Figure 2. Sulfate to Iron Ratio versus Time
May 3, 1983
The earth-moving operations of the mining industry continually expose deeply buried minerals to weathering. Pyrite material contained in these mine spoils reacts with oxygen and water to produce sulfuric acid and solable iron salts. These compounds dissolve in water and produce acid mine drainage.
The extent of this pollution is awesome. Annually approximately 500 billion gallons of mine drainage containing 5 to 10 million tons of acid pollute over 10,000 miles of surface streams and more than 15,000 acres of impounded waters. I Methods to eliminate acid pollution can be divided into three procedures:
The success of these treatments is confused by lack of related studies. Each technique has data which supports its effectiveness; however, the tests are normally run on different mine spoils under different conditions. This makes comparative evaluations very difficult. The mine operator does not know what to expect from these treatments; or how they eliminate the acid pollution.
The object of this research project is to study acid mine drainage production in a controlled experimental situation and to evaluate the effectiveness of a variety of treatment techniques.
The experimental design carefully tests the effectiveness of the various abatement techniques against a set of controls in an idealized conditions. Acid producing material is contained in 35 gallon white plastic barrels. Each barrel is fitted with a plastic distribution plate supported above a flow plate to an exit port. Liquid flows through the material, passes through the distribution plate down the flow plate, through the exit port equipped with an air trap and collects in a sealed 5 gallon plastic can. This design insures that water flow and air flow are single directional. (See Figure 1) The particle size is less than 0.1 the diameter of the barrel which insures negligible water channeling effects. Twelve of these barrels were set up.
The acid producing material used in these experiments is cleaning plant wastes from Island Creek Coal Company's Alpine mine in Dobbin, WV. A complete characterization of the material can be seen in Table 1.
The previously described columns were each filled with three hundred pound samples of thoroughly mixed cleaning plant waste. The twelve barrels were divided into four groups of three barrels each. Of these four groups, three were treated with ameliorates and one group was used as a control. The ameliorates used. were:
The barrels were arranged in mixed positions so the possibility of preferential precipitation during normal weathering conditions was eliminated. Effluents from the barrels were collected 2 days after each rainfall
Figure 2. Sulfate to Iron Ratio versus Time
Table 1
Table 2
event and analyzed for pH, sulfate, iron, manganese, calcium, and magnesium ions. The acidity of the leachate was determined by titration with standard sodium hydroxide and the neutralization in calcium carbonate equivalents to pH: 5.5, 7, and 8.3 is recorded.
Part I - Evaluation Procedure for Abatement Techniques.
It is very difficult to analyze raw data and make a conclusive statement as to reaction consistency with time; however, we can accomplish this by comparing derived equations which fit the data. Equations can be fitted to these trends by standard computer procedures and these equations will be used to predict effects. One such procedure commonly used to test the effectiveness of treatments compares the cumulative effect of a system against the' time period of data collected. With our data this is accomplished by plotting cumulated acid contributions as indicated by sulfate concentration versus time. These data are easily computed to equations of form.
y = m t b + C
A set of equations is determined for each treatment and control for the time period of data collected. By comparing the derived relationship for the treatment to the control, we can evaluate the effectiveness of the treatment. These comparisons are usually made by dividing the area under the curve for the treatment by the area under the curve for the control. The closer the number is to one, the lower the effectiveness of the treatment. Equations, correlation coefficients, and areas of the curves derived for the data can be seen in the Appendices.
Table 2 contains the ratio of the integrated areas of the treatments to the control. The effectiveness of the various treatments can be seen by comparing the ratio of the areas.
A more familiar procedure to compare abatement techniques is to determine the calcium carbonate equivalent required to neutralize the effluent stream. The procedure used in these comparisons is the same as that used above: 1) cumulative data is plotted against time, 2) an equation is fitted to the points, 3) the equations are integrated, and 4) the ratio of the areas of the treatments to the controls are used to scale the effectiveness.. Again, the closer the ratio is to one, the lower the effectiveness of the treatment. Table 3 lists those ratios and again a similar trend is seen as indicated by the sulfate equation.
Part II - Evaluation Procedure for Testing the Acid Producing Reaction
By studying its control data, changes in the acid production reactions processes can be followed. Acid production reactions can be monitored by comparing the ratios of ion concentrations of species dependent upon pyrite oxidation. These two ions are sulfate and total iron. A simple procedure to determine this relationship is 1) to derive the cumulative equations for iron production and sulfate production and 2) subtract the consecutive points of each plot for the same time period, which produces the concentration of each ion component for that time period and divide the concentration of one ion by the other. If a single acid producing reaction dominates in acid mine drainage formation, then the ratios of these two ions contributors will be nearly constant. If more than one acid producing process is dominant the slope will not be zero. The result of this procedure can be seen in Figure 2.
Table 3
There are no conclusions in this. report because as noted the work is still in progress; however, several hypotheses can be drawn from the interpreted data.
First, as the plot of the ratio of [SO 4- 2 ]/[Fe] demonstrates; more than one chemical process is responsible for acid production. The Plot would be linear with zero slope if a single process were responsible for acid production. The appearance of an apex shows that at least one other process which produces high concentrations of soluble sulfate compared to soluble iron is occurring. This phenomena took place in late September and early October. Since reaction products leach from the barrel about a month after the reaction, it is probable that this disproportionate sulfate/iron reaction takes place in August. Because there were several high temperature days with moderate rainfall during August, it is hypothesized that the second reaction is a low energy combustion of pyrite. (Data used to determine the temperature dependence on the reaction rate for pyrite calculation show us similar phenomena.) If this is true, it is possible to eliminate this reaction by keeping the surface of pyrite containing material cool. This can be accomplished by deep burial of toxic material as suggested by the task force recommendations.
Second, comparisons of the effectiveness of the various treatments can be seen by reading the ratios of areas under the derived curve portions describing the cumulative sulfate concentration eluted from the barrels. (See Table 2) As can be seen, the ratios can be ranked from least effective to most effective:
(Note, these are results of incomplete studies and are being reported as an update on progress on these studies, not as conclusions.) The sulfate equations results are duplicated by the calcium carbonate equivalent equations.
There is one particularly with the ag lime data that merits discussion: the discrepancy between the areas under the cumulative volume dependent equation and the volume independent equation. . It appears that the volume flow from the ag lime samples is about 60% that of the control and other test samples. The decrease in volume will be reflected in the cumulative volume independent equations, but not in the cumulative Volume dependent equations. This difference may be due to the hydration of iron oxides which are seen in the bed material or in the formation of calcium sulfate hydrate crystals. In no way is this due to flooding of the ag lime samples. The cumulative volume dependent sulfate equations and the limestone equivalent equations show that the ag lime treatment is not as effective in improving water quality as the other treatments. So, any improvement of the water reservoir from ag lime treatment is due to decreased volume of the elutent, not to chemical improvement of the water quality. This is in agreement with a recent EPA report. 3
Volume Dependent Cumulative Function for Sulfur
Volume Independent Cumulative Functions for Sulfate
Volume Dependent Cumulative Functions for CaCO3
Volume Independent Cumulative Function for CaCO 3
Volume Dependent Cumulative Function for Total Iron
Volume Independent ````````````````````````````````````````````````Cumulative Function for Total Iron
