W54”: v “'v - v Vvv vyvvw v _.._‘-- ..t'r . . I -—-v “‘-‘— -x- b .. - .J ,--;..‘(» u- 5 ' 11. :9; {gas " “1‘ “1---- #11931‘2". ..." 40151. :'.T:J1:113" 1.511" " '5 ' 1.4 - 4.644544}; :44 1'1": 1:1 ., 111m '-. '..]; ‘4‘ :1. 31V' J‘ ' 41444421144: . 4 4.4'5 "1751.571 5541 41' :‘4 4. ..4: 44 4°. 1 ‘ .flvwg" "a '5'. ’51 £' ' L k§:1.""fl f 4 .4,.... ' " 4 4:. ’44‘444 3:1 :4'31. 1,. n ‘11:”; .311 , ‘1: #119114 ME“ _. "i. ":4 "' 1'1" "' . 1.1': "13 0" {1% i1": ‘_.5 .1..." ":1. l'I"<""1l1"' 11. U14.“ 1141‘! 115.1 3511:; ‘1. '14'4' "":i"" I 0 1'. “ 1" "'5""'4454"'111:1P'1&",;1 ' ' 1)" £3 -4 4 '4 " ‘.‘-4}) . 341;, - - . 2.1111: , 1 . 1 .. . f.:_i..;l. n- 1.: 51:11, 0.. ".1 ‘ ‘1" 4'1"" ~"."""':1"".'4l"'1' '."1'4 54:""5w‘n 5:114:14; O 4 ’1 J ' ' ' ' ".11" E 1' V , "5.4-. '. "",.-1"1' ”53;; 12.: '1 1171' '11' a. 113.0 .1: 11114114.“ '91“ .: 1111, ’.. " 'h‘, 4 3.1511 151“” "1 -: 4 1 1' :1 s 111511 ""‘\"""§'1' . I _, ¢ -¢ ?D.~§ ' a . . ,c .¥ . ‘M‘u. - .l‘ {:1 5 11 144.4414}. .. 4474,44 . .....1'411' 114 "‘4: f'ingi' 414:“: 4.. .~ I -- '.‘—‘.‘ . i . t . .. ,A, ... 1,31 ' 3*.“ s .. "54.4,44rjf4 4 . :“4'4 11414, 5:: 45:2: 1. 1:31:51?” '51" ’5' '3 441‘ 1, 11' 1111:4114 43".: 4 - 4:4... . 44441:.44: 44 = """'""|",'.1'.' 51,11,111”; 51111313. 1 4 153.3. _—. :r. 11111 5'51: | 1 1 \ ,‘V: 1"": '51 .13.“: 1 1' 544;! '11": I 4.4.5:. 1'4 ‘ ' 5" "1:13.t 4151: '1"? :1'1'11'11 “14‘ ' 43314.1 5"""111'1'3"11:1,'.1'0L 1' '1'c113151'1'1'115. 111.4115 "::';".;' 151' "111'“ "11451114 41511311.: 1"11""""'3‘5'1'l'l'1114'1j51 31:4.1..11.'r;1'l':1.3.; -N - €5.33 I“" "'?'r"4'"7 I 4'": v1 ' .4311 '.':fli? ' . '3"1:,.i' .::.H 'l{: 1"4,‘ 1.1.1'1 1.. ~1_ 42.3.4.1”: ' 1;..4 " _ . 44.4 5:14.! n,4 '4,,4' “‘5 't ‘3'”; " . "‘ P'i:?:4£§"l:5«'_ "4 -"‘ ' ‘ 4' 4" 44:. 444'44‘ 4 ”$15.3. '1'1'111'" '4 H 4 5; ' .' ”1:31 '1' I; ' 1,: I... 5": 50“?" [’115"'1111};,' 1:. .1; '4" 1'5: v’4"/"4 33111-5517“ )1 ((115115 155 "5‘ K“ '1 44.4411. .44.:4‘4‘4.‘ 44:4". : i';4'4 ’ 444. 4444'4-4 .... :H‘ '5 ‘4 .4' :“ ‘- 4 I I ' . 14 u l 5- 11.551111113514313 _.;'4‘:1 ‘11,)“ 4111:111.4.|1m4i "3:41 4.. i1 14"11..1,:|.f.1’1,:'1 \ 13111114311”. 1' i V.“ :3, .- . 1'H4011' 5::‘1' ." . :4'4‘4..'4'| v' I ' 5L 1""."' "21' ' ‘ ‘ 4 .' 4" I. 4.411521. .51; “£4444 .:""4.5:.5'445":‘ i 11"." 1‘ 3415' "544114.144 9444.1 :43 ""‘. 411:7: 4 J?“ . 31:14 ...: "H.131: :4: 4;: 554‘ ' ,‘5' "5" ’.':4' 4444 4:44 .. 4“.- 2.3.1"; 4:315, .4543 55555 5"4‘155" ‘4' - ”'44:: ",4',' F'5'."5' 411' 11:1'4.""44‘:"‘,.4' , '41.:"41," . '1' ' ' ' .141} '.'1’4" ' .'4'1|."'?1": 4'17 '.' :15" 1'51 :"|'" ' '15'1""4'"5'.‘.'14'.'"'41""" 1"5":'14"L"'5 ROM“ 11"" '1' " ' " F'fi'fififi'} 4444.44. :: 24:4...4 4:. 4 :144444. 31 .4. 14.44 444. ' "'"""4.:4."'"" .4; .4": ' " 1.. '4'. .4 ‘-"'!1‘1' 4'13'4- . " v4" 4' 1331114. '15...'.' '1“ ‘4' ,11 3: : 111,14 11193111,: ‘3 1'10" | {'U'Jr . ”Tm" - .: “174.944" ”4'"; "'44; 1:1'4"341‘2'3‘4'15'54""4554'": s ‘5‘ 4'4 5-35" 1)" 4: ' “:‘4' ' ' ‘.4'” 5111411”; ' 4 I '1' '11 ... ..4: . - 4.. :~ 1'" 4:441:44. 44: “44'1'41'. .345 44:94“ 4'44' 4 "'51. 1""1";;.1 ”15‘ 11:111‘ 145511 1111' "“15 11,331? [11“ ln ."11"""."'1.1' ”'11'15'1 1 ‘ ...1'1'1'14' 1|" I 1'] .“t'lm: " " . ..'..’ .4:51:""1: 44...‘4:.:14 4 lllllllllll lllllllllllllllllllllll 3 1293 10614 7501 LIBRXRY \ Michigan Sum: University This is to certify that the thesis entitled Optimization of Liquid-Liquid Extraction and Stripping Batch Systems for the Recycling of Aluminum Coagulants in Water Treatment Plants presented by Roger Mark Lemunyon has been accepted towards fulfillment i of the requirements for Master of Science degree in Civil and Sanitary Engineering kpfi/M ajor professor Date November l7, l977 0-7 639 OPTIMIZATION OF LIQUID-LIQUID EXTRACTION AND STRIPPING BATCH SYSTEMS FOR THE RECYCLING OF ALUMINUM COAGULANTS IN HATER TREATMENT PLANTS By Roger Mark Lemunyon A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Civil and Sanitary Engineering 1977 ABSTRACT OPTIMIZATION OF LIQUID-LIQUID EXTRACTION AND STRIPPING BATCH SYSTEMS FOR THE RECYCLING OF ALUMINUM COAGULANTS IN WATER TREATMENT PLANTS By Roger Mark Lemunyon A procedure was developed for the recovery of aluminum when the aluminum was used as a coagulant in water treatment plants. A liquid- ion exchange process was developed to extract the aluminum from the sludge effluent and to regenerate liquid alum for reuse. The process was developed initially using synthetic solutions. This allowed large quantities of sludge to be readily available and solution character- istics to be easily changed. Water treatment plant sludge was then used to predict countercurrent, continuous flow operation. The alum sludge was collected from the Tampa, Florida water treatment plant. The sludge was first reacted with sulfuric acid to dissolve the aluminum from the organic solids. The supernatant was then separated from the residual organic solids by sedimentation. When sedimentation was utilized, 85% of the available aluminum could be separated. The acidified aluminum was separated from the supernatant by a liquid-ion exchange process. A kerosene solution containing a 0.84 M solution of extractant was contacted with the supernatant in a l:l Roger Mark Lemunyon volume ratio. The extractant was an equal molar solution of mono- di(2-ethylhexyl)phosphoric acid (MDEHPA). A minimum of 99% of the aluminum reacted with the MDEHPA and became kerosene soluble, result- ing in separation from the supernatant. A two-stage, countercurrent extraction circuit was required for aluminum recovery. The kerosene and water were very insoluble and readily separated in a settler. The aluminum rich organic phase was contacted with 6 N H2304 to force the aluminum into the acid. The organiczacid volume ratio was 15:]. ‘The two-stage countercurrent circuit allowed 49000 mg/l of aluminum to enter the acid phase. The final alum concentra- tion was 49%. The recovered alum was successfully reused for coagu- lation of a raw water. The organic solution was recycled back to the extraction circuit and successfully reused. The overall aluminum recovery was greater than 84%. ACKNOWLEDGMENT The guidance which was provided by my committee chairman, David Cornwell, allowed the many impediments to be overcome in the quest to obtain this degree. I thank both Professor Mackenzie L. Davis for the encouragement given during the difficult times and the guidance provided by Professor Carl Cooper. I can never forget the iendship developed by working with Jim. Also, I will never forget the won rous help provided by D. P. and M. C. \\\ Finally, recognition is given to theTAmerdtfifirihdggljkujfi‘¥fi Association Research Foundation which supported this study. ii TABLE OF CONTENTS List of Tables List of Figures CHAPTER I INTRODUCTION .................. l. l. Description of the PrOblem ........ l. 2. Previous Alum Recovery Systems ...... l. 3. Rationale of Research .......... l. 4. Objectives of Research ........ .. . CHARACTERISTICS OF ALUMINUM COAGULANT SLUDGES LIQUID-ION EXCHANGE ............... 3.l. Introduction ............... 3.2. Terminology of Liquid-Ion Exchange . . . . 3.3. Historical Description of Liquid-Ion Exchange ................. 3.4. Application of Liquid-Ion Exchange to Aluminum Purification .......... 3.5. Theoretical Description of Liquid-Ion Exchange ................. 3.6. Choice of Diluent, Extractant, and Modify- » ing Agents ................ 3. 6. 1. Introduction ........... 3. 6. 2. Choice of Diluent ........ 3. 6. 3. Choice of Extractant ....... 3. 6. 4. Choice of Modifying Agents . . . . Selectivity ............... Solids Handling During Extraction . Data Development in Liquid- -Ion Exchange 3. 3. 3. 3. 0. Liquid- Ion Exchange Equipment ...... HEDmSI EXPERIMENTAL APPARATUS AND PROCEDURES ..... 4.]. General Description ........... 4.2. Aqueous Feed Solutions .......... Page d \l U'lU'lN-J \DthOtO CHAPTER 4. 2. l. Synthetic Feed Solutions ...... 4.2.2. Treatment Plant Alum Sludge Solu- tions ............... 4.3. Organic Feed Solutions ........... 4.3.l. Alkyl Phosphoric Acids ....... 4.3.2. Diluents .............. 4.4. Analytical Equipment and Techniques Aluminum Determinations ....... pH Determinations .......... Extractant Loss Determinations . . . Organic Loss Determinations Heavy Metals Determinations Dispersion and Continuous and Phase Determinations ........ Evaporation Loss Determinations Specific Gravity Determinations Polymer Optimization Determinations. hh-fi- «DD-b-b-h-F- b-b-D h-h-h-D-D-fi OGDV men-puma 4.5. Experimental Procedures and Equipment. . . . 4.5.l. Procedures ............. 4.5.2. Equipment ............. 5 ALUMINUM RECOVERY FROM SYNTHETIC FEED SOLUTIONS 5.1. Introduction ................ 5.2. Extraction of Aluminum by an Equal Molar Mixture of Mono- Di(2- ethylhexyl) Phosphoric Acid .................... 5. 2. Introduction ............ Kinetics Study ........... Initial Feed pH Considerations . . . Development of Extraction Equili- brium Curves ............ Modifying Agents .......... Selectivity of MDEHPA ....... U101 U‘ICI'IUT NN NNN O O O C 0 mm thE‘ 5.3. Stripping of Aluminum-Mono-Di(2-ethylhexyl) Phosphoric Acid ............... 5 3 l Introduction ............ 5.3.2. Kinetics Study ........... 5.3.3. Acid Evaluation .......... 5 3 4 Development of Str1pp1ng Isotherms 5.4. Selection of Support Extractants ...... iv Page CHAPTER APPENDIX A. B. 5.5. Selection of Support Diluents ........ 5.6. Solvent Loss Considerations . . . . ..... ALUMINUM RECOVERY FROM ALUM SLUDGE SOLUTIONS 6.l. Introduction ................ 6.2. Extraction of Aluminum from the Tampa Sludge. 6.2. .2 .2. .2. 1. Introduction ............ . Extraction After Sludge Acidification 3. Extraction Before Sludge Acification. 4. Summary ............... mmm 6.3. Stripping of Aluminum MDEHPA ........ 6.3.l. Kinetics Study ............ 6.3.2. Comparison of Stripping when Extrac- tion Occurs in the Monomer and Dimer Ranges ............. 6.3.3. Development of Stripping Equilibrium Curves ................ .4. Phase Dispersion Considerations ....... .5. Color Contamination Considerations ...... .6. Interfacial Phase Disengagement with Polymers OCESS DESIGN FOR ALUMINUM RECOVERY BY LIQUID-ION C Introduction ................ Sludge Pre-Treatment ............ Aluminum Extraction ............. Disposal of Aqueous Raffinate ........ Stripping of the Extract .......... Organichecycle ............... Recovered Alum Solution ........... Summary ................... \INNNNNNN I“? 030501 0 O bowasmwa—I O O O O O O O 0 CONCLUSIONS AND RECOMMENDATIONS . . . . . . . . . . 8.l. Summary of Alum Recovery by Liquid-Ion Exchange ................... 8.2. Suggested Future Research .......... Glossary for Liquid-Ion Exchange ......... Determination of Aluminum in an Organic Solvent by Atomic Absorption Spectrophotometry ..... . . . BIBLIOGRAPHY .' ........................ V PAGE 108 109 110 llS lZl LIST OF TABLES TABLE Page 4.l Tampa Sludge Analysis ................ 26 4.2 Physical and Chemical Properties of Alkyl Acid Phosphates ..................... 27 4.3 Solubility Properties of Alkyl Acid Phosphates . . . 28 4.4 Specific Gravities and Available Specifications of Diluents ..................... 29 5.1 Selectivity of MDEHPA: Aluminum Versus Heavy Metals .................... .- . . 51 5.2 Organic Loss Determinations ............. 72 6.l Extraction of Tampa Sludge with MDEHPA ....... 78 6.2 Stripping of MDEHPA in the Monomer and Dimer Extraction Ranges .................. 91 6.3 Stripping of Tampa Sludge - MDEHPA ......... 92 vi LIST OF FIGURES Schematic Diagram of Aluminum Extraction\by Liquid- Ion Exchange .................... Schematic Diagram of Alkyl Phosphoric Acids Impeller Speed versus Time to Reach Extraction Equilibrium .................... Percent Equilibrium Versus Impeller Speed in the Extraction Circuit ................. Initial Feed pH Versus Aluminum Extracted . . . . . Extraction Equilibrium Curves for Various Initial Feed pH's ..................... Synthetic Extraction Equilibrium Curves for Various Molarities MDEHPA ................. Final Raffinate pH Versus Initial Aluminum Concentration ................... Impeller Speed versus Time to Reach Stripping Equilibrium .................... Percent Equilibrium versus Impeller Speed in the Stripping Circuit ................. Comparison of Acid Type and Normality on Stripping Efficiency ..................... Acid Stripping as.a Function of HZSO4 Normality and Phase Ratios .................... Synthetic Stripping Equilibrium Curves for Various Phase Ratios of H2504 Acid ............. Comparison of Synthetic Extraction Equilibrium Curves Utilizing MDEHPA and DEHPA as Extractants ..... Comparison of Diluents on Extraction Efficiency vii ' Page 15 20 37 39 42 44 46 49 54 56 59 61 64 66 7O Figure 6.1 6.2 6.3 6.4 6.5 7.1 7.2 7.3 Impeller Speed Versus Time to Reach Equilibrium in the Tampa Sludge .................. Initial pH Versus Final pH for Varying Molarities of MDEHPA ..................... Initial Sludge pH Versus Extraction Efficiency . . . Molarities of MDEHPA to Extract All the Aluminum from the Tampa Sludge at the Raw Sludge ...... Aluminum Stripping as a Function of Extraction in the Monomer and Dimer Ranges ............ Flow.Diagram of Alum Recovery by Liquid-Ion Exchange for the Tampa, Florida Water Plant ......... Graphical Determination of Number of Stages Necessary for 99% Aluminum Extraction ......... .. . . . Graphical Determination of Number of Stages Necessary for Aluminum Stripping .......... viii Page 77 82 84 86 89 99 l02 lOS CHAPTER 1 INTRODUCTION 1.1. Description of the Problem Water treatment management has undergone vast changes since the enactment of PL 92-500 in l972. PL 92-500 has specified criteria of effluent quality, namely Best Practical Treatment Currently Available (BPTCA) by July l, 1977, Best Practical Treatment Economically Available (BPTEA) by July 1, l983, and a later zero discharge stipulation by the mid 19805. BPTCA is generally defined as the equivalent sedimentation or filtration process presently being practiced in the water treatment field. BPTEA is generally defined as treatment technology that has been demonstrated on an advanced laboratory or pilot plant scale to be technically and economically feasible. In order to meet BPTEA criteria, the water utility industry is being forced to abandon the practice of discharging sludge effluent back into the raw water source. At this time the most economical replace- ment for direct disposal of sludge to the source is dewatering the sludge followed by landfilling. It has been estimated that a 20 per- cent solids concentration is needed for landfilling the sludge in conjunction with other solid wastes, such as municipal refuse, and that a 40 percent solids concentration is needed for landfilling the sludge alone. Current technology has shown that vacuum filtration can result in a nearly 20 percent solids concentration when preceded by thickening. However, the process is very capital intensive and is generally shown I to be uneconomical. An alternative sludge disposal technique is required if sludges are to be economically and lawfully disposed. Traditionally, surface water treatment is accomplished by the application of coagulants to the raw water. Aluminum coagulants in the form of aluminum sulfate or alum as it is more commonly named are prevalent in water treatment plants throughout the United States. It has been estimated that l4 million tons of wet weight alum sludge are produced nationally each year [1]. Since alum sludges are of such low density and contain large amounts of water of hydration, dewatering costs are high. In turn, final sludge disposal costs are high. The zero discharge requirement of PL 92—500 makes alum recovery very attractive in two respects. First, alum recovery has the advan- tages of reducing coagulant costs and conserving the earth's resources. Secondly, the nature of the alum recovery process reduces sludge de- watering costs and the quantity of sludge requiring disposal. l.2. Previous Alum Recovery Systems Present alum recovery methods have not been able to compete economically with conventional alum disposal techniques. Roberts and Roddy [2] examined the recovery of alum on both a pilot and full scale process. The recovery was reported to be based on the following reaction: 2Al(OH)3(s) + 3H2504 = Alz(SO4)3 + 6H20 The pH range for complete dissolution was between 1.5 and 2.5 depending on the alkalinity of the water. It was estimated that the acid recovery method could reduce chemical costs by 70 percent. Isaac and Vahidi [3] studied alum recovery as a method of sludge disposal. Isaac tested the alkaline and acid methods of aluminum recovery. He found that caustic soda was never very satisfactory in aluminum dissolution. He also found that the organic color in the sludge was much more soluble in alkali than in acid. Using the acidic method for aluminum recovery, tests were then run on fresh sludge and anaerobically digested sludge. At a pH of 2.5, corresponding to 70 percent aluminum recovery, a 74 percent volume reduction of sludge was obtained. The researchers concluded that the pH should be lowered to about 3.0 for a recovery of about 60 percent to 65 percent of the alu- minum. This pH prevented organic color from dissolving to an excessive degree. Webster [4] found that if sulfuric acid were added to alum sludge to reduce the pH to about 2.4, a clustering effect of the floc particles took place with extremely rapid settling of the insol- uble matter. The supernatant liquor containing the alum represented about 80 percent recovery. Tomono [5] reported that the Higoshimurayama, Tokyo, Japan water treatment plant utilizes alum recovery for the purpose of meeting national effluent regulations. The regulations prohibit water treat- ment plants from discharging sludge to nearby waterways. The 230-mgd plant utilizes sulfuric acid to reduce the pH of the alum sludge and to dissolve the aluminum from the solids. The recovered alum super- natant is then recycled to be used as a coagulant. The alum concentra- tion ranges from l.O-l.5 percent as Al2(SO4)-18H20. Manganese, also dissolved during pH reduction, contaminates the alum supernatant and builds up after recycling. The alum then has to be disposed. This usually occurs after 3-4 recycles. Streicher [6] conducted pilot tests to determine the usefulness of acid recovery of aluminum followed by filter pressing the remaining organic sludge. The pH was reduced to l.5 to 2.5 by sulfuric acid. He found that when the ratio of Al(OH)3(s) to other suspended matter in the sludge was high, considerably less than stoichiometric amounts of sulfuric acid were required. If the ratio were low, more than stoi- chiometric amounts of acid were needed. Acid treatment resulted in reduction of sludge volume to less than 10% of the original volume and a concentration of the sludge to 20% solids. The alum recovery was 80% to 93%. The residual sludge was concentrated to 40% to 50% solids with the use of a filter press. Hesterhoff and Daly [7,8,9] conducted a complete study of various alum sludge dewatering facilities. They included pressure filtration with and without alum recovery, centrifugation, rotary vacuum filtra- tion, horizontal vacuum filtration with and without alum recovery, coagulation, filter pressing, and freeze-thawing. The studies showed alum recovery followed by horizontal vacuum filtration to be a workable process warranting economic consideration. Alum recovery varied from 50% to 90%. Coagulation basin sludge was thickened from an initial 4% to 6% solids to a final 2l% solids content by acid treatment. After filtration the solids content was 37%. However, because of the low alum dosage used fOr turbidity removal, the most economical method of alum sludge treatment was determined to be pressure filtration without alum recovery. 1.3. Rationale for Research The review of alum recovery literature indicated that acid dissolution techniques for alum recovery may be applicable for plants with a raw water of high quality. These plants would have a low concentration of contaminants in the acidified sludge. However, those plants with poor water quality that use large quantities of alum cannot successfully apply simple acid dissolution for alum re- covery methods. Simple acid dissolution techniques do not meet the major goals of an alum recovery process as outlined below: l. The recovery system must have a procedure to prevent contamination of the recovered alum other than by de- creasing aluminum dissolution. Presently by decreas- ing the overall recovery of aluminum, the make-up costs and costs for residual solids disposal are increased; 2. The recovery system must separate the residual solids from the acidified aluminum. Presently when sedimentation is used, a large amount of aluminum is lost in the wasted sludge. Hhen filtration is used, the costs for sludge disposal remain high; 3. The recovered alum must be in a suitable concentration for ease of monitoring, pumping and dosage control. l.4. Objectives of Research Since this research is the first year study of a continuing re- search project, it is best to outline the purpose in two phaSes: over-’ all and specific first year objectives. The purpose of the overall research is to develop an economical aluminum recovery system in potable water treatment. A liquid -ion exchange procedure will be utilized. To become successful, the recovery system must.meet the fbllowing objectives: l. The recovered aluminum should be contaminate-free; 2. The recovery process should dissolve all the aluminum and recover as much as feasible for reuse; 3. The recovery system should result in residual solids which can be easily disposed; 4. The recovery alum concentration should be comparable to commercial alum to allow direct reuse. The first year objective of this research was the static optimi- zation of the extraction and stripping conditions in liquid -ion exchange laboratory batch systems. The purpose of this approach was to evaluate the future feasibility of using liquid -ion exchange for alum recovery and recycling. The end result of the first year research was to be a specific recommendation concerning continuation of the iresearch on a continuous-flow basis in the laboratory. IThe research was conducted in the following chronologic order. First, the liquid-ion exchange process was evaluated on synthetic aluminum solutions. The advantage of this approach was that control and variable parameters could be easily changed. Large quantities of synthetic solutions were also readily available. Secondly, the liquid-ion exchange process was applied to an alum sludge. The result was final optimization of parameters that need to be involved in the liquid-ion exchange operation. CHAPTER 2 CHARACTERISTICS OF ALUMINUM COAGULANT SLUDGES Alum sludges are characteristically different than other coagulant sludges found in water treatment processes. Hater contained in the alum sludge is bound to the sludge itself through hydrogen bonding. The resultant sludge is of a high volume, low density aluminum~hydroxide water floc nature. Due to the hydrated nature of the sludge, conven- tional dewatering is difficult to apply and too expensive to justify. Disposal of alum sludges can be more easily accomplished when the volume of sludge to be disposed can be reduced. Treatment of alum sludge has primarily been accomplished by vacuum filtration, filter pressing, sand drying beds, or centrifugation. The most effective means of sludge dewatering are vacuum filtration and filter pressing. The solids are separated from the water, leaving a higher concentration of solids. However, there is an economical obstacle faced by these conventional processes. The water contained in the alum sludge is bound to the sludge itself through hydrogen bonding, making the high volume reduction of sludge nearly impossible to achieve. The following simplified reactions have been proposed to occur to account for the high volume, low density aluminumrhydroxide water floc formation: Alz(SO4)3-T4H20 + 6HCO3 = 2Al(OH)3(s) + 6C02 + l4H20 + 3304 or, in the absence of alkalinity ATZ(SO4)3-l4HZO a 2Al(0H)3(S) + 2H2504 + 8H20 Under equilibrium conditions, aluminum would exist primarily as insoluble Al(OH)3. However, researchers [l0] have shown that under nonequilibrium conditions existing in water treatment plants the floc species is not Al(OH)3 but a positively charged species. Two of the more accepted fbrms are Alx(OH)2 +0.5x 2.5x A18(OH) :3'0 by Matijevic [l2]. In all of the proposed species the by Brossett [ll], and aluminum is associated with a high concentration of hydrated water molecules. Due to the hydrated nature of the sludge, the conventional de- . watering processes are very expensive. An alternative to mechanically separating the hydrated water would be to dissolve the aluminum in solution, thereby releasing the hydrated water, and allowing the resid- ual solids to settle. The aluminum-hydroxide floc is amphoteric in that it will dis- solve in acid or alkali. Recent studies [3] have indicated that acid dissolution is more effective in dissolving the aluminum. If alum recovery is to compete favorably with existing sludge disposal processes, the cost savings must be large enough to offset the additional capital and operating costs required for recovery. CHAPTER 3 LIQUID-ION EXCHANGE 3.l. Introduction Solvent extraction is a unit process of extractive metallurgy. Mixtures of different substances are separated by treatment with a selective liquid solvent. At least one of the components of the mix- ture must be immiscible with the treated solvent so that at least two phases are formed over the entire range of operating conditions used. Liquid-ion exchange is a specific type of solvent extraction. 3.2. Terminology of Liquid-Ion Exchange Terms used in liquid-ion exchange unit operations were defined in Appendix A. Additionally, the more important terms and operations essential to a working knowledge were discussed in the body of this research paper. 3.3. Historical Description of Liquid-Ion Exchange Interest in liquid-ion exchange as a unit operation was intensified by its successful use in producing purified uranium compounds during the Second World War. From T942 to 1953, the Mallinckrodt Chemical Works [l3] operated a uranium refinery for the Atomic Energy Commission utilizing an ether extraction of uranium nitrate substantially as described by Peliot in 1842. A refinery placed into operation in 1953 by the National Lead Company [14] used tributylphosphate as the uranium 10 nitrate extractant in the same purification process. In this same period of time, the Bureau of Mines [15] started production to separate hafnium and zirconium under an Atomic Energy Commission project. The basic process used was developed by the Oak Ridge National Laboratory. Hafnium was extracted from hydrochloric acid as a thiocyanate complex by methyl isobutyl ketone. Concurrent with these extraction developments, systematic searches for other extractants were conducted by a number of research laborator- ies, particularly those of the Dow Chemical Company [16] and the Oak Ridge National Laboratory [17]. These studies led to the commerical use of octyl pyrophosphone acid for recovering by-product uranium from phosphoric acid in 1955, and the use of alkylphosphoric acids and aliphatic amines for recovering uranium and vanadium from sulfuric acid solutions in 1956. Also in 1956, a Bureau of Mines [18] process enabling separation of columbium (present name, niobium) and tantalum was put into commercial use. The tantalum and columbium form complexes in a hydrofluric acid-sulfuric acid solution and then are extracted into hexone. Liquid-ion exchange has been used since 1959 in processing tungsten ore for recovering thorium from uranium wastes. In 1963, General Mills [19] introduced their first extractant which would purify copper from leaching low-grade oxide ores. Copper purification by liquid-ion exchange has become so popular that today over 300,000 metric tons of cathode copper will be produced by liquid-ion exchange [20]. General Mills [21, 22, 23, 24] continues to research liquid-ion exchange and has become one of the leading authorities in this field. 11 In 1977, liquid-ion exchange is used in commercial plants for the recovery of uranium, molybdenum, vanadium, tungsten, thorium, copper, boron, tantalum, columbium, hafnium, aluminum, and zirconium. It has been shown in pilot plant work that liquid-ion exchange may be used fbr the recovery of nickel, cobalt, the rare earth metals, and iron. In principle, almost any metal and most of the non-metals can be separated and purified by a solvent extraction procedure. The question is often not how to do it technically, but how to do it economically. 3.4. Application of Liquid-Ion Exchange to Aluminum Purification The research literature indicates that liquid-ion exchange has been applied in attempts to purify aluminum in two cases, one in waste dump leaching solutions and the other in wastewater sludges. George [25] in 1968 reported that aluminum could be purified from iron acid leachings economically. The process description followed these essential steps: 1. Aluminum was extracted from an acidic sulfate solution essentially free of ferric iron, with a kerosene solu- tion of an alkyl phosphoric acid, preferably a mono- alkyl phosphoric acid. Extraction from acidic chloride or nitrate solutions was also possible. 2. Aluminum was stripped from the loaded organic extractant with 6-8 N HCl, to yield a solution of aluminum chloride rich in aluminum and depleted in H+. Simultaneously, the organic extractant was regenerated to the acid form and was recycled to the extraction circuit. 3. The aluminum chloride strip solution was gassed with HCl to restore the concentration of free acid to 6-8 N. Simultaneously, most of the aluminum was selectively precipitated as crystalline AlCl -6H 0. The precipitate was filtered and washed with fregh H81 and the filtrate and washings were recycled to the stripping circuit. The principal impurity in the filtrate was ferric iron. 12 The buildup of ferric iron was controlled by extracting Fe+3, as the chloride complex, from a small bleed stream, with a kerosene solution of tributyl phosphate or an alkyl amine. 4. The solid A1Cl ~6H O was thermally decomposed to produce alumina and regoveF HCl for reuse. Although the extensive research indicated the process to be economically justified, no information could be found as to whether the process was ever put into commerCial operation. Cornwell, [26] in 1975, reported a procedure developed for the economical recovery of aluminum when the aluminum was used as a coagu- lant for phosphorus removal from domestic wastewater. After thickening of the sludge, the chemical-organic sludge was reacted with sulfuric acid to dissolve the aluminum and phosphorus. In order to separate the aluminum from the phosphorus, a liquid-ion exchange process was utilized. A kerosene solution of mixed mono- and di(2-ethy1hexy1) phosphoric acid (MDEHPA) was contacted with the aluminum-phosphorus solution. The alkyl phosphates reacted with the aluminum causing the metal to become kerosene soluble. The aluminum rich kerosene phase was then contacted with sulfuric acid allowing the aluminum to transfer from the kerosene to the acid phase where it was recovered as alum. The work was not extended outside the laboratory. 3.5. Theoretical Description of Liquid-Ion Exchange LiqUid-ion exchange is the separation of cationic or anionic solutes from a liquid phase solution by contact with another immiscible liquid solution. The theory of operation is dependent on the differ- ential solubilities of individual species in the two liquid phases. The process is very similar to resin-ion exchange. In liquid-ion 13 exchange, a small quantity of an organic soluble chemical called the extractant is dissolved in a second organic liquid called the diluent. The mixture is often referred to as the organic phase or solvent. The solution which is contacted with the organic phase during extraction is called the aqueous phase. For water treatment applications, the aqueous solution is the alum sludge. The extractant can be one of two general types. The amines act by forming an organic soluble salt with anions while alkyl phosphates react with cations. The diluent is some inert hydrocarbon, such as kerosene, which serves as acarrier medium for the extractant. During the extraction operation the extractant reacts chemically with the desired metal in the aqueous phase forming a new compound which is soluble in the inert diluent. A figurative representation of the process involved in the aluminum extraction is shown in Figure 3-1. The organic and aqueous phases are mixed such that small aqueous droplets are formed in the organic continuous phase. The extractant in the diluent contains a nonpolar part (the wavy lines in Figure 3-1A) causing the extractant to remain organic soluble, and a polar group (represented by the circle at the end of the wavy line in 3-lA) which sticks out into the aqueous phase and is the active site for aluminum complexation. The active site of the extractant ionizes. The aluminum-ion is exchanged for the H+ ion such that the H+ enters the aqueous phase and the A13+ moves to the organic-aqueous interface (Figure 3-1B). When mixing is stopped and the dispersed phases are allowed to settle, rapid coalescence takes place. The result is a separation into an aluminum- rich organic phase an an aluminum-free aqueous phase (Figure 3-1C). 14 Figure 3-1. Schematic representation of aluminum extraction by ' liquid-ion exchange. Organic Fe” - _ Aqueous Z‘p/Il/—\ 16 The aqueous phase which contains the aluminum to be extracted is called the feed. The aluminum-rich organic phase is called the extract and the aluminum-free aqueous phase is called the raffinate. The stripping circuit is operated in much the same way as the extraction circuit except that a stripping agent is chosen which causes the aluminum to leave the organic and enter the strip phase. A sulfuric acid solution has been shown successful in this research as a stripping agent. As a result, aluminum in the organic phase exchanges for protons in the acid phase resulting in aluminum sul- fate and regenerated solvent. It is usually operated in an organiczacid volume ratio of 3:1 or greater so that aluminum is correspondingly concentrated for reapplication as a coagulant. 3.6. Choice of Diluent, Extractant and Modifying Agents 3.6.1 Introduction The purpose of the diluent is to serve as the carrier for the extractant. The purpose of the extractant is to selectively remove the desired metal from the aqueous phase. The purpose of the modify- ing agent is to enhance the effectiveness and efficiency of the diluent and extractant. 3.6.2 Choice of Diluent The important parameters for diluents in liquid-ion exchange are: 1. Stability of diluent. There should be a minimum of evapor- ation losses and chemical interaction with other substances present; 2. Differential density of organic phase to aqueous phase. Allows fbr minimum settling area and time after mixing; 17 3. Low organic solubility in the aqueous phase. Minimizes losses of diluent which is an important economic consider- ation; 4. Minimum entrainment losses. Also an economic.consideration. Several diluents which are suitable for extraction systems are available. Specific diluents are usually chosen for each particular application. Kerosene is probably the most common diluent in extraction processes. Kerosene is used due to its availability, comparatively low cost, and relative safety in handling. Other fractionalized crude oil derivatives are now available commerically. Manufacturers closely control specifications so that uniformity of the product is maintained. The high, narrow boiling range and higher flash point in comparison with commercial kerosene have significantly reduced evaporation losses and fire hazards. An important property which exhibits more favorable results than kerosene is the faster rate of phase disengagement which results in minimum solvent entrainment losses and lower settling area requirements [27]. 3.6.3. Choice of Extractant The important parameters for extractants in liquid-ion exchange are: 1. Large number of active sites. Allows high concentration of extractant to be dissolved in the diluent; 2. Affinity of active site for the ion to be removed. Allows large number of desired ions to be complexed; 3. Selectivity of the active site. Allows the desired ion to be removed and other ions to remain in solution; 4. The degree of cross linking or polymerization of the extractant in the diluent. Allows high concentrations of the desired ion to be removed for a small concentra- tion of extractant. 18 Extractants are not easily chosen. There is an almost unlimited number of choices facing the researcher. Some extractants can be eliminated because they are not commercially available. Of the extrac- tants commercially available, two general groups emerge. The amines act by fOrming an organic soluble salt with anions while alkyl phos- phates react with cations and, therefore, were chosen for study in aluminum extraction. Typical molecules of alkyl phosphoric acids are shown in Figure 3-2. 3.6.4. Modifying Agents In addition to the diluent and the active extractant in liquid- ion exchange, there is often a third reagent called a modifier. The modifier may be added for one or more of three basic reasons. First is the synergistic effect that they may have on some extraction pro- cesses. Second, the modifier may improve phase separation. A third reason for the addition of modifiers is to prevent the formation of some insoluble compounds in the organic phase. For example, the addition of tributyl phosphate will prevent the formation of some insoluble sodium alkyl phosphate salts. Other modifying agents include isopropyl alcohol, phosphine oxides, and phosphonates. 3.7. Selectivity Whether or not liquid-ion exchange can be adopted for a specific application usually depends on the ability of the reagents to remove the desired ion and leave unwanted ions in solution. Selectivity in extraction is expressed by the selectivity ratio. The selectivity ratio is the extraction coefficient of the desired ion to be removed 19 Figure 3-2. Schematic diagram of alkyl phosphoric acids. a) Mono(2-ethylhexyl) phosphoric acid b) Di(2-ethy1hexyl) phosphoric acid 20 We owe orig-CH3 9H2 21 in the circuit divided by the extraction coefficient of the unwanted ion. The greater the magnitude of the ratio, the more pronounced the separation of the desired and undesired ions. Prognoses made from separation factors are trueonly if the extraction coefficients are nonvariant over the entire range of com- positions. Careful analysis of initial concentrations and synergistic effects must be considered when utilizing selectivity ratios. 3.8. Solids Handlings During Extraction The literature indicates mixed success in operating extraction circuits with solids. Although inert solids tend to decrease phase- disengagement rates and increase loss of solvent by entrainment, slurries containing a few percent solids are employed in refining uranium by tributyl phOSphate and in recovering by-product uranium from phosphoric acid [16, 28]. Other laboratory investigations of uranium leach slurries have reported organic feed losses too high or recovery schemes too cumbersome fbr commercial adoption [29, 30]. No literature can be cited which attempted to incorporate organic solids into the extraction circuit as was attempted in this research. 3.9. Data Development in Liquid-Ion Exchange Due to the high number of solvent-extractant-modifier combinations possible some preliminary exploratory work is often needed. The most common procedure for starting the study is to perfbrm what are called “batch shake out tests." The organic phase mixture is shaken in a separatory funnel with some of the solution from which the extraction is to be made. After one or two minutes of shaking the phases are 22 allowed to separate and each phase is analyzed. This gives a guide as to what solvent combinations warrant further investigation. The visual observations made in this step regarding rate of phase disen- gagement, interfacial scum, insoluble compounds, etc., are very important. After choosing some potential combinations, further extraction evaluation must be made and development of the fundamental design accomplished. This phase of the analysis can be divided into two steps. The first consists ofbdevelopment of distribution isotherms and the second consists of laboratory testing of the continuous flow process. Data for distribution isotherms may be obtained in two ways: 1) single contacts of the aqueous feed with different volumes of organic feed, and 2) single contacts of an aqueous feed consecutively with fresh organic feed. Both procedures are used in the laboratory to design continuous flow processes. The first does have limitations. The limitation of the first method is that the pH of the feed changes as it is contacted with different volumes of organic solvent. Since the extraction reaction is pH dependent (extraction increases with increasing pH) the resultant equilibrium data points are not representative of continuous flow operation. As a result, the second method was chosen in developing distri- bution isotherms in this research. Slight modification was utilized by maintaining the same initial aqueous feed pH. After the development of distribution isotherms, design of the continuous flow process must proceed. This can be accomplished by 23 utilizing McCabe-Thiele diagrams. These diagrams can be used to pre- dict the number of countercurrent stages required to achieve any given percentage extraction of the desired metal. 3.10. Liquid-Ion Exchange Equipment The extractive metallurgy industry has primarily adopted the mixer-settler concept for equipment design [31]. There are a number of important reasons for this. Differential contact extractors such as spray columns, plate columns, and packed columns are all charac- terized by incomplete separation of the two phases after mixing. Some of the dispersed phase will be carried along by movement of the con- tinuous phase. The net effect is a loss in efficiency.' Mixer-settlers achieve higher separation and approach ideal stages much better. Another reason in favor of mixer-settlers is that the device is amenable to handling slight amounts of solids which sometimes occur in feed solutions. They also involve relatively small capital costs and are very simple to operate. With proper design all flows can be ob- served by the operator and sampled as required. They can be started up or shut down without any control problems. The third and perhaps most compelling reason for using mixer- settler equipment is that a fast transition can be achieved from laboratory evaluation to operating the process in a continuous-flow system. Laboratory equipment is available which can allow direct scale-up to pilot plant studies. This equipment is specifically manufactured with this goal in mind. This allows final design para- meters to be developed during the laboratory evaluation. CHAPTER 4 EXPERIMENTAL APPARATUS AND PROCEDURES 4.1. General Description The extraction and stripping experiments were divided into two categories: tests on synthetic feed solutions, and tests on treatment plant alum sludges. All of the extraction and stripping experiments were performed in batch mixer-settler units. Research as previously discussed has shown that batch mixer-settler data is reliable to pre- dict and design continuous flow operations. 4.2. Aqueous Feed Solutions 4.2.1. Synthetic Feed Solutions All synthetic feed solutions were prepared with reagent grade chemicals and distilled water. All glassware and operation materials were rinsed with tap water, cleaned with either dichromate acid or soap cleaning solution, rinsed three times with tap water, and three rinses with distilled water. Glassware that had contained organic matter was rinsed with acetone during the cleaning process. Aluminum potassium sulfate (185.74 grams per liter) was used as the source of aluminum for preparation of 10,000 mg/l aluminum stock solution. Concentrated nitric acid (15 ml/l) was added to facilitate dissolution. All synthetic solutions were prepared from this stock. Sulfuric acid and/or potassium hydroxide were used for pH adjustments. The synthetic solutions allowed large quantities of feed solutions to 24 25 be made up at one time. Feed parameters could be easily varied. 4.2.2. Treatment Plant Alum Sludge Solutions Due to the variability of alum sludges from all geographical locations in the United States, one sludge had to be chosen for the alum sludge studies. The Tampa, Florida sludge was selected. The reasons were that the sludge had a high organic solids percentage, was extremely colored, and contained a very high aluminum concentra- tion (3000-3500 mg/l A13+). The 65-mgd Tampa water treatment plant utilizes an average 100 mg/l liquid alum and 4-mg/l sodium silicate for coagulation and settling. The sludge contains 0.6% solids from the sedimentation tank, 1.7% solids from the lagoons, and nearly 20% solids after de- watering on sand drying beds. The sludge samples shipped to the laboratory were pumped from the bottom of a wet well immediately following the lagoons. A typical analysis of the sludge is presented in Table 4.1. This sludge was felt to characterize the worst conditions that would be met in practical application. With the solids, organic losses could be quantitatively measured in the extraction circuit. Iron, manganese, and color contamination problems were expected to be encountered. The high aluminum concentration would test the efficiency of the liquid-ion exchange process. Upon receiving sludge samples, immediate refrigeration was employed until the samples were utilized in the experiments. Samples were acidified with concentrated sulfuric acid to dissolve the aluminum from the solids where applicable. 26 TABLE 4.1 Tampa Sludge Characteristics Parameter Dissolvable Inorganic Solids, % 61 Non-Dissolvable Inorganic Solids, % 6 Dissolvable Organic Solids, % 25 Non-Dissolvable Organic Solids, % 8 Suspended Solids, % 1.6 Total Aluminum Concentration, mg/l 3300 pH 6.58 4.3. Organic Feed Solutions 4.3.1. Alkyl Phosphoric Acids Three alkyl phosphates were evaluated during this research. Two different alkyl phosphoric acids were supplied by Stauffer Chemicals, Eastern Research Center, Westport, Connecticut. One was di(2-ethy1hexyl) phosphoric acid (DEHPA) and the other an equal molar mixture of mono- and di(2-ethy1hexyl) phosphoric acid (MDEHPA). Octylphenol acid phosphate was supplied by Mobil Chemical Company, Phosphorus Division, Richmond, Virginia. Properties of the three acids are shown in Table 4.2 and Table 4.3. All of the chemicals were commercially available in large quantities at the time of this research. The acids were used as directly supplied by the manufacturers since the chemicals would not be further purified in full-scale operation. The alkyl phosphoric acid solutions were prepared by volumetrically measuring out the appro- priate amount of solution and diluting in the proper diluent. A11 molar solutions of alkyl phosphates were reported as formal weights based on the average molecular weights in Table 4.2. 27 .mnmp xgmzenmd .mwcwms_> .ccosguwm .mpmuwsmcu magosqmoga .acmasou PmuPEmgu Fwnozrt .mump um3m3< .uaowuumccou .pcoaummz .mmumcamoca uwo< Fxxp< .xcmasou qumemzu commampmr "mucaom cm» “swab cwpom Agomov omo.p ..... mum armpmcamosa ku<, -wsmm Pocmgq—xuuo aeo moma agate: moeaeamaea copou Pmmexsa owmwumqm acmucma .Poz mmmsm>< Pxxp< mmpmcamozm u_u< PAxF< mo mmwucmaoca pmqumcu use pmuwnga ~.¢ m4m

.ugosnuHm .mHmuHEmsu maeocamoga .xcmasou HmuHEmsu Fwnozrr .mnmp .um:m:< .pzuwuumccou .usoapmmz .mwpmgamoga cwu< FaxH< .xcmasou Hmowsmsu gmmmsmumr .Hmueaom mpummm u m mpasHomcH u H mpnzpom u m Hmcoo H H H H --- m m H aropaeamoea eHo< Poem;a_xuuo m m m --- m m m m ruwu< owcosamoga apaxmgpxguormv HAJHmHo m m m 1.. m m m H ruwu< uwsocqmoga HHoneHzeoo-~v Hxxpmwouocoz umm m cue wcmmosmx mcmxmz epuo mcopmu< Honoup< Lopez mopmgamoga unscmx unscmx ”AxH< maoaeamaea aHo< HaxH< Ho maHHeoaaea HHHHHasHam m.a HHm1800) Fe (II) 475 5.0 170 Fe (III) 475 19.0 30 -Cr (VI) 475 10 O 75 Two Cu (11) 640 140 11 Cd (II) 640 70 22 Mn (II) 640 80 20 Zn (II) 640 10 178 Fe (II) 640 85 19 Fe (III) 640' 320 ‘ 2 640 210 7 Cr (VI) application, selectivities are usually much higher than in the labora- tory [22]. As a result, no contamination of heavy metals would be expected in operation of the aluminum recovery system. 52 5.3. Stripping of Aluminum - Mono-Di(2-ethylhexyl) Phosphoric Acid 5.3.1. Introduction Stripping of aluminum can be accomplished by either acid or alkali. The alkali has had mixed success in the literature and was not attempted in this research. Acid solutions studied were hydrochloric and sulfuric acids. 5.3.2. Kinetics Study The first tests conducted in the stripping circuit were kinetic tests. These were carried out in the same manner as the extraction kinetic tests. The results of varying impeller speeds to achieve equilibrium can be seen in Figure 5.7. A 6 N H2504 acid solution was contacted with loaded organic containing 640 mg/l aluminum at a 3:1 phase ratio (organiczaqueous). Complete stripping equilibrium was achieved at an impeller speed of 800 ft/min with a mix time of 12 minutes. Impeller speeds greater than 800 ft/min gave no higher stripping efficiencies, but did achieve equilibrium in a shorter time of 8 minutes. The reaction initially proceeded more quickly than the extraction process; approximately 90% equilibrium at the end of 2 minutes as compared with the extraction process being 75% complete in 2 minutes. Figure 5.8 shows the percent equilibrium versus impeller speed in the stripping circuit. The stripping process equilibriated at 800 ft/min and increasing of the impeller speed did not enhance the reaction. Higher equilibrium percentages were achieved with the strip- ping reactions than could be achieved with the extraction reaction at the lower impeller speeds. At 200 ft/min, the stripping equilibrium 53 Figure 5.7. Impeller speed versus time to reach stripping equilibrium. Organic = 640 mg/l A13+, Acid = 6 N H2504, 3:1 phase ratio. 2H: wsz xH: 54 .om .mN .DN .3 .2 . .m .o H H H q H H .mb ... .oo 2H5: 08H 0 ES: o8 x ZED... 8m 4 -.ma 2H5: HHS a .om .mm L11 “1‘1 if Lil PH‘ if [HOOn ”0188111003 lNBOUBd 55 Figure 5.8. Percent equilibrium versus impeller speed in the stripping circuit. Mix time = 15 minutes, Organic = 640 mg/l Al +, Acid = 6 N H2804, 3:1 phase ratio. PERCENT EQUILIBRIUM 56 1001 95. 90. 85. 80. 75. 70. P 65. l l l l 200. 400. 600. IMPELLER SPEED eoo . FT/MIN 1000. 57 had reached 92% while the extraction equilibrium was only 75%. All further tests were evaluated at 800 ft/min with a mix time of 15 minutes to ensure equilibrium conditions had been reached. 5.3.3. Acid Evaluation The next tests evaluated the performance of different acids in the stripping circuit. Figure 5.9 shows the results using hydro- chloric and sulfuric acid. Loaded organic containing 640 mg/l alu- minum was contacted for 20 minutes at a 3:1 phase ratio. The results indicate that 6 N HCl and 9 N H2504 are equally effective in stripping the aluminum loaded organic phase. Based on the cost of each acid, sulfuric acid was chosen as the stripping acid of choice in all further experiments. Sulfuric acid also supplies the sulfur for regenerated aluminum sulfate, the alum coagulant used in water treatment facilities. The hydrochloric strip solutions were yellowish to orange while the sulfuric strip solutions were crystal clear. The next experiments were conducted to evaluate acid stripping as a function of acid normality and phase ratio. The results are shown in Figure 5.10. Loaded extract containing 640 mg/l aluminum was contacted with sulfuric acid normalities ranging from 3 N to 12 N . and with phase ratios of 3:1, 5:1, 10:1, 15:1, and 20:1. The 9 N H2504 acid at a phase ratio of 3:1 was shown to have the highest stripping efficiency. It would be expected that lower phase ratios (i.e., l:l) would give even better stripping efficiencies. 5.3.4. Development of Stripping Equilibrium Curves The final tests on the synthetic solutions involved establishing stripping distribution curves. Varying concentrations of organically 58 Figure 5.9. Comparison of acid type and normality on stripping efficiency. Mix time = 20 minutes, Organic = 640 mg/l A13+, 3:1 phase ratio. 59 .v— >HHH¢zmoz oHua .2 .m .m H H H oHum Ho: oHuc vomu: 4 Hu .2. .oo .mo .om .mm .9: 03861818 NONINO1H 1N33836 60 Figure 5.10. Acid stripping as a function of H SO normality and hase ratio. Mix time = 15 minutes, Organ c = 640 mg/l A1 +, Acid = 3 N - 12 N H2504, Phase Ratio = 3:1 - 20:1. eomm: >HHchmoz .4H .NH .oH .m .o .a .u .o 61 H H H H H H H .Ow oHEm Has: :8 x SHE HE: :E e -.8 2:; ”acre :2 x 2:; HE: Ham a 1.2 SHE ”.55: a I .om l .om JODH 03ddIHlS HONINDWU lNBDHEd 62 loaded aluminum, contacted in the monomer extraction range, were con- tacted with 6 N sulfuric acid. Six isotherms were developed by varying the organic to aqueous phase ratio. High phase ratios were desirable in order to concentrate the aluminum to the concentration of commercial liquid alum (49000 mg/l Al3+). The results can be seen in Figure 5.11. Each curve illustrates an initial segment where essentially all the aluminum was stripped. A fairly linear portion was next exhibited. The curves were not seen to level off. This is due to the high alu- minum concentration required initially in the organic phase and diffi- culties in achieving this in the laboratory. A concentration of stripped aluminum equal to 64000 mg/l was achieved with a 40:1 phase ratio. 5.4. Selection of Support Extractants While the research of Cornwell [26] and this study fully support MDEHPA as being an effective extractant, a need was felt to have a second extractant that would have applicability in aluminum complexing. TWo other extractants were evaluated for use in the alum recovery process. Octylphenol acid phosphate was analyzed for use as an extractant. Problems were encountered in dissolving the octylphenol acid in the diluent. The manufacturer,_ Mobil Chemical Company, was notified by the researcher and currently is undertaking research to enhance their product's solubility. Experiments on this extractant were discontinued at that time. When and if Mobil can resolve this problem, octylphenol acid phosphate may show applicability in aluminum extraction. Di(2-ethylhexyl) phosphoric acid (DEHPA) was evaluated for use 63 Figure 5.11. Synthetic stripping equilibrium curves for various phase ratios of H S0 acid. Mix time = 15 minutes, Organic = 500-5000 mgi1 A13+, Phase Ratios - 3:1 - 40:1. G FlLUHINUll / L RECOVERED FlLUl‘l 70- SO- 50- 40. 30. 20. to. 64 ‘6 3* ‘0 IX 9 [3 l 1.0 40:1 20:1 15:1 10:1 5:1 3:1 PHHSE PHRSE PHRSE PHRSE PHHSE PHHSE l 2.0 RRTIO RRTIO RHTIO RRTIO RHTIO RRTIO 3.0 0~HLUUINUM / L EXTRFlCT 65 Figure 5.12. Comparison of synthetic extraction equilibrium curves utilizing MDEHPA and DEHPA as extractants. Mix time = 15 minutes, A13+ = 500-600 mg/l, 1:1 phase ratio. G RLUMINUH / L EXTRFlCT 66 0-4 M + x r- x x Hr. E] NDEHPH X DEHPFI x x x 0-1 N l l l 1 . 1 0. 1.0 2.0 3.0 4.0 5.0 G RLUHINUM / L RHFFINRTE - 67 as an extractant. Equilibrium isotherms were established and compared to the MDEHPA isotherms already established. The results are shown in Figure 5.12.’ The results indicate that DEHPA is at least equal and probably better in extraction effectiveness than MDEHPA. However, addi- tion of a neutral compound, tributyl phosphate, was needed to help prevent a third phase fbrmation during settling. The third phase con- tained both aqueous and organic droplets that had a tendency not to- disengage. The phase disengagement time was also longer. As a result, DEHPA was concluded to be an extractant worthy of consideration as a backup to MDEHPA. The extractant would require more control during operation in order to prevent third phase forma- tion. No further studies were undertaken during this research. 5.5. Selection of Support Diluents This and other liquid-ion exchange research has successfully used kerosene as the diluent for the organic phase. Recent research has developed other fractionalized crude oil deriviatives for use as diluents. Specifications are closely controlled so that uniformity of the product can be maintained. This is not done with kerosene. Evaporation and fire hazards have been minimized with these diluents. Manufacturers also claim faster rates of phase disengagement which result in minimum solvent entrainment losses and lower settling area requirements. Two of these diluents, Kermac 4708 and 627, were evaluated for use as organic solvents. Both Kermac 4708 and Kermac 627 were manu- factured by Kerr-McGee Refining Corporation. Extraction isotherms were established in exactly the same manner as previously outlined 68 for the MDEHPA, but the new diluents were substituted for kerosene. The results of these diluents on extraction efficiency can be seen in Figure 5.13. The results indicate that these two diluents were equally as effective as kerosene in aluminum extraction. Phase disengagement rates were next evaluated. This is an impor- tant parameter for determining settler area and detention time. 0b- servations revealed that all three of the diluents exhibit rapid phase disengagement. Secondary break was complete in 3-8 minutes for the Kerr-McGee products and in approximately 10 minutes for kerosene. Evaporation losses were evaluated for four days for the three diluents. The tests were not run under controlled conditions so the results are only relative. Both Kerr-McGee diluents had losses of 160 m1/day/m2 compared to 460 ml/day/m2 for kerosene. The physical and chemical characteristics reported by the manu- facturer in conjunction with the demonstrated effectiveness in opera- tion indicates that either Kerr-McGee diluent could and should be used as a replacement for kerosene in future research. The costs for the Kerr-McGee products were very comparable to kerosene at the time of this research. The only reason these diluents were not utilized further was their reduced availability in the time restraints imposed. 5.6. Solvent Loss Considerations Loss of the diluent or extractant into the aqueous phase during extraction is an important economic consideration. Losses within the circuit can occur by evaporation, by chemical attack, or by entrain- ment of the organic phase into the aqueous phase during mixing. The first two sources of loss are minimal especially where commercial 69 Figure 5.13. Comparison of diluents on extraction efficiency. Mix time = 15 minutes, Al3+ = 500-6000 mg/l, l:l phase ratio. G HLUMINUM / L EXTRHCT 1.75 1.50 1.25 1.00 0.75 0.50 0.25 0.0 70 r ., 0.2 M -—ah—4& 0.1 fl _‘__ El KEROSENE DILUENT X KERMRC 4708 DILUENT t KERl‘lRC 627 DILUENT 1 1 1 . 1 1 0. 1.0 2.0 3.0 4.0 5.0 G RLUMINUM / L RRFFINRTE 71 diluents are used that specifically try to eliminate these factors. The loss of solvent by entrainment in the raffinate constitutes the largest source of losses within the circuit. Entrainment losses are in turn related to the dispersion phase during mixing. Research has shown that when the extraction circuit is operated organic continuous, aqueous dispersed, organic losses can be minimized. In order to monitor organic losses, two analytical methods were employed. One utilized total organic carbon analysis to determine the combined diluent and extractant losses. The second method analyzed extractant losses only by a spectrophotometric method. In this way both diluent and extractant losses could be quantitatively measured. Synthetic solutions of different aluminum concentrations, dilu- ents, and molarities of MDEHPA were contacted at a 1:1 phase ratio. All extraction circuits were operated organic continuous. The raf- finate solutions were analyzed for total organic carbon and MDEHPA concentrations. Table 5.2 shows the results. The results indicated that most of the organic losses were associated with the diluent (83% as mg/l C). Aluminum concentration, the type of diluent, or the molarity of extractant did not have any effect on organic losses. Another important result was that the MDEHPA extractant entrainment was 180 mg/l as extractant. It was felt that much of the organic losses were unavoidable due to the nature of the settling apparatus. The organic solution adhered to the sidesof the separatory funnel and were flushed out with the recovered raffinate solutions. Also it was believed that organic losses would taper off after repeated contacts with feed and acid solutions due to impurities in the diluents initially. Another 72 ao.o mHaH mm oaH ooom Hme descox «axmoz z H.o H< ooom mm.o oamH oHN Doe ooHN Hue oaecog Harmaz z «.0 :H 88 ao.o oaou om oHH ooHN HOHe oaseox F H mmHoz m.o ¢.o , l m.o «.0 do .oooH 87 6.2.4. Summary The major problem associated with the introduction of solids into the extraction circuit was the scum layer formed at the organic- aqueous interface. This problem was anticipated since other extrac- tive metallurgic processes had reported similar problems in operation [29, 30]. These solids accumulated at the interface and did not seem to interfere with the extraction operation. A possible problem associated with this scum build up may result if the layer keeps increasing in thickness. Research with slurry operations of this kind have shown that the thickness of the layer reaches a certain level and then increases only slowly with extended operation [34]. A continuous flow study is necessary to fully evaluate this possible complication. 6.3. Stripping of Aluminum-MDEHPA 6.3.1. Kinetics Study The initial stripping tests on the Tampa sludge was a kinetic study. The results indicated that the sludge stripping circuit opera- ted in exactly the same manner as the synthetic feed solutions. A mix time of 12 minutes at 800 ft/min resulted in equilibrium being reached. This agreement was expected since no solids were present in the stripping circuit. 6.3.2. Comparison of Stripping when Extraction Occurs in the Monomer and Dimer Ranges As was stated earlier the stripping circuit was greatly depen- dent on whether the extraction was operated in the monomeric or dimeric ranges during extraction. Figure 6.5 exhibits the stripping as Figure 6.5. 88 Aluminum stripping as a function of extraction in the nonomer and dimer ranges. 1) 2) 3) Monomer extraction range - 0.35 M MDEHPA contacted in two stages with the Tampa sludge (pH = 2.0) in a 1:1 phase ratio. Final A13+ concentration in the organic phase = 2300 mg/l. Dimer extraction range - 0.52 M MDEHPA contacted in one stage with the Tampa sludge (pH = 2.0) in a 2:1 phase ratio. Final Al3+ concentration in the organic phase = 1500 mg/l. Stripping conditions - 3 N H 50 2 4, 3:1 phase rat1o. G HLUMINUM / L RECOVERED RLUM 6. 1. 89 E] A l EXTRRCTION IN THE MONUMER RRNGE EXTRRCTION IN THE DIMER RRNOE ......— l I l 0.2 0.3 0.4 G HLUMINUM / L EXTRHCT 0.5 90 functions of extraction in the monomeric or dimeric ranges. One aluminum loaded organic sample contained 0.52 M MDEHPA contacted in one stage with the Tampa sludge at a phase ratio of 2:1. This repre- sented the extration circuit in the dimer range. The loaded organic phase contained 1500 mg/l aluminum. The other organic sample con- tained 0.35 M MDEHPA contacted in two stages with the Tampa sludge at a phase ratio of 1:1. This represented the monomeric extraction cir- cuit. This organic phase contained 2300 mg/l aluminum. All of the organic was recycled to the next contact stage during the stripping circuits. Both organic phases were then contacted with 3 N H2S04 at a 3:1 phase ratio. Table 6.2 summarizes the equilibrium concentra- tions, stripping coefficients, and the cumulative percents of alu- minum stripped for the two extraction ranges. The figure clearly shows the higher stripping efficiencies achieved when the extraction circuit was Operated in the monomeric range. $0 With a phase ratio of 3:1, 3 N H could strip 100% of the 2 4 aluminum in the loaded Tampa organic (2300 mg/l A1) in four contacts when the extraction circuit was operated in the monomeric range. The same acid, operating under the exact conditions, could only strip 91% 13+) in five contacts of the aluminum loaded organic phase (1500 mg/l A when the extraction was operated in the dimer range. This clearly shows the advantages of operating the extraction circuit in the * monomer ranges. 6.3.3. Development of Stripping Equilibrium Curves All stripping equilibrium curves utilized Tampa sludge which had been extracted in the monomer range. Molarities of extractant 91 --- --- o o m ooH --- me o 8 mm o.m mm mH m mm o.o cam oe N ma m.am came ONH Hum aom~= zm H .mmmmw stoco: Hm N.o om ocH m cm ¢.o on omH e mm ~.H oHN oHH m cm m.H omv oem N «H m.m comm can Hum aomu: zm H mad. acmugma .eoachHm . HHHHEH HHHHEH Homeccem Ech53H< chHoHHHmou E==H53H< EchsaH< HuHuHumH=e=u mcHaaHeum nonaHHHm HueHme oHumm omega uHu< uumucou momcum :oHHumguxm HmEHo ecu Hoeocoz on» :H oumm H \ :DZHzaHm a 0. 3.0 4.0 8.0 O HLUMINUM / L EXTRHCT 2.0 1.0 0. 106 phase was recycled to obtain higher feed aluminum concentrations in the extraction circuit. The final alum concentration was 49% as alum. At the organic:acid flow rate of 15:1, there would be 9070 gpd of recovered alum and 136000 gpd organic solvent. Mix times of 15 minutes would be required in each mixer and settler. Each tank would have a volume of 1510 gallons. The total acid requirement would be 1515 gpd of concentrated sulfuric acid (11.24 tons per day). 7.6. Organic Recycle The stripped organic phase was recirculated to the extraction circuit.) The extractant successfully reproduced the initial extrac- tion efficiencies. Since the solvent contains excess acidity, an additional stage may be beneficial. The solvent could be washed with water at a 1:1 phase ratio before recirculation. A portion of the aqueous stream could be continually bled off and directed to the acidifer to help lower the pH of the sludge solution. Overall, 11315 gal or organic solution are needed in the process system at one time. The solvent would cycle through the system 12 times per day. 7.7. Recovered Alum Solution The total volume of recirculated alum was 9070 gpd. The concen- tration was 4.9% A13+ , the same as that of commercial liquid alum. The recovered alum was successfully reused for coagulation of a raw water. It was calculated that 6.3 gpm pump would be required for alum recirculation. 107 7.8. Summary The percentages of aluminum recovered in the acidification, extraction, and stripping stages were 85%, 99%, and 100% respectively. This resulted in an overall recovery of 84%. CHAPTER 8 CONCLUSIONS AND RECOMMENDATIONS The stated objective of this research was to evaluate the future feasibility of using liquid-ion exchange for alum recovery and recycling in water treatment plants. A process was developed whereby contaminate free alum could be recovered from water effluent and reused as a coagu- lant. In this chapter a summary of the process design for alum recovery is presented. This is fallowed by recommendations for future research. 8.1. Summary of Alum Recovery by Liquid-Ion Exchange At a pH of 2, essentially all the aluminum present in the sludge was dissolved. During pH reduction, the organic solids would floc- culate and settle. The dissolved aluminum could be separated from the solids by sedimentation. Sedimentation resulted in 85% aluminum separation. The acidified aluminum was separated from the supernatant by a liquid-ion exchange process. The supernatant was contacted with kerosene containing a 0.84 M solution of extractant in a 1:1 volume ratio. The extractant was an equal molar mixture of mono- and | di(2-ethylhexy1) phosphoric acid (MDEHPA). A minimum of 99% of the aluminum reacted with the MDEHPA and became kerosene soluble, result- ing in separation from the supernatant. IA two stage, countercurrent extraction circuit would be required for aluminum recovery. Additionally, several findings pertaining to the extraction 108 109 circuit were reported. Iron (II), manganese (II), and color did not contaminate the kerosene-aluminum solution. Organic losses were mini- mized by operating the circuit organic continuous. When the extraction circuit operated in the monomer range of the extractant, stripping efficiencies were greatly enhanced. Commercial diluents, especially manufactured for liquid-ion exchange processes, lowered organic losses of evaporation and entrainment when compared with kerosene. No modifying agents were required far the aluminum-MDEHPA complexing. Di(2-ethylhexyl) phosphoric acid (DEHPA) was concluded to be an extrac- tant of practical utility in the extraction process. The aluminum rich organic phase was contacted with 6.N H2504 to farce the aluminum into the acid phase. The organic:acid volume ratio was 15:1. Essentially 100% of the aluminum entering the strip- ping circuit went into the acid phase when operated in a two-stage countercurrent circuit. The final alum concentration was 49% as alum. The organic solution was recycled and reused in the extraction circuit. The recovered alum was successfully reused as a coagulant in the water treatment plant. . The overall aluminum recovery was 84%. 8.2. Suggested Future Research It is the recommendation of this research that the alum recovery process be continued during the second year of the research grant period. Application should be extended to a laboratory, continuous flow process. APPENDIX APPENDIX A APPENDIX A GLOSSARY FOR LIQUID-ION EXCHANGE Alkyl Phosphate. - A long-chaned phOSphoric acid of the general form R =P(O)OH. Each R group may be on 8-12 carbon chain or one R mgy be an additional OH group. Antisynergism. - Suppression of extraction caused by using a combination of extractants or diluents; antonym of synergism. Aqueous Phase. - Aqueous solution containing the salute to be extracted. Carrier. - Inert organic solvent in which an active organic extractant is dissolved; also referred to as the diluent, or solvent. Coalescence. - Growth or combination of small dispersed droplets into larger drops. Cocurrent multistage contact. - Sgg_Crosscurrent extraction. Combination extractant. - Organic solution of two or more extractants. Compartment-type mixer-settler. - Multiple-stage contactor featuring adjacent compartments sharing common interior walls. Contactor. - Device for dispersing and disengaging immiscible solu- tions; extractor. May be single stage, as in a mixer-settler, or multiple stage, as in columns and certain centrifuges. Continuous phase. - Bulk component that contains droplets of the dis- persed component in a mixture of two immiscible solutions. Countercurrent extraction. - Multistage extraction in which the aqueous and organic solutions flow in opposite directions. Crosscurrent extraction. - Treatment of a batch of aqueous solution by repeated contacts with fresh organic extractant; also called simple multiple contact, cocurrent multistage contact, and multi- stage cocurrent extraction. Differential extraction. - Procedure for extracting a batch of aqueous salution by continuously feeding and simultaneously withdrawing organic feed from a contractor; differs from crosscurrent extraction in that organic feed is introduced continuously instead of batchwise. llO 111 Diluent. - §g§_Carrier. Dispersed_phase. - Component that is diffused as droplets throughout the continuous component in a mixture of two immiscible solutions. Dispersion. - Mixture of immiscible phases in which one phase is diffused throughout the other (continuous phase). Distribution coefficient. - See extraction coefficient, E°. or strip- ping coefficient, S. Distribution isotherm. - Graphical representation of isothermal equilibrium concentrations of a metal solute in aqueous and organic solutions over an ordered range of conditions in extrac- tion (extraction isotherm) or stripping (stripping isotherm). Also equilibrum curve or distribution curve. Emulsion. - A mixture consisting of small droplets of one liquid dispersed in a continuum of another immiscible liquid. Equilibrate. - To disperse and disengage aqueous and organic solutions for the purpose of determining the equilibrium concentrations of metal solute in the respective phases. Equilibrium curve - See distribution isotherm. Extract. - Organic phase after extraction (loaded solvent). The solu- tion into which transfer of a metal solute is effected; used as a verb to describe transfer of a metal solute between two immiscible liquids. Extractant. - Organic soluble compound which causes distribution of'the metal solute to favor the solvent phase; chelating compound. See alkyl phosphate. Extraction coefficient, E°. - Ratio of the metal concentration in the organic extract to the metal concentration in the aqueous raffinate. Extraction isotherm. - Sgg_Distribution isotherm. Extractor. - Synonym for contactor, or mixer. Feed. - Aqueous solution containing the metal solute to be extracted. Flooding. - Discharge of mixed phases from one or both exit ports of a contactor. Flooding may occur in a single-stage contactor and in any or all stages of a series of contactors. Fractional_(double)solvent extraction. - Process in which two immis- cible organic liquids (double solvents) are passed countercur- rently through a multistage contactor to separately extract metal solutes from an aqueous feed introduced at an intermediate stage (usually the middle stage). 112 Internal mixer-settler. - Contacting device in which the mixer (usually shrouded) or mixer compartment is within the settler. Internal recycle. - Circulation of aqueous or organic solutions from a settler to the mixer in the same stage far control of the phase ratio during mixing independently of the feed ratios. Inversion. - Change in continuous phase from organic to aqueous or vice versa; breaking an emulsion by treatment with an excess of the dispersed phase. Liquid-Ion Exchang_. - Solvent extraction when solute transfer involves the exchange of cations or anions between phases. Loaded organic. - Organic solvent containing metal solute after con- tacting the aqueous feed liquor; the extract. ‘ Loadinggcapacity. - Saturation limit of metal solute in organic or strip liquor. McCabe-Thiele diagram. - A composite plot of the distribution isotherm and the operating line. It is used for estimating the theore- tical extraction stages required to obtain specific results in a solvent extraction system. The diagram can be prepared for either extracting or stripping operations. Mixer-settler. - Device for liquid-liquid extraction comprising separate mixing and settling compartments. Modifying agent. - Substance added to an organic solution to increase the solubility of the extractant or of salts of the extractant that form during extraction or stripping. Operating line. - Curve depicting the relationship between the metal salute content of organic and aqueous solutions in a countercur- rent system. In a McCabe—Thiele diagram (Cartesian coordinates) of solute distribution between two immiscible phases, the opera- ting 1ine is linear with a slope equal to the ratio of feed to solvent. The line contains points representing the solute con- centration in the influent and effluent streams throughout the system. Organic phase. - Organic solvent. Phase inversion. - Reversal of the continuous and dispersed phases, Sgg,lnversion. Phase ratio. - Volume ratio of the Organic solvent to the aqueous feed. Phase separation. - Separation of immiscible solutions into separate layers do to differences in specific gravity. Primary break. - Separation of a dispersion into two layers with a distinct common boundary. 113 Raffinate. - The liquid phase from which solute has been removed by single- or multiple-stage contacting with an immiscible solvent. Reextraction. - Sag stripping. Scrub. - Removal of impurities from the solvent prior to recirculation of the solvent into the extraction stage. The scrub stage usually follows the stripping operation. Secondary break. - Coalescence and separation of a fine dispersion present in either or both phases after the primary break. Sedimentation. - See Phase separation. Selective extraction. - The specific removal of a desired solute from a feed solution containing two or more solutes. Selectivity. - Ability to extract one solute from a mixture of solutes. Selectivity coefficient. - Ratio of the extraction coefficients of txo substances, used to express selectivity. Designated by or SB Settling. - Separation of dispersed immiscible liquids by coalescence and sedimentation. Solvent. - In liquid-liquid extraction, the liquid phase that pre- ferentially dissolves the extractable solute from the feed. Often the term is used to describe the organic phase. Solvent extraction. - Separation of one or more metallic solutes from a mixture by mass transfer between innfiscjble phases in which at least one phase is organic liquid. Stage. - Single contact (dispersion and disengagement); sometimes refers to a theoretical stage which is a contact that attains equilibrium conditions. Involves one mixer and settler. Stagegefficiency. - Ratio of actual mass transfer in a specific stage to theoretical transfer in that stage under equilibrium conditions. Stripping. - Removal of extracted metal solute from loaded organic extract; reextraction; back extraction. Selective stripping refers to separate removal of specific metal solutes from an extract containing more than one metal solute. Also called back extraction or reextraction. Stripping coefficient, 5:. - Ratio of the metal concentration in the aqueous extract to the metal concentration in the organic raffinate. 114 Strippinggjsotherm. - Sgg_Distribution isotherm. Synergism. - Cooperative effect of two or more extractants that exceeds the sum of the individual effects. Wash. - Removal of contaminationg solutes from organic solution; scrub. APPENDIX B 11. APPENDIX B DETERMINATION OF ALUMINUM IN AN ORGANIC SOLVENT BY ATOMIC ABSORPTION SPECTROPHOTOMETRY Introduction 3+ by atomic A method was developed for determination of aluminum absorption spectrophotometry. Liquid-ion exchange was utilized to complex aluminum in the aqueous solution into an organic phase. Kerosene was used as the organic solvent for extraction. An equal molar mixture of mono-di(2-ethylhexyl)phosphoric acid (MDEHPA) was the extractant of choice. 3 Experimental A. Equipment - An Instrument Laboratories Model 151 atomic absorp- ‘tion spectrophotometer with a premixed burner was used. A hollow cathode lamp served as the light source and the wave- length used was 3092.7 A. Samples were aspirated by a nitrous oxide - acetylene flame and nitrous oxide burner head system. 8. Reagents - Commercial kerosene (number one fuel oil) and aluminum potassium sulfate were used for standard solutions. Deionized water was used for preparing standard solutions. MDEHPA, as supplied by Stauffer Chemical Company, was utilized. C. Standard Aluminum Solutions - 9.287 grams aluminum potassium sulfate was dissolved in deionized water to 1 liter and the pH adjusted to 2.0 with sulfuric acid. The aluminum concentration was 500 ppm Al3+. 115 116 D. Extraction Solution - 266 m1 of MDEHPA was dissolved in kero- sene to 1 liter (1.0 M solution of MDEHPA). E. Standard Organic Solution - Equal volumes of standard aluminum solution and extraction solution were mixed in batch reactors for 15 minutes at 800 ft/min. The solution was allowed to set- tle in a separatory funnel for 30 minutes. The organic solu- tion was drawn off and used as the 500 ppm Al3+ in the organic solution. Appropriate dilutions were made from this stock. The solution is stable for up to 3 months if tightly sealed. F. Operating Conditions - The operational parameters were so adjusted that optimum sensitivity could be obtained. These parameters were: current in the hollow cathode lamp - 13mA burner height - 10 (scale unit of the instrument) nitrous oxide flow rate - 11 SCFH oxygen-acetylene flow rate - so adjusted that the red feather of the flame was 3/4 - 1 inch in height rate of aspiration - 2-3 ml/min slit width - 320 0. Typical Absorption Curve for Aluminum Determination - See Figure 1. III. Procedure Unknown aqueous samples containing up to 5000 ppm Al3+ were contacted with the 1.0 M extraction solution in a 1:1 phase ratio. Samples were mixed for 15 minutes at 800 ft/min and allowed to separate. The organic phase was measured by atomic absorption spectrophotometry 117 Figure 1. Standard curve for aluminum determinations in the organic phase. 118 0.5 _ l l «1 01 c: c: SiINO N01188OSQU 41 fi' 0 0.1 H 100. 150. 200. 250- 300. 3507 50. PPM HLUMINUM IV. VI. 119 and compared to known standards. Interferences Other metal ions are not extracted by this procedure. ratios were reported: Metal Cu2+ Cd2+ Mn2+ Zn2+ Fe2+ Fe3+ Cr6+ Selectivity_Ratio 190 300 360 >1800 170 30 75 Effects of Solids in Unknown Solution Selectivity Aqueous solutions containing up to 3% solids had no effect on aluminum determinations. Aluminum determinations were conducted on the Tampa, Florida water treatment alum sludge. Alum was utilized as a coagulant in water purification. The aluminum concentration was found to be 3000 ppm A13+z Test mmth—I Aluminum Concentration, ppm 3300 3300 3250 3300 3300 3350 This compared very well with other techniques for aluminum determinations attempted. 120 VII. Discussion Liquid ion exchange is a very useful technique for aluminum determinations. The sensitivity in an organic solvent is vastly improved over aqueous techniques. This advantage is due to the improvement of the flame condition which increases the noise to signal ratio. BIBLIOGRAPHY 10. 11. 12. 13. BIBLIOGRAPHY Susan, J., "Characterization of Acid Treated Alum Sludges for Alum Recovery in Water Treatment Plant," Masters Thesis, Department of Civil and Sanitary Engineering, Michigan State University, 2 (August 1 977). Roberts, J. M. and C. D. Roddy, “Recovery and Reuse of Alum Sludge at Tampa," J. Am. Water Works Assn., 62, 857 (1960). Isaac, P. C. G. and I. Proc. Soc. Water Treat. Vahadi, “The Recovery of Alum S1udge,‘I Exam., 16, 91 (1969). Webster, J. A., "Operation and Experience at Daer Water Treat- ment Works, with Special Reference to the Use of Activated Silicia and the Recovery of Alum from Sludge,“ J. Inst. Water Mg 20: 2, 167 (1966). Tomons, K., Chief Design Section, Tokyo Water Works, Personal Communication (April 23, 1977). Am. Water Works Assn. Research Committee, "Disposal of Wastes from Water Treatment P1 Assn., 6l;11, 619 (1969). Westerhoff, G. P., and ants--Part 2," J. Am. Water Works M. P. Daly, "Water Treatment Plant Disposal--Part 2," J. Am. Water Works Assn., 6636, 378 (1974). Westerhoff, G. P., and M. P. Daly, "Water Treatment Plant Disposa1--Part 3," J. Am. Water Works Assn., 6657, 441 (1974). Am Water Works Assn. Research Committee, "Disposal of Wastes from Water Treatment Pl 61311, 619 (1969). ants--Part 2," J. Am. Water Works Assn., Am. Water Works Assn. Committee., ”State of the Art of Coagu- lation," J. Am. Water Works Assn., 66;2, 99 (1971). Brosset, C., et al., "S XI, The Aluminum Ion Al Matijevic, E., et a1. lation (III Aluminum)," Morrison, G. H., and H. ta dies on the Hydrolysis of Metal Irons: ," Acta. Chem. Scand., 8, 1917 (1954). “Detection of Ion Hydrolysis by Coagu- J. Phys. Chem. , 65, 826 (1961). Freiser, Solvent Extraction in Analy- tical Chemistry, John Wiley and Sons, Inc., New York, NTY., pp, 121 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 122 Thayer, H. E., "The Newest United States Uranium Processing Plant,“ Proc. 2nd UN International Conference onithe Peaceful Uses of Atomic Energy, Geneva, Vol. 4, 22-29 (1959). Stickney, W. A., “Zirconium-Hafnium Separation," Bureau of Mines Rept. of Investigations 5499, 22 pp., (1959). Greek, 8. F., and C. W. Allen, and D. Tynan, "Uranium Recovery from Wet Process Phosphoric Acid,“ Ind. Eng. Chem., vol. 49, 628 (1957). Black, K. I., and J. Koslon, and J. D. Moore, "Design and Operation of a Uranium Processing Mill Using Liquid-Ion Exchange (Solvent Extraction)," Proc. 2nd UN International Conference on the Peaceful Uses of Atomic Energy, Geneva, Vol. 3, 488 (1958). Carlsan, c. w., and R. H. Nielsen, "Columbium and Tantalum Oxides by Liquid-Liquid Extraction," J. Metals, 16, 472 (1960). Swanson, R. R., and D. W. Agers, ”A New Reagent for the Extrac- tion of Copper,“ Presented at the AIME Meeting, New York, (February, 1964). "LIX 70:- A Major Advance in Liquid-Ion Exchange Technology," 100th Annual AIME Meeting, New York, (February 1971). Agers, D. W., and C. R. Merigold, and J. E. House, “LIX 64N - The Recovery of Copper from Ammoniacal Leach Solutions," 99th Annual AIME Meeting, Denver, (February 1970). Agers, D. W., and C. R. Merigold, and J. E. House, "Copper Recovery from Acid Solutions Using Liquid-Ion Exchange," Mining Engineering, (December 1965). Agers, D. W., and E. R. DeMent, "LIX 64 as an Extractant for Copper Pilot Plant Data," Presented at a Joint Meeting Minerals Conference of the Society of Mining Engineers of AIME, (September 1967 . DeMent, E. R., and C. R. Merigold, "LIX 64N - A Progress Report on the Liquid-Ion Exchange of Copper," 99th Annual AIME Meeting, Denver, (February 1970). George, D. R., and L. E. lame, and S. R. Crane, and K. 8. Higbie, "Recovery and Production of Alumina from Waste Solu- tions by Solvent Extraction," J. of Metals, 59 (1968). Cornwell, D. A., “Recycling of Alum Used for Phosphate Removal in Domestic Wastewater Treatment," J. Water Pollution Control Federation, 49:4, 600 (1977). 27. 28. 29. 30. 31. 32. 33. 34. 35. 123 Kerr-McGee Refining Corporation, "Kermac 4708," Manufacturers Bulletin SR-13, Oklahoma City, Oklahoma, (October 1974). Burger, J. C., and J. McN. Jerdine, "Canadian Refining Practice in the Production of Uranium Trioxide by Solvent Extraction with Tributylphosphate," Proc 2nd UN International Conference on the Peaceful Uses of Atomic Energy, Geneva, Vol. 4, 3 (1958). Brown, K. 8., and C. F. Coleman, “Solvent Extraction in Ore Processing," Chapter in Progress in Nuclear Energy, Series III, Vol. 2, Pergamon Press, New York, N.Y., 3 (1958). Ellis, 0. A., ”Uranium Recovery by Solvent Extraction," . Uranium Ore Processing, Addison Wesley Publishing Company, Reading, Mass., Chapter 10, 237 (1958). EPA, "Solvent Extraction Status Report," EPA R2-72-073, Ruther F. Mayhue National Experimental Research Center (1972). Ashbrook, A. W., "Spectrophotometric Determination of Small Amounts of Reagents Used in Solvent Extraction Processes," Analytica Chimica Acta., 66, 123 (1972). Blake, C. A., and C. F. Baes, and K. 8. Brown, "Solvent Extraction with Alkyl Phosphoric Compounds," Ind. Eng. Chem., 66, 1756 (1958). Hurst, F. J., and D. J. Crouse, and K. 8. Brown, "Recovery of Uranium from Wet Process Phosphoric Acid," Ind. Eng. Chem., 11, 127 (1972). Hazen, W. C., "Solvent Extraction Techniques," Solvent Extrac- tion Research Bulletin Number T4—832, Denver Equipment Division, Denver, Colorado. "Immmm‘5