Environmental Engineering Reference
In-Depth Information
2.4. Jarosite Waste Composite Characterization
The experimental data used to do the modeling as per mixture design is
shown in Table 2. The properties of jarosite waste composite products were
used in modeling and optimization. Properties such as density, shrinkage,
water absorption capacity and compressive strength were examined as per
ASTM (C67-99a) which is equivalent to that of Indian standard IS 3495 (part
3): 1992. Toxicity leachate characteristics of heavy metals in the jarosite waste
composites, were studied as per the USEPA approved toxicity leachate
characteristics procedure (TCLP) using Zero Head Space Extractor, Millipore,
USA.
2.5. Data Analysis
The proportions for the 3-component mixture (jarosite waste, clay and
CCRs) experiment initially were selected in terms of weight. The total weight
of the mixture was kept constant at 1 kg as required by the model. Since the
mass fractions must sum to unity, the component variables in a mixture
experiment are not independent. The minimum and maximum levels of each
component were chosen based on results of preliminary experiments. The
levels of the 3 mixture components are shown in Table 2.
2.6. Model Fitting and Validation
To verify the adequacy of the chosen model quantitatively and
graphically, the polynomial models described in classic mixtures approach
were fitted to data using analysis of variance (ANOVA) and least squares
techniques (Myers and Montgomery, 1995). From ANOVA the significance of
the treatment (mixture proportion) effect was obtained. Sequential F-tests were
performed, initially using linear model and adding terms. The degree of
freedom for each source was denoted as DF. The F-statistic was calculated for
each type of model, and the highest order model with significant terms to give
higher order equations. Significance was judged by determining the
probability that the F-statistic calculated from the data exceeded the theoretical
value (Hogg and Ledolter , 1989) When the probability was less than 0.05
(99.9% confidence level) or 0.01 (99.98% confidence level) termed as
significant results and their inclusion improved the model to compare the
Search WWH ::




Custom Search