta, and D. S. Lane. Methods for Assessing the Polishing Characteristics of Coarse Aggregates for Use in Pavement Surface Layers. Research Report VTRC 10-CR7. Virginia Transportation Research Council, 2010. 20. Wang, L., D. S. Lane, Y. Lu, and C. Druta. Portable Image Analysis System for Characterizing Aggregate Morphology. In Transportation Research Record: Journal of the Transportation Research Board, No. 2104, Transportation Research Board of the National Academies, Washington, D.C., 2009, pp. 3-11. TRB 2011 Annual Meeting Paper revised from original submittal.