QUANTITATIVE PETROGRAPHIC EVALUATION OF FINE AGGREGATE
Peter P. Hudec1 and Samuel Boateng2
ABSTRACT
Petrographic examination of coarse aggregate is accepted as a viable method for predicting its durability in concrete. No similar quantitative method existed for fine aggregate, although ASTM 295 does yield preliminary informaton of durability of fine aggregate. This paper proposes a quantitative petrographic analysis that can be used to quickly assess the natural fine aggregate's durability when used in concrete. The method of analysis is essentially the Ministry of Transportation, Ontario LS-616, which separates the fine aggregate into silicate, carbonate, shale, and chert fractions. The weighted percentages of these fractions were statistically compared to standard physical tests, including water adsorption and absorption, dry density, magnesium sulfate loss, and micro-Deval abrasion loss. Stepwise multiple regression equations were developed, which allowed calculation of expected magnesium sulfate and micro-Deval abrasion losses based on the proportion of the petrographic types. Mathematical and statistical techniques were also used to establish a Petrographic Number for Sand (PNS) which allows the classification of aggregate into good or poor categories based on their petrographic composition.
KEYWORDS: petrographic analysis, fine aggregate, microdeval abrasion, regression, prediction, durability, multivariate analysis
Hudec, P.P. and Boateng, S., 1995, Quantitative Petrographic Evaluation of Fine Aggregate, Cement, Concrete and Aggregates, CCAGDP Vol. 17, No. 2, pp. 107-112.
1
---------- Professor, Department of Geology, University of Windsor, Windsor, Ontario, Canada.2
---------- Ph.D. Candidate, Dept. of Geological Engineering, University of Missouri- Rolla, Rolla, Missouri.
INTRODUCTION
Petrographic analyses are used to describe and classify coarse aggregates according to rock type, mineralogic composition, texture and structure and also to detect defects and weaknesses that may have direct effects on durability. The petrographic technique is also useful for evaluating the quality of coarse aggregates in conjunction with standard abrasion and soundness tests [1]. The Ministry of Transport, Ontario (MTO) uses a quantitative petrographic analysis procedure (MTO Method LS-609, 1994) to assign a 'Petrographic Number' or PN to a coarse aggregate sample; a maximum PN is designated beyond which the aggregate is not accepted for a particular concrete use. The PN has proven to be a reliable and rapid method for classifying coarse aggregates, as shown by its good relationship to other physical tests [2]. Rogers [3] traces the development of the petrographic examination of aggregate and concrete in Ontario.
Even though fine aggregates play an important role in concrete durability, relatively little work has been done on the use of the MTO petrographic technique in the appraisal of fine aggregate quality. The presence of mica in fine aggregate may, for instance, lead to lower compressive strength of concrete [4] by inhibiting good aggregate-cement bond. According to Dolar-Mantuani [5], some alkali-reactive particles, such as opal, are more reactive in smaller sizes (<600µm) and their presence in fine aggregate may result in expansion and cracking when used in concrete. Shale particles in natural sand are known to cause surface pop-outs in concrete and bituminous surfaces.
The MTO has adopted a qualitative method for analysis of fine aggregate [6]. The method basically groups the aggregate into carbonate, silicate, chert, shale and cemented fractions on all of the standard sieve sizes. The weighted percentage of these fractions is reported, but no passing limits are applied. The purpose of this study was to quantify the LS-616 method somewhat along the lines of the LS-609 method, and derive a Petrographic Number for (natural) Sand (PNS) which could be used to estimate the probable quality of the aggregate. The work being reported here is part of a larger study by Boateng [7]. To better understand the results of the study, a description of the petrographic analysis of coarse aggregate, the MTO Method LS609, will be outlined, and its relationship to other durability tests presented.
Quantitative Petrographic Evaluation of Coarse Aggregate
The MTO Method LS609 [8] was developed initially in 1948 by Bayne and Greenland [9] who subdivided the aggregate into three groups: satisfactory, questionable, and unsatisfactory. The subdivisions were made on basis of a service record of the material in concrete and asphalt pavements. Numeric weighting was experimented with during the 1950's, and a workable method developed in 1955 [1]. The method was most recently evaluated in 1981 [10,11,12], and some modifications were adopted.
All coarse aggregates are subdivided into good, fair, poor, and deleterious categories, and weighting factors of 1, 3, 6, and 10 are assigned to each respectively. The percent weight of each category is multiplied by the factor, and the product summed. For example, an aggregate sample consisting of 70 percent good, 20 percent fair, 8 percent poor, and 2 percent deleterious (70x1 + 20x3 + 8x6 + 2x10) would have a PN of 198. Aggregates with PN > 200 perform poorly, whereas those with PN <140 generally perform well in concrete.
Petrographic Evaluation of Fine Aggregate
An attempt was made to apply the above technique for coarse aggregate to the fine aggregate. The early attempts failed mostly because it was difficult to correlate the field performance of concrete to its fine aggregate content. Very few examples exist where fine aggregate can be considered as the prime cause of concrete deterioration. The only exceptions are alkali reactivity of certain silicate particles in sand, and surface pop-outs traced to shale content of the sand.
In 1983, a procedure was developed [8] for qualitative assessment of fine aggregate. A minimum of 200 particles from each sieve size are placed on a glass slide and examined under a binocular microscope equipped with a mechanical linear stage. Point counting method is employed in determining percentages of various rock and mineral types. Silicate and carbonate rocks and minerals in the sand are counted separately, and are considered to be innocuous. Minerals and rocks such as shale or clay, mica or mica schist, chert and jasper, and contaminats (glass, slag, coal, etc.) are considered problematic. They may indicate potential problems, such as lack of freeze-thaw durability and a potential for alkali-aggregate reaction. Relative percentage of these groups is determined. No specifications exist; the results are considered as advisory, and used to suggest further testing of suspect sand.
The intent of this study was to quantify the results of the MTO Method LS-616 so that some limits, similar to those for the coarse PN, could be assigned.
SAMPLING AND TESTING
Thirty natural fine aggregates were sampled from various locations in southern Ontario. All samples are from glacio-fluvial deposits of Quaternary age, and include glaciolacustrine, and ice-contact fluvial deposits. The sampling was done in two stages: The original sampling was done by the MTO to collect samples for the micro-Deval abrasion study. The magnesium sulfate soundness and micro-Deval abrasion test results were obtained from this set of samples at the MTO laboratories. The sites were subsequently re-sampled by the first author to provide additional material for testing. The physical tests such as the petrographic analysis, adsorption, and absorption were done on the second set at Windsor, and a comparison micro-Deval test was done at MTO.
Petrographic Analysis
The petrographic analysis was conducted according to the Ministry of Transport, Ontario (MTO) procedure for petrographic analysis of fine aggregates (MTO method LS-616). The test was done on the collected material on minimum of 200 particles on each of the standard sieve sizes:
| Sieve size, mm | Sieve No. | |||
| Pass | Retain | Pass | Retain | |
| 4.75 | 2.36 | 4 | 8 | |
| 2.36 | 1.18 | 8 | 16 | |
| 1.18 | 0.6 | 16 | 30 | |
| 0.6 | 0.3 | 30 | 50 | |
| 0.3 | 0.15 | 50 | 100 | |
| 0.15 | 0.08 | 100 | 200 | |
All particles were identified under a binocular microscope and the total count for each rock or mineral type was recorded. Experience in our lab with most sands has shown that the minerals passing the No.100 had little effect on the final results of the LS-616 test. The analysis was difficult to perform, because of the small particle size and difficulty in accurately identifying the mineralogic or petrographic charateristis of the particles. Also, quartz and carbonate minerals tended to dominate the finer sizes - material that is largely innocuous in mortar. Mica is a known deleterious component in sands and does concentrate in the fine sizes. It was analyzed, but preliminary statistical evaluation showed it to play no role in sand quality calculations. Consequently, for the purposes of this paper, the material passing the No.100 sieve was omitted from the analysis.
The percentage of each material type was calculated for each size fraction. The weighted percentage of a material type (for example, shale) for each sieve fraction was calculated as shown below:
Sum of weighted percent of shale on each sieve = (x.y)/100 ------ (1)
where,
x= percent of total sample retained on a particular sieve
y=percent shale in that fraction
Similar calculations were done for silica, carbonate, chert, and cemented particle fractions of the fine aggregate. The results are reported in Table 1.
Water Adsorption
The adsorption test was conducted on 200 g of the fine aggregate sample, graded according to ASTM C227. Prior to testing the samples were washed over a 75-µm sieve to remove surface coatings and oven-dried at 110±5oC. While still warm, each sample was tightly sealed in a zip-lock plastic freezer bag to avoid adsorption of atmospheric moisture while cooling to room temperature. The cooled sample was weighed, and placed in an airtight humidity chamber. A relative humidity of 95 percent was maintained by means of a saturated aqueous solution of hydrated cupric sulfate (CuSO4.nH2O) at a temperature of 20±2oC for seventy-two hours.
Adsorption was reported as the percentage weight change over the initial dry weight, and is given in Table 1. The adsorption data of eight control samples indicated a mean adsorbtion of 0.541% with a standard deviation of 0.017%, as shown in Table 1a.
Water Absorption and Bulk Density
Following the adsorption test, water absorption and bulk density of the samples was determined according to ASTM C128. The results are given in Table 1.
Micro-Deval Abrasion Test
The micro-Deval Abrasion test is a modification of a similar test developed in France [13,14] for the evaluation of coarse aggregates. The test was performed in the MTO labs in Toronto as follows: 700 g of the sample was washed over a 75-µm sieve and oven-dried. A 500±5 g of the oven-dried sample, graded according to ASTM C227, was used for the test. Before tests began, each sample was immersed in water for 24 hours at room temperature. Excess water was drained and the sample was placed in a steel jar with 1250 g of steel balls (with a diameter of 9.5mm each) and 750ml of water. The steel jar was sealed and rotated at 100 rev/min for fifteen minutes. The sample was washed over the 75-µm sieve again and oven-dried to constant weight. The micro-Deval abrasion loss was reported as the percent change in weight relative to the original sample.
As mentioned above, the two sets of samples were collected from the same locations two years apart. Boateng [7] investigated the reproducibility of the micro-Deval abrasion test by comparing two sets of results. Although samples were taken from different location in the same pit, and were not fully identical, the tests correlated at R = 0.89, at a significance level of 99 percent and a slope of 0.96. The slope of one is ideal. Because of this correlation, it was felt that the results from the two sets of samples are directly comparable. The results of both tests are given in Table 1.
Magnesium Sulfate Test
The magnesium sulfate test was conducted on the first set of samples at the MTO laboratories according to the LS-606 method.
The results of the above tests are given in Table 1.
ANALYSIS OF RESULTS
The results of the tests were subjected to bi-variate and multi-variate analysis to determine relationships among them. The primary purpose was to establish an empirical formula which would allow the petrographic analysis results on the fine aggregate to predict its quality.
The quality of sand should be related to the proportion of poor aggregate that it contains. In the case of Ontario aggregates, the major poor material in the sand is shale and chert derived from the Paleozoic formations by glacial action.
Various empirical combinations and ratios of the carbonate, silicate, shale, and chert percentages were tried to obtain the best linear, logarithmic, or quadratic correlation with other measured properties of the sand. The single number so obtained was designated as the Petrographic Number of Sand (PNS). The best correlations were obtained where the PNS was obtained by the following empirical formula:
PNSemp = 100*Log10(1+(shale% + chert%)/silica%) ---- (2)
The equation suggests that the ratio of combined shale and chert to the silica content of the sand should give a good indication of the fine aggregate quality. Carbonate content seems to have no role in sand durability.
A more accurate method of computing the PN for fine aggregate is available through the use of multiple regression analysis. micro-Deval data was used as a dependent variable, since the micro-Deval results have been shown to be good indicators of fine aggregate durability (Rogers et al, 1990), and the petrographic data as the independent variables. The equation so derived is:
PNSreg = 7.8 + 0.511*chert% + 0.396*shale% R= 0.88 (3)
The equation shows that the chert and shale content alone are the main quality determinants for natural sand durability. Silica and carbonate content were found to play no significant role.
If the fine sand petrographic data correlates with micro-Deval abrasion, it can be used to predict the expected micro-Deval abrasion loss. To derive this relationship, a step-wise regression analysis was performed, with the measured micro-Deval data as dependent variable, and the petrographic data as independent variables (equation 4). Note that shale, chert, and carbonate are the picked independent variables. The equation has a high, positive correlation coefficient of 0.90. The derived equation is:
CalcMDA = 6.6 + 0.431*chert% + 0.416*shale% + 0.0376*carbonate% (4)
In a similar way, the calculated MgSO4 loss can be predicted from the petrographic data, as shown in Table 2. In this case, shale and chert content are seen as the main influence on the sulfate loss. The correlation coefficient is highly significant at 0.76, and the equation derived is:
CalcMgSO4 = 9 + 0.396*Chert% + 0.452*Shale% (5)
Note that equations 3, 4, and 5 are somewhat similar, and that they are heavily dependent on the shale and chert content. These equations were derived for glacially-derived deposits in SW Ontario. For other geologic environments, different deleterious materials can be similarly identified, and similar equations derived.
The four equations were used to calculate the two types of Petrographic Number of Sand (PNS), and the predicted micro-Deval and Magnesium sulfate results based on the petrographic analysis. The results are presented in Table 2.
A correlation matrix was computed to determine the relationships of the calculated variables to each other and to the physically determined ones, and is shown in Table 3. Most relationships are significant to the 99.9% level and least significance is at 98.6% level.
The graphical relationships between the calculated parameters and the physically measured ones are given in Figures 1 through 4. Figure 1 illustrates the relationship between the PNS calculated by the empirical and the regression equations. As can be seen, there is a very good correlation between the two. This suggests that either method can be used to get a good estimate of the PNS. Moreover, as the petrographic database from which the regression PNS are calculated increases, Equation 2 can be continually refined to give a more accurate relationship. This is not the case with a fixed relationship expressed in Equation 1.
Figure 2 gives the relationship between the two PNS types and the micro-Deval abrasion results. The graph shows that both PNSs are reliable indicators of fine aggregate durability as expressed by the micro-Deval abrasion results.
Likewise, Fig. 3 shows the relationship between the two petrographic numbers and the magnesium sulfate results. Both figures confirm that the PNS method can reliably establish the quality of fine aggregate.
If a permissible limit for magnesium sulfate loss of 15% is taken (ASTM C33), then the equivalent limit of PNSemp and PNSreg are 16 and 10 respectively. Likewise, if a permissible limit for micro-Deval abrasion of 20% is taken (CSA 1994 for concrete), then the equivalent limits of PNSemp and PNSreg are 30 and 26 respectively.
Figures 4 and 5 show the relationship between laboratory test resuls for micro-Deval abrasion and magnesium sulfate tests, compared to values determined for those tests obtained by regression calculations based on the petrographic analysis results. The regression line in both instances has a slope of 1, indicating that there is 1:1 correspondence, on the average, between laboratory measured and statistically calculated results.
DISCUSSION AND CONCLUSIONS
The results presented show that quantitative petrographic analysis of fine aggregate data can be quantified to help identify sand quality. The method is relatively simple, rapid, and requires minimal equipment. It does, however, require a petrographer who can differentiate various lithic types found in natural sand. The indicated Petrographic Number for Sand (PNS) is either or 26 or 30, depending on the PNS calculation method used, using micro Deval abrasion results as a guide. The PNS method can be adapted to field use for quick estimate of the fine aggregate quality. The quantitative Petrographic Number for Sand can also be used to predict expected results of other quantitative laboratory tests.
The samples studied and equations derived were for the glacially-derived deposits of Southwestern Ontario, where the principal deleterious components in sand are shale and chert. Other areas with different lithologic components in sand, such as mica and volcanics, may require similar studies to establish the empirical and regression formulas that can be applied to their specific lithologic material mix. Alternately, a larger study of sands from differing geologic environments may yield more general equations which would contain all potential deleterious components.
The multiple regression method of determining the Petrographic Number of sand is preferable to the empirical formula approach, since it can be continually improved as new data is added. PNS could, for instance, be applied to specific class of deposits, or to a regional group of deposits to determine the permissible content of undesirable material as a quality control measure.
ACKNOWLEDGMENTS
The research described above was supported by a grant from the National Science and Engineering Research Council of Canada. Chris Rogers of the Ontario Ministry of Transportation provided the initial samples and test results, and also performed the micro-Deval abrasion tests on the second set of samples.
REFERENCES
ASTM C-227, 1991. Standard Test Method for Potential Alkali Reactivity of Cement Aggregate Combinations (Mortar Bar Method). Annual Book of ASTM Standards, vol.04.02, pp.126-130.
Bayne, R.L. and Brownridge, F.C., 1955. Petrographic Analysis for Determining Quality of Coarse Aggregates. Proceedings of the 36th Convention, Canadian Good Roads Association, pp.114-122.
Boateng, S., 1992. Petrographic Analysis and Durability of Fine Aggregates. Unpublished M. A. Sc. Thesis, University of Windsor, 139p.
Dolar-Mantuani, L., 1983. Handbook of Concrete Aggregates -- A Petrographic and Technological Evaluation. Noyes Publications, 345p.
Hudec, P. P., 1983; Aggregate tests - their relationship and significance; Durability of Building Materials, Elsevier Publishing Co., v. 1, pp. 275-300.
MTO Method LS-609, 1994, Procedure for the Petrographic Analysis of Coarse Aggregate. Laboratory Testing Manual, Ministry of Transportation, Ontario, vol. II, 20p.
MTO Method LS-616, 1994. Procedure for the Petrographic Analysis of Fine Aggregate. Laboratory Testing Manual, Ministry of Transportation, Ontario, vol.II, 6p.
MTO Method LS-619, 1994. Method of Test for the Resistance of Fine Aggregate to Degradation by Abrasion in the Micro-Deval Apparatus, Laboratory Testing Manual, Ministry of Transportation, Ontario, vol.II, 7p.
Rogers, C.A., Bailey, M.L., and Price, B., 1990. Micro-Deval Test for Evaluating the Quality of Fine Aggregate for Concrete and Asphalt. MTO Draft Report, Ministry of Transportation, Ontario, 26p.
Schmitt, J.W., 1990. Effects of Mica, Aggregate Coatings, and Water Soluble Impurities on Concrete. Concrete International, The Magazine of the American Concrete Institute. vol.12, No.12, pp.54-58.
Table 1. Results of Petrographic Analysis and Physical Durability Tests
| SAMPLE | SILICATE | CARBONATE | SHALE | MICA | CHERT | CEMENTED PARTICLES | MDA MTO |
MDA WINDS |
MgSO4 | ABSORB | ADSORB | BulkDens |
|
9100 |
24.00 |
73.76 |
0.21 |
0.00 |
2.03 |
0.00 |
9.6 |
9.9 |
6.5 |
0.66 |
0.16 |
2.658 |
|
9102 |
21.19 |
67.20 |
0.00 |
0.00 |
11.61 |
0.00 |
14.9 |
16.0 |
18.5 |
1.24 |
0.37 |
2.645 |
|
9109 |
30.46 |
55.52 |
1.88 |
0.00 |
11.34 |
0.81 |
11.3 |
10.4 |
13.2 |
0.70 |
0.23 |
2.654 |
|
9121 |
16.16 |
74.74 |
0.00 |
0.00 |
9.09 |
0.00 |
15.5 |
17.9 |
16.2 |
1.23 |
0.19 |
2.643 |
|
9122 |
22.61 |
63.70 |
0.00 |
0.21 |
13.48 |
0.00 |
14.4 |
13.2 |
13.8 |
1.09 |
0.18 |
2.658 |
|
9123 |
17.79 |
70.89 |
1.92 |
0.00 |
8.56 |
0.83 |
13.7 |
15.0 |
9.2 |
6.00 |
0.23 |
2.634 |
|
9125 |
26.89 |
65.96 |
0.00 |
0.00 |
6.73 |
0.42 |
15.6 |
14.3 |
15.0 |
1.56 |
0.26 |
2.605 |
|
9126 |
38.25 |
56.34 |
0.00 |
0.00 |
5.40 |
0.00 |
7.9 |
7.0 |
5.5 |
0.40 |
0.12 |
2.672 |
|
9129 |
28.13 |
66.57 |
0.00 |
0.00 |
5.31 |
0.00 |
12.2 |
11.7 |
11.6 |
1.09 |
0.18 |
2.677 |
|
9130 |
24.48 |
22.51 |
39.70 |
0.00 |
11.89 |
1.42 |
31.3 |
26.6 |
33.9 |
3.70 |
1.02 |
2.429 |
|
9132 |
16.37 |
49.37 |
0.38 |
0.00 |
33.88 |
0.00 |
22.3 |
15.2 |
20.6 |
1.76 |
0.17 |
2.625 |
|
9135 |
19.64 |
58.75 |
16.44 |
0.00 |
5.17 |
0.00 |
12.8 |
11.4 |
13.1 |
1.25 |
0.29 |
2.608 |
|
9138 |
21.02 |
68.14 |
0.83 |
0.00 |
8.98 |
1.03 |
10.9 |
12.8 |
12.7 |
1.13 |
0.26 |
2.623 |
|
9143 |
24.34 |
67.01 |
1.84 |
0.00 |
6.63 |
0.19 |
10.9 |
12.6 |
10.7 |
0.76 |
0.18 |
2.657 |
|
9144 |
99.69 |
0.31 |
0.00 |
0.00 |
0.00 |
0.00 |
4.0 |
6.2 |
6.4 |
0.48 |
0.08 |
2.660 |
|
9147 |
18.17 |
72.55 |
0.00 |
0.00 |
9.07 |
0.21 |
15.0 |
16.3 |
14.8 |
1.25 |
0.13 |
2.634 |
|
9154 |
15.82 |
73.07 |
0.00 |
0.00 |
10.97 |
0.14 |
18.7 |
20.2 |
16.0 |
1.67 |
0.16 |
2.620 |
|
9157 |
15.54 |
79.88 |
0.00 |
0.00 |
4.58 |
0.00 |
9.3 |
13.7 |
12.3 |
0.97 |
0.21 |
2.642 |
|
9158 |
48.86 |
50.56 |
0.00 |
0.00 |
0.00 |
0.58 |
11.8 |
11.5 |
17.8 |
1.07 |
0.27 |
2.646 |
|
9159 |
98.61 |
0.00 |
0.00 |
0.00 |
0.00 |
1.39 |
6.0 |
6.1 |
4.3 |
0.521 |
0.15 |
2.654 |
| SAMPLE | SILICATE | CARBONATE | SHALE | MICA | CHERT | CEMENTED PARTICLES | MDA MTO |
MDA WINDS |
MgSO4 | ABSORB | ADSORB | BulkDens |
|
9165 |
47.52 |
44.79 |
1.65 |
0.00 |
5.21 |
0.84 |
17.6 |
18.8 |
21.3 |
1.420 |
0.44 |
2.628 |
|
9173 |
56.23 |
37.04 |
0.75 |
0.00 |
5.77 |
0.21 |
10.6 |
8.6 |
11.2 |
0.725 |
0.14 |
2.676 |
|
9177 |
23.07 |
73.23 |
0.00 |
0.00 |
3.50 |
0.21 |
8.5 |
8.3 |
5.7 |
0.402 |
0.08 |
2.689 |
|
9183 |
99.79 |
0.00 |
0.00 |
0.21 |
0.00 |
0.00 |
6.7 |
5.7 |
5.5 |
0.746 |
0.07 |
2.627 |
|
9187 |
99.58 |
0.00 |
0.00 |
0.42 |
0.00 |
0.00 |
8.7 |
5.3 |
13.1 |
0.583 |
0.06 |
2.716 |
|
9188 |
99.81 |
0.19 |
0.00 |
0.00 |
0.00 |
0.00 |
6.9 |
6.0 |
10.2 |
0.543 |
0.08 |
2.688 |
|
9189 |
99.38 |
0.00 |
0.00 |
0.63 |
0.00 |
0.00 |
6.7 |
6.6 |
8.3 |
0.543 |
0.14 |
2.675 |
|
9190 |
78.71 |
0.94 |
0.00 |
0.21 |
3.48 |
0.00 |
6.3 |
6.6 |
4.1 |
0.361 |
0.08 |
2.670 |
|
9232 |
99.72 |
0.00 |
0.00 |
0.00 |
0.00 |
0.28 |
7.2 |
6.4 |
14.3 |
0.766 |
0.24 |
2.661 |
|
9243 |
99.17 |
0.00 |
0.00 |
0.00 |
0.83 |
0.00 |
6.7 |
14.1 |
7.0 |
0.281 |
0.12 |
2.705 |
Table 1a Mean and standard deviation of adsorption test runs of 'standard' sand.
|
Run No. |
% Adsorbed |
|
1 |
0.56 |
|
2 |
0.54 |
|
3 |
0.57 |
|
4 |
0.55 |
|
5 |
0.53 |
|
6 |
0.52 |
|
7 |
0.54 |
|
8 |
0.52 |
|
Mean |
0.541 |
|
St. Dev. |
0.017 |
Table 2: Calculated Parameters from Fine Aggregate Petrographic Analysis
(For legend to olumn headings, see Table 3.)
|
Sample No |
MDAMTO | MDA Calc | PNSEmp | PNSReg | MgSO4 Calc |
|
9100 |
9.6 |
10.3 |
3.9 |
8.9 |
9.9 |
|
9102 |
14.9 |
14.1 |
19.0 |
13.7 |
13.6 |
|
9109 |
11.3 |
14.4 |
15.7 |
14.3 |
14.3 |
|
9121 |
15.5 |
13.3 |
19.4 |
12.4 |
12.6 |
|
9122 |
14.4 |
14.8 |
20.3 |
14.7 |
14.3 |
|
9123 |
13.7 |
13.8 |
20.1 |
12.9 |
13.3 |
|
9125 |
15.6 |
12.0 |
9.7 |
11.2 |
11.7 |
|
9126 |
7.9 |
11.1 |
5.7 |
10.6 |
11.1 |
|
9129 |
12.2 |
11.4 |
7.5 |
10.5 |
11.1 |
|
9130 |
31.3 |
29.1 |
49.2 |
29.6 |
31.7 |
|
9132 |
22.3 |
23.2 |
49.0 |
25.2 |
22.6 |
|
9135 |
12.8 |
17.9 |
32.2 |
17.0 |
18.5 |
|
9138 |
10.9 |
13.4 |
16.6 |
12.7 |
12.9 |
|
9143 |
10.9 |
12.7 |
13.0 |
11.9 |
12.5 |
|
9144 |
4.0 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9147 |
15.0 |
13.2 |
17.6 |
12.4 |
12.6 |
|
9154 |
18.7 |
14.1 |
22.9 |
13.4 |
13.3 |
|
9157 |
9.3 |
11.6 |
11.2 |
10.1 |
10.8 |
|
9158 |
11.8 |
8.5 |
0.0 |
7.8 |
9.0 |
|
9159 |
6.0 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9165 |
17.6 |
11.2 |
5.9 |
11.1 |
11.8 |
|
9173 |
10.6 |
10.8 |
4.8 |
11.0 |
11.6 |
|
9177 |
8.5 |
10.9 |
6.1 |
9.6 |
10.4 |
|
9183 |
6.7 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9187 |
8.7 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9188 |
6.9 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9189 |
6.7 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9190 |
6.3 |
8.1 |
1.9 |
9.6 |
10.4 |
|
9232 |
7.2 |
6.6 |
0.0 |
7.8 |
9.0 |
|
9243 |
6.7 |
7.0 |
0.4 |
8.2 |
9.3 |
Table 3. Correlation Matrix of Measured and Calculated Fine Aggregate Parameters
|
ABSORB |
ADSORB |
MgSO4 CALC |
MDA CALC |
DRYDENS |
MDAMTO |
MDAWIN |
MgSO4 |
PNSEMP |
PNSREG |
|
|
ABSORB |
1.00 |
0.52 |
0.53 |
0.55 |
-0.57 |
0.60 |
0.58 |
0.44 |
0.56 |
0.52 |
|
ADSORB |
0.52 |
1.00 |
0.78 |
0.71 |
-0.88 |
0.77 |
0.72 |
0.82 |
0.59 |
0.70 |
|
MgSO4 CALC |
0.53 |
0.78 |
1.00 |
0.97 |
-0.83 |
0.87 |
0.70 |
0.76 |
0.93 |
0.99 |
|
MDA CALC |
0.55 |
0.71 |
0.97 |
1.00 |
-0.78 |
0.90 |
0.76 |
0.74 |
0.97 |
0.98 |
|
DRYDENS |
-0.57 |
-0.88 |
-0.83 |
-0.78 |
1.00 |
-0.79 |
-0.71 |
-0.73 |
-0.71 |
-0.77 |
|
MDAMTO |
0.60 |
0.77 |
0.87 |
0.90 |
-0.79 |
1.00 |
0.89 |
0.89 |
0.86 |
0.88 |
|
MDAWIN |
0.58 |
0.72 |
0.70 |
0.76 |
-0.71 |
0.89 |
1.00 |
0.79 |
0.72 |
0.70 |
|
MgSO4 |
0.44 |
0.82 |
0.76 |
0.74 |
-0.73 |
0.89 |
0.79 |
1.00 |
0.69 |
0.75 |
|
PNSEMP |
0.56 |
0.59 |
0.93 |
0.97 |
-0.71 |
0.86 |
0.72 |
0.69 |
1.00 |
0.96 |
|
PNSREG |
0.52 |
0.70 |
0.99 |
0.98 |
-0.77 |
0.88 |
0.70 |
0.75 |
0.96 |
1.00 |
|
LEGEND |
|
|
ABSORB |
Absorption, % |
|
ADSORB |
Adsorption, % |
|
MgSO4 CALC |
Calculated MgSO4 |
|
MDA CALC |
Calculated micro-Deval Abrasion |
|
DRYDENS |
Dry Density |
|
MDAMTO |
micro-Deval Abrasion on original MTO samples |
|
MDAWIN |
mico-Deval Abrasion, Windsor Samples |
|
MgSO4 |
Magnesium Sulfate Loss, MTO samples |
|
PNSEMP |
Petrographic Number, Sand, empirical equation |
|
PNSREG |
Petrographic Number, Sand, Regression Equation |
FIGURES
|
|
