<%BANNER%>

Novel Technique for the Separation of Dolomite from Phosphate Rock

Permanent Link: http://ufdc.ufl.edu/UFE0041349/00001

Material Information

Title: Novel Technique for the Separation of Dolomite from Phosphate Rock
Physical Description: 1 online resource (43 p.)
Language: english
Creator: Antony, Abbin
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: dolomite, flotation, phosphate, reactive, sluice
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphate is a major ingredient in the manufacture of fertilizers. United States is the second largest phosphate producer and 85% of the total domestic output is from Florida and North Carolina. Depletion of phosphate resources is predicted in the next 100 years. Dolomite is recognized as a major impurity in the ore. Even though several techniques have been suggested by researchers, they have been found to have low separation efficiency in terms of grade and recovery or high process operation cost. Elaborate research by Ayman A. El-Midany under the guidance of Professor Hassan El-Shall at University of Florida led to discovery and development of a technique to separate dolomite from phosphate employing the property that carbonate rocks reacts with acid and nucleates CO2 gas. Coating the particles with poly vinyl alcohol (PVA) enables retaining the CO2 gas generated and enables flotation of dolomite (carbonate) particles. Beaker tests revealed significant separation and recovery. The work presented here involves the various stages in the translation of the beaker process into a continuous process. It involves analyzing equipment designs proposed. Development of each design is carried out based on observations in the preceding design starting with the beaker test. Once a design is confirmed, various parameters influencing the performance are also investigated. Designs analyzed in order are : i. Column ii. Screw Pipe iii. Vibrating Sluice The feed preparation and analysis techniques employed are consistent with those used by Ayman A. El-Midany for his research work.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Abbin Antony.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Svoronos, Spyros.
Local: Co-adviser: El-Shall, Hassan E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041349:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041349/00001

Material Information

Title: Novel Technique for the Separation of Dolomite from Phosphate Rock
Physical Description: 1 online resource (43 p.)
Language: english
Creator: Antony, Abbin
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: dolomite, flotation, phosphate, reactive, sluice
Chemical Engineering -- Dissertations, Academic -- UF
Genre: Chemical Engineering thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Phosphate is a major ingredient in the manufacture of fertilizers. United States is the second largest phosphate producer and 85% of the total domestic output is from Florida and North Carolina. Depletion of phosphate resources is predicted in the next 100 years. Dolomite is recognized as a major impurity in the ore. Even though several techniques have been suggested by researchers, they have been found to have low separation efficiency in terms of grade and recovery or high process operation cost. Elaborate research by Ayman A. El-Midany under the guidance of Professor Hassan El-Shall at University of Florida led to discovery and development of a technique to separate dolomite from phosphate employing the property that carbonate rocks reacts with acid and nucleates CO2 gas. Coating the particles with poly vinyl alcohol (PVA) enables retaining the CO2 gas generated and enables flotation of dolomite (carbonate) particles. Beaker tests revealed significant separation and recovery. The work presented here involves the various stages in the translation of the beaker process into a continuous process. It involves analyzing equipment designs proposed. Development of each design is carried out based on observations in the preceding design starting with the beaker test. Once a design is confirmed, various parameters influencing the performance are also investigated. Designs analyzed in order are : i. Column ii. Screw Pipe iii. Vibrating Sluice The feed preparation and analysis techniques employed are consistent with those used by Ayman A. El-Midany for his research work.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Abbin Antony.
Thesis: Thesis (M.S.)--University of Florida, 2009.
Local: Adviser: Svoronos, Spyros.
Local: Co-adviser: El-Shall, Hassan E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041349:00001


This item has the following downloads:


Full Text

PAGE 1

1 NOVEL TECHNIQUE FOR THE SEPA RATION OF DOLOMITE FROM PHOSPHATE ROCK By ABBIN ANTONY A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Abbin Antony

PAGE 3

3 To all who contributed and guided me in the accomplishment of this research work

PAGE 4

4 ACKNOWLEDGMENTS To begin with I would like to recognize all the faculty members in the Department of Chem ical Engineering at University of Florida. I thank my chair Dr. Spyros A. Svoronos of the Department of Chemical Engineering and my co chair Dr. Hassan El Shall of the Material Science Engineering Department for giving me the opportunity to indulge i n this research work. I would like to recognize my team members Lokendra kumar Bengani, Chris Brave and Kate Ob rero for their significant contribution towards the completion of the research work I also express gratitude to Ayman A. El Midany, as this is a contin uation of his research work. I am grateful to the staff at Particle Engineering Research Center (PERC) for providing guidance and resources for the research work. I especially thank Gary Scheiffle and Gill Brubaker for their assistance and support. I woul d like to show appreciation to the Mosaic Company for funding this research work. Finally, I would like to thank my family for their love, encouragement and guidance

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF OBJECTS ................................ ................................ ................................ ......... 9 LIST OF ABBREVIATIONS ................................ ................................ ........................... 10 ABSTRACT ................................ ................................ ................................ ................... 11 CHAPTER 1 LITERATURE REVIEW ................................ ................................ .......................... 13 Introduction ................................ ................................ ................................ ............. 13 Separation Techniques ................................ ................................ ........................... 14 I.M.C.F Process ................................ ................................ ................................ 14 U.S.B.M Process ................................ ................................ .............................. 15 U.F Process ................................ ................................ ................................ ...... 17 U.A Process ................................ ................................ ................................ ..... 18 T.V.A Process ................................ ................................ ................................ .. 19 CLDRI Process ................................ ................................ ................................ 20 Reactive Flotation ................................ ................................ ................................ ... 22 2 MATERIALS AND METHODS ................................ ................................ ................ 23 Feed Preparation ................................ ................................ ................................ .... 23 Coating ................................ ................................ ................................ ................... 23 Reactive Flotation ................................ ................................ ................................ ... 24 Analysis ................................ ................................ ................................ .................. 24 3 EXPERIMENTAL RESULTS WITH BEAKER TESTS ................................ ............. 25 4 NEW EQUIPMENT DESIGNs AND EXPERIMENTAL RESULTS .......................... 27 Column ................................ ................................ ................................ ................... 27 Screw Pipe ................................ ................................ ................................ .............. 29 Vibrating Sluice ................................ ................................ ................................ ....... 31 Co Current ................................ ................................ ................................ ........ 31 Counter Current ................................ ................................ ................................ 32 5 CONCLUSIONS AND RECCOMENDATIONS FOR FUTURE WORK ................... 40

PAGE 6

6 LIST OF REFERENCES ................................ ................................ ............................... 41 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 43

PAGE 7

7 LIST OF TABLES Table page 2 1 Size Distribution and composition of Barrel Content ................................ ........... 23 3 1 R esults on beaker test for different particle size ................................ ................. 26 4 1 Results on column test for different flow rates ................................ .................... 29 4 2 Results on count er current vibrating sluice for different feeding techniques, flow techniques, coating techniques and amount of coating ............................... 37 4 3 Results on counter current vibrating sluice for different coating agents .............. 37 4 4 Results on counter current vibrating sluice for different concentrations of PVA .. 39

PAGE 8

8 LIST OF FIGURES Figure page 1 1 Flow sheet of IMCF Process ................................ ................................ ............... 15 1 2 Flow sheet of USBM Process ................................ ................................ ............. 16 1 3 Flow sheet of UF Process ................................ ................................ ................... 17 1 4 Flow sheet of UA Process ................................ ................................ .................. 18 1 5 Flow sheet of TVA Process ................................ ................................ ................ 19 1 6 Flow sheet of CLDRI Process ................................ ................................ ............. 21 4 1 Experimental setup of column for reactive flotation ................................ ............ 27 4 2 Experimental setu p of screw pipe for reactive flotation. ................................ ...... 30 4 3 Concept of co current flow ................................ ................................ .................. 31 4 4 Concept of counter current flow ................................ ................................ .......... 33 4 5 Experimental setup of vibrating sluice for reactive flotation. ............................... 34 4 6 Process stream inlet ................................ ................................ ........................... 35 4 7 Weir and trap ................................ ................................ ................................ ...... 35 4 8 Point feed and line feed techniques respectively. ................................ ............... 36 4 9 Tailings and concentrate collected during experimentation. ............................... 36

PAGE 9

9 LIST OF OBJECTS Object page 3 1 Video of Beaker test ................................ ................................ ........................... 25 4 1 Video of Counter current vibrating sluice ................................ ............................ 34

PAGE 10

10 LIST OF ABBREVIATION S C Continuous Flow LF Line Feed M Mixing P Pulsating Flow PF Point Feed PVA Poly Vinyl Alcohol RF Reactive Flotation SC Spray Coating

PAGE 11

11 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science NOV EL TECHNIQUE FOR THE SEPA RATION OF DOLOMITE F ROM PHOSPHATE ROCK By Abbin Antony De cember 2009 Chair: Spyros A. Svoronos Co chair: Hassan El Shall Major: Chemical Engineering Phosphate is a major ingredient in the manufacture of fertilizers. United States is the second largest phosphate producer and 85 % of the total domestic output is from Florida and North Carolina. Depletion of p hosphate resources is predicted in the next 100 years. Dolomite is recognized as a major impurity in the ore. Even though several techniques have been suggested by researchers, they have been found to have l ow separation efficiency in terms of grade and recovery or high process operation cost. Elaborate research by Ayman A. El Midany under the guidance of Professor Hassan El Shall at University of Florida led to discovery and development of a technique to sep arate dolomite from phosphate employing the property that carbonate rocks reacts with acid and nucleates CO 2 g as. Coating the particles with poly vinyl a lcohol (PVA) enables retaining the CO 2 gas gene rated and enables flotation of d olomite (carbonate) part icles. Beaker tests revealed significant separation and recovery. The work presented here involves the various stages in the translation of the beaker process into a continuous process It involves analyzing equipment designs

PAGE 12

12 proposed. Development of each design is carried out based on observation s in the pre ceding design starting with the beaker test. Once a design is confirmed, various parameters influencing the performance are also investigated. Designs analyzed in order are : i. Column ii. Screw Pipe iii. Vibratin g Sluice The feed preparation and analysis techniques em ployed are consistent with those used by Ayman A. El Midany for his research work.

PAGE 13

13 CHAPTER 1 LITERATURE REVIEW Introduction Phosphate is an inorganic chemical and a salt of phosphoric acid. Elemental phosphorous was discovered accidently in the pursuit to make gold by a German chemist in 1669. Phosphate even though has many applications, as in treatment of potable water and cleaning solutions, it holds greater recognition in the agricultural industry because of it role in fertilizers and animal feed supplements. About 90% of the phosphate rock output in the world is used for fertilizer production. The United States is recognized as the second largest phosphate producer. Florida and North Carolina accou nt for more than 85 percent of the total domestic output. The phosphate industry in Florida is suffering a depletion of phosphate rich ore. Based on consumption rate in 2007 the supply of phosphate ore was estimated to run out in 345 years worldwide. Howev er scientists claim that a peak in consumption would occur in 30 years and the reserves will be depleted in 50 to 100 years. Based on the present mining rate reserves in Central Florida will last for the next 10 years and mining would move further south a nd southeast where the matrix is leaner in grade and higher in dolomite. Dolomite (Ca, Mg)CO 3 is identified as a highly taxing impurity in the ore. The proportion of carbonate present is directly proportional to the consumption of sulfuric acid in fertiliz er manufacture. Furthermore MgO forms a gel (increases viscosity) and reduces filtration capacity. Hence MgO content of dolomite is an important index in evaluating the quality of phosphate concentrate. Acidulation of phosphate rock requires feed of MgO co ntent less than 1% (by mass).

PAGE 14

14 Phosphate rock is observed to exist as alternate layers of high and low grade. Unavailability of economically feasible dolomite separation techniques has led to mining techniques that bypass high dolomitic zone s This results in wastage of 50% of the phosphate resource and furthermore discarding dolomite pebbles leads to wastage of about 13% Separation Techniques The phosphate industry is interested in a process that is efficient (high removal and recovery) ensures economica l feasibility (low operation cost) and is environmentally sound (low toxicity). In the recent decade s everal separation techniques have been proposed by researchers for the separation of dolomite from phosphate rock [El Shall et al., 2004]. I t is worthy t o mention in advance that all of the enlisted processes achieved removal by significantly sacrificing recovery I.M.C.F P rocess T he I.M.C.F p rocess [Snow, 1979; 1982; Lawver and Snow, 1980] (Figure 1 1), is a cationic process for silica free ores. H igh Mg O content pebbles are initially cycled through rod mills and s creens (24 35 Mesh) (0.707 0.420 mm). The screened particles are de slimed (150 Mesh) (0.10 5 mm) followed by flotation of s ilica from the stream using an amine condensate. The dolomite separa tion feed is dewatered, conditioned (20 30 seconds) at a slightly acid pH with tallow amine acetate and diesel fuel and is then subjected to rougher, cleaner and re cleaner flotation of dolomite from phosphate. Resulting p hosphate concentrate contained ~ 3 1% P 2 O 5 < 1% MgO and ~ 3% i nsoluble s Overall P 2 O 5 recovery of the process is estimated to be 55 60%.

PAGE 15

15 Figure 1 1 Flow sheet of IMCF Process U.S.B.M Process In the U.S. B.M (United States Bureau of Mines) p rocess [ Davis et al., 1984] (Figure 1 2), h i gh MgO pebbles are cycled through hammer mills and s creens (28 Mesh) (0.595 mm). The screened particles are de slimed (150 Mesh) (0.105 mm) and scrubbed with NaOH (1 lb/ton) for 20 minutes and the stream is again de slimed (150

PAGE 16

16 Mesh) (0.105 mm). This is fo llowed by conditioning of the stream for 5 minutes with NaOH, oleic acid and fuel oil at pH 9 9.2. The stream is subjected to rougher and cleaner flotation to float silica and dolomite from phosphate. Resulting p hosphate concentrate contained ~ 29% P 2 O 5 < 1.66% MgO and 5.2% i nsolubles. Overall recovery of the proce ss is estimated to be around 53 %. Figure 1 2. Flow sheet of USBM Process

PAGE 17

17 U.F Process This is a two stage condition process [Moudgil, 1988] (Figure 1 3) where h igh MgO pebbles are cycled thr ough rod mills and s creens (65 Mesh) (0.210 mm). The screened particles are de slimed (150 Mesh) (0.105 mm) and conditioned (2.5 minutes) at a pH 10. The concentrate is reconditioned ( 30 seconds) with sulfuric acid (maintains pH 4 5). This is followed by f lotation to remove dolomite. Final processing involves cationic flotation of the fine silica from the dewatered dolomite flotation cell. Resulting p hosphate concentrate contained ~ 28% P 2 O 5 ~ 1% MgO and 2.9% i nsolubles. Overall recovery of the process is estimated to be around 35.7%. Figure 1 3. Flow sheet of UF Process

PAGE 18

18 U.A Process In the U.A Process [Hanna and Anazia 1990] (Figure 1 4), h igh MgO pebbles are cycled through rod mills and s creens (35 Mesh) (0.420 mm). The particles are de slimed (150 Mes h) (0.105 mm), scrubbed for 10 minutes and de slimed again (325 Mesh) (0.044 mm). The stream is now subjected to con ditioning at a pH 5.5 6 with f atty acid, sulfuric acid, pine o il and NaOH followed by flotation. The underflow is further conditioned with sodium silicate and fatty acid and exposed to flotation. The results were observed to be inconsistent and were reported [El Shall et al 1994] Figure 1 4. Flow sheet of UA Process

PAGE 19

19 T.V.A Process In the T.V.A process [Hsieh and Lehr, 1985] (Figure 1 5), h igh MgO pebbles are cycled through ball mills and s creens (48 Mesh) (0.297 mm). The particles are de slimed at 400 Mesh ( 0.037 mm). The stream is scrubbed for 5 minutes and de slimed again at 400 Mesh. This is followed by flotation with di phosphonic aci d, oleic acid and p ine oil. The concentrate is exposed to silica flotation with an appropriate amine collector. Resulting p hosphate concentrate contained ~ 30% P 2 O 5 ~ 1.1% MgO and 3 .2% i nsolubles. Overall recovery of the process is estimated to be around 64%. Figure 1 5. Flow sheet of TVA Process

PAGE 20

20 CLDRI Process China is recognized as the l eading producer of phosphate. 90 % of the phosphate reserve s in China are observed to have high dolomitic content The CLDRI Process (Figure 1 6) was developed at China Lianyungang Design and Research Institute and it ensured MgO conte nt less than 0.1 % with P 2 O 5 loss of 6%. The process was comprised of two steps. The first step involved grinding of the feed particles for liberation and the second step i nvolved de slimin g which in turn improved flotation recovery and lower ed production cost These flotation technologies have successful industrial applications [El Shall et al., 1996]. The China Lianyungang Design and Research Institute had conducted extensive studies on Fl orida high dolomitic pebble with the objective of using the fine particle flotation technology to recover phosphate. P ilot scale testing was carried out jointly by the Research Institute Jacobs engin eering group and IMC phosphate company at engineering (based on pilot plant test results) showed that fine particle flotation dolomitic phosphate pe bbles [Geo et al., 2002] In the C fine flotation process, by which the percentage of MgO could be reduced to less than 1 %, concerns were on handling of the fine concentrate s (dewatering and transport), capital cost s because of grinding and multiple proc ess steps and operating cost s as the reagent consumption is directly proportional to the MgO content in the phosphate rock

PAGE 21

21 Figure 1 6. Flow sheet of CLDRI Process

PAGE 22

22 Reactive Flotation l ed involves combining two u nit operations i.e. reaction and flotation into a single unit Basic C oncept : It is well known that carbonate minerals (d olomit e) react with acid and nucleate CO 2 gas at the particle solution interface. 2 Entraining the liberated CO 2 at the surface via a suitable coating agent renders the particle more buoyant providing selective separation i .e. dolomite (carbonate) from phosphate rock (non carbonate). Extensive research by [El Midany et al, 2004] identified poly vinyl a lcohol (C 2 H 4 O) x as an effective coating agent that can retain CO 2 gas. This was because of the impermeability of PVA to CO 2 gas, permeability to sulfuric acid and flexibility of PVA coating. Experimental results reveal 3 % PVA solution (coating agent) and 3% H 2 SO 4 as optimum parameters. Beaker tests in this research confirmed flotation of particles as large as 10mm and p hosphate concentrates of less than 1% MgO with relatively high recovery were obtained

PAGE 23

23 C HAPTER 2 MATERIALS AND METHOD S Feed Prepa ration Phosphate rock received from Mosaic was sampled from the barrels in a manner that ensure s uniform distribution. It involved splitting the content into four groups followed by mixing two of the diagonal groups which was split again into 4 groups and the procedure was followed till manageable amount of rock sample is available for grinding. Direct a nalysis (technique explained later) of a sample revealed the following size distribution and composition Table 2 1. Size Distribution and composition of Barrel Content Properties +9.51mm 9.51+4.76mm 4.76+2mm 2+1mm 1mm Size Distribution,% 18.10 36.50 31.80 11.20 02.37 P 2 O 5, % 20.70 20.50 24.90 24.80 19.40 MgO, % 04.20 03.94 01.06 00.81 10.30 Rocks of size +4.76 mm in the sample were observed to have high MgO content. Hence feed preparation for all experiments involve d crushing of rocks of size +4.76mm from the sampl e. The sizes to which the rocks we re crushed are 9.51+4.76 mm, 4 .76+2mm, 2+1mm and 1mm. Crushing wa s carried out in a disc m ill. Crushed particles were washed to remove fines (de slimming) Coating Cr ushed particles we re coated with 3% Poly Vin y l Alcohol (PVA) [Optimum value determined through research ( El Midany et al )] Using >3% PVA in experiments wa s observed to be uneconomical as it further reduced the rate of diffusion of 3 % sulfuric acid, whereas <3% PVA revealed significant decrease in flotation due to rupture of PVA coating. For the preparation of 3% PVA solution, a calculated amount of PVA powd er was heated in distilled water for the duration of one hour which was followed by

PAGE 24

24 standardiz ation in flasks. Coating wa s carried out either by mixing or spraying. Experiments carried out reveal ed s pray coating technique more preferable as it i ncrease d co verage and ensured uniform coating Reactive Flotation The coated particles were introduced to 3% H 2 SO 4 in the process equipment (beaker / column / screw pipe / co current vibrating sluice / counter current vibrating s luice) Particles that float are refer red to as t ailings ( d olomite rich ) and the non floatin g particles are referred to as concentrates (p hosphate rich). The collected particles we re washed, dried, weighed and analyzed. Analysis Analysis involves determining percent composition of P 2 O 5 Mg O an d acid i nsolubles in the feed, tailing s and concentrate The dried samples we re crushed to fine powder and riffled t o weights less than 1 gm. These we re then dissolved in aqua regia. The resulting solution wa s filtered, standardized and forwarded for spect roscopy analysis. The filter paper containing insoluble material was dried, charred and weighed for calculating acid i nsolubles.

PAGE 25

25 CHAPTER 3 EXPERIMENTAL RESULTS WITH BEAKER TEST S Simple beaker tests have been used as reference Feed particles coated with P VA were immersed in a beaker containing 3% H 2 SO 4 solution. Amount of p articles employed were such that they form a monolayer at the bottom of the beaker. As the reaction proceeded particles we re ob served to rise to the surface. These particles were identi fied as tailings ( dolomite rich ) and continuously collected The duration of the experiment wa s roughly 30 minutes. P ar ticles remaining at the bottom we re the dolomite lean concentrates The tailings and concentrate collected we re dried and analyzed. Durin g experimentation it wa s observed that tailings if not collected quickly tend to drop. However a s the reaction proc e e d ed, these particles tend ed to float again i.e. there wa s continuous floating and sinking of dolomite particles. Significant amount of dolo mite particl es at the bottom of the beaker we re observed to have bubbles attached to them and d id not readily float. These particles tend ed to float over a long er duration of time or when the monolayer wa s exposed to slight shocks These particles were ide ntified as heavy tailings in later equipment designs and contribute significantly towards MgO removal Beaker tests were carried out for particles of size ranges 9.51mm, 4.76+2mm, 2+1mm and 1mm coated with 3 % PVA in the proportion 3 lb PVA/ton f eed and introducing to 3% sulfuric acid The feed was prepared as mentioned earlier by crushing sampled particles from Mosaic barrel of size >4.76 mm in disc mills followed by de slimming. Object 3 1. Video of Beaker test

PAGE 26

26 Table 3 1 Results on beaker test for d ifferent particle size Particle Size Range (mm) 9.51 9.51 + 4.76 4.76 + 2 2 + 1 1 Feed Composition (%) P 2 O 5 19.20 20.60 18.70 19.70 19.90 MgO 0 4.12 0 3.92 0 4.42 0 3.89 0 3.83 Acid Insoluble 0 9.02 0 9.02 15.20 11.7 TBD Concentrate (%) 62.60 89. 70 83.60 69.10 72.00 P 2 O 5 21.70 20.80 19.60 23.70 18.00 MgO 0 1.30 0 1.91 0 1.72 0 0.52 0 2.76 Acid Insoluble 13.40 0 9.42 13.10 11.60 TBD Tailings (%) 32.90 0 3.47 0 8.97 25.90 15.35 P 2 O 5 10.10 0 1.27 0 1.37 0 6.95 24.80 MgO 0 8.66 13.60 15.20 11.00 0 6.59 Acid Insoluble 12.80 10.20 14.90 11.40 10.90 P 2 O 5 Recovery 70.70 90.60 87.50 83.20 65.20 MgO Removal 80.30 56.40 67.50 91.90 48.20 Results reveal ed low removal in large feed particles of size 9.51+2mm and low recovery for particles of size 1mm. I n large particles ( 9.51 +4.76 mm) dolomite maybe entrapped within the complex es of the phosphate rock This prevented sulfuric acid from interacting with dolomite to generate CO 2 gas Furthermore, low surface to mass ratio for large particles trans lated to low bubble generation and hence low removal Particles of size 1mm tended to have high surface to mass ratio and hence large amount of PVA was required for effective coating (uneconomical). F roth formation was observed during experimentation which res ult ed in entrapment of p hosphate par ticles in the froth and explained the low recovery. Particles of size 2+1mm display ed satisfactory removal and recovery and hence have been used as standard feed particle size in all the following designs. It is worthy to note that beaker tests on particles of size 4.76+2 mm yielded removal of just 67.5%.

PAGE 27

27 Process Stream Inlet Feed Point Tailings Overflow Wire Mesh (Column Base) CHAPTER 4 NEW EQUIPMENT DESIGN S AND EXPERIMENTAL RES ULTS Column Columns are the most commonly used equipment in industries for floatation processes. Hence a column wa s t he first among the design s investigated A bench scale column was installed and feed particles coated with PVA were introduced through a rock feeder at the top of the column and we re exposed to a continuous counter current stream of 3% H 2 SO 4 which was intr oduced at the foot of the column at a controlled flow rate employing a centrifugal pump as shown in the figure (Figure 3 1) Tailings overflow whereas the concentrate is collected at the column base Figure 4 1. Experimental setup o f colu mn for reactive flotation

PAGE 28

28 Two mod els of actual column base were tried out i. Fluidized bed ii. Wire Mesh Experimentation on the model employing the concept of a fluidized bed to generate the effective base revealed significant attrition between the packing materi al of the bed and the coated particles which disrupted the stability of the bubbles and resulted in almost no flotation. Hence experiment ation on the column was carried out with the installation of a wire m esh as the eff ective base of the column i.e. while the tailing s overflow ed from the top of the column, concentrate wa s collected on the wire mesh. The effective height of the column wa s recognized as the length from the top of the column (overflow point) to the wire m esh and not to the actual foot of the column (inlet for 3% H 2 SO 4 ) During experimentation with effective column length of ~75 cm it was observ ed that particles float ed to a height of around 15 20 cm from the column base, remained suspended for a few seconds and then tend ed to sink. A possible explanation could be that the pressure exerted by the liquid column (~ 55 cm) accompanied with the flow to overcome this pressure tends to rupture the bubbles generated on the particles at the column base. Reducing the effective length improved the s ituati on Experiments were carried out for an effective column length of ~15cm at different flow rates with coated particles of size 2+1mm prepared by crushing barrel sample of size >4.76 mm followed by de slimming. Note that appreciable losses were recorded in column test because of extensive dissolution by sulfuric acid.

PAGE 29

29 Table 4 1. Results on column test for different flow rates Flow Rate (LPM) 5 10 15 20 Feed Composition (%) P 2 O 5 22.60 22.60 22.60 22.60 MgO 0 3.01 0 3.01 0 3.01 0 3.01 Acid Insoluble 0 7.03 0 7.03 0 7.03 0 7.03 Concentrate (%) 88.60 86.90 84.20 81.30 P 2 O 5 21.30 20.10 22.80 23.40 MgO 0 1.93 0 1.92 0 2.17 0 2.03 Acid Insoluble 0 7.50 0 8.48 0 8.43 0 7.96 Tailings (%) 0 2.53 0 2.97 0 3.05 0 4.79 P 2 O 5 15.70 17.40 11.50 12.20 MgO 0 5.40 0 4.46 0 6.95 0 6. 65 Acid Insoluble 0 7.33 0 7.01 0 9.27 0 7.48 P 2 O 5 Recovery (%) 83.6 0 77.1 0 85.1 0 84.2 0 MgO Removal (%) 43.3 0 44.5 0 39.1 0 45.1 0 Higher flow rates tend to rupture the bubbles and the tailing s consisted of particles that have been carried solely by the flo w and not due to bubble formation. Generally unsatisfactory results made us abandon the column and we moved forward to analyze horizontal equipment designs. S crew Pipe To avoid influence of pressure of liquid column and high flo w rate to overcome the press ure, e quipment with horizontal flow was proposed. A screw pipe with internal threading and of length ~1.25 m was installed at an angle of ~ 5 o and wa s connected to a mechanism that sets the pipe rotating in a manner such that particles introduced at the lo wer end are carried to the upper end as shown in Figure 3 2 The process liquid wa s introduced at the upper end of the screw pipe and a circular weir installed at the lower end regulat ed the liquid level in the pipe. The feed of coated particles introduce d at the lower end under the influence of rotation and the presence of screw threading inside the pipe moves to the upper end and is exposed to a counter current stream of

PAGE 30

30 process liquid (3% H 2 SO 4 ). The process liquid which flows towards the lower end unde r the influence of gravity carries the tailings over the circular weir and the concentrate is obtained at the upper end. The actual distance travelled by the coated particles is >> 1.25 m because of the internal threading. As the coated feed particles trav el between the threading there wa s significant contact between the particle s and the pipe surface which le d to rupture of bubbles. Hardly any tailings were collected during the experimentation. Furthermore the pipe threading prove d a hindrance for hea vy ta ilings, a concept coined later in the next horizontal equipment design tested i.e. a vibrating sluice Figure 4 2. Experimental setup of screw pipe for reactive flotation. Process Liquids Tailings Concentrate Rotation Mechanism Circular Weir

PAGE 31

31 Vibrating Sluice Two concepts of flow pattern w ere inv estigated in the Vibrating Sluice i. Co Current Flow ii. Counter Current Flow Co Current Co current flow is that in which the channel vibrations that influenced the movement of the concentrate and the process stream (3% H 2 SO 4 ) flow that influence d the movement o f the tailings is in the same direction Figure 4 3. Concept of co current flow A channel protected with acrylic material (H 2 SO 4 resistant) wa s attached to a vibrating rock feeder. The channel wa s placed at an inclination of ~1 o and the process liq uid wa s introduced at the upper end of the channel through a flow distribu tor. PVA coated feed particles were introduced at the upper end th rough a rock feeder. A skimmer wa s placed at the lower end to collect the tailings. Vibration of the sluice in the d i rection of the liquid flow tended to carry the particles down the channel along with the process liquid. During the mo vement of the feed particles in the channel it wa s

PAGE 32

32 separated into tailings and concentrate which we re collected separately at the lower e nd with the assistance of a skimmer There wa s significant engineering difficulty to maintain the liquid level in the channel and hence in placing the skimmer. Appreciable amount of particles with small bubbles that do not float to the gas liquid interface and hence getting collected among concentrate we re observed. It is here that the significance of these particles were recognized The term tailings were to be sub categorized into light tailings and heavy tailings. Dolomite particles that clustered to larg e bubbles and readily float ed have been termed as light t ailings whereas those with small bubbles (signifies some presence of dolomite) were referred to as heavy t ailings. Among the particles that do not reach the gas liquid interface (concentrates and hea vy tailings) h eavy tailings we re observed to be significantly influenced by fluid flow in relative to channel vibrations, unlike concentrate which was more influenced by the channel vibrations Reconsidering the beaker test particles referred to as heavy t ailings can be identified as those particles that tend to float over a longer duration of time. Ideal performance of our equipment would involve the capture of these heavy t ailings This l ed to the concept of employing c ounter current flow in the channel Counter Current The concept of counter current flow involved employing channel vibration that influenced the movement of concentrate in the opposite direction with respect to the direction of process stream flow that influenced the tailings. Tailings as mentioned earlier have been categorized as light and heavy tailings. Light tailings were solely influenced by the liquid flow whereas the heavy tailings were influenced by both channel vibrations and process stream flow but with a greater inclination to th e latter.

PAGE 33

33 Figure 4 4. Concept of c ounter c urrent flow A channel protected with acrylic material (H 2 SO 4 resistant) is attached to the base of a vibrating rock feeder as shown in Figure 3 5 T he channel wa s placed at an inclination of ~ 3 o and t he process liquid was introduced as a combination of continuous ( employing a centrifugal pump) and pulsating ( employing a peristaltic pump) flow at the upper end of the channel (Figure 3 6 ) liquid level in the channel nor the movement of the concentrate about to leave the channel. Both the liquid flows (continuous and pulsating) we re against the vibration of the channel. Pulsating flow wa s introduced to damp the vibrations transmitted to the liquid su rfa ce by the vibrating channel. A weir with a small opening at the bottom was installed at the lower end to facilitate exposure of heavy tailings to significant flow r ate. A differential level trap wa s set up at the lower end before the we ir to entra p the hea vy tailings (Figure 3 7).

PAGE 34

34 Figure 4 5 Experimental setup of vibrating sluice for reactive flotation A) Process stream inlet. B) Weir and Trap. C) Feed Point. D) Tailings. E) Concentrate. Object 4 1. Video of Counter curre nt vibrating sluice A B C E D Process Liquid + Light Tailings + Heavy Tailings Concentrate

PAGE 35

35 Figure 4 6 Process stream inlet Figure 4 7 Weir and trap The feed particles coated with PVA we re pre treate d with the process liquid and we re introduced at the lower end at a d istance ~ 7cm in front of the trap Pre treatment of the feed particles wa s observed to en hance the separation and reduced build up of parti cles in the channel (Figure 3 8 ). It was o bserv ed the introduction of the feed as a line hindered the movement of th e heavy tailings and hence point feeding was adopted (Figure 3 8 ) The effective length of the vibrating sluice is the distance from the liquid distributor to the feed point (~50 cm). As the feed particles move d to the upper end, they were exposed to a co unter current process stream (3% H 2 SO 4 ). The stream carried the light tailings (Figure 3 9 ) Continuous Flow Pulsating Flow Opening ~1mm Heavy Tailings

PAGE 36

36 over the weir and the heavy tailings (Figure 3 7 ) into the trap. The concentrate moved to the upper e nd and was collected across the liquid distributor (Figure 3 9 ) Figure 4 8 Point f eed an d line f eed technique s respectively Figure 4 9 Tailings and concentrate collected during experimentation. E xperiments were carried out employing crushed particles of size 2+1mm and 4.76+2mm varying the feeding technique, flow technique, coating technique, coating agent and amount of coating employed LF Line Feed; PF Point Feed P Pulsating Flow; C Continuous Flow M Mixing; SC Spray Coating Feed Pre T reatment

PAGE 37

37 Table 4 2. R esult s on counter current vibrating sluice for different feed ing technique s flow technique s coating technique s and amount of coating Feed Particle Size 2+1mm 2+1mm 2+1mm 2+1mm Feed Technique LF LF PF PF Flow Type P P+C P+C P+C Coating Technique M M SC SC Coating Material (lb PVA / ton feed) PVA (3) PVA (3) PVA (3) PVA (6) Feed Composition (%) P 2 O 5 2 1.90 21.90 21.90 21.90 MgO 03.75 03.75 03.75 03.75 Acid Insoluble 11.30 11.30 11.30 11.30 Concentrate (%) 79.70 69.40 58.48 67.80 P 2 O 5 24.10 23.20 2 5.82 25.23 MgO 02.36 01.59 01.46 01.19 Acid Insoluble 10.00 12.39 11.69 11.13 Tailings (%) 13.30 22.00 33.64 24.71 P 2 O 5 13.90 11.60 16.09 12.06 MgO 07.08 07.73 06.13 08.22 Acid Insoluble 11.80 14.26 10.73 11.06 P 2 O 5 Recovery (%) 87.80 73.80 69. 10 78.20 MgO Removal (%) 49.80 70.50 77.00 78.40 Table 4 3. Results on c ounter current vibrating sluice for different coating agents Feed Particle Size 2+1mm 2+1mm 4.76+2mm Feed Technique PF PF PF Flow Type P+C P+C P+C Coating Technique NA M SC Coating Material (lb PVA / ton feed) NIL Fatty Acid PVA (3) Feed Composition (%) P 2 O 5 2 1.90 21.90 22.30 MgO 03.75 04.74 04.38 Acid Insoluble 11.30 10.13 10.13 Concentrate (%) 90.00 63.32 67.00 P 2 O 5 17.80 22.01 23.29 MgO 01.98 03.42 01.62 Acid I nsoluble 12.12 10.23 11.23 Tailings (%) 03.47 19.70 28.53 P 2 O 5 09.25 08.36 11.71 MgO 08.13 10.47 08.73 Acid Insoluble 06.56 06.32 09.70 P 2 O 5 Recovery (%) 73.15 63.63 70.27 MgO Removal (%) 51.20 54.38 75.23

PAGE 38

38 Experiments reveal point feed more effe ctive than line feed. Combination of pulsating and continuous flow wa s found to be more effective than pulsating flow alone. Pulsating flow alone tends to leave the tailings suspended and do not move with the flow. S pray coating of feed particles wa s found to be more effective than mixing as it ensures uniform and eff ective coating of all particles Increasing the amount of PVA used to 6 lb PVA/ton of feed brought about dramatic improvement in recovery and removal. This however is not economically feasible. In the experiment involving the use of no coating the feed wa s introduced at a distance of ~1cm from the trap. Tailings collected were mainly heavy tailings and this reveals the performance capability of this equipment Use of fatty acid seems to be coun ter productive The process stream is observed to strip the particle surface of fatty acid and resulted in an oily stream. Experiments on the equipment with particles of size 4.76+2mm revealed significa nt removal in comparison to beaker tests This result revealed the effectiveness of the equipment as it is capable of removing particles that tend to slightly float (Heavy tailings) The presence of significant amount of dolomite in heavy tailings is further confirmed by this result. Ability to use particles larger than 2 mm is highly advantageous from the industrial perspective as it reduces crushing cost. Experiment were not carried with particl es larger than 4.76 because of the engineering limitations in maintaining the desired liquid level in the channel for this model Parameters observed to influence the performance of the process are the angle of the channel, flow rate, pulse rate, feed rate, vibration rate of the channel, size of the opening at t he bottom of the weir (regulates the flow rate exposed to the heavy tailings),

PAGE 39

39 fluid level in the channel, distance of the feed point from the trap and the extent of pre treatment of the coated feed particles. Experiments were also tried for different concentrations of PVA coating employing crushed particles of size 2+1mm LF Line Feed; PF Point Feed P Pulsating Flow; C Continuous Flow M Mixing; SC Spray Coating Table 4 4 R esults on counter current vibrating sluice for different concentration s of PVA Feed Particle Size 2+1mm 2+1mm 2+1mm Feed Technique PF PF PF Flow Type P+C P+C P+C Coating Technique SC SC SC Coating Material (lb/ton feed) PVA (2) PVA (2) PVA (3) Concentration of Coating PVA (%) 1 2 2 Feed Composition (%) P 2 O 5 22.01 22.01 22.01 MgO 03.50 03.50 03.50 Acid Insoluble 10.13 10.13 10 .13 Concentrate (%) 77.96 76.08 74.50 P 2 O 5 22.11 21.73 21.60 MgO 02.45 02.20 02.03 Acid Insoluble 10.17 09.75 10.78 Tailings (%) 13.89 15.97 18.01 P 2 O 5 08.97 09.01 08.99 MgO 09.80 09.65 10.23 Acid Insoluble 07.01 07.95 08.75 P 2 O 5 Recovery (% ) 78.31 75.11 73.11 MgO Removal (%) 45.42 52.23 57.83 The results were u nimpressive because low concentration of PVA does not guarantee strong binding of bubbles on the particle surface. Furthermore rupture of coating takes place. The r esults also rev eal ed that coating of the particles prevents dissolution of phosphate by sulfuric acid

PAGE 40

40 C HAPTER 5 CONCLUSIONS AND RECC OMENDATION S FOR FUTURE WORK The equipment designing was carried out crudely. One of the major concerns is the amount of PVA used. Optimu m results were obtained for 6 lb PVA / ton feed. Reducing this would enhance the favorability of the process. The model employed is crude and has limitations. However impressive results were observed for particles with no PVA coating and particles of size 4.76+2 mm (relative to beaker tests) Design limitations hindered testing of particles larger than 4.76 mm. Coating and m ixing technique should be further investigated to bring about optimum and uniform coverage. Various parameters such as flow rate (cont inuous + pulsating), frequency of vibration, feed rate, channel length and l iquid level in channel are potential parameters influencing the performance of the equipment and should be investigated on a better pilot scale equipment that readily allows the m anipulation of these parameters.

PAGE 41

41 LIST OF REFERENCES Andreyeva, Anikeyeva, Lirova, Tager 1973. Investig ation of PVA solutions. Polymer Science USSR 15 (8) Becker P.,1989. Phosphates and phosphoric acid: Raw materials technology and economics of the wet process. Fertilizer Science and Technical Series, 2nd ed., 6, Marcel Dekker Inc. Davis, B.E., Liewellyn, T.O., Smith, C.W, 1984. Continuous beneficiation of dolo mitic phosphate rocks. United States Bureau of Mines RI 8903. El Midany, 2004, Seperating Do lomite from Phosphate rock by reactive flotation: Fundamentals and Application. PhD Thesis, Gainesville, University of Florida El Midany, H. El Shall, R. Stana and B. Moudgil, 2006. Selective Seperation of carbonate Minerals by reactive flotation. El M idany, H. El Shall, R. Stana and B. Moudgil, 2004. Mechanisms involved in reactive flotation of dolomite. El Shall, 1994, Evaluation of dolomite separation techniques. 02 094 108 El Shall, El Midany, R.Stana and B.Moudgil, 2006. One more solution to Do lomite/Apatite separation problem. El Shall, Bogan, 1990. Characterization of future Florida phosphate resources. 2 nd progress report, FIPR Project 89 02 082R El Shall, Zhang, Abdel Khalek, El Mofty, 2004. Beneficiation technolo gy of phosphates: Challen ges and Solutions. Minerals and Metallurgical Processing 21 (1), 17 26. Finch, C.A., 1973. Polyvinyl alcohol; properties and applications. John Wiley and Sons, New York. Geo, Zheng, Gu, Z., 2002. Review of beneficiation technology for Florida high dolomi te pebble. In: Beneficiation of Phosphates: Fundamentals and Technology. Gu, Z., Geo, Z., Hwang, C.L., 1999. Development a new technology for beneficiation of Florida dolomitic phosphate resources. FIPR 02 129 167

PAGE 42

42 Hanna J., Anazia I 1990. Selective f lotation of dolomitic limestone impurities from Florida phosphat es. University of Alabama, FIPR 002 066 089. Hsieh, S.S., Lehr, J.R., 1985. Beneficiation of dolomite Idaho phosphate rock by the TVA diphosphonic depressant process. Mineral and Metallurgica l Processing 2, 10 13 H. Sis, S. Chander, 2003. Reagents used in the flotation of phosphate ores: a critical review Minerals Eng. 16 (2003) 577 585. Kiselev, A.V., Rygin, V.I., 1972. Infrared spectra of surface compounds. John Wiley, N ew York Moudgil B.M., 1988. Separation of dolomite from the south Florida phosphate rock. University of Florida, FIPR 02 023 066. Rule, A.R., 1982. Application of carbonate silica flotation techniques to Western phosphate materials United Sta tes Bureau of Mines RI 87 28. Snow, R.E., 1979. Beneficiation of phosphate ore. U.S.Patent 4,144,969. Snow, R.E., 1982. Flotation of phosp hate ore containing dolomite. U.S.Patent 4,364,824. Yu, Q., Ye, Y., Miller, J.D., 1990. A study of surfactant/oil emulsions for fine coal flo tation. pp. 345 355 Zhang, P., Yu, Y., Bogan, M., 1997. Challenging the Crago double float process. Minerals Eng. 10 (9), 983 994. Zheng, X., Smith, R.W., 1997. Dolomite depressants in the flotation of apatite and cellophane from dolomite. Minerals Eng. 10 (5), 537 545.

PAGE 43

43 BIOGRAPHICAL SKETCH Abbin Antony was born in Hofuf, Saudi Arabia in November 1985. He was enrolled at TKM College of Engineering affiliated to University of Kerala in India from 2003 to 2007. ent In Plant Training at Hindustan Organic Chemical Ltd. and also worked on a project involving optimization of a production unit at Kochi Refineries Bharat Petroleum Corp Ltd. in Kerala, India. He n 2007 and enrolled for A. Svoronos and Professor Hassan El Shall on translating a batch technique for separation of dolomite from phosphate developed by a student of Profes sor Hassan El Shall into a continuous process.