TN 295 .U4 ■ ■ ■ ■ ' ■ I ■ •: fc, I ■ w "of °c ,- P V > * V o o » • * * "o^ .^ / a>- o -o. » - .-v ■» o 0^ .'• ^**TT,-»' ^ ^^ c°\* 4* <, *-- TTi* ,0 % s*>/r??7L, ■> o o, *o . * * A 1>* ^^ * v .•-•- "^ O , '' ^ ~ J > A . t < » „ *^- a\ - - - ^ j <$> *" 1 * .^ ♦ V ^ -"^JS^NF 4 4; ^ 4^' ,* .. ^ * P v ..'^% ° •>6* ^ ^. '. %■_"""' -p G v o *<. . , « A <► "0/ <►». "' 1* • 1 •• %S •' **- ^% ^yJ^rX ^f. *^ V ^ ^-^ -.iSK: ^^ °»yj^^ 4* ^ M v o £^mm } fs*tikS JPslMkS o V**.,i- V ^ 1 v . » • ■ 3 ^ ^^ !• S\ ■ *bv* ^ **! 6^ ^ iV •••' r.* 4 ,6 *^ J - VV '°^i% < O. *o , ^ » ^ ..»• aG o. *o . G \ V ~— ° ^0< V v .*«^L V> «, ' V ,iR^^V<" "^A A"^ » 6^ ^ 1 .^ **. °.^W' A^ ^ ' * av r» c~ .n <. ^TV.« .0 .* r> v o V V V*7... % ^0 N ^..•.".•..^ '^ <.4 h ■*.^ :]»: *W G^ 'o. -?.T- A »P^. V * V c°"«* **>. <, *-7VT s ,G <* ^ % <* *'TT« 5 .0 \. y • "^0* •' < ^»^ "of »o _^0^ . • G v ^3, 'o . » « A <, \-^- / ^/^?/. ^^^z v^y v^^>° \^\* . A v , o » o , IC 9095 Bureau of Mines Information Circular/1986 Safe Cleaning of State of Maine Filters Using EDTA-Type Chelating Agents By John E. Pahlman and Daniel N. Tallman UNITED STATES DEPARTMENT OF THE INTERIOR c -^JjtaM^. B^g^jj^^f^2 Information Circular 9095 Safe Cleaning of State of Maine Filters Using EDTA-Type Chelating Agents By John E. Pahlman and Daniel N. Tallman UNITED STATES DEPARTMENT OF THE INTERIOR Donald Paul Hodel, Secretary BUREAU OF MINES Robert C. Horton, Director \\r v) M - «s 4 Library of Congress Cataloging in Publication Data: Pahlman, J. E. (John E.) Safe cleaning of State of Maine filters using EDTA-type chelating agents. (Information circular ; 9095) Bibliography: p. 12 . Supt. of Docs, no.: I 28.27:9095 1. Filters and filtration. 2. Ethylene-diaminetetraacetic acid. 3. Ore dressing. 4. Calcium carbonate. I. Tallman, Daniel N. II. Title. III. Series: Information circular (United States. Bureau of Mines) ; 9095. TN295.U4 [TN535] 622 s [669'.22] 86-600226 CONTENTS Abstract 1 Introduction 2 Acknowledgment 2 State of Maine filter system 3 Characterization of filter blinding 5 Laboratory test procedures and results 7 Filter cleaning strategies with chelating agents 9 Field test procedures and results 9 Summary and conclusions 12 References 12 ILLUSTRATIONS 1. State of Maine, 300-st/d, filter system 3 2. Generalized flow diagram for State of Maine filter system operation 4 3. Schematic diagram of State of Maine filter unit 4 4. SEM micrograph of blinded filter socks 6 5. EDAX data for blinded filter sock 7 6. Effect of Na4EDTA solution concentration and agitation on CaC03 removal.. 8 7. Schematic diagram of Na4EDTA recirculation system 10 8. Weight of CaC03 removed from blinded filters as a function of time 10 9. Schematic diagram of in situ Na4EDTA system plumbing modifications 11 TABLE 1. Scale removal data from laboratory evaluation of chelating agents 7 UNIT OF MEASURE ABBREVIATIONS USED IN THIS REPORT °c degree Celsius mg/L milligram per liter g gram min minute gal gallon mL milliliter gal/min gallon per minute pet percent g/L gram per liter psig pound per square inch, gauge h hour rpm revolution per minute in inch st/d short ton per day KeV kilo electron volt tr oz/st troy ounce per short L liter ton SAFE CLEANING OF STATE OF MAINE FILTERS USING EDTATYPE CHELATING AGENTS By John E. Pahlman 1 and Daniel N. Tallman 2 ABSTRACT An increasing number of gold and silver mining operations are employ- ing cyanide solutions to effectively dissolve silver and gold from finely disseminated ores. Many of these operations use small, packaged processing systems like the State of Maine filter system, which employs zinc precipitation of gold and silver. Solution-clarifying filters in these systems become blinded with calcium carbonate (CaC0 3 ) scale and require frequent (every 2 to 10 days) dismantling of the filter units for acid cleaning. This is a hazardous operation as it is not compat- ible with the alkaline cyanide leach solutions and may result in the evolution of toxic hydrogen cyanide (HCN) gas during cleaning and during any accidental acid spill. The Bureau of Mines has investigated an alternative, nonacidic method of cleaning blinded State of Maine filters that employs ethylene- diamine-tetraacetic-acid (EDTA) type chelating agents to dissolve the scale. In situ cleaning of a 150-st/d State of Maine filters with EDTA- type chelating agents was successfully demonstrated. Greater than 90 pet of the CaC0 3 scale was removed in about 10 min with circulation of 31 L of 2.5-pct-EDTA solution through the filter unit. The EDTA cleaning method not only eliminates the cyanide hazard in filter clean- ing but is advantageous from the standpoint of operating costs and productivity. Supervisory physical scientist. 2 Research chemist (now with Economics Laboratory, Eagan, MN). Twin Cities Research Center, Bureau of Mines, Minneapolis, MN. INTRODUCTION Cyanide solutions are widely used in precious metal ore processing to effec- tively dissolve finely disseminated gold and silver. Chamberlain (JJ 3 identified over 80 operations that were actively leaching or seriously planning on cyanide leaching of gold or silver by either dump, heap, or in situ leaching. Contact of cyanide-containing solutions with free silver and gold results in the following reactions (2-3) : 2Ag -+ 4NaCN + 2 + H 2 + 2NaAg(CN) 2 + H 2 2 + 2NaOH (A) 4Ag + 8NaCN + 2 + 2H 2 + 4NaAg(CN) 2 + 4NaOH. (B) and These reactions strongly depend upon the amount of oxygen in solution. Thus, oxy- gen is added to the leach solution by bubbling air through it and/or by spray- ing it onto heaps. Recovery of gold and silver from preg- nant leach solutions can be accomplished by carbon adsorption with subsequent stripping and electrowinning of the gold and silver, or by the Merrill-Crowe re- covery process, which involves precipi- tation of gold and silver with zinc dust. Both methods have their respective ad- vantages and disadvantages. Selection of the recovery method depends upon the size and specific conditions of the leach- ing operations and facilities already available. One disadvantage of the zinc precipitation method, with respect to the carbon adsorption method, is the need to filter out suspended particles from the pregnant leach solution before zinc pre- cipitation. This is necessary to prevent the coating of the zinc particles by the suspended particles and consequent retardation of the precipitation reac- tion. Capital expenditures to establish a conventional mill circuit system based on the Merrill-Crowe process are high, but there are now on the market, small prepackaged mill circuit systems that are designed and priced for small leaching operations. One of these is the State of Maine (SOM) filter system. One problem encountered with opera- tional SOM filter systems is irreversible plugging of tbe filter every 2 to 10 days of operation. This necessitates disman- tling of the filter unit and cleaning of the filters with acid, a potentially hazardous operation because of the evolu- tion of toxic HCN gas. Mine personnel who breathe or come into skin contact with the toxic HCN gas can develop cyano- sis. HCN is a chemical asphyxiant that exerts its effect by inhibiting the meta- bolic enzyme systems (4^). Mild cyanosis results in nausea, while severe cyanosis can result in asphyxia and death. As part of its present research, the Bureau of Mines sought to determine what caused the blinding of the SOM filters, and to eliminate the cyanide hazard by devising a safe method of filter cleaning that does not evolve toxic HCN gas. ACKNOWLEDGMENT The authors wish to thank Charles Esca- pule, Louis Escapule, Bailey Escapule, and other personnel at Tombstone Silver Mines, Inc., for allowing the Bureau to fer to items in the list of references at the end of this report. field test the EDTA cleaning method on Tombstone's State of Maine filters, and for their assistance in setting up the necessary equipment to perform the tests. The authors also wish to thank Dow Chemi- cal Co. for supplying the Bureau with many of the chelating agents employed in this study. STATE OF MAINE FILTER SYSTEM Figure 1 shows a 300-st/d SOM filter system. Typically, a large SOM filter system consists of two filter units for clarification and two filter units for zinc precipitation. Depending upon the size of the leaching operation, smaller filter units containing 72 filter socks or larger filter units containing 120 filter socks are employed. For continu- ous operation, a system will have two filter units in parallel for both the clarification and zinc precipitation operations so that one filter can be cleaned while the other is in operation. The SOM filter system also consists of a deaeration tank, a zinc dust feeder, and the necessary vacuum and liquid pumps, valves, and other plumbing to carry out the process. Figure 2 gives a generalized flow dia- gram for the SOM filter system operation. First, pregnant leach solution is pumped from the holding pond to the SOM filter system. Next, suspended particulates are removed by the first clarifying SOM fil- ter unit. The filtered solution then passes through the deaeration tank and zinc dust is added before the solution enters the second SOM filter unit, where silver (or gold) precipitation proceeds by the following reaction in the second SOM filter unit: 2NaAg(CN) 2 + 2Zn ■* Na 2 Zn(CN) 4 + 2Ag (C) 2NaAu(CN) 2 + 2Zn •> Na2Zn(CN) 4 + 2Au (D) Failure to remove oxygen from the leach solution results in the redissolving of precipitated silver and gold and precipi- tation of zinc oxide (ZnO) in the second SOM filter unit. FIGURE 1.— State of Maine, 300-st/d, filter system. To heap <- NaCN CaO -J I- Barren solution makeup tank Zn dust feeder Precipitation SOM filter unit Smelting furnace I Ag, Au precipitate FIGURE 2.— Generalized flow diagram for State of Maine filter system operation. The zinc replaces the silver (gold) in the cyanide complex. Enough free cyanide must be present in solution to form a soluble zinc complex to keep the zinc metal surface free to contact fresh solu- tion. Hydrogen gas is produced by zinc oxidation as a byproduct of the precip- itation of silver (gold). Finally, the silver-free (gold-free) barren solution is pumped to a makeup tank where cyanide level and pH are adjusted before again being sprayed on the leach pad. The pre- cipitate is dried and smelted to produce an ingot of silver (gold) . The SOM filter units are basically swimming pool filter canisters adapted for the SOM filter systems. (See figure 1.) The 150-st/d SOM filter system em- ploys a smaller filter unit that contains 72 nylon filter socks about 13 in long, while the 300-st/d SOM filter system em- ploys a larger filter unit containing 120 nylon filter socks of similar size. Fig- ure 3 shows a schematic diagram of a SOM filter unit. Turbid pregnant leach solu- tion is pumped into the first SOM filter unit and forced through the filter socks. Bumping handle FIGURE 3.— Schematic diagram of SOM filter unit. Clay and other suspended particles are retained, and the clarified solution then passes to the deaeration tank. Zinc dust is added to the deaerated and clarified pregnant leach solution, which is then pumped into the second SOM filter where precipitated silver (gold) is recovered. The drain is employed to remove spent diatomaceous earth and precipitated metal values from the clarification and precip- itation filters, respectively. Diatomaceous earth is used as a precoat on the filter socks of the clarifying filter to facilitate filtering. When the diatomaceous earth is blinded due to en- trapped clay and suspended particles, the filter is manually "bumped" (slight re- verse flow generated through the filter) to redistribute the diatomaceous earth, and filtering is continued. When "bump- ing" no longer improves solution flow, the mixture of diatomaceous earth, clay, and suspended particles is "bumped" off the socks and removed from the filter via the drain. A new diatomaceous earth pre- coat is applied, and the procedure is started again. All this is done without opening the filter units. The filter units are opened for cleaning only when the filter socks are irreversibly blinded. Depending upon varying condi- tions such as water hardness and tempera- ture, this may be necessary in as little as 2 days. From the standpoint of health hazard (the less frequent the exposure, the better) and production (the less downtime, the better), it would be ideal if opening of the filter units were only required when a filter sock was rup- tured and needed replacement. Irrevers- ible blinding of the filter socks of the precipitation filter units also occurs, but much less frequently. The common procedure for cleaning the filter socks is to (1) drain the filter canister, (2) open the filter assembly, (3) raise the filter tubes and socks out of the unit to drain off any solu- tion, (4) rinse the socks with water, (5) rub the socks, (6) soak overnight in a dilute acid bath (HC1), and finally (7) rub and rinse the filter socks before replacing them in the filter unit again. An alternate procedure for clean- ing SOM filters consists of dipping them in an alkaline hypochlorite solution after following the first four steps of the above procedure, then dipping them in an acid bath, and finally dipping them in a lime solution before replacing them in the filter unit. Dipping the filters in the alkaline hypochlorite solution should oxidize any cyanide to innocuous CNO". Incomplete oxidation and hydrolysis in this step result in the evolution of HCN and cyanogen chloride (CNC1) in the acid bath. CNC1 is also toxic and can cause severe chemical burns and pulmonary edema. CHARACTERIZATION OF FILTER BLINDING Blinded filter socks were cut into 3/4- by 3/4-in (1/4-g) coupons for analysis to determine the nature of the substance plugging the filters. Several of these coupons were prepared for examination under the scanning electron microscope (SEM). Other coupons were reacted with 100 mL of 1:1 HC1 solution overnight. Examination of the blinded filter sock under the SEM showed the presence of many crystallites of an unknown compound (fig. 4) . Elemental analysis of this blinded filter sock using energy dispersive an- alysis of X-rays (EDAX) indicated that these crystallites were mainly composed of calcium with traces of zinc, iron, and silver (fig. 5). The presence of silicon, aluminum, and magnesium is due to the clay and/or diatomaceous earth trapped on the filter. Reacting four scaled filter coupons overnight with 100 mL of 1:1 HC1 solution resulted in a solution containing about 3.6 g/L Ca, about 30 mg/L Zn, and about 20 mg/L Fe. Evolution of gas bubbles during acid attack of the filter coupons indicated that the plugging substance was a carbon- ate compound. From the acid-digestion analysis data, the SEM examination, and the EDAX analysis data, it was concluded that the substance that irreversibly plugs the filter is mainly CaC0 3 scale. From the standpoint of clay and suspended particles removal, the SOM fil- ter unit appears to be more than adequate when precoated with diatomaceous earth. Thus, alternative filter types that can be cleaned by backwashing were not FIGURE 4.— SEM micrograph of blinded filter socks (X600). _ 1 1 I 1 _ - Co 1 - - - - Al M9 .d Ag [ ill Co A - / Uy/^ Vi/Wntv, Fe -vkW^Wj v^H\ X-RAY ENERGY, KeV FIGURE 5.— EDAX data for blinded filter sock. evaluated, as these types of fil- ters would not solve the problem of calcium-scaled filters. Calcium scaling would also blind other filters and would probably not be removed by backwashing. Based on discussions with mining com- pany personnel who were using SOM filter units and on the nature of the plugging scale, it was concluded that cleaning of these filters with nonacidic solutions (1) should be possible and (2) would drastically reduce the cyanide hazard. Cleaning of the filters in situ — i.e., without opening up the filter units — would be advantageous not only by elimi- nating the health hazard to mining com- pany personnel but also by increasing productivity, LABORATORY TEST PROCEDURES AND RESULTS Although there are many scale inhibi- tors and removers on the market, the present research was restricted to an evaluation of the common EDTA-type che- lating agents as to their effectiveness in removing scale from blinded filters. The seven chelating agents tested were (1) NasDTPA, a pentasodium salt of diethylene-triamine-pentaacetic acid; (2) Na4EDTA, a tetrasodium salt of ethylene-diamine-tetraacetic acid; (3) Na3HEDTA a trisodium salt of n-hydroxyethyl-ethylene-diamine-tri- acetic acid; (4) (NH.4)2EDTA diammonium, disodium salt of ethylene-diamine-tetraacetic acid; (5) (NH.4)4EDTA, tetraammonium salt of ethylene-diamine-tetraacetic acid; (6) (EGTA) ethylene-glycol-bis(amino- ethyl)-tetraacetic acid; and (7) (TEA) tri-ethanolamine. Initial evaluation tests involved overnight shaking of 100 mL of 5-pct so- lution of each chelating agent with four scaled filter coupons (total weight, about 1 g). After this chelation treat- ment, the residual filter coupons were dried and then reacted with 100 mL of 1:1 HC1 to remove any residual scale, and thus to enable the determination of the extraction efficiency of each chelating agent. By employing a large excess of reagent for a relatively long time, a "pseudo-equilibrium" is set up and the selectivity of each reagent established. Results of these tests are shown in table 1, where the respective pH values of each 5-pct chelating agent solution and the percent removal of calcium, zinc, and iron are given. Except for TEA, which removed neglig- ible calcium, the other six chelating agents removed from 96.9 to 99.3 pet of the calcium present, thereby indicating that all these agents could be used TABLE 1. - Scale removal data from laboratory evaluation of chelating agents Reagent Na 5 DTPA Na 4 EDTA Na 3HEDTA (NH 4 ) 2EDTA. . (NH 4 )4EDTA. . EGTA. TEA.. pH 12.5 12.5 12.5 5.2 '12. 1 8.9 '12.1 12.4 10.0 Removal, pet Ca 98.8 99.3 98.3 99.1 98.6 99.2 98.7 96.9 .3 Zn 63.5 64.6 63.5 84.0 75.8 72.1 75.0 12.9 2.7 Fe 6.4 3.6 13.6 21.8 2.2 13.0 2.2 1.6 5.4 'Adjusted pH. to remove calcium scale from the SOM filters in reasonable time. The best overall chelating agents for calcium, zinc, and iron removal are (NH4)2EDTA and (NH 4 ) 4 EDTA; however, the low pH of their solutions is not compatible with the nor- mal cyanide leaching solutions at pH 12. (NH 4 ) 2 EDTA and (NH 4 ) 4 EDTA do not perform as well at pH 12.1 (table 1), although the difference is slight. They are not, however, enough better at pH 12.1 than the other four good chelating agents to justify their extra cost and the trouble and cost of raising their pH to 12.1. Na 4 EDTA was chosen for further labora- tory testing and field evaluation studies based on the ease of solution prepara- tion, pH of chelating agent solution, calcium removal efficiency, zinc removal efficiency, and reagent cost. Tests were conducted using Na 4 EDTA to determine both the concentration of solu- tion required to remove the scale in a reasonable amount of time and whether so- lution agitation was advantageous. For these tests, two 250-mL aliquots each of 2.5-pct, 5.0-pct, and 10.0-pct Na 4 EDTA were reacted with eight scaled filter coupons (2 g) . For analysis, 25 mL of solution was withdrawn after 5, 10, 20, 40, and 60 min of reaction. One set of 2.5-pct, 5.0-pct, and 10-pct Na 4 EDTA so- lutions was stirred at a rate of 100 rpm, while the other set was not stirred. The reacted coupons were dried and then immersed in 100 mL of 1:1 HC1 solution to determine calcium removal efficiency at each time interval for each solution concentration. Figure 6 graphically shows the results of the tests. For nonstirred solutions, 95-pct calcium removal was achieved in 60, 45, and 40 min, respectively, for the 2.5-pct, 5.0-pct, and 10.0-pct-Na 4 EDTA solutions. For stirred solutions, 95-pct calcium removal was achieved in less than 10 min for all three solution concentra- tions. On the basis of these tests, it was concluded that the filter unit could be cleaned in about 10 min with any of the three solution concentrations if agi- tation was provided. Further evidence of the ability of Na 4 EDTA solutions to remove scale from the SOM filters was obtained by coating FIGURE 6.— Effect of Na 4 EDTA solution concentration and agitation on CaC0 3 removal. the outside of a 3-in-long specimen of filter sock with finely ground scale until a vacuum reading of 2.42 psig was obtained when attempting to draw water through the filter sock. A total of 5 g scale, which was obtained from other areas of the leaching and recovery cir- cuits, was coated on the filter sock. This scale's composition was similar to that present on plugged filters. Next, 100 mL of a 10-pct-Na 4 EDTA solution was heated to 80° C and then introduced into the test compartment on the outside of the plugged filter sock. Initially, the solution was very slowly drawn through the filter sock; however, within 2 min the vacuum reading was greater than 13.7 psig and the scale was almost totally removed. FILTER CLEANING STRATEGIES WITH CHELATING AGENTS Four main strategies are proposed for cleaning State of Maine filters using chelating agents: 1. The chelating agent solution is simply substituted for the HC1 solution presently used to clean the filters. The disadvantage of this method with respect to the following strategies is that the operators are still exposed to cyanide solutions as they dismantle the filter units, remove clogged filters, and then rinse and rub them before dipping them into the chelating agent solution. 2. The mixture of diatomaceous earth, clay, and suspended particles is drained from the filter unit. The unit is then disconnected from the filter system and reconnected to a separate recirculating pump system with hoses leading to and from a reservoir of chelating agent solu- tion. The filters are cleaned inside the filter unit by circulating the chelating agent solution throughout the unit. This treatment method is safer than method 1, since the operator does not have to re- move and handle the filter. A second advantage is that the extra wear on the filter units associated with torquing down the bolts is avoided since the fil- ters are not removed from the filter unit. 3. Blinded filters are cleaned in situ by circulating chelating agent solution through the filter unit after drain- ing the leach solution. This method has the same advantages of method 2 with respect to method 1. The advantages of this treatment method over method 2 are that the filter unit does not have to be disconnected, moved, and reconnected. The plumbing modifications needed to add a recirculating loop for the chelating agent solution are minimal. 4. In this method, scale formation is controlled or avoided by continuously metering small doses of chelating agent to the pregnant leachate as it enters the filter unit. The effect of this treat- ment on the Merrill-Crowe process over extended periods of time is not known. Elevated levels of soluble EDTA-calcium complexes can be expected in the leaching system. An advantage of this method over the other three methods is that there would be no downtime required for filter cleaning. A second advantage of this treatment method over treatment methods 2 and 3 is that a separate recirculation system for filter cleaning is not re- quired. Plumbing modifications of the system for slowly dosing the chelating agent are minor. FIELD TEST PROCEDURES AND RESULTS Five tests were conducted to validate laboratory results on actual filter units. Two tests each were conducted to evaluate treatment method 1 and treatment method 2, respectively, while one test was conducted to evaluate treatment method 3. these filters was low, these tests indi- cated that the chelating agent solution could be used, rather than the HC1 solution, for filter cleaning. Unfortu- nately, more filters with heavy scaling were not available at the time of testing. Method 1 (Test 1 and 2) Method 2 (Tests 3 and 4) Two filters from a 300-st/d unit (120 filter socks) were cleaned by dunking them into a 30-gal reservoir of 2.5- pct-Na 4 EDTA solution for 15 min. The filters appeared to be relatively clean before the tests, though 7.1 g and 2.4 g CaC0 3 were removed from them in less than 15 min. Although the amount of scale on Two filters from a 150-st/d unit (72 filter socks) were cleaned using treatment method 2. Figure 7 shows the Na4EDTA recirculating system. It con- sists of a solution reservoir (25 gal) , an intake hose, a discharge hose, and a recirculating pump to pump the cleaning solution through the filter unit. With 10 FIGURE 7.— Schematic diagram of Na 4 EDTA recirculation system. a scaled filter in place (test 3), the filter unit was connected to the dis- charge hose and the recirculating pump. The intake hose was connected to the pump. Water was first circulated through the filter to determine the pressure drop across the blinded filter. A steady pressure drop of 19 psig was observed. The water was drained from the filter unit, and pumping of Na4EDTA solution through the filter unit was begun. Dur- ing the first 10 min of the test, the solution coming into the reservoir through the discharge hose was turbid. (It was clear for water circulation.) During this time, the pressure drop across the filter decreased fairly quickly from 19 psig to 3-5 psig, and re- mained in that range until the completion of the test. Figure 8 shows the amount of CaC03 dis- solved as a function of time for tests 3-5. In test 3, CaC0 3 weight removal was calculated from the calcium analysis of the Na4EDTA reservoir solution samples taken 1.5, 2.5, 4.5, 10, 20, and 30 min after the Na4EDTA solution was first in- troduced into the filter unit (figure 8, curve A). As shown, the cleaning was about 90 pet complete in 10 min, and a total of 87.4 g CaC03 was removed from the filter. For all practical purposes, the filter is probably unblinded in the time it takes to fill the filter, about 1 to 2 min. It appears that some scale particles are loosened from the filter but do not immediately dissolve. This gives rise to the initial turbidity of the Na4EDTA solution in the reservoir; as a result, initial values calculated for calcite removal based on analysis of 1 1 80 A A - 70 B^~- — " 60 "A J - 50 KEY ° Test 3 (A) A Test 4 IB) - 40 • Test 5 IC) - 30 - 20 10 C • I 1 1 I I - FIGURE 8.— Weight of CaC0 3 removed from blinded filters as a function of time. dissolved calcium are lower than the amount of scale actually removed from the filter. Test 4 was similar to test 3, except that the filter used was not as scaled; the same cleaning solution was used as in test 3. But in this test only 20 g CaC0 3 was removed from the filter, with clean- ing essentially complete in less than 5 min (figure 8, curve C) . This differ- ence can be attributed to the fact that clean water pressure drop across this partially blinded filter was only 2 to 3 psig over that for the filter after it had been cleaned with Na4EDTA. Method 3 (Test 5) The required Na4EDTA recirculating sys- tem was connected directly to the exist- ing plumbing of an operating 150-st/d filter unit. Figure 9 shows a schematic of the Na4EDTA system plumbing with respect to that of the operating unit. The filter in the unit was not totally blinded, as a pressure drop of only 10 psig was observed for an operating flow rate of 35 to 40 gal/min. To see a 11 I From No, EDTA FIGURE 9.— Schematic diagram of in situ Na 4 EDTA system plumbing modifications. greater change in calcium concentra- tion throughout the test, the volume of Na4EDTA being recirculated was decreased to about 31 L (8.2 gal). A 5/8-in-diam garden hose was used instead of the 1- and 2-in-diam hoses used in testing of method 2. Samples were taken from the solution reservoir 1.5, 2.5, 4, 6, 8, 10, 12, 15, and 20 min after introduction of the cleaning solution into the filter unit. In contrast to previous tests, the filter socks were not rinsed to remove adherent diatomaceous earth after the leach solution was drained from the fil- ter unit. A fair amount of diatomaceous earth precoat probably remained on the filter. This same procedure would be used in future cleaning operations. In test 5, using method 3, cleaning was slower than for tests 3 and 4 using meth- od 2; CaC03 was still being removed, although slowly, at the end of the 20-min test period (figure 8, curve B) . This ob- served decrease in cleaning efficiency may be attributed to the slower recircu- lation rate employed in these tests; however, it is more likely attributable to the presence of the diatomaceous earth remaining on the filter socks. The po- rous structure of the diatomaceous earth is what makes it useful as a precoat, and the cleaning rate for removing calcite from diatomaceous earth is probably dif- fusion limited. SEM examination of dia- tomaceous earth from a scaled filter showed the presence of calcite in its pores. Thus, the observed increase in calcite removal from 15 to 20 min into the test is believed to be attributable to calcite removal from the diatomaceous earth remaining in the filter unit. Cleaning the diatomaceous earth of CaC03 does not appear to be important, since the diatomaceous earth is discarded anyway. Assuming that much of the cal- cium dissolved after 15 min of testing is attributable to removal of calcite from the diatomaceous earth, >90 pet of the calcite on the filter is removed in 10 min. As was the case in the tests of method 2, the reservoir was turbid in the first few minutes of the operation. In fact, calcite particles could actually be seen in the 1.5-min sample of Na4EDTA so- lution as it was being collected. By the time the samples were filtered for analy- sis, the turbidity of the 1.5-min sample had disappeared. The high calcite re- moval value for the 1.5-min sample indi- cates that the filter is probably in effect unblinded within several minutes, even though not all the calcite that is removed from the filter is dissolved. Total calcite removal from this filter was 73.4 g. Method 4 Method 4 was not field tested as part of this investigation; however, bench- scale leaching tests were conducted at the mine to determine the effect, if any, of Na4EDTA on cyanide leaching. For these tests, 40 g of minus 200-mesh silver ore, 75 mL of cyanide leaching so- lution (0.05 pet CN~), and either 12.5 mL of distilled water or 12.5 mL of 2.5-pct Na4EDTA were shaken in stoppered flasks for 1 h. The amount of Na4EDTA added was that amount needed to complex all the calcium added in the leach solution for pH control. Silver extractions of 1.5 tr oz/st were obtained in both leach tests, thereby indicating that the pres- ence of chelating agent in the leaching circuit should not adversely affect the cyanide leaching process. Any additional problems in the leaching and recovery processes attributable to dosing small amounts of Na4EDTA solution into filter units will become apparent only after a long-term evaluation of method 4. 12 SUMMARY AND CONCLUSIONS In situ cleaning of SOM filters with ethylene-diamine-tetraacetic acid (EDTA) chelating agent has been successfully demonstrated. Greater than 90 pet of the calcium-base scale was removed in about 10 min with circulation of about 31 L of 2.5-pct-Na 4 EDTA solution through the 150- st/d-size filter unit. This nonacidic method replaces the old acid (HC1) clean- ing method, in which the filter unit had to be dismantled and the filter socks re- moved, rinsed, rubbed, soaked in dilute acid, and then rubbed and rinsed again before being reinstalled in the filter unit. This eliminated the health and safety problem associated with acid cleaning of the filters; i.e., toxic HCN evolution during cleaning and during an accidental acid spill. In addition to the elimination of the cyanide hazard in filter cleaning, in situ cleaning of the filters with EDTA solutions is advantageous from the stand- point of operating costs and productiv- ity. By not having to open the filter units to clean the filters, the wear and tear on the filter socks (due to rubbing and acid degradation) is eliminated, and the wear and tear on the filter unit (degradation of rubber gaskets and crack- ing of the filter unit tops due to over- torquing bolts) is drastically reduced. Furthermore, the downtime for a filter is reduced from 30 to 40 min to 10 to 15 min by the EDTA cleaning method. Since lime is often added to maintain high pH in cyanide leaching systems, calcite (CaC03) scaling of filters and plugging of sprayer nozzles and plant piping are common problems for many cya- nidation operations. The carbon-in-pulp (CIP) system for recovery of silver and gold from pregnant cyanide leach solu- tions also has scaling problems. The carbon particles are rendered ineffective due to calcite scaling and must be periodically cleaned. Use of chelating agents should be generally applicable to cleaning of other filter types, un- plugging sprayer nozzles and plant pip- ing, and cleaning carbon particles as well as to the cleaning of State of Maine filters. REFERENCES 1. Chamberlain, P. G. , and M. G. Po- jar. Gold and Silver Leaching Practices in the United States. BuMines IC 8969, 1984, 47 pp. 2. Barsky, G. , S. J. Swainson, and N. Hedley. Dissolution of Gold and Silver in Cyanide Solutions. Trans. AIME, v. 112, 1934, pp. 660-677. 3. Duncan, D. M. Open Pit Gold Min- ing at Cortez. Ch. in Colorado Mining Association 1974 Mining Yearbook. CO Min. Assoc, 1974, pp. 92-94. 4. Mackinson, F. W. , R. S. Stricoff, and L. J. Partridge (eds.). Occupa- tional Health Guidelines for Chemical Hazards. Arthur D. Little, Inc., DHHS(NIOSH) 81-123, 1981, 5 pp. 4 U.S. GOVERNMENT PRINTING OFFICE: 1986—605-017/40,069 U.S. Department of the Interior Bureau of Mines— Prod, and Distr. Cochrans Mill Road P.O. Box 18070 Pittsburgh. Pa. 15236 AN EQUAL OPPORTUNITY EMPLOYER OFFICIAL BUSINESS PENALTY FOR PRIVATE USE. 1300 ] Do not wish to receive this material, please remove from your mailing list. ^ Address change* Please correct as indicated* C ft* V « ^IIB^ .- u ^p ^ .> a <, -*£*' .^ v ^ -:«^i* ^ - -v **w> ,* / ^ • ^ v g°*.c^*°o ^ ^ L" * *bp t> -.,,. ^ o *„„„ ^o* *p^ V ^ ^O *T^%* (Y> * '^^T'" 4X.P iV^. .o 15 - « v 9* * o » » \- /♦c^:-% r ^6* 4 . o « o „ "^^ <> *'..' A^ * ^> H ' -0 a0^ 0- 'o 'o. t . <> * c° , •p* •« * V cO""* "^ -p -^.^ V> A 0* . * ^ w ^ v % A ■V w ^./ o v o^ .^••« *o •A <* 1 r *. c°" ,C^!^ "°o >* .^^% % & o_ ^ * » V7y ' ' ^'' - O^ * * .To ' ol'i 1 ' ■ J'X °WW< : <^\ J -3K ; j^X °ww*' ^ v \ j ^ "bV" •^^ :M%&:* ^ \W^/ vs?*v V^v v^v "bV •p. « ^/•^^V ►*• V °^ •' <^ "i ^^ v ^9 ^ •*#*• A>- ^ -o . , - -v .*^* % ■*0* r * *W a ' V. *••'•' .V" .•!••-. V- V .«i!nL' O N O * -0 .* .... *^* °Mm ^ y.% 9 «**->. <& ^ i* . i • . <*-„ *- J* <"> 1 *!,*^* > V 5 s **'. 4 ^ •" A *^W -. ^ yJ^L^ ^ V -^*'V ",p*v^'-"*o- V . t • « "^ "b^ 4* • »P^ *> u ^ ** I » «J> °^>. " 1( ■A o^ ^> ' V' .«•!!'♦ c\ ,0 % . * • °. vv ^^ vv c''^ °W%fVW* A v *^ ■;■'' ^..^:^\ /^^.\> /.^^\ /^^^°o 4**\^%% V 4 LIBRARY OF CONGRESS 002 955 928 4 v'i ■j