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

-------

-------

-------

-------