450390007
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
EPA-450/3-90-007
July 1991
Air
Guidance Document for
Selecting Antiskid Materials
Applied to Ice- and
Snow-Covered Roadways
j
-------
a
*
•* *
-------
EPA-450/3-90-007
Guidance Document
for Selecting Antiskid Materials
Applied to Ice- and Snow-Covered Roadways
Emission Standards Division
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
July 1991
-------
DISCLAIMER
This report has been reviewed by the Emission Standards Division
of the Office of Air Quality Planning and Standards, EPA, and
approved for publication. Mention of trade names or commercial
products is not intended to constitute endorsement or
recommendation for use. Copies of this report are available
through the Library Services Office (MD-35), U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711, or from
National Technical Information Services, 5285 Port Royal Road
Springfield, VA 22161. '
-------
CONTENTS
Preface jjj
Figures \\ \\\
Tables jx
1. Introduction 1_1
2. Current Practices for Skid Control 2-1
2.1 Antiskid materials 2-1
2.2 Application rates and procedures 2-13
2.3 Removal techniques 2-25
3. Measurement Methods for Physical Properties 3-1
3.1 Silt measurement methods 3-1
3.2 Durability measurement methods 3-2
3.3 Analysis methods for other properties 3-7
3.4 Measurement methods for deicing
chemicals 3.7
3.5 Modifications to standard test methods 3-10
4. Selection Criteria 4_1
4.1 Potential for dust emissions 4-1
4.2 Effectiveness of antiskid materials 4-3
4.3 Material durability 4-16
4.4 Initial acceptability criteria 4-21
5. Materials Evaluation 5-1
5.1 Origin of material samples 5-1
5.2 Material properties 5-1
5.3 Materials classification analysis .' 5-6
6. Simulated Traffic Tests 6-1
6.1 Material selection 6-1
6.2 Traffic tests ' 6-2
6.3 Small scale experiments 6-17
6.4 Data analysis 6-21
7. Conclusions and Recommendations 7-1
7.1 Conclusions 7-1
7.2 Final material selection criteria 7-3
7.3 Recommendations 7.4
8. References 8-1
-------
CONTENTS (continued)
Appendices
A. Survey procedures A-1
B. Bibliography ......... B-l
C. Alternative skid control measures C-1
D. ASTM silt analysis methods D-1
E. ASTM method for the Los Angeles Abrasion test F-1
F. State derived aggregate durability tests . . ^ F-1
G. ASTM Method E 384-89 for Vickers hardness G-1
H. ASTM Method E 534-86 for determination of percent
insoluble matter in sodium chloride H-3
I. Cost considerations for antiskid materials 1-1
VI
-------
FIGURES
Number
1 -1 Total particulate emission factor vs. average silt loading for an
artificially loaded paved road 1-2
1 -2 Diagram of street surface/atmospheric exchange of particulate
matter 1-4
4-1 Sieve analysis data of antiskid material in Helena parking lot 4-2
4-2 Total loading and silt loading vs. number of vehicle passes for
artificially loaded paved road 4-4
4-3 Profiles of total particulate concentration for artificially loaded
paved road 4-5
4-4 Increase in total surface loading vs. time after application of
salt or sand 4-6
4-5 Increase in silt loading vs. time after application of'salt or sand .... 4-7
4-6 TSP air quality impact of salt and sand application over time 4-8
4-7 Coefficient of friction (f) vs. number of wheel passes for four
antiskid materials compared on an equal volume basis 4-10
4-8 Skid resistance of limestone material vs. percent +200 mesh
silica • 4-11
4-9 . Average British portable skid number vs. rate of aggregate
application for various antiskid materials (30°F) "... 4-13
4-10 Average British portable skid number vs. rate of aggregate
application for various antiskid materials (0°F) 4-14
4-11 Average stopping distances determined on an ice track for
various antiskid materials 4-15
4-12 Mineral wear rate vs. Vickers hardness for SiO2 abrasive 4-18
5-1 . Scatterplot of silt content vs. LA abrasion loss 5-13
5-2 Scatterplot of percent void fraction vs. particle shape index 5-15
5-3 Percent deviation from mean value by cluster type - 5-17
6-1 MRI wheel passage machine 6-3
6-2 Percent moisture vs. equivalent vehicle passes for Samples 1-3 ... 6-11
6-3 Percent moisture vs. equivalent vehicle passes for
Samples 4 and 5 6-12
6-4 Percent moisture vs. equivalent vehicle passes for
Samples 8 and 9 6-13
6-5 Percent silt (% < 75 /imP) content vs. equivalent vehicle
passes for Samples 1, 1D, and 2 6-14
VII
-------
FIGURES (continued)
6-6 Percent silt (% < 75 /xmP) content vs. equivalent vehicle
passes for Samples 3, 4, and 5 6_15
6-7 Percent silt (% < 75 p.mP) content vs. equivalent vehicle passes'
for Samples 8 and 9 g-16
6-8 Correlation of change in mass median particle diameter with
LA abrasion loss g_2g
7-1 Decision "tree" for antiskid material selection 7-5
VIII
-------
TABLES
Number
2-1 Results of surveys of antiskid measures 2-2
2-2 Standard U.S. Screen Scales 2-6
2-3 Desirable characteristics of abrasives 2-7
2-4 Composition of water insolubles in rock salt 2-10
2-5 Annual average purity of rock salt, dry basis 2-10
2-6 Qualitative evaluation of chemicals and abrasives used in
snow and ice control 2-12
2-7 Chemical and sand use 2-14
2-8 State snow and ice control materials use 2-17
2-9 Highway salt sales by U.S. members of the Salt Institute (1989) .. . 2-21
2-10 Stormfighting guidelines of the Salt Institute 2-22
2-11 Guidelines for chemical application rates 2-23
3-1 Summary of instrumental methods of particle size analysis 3-3
3-2 Summary of test methods related to antiskid materials 3-8
3-3 Modifications to ASTM method C 131-89 3-11
3-4 Modifications to ASTM method C 131-89 for aggregate material
< 2.36 Mm 3-12
3-5 Silt analysis procedures 3-15
3-6 Modifications to ASTM E 384-89 to incorporate fine aggregate
materials [ 3-16
3-7 Modifications to ASTM E 534-86 to include calcium magnesium
acetate (CMA) 3-18
4-1 Results of impact tests 4-10
4-2 Material specifications for Alaska tests 4-12
4-3 Average mineral wear for different abrasive types 4-17
4-4 Acceptability criteria for antiskid materials 4-21
5-1 Sample inventory 5-2
5-2 Applicable test methods for antiskid materials 5-5
5-3 Results of laboratory evaluation 5-7
5-4 Results of materials evaluation 5.9
5-5 Range percent values for modified ASTM methods 5-11
5-6 Linear correlation coefficients (r)—material properties 5-12
5-7 Breakdown of test materials by characteristic type 5-16
5-8 Summary statistics—material properties by characteristic
cluster type 5-19
6-1 Antiskid materials exposed to simulated traffic 6-2
IX
-------
TABLES (continued)
6-2 Silt analysis procedures g_g
6-3 Moisture analysis procedure g_g
6-4 Sonic sieving procedure ' ] g_g
6-5 Test track material balance ' 6-18
6-6 Particle size data for traffic test No. 6 (rock salt) . . . . . . . . . . . . . . 6-19
6-7 Results of chemical analyses for NaCI tests 6-19
6-8 Results of small scale experiments 6-22
6-9 Particle size data for small scale experiments .'.... 6-23
6-10 Mass median diameter (/imP) of abrasive samples tested on
wheel passage machine 5.24
6-11 Linear correlation between MMD change and test material
properties 6_24
6-12 Mass balance for rock salt (NaCI) tests 6-27
7-1 Revised selection criteria for antiskid abrasives 7.3
-------
SECTION 1
INTRODUCTION
Several areas of the country which are in violation of the National Ambient Air
Quality Standard for PM10 (airborne particles less than or equal to 10 /zmA in diameter)
have conducted studies to identify the sources of these emissions. One source of PM10
emissions that has been identified in several of these studies is the resuspension of
antiskid material applied to active roadways.
Antiskid materials may consist of sand, stone, cinders, or other materials applied
to the road surface to improve traction on snow- and ice-covered roadways. "Deicers,"
on the other hand, serve to restore the traction associated with the road surface itself.
Prior research has established a direct relationship between the loading of fines
on a paved roadway and the PM10 emissions generated by vehicular traffic. In the initial
study, a salt/sand spreader was used to load a test road with: (a) pulverized top soil for
one test series; and (b) limestone gravel fines for a second test series (Cowherd et al.,
1977). Exposure profiling was used to quantify emissions over periods of 30 to 60 min
under controlled traffic conditions. The resulting relationship (Figure 1-1), between silt
loading and the total particulate emission factor represents conditions that closely parallel
the application and resuspension of antiskid materials consisting of insoluble abrasives.
The relationship between silt loading and PM10 emissions is reflected in the EPA-
recommended PM10 emission factors for paved urban roads. This relationship was
developed from a data base encompassing the results of tests conducted at eight sites
ranging from a freeway to a rural town road. According to EPA Publication AP-42, the
quantity of dust emissions from vehicle traffic on a paved roadway per vehicle kilometer
of travel may be estimated using the following empirical expression (USEPA, 1985):
0.8
,0.5j
where: e = PM10 emission factor (g/VKT)
L = total road surface dust loading (g/m2)
s = surface silt content (fraction of particles < 75 /zm in physical
diameter)
1-1
-------
10
IV)
8
£ 7
I
u
9
U_
c
O
E
LU
0)
o
"t:
0)
0.
15
0
0
Typical
Range for
Anti-Skid
Abrasives
10
Site: Stillwell Avenue
30 40 50 60 .70 80 90
Average Silt Loading (gm/m2) Excluding Curbs
Run?
100 110 120
Figure 1-1.
Total particulate emission factor vs. average silt loading for an artificially
loaded navfid rnad
-------
The total loading (excluding litter) is measured by sweeping and vacuuming lateral
strips of known area from each active travel lane. The silt fraction is determined by
measuring the proportion of loose dry road dust that passes a 200 mesh screen, using
a modified version of ASTM C 136 method. Silt loading is the product of total loading
and silt content.
Recently, in the absence of specific emission test results for antiskid materials,
PM10 emission factors from sanding and salting were estimated as a "gap filling" exercise
(Grelinger et al., 1988). The emission factor for sanding assumed that all of the PM10
content of typical road sand would be suspended. In the case of salting, the emission
factor was based on the assumption that: (a) 5% of the salt takes the form of a dried film
on the pavement; and (b) 10% of the film is liberated as PM10 particles. Because of the
uncertainties in these underlying assumptions, a low quality rating (E) was assigned to
the estimated emission factors.
It is clear from Equation (1-1) that techniques for controlling PM10 emissions
resulting from antiskid materials should be aimed at minimizing silt loading on the traveled
portion of the roadway (Figure 1-2). Specifically, reduced silt loadings may be expected
to result from snow/ice control programs that encompass improvements in three areas:
the properties of antiskid materials applied; the application protocols and procedures;
and the procedures for removal of the antiskid material from roadways. For example,
emission reductions may result from: substitution of different deicing materials for salt
and the use of antiskid materials that have been tested for durability and silt content;
lower application rates; and application of material to fewer roadways. Removal methods'
focus on the following roadway cleaning procedures: broom sweeping (wet or dry),
vacuuming, or water flushing. Water flushing (during periods with ambient temperatures
above freezing) followed by broom sweeping has proven to be most effective in capturing
silt-sized particles present in paved road surface material (Cowherd et al., 1988).
Local, state, and regional air pollution control agencies have requested information
on how to identify an appropriate antiskid material that is both durable and effective and
that produces fewer PM10 emissions. Thus the primary purpose of this study is to provide
guidance on methods to determine: (a) the physical properties and durability of antiskid
material selected for use on ice- and snow-covered roadways (see Figure 7-1); and
(b) criteria for defining the elements of an effective PM10 emission control strategy
associated with use of antiskid materials (see Table 7-1).
To meet the above objectives, the program was conducted in two phases
consisting of: (a) a literature search, telephone survey, and engineering analysis; and
(b) a laboratory evaluation of typical antiskid materials. Phase 1 of the study was
intended to collect sufficient background information to derive preliminary selection criteria
for antiskid materials based on existing data. The second phase of the program involved
1-3
-------
PARTICULATE ENTRAINMENT FROM URBAN STREETS
Background
Local
Vehicles
(Exhaust)
DEPOSITION
Sanding,
Salting,
Spills
Ground- Level
Suspended
Particulates"
(h< 10m)
Urban
Sources—
Conventional
& Fugitive
ENTRAINMENT
(By Wind & Vehicle Motion)
Vehicular Deposits
(Carryout from Unpaved
Areas, Tire, Wear, Oil,etc. )
Runoff. Mechanical Removal
(Sewers) • (Street Cleaners)
Figure 1-2. Diagram of street surface/atmospheric exchange of paniculate matter.
1-4
-------
an experimental study to: (a) determine the physical properties of typical antiskid
materials; (b) conduct suitable tests of selected materials to determine their potential for
the generation of PM10; and (c) refine the material selection criteria developed previously.
Each phase of the program is discussed in Sections 2 through 8 below.
Two survey methods were used in this study to gather the information needed to
meet the study objectives. First, an extensive literature survey was performed by on-line
accession of computerized informational data bases. Then a direct telephone survey was
targeted to states and municipalities known to be concerned with antiskid control was
performed. The methodologies for these surveys are described in detail in Appendix A.
For the purpose of describing abrasive materials, the following definitions will
apply:
1. Antiskid material: materials such as sand, stone, cinders, etc., spread on
an ice or snow-covered roadway to improve traction.
2. Deicing chemical: salts or similar compounds that control snow and ice by
preventing the formation of ice films, weakening the bond between snow
and the road surface, and/or by melting snow.
3. Antiskid abrasive: solid aggregate materials such as natural sand,
manufactured sand, or cinders applied to a road surface to provide an
immediate increase in skid resistance.
4. Sand: shall mean either "fine aggregate" or "manufactured sand,"
according to American Society for Testing and Materials definitions which
follow.
5. Fine aggregate: aggregate passing the 3/8-in sieve and almost entirely
passing the No. 4 (4.75 mm) sieve and predominantly retained on the
No. 200 (75 Aim) sieve.
6. Manufactured sand: the fine material resulting from the crushing and
classification by screening, or otherwise, of rock, gravel, or blast furnace
slag.
The organization of this document by section is as follows:
• Section 2 describes current practices for skid control.
Section 3 evaluates measurement methods for the physical properties of
antiskid materials.
1-5
-------
Section 4 develops the rationale for selection of effective antiskid materials
based on available literature.
Section 5 presents the physical and chemical properties of typical antiskid
materials obtained from local sources and state regulatory agencies.
Section 6 provides the results of the experimental testing program.
Section 7 presents the conclusions and recommendations derived from this
study.
Section 8 lists the references used in the study.
MRI-M\R9710-06.05 ~\ -Q
-------
SECTION 2
CURRENT PRACTICES FOR SKID CONTROL
Abrasives and deicing chemicals are the most commonly used antiskid materials.
These materials prevent formation of ice, melt ice that has formed; prevent buildup of
snowpack, or create a higher coefficient of friction for vehicle tires.
Some abrasives used include sand and cinders which provide a gritty surface on
snow and ice-covered roadways. Deicing chemicals are freezing point depressants that
melt ice and snow to achieve a wet surface to be maintained until drying occurs. The two
most commonly used deicing chemicals are rock salt (94% to 99% sodium chloride,
NaCI) and calcium chloride (CaCy.
Eleven states and municipalities were surveyed by telephone (some with multiple
contacts) to gather information on current practices for skid control, especially as related
to impact on dust production. In addition, a comprehensive literature review was
performed to document the use and effectiveness of abrasives, chemicals and other
antiskid materials, and methods. The survey procedures used in the program are
outlined in Appendix A with a bibliography of documents reviewed but not actually cited
in this report contained in Appendix B. Appendix C provides a brief overview of other
techniques potentially useful for ice and snow control.
This section documents results from the literature review, a review of results from
a survey undertaken by the South Dakota Department of Transportation and the
telephone survey. In this section the word "salt" applies only to rock salt; other chemical
salts such as calcium chloride are referred to by their chemical name.
2.1 ANTISKID MATERIALS
2.1.1 Survey Results
Table 2-1 presents a summary of the antiskid materials used by the two cities and
26 states who were surveyed by telephone and by the South Dakota questionnaire.
Almost all state and local agencies used rock salt and sand. One state, New Jersey,
used rock salt exclusively, stating that it was required for motorist safety and that sand
clogged drainage inlets. Many states mixed sand and salt in various proportions. Sand
was never used in urban areas exclusively. Higher proportions of salt to sand were used
in urban areas while lower proportions were applied in rural areas where traffic intensities
2-1
-------
10
Indianapolis
Minneapolis
Alaska
Connecticut
Delaware
Idaho
Indiana
Colorado
Salt
CaCI2 (32%)
Sand
Salt
Portland cement
Sand
Salt, sand, urea, "Qulcksalt,"
crushed stone
Salt
CaCI2
Screened pit
Crushed quarry
Washed stone sand
Washed concrete sand
Screened pit
Crushed pit
Crushed quarry
Screened cinders
Quarry cinders
Sand
Slag
Cinders
Salt
Sand
27.05/ton salt NS
0.38/gal CaCI2
2.00/ton sand
25.00/ton salt 2
NS 2
34.79/ton salt 5
98.04/ton CaCU
8.43/yd3 sand
NS • 4
NS No specification
NS
NS
NS
NS
NS 3, 6, or 7, dependent on
NS sand class
NS
NS No specification
NS
xJluvmy IIIUUIUU
Indiana DOT
Section 903 of
highway manual
ASTM C136
NS
AASHTOT-11
•
NS
NS
NS
NS
comments
Salt/sand Is used in outlying areas only;
only salt applied to city roads since sand
clogs sewers.
Washed sand must drain 12 h; sand
must be free of organic Impurities and
colorplate < No. 2; maximum 2,5%
shale, strength ratio of 1 .00+ as
measured by ASTM C87-83.
Durability-based on shaker test from
Washington state.
"Clean, hard, and durable" sand.
No clay, dirt, loam, or frozen lumps.
Washed sand must sit for 24 h.
Iowa
Screened pit
NS
3
(continued)
NS
No crushed material used.
-------
TABLE 2-1 (continued)
CO
State/city
Maine
Massachusetts
Michigan
Minnesota
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New York
Types of material
Screened pit
Crushed pit
Screened pit
Washed pit
Screened pit
Screened/washed pit
Screened pit sand
Pit run
Screened pH
Crushed pit
Maintenance aggregate (> '/»*)
Screened pit
Boiler slag
Screened pit
Screened pit
Washed sand
Salt
Pit run
Screened pit
Crushed pit
Crushed quarry
Iron ore tailings
Crushed slag
Crushed cinders
Cost ($)
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
4.25/ton sand
NS
NS
NS
NS
NS
NS
Abrasives (maximum %
< 200 mesh)
No specification
3
3
NS
10 < 100 mesh
10
6
a
12 < 4 mesh
5
Sieving method
NS
NS
NS
NS
NS
NS
NS
NS
New York DOT
703-1 P
703-2P
Comments
—
—
No crushed stone; < 7% loss by
washing; similar to 2 NS materials In
aggregate table. Fineness modulus 2.50
to 3.35.
—
Unit weight tested for, but no
specifications.
No crushed material.
—
Must be washed; sharp "feel" desirable.
No sand used because of clogging of
Inlets and drainage, and motorist safety.
Must be free of too much clay, loam,
etc.; particles must not degrade In
service and must visually contrast with
Ice/snow.
(continued)
-------
TABLE 2-1 (continued)
State/city
Ohio
Oregon
Utah
Vermont
K
Washington
West Virginia
Wisconsin
Wyoming
_
NS = Not specified.
Types of material
Screened pit
Crushed quarry
Cinders
Blast furnace agg
Slag
Crushed quarry
NS
Screened pit
Crushed quarry
Pit run
Crushed pit
Crushed quarry
Pit run
Screened pit
Crushed stone
Boiled slag
Pit run
Screened pit
Screened pit
Crushed pit
— -
Cost ($)
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
Abrasives (maximum %
< 200 mesh)
30 < 50 mesh
Sieving method
NS
Comments
10 or 15
10
4 or 5
8 < 100 mesh
5 < 100 mesh
15
DSHD 101
MSHTOT-11
NS
NS
NS
NS
NS
NS
Cubical aggregate specified.
Portion through No. 40 sieve shall be
nonplastic by AASHTO HTO T-90; sand
equivalent 80+ (UDOT 8-938); AASHTO
T-19 loose weight.
Minimum of 1 face for aggregate.
Low fines present fewer problems with
water retention and plugging.
Plastic Index < 6.
Free of organic material.
-------
are less. Agencies typically reported their selection of antiskid materials was based on
what was available (and, presumably, at reasonable cost).
2.1.2 Abrasive Usage and Specification
As stated in Section 1, the potential for dust emissions from abrasives can be
estimated from the silt fraction (i.e., percentage that passes a No. 200 mesh or < 74 )m
diameter). Table 2-2 compares U.S. standard mesh sizes and gives the nominal aperture
widths. The entire PM10 fraction contained in the silt of the applied abrasive can be
assumed to become airborne. Ratios of PM10/PM75 for common abrasives have not been
reported in the literature. However, an analysis of naturally-occurring, western sandy soils
produces an average ratio of 0.0026. This ratio times the silt fraction times the quantity
of sand applied gives an estimated 7.5 g/metric ton (0.018 Ib/ton) emission factor for road
sanding developed by Grelinger (1988).
Another measure of dust potential is the hardness or durability of the abrasive
material. Soft particles tend to crumble as they are abraded between vehicle tires and
pavement, while hard, angular particles tend to remain integral.
Sand is the most common abrasive applied to roadways. It is a cheap commodity
that is taken from pits, quarries, rivers and beaches. Sand sources are often close to
roadways, reducing transportation costs. Some states maintain their own sources for
road sand. Abrasives such as sand not only produce initial surface friction, but also
break up ice or packed snow as vehicles travel over the roads.
"Good" abrasives are described by Keyser (1973) as having "resistance to
degradation, angular shape, dark color and uniform grain size." He also concludes that
fines passing a No. 50 sieve contributes almost nothing to skid resistance. Tests
reported by Hegmon and Meyer (1968) indicate that the highest coefficients of friction are
produced by particles between the No. 8 and the No. 16 sizes, and that a slight
improvement in skid resistance is realized when particles passing a No. 50 sieve are
removed. Table 2-3 provides desirable characteristics of antiskid abrasives (Schneider,
1959).
Most agencies specify a top size of 3/8 in diameter for abrasives, but the maximum
fine particle (passing a 200 mesh or < 75 urn diameter) specification for sand ranges
from 2% to 15%. Some agencies reported that no specifications for fines are used, but
sometimes the sand must be washed.
Certain agencies specify "clean, hard, and durable" material with a "sharp" feel,
but most do not have quantitative specifications for hardness, durability, or number of
faces. One state reported the use of the Los Angeles abrasion test by cities, and said
the same
2-5
-------
Microns
1
2.5
5
10
20
37
44
53
62
74
88
105
125
149
177
210
250
—
Inches
.00004
.0001
.0002
.0004
.0008
.0014
.0017
.0021
.0024
.0029
.0035
.0041
.0049
. .0059
.0070
.0083
.0098
i.—
U.S.
standard
12,500
5,000
2,500
1,250
625
400
325
270
230
200
170
140
120
100
80
70
60
Microns
297
350
420
500
590
710
840
1,000
1,190
1,410
1,680
2,000
2,380
2,830 .
3,360
4,000
4,760
Inches
.0117
.0138
.0165
.0197
.0232
.0280
.0331
.0394
.0469
.0555
.0661
.0787
.0937
.111
.132
.157
.187
Uq
.0.
standard
50
45
40
35
30
25
20
18
16
14
12
10
8
7
6
5
4
2-6
-------
TABLE 2-3. DESIRABLE CHARACTERISTICS OF ABRASIVES
(SCHNEIDER, 1959)
Characteristic
Rationale
Great resistance to compression,
crushing, impact, and grinding
Resist degradation under the action of
traffic; could be recovered in spring
Angular shape
Greater stability; prevent its being blown
away
Darkish color
Absorbs heat to melt itself into the
surface of ice
Uniform grain size
Uniform spreading pattern; less likely to
damage equipment
pits tested also supply sand for both local and state highways. Occasionally sand
rejected for use in concrete or mortar is used for road antiskid material.
Cinders (steel and silver slag; boiler bottom ash) are used by several agencies.
One state commented that cinders are the best abrasive since they are black and fine,
and absorb sunlight to melt ice and snow. But one agency said they are not able to use
copper slag because of heavy metal content. Another said that if citizens can "see" the
deposited material, they credit the highway department with "doing their job." An
additional comment was that this precludes the use of "white" limestone that cannot be
seen on snow and ice.
A study done by Eck (1986) reported that four tested abrasives—boiler house
cinders, coke cinders, sand and stone—are essentially the same in effectiveness and that
selection should be based on cost factors. Eck also reported that one state used
sawdust as an antiskid material and New York reported using iron ore tailings. Montana
reported they did not reuse reclaimed sand from roads because of the size gradation
requirement, but another state said that up to 10% of reclaimed sand can be used.
Objections to abrasive use found in the literature search include:
1. Cinders are quite bulky and frequently nonuniform in size, causing
problems with delivery systems.
2. Cinders are usually acidic and can cause corrosion.
3. Cinders are easily blown away because of their light weight.
2-7
-------
4. Cinders often contain significant levels of toxic heavy metals.
5. Both cinders and sand are deposited in catch basins and drain conduits
requiring laborious efforts to remove them. conauus,
6. Coarse sand (> 3/8 in) can break windshields by being thrown into the air
7. Sand smothers roadside vegetation.
8. Sand causes silt buildup in ponds and lakes.
9. All abrasives form a thick ice mat that will melt near warm spots such as
manholes, causing deep holes and ruts that are traffic hazards.
1 0. All abrasives require heavy application amounts (in 'comparison with deicinq
chemicals) to achieve good skid resistance.
11. All abrasives should be cleaned from roadways after storm periods.
2.1.3 Deicino Chemicals
a H K The P°tential ,for dust emissions from deicing chemicals is based on the amount
and type of chemical that can be suspended by tire and wind action. Deicing chemicals
form a hqu.d solution as ice is melted. Much of this water/chemical solution will run off
the roadway, but some will be sprayed into the air by traffic. Evaporated droplets may
constitute a source of airborne particulates, but a larger source is likely to be the film of
solid residue left on the pavement surface upon evaporation of any remaining
water/chem,cal solution. This dry film residue will be a source of suspended paniculate
upon contact with vehicle tires and wind. y«uwJiaie
Rock salt, the most commonly used deicing chemical, achieves good skid control
results with far less quantity than abrasives. It is also the cheapest deicing chemical
available, but does cause corrosion of vehicles and metal structures. Major rock salt
deposits are found in New York, Michigan, Kansas, Texas, and Louisiana. Solar salt
ponds on the Great Salt Lake also supply some road deicing salt. Salt used in the
shi r edfro6 °f Ma'n6> h°wever> C0mes from Spain and Canada- Salt used in Alaska is
Much of the sodium chloride in rock salt is believed to wash off the roadway since
salt is very soluble in water. Even some suspended impurities in rock salt are likely to
run off with the melted snow/ice. However, some fraction of sodium chloride and the
insoluble impurities in rock salt are likely to be deposited on the pavement upon roadway
drying, to be suspended later by tire or wind action. Few data were found in the literature
concerning quantities of sodium chloride or insoluble matter left on roadways after
2-8
-------
deposition of rock salt, or the size fraction of this material when suspended. One study
(Keyser, 1981) claimed that an advantage of rock salt was that it "freezes dry on
pavement surface."
An estimate of PM10 emissions from the sodium chloride in rock salt has been
reported as 10 Ib/ton of salt applied, based on an estimate of 5% of the salt remaining
as a dried film on the road pavement, and 10% of this film being suspended as PM10
(Grelinger et al., 1988). No field data were available for validation of this estimate.
The sodium chloride content is about 99% for rock salt mined in Louisiana, 98%
in New York and Michigan, and 94% to 98% in Kansas. Proportions of the remaining
natural impurities found in rock salt are shown in Table 2-4. Table 2-5 shows that small
particles of rock salt are richer in NaCI than larger particles.
Calcium sulfate, the major impurity in rock salt, is present as small lath-like crystals
or crystal fragments of 100 /im average length which are slowly and difficultly soluble.
Highly dissolved sulfate occurs with fine grinding and long dissolving time (> 30 min) of
rock salt (Kaufmann, 1968). Both of these conditions are satisfied with salting of
roadways.
Other impurities in rock salt, including silica, alumina, ferric iron, and dolomite, are
not water soluble. Many quartz grains are very small (< 10 pm in diameter). It is
possible that only the truly insoluble impurities left on roadways after runoff and
evaporation will constitute a significant source of suspended particulate. Table 2-5 shows
how the purity of rock salt is dependent on its source.
Additional impurities are added by salt mines- for anticaking properties.
Chromates, phosphates, and similar compounds may also be added to rock salt as
possible corrosion inhibitors.
Calcium chloride is the second most common deicing chemical. It is often mixed
with sand and rock salt, either as a pellet or as a liquid, to produce a more effective
deicing mixture at low ambient temperatures. The eutectic temperature1 of calcium
chloride is -55°C (-67°F), considerably lower than that of sodium chloride. Connecticut
reported that calcium chloride is applied only in watershed areas as three parts salt and
one part calcium chloride. Indianapolis reported 5% of calcium chloride is mixed with
95% salt for application to city streets.
One report noted that calcium chloride added to rock salt produced less visible
salt residue both on the pavement and shoulders. The Legislative Research Council of
Massachusetts (1965) stated that "this salt residue is apparent on the pavement for
several days after a storm; it was, as late as spring, still obvious on the shoulders and
1 Eutectic temperature is the lowest point at which equilibrium is achieved between the
solid and liquid phases.
•
2-9
-------
TABLE 2-4. COMPOSITION OF WATER INSOLUBLES IN ROCK SALT
(PERCENT BY WEIGHT)
Avery Island, . Detroit,
Louisiana Michigan
2.56
0.12
0.61
Trace
96.66
3.06
0.73
1.71
6.37
86.85
Quartz (silica), SiO
Iron oxide, Fe,0
Magnesium carbonate, MgCO
Calcium carbonate, CaCO
Calcium sulfate, CaSO4
Reference: Kaufmann (1968).
Retsof, New
York
•
8.61
3.04
4.14
2.50
81.16
Kansas
1.4
0.7
TABLE 2-5. ANNUAL AVERAGE PURITY OF ROCK SALT, DRY BASIS
(PERCENT BY WEIGHT)'
Sizeb
No. 2
No. 1
CC
FC
Reference: Kaufmann (1968).
Detroit, Michigan Retsof, New York Avery Island, Louisiana
(12-year average) (14-year average) (representative composite)
96.705 97.658 99.073
97.388 98.101 99.094
98.193 98.320 99.144 (A size)
98.212 98.325 99.227 (C size)
b No. 2 = 0.42-0.50 in; No. 1 = 0.285-0.42 in; CC = 0.075-0.285 in; FC = < 0.075
2-10
-------
mall along the pavement edge in the other two sections. It is assumed that it was
because of the use of calcium chloride that very little of this salt recrystallization was
noted in the chemical mix test section."
Prilled urea and Quicksalt® (a reduced corrosion deicer) were the other two
chemicals found to be used by agencies during the telephone survey. Both were used
in Alaska for specialized purposes. The urea was obtained from a plant on the Kenai
peninsula and shipping costs were presumably less than the shipping costs of salt from
Seattle. The literature produced names of other chemicals used for roadway deicing, and
these included liquid chemicals such as ethylene glycol, propylene glycol, and alcohols
which would not produce any particulate emissions. These are discussed further in
Appendix C.
Objections to common chloride salt deicing chemicals include:
1. Vehicle bodies and bridges are corroded.
2. Concrete surfaces deteriorate.
3. Vegetation is damaged.
4. Water supplies are contaminated.
2.1.4 Mixtures of Abrasives and Chemicals
The most common deicing mixtures are those of rock salt, sand, and calcium
chloride. Table 2-6 presents a qualitative evaluation of chemicals and abrasives used for
snow and ice control (Keyser, 1981). Sand is almost never applied by itself, but is mixed
with a minimum of 5% rock salt to keep moisture in the sand from freezing. If melting
action is desired, higher concentrations of salt are used. Salt and sand are usually
premixed before a particular storm is forecast and left to set and cure for "better" action.
During the winter of 1988/1989, a 10% salt/90% sand mixture was tried in Denver,
Colorado area, but the telephone responder said this mixture did not give good
performance, so they have returned to an 18% salt/82% sand mixture. In Maine, 30 to
110 Ib of rock salt are mixed per cubic yard of sand. In Connecticut, 300 Ib of salt are
mixed with 1,264 Ib of sand for a 2-lane mile application on multilane systems. Eck
(1986) reported, "the overall mean quantity of chemical used per cubic yard of abrasive
is 225 Ib..."
Salt cannot be used in very cold temperatures because its eutectic temperature
is -21.1°C (-5.8°F). Addition of calcium chloride with a much lower eutectic temperature
of -55°C (-67°F) makes a mixture able to melt snow and ice at lower temperatures and
is also believed to make the mixture "stick better" to pavement. The residue of calcium
chloride has been reported to provide "sufficient concentration to prevent ice bonding
for up to 6 days after treatment" (Transportation Research Board, 1974 and 1984).
2-11
-------
Sodium chloride (rock salt), To melt snow and Ice. Very effective when the temperature
Calcium chloride, CaCI2
. Provides Immediate traction
Is above -3.8°C; effective between Salt particles bore, penetrate, and
-3.8° and -9.5°C; marginal between undercut the ice layer.
-9.5° and -12.3°C; and not effective Freezes dry on pavement surface
below -12.3°C. Low cost.
To melt snow and Ice. Normally used when the temperature High rate of solution
'8™ !,°~ -12'3°C: effective down to Liberates heat on going Into solution.
-29.1 C; marginal between -29.1" Effective at low temperatures
and -34.7°C; and not effective below
-34.7°C.
Mixtures of NaCI and CaCL To melt snow and Ice.
In cold weather, down to -17.9°C
when snow and Ice must be melted
In a short time.
.it. Mixtures of abrasives and
to salt; abrasives treated with
salt
Abrasives
High rate of solution.
Effective at low temperatures.
Chemical more stable on the road.
To increase the sliding In very cold weather, when salt Is not Free-flowing material
friction immediately. effective or where clean plowing is No freezing of stockpiles
Impossible if Immediate protection Is Abrasives more stable on the road.
necessary. Qu|ck anchoring of abrasive to the
road.
Improve skid resistance immediately.
Low rate of solution.
Ineffective at very low temperatures.
High cost.
Melting action could take place at the
ice surface.
Pavements remain wet.
High cost.
Pavement stays wet longer than with
CaCI2.
Creates spring cleanup problems.
Does not remove Ice or snow which
causes slipperiness.
May damage vehicles traveling at
high speed.
-------
2.2 APPLICATION RATES AND PROCEDURES
The U.S. Transportation Research Board (TRB) has twice printed the manual on
"Minimizing Deicing Chemical Use" (1974 and 1984) that acknowledges that application
rates are highly dependent on several factors. These include:
1. Level of service required.
2. Weather conditions and their change with time.
3. State and characteristics of chemicals used.
4. Time of application.
5. Traffic density at time of, and subsequent to, chemical application.
6. Topography and the type of road surface.
Table 2-7 presents a summary of chemical and sand use and application rates for
27 states, four Canadian provinces, and certain toll roads, counties, and cities published
in 1974. The most comprehensive survey of deicing chemical and abrasive usage in the
U.S. was collected by the Salt Institute (1984) for two winters in the early 1980s. Data
collected in this survey are summarized in Table 2-8. Finally, highway salt sales by
members of the Salt Institute are presented for 12 yr in Table 2-9 (Salt Institute, 1989).
Rock salt is applied routinely in cities with high traffic volumes, on icy spots such
as hills and entrances to intersections, or where storm sewers are located. The telephone
survey found application rates of 350 Ib/lane-mile for Maine and 432 lb/2-lane mile in
Connecticut for multilane systems with high traffic intensity.
A salt/sand mixture (18% rock salt, 88% sand by volume) applied at 300 Ib/lane-
mile is recommended in the Denver, Colorado area, but Colorado sand trucks do not
have-control levers for measured application intensities. Table 2-10 illustrates the
recommended salting/sanding rates of the Salt Institute (1986).
Recommended rates of sand application found in the literature and phone survey
ranged from .5 to 2 ton/lane mile. Maine applies a mostly sand mixture (30 to 110 Ib of
salt/1 yd3 of sand) at 1 ydVlane mile. Connecticut reported the application of 300 Ib of
salt mixed with 1,264 Ib of sand for a 2-lane mile application on multilane systems. Eck
(1986) reported that the overall mean quantity of chemical used per cubic yard of
abrasives is 225 Ib, with Canadian usage appreciably higher at a mean of 335 Ib. The
chemicals used for these means were not given. [The conversion factor from cubic yards
ofsand to tons is approximately 1.3.] Finally, the American Association of State Highway
and Transportation Officials have published guidelines on the use of antiskid materials
based on road and environmental conditions. These are shown in Table 2-11 (AASHTO,
1976).
To minimize deicing chemical use for environmental reasons, the Transportation
Research Board (1974 and 1984) reports that agencies use liquid chemicals such as
calcium chloride applied to rock salt, control application rates using calibrated spreaders,
2-13
-------
Agency
^" .•!•.
STATES
California
Connecticut
Idaho
Illinois
Kansas
Maine
Massachusetts
Minnesota
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New York
North Dakota
Ohio
NaCI
Year (tons)
69/70 22,000
72/73 31,000
72/73 87,600
8,000
69/70 309,900
70/71 256,411
71/72 280,930
69/70 74,846
70/71 88,817
71/72 85,692
70/71 172,000
71/72 196,000
72/73 127,000
69/70 172,720
70/71 148,712
71/72 116,664
69/70 76,765
70/71 60,709
71/72 68,837
2,850
72/73 18,000
69/70 153,742
-70/71 144,982
71/72 157,238
72/73 96,983
35,000
69/70 192,384
70/71 205,744
71/72 315,000
70 4,800
71 4,100
72 4,000
69/70 510,000
70/71 470,000
71/72 424,000
CaClg Sand
(tons) (tons) Other
5,480
3,380
3,390
208,000 36,000
195,000 36,000
116,000 26,000
2,906
2,367
1 ,381 283,552
5,942
5,069
4,456
45
800
8,000
2,285
2,535
4,920
1.000 32,000
1-375 17,000
650 174,000
2,600
2,300
3,000
Tons/lane
mile Application rates
400 to 600 Ib/mile on ice and
compacted snow
400 lb/2-lane mile
500 lb/2-lane mile
500 to 1 ,000 lb/2-lane mile of 3:1
NaCI-CaClg; abrasives with 10%
to 15% salt at 1,500 Ib/mile
19.4
23.0
22.2
600 to 800 Ib/lane mile sand-salt;
4.27 400 to 500 Ib/lane mile salt
10.38
400 lb/2-lane mile (maximum)
100 to 300 Ib/lane mile
200 to 500 lb/2-lane mile
0.5 yd3/2-lane mile
19.5
18.3
19.8
12.2
150 to 500 Ib/lane mile
200 to 600 lb/2-lane mile
9.0
12.7
11.7
10.6
(continued)
2-14
-------
TABLE 2-7 (continued)
Annual totals
Agency
Oregon
Pennsylvania
South Dakota
Vermont
Virginia
Washington
West Virginia
Year
69/70
70/71
71/72
69/70
70/71
71/72
71
72
73
69/70
70/71
71/72
72/73
69/70
70/71
71/72
70/71
71/72
72/73
69/70
70/71
71/72
NaCI
(tons)
810
710
720
565,750
738,800
652,500
1,536
1,890
2,804
110,000
88,000
97,000
88,260
56,097
66,243
44,518
7,051
15,251
16,665
152,453
148,570
112,244
CaClg Sand
(tons) (tons)
180
260
None
26,500
31,100
29,550
323 40,507
368 43,400
380 46,400
22,211 252,568
18,987 223,028
9,151 112,573
10,045
7,325
2,380
Tons/lane
Other mile Application rates
400 Ib/lane mile max.
60 to 80 Ib/Type 1
CaCl^on sand
300 to 500 lb/2-lane mile
21 .1 300 to 800 lb/2-lane mile;
16.3 546 lb/2-lane mile (average)
17.6
16.0
25°to32°F: 200 to 250 Ib/lane
mile NaCI; 10° to 25°F: 250 to
300 Ib/lane mile 3:1 NaCI-CaClg;
< 10°F: 350 to 450 Ib/lane mile
CaC^ flake or 300 to 375 Ib/lane
mile CaClg pellet
500 to 600 Ib/mile
Wyoming
PROVINCES
Alberta
B.C.
H.S.
Ontario
69/70 16,494
70/71 26,910
71/72 23,461
45,000
69/70 59,448
70/71 121,328
71/72 145,755
70/71 363,264
71/72 419,137
72/73 336,959
324
240
50
7.1
13.9
15.4
2,000 Ib sand-salt per 2-lane mile
500 lb/2-lane mile NaCI max.
100 to 300 Ib/lane mile
300 to 600 lb/2-lane mile; 70/71
ave.: 490; 71/72 ave.: 450
Sand: 2,000 lb/2-lane mile;
salt: 450 lb/2-lane mile
(continued)
2-15
-------
TABLE 2-7 (continued)
Agency
Year
NaCI
(tons)
44,000
TOLL ROADS
Roads
Illinois State Toll
Highway Authority
New York Thruway 69/70 107,958
70/71 120,486
71/72 121,904
Ohio Turnpike
COUNTY
Hennepin
(Minnesota)
CITIES
Milwaukee
New York City
Seattle
Toronto
69/70 26,645
70/71 27,162
71/72 24,638
70/71 11,252
71/72 7,541
72/73 5,260
Sand
(tons) (tons)
Tons/lane
Other mile
691
7,740
37.2
41.6
42.3
1,335
1,363
837
29,498
69/70
70/71
71/72
70/71
71/72
72/73
69/70
70/71
71/72
70/71
71/72
72/73
39,951
43,257
38,940
128,928
129,946
30,300
27
2,320
4,120
65,570
66,141
38,000
19
3
44
Application rates
500 to 600 Ib/lane mile
400 Ib/mile for snow; 200 Ib/mile
of 2:1 NaCI-CaClg for freezing rain
250 Ib/lane mile max.
100 to 450 Ib/lane mile
1 oz/yd2 (salt only)
600 lb/2-lane mile
700 Ib/lane mile
2-16
-------
TABLE 2-8. STATE SNOW AND ICE CONTROL MATERIALS USE (SALT INSTITUTE, 1984)
Winter 1981-1982
Calcium chloride
State
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Idaho
Illinois
Indiana
Iowa
Kansas
Total lane
miles Pavement miles
1,020
16,650
34,852
53,000
32,000
10,160
9,971
41,809
11,512
38,515
31,036
24,300
22,371
900
0
0
12,900
32,000
0
2,543
13,900
0
14,559
0
5,688
Salt
(tons)
450
394
2,510
13,600
22,460
103,201
8,913
18,500
11,000
304,184
313,365
64,000
35,490
Dry
(tons)
300
44
576
25
0
600
0
400
0
280
204
2,260
0
Liquid
(gai)
250,000
0
0
0
0
0
0
0
0
228,500
88,252
24,000
0
Abrasives
(tons)
18,225"
56,714
16,597
200,000
351,519
380,641"
18,000
12,200
228,000
278,193
145,000
Salt
(tons)
355
413
856
10,896
51,934
7,055
8,200
11,000
206,000
116,650
60,400
31,630
Winter 1982-1983
Calcium
Dry
(tons)
250
25
246
0
600
0
60
0
520
44
2,225
0
chloride
Liquid
(gai)
200,000
0
0
0
0
0
0
0
115,150
90
9,500
0
Abrasives
(tons)
18,225"
56,000
9,155
434,837
169,598"
7,714
7,800
192,000
108,619
116,000
65,000
(continued)
-------
TABLE 2-8 (continued)
State
Kentucky
Louisiana
Maine
Maryland
Massachusetts
Michigan
Minnesota
ro
J^ Mississippi
00
Missouri
Montana
Nebraska
Nevada
New Hampshire
New Jersey
New Mexico
Total lane
miles
53,846"
38,191
7,877
14,600
12,000
13,667
28,724
23,391
69,664
18,790
22,000
12,608
8,630
10,366
27,450
Calcium chloride
Pavement miles
4,966b
0
1.178
0
4,695
4,162
0
18,790
11,340
8,406
10,366
20,000
Salt
(tons)
73,275
827
51,676
155,758
262,000
397,000
118,587
332
90,963
2,817
22,221
8,500
138,692.
54,500
16,000
Dry
(tons)
550
0
600
1,914
3,600
210
310
4,615
16
464
0
293
980
0
Liquid
(gai)
0
0
0
0
0
0
6
34,642
0
0
273,000
0
Abrasives
(tons)
0
519,750
71,400
184,000
17,000
291,204
160,000
84,877
51,000
210,135
12,400
64,000
Salt
(tons)
32,964
23
49,202
82,499
178,500
229,000
127,957
285
75,111
3,245
24,899
9,831
93,813
35,700
23,000
Winter 1982-1983
Calcium chloride
Dry
(tons)
107
0
535
978
2,842
267
280
3,373
28
556
0
228
580
0
Liquid
(gai)
o
n
0
0
0
o
o
25,974
0
o
228,000
0
Abrasives
(tons)
472,500"
30,859
95,000
10,000
288,893
ODD nnn
86,734
51 000
1*51 ?99
4200
80,000
(continued)
-------
TABLE 2-8 (continued)
Winter 1981-1982
Calcium chloride
State
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
ro Pennsylvania
«O Rhode Island
South Carolina
South Dakota
Tennessee
• Utah
Vermont
Virginia
Washington
Total lane Salt
miles Pavement miles (tons)
29,780
112,573
15,800
42,192
25,935
17,895
77,000
3,015
84,450
18,216
25,087
22,000
6,079
112,814
16,778
29,780
23,342
0
0
3,015
3,450
25,087
6,079
17,350
0
443,000
45,264
8,222
401,285
9,300
0
500,000
56,280
1,988
4,345
51,000
79,540
71,904
178,500
10,000
Dry
(tons)
800
192
0
1,175
0
0
5,000
88
388
7
740
0
0
2,000
0
Liquid
(gai)
0
0
69,500"
224,014
449,016
0
0
0
0
0
0
0
0
0
Abrasives
(tons)
899,000 .
35,000
239,087
36,000
1,200,000
86,094
10,000
52,920
51,000
125,000
154,648*
400,000
360,445
Salt
(tons)
300,000
36,573
8,719
184,341
18,770
456
231,000
29,297
897
3,697
79,720
65,647
95,000
7,500
Winter
Calcium
Dry
(tons)
300
160
217,800b
195
0
5,000
132
286
5
0
0
1,000
0
1982-1983
chloride
Liquid
(gao
0
0
0
113,291
424,988
0
0
0
0
0
0
0
0
0
Abrasives
(tons)
460,000
38,000
128,976
72,000
655,000
45,000
4,750
45,002
125,000
106,654s
255,000
203,531
(continued)
-------
TABLE 2-8 (continued)
State
West Virginia
Wisconsin
Wyoming
Total lane
miles
70,000
25,774
15,743
Winter 1981-1982
Calcium chloride
Pavement miles
21,000
25,774
0
======
Salt
(tons)
90,636
236,790
5,000
=====
Dry
(tons)
1,009
755
0
=====
Liquid
(gai)
0
70,000
0
•
Abrasives
(tons)
239,655
26,585
113,000
==____
Salt
(tons)
52,709
229,803
6,340
Winter 1982-1983
Calcium
Dry
(tons)
314
648
0
chloride
Liquid
(gai)
0
40,000
0
Abrasives
(tons)
178,642
25,913
127,000
• Cubic yards of abrasives converted to tons on the basis of 1 yd3 equals 2,700 Ib-for calculation purposes only.
b Includes parkways.
8
-------
TABLE 2-9. HIGHWAY SALT SALES BY U.S. MEMBERS
OF THE SALT INSTITUTE (1989)
Year of record
7/77-6/78
7/78-6/79
7/79-6/80
7/80-6/81
7/81-6/82
7/82-6/83
7/83-6/84
7/84-6/85
7/85-6/86
7/86-6/87
7/87-6/88
7/88-6/89
Quantity sold (1 03 tons)8
10,900
11,300
8,110
7,240
9,790
7,270
10,300
10,700
11,100
9,460
11,100
10,200
8 Rounded to three significant figures.
2-21
-------
TABLE 2-10. STORMFIGHTING GUIDELINES OF THE SALT INSTITUTE
Environmental condition
^•«^™B.^^M
CONDITION 1:
Temperature
Near 30
Precipitation
Snow,, sleet, or freezing rain
Road Surface
Wet
CONDITION 2:
Temperature
Below 30 or falling
Precipitation
Snow, sleet, or freezing rain
Road Surface
Wet or sticky
CONDITION 3:
Temperature
Below 20 and falling
Precipitation
Dry snow
Road Surface
Dry
CONDITION 4:
Temperature
Below 20
Precipitation
Snow, sleet, or freezing rain
Road Surface
Wet
CONDITION 5:
Temperature
Below 10
Precipitation
Snow or freezing rain
Road Surface
Accumulation of packed snow or ice
Recommended application
If snow or sleet, apply salt at 500 Ib per
two-lane mile. If snow or sleet continues
and accumulates, plow and salt
simultaneously. If freezing rain, apply salt
at 200 Ib per two-lane mile. If rain
continues to freeze, reapply salt at 200 Ib
per two-lane mile.
Apply salt at 300 to 800 Ib per two-lane
mile, depending on accumulation rate. As
snowfall continues and accumulates, plow
and repeat salt application. If freezing rain,
apply salt at 200 to 400 Ib per two-lane
mile.
Plow as soon as possible. Do not apply
salt. Continue to plow and patrol to check
for wet, packed, or icy spots; treat them
with heavy salt applications.
Apply salt at 600 to 800 Ib per two-lane
mile, as required. If snow or sleet
continues and accumulates, plow and salt
simultaneously. If temperature starts to
rise, apply salt at 500 to 600 Ib per two-
lane mile, wait for salt to react before
plowing. Continue until safe pavement is
obtained.
Apply salt at rate of 800 Ib per two-lane
mile or salt-treated abrasives at rate of
1,500 to 2,000 Ib per two-lane mile. When
snow or ice becomes mealy or slushy,
plow. Repeat application and plowing as
necessary.
NOTE:
The light, 200-lb application called for in Conditions 1 and 2 must be repeated
often for the duration of the condition.
2-22
-------
TABLE 2-11. GUIDELINES FOR CHEMICAL APPLICATION RATES (AASHTO, 1976)
ro
CO
Weather conditions
Temperature
30°F and
above
25°-30°F
20°-25°F
15°-20°F
Below 15°F
Pavement
conditions
Wet
Wet
Wet
Dry
Wet
Dry
Precipitation
Snow
Sleet or freezing rain
Snow or sleet
Freezing rain
Snow or sleet
Freezing rain
Dry snow
Wet snow or sleet
Dry snow
Application rate (pounds of material per mile of two-lane road or two lanes of divided)
Low- and high-speed
muHilane divided
300 salt
200 salt
Initial at 400 salt;
repeat at 200 salt
Initial at 300 salt;
repeat at 200 salt
Initial at 500 salt;
repeat at 250 salt
Initial at 400 salt;
repeat at 300 salt
Plow
500 of 3:1 salt/
calcium chloride
Plow
Two- and three-
lane primary
300 salt
200 salt
Initial at 400 salt;
repeat at 200 salt
Initial at 300 salt;
repeat at 200 salt
Initial at 500 sari;
repeat at 250 salt
Initial at 400 salt;
repeat at 300 salt
Plow
500 of 3:1 salt/
calcium chloride
Plow
Two-lane
secondary
300 salt
200 salt
Initial at 400 salt;
repeat at 200 salt
Initial at 300 salt;
repeat at 200 salt
1,200 of 5:1 sand/
sari; repeat same
Plow
1,200 of 5:1 sand
Plow
Instructions
• Watt at least 0.5 h before plowing
• Reapply as necessary
• Watt at least 0.5 h before plowing; repeat
• Repeat as necessary
• Watt about 0.75 h before plowing; repeat
• Repeat as necessary
• Treat hazardous areas with 1 ,200 of 20:1
sand/salt
• Watt about 1 h before plowing; continue plowing
until storm ends; then repeat application
• Treat hazardous area with 1 ,200 of 20:1 sand/salt
-------
,c,ng chemical use is often considered not a good idea for safe^and cosi
ei"9 travelled' often <** 1 to 3 * wide Sen °
operations are done with the same truck during the
anrf '^ T0^^0 Path bei"9 travelled' often <** 1 to 3 * wide Sen sno±0°irla
and saltin/sandin oerat ™
2.3 REMOVAL TECHNIQUES
can be reduced if the fine materials left on roadways after
Quick cieanup
cars toHc^%(Z8? ^ °!bS (1985) 3ttached 9'ass P|ates to the rear bumpers of
cars to collect samples of antiskid materials. After exposure of the glass plates while
dnving on wet roadways, light transmission through the plates was measured og^e a
sem,quantrtat,ye est,mate of the comparative -dirtiness" of the roads These stud^s
reported that salted roads were "dirtier" than sanded ones as dSermined by S
transmission through the collection plates. aeiermmea Dy light
Many agencies contacted by phone stated that they clean up abrasives
to roadways, mostly during the spring, following the end of'the sno£SST
FPhnf^tV rban/°ads with vacuum sweePers- Cleanup operations begin ?n fate
February to keep strong spring winds from picking up the remaining sand on roadways
caus.ng a sandblasting action. Wyoming representatives, however, Responded ?haUhey
dean roadways whenever the weather permits if there is a heavy buildup of sand In rural
n SH Hy°T9't 'f nd JS SW6pt t0 the Side °f the road- while in urban areas,' sandTs
p,cked up by street cleaners. In season, Colorado cleans their highways at night (in the
'13"1'6'13"06 ^ MiSS°Uri "r°^ine|y ^a^s up the"
°Perations in cities is to prevent abrasives from
™ K 'nt° C3tCh baS'nS °r Sewers" Catch basins are emP^d at considerable
expense. Abras.ves may plug combination storm/sanitary sewer systems if there are no
2-24
-------
catch basins and streets are not cleaned. Cleanup operations in Maine and Minneapolis
also include the washing of bridges with pressure hoses.
2-25
-------
-------
SECTION 3
MEASUREMENT METHODS FOR PHYSICAL PROPERTIES
As stated in Section 1, the amount of PM10 generated from paved roadways is
directly related to the silt loading of the surface material available for reentrainment by
passing vehicles. Abrasives and deicing chemicals add temporary, but substantial,
amounts of silt-size particles to the road which, if not removed, will increase PM10
emissions and their associated air quality impact. These fine particles are either applied
directly to the road surface with the abrasive or chemical during initial treatment or are
created through attrition, dissolution, etc., while on the road surface.
The potential of a particular material to provide additional silt loading on a road
surface (and thus PM10) is directly related to its physical (or chemical) properties. In this
section, various methods for measuring the properties of antiskid materials are presented.
These methods can be divided into three general categories which include: silt
measurement methods; durability measurement methods; and measurement of other
applicable properties. Although this discussion will be directed mainly to abrasives, some
discussion of deicing chemicals will also be provided in Section 3.4.
Finally, a number of the standard methods found in the literature are not directly
applicable to some types of antiskid materials. For this reason, modifications to the
standard techniques were required in certain cases. These modified methods are
discussed in Section 3.5.
3.1 SILT MEASUREMENT METHODS
In general, there are two basic techniques for measuring the silt content (% less
than 200 mesh or 75 /um) of a material: mechanical classification and instrumental
methods. Each will be discussed as related to current practice.
3.1'.1 Mechanical Classification Methods
The most widely used technique for determination of silt content is through a
combination of wet and dry sieving according to American Society of Testing and
Materials (ASTM) Methods C 136 and C 117 (ASTM, 1984a and 1987). These methods
correspond to American Association of State Highway and Transportation Officials
(AASHTO) Methods T-27 and T-11, respectively (AASHTO, 1984; AASHTO, 1985a).
Copies of the ASTM methods are included in Appendix D.
3-1
-------
throuah a 200 m,,h ^™ f*™*™ methods- ™ aggregate sample is first washed
nhv^.H- tmes5 sieve to determine the amount of material less than 75 MM in
physical d,ameter. The remaining sample is then dry-sieved to determine the gradation
?nr?m L 'n T^ S'Ze ranges as we" as ™y residual m^ial passing he
^ tthT8,?' ™*wei9ht of the < 200 mesh material determined by washfng is
S±? M ^f dUrinQ dfy SieVing t0 determine the total silt content of the orighaJ
X PH ( t the,method used bV MRI in ^e development of the PM10 emission
co^t of tqhU±n A*™0"'* dry SieVin9 t0 determine Silt content-> The ^Proximate 1989
cost of the two ASTM analyses at a local laboratory in Kansas City is $35.
„«* /he ^°Ve techniclue is used bV most transportation agencies for determining the
gradation of aggregate materials, including those used for skid control. Size gradation
specifications= are normally developed for aggregates used in Portland cement concrete
mixes for road paving which rely heavily on the ASTM or AASHTO methods.
3.1.2 Instrumental Methods for Silt Determination
Although not as commonly used, various instrumental methods are available for
the determination of the silt content of finely divided particulate matter. In general these
methods include: microscopy, sensing zone instruments, elutriation and centrifugal
classification, mercury intrusion, gravity sedimentation, centrifugal sedimentation
hydrodynamic chromatography, and mercury porosimetry. Table 3-1 provides a further
classification of instruments falling into each of the above categories as well as examples
of commercially available equipment.
Although specifications vary from device to device, the above instruments are
hmited to measurement of particles < ~ 200 /tm in diameter. These instruments can
however, offer the advantage of automatic operation (thus reducing analytical error) as
well as providing data on particle size distribution, surface area, etc., which may be useful
in material selection. In general, instrumental analysis methods are relatively expensive
and thus are not widely used by transportation agencies.
3.2 DURABILITY MEASUREMENT METHODS
In general, there are four basic methods which can be used for determining the
durability of an aggregate material. These include abrasion and impact tests- wet
aggregate durability tests; freeze-thaw tests; and petrographic methods. Although these
tests were originally developed to measure the durability of aggregates used in concrete
mixes for road paving, they are also useful in the evaluation of skid control abrasives (see
Section 5). The following describes the various methods included in the above
categories.
3-2
-------
TABLE 3-1. SUMMARY OF INSTRUMENTAL METHODS OF
PARTICLE SIZE ANALYSIS
Microscopic methods:
Optical microscopy
Electron microscopy (TEM and SEM)
Automated image analysis (e.g., LaMont Analyzer)
Sensing zone methods:
Electrical resistance (e.g., Coulter Counter)
Optical (including laser)
Photoextinction (e.g., HIAC PA 720)
Forward and right-angle scattering (e.g., L&N Microtrac)
Acoustic
Elutriation and centrifugal classification methods:
Laminar flow (e.g., Roller analyzer)
Cyclone-elutriators (e.g., Cyclosizer)
Centrifugal classifiers (e.g., Bahco Micro-Particle Classifier)
Gravity sedimentation methods:
Turbidimetry (e.g., Wagner Turbidimeter)
Sedimentation balances (e.g., Micromerograph)
X-ray (e.g., SediGraph 5000D)
Photoextinction (e.g., SediGraph-L)
Centrifugal sedimentation methods:
Mass accumulation (e.g., MSA)
Photoextinction (e.g., Joyce-Loebl Disc Centrifuge)
X-ray
Hydrodynamic chromatography
Mercury porosimetry
3-3
-------
3-2-1 Abrasion and Impact Tests
Anqe'f ADr^inn Tpst is opined in ASTM Method C 131 and AASHTO
fmaJ't'Ze C03rSe a"regate (AS™' 1989= AASHTO, 1983). The Los
ndf t0 meaSUre the de9radati°n of washed mineral aggregates of
SI2H (e'9H-4-75 t0 2'36 mm) due t0 a ^bination of abrasion (o
sieveaftmnft T^9 '" * ^^ ba" m'U Each graded material sa™Ple Is
sieved after milling to determine the amount of fines generated by the process The
analyzed ^ *** '" *" "'" ^^ *" ^ ^ ^^ °" 'he '
A copy of the Los Angeles abrasion test has been included in Appendix E The
approbate 1989 cost of a standard ASTM Method 131 or AASHTO Me hod T 96
analysis is $50 at a local laboratory in Kansas City.
The Aqqreqate Impact Tasf is given in British Standard 812 and also used in the
fntSt 76Searf PUrP°SeS (W°0dside and Peden' 1983= He9m°n and Meyer
« ! ;,3 ! ' ag9re9ate samP|e is P|a^ed in an instrument where a standard
dp rePftedly dropped onto the material thus attenuating the sample. Degradation
determined by differences in gradation before and after impact (Hegmon and Meyer"
Peden QR^^^^1^ 'S a'S° 9iV9n in'British Standard 812 (Woodside and
Peden, 1 983) In this test, an aggregate sample is crushed at a constant rate so that the
compressiye force at the end of the 10-min test period is 40 tonnes (44 tons) The
^nnJ^f riSt t0 thV°rmation of a cushioning layer formed from the matrix
composed of weaker crushed material, which effectively dampens the effect of loading
on the enclosed sample. Both the aggregate crushing test and aggregate impact test
are used extensively in Europe to determine the durability of aggrega?e us^d Tn road
pavement (Woodside and Peden, 1983; Anon, 1979).
IAI H I?6 va'u|s °btained from the above tests were correlated with one another by
Woods.de and Peden (1 983) using experimental data for 1 3 different aggregate materials
used in paving mixes. The correlation of the aggregate crushing value (ACV) with the Los
Angeles abrasion value (LAAV) is given by the following equation (Woodside and Peden
I s7OOj . . '
ACV = 0.8013 LAAV + 3.5853 (3_1)
The correlation coefficient for the above relationship is 0.9042.
3-4
-------
Using the same data set, the correlation of aggregate impact value (AIV) with the
Los Angeles abrasion value (LAAV) was determined by MRI to be:
AIV = 0.6535 LAAV + 3.1778 (3-2)
The correlation coefficient for the above relationship is 0.9723. As shown by
Equations (3-1) and (3-2), the results of the three abrasion/impact tests are intercorrelated
with one another. This intercorrelation will be discussed in more detail later in this report.
•
3.2.2 Wet Aggregate Durability Tests
All wet aggregate durability tests are based on the basic techniques outlined in
AASHTO Method T-210 and ASTM Method D 3744 for fine aggregate material (AASHTO,
1964; ASTM, 1985c). Variations of these methods have been developed by Maine,
Washington (Method 113A), Oregon, and Alaska (Method T-13). The purpose of this
method is to determine the durability of an aggregate material based on its relative
resistance to the production of detrimental, claylike fines when subjected to mechanical
agitation in water.
In the basic method, a dry graded sample is placed in a closed vessel along with
1000 mL of distilled (or deionized) water and agitated on a sieve shaker for 2 min. After
agitation, the degraded material is washed through a 200 mesh screen and the material
> 200 mesh is dried and dry-sieved to obtain the size distribution. After sieving, the
individual separates are recombined, a calcium chloride solution added to a subsample
of the material, and the material reshaken and allowed to settle. The "clay" layer and
"sand" layer is then measured with the "durability index" calculated from the values
obtained.
A number of variations to the above method have been developed by state
transportation agencies. These variations include changes in both methodology or in the
calculation procedure. Copies of the various state-generated wet durability test methods
are included in Appendix F.
The specification of a minimum durability value for paving aggregates using the
above tests will vary from state to state. For example, Washington specifies a minimum
degradation value of 50 (using its particular test method) as compared to Maine which
specifies a minimum value of 40 (using its method) (Norburg, 1981). It would be
expected that these specifications may also be used as general guidelines for the
selection of antiskid abrasives as well.
3.2.3 Freeze-Thaw Tests
Another measure of aggregate durability is its susceptibility to fracture during
freeze-thaw cycles. This is referred to as "soundness" and is determined using the
sodium/magnesium sulfate test specified in ASTM C 88 and AASHTO T-104 (ASTM, 1983;
3-5
-------
AASHTO, 19855). The 1989 cost of analysis for either the ASTM or AASHTO methods
is approximately $65 at a local laboratory in Kansas City.
The above methods estimate the soundness of individual aggregate size separates
by repeated immersion in a saturated salt solution followed by oven drying to partially (or
completely) dehydrate the salt precipitate in the permeable pore spaces The internal
expansive force, derived from the rehydration of the salt upon reimmersion, simulates the
expansion of water on freezing. After the final immersion/drying cycle, degradation is
determined by sieving the sample over the same sieve on which it was retained before
the test to determine the percent loss. Since most abrasives used for skid control are
mixed with salt either before or after application, this test may be of particular interest in
determining the durability of such materials in service.
3.2.4 Petroaraphic Methods
The final durability measurement techniques to be discussed are petrographic
methods which evaluate the mineral composition of fine aggregate materials. The
nomenclature used to describe the constituents of mineral aggregates is provided in
ASTM Method C 294 with the petrographic examination of this material described in
ASTM Method C-295 (ASTM, 1969; ASTM 1985a). The approximate cost of analysis for
- ASTM C-295 is $2,000 for a laboratory located in Chicago.
The above ASTM methods are only intended to determine the mineral content of
a particular material and thus do not attempt to relate mineralogy to durability. However
the Petrographic Number Method has been developed in Canada for such a
determination (Hudec, 1984). This method is basically a subjective method of classifying
crushed rock particles into four discrete categories of good, fair, poor, and bad
(deleterious) aggregate (1 = good; 3 = fair; 6 = poor; 10 = bad).
Hudec (1984) related the petrographic number (PN) determined by the above
method to various other measured parameters by step-wise multiple linear regression.
The results of this analysis obtained the following relationship:
PN = 4.406 + 0.144 Ab + 1.805 Ad - 0.165 Gr - (3-3)
(2.014Ad/VAd) -l.004Hd
where: PN = petrographic number (dimensionless)
Ab = low intensity abrasion.value for < 6.7 mm material (dimensionless)
Ad = adsorbed water (%)
Gr = grain size (mm)
3-6
-------
VAd = vacuum water adsorption (%), by boiling method
Hd = hardness
As shown by the above relationship, the durability (as determined by petrography)
of an aggregate decreases with its susceptibility to abrasion and water absorption and
increases with grain size and hardness of the particles.
3.3 ANALYSIS METHODS FOR OTHER PROPERTIES
In addition to the above, there is a wide variety of other properties related to the
durability and fines-generating potential of abrasives used in ice and snow control. These
are too numerous to describe in detail here. However, Table 3-2 provides a summary of
the various methods and the basis for their applicability to skid control abrasives.
Analysis cost estimates are also provided in Table 3-2 based on a laboratory located in
either Kansas City or Chicago.
Like those described above, most of the methods described in Table 3-2 are
related to the selection of aggregates for use in portland cement concrete mixtures. A
standard specification for concrete aggregates is published in ASTM Method C 33 which
can be used as a general guide for material selection (ASTM, 1985b). Please note,
however, that some transportation agencies use rejected concrete aggregate (e.g., sand)
as a skid control abrasive.
3.4 • MEASUREMENT METHODS FOR DEICING CHEMICALS
Rock salt used for road deicing purposes is composed mostly of NaCI which is
believed not to remain on the road surface in any substantial amount after the storm. The
NaCI is either washed away with the storm runoff, entrained by passed vehicles, or
retained by the pavement itself. However, rock salt contains appreciable quantities (up
to 6% for Kansas salt) of water-insoluble mineral matter which can remain on the road
surface for extended periods (Kaufmann, 1968). This insoluble mineral matter consists
of very fine particles of CaS04, silica, alumina, ferric oxide, and dolomite. Therefore, in
addition to gradation (see Section 3.1 above), the percent insoluble matter in rock salt
is important in the generation of surface silt loading and thus PM10.
To determine the percent insoluble material in rock salt using ASTM E 534, 25 g
of sample is placed in 600 mL of heated water and stirred for a period of 60 min (ASTM,
1986). After stirring, the mixture is wet filtered to obtain the percent insoluble matter by
gravimetric analysis. General specifications for sodium chloride are provided in ASTM
Method D 632 (ASTM, 1984b).
3-7
-------
Method designation Title of test method
Summary of test method
Significance of test to antiskid
materials
ASTM C 29
(AASHTO-T-19)
ASTM C 128
ASTM C 142
CO
do
ASTM C 289
ASTM C 672
Unit Weight and Voids in
Aggregate
Specific Gravity and
Absorption of Fine
Aggregate
Clay Lumps and Friable
Particles In Aggregates
Potential Reactivity of
Aggregates (Chemical
Method)
Scaling Resistance of
Concrete Surfaces
Exposed to Delclng
Chemicals
A standard metal cup is sequentially filled with aggregate
and repeatedly tamped with a rod. The weight of the
aggregate In the cup Is then measured with the percent
voids calculated based on the dry bulk density of the
aggregate and the equivalent weight of water occupying
the cup volume.
The aggregate to be tested is immersed in water for a
period of 24 h after which time it Is partially air-dried
such that It no longer holds a cylindrical shape. A 500-g
sample of this material is then Introduced Into a pycnom-
eter and filled with water. The pycnometer Is weighed
both with the aggregate sample/water mixture and with
an equivalent volume of water. The sample Is removed
from the pycnometer. oven-dried, and the total moisture
determined by gravimetric analysis. The bulk specific
gravity and water absorption Is then calculated from the
test results.
A dry-sieved (Method C 117) aggregate sample
(> 1.18 mm) Is Immersed In water for a period of 24 h,
after which the lumps and friable particles are Individually
broken by hand. The material is then wet-sieved and
dried to determine the percent of the original sample
consisting of lumps or friable particles.
A small aggregate sample containing particles ranging
from 150 to 300 iim Is immersed In a heated (80°C),
1.0 N solution of NaOH for a period of 24 h. At the end
of this period, the solution containing dissolved SI02 Is
. filtered from the aggregate, digested with concentrated
HCI, and the solid SI02 repeatedly filtered from the
reacted solution. The percent Si02 is calculated from the
results of gravimetric analysis.
In this method, concrete specimens are exposed to a 4%
solution of CaCI (or other delcers) and repeatedly frozen
and thawed for 16 to 18 h. The scaling of the specimen
is determined visually over the desired number of cycles
and a rating assigned from 0-5 (0 = no scaling) every
5 cycles (or every 25 cycles after 25 cycles have been
made).
Approximate cost of
analysis8
The void fraction of an aggre- $50
gate Is related to the angularity
of the particles. Highly angular
particles are desired for good
skid control.
Bulk specific gravity is needed $25
for Method C 29 above. In addi-
tion, since application of
abrasives Is normally metered
according to volume, bulk spe-
cific gravity can be used to
determine the application rate
on a weight basis. Finally, water
absorption is related to sound-
ness of an aggregate to with-
stand freeze-thaw cycles.
Clay lumps and friable particles '$25
contained in skid control abra-
sives have a greater potential for
the generation of fines when
applied to a road surface.
The silica (SiO2) content of a $150
fine aggregate used for skid
control has been determined to
be one of the most Important
factors in durability with high
silica particles being most
durable.
Reference
No.b
1.2
The ability of a road pavement $600
to withstand the application of (50 cycles)
delclng chemicals Is related to
pavement breakup (I.e., scaling)
and thus the production of addi-
tional fine particles for sus-
pension. This test is not directly
related to aggregate durability,
however.
(continued)
-------
TABLE 3-2 (continued)
Method designation Title of test method
Summary of test method
Significance of test to antiskid Approximate cost of Reference
materials
analysis*
No."
ASTM D 75
(AASHTO T-2)
ASTM D 3398
ASTM E 660
CO
AASHTO T-248
(ASTM C 702)
None
Sampling of Aggregates
Index of Aggregate
Particle Shape and
Texture
Accelerated Polishing of
Aggregates or Pavement
Surfaces Using a Small-
Wheel, Circular Track
Polishing Machine
Reducing Field Samples
of Aggregate to Testing
Size
Moh Hardness
Grab sampling procedures are outlined for static and
dynamic conditions.
A sample of aggregate Is placed In a cylindrical mold of
known volume and repeatedly tamped with a rod of
standard weight The weight of material Is determined
after 10 and 50 drops of the rod to calculate the particle
Index
Pavement (bituminous or concrete) specimens are
mounted on a track, over which four smooth, pneumatic
tires are continuously run. A polishing curve Is obtained
over a period of ~ 8 h or until friction measurements
show no substantial decrease with continued polishing.
(Note that although pavement specimens are normally
used, an Immobilized aggregate sample could also "be
evaluated by the above method.)
Sample splitting using a riffle and coning and quartering
are described.
Surface scratch hardness relative to empirical scale.
Collection of a representative Variable 1, 2
sample of antiskid materials.
Particle shape (high degree of $560 1
angularity) and texture (coarse) (7 fractions)
are Important In skid resistance.
This method Is an indirect
measure of these two properties.
The resistance of a skid control Unknown 1
material to wear and polishing
by the action of rotating tires Is
the most direct measure of
durability currently available.
Collection of a representative Variable 1,2
sample of antiskid material for
further analysis.
Material hardness Is directly $50 —
related to durability as measured
by other methods.
Based on 1989 prices of an independent materials testing laboratory located In either Kansas City or Chicago.
Reference 1 = 1990 Annual Book of ASTM Standards, American Society of Testing and Materials, Philadelphia, Pennsylvania, 1990.
Reference 2 = Standard Specifications for Transportation Materials and Methods of Sampling and Testing, Part II, Methods of Sampling and Testing, 14th Edition, American
Association of State Highway and Transportation Officials. Washington, D.C., August 1986.
-------
3.5 MODIFICATIONS TO STANDARD TEST METHODS
was presen/ed to the 9reates<
a«. i TheAfolloiwin9 sections describe ASTM methods modified by MRI. These methods
SI 96leS abraSi°n 'OSS: Silt C°ntent: Vickers hardness-' and insolubte matteMn
calcium magnes,um acetate. Copies of the standard ASTM methods for these
parameters are included in Appendices D, E, G, and H. me™vs for these
It can be safely assumed that ASTM methods are generally developed with
sufficient data to validate the technique and determine its accuracy. Su^ i^tffe c^?
however, for methods modified in the current program. For these method ^ accuracJ
could generally not be determined either from existing data or from d^ gener^"n ?S
study mstead, multiple analyses of commonly available materials were performed to
partially assess the reproducibility (i.e., precision) of each modified method This is
discussed in more detail in Section 5.2. "eu.uu. i nis is
3-5-1 Los Angeles Abrasion Loss
At present, the Los Angeles (LA) Abrasion test (ASTM C 131) is only applicable to
aggregate larger than 2.36 mm (0.093 in). Since many abrasives used for'ice and snow
control are finer than 2.36 mm, minor modifications to ASTM C 131 are required In
addition, since silt size particles are of most concern in the assessment of PM
emissions, the mcrease in the silt content after exposure in the LA test ™ach?ne was
,n Sani Cat0r0fdUrability' Themodificati°ns made to ASTM C l£ aresho^n
test data Produced by the procedure outlined in Table 3-3
'ded that further revisions to the method we^e necessary
4 Th. thH H ? in Tab'e 3'3 WaS altered Slj9ht|y to that ^hownTn
°d h°Wn 'n Table 3'4 provides Slj9ht|y more detail ^d a revised
,rt H .
calculation scheme which more closely reflects the intent of the standard technique
Table 3-4 was used for analysis of all antiskid abrasives in the remainder of the program.'
It should be noted, however, that further modification to Table 3-4 miqht be
recommended in future efforts based on the experience gained during Phase 2 For
example, all materials should be normalized to 0% initial silt content by washing prior to
3-10
-------
TABLE 3-3. MODIFICATIONS TO ASTM METHOD C 131-89
The standard ASTM methodology should be used with the following modifications:
5. Apparatus
5.4.1 The weight of the test sample placed in the Testing Machine shall be
5,000 ± 10 g along with a charge of six (6) standard spheres.
8. Procedure
8.1 Split the sample into two equal fractions according to Method C 702.
Sieve the first fraction according to the procedure outlined in Table A and record
the results. Weigh the charge of spheres to be introduced into the testing machine
and record the results.
8.2 Place the second sample fraction and the charge in the Los Angeles
testing machine and rotate the machine at a speed of 30 to 33 rpm for
500 revolutions. After the prescribed number of revolutions, discharge the material
from the machine and sieve the sample material using the procedure shown in
Table A. Record the results.
TABLE A. SAMPLE SIEVING PROCEDURE
1. Select the appropriate 8-in diameter. 2-in deep sieve sizes. Recommended U.S. Standard Series sizes are 3/8 in,
No. 4, No. 20, No. 40, No. 100, No. 140, No. 200, and a. pan. Comparable Tyler Series sizes can also be utilized.
The No. 20 and the No. 200 are mandatory. The others can be varied if the recommended sieves are not available
or if buildup on one particular sieve during sieving indicates that an intermediate sieve should be inserted.
2. Obtain a mechanical sieving device such as vibratory shaker or a Roto-Tap.
3. Clean the sieves with compressed air and/or a son brush. Material lodged in the sieve openings or adhering to the
sides of the sieve should be removed frf possible) without handling the screen roughly.
4. Obtain a scale (capacity of at least 1,600 g) and record make, capacity, smallest division, date of last calibration, and
accuracy (if available).
5. Tare sieves and pan. Check the zero before every weighing. Record weights.
6. After nesting the sieves in decreasing order with pan at the bottom, dump dried laboratory sample (probably
immediately after moisture analysis) into the top sieve. Brush fine material adhering to the sides of the container
into the top sieve and cover the top sieve with a special lid normally purchased with the pan.
7. Place nested sieves into the mechanical device and sieve for 20 min. Remove the pan containing the minus
No. 200 fraction and weigh. Replace pan beneath the sieves and sieve for another 10 min. Remove pan and
weigh. When the differences between two successive pan sample weighings (where the tare of the pan has been
subtracted) is less than 3.0%, the sieving is complete.
8. Weigh each sieve and its contents and record the weight Check the zero before every weighing.
9. Collect the laboratory sample and place the sample in a separate container if further analysis is expected.
3-11
-------
5 nnn + In J^ ^l °f ^ test Sample placed in the Testin9 Machine shall be
5,000 ± 10 g along with a charge of six (6) standard spheres.
8. Procedure
1^ the S3mple int° two equal fractions according to Method C 702
achcordin9 to «* procedure outlined in Table A and record
^ * ^^ * * '^^ lrt°
the resuts
and recf'd
i n« An8'2. Place+.the Second sample fraction
-------
TABLE 3-4 (continued)
9. Calculation
9.1 Determine the abrasion loss as a percentage using the following
equation:
aL =
S, - S,
x 100
where:
aL = LA abrasion loss (weight percent)
Sj = Weight percent of material > 200 mesh (75 /im) in
original sample
S, = Weight percent of material > 200 mesh (75 /JITI) in
sample after exposure in Los Angeles Testing Machine
3-13
-------
The calculation scheme shown in Table 3-4 would also be revised to the following form:
Wt
x 100 (3.4)
Where: aL = LA abrasion loss (weight %)
*"
W, = Total weight of original sample (g)
3.5.2 Silt Content
silt conh^H T'- hf US6d a m°dified version of AS™ c 136 for determination of
nit i K y °n dry sieving'
-------
TABLE 3-5. SILT ANALYSIS PROCEDURES
1. Select the appropriate 8-in diameter, 2-in deep sieve sizes. Recommended U.S.
Standard Series sizes are 3/8 in, No. 4, No. 20, No. 40, No. 100, No. 140,
No. 200, and a pan. Comparable Tyler Series sizes can also be utilized. The
No. 20 and the No. 200 are mandatory. The others can be varied if the
recommended sieves are not available or if buildup on one particular sieve
during sieving indicates that an intermediate sieve should be inserted.
2. Obtain a mechanical sieving device such as vibratory shaker or a Roto-Tap.
3. Clean the sieves with compressed air and/or a soft brush. Material lodged in
the sieve openings or adhering to the sides of the sieve should be removed (if
possible) without handling the screen roughly.
4. Obtain a scale (capacity of at least 1,600 g) and record make, capacity, smallest
division, date of last calibration, and accuracy (if available).
5. Tare sieves and pan. Check the zero before every weighing. Record weights.
6. After nesting the sieves in decreasing order with pan at the bottom, dump dried
laboratory sample (probably immediately after moisture analysis) into the top
sieve. Brush fine material adhering to the sides of the container into the top
sieve and cover the top sieve with a special lid normally purchased with the pan.
7. Place nested sieves into the mechanical device and sieve for 20 min. Remove
the pan containing the minus No. 200 fraction and weigh. Replace pan beneath
the sieves and sieve for another 10 min. Remove pan and weigh. When the
differences between two.successive pan sample weighings (where the tare of
the pan has been subtracted) is less than 3.0%, the sieving is complete.
8. Weigh each sieve and its contents and record the weight. Check the zero
before every weighing.
9. Collect the laboratory sample and place the sample in a separate container if
further analysis is expected.
3-15
-------
ftsu m H H-30 apPr°Priate eP°*y ™*
tne surface of the mold according to standard techniques.
. Grind and
8. Procedure
l K f°Ur (4) partides from each mold for ana|ysis. The particles
^^f5 KematLVe °f the bU'k material Present in the specimen. Variations in
reflec ance, pol.sh.ng hardness, color, and anisotropism should be noted and
particles excluded which do not represent the bulk of the material present For
materials of greatly varied lithologic character, particles with different lithology
should be selected which represent the entire sample.
8.2 Delete
fn °f fiVe (5) indentations on each of the four particles selected
for analysis o obtam a total of 20 indentations for each polished section. Read *he
ftvo Diagonals of the .ndentation to within 0.25 ,m or 0.4%, whichever is larger and
determine the average of the diagonal lengths. For materials with very low
reflectance values (e.g., calcite, quartz, etc.), indentation measurements should be
oThP^rir. Uf *9 aP appr°Priate microscope equipped with dark field illumination or
otner special types of lighting arrangements.
8.4.8 Indentations which exhibit obvious problems shall not be measured (For
example, an indentation that shows cracks at its edge.) Instead, repeat the
-P-entative particle
8.5.2 Compute the mean HV for the polished section by calculating the Vickers
Ssfrrfr fT^ f°r, eaCh,°f the 2° indentations fror" the equation given in Section
4.3 1 .1 or from the information given in Table 2. Average the 20 measurements to
3I?K uw ?f " °f the Sample> To obtain the hardness number from the table
read the HV (1 g() corresponding to the average of the two measured diagonal '
lengths (in micrometers) and multiply by the test load in grams-force
3-16
-------
3.5.4 Percent Insoluble Matter in Calcium Magnesium Acetate
Currently there is no ASTM method for the determination of insoluble matter in
calcium magnesium acetate (CMA). Therefore, a technique was developed by altering
ASTM E 534 for the determination of insolubles in sodium chloride. These modifications
were based on a technique used by a manufacturer of CMA for product quality control.
The modifications to ASTM E 354 are shown in Table 3-7.
3-17
-------
TABLE 3-7. MODIFICATIONS TO ASTM E 534-86 TO INCLUDE
CALCIUM MAGNESIUM ACETATE (CMA)
The standard ASTM methodology should be used with the following modifications:
8. Preparation of Stock Solution
8.1 Split sample with a riffle to obtain a 100-g sample.
8.2 Weigh out a 30-g sample of material and identify appropriately.
8.3 Place 100 mL of water at a temperature of 37°C (100°F) in the high-
speed blender. (Note that a magnetic stirrer may be substituted for the hiah
speed blender.) a
8.4 Slowly add the CMA to the blender and blend on slow speed for 1 min.
8.5 Test for water insolubles as described in Sections 17 to 24.
20. Procedure for Calcium Magnesium Acetate
20.1 Transfer the sample prepared in accordance with 8.1 to 8.5 above to a
1-L Erlenmeyer flask, washing out the blender with 100 mL of water (at 37°C)
to give a total volume of 200 mL
20.2 Stir on a magnetic stirrer for 1-h at a constant temperature of 37°C
(100°F). Adjust the stirrer speed to give maximum agitation without danger of
losing the sampling due to splashing. Place a beaker over the top of the
flask while stirring.
20.3 After the 1-h mixing period has elapsed, remove the flask from the
magnetic stirrer and allow to cool to room temperature.
20.4 Filter the solution by vacuum through a previously dried (110°C for 1 h)
and accurately weighed filter disk using the Parabella funnel. Quantitatively
transfer all insoluble matter to the filter paper with distilled water from a wash
bottle making sure the sides of the flask are flushed and the rinse allowed to
filter.
20.5 Dry the filter disk at 110°C for 1 h and cool in a desiccator. Weigh the
disk on an analytical balance to the nearest 0.1 mg.
20.6 Calculate the percent insoluble matter according to Section 22.
3-18
-------
SECTION 4
SELECTION CRITERIA
In this section, the potential of antiskid materials to generate PM10, their
effectiveness and durability in service, and initial criteria for material selection will be
discussed as background to Phase 2. Applicable cost data related to both traditional
abrasives and deicers, as well as alternative skid control measures (e.g., embedded pipe
pavement heating), have been provided in Appendix I.
4.1 POTENTIAL FOR DUST EMISSIONS
As stated elsewhere in this document, the amount of PM10 emitted per vehicle-mile
traveled is a direct function of the silt loading on the road surface. Under normal
conditions, barring other loading increases such as mud-dirt carryout, a type of loading
"equilibrium" is established on the road surface where the amount of new material
deposited is balanced with the amount of fines being resuspended as dust. This
equilibrium is disturbed, however, when antiskid materials are applied which provide
temporary but substantial increases in silt loading and resuspended dust. In this section,
applicable data related to silt loading increases and associated ambient air quality impact,
due to antiskid materials will be presented.
As determined from available literature, only limited data exist on the air quality
impact associated with the application of antiskid materials to paved roadways. Three
studies were identified which relate to this topic. One additional study (conducted in
Denver during the winter of 1988-1989) was also found, but no firm conclusions could be
drawn from the test results.
The first study of interest was conducted in a parking lot in Helena, Montana
(Brant, 1972). In this study, a surface sample of sand applied to the parking lot at the
end of the winter snow season was collected. The gradation of this sample was
compared to that of the original stockpiled material. The results are shown in Figure 4-1.
• As can be seen from Figure 4-1, the percentage of material less than 200 mesh
(i.e., silt) increased from approximately 2% to approximately 13% over the course of the
winter. In addition, the material applied to the parking lot was reported to contain a high
percentage of quartz which would indicate good durability characteristics. These results
would indicate that even relatively hard, durable material can produce substantial
increases in silt loading as a result of vehicular traffic. It should be considered, however,
4-1
-------
Helena Parking Lot
0
1/2" 3/8" 4M 10 20 30 40 50 60 80 100 200
Sieve Size
Figure 4-1. Sieve analysis data of antiskid material
in Helena parking lot.
4-2
-------
that the traffic volume in a parking lot is much lower and traveling at lower speeds than
would be the case for a paved city street. Therefore, the increase in silt content observed
could be even greater than that reported for the parking lot studied.
The second program considered was a portion of the original MRI paved road
study performed in 1977 (Cowherd et al., 1977). In this study, a truck-mounted spreader
(similar to that used for winter sanding) was used to artificially load a paved city street
with a fine limestone aggregate. The loading and gradation of the surface material were
measured for the four exposure profiling tests conducted during the program. Plots of
total loading and silt loading vs. number of vehicle passes are presented in Figure 4-2
with the exposure profiling results for total particulate shown in Figure 4-3.
As shown by Figure 4-2, both total loading and silt loading on the road decrease
monotonically with vehicle passes after application of limestone gravel fines. This trend
is also indicated by the dust concentrations shown in the profiles in Figure 4-3.
The final data set is from the salting and sanding demonstration study performed
in Denver during the winter of 1980-1981 (PEDC6, 1981). Two study areas were used.
The first study area, located in Lakewood, evaluated the application of salt as compared
to the application of sand along 20th Avenue. The second study area evaluated the
sanding of two sections of Morrison Road in Denver with and without poststorm street
cleaning. Application rates or properties of the antiskid materials used in the study were
not specified.
Air quality measurement data collected during the Denver program did not show
much correlation with either the application of different antiskid materials or for poststorm
cleaning. However, the data from the Lakewood study area were found to be of limited
usefulness in defining general trends for the current program.
Figure 4-4 provides -a plot of the increase in total surface loading (as related to
"baseline" conditions) vs. time after application of sand or salt as calculated by MRI from
the raw data. Figure 4-5 shows a similar plot for silt loading. Finally, Figure 4-6 provides
air quality measurements for TSP as determined at the near-street monitoring site.
As indicated by these figures, the data are scattered and show a high degree of
variability which makes interpretation difficult. However, for sand application there is
some limited evidence of air quality improvement that can be associated with an apparent
decay in silt loading during the period following sand application.
4.2 EFFECTIVENESS OF ANTISKID MATERIALS
The main purpose of applying abrasives to ice and snow covered roadways is to
improve traction or the skid resistance of the surface. The degree to which traction is
improved is directly related to the type and properties of the antiskid material and the
4-3
-------
(Q
i
W
c*
OT o"
-* 0)
S a
&>
-a
o>
0)
CD
V)
OQ
(O
CO
T"
CD
W
H
a
: 0)
°
5
(Q
, -
-------
SITE: SMllwell Avenue
SURFACE LOADING: Gravel Fines
01
O Run 11
a Run 12
A Run 13
v Run 14
Silt
Loading
(gny/m2)
94.2
86.4
76.4
30.5
ISOKINETIC PARTICULATE CONCENTRATION (ma/™3) odjutted to 250 vehicle
Figure 4-3. Profiles of total particulate concentration for artificially loaded paved road.
-------
12,
10 -
t 8
.Q
m
O
D)
C
T3
CO
O
3
c
CD
CO
s
O
- 4
-
• Sanding
• Salting
J.
0
10 20
Time (Days) After Application
30
Figure 4-4. Increase in total surface loading vs. time after
application of salt or sand.
4-6
-------
30
D)
I
O
_c
0)
03
£
o
c
©
•
Sanding
Salting
10
_L
10 20
Time (Days) After Application
30
Figure 4-5. Increase in silt loading vs. time after
application of salt or sand.
4-7
-------
120
m
1
c
o
I
i-
o
O
si
—"
90
60
^30
0
Near-Street TSP (Salt)
Near-Street TSP (Sand)
J_
_L
_L
2345678
Days Since Application of Salt or Sand
Figure 4-6. TSP air quality impact of salt and sand application over time.
4-8
-------
amount of material applied per unit surface area. In this section, the effectiveness of
various antiskid materials will be discussed as related to the generation of PM10.
Skid resistance is the force developed between the tire and pavement surface
during braking and is a function of: speed; tire temperature, design, and pressure;
pavement surface texture; and surface wetness. Frictional force is commonly associated
with skid resistance and is mathematically defined as:
F = fxP (4-1)
where: F = Frictional force
f = The coefficient of friction
P = The load perpendicular to the tire/surface interface
Since skid resistance is a function of several independent variables, simple models
cannot be used for its measurement. Therefore, most transportation agencies have
adopted ASTM Method E 274 which slides a standard locked tire along an artificially
wetted (or ice/snow covered) pavement at a constant speed (usually 40 mile/h). The
measurements are reported in terms of skid number (SN) which is computed as follows:
SN = 100 f = 100 T/L (4-2)
where: SN = Skid number at 40 mph
T{ = Tension force to pull the locked tire across the pavement (Ib)
L = Load (weight) on the tire (Ib)
A measure of surface frictional properties can also be determined using the British
pendulum tester per ASTM "Method E 303. The British pendulum tester is a dynamic
pendulum impact-type tester used to measure the energy loss when a rubber slider is
propelled over a test surface. The values measured (British Pendulum Number or BPN)
represent the frictional properties of the surface not related to other slipperiness
measuring equipment.
A number of studies have investigated the ability of various antiskid materials to
increase skid resistance or the friction coefficient. Hegmon and Meyer (1968) compared
the coefficient of friction of four antiskid abrasives as tested on a circular test track in a
cold room. The abrasives tested were: boiler cinders (0.9% < 200 mesh); coke cinders
(0.4% < 200 mesh); natural sand (2.6% < 200 mesh); and crushed limestone (0%
< 200 mesh). Each material was also subjected to an impact (drop) test of 8,650 ft-lb/ft2.
The test results for the friction measurements conducted are shown in Figure 4-7 (on an
equivalent volume basis) and the results of impact tests shown in Table 4-1.
As shown by the above data, sand has the overall highest coefficient of friction and
the highest breakdown strength under applied load. Limestone is the second best
material with respect to both friction and breakdown strength with cinders being least
4-9
-------
0.4
c
g
?0.3
o
S 0.2
'o
o
O
0.1
0
Sand
Stone
Boilerhouse Cinders
Coke Cinders
_L
8 16 24 32
Number of Wheel Passes
40
Figure 4-7. Coefficient of friction (f) vs. number of wheel passes for
four antiskid materials compared on an equal volume basis.
TABLE 4-1. RESULTS OF IMPACT TESTS
Type of material
Percent breakdown8
Boiler cinders
Coke cinders
Natural sand
Limestone
5.3
3.9
0.7
0.8
Breakdown is defined as the percent change in the
sum of the cumulative percentages retained on each
of the following sieve sizes before and after impact:
3/4 in; 3/8 in; No. 4; No. 8; No. 16; No. 30; No. 50;
and No. 100.
acceptable. This study also found that the highest friction coefficients were obtained for
materials with gradations between 4 and 16 mesh with finer (i.e., -50 mesh) particles
being substantially less effective.
In another study performed by Furbush et al. (1972), pavement skid resistance (no
ice or snow) was related to both aggregate mineralogy and particle size. These
investigators found that medium to coarse grain sandstones with high levels of quartz
exhibited the highest skid numbers (i.e., 65-70 at 40 mph) and soft aggregates such as
limestone polish rapidly and exhibit poor skid resistance. These results are illustrated in
Figure 4-8 which show the skid resistance of limestone containing various quantities of
+200 mesh silica. As shown by these data, the skid resistance increases substantially
with increasing silica content. The report also concludes that a Moh Hardness of 5-7 is
desirable for good skid resistance.
4-10
-------
JD
E
80
70
60
50
CD
O
I 40
CO
I 30
TJ
2
co 20
10
o o
i Projected Intervals
Road Friction Tester Numbers are Corrected to a Temperature Base
of 70° F. (Temperature Gradient: 3 SN/10° F.)
10 15 20 25 30 35 40
" Percent of Plus 200 Sieve Size Silica
45
50
Figure 4-8. Skid resistance of limestone material vs.
percent +200 mesh silica (Furbush, 1972).
4-11
-------
The final study found in the literature was performed by the State of Alaska
(Connor and Gaffi, 1982) to determine optimum specifications for sand used in skid
control. This study determined the BPN of different gradations of four types of aggregate
material as determined using the British Pendulum Test at various temperatures in a cold
room. Vehicle stopping distances on an outdoor ice track were also determined for the
same materials. The materials tested were: fractured stone, pit-run stone, concrete
aggregate, and coal cinders. The specifications for this material are provided in
Table 4-2. Representative data for these tests are shown in Figures 4-9, 4-10, and 4-11
for the cold room and vehicular tests, respectively, at various application rates.
TABLE 4-2. MATERIAL SPECIFICATIONS FOR ALASKA TESTS
Type of material Percent fracture Percent < 200 mesh
Fractured stone 87 1
(Alaska maintenance)
Pit-run stone 50 0
Concrete aggregate 65 1
Coal ash — 20
As shown by Figures 4-9, 4-10, and 4-11, the use of coal ash had the highest skid
resistance and shortest stopping distance of all the materials tested with concrete sand
being a good second choice. Also, the data show that materials in the range of about
16 mesh provided the best skid resistance as compared to coarser materials. Finally, it
was also found that angular particles (high percentage of fracture) were more effective
than rounded aggregate (e.g., fractured stone vs. pit run).
It should be noted that the coal ash used in the above tests contained a high
percentage of < 200 mesh material which would indicate a high potential for PM10
production in service. Thus, from an emissions standpoint, coal ash would not be
suitable for use in NAAQS nonattainment areas. The above results would indicate,
therefore, that a good quality, washed construction aggregate (e.g., sand) would be the
best choice with respect to both maximum effectiveness and minimum PM10 production
of the materials tested.
From the above data it can be seen that the effectiveness of a particular material
to increase skid resistance is a function of particle size, shape, hardness, and durability.
These results would indicate that a coarse grained (16 to 50 mesh), hard (Moh Hardness
5-7) material with a high degree of angularity (80+% fracture) and durability (< 0.7%
impact degradation) would be most appropriate for skid control on ice and snow covered
4-12
-------
70 r
CO
o
60
UJ
CC
U.
O
50
CO
UJ
m
a:
o
0.
cc
m
30
State Maintenance Sand
CLEAR ICE
ASPHALT SHEET BPN = 100
TEMPERATURE = SOT
0.1 0.2 0.3 _ 0.4
RATE OF APPLICATION CLB/FT2)
Figure 4-9. Average British portable skid number vs. rate of aggregate
application for various antiskid materials (30°F).
4-13
-------
70 r
CO
CD
o 80
co
LL.
a
i 50
CO
LU
m
o
a.
a:
ca
30
ASPHALT SHEET BPN » 93
TEMPERATURE » 0*F
0.1 0.2 0.3 _ 0.4
RATE OF APPLICATION (LB/FT2)
Figure 4-10. Average British portable skid number vs. rate of aggregate
application for various antiskid materials (0°F).
4-14
-------
280 r
240 -
Q MAINT. SAND o *4-t10 CURSHED
o 3/8-4=4 PIT RUN H CONCRETE SAND
3/8-*4 CRUSHED W COAL ASH
3 4
STOPS
Figure 4-11. Average stopping distances determined on an ice track
for various antiskid materials.
4-15
-------
roadways. Particle size, shape, and hardness are also related to durability as will be
discussed below.
4.3 MATERIAL DURABILITY
As mentioned previously, there were no data located in either the literature search
or the telephone survey which relate silt generation (and the associated PM10 emissions)
with the durability of antiskid materials. All work in this area has focused on construction
aggregates used in bituminous and concrete paving mixtures. Although not directly
applicable to antiskid abrasives per se, these data can be used to provide guidance in
proper material selection. The following discusses the physical and mineral properties
of antiskid materials as related to their ability to resist abrasion, impact, and crushing.
4.3.1 Aggregate Physical Properties
The three most important physical properties for good aggregate durability are
hardness, particle shape, and particle size. Each is discussed below.
Material hardness is probably the single most important factor in aggregate
durability. If a material is very hard it is capable of resisting abrasion, impact, and
crushing under applied mechanical loads. Hardness is, of course, a direct function of
particle mineralogy as will be presented later.
To determine the effect of hardness on abrasion resistance, controlled experiments
were performed by Stiffler (1969) where a mineral surface was wear tested with a
commercial abrasive (e.g., SiO2) through the action of a rolling loaded wheel applied to
the specimen. In these tests, Stiffler found that the volume of mineral removed from the
surface can be reasonably predicted based on Vick'ers hardness. The average data
collected in the study are shown in Table 4-3.
As shown by the data in Table 4-3, the amount of wear is inversely proportional
to the hardness of the mineral. This is illustrated in Figure 4-12 for different minerals
abraded with SiO2. These data would suggest that an antiskid abrasive with a Vickers
hardness > ~ 1,000 kg/mm2 should be acceptable for good durability. [Note that the
study performed by Furbush et al. (1972) recommended a Moh Hardness of 5-7 for good
skid resistance and the study by Hudec (1984) showed durability increasing with
hardness.]
With respect to particle shape, highly angular particles are almost universally
recommended for good durability in all applications tested (Dhir et al., 1971; Furbush
et al., 1972; Brant, 1972; Anon., 1979; Havens and Newberry, 1982; Connor and Gaffi,
1982). Without detailed petrographic examination, the determination of particle shape is
difficult, however.
4-16
-------
TABLE 4-3. AVERAGE MINERAL WEAR FOR DIFFERENT ABRASIVE TYPES
Abrasive wear (mm3 x 1 0"2)
Mineral
CaC03(c)«
CaC03(i)b
Slag
SiOa(f)c
Mullite
Si02(c)d
MgO
Zr02
SiC
AI203
Melt temp.
825
825
1400
1700
1810
1700
2620
2650
2200
2000
Modulus
(psixlO6)
—
—
—
6
21
7
42
21
50
.45
Yield stress
(psixlO4)
—
—
—
—
0.9
0.7
1.5
2.0
5.0
4.0
Specific
gravity
2.70
2.70
2.70
2.10
2.95
2.65
3.60
5.70
3.00
4.00
Vickers hardness
(kg/mm2)
460
400
620
1100
1720
2000
1240
1700
4500+
3300
AI202
42.2
38.0
24.0
11.0
11.0
7.7
3.4
2.4
0.7
0.3
Si02
30.7
40.9
16.3
11.0
6.5
6.5
1.3
0.6
0.4
0.7
MgO
13.5
12.5
9.4
2.6
3.1
1.7
0.2
0.2
0.07
0.03
a Crystallike with cleavage planes.
b Limestone chips.
0 Clear fused-quartz rod.
d Glasslike lump.
-------
50
05
CD
1 I I 11
iC03(
CaCO3(c)
\
Slag
SiO2(f)-o'
1 I i i i 11
Mullite
SIO2 (c)
45C
MgO-o
ZrO2-o -
i i i i i i i ti i i i i i i 11
100 1000
Vickers Hardness (kg/mm2)
10000
Figure 4-12. Mineral wear rate vs. Vickers hardness for SiO2 abrasive.
4-18
-------
Havens and Newberry (1982) determined that the shape of aggregate particles can
be determined indirectly from the measurement of void fraction using a version of ASTM
Method C-29 (see Section 3.3). They found that a void fraction > 50% indicates a greater
degree of disorder in particle shape and/or texture and thus is indicative of angular
particles. Also, Connor and Gaffi (1982) found that a minimum fracture of 80% on one
face for the material retained on the No. 10 sieve should be adequate for good skid
resistance.
The final physical property of importance in aggregate durability is particle size.
As stated in Section 3, Hudec (1984) determined that the durability of an aggregate
material (as determined by petrography) decreases with grain size. This is illustrated by
the relationship provided in Equation (3-3). When effectiveness is also considered (see
Section 5.2 above), particle sizes in the range of about 16 mesh seem to be exhibit the
highest skid resistance. It would seem that this specification would also be applicable to
good durability as well.
4.3.2 Particle Mineralogy
The hardness of an aggregate material is directly related to its mineralogy.
Numerous studies have investigated particle mineralogy with respect to durability in
paving mixtures (Dhir et al., 1971; Furbush etal.,'1972; Anon., 1979; Woodside and
Peden, 1983; Hudec, 1984; Goswami, 1984; Dubberke and Marks, 1985). These studies
have shown that aggregates containing high percentages of siliceous minerals such as
quartz, granite, chert, greywacke, etc., show the highest durability with respect to
abrasion, impact, and crushing (Stiffler, 1969; Anon., 1979; Hudec, 1984; Goswami, 1984).
To produce durable pavement mixtures, many transportation agencies have
adopted standards for aggregates which include minimum contents of various minerals.
For example, the State of Kentucky has specified a minimum quartz content of sand to
be 90% by visual count (e.g., ASTM Method C-295) or 94% by chemical analysis (Havens
and Newberry, 1982). Also, the U.S. Transportation Research Board (TRB) has
established a durability classification system for aggregates used in pavement mixtures
with respect to skid resistance. These classifications are as follows (Anon., 1979):
Group I—Outstanding Polish Resistance. Aggregates in this group are
heterogeneous combinations of hard minerals with a coarse-grained
microstructure of hard particles bonded together with a slightly softer matrix.
Examples include emery or industrial abrasives which are normally too
expensive for use in road pavement mixtures.
Group II—Above Average Polish Resistance. This group comprises a minority
of aggregate types used for highway paving, including blast furnace slags,
expanded shales, and crushed sedimentary rocks such as sandstone, arkose,
greywacke, and quartzite. Recommended for use on high traffic volume
roads.
4-19
-------
Group III—Average Polish Resistance. This group consists of aggregates
currently used in pavement construction, including crushed, dense igneous
and metamorphic rocks of the granite, granite gneiss, and diorite types.
Exhibits satisfactory skid resistance for all but the most severe conditions.
Group IV—Below Average Polish Resistance. All remaining aggregate mineral
types, except for Group V. Acceptable for use on low volume pavement
surfaces and where skid resistance requirements are below normal.
Group V—Low Polish Resistance. Typical of this group are the carbonate
minerals containing low levels of siliceous materials and uncrushed gravels.
Based on the above information, it can be concluded that aggregates used for skid
control purposes should be a Group III material or better (according to the TRB
classification) with a silica content of around 90%. These specifications should assure a
highly durable material with a low tendency for fines generation when exposed to
vehicular traffic. (Note, however, that some transportation agencies use aggregate
rejected for use in pavement as antiskid materials.)
4.3.3 Resistance to Abrasion. Impact, and Crushing
The ability of an aggregate to withstand the effects of abrasion, impact, and
crushing is a function of the material properties discussed above. Therefore, as a
measure of resistance to these effects, a number of standard tests have been developed
as discussed previously in Section 3. Based on MRI's analysis of the available literature,
it was found that the Los Angeles abrasion test is probably the best overall indicator of
aggregate durability for antiskid abrasives. This is illustrated below.
With respect to overall durability of construction aggregates, the work performed by
Woodside and Peden (1983) contains the most comprehensive comparison of aggregate
properties found in the literature. In this work, 13 aggregate samples were evaluated for
10 different strength parameters. Using this data set, MRI performed a multiple linear
regression analysis of selected parameters to determine their relationship to the
applicable Los Angeles abrasion value:
LAAV = 11.9 + 0.611 ACV + 1.10 AIV + 0.0209 S - 0.0837 F - 0.481 W - 6.06 SG (4-5)
where: LAAV = Los Angeles abrasion value (per ASTM C 131)
ACV = Aggregate crushing value, % (per British Standard 812)
AIV = Aggregate impact value, % (per British Standard 812)
S = Soundness, % (per ASTM C 88)
F = Flakiness, % (per British Standard 812)
W = Water absorption, % (per British Standard 812)
SG = Specific gravity (per British Standard 812)
4-20
-------
The correlation coefficient for the above regression is 0.990; this result indicates that
the combination of the six strength parameters are highly related to.the Los Angeles
abrasion value. Thus, these results indicate that a single test can reasonably be used as
an overall .measure of aggregate durability. It was therefore concluded that Phase 2 of
the program should include the determination of durability using the Los Angeles
Abrasion Method and that the results of this test be used to compare different aggregate
samples as related to silt generation under simulated traffic conditions.
4.4 INITIAL ACCEPTABILITY CRITERIA
Based on an engineering evaluation of information obtained from the literature
search and telephone survey, initial selection criteria have been developed for antiskid
materials that have a low target potential for silt generation in service. For example, these
criteria can probably be met by a washed construction aggregate (sand). Selection of
less durable abrasives because of cost/availability considerations would entail higher PM10
emissions unless mitigated by more extensive cleanup procedures. These criteria are
provided only as preliminary guidelines pending further laboratory evaluations in Phase 2
(see Section 6). The acceptability criteria derived from available data are shown in
Table 4-4.
TABLE 4-4. ACCEPTABILITY CRITERIA FOR ANTISKID MATERIALS
Type of Material or application
material parameter
Recommended
value
Abrasives General gradation
Silt content as applied
Moh hardness
Vickers hardness
Fracture
Void fraction
Quartz content
Salts Percent insoluble matter
> 16 mesh
< 1%
5-7
> 1,000 kg/mm2
> 80%
> 50%
> 90%
< 2%
4-21
-------
-------
SECTION 5
MATERIALS EVALUATION
Two different sample sets were evaluated in the program to develop a data base
for typical antiskid materials. The first set consisted of aggregate materials commonly
used in construction or readily available by-products of other processes. In the second
set were materials used for ice and snow control in PM10 nonattainment areas.
The following sections describe origin of the samples obtained and their physical
properties. Analysis of physical parameters resulted in the definition of five distinct
groupings (or clusters), which were used to select suitable samples for traffic tests in
MRI's cold room.
5.1 ORIGIN OF MATERIAL SAMPLES
The first sample set consisted of six different construction and by-product materials.
Five of these materials were obtained from local sources with the remaining sample
provided by a state regulatory agency. These samples were used for both a laboratory
evaluation of the modified test methods described in Section 3.5 and simulated traffic
tests. The first samples consisted of: calcium magnesium acetate (CMA); road salt (i.e.,
NaCI); ice control sand; crushed limestone; power plant bottom ash; and river rock from
Presque Isle, Maine. These samples are described in Table 5-1 along with the quantities
received.
In the second set of samples were 28 different materials obtained from state and
local regulatory agencies located in PM10 nonattainment areas. These samples consisted
of a wide range of natural and manmade aggregates available in their .local jurisdiction.
Of the original 34 materials, 28 were finally selected by the EPA Work Assignment
Manager (WAM) for physical/chemical analysis in the program. An inventory of the .
materials received along with those selected for analysis are also shown in Table 5-1.
5.2 MATERIAL PROPERTIES
Each material in both sample sets were analyzed for up to nine different properties,
as applicable. These properties included: Los Angeles abrasion loss; wet aggregate
durability; quartz content; silt content; Vickers hardness; unit weight; void fraction; particle
shape index; and percent insoluble matter. Table 5-2 provides the ASTM
5-1
-------
Sample origin
TABLE 5-1. SAMPLE INVENTORY
—
Sample designation Quantity received
Sample properties
fo
Chevron Chemicals
Missouri Dept. of Trans.
(DOT)
Holiday Sand & Gravel Ice control sand8
Calcium magnesium acetate8 50 Ib
Road salt8 3 x 75 Ib
Astro Quarries
Nearman Power Plant,
Kansas
Maine DEP
Colorado APCD
Wyoming DEQ
Montana Dept. of Health
& Environmental Services
Crushed limestone8
Bottom ash8
Maine DOT8
Presque lsleb
Department of Health13
Scoria-Sheridan*5
Washed-Sheridan"
Kalispell—new"
Kalispell—old
2x100lb
2 x 70 Ib
3 x 50 Ib
2 x 65 Ib
2 x 55 Ib
2x100lb
2 x 200 Ib
2 x 200 Ib
65 + 75 Ib
73 Ib
Round pellets
Carey salt in piles
Washed river sand; < 0.1%
silt
Nominal 1/4 in
Cinders from Wyoming coal
Presque Isle stockpile; "dirty"
sand
City stockpile; cleaner and
more durable
Washed squeege; 3/8 in
nominal
Dark red material
Washed sand; 95% > 3/8 in;
2% to 7% silt; wear % < 35
Washed and screened
Pit run material
(continued)
-------
TABLE 5-1 (continued)
Sample origin
Sample designation
Quantity received
Sample properties
Montana Dept. of Health
& Environmental Services
V
CO
Kalispell—2 yr oldb 72 Ib
Kalispell Dept. of Highways 90 + 80 Ib
Libby, Montana, St. Hwy. 56 + 48 Ib
Shop"
Libby, Montana, County Shop 48 + 56 Ib
Libby, Montana, City Sandb 70 + 69 Ib
Libby, Montana, City Crush6 57 + 68 Ib
Columbia Falls—Oldb
Columbia Falls—Newb
Missoula City—90
Missoula City—pre 1987b
Missoula—Montana State
Hwy.
Missoula County—90b
60 + 61 Ib
65 + 63 Ib
70 + 100 Ib
~ 100 Ib
~ 100 Ib
~ 100 Ib
Pit run material
Pit run material
Pit run material; screened
Pit run; screened to 1 in
Pit run material; screened
Crushed and screened to
Vain
Pit run material
Washed and screened
Wet durability > 70; < 3% silt
Unknown
4% to 10% silt
< 6% passing No. 100 screen
(continued)
-------
TABLE 5-1 (continued)
Sample origin
Sample designation
Quantity received
Sample properties
V1
Montana Dept. of Health
& Environmental Services
City of Thompson Falls,
Montana
City of Steamboat
Springs, Colorado
Butte/Silver Bow City/County
Sand Pile"
Ronan City Sandb
MDOH-Nine Pipes6
BIA—Lame Deer
St. Xavier"
Shipping Nos. 301663,
301674, and 301685"
Sand"
2 x 50 Ib
150lb
150lb
200 Ib
100lb
54 + 49 + 70 Ib
~ 50 Ib
Pit run material
Pit run material
Reject paving material
Unknown
Reject paving material
Pit run material
"Coarse" sand
Sand"
Squeegeb
Scoriab
~50lb
2 x 50 Ib
2 x 30 Ib
"Fine" sand
Unknown
Unknown
a Samples split and analyzed during laboratory evaluation.
b Samples split and analyzed during materials evaluation.
-------
TABLE 5-2. APPLICABLE TEST METHODS FOR ANTISKID MATERIALS
Type of material
Abrasives
Deicers
Material property
LA abrasion lossb
Wet durability6
Quartz content15
Silt content0
Vickers hardnessd
Void fraction6
Particle shape and
texture6
Insoluble matter—
NaClc
Insoluble matter—
CaMg acetate6
ASTM method
C131
D3744
C289
C 136
C117
E384
C29
C128
D3398
E534
E534
Method modified3
Yes
No
No
Yes
Yes
Yes
No
No
No
No
Yes
a See Section 3.5 for discussion of modified ASTM methods.
b Analyses performed by commercial testing laboratories.
Analyses performed by MRI. CaMg acetate = calcium magnesium acetate
(CMA).
d Analyses performed by Dr. Dick Hagni, University of Missouri-Rolla.
5-5
-------
methods used for each analysis along with an indication of whether the method was
modified for application to antiskid materials.
The test data obtained for both sample sets are provided in Tables 5-3 and 5-4
For cases where a modified ASTM method is used, Table 5-3 shows the results of
triplicate analyses performed on the first set of six materials. Also noted in these tables
are materials selected for further testing under simulated traffic conditions in MRI's cold
room.
For those samples where triplicate analyses were performed, range percent is
provided as a measure of method precision. As shown in Table 5-5, the "best" and
"worst" precision was obtained for the same test—LA abrasion loss. The results indicate
that additional development is probably needed on this method as well as the other
methods to improve reproducibility.
5.3 MATERIALS CLASSIFICATION ANALYSIS
The laboratory data generated in this study were subjected to exploratory statistical
analysis techniques. The effort was guided by two general questions. These are:
1. For the 25 materials considered in this program, do the physical parameters
(i.e., material properties) exhibit significant interrelationships?
2. If the material properties show significant intercorrelation, can these
relationships be used to define distinct material types?
These questions were addressed using linear correlations as well as multivariate
pattern recognition techniques—principal components analysis (PCA) and cluster analysis.
Details of these techniques can be found in many standard statistical texts. The following
concentrates on presentation of the results indicated by application of each of the
techniques to the data sets developed in this study.
5.3.1 Linear Correlation—Material Properties
Table 5-6 presents the correlation matrix of material properties; it is based on the
laboratory data for the 25 materials (see Tables 5-3 and 5-4). (Note: Both deicers and
limestone were eliminated from this analysis.) Examination of the matrix indicates two
main points. First, there apparently is a set of interrelationships between material silt
content and the corresponding abrasion and wet durability scores. The positive
correlation between silt and abrasion suggests that materials with high silt content will
readily abrade (i.e., produce substantial additional "fines" when subjected to mechanical
forces found in the LA abrasion machine). Of course, the linear correlation also implies
that the opposite condition—low silt content, low abrasion—holds true. Figure 5-1
presents a scatterplot of the correlation between silt content and the abrasion index.
5-6
-------
TABLE 5-3. RESULTS OF LABORATORY EVALUATION
Material sample
Washed construction sand (Holiday S&Q
Ice control sand)
• Sample 1
• Sample Z
• Sample 3
• Average
Maine DOT Presque Isle stockpile'
• Sample 1
• Sample 2
• Sample 3
• Average
Crushed limestone
• Sample 1
• Sample 2
cn
' • Sample 3
• Average
Boiler bottom ash (cinders)
• Sample 1
• Sample 2
• Sample 3
• Average
Los Angeles
abrasion loss Wet aggregate Quartz content0 Silt content
(wt %)* durability lndexb (mmol/L) (wt %)d
.
0.90 95 28.1 0.02
1.43 - _ _
1.88 _ _ _
1.33 _ _ _
8.35 60 19.8 4.31
8.04 _ _ _
7.10 _ _ _
7.83 _ _ _
7.03 43 None detected 1.58
7.83 — _ _
7.08 — _ _
7.31 _ _ _
7.21 68 39.4 4.41
7.21 - _ _
7.42 — — —
7.28 _ _ _
Vlckers
hardness
(kg/mm2)*
1,238
798
755
930
928
775
859
854
85.5
105
93
95
343
258
428
343
Particle Insoluble
Unit weight Void fraction shape matter
(kg/m3)' (wt%)' Index" (%)h
1.799 30.2 8.3 NA
- - - NA
— - — NA
- — — NA
1.859 23.1 8.1 NA
- - - NA
- - - NA
- - - NA
1.497 40.3 12.1 NA
— — - NA
— — - NA
— — - NA
1.223 48.0 21.1 NA
— — — NA
- - - NA
- - - NA
(continued)
-------
TABLE 5-3 (continued)
V
oo
Material sample
MO DOT road salt1
• Sample 1
• Sample 2
• Sample 3
*
• Average
Calcium magnesium acetate (CMA)
• Sample 1
• Sample 2
• Sample 3
• Average
—
Los Angeles
abrasion loss
fwt%)«
0.68
•
-
0.30
0.31
-
0.30
======
Wet aggregate
durability lndexb
NA
NA
NA
NA
NA
NA
NA
NA
—
Quartz content0 SIM content
(mmol/L) (wt %)d
NA 0.02
_
NA —
NA _
NA 0.02
NA _
NA _
NA _
Vlcker's
hardness
(kg/mm2)0
NA
NA
NA
NA
NA
NA
NA
NA
—
Unit weight
fkg/m3)'
NA
NA
NA
NA
NA
NA
NA
NA
"— ii-L'l'l ••!•
Void fraction
NA
NA
NA
NA
NA
. NA
NA
NA
Particle
shape
IndexS
NA
NA
NA
NA
NA
NA
NA
NA
— ' - ~
Insoluble
matter
0.58
0.50
1.23
0.77
2.03
1.72
1.88
1.87
a Per modified ASTM Method C-131.
Dlmenslonless. Per ASTM Method D-3744.
e Per ASTM Method C-289.
Silt - % < 200 mesh or 75 MmP. Per ASTM Methods C-138 and C-117 as modified by MRI.
Per modified ASTM Method E-384. Average of 20 Indentations per polished section. NA - not applicable.
Per ASTM Methods C-29 and C-128. NA - not applicable.
9 Dlmenslonless. Weighted average per ASTM Method D-3388. NA • not applicable.
h Road salt (Nad) per ASTM E-584. CMA per modified ASTM E-584. NA - not applicable.
1 Samples exposed to simulated traffic.
-------
TABLE 5-4. RESULTS OF MATERIALS EVALUATION
CD
Material sample
Presque Isle, Maine— City
Stockpile
Colorado APCD— Washed
Squeegeh
Sheridan, Wyoming— Scoriah
Sheridan, Wyoming— Washed
Sand
Kalispell, Montana— "New"
Kalispell, Montana— "Old"
Libby, Montana— State
Highway Shop
Libby. Montana— City Sand
Llbby, Montana— City
Crushed Aggregate
Columbia Falls, Montana—
"Old-
Columbia Falls, Montana—
"New"h
Missoula, Montana— County
1990
Missoula. Montana— Pre-1987
Butte/Sllver Bow, Montana—
City/County Sand Pile
St Xavler, Montana
Thompson Falls. Montana
Steamboat Springs,
Colorado— "Coarse" Sandh
Steamboat Springs,
Los Angeles
abrasion loss
(wt %)'
4.01
3.84
13.1
5.83
5.49
7.06
7.82
6.51
6.41
6.30
3.62
5.60
17.1
5.85
3.62
11.6
3.95
3.89
^
Wet aggregate
durability indexb
68
97
90
94
68
45
72
86
75
78
63
62
56
85
87
56
85
78
Quartz content0
(mmol/L)
17.4
15.9
155.3
10.0
36.5
29.1
28.9
31.6
22.7
33.8
32.5
34.4
22.6
11.4
187.8
27.5
34.2
22.3
Silt content
(wt %)d
1.24
0.09
4.90
1.82
0.99
6.82
3.56
4.91
3.02
1.97
2.47
1.77
5.26
1.43
4.14
8.93
0.25
(continued)
1.22
=====
Vlckers hardness
(kg/mm2)8
88.7
1.042
607
443
148
1,052
465
840
676
955
765
280
755
620
828
689
554
493
Unit weight
1,750
1,627
1.348
1.789
1.651
1,863
1.776
1.916
1.633
1,784
1.641
1,698
1,654
1.656
1.683
1,816
1,669
1.715
Void fraction
(wt %)'
342.8
36.4
33.2
32.4
32.8
25.4
29.7
26.5
38.5 .
26.5
38.5
34.6
33.0
30.9
33.1
26.1
33.2
33.2
Particle shape
Index9
11.9
15.5
12.7
15.3
11.1
12.8
8.67
8.07
21.2
10.7
20.5
17.6
6.5
13.6
14.6
11.3
8.6
mo
Colorado—"Fine" Sand
-------
TABLE 5-4 (continued)
V
— L.
o
Los Angeles
abrasion loss
Material sample (wt. %)*
Steamboat Springs, 6.49
Colorado— Squeege
Steamboat Springs, 8.72
Colorado — Scoria^
Ronan, Montana— City Sand 10.7
MDOH— Nine Pipes 8.24
Wet aggregate Quartz content" Silt content VIcker's hardness Unit weight
durability Index" (mmol/L) (wt %)d (kg/mm2)" (kg/mV
80 31.6 1.17 503 1,651
100 39.9 2.28 458 856
73 20.3 4.99 408 1,855
72 32.2 4.21 1.163 1,742
Void fraction Particle shape
(wt %)' Index0
34.7 14.2
45.1 27.4
27.5 8.95
* Per modified ASTM Method C-131.
b Dimensionless. Per ASTM Method D-3744.
c Per ASTM Method C-289.
d Silt = % < 200 mesh or 75 /jmP. Per modified ASTM Method C-136.
e Per modified ASTM Method E-384. Average of 20 Indentations per polished section.
' Per ASTM Methods C-29 and C-128.
8 Dimensionless. Weighted average per ASTM Method D-3398.
h Samples exposed to simulated traffic.
-------
TABLE 5-5. RANGE PERCENT VALUES FOR MODIFIED ASTM METHODS
Parameter Sample identification Range percenf
LA abrasion loss Washed construction sand 57.1
Maine DOT Presque Isle 16.0
Crushed limestone 10.9
Boiler bottom ash (cinders) 2.9
Vickers hardness Washed construction sand 51.9
Maine DOT Presque Isle 17.9
Crushed limestone 20.5
Boiler bottom ash (cinders) 49.6
Insoluble matter Calcium magnesium acetate 16.6
Range percent = h®h value-'ow vaiue x 100
mean value
5-11
-------
ro
Material property
LA abrasion loss
Wet aggregate durability
Quartz content
Silt content
Vickers hardness
Unit weight
Void fraction
Particle shape index
-^ — •
abrasion durability index
1.0
-0.34
0.09
•0.68"
-0.06
-0.16
-0.11
-0.13
1.0
0.23
-0.55a
0.07
-0.35
0.32
0.18
content
1.0
0.21
0.06
-0.25
0.09
0.08
content
•
1.0
0.18
0.12
-0.34
-0.11
v iv_-r\ci o
hardness
1.0
0.29
-0.38
-0.27
Ul III
weigh
1.0
-0.84"
-0.72"
Underlined coefficients significant beyond the 99% level.
1.0
0.79a
1.0
-------
10
8 -
r e i-
~c
O
O
±=1 4 h
CO
2 -
0
0
5
10
15
20
Abrasion Index (Wt. %)
Figure 5-1. Scatterplot of silt content vs. LA abrasion loss.
5-13
-------
The second point indicated by the correlations in Table 5-6 involves the apparent
interrelationships between unit weight, void fraction, and the particle shape index The
correlations are highly significant in a statistical sense and, from a physical viewpoint are
internally consistent. For example, the negative correlations between unit weight (i e
density in mass/volume) and void fraction (wt. %) imply that a unit volume of less dense
material contains a high percentage of "air space," or alternatively, high density materials
contain relatively little air. Similarly, the positive relationship between shape index and
void fraction indicates that materials composed of particles that are more angular
necessarily exhibit a greater void fraction. Figure 5-2 presents an example scatterplot of
the correlation between void fraction and particle shape index.
5.3.2 Multivariate Analyses
Principal component analysis (PCA), as used in this study, is based on the
correlation matrix (Table 5-6). In essence, PCA serves to formally define the observations
made above concerning the underlying patterns inherent in the correlation matrix
Specifically, the PCA results indicate that the variation in the eight key abrasive properties
can be compactly expressed in terms of three composite variables. These composite
variables may be described as follows:
1. Variable l-Particle size/shape including void fraction, unit weight and
particle shape. '
2. Variable Il-Degradability including silt content, abrasion loss and wet
aggregate durability.
3. Variable Ill-Silica/hardness including quartz content and Vickers hardness.
From an analytical perspective, the important result of the PCA is that the
composite variables contain the physical information or "signal" exhibited in the raw
material properties data. From a practical viewpoint, the composite variables (from PCA)
form a convenient basis for determining whether the sample test materials, in fact can
be grouped into a relatively small number of distinct material "types."
To address the question of grouping materials into types, the multivariate statistical
technique commonly referred to as cluster analysis was used. The results of this analysis
suggest that based on experimentally determined physical properties, the test materials
can be placed into five types (or clusters). Table 5-7 shows the breakdown of test
materials into characteristic types.
In order to gain some appreciation for the differences between material types
average values were calculated for each type and for each of the material properties (i e '
abrasion through particle shape index). In turn, these average values are expressed as
a percentage of the respective overall means (i.e., average values based on all 25 test
materials). Figure 5-3 provides a graphical representation of these results.
5-14
-------
20
10 15 20 25
Particle Shape index (dimensionless)
Figure 5-2. Scatterplot of percent void fraction vs. particle shape index.
30
5-15
-------
V1
_x
O)
Presque Isle, ME—City Stockpile
Sheridan, WY-Washed Sand
Kallspell, MT—"New"
Llbby, MT-State Highway
Libby, MT—City Sand
Libby, MT—City Crushed Aggregate
Columbia Falls, MT—"Old"
Columbia Falls, MT—"New"
Missoula, MT—1990
Butte/Silver Bow, MT—Sand
Steamboat Springs, CO—"Fine Sand"
Steamboat Springs, CO—Squeege
Steamboat Springs, CO-Scorla Sheridan, WY-Scoria CO APCD-Washed Squeege
Boiler Bottom Ash St. Xavler, MT
Kalispell, MT—"Old-
Steamboat Springs, CO-«Coarse" Sand Missoula, MT-Pre-1987
Washed Construction Sand Thompson Falls-MT
Ronan, MT—City Sand
MOOT—Nine Pipes
Maine DOT Presque Isle
-------
200
150
03
CD
0
» [
CD
•3 100
03
Cl
CD
Q
"c
CD
0
CD
°~ 50
n
A
i i i i ^
*
A ~
+ .
• o
• o •
* Q • Q
- S A $ *
n
* ?
A s" A A
i i i f i
* PSHAPE
+ VOIDFRAC
n UNITWGT
a VICKERS
A SILT
A QUARTZ
O WETDUR
© ABRASION
C1 C2 C3 C4 C5
Cluster Type
Figure 5-3. Percent deviation from mean value by cluster type.
5-17
-------
Examination of Figure 5-3 suggests that the material types can be characterized
as follows:
C1-Materials exhibiting relatively low silt (0.99% to 4.9%) and quartz
(10to36mmol/L) contents; other material properties near the overall
means (see Table 5-8).
C2—Coarse and soft materials as indicated by high particle shape
(21 to 27) and void fraction (45% to 46%) indices along with relatively low
unit weight (860 to 1200 kg/m3) and Vickers hardness values
(340 to 460 kg/mm2).
C3-Materials that exhibit both high quartz (150 to 190 mmol/L) and high
silt (4.1% to 4.9%) contents; other material properties near the overall
means (see Table 5-8).
C4—Relatively hard materials as indicated by low silt content (0 02% to
0.25%), low abrasion loss (0.90% to 4.0%), and high values for the Vickers
hardness (550 to 1200 kg/mm2) test.
C5—Finely divided materials that are easily abraded; indicated by high silt
content (4.3% to 8.9%) and high abrasion losses (7.1% to 17%).
Summary statistics for the entire experimental data set as tabulated by cluster tvoe
are shown in Table 5-8.
5-18
-------
TABLE 5-8. SUMMARY STATISTICS-MATERIAL PROPERTIES BY CHARACTERISTIC CLUSTER TYPE
CO
Parameter/type-
Mean
C1
C2
C3
C4
C5
E
Minimum
C1
C2
C3
C4
C5
E
Maximum
C1
C2
C3
C4
C5
E
LA abrasion loss
(wt%)
5.85
7.98
8.38
2.90
11.0
6.89
3.62
7.21
3.82
0.90
7.06
0.90
8.24
8.72
13.1
3.95
17.1
17.1
Wet aggregate
durability Index
75.5
84.0
88.5
92.3
58.0
75.7
82.0
68.0
87.0
85.0
45.0
45.0
94.0
100.0
90.0
97.0
73.0
100.0
Quartz content
(mmoL/U
26.6
39.6
171.8
25.4
25.0
38.8
10.0
39.4
155.0
15.9
19.6
10.0
36.5
39.9
187.8
34.2
29.1
187.8
Silt content
(wt%)
2.29
3.34
4.52
0.12
6.06
3.05
0.99
2.28
4.14
0.02
4.31
0.02
4.91
4.41
4.90
0.25
8.93
8.93
Tickets hardness
(kg/mm2)
572
400
718
945
766
654
88.7
343
607
554
408
88.7
1,163
458
828
1,238
1,052
1,238
Unit weight
(Vg/rrO
1,723
1,040
1,516
1,698
1,809
1,666
1,633
856
1,348
1,627
1,654
856
1,916
1,223
1,683
1,799
1,863
1,916
Void fraction
(wt%)
32.2
45.8
33.2
33.3
27.0
32.5
26.5
45.1
33.1
30.2
23.1
23.1
38.5
46.0
33.2
38.4
33.0
46.0
Particle shape
Index
13.3
24.2
13.6
10.1
9.53
13.1
8.07
21.1
12.7
6.3
6.5
6.30
21.2
27.4
14.8
15.5
12.8
27.4
-------
-------
SECTION 6
SIMULATED TRAFFIC TESTS
Two sets of experiments were performed in MRI's cold room. In the first set,
selected antiskid materials were exposed to simulated traffic conditions on a wheel
passage machine designed and constructed by MRI. In the other experiments, small
scale tests were performed on common chemical deicers to determine the amount of
solid material potentially available for resuspension. Both are described in detail below.
6.1 MATERIAL SELECTION
A total of seven abrasives and one deicer were evaluated on MRI's wheel passage
machine. For the abrasives, at least one sample from each of the five material groups
determined from the cluster analysis (Section 5.3) were tested as selected by the EPA
work assignment manager (WAM). Table 6-1 lists the specific antiskid materials tested
in the program.
For the seven abrasives listed in Table 6-1, a mixture of 90% abrasive and 10%
NaCI (MO DOT road salt) was used for test purposes. This abrasive/deicer mixture
reflects that typically used by local transportation agencies for ice and snow control as
determined from the literature search and telephone survey.
With respect to the quantity of material applied during the wheel passage tests, six
equivalent applications of 4,000 Ib/lane mi were selected for all antiskid abrasives tested.
This rate is indicative of the maximum application currently used by transportation
agencies contacted in the telephone survey. The maximum application was used in the
study to provide a sufficient amount of material on the track for adequate sample
collection and analysis. In the case of the single deicer tested on the wheel passage
machine, six applications of the AASHTO recommended rate of 500 Ib/lane mi were used
during simulated traffic testing.
In addition to the simulated traffic tests, three common chemical deicers were also
evaluated in small scale experiments. The purpose of these experiments was to
determine the amount of solid material potentially available for resuspension on paved
roadways as well as to evaluate applicable sampling techniques for use in the simulated
traffic tests. One equivalent application of 500 Ib/lane mi, plus five reapplications of
250 Ib/lane mi, was used for the deicers tested.
6-1
-------
TABLE 6-1. ANTISKID MATERIALS EXPOSED TO SIMULATED TRAFFIC8
======= =g"
Test No- Sample No. Sample identification
1 &1Db
2
3
4
5
6
7
8
1 &1D
2
3
4
5
6
8
9
Holiday Sand and Gravel ice
control sand0
Maine DOT Presque Isle stockpile0
Steamboat Springs scoriad
Columbia Falls newd
Colorado APCD washed squeeged
MO Department of Transportation
(DOT) road salt0
Steamboat Springs coarse sandd
Sheridan, WY scoriad
All abrasives combined with NaCI in a 90:10 mixture.
Test No. 1 was performed with ice on the test track as was the case in
the other tests. Test 1D was performed dry without ice on the track.
See Table 5-3 for material properties.
See Table 5-4 for material properties.
The deicers evaluated in the small scale tests were MO DOT road salt; CMA- and
Peladow® (a commercial CaCI compound). The properties of Peladow® are'as follows:
Manufacturer: Dow Chemical USA
Lot Number: ML861006
Assay: 91.0% CaCI2 (ASTM E 449); 2.16% KCI; 1.7% NaCI; and 003%
MgCI2
6.2 TRAFFIC TESTS
A total of eight simulated traffic tests were performed in MRI's cold room The
following describes the test apparatus, protocol, and results obtained in the experimental
program.
6.2.1 Description of Test Apparatus
The MRI wheel passage machine, shown in Figure 6-1, was designed and built to
duplicate the forces exerted on road surfaces by a conventionally loaded tire. The
6-2
-------
Figure 6-1. MRI wheel passage machine.
6-3
-------
machine was a modified design of the standard circular track polishing machine (ASTM
E660) used in pavement surface wear research.
The wheel passage machine was designed with four wheel assemblies attached
to a central frame. The central frame is chain-driven by a gear reduction/motor unit The
final rotational speed is accomplished with the use of different-sized sprockets Identical
gears are used on the driver and the driven to obtain the required 32 rpm. A 8-1 ratio
in gear size is used to obtain the 4 rpm required in the first stage of the test.
The desired normal load is induced by two 375 Ib/in compression springs located
on each wheel assembly. The springs were adjusted to deliver a footprint pressure of
28 to 30 psi on the track surface. Each wheel assembly pivots somewhat reducing the
amount of side abrasion at the tire-track interface caused by the small turning radius of
the machine. To obtain full track coverage by the tires, two of the four wheels were offset
from one another.
For the purpose of these particular tests, the track of the wheel passage machine
was reconstructed using a 2-in thick layer of Type 1 Portland cement concrete having a
50/50 coarse/fine sand mixture. The aggregate used in the mix design had a maximum
size of 3/8 in limestone with a rated slump of 2.5 ± 1 in and a compressive strength
greater than 4000 psi. The concrete was non air-entrained and had a density of
145 Ib/ft. The concrete track was cured for many months before testing.
The concrete track was sandblasted and epoxied to the base plate of the machine
Masonite shielding was installed around the inside and outside diameters of the track and
extended high enough to contain debris from leaving the track area. Interior curbing was
built using a conventional sand cement mixture. The curbing forced paniculate to drop
back into the tracks of the tires.
After Test No. 6 the track was reconditioned by again sandblasting the surface and
laying a new, thin coat of the same sand—Portland cement mix as above onto the
cleaned track. The cure time was about 30 days. This technique proved to be
unsatisfactory because of the friability of the top layer on the track. This friability caused
substantial problems during Test Nos. 7 and 8 as discussed below.
6.2.2 Test Protocol
6.2.2.1 Pretest Preparations--
Prior to each test, the track surface was cleaned with a wire brush followed by
water and ethyl alcohol rinses. The surface was then visually inspected to assure that all
tire debris and other loose material had been removed. If required, the brushing/rinsing
procedure was repeated to obtain a clean track surface free of residue.
After cleaning the test track, the cold room temperature was decreased to 15°F
overnight. Approximately 2 L of chilled water was then added to the track surface and
6-4
-------
allowed to freeze overnight. This produced a layer of ice approximately 0.125 in thick
suitable for testing. The cold room temperature was then raised to 25°F and the
equipment allowed to equilibrate prior to testing.
Finally, to complete pretest preparations, the abrasive and/or deicer sample to be
evaluated was split from the bulk material using a riffle, placed in the cold room, and
equilibrated to a temperature of 25°F. Six equal quantities of test material were then
weighed into aluminum containers and the data recorded appropriately. The remaining
bulk material (after splitting out the track sample) was analyzed for silt content using the
method outlined previously in Table 3-5, which is reproduced here as Table 6-2 (Cowherd
etal., 1988).
6.2.2.2 Test Procedure-
To begin the test, the contents of one sample container were applied to the track
surface by broadcasting. The wheel passage machine was then started and the center
shaft rotated at 4 rpm for a period of 1 h. During this period, the ice layer started to
become "slushy" and/or "muddy" with some free brine present.
At the end of the first hour, the material was manually redistributed (by scraping)
into the center of the track and the contents of the second container applied to the
surface. The machine was then restarted for another one hour period with the material
again being redistributed at the end of the period. This procedure was repeated four
additional times until all six samples had been applied to the test track.
After the final application of antiskid material (i.e., after 6 h of run time), the
machine was allowed to operate continuously until an equivalent of 5,000 cumulative
vehicle passes was reached. During this period, the material was redistributed into the
center of the track at approximately 60-min intervals. At the end of 5,000 passes, the
machine was stopped and heat lamps applied overnight to melt any remaining ice and
evaporate the standing water. The cold room temperature was also adjusted to 35°F for
the remainder of the test. The drying procedure produced a visually dry but "pasty"
material on the track surface with a high surface moisture content (see Section 6.2.4
below).
After completion of the first 5,000 passes, a small grab sample was collected from
the track for moisture analysis. (Note that silt analysis of this small sample was not
feasible.) The machine was also reconfigured to rotate the center shaft at 32 rpm for the
remainder of the test duration.
After the first track sample was collected at 5,000 passes, the machine was
restarted and allowed to run for an additional 10,000 vehicle passes. During this period,
the apparatus was stopped approximately every half hour and the material redistributed
into the center of the track by gentle brushing. The machine then operated in 10,000'
pass increments until a maximum of 65,000 equivalent vehicle passes was reached and
the test terminated.
6-5
-------
Select the appropriate 8-in diameter, 2-in deep sieve sizes. Recommended U S
Standard Series sizes are 3/8 in, No. 4, No. 20, No. 40, No. 100, No 140
No. 200, and a pan. Comparable Tyler Series sizes can also be utilized ' The
No. 20 and the No. 200 are mandatory. The others can be varied if the
recommended sieves are not available or if buildup on one particular sieve
during sieving indicates that an intermediate sieve should be inserted.
2. Obtain a mechanical sieving device such as vibratory shaker or a Roto-Tap.
3. Clean the sieves with compressed air and/or a soft brush. Material lodged in
the sieve openings or adhering to the sides of the sieve should be removed (if
possible) without handling the screen roughly.
4. Obtain a scale (capacity of at least 1,600 g) and record make, capacity smallest
division, date of last calibration, and accuracy (if available).
5. Tare sieves and pan. Check the zero before every weighing. Record weights.
6. After nesting the sieves in decreasing order with pan at the bottom dump dried
laboratory sample (probably immediately after moisture analysis) into the top
. sieve. Brush fine material adhering to the sides of the container into the top
sieve and cover the top sieve with a special lid normally purchased with the pan.
7. Place nested sieves into the mechanical device and sieve for 20 min Remove
the pan containing the minus No. 200 material and weigh. Replace pan beneath
the sieves and sieve for another 10 min. Remove pan and weigh. When the
differences between two successive pan sample weighings (where the tare of
the pan has been subtracted) is less than 3.0%, the sieving is complete.
8. Weigh each sieve and its contents and record the weight. Check the zero
before every weighing.
9. Collect the laboratory sample and place the sample in a separate container if
further analysis is expected.
6-6
-------
At the end of each 10,000 pass increment, two material samples were collected
from different portions of the test track by a combination of hand sweeping and
vacuuming. Two equal areas of the track surface were sampled in a manner similar to
that used in traditional paved road studies (Cowherd et al., 1988). In this case, however,
a small brush and pan followed by a small, battery powered vacuum (Mini Vac® equipped
with a 37-mm filter cassette and Millipore Type AA filter) was used for sample collection.
The two samples were analyzed separately for moisture content and then combined for
silt analysis. (Note that due to the small sample size, the standard moisture and silt
analysis methods were modified as described later in this report.)
Finally, at the completion of each test, all material remaining on the track surface
was vacuumed into "tared bags and analyzed for silt content. In this case, however, a
larger hand-held vacuum (Dirt Devil®) was used for sample collection. Standard analysis
procedures were used for the determination of silt content as shown previously in
Table 6-2.
6.2.3 Analytical Methods
To meet the unique analytical requirements of the program, several special
methods were used to analyze the samples collected from the test track. The following
outlines the modified analytical methods used during the simulated traffic tests.
6.2.3.1 Analysis of Abrasives-
At 15,000 passes, and every 10,000 passes thereafter, the two small samples
collected from the test track were analyzed for moisture and silt content using modified
versions of the standard analytical methods for these parameters. In the case of moisture
content, the standard MRI technique (Table 6-3) was followed except that the drying time
was limited to 1 h. This was deemed appropriate due to the small size of the samples
collected and the substantial amount of organic tire debris (e.g., carbon black) visibly
present in the samples.
For determination of silt content, the abrasive material was hand sieved by
brushing through a 200 mesh screen. Brushing of the sample was found necessary
since the small particles contained in the samples tended to agglomerate and blind the
larger screens when a full sieve stack was attempted. Otherwise, the same weighing and
data recording procedure outlined in Table 6-2 was used.
6.2.3.2 Deicer Analyses-
For road salt, the samples collected from the test track were extremely small (i.e.,
a few tenths of a gram) and appeared to contain substantial quantities of tire wear
particles. Therefore, special methods were used to evaluate ooth the particle size
distribution and chemical (i.e., NaCI) content of the samples collected.
Each sample was sonically sieved using the general technique outlined in Table 6-4
(Kinsey and Coveney, 1986). In this case, however, no mechanical sieving was
6-7
-------
1.
2.
3.
4.
5.
6.
7.
8.
TABLE 6-3. MOISTURE ANALYSIS PROCEDURE
1 _____^
Preheat the oven to approximately 110°C (230°F). Record oven temperature.
Tare the laboratory sample containers which will be placed in the oven Tare
the containers with the lids on if they have lids. Record the tare weiqhtfe)
Check zero before weighing.
Record the make, capacity, smallest division, and accuracy of the scale.
Weigh the laboratory sample in the container(s). Record the combined
weight(s). Check zero before weighing.
Place sample in oven and dry overnight.8
Remove sample container from oven and (a) weigh immediately if uncovered
being careful of the hot container; or (b) place tight-fitting lid on the container
. and let cool before weighing. Record the combined sample and container
weight(s). Check zero before weighing.
Calculate the moisture as the initial weight of the sample and container minus
the oven-dried weight of the sample and container divided by the initial
weight of the sample alone. Record the value.
Calculate the sample weight to be used in the silt analysis as the oven-dried
weight of the sample and container minus the weight of the container
Record the value.
Dry materials composed of hydrated minerals or organic materials like coal and
certain soils for only 1V2 h. Because of this short drying time, material dried for
only 11/2 h must not be more than 2.5 cm (1 in) deep in the container
6-8
-------
TABLE 6-4. SONIC SIEVING PROCEDURE
1. Obtain a sonic sieving apparatus.
2. Select the appropriate sieves that are optionally available for use with the above machine. The
sieves commonly utilized are the 53-/jm, 20-/im, and 10-pm sieves.
3. Obtain an analytical balance with the smallest division being 0.01 mg.
4. The material to be sieved on this machine is bottom pan (< 75 /im) catch from the mechanical
silt analysis procedure (Table 2-8).
5. Clean the sieves in a low wattage (< 800 watts) ultrasonic bath, taking care to immerse the
sieves edgewise only in the solvent. The use of Freon T.F. as a solvent is recommended
because of its fast drying time. Once the sieves are clean, handle them only with cloth gloves
to prevent contamination and static charge buildup. Latex gloves are not recommended.
6. Tare weigh the sieves and the catch pan. Weigh each sieve and the catch pan three times,
alternating sieves between each weighing, and record each weight. Calibrate the balance with
Class S weights prior to each weigh period, and periodically check the balance during its use.
7. After nesting the sieves in decreasing order of size, place the sample into the top sieve. The
sample should weigh between 0.5 and 1.0 g. The sample weigh boat must be tare weighed
prior to receiving the sample and again after the sample is introduced to the top sieve. The
difference is subtracted from the sample weight and is recorded as material lost due to
handling. Due to the hygroscopic nature of oven-dried soils, the sample will gain moisture.
Thus the work must be done quickly without stopping for any length of time during the entire
test cycle.
8. Once the material is placed on the top sieve, cover the top sieve with the sound wave
generating diaphragm and place the sieves in the sonic shaker. The total sieving time is
5.0 min. With the sonic shaker in the sieve mode and the amplitude set on 2, sieve the
material. Increase the amplitude to 5 after 1 min. After 1.5 min (elapsed time), switch the
machine to the sift/pulse mode until the end of the test. If during the sift/pulse mode
appreciable amounts of material cqllect on the sieve wall, carefully tap the sides where the
particulate is adhering using a wooden stick.
9. After the 5-min sieving, remove the sieves and promptly weigh each sieve as rapidly as
possible. Repeat the weighing two more times. Record each weight and do not interrupt this
procedure.
10. Calculate the average tare and final weights of each sieve and pan. Subtract the tare from the
final weight to find the average amount retained on each sieve and pan. Calculate both the
mass and the percentage retained. Record these values on the data sheet.
6-9
-------
performed to remove larger particles prior to sonic sieving. Instead, a 75-um (200 mesh)
screen was used as the top size which was followed by a 53- and 20-,Tm screen and
associated pan Also, the entire sample was analyzed instead of the amounts specified
in otep / of Table 6-4.
Since samples appeared to contain tire debris, a chemical analysis was performed
to determine the NaCI content of each deicer sample. This analysis was performed bv
an aqueous extraction of the entire sample (recombined after sonic sieving) followed bv
an analysis for Na+ and CI'. The analytical results showed that the water soluble fraction
of each sample (total dissolved solids) was composed of almost 100% NaCI with a small
amount of other undefined material (assumed to be either track or tire debris)
6.2.4 Test Results
6.2.4.1 Abrasive Test Results-
The moisture and silt data for the seven antiskid abrasives tested are shown in
hgures 6-2 to 6-7. In these figures, data for the original bulk material along with that for
the 6 small track samples are plotted against equivalent vehicle passes. The individual
data points have also been connected to produce an overall "degradation curve" for
each test.
As shown by the above figures, both moisture and silt content vary substantially
as a unction of vehicle passes. For example, the moisture content of the material rises
rapidly at first, dries due to wheel friction, and stabilizes at about 25,000 passes.
In the case of silt content, the results would indicate a rapid increase in fines
generation until around 35,000 passes, at which point silt content tends to level off Thus
if additional testing was to be performed, the test could be terminated at 35 000 passes
without adversely affecting data quality. In addition, the silt data also show that Sample 1
(from data cluster 4-see Section 5.3) exhibits the lowest overall increase in fines as
compared to Sample 2 (from cluster 5) which exhibits the highest increase The
significance of these results is discussed later in this report.
It should also be noted that two different traffic tests were performed on the one
of the antiskid abrasives (i.e., Holiday Sand and Gravel ice control sand) shown in
Figures 6-2 and 6-5. Sample I was ice control sand tested using the method described
above whereas Sample 1D indicates the same material evaluated without ice on the track.
As shown by the silt data in Figure 6-5, the amount of fine material in Sample 1 is
higher than Sample 1D at traffic levels above about 15,000 passes. These results might
suggest that abrasive wear during exposure to the ice melting process could be more
pronounced as compared to the "no-ice" case.
A mass balance was also prepared for each abrasive tested using the experimental
data. In this determination, the original amount of material applied to the track was
6-10
-------
PERCENT MOISTURE VS VEHICLE PASSES
LU
£
Z)
J-
c/2
O
2
LJ
O
<
ll.
QL
W
Z
LJ
O
a
LJ
QL
0
SAMPLE 1
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
+ SAMPLE 2
o
60
SAMPLE 3
Figure 6-2. Percent moisture vs. equivalent vehicle passes for
Samples 1-3. (See Table 6-1 for sample identification.)
-------
PERCENT MOISTURE VS. VEHICLE PASSES
9
to
CsL
h-
co
o
111
O
to
h-
LJ
O
a
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
SAMPLE 4 + SAMPLE 5
60
Figure 6-3. Percent moisture vs. equivalent vehicle passes for Samples 4
and 5. (See Table 6-1 for sample identification.)
-------
PERCENT MOISTURE VS. VEHICLE PASSES
O)
_JL
CO
UJ
a.
13
CO
o
5
UJ
o
<
U.
a
CO
H
z
UJ
o
a:
UJ
a.
0
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
SAMPLE 8 + SAMPLE 9
60
Figure 6-4. Percent moisture vs. equivalent vehicle passes for Samples 8
and 9. (See Table 6-1 for sample identification.)
-------
PERCENT SILT VS. VEHICLE PASSES
h-
LJ
H
Z
O
o
K-
00
H
Ld
O
tt
LJ
Q.
D SAMPLE 1
T
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
+ SAMPLE ID
o
r
60
SAMPLE 2
Figure 6-5. Percent silt (% < 75 /imP) content vs. equivalent vehicle passes for
Samples 1, 1D, and 2. (See Table 6-1 for sample identification.)
-------
PERCENT SILT VS. VEHICLE PASSES
S
on
LU
H
z
o
o
H
_J
CO
H
•z.
u
o
a
LJ
Q.
20 -
10 -
0
D
SAMPLE 3
n 1 r
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
+ SAMPLE 4
o
60
SAMPLE 5
Figure 6-6. Percent silt (% < 75 /jmP) content vs. equivalent vehicle passes for
Samples 3, 4. and 5. (See Table 6-1 for samole identification.^
-------
PERCENT SILT VS. VEHICLE PASSES
2
05
LJ
h-
z
O
O
h-
h-
LU
O
a
LJ
a.
10
0 -i
n r
20 40
(Thousands)
EQUIVALENT VEHICLE PASSES
SAMPLE 8 + SAMPLE 9
60
Figure 6-7. Percent silt (%•< 75 umP) content vs. equivalent vehicle passes for
Samples 8 and 9. (See Table 6-1 for sample identification.)
-------
compared to that recovered by sampling both during and after testing. The results of the
mass balances performed are shown in Table 6-5.
As shown in Table 6-5, a certain portion of the original sample mass was lost to
the cold room environment for all abrasives tested except for Samples 8 and 9 (Test
Nos. 7 and 8). In these tests, more material was recovered from the track than was
originally applied. This apparent increase in sample mass was probably due to the
production of paving wear debris during testing. (Note that increased paving wear was
presumably caused by resurfacing of the test track prior to Test No. 7—see Section 6.2.1
above.) Since the contribution of track wear to the overall silt content cannot be
distinguished from that of the abrasive itself, the results of these two experiments are
obviously invalid and thus were not used in the data analysis.
6.2.4.2 Deicer Test Results-
The particle size data for Test No. 6 of rock salt are shown in Table 6-6 as a
function of vehicle passes. These data represent the particle size distribution of the entire
sample collected from the test track as determined by sonic sieving.
After sonic sieving, each track sample was recombined and chemically analyzed
for total dissolved solids, Na+, and CI" content. The results of these analyses are shown
in Table 6-7. Also shown in Table 6-7 is the percent NaCI in each sample as calculated
from the analytical data. As seen from these results, only a relatively small portion (i.e.,
approximately 30% or less) of each sample was composed of NaCI with the remainder
consisting of insoluble material (e.g., tire and/or track debris).
6.3 SMALL SCALE EXPERIMENTS
As an adjunct to the simulated traffic tests, small scale experiments were also
conducted in MRI's cold room to: (1) investigate suitable sampling methods for use in
the simulated traffic tests; and (2) determine the amount of solid material potentially
available for resuspension resulting from the application of common deicers. The
following describes the apparatus, protocol, and results associated with the small scale
experiments.
6.3.1 Description of Apparatus
The test apparatus used in the small scale experiments consisted of a series of 6
x 12 x 3 in solid blocks constructed of Portland cement concrete. Each block contained
a depressed area (or "well") having approximate internal dimensions of 4.5 in wide x
10.5 in long x 0.75 in deep. The well provided an area of 0.328 ft2 for the growth of ice.
Up to three blocks were used during each test.
In addition to the blocks, heat lamps were also used to melt any remaining ice and
evaporate the water. A modified ink roller was also provided to crush and distribute the
deicer particles over the internal block surface and to simulate vehicles moving over a
6-17
-------
CD
oo
TABLE 6-5. TEST TRACK MATERIAL BALANCE
Test Sample
No. No.
1
1D
2
3
4
5
6
7
1
1
2
3
4
5 i
4
6 I
8 J
8
3rial identification
y Sand & Gravel
itrol sand
y Sand & Gravel
itrol sand
DOT sand—
le Isle
boat Springs
bia Falls— new
do APCD washed
je
)T road salt
3oat Springs
Total material applied to
test track (g)a
816
816
816
816
816
816
102
774C
Total material recovered
from test track (g)
521.5
716.2
471.3
613.6
707.5
717.1
23.7
1214.7
g
294.5
99.8
344.7
202.4
108.5
98.9
78.3
<440.7>
Percent
36.1
12.2
42.2
24.8
13.3
12.1
76.8
<157>
scoria
squeege
coarse sand
Sheridan, Wyoming,
scoria
774°
989.6
<215.6> <128>
Equivalent to 6 applications of a 90:10 mixture of abrasive and rock salt (NaCI) at 4 000 Ib/lane mi
^ Qr\r\lir»atirtr» *-\f rr\r+ls o*-»lf /Kl«^l\ «* ff\r\ IU./IA~, • .*•__•'.* . .. ..
• • — - — —-.— . _v. . ^r vtl !•_• I X»TV^I\ V^«^ll. t I XC4V/II C»l ^f\J\J\J IL.
application of rock salt (NaCI) at 500 Ib/lane mi + 5 applications of 250 Ib/lane mi each.
Losses to cold room environment regardless of fate; < > indicate gain in material mass.
c After rebuilding of test track and slight change in track area.
-------
TABLE 6-6. PARTICLE SIZE DATA FOR TRAFFIC TEST NO. 6 (ROCK SALT)
Equivalent
vehicle passes
15,000
25,000
35,000
45,000
55,000
65,000
Post-test
Sample weight
(g)a
0.4123
0.3915
0.1865
0.2033
0.2854
0.3771
21 .8b
< 75 ^n
51.1
47.6
49.8
42.1
44.4
44.2
48.2
Weight percent < stated
iP < 53 \irnP
37.7
33.5
' 35.3
28.6
30.2
26.9
NAb
size"
< 20 jimP
4.36
0.778
1.63
1.33
• 2.28
1.98
NAa
* Total mass recovered from test track.
b NA = not available.
TABLE 6-7. RESULTS OF CHEMICAL ANALYSES FOR NaCI TESTS
Sample
identification
1 5,000 passes
25,000 passes
35,000 passes
45,000 passes
55,000 passes
65,000 passes
Post-test
sweepings
Total dissolved
solids (wt %)'
30.0
25.3
48.0
26.3
28.6
34.1
39.6
Na+ content
(wt %)b
55.6
57.4
29.2
58.2
59.3
56.9
58.8
Cr content Overall NaCI
(wt %)b content in sample0
34.4
35.9
16.8
32.9
35.9
35.5
16.5
27.0
23.6
22.1
24.0
27.3
31.6
29.8
Percent water-soluble fraction in as-received sample.
Percent Na+ and CI" in water-soluble fraction of sample.
c Values calculated from data in other columns.
6-19
-------
paved road. Finally, large plastic syringes fitted with large bore needles were used to
remove wash water from the test blocks.
6.3.2 Test Protocol
In preparation for testing, a layer of ice was grown in the well of the test blocks.
This was accomplished by first cleaning each block and placing it in a freezer
compartment overnight. A container of deionized water was also chilled in a refrigerator
set at 40°F. After equilibration, 100 cm3 of the chilled water was added to each block and
allowed to freeze overnight. Also during this period, the 37-mm, Millipore Type AA filters
for the Mini Vac® were equilibrated at constant temperature and humidity (i.e., 23° ± 1°C
and 45% ± 5% relative humidity) and tare weighed in preparation for testing.
To conduct each test, the blocks were removed from the freezer, placed in the
cold room, and equilibrated to a temperature of 35°F. Next, 4.11 g of deicer was
distributed evenly across the surface of each block. (Note that this amount of deicer is
equivalent to an initial application of 500 Ib/lane mi plus five reapplications of 250 Ib/lane
mi per AASHTO recommendations for application of deicing salts.) The modified ink roller
was then used to crush the deicer particles and spread the deicer/brine mixture equally
across the test surface. Any deicer/brine remaining on the rojler was washed onto the
block with deionized water. The block was dried by heat lamps overnight.
Dry residue on the blocks was quantified by dry surface sampling followed by wet
flushing. Dry sampling was used to estimate the amount and size of fine material
potentially available for resuspension. Wet flushing was conducted to determine the
amount of deicer potentially removed by post-storm runoff.
For dry sample recovery, a coarse brush was used to loosen the particles
adhering to the block surface. The solid material was then removed by a combination
of brushing and dry vacuuming using the Mini Vac® described previously. The quantity
and particle size distribution of the samples collected were determined by gravimetric
analysis followed by sonic sieving of the entire sample (see Section 6.2.3 above).
After removal of the dry residue from the test blocks, a series of three hot water
washes were performed. This was accomplished by first heating a 1-L beaker of
deionized water to a boil. The water was allowed to cool slightly and transferred to a
plastic squirt bottle. Each block was then flushed with approximately 360 cm3 of hot
water. Brine was removed from the block by a large hypodermic syringe and
quantitatively transferred to tared glass beakers. The beakers were dried in a laboratory
oven (110°C or 230°F) overnight, cooled in a desiccator, and the residue determined by
gravimetric analysis.
6-20
-------
6.3.3 Test Results
The results of the small scale experiments are shown in Tables 6-8 and 6-9 for the
gravimetric and particle size analyses, respectively. Individual data are shown in these
tables along with an average for each test performed.
As can be seen from Tables 6-8 and 6-9, CMA has the highest potential for dust
resuspension of the three deicing compounds tested. Approximately 35% of the total
deicer applied ends up as solid residue on the test surface with over 50% of this material
consisting of silt size particles. Thus, CMA appears to have the highest potential for PM10
emissions as compared to the other deicers tested.
Table 6-8 is also very informative with respect to the potential for post-storm runoff.
As shown, rock salt has the highest wet runoff potential (i.e., average of 30.2%) of the
three deicers tested with CMA being a close second. CaCI2 was shown to have little or
no potential for post-storm runoff. These results could have substantial implications in
the design of future snow and ice control programs which may rely on pretreatment to
prevent rather than mitigate the loss of traction on paved roadways.
6.4 DATA ANALYSIS
The following subsections provide the re'sults of MRI's analysis of the test data
derived from the experimental program. Each test type is discussed separately.
6.4.1 Abrasive Tests
One of the objectives of this study was to determine whether the apparent
degradation (i.e., grinding) produced by simulated traffic could be quantitatively related
to material properties (e.g., LA. abrasion loss, quartz content, etc.). In order to perform
this analysis, it first was necessary to derive an indicator of the change in particle size
distribution between the initial and post test states. There are a number of different
formulations that could be used. One simple example, would be the ratio of silt content
after a given number of passes to the initial material silt content. This study adopts a
somewhat broader indicator—the change in material mass median diameter (MMD)
between the initial state and final state (i.e., after 65,000 simulated vehicle passes). The
principal advantage of using the mass median diameter over a single size "cut" (e.g., silt
content) is that by definition the MMD considers the entire size distribution. In this sense
the MMD should better reflect the overall impact/degradation effects of simulated traffic
on the various test materials.
Table 6-10 presents both the MMD data and the change in MMD, expressed as
a ratio, for each of the five valid track tests. The MMD data were developed by fitting the
mechanical sieving results to a presumed lognormal distribution. (Note that the data
presented do not include Tests 7 and 8 as indicated in Section 6.2.4.1—these tests
6-21
-------
TABLE 6-8. RESULTS OF SMALL SCALE EXPERIMENTS
18
Delcer tested
Rock saltc
(NaCI)
CaMg Acetate"
Calcium
chloride"
Block ID
A
B
C
Average
1
2
Average
A
B
C
Average
Original
amount
applied
(9)
4.11
4.11
4.11
4.11
4.11
4.11
4.11
4.11
4.11
4.11
4.11
Amount of sample recovered bv method6
Application
rate
(g/m2)'
135
135
135
135
135
135
135
135
135
135
135
Dry recovery
grams
0.3786
0.4600
0.2501
0.3629
1.408
1.502
1.455
0.0324
0.0335
0.0250
0.0303
weight %
9.2
11
6.1
8.8
34
37
35
0.79
0.81
0.61
0.74
Wash No. 1
grams
0.9000
0.8630
0.851 1
0.8714
0.8206
1.004
0.9122
0.2062
0.2895
0.2208
0.2388
weight %
22
21
• 21
21
20
24
22
5.0
7.0
5.3
5.8
Wash
> grams
0.2818
0.2133
0.2508
0.2486
0.1117
0.1341
0.1229
0.0996
0.1398
0.1540
0.1311
No. 2
weight %
6.9
5.2
6.1
6.1
2.7
3.3
3.0
2.4
3.4
3.8
3.2
Wash No. 3
grams
0.1321
0.0940
0.1032
0.1098
0.0421
0.0330
0.0376
0.0835
0.1018
0.1038
0.0964
weight %
3.2
2.3
2.5
2.7
1.0
0.80
0.91
2.0
2.5
2.5
2.3
Total %
recovered
41
40
35
39
58
65
62
10
14
12
12
Equivalent to an initial application of 500 Ibs/lane mile plus five (5) reapplications of 250 Ibs/lane mile according to AASHTO recommendations for deiclna
salts. a
Dry recovery was performed by mechanical removal of the surface material followed by dry vacuuming. Wet recovery was conducted by flushing the
surface with hot delonlzed water (see text for details). Wt. % = percent of original sample recovered from surface.
c Block A = mostly large salt particles; Block B = mixture of large and small particles; Block C = mostly small salt particles.
Homogeneous round pellets. No melting of ice was observed after application of CMA; Ice melted by combination of warm water and heat lamps.
8 Homogeneous round pellets of Peladow*.
-------
TABLE 6-9. PARTICLE SIZE DATA FOR SMALL SCALE EXPERIMENTS
(SONIC SIEVING RESULTS)
Sample
Deicer tested ID
Rock salt (NaCI) Block A
Block B
Block C
Average
CaMg Acetate Block 1
Block 2
Average
Calcium chloride (CaCyd Block A
Block B
Block C
Average
Total
dry
loading
(g/my
12
15
8.2
12
46
49
48
1.1
1.1
0.82
0.99
Percent
< 75 (imP <
71
62
58
63
55
56
56
NA
NA
NA
—
< stated
53 jimP
51
42
44
46
39 •
40
39
NA
NA
NA
—
size"
< 20 umP
6.8
4.9
29
13
6.5
3.1
4.8
NA
NA
NA
—
' Calculated from amount of material recovered from surface by mechanical removal
followed by dry vacuuming. Total initial loading applied to surface = 135 g/nf (4.11 g of
deicer).
b (imP = micrometers physical particle diameter. NA = not available (insufficient sample
for analysis).
c Peladow* pellets.
6-23
-------
TABLE 6-10. MASS MEDIAN DIAMETER (MmP) OF ABRASIVE SAMPLES
TESTED ON WHEEL PASSAGE MACHINE
Test
No.
1
2
3
4
5
Sample
identification
Holiday Sand & Gravel
Ice control sand
Maine DOT Presque Isle
Stockpile
Steamboat Springs
Scoria
Columbia Falls
New
Colorado APCD
Washed squeege
Initial
MMD
900
2,400
1,500
7,000
4,800
Post-test
MMD
190
10
37
350
95
Change
(initial/post-test)
4.74
240
40.5
20.0
50.5
showed as a net gain in the material mass recovered from the test track, apparently the
result of track resurfacing.)
To determine whether quantitative relationships exist between change in MMD and
material properties, both linear correlation and simple regression were used Table 6-11
shows the correlations between MMD change and material property. Note that all data
were transformed using natural logarithms, prior to computing the correlations.
TABLE 6-11. LINEAR CORRELATION BETWEEN
MMD CHANGE AND TEST MATERIAL PROPERTIES
Material property
Correlation with MMD change
LA abrasion loss
Wet aggregate durability
Quartz content
Vickers hardness
Lnit weight
Void fraction
Particle shape index
0.866
-0.298
0.084
-0.696
-0.479
0.109
0.555
6-24
-------
Based on the above correlations, it is evident that degradation produced by
simulated traffic (i.e., MMD change) is best related to abrasion loss. Figure 6-8 presents
this correlation graphically. Alternatively, the relationship can be presented in the
following regression format.
A MMD = 5.46 • (Abrasion Loss)1'35 (6-1)
This relationship accounts for 75% of the variation (i.e., R2 = 0.75) in MMD change
over the five valid test runs.
6.4.2 Deicer Tests
The results of the single deicer test conducted on the wheel passage machine
were analyzed to determine if any discernable trends could be derived from the data.
This analysis yielded no conclusions which could be considered physically meaningful.
Therefore, these results were combined with other limited data from the small block
experiments and analyzed further.
To determine the amount of deicer potentially available for resuspension as PM10
on paved roadways, a mass balance was performed using data from tests of MO DOT
_ rock salt. In this analysis the total amount of sample recovered from the test surface (i.e.,
track or test block) was compared to that originally applied. The NaCI and silt content
of the samples were then combined to estimate the percent of the total NaCI in the silt
size range. (Note that the remaining material recovered from the wheel passage machine
was probably composed of tire wear, track debris, and other insoluble material generated
during testing.) The results of this analysis are shown in Table 6-12.
As noted from Table 6-12, the mass balance for the wheel passage tests and small
block tests compare favorably with one another. These results indicate that between 3%
and 6% of the total NaCI applied (i.e., 60 to 120 Ib/ton) ends up as silt size particles and
thus available for resuspension as PM10. (Note that the NaCI applied = total deicer
applied less the percent insoluble matter in the deicer.) Comparing these values to the
percent insoluble matter in the original deicer (i.e., average of 0.77%) would indicate that
substantially more silt (i.e., a factor of > 5) is potentially created by the application of rock
salt than could be estimated from material properties alone.
Using the results of the wheel passage test shown in Table 6-12, the PM10
emissions associated with the application of rock salt were estimated using the current
AP-42 emission factor (Equation 1-1). This estimate was compared to similar emissions
determined by the calculation scheme developed during the "gap filling" exercise of
Grelinger et al. (1988). Here the following assumptions were used: a 1-mi stretch of
4-lane road; a lane-width of 12 ft; an average daily traffic (ADT) of 40,000 (i.e., 10,000
vehicles/lane-day); and one application of rock salt at a rate of 500 Ib/lane-mile plus'five
reapplications at 250 Ib/lane-mile performed within a 24-hour period.
6-25
-------
O5
31
(Q*
i
9
oo
0:0
to o
II
sa
si'
£§•
p)
05 3
CTCQ
J3 ®
W -•
O 3
3 3
II
0)
CL
(O
T3
03
g.
-------
TABLE 6-12. MASS BALANCE FOR ROCK SALT (NaCI) TESTS
Average
Total NaCI Total dry sample Total NaCI silt content of
applied to recovered from in sample sample
Test test surface test surface recovered recovered
configuration (gm)a
O)
ro
•M
Wheel passage
machine
Small block
tests
a Weight of deicer
101
4.08
less the
Grams
23.7
0.363
Weight % Grams Weight %b (Weight %)
23.3 6.99 6.89 48. 1d
8.89 0.363° 8.89 63.5f
Total NaCI on
test surface
in silt size ranae0
Grams Weight %
3.36 3.31
0.230 5.65
percent insoluble matter.
b Weight % = Weight percent of total NaCI applied to test surface.
c Material considered available for resuspension as PM10 using AP-42 predictive equation.
d Weighted average calculated from data in Table 6-6.
9 All material recovered assumed to be NaCI.
' Arithmetic average shown in Table 6-9. Three significant figures.
-------
The above assumptions, together with Equation (1-1) and Table 6-12 result in an
emission rate of 470 Ib/lane-day. This should be compared to approximately
8.8 Ib/lane-day for similar calculations performed using the "gap filling" document. The
great disparity (i.e., a factor of 52) between the two estimates stresses the need for an
emission factor based on actual field testing.
Another factor of note pertains to the contribution of paving and tire wear to the
increase in silt content due to application of antiskid deicers. These materials normally
contribute to the total silt loading on paved roads to some extent. However, what is not
currently known is how wear debris change as a result of deicer application.' Therefore
a total increase in silt loading of 6% to 11% (i.e., % total sample recovered x silt content)
might be expected to more appropriately account for additional pavement/tire debris
generated by the use of deicers on paved roads.
6-28
-------
SECTION 7
CONCLUSIONS AND RECOMMENDATIONS
A number of conclusions and recommendations were developed in the preparation
of this guidance document. These are presented in the following subsections.
7.1 CONCLUSIONS
The following conclusions were reached as part of Phase 1:
1. The application of antiskid materials causes a temporary, but substantial
increase in surface silt loading on paved roads.
2. The level of PM10 emissions generated by traffic resuspension of (dry) roadway
surface dust is directly dependent on the surface silt loading.
3. The tendency for an antiskid abrasive to generate silt-sized particles is a
function of its durability.
4. Particles smaller than 50 mesh (approximately 300 Aim in physical diameter)
have been found to be relatively ineffective in increasing the coefficient of
friction on paved roads.
5. Of all the durability test methods surveyed, the Los Angeles abrasion test
seems to be the most appropriate for the measurement of overall aggregate
durability.
6. A good quality, washed construction aggregate (e.g., sand) appears to be the
best choice with respect to maximum effectiveness as an antiskid abrasive with
low silt generation potential.
7. Salt is effective as an antiskid material above about 20°F with minimal cleanup
requirements but with significant (albeit poorly known) potential for PM10
emissions as well as other adverse environmental effects such as corrosivity
and ecological stress.
7-1
-------
8. Antiskid materials are frequently applied at loadings well above recommended
levels because of public perception that effectiveness is proportional to the
visible amount of surface loading.
9. Excess silt loadings (and thus PM10 emissions) associated with antiskid
materials result primarily from over application and noncompliance with
recommended fines and durability specifications for antiskid abrasives.
The following conclusions were reached as part of Phase 2 of the study:
1. Antiskid abrasives typically used in PM10 nonattainment areas can be
statistically grouped into five basic categories (or clusters) based on the eight
key material properties measured in the program as follows:
C1—Materials exhibiting relatively low initial silt (0.99% to 4.9%) and
quartz (10 to 36 mmol/L) contents; other material properties near the
overall means.
C2—Coarse and soft materials as indicated by high particle shape
(21 to 27) and void fraction (45% to 46%) indices along with relatively
low unit weight (860 to 1200 kg/m3) and Vickers hardness (340 to 460
kg/mm2) values.
C3—Materials that exhibit both high quartz (150 to 190 mmol/L) and
high silt (4.1% to 4.9%) contents; other material properties near the
overall means.
C4—Relatively hard materials as indicated by low silt content (0.02% to
0.25%), low abrasion loss (0.90% to 4.0%), and high values for the
Vickers hardness test (550 to 1200 kg/mm2).
C5—Finely divided materials that are easily abraded; indicated by high
silt content (4.3% to 8.9%) and high abrasion losses (7.1% to 17%).
2. Based on the limited data collected in the simulated traffic tests, materials in
cluster C4 tended to have the lowest overall potential for the generation of
PM10 emissions as compared to other groups. Alternately, the material from
cluster C5 showed the greatest potential for PM10 emissions in these tests as
predicted by the current AP-42 emission factor equation.
3. In the small scale experiments of three chemical deicers, calcium magnesium
acetate (CMA) had higher PM10 emissions potential than rock salt and calcium
chloride (i.e., 27 g/m2 sL for CMA vs. 7.6 g/m2 and < 1 g/m2 for NaCI and
CaCI2, respectively). On the other hand, rock salt had the highest post-storm
7-2
-------
(wet) runoff potential (i.e., 30% wet recovery for NaCI vs. 26% and 11 % for the
other deicers) in these tests.
4. The best overall indicator of silt generation potential found in the five valid
traffic tests of abrasives appears to be Los Angeles abrasion loss. However,
Vickers hardness and particle shape index may also be useful in predicting
overall material behavior for the abrasives tested.
5. Increases in surface silt loading for rock salt were found to be a factor of > 5
higher than could be predicted from the insoluble matter in the material alone.
6. PM10 emissions calculated using the traffic test results for rock salt were a
factor 50 higher than previously estimated. This underestimation could affect
the preparation and/or implementation of PM10 SIPs for certain state and local
jurisdictions.
7. Contributions to silt loading from enhanced tire and pavement wear could not'
be quantified from existing data.
7.2 FINAL MATERIAL SELECTION CRITERIA
It is difficult to formulate definitive quantitative selection criteria for antiskid materials
from available data. Still, one can provide general ranges of key parameters for material
groups (clusters) which exhibit high or low increases in silt content in service, thus
implying high or low durability. In this sense, materials in these ranges could be classified
as "acceptable" and "nonacceptable." Currently only antiskid abrasives have a data
base adequate to provide selection criteria based on material properties. These are
shown in Table 7-1.
TABLE 7-1. REVISED SELECTION CRfTERIA FOR AhfTISKlD ABRASIVES
Acceptable materials* Unacceptable materials'5
Measurement parameter Units Range of values Mean Range of values Mean
Modified Los Angeles abrasion
loss
Initial silt content0
Vickers hardness
Particle shape index
Weight %
Weight %
kg/mm
Dimensionless
0.9-4
0.02-0.3
500-1,200
6.3-15
3
0.1
1,000
10
7-17
4-9
. 400-1,000
6.5-13
11
6
800
9
a Based on data for cluster C4.
Based on data for cluster C5.
0 This parameter is coupled to LA abrasion loss and thus included in the material selection criteria.
7-3
-------
w , u J"1 tS the upPer and lower bounds of LA abrasion loss, initial silt content
Vickers hardness, and particle shape index for acceptable and nonacceptable materials'
respectively. If the properties of a particular material fall between the ranges indicated'
the acceptability of the material must be considered as "questionable - and good
engineering judgment should be employed before the material is used for ice and snow
control in a PM10 nonattainment area. Other qualitative factors discussed in Section 4 of
this report can also be incorporated into the decision-making process.
To further assist in the decision-making process, a decision "tree" has been
developed for material selection as shown in Figure 7-1. As a starting point in this
analysis, the service conditions must first be defined. These conditions take into account
road surface characteristics (e.g., slope, pavement composition, presence of
curbs/shoulders), traffic parameters (e.g., volume, route speed, acceleration/deceleration)
and meteorological conditions surrounding snow/ice events (e.g., temperature'
evaporation rates). The service conditions will also vary from location to location within
a particular geographical area.
From that point, candidate materials can be evaluated based on silt generation
potential, cost, availability, etc., and the most promising material(s) selected for field
study. The above analysis of the acceptability of candidate materials is highly dependent
on local conditions and should be performed on a case-by-case basis The criteria
shown in Table 7-1 and the decision tree shown in Figure 7-1 should assist in the overall
selection process.
7.3 RECOMMENDATIONS
Based on study results, the following recommendations for further investigation were
1. A revised Los Angeles abrasion test method should be developed to improve
precision and better determine the silt generation potential of antiskid
abrasives. Applicable modifications to this method are discussed in
Section 3.5.1.
2. The current method for field measurement of silt loading should be subjected
to collaborative testing under wintertime snow/ice control conditions to
determine existing method reproducibility and any needed modifications to
improve reproducibility.
3. Additional study of surface loading dynamics following the application of
antiskid materials is needed in order to better understand the PM10 emission
impacts associated with various antiskid program scenarios. This could be
accomplished by a well designed and implemented surface sampling program
performed before, during, and after individual storm events at selected
locations in PM10 nonattainment areas.
7-4
-------
IDENTIFY
SERVICE
CONDITIONS
Level of Antiskid
Control Required
Traffic Volume,
Lanes, Speed
Temperature and
Environmental
Conditions
Use AASHTO,
Federal, and State
Guidelines
Define Candidate
Materials for
Consideration
FOR EACH CANDIDATE MATERIAL
Define Material
Specifications to
Reduce Silt Loadings
Define Application Levels
and Procedures to Reduce Silt Loadings
• Particle Size
Particle Morphology
• Hardness
• Durability
% Insoluable Matter
• Others
Define Clean-up and
Other Mitigation Procedures
for Reducing Silt Loading
• Application Rate of Abrasives
• Application Rate of Salts
Equipment Calibration/Maintenance
• Development of Protocol and
Documentation for
Equipment Operators
• Monitoring of Compliance
w/Specifications
Clean-up Frequency and Time After Storm
• Type of Equipment Used
• Monitoring of Clean-up Effectiveness
Select Material for
Field Study and Evaluation
69-21 SEV klnsgr 11/3/89
Figure 7-1. Decision "tree" for antiskid material selection.
-------
4. Further study of the fate and transport of salt(s) applied for antiskid control is
needed to determine an applicable emission factor based on field testing A
mass balance approach using a combination of surface and air sampling
should be used in this study to quantify the PM10 emissions associated with
deicmg salts.
5. Common post-storm cleanup practices should be studied to determine the net
air quality benefits of cleanup at various points after application as a function
of antiskid material specifications.
6. The contribution of tire and pavement wear to surface silt loading (sL) should
be quantified for the application of antiskid abrasives and deicers. If increases
in sL are observed, methods for prevention/mitigation should be investigated
7-6
-------
SECTION 8
REFERENCES
American Association of State Highway and Transportation Officials (1964). Standard
Method of Test for Production of Plastic Fines in Aggregates. Method T-210-64, in
Standard Specifications for Transportation Materials and Methods of Sampling and
Testing, Part II: Methods of Sampling and Testing, Fourteenth Edition, Washington, D.C.,
August 1986.
American Association of State Highway and Transportation Officials (1976). AASHTO
Maintenance Manual. 1st Edition, Washington, D.C., February.
American Association of State Highway and Transportation Officials (1983). Standard
Method of Test for Resistance of Abrasion of Small Size Coarse Aggregate by Use of the
Los Angeles Machine. Method T 96-83, in Standard Specifications for Transportation
Materials and Methods of Sampling and Testing, Part II: Methods of Sampling and
Testing, Fourteenth Edition, Washington, D.C., August 1986.
American Association of State Highway and Transportation Officials (1984). Standard
Method of Test for Sieve Analysis of Fine and Coarse Aggregates. Method T 27-84, in
Standard Specifications for Transportation Materials and Methods of Sampling and
Testing, Part II: Methods of Sampling and Testing, Fourteenth Edition, Washington, D.C.,
August 1986.
American Association of State Highway and Transportation Officials (1985a). Standard
Method of Amount of Material Finer than 75 /iim Sieve in Aggregate. Method T 11-85, in
Standard Specifications for Transportation Materials and Methods of Sampling and
Testing, Part II: Methods of Sampling and Testing, Fourteenth Edition, Washington, D.C.,
August 1986.
American Association of State Highway and Transportation Officials (1985b). Standard
Test Method for Soundness of Aggregates by Use of Sodium Sulfate or Magnesium
Sulfate. Method T-104, in Standard Specifications for Transportation Materials and
Methods of Sampling and Testing, Part II: Methods of Sampling and Testing, Fourteenth
Edition, Washington, D.C., August 1986.
8-1
-------
American Society of Testing and Materials, Subcommittee C09.02.06 (1969). Standard
Descriptive Nomenclature of Constituents of Natural Mineral Aggregates
Method C 294-69, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.03.05 (1983). Standard
Test Method for Soundness of Aggregates by Use of Sodium Sulfate or Magnesium
Sulfate. Method C 88-83, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.03.05 (1984a). Standard
Method for Sieve Analysis of Fine and Coarse Aggregates. Method C 136-84a
Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee D04.31 (1984b). Standard
Specification for Sodium Chloride. Method D 632-84, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.02.06 (1985a). Standard
Practice for Petrographic Examination of Aggregates for Concrete. Method C 295-85
Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.03.05 (1985b). Standard
Specification for Concrete Aggregates. Method C 33-85, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee D04.51 (1985c). Standard Test
Method for Aggregate Durability. Method D 3744-85, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Committee E-15 (1986). Standard Test
Method for Chemical Analysis of Sodium Chloride. Method E 534-86, Philadelphia,
Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.03.05 (1987). Standard
Test Method for Materials Finer than 75 /xm (No. 200) Sieve in Mineral Aggregates by
Washing. Method C 117-87, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Subcommittee C09.03.05 (1989). Standard
Test Method for Resistance to Degradation of Small-Size Coarse Aggregate by Abrasion
and Impact in the Los Angeles Machine. Method C 131-89, Philadelphia, Pennsylvania.
American Society of Testing and Materials, Committee E-4 (1989). Standard Test Method
for Microhardness of Materials. Method E 384-89, Philadelphia, Pennsylvania.
Anonymous (1979). Aggregates for Road Pavements. Hwys. Pub. Wks., 47:1835,16-19,
October.
Anonymous (1988). "How Highway Departments Deal with Nature's Forces," Better
Roads, 58:1, 19-21, January.
8-2
-------
Baboian, R. (1988). The Automotive Environment. Automotive Corrosion by Deicing
Salts, National Association of Corrosion Engineers, Houston, Texas.
Blackburn, R. R., et al. 1978. "Physical Alternatives to Chemicals for Highway Deicing,"
U.S. Department of Transportation, Federal Highway Administration, Washington, D.C.
Brant, L A. (1972). Winter Sanding Operations and Air Pollution. Public Works, 103:9.
94-7, September.
Brown, M. G. (1988). Corrosion of Highway Appurtenances Due to Deicing Salts.
Automotive Corrosion Due to Deicing Salts, National Association of Corrosion Engineers,
Houston, Texas.
Chollar, B. H. (1984). Federal Highway Administration Research on Calcium Magnesium
Acetate—An Alternative Deicer. Public Roads, 47:4, 113-118.
Connor, P. E., and R. Gaffi (1982). Optimum Sand Specifications for Roadway Ice
Control. State of Alaska, Department of Transportation and Public Facilities, Fairbanks,
Alaska, June.
Cowherd, Chatten, et al. (1977). Quantification of Dust Entrainment from Paved
Roadways. EPA-450/3-77-027, U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, July.
Dhir, R. K., et al. (1971). Study of the Aggregate Impact and Crushing Value Tests. Hwy
Engr., 18:11, 17-27, November.
Dubberke, W., and V. J. Marks (1985). The Effect of Deicing Salt on Aggregate Durability.
Transportation Research Record, 1031:27-34.
Dunn, S. A., and R. U. Schenk (1980). Alternatives to Sodium Chloride for Highway
Deicing. Transportation Research Record 776, 12-15.
Eck, R. W., et al. (1986). Natural Brine as an Additive to Abrasive Materials and Deicing
Salts. WVDOH Research Project 75, West Virginia Department of Highways, May.
Furbush, M. A. (1972). Relationship of Skid Resistance to Petrography of Aggregates.
PB-220 071, Federal Highway Administration, Washington, D.C., July.
Goswami, S. C. (1984). Influence of Geological Factors on Soundness and Abrasion
Resistance of Road Surface Aggregates: A Case Study. Bull. Inter. Assn Eng Geol
30, 59-61, December.
8-3
-------
Grelinger, M. A., et al. (1988). Gap Filling PM10 Emission Factors for Selected Open Area
Dust Sources. Final Report, EPA Contract No. 68-02-4395, Work Assignment 6, Midwest
Research Institute, Kansas City, Missouri, February 9.
Havens, J. H., and D. C. Newberry (1982). Aggregate Shape and Skid Resistance.
UKTRP-82-1, Kentucky Transportation Cabinet, Frankfort, Kentucky, January.
Hegmon, R. R., and W. E. Meyer (1968). The Effectiveness of Antiskid Materials. Highw.
Res. Rec. 227, Highway Research Board, NAS-NRC, 50-56.
Helmers, G., and U. Ytterbom (1986). Effects of Salt on the State of Dirtiness of the
Road. NTIS No. PB86-201167, Statens Vaeg-och Trafikinstitut, Sweden.
Hudec, P. P. (1984). Quantitative Petrographic Analysis of Aggregate. Bull. Inter. Assn.
Eng. GeoL, 29, 381-385, June.
Iowa Department of Transportation (1980). Deicing Practices in Iowa: An Overview of
Social, Economic, and Environmental Implications. Report No. 17-T68PP/9:D368,
Planning and Research Division, Office of Project Planning Highway Division, Office of
Maintenance, for the Iowa General Assembly House of Representatives, Des Moines,
Iowa, January.
Jorgensen, R., et al. (1964). Nonchemical Methods of Snow and Ice Control on Highway
Structures. Report No. 4, National Cooperative Highway Research Program, Highway
Research Board, Washington, D.C.
Kaufmann, D. W. (1968). Sodium Chloride—The Production and Properties of Salt and
Brine, American Chemical Society, Hafner Publishing Company, New York.
Keyser, J. H. (1973). Deicing Chemicals and Abrasives: State of the Art. Highw. Res.
Rec. 425, Highway Research Board, 36-51.
Kinsey, J. S., and R. M. Coveney, Jr. (1986). Mineral Characterization of Selected Soil
Samples. Final Report, MRI Project 8466-L to New Mexico State University, Las Cruces,
New Mexico, January 14.
Legislative Research Council (1965). The Use and Effects of Highway Deicing Salts,
Commonwealth of Massachusetts, Boston, Massachusetts, January.
McDonald, R. D. (1981). Automotive Underbody Corrosion Testing.
McElroy, A. D., et al. (1988). Comparative Evaluation of Calcium Magnesium Acetate and
Rock Salt, Transportation Research Record 1157, 12-19.
8-4
-------
Minsk, D. L (1981). Snow Removal Equipment, in Handbook of Snow-Principals,
Processes, Management, and Use. Pergamon Press, New York.
Murray, Donald M., and U. F. W. Ernst (1976). An Economic Analysis of the Envi-
ronmental Impact of Highway Deicing. EPA-600/2-76-105, U.S. Environmental Protection
Agency, Cincinnati, Ohio, May.
Norburg, C. H. (1981). The Washington Degradation Test Maine Variation. Technical
Paper 81 -9, Materials Research Division, Maine Department of Transportation, Augusta,
Maine.
Oeberg, G., et al. (1985). Experiments With Unsalted Roads. NTIS No. PB86-177409,
Statens Veag-och Trafikinstitut, Sweden.
PEDCo Environmental, Inc. (1981). Denver Demonstration Study. Colorado Division of
Air Pollution Control, Denver, Colorado, October.
Rainiero, J. M. (1988). Investigation of the Ice-Retardant Characteristics of Verglimit-
Modified Asphalt. Transportation Research Board, 1157. 44-53.
Salt Institute (1984). Survey-of Salt, Calcium Chloride, and Abrasive Use in the United
States and Canada. Alexandria, Virginia.
Salt Institute (1986). The Snowfighter's Handbook. Alexandria, Virginia.
Salt Institute (1989). Salt Institute Statistical Report Analysis, Total Sales by United States
Members. Alexandria, Virginia.
Schneider, T. R. (1959). Schneeverwehungen und Winterglatte. Interner Bericht Nr. 302.
Eidgenossisches Institut fur Schnee-und Lawinenforschung.
Schneider, T. R. (1960). The Calculation of the Amount of Salt Required to Melt Ice and
Snow on Highways. National Research Council of Canada, Technical Translation
No. 1004.
Sheeny, J. P., et al. (1968). Handbook of Air Pollution. 999-AP-44, U.S. Public Health
Service, Washington, D.C.
Slick, D. S. (1988). Effects of Calcium Magnesium Acetate on Pavements and Motor
Vehicles. Transportation Research Record 1157, 27-30.
Stiffler, A. K. (1969). Relation Between Wear and Physical Properties of Roadstone. Hwy.
Res. Board Special Reports, 101. 56-68.
8-5
-------
Transportation Research Board (1974 and 1984). Minimizing Deicing Chemical Use N24
Final Report, Washington, D.C.
U.S. Environmental Protection Agency (1972). A Search: New Technology for Pavement
Snow and Ice Control. EPA-R2-72-125, Office of Research and Monitoring US
Environmental Protection Agency, Washington, D.C., December.
U.S. Environmental Protection Agency (1985). Compilation of Air Pollution Emission
Factors, AP-42. U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina.
Welch, Bob H., et al. (1976). Economic Impact of Highway Snow and Ice Control—State
of the Art. Report No. FHWA-RD-77-20, Utah Department of Transportation Research and
Development Section, September.
Wood, F. (1983). Noncorrosive Winter Maintenance Workshop, Frank Wood of Salt
Institute. Method 10-26-83, in Proceedings of the Northstar Workshop on Noncorrosive
Winter Maintenance, FHWA/MN/RD-84/03, Federal Highway Administration, Washington
D.C., October. '
Wood, F. (1983). Proceedings of the Northstar Workshop on Noncorrosive Winter
Maintenance (Technical Presentations). Report No. FHWA/MN/RD-84/03, Minnesota
Department of Transportation, 56-67, October.
Woodside, A. R., and R. A. Peden (1983). Durability Characteristics of Roadstone
Quarry Mgt: Prod., 10:8, 493-494/497-498, August.
Wyoming Highway Department Maintenance Manual. (1984). Wyoming Department of
Transportation.
8-6
-------
APPENDIX A
SURVEY PROCEDURES
A-1
-------
The following sections outline the literature search and telephone survey performed
during the study.
A.1 LITERATURE SEARCH
The results of three literature searches were examined for references pertinent to
the current study. From a previous program, MRI had compiled a data base of abstracts
and citations on: ice disbonding; laboratory studies on ice; salt effects on soils
pavements, steel corrosion, and plants; and the properties and uses of deicing chemicals'
The abstracts had been compiled from the STN CA file, NTIS, and TRIS data bases. The
most recent references in the MRI data base were published in 1987.
Additional literature searches were done in August and September 1989 on the
hardness and tendency of antiskid materials to produce fines. The more extensive on-line
search in August focused on the TRIS data base, although a few retrievals were made
from the NTIS data base. The strategy combined the concept of the materials (key
words: antiskid or abrasives or sand or .stone or cinders or slag) used for
highway/road/street snow or ice removal (controlling or deicing or deicers or salt or melt)
with key words for the following concepts:
• Skid resistance (including coefficient of friction and traction).
• Application/spreading rates or costs.
• The dustiness concept (key words: attrition or fines or dusts or silt or
gradation or grade or sizing or sieve or sieving or screening or grain size or
granulometry or durability or stability or petrography or degradation).
• Cleaning or flushing the materials.
• Alternatives or substitutes.
The strategy, of course, allowed for appropriate truncation and proximity operators
for the key word relationships.
In September, the data bases Compendex Plus (Engineering Abstracts and
Engineering Meetings), TRIS, IMS International Standards and Specifications, and
Standards and Specifications were searched for reviews on aggregate hardness testing
procedures. The search combined the following three sets of key words with the key
word review:
1. Road/highway/street use of aggregate.
2. Test or standard or specification.
A-2
-------
3. Hardness or Mohs or Knoop or scleroscopy or crush resistance or wear
resistance or abrasion resistance.
From the above search 121 documents were identified, collected, and reviewed for
pertinent data. The documents collected are listed under references in Section 8 or in
the bibliography in Appendix B.
A.2 TELEPHONE SURVEY
Nine states and two municipalities were surveyed by telephone (some with multiple
contacts) to gather information on current ice and snow control practices, especially as
related to the impact of antiskid materials on dust production. The telephone survey was
based on the questionnaire found in Figure A-1. This six-page form permitted the
recording of comprehensive information on types of deicing chemicals and abrasives,
factors associated with selection of particular antiskid measures, application methods and
rates, and cleanup operations. Often the state or municipal official surveyed had partial
information, so only the appropriate sections of the questionnaire were completed.
Several states also submitted specifications for purchase of antiskid materials.
Early in the program it was determined that the South Dakota Department of
Transportation was accumulating information on abrasives. Over 20 states had
responded with information on specifications for particle size distributions and other
physical characteristics of abrasives. The questionnaire used in this effort appears in
Figure A-2.
The purpose of the South Dakota survey was to develop background information
for a new specification for antiskid aggregates based on reducing the number of broken
windshield complaints and achieving high skid resistance. The sand in the South Dakota
survey was classified into five categories: pit run sand, screened pit run sand, crushed
pit run aggregate, crushed quarry aggregate, and other. Also, some respondents
provided their particular material specifications as attachments. The results of the
telephone and South Dakota surveys are provided in Section 2.1 of the main report.
A-3
-------
TELEPHONE SURVEY QUESTIONNAIRE
Selection and Usage of Anti-Skid Materials for Ice and Snow Control
Agency Surveyed: Date:
Person Contacted: Tele. No. ( )
Mailing Address of Agency:
1. What types and amounts of deicing chemicals and abrasives have you used in
each of the last 3 years? How many lane-miles of road were treated by each type
of chemical and abrasive? What is the cost of these materials and the "nominal"
application rate (Ibs/lane-mile)?
UNIT COST
($/TON) OR NOMINAL
QUANTITY TOTAL APPLICATION
MATERIAL (TONS OR LANE-MILES ANNUAL RATE (LBS/
YEAR TYPE GALLONS) TREATED COST ($K) LANE-MILE)
Figure A-1. Telephone survey questionnaire.
A-4
-------
2. Are chemicals and abrasives mixed prior to application and, if so, in what
proportions? —
3. What factors (e.g., cost, availability, effectiveness, etc.) are used to select a
particular abrasive or chemical for use in ice and snow control?
a. Cost Considerations:
Abrasives: __^_ ——
Chemicals:
b. Availability Considerations: What are the sources of the materials currently
used?
Abrasives:
Chemicals:
c. Driving Improvement Effectiveness:
MATERIAL TYPE ROAD CONDITION EFFECTIVENESS
How dp you define effectiveness?.
Figure A-1 (Continued)
A-5
-------
d. Dust Generation Potential:
e. Other Considerations (e.g., durability; temperature; corrosivity; environmental
conditions; etc.):
4. What materials, other than those currently being used, have been considered for
use in ice and snow control?
Why were these rejected?.
5. Do you have specifications (e.g., gradation, durability, etc.) and standard test
methods for the abrasives and chemicals used for ice & snow control?
If so, what do they generally consist of? ____
6. Could you send a copy of the above specifications and test methods by first class
mail?
7. What is the basis for a particular road being treated (traffic volume, rural vs. urban
routes, school route, public demand, accident rates, etc.)?
8. Are abrasives/chemicals applied at different rates? If so, what is the
basis for selecting the application rate?
Figure A-1 (Continued)
A-6
-------
9. Have directives been published and distributed to operating personnel regarding
the current policy on the use of abrasives and chemicals for ice and snow control?
If so, could you send a copy by first class mail?
10. Do you require detailed reports by equipment operators on the quantity of
abrasives and chemicals used? If so, is this information correlated
with temperature, traffic, snowfall, humidity, or other factors effecting the quantities
used?
11. What type of distribution equipment is used to apply abrasives/chemicals to
different road types?
MATERIALS DISTRIBUTION
TYPE EQUIPMENT
a. State Highways (rural areas):
b. Expressways:
c. City Streets:
d. Major Intersections/Bridges:
e. Other :
12. How are the application rates and spreading pattern set on the above equipment?
13. Is abrasive/chemical use and durability checked after the storm to determine the
effectiveness of treatment, adherence to application rate specifications, generation
of fines, etc.? If so, how is this performed?
Could you send a copy of the above method(s) by first class mail?
Figure A-1 (Continued)
A-7
-------
14. What specific type of equipment (e.g. broom sweeping, flushing, vacuuming) is
used to clean up the abrasive/chemical residue from the roads after the storm?
How soon is clean-up started after the storm? How often does the clean-up
equipment operate over the same route after the storm?
TIME
REMOVAL AFTER REMOVAL
EQUIPMENT STORM FREQUENCY
a. State Highways (rural areas):
b. Expressways:
c. City Streets:
d. Major Intersections/Bridges:
e. Other
15. Is the cleanliness of the road surface checked after post-storm (or spring) clean-
up is completed? |f So, how is this performed?
16. Considering variations in storm frequency from year-to-year, do you believe that
your annual use of abrasives/chemicals is rising, falling, or staying about the
same? If your consumption is falling, what specific practices have you
used to minimize anti-skid material usage?
Figure A-1 (Continued)
A-8
-------
OTHER COMMENTS
MATERIALS TO BE SENT TO MRI:
Person Completing Questionnaire: Date:
FOLLOW-UP CONVERSATION:
Person Performing Folfow-Up:
Date of Follow-Up:
Figure A-1 (Concluded)
A-9
-------
SD:DOT
Sanaing Materials Questionaire Agency:
(1) Check cne types of materials used tor Winter Operations Sau
Pic Run Sana Crushea Pic Kun Afeg
Screenea Pic Run Sana Crusned Quarry Aggf
Ocners (Specify) &&
(2) Wnicn of tnose cnecKea proviaes tne oest sicia resistance?
Provide specifications ror material y-9u purcnase from commercial
sources.
% Passing •
(4) Proviae specifications ror material-you produce (crusn/screen) witn
your own rorces - If screening/crusnin^ is not necessary, what is
normal maximum particle size.
% Passing
(5) Do you nave a significant numoer of broken windsniela complaints witn
tne materials you use?
(6) If you use both natural (round) or crushed materials is there a
difrerence in number or broken windshield complaints between tne two?Yes
Complaints more common wicn wnicn material?
(7) Do your specifications include requirements for Unit Weight? - No.
(8) Do your specifications ror crusned pit run or quarry aggregate require
a minimum percentage or fracturea faces? No Minimum % Fracture
on face(s) for material retained on tne
(one-two-more) (sieve.
sieve.
Name: . — . Title: .
Address: : Phone:
Figure A-2. Sanding materials questionnaire.
A-10
-------
APPENDIX B
BIBLIOGRAPHY
B-1
-------
-------
Tremayne, J. E. 1938. "Calcium Chloride Treatment of Abrasives Used in Ice Control,"
Can. Eng'r., 75:8-10, November 15.
Schneider, T. R. 1952. "Snowdrifts and Winter Ice on Roads," NRC TT-1038, National
Research Council of Canada, Ottawa, Technical Translation of original report from the
Eidgenossissches Institut fur Schnee-und Lawinenforschung, Intemer Bericht NR. 302,
141, 1959.
Highway Research Board. 1954. "Recommended Practice for Snow Removal and
Treatment of Icy Pavements," Current Road Problems, No. 9, 3rd Rev., Dept. Maint.,
Comm. Snow and Ice Control, Highway Research Board, NAS-NRC, Washington, D.C.
Nichols, R. J., and W. I. J. Price. 1956. "Salt Treatment for Clearing Snow and Ice," The
Surveyor, 115:886-888.
Wirshing, R. J. 1957. "Effect of Deicing Salts on Corrosion of Automobiles," Bull. 150,
Highway Research Board, NAS-NRC, Washington, D.C.
Brohm, D. R, and H. R. Edwards. 1960. "Use of Chemicals and Abrasives in Snow and
Ice Removal From Highways," Res. Bull. 252, Highway Research Board, NAS-NRC Publ.
761, Washington, D.C.
Webster, H. A. 1961. "Automobile Body Corrosion Problems," Corrosion, 17:8, 9-12.
Highway Research Board. 1962. "Current Practices for Highway Snow and Ice Control,"
Current Road Problems, No. 9, 4th Rev., Highway Research Board, NAS-NRC,
Washington, D.C.
Himmelman, B. F. 1963. "Ice Removal on Highways and Outdoor Storage of Chloride
Salts," Highw. Res. Rec. 11, Highway Research Board, NAS-NRC, Washington, D.C.
Solvay Technical and Engineering Service. 1963. Calcium Chloride, 3rd Edition, Bulletin
No. 16.
Jorgensen, R., et al. 1964. "Nonchemical Methods of Snow and Ice Control on Highway
Structures," Report No. 4, National Cooperative Highway Research Program, Highway
Research Board, Washington, D.C.
Lang, C. H., and W. E. Dickinson. 1964. "Snow and Ice Control with Chemical Mixtures
and Abrasives," Highw. Res. Rec. 61, Highway Research Board, NAS-NRC, Washington,
D.C., pp. 14-18.
Fromm, H. J. 1967. "The Corrosion of Autobody Steel and the Effects on Inhibited
Deicing Salts," Rep. No. RR135, Department of Highways, Toronto, Ontario.
B-2
-------
National Research Council. 1967. "Manual on Snow Removal and Ice Control in Urban
Areas," Tech. Memo. No. 93, NRC 9904, National Research Council of Canada Ottawa
Ontario.
Minsk, D. L. 1968. "Electrically Conductive Asphalt for Control of Snow and Ice
Accumulation," Highw. Res. Rec., No. 227, Highway Research Board, Washington D C
pp. 57-63. • • •-
Road Research Laboratory. 1968. "Salt Treatment of Snow and Ice on Roads," Road
Note No. 18, Dept. Science and Industrial Research, 2nd Ed., Her Majesty's Stationery
Off., London.
Dunnery, D. A. 1970. "Chemical Melting of Ice and Snow on Paved Surfaces," Highway
Research Board Special Reports No. 115, 172-176.
Spellman, D. L, and R. F. Stratfull. 1970. "Chlorides and Bridge Deck Deterioration "
Highw. Res. Rec. 328, Highway Research Board, NAS-NRC, Washington, D.C.
Byrd, Tallamy, et al. 1971. "Snow Removal and Ice Control Techniques at
Interchanges," Highw. Res. Rec. 127, Highway Research Board, NAS-NRC, Washington,
u.c.
Stewart, C. 1971. "Deterioration in Salted Bridge Decks," Spec/a/ Report 116 Highway
Research Board, NAS-NRC, Washington, D.C.
Struzeski, E. 1971. "Environmental Impact of Highway Deicing," Final Report 11040
GKK, U.S. Environmental Protection Agency, Edison, New Jersey, June.
Anon. 1973. "Recycled Winter Sand Tested in Connecticut," Rural and Urban Roads
11:2, 51, February.
Murray, D. M., and M. R. Eigerman. 1973. "A Search: New Technology for Pavement
Snow and Ice Control," EPA-R2-72-125, U.S. Environmental Protection Agency
Washington, D.C., December.
Richardson, D. L, et al. 1974. "Manual for Deicing Chemicals: Application Practices,"
EPA-670/2-74-045, National Environmental Research Center, U.S. Environmental
Protection Agency, Cincinnati, Ohio, December.
NACE Group Committee T-3. 1975. "Deicing Salts, Their Use and Effects," Matls Pert
14:9-14, April.
American Association of State Highway and Transportation Officials. 1976 "AASHTO
Maintenance Manual," First Edition, AASHTO, Washington, D.C., February.
B-3
-------
Welch, B. H., et al. 1976. "Economic Impact of Highway Snow and Ice Control-State-of-
the-Art," Report No. FHWA-RD-7720, Federal Highway Administration, Offices of Research
and Development, Washington, D.C.
McBride, J. C., et al. 1977. "Economic Impact of Highway Snow and Ice Control ESIC--
User's Manual," Report No. FHWA-RD-77-96, Federal Highway Administration, Offices of
Research and Development, Washington, D.C.
Dunn, S. A., and R. U. Schenk. 1979. "Alternative Highway Deicing Chemicals," in Snow
Control and Ice Control Research, Special Report 185, Transportation Research Board,
Washington, D.C.
Hu, A. C. 1979. "The Effect of Chloride Concentration on Automobile Stopping
Distance," \nSnowControlandlce Control Research, Special Report 185, Transportation
Research Board, Washington, D.C.
Dunn, S. A., and R. U. Schenk. 1980. "Alternate Highway Deicing Chemicals,"
FHWA-RD-79-108, Federal Highway Administration, Offices of Research and Development,
Washington, D.C., March.
Baboian, R. 1981. "The Automotive Environment," In Automotive Corrosion by Deicing
Salts, National Association of Corrosion Engineers, Houston, Texas.
Brown, M. G. 1981. "Corrosion of Highway Appurtenances Due to Deicing Salts," in
Automotive Corrosion by Deicing Salts, National Association of Corrosion Engineers,
Houston, Texas.
Cook, A. R. 1981. "Deicing Salts and the Longevity of Reinforced Concrete," in
Automotive Corrosion by Deicing Salts, National Association of Corrosion Engineers,
Houston, Texas.
Fromm, H. J. 1981. "Winter Maintenance Practice and Research in Ontario," in
Automotive Corrosion by Deicing Salts, National Association of Corrosion Engineers,
Houston, Texas.
Gray, D. M., and D. H. Male, Editors. 1981. Handbook of Snow-Principles, Processes,
Management, and Use, Pergamon Press, New York.
McDonald, R. D. 1981. "Automotive Underbody Corrosion Testing," in Automotive
Corrosion by Deicing Salts, National Association of Corrosion Engineers, Houston, Texas.
Passaglia, E., and R. A. Haines. 1981. "The National Cost of Automobile Corrosion," in
Automotive Corrosion by Deicing Salts, National Association of Corrosion Engineers,
Houston, Texas.
B-4
-------
Wood, F. O. 1981. "Survey of Salt Use for Deicing Purposes," In Automotive Corrosion
by Deicing Salts, National Association of Corrosion Engineers, Houston, Texas.
Minsk, D. L 1982. "Optimizing Deicing Chemical Application Rates," CREEL Report 82-
18, Federal Highway Administration, Washington, D.C., August.
Zaman, M. S., et al. 1982. "Prediction of Deterioration of Concrete Due to Freezing and
Thawing and to Deicing Chemical Use," American Concrete Institute Journal 79-1 56-58
January-February. '—' '
Zaman, M. S., et al. 1982. "Prediction of Deterioration of Concrete Due to Freezing and
Thawing and to Deicing Chemical Use," American Concrete Institute Journal 6*502-504
November-December.
Eck, R. W., et al. 1983. "Evaluation of the Effect of Natural Brine Deicing Agents on
Pavement Materials," Transportation Research Record, 933:24-31.
Huisman, C. L. 1983. "Cost Effective Roadway Deicing Using: Abrasives, Salt and
Calcium Chloride," Presented at Iowa Snow Conference, Ames, Iowa.
Connecticut Department of Transportation. 1984. "Snow and Ice Control Policy," State
of Connecticut Bureau of Highways.
Anderson, R. W. 1985. "Safety Restoration During Snow Removal: Problems and
Corrective Guidelines for Post Snow Sjorm Cleanup Operations," Transafety Reporter
3:12, 4-5, December.
Saarela, A. 1985. "Future Views on Road Maintenance," Tie ja Liikenne, 55:8, 329-331.
Venermo, V., et al. 1985. "Katujen liukkaudentorjunta (Antiskid Treatment of Streets) "
Tie ja Liikenne, 55:8, 332-335.
Eck, R. W., et al. 1987. "Natural Brine as an Additive to Abrasive Materials and Deicing
Salts," Transportation Research Record 1127, 16-26.
Pitt, J. M., et al. 1987. "Sulfate Impurities From Deicing Salt and Durability of Portland
Cement Mortar," Transportation Research Record 1110, 16-23.
Anonymous 1988. "Winter Maintenance-Learning Finland's Methods " Better Roads
58:6, 22-26, June.
Hiatt, G. F. S., et al. 1988. "Calcium Magnesium Acetate: Comparative Toxicity Tests
and an Industrial Hygiene Site Investigation," Transportation Research Record 1157 20-
26.
B-5
-------
Jokinen, M. 1988. "Katupolun Vahentamista Tutkfttu Turussa (Decreasing Street Dust
Problems-A Study in Turku)," Tie ja Liikenne, 58:6, 22-24.
McElroy, A. D., et al. 1988. "Comparative Study of Chemical Deicers," Transportation
Research Record 1157, 1-11.
McElroy, A. DM et al. 1988. "Study on Wetting Salt and Sand Stockpiles with Liquid
Calcium Chloride," Transportation Research Record 1157, 38-43.
Nadezhdin, A., D. A. Mason, B. Malric, D. F. Lawless, and J. P. Fedpsoff. 1988. "The
Effect of Deicing Chemicals on Reinforced Concrete," Transportation Research Record
1157, 31-37.
Rainiero, J. M. 1988. "Investigation of the Ice-Retardant Characteristics of Verglimit-
Modified Asphalt," Transportation Research Record 1157, 44-53.
Slick, D. S. 1988. "Effects of Calcium Magnesium Acetate on Pavements and Motor
Vehicles," Transportation Research Record 1157, 27-30.
Anderson, R. C., and C. Auster. "Costs and Benefits of Road Salting," Environ. Affairs,
3:1, 129-144.
B-6
-------
-------
APPENDIX C
ALTERNATIVE SKID CONTROL MEASURES
C-1
-------
ALTERNATIVE SKID CONTROL MEASURES2
Essentially all roadway ice and snow control measures in the United States consist
of the long-established methods of plowing, sanding, and salting. Alternative measures
to remove snow and ice from roadways have been extensively studied in the literature
particularly to identify methods that will reduce environmental damage resulting from the
application of chloride salts. These alternative measures include other deicing chemicals
as well as pavement coatings to prevent ice adhesion, resistance heating, mechanical ice
removal methods, and new tire and vehicle design. The following sections describe
studies of many of these alternative methods.
C.1 ALTERNATIVE MATERIALS
The Federal Highway Administration has conducted an extensive search (Dunn,
1980) to find road deicing chemicals to replace sodium chloride. Criteria for selection
included a decrease in water solubility and freezing point, corrosivity, toxicity, relative cost
(or cost potential), effect on soils, plants and water supplies, and flammability. The
chemicals analyzed for deicing effectiveness consisted of both inorganic salts and organic
compounds.
•
Inorganic salts (ionic compounds) are sodium-containing salts, salts containing
chloride, nitrate or sulfate, sodium salts, carbonic and phosphoric acids, potassium salts
of carbonic acid, potassium salts of phosphoric acid, tetrapotassium pyrophosphate,
ammonium salts of phosphoric acid and ammonium salts of carbonic acid. Metal organic
salts were also studied, as was glycine.
Nonionic deicers that were investigated by Dunn (1980) included methanol, ethanol
and isopropanol, acetone, urea, formamide, dimethyl sulfoxide (DMSO) and ethyl
carbamate (urethane). Two chemicals, calcium magnesium acetate (CMA) and methanol,
were also field tested on a road, highway, sidewalk and parking ramp.
CMA and methanol were found to be as effective as sodium chloride, but without
its drawbacks. Methanol reacts almost immediately to melt snow and ice, but is less
persistent than sodium chloride. The eutectic temperature of methanol '(-120°C or
-184°F) is much lower than that of sodium chloride, and thus methanol works well at low
ambient temperatures (i.e., below -7°C or 20°F) when the melting efficiency of sodium
chloride diminishes.
The application of methanol to roadways would not create a source of particulate
emissions, but would be a source of short term volatile organic compounds (VOCs) to
persons applying the chemical because of its relatively high evaporation rate. For
workers exposed to methanol vapors, the Occupational Safety and Health Administration
has established an 8-h permissible exposure limit (PEL) of 200 ppm and a 15-min short
References listed in Section 8 of main report.
C-2
-------
term exposure limit (STEL) of 250 ppm. Also methanol can be absorbed by skin contact
and has been found to be toxic to some animal species. This chemical is also very
flammable during storage, but once applied to a road, will not ignite even when puddled
on a snow covered road (Dunn, 1980).
The other candidate deicer, CMA, acts at about the same rate as sodium chloride
in the temperature range of common activity and shows about the same persistence.
However, CMA requires free water to be effective in melting snow and ice. CMA has
other advantages:
1. Braking traction and skidding friction are about the same as NaCI.
2. Corrosion is retarded.
3. Soils are not affected.
4. Drinking water supplies are not harmed.
N
5. Unpurified CMA grains can be dark in color to enhance insolation and
accelerate melting.
Dunn's study (1980) also reported that the acetate gradually decomposes to yield
carbon dioxide and water, whereas the calcium and magnesium are reconverted to
limestone, which would potentially be a source of fugitive particulate emissions. The
unpurified CMA reaction product also contains calcium and magnesium carbonates (and
some oxalates) as the insoluble impurities. However no data were found in the literature
to indicate the particle sizes of these impurities or of the limestone residue of CMA. CMA
is mildly basic, and some respiratory protection against dust would be desirable for
workers handling the chemical.
Chollar (1984) noted that according to theoretical considerations, the weight ratio
of CMA to sodium chloride to obtain equal deicing capabilities is 1.7 to 1.0. Because of
the significantly higher cost of CMA compared to rock salt, Chollar foresaw future use of
CMA only in certain areas, such as bridge decks, in urban areas of high traffic volume,
and in areas of possible contamination of water supplies by sodium.
Table C-1 presents a summary of alternative deicing chemicals prepared by the
Iowa Department of Transportation (1980). The Iowa report discussed advantages and
disadvantages of each chemical based on effectiveness, cost, toxicity, and corrosiveness.
Only five chemical alternatives are proposed, since Iowa stated that other alternative
deicing chemicals had been tested and were not recommended because of "problems
with safety, availability, economy and/or ecology." The nonrecommended chemicals
included: ammonium acetate; ammonium nitrate; ammonium sulfate; alcohols; glycols;
sodium formate or calcium formate; and ammonium carbamate.
C-3
-------
TABLE C-1. ALTERNATIVE DEICING CHEMICALS (IOWA. 1980)
^^™""^~~' a^^^^^^^^^s - -T^^a^^ai^^^^^sa^^agg^^g'^^^Bg5g--B—BL •••.....^——-^^—___—_^____^^
Alternative chemical Advantages Disadvantages and costs'
Formamide
Urea
Tetrapotassium
pyre-phosphate (TKPP)
Formamide-urea-water
(75%) (20%) (5%)
(liquid)
Tripotassium
Phosphate-formamide
(75%) (25%)
(pellet)
Noncorrosive
Relatively nonslippery on pavement
Nontoxic unless biodegraded
Effective for use in automatic ice
prevention systems
Noncorrosive
Nonconductive
Has higher freezing point
10 to 15 times more costly
Toxicity no greater
Harmful, ecological effects no greater
At very low temperatures outperforms
sa(ts
Less effective
1Vi to 2 times as much required for
same effect
5 to 10 times more costly
Can be toxic
Acceptable deicing chemical only
above 15°F(-9°C)
Runoff must be controlled, or only
use in areas where runoff will not
enter critical waterways
Corrosive effect on vehicular steel
15+ times as costly
Costs 10+ times as much
Considerably less corrosive
Nontoxic if not biodegraded
Effective for automatic ice prevention
systems
Can be applied with existing equipment Minor scaling of concrete may result
Will melt ice at -10°F (-23.3°C)
Effective at lower temperatures at
reduced rate
Corrosive effects acceptable . ,
Nontoxic
May promote vegetation growth
Costs 10 to 15 times as much
Cost comparisons relative to rock salt.
C-4
-------
The Iowa report also identified two pavement coating formulas to reduce the
adherence of ice on pavements without compromising pavement skid resistance qualities.
These included a modified traffic paint containing silicone rubber and a waterproofing
compound combined with silicone rubber. Applications of these two pavement coatings
were reported to have an estimated effective wear life of only one to two months on
tested roads (150,000 to 300,000 vehicles passes). Another reported drawback is the
release of flammable vapors into the atmosphere during application of these pavement
coatings.
C.2 OTHER TECHNIQUES
According to Eck (1986), a properly designed snowplow can remove about 95% of
a 2-in snowfall, with the remaining 5% removable with a small amount of rock salt. Seattle
was reported to reduce chemical use by early plowing, under conditions of moderate to
heavy snowfall (TRB, 1974). Wyoming (1984) also reported that wet snow usually
requires immediate plowing and sanding, but dry snowfall should be allowed to build up
to 1 to 2 in before plowing. Mechanical devices for snow/ice removal have been
extensively studied as shown in Figure C-1 from Minsk (1981) and in Table C-2from EPA
(1972).
•
The Iowa Department of Transportation (1980) studied alternative deicing measures
to reduce corrosion. These included bridge heating (including mobile thermal units), and
application of ultrasonic (vibrational) and electromagnetic energy.
Wood (1983) also suggested a number of alternative methods to remove (or prevent
the buildup of) snow and ice on roadways. These included electrically heated
pavements, earth-heated pavements and geothermal heating, urethane foam and
styrofoam insulation, air jet plows, infrared heat lamps and underbody-blade trucks.
Rainiero (1988) and others have also investigated the use of pavement incorporating
encapsulated pellets of calcium chloride. This material, referred to as Verglimit, has been
tested in the United States, Canada, and Europe, with some success.
There have been a number of experiments with heated pavements, especially in
England. According to EPA (1972), there are two types of road heaters: fluid circulating
systems and electrical resistance heaters. Flufd circulating systems usually consist of
pipes embedded in pavement that carry a heated, low-freezing point liquid. Pipes are
closely spaced (about 1 to 1.5 ft apart) at the time the pavement is laid. Electrical
pavement heaters include high voltage, insulated cable systems; low voltage systems
using uninsulated metal mesh; and electrically conductive pavements.
Other alternative approaches have entailed the use of vibrational and
electromagnetic energy. Both methods are very inefficient because the energy
requirements are too great.
C-5
-------
Displacement Plows
Side-mounted (wings)
Front-mounted
Underbody
Trailing
I
V-blade
One-way
Truck- Drags
mounted
Reversible
Fixed
Swivel Rollover
(a) Displacement plows
Rotary Plows
I— 1
Two-element
Single-element
I
Auger
Cutter
!- Horizontal axis
- Vertical axis
Swept-back axis
Scoop wheel
Drum
-Helical
- Horizontal
(b) Rotary plows
Specialized Equipment
Pure blowers
- Compressor fed
(compressed air jet)
~ Combustion jet
(jet engine)
Power brooms
Hybrid machines
Ice removal devices
~ Combination blade
and impeller
*~ Combination blade
and cutter
"" Combination broom
and blower
' Combination blade
and blower
- Wobble wheel
- Spiral rolls
- Scarifier (serrated blade)
- Hammers
(c) Specialized equipment
Figure C-1. Family tree of snow removal equipment (Minsk, 1981).
C-6
-------
TABLE C-2. MECHANICAL DEVICES FOR SNOW REMOVAL AND
ICE CONTROL (EPA, 1972)
1. Blade or displacement plows
1.1 Front-mounted
One-way blade
Fixed position (right or left cast)
Reversible (swivel or roll-over)
V-blade
1.2 Underbody
Road grader
Truck-mounted
1.3 Side-mounted (wings)
1.4 Trailing
2. Rotary snow blowers
2.1 Two-element (impeller)
Auger
Horizontal axis
Vertical axis
Swept-back axis
Cutters
Helical
Horizontal breakers (rakes)
2.2 Single-element (no impeller)
Scoop wheel
Drum
3. Pure blowers
3.1 Compressed air jet (compressor-fed)
3.2 Combustion jet (jet engine)
4. Power brooms (rotary brush)
•
5. Specialized ice removing equipment
5.1 Ice crushing rollers (pressure cutting edges)
Wobble wheel
Spiral rolls
Serrated blade
5.2 Ice melting machines
6. Combination equipment
6.1 Blade and impeller
6.2 Blade and cutter
6.3 Blade and compressed air
6.4 Broom and blower
6.5 Burner and blower
6.6 Burner, broom, and vacuum
C-7
-------
A comprehensive study led by Blackburn (1978) produced the document, « Physical
Alternatives to Chemicals for Highway Deicing." This U.S. Department of Transportation
document reported on practical and economical alternative means of highway deicinq
Twenty-one deicing concepts were identified and evaluated for their potential usefulness
The selection of concepts was guided by a consideration of three separate steps that are
generally required of any successful method to remove snow and ice from roadways:
1. Prevent, reduce, or eliminate the bond between the pavement and the
ice/snow.
2. Break and dislodge the ice/snow.
3. Remove the ice/snow fragments from the traveled way.
A rating scheme was developed by Blackburn (1978) to compare the different
physical alternatives to deicing chemicals and to select the most feasible methods for
further study. A five-category rating scheme was used:
1. Technical feasibility.
2. Operational feasibility-maintenance operations.
3. Operational feasibility-traffic operations and safety.
4. Economic acceptability.
5. Environmental acceptability.
This rating scheme was applied to the identified alternative physical deicing methods and
the results are summarized in Table C-3. Unadjusted, or raw, ratings "are the best
available measures of the feasibility/acceptability of a concept, based on the rating for
. each category, arrived at upon objective consideration of many factors." Adjusted ratings
incorporate a measure of uncertainty to include the reliability (or certainty) of the ratings
A rating of 10 is best, whereas a rating of 0 is the least desirable rating.
About 75% of the investigated alternatives have low ratings resulting from excessive
power requirements, size and bulk of the equipment required, low efficiency high costs
safety factors, and potential damage to the environment (including pavement)'
Unadjusted overall ratings of less than 5.5 identify concepts considered unacceptable
Four candidate concepts had ratings between 5.5 and 6.0 and were noted as deserving
additional consideration. These included release agents in the surface course (e g
Verglimit-modified asphalt), mechanical waves (ultrasonics), rotary blowers and plow
blades with air jets. '
C-8
-------
TABLE C-3. RATING OF CONCEPTS FOR ALTERNATIVE PHYSICAL DEICING METHODS (BLACKBURN. 1978)
o
Category rating values'
Concept
Step 1 : Prevention, reduction, or elimination of adhesive bond of
ice to pavement
• Release agents on pavement surface
• Gas releasing agents in surface course
• Release agents In surface course
• Temporary melting at Interface with energy
delivered by:
— Current In surface course
— Visual range electromagnetic radiation
—Microwave electromagnetic radiation
— Induced current at the interface
—Induced current In surface course
• Gas formation at the Interface by electrolysis
Step 2: Breaking and dislodging of the Ice/snow layer
• Mechanical devices producing steady or Impact
stresses
• Mechanical waves (ultrasonic)
• Stresses under tire passages
• Variable electromagnetic fields
Technical
9.25
8.00
6.50
3.75
3.50
3.00
2.50
2.50
2.50
8.25
5.25
5.60
3.00
feasibility
7.45
4.23
3.88
2.38
1.05
0.98
0.48
0.58
0.26
6.08
1.13
1.05
0.36
Operational feasibility
maintenance
operations
6.82
6.56
6.21
5.57
5.24
4.92
4.68
4.68
6.10
5.71
"
5.86
5.23
4.63
(continued)
3.90
3.16
3.16
1.82
0.52
0.57
0.48
0.48
0.63
2.61
0.65
1.65
0.47
Operational
feasibility traffic
operations and
safety
3.67
4.33
3.87
4.33
4.33
4.33
4.33
4.33
4.33
5.56
7.74
3.54
6.76
(
1.78
0.43
0.56
0.83
0.43
0.43
0.43
0.43
0.43
3.69
0.77
1.58
0.67
Economic
acceptability
.
6.16
7.48
6.16
4.78
4.15
3.73
4.90
4.69
4.99
4.46
2.67
7.07
1.78
0.66
0.76
0.75
0.48
0.41
0.37
0.49
0.61
0.50
0.44
0.26
0.72
0.18
Environmental
acceptability
8.40
7.60
7.20
8.80
6.80
6.80
6.40
6.40
8.80
4
7.20
6.80
6.00
6.80
•
3.40
1.48
2.64
0.88
0.68
0.68
0.64
0.64
0.88
0.72
0.68
2.40
0.68
Overall rating
values'
(average)
6.86
6.79
5.95
5.45
4.80
4.56
4.56
4.52
5.34
6.24
5.66
5.47
4.59
3.44
2.01
2.20
1.28
0.62
0.60
0.50
0.55
0.54
2.71
0.70
1.48
0.47
-------
TABLE C-3 (continued)
Category rating values'
Operational
Concept
• Gas formations by electrolysis at Ice/snow-to-
pavement Interface
• Heated plow blades
• Jets of gas or liquids driven by compressors and
pumps
Step 3: Removal of the Ice/snow from the traveled way
• Blade action
o
ji^ • Rotary blowers
• Plow blades with air jets
• Traffic action
• Sweeper action
Technical feasibility
2.75
3.00
3.00
9.50
7.50
7.00
5.50
5.50
0.43
0.36
0.46
9.05
7.25
3.30
1.05
4.70
Operational feasibility
maintenance
operations
4.35
3.91
3.48
5.58
5.25
5.58
5.23
4.92
0.45
0.53
0.43
4.20
4.59
0.87
1.65
3.44
=====
feasibility traffic
operations and
safety
6.76
5.74
4.30
5.18
5.74
4.64
3.54
5.56
=====
0.67
0.91
0.87
3.99
4.59
0.93
1.58
3.34
========
Economic
acceptability
1.78
1.89
2.68
6.74
3.48
4.89
7.07
4.53
========
0.18
0.21
0.27
5.52
2.48
0.49
0.72
1.50
Environmental
acceptability
6.80
6.80
5.60
7.60
7.20
6.20
6.00
6.40
0.68
0.68
0.56
7.60
7.20
0.62
2.40
4.12
Overall rating
values"
(average)
4.49
4.27
3.81
6.92
5.83
5.66
5.47
5.38
0.48
0.54
0.52
6.07
5.22
1.24
1.48
3.42
==
• Numbers appearing In left column of each category are unadjusted rating values; adjusted rating values appear In right column of each category.
-------
Four closely related concepts had unadjusted ratings above 6 and were evaluated
as the most promising alternative deicing methods: release agents on the pavement
surface, gas-releasing agents in the surface course, mechanical deicing methods
producing steady or impact stresses, and blade action.
One of the most promising concepts was actually tested by Blackburn (1978)—
mechanical devices (rotating metal discs) that put steady stresses on the ice/snow
pavement bond. Various disc configurations, an array of two discs, and a simulated plow
blade were laboratory tested. Results from these tests were mixed, but suggested that
ice was difficult to remove completely by mechanical devices producing steady stresses.
The adhesive bond of ice to some materials can exceed the cohesive strength in the ice
itself, and may be supplemented by mechanical interlocking to pavement surface
asperities. Discs need to penetrate through to the ice-pavement interface, but disc depth
proved difficult to control. This problem could possibly cause more than superficial
pavement damage by field equipment.
C.3 CONCLUSIONS
Of the alterative deicers identified above, only CMA shows some limited potential
for applications such as bridges, intersections.and the like. For this reason, CMA was
compared to two common deicers (rock salt and CaClg) in small scale experiments. The
results of these experiments can be found in Section 6.3.3 of the main text.
With regard to other alternative methods of ice and snow control, new techniques
for mechanical ice disbonding appears to be the most promising. An evaluation of these
techniques was beyond the scope of this particular study and thus were not evaluated.
C-11
-------
C-12
-------
APPENDIX D
ASTM SILT ANALYSIS METHODS
D-1
-------
-------
Designation: C 117 - 87
Standard Test Method for
Materials Finer than 75-jim (No. 200) Sieve in Mineral
Aggregates by Washing1
This standard is issued under the fixed designation C 117; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (<) indicates an editorial change since the last revision or reapproval.
This test method has been approved for use by agencies of the Department of Defense and for listing in the DoD Index of Specifications
and Standards.
1. Scope
1.1 This test method covers determination of the amount
of material finer than a 75-um (No. 200) sieve in aggregate
by washing. Clay particles and other aggregate particles that
are dispersed by the wash water, as well as water-soluble
materials, will be removed from the aggregate during the test.
1.2 Two procedures are included, one using only water for
the washing operation, and the other including a wetting
agent to assist the loosening of the material finer than the
75-um (No. 200) sieve from the coarser material. Unless
otherwise specified, Procedure A (water only) shall be used.
1.3 The values stated in acceptable metric units are to be
regarded as the standard.
1.4 This standard may involve hazardous materials, oper-
ations, and equipment. This standard does not purport to
address all of the safety problems associated with its use. It is
the responsibility of the user of this standard to establish
appropriate safety and health practices and determine the
applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 ASTM Standards:
C 136 Method for Sieve Analysis of Fine and Coarse
Aggregates2
C 6 70 Practice for Preparing Precision and Bias State-
ments for Test Methods for Construction Materials2
C 702 Practice for Reducing Field Samples of Aggregate to
Testing Size2
D 75 Practice for Sampling Aggregates3
E 11 Specification for Wire-Cloth Sieves for Testing
Purposes4
2.2 AASHTO Standard:
T 11 Method of Test for Amount of Material Finer than
0.075-mm Sieve in Aggregate5
' This test method is under the jurisdiction of ASTM Committee C-9 on
Concrete and Concrete Aggregates and is the direct responsibility of Subcommittee
C W 03.05 on Methods of Testing and Specifications for Physical Characteristics
°f Concrete Aggregates.
Current edition approved Feb. 9, 1987. Published March 1987. Originally
Published as C 117 - 35 T. Last previous edition C 117 - 84.
An alternative procedure that allows the use of a wetting agent was added to the
""rent edition of this test method.
1 Annual Book of ASTM Standards. Vol 04.02.
3 Annual Book of ASTM Standards. Vols 04.02 and 04.03.
'Annual Book of ASTM Standards. Vols 04.02 and 14.02.
Available from the American Association of State Highway and Transpora-
"oi Officials, 444 N. Capitol St., NW, Suite 225, Washington, DC 20C01.
3. Summary of Method
3.1 A sample of the aggregate is washed in a prescribed
manner, using either plain water or water containing a
wetting agent, as specified. The decanted wash water, 'con-
taining suspended and dissolved material, is passed through a
75-um (No. 200) sieve. The loss in mass resulting from the
wash treatment is calculated as mass percent of the original
sample and is reported as the percentage of material finer
than a 75-um (No. 200) sieve by washing.
4. Significance and Use
4.1 Material finer than the 75-um (No. 200) sieve can be
separated from larger particles much more efficiently and
completely by wet sieving than through the Use of dry
sieving. Therefore, when accurate determinations of material
finer than 75 um in fine or coarse aggregate are desired, this
test method is used on the sample prior to dry sieving in
accordance with Method C 136. The results of this test
method are included in the calculation in Method C 136,
and the total amount of material finer than 75 um by
washing, plus that obtained by dry sieving the same sample,
is reported with the results of Method C 136. Usually the
additional amount of material finer than 75 um obtained in
the dry sieving process is a small amount. If it is large, the
efficiency of the washing operation should be checked. It
could, also, be an indication of degradation of the aggregate.
4.2 Plain water is adequate to separate the material finer
than 75 um from the coarser material with most aggregates.
In some cases, the finer material is adhering to the larger
particles, such as some clay coatings and coatings on
aggregates that have been extracted from bituminous mix-
tures. In these cases, the fine material will be separated more
readily with a wetting agent in the water.
5. Apparatus and Materials
5.1 Balance—A balance or scale readable and accurate to
0.1 g or 0.1 % of the test load, whichever is greater, at any
point within the range of use.
5.2 Sieves—A nest of two sieves, the lower being a 75-um
(No. 200) sieve and the upper a 1.18-mm (No. 16) sieve,
both conforming to the requirements of Specification Ell.
5.3 Container—A pan or vessel of a size sufficient to
contain the sample covered with water and to permit
vigorous agitation without loss of any pan of the sample or
water.
5.4 Oven—An oven of sufficient size, capable of main-
taining a uniform temperature of 110 ± 5*C (230 ± 9*F).
D-2
-------
C117
5.5 Wetting Agent—Any dispersing agent, such as liquid
dishwashing detergents, that will promote separation of the
fine materials.
NOTE 1—The use of a mechanical apparatus to perform the washing
operation is not precluded, provided the results are consistent with those
obtained using manual operations. The use of some mechanical washing
equipment with some samples may cause degradation of the sample.
6. Sampling
6.1 Sample the aggregate in accordance with Practice
D 75. If the same test sample is to be tested for sieve analysis
according to Method C 136, comply with the applicable
requirements of that method.
6.2 Thoroughly mix the sample of aggregate to be tested
and reduce the quantity to an amount suitable for testing
using the applicable methods described in Methods C 702. If
the same test sample is to be tested according to Method
C 136, the minimum mass shall be as described in the
applicable sections of that method. Otherwise, the mass of
the test sample, after drying, shall conform with the fol-
lowing: . .
Nominal Maximum Size
Minimum Mass, g
2.36 mm (No. 8) 100
4.75 mm (No. 4) 500
9.5 mm (H in.) .1000
19.0 mm (% in.) 2500
37J mm (IVi in.) or larger 5000
The test sample shall be the end result of the reduction.
Reduction to an exact predetermined mass shall not be
permitted.
7. Selection of Procedure
7.1 Procedure A shall be used, unless otherwise specified
by the Specification with which the test results are to be
compared, or when directed by the agency for which the
work is performed.
8. Procedure A—Washing with Plain Water
8.1 Dry the test sample to constant mass at a temperature
of 110 ± 5"C (230 ± 9'F). Determine the mass to the nearest
0.1 % of the mass of the test sample.
8.2 If the applicable specification requires that the
amount passing the 75-um (No. 200) sieve shall be deter-
mined on a portion of the sample passing a sieve smaller
than the nominal maximum size of the aggregate, separate
the sample on the designated sieve and determine the mass
of the material passing the designated sieve to 0.1 % of the
mass of this portion of the test sample. Use this mass as the
original dry mass of the test sample in 10.1.
NOTE 2—Some specifications for aggregates with a nominal max-
imum size of 50 mm or greater, for example, provide a limit for material
passing the 75-um (No. 200) sieve determined on that portion of the
sample passing the 25.0-mm sieve. Such procedures are necessary since
it is impractical to wash samples of the size required when the same test
sample is to be used for sieve analysis by Method C 136.
8.3 After drying and determining the mass, place the test
sample in the container and add sufficient water to cover it.
No detergent, dispersing agent, or other substance shall be
added to the water. Agitate the sample with sufficient vigor
to result in complete separation of all particles finer than the
75-um (No. 200) sieve from the coarser particles, and to
bring the fine material into suspension. Immediately pour
the wash water containing the suspended and dissolved soli<
over the nested sieves, arranged with the coarser sieve on to
Take care to avoid, as much as feasible, the decantation <
coarser particles of the sample.
8.4 Add a second charge of water to the sample in t!
container, agitate, and decant as before. Repeat this oper
tion until the wash water is clear.
NOTE 3—If mechanical washing equipment is used, the charging
water, agitating, and decanting may be a continuous operation.
8.5 Return all material retained on the nested sieves t
flushing to the washed sample. Dry the washed aggregate •
constant mass at a temperature of 110 ± 5*C (230 ± 9°F) ar
determine the mass to the nearest 0.1 % of the original ma
of the sample.
9. Procedure B—Washing Using a Wetting Agent
9.1 Prepare the sample in the same manner as f<
Procedure A.
9.2 After drying and determining the mass, place the te
sample in the container. Add sufficient water to cover tf
sample, and add wetting agent to the water (Note 4). Agita
the sample with sufficient vigor to result in complete sep;
ration of all particles finer than the 75-um (No. 200) SICN
from the coarser particles, and to bring the fine material ini
suspension. Immediately pour the wash water containing tr
suspended and dissolved solids over the nested sieve
arranged with the coarser sieve on top. Take care to avoid,;
much as feasible, the decantation of coarser particles of tf
sample.
NOTE 4—These should be enough wetting agent to produce a sm;
amount of suds when the sample is agitated. The quantity will deper
on the hardness of the water and the quality of the detergent Excessi-
suds may overflow the sieves and carry some material with them..
9.3 Add a second charge of water (without wetting agen
to the sample in the container, agitate, and decant as befor
Repeat this operation until the wash water is clear.
9.4 Complete the test as for Procedure A.
10. Calculation
10.1 Calculate the amount of material passing a 75-ui
(No. 200) sieve by washing as follows:
A = [(B - Q/B] x 100
where:
A = percentage of material finer than a 75-um (No. 20<
sieve by washing,
B = original dry mass of sample, g, and
C = dry mass of sample after washing, g.
11. Report
11.1 Report the percentage of material finer than tr
75-nm (No. 200) sieve by washing to the nearest 0.1 ?
except if the result is 10 % or more, report the percentage :
the nearest whole number.
11.2 Include a statement as to which procedure was use'
12. Precision and Bias
12.1 The estimates of precision of this test method liste
in Table 1 are based on results from the AASHTO Materia
Reference Laboratory Reference Sample Program, wii
testing conducted by this test method and AASHTO Methc
D-3
-------
48IP C117
TABLE 1 Precision
c,,~i,,* Acceptable Range
Standard
Delation (IS)-, X
Coarse Aggregate0
Single-Operator Precision
Multilaboratory Precision
Fine Aggregate0
Single-Operator Precision
Multilaboratory Precision
0.10
0.22
0.15
0.29
0.28
0.62
0.43
0.82
* These numbers represent the (IS) and (D2S) limits as described in Practice
C 670.
8 Precision estimates are based on aggregates having a nominal maximum size
of 19.0 mm (Vt in.) with toss than 1.5% finer than the 75-tim (No. 200) sieve.
c Precision estimates are based on fine aggregates having 1.0 to 3.OX finer
than the 75-nm (No. 200) sieve.
F 11. The significant differences between the methods at the test results from 40 to 100 laboratories.
ime the data were acquired is that Method T 11 required, 12.2 Bios—Since there is no accepted reference material
ind Method C 117 prohibited, the use of a wetting agent. suitable for determining the bias for the procedure in this test
The data are based on the analyses of more than 100 paired method, no statement on bias is made.
The American Society lor Testing and Materials takes no position respecting the validity of any patent rights asserted in connection
with any item mentioned in this standard. Users of this standard are expressly advised that determination of the validity of any such
patent rights, and the risk of infringement of such rights, are entirely their own responsibility.
This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
U not revised, either reapproved or withdrawn. Your comments are invited either tor revision of this standard or for additional standards
and should be addressed to ASTM Headquarters. Your comments will receive careful consideration at a meeting of the responsible
technical committee, which you may attend. If you feel that your comments have not received a fair hearing you should make your
views known to the ASTM Committee on Standards, 7976 Race St., Philadelphia, PA 79703.
D-4
-------
Designation: C 136 - 84a
Standard Method for
Sieve Analysis of Fine and Coarse Aggregates1
the/lxeddesi8nati°n C 136; the number immediately following the designation indicates the year of
, * ^VlS'0n; u' yeaf °f "* rtVisi0n- A number in P*™""** indicates The year of las, reapprova A
superscnpt epsilon (<) indicates an eduonal change since the last revision or reapproval.
or "' *y
-------
C136
all sizes of samples, since the large sieving area needed for practical
sieving of a large nominal size coarse aggregate very likely could result in
loss of a portion of the sample if used for a small sample of coarse
aggregate or fine aggregate.
5.4 Oven—An oven of appropriate size capable of main-
taining a uniform temperature of 110 ± 5°C (230 ± 9°F).
6. Sampling
6.1 Sample the aggregate in accordance with Practice
D 75. The weight of the field sample shall be the weight
shown in Practice D 75 or four times the weight required in
6.4 and 6.5 (except as modified in 6.6), whichever is greater.
6.2 Thoroughly- mix the sample and reduce it to an
amount suitable for testing using the applicable procedures
described in Methods C 702. The sample for test shall be
approximately of the weight desired when dry and shall be
the end result of the reduction. Reduction to an exact
predetermined weight shall not be permitted.
NOTE 3—Where sieve analysis, including determination of material
finer than the 75-um sieve, is the only testing proposed, the size of the
sample may be reduced in the field to avoid shipping excessive
quantities of extra material to the laboratory.
6.3 Fine Aggregate—The test sample of fine aggregate
shall weigh, after drying, approximately the following
amount:
Aggregate with at least 95 % passing a 2.36-mm (No. 8) sieve 100 g
Aggregate with at least 85 % passing a 4.75-mm (No. 4) sieve 500 g
and more than 5 % retained on a 2.36-mm (No. 8) sieve
6.4 Coarse Aggregate—The weight of the test sample of
coarse aggregate shall conform with the following;
Nominal Maximum Size,
Square Openings, mm (in.)
9.5 (>/•)
12.5 0/2)
19.0(y<)
25.0(1)
37.5(1 'A)
50(2)
63 (2'A)
75(3)
100(4)
1 12 (4'/2)
125 (5)
150(6)
Minimum Weight
of Test Sample, kg (!b)
1(2)
2(4)
5(11)
10 (22)
15 (33)
20(44)
35(77) \
60(130)
100(220)
150 (330)
200 (440)
300 (660)
500(1100)
6.5 Coarse and Fine Aggregate Mixtures—The weight of
the test sample of coarse and fine aggregate mixtures shall be
the same as for coarse aggregate in 6.4.
6.6 The size of sample required for aggregates with large
nominal maximum size is such as to preclude testing except
with large mechanical sieve shakers. However, the intent of
this method will be satisfied for samples of aggregate larger
than 50 mm nominal maximum size if a smaller weight of
sample is used, provided that the criterion for acceptance or
rejection of the material is based on the average of results of
several samples, such that the sample size used times the
number of samples averaged equals the minimum weight of
sample shown in 6.4.
6.7 In the event that the amount of material finer than the
75-um (No. 200) sieve is to be determined by Test Method
C 117, proceed as follows:
6.7.1 For aggregates with a nominal maximum size of
12.5 mm (1/2 in.) or less, use the same test sample for testing
by Test Method C 117 and this method. First test the sample
in accordance with Test Method C 117 through the final
drying operation, then dry sieve the sample as stipulated in
7.2 through 7.7 of this method.
6.7.2 For aggregates with a nominal maximum size
greater than 12.5 mm (1/2 in.), a single test sample may be
used as described in 6.7.1, or separate test samples may be
used for Test Method C 117 and this method.
6.7.3 Where the specifications require determination of
the total amount of material finer than the 75-um sieve by
washing and dry sieving, use the procedure described in
6.7.1.
7. Procedure
7.1 Dry the sample to constant weight at a temperature of
110±5'C(230±9T).
NOTE 4—For control purposes, particularly where rapid results are
desired, it is generally not necessary to dry coarse aggregate for the sieve
analysis test The results are little affected by the moisture content
unless: (I) the nominal maximum size is smaller than about 12.5 mm
0/2 in.); (2) the coarse aggregate contains appreciable material finer than
4.75 mm (No. 4); or (3) the coarse aggregate is highly absorptive (a
lightweight aggregate, for example). Also, samples may be dried at the
higher temperatures associated with the use of hot plates without
aJTecting results, provided steam escapes without generating pressures
sufficient to fracture the particles, and temperatures are not so great as to
cause chemical breakdown of the aggregate.
7.2 Suitable sieve sizes shall be selected to furnish the
information required by the specifications covering the
material to be tested. The use of additional sieves may be
desirable to provide other information, such as fineness
modulus, or to regulate the amount of material on a sieve.
Nest the sieves in order of decreasing size of opening from
top to bottom and place the sample on the top sieve. Agitate
the sieves by hand or by mechanical apparatus for a
sufficient period, established by trial or checked by measure-
ment on the actual test sample, to meet the criterion for
adequacy or sieving described in 7.4.
7.3 Limit the quantity of material on a given sieve so that
all particles have opportunity to reach sieve openings a
number of times during the sieving operation. For sieves
with openings smaller than 4.75-mm (No. 4), the weight
retained on any sieve at the completion of the sieving
operation shall not exceed 6 kg/m2 (4 g/in.2) of sieving
surface. For sieves with openings 4.75 mm (No. 4) and
larger, the weight in kg/m2 of sieving surface shall not exceed
the product of 2.5 x (sieve opening in mm). In no case shall
the weight be so great as to cause permanent deformation of
the sieve cloth.
NOTE 5—The 6 kg/m2 amounts to 194 g for the usual 203-mm (8
in.) diameter sieve. The amount of material retained on a sieve may be
regulated by (1) the introduction of a sieve with larger openings
immediately above the given sieve or (2) testing the sample in a number
of increments.
7.4 Continue sieving for a sufficient period and in such
manner that, after completion, not more than 1 weight % of
the residue on any individual sieve will pass that sieve during
1 min of continuous hand sieving performed as follows:
Hold the individual sieve, provided with a snug-fining pan
and cover, in a slightly inclined position in one hand. Strike
the side of the sieve sharply and with an uoward motion
against the heel of the other hand at the rate of about 150
D-6
-------
C136
times per minute, turn the sieve abouf one sixth of a
revolution at intervals of about 25 strokes. In determining
sufficiency of sieving for sizes larger than the 4.75-mm (No.
4) sieve, limit the material on the sieve to a single layer of
particles. If the size of the mounted testing sieves makes the
described sieving motion impractical, use 203-mm (8 in.)
diameter sieves to verify the sufficiency of sieving.
7.5 In the case of coarse and fine aggregate mixtures, the
portion of the sample finer than the 4.75-mm (No. 4) sieve
may be distributed among two or more sets of sieves to
prevent overloading of individual sieves.
7.5.1 Alternatively, the portion finer than the 4.75-mm
(No. 4) sieve may be reduced in size using a mechanical
splitter according to Methods C 702. If this procedure is
followed, compute the weight of each size increment of the
original sample as follows:
W, _
tained, or percentages in various size fractions to the nearest
0.1 % on the basis of the total weight of the initial dry
sample. If the same test sample was first tested by Test
Method C 1 17, include the weight of material finer than the
75-um (No. 200) size by washing in the sieve analysis calcu-
lation; and use the total dry sample weight prior to washing
in Test Method C 1 1 7 as the basis for calculating all the
percentages.
8.2 Calculate the fineness modulus, when required, by
adding the total percentages of material in the sample that is
coarser than each of the following sieves (cumulative per-
centages retained), and dividing the sum by 100: 150-um
(No. 100), 300-um (No. 50), 600-um (No. 30), 1.18-mm
(No. 16), 2.36-mm (No. 8), 4.75-mm (No. 4), 9.5-mm
(3/8-in.), 19.0-mm (3/4-in.), 37.5-mm (I'/z-in.), and larger,
increasing in the ratio of 2 to 1.
where:
A = weight of size increment on total sample basis,
Wl = weight of fraction finer than 4.75-mm (No. 4) sieve in
total sample,
W2 = weight of reduced portion of material finer than
4.75-mm (No. 4) sieve actually sieved, and •
B = weight of size increment in reduced portion sieved.
7.6 Unless a mechanical sieve shaker is used, hand sieve
particles larger than 75 mm (3 in.) by determining the
smallest sieve opening through which each particle will pass.
Start the test on the smallest sieve to be used. Rotate the
particles, if necessary, in order to determine whether they
will pass through a particular opening; however, do not force
particles to pass through an opening.
7.7 Determine the weight of each size increment by
weighing on a scale or balance conforming to the require-
ments specified in 5.1 to the nearest 0.1 % of the total
original dry sample weight. The total weight of the material
after sieving should check closely with original weight of
sample placed on the sieves. If the amounts differ by more
than 0.3 %, based on the original dry sample weight, the
results should not be used for acceptance purposes.
7.8 If the sample has previously been tested by Test
Method C 117, add the weight finer than the 75-um (No.
200) sieve determined by that method to the weight passing
the 75-um (No. 200) sieve by dry sieving of the same sample •
in this method.
3. Calculation
8.1 Calculate percentages passing, total percentages re-
9. Report
9.1 Depending upon the form of the specifications for use
of the material under test, the report shall include the
following:
, 9.1.1 Total percentage of material passing each sieve, or
9.1.2 Total percentage of material retained on each sieve,
or
9.1.3 Percentage of material retained between consecutive
sieves.
9.2 Report percentages to the nearest whole number,
except if the percentage passing the 75-um (No. 200) sieve is
less than 10 %, it shall be reported to the nearest 0.1 %.
9.3 Report the fineness modulus, when required, to the
nearest 0.01.
10. Precision
10.1 The estimates of precision of this method listed in
Table 1 are based on results from the AASHTO Materials
Reference Laboratory Reference Sample Program, with
testing conducted by this method and AASHTO Method
T 27. While there are differences in the minimum weight of
the test sample required for other nominal maximum sizes of
aggregate, no differences entered into the testing to affect the
determination of these precision indices. The data are based
on the analyses of more than 100 paired test results from 40
to 100 laboratories. The values in the table are given for
different ranges of percentage of aggregate passing one sieve
and retained on the next finer sieve.
D-7
-------
C136
TABLE 1 Precision
Coarse Aggregates: °
Single-Operator
Precision
Multilaboratory
Precision
x
Fine Aggregates:
Single-Operator
Precision
. Multilaboratory
Precision
•* These numbers represent, respe(
9 These numbers represent, resp*
0 These values are frnm^f*- b8S8
T> of Size Fraction
Between Consecutive
Sieves
Oto3
3 to 10
10 to 20
20 to 50
Oto3
3 to 10
10 to 20
20 to 30
30 to 40
40 to 50
Oto3
3 to 10
10 to 20
20 to 30
30 to 40
40 to 50
Oto3
3 to 10
10 to 2tt
20 to 30
30 to 40
40 to 50
lively, the (1S) and (D2S) limit
lively, the (1S X) and (D2S %)
Coefficient of Standard
Variation Deviation
(1SX).Xa (1S),X*
30°
1.4°-
0.95
35°
1.06
1.66
2.01
2.44
3* o
.15
01A
, !*»
0.43
0.60
0.64
0.71
0.21
0.57
0.95-
1.24
1.41
s as described in Practice C 670.
limits as described in Practice C 670.
lominal maximum size of 19.0 mm (% in.).
Acceptable Range of Test Results
% of Avg (D2S)/ %
85°
i no
*».u
0 7
£./
3.9
99°
o n
O.O
c 7
3,1
6n
.y
9.0
0.4
If
./
Irt
.8
A.O
0.6
1C
.O
2.7
1 G,
0.3
A 1
4.0
M,,,,.,,
-------
-------
APPENDIX E
ASTM METHOD FOR THE LOS ANGELES ABRASTION TEST
E-1
-------
-------
Designation: C 131 - 89
Standard Test Method for
Resistance to Degradation of Small-Size Coarse Aggregate
by Abrasion and Impact in the Los Angeles Machine1
This standard is issued under the fixed designation C 131; the number immediately following the designation indicates the year of
original adoption or. in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapprovaL A
superscript epsilon (t) indicates an editorial change since the last revision or reapproval.
1. Scope
1.1 This test method covers a procedure for testing sizes of
coarse aggregate smaller than 1'/: in. (37.5 mm) for resist-
ance to degradation using the Los Angeles testing machine.
NOTE 1—A procedure for testing coarse aggregate larger than Vt in.
(19 mm) is covered in Test Method C 535.
2. Referenced Documents
2.1 ASTM Standards:
C 136- Method for Sieve Analysis of Fine and Coarse
Aggregates2
C 535 Test Method for Resistance to Degradation of
Large-Size Coarse Aggregate by Abrasion and Impact in
the Los Angeles Machine2
C 670 Practice for Preparing Precision and Bias State-
ments for Test Methods for Construction Materials2
C 702 Practice for Reducing Field Samples of Aggregate to
Testing Size2
D 75 Practice for Sampling Aggregates2
E 11 Specification for Wire-Cloth Sieves for Testing
Purposes3
3. Summary of Test Method
3.1 The Los Angeles test is a measure of degradation of
mineral aggregates of standard gradings resulting from a
combination of actions including abrasion or attrition,
impact, and grinding in a rotating steel drum containing a
specified number of steel spheres, the number depending
upon the grading of the test sample. As the drum rotates, a
shelf plate picks up the sample and the steel spheres, carrying
them around until they are dropped to the opposite side of
the drum, creating an impact-crushing effect. The contents
then roll within the drum with an abrading and grinding
action until the shelf plate impacts and the cycle is repeated.
After the prescribed number of revolutions, the contents are
removed from the drum and the aggregate portion is sieved
to measure the degradation as percent loss.
1 This test method is under the jurisdiction of ASTM Committee C-9 on
Concrete and Concrete Aggregates and is the direct responsibility of Subcommittee
C09.03.05 on Methods of Testing and Specifications for Physical Characteristics of
Concrete Aggregates.
Current edition approved June 15, 1989. Published June 1989. Originally
Published as C 131 - 37 T. Last previous edition C 131 - 81 (1987).
2 Annual Book of ASTM Standards, Vols 04.02 and 04.03.
3 Annual Book of ASTM Standards, Vol 14.02.
4. Significance and Use
4.1 The Los Angeles test has been widely used as an
indicator of the relative quality or competence of various
sources of aggregate having similar mineral compositions.
The results do not automatically permit valid comparisons to
be made between sources distinctly different in origin,
composition, or structure. Specification limits based on this
test should be assigned with extreme care in consideration of
available aggregate types and their performance history in
specific end uses.
5. Apparatus
5.1 Los Angeles Machine—The Los Angeles testing ma-
chine, conforming in all its essential characteristics to the
design shown in Fig. 1, shall be used. The machine shall
consist of a hollow steel cylinder, closed at both ends, having
an inside diameter of 28 ± 0.2 in. (711 ± 5 mm), and an
inside length of 20 ± 0.2 in. (508 ± 5 mm). The cylinder
shall be mounted on stub shafts attached to the ends of the
cylinder but not entering it, and shall be mounted in such a
manner that it may be rotated with the axis in a horizontal
position within a tolerance in slope of 1 in 100. An opening
in the cylinder shall be provided for the introduction of the
test sample. A suitable, dust-tight cover shall be provided for
the opening with means for bolting the cover in place. The
cover shall be so designed as to maintain the cylindrical
contour of the interior surface unless the shelf is so located
that the charge will not fall on the cover, or come in contact
with it during the test. A removable steel shelf extending the
full length of the cylinder and projecting inward 3.5 ± 0.1 in.
(89 ± 2 mm) shall be mounted on the interior cylindrical
surface of the cylinder, in such a way that a plane centered
between the large faces coincides with an axial plane. The
shelf shall be of such thickness and so mounted, by bolts or
other suitable means, as to be firm and rigid. The position of
the shelf shall be such that the distance from the shelf to the
opening, measured along the outside circumference of the
cylinder in the direction of rotation, shall be not less than 50
in. (1.27m).
NOTE 2—The use of a shelf of wear-resistant steel, rectangular in
cross section and mounted independently of the cover, is preferred.
However, a shelf consisting of a section of rolled angle, properly
mounted on the inside of the cover plate, may be used provided the
direction of rotation is such that the charge will be caught on the outside
face of the angle. If the shelf becomes distorted from its original shape to
such an extent that the requirements given in XI.2 of the Appendix to
this method are not met, the shelf shall either be repaired or replaced
before additional tests are made.
E-2
-------
C131
FILLER PLATE OF SAME
THICKNESS OF GASKET
-STEEL WALL 5 THICK
.',—GASKET
jfo.--FILLER PLATE THICKNESS
IVS =J + THICKNESS OF GASKET
3f » f» 20'
STEEL SHELF —
?" s PLATE COVER
ALTERNATE DESIGN
OF ANGLE SHELF
PREFERRED DESIGN -NOT LESS THAN 50'
OF PLATE SHELF AND COVER \ MEASURED ON
\ OUTSIDE OF DRUM
STEEL OR ROLLED STEEL ,-"". ^^
DIRECTION
ROTATION—
.SHAFT BEARING WILL BET^>- ! .-:*(
\M3UNTED ON CONCRETE/ "- = ~-~-~^-~' ' \
' PIERS OR OTHER RIGID/ CONCRETE PlER ,'\
SUP°ORTS
Metric Equivalents
in. 1/4
mm 6.4
Vi
12.7
1
25.4
3Vi
89
4
102
6
152
7V2
190
20
508
28
711
50
1270
FIG. 1 Los Angeles Testing Machine
5.1.1 The machine shall be so driven and so counterbal-
anced as to maintain a substantially uniform peripheral
speed (Note 3). If an angle is used as the shelf, the direction
of rotation shall be such that the charge is caught on the
outside surface of the angle.
NOTE 3—Back-lash or slip in the driving mechanism is very likely to
furnish test results which are not duplicated by other Los Angeles
machines producing constant peripheral speed.
5.2 Sieves, conforming to Specification Ell.
5.3 Balance—A balance or scale accurate within 0.1 % of
test load over the range required for this test.
5.4 Charge—The charge shall consist of steel spheres
averaging approximately F7/32 in. (46.8 mm) in diameter
and each weighing between 390 and 445 g.
5.4.1 The charge, depending upon the grading of the test
sample as described in Section 7. shall be as follows:
Grading
A
B
C
D
Number of
Spheres
12
11
Weight of
Charge, g
5000 ± 25
4584 ± 25
3330 ± 20
2500 ± 15
NOTE 4—Steel ball bearings l'Vi6 in. (46.0 mm) and 1V8 in. (47.6
mm) in diameter, weighing approximately 400 and 440 g each,
respectively, are readily available. Steel spheres PV32 in. (46.8 mm) in
diameter weighing approximately 420 g may also be obtainable. The
charge may consist of a mixture of these sizes conforming to the weight
tolerances of 5.4 and 5.4.1.
6. Sampling
6.1 The field sample shall be obtained in accordance with
Practice D 75 and reduced to test portion size in accordance
with Methods C 702.
7. Test Sample
7.1 The test sample shall be washed and oven-dried at 221
to 230T (105 to 110°C) to substantially constant weight
(Note 5), separated into individual size fractions, and recom-
bined to the grading of Table 1 most nearly corresponding to
the range of sizes in the aggregate as furnished for the work.
The weight of the sample prior to test shall be recorded to the
nearest 1 g.
8. Procedure
8.1 Place the test sample and the charge in the Los
Angeles^testing machine and rotate the machine at a speed of
30 to 33 rpm for 500 revolutions. After the prescribed
number of revolutions, discharge the material from the
E-3
-------
C131
TABLE 1 Gradings of Test Samples
Sieve Size (Square
Passing
37.5 mm (11/2 in.)
25.0 mm (1 in.)
19.0 mm PA in.)
12.5 mm (Vi in.)
9.5 mm (3/« in.)
6.3 mm (1A in.)
4.75-mm (No. 4)
Total
Openings)
25.0 mm (1 in.)
19.0 mm (%in.)
12.5 mm 0/2 in.)
9.5 mm (% in.)
6.3 mm (Vt in.)
4.75-mm (No. 4)
2.36-mm (No. 8)
Weight of Indicated Sizes, g
Grading
ABC
1 250 ± 25 •
1 250 ± 25
1 250 ± 10 2 500 ± 10
1 250 ± 10 2500 + 10
2500 + 10
2500 + 10
5 000 ± 10 5 000 ±10 5 000 ± 10
D
5000± 10
5 000 ± 10
machine and make a preliminary separation of the sample
on a sieve coarser than the 1.70-mm (No. 12). Sieve the finer
portion on a 1.70-mm sieve in a manner conforming to
Method C 136. Wash the material coarser than the 1.70-mm
sieve (Note 5), oven-dry at 221 to 230T (105 to 110'C) to
substantially constant weight, and weigh to the nearest 1 g
(Note 6).
NOTE 5—If the aggregate is essentially free of adherent coatings and
dust, the requirement for washing before and after test may be waived.
Elimination of washing after test will seldom reduce the measured loss
by more than about 0.2 % of the original sample weight.
NOTE 6—Valuable information concerning the uniformity of the
sample under test may be obtained by determining the loss after 100
revolutions. This loss should be determined without washing the
material coarser than the 1.70-mm sieve. The ratio of the loss after 100
revolutions to the loss after 500 revolutions should not greatly exceed
0.20 for material of uniform hardness. When this determination is
made, take care to avoid losing any part of the sample; return the entire
sample, including the dust of fracture, to the testing machine for the
final 400 revolutions required to complete the test.
9. Calculation
9.1 Express the loss (difference between the original
weight and the final weight of the test sample) as a
percentage of the original weight of the test sample. Report
this value as the percent loss.
NOTE 7—The percent loss determined by this method has no known
consistent relationship to the percent loss for the same material when
tested by Test Method C 535.
10. Precision
10.1 For nominal 19.0-mm (%-in.) maximum size coarse
aggregate with percent losses in the range of 10 to 45 %, the
multilaboratory coefficient of variation has been found to be
4.5 %.4 Therefore, results of two properly conducted tests
from two different laboratories on samples of the same
coarse aggregates should not differ from each other by more
than 12.7 %4 of their average. The single-operator coefficient
of variation has been found to be 2.0 %.4 Therefore, results
of two properly conducted tests by the same operator on the
same coarse aggregate should not differ from each other by
more than 5.7 % of their average.4
10.2 Bias—Since there is no accepted reference material
suitable for determining the bias for this procedure, no
statement on bias is being made.
'These numbers represent, respectively, the (IS%) and (D2S%) limits as
described in Practice C 670.
E-4
-------
C131
APPENDIX
(Nonmandatory Information)
XL MAINTENANCE OF SHELF
Xl.l The shelf of the Los Angeles machine is subject to
severe surface wear and impact. With use, the working sur-
face of the shelf is peened by the balls and tends to develop a
ndge of metal parallel to and about 1V* in. (32 mm) from the
junction of the shelf and the inner surface of the cylinder If
the shelf is made from a section of rolled angle, not only may
this ndge develop but the shelf itself may be bent longitudi-
nally or transversely from its proper position.
XI.2 The shelf should be inspected periodically to deter-
mine that it is not bent either lengthwise or from its normal
radial position with respect to the cylinder. If either condi-
tion is found, the shelf should be repaired or replaced before
further tests are made. The influence on the test result of the
ndge developed by peening of the working face of the shelf is
not known. However, for uniform test conditions it is
recommended that the ridge be ground off if its height
exceeds 0.1 in. (2 mm).
E-5
-------
APPENDIX F
STATE DERIVED AGGREGATE DURABILITY TESTS
F-1
-------
The following pages contain state-derived durability tests for construction
aggregtes based on AASHTO Method T-210 and ASTM Method D 3744. Test methods
derived by Washington, Main, and Alaska are provided.
F-2
-------
WASHINGTON DEGRADATION TEST PROCEDURE
Following is the procedure for performing the Washington
Degradation Test, Revised 1962, as modified by the Maine Depart-
ment of Transportation.
Equipment
Balance, minimum 800 gin capacity, sensitive to 0.1 gm
Sieve Shaker - Soiltest =CL300
Plastic Canister, 6" high 7%" diameter, Tupperware, with cover
'Sand Equivalent Cylinder
Sand Equivalent Stock Solution
Sieves - 1/2", 1/4", #10, £200
Graduate 500 ml, 10 ml
Interval Timer
Funnel, 9"
Squeeze Bottle, 500 ml
Test Method
The material to be tested shall pass the 1/2" sieve, be washed
over a #10 sieve and dried to constant weight. Make up sample
graded as follows:
1/2" - 1/4" 500 grams
1/4" - U.S. £10 500 grams
F-3
-------
Part I
Place sample in a 7 1/2" diameter x 6" high plastic canister
(tupperware), add 200 cc water, cover tightly, and place in a
Tyler Portable Sieve Shaker (Soiltest #CL-300, 305, suitably
motored to provide oscillation described below). *Run shaker for
twenty minutes at 215 oscillations per minute with a 2 1/8" throw
on the cam. At the conclusion of shaking time, empty the canister
into nested #10 and =±200 sieves, placed in a funnel over a 500 ml
graduate to catch all water. Wash out the canister and continue
to wash the aggregate with fresh water from squeeze bottle until the
graduate is filled to the 500 ml mark.
Caution: The aggregate may drain 50 - 100 ml of water after
washing has been stopped. Save all aggregate!
Pour 7 ml of sand equivalent stock solution (see AASHTO T-176-73)
into a sand equivalent cylinder. Bring all solids in the wash
water into suspension by capping the graduate with the palm of the
hand, then turning the cylinder upside down and right side up as
rapidly as possible about-ten times. Immediately pour the liquid
into the sand equivalent cylinder to the 15" mark.
Cap cylinder with stopper, hand or other suitable means, and
mix the contents of the sand equivalent cylinder by alternately
turning the cylinder upside down and right side up, allowing the
bubble to transverse completely from end to end. Repeat this
cycle twenty times as rapidly as possible.
*It is essential that the portable field sieve shaker be checked
against the Central Lab shaker by running degradation values on check
samples.
F-4
-------
At the conclusion of the mixing time, pleace the cylinder on
the table, remove the stopper and start the timer. After twenty
minutes read and record the height of the sediment column to the
nearest 0.1 inch.
Part II
Place the aggregate retained on the #10 and #200 sieves in
oven until dry, then sieve and record the weights retained on
U.S. #10 and #200 sieves. Loss through each sieve is determined
by subtraction from original weight, and recorded to nearest gram.
Calculations :
Calculate the degradation factor by the following formula:
D =
0.3 M.OOV^ + 0.7
f- 0.4H
+ 0. 6H
X 100
Where D = degradation factor
L
200 = grams lost through #200 sieve
L
10 = grams lost through #10 sieve
H = height of sediment in tube
This formula gives a weight of 30 percent to the ratio of the loss
through the #200 and #10 sieves, and 70 percent to the quality of
the fines as determined by the cleanness portion of the test.
Values will range from 0 to 100, with high values being best materials
F-5
-------
MAINE TEST METHOD FOR
DETERMINING AGGREGATE DEGRADATION
1. SCOPE:
A. This test method details a procedure for determining the
susceptibility of an aggregate to degrade and the quality
of the fines produced by self-abrasion in the presence
of water.
2. APPARATUS:
A. Balance - 800 gram capacity, sensitive to 0.1 gm.
B. Sieve shaker - Tyler portable model, +_ 1 3/4" throw on
cam at +.300 oscillations per minute.
C. Plastic Canister - 7 1/2" in diameter x 6" high ("Tupcerware" )
with cover.
D. Sand Equivalent Cylinder.
E. Sand Equivalent Stock Solution (AASHTO T-176-73).
F. Sieves 1/2", 1/4" - U.S. No. 10 and U.S. No. 200.
G. Graduates - 500 ml tall form, 10 ml.
H. Interval Timer.
I. Funnel, 9" .
J. Squeeze Bottle, 500 ml.
3. PROCEDURE:
A. Sieve the material to be tested through the 1/2" sieve,
wash over a No. 10 sieve and dry to constant weight.
B. 1000 g sample of the aggregate graded as follows:
1/2 in. - 1/4 in 500 g
1/4 in. - U.S. No. 10 500 g
F-6
-------
C. Place sample in the plastic canister, add 200 cc of water,
cover tightly and place in sieve shaker.
D. Agitate the material for 20 minutes.
E. Empty the canister into nested No. 10 and No. 200 sieves
placed in a funnel over a 500 ml graduate to catch all the
water.
F. Wash out the canister and continue to wash the aggregate
with fresh water from the squeeze bottle until the graduate
is filled to the 500 ml mark. (The aggregate may drain
50-100 ml of water after washing has been stopped).
G. Pour 7 ml of sand equivalent stock solution into a sand
equivalent cylinder.
H. Bring all solids in the graduate into suspension by capping
the graduate with the palm of the hand and turning it
upside down and back as rapidly as possible about 10 times.
I. Immediately decant into the sand equivalent cylinder to
the 15" mark and insert stopper in the cylinder, or cover
with hand.
J. Mix the contents of the cylinder by alternately turning
the cylinder upside down and right side up, allowing the
bubble to transverse from end to end. Repeat this cycle
20 times as rapidly as possible.
K. Place the cylinder on the table, remove stopper and start
timer. After 20 minutes read and record the height of
the sediment column to the nearest 0.1".
4. CALCULATIONS:
A. Calculate the degradation factor by the following formula;
DF = (15+1 75H) x 10° or use Table 1
where :
DF = Degradation Factor
K = Height of Sediment in Tube.
B. Values may range from 0 to 100, with high values being
best materials.
5. REPORTS:
A., All test rasults shall be reported on Form TL-83(1/31),
F-7
-------
Alaska Test Method T-13
Rev. 7-79
STANDARD METHOD OF TEST FOR
DEGRADATION OF AGGREGATES
A. SCOPE:
This method of test covers a procedure for determining the susceptibility of an aggregate to
degradation during agitation in water.
B. APPARATUS:
1. Balance — The balance shall be sensitive to 1.0 gram with a minimum capacity of 500 grams.
2. Sieves — The sieves shall conform to AASHTO M-92. Sizes 1/2-inch, 1/4-inch, No. 10 and No.
200 are required.
3. Sieve Shaker — The sieve shaker shall be a portable model suitably "motorized to provide
300+10 oscillations per minute with a 1 3/4-inch throw on the cam. It shall be calibrated
against the sieve shaker at the Regional Materials Laboratory before use.
4. - Plastic Canister — The plastic canister shall be "Tupper Ware", 7 1/2" in diameter and 6" in
height, having a flat bottom.
5. Miscellaneous — Standard Sand Equivalent Cylinder, Sand Equivalent Solution*, funnel with
9" mouth, ring and ring stand, polyethylene wash bottle, 500 ml. graduate, rubber or cork stop-
pers, 10 ml. capacity graduated cylinder, drying oven.
C. SAMPLE PREPARATION:
1. Crush a representative sample of the plus No. 4 material to be tested to pass the 1/2 inch
sieve. There will be some material that will pass through the crusher uncrushed.
2. Wash over a No. 10 sieve and dry to a constant weight.
- 3. Separate the sample into two sizes, 1/2-inch to 1/4-inch and 1/4-inch to No. 10, and weigh out
500 gms. of each size to the nearest 1.0 gm.
D. PROCEDURE:
1. Place both sample portions in the plastic canister, add 200 ml. of water, and cover tightly.
Place the canister in the shaker and run for 20 minutes.
2. Stack a No. 10 and No. 200 sieve inside the large funnel. Support the funnel with a ring and ring
stand over the 500 ml. graduate.
* Sand Equivalent Stock Solution to be supplied by the Central Material Laboratory
(7-79)
F-8
-------
Alaska Test Method T-13, Cont'd
3. Wash the contents of the canister over the No. 10 and No. 200 sieves and continue washing
until wash water has reached the 500 ml. mark on the graduate. Washing may be facilitated by
agitating the nest of sieves. In instances where highly degradable materials are encountered
and 'the sample cannot be washed clean with 500 ml. of water, continue washing using water
sparingly, until sample is washed clean. In no instance, however, shall the total volume of
water be greater than 1 000 ml. Allow the wash water to settle until clear, then siphon or pipette
the extra water to 500 ml. being careful not to:disturb the settled material.
4. Pour 7 ml. of stock sand equivalent solution into the sand equivalent cylinder. Bring all solids in
the wash water into suspension by clapping the graduate with the palm or a rubber stopper,
then turn the graduate upside down and right side up 10 times. Immediately fill the sand
equivalent cylinder to the 15" mark and insert rubber stopper in cylinder.
5. Mix the contents of the sand equivalent cylinder by alternately inverting and righting the
cylinder allowing the bubble to traverse from one end to the other and back again. This is one
cycle. Repeat this cycle 20 times as rapidly a£ possible.
6. Remove the stopper and place the cylinder on'the table. Allow the cylinder to set undisturbed
for 20 minutes, then read and record the height of the sediment to the nearest 0.1 inch.
E. CALCULATIONS:
1. Calculate the degradation factor by the following formula:
D= - = - (100)
15 + 1.75 H
Where: D = Degradation factor
H = Height of sediment in cylinder.
or use Table I.
2. Values may range from 0 to 100, with high values being more suitable materiai.
F. REPORTING RESULTS:
Use appropriate lab worksheet, and report date on form 25-229.
(7-79)
F-9
-------
Alaska Test Method T-13, Ccnt'd
TABLE I
DEGRADATION VALUE "D"
D =-
15 — H
15 +1.75H
(100)
H
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
2.3
2.4
2.5
2.6
2.7
2.8
2.9
3.0
D
98
96
95
93
91
90
88
87
85
84
82
81
79
78
77
75
74
73
71
70
69
68
67
66
65
63
62
61
60
59
H
3.1
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
4.9
5.0
5.1
5.2
5.3
6.4
5.5
4.5
5.7
5.8
5.9
6.0
D
58
57
56
55
54
54
53
52
51
50
49
48
48
47
46
45
44
44
43
42
41
41
40
39
39
38
37
37
36
35
H
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7.0
7.1
7.2
7.3
7.4
7.5
7.6
7.7
7.8
7.9
8.0
8.1
8.2
8.3
8.4
8.5
8.6
8.7
8.8
8.9
9.0
D
35
34
33
33
32
32
31
30
30
29
29
28
28
27
27
26
26
25
25
24
24
23
23
22
22
21
21
20
20
20
H
9.1
9.2
9.3
9.4
9.5
9.6
9.7
9.8
9.9
10.0
10.1
10.2 .
10.3
10.4
10.5
10.6
1.7
10.8
10.9
11.0
11.1
11.2
11.3
11.4
11.5
11.6
11.7
11.8
11.9
12.0
D •
19
19
18
18
17
17
17
16
16
15.
15
15
14
14
13
13
13
12
12
12
11;
1t
1V
id
10
10
9
9
9
8
H
12.1
12.2
12.3
12.4
12.5
12.6
12.7
12.8
12.9
13.1
13.1
13.2
13.3
13.4
13.5
13.6
13.7
13.8.
13.9
14.0
14.1
14.2
14.3
14.4
14.5
14.6
14.7
14.8
14.9
15.0
D
8
8
7
7
7
6
6
6
6
5
5
5
4
4
4
4
3
3
3
3
2
2
2
1
1
1
1
0
0
0
F. REPORTING RESULTS:
Report data using form 25-229
F-10
(7-79)
-------
APPENDIX G
ASTM METHOD E 384-89 FOR VICKERS HARDNESS
G-1
-------
-------
Designation: £ 3i4 - 69
Standard Test Method for
Microhardness of Materials1
This standard is issued under the fixed designation E 384: the number immedialelv following the designation indicates the vear of
original adopt.on or, m the case of rev,sion. the year of last revision. A number in parentheses indicates the year oflast reapproval A
superscript cpsilon (<) indicates an editorial change since the last revision or reapproval.
1. Scope
1.1 This test method covers determination of the micro-
hardness of materials, the verification of microhardness
testing machines, and the calibration of standardized test
blocks. Appendix XI provides guidance in establishing
correlation of microhardness tests when two or more labora-
tories are involves.
1.2 This standard may involve hazardous materials, oper-
ations, and equipment. This standard does not purport to
address all of the safety problems associated with its use. It is
the responsibility of the user of this standard to establish
appropriate safety and health practices and determine the
applicability of regulatory limitations prior to use.
2. Referenced Documents
2.1 ASTM Standards:
E 3 Methods of Preparation of Metallographic Specimens2
E 122 Practice for Choice of Sample Size to Estimate the
Average Quality of a Lot or Process3
E 175 Definitions of Terms Relating to Microscopy4
E69] Practice for Conducting an Imerlaboratory Test
Program to Determine the Precision of Test Methods3
A. TEST PROCEDURE
3. Scope
3.1 This procedure covers tests made with the Knoop and
Vickers indenters under test loads in the range from 1 to
1000 gf.
NOTE 1—While Committee E-4 is primarily concerned with metals.
the test procedures described are applicable to all types of materials, and
not merely to metals.
4. Terminology
4.1 Definitions:
4.1.1 microhardness test—a microindentation hardness
test using a calibrated machine to force a diamond indenter
of specific geometry, under a test load of 1 to 1000 gf. into
the surface of the test material and to measure the diagonal
or diagonals optically.
4.1.2 Knoop hardness number (//A')—the number ob-
tained by dividing the applied load in kilograms-force by the
projected area of the indentation in square millimetres.
' This tcsl method is under the jurisdiction of ASTM Committee E-4 on
Metallography and is the direct responsibility of Subcommittee E04.05 on
Micrchardness.
Current edition approved Oct. I". 1989. Published April 1990 Originally
published as E 384 - 69 Las: previous edition E 384 - 84. "
; Annual Book of ASTM Standard*. Vol 03.01.
"' Aniihu! Book 01 ASTM Standards. Vol 14.02.
- Annual Book of ASTM Standards. Vo! 14.01.
computed from the measurement of the long diagonal of the
indentation. It is assumed that the indentation is an imprint
of the undeformed indemer (see Fig. 1).
Discussion—The Knoop hardness number is computed
from the following equation:
HK = P/Ap = P/d-c = P/Q.01028d- = 14.229P/d2 (1 >
where:
P = load, kgf,
IP = projected area of indentation, mm2,
d - length of long diagonal, mm. and -
c = indenter constant relating projected area of the inden-
tation to the square of the length of the long diagonal.
Discussion—Since the units normally used are grams-
force and micrometres rather than kilograms-force and
millimetres, the equation for the Knoop hardness number
can be expressed conveniently as:
A
HK = 14229/Va',2
(2)
where:
.P, = load, gf, and
d\ = length of long diagonal urn.
Discussion—The Knoop hardness numbers are given in
Table 1 for a test load of 1 gf. For obtaining hardness
numbers when other test loads are used, the Knoop hardness
number, obtained from Table 1. is multiplied by the test load
in grams-force.
4.1.3 Vickers hardness number (HV)—the number ob-
tained by dividing the applied load in kilograms-force by the
surface area of the indentation in square millimetres com-
puted from the mean of the measured diagonals of the
indentation. It is assumed that the indentation is an impnni
of the undeformed indenter (Fig. 2).
Discussion—The Vickers hardness number is computed
from the following equation:
HV = P/AS = 2P sm(al2)/d- = 1.8544P/V2 (>•
where:
P = load, kgf,
As = surface area of indentation, mm2.
d = mean diagonal of indentation, mm. and
a = face angle of indenter = 136°.
Discussion—Since the units normally used are grams-
force and micrometres rather than kilograms-force and
millimetres, the equation for the Vickers hardness number
can be expressed conveniently as:
HV= 1854.4/'l/c'l: (-!•
where:
P, = load, gf, and
d-t = mean diagonal of indentation, urn.
Discussion—The Vickers hardness numbers are given ir.
Table 2 for a test load of 1 gf. For obtaimnc hardness
G-2
-------
E384
numbers when other test loads are used, the Vickers hardness
number obtained from Table 2 is multiplied by the test load
in grams-force.
-i. 1A verification—checking or testing to assure conform-
ance with the specification.
4.1.5 calibration—determination of the values of the
significant parameters by comparison with values indicated
bv a reference instrument or bv a set of reference standards.
5. Apparatus
5.1 Testing Machine—Equipment for microhardness
testing usually consists of a testing machine that supports the
specimen and permits the indenter and specimen to be
brought into contact gradually and smoothly under a prede-
termined load. The design of the machine should be such
that no rocking or lateral movement of the indenter or
specimen is permitted while the load is being applied or
removed. A measuring microscope is usually mounted on
the machine in such a manner that the indentation may be
readily located in the field of view.
5.1.1 Load Application:
5.1.1.1 The plane of the surface of the specimen shall be
perpendicular to the axis of the direction of the load
application.
5.1.1.2 The indenter shall contact the specimen at a
velocity in the.range from 15 to 70-u,m/s.
5.1.1.3 The time of application of the full test load shall be
10 to 15s unless otherwise specified.
5.1.2 Vibration Control:
5.1.2.1 The microhardness testing machine shall be lo-
cated in an area as free from vibrations as possible in order to
avoid erroneous results. When test loads of 100 gf to less
than 500 gf are employed, the machine shall be isolated in
such a manner that vibrational accelerations of 0.005 g (2
m./s:) or larger are not applied to the machine frame. For
test loads less than 100 gf the maximum allowable vibra-
tional accelerations inadvertently reaching the machine
should be held proportionately lower than the 0.005 g limit.
5.1.2.2 Operators of microhardness testing machines in
which the indenter contacts and leaves the work surface by
manual control should take care to perform all operations
slowly and smoothly.
5.1.2.3 At no time during the period of indenter contact
should the operator inadvertently contact the machine frame
so as to subject it to vibrational accelerations in excess of
0.005 g.
5.2 Indenters:
5.2.1 Knoop Indenter—A highly polished, pointed,
rhombic-based, pyramidal diamond with included longitu-
dinal edge angles of 172° 30 min (±5 min) and 130° 0 min.
The indenter constant, c, shall differ from 0.07028 by not
more than 1 %.
5.2.1.1 The four faces of the indenter shall be equally
inclined to the axis of the indenter (within ±30 min) and
shall meet at a sharp point. The line of junction between
opposite faces (offset) shall be not more than 1.0 um in
length for indentations greater than 15 um in length, as
shown in Fig. 3. For shorter indentations the offset should be
proportionately less.
5.2.1.2 The diamond shall be examined periodically; and
if it is loose in the mounting material, chipped or cracked, it
should be replaced.
5.2.2 Dickers Indenter—A highly polished, pointed.
square-based pyramidal diamond with face angles of 136° 0
min (=30 min).
5.2.2.1 The four faces of the indenter shall be equally
inclined to the axis of the indenter (within ±30 min) and
shall meet at a sharp point. The line of junction between
opposite faces (offset) shall be not more than 0.5 um in
length, as shown in Fig. 4.
5.2.2.2 The diamond should be examined periodically:
and if it is loose in the mounting material, chipped or
cracked, it should be replaced.
5.3 Measuring Equipment—The measuring microscope at
highest magnification shall be graduated in 0.5-um or
smaller divisions.
5.3.1 Microscope Objectives and Ocular—The optical
portion of the measuring equipment shall have Kohler or
Abbe-Nelson illumination (Annex Al).*
5.3.1.1 For maximum resolution, provisions shall be
made for the adjustment of the illumination intensity and
concentration, the aperture diaphragm, and the field dia-
phragm. The ocular shall be of the filar micrometer type as
defined in Definitions E 175. The objective shall be of a type
for use without a cover glass.
5.3.1.2 Magnification for range of indentation length is
indicated as follows:
Magnification
Indentation Length, urn
Less than 76
"6 to 125
Greater than 125
800
600
mm
400
300
200
5.3.1.3 Illumination—Proper illumination is necessary in
order to obtain optimum resolution from a microscope.
There are two systems which give proper illumination.
Abbe-Nelson or "critical" is the system in which the image of
the illuminating source is focused in the plane of the
specimen. Kohler illumination is the system in which the
illuminating source is imaged at the near focal plane of the
objective lens. (Abbe-Nelson and Kohler illumination ad-
justment techniques are given in Annex Al.)
6. Specimen Mounting
6.1 Mounting is recommended for convenience in surface
preparation and testing. The specimen should be adequately
supported in the mounting medium.
6.2 Where edge retention is not essential, the specimen
may or may not be mounted. If mounted, a softer mounting
material may be used. However, the material must have
sufficient rigidity so that no movement of the specimen can
occur during the application of load.
7. Test Specimens
7.1 Surface Preparation:
7.1.1 Degree of Finish—The degree of surface finish
required can vary with the several loads and magnifications
used in microhardness testing. However, in all tests the
perimeter of the indentation must be clearly defined in the
field of the microscope.
7.1.2 Grinding and Polishing—When grinding or pol-
ishing, or both is necessary in specimen preparation, care
G-3
-------
E384
should be taken to minimize heating and distorting the
specimen surface. Polishing should be performed according
to the procedures outlined in Methods E 3.
8. Procedure
8.1 Support the specimen so that the test surface is normal
to the axis of the indenter and the ends of the diagonals are
clearly defined (see 8.4).
8.2 Space the indentations so that the distance between
any two indentations is greater than twice the extent of any
stress deformation (cold working, "butterfly" fractures, etc)
that may occur as a result of the indentation process so that
there shall be no overlap of the deformation between two
indentations.
8.3 Adjust optical equipment (Section 5).
8.4 Measure Indentation—Having completed the adjust-
ment of the optical system, when applicable, and having
selected the correct objective, bring the ends of the diago-
nal^) into sharp focus. Examine the indentation for sym-
metry prior to measurement.
8.4.1 Knoop Indentation—If one leg of the long diagonal
is more than 20 % longer than the other, or if the ends of the
diagonal are not both in the field of focus, the surface of the
specimen may not be normal to the axis of the indenter.
Align the specimen surface properly, and make another
indentation.
8.4.2 Vickers Indentation—If one leg of a diagonal is
noticeably longer than the other leg of the same diagonal.
resulting in a deformed indentation, misalignment is prob-
ably present and should be corrected before proceeding with
any measurements.
8.4.3 To determine the source of misalignment, rotate the
specimen 90 or 180° and make a second indentation. If the
deformed shape of the second indentation has also rotated,
the surface is at fault and shall be properly aligned. If the
deformed shape has not rotated, examine the indenter seat
for din or nicks which may cause misalignment of the
indenter.
NOTE 2—It should be pointed out that there are a few materials that
may show deformed indentations even though the specimen and
indenter are properly aligned. Therefore, in the situation where realign-
ment has not corrected the problem, use another material known to give
symmetrical indentations to check the alignment of the setup.
8.4.4 Make any micrometer screw motion required for the
measurement of the indentation in the same direction to
eliminate errors due to the backlash of the screw thread.
8.4.5 Bring the edge of the ocular or graduation line just
into contact with the end of the diagonal. This is particularly
important since the precision of the measurement of the
diagonal is dependent upon the initial positioning of the
graduation line. To eliminate the influence of the thickness
of the line, always use the same edge of the graduation line.
8.4.6 Knoop Indentation—Read the long diagonal of the
indentation to within 0.25 urn or 0.4 %, whichever is larger.
8.4.7 Vickers Indentation—Read the two diagonals of the
indentation to within 0.25 urn or 0.4 %. whichever is larger,
and determine the average of the diagonal lengths.
8.5 Determination of Hardness Numbers:
8.5.1 Compute the Knoop hardness number from the
equation given in 4.2.1.1 or from the information given in
Table 1. To obtain the hardness number from the table, read
the HK (1 gf) corresponding to the measured diagonal length
in micrometres and multiply by the test load in grams-force.
8.5.2 Compute the Vickers hardness number from the
equation given in 4.3.1.1 or from the information given in
Table 2. To obtain the hardness number from the table, read
the HV (1 gf) corresponding to the average of the measured
diagonal lengths in micrometres and multiply by the test
load in grams-force.
9. Precision and Bias5
9.1 The precision and bias of microhardness measure-
ments depend on strict adherence to the stated test procedure
and are influenced by instrumental and material factors and
indentation measurement errors.
9.2 The consistency of agreement for repeated tests on the
same material is dependent on the homogeneity of the
material, reproducibility of the hardness tester, and consis-
tent, careful measurement of the indents by a competent
operator.
9.3 Instrumental factors that can affect test results in-
clude: accuracy of loading, inertia effects, speed of loading.
vibrations, the angle of indentation, lateral movement of the
indenter or sample, indentation and indenter shape devia-
tions.
9.3.1 Vibrations during indenting will produce larger
indents with the influence of vibrations becoming larger as
the load decreases (1, 2).6
9.3.2 The angle between the indenter and specimen sur-
face should be within 2° perpendicular. Greater amounts of
tilting produces nonuniform indentations and invalid test
results.
9.4 Material factors that can affect test results include:
sample homogeneity, orientation/texture effects, improper
specimen preparation, low specimen surface reflectivity, and
transparency of the specimen.
9.4.1 Residual deformation from mechanical polishing
must be removed, particularly for low-load testing.
9.4.2 Distortion of the indent shape due to either crystai-
lographic or microstructural texture influences diagonal
lengths and the validity of the calculated hardness.
9.4.3 Metal deformation during indentation can produce
ridging around the indent periphery that will affect diagonal
measurement accuracy.
9.4.4 Testing of etched surfaces, depending on the extern
of etching, can produce results that are different from those
obtained on unetched surfaces (1).
9.5 Measurement errors that can affect test results in-
clude: inaccurate calibration of the measuring device, inade-
quate resolving power of the objective, insufficient magnifi-
cation, operator bias in sizing the indents, poor image
quality, and nonuniform illumination.
9.5.1 The accuracy of microindentation hardness testing
is strongly influenced by the accuracy to which the indents
can be measured.
9.5.2 The error in measuring the diagonals increases a>
5 Supporting daia have been filed al ASTM Headquarters and may be obtains
b\ requesting RR: E-04.
* The boldface numbers in parentheses refer to the list of references al the •:'•'
of this standard.
G-4
-------
E384
;he numerical aperture of the measuring objective decreases
13. 4).
9.5.3 Bias is introduced if the operator consistently
undersizes or oversizes the indents.
9.6 Some of the factors th.at affect test results produce
systematic errors that influence all test results while others
primarily influence low-load test results. Some of these
problems occur continually, others may occur in an unde-
fined, sporadic manner. Low load hardness tests are influ-
enced by these factors to a greater extent than high load tests.
9.6.1 Due to the relationship between the diagonal length
and the calculated Knoop or Vickers hardness (see Eq 2 and
-l). measurement errors of the same magnitude cause propor-
tionally greater hardness variability as the indent size de-
jreases.
9.7 An interlaboratory test program was conducted in
accordance with Practice E 691 to develop information
regarding the precision, repeatability, and reproducibility of
the measurement of Knoop and Vickers indentations made
using loads of 25, 50. 100, 200. 500. and 1000 gf on three
ferrous and four nonferrous samples.5 Twelve laboratories
measured the indents, five of each type at each load on each
-ample. Additional details of this study are given in Ap-
pendix X2.
9.7.1 Tests of the three ferrous samples revealed that nine
aboratories produced similar measurements while two labo-
•atories consistently undersized the indents and one labora-
tory consistently oversized the indents. These latter results •
.vere most pronounced as the load decreased and sample
hardness increased (that is. as the diagonal size decreased)
and were observed for both Vickers and Knoop indents.
Results for the lower hardness nonferrous indents produced
->etter agreement. However, none of the laboratories that
obtained higher or lower results on the ferrous samples
measured the nonferrous indents.
9.7.2 Repeatability Interval—The difference due to test
error between two test results in the same laboratory on the
^ame material increases with increasing specimen hardness
and with decreasing test load (see X2.5.4).
9.7.3 Reproducibility Interval—The difference in test re-
sults on the same material tested in different laboratories
increased with increasing specimen hardness and with de-
creasing test load (see X2.5.5).
9.7.4 The within-laboratory and between-laboratory pre-
:ision values improved as specimen hardness decreased and
test load increased. The repeatability interval and reproduc-
ihility interval were generally larger than the precision
estimate, particularly at low test loads and high sample
hardnesses.
10. Conversion to Other Hardness Scales or Tensile-
Strength Values
10.1 There is no generally accepted method for accurate
conversion of Knoop or Vickers hardness numbers to other
hardness scales or tensile-strength values. Such conversions
are limited in scope and should be used with caution, except
'or special cases where a reliable basis for the conversion has
ieen obtained by comparison tests.
11. Report
11.1 The report shall include the following:
11.1.1 Hardness number,
11.1.2 Test load,
11.1.3 Time of full load application if other than 10 to 15
s.
11.1.4 Magnification, and
11.1.5 Any unusual conditions encountered during the
test.
11.2 The hardness number shall be reported as HK for
Knoop hardness or HV for Vickers hardness. The load shall
be reported in grams-force by subscript notation following
the HK or HV, for example. 400 HK100.
B. VERIFICATION OR MICROHARDNESS TESTING
MACHINES
12. Scope
12.1 Part B covers two procedures for the verification of
microhardness testing machines and a procedure that is
recommended for use to confirm that the machine has not
become maladjusted in the intervals between periodic rou-
tine checks. The two methods of verification are:
12.1.1 Separate verification of load, indenter. and mea-
suring microscope. This procedure is mandatory only for
new and rebuilt machines and is the responsibility of the
manufacturer.
12.1.2 Verification by standardized test block method.
This procedure shall be used for verifying machines in
service.
13. General Requirements
13.T Before a microhardness testing machine is verified
the machine shall be examined to ensure that:
13.1.1 The machine is properly set up, and
13.1.2 The load can be applied and removed without
shock or vibration in such a manner that the readings are not
influenced.
14. Verification
14.1 Separate Verification of Load. Indenter, and Mea-
suring Microscope—This procedure is mandatory for new
and rebuilt machines and is the responsibility of the manu-
facturer.
14.1.1 Load—When the load is applied by a lever system
with dead weights, the combined accuracy of lever distances
and dead weights shall be ±0.2 % at the indenter. When the
load is applied by direct loading with dead weights, the
weights shall be accurate to ±0.2 %.
14.1.2 Indenter—The form of the diamond indenter shall
be verified by direct measurement of its shape or by
measurement of its projection on a screen.
14.1.2.1 For laboratory or routine tests the indenter shall
meet the requirements of 5.2.
14.1.2.2 Knoop diamond indenters used for calibrating
standardized hardness test blocks shall have included longi-
tudinal edge angles of 172°. 30 min (±5 min) and 130°, 0
min (±5 min), and the offset shall not be more than 0.5 urn.
14.1.2.3 Vickers diamond indenters used for calibrating
standardized hardness test blocks shall have face angles of
136°, 0 min (±15 min), and the offset shall not be more than
0.25 urn.
14.1.3 Measuring Microscope—The measuring micro-
scope or other device for measuring the diagonals of the
G-5
-------
E384
indentation shall be calibrated against an accurately ruled
line scale (stage micrometer). The errors of the line sca~le shall
not exceed 0.05 urn or 0.05 % of any interval, whichever is
greater. The measuring microscope shall be calibrated in the
range of its use, and a calibration factor shall be chosen such
that the error shall not exceed ±0.5 %.
14.2 Verification by Standardized Test Block Method:
14.2.1 A microhardness testing machine may be checked
by making a series of indentations on standardized hardness
test blocks.
14.2.2 Make a minimum of five indentations on a test
block using a test load applied for 10 to 15 s.
14.2.3 Consider the microhardness testing machines veri-
fied if the mean diagonal for five indentations meets the
requirements of Section 15.
14.2.4 Verification with standardized test blocks is not
recommended if the combination of hardness and test load
results in indentations having diagonals less than 20 urn
because, under such conditions, the error in the standard
micrometer microscope can represent a significant per-
centage of the diagonal length. Under these circumstances
this method of verification does not offer a reliable means of
verifying the equipment.
15. Repeatability and Error
15.1 Repeatability:
15.1.1 For each standardized block, let d}. d2 d5 be
the diagonal lengths of the indentations, arranged in' in-
creasing order of magnitude.
15.1.2 The repeatability of the machine is expressed by
the quantity d$ — d}.
15.1.3 The repeatability is considered satisfactory if it
meets the condition given in Table 3
15.2 Error:
15.2.1 The error of the machine is expressed by the
quantity 3, - d. and 3 = (d, + d2 .. . ds)/5. and d, is the
reported mean diagonal length of the standard indentations.
15.2.2 The mean diagonal. 3. shall not differ from the
mean diagonal. ds. by more than 2 % or 0.5 urn. whichever is
greater.
C. CALIBRATION OF STANDARDIZED HARDNESS TEST
BLOCKS FOR MICROHARDNESS MACHINES
16. Scope
16.1 Part C covers the calibration of standardized hard-
ness test blocks for the verification of microhardness testing
machines as described in part B.
17. Manufacture
17.1 Each metal block to be standardized shall be not less
than Vi in. (6 mm) in thickness.
17.2 Each block shall be prepared to give the necessary
repeatability.
17.3 Each block, if ferromagnetic, shall be demagnetized
by the manufacturer and maintained in that condition by the
user.
17.4 The supporting surface of the test block shall have a
fine ground finish and the block shall be parallel to the test
surface to ±0.0005 in./in. or-±l min 40 s angular tolerance.
17.5 The test surface shall be polished and free from
scratches that would interfere with measurements of the
diagonals of the indentations.
17.5.1 The mean surface-roughness-height rating shall not
exceed 4 uin. (0.001 u.m), center-line average.
18. Standardizing Procedure
18.1 Calibrate the standardized hardness test blocks on a
microhardness testing machine verified in accordance with
the requirements of 13.1.
18.2 The mechanism that controls the application of load
shall employ a device to maintain a constant velocity of the
indenter.
18.3 Apply the full load for 10 to 15 s.
18.4 Make at least four groups, consisting of five indenta-
tions each, on each test block.
18.5 Adjust the illuminating system of the measuring
microscope to give uniform intensity over the field "of view
and maximum contrast between the indentation and the
undisturbed surface of the block (see 5.3.1.3 and Annex A1).
18.6 Check the measuring microscope with a stage mi-
crometer, or by other suitable means, to ensure that the
difference between readings corresponding to any two divi-
sions of the instrument is correct within ±0.5 urn.
19. Repeatability
19.1 Lei d}, d2 dn be the mean values of the
measured diagonals as determined by one observer, arranged
in increasing order of magnitude.
19.2 The repeatability of the hardness readings on the
blocks is expressed as (dw - d,). when ten readings have
been made, or 1.32 (d$ - <•/,) when five readings are-taken on
the block.
19.3 Unless the repeatability of hardness readings as
measured by the mean diagonals of five or ten indentations
is within the limits given in Table 4. the block cannot be
regarded as sufficiently uniform for standardization pur-
poses.
20. Marking
20.1 Each block shall be marked with an appropriate
identifying serial number.
21. Certification
21.1 The certificate accompanying each standardized
hardness test block shall include:
21.1.1 Arithmetic mean of the hardness values found in
the standardization test followed by the tolerance range for
each group of five indentations.
21.1.2 Test load.
21.1.3 Serial number of test block.
21.1.4 Name of the certifying organization, and
21.1.5 Magnification.
G-5
-------
E384
FIG. 1 Knoop Indenter
OFFSET
I.O/im. max.
FIG. 3 Knoop Indenter Offset
-I3S'
FIG. 2 Vickers Indenter"
OFFSET
0.5/lm ma*.
FIG. 4 Vickers Indenter Offset
G-"
-------
E384
TABLE 1 Knoop Hardness Numbers for Load of 1 gf
Diagonal o
Impression
\im
1
2
3
4
5
e
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
47
48
49
50
51
52
53
54
55
56
57
58
59
60
f
0.0
14230
3557
1581
889.3
569.2
395.2
290.4
222.3
175.7
142.3
117.6
98.81
84.20
72.60
63.24
55.58
49.24
43.92
39.42
35.57
32.27
29.40
26.90
24.70
22.77
21 .05
19.52
18.15
16.92
15.81
14.81
13.90
13.07
12.31
11.62
10.98
10.39
9.854
9.355
8.893
8465
8.066
7.695
7.350
7.027
6.724
6.441
6.176
5.926
5.692
5.471
5.262
5.065
4.880
4.704
4.537
4.379
4.230
4.088
3.952
Knoop Hardness Number for Diagonal Measured to 0.1 (im
0.1
11760
3227
1481
846.5
547.1
382.4
282.3
216.9
171.8
139.5
115.5
97.19
82.91
71.57
62.40
54.89
48.66
43.43
39.00
35.22
31.96
29.13
26.67
24.50
22.59
20.89
19.37
18.02
16.80
15.71
14.71
13.81
12.99
12.24
11.55
10.92
10.34
9.802
9.307
8.849
8.423
8.028
7.660
7.316
6.996
6.695
6.414
6.150
5.902
5.669
5.449
5.242
5.046
4.082
4.687
4.521
4.364
4.215
4.074
3.939
0.2
9881
2940
1390
806.6
526.2
370.2
274.5
211.6
168.1
136.8
113.4
95.60
81.66
70.57
61.59
54.22
48.10
42.96
38.60
34.87
31.66
28.87
26.44
24.30
22.41
20.73
19.23
17.23
16.89
15.60
14.62
13.72
12.91
12.17
11.48
10.86
10.28
9.751
9.260
8.805
8.383
7.990
7.624
7.283
6.965
6.666
6.387
6.125
5.878
5.646
5.428
5.222
5.027
4.844
4.670
4.505
4.349
4.201
4.060
3.926
0.3
8420
2690
1307
769.5
506.2
358.5
267.0
206.5
164.5
134.1
111.4
94.05
80.44
69.58
60.78
53.55
47.54
42.49
•38.20
34.53
31.36
28.61
26.21
24.10
22.23
20.57
19.09
17.77
16.57
15.60
14.52
13.64
12.83
12.09
11.42
10.80
10.23
9.700
9.213
8.761
8.342
7.952
7.589
7.250
6.934
6.638
6.360
6.099
5.854
5.624
5.407
5.202
5.009
4.826
4.653
4.489
4.334
4.188
4.046
3.913
0.4
7260
2470
1231
735.0
488.0
347.4
259.8
201.7
161.0
131.6
109.5
92.54
79.24
68.62
60.00
52.90
47.00
42.03
37.81
34.19
31.07
28.36
25.99
23.90
22.05
20.42
18.95
17.64
16.46
15.40
14.43
13.55
12.75
12.02
11.35
10.74
10.17
9.650
9.166
8.718
8.302
7.915
7.554
7.218
6.903
6.609
6.333
6.074
5.831
5.602
5.386
5.182
4.990
4.808
4.636
4.473
4.319
4.172
4.033
3.900
0.5
6324
2277
1162
702.7
470.4
336.8
253.0
196.9
157.7
129.1
107.6
91 .07 .
78.07
67.68
59.23
52.36
46.46
41.57
37.42
33.86
30.78
28.11
25.77
23.71
21.88
20.26
18.82
17.52
16.35
15.30
14.34
13.47
12.68
11.95
11.29 ;
10.68
10.12
9.600
9.120
8.675
8.262
7.878
7.520
7.185
6.873
6.581
6.306
6.049
5.807
5.579
5.365
5.162
4.971
4.790
4.619
4.457
4.304
4.158
4.019
3.887
0.6
5558
2105
1098
672.4
453.7
326.7
246.3
192.4
154.4
126.6
105.7
89.63
76.93
66.75
58.47
51.64
45.94
41.13
37.04
33.53
30.50
27.86
25.55
23.51
21.71
20.11
18.68
17.40
16.24
15.20
14.25
13.39
12.60
11.89
11.23
10.52
10.06
9.550
9.074
8.632
8.222
7.841
7.485
7.153
6.843
6.552
6.280
6.024
5.784
5.557
5.344
5.143
4.953
4.773
4.603
4.442
4.289
4.144
4.006
3.875
0.7
4924
1952
1039
644.1
437.9
317.0
240.0
188.0
151.2
124.3
103.9
88.22
75.81
65.85
57.73
51.02
45.42
40.69
36.66
33.21
30.22
27.61
25.33
23.32
21.54
19.96
18.54
17.27
16.13
15.10
14.16
13.31
12.53
11.82
11.16
10.56
10.01
9.501
9.028
8.590
8.183
7.804
7.451
7.121
6.813
6.524
6.254
6.000
5.761
5.536
5.323
5.123
4.934
4.756
4.586
4.426
4.274
4.129
3.992
3.362
0.8
4392
1815
985.4
617.6
423.0
307.7
233.9
183.7
148.2
122.0
102.2
86.85
74.72
64.96
57.00
50.41
44.91
40.26
36.29
32.89
29.94
27.37
25.12
23.14
21.38
19.81
18.41
17.15
16,01
15.00
14.07
13.23
12.45
11.75
11.10
10.51
9.958
9.452
8.983
8.548
8.144
7.768
7.417
7.090
6.783
6.497
6.228
5.975
5.737
5.514
5.303
5.104
4.916
4.738
4.570
4.410
4.259
4.115
3.070
3 849
0.9
3942
1692
935.5
592.6
408.8
298.9
228.0
179.6
145.2
119.8
100.5
85.51
73.65
64.09
56.28
49.82
44.41
39.83
35.93
32.57
29.67
27.13
24.91
22.95
21.21
19.66
18.28
17.04
15.92
14.90
13.98
13.15
12.38
11.68
11.04
10.45
9.906
9 403
8.938
8.506
8.105
7.731
7.383
7.058
6.754
6469
6.202
5.951
5.714
5492
5.282
5.085
4.898
4.721
4.554
4.395
. 4,244
4.102
3.966
3.837
G-8
-------
W E 384
TABLE 1 Continued
Diagonal of
Impression.
nm
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
Knoop Hardness Number for Diagonal Measured to 0.1 urn
0.0
3.824
3.702
3.585
3.474
3.368
3.267
3.170
3.077
2.989
2.904
2.823
2.745
2.670
2.598
2.530
2.463
2.400
2.339
2.280
2.223
2.169
2.116
2.065
2.017
1.969
1.924
1.880
1.837
1.796
1.757
1.718
1.681
1.654
1.610
1.577
1.544
1.512
1.482
1.452
1.423
1.395
1.363
1.341
1.316
1.291
1.266
1.243
1.220
1.198
1.176
1.155
1.134
1.114
1.095
1.076
1.057
1.039
1.022
1.005
0.9881
0.1
3.811
3.690
3.574
3.463
3.357
3.257
3.160
3.068
2.980
2.896
2.815
2.737
2.663
2.591
2,523
2.457
2.394
2.333
2.274
2.218
2.163
2.111
2.060
2.012
1.965
1.919
1.876
1.833
1.792
1.753
1.715
1.677
1.642
1.607
1.573
1.541
1.509
1.479
1.449
1.420
1.392
1.365
1.339
1.313
1.288
1.264
1.240
1.218
1.195
1.174
1.153
1.132
1.112
1.093
1.074
1.056
1.038
1.020
1.003
0.9865
0.2
3.799
3.678
3.562
3.452
3.347
3.247
3.151
3.059
2.971
2.887
2.807
2.730
2.656
2.584
2.516
2.451
2.387
2.327
2.268
2.212
2.158
2.106
2.056
2.077
1.960
1.915
1.871
1.829
1.788
1.749
1.711
1.674
1.638
1.604
1.570
1.538
1.506
1.476
1.446
1.417
1.389
1.362
1.336
1.311
1.286
1.262
1.238
1.215
1.193
1.172
1.151
1.130
1.110
1.091
1.072
1.054
1.036
1.018
1.001
0.9848
0.3
3.787
3.666
3.551
3.442
3.337
3.237
3.142
3.050
2.963
2.879
2.799
2.722
2.648
2.577
2.509
2.444
2.381
2.321
2.263
2.207
2.153
2.101
2.051
2.002
1.956
1.911
1.867
1.825
1.784
1.745
1.707
1.670
1.635
1.600
1.567
1.534
1.503
1.473
1.443
1.413
1.387
1.360
1.333
1.308
1.283
1.259
1.236
1.213
1.191
1.170
1.149
1.128
1.108
1.089
1.070
1.052
1.034
1.017
0.9993
0.9832
0.4
3.774
3.654
3.540
3.431
3.327
3.227
3.132
3.041
2.954
2.871
2.791
2.715
2.641
2.571
2.503
2.438
2..37S
2.315
2.257
2.201
2.147
2.096
2.046
1.998
1.951
1.906
1.863
1.821
1.780
1.741
1.703
1.667
1.631
1.597
1.563
1.531
1.500
1.470
1.440
1.412
1.384
1.357
1.331
1.305
1.281
1.257
1.234
1.211
1.189
1.167
1.147
1.126
1.106
1.087
1.068
1.050
1.032
1.015
0.9981
0.9816
0.5
3.762
3.643
3.529
3.420
3.317
3.218
3.123
3.032
2.946
2.863
2.783
2.707
2.634
2.564
2.496
2.431
2.369
2.309
2.251
2.196
2.142
2.091
2.041
1.993
1.946
1.902
1.858
1.817
1.776
1.737
1.700
1.663
1.628
1.593
1.560
1.528
1.497
1.467
1.437
1.409
1.381
1.354
1.328
1.303
1.278
1.255
1.231
1.209
1.187
1.165
1.145
1.124
1.105
1.085
1.067
1.048
1.031
1.013
0.9964
0.9799
0.6
3.750
3.631
3.518
3.410
3.306
3.208
3.114
3.024
2.937
2.855
2.776
2.700
2.627
2.557
2.490
2.425
2.363-
2.303
2.246
2.190
2.137
2.086
2.036
1.988
1.942
1 .897.
1.854
1.813
1.772
1.733
1.696
1.659
1.624
1.590
1.557
1.525
1.494
1.464
1.434
1.406
1.378
1.352
1.326
1.301
1.276
1.252
1.229
1.206
1.185
1.163
1.142
1.122
1.103
1.083
1.065
1.047
1.029
1.012
0.9947
0.9783
0.7
3.738
3.619
3.507
3.399
3.296
3.198
3.105
3.015
2.929
2.846
2.768
2.692
2.620
2.550
2.483
2.419
2.357
2.297
2.240
2.185
2.132
2.080
2.031
1.9S3
1.937
1.8S3
1.850
1 .809
1.768
1.730
1.692
1.656
1.621
1.587
1.554
1.522
1.491
1.461
1.431
1.403
1.376
1.349
1.323
1.298
1.274
1.250
1.227
1.204
1.182
1.161
1.140
1.120
1.101
1.082
1.063
1.045
1.027
1.010
0.9931
0.9767
0.8
3.726
3.608
3.496
3.389
3.286
3.189
3.953
3.006
2.921
2.839
2.760
2.685
2.613
2.543
2.476
2.412
2.351
2.292
2.234
2.179
2.127
2.075
2.026
1.979
1.933
1.889
1.846
1.804
1.765
1.726
1.688
.1,652
1.617
1.583
1.550
1.519
1.488
1.458
1.429
1.400
1.373
1.346
1.321
1.296
1.271
1.247
1.224
1.202
1.180
1.159
1.138
1.118
1.099
1.080
1.161
1.043
1.025
1 .008
0.9914
0.9751
0.9
3.714
3.596
3.485
3.378
3.276
3.179
3.086
2.997
2.912
2.831
2.752
2.677
2.605
2.536
2.470
2.406
2.345
2.286
2.229
2.174
2.121
2.070
2.021
1.974
1.928
1.884
1.842
1.800
1.761
1.722
1.685
1.649
1.614
1.580
1.547
1.515
1 .485
1.455
1.425
1.398
1.370
1344
1.318
1.293
1.269
1.245
1.222
1.200
1.178
1.157
1.136
1.116
1.097
1.078
1.059
1.041
1.024
1.006
0.9898
0.9735
G-9
-------
E384
TABLE 1 Continued
Diagonal of
Impression,
pm
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
Knoop Hardness Number for Diagonal Measured to 0.1 ^m
— _ __
0.9719
0.9560
0.9405
0.9254
0.9107
0.8963
0.8822
0.8685
0.8551
0.8420
0.8291
0.8166
0.8044
0.7924
0.7807
0.7693
0.7581
0.7472
0.7365
0.7260
0.7157
0.7057
0.6958
0.6862
0.6768
0.6675
0.6585
0.6496
0.6409
0.6324
0.6241
0.6159
0.6078
0.6000
0.5923
0.5847
0.5773
0.5700
0.5628
0.5558
0.5489
0.5422
0.5356
0.5290
0.5226
0.5164
0.5102
0.5041
0.4982
0.4924
0.4866
0.4810
0.4754
0.4700
0.4646
0.4594
0.4542
0.4491
0.4441
0.4392
0.1
0.9703
0.9544
0.9390
0.9239
0.9092
0.8948
0.8808
0.8671
0.8537
0.8407
0.8279
0.8154
0.8032
0.7913
0.7796
0.7682
0.7570
0.7461
0.7354
0.7249
0.7147
0.7047
0.6949
0.6852 .
0.6758
0.6666
0.6576
0.6487
0.6401
0.6316
0.6232
0.6151
0.6071
0.5992
0.5915
0.5839
0.5765
0.5693
0.5621
0.5551
0.5483
0.5415
0.5349
0.5284
0.5220
0.5157
0.5096
0.5035
0.4976
0.4918
0.4860
0.4804
0.4749
0.4694
0.4641
0.4588
0.4537
0.4486
0.4436
0.4387
0.2
0.9687
0.9529
0.9375
0.9224
0.9078
0.8934
0.8794
0.8658
0.8524
0.8394
0.8266
0.8142
0.8020
0.7901
0.7784
0.7670
0.7559
0.7450
0.7343
0.7239
0.7137
0.7037
0.6939
0.6843
0.6749
0.6657
0.6567
0.6479
0.6392
0.6307
0.6224
0.6143
0.6063
0.5984
0.5907
0.5832
0.5758
0.5685
0.5614
0.5544
0.5476
0.5408
0.5342
0.5278
0.5214
0.5151
0.5090
0.5030
0.4970
0.4912
0.4855
0.4799
0.4743
0.4689
0.4636
0.4583
0.4532
0.4481
0.4431
0.4382
0.3
0.9671
0.9513
0.9359
0.9209
0.9063
0.8920
0.8780
0.8644
0.8511
0.8381
0.8254
0.8129
0.8008
0.7889
0.7773
0.7659
0.7548
0.7439
0.7333
0.7229
0.7127
0.7027
0.6929
0.6834
0.6740
0.6648
0.6558
0.6470
0.6383
0.6299
0.6216
0.6134
0.6055
0.5976
0.5900
0.5825
0.5751
0.5678
0.5607
0.5537
0.5469
0.5402
0.5336
0.5271
0.5208
0.5145
0.5084
0.5024
0.4964
0.4906
0.4849
0.4793
0.4738
0.4684
0.4630
0.4578
0.4526
0.4476
0.4426
0.4377
0.4
0.9655
0.9498
0.9344
0.9195
0.9049
0.8906
0.8767
0.8631
0.8498
0.8368
0.8241
0.8117
0.7996
0.7877
0.7761
0.7648
0.7537
0.7429
0.7322
0.7218
0.7117
0.7017
0.6920
0.6824
0.6731
0.6639
0.6549
0.6461
0.6375
0.6290
0.6208
0.6126
0.6047
0.5969
0.5892
0.5817
0.5743
0.5671
0.5600
0.5531
0.5462
0.5395
0.5329
0.5265
0.5201
0.5139
0.5078
0.5018
0.4959
0.4900
0.4843
0.4787
0.4732
0.4678
0.4625
0.4573
0.4521
0.4471
0.4421
0.4372
0.5
0.9639
0.9482
0.9329
0.9180
0.9034
0.8892
0.8753
0.8617
0.8485
0.8355
0.8229
0.8105
0.7984
0.7866
0.7750
0.7637
0.7526
0.7418
0.7312
0.7208
0.7107
0.7007
0.6910
0.6815
0.6721
0.6630
0.6540
0.6452
0.6366
0.6282
0.6199
0.6118
0.6039
0.5961
0.5885
0.5810
0.5736
0.5664
0.5593
0.5524
0.5455
0.5389
0.5323
0.5258
0.5195
0.5133
0.5072
0.5012
0.4953
0.4895
0.4838
0.4782
0.4727
0.4673
0.4620
0.4568
0.4516
0.4466
0.4416
0.4367
0.6
0.9623
0.9467
0.9314
0.9165
0.9020
0.8878
0.8739
0.8604
0.8472
0.8343
0.8216
0.8093
0.7972
0.7854
0.7738
0.7626
0.7515
0.7407
0.7301
0.7198
0.7097
0.6997
0.6900
0.6805
0.6712
0.6621
0.6531
0.6444
0.6358
0.6274
0.6191
0.6110
0.6031
0.5953
0,5877
0.5802
0.5729
0.5657
0.5586
0.5517
0.5449
0.5382
0.5316
0.5252
0.5189
0.5127
0.5066
0.5006
0.4947
0.4889
0.4832
0.4776
0.4721
0.4668
0.4615
0.4562
0.4511
0.4461
0.4411
0.4363
0.7
0.9607
0.9451
0.9299
0.9150
0.9005
0.8864
0.8726
0.8591
0.8459
0.8330
0.8204
0.8080
0.7960
0.7842
0.7727
0.7614
0.7504
0.7396
0.7291
0.7188
0.7087
0.6988
0.6891
0.6796
0.6703
0.6612
0.6523
0.6435
0.6349
0.6265
0.6183
0.6102
0.6023
0.5946
0.5869
0.5795
0.5722
0.5650
0.5579
0.5510
0.5442
0.5375
0.5310
0.5246
0.5182
0.5120
0.5060
0.5000
0.4941
0.4883
0.4827
0.4771
0.4716
04662
0.4609
0.4557
04506
0.4456
0.4406
0.4358
0.8
0.9591
0.9436
0.9284
0.9136
0.8991
0.8850
0.8712
0.8577
0.8446
0.8317
0.8191
0.8068
0.7948
0.7831
0.7716
0.7603
0.7493
0.7386
0.7281
0.7177
0.7077
0.6978
0.6881
0.6786
0.6694
0.6603
0.6514
0.6426
0.6341
0.6257
0.6175
0.6094
0.6015
0.5938
0.5862
0.5787
0.5714
0.5643
0.5572
0.5503
0.5435
0.5369
0.5303
0.5239
0.5176
0.5114
0.5054
0.4994
0.4935
0.4878
0.4821
0.4765
0.4711
04657
0.4604
0.4552
04501
0.4451
0.4401
0.4353
0.9
0.9576
0.9420
0.9269
0.9121
0.8977
0.8836
0.8698
0.8564
0.8433
0.8304
0.8179
0.8056
0.7936
0.7819
0.7704
0.7592
0.7483
0.7375
0.7270
0.7167
0.7067
0.6968
0.6872
0.6777
0.6684
0.6594
0.6505
0.6418
0.6332
0.6249
0.6167
0.6086
•0.6008
0.5930
0.5854
0.578P
0.5707
0.5635
0.5565
0.5496
0.5429
0.5362
0.5297
0.5233
0.5170
0.5106
0.5047
0.4986
0.4929
0.4872
0.4815
0.475C
0.4705
0.4652
0.459S
04547
0 4496
0.4446
0.43S'
0 434f
U- !
-------
E384
TABLE 1 Continued
Diagonal of
Impression.
^m
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
Knoop Hardness
0.0
0.4343
0.4296
0.4249
0.4203
0.4158
0.4113
0.4069
0.4026
0.3983
0.3942
0.3900
0.3860
0.3820 '
0.3781
0.3742
0.3704
0.3663
0.3630
0.3593
0.3557
0.1
0.4339
0.4291
0.4244
0.4198
0.4153
0.4109
0.4065
0.4022
0.3979
0.3937
0.3896
0.3856
0.3816
0.3777
0.3738
0.3700
0.3663
0.3626
0.3590
0.3554
0.2
0.4334
0.4286
0.4240
0.4194
0.4149
0.4104
0.4060
0.4017
0.3975
0.3933
0.3892
0.3852
0.3812
0.3773
0.3734
. 0.3696
0.3659
0.3622
0.3586
0.3550
0.3
0.4329
0.4282
0.4235
0.4189
0.4144
0.4100
0.4056
0.4013
0.3971
0.3929
0.3888
0.3848
0.3808
0.3769
0.3731
0.3693
0.3655
0.3619
0.3582
0.3547
Number for Diagonal Measured to 0.1 urn
0.4
0.4324
0.4277
0.4230
0.4185
0.4140
0.4095
0.4052
0.4009
0.3967
0.3925
0.3884
0.3844
0.3804
0.3765
0.3727
0.3689
0.3652
0.3615
0.3579
0.3543
0.5
0.4319
0.4272
0.4226
0.4180
0.4135
0.4091
0.4047
0.4005
0.3962
0.3921
0.3880
0.3840
0.3800
0.3761
0.3723
0.3685
0.3648
0.3611
0.3575
0.3540
0.6
0.4315
0.4268
0.4221
0.4176
0.4131
0,4087
0.4043
0.4000
0.3958
0.3917
0.3876
0.3836
0.3796
0.3757
0.3719
0.3681
0.3644
0.3608
0.3572
0.3536
0.7
0.4310
0.4263
0.4217
0.4171
0.4126
0.4082
0.4039
0.3996
0.3954
0.3913
0.3872
0.3832
0.3792
0.3754
0.3715
0.3678
0.3641
0.3604
0.3568
0.3533
0.8
0.4305
0.4258
0.4212
0.4167
0.4122
0.4078
0.4034
0.3992
0.3950
0.3909
0.3868
0.3828
0.3789
0.3750
0.3712
0.3674
0.3637
0.3600
0.3564
0.3529
0.9
0.4300
0.4254
0.4207
0.4162
0.4117
0^4073
0.4030
0.3988
0.3946
0.3905
0.3864
0.3824
0.3785
0.3746
0.3708
0.3670
0.3633
0.3597
0.3561
0.3525
G-11
-------
E384
TABLE 2 Vickers Hardness Numbers for Load of 1 gf
Diagonal of
Impression, (irr
1
2
3
A
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
36
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
' 0.0
1854.6
463.0
206.9
115.17
74
51.51
37.84
28.97
22.89
18.54
15.33
. 12.88
10.97
9.461
8.242
7.244
6.416
5.723
5.137
4.636
4.205
3.831
3.505
3.219
2.967
2.743
2.544
2.365
2.205
2.060
1.930
1.811
1.703
1.604
1.514
1.431
1.355
1.284
1.219
1.159
1.103
1.051
0.003
0.9578
0.9157
0.8764
0.8395
0.8048
0.7723
0.7417
0.7129
0.6858
0.6602
0.6359
0.6130
0.5913
0.5708
0.5512
0.5327
0.5151
Vickers Hardness Number for Diagonal Measured to 0.1 urn
0.1
1533.5
420.0
193.3
110.29
71
49.83
36.79
28.26
22.39
18.18
15.05
12.67
10.81
9.327
8.133
7.154
6.342
5.660
5.083
4.590
4.165
3.797
3.475
3.193
2.943
2.722
2.525
2.348
2.190
2.047
1.917
1.800
1.693
1.595
1.505
1.423
1.347
1.277
1.213
1.153
1.098
1.046
0.9983
0.9535
0.9117
0.8726
0.8359
0.8015
0.7692
0.7388
0.7102
0.6832
0.6577
0.6336
0.6108
0.5892
0.5688
0.5493
0.5309
0.5134
0.2
1288.1
383.1
181.1
105.58
68
48.24
35.77
27.58
21.91
17.82
14.78
12.46
10.64
9.196
8.026
. 7.066
6.268
5.598
5.030
4.545
4.126
3.763
3.445
3.166
2.920
• 2.701
2.506
2.332
2.175
2.033
1.905
1.788
1.682
1.585
1.497
1.415
1.340
1.271
1.207
1.147
1.092
1.041
0.9936
0.9492
0.9077
0.8688
0.8324
0.7982
0.7661
0.7359
0.7074
0.6805
0.6552
0.6312
0.6086
0.5871
0.5668
0.5475
0.5291
0.5117
0.3
1097.5
350.3
170.3
100.01
66
46.72
34.80
26.92
21.44
17.48
14.52
12.26
10.48
9.068
7.922
6.979
6.196
5.537
4.978
4.500
4.087
3.729
3.416
3.140
2.897
2.681
2.488
2.315
2.160
2.020
1.893
1.777
1.672
1.576
1.488
1.407
1.333
1.264
1.201
1.142
1.087
1.036
0.9891
0.9449
0.9036
0.8650
0.8288
0.7949
0.7630
0.7329
0.7046
0.6779
0.6527
0.6289
0.6064
0.5850
0.5648
0.5456
0.5273
0.5100
0.4
946.1
321.9
160.4
95.78
63.59
45.27
33.86
26.28
20.99
17.14
14.27
12.06
10.33
8.943
7.819
6.895
6.125
5.477
4.927
4.456
4.049
3.696
3.387
3.115
2.874
2.661
2.470
2.299
2.145
2.007
1.881
1.766
1.662
1.567
1.480
1.400
1.326
1.258
1.195
1.136
1.082
1.031
0.9845
0.9407
0.8997
0.8613
0.8254
0.7916
0.7599
0.7300
0.7019
0.6754
0.6503
0.6266
0.6042
0.5830
0.5628
0.5437
0.5256
0.5083
0.5
824.2
296.7
151.4
91.57
61.30
43.89
32.97
25.67
20.55
16.82
14.02
11.87
10.17
8.820
7.718
6.811
6.055
5.418
4.877
4.413
4.012
3.663
3.358
3.089
2.852
2.641
2.452
2.283
2.131
1.993
1.869
1.756
1.652
1.558
1.471
1.392
1.319
1551
1.189
1.131
1.077
1.027
0.9800
0.9364
0.8957
0.8576
0.8219
0.7883
0.7568
0.7271
0.6992
0.6728
0.6479
0.6243
0.6020
0.5809
0.5609
0.5419
0.5238
0.5066
0.6
724.4
274.3
143.1
87.64
59.13
42.57
32.10
25.07
20.12
16.50
13.78
11.68
10.03
8.699
7.620
6.729
5.986
5.360
4.827
4.370
3.975
3.631
3.329
3.064
2.830
2.621
2.434
2.267
2.116
1.980
1.857
1.745
1.643
1.549
1 .463
1.384
1.312
1.245
1.183
1.125
1.072
1.022
0.9755
0.9322
0.8918
0.8539
0.8184
0.7851
0.7538
0.7243
0.6965
0.6702
0.6455
0.6220
0.5999
0.5788
0.5589
0.5400
0.5220
0.5050
0.7
641.6
254.4
135.5
83.95
57.07
41.31
31.28
24.50
19.71
16.20
13.55
11.50
9.880
8.581
7.523
6.649
5.919
5.303
4.77ff
4.328
3.938
3.599
3.301
3.039
2.808
2.601
2.417
2.251
2.102
1.968
1.845
1.734
1.633
1.540
1.455
1.377
1.305
1.238
1.177
1.119
1.066
1.017
0.9710
0.9281
0.8879
0.8503
0.8150
0.7819
0.7507
0.7214
0.6938
0.5677
0.6431
0.6198
0.5977
0.5768
0.5570
0.5382
0.5203
0.5033
0.8
572.3
236.5
128.4
80.48
55.12
40.10
30.48
23.95
19.31
15.90
13.32
11.32
9.737
8.466
7.428
6.570
5.853
5.247
4.730
4.286
3.902
3.587
3.274
3.015
2.786
2.582
2.399
2.236
2.088
1.955
1.834
1.724
1.623
1.531
1.447
1.369
1.298
1.232
1.171
1.114
1.061
1.012
0.9666
0.9239
0.8840
0.8467
0.8116
0.7787
0.7477
0.7186
0.6911
0.6652
0.6407
0.6175
0.5956
0.5748
0.5551
0.5363
0.5186
0.5016
0.9
513.7
220.5
121.9
77.23
53.27
38.95
29.71
23.41
18.92
15.61
13.09
11.14
9.598
8.353
7.335
6.493
5.787
5.191
4.683
4.245
3.866
3.536
3.246
2.991
2.764
2.563
2.382
2.220
2.074
1.942
1.822
1.713
1.614
1.522
1.439
1.362
1.291
1.225
1.165
1.109
1.056
1.008
0.9622
0.9198
0.8802
0.8430
0.8082
0.7755
0.7447
0.7158
0.6884
0.6627
0.6383
0.6153
0.5934
0.5728
0.5531
0.5345
0.5166
0.5000
G-12
-------
E384
TABLE 2 Continued
Diagonal of
•noression, jam
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
VicKers Hardness. Number for Diagonal Measured to 0.1 nm
0.0
0.4984
0.4824
0.4672
0.4527
0.4389
0.4257
0.4131
0.4010
0.3895
0.3784
0.3679
0.3577
0.3480
0.3386
0.3297
0.3211
0.3128
0.3048
0.2971
0.2897
0.2826
0.2758
0.2692
0.2628
0.2567
0.2507
0.2450
0.2395
0.2341
0.2289
0.2239
0.2191
0.2144
0.2099
0.2055
0.2012
0.1971
0.1931
0.1892
0.1854
0.1818
0.1782
0.1748
0.1715
0.1682
0.1650
0.1620
0.1590
0.1561
0.1533
0.1505
0.1478
0.1452
0.1427
0.1402
0.1378
0.1355
0.1332
0.1310
0.1288
0.1
0.4967
0.4809
0.4657
0.4513
0.4376
0.4244
0.4119
0.3999
0.3884
0.3774
0.3668
0.3567
0.3470
0.3377
0.3288
0.3202
0.3120
0.3040
0.2964
0.2890
0.2819
0.2751
0.2685
0.2622
0.2561
0.2501
0.2444
0.2389
0.2336
0.2284
0.2234
0.2186
0.2139
0.2094
0.2050
0.2008
0.1967
0.1927
0.1888
0.1851
0.1814
0.1779
0.1745
0.1711
0.1679
0.1647
0.1617
0.1587
0.1556
0.1530
0.1502
0.1476
0.1450
0.1424
0.1400
0.1376
0.1352
0.1330
0.1307
0.1236
0.2
0.4951
0.4793
0.4643
0.4499
0.4362
0.4231
0.4106
0.3987
0.3872
0.3763
0.3658
0.3557
0.3461
0.3368
0.3279
0.3194
0.3111
0.3032
0.2956
0.2883
0.2812
0.2744
0.2697
0.2616
0.2555
0.2496
0.2439
0.2384
0.2331
0.2279
0.2230
0.2181
0.2135
0.2090
0.2046
0.2004
0.1963
0.1923
0.1884
0.1847
0.1811
0.1775
0.1741
0.1708
0.1676
0.1644
0.1614
0.1584
0.1555
0.1527
0.1500
0.1473
0.1447
0.1422
0.1397
0.1373
0.1350
0.1327
0.1305
0.1284
0.3
0.4935
0.4778
0.4628
0.4485
0.4349
0.4219
0.4094
0.3975
0.3861
0.3752
0.3648
0.3548
0.3451
0.3359
0.3270
0.3185
0.3103
0.3025
0.2949
0.2876
0.2806
0.2738
0.2672
0.2609
0.2549
0.2490
0.2433
0.2378
0.2325
0.2274
0.2225
0.2177
0.2130
0.2085
0.2042
0.2000
0.1959
0.1919
0.1881
0.1843
0.1807
0.1772
0.1738
0.1705
0.1672
0.1641
0.1611
0.1581
0.1552
0.1524
0.1497
0.1470
0.1445
0.1419
0.1395
0.1371
0.1348
0.1325
0.1303
0.1281
0.4
0.4919
0.4762
0.4613
0.4471
0.4336
0.4206
0.4082
0.3964
0.3850
0.3742
0.3638
0.3538
0.3442
0.3350
0.3262
0.3177
0.3095
0.3017
0.2941
0.2869
0.2799
0.2731
0.2666
0.2603
0.2543
0.2484
0.2428
0.2373
0.2320
0.2269
0.2220
0.2172
0.2126
0.2081
0.2038
0.1995
0.1955
0.1915
0.1877
0.1840
0.1804
0.1769
0.1734
0.1701
0.1669
0.1638
0.1608
0.1578
0.1549
0.1521
0.1494
0.1468
0.1442
0.1417
0.1393
0.1369
0.1345
0.1323
0.1301
0.1279
0.5
0.4903
0.4747
0.4599
0.4457
0.4322
0.4193
0.4070
0.3952
0.3839
0.3731
0.3627
0.3528
0.3433
0.3341
' 0.3253
0.3169
0.3087
0.3009
0.2934
0.2862
0.2792
0.2725
0.2660
0.2597
0.2537
0.2478
0.2422
0.2368
0.2315
0.2264
0.2215
0.2167
0.2121
0.2077
0.2033
0.1991
0.1951
0.1911
0.1873
0.1836
0.1800
0.1765
0.1731
0.1698
0.1666
0.1635
0.1605
0.1575
0.1547
0.1519
0.1492
0.1465
0.1440
0.1414
0.1390
0.1366
0.1343
0.1321
0.1299
0.1277
0.6
0.4887
0.4732
0.4584
0.4444
0.4309
0.4181
0.4058
0.3941
0.3828
0.3720
0.3617
0.3518
0.3423
0.3332
0.3245
0.3160
0.3079
. 0.3002
0.2927
0.2855
0.2785
0.2718
0.2653
0.2591
0.2531
0.2473
0.2417
0.2362
0.2310
0.2259
0.2210
0.2163
0.2117
0.2072
0.2029
0.1987
0.1947
0.1907
0.1869
0.1832
0.1796
0.1762
0.1728
0.1695
0.1663
0.1632
0.1602
0.1572
0.1544
0.1516
0.1489
0.1463
0.1437
0.1412
0.1388
0.1364
0.1341
0.1318
0.1296
0.1275
0.7
0.4871
0.4717
0.4570
0.4430
0.4296
0.4168
0.4046
0.3929
0.3817
0.3710
0.3607
0.3509
0.3414
0.3323
0.3236
0.3152
0.3072
0.2994
0.2919
0.2847
0.2778
0.2711
0.2647
0.2585
0.2525
0.2467
0.2411
0.2357
0.23C5
0.2254
0.2205
0.2158
0.2112
0.2068
0.2025
0.1983
0.1943
0.1904
0.1866
0.1829
0.1793 .
0.1758
0.1724
0.1692
0.1660
0.1629
0.1599
0.1569
0.1541
0.1513
0.1486
0.1460
0.1434
0.1410
0.1385
0.1362
0.1339
0.1316
0.1294
0.1273
0.8
0.4855
0.4702
0.4556
0.4416
0.4283
0.4156
0.4034
0.3918
0.3806
0.3699
0.3597
0.3499
0.3405
0.3314
0.3227
0.3144
0.3064
0.2986
0.2912
0.2840
0.2771
0.2705
0.2641
0.2579
0.2519
0.2461
0.2406
0.2352
0.2300
0.2249
0.2200
0.2153
0.2108
0.2063
0.2021
0.1979
0.1939
0.1900
0.1862
0.1825
0.1789
0.1755
0.1721
0.1688
0.1657
0.1626
0.1596
0.1567
0.1538
0.1511
0.1484
0.1457
0.1432
0.1407
0.1383
0.1359
0.1336
0.1314
0.1292
0.1271
0.9
0.4640
0.4687
0.4541
0.4403
0.4270
0.4143
0.4022
0.3906
0.3795
0.3689
0.3587
0.3489
0.3396
0.3305
0.3219
0.3136
0.3056
0.2979
0.2905
0.2833
0.2765
0.26S8
0.2634
0.2573
0.2513
0.2456
0.2400
0.2346
0.2294
0.2244
0.2196
0.2149
0.2103
0.2059
0.2016
0.1975
0.1935
0.1896
0.1858
0.1821
0.1786
0.1751
0.1718
0.1685
0.1654
0.1623
0.1593
0.1564
0.1535
0.1508
0.1481
0.1455
0.1429
0.1405
0.1381
0.1357
0.1334
0.1312
0.1290
0.1269
G-13
-------
E384
TABLE 2 Continued
Diagonal of
Impression, jim
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155.
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
Vickers Hardness Number for
0.0
0.1267
0.1246
0.1226
0.1206
0.1187
0.1168
0.1150
0.1132
0.1114
0.1097
0.1081
0.1064
0.1048
0.1033
0.1018
0.1003
0.0988
0.0974
0.0960
0.0946
0.0933
0.0920
0.0907
0.0894
0.0882
0.0870
0.0858
0.0847
0.0835
0.0824
0.0813
0.0803
0.0792
0.0782
0.0772
0.0762
0.0752
0.0743
0.0734
0.0724
0.0715
0.0707
0.0698
0.0690
0.0581
0.0673
0.0665
0.0657
0.0649
0.0642
0.0634
0.0627
0.0620
0.0613
0.0606
0.0599
0.0592
0.0585
0.0579
0.0572
0.1
0.1265
0.1244
0.1224
0.1204
0.1185
0.1166
0.1148
0.1130
0.1113
0.1096
0.1079
0.1063
0.1047
0.1031
0.1016
0.1001
0.0987
0.0972
0.0958
0.0945
0.0931
0.0918
0.0906
0.0893
0.0881
0.0869
0.0857
0.0845
0.0834
0.0823
0.0812
0.0802
0.0791
0.0781
0.0771
0.0761
0.0751
0.0742
0.0733
0.0724
0.0715
0.0706
0.0697
0.0689
0.0680
0.0672
0.0664
0.0656
0.0649
0.0641
0.0633
0.0626
0.0619
0.0612
0.0605
0.0598
0.0591
0.0585
0.0578
0.0572
0.2
0.1262
0.1242
0.1222
0.1202
0.1183
0.1164
0.1146
0.1128
0.1111
0.1094
0.1077
0.1061
0.1045
0.1030
0.1015
0.1000
0.0985
0.0971
0.0957
0.0943
0.0930
0.0917
0.0904
0.0892
0.0880
0.0868
0.0856
0.0844
0.0833
0.0822
0.0811
0.0801
0.0790
0.0780
0.0770
0.0760
0.0750
0.0741
0.0732
0.0723
0.0714
0.0705
0.0696
0.0688
0.0680
0.0671
0.0663
0.0656
0.0648
0.0640
0.0633
0.0625
0.0618
0.0611
0.0604
0.0597
0.0591
0.0584
0.0578
0.0571
0.3
0.1260
0.1240
0.1220
0.1200
0.1181
0.1163
0.1144
0.1127
0.1109
0.1092
0.1076
0.1059
0.1044
0.1028
0.1013
0.0998
0.0984
0.0970
0.0956
0.0942
0.0929
0.0916
0.0903
0.0891
0.0878
0.0866
0.0855
0.0843
0.0832
0.0821
0.0810
O.U800
0.0789
0.0779
0.0769
0.0759
0.0749
0.0740
0.0731
0.0722
0.0713
0.0704
0.0695
0.0687
0.0679
0.0671
0.0663
0.0655
0.0647
0.0639
0.0632
0.0625
0.0617
0.0610
0.0603
0.0597
0.0590
0.0583
0.0577
0.0570
0.4
0.1258
0.1238
0.1218
0.1198
0.1179
0.1161
0.1143
0.1125
0.1108
0.1091
0.1074
0.1058
0.1042
0.1027
0.1012
0.0997
0.0982
0.0968
0.0954
0.0941
0.0928
0.0915
0.0902
0.0889
0.0877
0.0865
0.0854
0.0842
0.0831
0.0820
0.0809
0.0798
0.0788
0.0778
0.0768
0.0758
0.0749
0.0739
0.0730
0.0721
0.0712
0.0703
0.0695
0.0686
0.0678
0.0670
0.0662
0.0654
0.0646
0.0639
0.0631
0.0624
0.0617
0.0610
0.0603
0.0596
0.0589
0.0583
0.0576
0.0570
Diagonal Measured to 0.1 urn
0.5
0.1256
0.1236
0.1216
0.1196
0.1177
0.1159
0.1141
0.1123
0.1106
0.1089
0.1072
0.1056
0.1041
0.1025
0.1010
0.0995
0.0981
0.0967
0.0953
0.0939
0.0926
0.0913
0.0901
0.0888
0.0876
0.0864
0.0852
0.0841
0.0830
0.0819
0.0808
0.0797
0.0787
0.0777
0.0767
0.0757
0.0748
0.073S
0.0729
0.0720
0.0711
0.0702
0.0694
0.0685
0.0677
0.0669
0.0661
0.0653
0.0645
0.0638
0.0631
0.0623
0.0616
0.0609
0.0602
0.0595
0.0589
0.0582
0.0576
0.0569
0.6
0.1254
0.1234
0.1214
0.1194
0.1176
0.1157
0.1139
0.1121
0,1104
0.1087
0.1071
0.1055
0.1039
0.1024
0.1009
0.0994
0.0979
0.0965
0.0952
0.0938
0.0925
0.0912
0.0899
0.0887
0.0875
0.0863
0.0851
0.0840
0.0829
0.0818
0.0807
0.0796
0.0786
0.0776
0.0766
0.0756
0.0747
0.0737
0.0728
0.0719
0.0710
0.0701
0.0693
0.0684
0.0676
0.0668
0.0660
0.0652
0.0645
0.0637
0.0630
0.0623
0.0615
0.0608
0.0601
0.0595
0.0588
0.0581
0.0575
0.0569
0.7
0.1252
0.1232
0.1212
0.1193
0.1174
0.1155
0.1137
0.1120
0.1102
0.1086
0.1069
0.1053
0.1037
0.1022
0.1007
0.0992
0.0978
0.0964
0.0950
0.0937
0.0924
0.0911
0.0898
0.0886
0.0874
0.0862
0.0850
0.0839
0.0828
0.0817
0.0806
0.0795
0.0785
0.0775
0.0765
0.0755
0.0746
0.0736
0.0727
0.0718
0.0709
0.0701
0.0692
0.068
-------
E384
TABLE 2 Continued
Diagonal of
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
TABLE
Hardness Range
of Standardized
Test Blocks
Vickers Hardness Number for Diagonal Measured to 0.1 (im
0.0
0.0566
0.0560
0.0554
0.0548
0.0542
0.0536
0.0530
0.0525
0.0519
0.0514
0.0508
0.0503
0.0498
0.0493
0.0488
0.0483
0.0478
0.0473
0.0468
0.0464
0.1
0.0565
0.0559
0.0553
0.0547
0.0541
0.0535
0.0530
0.0524
0.0519
0.0513
0.0508
0.0503
0.0497
0.0492
0.0487
0.0482
0.0477
0.0473
0.0468
0.0463
3 Repeatability of
0.2
0.0565
0.0559
0.0553
0.0547
0.0541
0.0535
0.0529
0.0524
0.0518
0.0513
0.0507
0.0502
0.0497
0.0492
0.0487
0.0482
0.0477
0.0472
0.0467
0.0463
Machines
0.3
0.0564
0.0558
0.0552
0.0546
0.0540
0.0534
0.0529
0.0523
0.0518
0.0512
0.0507
0.0502
0.0496
0.0491
0.0486
0.0481
0.0476
0.0472
0.0467
0.0462
0.4 0.5
0.0564 0.0563
0.0557 0.0557
0.0551 0.0551
0.0545 0.0545
0.0540 0.0539
0.0534 0.0533
0.0528 0.0528
0.0522 0.0522
0.0517 0.0516
0.0512 0.0511
0.0506 0.0506
0.0501 0.0500
0.0496 0.0495
0.0491 0.0490
0.0486 0.0485
0.0481 0.0480
0.0476 0.0475
0.0471 0.0471
0.0466 0.0466
0.0462 0.0461
0.6
0.0562
0.0556
0.0550
0.0544
0.0538
0.0533
0.0527
0.0521
0.0516
0.0510
0.0505
0.0500
0.0495
0.0490
0.0485
0.0480
0.0475
0.0470
0.0465
0.0461
TABLE 4 Repeatability
Repeatability of the
Machine Should Be
Less
Than. *
For loads from 1 gf to <500 gf:
Knoop:
100 to 250. incl
Over 250 to 650, incl
Over 650
VicKers:
100 to 240. inc.
Over 240 to 600. incl
Over 600
For loads from 500 gf to
gf. incl:
Knoop:
100 to 250. incl
Over 250 to 650. inci
Over 650
Vickers:
100 to 240. incl
Over 240 to 600, incl
Over 600
1000
6
5
4
6
5
4
5
4
3
5
4
3
A.B.C
A.B
A.a
A.B.C
A.B
A.a
A.B.C
A. a
A.B
A.B.C
A.B
A.B
0.7
0.0562
0.0556
0.0550
0.0544
0.0538
0.0532
0.0526
0.0521
0.0515
0.0510
0.0505
0.0499
0.0494
0.0489
0.0484
0.0479
0.0474
0.0470
0.0465
0.0460
of Hardness
Hardness Range
for Standardized
Test Blocks
For loads from 1 gf to
Knoop:
100 to 250, incl
Over 250 to 650,
Over 650
Vickers:
100 to 240. incl
Over 240 to 600.
Over 600
For loads from 500 gf
gf. ind:
Knoop:
100 to 250. incl
Over 250 to 650.
Over 650
Vickers:
100 to 240. ind
Over 240 to 600.
Over 600
<500 gf:
inc.
ind
to 1000
ind
ind
0.8
0.0561
0.0555
0.0549
0.0543
0.0537
0.0531
0.0526
0.0520
0.0515
0.0509
0.0504
0.0499
0.0494
0.0489
0.0484
0.0479
0.0474
0.0469
0.0465
0.0460
Readings
0.9
0.0560
0.0554
0.0548
0.0542
0.0537
0.0531
0.0525
0.0520
0.0514
0.0509
0.0504
0.0498
0.0493
0.0488
0.0483
0.0478
0.0474
0.0469
0.0464
0.0459
Repeatability of me
Test Block Readinas
Shall Be Less
$A.B.C
4*.s
3*.a
-
§A.B.C
4».s
OA3
4».a.c
3A.B
2A.B
£*.B.C
3*-«
ZA.B
Thar,, %
* a = I
+ d2 + .
- ds)/5.
8 In all cases the repeatability is the percentage given of 1 urn, whichever is
greater.
c Due to the nature of matenals currently available for test blocks in 100 HK to
250 HK and 100 HV to 240 HV ranges, percentage values noted represent
repeatability of averages of 2 or more groups of 5 indentations each.
* d, = (a, + d2 + . .. + d5)/5.
B In all cases the repeatability is the percentage given or 1 jim. whichever is
greater.
° Due to the nature of matenals currently available for test blocks in 100 HK to
250 HK and 100 HV to 240 HV ranges, percentage values noted represent
repeatability of averages of 2 or more groups of 5 indentations each.
G-15
-------
E384
ANNEX
(Mandatory Information)
Al. ADJUSTMENT OF ABBE-NELSON OR KOHLER ILLUMINATION SYSTEMS
A 1.1 While some optical systems are permanently
aligned, others have means of minor adjustments. To gain
the utmost in resolution the operator should make the
following adjustments:
A 1.1.1 Abbe-Nelson Illumination:
A 1.1.1.1 Focus to critical sharpness the surface of a flat
polished specimen.
A 1.1.1.2 Center the illuminating source.
A 1.1.1.3 Centrally align the field and aperture dia-
phragms.
A 1.1.1.4 Adjust the lamp so that the filament is in sharp
focus in the specimen plane.
A 1.1.1.5 Close the field diaphragm so that a thin, dark
ring rims the field of view.
A1.1.1.6 Close the aperture diaphragm until the glare just
disappears. Never close the diaphragm to the point where
diffraction phenomena appear.
A 1.1.1.7 Place a diffusing disk in back of the field
diaphragm if the lamp is not a ribbon-filament type.
A 1.1.1.8 If the light is too strong for eye comfort, reduce
the intensity by the use of an appropriate neutral density
filter or rheostat control.
A1.1.2 Kohler Illumination:
A 1.1.2.1 Focus to critical sharpness the surface of a flat
polished specimen.
A 1.1.2.2 Center the illuminating source.
A 1.1.2.3 Centrally align field and aperture diaphragms.
A 1.1.2.4 Open the field diaphragm so that it just disap-
pears from the field of view.
A 1.1.2.5 Remove the eyepiece and examine the rear focal
plane of the objective. If all the components are in their
proper places, the source of illumination and the aperture
diaphragm will appear in sharp-focus.
A 1.1.2.6 Full-aperture diaphragm is preferred for max-
imum resolving power. If glare is excessive, reduce the
aperture; but never use less than the 3/4 opening since
resolution would be decreased and diffraction phenomena
could lead to false measurements.
A 1.1.2.7 If the light is too strong for eye comfort, reduce
the intensity by the use of an appropriate neutral density
filler or rheostat control.
APPENDIXES
(Nonmandatory Information)
XI. CORRELATION OF MICROHARDNESS
XI.1 Scope
XI. 1.1 This procedure provides guidance in establishing
correlation of microhardness tests when two or more labora-
tories are involved.
XI.2 Significance
X1.2.1 Test Method E 384 establishes methods for deter-
mining microhardness of materials, verifying microhardness
testing machines, and calibrating standardized test blocks.
This appendix supports the method by guiding laboratories
wishing to correlate their microhardness test data.
XI .3 Correlation Procedure
XI.3.1 All laboratories shall first establish that their test
equipment conforms to the requirements in Test Method
E384.
XI.3.2 The specimens to be tested shall be taken from
adjoining areas of the larger sample after all processing has
been completed and prior to being sent to the cooperating
laboratories for preparation and testing.
XI.3.3 The specimens shall be prepared for microhard-
ness by the two or more laboratories using essentially the
same procedures. If the specimens are capable of being
TEST DATA BETWEEN LABORATORIES
prepared as metallographic specimens, established ASTM
procedures shall be maintained uniformly among the labora-
tories as follows:
X 1.3.3.1 The same surfaces shall be exposed for the
microhardness test. This is to ensure that gram direction, if a
characteristic, is taken into consideration.
XI.3.3.2 The surface preparation of the specimens shall
be in accordance with Methods E 3.
XI.3.4 All laboratories shall calibrate the optics of their
test apparatus with the same stage micrometer.
Xl.3.5 The indentations shall be oriented the same rela-
tive to grain direction in order to avoid differences in results
arising from this factor.
XI.3.6 The method of measuring the indentations shall
be established prior to making the tests. It shall be the most
accurate method as described by the equipment manufac-
turer.
XI.3.7 A minimum number of indentations shall be
established. This shall conform to acceptable statistical
methods of analysis, in accordance with Recommended
Practice E 122.
Xl.3.8 Each test specimen shall be indented and mea-
sured by the laboratory having prepared it. then sent with the
G-16
-------
E384
data for testing in the other laboratory or laboratories.
XI.3.8.1 After the specimens have been exchanged, each
laboratory shall measure and record the indentations applied
by the originating laboratory in a manner identical to the
initial measurements.
XI.3.8.2 Each laboratory shall then repeat the indenta-
tion and measuring procedures as performed in XI.3.5 and
XI.3.6, before sending the data and specimen to the re-
maining laboratory or laboratories.
XI.3.8.3 Each laboratory shall determine a set of micro-
hardness values from the specimen they prepared, as well as
sets of values they obtained by indenting and measuring
specimens prepared by the other laboratory or laboratories.
XI.3.9 All data shall then be analyzed by the same
acceptable statistical methods to establish the limits of
agreement that are attainable between the two laboratories.
As a minimum the following statistical data shall be evolved:
X = mean
a = standard deviation
a/% = standard error of the mean
XI .4 Referee
XI.4.1 If the laboratories cannot establish an acceptable
correlation through this procedure, it will be necessary to
introduce an independent laboratory to act as the referee.
X2. RESULTS OF INTERLABORATORY TEST OF THE MEASUREMENT OF MICROINDENTATIONS (5, 6)
X2.1 Introduction
X2.1.1 This interlaboratory test program was conducted
to develop precision and bias estimates for the measurement
of both Knoop and Vickers indentations using loads of 25 to
1000 gf for ferrous and nonferrous specimens covering a
wide range of hardness.
X2.2 Scope
X2.2.1 This interlaboratory test program provides infor-
mation on the measurement of the same indents by different
laboratories in accordance with the procedures of Practice
E691.
X23 Referenced Documents
X2.3.1 AST.Vf Standards:
E 691 Conducting an Interlaboratory Test Program to
Determine the Precision of Test Methods.7
X2.4 Procedure
X2.4.1 Five indents were made under controlled condi-
tions at each losd (25, 50, 100, 200, 500, and 1000 gf), with
both Knoop and Vickers indenters, using three ferrous and
four nonferrous speciirans.
X2.4.2 Twelve laboratories measured the indents on the
ferrous specimens and twelve laboratories measured the
indents on the nonferrous specimens (only two laboratories
measured the indents on both the ferrous and nonferrous
specimens).
X2.4.3 Each laboratory used the same stage micrometer
to calibrate their measuring device.
X2.4.4 Results were tabulated and analyzed according to
Practice E691.
X2.5 Results
X2.5.1 For the three ferrous specimens, results from nine
laboratories showed general agreement as to the diagonal
sizes. Two other laboratories consistently undersized the
indents (higher hardness) and one laboratory consistently
oversized the indents (lower hardness). This bias was ob-
served with both Vickers and Knoop indents sized by these
laboratories with the degree of bias increasing as the indent
7 Annual Book ofASTM Standards. Vol 14.02.
size decreased and the specimen hardness increased. Test on
the four nonferrous specimens produced general agreement
but none of the three laboratories that produced biased
results for the ferrous samples measured the nonferrous
specimens.
X2.5.2 For the Vickers test data, the calculated hardness
increased with increasing load and then became reasonably
constant. This trend was apparent in the data from the nine
consistent laboratories (ferrous specimens) and for the labo-
ratory that oversized the indents. The two laboratories that
undersized the indents became relatively constant. The load
at which the hardness became relatively constant increased
with increasing specimen hardness. For specimens below
about 300 HV, there was relatively little difference in HV
over the test load range.
X2.5.3 For the Knoop test data, all of the laboratories
agreed that the hardness decreased continually with in-
creasing test load and then became reasonably constant. The
difference in HK values between low loads and high loads
increased with increasing specimen hardness. For specimens
with hardnesses below about 300 HK, the difference in
hardness was quite small over the test load range.
X2.5.4 Repeatability Interval—The difference due to test
error between two test results in the laboratory on the same
material was calculated using the (Sr)j values, the pooled
within-laboratory standard deviation. (Sr)j increased with
diagonal size and the relationship varied for each material
and test type. Table X2.1 lists regression equations that show
the relationship between (Sr)j and the diagonal length in
micrometers. The repeatability interval, /(/•),, was calculated
based on the relationships in Table X2.1. Because the
repeatability intervals are also a function of diagonal length,
regression equations were also calculated, Table X2.2. The
repeatability intervals, in terms of Knoop and Vickers values
for ferrous and nonferrous specimens, are shown in Figs.
X2.1 toX2.4.
X2.5.5 Reproducibility Interval—The difference in test
results on the same material in different laboratories was
calculated using the (SR)j values, the between-laboratory
estimate of precision. (SR)j increased with diagonal size and
the relationship varied for each material and test type. Table
X2.3 lists the regression equations that show the relationship
3-17
-------
QlP E384
TABLE X2 1 I Ry»ton«MP *etwfen Dia9°na' ^ngth and (Sr)/, the TABLE X2.3 Relationship Between Diagonal Length and (SB), the
Pooled Within-Laboratory Standard Dev.at.on Between-Laboratory Estimate of Precision '
Material
Ferrous
Ferrous
Nonferrous
Nonferrous
TABLE X2.2
Material
Ferrous
Ferrous
Nonferrous
Nonferrous
Test Regression Equation Correlate
Coefficient
Vickers (S,), = 0.231 + 0.00284d, 0.535
Knoop (S,); = 0.216 + O.OO&J, 0.823
Vickers (S,)y = 0.373 + O.OOSd, 0.862
Knoop (5^ = 0.057 + 0.0177(7, 0.8196
Relationship Between the Diagonal Length and /(r),,
the Repeatability Interval (±)
Test Regression Equation
Vickers l(r\ = 0.653 + O.OOSd,
Knoop /(r), = 0,61 4 + 0.01 7c7,
Vickers /(r); = 1 .0556 •*- 0.0226<7,
Knoop l(r)t = 0.161 + O.OStf,
Material
Ferrous
Ferrous
Nonferrous
Nonferrous
TABLE. X2.4
Matenal
Ferrous
Ferrous
Nonferrous
Nonferrous
Test Regression Equation Correlation
Coefficient
Vickers (SH}; = 0.31 + 0.004d, 0.747
Knoop (SH), = 0.333 + 0.007,2, 0.899
Vickers (SH); = 0.357 + 0.01 56d, 0.8906
Knoop (5,,}. = 0.378 + 0.01 77rf, 0.8616
Relationship Between the Diagonal Length and /(ft),,
the Reproducibility Interval (±)
Test Regression Equation
Vickers /(fl); = 0.877 + 0.0113d,
Knoop /(«), = 0.946 + 0.01 98cf,
Vickers /(fl); = 1.0103 + 0.0441?,
Knoop /(fl). = 1 .07 + O.OSo7,
between (Sx)j and the diagonal length in micrometers. The
reproducibility intervals, ](R )r was calculated based on the
relationships shown in Table X2.3. Because the reproduc-
ibility intervals are also a function of diagonal length,
regression equations were also calculated. Table X2.4. The
reproducibility intervals, in terms of Knoop and Vickers
values for the ferrous and nonferrous specimens, are shown
in Figs. X2.! to X2.4.
X2.5.6 The within-laboratory and between-laboratorv
precision values were calculated from (I7X%)^ and (VL(%)).
which are the coefficients of variation for within-laboratory
and between-laboratory tests. Both are a function of the
length of the length of the diagonal. The within-laboratory
and between-laboratory precision values were relatively sim"-
ilar for both Vickers and Knoop test data, either ferrous or
.nonferrous. In general, the repeatability intervals and repro-
ducibility intervals were larger than the precision estimates.
particularly at low test loads and high sample hardnesses.
*OO 5OC tOO
KHOOf MJMOMCSS(HK)
FIG. X2.1 Repeatability and Reproducibility Intervals in Terms of
Vickers Hardness (±) for the Ferrous Samples as a Function of
Test Load and Specimen Hardness
SCO 400 SCO »
-------
E384
MONftMKXB SAhtPLES
200 900
HAKONESS(MV)
100 200 100
KNOOP HARDNESS INK)
100 200 300
VKKEB5 HAHONESS IMVI
FIG. X2.3 Repeatability and Reproducibility Intervals in Terms of
Vickers Hardness (±) for the Nonferrous Samples as a Function of
Test Load and Specimen Hardness
NOM*t»«OU5 SAMPLES
KX> 200 SOO
KNOOP HAMOMFSS INK)
FIG. X2.4 Repeatability and Reproducibility Intervals in Terms of
Knoop Hardness (±) for the Nonferrous Samples as a Function of
Test Load and Specimen Hardness
REFERENCES
(I) Campbell. R. F.. et al.. "A New Design of Microhardness Tester
and Some Factors Affecting the Diamond Pyramid Hardness
Number at Light toads." Trans. ASM. Vol 40. 1948. pp. 954-982.
(2) Kennedy. R. G.. and Marrotte. N. W.. "The Effect of Vibration on
Microhardness Testing," Materials Research and Standards. Vol 9.
November 1969. pp. 18-23.
(3) Brown. A. R. G.. and Ineson. E.. "Experimental Survey of
Low-Load Hardness Testing Instruments." J.I.S.I.. Vol 169 1951
pp. 376-388.
(4) Thibault. N. W.. and Nyquist. H. L.. "The Measured Knoop
Hardness of Hard Substances and Factors Affecting Its
Determination." Trans. ASM. Vol 38. 194. pp! 271-330.
(5) Tarasov, L. P., and Thibault. N. W.. "Determination of Knoop
Hardness Numbers Independent of Load." Trans. ASM. Vol 38.
194. pp. 33i-353.
(6) Vander Voort. G. F., "Results of an ASTM E04 Round Robin on
the Precision and Bias of Measurements of Microindentation
Hardness.
G-1S
-------
APPENDIX H
ASTM METHOD E 534-86 FOR DETERMINATION
OF PERCENT INSOLUBLE MATTER
IN SODIUM CHLORIDE
H-1
-------
-------
Designation: E 534 - 86
Standard Test Methods for
Chemical Analysis of Sodium Chloride1
This standard is issued under the fixed designation E 534: the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (<) indicates an editorial change since the last revision or reapproval.
1. Scope
1.1 These test methods cover the chemical analyses usu-
ally required for sodium chloride.
1.2 The analytical procedures appear in the following
sections:
Sample Preparation
Moisture
Water Insolubles
Calcium and Magnesium
Sulfate
Reporting of Analyses
Section
5 to 9
10 to 16
17 to 24
25 to 31
32 to 38
39 to 41
1.3 The values stated in SI units are to be regarded as the
standard.
1.4 This standard may involve hazardous materials, oper-
ations, and equipment. This standard does not purport to
address all of the safety problems associated with its use. It is
the responsibility of the user of this standard to establish
appropriate safety and health practices and determine the
applicability of regulatory limitations prior to use.
2. Referenced Documents
- 2.1 ASTM Standards:
D 1193 Specification for Reagent Water
E 180 Practice for Determining the Precision Data of
ASTM Methods for Analysis and Testing of Industrial
Chemicals3
E 200 Practice for Preparation, Standardization, and
Storage of Standard Solutions for Chemical Analysis3
3. Significance and Use
3.1 Sodium chloride occurs in nature in almost unlimited
quantities. It is a necessary article of diet as well as the source
for production of many sodium compounds and chlorine.
The methods listed in 1.2 provide procedures for analyzing
sodium chloride to determine if it is suitable for its intended
use.
4. Reagents
4.1 Purity of Reagents—Unless otherwise indicated, it is
intended that all reagents should conform to the specifica-
tions of the Committee on Analytical Reagents of the
' These test methods are under the jurisdiction of ASTM Committee £-15 on
Industrial Chemicals and are under the direct responsibility of Subcommittee
El5.52 on Alkalies.
Current edition approved July 25. 1986. Published September 1986. Originally
published as E 534- 75. Last previous edition E 534- 75(1981).
: Annual Book of ASTM Standards. Vol 11.01.
1 Annual Book of ASTM Standards. Vol 15.05.
American Chemical Society, where such specifications are
available.4
4.2 Purity of Water—Unless otherwise indicated, refer-
ences to water shall be understood to mean Type II or III
reagent water conforming to Specification D 1193.
SAMPLE PREPARATION
5. Scope
5.1 This test method covers preparation of a sample thai
will be as representative as possible of 'the entire bulk
quantity. The results of any analysis pertain only to the
sample used.
6. Apparatus
6.1 Coarse Grinder.
6.2 High-Speed Blender.
6.3 Oven.
6.4 Riffle Sampler.
6.5 Scale.
7. Reagents
7.1 Hydrochloric Acid, Standard 1 N HC1—See Practice
E200.
8. Rock and Solar Salt Stock Solutions
8.1 Mix and split sample to 500 g, using the riffle sampler.
8.2 If sample appears wet, dry at 110°C for 2 h.
8.3 Grind the sample to -8.mesh in the coarse grinder.
8.4 Mix ground sample well and weigh out a 25.0-g
representative portion for rock salt or 50.0 g for solar salt.
8.5 Place 200 mL of water in the high-speed blender and
start at low speed.
8.6 Slowly add the salt sample to the high-speed blender
and blend for 5 min.
8.7 Test for water insolubles as described in Sections 17 to
24.
8.8 Save filtrate from water insolubles test and dilute in a
volumetric flask to 1 L with water as a stock solution for
subsequent analyses.
9. Evaporated and Purified Salt Stock Solutions
9.1 Mix and split the sample to 100 g for evaporated sail.
or 200 g for purified evaporated salt.
9.2 Transfer to a 1-L volumetric flask.
4 "Reagent Chemicals. American Chemical Society Specificaiions." Amerw-i''
Chemical Society. Washington. DC. For suggestions on testing of reagents n^
listed by the American Chemical Society, see "Reagent Chemicals and Standards-
by Joseph Rosin. D. Van Nostrand Co.. Inc., New York. NY. and The "UniK*
States Pharmacopeia."
H-2
-------
E534
9.3 Add 800 mL of water and allow the salt to dissolve.
9.4 Add 2 mL of concentrated HC1 to dissolve any water
-soluble calcium salts, particularly calcium carbonate.
9.5 Dilute to volume with water and use as a stock
olution for subsequent analyses.
MOISTURE
11). Scope
10.1 This test method determines free moisture in the salt
•ver a concentration range from 0.00 to 0.04 %. It does not
.-etermine occluded moisture trapped within the salt crystals.
The procedure is based on weight loss after a sample is
-;ated to volatize moisture.
II. Apparatus
11.1 Analytical Balance.
11.2 Desiccator.
11.3 Oven.
12. Procedure, Rock and Solar Salt
12.1 Weigh 100 g of salt to the nearest 0.05 g into a
previously dried and tared moisture dish.
12.2 Dry at 110°Cfor2h.
12.3 Cool in a desiccator and weigh.
13. Procedure, Evaporated and Purified Evaporated Salt
13.1 Weigh 20 g of salt to the nearest 0.001 g into a
?reviously dried and weighed glass weighing bottle and
:over.
13.2 Dry at HOT for 2 h.
13.3 Cool in a desiccator, replace cover, and weigh.
14. Calculation
14.1 Calculate the percentage of moisture as follows:
the values shown in Table 2. Two such averages should be
considered suspect (95 % confidence level) if they differ by
more than the values in this table (Note 1).
16.1.2 Reproducibility (Multilaboralory)—The standard
deviation of results (each the average of duplicates), obtained
by analysts in different laboratories has been estimated at the
values given in Table 2. Two such averages should be
considered suspect (95 % confidence level) if they differ by
more than the values in this table (Note 1).
NOTE 1—The above precision estimates are based on an
interlaboratory study5 with five samples of sodium chloride covering the
above ranges of moisture. One analyst in each of ten laboratories
performed duplicate determinations and repeated 1 day later.
16.2 The bias of this test method has not been deter-
mined.
WATERINSOLUBLES
17. Scope
17.1 This gravimetric method determines only the
amount of insolubles present in sodium chloride which will
not dissolve in water.
18. Apparatus
18.1 A nalytical Balance.
18.2 Desiccator.
18.3 Magnetic Stirrer-wiih Stirring Bar.
18.4 Parabella Filter Funnel Assembly,6 1000-mL, or its
equivalent with 0.3-um glass fiber filter disk.
19. Reagents
19.1 Silver Nitrate, Standard Solution, 0.1 TV AgNO3—
See Practice E 200.
Moisture, % = - x 100
D
•vhere:
•i = loss of weight on drying, g, and
S = weight of sample, g.
15. Report
15.1 Report the moisture content to the nearest decimal
percentage shown in Table 1. Duplicate determinations that
agree within the percent absolute as specified in Table 1 are
acceptable for averaging (95 % confidence level).
16. Precision and Bias
16.1 The following criteria should be used in judging the
acceptability of results:
16.1.1 Repeatability (Single Analyst)—The standard de-
'•lation of results (each the average of duplicates), obtained
^;> the same analyst on different days, has been estimated at
TABLE 1 Percentage of Moisture
Range, %
0.003 to 0.004
0.025 to 0.035
Report to. %
0.001
0.001
Checking
Limits
0.004
0.020
20. Procedure, Rock and Solar Salts
20.1 Transfer a sample prepared in accordance with 8.1 to
8.6 to a l-L Erlenmeyer flask, washing out the blender with
100 mL of water. Add 300 mL of water to give a total of 600
rnL of water added.
20.2 Stir on a magnetic stirrer for I h. Adjust the stirrer
speed to give maximum agitation without danger of losing
any sample due to splashing. Place a beaker over top of the
flask while stirring.
20.3 Filter the solution by vacuum through a previously
dried (110°C for I h) and accurately weighed filter disk using
the Parabella funnel. Transfer all insolubles to the paper and
wash free of chlorides with water until the filtrate shows no
turbidity when tested with O.I N AgNO3 solution.
20.4. Dilute filtrate and washings to I L with water in
volumetric flask.
20.5 Dry the filter disk at 110°C for I h.
20.6 Cool in a desiccator and weigh the disk on an
analytical balance.
20.7 Save the filtrate for subsequent analyses.
3 Supporting data are available at ASTM Headquarters. Request RR: E-15-
1023.
6 Fisher Scientific No. 9-730-200 has been found satisfactory.
H-3
-------
4S|P E534
TABLE 2 Precision Estimates (Moisture)
Range, %
0.003 to 0.004
0.025 to 0.035
Standard
Deviation
0.00223
0.00428
Repeatability
Degrees of
Freedom
18
30
95%
Confidence
Level
0.007
0.01
Standard
Deviation
0.00322
0.0138
Reproducitklity
Degrees of
Freedom
e
9
95%
Confidence
Level
0.010
0.04
21. Procedure, Evaporated and Purified Evaporated Salts
21.1 Place a well mixed sample in a 2-L beaker. Use 100-g
sample for evaporated or 200 g for purified evaporated salt.
21.2 Add 750 mL of water.
21.3 Mix with a mechanical stirrer until solution is
complete.
21.4 Filter the solution by vacuum through a previously
dried (110°C for 1 h) and accurately weighed filter disk using
the Parabella funnel. Transfer all insolubles to the paper and
wash free of chlorides with water until the filtrate shows no
turbidity when tested with 0.1 N AgNO3 solution.
21.5 Dry the filter disk at 110'C for 1 h.
21.6 Cool in a desiccator and weigh on an analytical
balance.
21.7 Dilute the filtration and washings to 1 L with water
in a volumetric flask and reserve for subsequent analyses.
22. Calculation
22.1 Calculate the percentage of water insolubles as fol-
lows:
Insolubles, % = - x 100
B
where:
A = increase in weight of filter disk, g, and
B = sample weight, g.
23. Report
23.1 Report the percentage of water insolubles in the
nearest decimal percentage shown in Table 3. Duplicate
determinations that agree within the percent relative as
specified in Table 3 are acceptable for averaging (95 %
confidence level).
24. Precision and Bias
24.1 The following criteria should be used in judging the
acceptability-of results:
24.1.1 Repeatability (Single Analyst)—The coefficient of
variation of results (each the average of duplicate determina-
tions), obtained by the same analyst on different days, was
estimated to be 23.5 % relative at 57 degrees of freedom.
TABLE 3 Percentage of Water Insolubles
Range.'.
Report to,'
Checking
Limits
Two such averages should be considered suspect (95 ^
confidence level) if they differ by more than 70 % relative
(Note 2).
24.1.2 Reproducibility (Multilaboratory)—The coeffi-
cient of variation of results (each the average of duplicate
determinations), obtained by analysts in different laborato-
ries has been estimated at the values given in the table belo\\.
Two such averages should be considered suspect (95 ~
confidence level) if they differ by more than the values in
Table 4 (Note 2).
NOTE 2—The above precision estimates are based on an
interlaboratory study5 with six samples of sodium chloride covering the
above ranges of water insolubles. One analyst in each often laboratories
performed duplicate determinations and repeated 1 day later. Practice
E 180 was used in developing these precision estimates.
24.2 The bias of this test method has not been deter-
mined.
CALCIUM MAGNESIUM
25. Scope
25.1 This test method covers the EDTA titrimetric deter-
mination of calcium and magnesium and the EDTA
titrimetric determination of calcium. The magnesium con-
tent is determined by difference.
26. Apparatus
26.1 Magnetic Stirrer with Stirring Bar.
27. Reagents
27.1 Eriochrome Black T Indicator Solution, Hydro.\\
Naphthol Blue, or its equivalent.
27.2 Murexide (Ammonium Purpitrate) Indicator Solu-
tion, or its equivalent.
27.3 EDTA Standard Solution (1 mL = 0.400 mg cal-
cium)—Dissolve 4.0 g of disodium dihydrogen ethyleru-
diaminetetraacetate (EDTA) in 1 L of water. Standardize th:--
solution against a standard calcium solution prepared b'-
dissolving 1.000 g of CaCO? and 2 mL of HC1 in water and
diluting to 1 L with water in a volumetric flask. Obtain an
exact factor for the EDTA solution. This factor is equal to
the milligrams of calcium equivalent to 1.00 mL of EDTA
solution. See Practice E 200.
TABLE 4 Precision Estimates (Water Insolubles)
0.002 to 0.005
0.01 to 0.04
0.15 to 0.35
0.001
0.01
0.01
65
65
65
Range. %
0.002 to 0.005
0.01 to 0.04
0.15 to 0.35
Coeffi-
cient of
Variation
91.7
42.2
20.5
Degrees
of
Freedom
9
8
9
Range. 95 %
Confidence
Level
300. relative
140, relative
65. relative
H-4
-------
E534
Factor = —
II' = calcium in aliquot, mg, and
!' = EDTA solution required for titration.mL.
27.4 Ammonium Chloride-Ammonium Hydroxide Solu-
:n>n—Add 67.5 g of ammonium chloride (NH4C1) to 570
mL of ammonium hydroxide (NH4OH) contained in a 1-L
\olumetric flask. Reserve this solution for use as described in
:".6 and 27.7.
27.5 Potassium Cyanide Solution (50 g/L)—Dissolve 50 g
of potassium cyanide (KCN) in water and dilute to 1 L with
water. Store in a borosilicate glass bottle. Caution: Potassium
i:\anide is extremely poisonous.
27.6 Magnesium Sulfate Solution (2.5 g/L)—Dissolve 2.5
g of MgSO4-7H2O in water and dilute to volume with water
in a 1-L volumetric flask. Determine the volume of EDTA
solution equivalent to 50 mL of MgSO4 solution as follows:
Pipet 50.0 mL of MgSO4 solution into a 400-mL beaker.
Add 200 mL of water and 2 mL of NH4C1:NH4OH solution
l27.4). Add 1 mL of KCN solution and a sufficient amount
of Eriochrome Black T Indicator solution or its equivalent.
Titrate the solution with EDTA solution while stirring with a
magnetic stirrer to the true blue end point. This gives the
\olume of EDTA solution equivalent to 50.0 mL of MgSO4
solution.
27.7 Buffer Solution—Pipet 50 mL of MgSO4 solution
into the volumetric flask containing the remaining
\H4C1-NH4OH solution (27.4). Add the exact volume of
EDTA solution equivalent to 50 mL of the MgSO4 solution.
Dilute to 1 L with water. Store the solution in a polyethylene
bottle.
27.8 Potassium Hydroxide Solution (600 g/L)—Dissolve
! M) g of potassium hydroxide (KOH) in 250 mL of water.
Cool and store in a polyethylene bottle.
28. Procedure
28.1 Using Table 5 as a guide, pipet two aliquots of stock
solution into 400-mL beakers to give a liter between 1 and
10 mL of standard EDTA solution. One aliquot is used to
determine calcium and the other for total calcium and
magnesium.
28.2 Dilute to 200 mL with water, if necessary, and place
on magnetic stirrer.
28.3 Total Calcium and Magnesium:
28.3.1 Add 5 mL of buffer solution, 1 mL of KCN
solution, and a sufficient amount of Eriochrome Black T
Indicator Solution or its equivalent.
28.3.2 Titrate with standard EDTA solution to a true blue
color.
28.3.3 Record the millilitres used as Titration 1 (7,).
TABLE 5 Stock Solutions (Calcium and Magnesium)
Stock Solution
Kansas rock salt
Nonhem rock salt
Southern rock salt
Evaporated salt
Punfied salt
Solar satt
Aliquot, mL
10
25
50
50
200
100
28.4 Calcium Onlr.
28.4.1 Add 2 mL of KOH solution, 1 mL of KCN
solution and stir for about 2 min to precipitate magnesium.
28.4.2 Add a sufficient amount of murexide solution or
an equivalent calcium indicator solution.
28.4.3 Titrate with standard EDTA solution to a true blue
color.
28.4.4 Record the millilitres used as Titration 1 ( T2).
29. Calculation
29.1 Calculate the weight percent of calcium as follows:
• u o, (7",) (factor) (0.1)
Ca. weight % = —•
o
where:
T2 = EDTA used to titrate calcium only, mL.
5 = weight of salt in aliquot, g.
29.1.1 See 27.3 for factor.
29.2 Calculate the weight percent of magnesium as fol-
lows:
Mg, weight % =
(7, - r.) (factor) (0.6064) (0.1)
where:
7, = EDTA used to titrate total calcium and magnesium,
mL, and
S = weight of salt in aliquot, g.
30. Report
30.1 Report the percentage of calcium to the nearest
0.001 %. Duplicate determinations that agree within 10 %
relative are acceptable for averaging (95 % confidence level)
(see Note 3).
30.2 Report the percentage of magnesium to the nearest
percent given in Table 6. Duplicate determinations which
agree within the percent absolute as specified in Table 6 are
acceptable for averaging (95 % confidence level) (see Note 4).
31. Precision and Bias
31.1 The following criteria should be used in judging the
acceptability of calcium results:
31.1.1 Repeatability (Single Analyst)—The coefficient of
variation of results (each the average of duplicate determina-
tions) obtained by the same analyst on different days, was
estimated to be 6.34 % relative at 48 degrees of freedom.
Two such averages should be considered suspect (95 %
confidence level) if they differ by more than 18 % relative.
31.1.2 Reproducibility (Multilaboralory)—The coeffi-
cient of variation of results (each the average of duplicate
determinations), obtained by analysts in different laborato-
ries has been estimated to be 9.82 % relative at 9 degrees of
freedom. Two such averages should be considered suspect
TABLE 6 Percentage of Magnesium
Range, %
Report to,
Check Limits
0.001 to 0.003
0.020 to 0.025
0.001
0.001
0.002
0.010
n-o
-------
(95 % confidence level) if they differ by more than 30 %
relative.
NOTE 3—The above precision statements are based on an
imerlaboratory study5 with five samples of sodium chloride covering the
range from 0.036 to 0.909 % calcium. One analyst in each of eight
laboratories performed duplicate determinations and repeated 1 day
later. Practice E 180 was used in developing these precision estimates.
31.1.3 The bias of this test method has not been deter-
mined.
31.2 The following criteria should be used in judging the
acceptability of magnesium results:
31.2.1 Repeatability (Single Ana!vst)—The standard de-
viation results (each the average of duplicates), obtained by
the same analyst on different days, has been estimated at the
values shown in Table 7. Two such averages should be
considered suspect (95 % confidence level) if they differ by
more than the values in Table 7 (Note 4).
31.2.2 Reproducibility (Multilaboratory)-The standard
deviation of results (each the average of duplicate determina-
tions), obtained by analysts in different laboratories has been
estimated at the values given in Table 7. Two such averages
should be considered suspect (95 % confidence level) if they
differ by more than the values in Table 7 (Note 4).
NOTE 4—The above precision statements are based on an
Intel-laboratory study5 of six samples of sodium chloride covering the
above ranges of magnesium. One analyst in each of ten laboratories
performed duplicate determinations and repeated 1 day later. Practice
E 180 was used in developing these precision estimates.
31.2.3 The bias of this test method.has not been deter-
mined.
SULFATE
32. Scope
32.1 This test method covers the gravimetric determina-
tion of the sulfate content of sodium chloride.
33. Apparatus
33.1 Gooch Asbestos Slurry.
33.2 Gooch Filtering Crucible and Holder.
33.3 Muffle Furnace.
33.4 Oven.
34. Reagents
34.1 Barium Chloride. Standard Solution 10% Bad,—
See Practice E 200.
34.2 Hydrochloric Acid Standard Solution 1 N HC1—See
Practice E 200.
34.3 Methyl Orange Indicator solution—See Practice
35. Procedure
35.1 Using Table 8 as a guide, pipet the recommended
aliquot of stock solution into a 400-mL beaker
35.2 Dilute to 200 mL, add a few drops of methyl orange
indicator solution and acidify with 1 mL of HC1 (1 + 1) if
necessary.
35.3 Heat solution gently to boiling and add 10 mL of
BaCl2 solsution dropwise while stirring.
35.4 Digest on a hot plate below the boiling point for 30
mm.
E534
35.5 Cool overnight.
35.6 Filter through a tared Gooch crucible previousk
prepared with an asbestos mat and ignited in a muffle
furnace at 800°C for 30 min. Transfer all the precipitate to
the crucible with a rubber policeman. Wash with portions of
hot water until washings are free of chlorides.
35.7 Dry the crucible at 110°C for 15 min, then ignite in j
muffle furnace at 800°C for 30 min.
35.8 Cool in a desiccator and reweigh.
36. Calculation
36.1 Calculate percentage of sulfate as follows:
Sulfate, % ='-x0.4115 x 100
D
where:
A = weight of precipitate, g, and
B - weight of salt in aliquot, g.
37. Report
37.1 Report the percentage of sulfate to the nearest
0.001 %. Duplicate determinations that agree within ">0 ^
relative are acceptable for averaging (95 % confidence level.
(see Note 5).
38. Precision and Bias
38.1 The following criteria should be used in judgin° the
acceptability of sulfate results:x °
38.1.1 Repeatability (Single Analyst)—The coefficient of
variation of results (each the average of duplicate determina-
tions) obtained by the same analyst on different days \va«
estimated to be 8.03 % relative at 60 degrees of freedom
Two such averages should be considered suspect (95 "
confidence level) if they differ by more than 23 % relative
38.1.2 Reproducibility (Multilaboratoryj—The coeffi-
cient of variation of results (each the average of duplicate
determinations) obtained by the same analyst on different
days, was estimated to be 8.01 % relative at 9 degrees of
freedom. Two such averages should be considered suspec:
(95 % confidence level) if they differ by more than 26 ~
relative.
NOTE 5—The above precision estimates are based on an interlabor-i-
tory study of s.x samples of sodium chloride covenng the range frorr.
0.016 to 2 030 % sulfate. One analyst in each of ten laboratory
performed duplicate determinations and repeated 1 day later Practice
E 180 was used in developing these precision estimates.
38.1.3 The bias of this test method has not been deter-
mined.
REPORTING OF ANALYSES
39. Scope
39.1 Analyses should be reported on a dry basis I:
analyses are on an as received sample, correction should be
made by converting to a dry basis. Sodium chloride purity is
determined by subtracting the total percentage of impurities
from 100. Moisture should be reported as a separate value.
H-6
-------
® E534
TABLE 7 _ Precision Estimates (Calcium and Magnesium)
Range, %
0.001 to 0 003
0.02 to 0.025
Standard
Deviation
0.00066
0.033
Repeatability
Degrees of
Freedom
28
26
Range. 95 %
Confidence
Level
0.002
0.096
Standard
Deviation
0.00147
0.0042
Reproduce bility
Degrees of
Freedom
8
8
Range, 95 %
Confidence
Level
0.005
0.014
TABLE 8 Stock Solutions (Sulfate)
Stock Solution
?OCK salt
^aporated salt
Untied evaporated salt
50:5' salt
Aliquot, ml
40
100
200
100
40.3 Report evaporated salt impurities to the third dec-
imal place and salt purity, by difference, to the second
decimal place.
40.4 Report purified salt impurities to the fourth decimal
place and salt purity, by difference, to the third decimal
place.
41. Conversion Factors
40. Procedure
40.1 Convert sulfate to calcium sulfate and the unused
calcium to calcium chloride unless the sulfate in sample
exceeds the quantity necessary to combine with the calcium.
In this case, convert the calcium to calcium sulfate and the
unused sulfate first to magnesium sulfate, and the remaining
sulfate, if any, to sodium sulfate. Convert the unused
rrugnesium to magnesium chloride.
40.2 Report rock and solar salt impurities to the second
decimal place and salt purity, by difference, to the first
decimal place.
The American Society for Testing and Materials takes no position respecting the validity ol any patent rights asserted in connection
with any item mentioned in this standard. Users ol this standard are expressly advised that determination of the validity of any such
patent rights, and the risk ot infringement of such rights, are entirely their own responsibility.
This standard is subject to revision at any time by the responsible technical committee and must be reviewed every five years and
if not revised, either reapproved or withdrawn. Your comments are invited either lor revision of this standard or for additional standards
and should be addressed to ASTM Headquarters. Your comments will receive careful consideration at a meeting of the responsible
technical committee, which you may attend. H you feel that your comments have not received a tair hearing you should make your
views known to the ASTM Committee on Standards, 1916 Race St., Philadelphia, PA 79703.
BaSO4
BaSO4
BaSO4
Ca
Ca
CaSO4
CaSO4
CaSO4
CaSO4
Mg
MgCl2
MgCl2
MgSO4
MgSO4
MgS04
SO4
x (
X (
X (
X .
x (
x (
x (
X (
X
X
X
X
X
X
X
X
).5832
),6086
!.3967
J.6064
J.2944
J.8153
J.8842
.0434
i.9173
.4296
.2641
.4299
.1311
.1800
.4173
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
S04
CaS04
Na,SO4
CaS04
Mg
Ca
CaCl,
MgS04
Na,S04
MgC!2
CaSO4
MgS04
CaSO4
MgO2
Na2SO4
CaS04
H-
-------
APPENDIX I
COST CONSIDERATIONS FOR ANTISKID MATERIALS
1-1
-\
-------
-------
COST CONSIDERATIONS FOR ANTISKID MATERIALS1
1.1 PURCHASE COSTS
The cost of a deicing material is dependent on availability, abundance, and amount
needed. The most readily available and widely used are rock salt and abrasives treated
as necessary with calcium chloride. A comparison of cost relating these to other
materials can be seen in Table 1-1.
Due to increased costs, alternative methods to salt and abrasives are often
avoided. Calcium magnesium acetate (CMA) is a favorable alternative to salt with regard
to its noncorrosive effects and high solubility, yet the purchase price is approximately
34 times that of salt. This would suggest the limited use of the product for areas such
as bridge decks where corrosion is a severe problem. Other chemicals that are inhibited
by cost are aluminum chloride and lithium chloride costing 20 and 333 times that of salt,
respectively.
I.2. APPLICATION AND CLEANUP COSTS
Purchase price is not the only consideration when implementing an ice and snow
control program. The application of salts has been estimated at $200 million a year (Iowa
Department of Transportation Planning and Research Division, 1980). This includes the
purchase of equipment and labor costs. Abrasives also require spring cleaning,
estimated at an annual $4 million (Anonymous, 1988). A favorable alternative method
would be one that was inexpensive to implement, yet required no cleanup.
Many of the alternative methods under consideration require large initial
investments and continued maintenance, thus not competing with the present
salt/abrasive system. Tables I-2 and I-3 show relative costs of some alternative deicing
chemicals and snow/ice removal methods, respectively. Embedded pipes, for example,
cost approximately $4.00 per square foot for a completely self-contained system. Annual
operational costs vary due to the heat source used. If purchased steam power is used,
one could expect a cost in excess of $0.133 per square foot (Jorgensen, 1964).'
Escalated costs would also result from repairs that require stripping of pavement.
I.3 CORROSION COSTS
Damage due to chemicals has been observed in automobiles, bridges, and in the
roadside environment. Corrosion to automobiles has been a widely researched topic, yet,
it is still hard to pinpoint in exact dollar figures the amount of damage (much of which is
vehicle depreciation). The following quotation describes the corrosion process, "After a
road salt spray, the automobile owner will observe a white salt film during the low
References listed in Section 8 of main report.
I-2
-------
TABLE 1-1. COST DATA FOR DEICING CHEMICALS
Material
Sodium chloride
Potassium chloride
Ammonium sulfate
Calcium chloride
Methanol
Urea
Urea:Ca formate (2:1)
Magnesium chloride
Fertilizer
Aluminum chloride
Ethylene glycol
Urearcalcium formateiformamide (1:1:1)
Calcium magnesium acetate (CMA)
Tetrapotassium pyrophosphate
Propylene glycol
Lithium chloride
Sand
Aggregate
Aggregate lime
Relative
cost8
1X
4X
4X
8X
8X
11X
14X
14X
19X
20X
27X
27X
34X
41 X
SOX
333X
—
—
—
Representative
cost ($/ton)
25-34.79
30.00
31.00
98.04
9.00
67.00
320.00
0.72/gal
600.00
1740.00
2-4.25
6.10
4.76
Reference
Wood (1983); phone survey
Wood (1983); Welch (1976)
Wood (1983); Welch (1976)
Wood (1983); phone survey
Wood (1983)
Wood (1983); Welch (1976)
Wood (1983)
Wood (1983); Anonymous (1988)
Wood (1983)
Wood (1983); Welch (1976)
Wood (1983); Transportation Research Board (1984)
Wood (1983)
Wood (1983); McElroy (1988)
Wood (1983)
Wood (1983)
Wood (1983); Welch (1976)
Phone survey
Welch (1976)
Welch (1976)
Compared to sodium chloride (rock salt).
-------
TABLE 1-2. RELATIVE CAPITAL COST OF
ALTERNATIVE DEICING CHEMICALS
Relative
cost
_ Chemical _ ($/lb)a
Methanol 0.7X
Ethanol 2X
Isopropanol 1 .6X
Acetone 1 .9X
Urea 1X
Formamide 4.4X
Dimethyl sulfoxide 5.9X
Ethyl carbonate (urethane) 4X
Verglimit-modified asphalt 0.5X
a Cost relative to urea (1979 dollars).
TABLE I-3. RELATIVE CAPITAL COST OF
ALTERNATIVE SNOW/ICE
REMOVAL METHODS
•Relative
cost
Method
Modified traffic paint 0.04X
Waterproofing compound (with silicone rubber) 0.06X
UCAR 0.1 2X
Infrared generator 3.1X
Embedded pipes • 1X
Embedded electrical elements 1 .4X
Conductive asphalt 0.4Xb
a Cost relative to embedded pipes.
b Cost will also include the purchase of a protective
coating.
I-4
-------
humidity hours, as the humidity increases it disappears, forming a corrosive liquid"
(Baboian, 1988). This process continues from day to night, and is accelerated with
warmer temperatures. It has been estimated that $200 to $300 in damage is done per
vehicle each year due to road salt (EPA, 1972).
Damage to bridge decks results from: scaling (due to chloride concentrations that
causes the freezing point of layers to be different and thus freeze at different times);
reinforcing steel corrosion (from chlorine ion concentration); dealimentation (cracks
resulting from pressure since the volume of iron oxide is larger than that of the original
steel); and spalling (occurring over reinforcing steel with portions of concrete removed)
of pavement (Welch, 1976). Another corrosive effect can be seen in the deterioration of
road markings such as paints, plastic, and tape (Slick, 1988). As an example, Brown
(1988) estimated an average repair cost of $40 per square yard (not including traffic
control) for bridge decks.
Other than fugitive dust, it is difficult to assess the cost of environmental damage
due to deicing chemicals. Damage can be observed in roadside vegetation and water.
Damage to trees and vegetation has been estimated at $50 million a year (Wood, 1983).
Water pollution costs weigh most heavily in the contamination of wells that require
replacing and also medical bills, not to mention court costs. Table I-4 illustrates
approximate costs of corrosion and environmental damages due to deicing salts.
I.4 GENERAL COST CONSIDERATIONS
As can be observed by comparing Tables I-2 and !-3, the capita! cost of alternative
ice and snow control methods place their possibility for use in the distant future.
Favorable aspects, such as noncorrosive and nonpplluting characteristics, have not as
yet outweighed the need for an economically feasible method of snow and ice control.
However, the unfavorable affects of salt and abrasives will promote the continued search
for more favorable cost-effective alternatives.
For now, it can be surmised that the most cost-effective ice/snow control method
is the continued use of rock salt and abrasives, it is advisable that calcium chloride be
used only when conditions warrant. The use of a substitute chemical (e.g., CMA) may
be necessary on limited areas such as bridge decks to inhibit the corrosive effects that
result from the use of salts and chlorides.
I-5
-------
TABLE 1-4. RECENT ESTIMATES OF CORROSION AND
ENVIRONMENTAL COSTS DUE TO CHEMICAL USE3
Infrastructure Environmental
Auto corrosion damage damage ($106/yr)
($106/yr) ($106/yr)
2,000
643
70,000b
500
160
200°
210
12d
a 1976-1981 dollars.
b McDonald, 1981.
c Murray & Ernst, 1976.
d Iowa DOT, 1980.
I-6
_v
-------
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA - 450/3-90-007
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Guidance Document for Selecting Antiskid Materials
Applied to Ice- and Snow-Covered Roadways
5. F-PORT DATE
Jnly-1-QQI
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Emission Standards Division (MD-13)
Office Of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-DO-0123
Work Assignment No. 6
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report describes a study whose primary purpose was to provide guidance on
methods to determine: (1) the physical properties and durability of antiskid
material selected for use on ice and snow-covered roadways; and (2) criteria for
defining the elements of an effective PM,n>emission control strategy associated
with the use of antiskid materials. This report is expected to be used by local,
State, and regional air pollution control agencies to identify antiskid materials
that are both durable and effective and produce fewer PM,n emissions.
Phase 1 of the study collected background information to derive preliminary
selection criteria for antiskid materials based on existing data. Phase 2 was a
laboratory evaluation to: (a) determine the physical properties of typical antiskid
materials and (b) conduct suitable tests of antiskid materials to' determine their
potential for the generation of PM,Q emissions. Based on limited data, the class
of materials that is relatively hard as indicated by low silt content, low abrasion
loss and high values for the Vickers hardness test, had the lowest overall potential
for silt generation. The best overall indicator of silt generation potential appear-
ed to be the Los Angeles abrasion loss.'
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Air pollution
Fugitive emissions
PM,Q emissions
Antiskid materials
Deicing chemicals
Paved roads
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS / This Report;
Unclassi f i ed
21. NO. OF PAGES
20. SECURITY CLASS /Tliispage.i
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4 — 77) PREVIOUS EDITION is OBSOLETE
-------
-------
-------
-------
|