AUSTRALIA Patent act 1990 COMPLETE SPECIFICATION INNOVATIVE PATENT A Low Cost Rapid Neutralisation Process for Alkali By-Products and Waste Materials The following statements are a full description of this invention including Title, Invention Description, Best Method of Applying the Invention, Claims, Abstract plus pertinent Figures and Tables. Customer Registration #3810341566 Green Technologies International Provisional Application #2011901937 1 AUSTRALIA Patent act 1990 DESCRIPTION OF THE INNOVATIVE PATENT A Low Cost Rapid Neutralisation Process for Alkali By- Products and Waste Materials 1. There are several problems with traditional methods for alkali neutralisation such as the use of mineral acids. These include worker exposure to potential acid burns, inhalation of toxic vapors in the workplace, equipment corrosion, and overall resultant costs. This innovative neutralisation process eliminates these dangers and costs through the use of non-toxic, relatively inexpensive reagents. There have been some non traditional approaches to neutralising some caustic by-products such as spent Bauxite. These involved the addition of large volumes of seawater with long reaction times, poor neutralisation results and large volumes of waste water to handle. The application of calcium and magnesium brines also leads to excess process time, poor and uncontrolled pH reduction results, and, excess liquid to handle. 2. In this innovative application, the neutralisation reagents possible are dry or liquid forms of soluble alums, such as sodium, potassium, ferric, and ammonium alum. The highest positive cation valent alum is the best to use, such as sodium alum. Other highly water soluble reagents may be used such as the di-chloride of calcium, magnesium or the sulfates of anhydrous or hydrated magnesium and aluminum. Aluminum alum, aluminum sulfate or hexahydrate aluminum chloride may also be used. The typical amount of reagent needed and cost, Provisional Application #2011901937 2 depends on the reagent used, results, and final pH desired is on the order of 1 3% by weight per metric ton of the waste or by-product being treated. Currently, this equates to about $5 Australian, per metric ton of waste in reagent cost (2012). 3. The neutralisation process is innovatively optimized through a designed approach to determining the alkalinity of the waste using the PNDX value for the waste or by-product. This is the Process Neutralisation Index. This is determined by integrating the curve generated between pH 13 and pH 6 using a 1 Normal solution of hydrochloric acid. Should the waste not test at pH 13 or higher initially, the curve generated from the starting point to pH 6 will be integrated to yield a PNDX value for the waste. A process regression equation is developed using the reagent or reagents to be used and the PNDX value of the by-product waste. In this way, a desired process end pH point can be maintained on a continuous flow of the waste to treatment ratio. Figure 1 found in Drawings, shows an example of this PNDX calculation. The alkaline material in question was titrated with 1 Normal hydrochloric acid and the neutralisation curve plotted. The curve was fitted to a high correlationship equation (r = -. 99) and then the area under the curve between pH 6 and 13 was integrated. The area value derived is the PNDX for this waste. If the initial pH is not 13 use the range 6 to the starting pH. 4. Due to the nature of the first order reaction between the process neutralisation salts used and the alkali by-product or waste, the waste or by-product mixture neutralisation process can be rapidly completed using an in-line mixer such as a simple static mixing unit for flowable wastes. 5. Static mixing systems which use fixed blades to mix wastes containing available free water will rapidly contact the reacting salt(s) with the alkali waste; an example is seen in Drawings Figure 2; the flow of alkali waste slurry enters in one end and reacting dry salt(s) is fed into the flow at right angles which is one of several feed arrangements. 6. Alkali wastes that do not have readily available free water can be rapidly and thoroughly blended with reacting di and tri valent salt(s) through the use of a plow/paddle mixer. Drawings Figure 3 shows an example of a commercially available plow/paddle mixer. This type of mixer's insightful design incorporates Provisional Application #2011901937 3 a cylindrical mixing chamber and high intensity plows or paddles to create a fluidised mixing zone. The mixer shaft operates at higher speeds than a traditional batch mixer and when combined with the cylindrical vessel design, delivers excellent mixing results. The fluidised mixing bed created by the mixing agitator is ideal for mixing solids to solids, solids to liquids and liquids to solids. 7. A factorial design testing approach is suggested to be applied to testing an alkaline waste in order to determine a control regression equation that contains the independent variables, e.g. PNDX of an alkaline material, the level and type of reagent(s) used, possibly mixing time, etc., and the desired dependent variable or variables, e.g. desired treated pH. A factorial experimental design is one in which all levels of each independent variable are combined with all levels of the other independent variables, and allows the investigation of the separate main effects and interactions of two or more independent variables, two or more independent variables that are completely crossed or tested with each other. The resulting regression equation from a PNDX analysis could be of varying best fit forms. The example shown in Drawings, Figure 1 has the form: Y(pH @ 15 mins) = a + bX + cX2+ dX3 ; correlation coefficient = 0.99 Standard Error of Y values = 0.11 The actual curve equation description is Y = 87.6 - 26.9X + 2.99 X 2 - 0.11 X 3 Table of results from regression equation is shown in Table 1. X = pH value, Y = amount of added 1 Normal solution of hydrochloric acid to the caustic waste or by-product. Table 1 X = 6.0 Y = 10.1 X = 8.1 Y = 7.29 X = 9.6 Y = 7.19 8. Three different soluble multi-valent salts were examined against three different alkaline wastes. This was to determine and demonstrate the differing Provisional Application #2011901937 4 effectiveness of Aluminum versus Magnesium and Calcium soluble salts as measured by addition levels of each cation, not the salts themselves. See Table 2 for the Factorial Experimental Design applied and Table 3 for the statistical results. 9. In a factorial statistical study of how Ca+2, Mg+2 and Al+3 neutralises high PNDX wastes, two and three PNDX level wastes were determined (31, 31.75, 45) and used to be treated with three differing "Low Cost, Rapid Process" eligible reagents. The value of each added reagent cation is based on the exact amount of each cation contained in the treatment chemical and quantity added to each test sample. For examples: Al+3 is 11.9% by weight of Ammonium Aluminum Sulfate and Mg+2 is 9.85% by weight of Magnesium Sulfate, 7 H20. Ca+2 is 36% by weight of Calcium Chloride. Table 2 shows the data, inputs for all of the factorial tests. Factorial results are shown in Table 3. 10. Table 2 Factorial Test Design to Measure Cation Effectiveness 0.296 0.54 0.063 31 0.16 8.5 0.197 0.36 0.126 37.5 0.071 8.3 0.099 0.18 0.064 44 0.0178 9.9 0.296 0.18 0.064 44 0.054 8.55 0.099 0.54 0.064 44 0.054 9.5 0 .099 0.54 0.189 31 0.054 6.0 0.197 0.36 0.189 37.5 0.071 8.2 0.099 0.18 0.189 45 0.0178 8.2 0.296 0.18 0.189 31 0.053 5.8 0.296 0.54 0.189 45 0.16 7.5 0.246 0.001 0.095 45 0.001 8.8 0.246 0.18 0.095 45 0.0443 8.7 0.296 0.36 0.001 45 0.001 9.6 0.247 0.36 0.063 37.5 0.0889 8.6 0.296 0.36 0.063 45 0.107 8.55 0.296 0.18 0.063 45 0.053 8.6 0.296 0.18 0.191 45 0.053 7.5 0.296 0.54 0.191 45 0.16 7.7 0.099 0.54 0.191 31 0.053 6.0 0.099 0.54 0.063 31 0.053 8.1 0.099 0.54 0.064 45 0.053 9.5 0.099 0.54 0.064 45 0.053 9.6 Mg+2 Ca+2 A1+3 PNDX Mg*Ca pH(1 5min) X1 X2 X3 X4 X1*X2 Y 5 11. Table 3 Operational Factorial Regression Report Dependent pH(@15 mins of reaction time) = Y pH(15 min)= 6.95 - 5.17(Mg+2 gms) - 1.47(Ca+2 gms) - 13.15 (AJ+3 gms) + 0.10 (PNDX) + 7.8 (Ca+2 * Mg+2) Y = Dependent pH (@15 min) Regression Equation Section Independent Regression Standard T-Value Prob Decision Power Variable Coefficient Error (Ho: B=0) Level (5%) (5%) Intercept 6.957328 1.095927 6.3484 0.000010 Reject Ho 0.999965 Mg+2 gms -5.174236 1.904435 -2.7169 0.015234 Reject Ho 0.722827 Ca+2 gms -1.46881 1.09554 -1.3407 0.198738 Accept Ho 0.242934 AI+3 gms -13.15224 1.830659 -7.1844 0.000002 Reject Ho 0.999999 PNDX 9.503399E-02 1.882131 E-02 5.0493 0.000119 Reject Ho 0.997180 Mg*Ca 7.828239 4.333831 1.8063 0.089715 Accept Ho 0.396804 R-Squared 0.876575 Analysis of Variance Section Sum of Mean Prob Power Source DF Squares Square F-Ratio Level (5%) Intercept 1 1500.677 1500.677 Model 5 24.60388 4.920775 22.74 0.000001 0.894393 Error 16 3.464305 0.2165191 Total (Adjusted) 21 28.06818 1.33658 Root Mean Square Error 0.4653161 R-Squared 0.8766 Mean of Dependent 8.259091 Adj R-Squared 0.8380 Provisional Application #2011901937 6 12. Explanation of Table 3 and resultant regression equation. The equation's R 2 value [coefficient of determination] is a high -0.88 (1.0 max possible); the ANOVA F value is a significant -22.7 which indicates one can trust the equation results as being statistically significant. This is since at 95% confidence the calculated F is 22.7 and is greater than the F critical table value for the Degrees of Freedom (DF, sample size) = 4.41. The ANOVA Probability value of 0.000001 indicates there's a 99.99+% probability that the F value is correct. The 5% Power value of 0.91 indicates that there is less than a 5% chance that at least one of the equation coefficients is zero. The least trustworthy equation value is for calcium but it is still well within a small range for the least effective cation. The regression equation reveals an approximate neutralisation effectiveness ratio for Al, Mg and Ca with that ratio being approximately 13.1 : 5.17: 1.47. Note however that calcium can only reduce caustic alkaline material's pH to about pH 10.3 due to the reversible solubility properties of calcium hydroxide. The use of a regression equation developed with a calculated PNDX for a specific waste will allow a continuous process flow control by feeding back the immediate post treatment pH to a computer control. This control loop can then adjust reagent input to obtain the desired immediate treated pH result and maintain quality control of the end processed material ready for environmentally safe ultimate disposal or a beneficial reuse. A large volume alkaline material that could be treated with this technology is caustic spent bauxite. But any alkaline material could potentially be neutralised to a desired pH with this innovative technology. 13. The "Best Method" for performing this innovative technology. The best method is performing a neutralisation of a caustic waste or by product is to determine the amount of reagent needed and by the calculating the PNDX of the caustic material. The following steps should be taken: Provisional Application #2011901937 7 Using a water soluble aluminum salt such as aluminum sulfate, anhydrous, to maximize the amount of Al 3 cation available in the reagent, run a step by step neutralisation of the caustic material generating a neutralisation curve such as in Figure 1 of Drawings. In this way find the amount of reagent required to obtain the desired end pH point equal to or greater than 6.0 after 15 minutes of mixing the reagent into the waste. Calculate the amount of Al+ 3 cation added in grams (see section 9 above for details). In addition, calculate the PNDX of the waste or by-product to be treated as outlined in section 3 above. Run a sufficient number of waste neutralisation tests to generate a table of at least 8 results. Run a regression analysis on the neutralisation data: use pHs at differing Al* 3 additions and the caustic material's PNDX. With a resultant R 2 typically of greater than 0.90 for this derived equation, you now have a tool with which to run a continuous, computer controlled neutralisation process for the specific caustic material under treatment. The resultant regression equation will be of the form: Y(pH @ 15 min) = C +/- nl(grams of Al*3 added) +/- n2(PNDX), but could be a differing equation format. This process control equation will allow ongoing process control with changes in reagent addition levels as the output pH of the treated material varies during processing. Provisional Application #2011901937