Download Modified Micro-Deval Procedure for Evaluating the Polishing

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.