INVESTIGATION OF TESTING METHODS TO DETERMINE LONG-TERM

DURABILITY OF WISCONSIN NATURAL AGGREGATE RESOURCES

GREGORY S. WILLIAMSON
Graduate Student in Civil and Environmental Engineering, Virginia Polytechnic Institute and State University
Blacksburg, VA, USA

RICHARD E. WEYERS, Ph.D.
Professor in Civil and Environmental Engineering, Virginia Polytechnic Institute and State University
Blacksburg, VA, USA

DAVID W. MOKAREM, Ph.D.
Research Scientist, Virginia Transportation Research Council
Charlottesville, VA, USA

DANIEL S. LANE
Research Scientist, Virginia Transportation Research Council
Charlottesville, VA, USA

ABSTRACT
The Wisconsin Department of Transportation (WisDOT) uses approximately 11,000,000
tons of aggregate per year for transportation projects. Therefore, being able to select durable
aggregates for use in transportation projects is of considerable importance. If the aggregate
deteriorates then the constructed facility will require premature repair, rehabilitation or
replacement. Realizing the importance aggregate durability and also those deficiencies in the
current WisDOT testing protocol may exist, it has been concluded that the durability-testing
protocol for Wisconsin aggregates needs to be updated.
This project has identified recent advances in the understanding and testing of aggregate
durability. An in depth literature review was conducted and from the compiled information a
laboratory testing program was developed.
From the test results it was found that the WisDOT aggregate testing protocol could be
reduced substantially by eliminating many of the testing requirements for aggregates that have
vacuum saturated absorptions of less than 2%. The Micro-Deval abrasion test is recommended
for inclusion in WisDOT testing protocols as a test to measure the abrasion resistance of
aggregate while the L.A. Abrasion test is better suited as a measure of aggregate strength.
Additional conclusions were made based on the durability testing conducted and an overall
testing protocol was developed.

Keywords: aggregate, durability, soundness, abrasion resistance, aggregate strength
Williamson, Weyers, Mokarem, and Lane 1
INTRODUCTION
Aggregate is the most fundamental component of construction. It is used as an unbound
or lightly bound material in base courses and as a bound material in bituminous and Portland
Cement Concretes (PCC). Aggregate constitutes in excess of 90% of the volume of base courses
and bituminous concrete and 60 to 80% of the volume of Portland cement concrete. Aggregate
is exposed to a number of physical and chemically degrading forces during processing,
transporting, and construction as well as during the constructed facilitys life. As the main load-
carrying component of unbound, bituminous concrete, and Portland cement concretes, if the
aggregate fails, the facility fails to perform its design intent.
Durability is a term used to define resistance to the chemical and physical forces of
degradation to which a material is subjected throughout its service life. Given WisDOTs
investment in its transportation infrastructure, it is important that the aggregate testing protocol
be appropriate for the assessment of the long-term durability of constructed facilities. An ideal
protocol would neither accept non-durable materials nor reject sufficiently durable materials.
To ensure the durability of pavements and structures, WisDOT has used a number of test
methods to assess aggregate quality and durability. The test methods include gradation,
plasticity, resistance to abrasion (impact), soundness, and freezing and thawing resistance. Some
of the tests have been developed as aggregate quality assurance tests and others have been
borrowed from other materials testing programs. These tests have been in use for well over 50
years and for the most part have served the highway industry well. However, these tests were
developed when high quality natural sources of aggregate were abundant and social and political
pressures on the use of industrial by-products and recycled/reclaimed materials were nonexistent.
In addition, some of these tests have been kept in use in the name of tradition and simplicity
rather than being replaced by other methods based on our ever-increasing understanding of the
science of aggregate durability. In light of these changes and advances in technology, WisDOT
sponsored a project to assess the adequacy of its aggregate testing protocol
BACKGROUND
The following discussion is presented to illustrate the underlying philosophy that was
used to develop the WisDOT aggregate durability testing protocol.
Williamson, Weyers, Mokarem, and Lane 2
WisDOT projects use aggregates as unbound pavement base courses, bituminous and
PCC pavements and structural concretes for bridges, box culverts, and retaining walls. Each one
of these facility components is exposed to a set of chemical and physical degrading forces during
construction and throughout the facilitys service life. For example, structural Portland cement
concrete aggregates are exposed to:
Abrasive forces while the aggregate is in a moist/wet state during stock piling,
transporting, batching, mixing, and placing of the concrete,
Tensile, shear, and compressive stresses during loading of the reinforced concrete
structure,
Chemical environments of a saturated solution of calcium hydroxide, sodium and
potassium hydroxide, and sulfates in the concrete,
Wetting and drying cycles of the concrete, and
Temperature changes including freezing and thawing of absorbed moisture of the
concrete.

With respect to temperature changes, an unconfined aggregate must be capable of
handling the stresses that are developed due to water freezing within the aggregate pore
structure. During the freezing of saturated aggregates bound together by Portland cement paste,
the pressure being created by the freezing water must not be sufficient to fracture the aggregate
nor be extruded into the surrounding cement paste at a rate which fractures the cement paste
(1). Of these three potential aggregate freezing and thawing destructive mechanisms, two are
related to the aggregate being bound within cement paste. An unbound aggregate test such as
soundness or the freezing and thawing test would not assess these aggregate performance related
aspects. This may be the reason why these tests have been poor predictors of the freezing and
thawing durability of certain aggregate types. It is noteworthy that the expulsion distance
mechanism may become more important in the future as more low permeable concretes are
produced which have denser aggregate-paste transition zones.
The above could be interpreted that it is necessary to test all concrete aggregates exposed to
freezing and thawing under saturated conditions in a concrete freezing and thawing test. This
will not be necessary because an aggregate with porosity below 0.3% produces expansions
within the elastic limits of the aggregate and surrounding cement paste (1). Thus, a concrete
Williamson, Weyers, Mokarem, and Lane 3
aggregate protocol may consist of a simple, precise test for water absorption and saturated
specific gravity to identify the very good and very poor aggregates and a freezing and thawing or
surrogate test for marginal aggregates.
However, since the freezing and thawing of base courses takes place in the unbound state,
the soundness test may be sufficient if the lack of precision aspects of this test can be addressed.
Also, PCC pavement aggregates need to be assessed for the propensity of D-cracking (durability
cracking), a somewhat different freezing and thawing property (2). D-cracking is cracking that
occurs along the edges or at the corners of concrete slabs due to the expansion of non-durable
coarse aggregates during freezing and thawing. The resulting deterioration is crescent shaped
cracks or spalls in the concrete.
Thus, the developed aggregate durability testing protocol is based on construction-service
life performance criteria. The protocol is then based on performance tests, which will
realistically simulate the field exposure and degradation process (3).
EXPERIMENTAL METHODS
Aggregate Performance Assessment Program
Aggregate performance durability issues may be categorized as physical or chemical.
Physical degradation mechanisms include:
Attrition during handling and construction,
Degradation under in-service loads, and
Environmental degradation from freezing and thawing, wetting and drying, and/or
thermal expansion and contraction.

Chemical degradation mechanisms include the following but are not assessed in this study:
Reactive oxides as CaO and MgO and sulfides as ferrous sulfide,
Alkali-silica reaction, and
Alkali-carbonate reaction.

Aggregate performance properties have a direct influence on the stability of aggregate
particles in the unbound and bound state. For example, an aggregate with a high porosity and
low permeability defines the aggregates freezing and thawing critical size whether in the
Williamson, Weyers, Mokarem, and Lane 4
unbound state or the bound state. An aggregate with a high porosity and high permeability may
not fracture as an unbound material but will degrade the binding forces in bituminous and PCC.
Thus, aggregate performance properties not only influence aggregate durability, but also the
durability of their inclusion material. Therefore, the proposed aggregate testing program is
presented for unbound, and bound aggregate such as bituminous concrete, PCC pavement, and
structural elements and their associated durability performance parameters.
Aggregate Physical Properties
Aggregate physical durability properties are typically inter-related. For example, both
freezing and thawing degradation mechanisms, aggregate fracture and degradation of binder-
aggregate forces occur when the aggregate has a high porosity. Porosity and absorption are
directly related, as are absorption and specific gravity. Aggregate that has a high specific gravity
generally has a low absorption. These aggregates would generally have a high strength, high
abrasion resistance, and a high resistance to dimensional changes. Relative to physical
degrading forces, some aggregates may be accepted based on specific combinations of specific
gravity and absorption. The physical performance characteristics of other aggregates may have
to be determined by abrasion, strength, and freezing and thawing testing. Thus, the objective of
the proposed testing program is to develop a tiered aggregate assessment protocol.
Aggregate Chemical Properties
While aggregate physical durability properties are generally inter-related, chemical
degradation mechanisms are typically dependent upon the mineralogical composition of the
aggregate. WisDOT officials have determined that chemical degradation mechanisms relating to
aggregate mineralogy are not a significant problem in Wisconsin. Therefore, the proposed
testing protocol does not contain any chemical durability testing.

LABORATORY TESTING
Current WisDOT Testing Protocol
WisDOTs current aggregate durability testing protocol is presented below in Table 1:

Williamson, Weyers, Mokarem, and Lane 5
Aggregate Test Selection
Selection of the tests was based upon the tests precision, repeatability, efficiency, and
predictive capabilities. Wisconsins current aggregate tests were also conducted for comparative
purposes. The tests selected were as follows:
Lightweight Pieces in Aggregate (ASTM C 123-98) (4)
Vacuum Saturated Specific Gravity and Absorption (Coarse)(Modified - ASTM C 127)
Sodium Sulfate Soundness (ASTM C 88) (4)
Frost Resistance of Aggregates in Concrete (ASTM C 666) (4)
Unconfined Freezing and Thawing of Aggregate (CSA A23.2-24A) (5)
Micro-Deval Abrasion (Coarse) (AASHTO TP 58) (7)
L.A. Abrasion (ASTM C 131-01) (4)
Aggregate Crushing Value (British Standard 813-Part 3) (6)

Crushed stone and gravel aggregates were all subjected to the same testing protocol. The
use of the same testing procedures for all sources will increase efficiency and will also
encompass the full spectrum of durability testing needs. Testing for Lightweight Pieces in
Aggregate is an important screening test used to determine the percentage of non-durable
aggregates for crushed stone and gravel samples, as excessive amounts of lightweight aggregate
will result in a reduction in durability. The test for Vacuum Saturated Specific Gravity (VSSG)
and Absorption (VSA) may be used as an indicator for aggregate soundness. VSSG and VSA
testing was selected in place of standard SG and absorption testing because it is a better predictor
of the long-term absorptions of aggregate in the field. Soundness tests that were investigated
were the Sodium Sulfate Soundness (ASTM C88), Unconfined Freezing and Thawing of
Aggregates (CSA A23.2-24A), and Frost Resistance of Aggregates in Concrete (ASTM C 666)
tests.
Unconfined Freezing and Thawing of Aggregate (CSA A23.2-24A) testing was carried out in
place of Soundness of Aggregates by Freezing and Thawing (AASHTO T 103) because it was
recommended over AASHTO T 103 and the Sulfate Soundness Test by Senior and Rogers (8).
The test has better precision and better correlation with field performance, which are among the
reasons cited for its recommendation. Additionally, the Canadian test requires only 5 freezing
and thawing cycles in comparison to the 25 required by AASHTO T 103.
Williamson, Weyers, Mokarem, and Lane 6
Testing of the resistance of aggregates to abrasion was conducted using both the L.A.
Abrasion test and the Micro-Deval test. The Micro-Deval test was selected because it can be
applied to both fine and coarse aggregates, and recent reports have shown that test results
correlate better with field performance records (9, 10, 11). Also, the L.A. Abrasion and Impact
Test includes the effects of impact on the aggregate sample, thus both the effects of abrasion and
impact are present in the test results. As a result, brittle aggregates may have a higher L.A.
Abrasion loss do to the impact forces in the test. There is no clear understanding on how to
interpret the results so that mass losses of the aggregate sample can be attributed to the abrasion
resistance or the impact resistance of the aggregate. If the Micro-Deval test is used, another test
must also be used to measure the aggregate strength. The British Aggregate Crushing test and
the L.A. Abrasion test have been chosen to assess the strength of the aggregate.
Although, not included in the testing program, aggregates which may be alkali carbonate
or alkali silica reactive are to be tested in accordance with ASTM C 1293 (Determination of
Length Change of Concrete Due to Alkali-Silica Reaction) (4). This recommendation is based
on the results of the literature review (12).
Laboratory Testing Program
Seventy natural aggregate samples, representing the full range of aggregate available in
Wisconsin, were collected for testing. Initially all 70 crushed stone and crushed gravel samples
were tested for VSSG and VSA in a test procedure similar to ASTM C 127. The modification
was to place the aggregate under a vacuum of 635 mm (25 in.) of mercury for 5 minutes prior to
saturating the aggregates. Aggregate saturation consisted of the introduction of tap water while
the aggregate was under vacuum and subsequent submersion in water for 24 +/- 1 hour. From
these results 30 aggregates were selected for further analysis throughout the range of the VSA
values. The selected aggregates were then subjected to the full suite of selected tests with one
exception. Only nine aggregates were tested in concrete for freezing and thawing durability
using ASTM C 666-97 (Resistance of Concrete to Rapid Freezing and Thawing) due to limited
time and resources. ASTM procedure C 666-97 was modified to 28 days of curing in lime
saturated water rather than the 14 days specified and the aggregate was saturated by soaking it in
water for 24 hours prior to the mixing of the concrete. These modifications were carried out to
Williamson, Weyers, Mokarem, and Lane 7
ensure complete saturation of the aggregate prior to freezing and thawing testing. This is
intended to simulate the worst-case field condition.
Selection of Aggregates
With the assistance of WisDOT officials, 70 natural aggregate samples were selected for
testing from across the state. Among the aggregates tested were unconsolidated deposits and
bedrock from different geologic groups. A variety of glacial deposits were tested because the
material often varies widely depending upon from which direction the depositing glacier
originated. It was also important to test bedrock from groups with good and poor field
performance ratings. The performance ratings of poor, intermediate, and good were based on
either field performance or test results. The performance ratings were designated by WisDOT
officials and should be verified due to their subjective nature. Figure 1 presents the sample site
locations throughout the state of Wisconsin for both the pits and quarries.
The importance of testing the range of aggregate durability performance based on field
performance ratings cannot be over emphasized. It is often very easy to identify good and poor
performing aggregates based on laboratory test results. It is much more difficult to identify
aggregates that have adequate field performance histories but would be classified as intermediate
aggregate based on laboratory testing.
For the 30 aggregate samples that were selected for further testing it was important to
ensure that the full range of aggregate qualities was reflected in the sample set with an emphasis
on intermediate quality aggregates. The distribution of the aggregate samples tested was as
follows: 7 Poor, 13 Intermediate, and 10 Good
Once the sample distribution had been determined, the individual samples within the
performance categories were selected based on VSA data, as there is a strong relationship
between VSSG and VSA. Aggregates with low, moderate, and high absorption values were
selected from each group in order to be certain that aggregate qualities can be identified without
any dependence on absorption i.e. a poor aggregate with a low absorption can be identified as
poor just the same as a poor aggregate with a high absorption.
Williamson, Weyers, Mokarem, and Lane 8
TEST RESULTS
Vacuum Saturated Specific Gravity and Absorption
VSSGs and VSAs were determined for all 70 natural aggregate samples using the
modified version of ASTM C 127. VSA was then plotted against VSSG to investigate the
relationship (see Figure 2). The use of the VSA as a preliminary durability indicator will be
investigated in comparisons with other test data. As shown, there is a general linear relationship
between VSA and VSSG for carbonate rocks. The relationship was not investigated for
igneous/metamorphic rocks because of their tendency to have low absorptions regardless of
specific gravity. The siliceous granites and metamorphic aggregates have lower VSAs and
VSSGs and are shown with VSAs of about 1% and VSSGs ranging between 2.55 and 2.65.
Lightweight Particles in Aggregate
The percentage of lightweight particles in aggregate was determined using WisDOT
modified ASTM procedure 123-98. The ASTM procedure was modified to test only coarse
aggregate material retained on the 3/8 in. sieve. WisDOT sets a limit of 5% and 2% chert, with a
SSD specific gravity of less than 2.45, by mass for standard concrete and pre-stressed concrete,
respectively. Chert is a white or buff colored siliceous material that is highly porous and non-
durable. Freezing and thawing deterioration caused by chert tends to manifest itself in the form
of popouts. The percentages of chert were determined by petrographic analysis from those
aggregate samples that contained lightweight material (See Table 2).
L.A. Abrasion
L.A. Abrasion tests were performed using 500 revolutions in accordance with ASTM C
131-01. Figure 3 presents the relationship between L.A. Abrasion and VSA. The failure limits
are shown as solid lines. The dashed line represents the range for which all L.A. Abrasion values
will lie for VSAs of less than 2%.
As demonstrated, aggregate with VSAs of less than 2% will have a L.A. Abrasion value
of less than 35%, which is substantially lower than the 50% loss failure criterion used by
WisDOT. It should be noted that only one aggregate had a L.A. Abrasion value of greater than
50%.
Williamson, Weyers, Mokarem, and Lane 9
Micro-Deval
Figure 4 presents the relationship between VSA and Micro-Deval % loss. A loss of 18%
was used as the failure criterion as recommended by Kandhal and Parker for HMA (13).
As demonstrated, a failure limit of 18% may be too stringent and a limit of 25-30% may
be more appropriate for Wisconsin aggregate. Also, for aggregate with VSA less than 2% only
one has a % loss greater than 12%. Therefore, it is reasonable to conclude that aggregate with
VSA less than 2% will have a low Micro-Deval value.
Aggregate Crushing Value
Figure 5 presents the Aggregate Crushing Values of the aggregates tested with respect to
VSA. The British Standard suggests that the allowable aggregate crushing loss should be based
on the parent material. Average losses range from 16% for igneous material to 27% for
argillaceous limestone.
With no discernable relationship or recommended failure criteria it is not possible to
draw any conclusions from this data. However, for aggregates with a VSA of less than 2%, the
Aggregate Crushing Value is less than 22%.
Strength and Abrasion Test Comparison
A comparison of L.A. Abrasion and Micro-Deval test results demonstrated that there is a
linear relationship although not a strong one (See Figure 6). Thus, it can be reasoned that the
two tests are measuring different aggregate properties.
Figure 7 presents the relationship between L.A. Abrasion and Aggregate Crushing Value.
As shown, the two tests have a distinct correlation implying that the L.A. Abrasion test is more a
test of aggregate strength than of abrasion resistance. The results indicate that it may be
necessary to perform both the L.A. Abrasion Test and the Micro-Deval Test or possibly the
Micro-Deval and Aggregate Crushing tests.
The Micro-Deval, Aggregate Crushing Value, and L.A. Abrasion test results demonstrate
that the tests are measuring different physical properties of the aggregate. The Micro-Deval Test
measures the abrasion resistance of the aggregate while the L.A. Abrasion and Aggregate
Crushing Value measure strength.
Williamson, Weyers, Mokarem, and Lane 10
Sodium Sulfate Soundness
Figure 8 presents the relationship between VSA and sodium sulfate soundness loss with a
failure criterion of 12% loss, which is currently being used by WisDOT. There is no apparent
correlation between the two data sets.
As shown aggregates with VSA of less than 2% will have sulfate soundness losses of less
than approximately 5%. It should also be noted that the soundness test was only able correctly
identify 2 out of the 7 poor aggregates tested.
Unconfined Freezing and Thawing
Unconfined Freezing and Thawing of Aggregate tests were conducted in accordance to
Canadian Standard A23.2-24A. The failure criterion is a 10% loss after five cycles of freezing
and thawing, as stated in the Canadian Standards Association specification. Figure 9 presents the
relationship between VSA and freezing and thawing loss.
There is no apparent relationship between VSA and unconfined freezing and thawing
loss. A full third of the aggregates tested were greater than the failure limit of 10%. Present
WisDOT specifications for unconfined freezing and thawing limit the loss to 18% for AASHTO
T 103, which is similar to the Candian Freeze/Thaw test. Thus, a 15% loss limit appears to be
reasonable.
Freezing and Thawing in Concrete
The freezing and thawing of concrete specimens containing Wisconsin aggregate was
conducted in accordance with ASTM C 666 (Resistance of Concrete to Rapid Freezing and
Thawing). The failure criterion for ASTM C 666 is a reduction in the fundamental transverse
frequency of greater than 40%. Due to time constraints only 9 of the 30 aggregates were tested.
The aggregates selected for testing represented all three performance categories with samples
having low, moderate, and high absorptions. Of the nine aggregate samples tested, none failed.
The poorest performing aggregate reflected a 30% reduction in transverse frequency. It is useful
to note, however, the deterioration that occurs throughout the test. Those aggregate samples
containing chert resulted in more popouts in the concrete specimens, which is generally more of
an aesthetics concern rather than a structural problem. It may be necessary to conduct freezing
and thawing in concrete tests where lightweight aggregate particles are a concern because one
Williamson, Weyers, Mokarem, and Lane 11
aggregate, sandstone containing in excess of 16% lightweight aggregate, disintegrated during
the testing and left voids in the concrete where the aggregate once was.
CONCLUSIONS AND RECOMMENDATIONS
1. The absorption of an aggregate, while not directly related to the quality of the aggregate, can
still be used as a preliminary indicator of durability. Aggregates with vacuum saturated
absorptions of less than 2% do not need to be tested for L.A. Abrasion loss, Micro-Deval
loss, Unconfined or Confined Freezing and Thawing tests. They will still, however, need to
be tested for Lightweight Pieces in Aggregate.
2. The inclusion of ASTM C 123 (Lightweight Pieces in Aggregate) in the WisDOT aggregate
durability testing protocol is necessary in order to quantify non-durable lightweight material
percentages. This is important particularly for gravel resources where high variability in
parent material is common. The WisDOT pre-established maximum allowable chert
percentages of 5% for normal concrete and 2% for pre-cast concrete are appropriate.
However, rather than specifying only chert with a SSD specific gravity of less than 2.45, it
would be best to limit all lightweight aggregate to 5%, and test those aggregates that fail in a
confined freezing and thawing test. For concrete structures where aesthetics are a concern
there should be no chert present in the aggregate to prevent popouts from occurring.
3. The L.A. Abrasion test was only able to identify the very worst aggregate sample as being
poor. This indicates that the L.A. Abrasion test does have some ability to predict
performance. From this data it can be concluded that the L.A. Abrasion test cannot directly
predict the overall performance of an aggregate, but it can accurately estimate a key
parameter, aggregate strength. The L.A. Abrasion test should continue to be used to evaluate
aggregate strength. Realizing that the Aggregate Crushing Value measures this same
parameter, it is not recommended that the Aggregate Crushing Value test be implemented at
this time.
4. The Micro-Deval test (AASHTO TP 58) with the recommended maximum allowable loss
limit of 18% rejects nearly 50% of all aggregates tested. It is clear, however, that more poor
aggregates are accurately classified than with the L.A. Abrasion test. For Wisconsin
aggregate it appears that a maximum allowable loss limit of 25% is more reasonable. The
Micro-Deval test should be added to WisDOT testing protocol to evaluate the abrasion
Williamson, Weyers, Mokarem, and Lane 12
resistance of aggregate, as this test more accurately models the degradation that occurs
during handling and mixing.
5. The Sodium Sulfate Soundness test (ASTM C 88) was able to correctly identify two out of
seven poor aggregates without rejecting any intermediate or good aggregates. The problem
with this test is that it is highly variable with a potential multi-laboratory difference between
two tests of 116%.
6. The Unconfined Freezing and Thawing test (A23.2-24A) data has no correlation with
Sodium Sulfate Soundness results. Due to the poor precision of the sodium sulfate test, it is
recommended that the unconfined freezing and thawing test be included in the WisDOT
aggregate testing protocol with an upper limit of 15% loss.
7. The Freezing and Thawing of Concrete test is recommended for aggregate that are to be used
in the bound state. This test helps to identify non-durable aggregates that may result in
popouts, aggregate deterioration, and cracking of concrete.
PROPOSED TESTING PROTOCOL
Based on the conclusions presented above proposed testing protocols have been
developed for unbound and PCC aggregates and are presented in Figures 10 and 11, respectively.
The testing protocols were developed with the aim of improving the aggregate durability testing
program while also reducing the amount of testing required. The results of the proposed testing
protocol in comparison with current WisDOT testing protocol are presented in Table 3. The
proposed testing protocol rejected more of the poor performing aggregate samples while
rejecting fewer good samples. Additionally, by using the proposed testing protocols the required
L.A. Abrasion testing can be reduced by 37% and the Unconfined Freezing and Thawing testing
can be reduced by 30% for the 30 aggregate samples tested.

Williamson, Weyers, Mokarem, and Lane 13
REFERENCES
1. Verbeck, G. and R. Landgren, Influence of Physical Characteristics of Aggregate on
Frost Resistance of Concrete, Proceedings, ASTM, Vol. 60, pp. 1063-1079.
2. Dubberke, W. and V. Marks, The Relationship of Aggregate Durability to Trace
Element Content, Interim Report on Project HR-266, Iowa DOT, p. 12, J an.
1984.
3. Frohnsdoff, G. and J . Clifton, Cement and Concrete Standards of the Future,
Workshop on Cement and Concrete Standards Report, NIST, p. 48, Oct. 1995.
4. ASTM, Annual Book of ASTM Standards. Vol. 04.02, ASTM, Philadelphia, PA,
2002.
5. CSA, Canada Materials and Methods of Concrete Construction/Methods of Test for
Concrete: National Standards of Canada, CSA-A23.1-00 and CSA-A23.2-00,
August 2001.
6. British Standard, Testing Aggregates Part 110: Methods for determination of
aggregate crushing value (ACV), BS 812-110 1990, 1998.
7. AASHTO, 2000, Standard Specifications for Transportation Materials and Methods
of Sampling and Testing. Part II Tests: 1136.
8. Senior, S. A. and Rogers, C. A., 1991-2, Laboratory Tests for Predicting Coarse
Aggregate Performance in Ontario, Transportation Research Record 1301,
Washington, D.C.
9. Folliard, K.J . and Smith, K.D., 2003, Aggregate Tests for Portland Cement Concrete
Pavements Review and Recommendations. National Cooperative Highway
Research Program, Research Reports Digest No. 281, Transportation Research Board,
Washington, D.C.
10. Kandhal, P.S., and Parker, F., 1998, Aggregate Tests Related to Asphalt Concrete
Performance in Pavements. National Cooperative Highway Research Program,
Report No. 405, Transportation Research Board, National Research Council,
Washington, D.C.
Williamson, Weyers, Mokarem, and Lane 14
11. Saeed, A., Hall, J . and Barker, W, 2001, Performance-related Tests of aggregates
for Use in Unbound Pavement Layers. NCHRP Report # 453, National
Academy Press, Washington, D.C.
12. Williamson, G., Weyers, R. E., Mokarem, D., Lane, D., and Cady, P., Investigation
of Testing Methods to Determine Long-Term Durability of Wisconsin Aggregate
Resources Including Natural Materials, Industrial By-Products, and
Recycled/Reclaimed Materials. August, 2005.
13. Kandhal, P.S., and Parker, F., 1998, Aggregate Tests Related to Asphalt Concrete
Performance in Pavements. National Cooperative Highway Research Program,
Report No. 405, Transportation Research Board, National Research Council,
Washington, D.C.




Williamson, Weyers, Mokarem, and Lane 15
TABLES AND FIGURES
Figure 1 Pit and Quarry Locations
Figure 2 Vacuum Saturated Absorption vs. Vacuum Saturated Specific Gravity
Figure 3 Vacuum Saturated Absorption vs. L.A. Abrasion
Figure 4 Vacuum Saturated Absorption vs. Micro-Deval
Figure 5 Vacuum Saturated Absorption vs. Aggregate Crushing Value
Figure 6 L.A. Abrasion vs. Micro-Deval
Figure 7 L.A. Abrasion vs. Aggregate Crushing Value
Figure 8 Vacuum Saturated Absorption vs. Sodium Sulfate Soundness
Figure 9 Vacuum Saturated Absorption vs. Unconfined Freezing and Thawing
Figure 10 Aggregate Durability Testing Flowchart for Unbound Aggregate
Figure 11 Aggregate Durability Testing Flowchart for Bound Aggregate in Portland
Cement Concrete
Table 1 Current WisDOT Testing Protocol
Table 2 Lightweight and Chert Percentages
TABLE 3 Current vs. Proposed Testing Protocol Comparison
Williamson, Weyers, Mokarem, and Lane 16

Quarry
Pit
FIGURE 1 - Pit and Quarry locations
Williamson, Weyers, Mokarem, and Lane 17
Vacuum Saturated Absorption vs. Vacuum Saturated Specific Gravity
R
2
=0.7539
0
1
2
3
4
5
6
7
2.5 2.55 2.6 2.65 2.7 2.75 2.8 2.85 2.9 2.95 3
Vacuum Saturated Specific Gravity
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n
Igneous/Metamorphic Carbonate Linear (Carbonate)

FIGURE 2 - Vacuum Saturated Absorption vs. Vacuum Saturated Specific Gravity




Vacuum Saturated Absorption vs. L.A. Abrasion
R
2
=0.572
0
1
2
3
4
5
6
7
0 10 20 30 40 50
L.A. Abrasion (% Loss)
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n

(
%
)
60

FIGURE 3 - Vacuum Saturated Absorption vs. L.A. Abrasion

Williamson, Weyers, Mokarem, and Lane 18
Vacuum Saturated Absorption vs. Micro-Deval
R
2
=0.8756
0
1
2
3
4
5
6
7
0 5 10 15 20 25 30 35 40 45
Micro-Deval (% Loss)
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n

(
%
)

FIGURE 4 - Vacuum Saturated Absorption vs. Micro-Deval




Vacuum Saturated Absorption vs. Aggregate Crushing Value (VT)
R
2
=0.547
0
1
2
3
4
5
6
7
10 12 14 16 18 20 22 24 26 28
Aggregate Crushing Value (% Loss)
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n

(
%
)
FIGURE 5 - Vacuum Saturated Absorption vs. Aggregate Crushing Value

Williamson, Weyers, Mokarem, and Lane 19
L.A. Abrasion vs. Micro-Deval
R
2
=0.5623
0
10
20
30
40
50
60
0 5 10 15 20 25 30 35 40 45
Micro-Deval (% Loss)
L
.
A
.

A
b
r
a
s
i
o
n

(
%

L
o
s
s
)

FIGURE 6 - L.A. Abrasion vs. Micro-Deval






L.A. Abrasion vs. Aggregate Crushing Value (VT)
R
2
=0.794
0
10
20
30
40
50
60
10 12 14 16 18 20 22 24 26 28
Aggregate Crushing Value (% Loss)
L
.
A
.

A
b
r
a
s
i
o
n

(
%

L
o
s
s
)

FIGURE 7 - L.A. Abrasion vs. Aggregate Crushing Value
Williamson, Weyers, Mokarem, and Lane 20
Williamson, Weyers, Mokarem, and Lane 21

FIGURE 8 - Vacuum Saturated Absorption vs. Sodium Sulfate Soundness







FIGURE 9 - Vacuum Saturated Absorption vs. Unconfined Freezing and Thawing FIGURE 9 - Vacuum Saturated Absorption vs. Unconfined Freezing and Thawing

Vacuum Saturated Absorption vs. Sodium Sulfate Soundness
R
2
=0.3249
0
1
2
3
4
5
6
7
0% 5% 10% 15% 20% 25% 30% 35%
Sodium Sulfate So
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n

(
%
)
undness (% Loss)
Vacuum Saturated Absorption vs.
0
1
2
3
4
5
6
7
0 2 4 6
Unconfined Freezing
V
a
c
u
u
m

S
a
t
u
r
a
t
e
d

A
b
s
o
r
p
t
i
o
n

(
%
)
Unconfined Freezing and Thawing
R
2
=0.1773
8 10 12 14 16
and Thawing (% Loss)
FIGURE 10 - Aggregate Durability Testing Flowchart for Unbound Aggregate

Absorption >2%
L.A. Abrasion >50%
Micro-Deval >25%

% Lightweight >5%
Yes
No
Yes
Yes
REJ ECT
ACCEPT
ACCEPT
No
No
Yes

Unconfined F&T >15%
START
No
Samples Accepted
(Based on WisDOT
performance ratings)

Poor 2
Intermediate 4
Good - 5
Samples Rejected
(Based on WisDOT
performance ratings)

Poor 5
Intermediate 2
Good - 0
Samples Accepted
(Based on WisDOT
performance ratings)

Poor 0
Intermediate 7
Good - 5

Unconfined F&T >15%
Yes
No
Williamson, Weyers, Mokarem, and Lane 22
illiamson, Weyers, Mokarem, and Lane 23

FIGURE 11 - Aggregate Durability Testing Flowchart for Bound Aggregate in Portland Cement Concrete
Absorption >2%
L.A. Abrasion >50%
Micro-Deval >25%

% Lightweight >5%
Yes
No
Yes
Yes
REJ ECT
ACCEPT
ACCEPT
No
START
No
F&T C 666 >40%
reduction in Freq. or
excessive Agg. Det.
Yes
No
Samples A
(Based on Wi
performa

Poor 2
Intermedia
Good - 5
ccepted
sDOT
nce ratings)
te 4
Samples Rejected
(Based on WisDOT
performance ratings)

Poor 5
Intermediate 2
Good - 0
Samples Accepted
(Based on WisDOT
performance ratings)

Poor 0
Intermediate 7
Good - 5
F&T C 666 >40%
reduction in Freq. or
excessive Agg. Det.
Yes
No
W
TABLE 1 Current WisDOT Testing Protocol
Test Failure Criterion
Lightweight Pieces in Aggregate (AASHTO T 113) (7) >5% Chert
Sodium Sulfate Soundness (AASHTO T 104) (7) >12% Loss
Soundness of Aggregates by Freezing and Thawing (AASHTO T 103) (7) >18% Loss
L.A. Abrasion (AASHTO T 96) (7) >50% Loss
Current WisDOT Testing Protocol


TABLE 2 - Lightweight and Chert Percentages
Sample # % Lightweight (3/8") % Chert Sample # % Lightweight (3/8") % Chert
2 12.91 0.00 24 2.14 0.23
4 0.00 0.00 55 4.62 1.62
6 0.00 0.00 57 9.09 0.00
8 2.25 0.21 59 9.20 0.00
10 0.00 0.00 60 3.08 2.50
50 16.19 0.00 31 8.38 1.78
52 7.80 0.00 33 0.5 0.00
11 3.41 1.58 36 0.00 0.00
12 2.35 0.56 39 0.00 0.00
13 0.5 0.00 40 3.86 3.86
14 0.5 0.00 42 5.45 1.25
16 0.5 0.00 48 4.79 0.65
19 1.28 0.23 64 5.38 0.00
21 0.5 0.00 68 0.95 0.85
22 5.24 1.83 69 0.5 0.00


TABLE 3 Current vs. Proposed Testing Protocol Comparison
Aggregate
Quality Rating
Rejected Accepted Rejected Accepted
Poor 3 4 5 2
Intermediate 1 12 2 11
Good 1 9 0 10
Current WisDOT Testing Protocol Proposed Protocol

Williamson, Weyers, Mokarem, and Lane 24