Title of Invention  A METHOD TO EVALUATE BENEFICIATION PLANT PERFORMANCE OF DIFFERENT FEEDS FOR A DEFINITE PERIOD AND A COMPARISON OF FEED LIBERATION CHARACTERISTICS. 

Abstract  This invention relates to a method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics comprising the steps of preparing a linear regression curve for three grade feeds of difference size fractions, determining impurity rejection values of reject, determining recovery of impurity fraction in the concentrate mathematically dependent on iron recovery, subsequently evaluating liberation characteristics of feed by employing an indicator (separation efficiency) being a function of mineral value recovery and impurity rejection in the concentrate, the said separation efficiency being determined by estimating impurity recovery mathematically and thus ensuring recovery of mineral values such as iron and maximum rejection of mineral as gangue on maintaining appropriate size fractions of feed and product streams, specific gravity cut and condition in the separation media. 
Full Text  FIELD OF THE INVENTION The present invention relates to a method of evaluation of liberation characteristics of metallic/non metallic ore feeds for gravity separation. More specifically the present invention relates to recovery of mineral values form ores by gravity separation on determination of efficiency values indicative of liberation characteristics of each feed at different size fractions and comparing liberation characteristics of two graded feeds on the basis of chemical analysis of the sized fraction collected form the feed and product streams and maintaining size fractions and specific gravity cut and condition accordingly to recover maximum metal values from ore feeds. BACKGROUND OF THE INVENTION Selection of technology in mineral beneficiation largely depends on the efficiency of the unit operation and the feed characteristics. These two parameters are interrelated. A unit operation is selected based on the size distribution of the feed and the physical property which is exploited to separate the values from the gangue mineral. In spite of the these measures, process efficiency is often controlled by the liberation characteristics of the feed. Therefore, the evaluation of the liberation characteristics of the feed is important to mineral engineers. To emphasize this point further; the liberation characteristics of the feed provides us invaluable information like: (i) yield of the desired quality concentrate, (ii) probable effect of size reduction on yield and finally (iii) feasibility of the entire process technology. Liberation characteristic of feed that is treated in a gravity separator is studied through (a) float  sink method and (b) microscopical analysis. In a coal washery, the float  sink analysis is known as washability test. Washability test of coal indicates the theoretical yield of the clean coal corresponding to the specific gravity cut for the desired clean coal ash level. In relative scale, higher theoretical yield indicates better liberation. Theoretical yield for different size fraction provides far more insight to liberation characteristics of the feed. Microscopical analysis is a part of very detail liberation study. It shows the prospect of mineral liberation with gradual size reduction. The said analysis is demerited that the process is very lengthy and time consuming. Therefore this exercise is usually done only when a new coal seam is made exposed for mining. However analysis of liberation characteristics of most of the metallic or nonmetallic ore poses serious challenge as the availability of heavy liquids required for carrying out the floatsink tests of these heavy minerals are rare. Among metallic minerals iron ore mining is most common. In iron ore beneficiation, sinkfloat tests for the feed and the products are not always feasible due to very high specific gravity of gangue particles. Clerici solution is used in few cases to carry out the sinkfloat tests of iron ores. The same solution cannot be used either repetitively or for bigger volume of sample, as the above solution is costly and unsafe for use. Sinkfloat test using ferrosilicon based heavy media is also not feasible as specific gravity cut for most of the iron ore is higher. As mentioned earlier the microscopical analysis can not be conducted on daily basis. Therefore, in an alternative approach the recovery of iron is considered as the yardstick for evaluation of liberation characteristics. This is an indirect method for evaluation of the liberation characteristics and the method is described in detail with iron ore as a case study. This method is universally applicable for all metallic/nonmetallic ore types. This method was used to evaluate the liberation characteristics of three different iron ore feeds. These feeds are high, medium and lowgrade iron ore feed with alumina content in the range of 2.22.8, 2.83.4 and 3.45.4% respectively. All these three feeds are in the size range of  10 + 0.5mm and were treated in a 300 t/hr capacity jig. This jig was optimised based on the fundamental understanding of iron ore jigging. To evaluate liberation characteristics of the three different types of feed, samples were collected from the feed, concentrate and the reject streams of the jig. Samples were drawn at an interval of two hours for two shifts in a day. All samples collected in a day were then mixed together to form a composite sample for the day. These composite samples were chemically analysed for total iron, alumina, silica, phosphorus and LOI (Loss on Ignition). Plant trial was carried out for long duration with each fed so as to generate sufficient data for subsequent statistical analysis. Average alumina content in the feed and jig products for each of the three feeds are plotted. Based on the alumina content in feed and products, iron recovery values were calculated for each day. Average iron recovery values in the concentrate were calculated for all the three feeds. However, these values can not be used for any comparison as the corresponding alumina content in the concentrate varies widely. An improved statistical method is used to normalize the recovery values with respect to the alumina content in the concentrates for all three feeds and then they were compared. In this method the alumina content of the concentrate and the corresponding recovery values were used to develop the basic linear regression model for all three feeds. In this type of model it is assumed that the expected response value (recovery) is a linear function of the independent value (alumina content in the concentrate). In the present invention correlation coefficient values were statistically tested for the correctness of the linearization model of recovery  grade relation. DESCRIPTION OF THE INVENTION One objective of the invention is to instead of studying liberation characteristic of ore feed for mineral beneficiation by gravity separation through the route of sinkfloat tests of the entire feed as well as different size fraction of the feed, the said liberation characteristics of ore minerals are studied through an indirect method. Another objective of the invention is to employ a statistical method to normalize the recovery values of iron in the concentrates with respect to the impurity content in the concentrates for three graded and comparing is possible only after developing a basic linear regression curves for the three feeds through functional linear relationship between response value (recovery) and independent value impurity level in the concentrate and statistically testing for the correctness of the linearizaion curve (model) for the recovery grade through correlation coefficient values. Yet another objective of the invention is to prepare the regression lines on the basis of intercept and the slope values, which are statistically tested for their correctness being a function of liberation characteristics and the effect of liberation characteristics are estimated based on the difference between the intercepts after normalizing the slopes of the two linear regression models. A still another objective of the invention is to regress the liberation characteristics of feeds by adjusting the slopes and then comparing the recovery difference during process of normalizing of regression lines for recovery. A still further objective of the invention is to determine impurity rejection values of reject such as alumina from chemical assay values of the feed and jig products for all the three types of feed namely high, mediumand lowgrade of iron ores when firstly the concentrate yield of the process of recovery is calculated as where f, c, t are the percentage of alumina in the feed, concentrate and reject respectively, the recovery of nonalumina fraction (Rc2) being determined by indicates the iron recovery and alumina rejection (Rt1) is determined by plotting iron recovery value against the impurity rejection value of all three feed types to generate trend lines for all types of feed which are indicative of their liberation characteristics. A still yet another objective of the invention is to evaluate the liberation characteristic of feed by employing an indicator called as separation efficiency being a function of mineral value recovery and impurity rejection in the concentrate, the separation efficiency being determined based on alumina content in the feed and in the products by estimating impurity recovery (Rc1) from Rc1 = 100  Rt1 and finally separation efficiency (x) from x = Rc2  Rc1, thus providing a reliable method to calculate the maximum recovery of mineral values such as iron and maximum rejection of impurity bearing mineral such as alumina bearing gangue minerals by keeping high separation efficiency through appropriate maintenance of specific gravity cut > 2.9 gm/cc of the media in the separation of mineral values by gravity separation fed with mineral feeds. A still further yet another objective of the invention is to recover mineral values form ores by gravity separation on determination of separation efficiency values indicative of liberation characteristics of each feed at different size fraction and comparing liberation characteristics of two feeds on the basis of chemical analysis of the size fraction collected from the feed and product streams thus not requiring sinkfloat tests and maintaining the size fraction and specific gravity cut accordingly to recover maximum metal values from ore feeds. Similar to the other mineral processing unit operations, in this case of developing regression model also the recovery  grade relations are linear. In linear models, the intercept and the slope values are the unknown parameters that determine the line. Therefore, in each case the intercept and the slope values were statistically tested for their correctness and also for their compatibility for comparison. The slope and the intercept value in each case is a function of the liberation characteristics as same jig operating condition was maintained in all three cases. In the present method at first slope of the two linear models are normalized and then the effect of the liberation characteristics were estimated based on the difference between the intercepts. In this study, the high  and the low  grade feeds are compared separately with the medium  grade feed as medium feed is considered as standard feed. Three major steps followed in the calculation procedures are: (a) establishing a linear relationship between the recovery and the concentrate grade, (b) fitting of regression lines through a pooled gradient value, and (c) finding out the separation between the lines which indicates the recovery difference between the two feed types under comparison. The present invention will be better understood from the following description with reference to the accompanying drawings in which Figure 1 represents average alumina content in feed and jig products for three different type of feeds according to the present invention. Figure 2 (a) represents difference in Fe (total) recovery between high and medium grade feed as shown in recovery concentrate Vs alumina concentrate linear curve according to the prior art. Figure 2 (b) represent difference in Fe (total) recovery between medium and lowgrade feed as shown in concentrate recovery Vs concentrate alumina liner curve according to the existing art. Figure 3 represents in plot of Fe recovery Vs alumina rejection for three different types of feed according to the invention. Figure 4 represent alumina content in different size fraction of mediumand lowgrade feeds. Figure 5 represents sizewise separation efficiency values for three different types of feed tested in the jigging plant according to the invention. The manner of fittment of regression lines with pooled gradients is schematically explained for the high and mediumgrade feeds in Figures 2(a) and 2(b). Figure 2(b) explains the recovery difference between the medium and the lowgrade feed. In both these figures, the regression lines are shown as line joining points. The dotted lines are derived from the main regression lines after they are fitted with pooled gradient. In case of Figure 2(a), first regression lines were drawn for the high and the mediumgrade feed relating iron recovery and alumina content in the concentrate. These regressin lines are marked by points in the figure. These regression lines in the present form cannot be used to compare in iron recovery at certain alumina level. It is possible only when the slope of these two lines is same. This is made possible through pooled gradient method. In this method, at first gradients of these two lines were statistically tested for their compatibility for finding out the pooled (common) gradients. In the next step, regression lines are modified with new slope with each line passing through the corresponding average iron recovery and alumina content in the concentrate. The fitted regression lines are used for comparison. Estimation of liberation characteristics through basic regression line method is far superior to the mean recovery method. Evaluation of liberation characteristics of ore feeds are carried through three different types of ore feeds that constitutes Raw Of Mine (ROM) which are shown in Table 1. Compositionally, the high grade ROM consisted of 60  80% hard ore and the rest is lateritic hard ore. Mediumgrade ROM contains 20  30% hard ore, 60  70% lateritic hard ore and 0  10% soft ore. The low  grade ROM contains 50  60% lateritic hard ore and 50  40% soft ore. In the present study  10 + 0.5 mm size fraction from three different types of ROM were treated in jig. These feeds ore termed as high , medium  and low  grade jig feed as they are derived from the high , medium  and low  grade ROM respectively. Remarks 14.12% more 8.52% less iron iron recovery in recovery in the the concentrate concentrate for for the high lowgrade feed grade feed as as compared to compared to the medium the medium grade feed, grade feed. The result show iron recovery is 14.12% more for the highgrade feed as compared to the mediumgrade feed. Conversely, iron recovery for the lowgrade feed is 8.52% less than the medium grade feed. In both cases the recovery difference remains same across different levels of alumina content in the concentrate. Thus, recovery difference as measure for evaluation of liberation characteristics of feed is often misleading. The linear relationship between the recovery and the concentrate grade is feed specific and it indicates liberation characteristics of feed. A gentle positive slope of the line indicates poor liberation in the feed. Conversely, a high positive slope of the line indicates that a part of the feed is liberated. In this study, regression lines showing linear relationship between recovery and grade for three different feeds are different (see Figure 2a & b). The highgrade feed shows the least liberation and the liberation characteristics are most favourable for the mediumgrade feed. However, during the process of normalization the slopes are adjusted and then compared the recovery difference; there by suppressed the vital liberation characteristics of feeds. Furthermore this method compares only the iron recovery difference instead of both, iron recovery and alumina rejection; this may as well mislead the inference on liberation characteristics of feed. Process of normalisation of regression curves are described as follows: At first correlation coefficients for both feed types were found out. Secondly, significance of the correlation coefficients was tested for each case. Thirdly, significance of the error values for both cases was tested. Fourthly, intercept value were calculated. In the fifth step, the gradient values were calculated for each case and the significance of their difference was tested. In the sixth step, a pooled gradient was calculated and the regression lines were modified with the pooled gradient value. With this step, the feed under comparison was normalized and was made ready for comparison. Finally, the separation between the lines was calculated and it indicates the concentrate yield difference at same level of alumina content in the concentrate. [Note: In the present invention recovery values were used for comparison as against yield value as shown in Table 3 for obvious reason]. PROPOSED METHOD Therefore, in the next step alumina rejection values were calculated from chemical assay values of the feed and the jig products for all three types of feed. In the first step the concentrate yield for the process is calculated as Where, f, c and t are the percentage of alunima in the feed, concentrate and reject respectively. The recovery of nonalumina fraction (Rc2) is In this case, (Rc2) value indicates the iron recovery Then alumina rejection (Rt1) is, Iron recovery is plotted against the alumina rejection for all three feed types in Figure 3. Trend line in each case is indicative of their liberation characteristics. The slope of the trend line for the high grade feed is the maximum and it suggests poor liberation for the highgrade feed as compared to other two feeds. Therefore, for better evaluation method used the liberation characteristics, which is a function of iron recovery and alumina rejection in the concentrate. Precisely, in the formula alumina recovery is used in place of alumina rejection to measure the efficiency. The method does not involve complexities of statistical analysis and also does not introduce error in the final result through normalisation step. The said indicator is called as separation efficiency and it was calculated based on alumina content in the feed and in the products. To calculate separation efficiency, alumina recovery (Rc1) is calculated as As per equation 5, the essential conditions for high separation efficiency are maximum recovery of iron minerals and maximum rejection of alumina bearing material which is gangue minerals. This is possible for feed with favourable liberation characteristics. Separation efficiency values for each day were calculated. It was found that the average separation efficiency value for the high, mediumand lowgrade feeds are 15.74, 25.44 and 28.38 respectively. Plant result show recovery for the highgrade feed is 14.12% higher than the medium grade feed. However the separation efficiency value for the highgrade feed is less than the mediumgrade feed. This is due to the difference in the method of calculation; the separation efficiency based calculation takes into account the recovery of iron and alumina and depicts better representation of the liberation characteristics than just iron recovery based analysis. Low separation efficiency of the highgrade feed is due to less rejection of alumina bearing minerals from the concentrate. Similarly, concentrate recovery of the mediumgrade feed is 8.52% higher than the lowgrade feed. Low yield for the lowgrade feed is due to its high alumina content in the feed. Figure 4 shows alumina content is high in all size fractions for lowgrade feed as compared to the mediumgrade feed. However, overall separation efficiency value for the lowgrade feed is higher than the mediumgrade feed and it indicates liberation characteristic of the lowgrade feed is better than the mediumgrade feed. This is also reflected in the size wise separation efficiency values as plotted in Figure 5. The invention as herein narrated with an exemplary embodiment should not be read and constructed in a restrictive manner as various modifications, alterations and adaptations are possible within the ambit and scope of the invention as defined in the appended claims. WE CLAIM: 1. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics comprising the steps of: employing a statistical method of normalizing the recovery values of iron concentrates with respect to impurity content in the concentrates for three high, medium and lowgraded ore feeds on basic linear regression curves through functional linear relationship between recovery and impurity level in the concentrate, determining impurity rejection values of reject such as alumina from chemical assay values of the feed and products (concentrates) for all the three types of feed when firstly the concentrate yield is determined from the equation 100 in which f, c, t are the percentage of impurity in the feed, the concentrate and reject respectively, the recovery of impurity fraction in the concentrate being determined by indicate the iron recovery and impurity rejection (Rt1) being determined from the equation plotting Iron recovery value against the impurity rejection value of the three feeds to generate trend lines for the three types of feed which are indicative of their liberation characteristics, ■ evaluating the liberation characteristics of feed employing an indicator called as separation efficiency being a function of mineral value recovery and impurity rejection in the concentrate, the said separation efficiency being determined based on impurity content in the feed and in the products by estimating impurity recovery (Rc1) from the equation Rc1 = 100  Rt1 and finally determining separation efficiency (%) from the equation x = Rc2  Rc1, thus providing a reliable and accurate method of maximum recovery of mineral values such as iron and maximum rejection of impurity bearing minerals as gangue mineral on maintaining appropriate size fraction of feed and product streams, appropriate specific gravity cut and conditions in the separation media according to evaluated separation efficiency of the feed indicative of liberation characteristics of the feed. 2. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in claim 1 wherein the corrections of basic linear regression curves for the recovery grade is secured on testing statistically through correlation coefficient values. 3. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in claims 1 and 2 wherein the said linear regression curves (model) are prepared on the basis of intercept and the slope values, being a function of liberation characteristics and the effect of liberation characteristics are estimated based on the difference between the estimated intercepts after normalizing the slopes of the linear regression models. 4. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics is suppressed by adjusting the slopes and then comparing the recovery difference during process of normalizing of regression curves for recovery. 5. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in the preceding claims wherein plant result show recovery for the highgrade feed as 14.14% higher than the medium grade feed, though the separation efficiency value for the highgrade feed is less than the mediumgrade feed when separation efficiency based evaluation takes into account the recovery of metal values such as iron and impurity rejection value such as alumina than evaluated only through iron recovery based analysis. 6. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in the preceding claims wherein concentrate recovery of the medium grade feed is 8.52% higher than the lowgrade feed, the low yield for the lowgrade feed being due to its high impurity content in the feed. 7. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in the preceding claims wherein overall separation efficiency value for the lowgrade feed is higher than the mediumgrade feed is better than the mediumgrade feed. 8. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in the preceding claims wherein in the gravity separation media specific gravity cut is maintained > 2.9. 9. A method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics as claimed in the preceding claims wherein the mineralogical and textural characteristics of the three different types of iron ore feed are as in Table 1 and the iron recovery values of those feeds are as in Table 2. ABSTRACT A METHOD TO EVALUATE BENEFICATION PLANT PERFORMANCE OF DIFFERENT FEEbS FOR A DEFINITE PERIOD AND A COMPARISON OF FEED LIBERATION CHARACTERISTICS This invention relates to a method to evaluate benefication plant performance of different feeds for a definite period and a comparison of feed liberation characteristics comprising the steps of preparing a linear regression curve for three grade feeds of difference size fractions, determining impurity rejection values of reject, determining recovery of impurity fraction in the concentrate mathematically dependent on iron recovery, subsequently evaluating liberation characteristics of feed by employing an indicator (separation efficiency) being a function of mineral value recovery and impurity rejection in the concentrate, the said separation efficiency being determined by estimating impurity recovery mathematically and thus ensuring recovery of mineral values such as iron and maximum rejection of mineral as gangue on maintaining appropriate size fractions of feed and product streams, specific gravity cut and condition in the separation media. 

01012kol2007correspondence others 1.1.pdf
01012kol2007correspondence others 1.2.pdf
01012kol2007correspondence others.pdf
01012kol2007description complete.pdf
1012KOL2007(01082012)CORRESPONDENCE.pdf
1012KOL2007AMANDED CLAIMS.pdf
1012KOL2007AMANDED PAGES OF SPECIFICATION.pdf
1012KOL2007CORRESPONDENCE 1.1.pdf
1012KOL2007CORRESPONDENCE.pdf
1012KOL2007DESCRIPTION (COMPLETE).pdf
1012KOL2007EXAMINATION REPORT.pdf
1012KOL2007GRANTEDABSTRACT.pdf
1012KOL2007GRANTEDCLAIMS.pdf
1012KOL2007GRANTEDDESCRIPTION (COMPLETE).pdf
1012KOL2007GRANTEDDRAWINGS.pdf
1012KOL2007GRANTEDFORM 1.pdf
1012KOL2007GRANTEDFORM 2.pdf
1012KOL2007GRANTEDSPECIFICATION.pdf
1012KOL2007PETITION UNDER RULE 137.pdf
1012KOL2007REPLY TO EXAMINATION REPORT 1.1.pdf
1012KOL2007REPLY TO EXAMINATION REPORT.pdf
Patent Number  254254  

Indian Patent Application Number  1012/KOL/2007  
PG Journal Number  41/2012  
Publication Date  12Oct2012  
Grant Date  10Oct2012  
Date of Filing  17Jul2007  
Name of Patentee  TATA STEEL LIMITED  
Applicant Address  JAMSHEDPUR831001, INDIA  
Inventors:


PCT International Classification Number  B01J8/32; B01J8/24  
PCT International Application Number  N/A  
PCT International Filing date  
PCT Conventions:
