<%BANNER%>

Modification and Characterization of Montmorillonite Clay for the Extraction of Zearalenone

Permanent Link: http://ufdc.ufl.edu/UFE0024977/00001

Material Information

Title: Modification and Characterization of Montmorillonite Clay for the Extraction of Zearalenone
Physical Description: 1 online resource (136 p.)
Language: english
Creator: Hue, Kerri-Ann
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: cation, design, ftir, montmorillonite, surfactant, xrd, zearalenone
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Mycotoxins are secondary metabolites of organisms belonging to the fungus kingdom. The cost associated with mycotoxin contamination in the USA and Canada is approximately US $5 billion. Zearalenone (ZEN), a resorcylic acid lactone, is produced by various members of the genus Fusarium. These fungi often colonize a variety of foods and feedstuffs including, corn, sorghum, wheat, oats, barley, and other cereal grains. This metabolite has estrogenic effects in farm animals with pigs being the most sensitive. ZEN induces hyperestrogenism and can cause infertility, reduced sex drive, fetal mummification, and abortions. Clays have successfully been used in the animal feed industry as an adsorbent and binders for certain small, water soluble mycotoxins. These mycotoxins are attracted to the electrical imbalance between the layers of the clays caused by isomorphic substitution of structural atoms. The mycotoxins are sequestered in the clay layers and pass harmlessly through the animal. However, ZEN is water insoluble and is not extracted easily with aluminosilicate clays. Therefore the modification of hydrated sodium calcium aluminosilicate (HSCAS) clays with organic cations has been proposed to render the clays hydrophobic and increase the ZEN binding capacity. The goal of this study was to develop a safe and cost effective organophilic material able to bind and extract zearalenone, to investigate the factors most important to extraction, and to investigate the fundamental properties between the clay-surfactant-mycotoxin systems that lead to extraction. The clay was modified by cation exchange reactions with tricaprylmethylammonium (TCMA) chloride and generic corn oil. The organophilic clays were then characterized using XRD, FTIR, and TGA analytical techniques. These techniques were used to determine the change in fundamental clay properties that would lead to the extraction of ZEN. Desorption studies were performed to determine any increase in toxicity that might be caused by washing of the clays or exposure to electrolytic solutions. Statistical design of experiments was used to determine the factors most influential during ZEN extraction. Modification by TCMA resulted in an increase in intergallery spacing of ~0.6nm. TGA and FTIR studies indicated intercalation of organic species within the clay layers. An increase in weight loss proportional to the amount of TCMA added was observed by TGA analysis. In addition to the peaks found in the natural clay, peaks at 2928 cm-1, 2852 cm-1, and 1466 cm-1, which belong to C-H asymmetric stretching, C-H symmetric stretching and CH2 scissoring respectively characteristic of TCMA were present. The clays developed were able to extract > 90% ZEN in vitro at pH 3 and pH7. The factors most important for extraction changed depending on the levels of parameters chosen. Mathematical models were developed that showed the relationship between the factors and the ZEN removal percentage. When exposed to electrolyte solutions ~1.5pmm of surfactant desorbed from the modified clay.
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 Kerri-Ann Hue.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: El-Shall, Hassan E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024977:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024977/00001

Material Information

Title: Modification and Characterization of Montmorillonite Clay for the Extraction of Zearalenone
Physical Description: 1 online resource (136 p.)
Language: english
Creator: Hue, Kerri-Ann
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: cation, design, ftir, montmorillonite, surfactant, xrd, zearalenone
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Mycotoxins are secondary metabolites of organisms belonging to the fungus kingdom. The cost associated with mycotoxin contamination in the USA and Canada is approximately US $5 billion. Zearalenone (ZEN), a resorcylic acid lactone, is produced by various members of the genus Fusarium. These fungi often colonize a variety of foods and feedstuffs including, corn, sorghum, wheat, oats, barley, and other cereal grains. This metabolite has estrogenic effects in farm animals with pigs being the most sensitive. ZEN induces hyperestrogenism and can cause infertility, reduced sex drive, fetal mummification, and abortions. Clays have successfully been used in the animal feed industry as an adsorbent and binders for certain small, water soluble mycotoxins. These mycotoxins are attracted to the electrical imbalance between the layers of the clays caused by isomorphic substitution of structural atoms. The mycotoxins are sequestered in the clay layers and pass harmlessly through the animal. However, ZEN is water insoluble and is not extracted easily with aluminosilicate clays. Therefore the modification of hydrated sodium calcium aluminosilicate (HSCAS) clays with organic cations has been proposed to render the clays hydrophobic and increase the ZEN binding capacity. The goal of this study was to develop a safe and cost effective organophilic material able to bind and extract zearalenone, to investigate the factors most important to extraction, and to investigate the fundamental properties between the clay-surfactant-mycotoxin systems that lead to extraction. The clay was modified by cation exchange reactions with tricaprylmethylammonium (TCMA) chloride and generic corn oil. The organophilic clays were then characterized using XRD, FTIR, and TGA analytical techniques. These techniques were used to determine the change in fundamental clay properties that would lead to the extraction of ZEN. Desorption studies were performed to determine any increase in toxicity that might be caused by washing of the clays or exposure to electrolytic solutions. Statistical design of experiments was used to determine the factors most influential during ZEN extraction. Modification by TCMA resulted in an increase in intergallery spacing of ~0.6nm. TGA and FTIR studies indicated intercalation of organic species within the clay layers. An increase in weight loss proportional to the amount of TCMA added was observed by TGA analysis. In addition to the peaks found in the natural clay, peaks at 2928 cm-1, 2852 cm-1, and 1466 cm-1, which belong to C-H asymmetric stretching, C-H symmetric stretching and CH2 scissoring respectively characteristic of TCMA were present. The clays developed were able to extract > 90% ZEN in vitro at pH 3 and pH7. The factors most important for extraction changed depending on the levels of parameters chosen. Mathematical models were developed that showed the relationship between the factors and the ZEN removal percentage. When exposed to electrolyte solutions ~1.5pmm of surfactant desorbed from the modified clay.
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 Kerri-Ann Hue.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: El-Shall, Hassan E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024977:00001


This item has the following downloads:


Full Text

PAGE 1

1 MODIFICATION AND CHARACTERIZATION OF MONTMORILLONITE CLAY FOR THE EXTRACTION OF ZEARALENONE By KERRI -ANN ALICIA HUE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF TH E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Kerri -Ann Alicia Hue

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS I would not have been able to accomplish this task without the love, support, and prayers of many people. I thank God for His Mercy and Grace. I thank my parents Gertina Guscott, Raymond Hue, Lily Hue, and Desmond McKenzie for their never ending love and support. Words can never express how much they have meant to me throughout this journey. I am blessed to have them in my life. They have shown me that with love, hard work, and perseverance anything is possible. I thank my grandmothers Olive Hue and Daisy Guscott. I hope that one day I can live up to the legacy they have left for their children and grandchildren. My other family members, brother, sisters, cousins, aunts, uncles, nieces and nephews have all contributed in some way to who I am today. I also thank my very close friends Dr. Breagin Riley, Sha Brooks, Dr. C armen Gaines, and Jnai Pittman who have provided emotional support throughout difficult times. I especially thank Dr. Hassan El -Shall. His guidance has been invaluable to me. He has taught me more about the scientific process than he might ever know. I ow e the success of this work and any future work to him. I thank Dr. Victor Jackson for his efforts in managing this project between PERC and Milwhite Inc. Special thanks to Dr. Derendorf, Dr. Rajendra Pratrap, and Dr. Sabarinath for the training and use of equipment without which I could not complete my research. Id like to thank Gary Schieffle and Gill Brubaker for their training and helpful discussions. I also thank my doctoral committee members, Dr. B. Koopman, Dr. R. Singh, Dr. E. Whitney, and Dr. C. Ba tich for their much appreciated comments and insight. Financial support from the University of Florida, the Particle Engineering Research Center, SEAGEP, and Milwhite Inc., was crucial for the completion of this work. I would also like to thank my group m embers, Dr. S. Tedeschi, Dr. R. Hamey, and Dr. S. Daosukho, who helped me navigate graduate school and PERC. Special thanks to Amit Singh who has been very helpful to

PAGE 5

5 me during this dissertation writing process. Id also like to thank some of the PERC students and postdocs: P. Carpinone, Dr. N. Stevens, and Dr. G Pyrgiotakis for being excellent sounding boards for ideas and for their willingness to help at a moments notice.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 8 LIST OF FIGURES ............................................................................................................................ 10 LIST OF ABBREVIATIONS ............................................................................................................ 14 CH A P T E R 1 INTRODUCTION ....................................................................................................................... 17 2 BACKGROUND AND LITERATURE SURVEY .................................................................. 22 2.1 Mycotoxin Contami nation Worldwide ................................................................................ 22 2.2 The Impact of Zearalenone Contamination ......................................................................... 23 2.3 Zearalenone Decontamination Research ............................................................................. 26 2.4 Aluminosilicate Clays for Zearalenone Adsorption ........................................................... 27 2.4.1 Clay Structure ............................................................................................................. 27 2.4.2 Organophilic Clays ..................................................................................................... 29 2.4.3 Organophilic Clays for the Extraction of Zearalenone ............................................ 32 2.5 Hypothesis for Clay Modification and Zearalenone Extraction ........................................ 37 3 EXPERIMENTAL AND CHARACTERIZATION METHODOLOGY ............................... 41 3.1 Clay Characterization ........................................................................................................... 41 3.2 IMTX Modification with TCMA ......................................................................................... 44 3.3 Characterization .................................................................................................................... 44 3.3.1 X Ray Diffraction ....................................................................................................... 44 3.3.2 Fourier Transform Infra red Spectroscopy (FTIR) .................................................... 44 3.3.3 Thermal Analysis ....................................................................................................... 45 3.3.4 UV -Vis and HPLC -FD for determining ZEN Extraction ........................................ 45 3.4 Zearalenone Extraction ......................................................................................................... 47 3.5 TCMA Adsorption/Desorpt ion Studies ............................................................................... 49 3.5.1 TCMA Adsorption Studies ........................................................................................ 49 3.5.2 TCMA Washing Cycle Desorption ........................................................................... 50 3.5.3 Electrolyte Desorption Studies .................................................................................. 50 4 RESULTS AND DISCUSSION ................................................................................................ 52 4.1 Characterization of IMTX .................................................................................................... 52 4.2 Clay Modification ................................................................................................................. 54 4.2.1 FTIR Results ............................................................................................................... 54 4.2.2. XRD Results .............................................................................................................. 55

PAGE 7

7 4.2.3. TGA Results .............................................................................................................. 55 4.2.4. Possible TCMA Conformation ................................................................................. 56 4.3 ZEN Extraction by TCMA IMTX ....................................................................................... 57 4.3.1 Extraction Results by UV -Vis Spectroscopy............................................................ 59 4.3 .2. IMTX TCMA -corn oil Extraction ........................................................................... 69 4.3.3. Extraction Results by HPLC FD .............................................................................. 81 4.4 TCMA Adsorption/Desorption Results ............................................................................... 87 4.4.1 TCMA Adso rption Isotherm .................................................................................... 102 4.4.2 Desorption by Washing............................................................................................ 104 4.4.3 Desorption by Electrolyte Solutions ....................................................................... 107 5 SUMMARY, CONCLUSIONS, AND FUTURE RECOMMENDATIONS ........................ 118 5.1 Summary of Results ............................................................................................................ 118 5.2 Conclusions ......................................................................................................................... 122 5.3 Recommendations for Future Work .................................................................................. 122 APPENDIX A ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RESPONSE AT PH 3 AS MEASURED UV-VIS SPECTROSCOPY ............................................................................. 123 B ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RESPONSE AT PH 7 AS MEASURED UV-VIS SPECTROSCOPY ............................................................................. 125 C ANA LYSIS OF VARIANCE FOR ZEN REMOVAL % RESPONSE AT PH 3 AS MEASURED BY HPLC -FD .................................................................................................... 127 D ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RESPONSE AT PH 7 AS MEASURED BY HPLC -FD .................................................................................................... 129 LIST OF REFERENCES ................................................................................................................. 131 BIOGRAPHICAL SKETCH ........................................................................................................... 136

PAGE 8

8 LIST OF TABLES Table page 2 1 Natural Occurrence of Zearalenone in Mixed Feeds (G. A. Bennett & Shotwell, 1979) ....................................................................................................................................... 24 2 2 Structure and Critical Micelle Concentration of exchanged surfactants (Lemke et al., 1998). ...................................................................................................................................... 35 3 1 23 factorial design used to determi ne the most significant factors for ZEN extraction. .... 48 4 1 FTIR Band Assignments of TCMA ...................................................................................... 59 4 2 FTIR Band assignments of IMTX an d Modified IMTX ..................................................... 59 4 3 Intergallery spacing of IMTX modified with varying amounts of TCMA. ........................ 59 4 4 23 factorial design matrix. ...................................................................................................... 65 4 5 ZEN experimental runs and extraction results by UV -Vis Spectroscopy. ......................... 70 4 6 Intergallery spacing of IMTX TCMA -corn oil modi fied clays. ......................................... 81 4 7 IMTX TCMA -corn oil experimental runs and extraction results by UV -Vis Spectroscopy........................................................................................................................... 83 4 7 ZEN experimental ru ns and extraction results by HPLC -FD .............................................. 88 4 8 Desorption results for TCMA IMTX modified clays exposed to electrolyte solutions at pH 3 and pH 7. ................................................................................................................. 117 A 1 Analysis of variance table [Partial sum of squares Type III] .......................................... 123 A 2 Statistical Results ................................................................................................................. 123 A 3 Diagno stic case statistics ..................................................................................................... 124 B1 Analysis of variance table [Partial sum of squares Type III] .......................................... 125 B2 Statistical Results ................................................................................................................. 125 B3 Diagnostic case statistics ..................................................................................................... 126 C2 Statistical Results ................................................................................................................. 127 C3 D iagnostic case statistics ..................................................................................................... 128

PAGE 9

9 D 2 Statistical Results ................................................................................................................. 129 D 3 Diagnostic case statistics ..................................................................................................... 130

PAGE 10

10 LIST OF FIGURES Figure page 1 1 2D model of zearalenone molecule (A. Dakovic et al. 2007). ........................................... 17 1 2 2D model of Aflatoxin B1 molecule. .................................................................................... 19 2 1 Layered structure of montmorillonite clays. ......................................................................... 28 2 2 Orientations of alkylammonium ions in the galleries o f layered silicates: monolayer, bilayers, pseudotrimolecular layers, and paraffin type arrangements of alkylammonium ions with different tilting angles of the alkyl chains ( Handbook of clay science, 2006). ................................................................................................................ 31 2 3 Variation of basal spacings of alkylammonium montmorillonites with chain length (nc) due to formation of monolayers, bilayers and pseudotrimolecular layers of alkylammonium ions in the interlayer spaces (de Paiva et al., 2008). ................................ 31 2 4 Probable arrangements of the organic cations in the clay interlayers of a Wyoming montmorillonite clay (SWy) modified with: a) trimethylphenylammonium (TMPA) and trimethylaamonium adamantine ( Adam) b) HDTMA (Sharmasarkar et al., 2000). ... 34 2 5 Adsorption isotherms for ZEN on clays exchanged with a variety of organophilic cations. .................................................................................................................................... 36 2 6 Structure of tricaprylmethylammonium (TCMA) chloride surfactant. ............................... 39 3 1 XRD scan for natural IMTX with an angular range of 2o o. .................................... 42 3 2 Molecular structure of crystal violet. .................................................................................... 43 3 3 Zearalenone calibration curve. .............................................................................................. 46 3 4 Schematic of the procedure for ZEN extraction. .................................................................. 49 4 1 XRD scan of IMTX clay with angular range 2o o. ................................................... 53 4 2 XRD scan of IMTX with angular range of 2o o. ....................................................... 54 4 3 Desorption of inorganic cations from IMTX after 1, 3, a nd 13 days of incubation. ......... 58 4 4 CV adsorption by IMTX after 1, 3, and 13 days incubation. .............................................. 58 4 6 FTIR analysis of IMTX and IMTX modified with TCMA ................................................. 61 4 7 XRD scans of IMTX and TCMA -IMTX clays with angular range 2o o. ................ 62 4 7 TGA curves of IMTX modified with varying amounts of TCMA. .................................... 63

PAGE 11

11 4 8 Pseudotrimolecular conformation of alkylammoniu m ions (de Paiva et al., 2008). .......... 64 4 10 ZEN calibration curves used to determine ZEN extraction. ................................................ 70 4 11 Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 3 ............................................................................................................................................... 71 4 12 Model graphs showing contour plot of ZEN removal % at pH 7as a function of factors A and B with factor C (ZEN dosage) held constant at 14ppm and interaction plot of the response data against factor A (%CEC replaced) for both levels of factor B (amount of clay added) ...................................................................................................... 72 4 14 Model graphs showing contour plot of ZEN removal % at pH 3as a function of factors A and C with factor B (amount of clay added) held constant at 87.5 mg and interaction plot of the response data against factor A (%CEC replac ed) for both levels of factor C (ZEN dosage) ............................................................................................ 74 4 14 Interaction plots showing the ZEN removal % against the % CEC replaced at high and low ZEN dosages when the amount of clay added is 50mg and 125mg. .................... 75 4 15 Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 7 ............................................................................................................................................... 76 4 16 Model graphs showing 2D contour plot of ZEN removal % at pH 7as a function of factors A and B with factor C (ZEN dosage) held constant at 14ppm and interaction plot of the response data against factor A (%CEC replaced) for both levels of factor B (amount of clay added) ...................................................................................................... 77 4 17 Response surface 3D plot of ZEN removal (%) as a function factors A and C. ................ 78 4 18 2D contour plots of ZEN removal % at pH 7 as a function of factors A and C and factors B and C. The amount of ZEN added is held constant at 87.5 mg. .......................... 79 4 19 XRD scans of IMTX, 40% CEC IMTX, and the IMTX TCMA Corn oil samples with an angular range of with angular range 2o ....................................................... 82 4 20 TGA curves for IMTX modified with 40% CEC TCMA and corn oil. .............................. 83 4 21 ZEN extraction by IMTX TCMA -corn oil modifie d clays. ................................................ 84 4 22 Calibration curves for ZEN extraction by HPLC -FD at pH 7 and pH 3. ........................... 88 4 23 Response surface 3D plot of ZEN remo val (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 3. The ZEN dosage is held constant at 8ppm ....................................................................... 89

PAGE 12

12 4 24 Model graps showing 2D contour lines of ZEN extraction at pH 3 as a function of factor B and factor A and interaction plot of ZEN removal against factor A. at high and low values of factor B ..................................................................................................... 90 4 25 Interaction plot s for ZEN removal against factor B at high and low values of factor A when ZEN dosage is 2ppm and 14 ppm ............................................................................... 91 4 26 Response surface 3D plot of ZEN removal (%) as a function the CEC replaced by T CMA (%) and the initial ZEN dosage at pH 3 ................................................................... 92 4 27 Model graphs showing 2D contour lines of ZEN extraction at pH 3 as a function of factor C and factor A and interaction plot of ZEN removal against factor A. at high and low values of factor C ..................................................................................................... 93 4 28 Response surface 3D plot of ZEN removal (%) as a function the initial ZEN dosage (ppm) and the amount of clay added (mg) at pH 3. ............................................................. 94 4 29 Model graphs showing 2D contour lines of ZEN extraction at pH 3 as a function of factor C and factor B and interaction plot of ZEN removal against factor B at high and low values of factor C. .................................................................................................... 95 4 30 Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 7 ............................................................................................................................................... 96 4 31 Model graphs showing 2D contour lines of ZEN extraction at pH 7 as a function of factor B and factor A and interaction plot of ZEN removal against factor A. at high and low values of factor B ..................................................................................................... 97 4 32 Response surface 3D plot of ZEN removal (%) as a function of ZEN dosage (ppm) and the CEC replaced by TCMA (%) at pH 7 ...................................................................... 98 4 33 Model g raphs showing 2D contour lines of ZEN extraction at pH 7 as a function of factor C and factor A and interaction plot of ZEN removal against factor A. at high and low values of factor C ..................................................................................................... 99 4 3 4 Response surface 3D plot of ZEN removal (%) as a function of ZEN dosage (ppm) and the amount of clay added (mg) at pH 7 ....................................................................... 100 4 35 Model graphs showing 2D contour lines of ZEN extraction at pH 7 as a function of factor C and factor B and interaction plot of ZEN removal against factor B. at high and low values of factor C ................................................................................................... 101 4 36 Calibration curve for potentiometric titration of TC MA with 100ppm SLS solution. .... 108 4 37 Calibration curve for potentiometric titration of TCMA with 0.0075M SLS solution. ... 108

PAGE 13

13 4 38 Adsorption isotherm for TCMA on IMTX at all concentrations and at high TCMA concentrations. ...................................................................................................................... 109 4 39 Potentiometric titration curve of 10ppm TCMA titrated with SLS with equivalence point (EP) ............................................................................................................................. 110 4 40 First derivative curve of 10ppm TCMA potentiometric titration curve according to volume. .................................................................................................................................. 110 4 41 First derivative cu rve of 2pmm TCMA potentiometric titration curve according to volume. .................................................................................................................................. 111 4 42 Potentiometric titration and negative first derivative curves of the supernatant after the 1st washing cycle of 100% CEC organophilic clay. ..................................................... 112 4 43 Negative first derivative potentiometric titration curves of the supernatant from the second and third washing cycles of the 100% CEC modified clays. ................................ 113 4 44 Potentiometric titration and negative first derivative curves of the supernatant after the 1st washing cycle of 40% CEC modified clay. ............................................................. 114 4 45 Negative first derivative of the potentiometric titration curve for 1ppm TCMA titrated with 100ppm SLS solution. .................................................................................... 115 4 46 Potentiometric titration and negative 1st derivative curves of duodenal electrolyte solution exposed to 100% CEC organophilic IMTX at pH7. ............................................ 116 4 47 Potentiometric titration and negative 1st derivative curves of gastric electrolyte solution expo sed to 100% CEC organophilic IMTX at pH3. ............................................ 117

PAGE 14

14 LIST OF ABBREVIATION S ANOVA analysis of variance CEC cation exchange capacity CV crystal violet DON deoxynivalenol FTIR fourier transform infrared spectroscopy HDA hexadecylami ne HDTMA hexadecyltrimethylammonium chloride HPLC -FD high performance liquid chromatography with fluorescence detection IMTX Improved Milbond TX MMT montmorillonite OCHRA ochratoxin SLS sodium lauryl sulfate TGA thermogravimetric analysis TCMA tricaprylme thylammonium chloride XRD x ray diffraction spectroscopy ZEN zearalenone

PAGE 15

15 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MODIFICATIO N AND CHARACTERIZATION OF MONTMORILLONITE CLAY FOR THE EXTRACTION OF ZEARALENONE By Kerri-Ann Alicia Hue December 2009 C hair: Hassan El Shall Major: Materials Science and Engineering Mycotoxins are secondary metabolites of organisms belonging to the fungus kingdom. The cost associated with mycotoxin contamination in the USA and Canada is approximately US $5 billion. Zearalenone (ZEN), a resorcylic acid lactone, is produced by various members of the genus Fusarium. These fungi often colonize a variety of f oods and feedstuffs including, corn, sorghum, wheat, oats, barley, and other cereal grains. This metabolite has estrogenic effects in farm animals with pigs being the most sensitive. ZEN induces hyperestrogenism and can cause infertility, reduced sex drive fetal mummification, and abortions. Clays have successfully been used in the animal feed industry as an adsorbent and binders for certain small, water soluble mycotoxins. These mycotoxins are attracted to the electrical imbalance between the layers of t he clays caused by isomorphic substitution of structural atoms. The mycotoxins are sequestered in the clay layers and pass harmlessly through the animal. However, ZEN is water insoluble and is not extracted easily with aluminosilicate clays. Therefore the modification of hydrated sodium calcium aluminosilicate (HSCAS) clays with organic cations has been proposed to render the clays hydrophobic and increase the ZEN binding capacity.

PAGE 16

16 The goal of this study was to develop a safe and cost effective organophili c material able to bind and extract zearalenone, to investigate the factors most important to extraction, and to investigate the fundamental properties between the clay -surfactant -mycotoxin systems that lead to extraction. The clay was modified by cation e xchange reactions with tricaprylmethylammonium (TCMA) chloride and generic corn oil. The organophilic clays were then characterized using XRD, FTIR, and TGA analytical techniques. These techniques were used to determine the change in fundamental clay prope rties that would lead to the extraction of ZEN. Desorption studies were performed to determine any increase in toxicity that might be caused by washing of the clays or exposure to electrolytic solutions. Statistical design of experiments was used to determ ine the factors most influential during ZEN extraction. Modification by TCMA resulted in an increase in intergallery spacing of ~0.6nm. TGA and FTIR studies indicated intercalation of organic species within the clay layers. An increase in weight loss prop ortional to the amount of TCMA added was observed by TGA analysis. In addition to the peaks found in the natural clay, peaks at 2928 cm1, 2852 cm1, and 1466 cm1, which belong to C H asymmetric stretching, C H symmetric stretching and CH2 scissoring res pectively characteristic of TCMA were present. The clays developed were able to extract >90% ZEN in vitro at pH 3 and pH7. The factors most important for extraction changed depending on the levels of parameters chosen. Mathematical models were developed th at showed the relationship between the factors and the ZEN removal percentage. When exposed to electrolyte solutions ~1.5pmm of surfactant desorbed from the modified clay.

PAGE 17

17 CHAPTER 1 INTRODUCTION Zearalenone (ZEN), a resorcylic acid lactone, is produced by various members of the genus Fusarium. These fungi often colonize a variety of foods and feedstuffs including, corn, sorghum, wheat, oats, barley, other cereal grains, and bread (Tanaka et al. 1988) This metabolite has estrogenic effects in livestock wi th swine being the most sensitive (Pitt, 2000) Although ZEN is often discussed with mycotoxins, it is biologically potent but not toxic. It is better to refer to ZEN as a nonsteroidal estrogen or mycoestrogen (J. W. Bennett & Klich, 2003) The structure of the zearalenone molecule is shown below in Figure 1 1. Concentrations as low as 1ppm can lead to hyperestrogenism in pigs and higher concentrations can cause infertility, reduced sex drive, fetal mummification, and abortions (Abbs et al. 2006b; Feng et al. 2008; Hadiani et al. 2003) Figure 1 1. 2D model of zearalenone molecule (A. Dakovic et al. 2007) The contamination of foods and feeds by ZEN has been reported throughout the world (Hadiani et al., 2003; Zinedine et al. 2007) Agricultural practices, climate, and storage conditions greatly influence the production of Fusarium fungi and subsequent ZEN c ontamination (Magan & Olsen, 2004) Although there are no concrete reports of ZEN contamination being harmful to hu man beings, there are some reports of the mycoestrogen being found in the endometrial tissue of women with endometrial cancer and one study has linked it to

PAGE 18

18 precocious puberty in Puerto Rico (Derodriguez, 1984; Derodriguez et al. 1985; Magan & Olsen, 2004) In a 2000 Joint Food and Agriculture/World Health Organization (FAO/WHO) Expert Committee on Food Additives report, the committee determined that ZEN could be evaluated based on the minimum dosage that causes hormonal effects in pigs. They suggested a tempor ary tolerable daily intake (TDI) of 0.2 g/kg of body weight. It was also reported that in five regional European and Middle Eastern diets the intake of ZEN was between 0.03 and 0.06 g/kg of body weight ( Fusarium toxins part 2: Zearalenone (zea) 2000) Due to the economic losses and the impact on livestock and human health there is great interest in the study of mycotoxins. It is reported that in the USA and Canada, the cost associated with the impact mycotoxins have on the feed and livest ock industry is approximately US $5 billion. Although there is little worry about food contamination in developed countries, in poorer countries mycotoxin contamination has a greater impact on the agricultural economy and animal and human health (Fokunang et al. 2006) In order to combat the effects of some mycotoxins, alluminosilicate clays have been successfully used in the animal feed industry as an adsorbent and binders for certain small, water soluble mycotoxins such as aflatoxins (Abbs et al. 2006a; Chestnut et al. 1992; A. Dakovic et al. 2007; Aleksandra Dakovic et al. 2005; Dixon et al. 2008) Studies have shown that hydrated sodium calcium aluminosilicates (HSCAS) and other layered clays can bind aflatoxins in the gastrointestinal tract and reduce their bioavailability and toxicity (Dixon et al., 2008; M. A. Abdel -Wahhab, 1999) Desheng and coworkers were able to extract greater than 80% of aflatoxin at pH 2 and greater than 90% at pH 8 from 10mL aqueous solutions containing 0 20 micrograms of aflatoxin B1 (AFB1) when 50mg calcium montmorillonite clay was added. In addition, in vivo studies showed a significant decrease in the

PAGE 19

19 adverse effects of AFB1 when the clay was added at an amount of 0.5 wt% of the total feed and the toxin was added to the feed at 200micrograms per kilogram of weight (Desheng et al. 2005) Using x ra y diffraction techniques (XRD), it was determined that the planar aflatoxin molecule enters directly into the interlayers of the IMTX and remain sequestered there (Dixon et al., 2008; Kannewischer et al. 2006) The structure of the aflatoxin B1 molecule is shown below in Figure 1 2. Figure 1 2. 2D model of Aflatoxin B1 molecule. Previous attempts to use adsorbents to prevent the effects of zearalenone in livestock have been largely ineffective. The alluminosilicate clays which were quite successful in binding aflatoxin were much less successful at extr acting ZEN (Abbs et al., 2006a; Abbs et al., 2006b; Afriyie Gyawu et al. 2005) Therefore the modification of aluminosilicate clays with organic cations has been proposed to r ender the clays hydrophobic, increase the intergallery spacing, and thus increase the ZEN binding capacity. In vitro studies have shown that modification of clays with cations containing 16 -carbon alkyl chains such as cetyl pyridinium (CP) and hexadecyltri methylammonium (HDTMA) have a greater ZEN binding capacity than unmodified clays (Afriyie Gyawu et al., 2005; Lemke et al. 1998; Lemke et al. 2001a) Although Zhang and coworkers demonstrated that surfactant modified clays are relatively stable i n aqueous salt

PAGE 20

20 solutions (Zhang et al. 1993) thes e studies do not accurately model the internal environment of animals. Therefore, there is some concern that noncovalently bonded surfactant molecules can desorb from the interlayer of the clays and be potentially toxic (Lemke et al., 2001a) Animals that were fed organophilic clays and ZEN contaminated feed showed increased feed toxicity, decreased weight gain, and refusal to eat (Afriyie Gyawu et al., 2005; L emke et al., 2001a) In attempts to decrease the toxicity of organophilic clays, Lemke and coworkers modified montmorillonite clay with hexadecylamine (HDA). They hypothesized that HDA which is a primary amine and has a higher mean lethal dose (LD50) valu e than quaternary amines would be able to exchange with the IMTX and produce a less toxic ZEN binding product. The results showed that the HCA modified clay had a lower ZEN binding capacity than the HDTMA modified clay. The goal of this work is to develo p a biocompatible organoclay able to bind and extract zearalenone in amounts greater than 90% and to investigate the fundamental relationships that exist between surfactant, clay, and toxin. In developing this clay, statistical design of experiments was us ed to study some of the factors involved in extraction. Hydrated sodium calcium alluminosilicate clays (HSCAS) were modified with tricaprylmethylammonium chloride (TCMA), soybean lecithin, and a combination of corn oil and TCMA. The clays were modified w ith organic species in varying amount of the clays cation exchange capacity (CEC). The clay counterions were exchanged with crystal violet dye and the amount of counterions released from the clay was determined using inductively coupled plasma (ICP) to de termine the cation exchange capacity of the HSCAS clay. X ray diffraction techniques were used to determine the intergallery d -spacing of the modified and unmodified clays. Fourier transform infrared spectroscopy was used to investigate the interaction bet ween

PAGE 21

21 organic species and clays. The thermal stability and amount of organic species adsorbed was quantified using thermogravimetric analysis (TGA). In addition, adsorption and desorption isotherms of surfactants on clay were obtained using potentiometric t itration techniques. A 23 factorial design was used to determine how parameters such as the amount of surfactant used for modification, ZEN dosage, and amount of clay influence ZEN adsorption. Statistical analysis provided a mathematical model to predict how each factor and any interactions between them would affect ZEN adsorption. To determine the amount of ZEN adsorbed and extracted by the clays in vitro, UV -visible spectroscopy and high performance liquid chromatography with fluorescence detection were used. This research will be described in greater detail in chapters 3 and 4.

PAGE 22

22 CHAPTER 2 BACKGROUND AND LITER ATURE SURVEY 2.1 Mycotoxin Contamination Worldwide Mycotoxins are toxic secondary metabolites produced by fungi. These toxins can have significa nt effects on human and animal health. The mycotoxins of most concern are aflatoxins, fumonisin, ochratoxin (OCHRA), T 2 toxin, deoxynivalenol (DON), and zearalenone (Fokunang et al., 2006) The fungi that produce these mycotoxins can be found in various food products such as cereals, grains, wheat, and barley. Mycotoxins can have a cumul ative effect in humans causing cancer, acute symptoms, and immune -deficiency diseases. In livestock, they can be linked to reduction in body weight gain and lower feed intake and efficiency (Watts et al. 2003) Accumulation of mycotoxins in animal tissue can be a source of exposure to humans who consume products originating from animals. It has been estimated that approximately 25% of food crops throughout the world is contaminated with mycotoxins. In the US a nd Canada alone the cost associated with contamination is $5 billion (Feng et al., 2008; Fokunang et al., 2006) Many strategies for cost effective detoxification of mycotoxin-contaminated feedstuff have been explored with little success. The most employed method of protection against mycotoxins is the addition of nonnutritive adsorptive materials to the feeds of livestock animals to reduce the bioavailability of mycotoxins in the gastrointestinal tract. This method has been successful for protection against aflatoxins and the smaller more water soluble mycotoxins. However, these materials have not been as effective for protection against ochratoxin, deoxynivalenol, and zearalenone which are all hydrophobic chemicals. Montmorillonite (MMT) clays have been used as feed additives to reduce th e bioavailability of some mycotoxins in livestock. Aflatoxin is a common mycotoxin found in

PAGE 23

23 animal feed. It is toxic and is one of the most carcinogenic substances known. Researchers have found that an inclusion of <1wt% of various clay types in the diet of livestock animals could produce a reduction in available aflatoxin. When MMT and hydrated sodium calcium aluminosilicate (HSCAS) clays were used the improvement ranged from 50% to 100% depending on the type of clay binder, the amount of Clay added, and the concentration of AFB1 in the diet (Desheng et al., 2005; Dixon et al., 2008; K annewischer et al., 2006) With the success of clays to extract aflatoxin, attempts were made to extract other mycotoxins such as ochratoxin (OCHRA), deoxynivalenol (DON), and zearalenone (ZEN). Ochratoxin has been linked to Balkan endemic nephropathy in humans and immunological suppression and cancer in animals (Fokunang et al., 2006) In studies conducted by Dakovic and coworkers, natural clays were able to adsorb 40% ochratoxin (in vitro) at pH 3 and 3% at pH 7 (A. Dakovic et al. 2003) Deoxynivalenol is quite likely the most widely occurring mycotoxin. It has been given the nickname vomitoxin due to the emetic effects it has on infected livestock. Outbreaks of vomitoxin has caused acute cases of mycotoxicosis in Asia (Fokunang et al., 2006) In a study conducted by Sabater -Vilar and coworkers to evaluate the various adsorbents available for mycotoxin extraction, they found that aluminosilicates, humic substances, and yeast cell walls were very poor adsorbents (8 15%) of DON at acidic and alkaline pHs. Activated charcoal performed the best and showed an average adsorption of 85% at both pHs. 2.2 The Impact of Zearalenone Contamination Zearalenone [3,4,5,6,9,10 hexahydro14,16-dihydroxy 3 -methyl 1 H 2 benzoxycyclotetradecin1,7(8 H ) dione], is considered a nonsteroidal mycotoxin or mycoestrogen. It is produced worldwide by Fusarium molds which thrive in warm moist climates. The study of ZEN as a mycotoxin began with observations made by researchers in the 1920s, J.B. Buxton noted increased occurrences of swelling and eversion of the vagina

PAGE 24

24 (hype restrogenism) in swine that were fed moldy corn and McNutt and coworkers observed inflammation in the vagina of corn -fed swine (Collins et al. 2006; Magan & Olsen, 2004) Over the next 40 years, researchers reported m ore instances of reproductive problems in swine that were fed moldy corn. A compound (F 2) was isolated from corn inoculated with Fusarium molds by Christensen et al. in 1965 and in 1966 Urry et al used NMR and mass spectrometry to isolate the same compound and named it zearalenone (Urry et al. 1966) In 1979, Bennett and Shotwell published a review of published data concerning contamination by ZEN throughout the world (G. A. Bennett & Shotwell, 1979) The data compilation can be seen in Table 2 1. ZEN contamination levels of <1ppm produced symptoms of hyperestrogenism in swine. Today data regarding ZEN contamination can be found on almost every continent in varying degrees. Contamination of crops by mycotoxins has been described as a three -stage process beginning with infection of the plant by fungal inoculum followed by growth of the fungi and finally production of the mycotoxin (Magan & Olsen, 2004) After ingestion by mouth, ZEN is Table 2 1. Natural Occurrence of Zearalenone in Mixed Feeds (G. A. Bennett & Shotwell, 1979)

PAGE 25

25 rapidly absorbed and metabolized by the liver and converted into -zearalenol, -zearalenol, and zearalanol. The toxic effect of zearalenone comes from its ability to assume a conformation similar to 17 -estradi ol which is critical for reproductive and sexual functioning in humans and animals. Although, the binding efficiency of all ZEN and its conjugates is less than 17 estradiol, they can bind to oestrogen receptors and cause female sex hormone responses (Hadiani et al., 2003; Magan & Olsen, 2004; Minervini & Dell'Aquila, 2008; Turcotte et al. 2005) Met abolism of the different conjugates of ZEN is species dependent. In swine where ZEN is most oestrogenic, ZEN is primarily found as -zearalenol or the parent compound. In rats, ZEN causes decreased fertility, resorption of fetuses, fetus deformity, and abortion when a high concentration of the toxin is administered (Turcotte et al., 2005) In female pigs, ZEN causes hyperestrogenism, reduced litter size, vulva enlargement, pseu dopregnancy, and loss of embryo. It has been suggested that the amount of ZEN in feed for pigs should not exceed 200 g per kg of body weight. The effects of ZEN can also be seen in boar reproductive organs where there can be decreased testes, cessation of spermatogenesis, and reduced libido. ZEN has shown less dramatic cases of hyperestrogenism in other livestock animals such as horses, poultry, and cattle (Lemke et al., 1998; Magan & Olsen, 2004; Minervini & Dell'Aquila, 2008) Further studies of ZEN have shown that the compound is teratogenic, carcinogenic, and mutagenic in animals (Aw et al. 1989) There is some debate about whether or not ZEN impacts human health due to the low doses that are found in foods consumed by people. Th is controversy extends to breast cancer which has been linked to estrogen exposure. Reports that 17 -estradiol has been linked to breast cancer led Ahamed and coworkers to investigate ZEN as a contributor to the overall estrogenic exposure in women (Ahamed et al. 2001; Shier et al. 2001) In addition, zearalenone has been

PAGE 26

26 implicated in the early pubertal development of children in Puerto Rico and Hungary. Researchers have presented data linking ZEN to cancer of the cervix, endometric hyperplasia, and esophageal cancer (Briones Reyes et al. 2007) Human exposure to ZEN can come from multiple sources, in addition to food ZEN has been used as a contraceptive and in estroge n therapy for women. At one point, zearanol was the most popular growth hormone used in the United States as a growth promoter in sheep and bovines (Price & Fenwick, 1985) It was used in animal husbandry and to promote weight gain in cattle and sheep before slaughter (Aw et al. 1989) 2.3 Zearalenone Decontamination Research Du e to the economic and health costs of ZEN contamination much research has gone in reducing and eliminating the effects of this mycoestrogen. One method involves prevention of mycotoxin production. Researchers have investigated developing crops that are res istant to fungus, reducing insect damage to plants, and controlling humidity and temperature during grain storage (Lemke et al., 1998) According to Whitlow et al, drought and insect damage are the most important factors involved in mold growth in fields. After harvest and during storage mold proliferation can occur. Therefore, grains should be stored in such a way to eliminate moisture migration, condensation, and leaks. Nevertheless, with all the recommended safeguards there is still mycotoxin contaminat ion of crops. In addition, in less developed countries the lack of knowledge and resources still allows for high levels of mycotoxin and ZEN contamination. Chemical methods and thermal treatment was employed in attempts to destroy ZEN. Although some resear chers reported that hydrogen peroxide and ammonium hydroxide was used to reduce the ZEN on contaminated corn, the amount of ZEN in the contaminated corn was not reported (G. A. Bennett & Shotwell, 1979) Later reports stated that propionic acid, acetic acid, hydrochloric acid, and hydrogen peroxide did not reduce toxin levels. In addition, high

PAGE 27

27 temperature treatments u p to 150oC also did not reduce toxin levels. Bennett et al reported that formaldehyde in both aqueous and vapor form was the most effective at destroying ZEN in corn spiked with the toxin (G. A. Bennett et al. 1979) Most recently, attempts have been made to reduce the bioavailability of ZEN for adsorption in the gastrointestinal tr act. Clays that were used to extract aflatoxin from livestock were investigated for the extraction of zearalenone. Researchers did not have the same success extracting zearalenone as they did with aflatoxin. Sabater -Vilar et al studied various adsorbents for the extraction of ZEN including mineral clays, humic substances, yeast cell -walls, commercially available mycotoxin binders and activated charcoal as the control substance. The in vitro tests determined that when the concentration of clay was added was 2.5 mg/mL, the most effective material for ZEN adsorption at alkaline and acid pHs was activated charcoal. This material exhibited 100% adsorption at any inclusion rate. However, activated charcoal is not an ideal food additive due to nutritional limitatio ns such as the adsorption of minerals, vitamins, and other nutrients. Of the other 20 materials tested, the unmodified materials were able to extract ~70% of ZEN (Sabater -Vilar et al. 2007) 2.4 Aluminosilicate Clays for Zearalenone Adsorption 2.4.1 Clay Structure Although there are many clay types and classifications, montmorillonite and hydrated sodium calcium aluminosilicates have been investigated most frequently for the extraction of mycotoxins. Aluminosilicate clays are comprised of tetrahedral silica sheets which sandwich an aluminum -oxygen-hydroxyl octahedral. The silicon atoms in the tetrahedral sheets are coordinated with oxygen atoms and three or four oxygen at oms are shared by the neighboring tetrahedra. In the octahedral sheets, the Al atoms are coordinated with six oxygen atoms or OH groups. Silicon atoms in the tetrahedral sheets are sometimes replaced with aluminum or

PAGE 28

28 magnesium atoms by isomorphous substitution, resulting in an excess of negative charges (van Olphen, 1976) These unbalanced charges are resolved by the presence of positively charged exchangeable counterions such as Na+, K+, and Ca++. The structure of a luminosilicate clays can be seen below in Figure 2 1. The total number of exchangeable ions can be quantified and is called the cation exchange capacity (CEC) and is expressed in terms of milliequivalent per 100 grams of dry clay. In a stack of clay laye rs, the counterions are located on both sides of the layer and therefore can be found on the external clay surface but also within the clay layers. Figure 2 1. Layered structure of montmorillonite clays. The most common method for determination of the cation exchange capacity of clays is by exchanging the cationic counterions with other organic cations. Although organic dyes such as rhodamine, thionine, and acridine orange have been used to determine CEC, methylene blue and crys tal violet are the most popular. Brindley and coworkers introduced methylene blue adsorption for CEC determination in 1970 and it remains the most popular method for CEC determination today (Lagaly, 1981) Rytwo and coworkers examined both crystal violet (CV)

PAGE 29

29 and methylene blue (MB) dyes for the determination of exchangeable ions in montmorillonite clays. The results showed that both dyes exchanged roughly the same amount of cations when the amount of dye added was slightly above the CEC of the clay. However, the CEC could not be determined accurately by the amount of dye adsorbed but rather should be determined by the exchanged cations. Of the two dyes, the report that CV may be preferabl e for CEC determination due to problems with incomplete displacement of the counterions at high MB concentrations (Rytwo et al. 1991) The cation exchange capacity for al uminosilicate clays ranges between 0.8 and 1.5 meq/100g of clay. Using x ray diffraction techniques and Braggs equation it is possible to determine the distance between the layers of clays. This distance is called the interlayer d -spacing or the basal sp acing. Most MMT and HSCAS clays have an interlayer spacing of 0.91.2 nm. This layer spacing is large enough to accommodate the smaller mycotoxins such as aflotoxin (~1 nm). In addition, it is presumed that the -dicarbonyl groups of the aflatoxins molecul es is essential for chemical (electrostatic) adsorption to phyllosilicate clays (Grant & Phillips, 1998; Sabater -Vilar et al., 2007) The laye red structure of these clays, their capacity to exchange ions, and hydrophilic nature make them good candidates for the extraction of polar mycotoxins. However, less polar toxins such as OCHRA, DON, and ZEN have shown lesser binding affinities to these mat erials. 2.4.2 Organophilic Clays Organophilic clays are most preferably produced by cationic exchange of interlayer cations with cationic organic species. With the discovery that interlayer cations could be exchanged with methylene blue and other dyes, res earchers began to investigate other cationic species such as alkylammonium ions. Since the 1950s many research groups have reported the modification of

PAGE 30

30 clays with quaternary ammonium salts (de Paiva et al. 2008; Jordan et al. 19 50) During cation exchange the interlayer counterions of the clay mineral are replaced by organic compounds in aqueous solution. The amount of exchange is determined by the length of the hydrocarbon chain and can be equal to or greater than the CEC of th e natural clay. In 1950, Jordan and coworkers reported the modification of MMT clays with commercial organic compounds. They observed the importance of the number of carbon atoms in the amine chain length. They determined that a carbon chain length of ten was needed to show organophilic properties and a length of 12 carbons was required for maximum swelling of the clays. They also observed an increase in the 001 or basal plane spacing of the clays on the order of 4 which corresponds to the van der Waal s diameter of a methyl group. These findings were taken as an indication of the orientation of the aliphatic chains that were intercalated within the clay layers. It was determined that the hydrocarbon chains were oriented in a flat monolayer along the sur face of the planes in a zig -zag conformation parallel to the surface of the clay layers (Jordan et al., 1950) Since Jordans work, many researchers have reported modification of aluminosilicate clays with alkylammonium surfactants and other cationic surfactants. More thorough models for the intercalation of primary and dialkylammonium cations were developed. The conformation of the organic species within t he clay layers is dependent on the surface charge and the length of the hydrocarbon chain (Handbook of clay science 2006) The alkylammonium ions can be arranged in a monolayer, bilayers, pseudotrimolecular layers, and paraffintype arrangements. The possible conformations are shown in Figure 2 2. The shorter organic ions form monolayers while longer ions form bilayers parallel to the surface of the clay (de Paiva et al., 2008) Pseudotrimolecular conformations are observed when the clay minerals are highly charged

PAGE 31

31 Figure 2 2. Ori entations of alkylammonium ions in the galleries of layered silicates: a) monolayer, b) bilayers, c) pseudotrimolecular layers, and d, e) paraffin type arrangements of alkylammonium ions with different tilting angles of the alkyl chains (Handbook of clay science, 2006) Figure 2 3. Variation of basal spacings of alkylammonium montmorillonites with chain length (nc) due to formation of monolayers, bilayers and pseudotrimolecular layers of alkylammonium ions in the interlayer spaces (de Paiva et al., 2008)

PAGE 32

32 and/or when the alkyl chai ns are long. The paraffin type structures are formed by quaternary ammonium ions with two or more hydrocarbon chains. Lagaly and coworkers related alkyl chain length to interlayer spacing and chain conformation, as chain length increases the interlayer spa cing increases and the conformations progress from monolayer to bilayer to pseudotrimolecular (Lagaly, 1986) .This trend can be seen in Figure 2 3. Vaia and coworkers investigated the interlayer structure of several modified layered silicates. The clay minerals were modified with primary amines and quaternary ammonium surfactants. They used fourier transfo rm infrared spectroscopy and XRD to provide insight into the interlayer structure and phase state of intercalated alkylammonium organoclays. They investigated how factors such as chain length, temperature, and packing density affect the interlayer structur e of Organophilic clays. They concluded that by monitoring frequency shifts in CH2 stretching and bending, they could determine that intercalated chains could exist in different states. They also determined that by decreasing packing density and chain length or by increasing temperature a more disordered and liquid like conformation was attained (Vaia et al. 1994) Organophilic clays are most often used commer cially to produce polymer nanocomposites. They are an ideal choice for this industry due to the low cost compared to other nanomaterials, the raw materials are ubiquitous and abundant, and they can be produced in existing large scale production facilities. 2.4.3 Organophilic Clays for the Extraction of Zearalenone Researchers postulated that the decrease in ZEN binding affinity to natural unmodified clays was due to the hydrophobic nature of ZEN and inability of the toxin to enter into the clay galleries. In addition, modified clays have been studied for the extraction of hydrophobic pollutants such as benzene, toluene, ethylbenzene, and xylene (BTEX) (Sabater -Vilar et al.,

PAGE 33

33 2007; Sharmasarkar et al. 2000) Sharmasarkar and coworkers mod ified MMT clay with primary amines (trimethylphenylammonium and trimethylaamonium adamantane) and hexadecyltrimethylammonium (HDTMA). From adsorption studies it was determined that the sorptive removal of the pollutants by adsorption mechanisms in the case of the primary amines and by partitioning in the case of the HDTMA -modified containing large chain amines. Figure 2 4 shows possible conformations of these organic species within the interlayers of a Wyoming montmorillonite clay. HDTMA modified clays were also investigated for the extraction of chromate ions. Krishna and coworkers found that the organoclay held chromate ions between the clay layers in the form of salts and determined that the modified clay could be used to concentrate and recover toxic hyd rophobic molecules from aqueous environments (Krishna et al. 2001) An other quaternary ammonium salt investigated for the extraction of BTEX hydrocarbons via an organoclay complex was tricaprylmethylammonium (TCMA) chloride. Lo et al were able to indicate intercalation of the surfactant using FTIR spectroscopy and adsorption of the hydrocarbons. Their results indicated that the more hydrophobic species had a higher affinity for adsorption by the organoclays (Lo et al. 1996) Due to the hydrophobicity of ZEN and the increased hydrophobic nature of organoclays it was postulated that the modified clays could be used to bind and extract the toxin. Lemke et al (1998) studied the extraction of zearalenone with clays modified with pyridium and ammonium based organic cations of varying chain lengths and sizes. The structure of these cations can be seen in Table 2 2. The clays were exchanged with various equivalents of each surfactant. The amounts were equal to 0.25, 0.50. 0.75, 1.0, and 1.5 of the cation exchange capacities (CEC) of the clay. After performing adsorption isotherm studies, the researchers reported the 1.5 CEC exchanged cetylpyridium -MMT clays showed the best binding capacity of ZEN. The adsorption

PAGE 34

34 isotherms of ZEN on organophilic clays can be seen in Figure 2 5. They found that the greater the hydrophobicity of the clay, the higher the binding capacity for ZEN (Lemke et al., 1998) Cetyltrimethyl ammonium bromide (CTAB) was also us ed to modify MMT clays for the extraction of ZEN. Feng et al (2008) studied several factors that might impact ZEN extraction by MMT clay exchanged with CTAB at amounts equal to 150% of the CEC of the clay. The factors studied included pH, adsorbent amount, temperature, interaction time, shaking speed, and initial ZEN concentration. The modified clay showed an increase in ZEN extraction over the natural clay mineral from 0.60 mg/g of clay to 8.83 mg/g. The desorption rate was between 0.9 and 5.8% depending on pH. Both natural and modified clays exhibited greater binding capacity at alkaline Figure 2 4. Probable arrangements of the organic cations in the clay interlayers of a Wyoming montmorillonite clay (SWy) modified with: a) trimethylphenylammonium (TMPA ) and trimethylaamonium adamantine (Adam) b) HDTMA (Sharmasarkar et al., 2000)

PAGE 35

35 Table 2 2. Structure and Critical Micelle Concentration of exchanged surfactants (Lemke et al., 1998) pHs. They postulate this is due to ZEN having a negative charge at higher pHs. ZEN molecules were adsorbed by the clays within 30 minutes and equilibrium was reached after 45 minutes. The researchers concluded that the modified clay could be employed for the removal of ZEN from aqueous solutions (Feng et al., 2008) Extensive desorption studies were performed by Zhang et al (1993) on organophilic clays. The clays were agitated in aqueous salt solutions containing various salts to determine if the surfactants would de sorb due to competitive adsorption mechanisms. Their short term (48 hours) data suggests that the amount of desorption is directly proportional to the amount of amines adsorbed and inversely proportional to chain length. The HDTMA modified clays exhibited less than 1% desorption when the added amount was less than 100% of the CEC of the clay. Above

PAGE 36

36 100% CEC of surfactant added, desorption increased to 10% of the adsorbed quaternary amines. Long term desorption studies showed no increase in desorption over t ime (up to 180 days) (Zhang et al., 1993) It is im portant to note that the electrolyte solutions used in these desorption studies do not mimic those found within the stomach. Figure 2 5. Adsorption isotherms for ZEN on clays exchanged with a variety of organophilic cations. Studies were performed at 37oC for 2 hours.Data represent the mean adsorption from three replicate experiments (Lemke et al., 1998) To further examine organophilic clays for the extraction of ZEN, Lemke and coworkers (2001) performed in vitro studies incorporating 0.25% and 0.5% by weight modified clays into ZEN -contaminated diets of mice. Two surfactants, hexadecylamine (HDA) and HDTMA, were exchanged at amounts equal to 100% of the CEC of the MMT clay. The HDA modified clay was prepared in attempts to take advantage of the proper ties of the organoclays and decrease the risk

PAGE 37

37 of toxicity associated with quaternary amines. They found that at 0.25% inclusion, both modified clays exhibited no significant change in the uterine weight of the mice as compared to the control. When the clay s were added at 0.5% inclusion, the mean uterine weights of the mice decreased without the addition of ZEN. However, when 35ppm ZEN was added to the diets, the uterine weights did not change statistically. In addition, the mice exhibited feed refusal when fed the clays modified with HDTMA. From these results, the researchers concluded that the modified clays were not effective in protecting the animals from the estrogenic effects of ZEN. They also hypothesized that the amine compounds desorbed from the cl ays and interacted with the membranes of the gastrointestinal tracts of the animals allowing for an increase in ZEN adsorption (Lemke et al., 2001a) 2.5 Hypothesis for Clay Modification and Zearalenon e Extraction As mentioned previously, there is a worldwide need for a feed additive that can reduce the bioavailability of ZEN and protect animals from the estrogenic effects of this mycotoxin. This additive must traverse the intestinal tract of animals en tering areas where the pH varies from pH 3 to pH 7. It must be nontoxic and the organic species should be stable in aqueous environments where there is an excess of cationic species that can compete for the anionic sites in the interlayers of the clays. Th e complications with the current organoclays involve ZEN efficiency in vivo and toxicity issues. It should be noted that the test performed by Lemke and coworkers involved uterine bioassays of rats which do not show the same susceptibility to ZEN as swine. In addition, the organophilic clays that have been tested for ZEN adsorption have been modified in amounts equivalent to 100% of the cation exchange capacity of the clay. There is a worldwide interest in developing a cost effective and nontoxic organocla y able to extract ZEN from the intestinal tract of livestock.

PAGE 38

38 Milwhite Inc, has been interested in developing such an organoclay. In this effort they have modified their own hydrated sodium calcium aluminosilicate clays (Improved Milbond TX) for the extra ction of ZEN. They modified IMTX by manually mixing with several commercially available ammonium surfactants. They were able to increase the ZEN binding capacity to approximately 85%. In other studies, organophilic clays were prepared by mixing IMTX with f atty acids (20% by weight) having chain length between 16 to 20 carbon atoms. These acids included, corn oil, olive oil, peanut oil, and tall oil. They also increased the ZEN binding capacity to ~85%. To reduce cost combinations of corn oil and surfactant were used to modify IMTX clays. These combinations removed no greater than 52% of ZEN. It was also concluded that it would be necessary to add between 35%40% corn oil by weight in order to extract >90% ZEN. After reviewing the previous work done by Milwhite, Inc., it was proposed to that IMTX be modified by tricaprylmethylammonium chloride (TCMA), a quaternary ammonium cation with three alkyl chains with ten carbon atoms. This particular surfactant was chosen due to previous studies indicating that it cou ld extract hydrophobic hydrocarbon molecules from aqueous solution when exchanged in amounts equal to 30% of the CEC of the clay (Lo et al., 1996) Most surfactant modified clays studied for ZEN extraction were exchanged with surfactants with one or two hydrocarbon chains. The additional alkyl chain of TCMA will present possibilities for different surfactant conformati ons and may allow for higher ZEN extraction efficiencies at lower surfactant concentrations. Figure 2 6 shows the structure of the TCMA surfactant molecule. Also, to reduce cost, modification of IMTX with a combination of surfactant and corn oil was propos ed.

PAGE 39

39 Figure 2 6. Structure of tricaprylmethylammonium (TCMA) chloride surfactant. In order to extract ZEN, the clays must be modified to expand the galleries of the clays and render them hydrophobic. As mentioned above, it will be necessary to study seve ral factors that can influence ZEN adsorption. The factors that will be studied are initial ZEN concentration, amount of organic species added for modification, pH, and the concentration of Clay added for extraction. Although other researchers have studied these factors, none have studied them simultaneously using design of experiments. Statistical design of experiments has the advantages of reducing time and cost while determining the interactions if any between the parameters being studied. The final resu lt will be a mathematical model that will relate the factors and indicate which are most important for ZEN extraction. Along with the extraction of ZEN, it is necessary to study the fundamental properties of the modified clays that enable extraction of the toxin. In summary, the background and fundamental literature survey about zearalenone, its economic and health impacts worldwide, methods for prevention of its estrogenic effects, the properties of clays and how modifications can lead to ZEN extraction ha ve been discussed. This (C 25 H 54 NCl)

PAGE 40

40 work focuses on accomplishing the goals of Milwhite Inc as mentioned above and investigating the fundamental properties of the modified clays that enable ZEN removal.

PAGE 41

41 CHAPTER 3 EXPERIMENTAL AND CHA RACTERIZATION METHODOLOGY The goa l of this project is to prepare organosilicate clays able to bind and extract the mycotoxin Zearalenone and to investigate which factors in the clay modification and the ZEN adsorption process have the greatest impact on the binding efficiency of the clays The clays are modified through a cation exchange process in which the counterions of the hydrated sodium calcium aluminosilicate clays provided by Milwhite, Inc. are replaced by cationic organic species. The clays are then characterized by x ray diffract ion (XRD) techniques, fourier transform infrared (FTIR) spectroscopy, and thermogravimetric analysis (TGA). ZEN adsorption was monitored using UV -visible spectrometry and high performance liquid chromatography (HPLC) with fluorescence detection. The factor s most important during modification and adsorption processes for ZEN extraction were determined using statistical analysis. 3.1 Clay Characterization Improved milbond TX (IMTX), a hydrated sodium calcium aluminosilicate clay was provided by Milwhite, Inc.. The cation exchange capacity (CEC) and the intergallery spacing of the natural IMTX clays were determined. X ray diffraction techniques (XRD) were used to determine inter gallery d -spacing the clay. XRD can be used to acquire information about the ch emical composition and crystallographic structure of a material. This method is the most useful for identifying and quantifying clay minerals (Handbook of clay science 2006) During an XRD scan, characteristic x rays are produced and are projected onto the sample. The x rays are then diffracted and counted by a detector. A typical scan is plotted as counts per second vs 2 theta. XRD analysis was radiation. Scans were conducted at 40kV and 20mA with an angular range of 2o o and a step size of 0.01. The d-spacing was calculated using Braggs Law:

PAGE 42

42 =2d*sin (3 1) In this equation i s the wavelength of xrays hitting the sample, d is the intergallery spacing, and is the angle between the incident rays and the scattering planes or in this case the 001 plane. For this instrument = 1.54. Figure 3 1 shows a characteristic scan of IM TX and indicates the location of the 001 plane peak. Figure 3 1. XRD scan for natural IMTX with an angular range of 2o o The d -spacing is calculated from the angle of the 001 plane peak. Milwhite, Inc., reported the CEC of IMTX to be 56 meq/100g of clay in one report and 139.9 mmol/100g in another report. Rytwo and coworkers determined that both crystal violet and m ethylene blue dyes could be used to determine the CEC of montmorillonite clays (Rytwo et al., 1991) To confirm the value reported by Milwhite, Inc., the CEC of IMTX was d etermined by crystal violet (CV) adsorption. The structure of the crystal violet molecule can be seen in Figure Improved Milbond TX 0 100 200 300 400 500 600 700 800 900 1000 2 12 22 32 42 52 62 2-theta CPS

PAGE 43

43 3 1. The counterions are exchanged with CV, which has a large binding affinity to the clay. The CEC was determined by the amount of displaced sodium (Na+), calcium (Ca++), and magnesium (Mg++) ions. This amount was determined by inductively-coupled plasma emission spectroscopy (ICP). Figure 3 2. Molecular structure of crystal violet. Suspensions were prepared in 100mL plastic bottles containing 0.5 grams of IMTX clay in 60 mL nanopure water. CV dye was added in amounts equal to 0.0, 0.3, 0.6, 0.9, 1.2, and 1.5 meq/g of clay. The suspension was kept at room temperature ~22oC and stirred. 10 mL aliquots were taken from each bottle after 1, 3, 7, 1 5 days and filtered through 0.25 m filters. The amount of dye in the filtrates was determined using UV -visible spectroscopy. Adsorption was measured at 588 nm using a Perkin Elmer Lambda 800 spectrometer. In addition, the amount of ions released was deter mined using a Perkin Elmer Optima 3200RL ICP spectrometer It was previously determined that the CEC should be determined by the concentration of inorganic ions released rather than the amount of dye adsorbed and the presence of organic dyes did not interf ere with the measurement of inorganic ions (Rytwo et al., 1991)

PAGE 44

44 3.2 IMTX Modification with TCMA To prepare for cation exchange 5g of IMTX was added to 250 ml of nanopure water at 45oC and allowed to swell for a minimum of 3 hours. Tricaprylmethylammonium chloride (C25H54ClN), a quaternary ammonium cation, was added dropwise to IMTX while stirring at amounts equal to 40%, 70%, 100%, 200%, and 300% of the cation exchange ca pacity (CEC) of the clay. The resulting IMTX agglomerates were washed with nanopure water and collected by centrifugation until no NaCl was detected with a 0.1M AgNO3 solution. The clays were then dried at 70oC for 24hours. After drying the modified clays were ground through high speed pulverization using a Wig-L Bug grinding mill. To reduce cost, IMTX was modified with a combination of corn oil and TCMA. The clays were prepared for exchange as mentioned above. TCMA was then added dropwise to the clay susp ension under vigorous stirring at amounts equal to 20% of the CEC of the clay. After 2 hours of mixing, store brand corn oil was added as received in amounts equal to 3%, 7%, and 10% by weight of the clay. The resulting IMTX agglomerates were washed with n anopure water and collected by centrifugation until no NaCl was detected with a 0.1M AgNO3 solution. The clays were then dried at 70oC for 24hours. 3.3 Characterization 3.3.1 X -Ray Diffraction Scans were conducted at 40kV and 20mA with an angular range of 2o o and a step size of 0.01 and the interlayer d-spacing was calculated using Braggs Law. 3.3.2 Fourier Transform Infrared Spectroscopy (FTIR) Infrared radiation wavelengths range from 4000 cm1 to 400cm1. When infrared radiation comes in contact with a sample, some of the radiation is adsorbed and some passes through the sample. FTIR detects the vibrational characteristics of chemical functional groups in a sample.

PAGE 45

45 When IR light interacts with a sample, chemical bonds of a functional group will be nd, stretch, and contract. A particular functional group will adsorb IR light in a specific wavenumber range regardless of the structure of the molecule. The intercalation of surfactant molecules within the clay layers was verified using FTIR. The natural IMTX and the TCMA modified clays were dried overnight in a vacuum oven at 80oC before analysis in the Nicolet Magma FTIR. The clays were ground with KBr crystals at an inclusion of 1wt%. Absorbance spectra were plotted with 256 scans and 4cm1 resolution. The spectrum of the pure surfactant was obtained by drying upon KBr crystals at 80oC in a vacuum oven overnight. The crystals were then ground and analyzed in the same manner as the clay samples. 3.3.3 Thermal Analysis Thermogravimetric analysis (TGA) and Simultaneous Differential Thermal Analysis (SDTA) was performed on a Mettler Toledo TGA/SDTA851. TGA records mass loss during a time or temperature profile. From this mass loss information about moisture content and phase changes which occur at specific temperatures can be attained. SDTA measures the mass loss and temperature difference between an unknown sample and a reference sample as a function of temperature. Weight loss data for IMTX, TCMA -modified clays, and TCMA -corn oil -modified clays was determi ned from 20 800oC in nitrogen atmosphere, with a heating rate of 20oC/min. 3.3.4 UV Vis and HPLC -FD for determining ZEN Extraction Methods for detecting Zearalenone include thinlayer chromatography, gas -liquid chromatography-mass spectrometry, and HPLC. Z EN extraction was determined using two methods Ultraviolet visible (UV -vis) spectroscopy and HPLC FD. UV -vis spectroscopy can only be used to detect larger quantities of ZEN therefore HPLC -FD must be used to confirm data collected by UV -Vis. In UVvis spec troscopy, photons of light in the ultraviolet and visible

PAGE 46

46 spectrum (200800nm) are passed through a sample and the absorbance is measured and compared to the initial intensity of light. This instrument can be used to quantify the amount of an unknown in a sample using the Beer Lambert Law, which states that the concentration of the unknown is directly proportional to the amount of light it absorbs. A calibration curve is determined by preparing 20ppm, 14ppm, 8ppm, and 1ppm ZEN solutions at pH7 and pH3. The absorbance of each sample was measured using a Perkin Elmer Lambda 800 UV -Vis spectrometer at 284nm. The ZEN concentration is plotted vs. the absorbance detected. From this curve and its characteristic equation that is found by linear regression, the ZEN contained in an unknown sample can be calculated. In the equation x is the ZEN concentration of the unknown and y is the absorbance of the unknown sample. An example of a ZEN calibration curve can be seen in Figure 3 3. The R2 value is an indicator of how well the regression line fits real data. A value of 1.0 indicates a perfect fit. ZEN Calibration Curve y = 0.0335x 0.0181 R2 = 0.9987 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 5 10 15 20 ZEN concentration (ppm) Absorbance Figure 3 3. Zearalenone calibration curve.

PAGE 47

47 HPLC is a column chromatography method that is used to separate, identify, and quantify compounds. The system consists of a column that contains chromatographic packing material, a pump that moves the mobile phase through the column and a detector that shows the retention times of the molecules. The HPLC -FD ZEN extraction data was collected following the AOAC official method for de termining ZEN by liquid chromatography zearalenol and zearalenone in corn, liquid chromatographic method") The instrument used was an Agilent 1100 Series HPLC with a symmetry 4.6x50mm (3.5micr onparticle size) column from Waters. The fluorescence detector was set at 274nm excitation wavelength and 440 nm emission wavelength. ZEN solutions were prepared in a 70 methanol: 30 water (v/v) mixture. The mobile phase was a 70:30 methanol:water mixture with a flow rate of 0.700 mL/min and a run time of 5 minutes. ZEN had a retention time of 2.5 minutes. Calibration curves were also determined on this instrument and used to calculate the unknown ZEN solutions. The concentration of ZEN in the solutions use d to produce the calibration curve was 0.01ppm, 0.02ppm, 0.1ppm, 0.2ppm, and 1ppm. 3.4 Zearalenone Extraction Statistical Design of Experiments (DOE) is the process of planning experiments so that the necessary data can be collected and analyzed by statis tical methods. The result leads to valid and objective conclusions (Montgomery, 2005) Some of the benefits of DOE are reductions in time, cost, and materials. It is possible to identify parameter values that will maximize desired results while determining parameters that do not have an impact on th e outcome. In addition, any interactions between multiple parameters can be investigated. To determine the important factors for ZEN extraction with surfactant modified IMTX, screening experiments were performed using a design of experiments. The factors i nvestigated

PAGE 48

48 were A: percentage of CEC replaced by TCMA, B: amount of modified clay added (mg), and C: ZEN dosage (ppm). A 23 factorial design was used with 3 midpoints. An example of the design is detailed in Table 3 1. 10mL aliquots of the required ZEN so lution was added to a glass vial containing the required amount of modified IMTX and shaken for 30 minutes using a Burrell wrist action shaker. After shaking, 1.5 mL of each solution and the standards were centrifuged at 12000rpm for 15 minutes. The absorb ance of each supernatant was then measured using a Perkin Elmer Lambda 800 UV -Visible spectrometer at 284nm and HPLC FD. All ZEN extraction experiments were carried out at pH3 and pH7 to mimic the pH variation found in swine intestines. A schematic of the procedure for ZEN extraction in vitro can be seen in Figure 3 4. A semiquantitative measurement was taken using a pHydrion QT 10 quaternary ammonium test kit. This kit consists of test strips that indicate the relative concentration of quaternary ammonium compounds by visual comparison of the reaction zone with the fields of color scale in the same manner of a pH strip. The test strips had a measuring range of 0 400ppm with a color grading scale every 100ppm. Table 3 1. 23 factorial design used to determine the most significant factors for ZEN extraction. Factor 1 Factor 2 Factor 3 A: CEC replaced B: Amount of Clay C: ZEN dosage (%) (mg) (ppm) 40 50 8 100 50 8 40 125 8 100 125 8 40 50 20 100 50 20 40 125 20 100 125 20 70 87.5 14 70 87.5 14 70 87.5 14

PAGE 49

49 Figure 3 4. Schematic of the procedure for ZEN extraction. 3.5 TCMA Adsorption/Desorption Studies 3.5.1 TCMA Adsorption Studies To examine the adsorption properties of TCMA on the IMTX clay, TCMA was added to IMTX in amounts equal to three times the CEC of natural IMTX. In this study, 5g of clay was added to 300mL of nanopure water and allowed to swell for three hours at 45oC. TCMA was added to the swollen clay at amounts equal to 40%, 70%, 100%, 150%, 170%, 200%, 250% and 300% of the CEC. Cation exchange continued for 2 hours. After which the agglomerated clays were allowed to settle. The aqueous portion was then centrifuged at 3000rpm for 20 minutes and Add clay to 10 mL ZEN solution Shake 30 mins. using wrist action shaker Allow to settle 15 mins. Centrifuge 30 mins at 12000 rpm Measure extraction using UV Vis spectro metry or HPLC -FD

PAGE 50

50 the supernatant analyzed to determine the amounts of TCMA not adsorbed onto the clay surfaces. For more quantitative measurements, aqueous potentiometric titrations were performed using a Metrohm 726 titroprocessor. Potentiometric titration is a volumetric method that measures the potential between two electrodes (re ference and indicator) as a function of the added reagent volume. The titration curve is sigmoid in shape and the maximum change in slope indicates the equivalence point where the amount of the titrant added is equal to the amount of analyte in the sample. The reference electrode was a silver/silver chloride electrode and the indicator electrode was an ionic surfactant electrode. The titrant was sodium lauryl sulfate (SLS) solution. 3.5.2 TCMA Washing Cycle Desorption Due to concern over desorption of quat ernary ammonium cations and the effects they may have on ZEN adsorption in vivo, steps were taken to investigate the desorption properties of TCMA from IMTX. As previously mentioned, after modification the organophilic clays were washed 3 times to remove a ny sodium chloride present. During the washing cycles the organophilic clays were agitated for 30 minutes using a wrist action shaker and then centrifuged. The supernatant was then tested by both the semiquantitative and potentiometric titration methods to determine the amounts of TCMA desorbed during the washing cycles. 3.5.3 Electrolyte Desorption Studies Studies were performed to investigate the desorption of amines in 0.2M sodium chloride solutions from MMT clays by Zhang and coworkers (Zhang et al., 1993) .. However, these studies did not model the gast rointestinal system and did not investigate the effect pH on desorption. In this study, electrolyte solutions were prepared to simulate gastric electrolyte solutions and duodenal electrolyte solutions. The solutions were prepared according to previous

PAGE 51

51 publ ished works (Avantaggiato et al. 2003) The gastric electrolyte solution contained 3.1 g/L NaCl, 1.1 g/L KCl, 0.15g/L CaCl2, and NaHCO3. The duodenal electrolyte solution contained 5.0g/L NaCl, 0.6 g/L, KCl, and 0.3g/L CaCl2. The suggested inclusion rate of organophilic clay in livestock feed is in the range of 0.05 0.2 wt%. Amounts equal to 0.2 wt% of the feed intake of swine (70kg per week) in the grower stage of development between 64 100 days old was added to 100mL of each electrolyte solution at pH 3 and pH7. The clay-electrolyte solutions were shaken on a rotary shaker for a minimum of 5 hours at a constant temperature at 39oC. The 100% CEC and 40% CEC organophilic clays were chose n for these studies. The electrolyte solutions were then titrated with sodium lauryl sulfate solution to determine any free TCMA molecules. These studies were performed in triplicate.

PAGE 52

52 CHAPTER 4 RESULTS AND DISCUSSI ON 4.1 Characterization of IMTX Underst anding clay properties is essential to the development of organoclays that serve the purpose of extracting the mycoestrogen Zearalenone. Researchers have attributed the extraction of other small water -soluble mycotoxins by clay to the hydrophilic nature an d layered structure of montmorillonite clays. Milwhite, Inc., has done extensive characterization of natural IMTX clay for various research endeavors. These characterization methods included XRD, determination of CEC by methylene blue adsorption, and therm al analysis by Differential Thermal Analysis (DTA) and Thermogravimetric Analysis (TGA). From XRD analysis, they determined IMTX to be comprised of quartz and montmorillonite clay. XRD analysis was performed to confirm the results found by Milwhite. The sc an of IMTX with characteristic peaks for quartz and montmorillonite are shown in Figure 4 1. The intergallery spacing was also determined using XRD and Braggs Law according to Equation 3 1. The intergallery or dspacing was 15.7 or 1.57 nm. This correspo nds to a 2 theta value of 5.6. The XRD scan of IMTX shown in Figure 4 2 with angular range 2o o was used to determine intergallery spacing. This gallery spacing is wide enough to accommodate the aflatoxin molecule but is not wide enough to sequester t he ZEN molecule. Milwhite, Inc., reported the CEC of IMTX to be 56.7 meq/100g of clay. For thorough analysis, the CEC was determined in this work by crystal violet adsorption. The amounts of sodium (Na+), calcium (Ca++), and magnesium (Mg++) ions displaced by CV were determined by ICP analysis. The cation exchange capacity was found to be 70 7 meq/100g. The ion displacement amount was disproportionate and were as follows Ca++ > Na+ > Mg++ from least to most. The total amounts of ions released were similar after 3 and 13 days of incubation. Figure

PAGE 53

53 4 3 shows the total amounts of inorganic cations displaced by CV after 1, 3, and 13 days of incubation. Figure 44 shows adsorption isotherms for CV onto IMTX after 1, 3, and 13 days incubation. CV adsorption isot herms show higher adsorption after 13 days incubation. It has been previously reported that the CEC should be determined by the total ions rather than the amount of dye adsorbed (Rytwo et al., 1991) Increased adsorption above the CEC of the clays is attributed to the CV molecules coating the surfaces of the IMTX at high concentrations over a long period of time (Hang & Brindley, 1970) ICP analysis determined that of the total cation exchange capacity approxi mately 53% is comprised of Ca++ ions, followed by 32% Na+ ions, and the remaining 15% Mg++ ions. 0 100 200 300 400 500 600 700 800 900 1000 2 12 22 32 42 52 62 2-theta CPS M M Q Q M MQ quartz M montmorillonite 0 100 200 300 400 500 600 700 800 900 1000 2 12 22 32 42 52 62 2-theta CPS M M Q Q M MQ quartz M montmorillonite Figure 4 1. XRD scan of IMTX clay with angular range 2o o. Q denotes the characteristic peaks for quartz and M denotes the characteristic peaks for montmorillonite.

PAGE 54

54 0 100 200 300 400 500 600 700 2 3 4 5 6 7 8 9 10 2-theta CPS Figure 4 2. XRD scan of IMTX with angular range of 2o o. The intergallery spacing corresponds to a peak at 5.6 2 4.2 Clay Modifica tion In order to extract Zearalenone it is necessary to first render the clay hydrophobic by cationic exchange of inorganic cations with a surfactant of sufficient chain length and expand the intergalleries of the clay to accommodate the ZEN molecule. The surface polarity of the natural IMTX attracts the surfactant molecules and allows for penetration into the intergallery spacing. The surfactant molecules can assume several conformations depending on the cation exchange capacity and the length of the hydro carbon chain(s) of the modifying species. 4.2.1 FTIR Results In this work, Improved Milbond TX, was modified with tricaprylmethylammonium chloride (TCMA). Modification of the clay was verified by FTIR spectroscopy. Figure 4 5 shows the infrared spectra of TCMA. The peaks at 2923cm -1 and 2853 cm1 are assigned to asymmetric and symmetric stretching C H. The peaks at 1467 cm -1 show CH2 scissoring, CH3

PAGE 55

55 scissoring at 1378 cm -1, and C N stretching at 1043 cm1. The peak assignments are listed in Table 4 1. F igure 4 6 shows the infrared spectra of IMTX and the TCMA IMTX organoclay. The peaks at 3630cm1, 3375cm1 and 1630cm1 are associated with OH stretching and interlayer H2O stretching in IMTX clay. The band at 1045 cm1 belongs to Si O deformation vibrati on. When IMTX is modified with TCMA water is displaced and there is a decrease in the magnitude and broadening of the water related peaks. The TCMA modified clays present peaks at 2928 cm1, 2852 cm1, and 1466 cm1, which belong to C H asymmetric stretchi ng, C -H symmetric stretching and CH2 scissoring respectively. This suggests that the surfactant is tethered to the clay interlayer surfaces. The peak assignments are listed in Table 4 2. 4.2.2. XRD Results XRD was used to determine the intergallery spacin g of the TCMA IMTX clays. The intergallery spacing of the pure IMTX clay is 15.7 When TCMA is added in amounts equal to 40% of the CEC of the clay, the XRD scan shows one distinct peak that corresponds to 15.7 as in the original IMTX and a broader peak at approximately 4 2 This indicates some intercalation of TCMA. These intercalated molecules expand the intergallery spacing while some clay particulates remain unchanged and retain the original gallery spacing. When TCMA was added in amounts equal to or greater than 100% of the CEC the intergallery spacing remains relatively unchanged at ~22. Table 4 3 shows the intergallery spacing of the TCMA -IMTX clays. Due to the insignificant change in the intergallery spacings of the 200%CEC and 300%CEC modified c lays no further studies were performed with clays modified with more than 100% CEC. 4.2.3. TGA Results Thermal analysis provides information on thermal stability as well as phase transformations. The TCMA modified clays exhibit three distinct mass loss regions. Mass loss

PAGE 56

56 below 250oC is characteristic of the evolution of adsorbed water and gaseous species. Between 250oC and 575oC organic substances are degraded. The third region indicates dehydroxylation of the aluminosilicate layers where chemically adso rbed hydroxyl groups are removed. The total mass loss for unmodified IMTX is ~10%. The total mass loss of the 40% CEC TCMA IMTX clays is ~17%, the 70% CEC TCMA IMTX is ~20%, and the 100% CEC TCMA IMTX is ~24%. These values are consistent with the TCMA to c lay ratios added during modification. These results imply that most if not all the TCMA is adsorbed onto the clay surfaces. In the first region the pure IMTX shows the most mass loss due to adsorbed water. This further verifies that the TCMA molecules dis place the adsorbed water within the clay. The modified clays show increased mass loss in the second mass loss region due to adsorbed TCMA. The mass loss is proportional to the amount of TCMA added during modification. 4.2.4. Possible TCMA Conformation Fro m the characterization data obtained, a hypothesis about the possible chain conformations within the aluminosilicate layers can be developed. As previously mentioned, the cation exchange capacity, the intergallery d -spacing, and the hydrocarbon chain lengt h of the alkylammonium is often used to determine the chain conformation. The previous studies on chain conformation have investigated single chain akylammonium cations or dialkylammonium cations. It was determined that short chain ( nc<12) surfactants would yield monolayer conformations resulting in d-spacings of 1.4nm and longer chain ions would yield bilayer conformations with d-spacings of 1.8 nm. The pseudotrimolecular conformation yields a basal spacing of ~2.2nm. In this study the surfactant used, TC MA, is a trialkylammonium cation with chain lengths containing 8 10 carbon atoms. The maximum achieved d-spacing is ~2.3 nm. As per Figure 2 3, a chain length of 10 should at most yield a bilayer arrangement and intergallery spacing of 1.8nm. However, the three hydrocarbon chains of TCMA allow for greater expansion

PAGE 57

57 of the layers to yield a basal spacing indicative of the pseudotrimolecular conformation. Therefore, it can be assumed that the conformation lies somewhere between the two and more likely resembl es a pseudotrimolecular conformation see in Figure 4 8. 4.3 ZEN Extraction by TCMA -IMTX In order to optimize the ZEN extraction experimental procedures, Statistical Design of Experiments is used to increase efficiency, decrease cost and time, and to simul taneously investigate some of the factors involved in the extraction process. This investigation reveals the parameters most involved in the extraction of ZEN and any interactions that may exist between them. In this study, factor A represents the amount o f surfactant added during the clay modification process (CEC replaced), factor B the amount of clay used for extraction, and factor C the initial ZEN dosage. The design chosen was a 23 factorial design with 3 midpoints. This design involves 3 parameters wi th high (+1) and low levels ( 1). The design runs encompass the various combinations that can arise between the 3 factors at both levels. These levels meet at a midpoint which is reproduced 3 times for statistical significance. The geometric view of a 23 f actorial design can be seen in Figure 4 9 and the experimental combinations are shown in Table 4 4. One of the basic principles of statistical design is randomization. This means that the individual runs of the experiment are randomly determined. For this work the randomization of the experimental runs was performed using Stateases Design Expert 7.1.3. For the experiments in which UV -Vis spectroscopy was used to determine ZEN extraction the amount of IMTX and ZEN dosage was increased to accommodate the l imitations of the instrument to detect ZEN at minute amounts.

PAGE 58

58 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 0 0.2 0.4 0.6 0.8 1 1.2 1.4 CV added (meq/g) Total Ions desorbed (meq/g) day 13 day 3 day 1 Figure 4 3. Desorption of inorganic cations from IMTX after 1, 3, and 13 days of incubation. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 crystal violet added (meq/g) crystal violet adsorbed (meq/g) Day1 Day3 Day 13 Figure 4 4. CV adsorption by IMTX after 1, 3, and 13 days incubation.

PAGE 59

59 Table 4 1. FTIR Band Ass ignments of TCMA Frequency (cm1) Tentative Assignment 2923 C H asymmetric stretching 2853 C H symmetric stretching 1467 CH2 scissoring 1378 CH3 bending 1043 C N stretching Table 4 2. FTIR Band assignments of IMTX and Modified IMTX Frequency (cm1) Tentative Assignment 3630, 3375 OH stretching, interlayer H2O stretching 2928 C H asymmetric stretching 2852 C H symmetric stretching 1630 Interlayer H2O deformation 1466 CH2 scissoring 1045 Si O stretching Table 4 3. Intergallery spacing of IM TX modified with varying amounts of TCMA. Surfactant Added (% CEC) Intergallery D Spacing () 0 15.7 40 15.7 70 16 100 21.4 200 22.9 4.3.1 Extraction Results by UV Vis Spectroscopy The experimental trials were performed as displayed in Table 4 5. S eparate experiments were performed for extraction at pH 7 and at pH 3. Figure 4 10 shows the calibration curves used to determine ZEN extraction. At 70% and 100% CEC, ZEN removal is greater than 90% for both pH values and is independent of ZEN dosage and t he amount of organophilic clay added. At

PAGE 60

60 722.66 1043.43 1378.20 1467.42 2853.94 2872.33 2923. 2956.11 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50Abs 500 1000 1500 2000 2500 3000 3500 4000 cm-1 cm1 3500 2500 1500 500 0.00 0.50 Absorbance 722.66 1043.43 1378.20 1467.42 2853.94 2872.33 2923. 2956.11 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50Abs 500 1000 1500 2000 2500 3000 3500 4000 cm-1 cm1 3500 2500 1500 500 0.00 0.50 Absorbance Figure 4 5. FTIR analysis of TCMA.

PAGE 61

61 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 KM 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers (cm-1) 1626.25 1869.01 3370.11 3628.22 1466.29 1483.18 2852.92 2928.05 3628.69 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 KM 1500 2000 2500 3000 3500 Wavenumbers (cm-1) 4000 500 0.0 1.1 wavenumber (cm1) KM 3500 1500 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 KM 500 1000 1500 2000 2500 3000 3500 4000 Wavenumbers (cm-1) 1626.25 1869.01 3370.11 3628.22 1466.29 1483.18 2852.92 2928.05 3628.69 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 KM 1500 2000 2500 3000 3500 Wavenumbers (cm-1) 4000 500 0.0 1.1 wavenumber (cm1) KM 3500 1500 IMTX TCMA IMTX Figure 4 6. FTIR analysis of IMTX and IMTX modified with TCMA. Inset shows magnified view of the scan from wavenumber 4000cm1 to 1200 cm1 IMTX TCMA IMTX

PAGE 62

62 Figure 4 7. XRD scans of IMTX and TCMA IMTX clays with angular range 2o o. 2 3 4 5 6 7 8 9 10 2-theta CPS 50% CEC Milbond 20% CEC 50% CEC TCMA 20% CEC TCMA Milbond 2 3 4 5 6 7 8 9 10 2-theta CPS 50% CEC Milbond 20% CEC 50% CEC TCMA 20% CEC TCMA Milbond 40% CEC TCMA 100% CEC TCMA IMTX Mass loss (%)

PAGE 63

63 Figure 4 7. TGA curves of IMTX modified with varying amounts of TCMA. 75 80 85 90 95 100 25 125 225 325 425 525 625 725 Temperature (C) IMTX 40%CEC 70%CEC 100%CEC Physically adsorbed water loss Organic species degradation Dehydroxylation of clay layers 75 80 85 90 95 100 25 125 225 325 425 525 625 725 Temperature (C) IMTX 40%CEC 70%CEC 100%CEC Physically adsorbed water loss Organic species degradation Dehydroxylation of clay layers Temperature (C) Mass loss (%)

PAGE 64

64 Figure 4 8. Pseudotrimolecular conformation of alkylammonium ions (de Paiva et al., 2008) Figure 4 9. Geometric view of the 23 factorial design. X 1 1 +1 X 2 1 +1 X 3 +1 1

PAGE 65

65 Table 4 4. 23 factorial design matrix. Run X1 X2 X3 1 1 1 1 2 1 1 1 3 1 1 1 4 1 1 1 5 1 1 1 6 1 1 1 7 1 1 1 8 1 1 1 9 0 0 0 10 0 0 0 11 0 0 0 40% CEC it is necessary to add greater amounts of clay to remove more than 90% ZEN. The lowest values of ZEN extraction occur when the amount of TCMA and clay added is the lowest and the ZEN dosage is highest. In most cases, ZEN extraction is increased at pH 3 w hich is consistent with previous results by Milwhite, Inc. ZEN extraction at pH 3: Three dimensional curves, projected two dimensional curves, and interaction curves were constructed to investigate ZEN extraction as a function of two parameters. The third parameter is held constant. From these curves it is possible to view the effect changing the two parameters might have on ZEN removal percentages. Figure 4 11 shows the 3D plot of ZEN extraction at pH 3 as a function of the amount of clay used for extracti on (factor B) and the TCMA modification amount (factor A). This curve shows that ZEN extraction is at its highest when the high values of both parameters are used and the lowest when the low values are used. It also shows that an increase in extraction can be achieved by increasing either parameter. However, the most dramatic increase occurs when factor A is increased. From this

PAGE 66

66 figure an estimate can be made on values of ZEN extraction that can be expected within the range of the high and low values of the factors examined in this study. Figure 4 12a shows the 2D contour lines of constant response (ZEN removal %) in the x1 (CEC replaced), x2 plane (amount of clay). The curvature of the contour lines indicates interactions between x1 and x2. This interactio n is confirmed in Figure 4 12b which illustrates the response against factor A (% CEC replaced) for both levels of factor B (amount of clay added). The crossing of the factor B lines is indicative of an interaction between factors A and B. The contour plot shows that when 70%100% of the CEC is replaced with TCMA, greater than 90% of ZEN is removed irrespective of the amount of clay added. However, as the amount of TCMA added during modification decreases the amount of clay needed to achieve >90% extraction must increase. This interaction is called an antagonistic interaction. At 40% CEC replaced and the high value for factor C (125mg), the amount of ZEN extracted is notably higher than at the low value of factor C (50mg). As the CEC replaced increases, the two lines converge and ZEN extraction is statistically the same. When the 3D response surface is considered as a function of factor C and factor A, the curve becomes somewhat flat indicating that there is less interaction between these two factors. The 3 D surface is shown in Figure 413. For this curve and the curves shown in Figure 4 14, the amount of clay added is held constant at 87.5 mg. In Figure 4 14a, the contours are plotted as a function of factors A and C. Again it is apparent that ZEN extractio n is >90% when the CEC replaced is greater than 70%. However, as the ZEN dosage increases, the extraction decreases at low TCMA modification amounts. The interaction plot in Figure 4 14b shows that ZEN extraction is lower when the initial dosage is at the high value of 20ppm and the CEC replaced is low. As expected, when the CEC is increased the two lines reach approximately the

PAGE 67

67 same extraction value. Although in this figure the lines representing the ZEN dosage do not cross, they are not exactly parallel i ndicating there is some interaction between the factors but not at this clay concentration of 87.5mg. Figure 4 15 shows interaction plots of ZEN extraction when the amount of clay added ranges from 50mg to 125mg. The interaction between factors A and C app ears when the amount of clay added decreases as shown in Figure 415a. This interaction can be considered negligible when the actual extraction values are considered. Extraction when factor A = 40% CEC, factor B = 50 mg, and factor C = 8ppm is ~87% and rem ains relatively unchanged at ~85% when factor C = 20ppm. At 125mg of clay added the two lines representing the high and low values of ZEN dosage approaches parallel which would indicate no interaction. Analysis of variance or ANOVA was used to determine wh ich parameters were most significant for ZEN extraction by testing for significant differences between means. Although researchers have studied many of the ZEN extraction parameters individually, DOE has allowed for the simultaneous investigation of severa l parameters. ANOVA analysis determined that of the three parameters investigated, the amount of clay added and the %CEC replaced by TCMA are significant to ZEN extraction. However, the %CEC replaced is the most significant factor. In addition, there is an antagonistic interaction between the two significant parameters. Therefore, as the %CEC replaced increases the amount of clay needed to extract ZEN decreases. The same is true in the reverse, as the %CEC replaced decreases; the amount of clay needed to ex tract a given amount of ZEN has to be increased. These relationships are represented by a mathematical model equation seen in Equation 4 1 where A represents the %CEC replaced with TCMA, B the amount of modified clay added for extraction, and C initial the ZEN dosage. The mathematical model describes the relative influence each parameter has on ZEN extraction efficiency. The

PAGE 68

68 magnitude of the coefficients indicates that the %CEC has the most dramatic influence on ZEN extraction and there are some slight inte ractions between the factors. The negative coefficients in the 4th and 6th terms imply an antagonistic relationship. This means a decrease in one term would require an increase in the other to achieve a particular response. The full ANOVA analysis can be f ound in Appendix A. ZEN removal = 74+ 0.46( A) + 0.18( B) 0.26( C) 2.95E 3( A*B)+9.16E 3( A*C)2.92E 3( B*C) (4 1) ZEN extraction at pH 7: At pH 7, the 3D curve shows the same trends as is seen at pH 3 when the response is plotted as a function of factors A and B. This can be seen in Figure 4 15. Figures 4 16a -b show the contour plot with curved lines and the interaction plot indicating an interaction between factors A and B. Here also extraction is >90% when TCMA replaced is greater than 70% and the factor B lines intersect at 100% CEC replaced by TCMA. When factor C (ZEN dosage) is considered against either factor A or factor B, the 3D response curve becomes flat. The curve of ZEN removal% as a function of ZEN dosage and the CEC replaced is shown in Figu re 4 17. The amount of clay added is kept constant at 87.5 mg. Again, it is apparent that ZEN extraction is greatest with the 100% CEC clays and CEC replacement greater than 70% will yield extraction amounts >90%. Also, at any given ZEN dosage, as you incr ease the amount of TCMA added for modification, the ZEN extraction efficiency increases. The contour plots in Figure 4 18a -b exhibit parallel straight lines which indicate no interaction between ZEN dosage and the other two parameters. Equation 4 2 shows t he mathematical model representing ZEN removal % at pH 7 under the conditions presented in this study. Unlike equation 4 1, this model does not include a C parameter because it was deemed insignificant to ZEN extraction. The negative coefficient in the

PAGE 69

69 thi rd term relays the antagonistic interaction between parameters A and B. The full ANOVA analysis can be found in Appendix B. 4.3.2. IMTX -TCMA -corn oil Extraction In attempts to lower cost for industrial applications, the 40% CEC clays are further modified with store brand corn oil at 3wt% and 10wt% of the clay. Previous work done by Milwhite, Inc. indicated that it would be necessary to modify IMTX clay with ~40% by weight corn oil in order to extract >90% ZEN. In this study, it i s hypothesized that the surfactant molecules will help facilitate the adsorption of the fatty acid molecules within the clay galleries by ion exchange. With increased organic content within the clay layers an expansion in the d spacing is expected. After modification with TCMA and corn oil, the clays are characterized using XRD and TGA. The intergallery spacing was determined using Braggs Law as in previous cases. The XRD scan of the IMTX TCMA -corn oil modified clays is shown in Figure 419. There is no statistical difference in the clay expansion between the original 40% CEC clays and those modified with corn oil. There is some increase in the peak width toward lower angles which could indicate some increase in the d -spacing of some of the clay particula tes but there is no indication of an overall expansion. Thermogravimetric analysis also indicates no statistical difference in mass loss. The characteristic three regions of mass loss are present; the first indicating the evolution of adsorbed water, the s econd the degradation of organic species, and finally the dehydroxylation of the aluminosilicate layers. Figure 4 20 shows the TGA curves for these modified clays. ZEN removal% = 42.91 + 1.12 ( A ) + 0.33 ( B ) 6.73E 3 ( A*B ) (4 2)

PAGE 70

70 y = 0.0323x R2 = 0.9969 y = 0.033x R2 = 0.9993 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 5 10 15 20 ZEN standards (ppm) absorbance (a.u.) pH 7 pH 3 Linear (pH 7) Linear (pH 3) Figure 4 10. ZEN calibration curves used to determine ZEN extraction. Table 4 5. ZEN experimental runs and extraction results by UV -Vis Spectroscopy. A:CEC replaced B:amount of clay C:ZEN dosage ZEN removal at pH 3 ZEN removal at pH 7 (%) (mg) (ppm) (%) (%) 40 50 8 86.86 77.59 40 125 8 95.30 91.29 40 50 20 89.05 72.85 40 125 20 85.32 88.60 70 87.5 14 95.37 96.68* 70 87.5 14 96.93* 96.48* 70 87.5 14 95.56 94.06 100 50 8 98.29* 98.77* 100 125 8 99.49* 100.00* 100 50 20 99.37* 98.05* 100 125 20 99.26* 97.70* std dev: 1.34 1.92 *Values calculated from th e linear regression are below the detection limit of the instrument

PAGE 71

71 Figure 4 11. Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 3. The ZEN dosage is held constant at 14ppm. 20.00 27.50 35.00 42.50 50.00 50.00 68.75 87.50 106.25 125.00 86 89.5 93 96.5 100 ZEN removal A: cec replaced B: amount of clay 50 40 100 12 5 86 100 pH 3 47.5 55 62. 5

PAGE 72

72 Figure 4 12. Model graphs showing a) contour plot of ZEN removal % at pH 7as a function of factors A and B with factor C (ZEN dosage) held constant at 14ppm b) interaction plot of the response data against factor A (%CEC replaced) for both levels of factor B (amount of clay added) A: CEC replaced 20.00 27.50 35.00 42.50 50.00 50.00 68.75 87.50 106.25 125.00 ZEN removal B: amount of clay 88.4 90.6 92.8 95.0 97.2 3 97.2 95.0 92.8 90.6 88.4 40 100 50 125 (a) B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction ZEN removal 84 88.25 92.5 96.75 101 A: CEC replaced 40 100 84 101 (b) st. dev: 1.01 125 mg clay 50 mg clay pH 3 47.50 55.00 62.50 47.50 55.00 62.50

PAGE 73

73 Figure 4 13. Response surface 3D plot of ZEN removal (%) as a function factors A and C. 20.00 27.50 35.00 42.50 50.00 8.00 11.00 14.00 17.00 20.00 87.0 90.3 93.5 96.8 100.0 ZEN removal A: cec replaced C: zen dosage 8 20 50 20 87 100 pH 3 20.00 27.50 35.00 42.50 50.00 8.00 11.00 14.00 17.00 20.00 87.0 90.3 93.5 96.8 100.0 ZEN removal A: cec replaced C: zen dosage 8 20 50 20 87 100 pH 3 40 100 47.5 55 62.5

PAGE 74

74 Figur e 4 14. Model graphs showing a) contour plot of ZEN removal % at pH 3as a function of factors A and C with factor B (amount of clay added) held constant at 87.5 mg and b) interaction plot of the response data against factor A (%CEC replaced) for both level s of factor C (ZEN dosage) 20.00 27.50 35.00 42.50 50.00 8.00 11.00 14.00 17.00 20.00 ZEN removal A: cec replaced C: zen dosage 89.3 91.3 93.4 95.4 97.4 3 20 8 50 20 89.1 91.3 93.4 95.4 97.4 20.00 27.50 35.00 42.50 50.00 8.00 11.00 14.00 17.00 20.00 ZEN removal A: cec replaced C: zen dosage 89.3 91.3 93.4 95.4 97.4 3 20 8 50 20 89.1 91.3 93.4 95.4 97.4 C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced 85.0 89.0 93.0 97.0 101.0 20 50 85 10120ppm ZEN 8ppm ZEN C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced 85.0 89.0 93.0 97.0 101.0 20 50 85 10120ppm ZEN 8ppm ZEN ZEN removal st. dev: 1.01 (b) (a) 40 100 47.50 55.00 62.50 47.50 55.00 62.50 40 100

PAGE 75

75 (a) (b) Figure 4 14. Interaction plots showing the ZEN removal % against the % CEC replaced at high and low ZEN dosages when the amount of clay added is a) 50mg and b) 125mg. C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 82.0 87.0 92.0 97.0 102.0 20 50 8220ppm ZEN 8ppm ZEN 10250 mg clay added C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 82.0 87.0 92.0 97.0 102.0 20 50 8220ppm ZEN 8ppm ZEN 10250 mg clay added C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 85.0 89.3 93.5 97.8 102.0 20 50 8520ppm ZEN 8ppm ZEN 102125 mg clay added C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 85.0 89.3 93.5 97.8 102.0 20 50 8520ppm ZEN 8ppm ZEN 102125 mg clay added 47.50 55.00 62.50 40 100 47.50 55.00 62.50 40 100

PAGE 76

76 Figure 4 15. Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 7. The ZEN dosage is held constant at 14ppm 20 27.5 35 42.5 50 50 68.75 87.5 106.25 125 75.00 81.00 87.00 93.00 99.00 ZEN removal A: cec replaced B: amount of clay 50 40 100 125 75 99 pH 7 47.5 55 62.5

PAGE 77

77 Figure 4 16. Model graphs showing a) 2D contour plot of ZEN removal % at pH 7as a function of factors A and B with factor C (ZEN dosage) held constant at 14ppm and b) interaction plot of the response data against factor A (%CEC replaced) for both levels of factor B (amount of clay added) 20.00 27.50 35.00 42.50 50.00 50.00 68.75 87.50 106.25 125.00 A: cec replaced B: amount of clay 79.2 83.1 87.0 91.0 94.9 3 40 100 50 125 79.2 83.1 87.0 91.0 94.9 (a) B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 72.00 79.50 87.00 94.50 102.00 20 50 72 102 B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 72.00 79.50 87.00 94.50 102.00 20 50 72 102 (b) pH 7 125 mg clay 50 mg clay st. dev: 1.92 47.50 55.00 62.50 40 100 47.50 55.00 62.50

PAGE 78

78 Figure 4 17. Response surface 3D plot of ZEN removal (%) as a function factors A and C. 20 27.5 35 42.5 50 8.00 11.00 14.00 17.00 20.00 82.00 86.25 90.50 94.75 99.00 ZEN removal A: cec replaced C: zen dosage pH 7 40 100 47.5 55 62.5

PAGE 79

79 Figure 4 18. 2D contour plots of ZEN removal % at pH 7 as a function of a) factor s A and C and b) factors B and C. The amount of ZEN added is held constant at 87.5 mg. 20.00 27.50 35.00 42.50 50.00 8.00 11.00 14.00 17.00 20.00 ZEN removal A: cec replaced C: zen dosage 85.3 87.9 90.6 93.3 96.0 3 pH 7 20 50 8 20 96.0 93.3 90.6 87.9 85.3 C: zen dosage 50.00 68.75 87.50 106.25 125.00 8.00 11.00 14.00 17.00 20.00 ZEN removal B: amount of clay 88.23 89.42 90.61 91.80 92.99 3 88.2 89.4 90.6 91.8 93.0 50 125 8 20 50.00 68.75 87.50 106.25 125.00 8.00 11.00 14.00 17.00 20.00 ZEN removal B: amount of clay 88.23 89.42 90.61 91.80 92.99 3 88.2 89.4 90.6 91.8 93.0 50 125 8 20 st. dev: 1.92 st. dev : 1.92 (a) (b) 47.50 55.00 62.50 40 100

PAGE 80

80 At pH 7 and 125mg, the IMTX TCMA -corn oil clays extracts roughly all the ZEN present. This is as expected since previous studies with TCMA alone added at 125mg remov ed upwards of 90% of the zearalenone added. The experimental trials and extraction results for the IMTX TCMA -corn oil experimental runs are shown in Table 47. At pH 7, when 50mg of modified clay is added, >93% toxin is removed with the 3wt% and 10wt% corn oil modified clays. At pH 3, extraction is highest at 3wt% corn oil when 10ppm ZEN is added but decreases to <90% at higher corn oil concentrations. ZEN extraction results by the modified clays are shown graphically in Figure 4 21. There are two possible explanations for the decrease in extraction at higher corn oil concentrations at pH 3. The first explanation was given by Lemke and coworkers regarding the need for negative charges for the bonding of the phenolic anion form of ZEN. They hypothesize that a t low pH, the anionic form of the ZEN molecule is unable to bond to excess OH edge site groups through ion dipole interaction (Lemke et al., 1998) The second is a matter of oversaturation of the clays with organic species. The TCMA within the clay layer s will attract any hydrophobic species and can not distinguish between ZEN and corn oil. The ZEN extraction studies imply that there is a maximum amount of ZEN that can be adsorbed at a certain CEC replacement value. The results of these studies imply that there is a maximum amount of hydrophobic species that can be adsorbed by the organophilic clays and although the ZEN extraction is increased by the presence of the corn oil. The adsorption plateaus and the ZEN and corn oil molecules have a somewhat antago nistic relationship. Further DOE studies would be needed to confirm this hypothesis.

PAGE 81

81 4.3.3. Extraction Results by HPLC -FD Due to the limitation of UV -Visible spectroscopy to accurately detect ZEN below 1ppm, the DOE trials were repeated and the ZEN extr action removal percentage was determined using HPLC with fluorescence detection (HPLC FD). Although, the parameters that were investigated remained unchanged the high and low values were adjusted to better represent real -world values. Due to the sensitivit y of the instrument, calibration curves could be determined to 0.01ppm. The calibrations curves for ZEN extraction with HPLC FD are shown in Figure 4 22. The experimental runs and extraction results can be seen in Table 4 7. Table 4 6. Intergallery spacin g of IMTX TCMA -corn oil modified clays. Sample d spacing () IMTX 15.7 40% CEC 15.7 IMTX 40% CEC 3wt% corn oil 15.4 IMTX 40% CEC 10wt% corn oil 16.4 As with the UV-Vis spectroscopy results, the 100% CEC and 70% CEC clays show the best (>95%) ZEN ext raction efficiency. However, when 25mg of 100% CEC clay is used to extract ZEN at 14ppm, the ZEN removal percentage drops to below 90% at pH 7. Under these conditions, the amount of clay added becomes a significant factor. The 40% CEC clays show the least ZEN removal efficiency especially when the ZEN added is 14ppm or the amount of clay added is 25mg. In addition, the effect of pH is more apparent. Lemke and coworkers reported markedly decreased ZEN adsorption at acidic pH on montmorillonite clay modified with various ammonium and pyridinium -based organic cations. This decline in adsorption was attributed to a need for some negative charge for site specific bonding of the phenolic anion associated with ZEN (Lemke et al ,. 1998)

PAGE 82

82 Figure 4 19. XRD scans of IMTX, 40% CEC IMTX, and the IMTX TCMA Corn oil samples with an angular range of with angular range 2o IMTX 2 3 4 5 6 7 8 9 10 2-theta CPS IMTX 20% CEC 3 wt% corn oil 10 wt% corn oil 40% CEC IMTX

PAGE 83

83 Figure 4 20. TGA curves for IMTX modified with 40% CEC TCMA and corn oil. Table 4 7. IMTX TCMA -corn oil experimenta l runs and extraction results by UV -Vis Spectroscopy amount of clay pH ZEN dosage ZEN removal (wt%) (mg) (ppm) (%) 3 50 7 8 93.57 10 50 7 8 *95.79 3 50 7 10 93.48 10 50 7 10 93.93 3 125 7 10 *99.88 10 125 7 10 *99.93 3 50 3 10 94.30 10 50 3 10 82.94 3 125 3 10 *100 10 125 3 10 92.79 *Values calculated from the linear regression are below the detection limit of the instrument 25 225 425 625 825 Temperature (C) 10% corn oil [mg] 3% corn oil [mg] 20% CEC Mass Loss (%) 40% CEC

PAGE 84

84 Figure 4 21. ZEN extraction by IMTX TCMA -corn oil modified clays. ZEN extraction at pH 3: Figure 4 23 shows the 3D response curve for ZEN removal at pH 3 as a function of the %CEC replaced by TCMA (factor A) and the amount of clay added (factor B). The results are similar to those achieved with UV -Vis spectroscopy. At low values of factor A and factor B the ZEN removal percentage is at its lowest between 67% 88% depending on the initial ZEN dosage. Figure 4 24a shows the 2D contour lines of Figure 4 23. The curved lines indicate an interaction between the factors. The sensitivity of this ins trument yields response values with a lower standard deviation of 0.07% for this test. The interaction plot shown in Figure 4 24b further implies interaction. The lines representing the ZEN removal at high and low values of factor B are not parallel. It is apparent that when 125mg of clay is added, ZEN extraction is >90% at any value of factor A. When 25mg of clay is added the CEC must 93.5 93.9 94.3 82.9 99.9 99.9 100.0 92.8 0.0 20.0 40.0 60.0 80.0 100.0 3wt% pH7 10wt% pH7 3wt% pH3 10wt% pH3 Corn Oil ZEN Removal (%) 50mg 125mg st. dev: 1.4

PAGE 85

85 increase in order to accomplish higher extraction amounts. Although there is indications of interactions, within the range of 2 14ppm ZEN dosage, the factor B lines do not converge as can be seen in Figure 4 25a -b. At both 2ppm and 14 ppm ZEN the 100% CEC lines indicate extraction >90%. However, at 40% CEC the extraction falls below 90% and is dramatically less at 14ppm ZEN dosage. These plots indicate only a slight interaction which is represented in the mathematical model by a very small coefficient relating these two terms. When factor C (ZEN dosage) is considered, interactions between this factor and factors A and B are se en. The 3D response surface shown in Figure 426 displays the ZEN removal percentage as a function of factors A and C. ZEN extraction is lowest at low A and high C values. The inverse is true for the greatest ZEN extraction percentage; the C value is low a nd the A value is high. According to Figure 4 27a, ZEN extraction is highest with the 70% CEC and 100% CEC at >90% for any ZEN dosage. However, the 40% CEC clays extract less ZEN when higher concentrations are added initially. Interaction between factors A and C is also suggested by Figure 4 27b. The lines representing ZEN dosage will meet when the amount of clay is increased. When any of the organophilic clays are added at an inclusion of 125mg, the extraction percentage is >90%. This is also shown in Figu re 4 28 which shows the 3D response surface for ZEN extraction as a function of factors B and C. However, as ZEN dosage increases and the amount of clay decreases, the ZEN extraction decreases. Figures 4 29a -b show the interaction between these two paramet ers. When the amount of clay increases the two factor C lines converge. An interaction is also apparent between all three factors. An increase in factors A and B result in higher ZEN extraction. However factor C has a deleterious effect on the overall extr action percentage as it increases. The mathematical model in Equation 4 3 shows how each parameter affects the ZEN extraction percentage. Factor C is by far the most significant term for

PAGE 86

86 ZEN extraction in this study. The comparatively large negative coeffi cient verifies the reduction in extraction caused by increased amounts of ZEN contamination. The interactions are all slight and the interaction between the three terms is even smaller. The full ANOVA analysis can be found in Appendix C. ZEN removal % =+83 .87+0.28 (A) + 0.11 (B) 3.34 (C) 2.07E 003 (A* B)+0.06 (A*C)+0.02 (B*C)4.15 004 (A*B*C) (4 3) ZEN extraction at pH 7: At pH 7, the trends are similar to those at pH 3. Figures 3031 show the characteristic 3D response curve for ZEN removal as a function of factors A and B, the contour lines, and the interaction plot. Again it is apparent that 125mg of any of the modified clays will extract >90% of the initial ZEN dosage. The same is true for the 100% CEC clays at any inclusion. Furthermore, the 3D response curve shown in Figure 4 30 appears almost flat. This is an indication of only a slight interaction between the factors. The interaction plot shows lines that are almost parallel as shown in Figure 31b. Therefore it can be concl uded that the interaction between factors A and B is very slight and the interaction lines do not intersect within the parameters of this experiment. The 3D response curve for factors A and C are also somewhat flat as shown in Figure 32. Figures 32 33 indi cate high ZEN extraction at low ZEN dosages and high clay concentration. As one factor decreases, the other must increase to achieve a desired result. However, the ZEN dosage has a greater impact on extraction. This will become apparent in Equation 4 4. W hen the CEC replaced by TCMA is increased to 100%, the interaction lines do not meet. However, the gap between the two narrow and the 14ppm line approaches the 2ppm line. The interaction plot for ZEN removal against factor B at high and low values of ZEN dosage is shown in Figure 4 33b. Factors B and C also show signs of interactions. The 3D response curve

PAGE 87

87 does not appear flat in Figure 4 34. In addition, the contour lines are curved. Low doses of ZEN are easily extracted by all clays while greater amounts of modified clay accomplish higher ZEN removal percentages as can be seen in Figure 35a. The interaction plot in Figure 4 35b does not interact. However, at higher TCMA replacement percentages the interaction lines approach one another at higher amounts of clay added. Equation 4 4 relates all the factors examined in this study and the interactions between them. ANOVA analysis has determined that for this study, the factor that has the most impact on ZEN extraction is factor C the initial dosage of ZEN, foll owed by the % CEC replaced by TCMA and finally the amount of clay added. The two factor interactions are all slight. However, the interaction between factors A and C is the greatest. Although there is a term for interactions between all three terms the mag nitude is so small that it can be deemed insignificant. The full ANOVA analysis can be found in Appendix D. ZEN removal = 92.73+0.14 ( A)+0.05( B)2.01( C)1.12E 3( A*B)+0.02( A*C)+8.00E 3( B*C)+8.88E 6( A*B*C) (4 4) The results of this study and this equation a lso suggest that for every combination of factors A and B there is some maximum amount of ZEN that can be extracted. They also show the importance of choosing parameters and levels specific to the application of the system being evaluated. Although the mod el can be used to predict possible outcomes, any dramatic changes in the levels of the parameters or the experimental methods can lead to unexpected outcomes. Therefore, it is necessary to carefully choose the model, parameters, and levels of an experiment 4.4 TCMA Adsorption/Desorption Results During this study many questions arose about TCMA adsorption and desorption onto IMTX clays. It was important to determine if there was total adsorption of TCMA during modification. Excess TCMA can increase overall cost and free surfactant molecules can present waste water toxicity issues. Also, there has been some research geared towards the recycling of

PAGE 88

88 y = 14.437x 0.7101 R2 = 0.9977 y = 17.936x 0.2013 R2 = 0.9999 0 5 10 15 20 25 30 35 40 0 0.5 1 1.5 2 ZEN concentration (ppm) Area pH 3 pH 7 Linear (pH 7) Linear (pH 3) Figure 4 22. Calibration curves for ZEN extraction by HPLC -FD at pH 7 and pH 3. Table 4 7. ZEN experimental run s and extraction results by HPLC FD A:CEC replaced B:amount of clay C:ZEN dosage ZEN removal at pH 7 ZEN removal at pH 3 (%) (mg) (ppm) (%) (%) 40 25 2 93.30 87.73 100 25 2 97.70 97.68 40 125 2 97.67 97.57 100 125 2 98.75 98.83 40 25 14 75.52 66.95 100 25 14 85.77 95.74 40 125 14 89.70 93.87 100 125 14 96.95 99.00 70 75 8 96.23 97.98 70 75 8 96.30 98.12 70 75 8 96.30 98.05 std dev: 0.043 0.070

PAGE 89

89 Figure 4 23. Response surface 3D plot of ZEN removal (%) as a function of t he amount of modified clay added (mg) for extraction and the CEC replaced by TCMA (%) at pH 3. The ZEN dosage is held constant at 8ppm 20 27.5 35 42.5 50 25 50 75 100 125 77.00 82.50 88.00 93.50 99.00 ZEN removal A: cec replaced B: amount of clay pH 3 40 47.5 55 62.5 100

PAGE 90

90 Figure 4 24. Model graps showing a) 2D contour lines of ZEN extraction at pH 3 as a function of factor B and factor A and b) interaction plot of ZEN removal against factor A. at high and low values of factor B. The ZEN dosage is held constant at 8ppm for both plots. 20 27.5 35 42.5 50 25.00 50.00 75.00 100.00 125.00 ZEN removal A: cec replaced B: amount of 80.93 84.53 88.13 91.72 95.32 3 20 50 A: CEC replaced B: Amoun t of IMTXdded (mg 25 125 80.93 84.53 88.13 91.72 95.32 pH 3 (a) B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction 66.00 74.50 83.00 91.50 100.00 3 3 3 20 50 66 100 B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction 66.00 74.50 83.00 91.50 100.00 3 3 3 20 50 66 100 A: CEC replaced ZEN removal (b) B: Amount of clay 25 mg clay 125 mg clay st. dev: 0.07 40 100 47.5 55 62.5 40 100 47.5 55 62.5

PAGE 91

91 Figure 4 25. Interaction plots for ZEN rem oval against factor B at high and low values of factor A when Zen dosage is a) 2ppm and b) 14 ppm A: cec replaced 25.00 50.00 75.00 100.00 125.00 Interaction B: amount of clay 66.00 74.50 83.00 91.50 100.00 25 125 66 100 A: cec replaced 25.00 50.00 75.00 100.00 125.00 Interaction B: amount of clay 66.00 74.50 83.00 91.50 100.00 25 125 66 100 ZEN removal ZEN removal 40% CEC 100CEC 100CEC 40% CEC 2ppm ZEN 14ppm ZEN (a) (b)

PAGE 92

92 Figure 4 26. Response surface 3D plot of ZEN removal (%) as a function the CEC replaced by TCMA (%) and the initial ZEN dosage at pH 3 The amount of modified clay added is kept constant at 75mg. 20.00 27.50 35.00 42.50 50.00 2.00 5.00 8.00 11.00 14.00 66.00 74.50 83.00 91.50 100.00 ZEN removal A: cec replaced C: zen dosage 14 2 20 50 66 100 20.00 27.50 35.00 42.50 50.00 2.00 5.00 8.00 11.00 14.00 66.00 74.50 83.00 91.50 100.00 ZEN removal A: cec replaced C: zen dosage 14 2 20 50 66 100 pH 3 40 47.5 55 62.5 100

PAGE 93

93 Figure 4 27. Model graphs showing a) 2D contour lines of ZEN extraction at pH 3 as a function of factor C and factor A and b) interaction plot of ZEN removal against f actor A. at high and low values of factor C. The amount of clay added is held constant at 75mg for both plots. 20 27.5 35 42.5 50 2.00 5.00 8.00 11.00 14.00 ZEN removal A: cec replaced 83.38 86.36 89.33 92.30 95.28 3 20 50 A: CEC replaced 2 14 pH 3 83.38 86.36 89.33 92.30 95.28 (b ) C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced 66.00 74.50 83.00 91.50 100.00 3 3 3 20 50 66 100 C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced 66.00 74.50 83.00 91.50 100.00 3 3 3 20 50 66 100 C:ZEN concentration ZEN removal 14 ppm ZEN 8 pp m ZEN (a) (b) st. dev: 0.07

PAGE 94

94 Figure 4 28. Response surface 3D plot of ZEN removal (%) as a function the initial ZEN dosage (ppm) and the amount of clay added (mg) at pH 3. The %CEC replaced by TCMA is held constant at 70%. 25.00 50.00 75.00 100.00 125.00 2.00 5.00 8.00 11.00 14.00 66.00 74.50 83.00 91.50 100.00 ZEN removal B: amount of clay C: zen dosage 14 2 25 125 66 100 25.00 50.00 75.00 100.00 125.00 2.00 5.00 8.00 11.00 14.00 66.00 74.50 83.00 91.50 100.00 ZEN removal B: amount of clay C: zen dosage 14 2 25 125 66 100 pH 3

PAGE 95

95 Figure 4 29.Model graphs showing a) 2D contour lines of ZEN extraction at pH 3 as a function of factor C and factor B and b) interaction plot of ZEN removal against fa ctor B at high and low values of factor C. The %CEC replaced by TCMA is held constant at 70% for both plots. 25.00 50.00 75.00 100.00 125.00 2.00 5.00 8.00 11.00 14.00 ZEN removal B: amount of clay C: zen dosa 84.15 86.96 89.77 92.58 95.39 3 25 125 A: Amount of Clay (mg) B: ZEN concentration (ppm) 2 14 pH 3 84.15 86.96 89.77 92.58 95.39 (c ) C: zen dosage 25.00 50.00 75.00 100.00 125.00 Interaction B: amount of clay ZEN removal 66.00 74.50 83.00 91.50 100.00 25 125 66 100 14ppm 2ppm (a) (b) st. dev: 0.07

PAGE 96

96 Figure 4 30. Response surface 3D plot of ZEN removal (%) as a function of the amount of modified clay added (mg) for extracti on and the CEC replaced by TCMA (%) at pH 7. The ZEN dosage is held constant at 8ppm 20.00 27.50 35.00 42.50 50.00 25.00 50.00 75.00 100.00 125.00 84 87.5 91 94.5 98 ZEN removal A: cec replaced B: amount of clay 20 50 25 125 84 98pH 7 40 100 47.5 55 62.5

PAGE 97

97 Figure 4 31. Model graphs showing a) 2D contour lines of ZEN extraction at pH 7 as a function of factor B and factor A and b) interaction plo t of ZEN removal against factor A. at high and low values of factor B. The ZEN dosage is held constant at 8ppm for both plots. 20.00 27.50 35.00 42.50 50.00 25.00 50.00 75.00 100.00 125.00 ZEN removal A: cec replaced B: amount of clay 86.65 88.89 91.13 93.37 95.61 3 20 50 25 125 95.61 93.37 91.13 88.89 86.65 pH 7 st. dev: 0.043 (a) (b) B: amount of clay 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 75.00 81.00 87.00 93.00 99.00 3 3 20 50 75 99 125mg clay 25mg clay 40 100 47.5 55 62.5 40 100 47.5 55 62.5

PAGE 98

98 Figure 4 32. Response surface 3D plot of ZEN removal (%) as a function of ZEN dosage (ppm) and the CEC replaced by TCMA (%) at pH 7. The amount of organophilic clay added is held constant at 75mg. 20.00 27.50 35.00 42.50 50.00 2.00 5.00 8.00 11.00 14.00 82 86.25 90.5 94.75 99 ZEN removal A: cec replaced C: zen dosage 14 2 20 50 82 99pH 7

PAGE 99

99 Figure 4 33. Model graphs showing a) 2D contour lines of ZEN extraction at pH 7 as a function of factor C and factor A and b) interactio n plot of ZEN removal against factor A. at high and low values of factor C. The amount of organophilic clay added is held constant at 75mg for both plots. 20.00 27.50 35.00 42.50 50.00 2.00 5.00 8.00 11.00 14.00 ZEN removal A: cec replaced C: zen dosage 85.21 87.82 90.42 93.02 95.62 3 20 50 2 14 95.62 93.02 90.42 87.82 85.21 pH 7 st. dev: 0.043 (a) C: zen dosage 20.00 27.50 35.00 42.50 50.00 Interaction A: cec replaced ZEN removal 75.00 81.00 87.00 93.00 99.00 20 50 75 99 14ppm ZEN 2ppm ZEN (b) 40 100 47.5 55 62.5 40 100 47.5 55 62.5

PAGE 100

100 Figure 4 34. Response surface 3D plot of ZEN removal (%) as a function of ZEN dosage (ppm) and the amount of clay added (mg) at pH 7. The %CEC replaced by TCMA is held constant at 70%. 25.00 50.00 75.00 100.00 125.00 2.00 5.00 8.00 11.00 14.00 80 84.75 89.5 94.25 99 ZEN removal B: amount of clay C: zen dosage 14 2 25 125 80 99 pH 7

PAGE 101

101 Figure 4 35. Model graphs showing a) 2D contour lines of ZEN extraction at pH 7 as a function of factor C and factor B and b) intera ction plot of ZEN removal against factor B. at high and low values of factor C. The %CEC replaced by TCMA is held constant at 70% f or both plots. 25.00 50.00 75.00 100.00 125.00 2.00 5.00 8.00 11.00 14.00 ZEN removal B: amount of clay C: zen dosage 83.479 86.4228 89.3666 92.3104 95.2543 25 125 2 14 95.25 92.31 89.37 86.42 83.48 C: zen dosage 25.00 50.00 75.00 100.00 125.00 Interaction B: amount of clay ZEN removal 75 81 87 93 99 25 125 75 99 2ppm ZEN 14ppm ZEN (a) (b) st. dev: 0.043 pH 7

PAGE 102

102 unused cationic surfactants. In addition, it was equally important to determine whether TCMA molecules were being desorbed during the organophilic clay washing process for the same reasons previously mentioned. Lemke and coworkers reported that desorption of alkylammonium molecules during the digestion process could have led to increased ZEN toxicity in rats (Lemke et al. 2001b) Desorption tests conducted by Zhang and coworkers did not accurately replicate the conditions the modified clays would encounter during the digest ion process (Zhang et al., 1993) Although it is di fficult to exactly replicate these conditions in vitro, steps were taken in this study to prepare two electrolyte solutions present during digestion. Desorption tests were run at temperature and pH conditions similar to those present within the stomach of swine. From these studies the viability of these organophilic clays for practical application can be evaluated. 4.4.1 TCMA Adsorption Isotherm The TCMA adsorption isotherm was determined at constant temperature (25oC) using the residue method. The concentr ation of TCMA adsorbed (meq/g) was plotted against the equilibrium concentration of surfactant (mmol/L). To determine the equilibrium concentrations of TCMA two calibration curves were constructed. The first was used to determine concentrations of TCMA bel ow 20ppm and the second to determine TCMA between 100 and 1000ppm. Figure 4 36 shows the calibration curve for low TCMA residues along with the characteristic equation determined by linear regression. The calibration curve and characteristic equation for t he higher TCMA residues is shown in Figure 4 37. The adsorption isotherm was fitted to the Langmuir model and the Freundlich model according to Equation 4 5 and Equation 4 6 respectively. In the Langmuir model, a is the amount of TCMA adsorbed (meq/g), am is the maximum adsorption of TCMA, k is the equilibrium constant, and ce is the equilibrium concentration of surfactant in solution (mmol/L). The Freundlich isotherm is an empirical model

PAGE 103

103 in which K represents the Freundlich constant, ce the equilibrium co ncentration of TCMA, and n is a power constant that is an indicator of the heterogenity of surface sites available for adsorption (Chen et a l. 1997; Praus & Turicova, 2007; Tahani et al. 1999) The adsorption isotherm for TCMA on IMTX and the two isotherm models are shown in Figure 438.The shape of the isotherm is characteristic of high affinity adsorption in which there is a domain of h ighly energetic adsorption at low surfactant concentrations. This type of adsorption gives the appearance of an intercept on the y axis. According to Patzko et al. who studied the adsorption of hexadecylpyridium chloride on montmorillonites, there are two parts of the adsorption isotherm. The first portion represents the adsorption due to ion exchange and the second is due to molecular adsorption (Patzko & Dekany, 1993) This implies that adsorption up to the CEC takes place wi thin the intergalleries of the clay and then on all surfaces of the clay at higher TCMA concentrations. In Figure 4 37a, at concentrations at or below the CEC the Langmuir and Freundlich models do not fit the data. However, as the amount of surfactant is i ncreased above the CEC, the Freundlich model is the best fit for the adsorption data. Figure 437b shows the TCMA -IMTX adsorption isotherm at high TCMA concentrations. It is also important to note that as the TCMA content increases above 100% CEC of IMTX, the hydrophobic interaction between the clay particulates also increases resulting in aggregates that can be larger than several millimeters in diameter. These aggregates have a great impact on the adsorption isotherm as can be seen by the larger error bar s at higher TCMA concentrations in Figure 4 37a b. Although, the better fit by the Freundlich equation can be attributed to the consideration of site heterogeneity, another consideration is that the Freundlich equation does not limit adsorption capacity. I n most

PAGE 104

104 instances this would be considered a limitation. However, the increased adsorption due to hydrophobic interactions is modeled up to 300% of the CEC. e e mkc kc a a 1 (4 5) n eKc a (4 6) Adsorption of alkylammonium cations beyo nd the CEC of the clay has been reported by other researchers (Handbook of clay science 2006; Kwolek et al. 2003; Tahani et al. 1999; Zhang et al., 1993) Adsorption on mineral surfaces up to the CEC is due to s urfactant ions adsorbing on the negatively charged surface sites. Beyond the CEC, additional adsorption is attributed to hydrophobic interactions of the nonpolar surfactant tails. Zhang and coworkers reported that hexadecyltrimethylammonium (HDTMA) modifie d clays were able to adsorb amounts greater than twice the cation exchange capacity of the clays. They also studied the adsorption of nonyltrimethylammonium (NTMA) and dodecyltrimethylammonium (DTMA) onto sodium montmorillonite clays. These clays showed de creased adsorption above 60% of the CEC of the clays. The additional adsorption of HDTMA onto the clay beyond the CEC was attributed to adsorption at nonexchangeable sites. They also reported that there was a critical carbon chain length of 12 carbon units as is the case with HDTMA, to achieve this excessive adsorption (Zhang et al., 1993) As previously discussed in this work, TCMA has three hydrocarbon tails each with 10 carbon units. The unique structure of this surfactant allows it to have properties similar to those of longer chain length. 4.4.2 Desor ption by Washing The semiquantitative measurements by the by the pHydrion QT 10 test strips indicated that in all samples, if any TCMA was present, it was at a concentration of less than 100ppm. For more accurate determinations, the amount of TCMA not adso rbed was determined by

PAGE 105

105 potentiometric titration. In order to quanitify the amount of TCMA present in aqueous solution, standards were prepared and a calibration curve was determined for each titration. Figure 4 38 shows a characteristic potentiometric titr ation curve for 10ppm TCMA titrated with 100ppm sodium lauryl sulfate (SLS) solution. As mentioned previously, the amount of TCMA in the solution is determined by the equivalence point. The equivalence point (EP) indicates the largest change in slope. This change in slope indicates the amount of the titrant equivalent to the concentration of TCMA in solution. In some cases multiple equivalence points are detected by the instrument. Therefore the derivative of the titration curve is taken and the highest pea k indicates which EP has the largest slope. Figure 4 39 shows the derivative of the curve in Figure 4 37. The volume corresponding to the highest EP peak is then used for calibration. Standards of 2ppm, 5ppm, 10ppm, and 20ppm were titrated and the equiva lence points were used to prepare a calibration curve. The calibration curve is shown in Figure 4 36. Each modified clay was subjected to agitation and centrifugation three times or until no sodium chloride could be detected by silver nitrate solution. 50 mL aliquots of the wash supernatants were adjusted to pH 10 by adding 80 L of 2N sodium hydroxide solution so that there would be no interference from amine hydrochlorides during titration. The titration results indicate little or no desorption of TCMA dur ing the washing process. In most cases, no equivalence points were detected. Therefore, if any TCMA is present in the supernatant, the concentration is less than the 2ppm standard which was detected by the instrument. The negative 1st derivative curve for 2ppm TCMA in Figure 4 41 shows two peaks where EPs were detected by the instrument. However, the second peak at 0.402mL SLS added is the highest and is used for calibration. When an EP was detected the amount calculated by the linear regression was always less than 2ppm. However, when the negative 1st derivative curve was considered, the peaks were not

PAGE 106

106 substantial especially when compared to the negative 1st derivative of the titration curve for 2pmm TCMA shown in Figure 4 45. Figure 4 42 shows the potenti ometric titration and negative 1st derivative curves for supernatant of the 100% CEC modified clays after being washed once. The instrument detected no equivalence points. This is supported by the negative 1st derivative curve which does not show a peak. T herefore it can be said that there was no desorption during the 1st wash cycle for the 100% CEC modified clays. The 2nd and 3rd washing cycles for the 100% CEC clays also showed no EPs. The negative 1st derivative curves are shown in Figure 4 43. Again, there are no peaks to indicate the presence of TCMA in solution. The 70% CEC modified clays also showed no desorption of TCMA during the washing cycle. This is evident by the lack of EPs detected and no peaks present in the negative 1st derivative curves. Ho wever, EPs were detected for some of the washing cycles of the 40% CEC clays. Figure4 44 shows the potentiometric and negative 1st derivative curves for the 40% CEC IMTX after the 1st wash. Although a change in slope was detected by the instrument, examina tion of the negative 1st derivative did not indicate any significant peaks. The amount of TCMA calculated by the linear regression is ~1.5 ppm. Figure 4 45 shows the negative 1st derivative curve of 1ppm TCMA. This plot clearly indicates a peak and equival ence point. It can then be determined that although the instrument detected an equivalence point for the first wash cycle of the 40% CEC modified clays, the change in slope was very slight as indicated by the lack of a peak in the negative 1st derivative p lot. Therefore it is fair to conclude that there is less than 1ppm TCMA in the supernatant. The other wash cycles did not have EPs. From this study it was determined that the various washing cycles of the modified clays did not perpetuate desorption of TCM A from the modified clays.

PAGE 107

107 4.4.3 Desorption by Electrolyte Solutions In the presence of gastric and duodenal electrolyte solutions there is some evidence of TCMA desorption. However, the desorption was not found to exceed 2ppm at either pH 3 or pH 7. The values for TCMA desorption from the 100% CEC and 40% CEC modified clays can be found in Table 4 8. The average desorption amount was ~1.5ppm. At pH 3 the desorption amount varied slightly as indicated by a consistently higher standard deviation of ~0.43ppm Figure 4 46 shows the potentiometric titration and corresponding negative 1st derivative curve for the 100% CEC IMTX clays that were exposed to the duodenal solution at pH 7. There is a definite change in slope. This is confirmed by the negative 1st deri vative curve which shows a peak at the EP. The same is true for Figure 4 47 which shows the potentiometric titration and negative 1st derivative curves for the 40% CEC clay exposed to gastric electrolyte solution. The mean lethal dose (LD50) for TCMA is 22 3 mg per kilogram of body weight when administered to rats orally. The values of desorbed surfactant are much less than the LD50 for this surfactant. However, it is important to note that the National Science foundation has recommended that when quaternary ammonium compounds are used in water purification facilities, the maximum amount of residues in drinking water be limited to 50 micrograms per liter (Bolto & Gregory, 2007) Therefore it would be necessary to determine what effects this desorption could have on livestock. As mentioned previously, the gastric electrolyte solution contained 3.1 g/L NaC l, 1.1 g/L KCl, 0.15g/L CaCl2, and NaHCO3 and the duodenal electrolyte solution contained 5.0g/L NaCl, 0.6 g/L, KCl, and 0.3g/L CaCl2. The difference in composition of the salt solutions did not have an effect on the amount of TCMA desorbed. Although the a verage desorption was the same for both pHs, the consistent difference in the standard deviations could indicate that pH is an important factor in the desorption of alkylammonium ions.

PAGE 108

108 y = 0.2913x 0.2041 R2 = 0.9998 0 1 2 3 4 5 6 0 5 10 15 20 Concnetration TCMA (ppm) SLS added (mL) Figure 4 36. Calibration curve for potentiometric titration of TCMA with 100ppm SLS solution. y = 0.014x 0.3712 R2 = 0.9978 0 2 4 6 8 10 12 14 16 0 200 400 600 800 1000 1200 TCMA concentration (ppm) SLS added (mL) Figure 4 37. Calibration curve for potentiometric titration of TCMA with 0.0075M SLS solution.

PAGE 109

109 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 0 0.5 1 1.5 2 2.5 3 Equilibrium Concentration (mmol/L) TCMA Adsorbed (meq/g) Langmuir Freundlich 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 0.55 0.75 0.95 1.15 1.35 1.55 1.75 1.95 2.15 2.35 2.55 Equilibrium Concentration (mmol/L) TCMA Adsorbed (meq/g) Langmuir Freundlich Figure 4 38 Adsorption isotherm for TCMA on IMTX a) at all concentrations and b) at high TCMA concentrations.

PAGE 110

110 220 270 320 370 420 470 520 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 SLS added (mL) mV EP (2.737 mL, 346.9mV) Figure 4 39. Pot entiometric titration curve of 10ppm TCMA titrated with SLS with equivalence point (EP). 0 50 100 150 200 250 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 SLS added (mL) dmV/dV Figure 4 40. First derivative curve of 10ppm TCMA potentiometric titration curve according to volume.

PAGE 111

111 Figure 4 41. First derivative curve of 2pmm TCMA potentiometric titration curve according to volume. -100 0 100 200 300 400 0 0.5 1 1.5 2 2.5 SLS added (mL) dmV/dV dmV/dV SLS added (mL)

PAGE 112

112 0 50 100 150 200 250 300 350 400 450 500 0 1 2 3 4 5 SLS added (mL) dmV/dV 60 70 80 90 100 110 120 130 140 150 160 mV dmV/dV 50 CEC wash 1 Figure 4 42. Potentiometric titration and negative first derivative curves of the supernatant after the 1st washing cycle of 100% CEC organophilic clay.

PAGE 113

113 0 20 40 60 80 100 120 140 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 SLS added (mL) dmV/dV wash 2 wash 3 Figure 4 43. Negative first derivative potentio metric titration curves of the supernatant from the second and third washing cycles of the 100% CEC modified clays.

PAGE 114

114 EP (0.305 mL, 162mV) 0 100 200 300 400 500 600 0 1 2 3 4 5 SLS added (mL) dmV/dV 80 90 100 110 120 130 140 150 160 170 180 mV dmV/dV 20CEC wash 1 EP Figure 4 44. Potentiometric titration and negative first derivative curves of the supernatant after the 1st washing cycle of 40% CEC modi fied clay.

PAGE 115

115 0 50 100 150 200 250 300 350 400 450 500 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 SLS added (mL) mV EP Figure 4 45. Negative first derivative of the potentiometric titration curve for 1ppm TCMA titrated with 100ppm SLS solution. Equivalence point is indicated.

PAGE 116

116 EP (0.239 mL, 261.4 mV) 0 20 40 60 80 100 120 140 160 180 200 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 SLS added (mL) dmV/dV 150 170 190 210 230 250 270 290 mV 1st derivative Duodenal Solution pH7 Duodenal Solution pH7 EP Figure 4 46. Potentiometric titration and negative 1st derivative curves of duodena l electrolyte solution exposed to 100% CEC organophilic IMTX at pH7.

PAGE 117

117 Figure 4 47. Potentiometric titration and negative 1st derivative curves of gastric electrolyte solution exposed to 100% CEC organophilic IMTX at pH3. Table 4 8. Desorpt ion results for TCMA -IMTX modified clays exposed to electrolyte solutions at pH 3 and pH 7. Sample Solution pH Desorption (ppm) 100% CEC gastric 7 1.50 0.14 100% CEC gastric 3 1.27 0.43 100% CEC duodenal 7 1.51 0.08 100% CEC duodenal 3 1.64 0. 44 40% CEC gastric 7 1.53 0.11 40% CEC gastric 3 1.36 0.37 40% CEC duodenal 7 1.58 0.13 40% CEC duodenal 3 1.51 0.39 EP (0.111 mL, 223 mV) 0 20 40 60 80 100 120 140 160 180 0 1 2 3 4 5 SLS added (mL) dmV/dV 130 140 150 160 170 180 190 200 210 220 230 mV 1st derivative gastric solution gastric electrolyte solution pH3 EP

PAGE 118

118 CHAPTER 5 SUMMARY, CONCLUSIONS AND FUTURE RECOMME NDATIONS 5.1 Summary of Results The goal of this study was to develop a safe and cost effect organophilic clay able to bind and extract the mycotoxin zearalenone at amounts greater than 90%, to investigate the factors most important to extraction, and to investigate the fundamental properties between the clay surfactant -mycoto xin systems that lead to extraction. In order to maximize ZEN extraction it is necessary to understand how the properties of the modified clays allow for the selective adsorption of ZEN. In order to accomplish the above research goal, several clay modifica tions and investigations were performed. The resulting data has led to the synthesis of several conclusions that will be discussed later in this chapter. IMTX clay was characterized and modified by cation exchange with TCMA, a quaternary ammonium surfacta nt in amounts equal to 40%, 70%, 100% and 200% of the cation exchange capacity of the clay. These clays were characterized to determine surfactant intercalation, intergallery spacing, and thermal stability. It was determined by FTIR that TCMA was indeed in tercalated into the clay interlayers by the presence of characteristic TCMA peaks along with peaks associated only with IMTX. The peaks at wavenumbers 2928 cm1, 2852 cm1, and 1466 cm1 belonging to C H asymmetric stretching, C H symmetric stretching and CH2 scissoring were associated with TCMA. FTIR also revealed the broadening and diminishing of characteristic water peaks at 3630 cm1 and 3375cm1 wavenumbers caused by the replacement of interlayer water with surfactant. XRD analysis identified the widen ing of the d-spacing of the clays from 1.6nm to 2.2nm upon intercalation. The intercalation of the surfactant and the

PAGE 119

119 widening of the galleries make IMTX favorable for ZEN extraction. The TCMA renders the clay hydrophobic which electrostatically attract s the nonpolar toxin and the newly widened galleries allow for sequestration and extraction. From the XRD data, it was determined not to employ clays modified with more than 100% CEC TCMA due to no considerable increase in the intergallery spacings of the modified clays. TGA identified the three major regions of mass loss of the modified clays and confirmed thermal stability if TCMA to upwards of 200oC. Mass loss below 250oC was characteristic of the evolution of adsorbed water and gaseous species. Between 250oC and 575oC organic substances were degraded. The third region indicated dehydroxylation of the aluminosilicate layers where chemically adsorbed hydroxyl groups are removed. The increase mass loss % in the second region of mass loss when the clays wer e modified with TCMA indicates further that the organic species is present with the clay layers. From the characterization results gained it was possible to surmise what the conformation of TCMA was within the galleries of the clay. Gallery spacings of 2.2nm are characteristic of a pseudotrimolecular conformation in which the surfactant has a somewhat zig -zag appearance. Although the three hydrocarbon chains of TCMA are relatively short (nc = 10 carbon chains) for gallery expansion, the structure of this surfactant allows for widening of the gallery in a manner characteristic of longer chain surfactants. After modification the clays were washed several times before drying to remove any nonadsorbed surfactant molecules and to remove salt. The aqueous exchang e solutions were tested before washing and after each wash cycle to determine if TCMA would desorb from the clays and potentially cause harm during the digestion process. Potentiometric titrations showed little to no desorption of TCMA left behind during m odification and the subsequent washing cycles.

PAGE 120

120 To determine the factors that most influence ZEN extraction, statistical design of experiments was used. The factors investigated were the % of the CEC that was replaced by TCMA, the amount of clay used for extraction, and the initial ZEN dosage. The design chosen was a 23 factorial design with three midpoints. This design investigates the high and low values for each parameter to determine the impact they have on ZEN removal percentages. All experiments were performed at pH 3 and pH 7 in order to better simulate the environments the modified clays will encounter during digestion. The initial ZEN extractions indicated that the %CEC replaced by TCMA was the most influential on the ZEN removal percentage. The amo unt of Clay added was also influential to a lesser degree. ZEN extraction percentages were >90% when the TCMA was added at or above 70% of the CEC at both pH 3 and pH 7. To extract >90% ZEN with 40% CEC modified clays it was necessary to increase the amoun t of clay present in the system. In this study, the amount of ZEN was found to not be significant for the removal of ZEN. There was some interaction between the 2 significant parameters; an increase in one could be balanced by a decrease in the other and vice versa. This investigation led to IMTX modifications at 40% CEC with an addition of corn oil at 3wt% and 10wt%. This new modification did not yield a significant increase in intergallery spacing or overall mass loss as determined by TGA. However the fa tty acids of the oil were intercalated within the clay layers and allowed for an increase in ZEN from ~70% to ~90% in some cases. The extraction decreased with increased corn oil added. This is attributed to saturation of the intergalleries with hydrophobi c species. These studies reveal that it is possible

PAGE 121

121 to use a combination of organic species to modify clays for the extraction of ZEN. The clays can be tailored to decrease cost or increase efficiency according to need. ZEN extraction was also performed at lower ZEN dosages and removal percentages were determined using HPLC -FD. These experiments were scaled down to accommodate the sensitivity of the instrument. As with the UV -Vis spectroscopy results, the 100% CEC and 70% CEC clays showed the best (>95%) ZEN extraction efficiency. However, when 25mg of 100% CEC clay is used to extract ZEN at 14ppm, the ZEN removal percentage drops to below 90% at pH 7. Under these conditions, the amount of clay added becomes a significant factor. The 40% CEC clays show th e least ZEN removal efficiency especially when the ZEN added is 14ppm or the amount of Clay added is 25mg. In addition, the effect of pH is more apparent. The mathematical model derived for this system shows interactions between all the parameters and reve al that the ZEN dosage is the most significant parameter. At low amounts of clay the amount of ZEN in the system becomes overpowering and dwarfs the effects of the other factors. This difference in results shows the importance of the choice of factors and their levels in a system and how desired outcomes can shift quickly depending on the circumstance. Further SDOE would be needed to further explore the intricacies of each parameter and their levels for ZEN extraction. Adsorption studies revealed that TCMA could be adsorbed onto IMTX at amounts equal to 3 times the cation exchange capacity of the clays. Adsorption beyond the CEC is due to hydrophobic interaction between the hydrocarbon tails of the surfactant once the surface charge had been neutralized. Des orption studies determined that washing of the organophilic clays did not increase desorption of TCMA. Large amounts of desorption would indicate instability of the clays. In addition, desorption could also lead to increased waste toxicity and increased co st.

PAGE 122

122 Desorption studies with stomach electrolyte solutions showed approximately 2ppm desorption of TCMA at pH 3 and pH 7. 5.2 Conclusions The major conclusions from this study include: Using IMTX modified with organic species, selective adsorption of Zearal enone at pH 3 and pH 7 was demonstrated. Extraction percentages of >90% were achieved. Statistical Design of Experiments was used to determine the factors important to ZEN extraction and any interactions between the two. Experimental studies should be spec ific to the environment. The modified clays show stability in various pH and electrolyte environments and can be considered biocompatible. 5.3 Recommendations for Future Work The following are several recommendations for future work: Further DOE should be performed to optimize the modified clays for extraction to reduce material cost and optimize extraction results Additional extraction studies should be performed in environments similar to those present during the digestion process of animals to eliminate the possibility of increased adsorption of ZEN caused by the desorption of surfactant. In vivo studies should be performed to determine any toxicity to animals caused by the modified clays

PAGE 123

123 APPENDIX A ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RE SPONSE AT P H 3 AS MEASURED UV-VIS SPECTROSCOPY A = CEC Replaced B = Amount of Clay C = ZEN dosage Response1 : ZEN removal ANOVA for selected factorial model Table A 1. Analysis of variance table [Partial sum of squares Type III] Sum of Mean F p value So urce Squares df Square Value Prob > F Model 252.15 6 42.02 41.48 0.0056 significant A 195.11 1 195.11 192.58 0.0008 B 15.49 1 15.49 15.28 0.0297 C 10.57 1 10.57 10.43 0.0482 AB 22.08 1 22.08 21.79 0.0186 AC 5.44 1 5.44 5.37 0.1033 BC 3.46 1 3.46 3.42 0.1617 Curvature 7.35 1 7.35 7.26 0.0742 not significant Residual 3.04 3 1.01 Lack of Fit 1.62 1 1.62 2.29 0.2697 not significant Pure Error 1.42 2 0.71 Cor Total 262.54 10 The Model F -value of 41.48 implies the model is significant. There is only a 0.56% chance that a "Model F -Value" this large could occur due to noise. Values of "Prob > F" less than 0.0500 indicate model terms are significant. In this case A, B, C, AB are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. The "Curvature F -value" of 7.26 implies there is curvature (as measured by difference between the average of the center points and the average of the factorial points) in the design space. There is only a 7.42% chance that a "Curvature F value" this large could occur due to noise. The "Lack of Fit F -value" of 2.29 implies the Lack of Fit is not signifi cant relative to the pure error. There is a 26.97% chance that a "Lack of Fit F -value" this large could occur due to noise. Non -significant lack of fit is good -we want the model to fit. Table A 2. Statistical Results Std. Dev. 1.01 R-Squared 0.9881 Mean 94.68 Adj R Squared 0.9643 C.V. % 1.06 Pred R Squared 0.5810 PRESS 106.93 Adeq Precision 17.426

PAGE 124

124 The "Pred R -Squared" of 0.5810 is not as close to the "Adj R Squared" of 0.9643 as one might normally expect. This may indicate a large block effect or a possible problem with your model and/or data. Things to consider are model reduction, response tranformation, outliers, etc. "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your ratio of 17.426 indicates an adequate signal. This model can be used to navigate the design space. Table A 3. Diagnostic case statistics Coefficient Standard 95% CI 95% CI Factor Estimate df Error Low High VIF Intercept 94.18 1 0.36 93.04 95.31 A 4.94 1 0.36 3.81 6.07 1.00 B 1. 39 1 0.36 0.26 2.52 1.00 C 1.15 1 0.36 2.28 0.017 1.00 AB 1.66 1 0.36 2.79 0.53 1.00 AC 0.82 1 0.36 0.31 1.96 1.00 BC 0.66 1 0.36 1.79 0.47 1.00 Center Point 1.84 1 0.68 0.33 4.00 1.00 Final Equation in Terms of Coded Factors: ZEN remo val = +94.18 +4.94 A +1.39 B 1.15 C 1.66 A B +0.82 A C 0.66 B C Final Equation in Terms of Actual Factors: ZEN removal = +73.95 +0.46 cec replaced +0.18 amount of clay 0.26 zen dosag e 2.95E 3 cec replaced amount of clay +9.16E 3 cec replaced zen dosage 2.92E 3 amount of clay zen dosage

PAGE 125

125 APPENDIX B ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RE SPONSE AT PH 7 AS ME ASURED UV-VIS SPECTROSCOPY A = CEC Replaced B = Amo unt of Clay C = ZEN dosage Response : ZEN removal ANOVA for selected factorial model Table B 1. Analysis of variance table [Partial sum of squares Type III] Sum of Mean F p value Source Squares df Square Value Prob > F Model 731.99 3 244.00 6 6.43 < 0.0001 significant A 515.00 1 515.00 140.21 < 0.0001 B 102.02 1 102.02 27.78 0.0019 AB 114.97 1 114.97 31.30 0.0014 Curvature 57.44 1 57.44 15.64 0.0075 significant Residual 22.04 6 3.67 Lack of Fit 17.79 4 4.45 2.10 0.3481 not significant Pure Error 4.24 2 2.12 Cor Total 811.47 10 The Model F -value of 66.43 implies the model is significant. There is only a 0.01% chance that a "Model F -Value" this large could occur due to noise. Values of "Prob > F" less than 0.0500 indicate model terms are si gnificant. In this case A, B, AB are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. The "Curvature F -value" of 15.64 implies there is significant curvature (as measured by difference between the average of the center points and the average of the factorial points) in the design space. There is only a 0.75% chance that a "C urvature F -value" this large could occur due to noise. The "Lack of Fit F -value" of 2.10 implies the Lack of Fit is not significant relative to the pure error. There is a 34.81% chance that a "Lack of Fit F value" this large could occur due to noise. Non -s ignificant lack of fit is good -we want the model to fit. Table B 2. Statistical Results Std. Dev. 1.92 R-Squared 0.9708 Mean 92.01 Adj R Squared 0.9562 C.V. % 2.08 Pred R Squared 0.8929 PRESS 80.72 Adeq Precision 18.287 The "Pred R -Squared" of 0. 8929 is in reasonable agreement with the "Adj R -Squared" of 0.9562. "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your ratio of 18.287 indicates an adequate signal. This model can be used to navigate the design

PAGE 126

126 space. Table B 3. Diagnostic case statistics Coefficient Standard 95% CI 95% CI Factor Estimate df Error Low High VIF Intercept 90.61 1 0.68 88.95 92.26 A 8.02 1 0.68 6.37 9.68 1.00 B 3.57 1 0.68 1.91 5.23 1.00 AB 3.79 1 0.68 5.45 2.13 1.00 Center Point 5.13 1 1.30 1.96 8.31 1.00 Final Equation in Terms of Coded Factors: ZEN removal = +90.61 +8.02 A +3.57 B 3.79 A B Final Equation in Terms of Actual Factors: ZEN removal = +42.9 +1.12 cec repl aced +0.33 amount of clay 6.74E 003 cec replaced amount of clay

PAGE 127

127 APPENDIX C ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RE SPONSE AT PH 3 AS ME ASURED BY HPLC FD A = CEC Replaced B = Amount of Clay C = ZEN dosage Response: ZEN removal ANOVA for selected factorial model Table C 1. Analysis of variance table [Partial sum of squares Type III] Source Sum of Squares df Mean Square F Value Prob > F Model 822.16 7 117.45 24181.74 < 0.0001 significant A 254.66 1 254.66 52432.47 < 0.0001 B 211.87 1 211.87 43622.14 < 0.0001 C 86.17 1 86.17 17741.25 < 0.0001 AB 130.84 1 130.84 26938.52 < 0.0001 AC 64.55 1 64.55 13289.16 < 0.0001 BC 46.10 1 46.10 9490.48 0.0001 ABC 27.97 1 27.97 5758.18 0.0002 Curvature 75.36 1 75.36 15516.47 < 0.0001 signi ficant Pure Error 9.714E 003 2 4.857E 003 Cor Total 897.53 10 The Model F -value of 24181.74 implies the model is significant. There is only a 0.01% chance that a "Model F -Value" this large could occur due to noise.Values of "Prob > F" less than 0.0500 indicate model terms are significant. In this case A, B, C, AB, AC, BC, ABC are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those requir ed to support hierarchy), model reduction may improve your model. The "Curvature F -value" of 15516.47 implies there is significant curvature (as measured by difference between the average of the center points and the average of the factorial points) in th e design space. There is only a 0.01% chance that a "Curvature F -value" this large could occur due to noise. Table C 2. Statistical Results Std. Dev. 0.070 R-Squared 1.0000 Mean 93.77 Adj R Squared 0.9999 C.V. % 0.074 Pred R Squared N/A PRESS N/A Adeq Precision 508.549 Case(s) with leverage of 1.0000: Pred R Squared and PRESS statistic not defined "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your ratio of 508.549 indicates an adequate signal. This model can be used to navigate the design

PAGE 128

128 space. Table C 3. Diagnostic case statistics Coefficient Standard 95% CI 95% CI Factor Estimate df Error Low High VIF Intercept 92.17 1 0.025 92.06 92.28 A 5.64 1 0.025 5.54 5.75 1.00 B 5.15 1 0.025 5 .04 5.25 1.00 C 3.28 1 0.025 3.39 3.18 1.00 AB 4.04 1 0.025 4.15 3.94 1.00 AC 2.84 1 0.025 2.73 2.95 1.00 BC 2.40 1 0.025 2.29 2.51 1.00 ABC 1.87 1 0.025 1.98 1.76 1.00 Center Point 5.88 1 0.047 5.67 6.08 1.00 Final Equation in Terms o f Coded Factors: ZEN removal = +92.17 +5.64 A +5.15 B 3.28 C 4.04 A B +2.84 A C +2.40 B C 1.87 A B C Final Equation in Terms of Actual Factors: ZEN remova l= +83.87 +0.28 cec replaced +0.11 amount of clay 3.34 zen dosage 2.07E 3 cec replaced amount of clay +0.062 cec replaced zen dosage +0.023 amount of clay zen dosage 4.15E 4 cec replaced amount of clay zen dosage

PAGE 129

129 APPENDIX D ANALYSIS OF VARIANCE FOR ZEN REMOVAL % RESPONSE A T PH 7 AS MEASURED BY HPLC FD A = CEC Replaced B = Amount of Clay C = ZEN dosage Response: ZEN removal ANOVA for selected factorial model Table D 1. Analysis of variance table [Partial sum of squares Type III] Sum of Mean F p value Source Squares df Square Value Prob > F Model 451.96 7 64.57 34573.07 < 0.0001 significant A 65.97 1 65.97 35322.98 < 0.0001 B 118.45 1 118.45 63423.78 < 0.0001 C 194.83 1 194.83 1.043E5 < 0.0001 AB 4.99 1 4.99 2674.34 0.0004 AC 18.03 1 18.03 9651.84 0.0001 BC 49.69 1 49.69 26608.22 < 0.0001 ABC 0.013 1 0.013 6.84 0.1203 Curvature 41.46 1 41.46 22203.05 < 0.0001 significant Pure Error 3.735E 3 2 1.868E 3 Cor Total 493.43 10 The Model F -value of 34573.07 implies the model is significant. There is only a 0.01% chance that a "Model F -Value" this large could occur due to noise. Values of "Prob > F" less than 0.0500 indicate model terms are significant. In this case A, B, C, AB, AC, BC are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. The "Curvature F -value" of 22203.05 implies there is significant curvature (as measured by difference between the average of the center points and the average of the factorial points) in the design space. There is only a 0.01% chance that a "Curvature F -value" this large could occur due to noise. Table D 2. Statistical Results Std. Dev. 0.043 R-Squared 1.0000 Mean 93.11 Adj R Squared 1.0000 C.V. % 0.046 Pred R Squared N/A PRESS N/A Adeq Precision 594.245 Case(s) with leverage of 1.0000: Pred R -Squared and PRESS statistic not defined "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. Your ratio of 594.245 indicates an adequate signal. This model can be used to navigate the design space.

PAGE 130

130 Table D 3. Diagnostic case statistics Coefficient Standard 95% CI 95% C I Factor Estimate df Error Low High VIF Intercept 91.92 1 0.015 91.85 91.98 A 2.87 1 0.015 2.81 2.94 1.00 B 3.85 1 0.015 3.78 3.91 1.00 C 4.93 1 0.015 5.00 4.87 1.00 AB 0.79 1 0.015 0.86 0.72 1.00 AC 1.50 1 0.015 1.44 1.57 1.00 BC 2.49 1 0. 015 2.43 2.56 1.00 ABC 0.040 1 0.015 0.026 0.11 1.00 Center Point 4.36 1 0.029 4.23 4.49 1.00 Final Equation in Terms of Coded Factors: ZEN removal = +91.92 +2.87 A +3.85 B 4.93 C 0.79 A B +1.50 A C +2.49 B C +0.040 A B C Final Equation in Terms of Actual Factors: ZEN removal = +92.73 +0.14 cec replaced +0.050 amount of clay 2.006 zen dosage 1.12E 3 cec replaced amount of clay +0.016 cec replaced zen dosage +7.00E 3 amount of clay zen dosage +8.88E 6 cec replaced amount of clay zen dosage

PAGE 131

131 LIST OF REFERENCES Abbs, S., Ouanes, Z., Salah -Abbs, J. b., Houas, Z., Oueslati, R., Bacha, H., et al. (2006a). The protective effect of hydrated sodium calcium aluminosilicate against haematological, biochemical and pathological changes induced by zearalenone in mice. Toxicon, 47(5), 567. Abbs, S., Salah -Abbs, J. B., Ouanes, Z., Houas, Z., Othman, O., Bacha, H., et al. (2006b). Pre ventive role of phyllosilicate clay on the immunological and biochemical toxicity of zearalenone in balb/c mice. International Immunopharmacology, 6(8), 1251. Afriyie Gyawu, E., Wiles, M. C., Huebner, H. J., Richardson, M. R., Fickey, C., & Phillips, T. D. (2005). Prevention of zearalenone -induced hyperestrogenism in prepubertal mice. Journal of Toxicology and Environmental Health-Part a -Current Issues, 68 (5), 353368. Ahamed, S., Foster, J. S., Bukovsky, A., & Wimalasena, J. (2001). Signal transduction thr ough the ras/erk pathway is essential for the mycoestrogen zearalenone induced cell -cycle progression in mcf 7 cells. Molecular Carcinogenesis, 30(2), 88 98. -zearalenol and zearalenone in corn, liquid chromatographic method. A OAC Official Methods of Analysis Avantaggiato, G., Havenaar, R., & Visconti, A. (2003). Assessing the zearalenone binding activity of adsorbent materials during passage through a dynamic in vitro gastrointestinal model. Food and Chemical Toxicology, 41(10 ), 1283. Aw, T. C., Smith, A. B., Stephenson, R. L., & Glueck, C. J. (1989). Occupational exposure to zeranol, an animal growth promoter. British Journal of Industrial Medicine, 46(5), 341346. Bennett, G. A., & Shotwell, O. L. (1979). Zearalenone in cerea l grains. Journal of the American Oil Chemists Society, 56 (9), 812819. Bennett, G. A., Shotwell, O. L., & Hesseltine, C. W. (1979). Detoxification of zearalenone contaminated corn. Journal of the American Oil Chemists Society, 56(2), A205 a205. Bennett, J W., & Klich, M. (2003). Mycotoxins. Clinical Microbiology Reviews, 16 (3), 497+. Bolto, B., & Gregory, J. (2007). Organic polyelectrolytes in water treatment. Water Research, 41(11), 2301. Briones Reyes, D., Gmez -Martinez, L., & Cueva Roln, R. (2007). Zearalenone contamination in corn for human consumption in the state of tlaxcala, mexico. Food Chemistry, 100(2), 693. Chen, G., Song, X., Zhao, Y., Han, B., & Yan, H. (1997). Adsorption enthalpy and adsorption isotherm of tetradecylpyridinium bromide on n a montmorillonite. Journal of Colloid and Interface Science, 186 (1), 206.

PAGE 132

132 Chestnut, A. B., Anderson, P. D., Cochran, M. A., Fribourg, H. A., & Gwinn, K. D. (1992). Effects of hydrated sodium calcium aluminosilicate on fescue toxicosis and mineral absorptio n. J. Anim Sci., 70 (9), 28382846. Collins, T. F. X., Sprando, R. L., Black, T. N., Olejnik, N., Eppley, R. M., Alam, H. Z., et al. (2006). Effects of zearalenone on in utero development in rats. Food and Chemical Toxicology, 44(9), 1455. Dakovic, A., Mati jasevic, S., Rottinghaus, G. E., Dondur, V., Pietrass, T., & Clewett, C. F. M. (2007). Adsorption of zearalenone by organomodified natural zeolitic tuff. Journal of Colloid and Interface Science, 311(1), 8 13. Dakovic, A., Tomasevic Canovic, M., Dondur, V. Rottinghaus, G. E., Medakovic, V., & Zaric, S. (2005). Adsorption of mycotoxins by organozeolites. Colloids and Surfaces B: Biointerfaces, 46 (1), 20. Dakovic, A., Tomasevic Canovic, M., Rottinghaus, G., Dondur, V., & Masic, Z. (2003). Adsorption of ochra toxin a on octadecyldimethyl benzyl ammonium exchangedclinoptilolite -heulandite tuff. Colloids and Surfaces B -Biointerfaces, 30 (1 2), 157165. de Paiva, L. B., Morales, A. R., & Valenzuela Daz, F. R. (2008). Organoclays: Properties, preparation and appli cations. Applied Clay Science, 42(1 2), 8. Derodriguez, C. A. S. (1984). Environmental hormone contamination in puerto rico. New England Journal of Medicine, 310(26), 17411742. Derodriguez, C. A. S., Bongiovanni, A. M., & Deborrego, L. C. (1985). An epide mic of precocious development in puerto rican children. Journal of Pediatrics, 107(3), 393396. Desheng, Q., Fan, L., Yanhu, Y., & Niya, Z. (2005). Adsorption of aflatoxin b1 on montmorillonite. Poult Sci, 84(6), 959961. Dixon, J. B., Kannewischer, I., Te norio Arvide, M. G., & Barrientos Velazquez, A. L. (2008). Aflatoxin sequestration in animal feeds by quality -labeled smectite clays: An introductory plan. Applied Clay Science, 40(1 4), 201. Feng, J., Shan, M., Du, H., Han, X., & Xu, Z. (2008). In vitro a dsorption of zearalenone by cetyltrimethyl ammonium bromide -modified montmorillonite nanocomposites. Microporous and Mesoporous Materials, 113(1 3), 99. Fokunang, C. N., Tembe -Fokunang, E. A., Tomkins, P., & Barkwan, S. (2006). Global impact of mycotoxins on human and animal health management. Outlook on Agriculture, 35(4), 247253. Grant, P. G., & Phillips, T. D. (1998). Isothermal adsorption of aflatoxin b1 on hscas clay. Journal of Agricultural and Food Chemistry, 46(2), 599605.

PAGE 133

133 Hadiani, M. R., Yazdanpa nah, H., Ghazi -Khansari, M., Cheraghali, A. M., & Goodarzi, M. (2003). Survey of the natural occurrence of zearalenone in maize from northern iran by thin layer chromatography densitometry. Food Additives and Contaminants, 20(4), 380385. Handbook of clay science (First Edition ed.)(2006). First Edition ed.). New York: Elsevier. Hang, P. T., & Brindley, G. W. (1970). Methylene blue absorption by clay minerals. Determination of surface areas and cation exchange capacities (clay -organic studies xviii). Cla ys and Clay Minerals, 18, 203212. Jordan, J. W., Hook, B. J., & Finlayson, C. M. (1950). Organophilic bentonites.2. Organic liquid gels. Journal of Physical and Colloid Chemistry, 54(8), 11961208. Kannewischer, I., Arvide, M. G., White, G. N., & Dixo, J. B. (2006). Smectite clays as adsorbents of aflatoxin b1: Initial steps. Clay Science, Japan, 12 (Supplement 2), 199204. Krishna, B. S., Murty, D. S. R., & Jai Prakash, B. S. (2001). Surfactant -modified clay as adsorbent for chromate. Applied Clay Science 20 (1 2), 65. Kwolek, T., Hodorowicz, M., Stadnicka, K., & Czapkiewicz, J. (2003). Adsorption isotherms of homologous alkyldimethylbenzylammonium bromides on sodium montmorillonite. Journal of Colloid and Interface Science, 264(1), 14. Lagaly, G. (1981). Characterization of clays by organic -compounds. Clay Minerals, 16(1), 1 21. Lagaly, G. (1986). Interaction of alkylamines with different types of layered compounds. Solid State Ionics, 22(1), 43 51. Lemke, S. L., Grant, P. G., & Phillips, T. D. (1998). Ads orption of zearalenone by organophilic montmorillonite clay. Journal of Agricultural and Food Chemistry, 46(9), 37893796. Lemke, S. L., Mayura, K., Reeves, W. R., Wang, N., Fickey, C., & Phillips, T. D. (2001a). Investigation of organophilic montmorilloni te clay inclusion in zearalenone -contaminated diets using the mouse uterine weight bioassay. Journal of Toxicology and Environmental Health, Part A, 62(4), 243 258. Lemke, S. L., Mayura, K., Reeves, W. R., Wang, N. Y., Fickey, C., & Phillips, T. D. (2001 b). Investigation of organophilic montmorillonite clay inclusion in zearalenone -contaminated diets using the mouse uterine weight bioassay. Journal of Toxicology and Environmental Health -Part A, 62 (4), 243258. Lo, I. M. C., Lee, C. H., & Liljestrand, H. M (1996). Tricaprylmethylammonium bentonite complexes as adsorbents for benzene, toluene, ethylbenzene and xylene. Water Science and Technology, 34(7 8), 319325.

PAGE 134

134 M. A. Abdel -Wahhab, S. A. N. H. A. A. (1999). Effect of aluminosilicates and bentonite on afl atoxin -induced developmental toxicity in rat. Journal of Applied Toxicology, 19(3), 199204. Magan, N., & Olsen, M. (2004). Mycotoxins in food: Detection and control USA: CRC Press LLC. Minervini, F., & Dell'Aquila, M. E. (2008). Zearalenone and reproduct ive function in farm animals. International Journal of Molecular Sciences, 9(12), 25702584. Montgomery, D. C. (2005). Design and analysis of experiments (Sixth Edition ed.): John Wiley & Sons, Inc. Opinion of the scientific committee on food (No. SCF/CS/CNTM/MYC/22Rev 3 Final)(2000). No. SCF/CS/CNTM/MYC/22Rev 3 Final): European Commission Health & Consumer Protection Directorate General. Patzko, A., & Dekany, I. (1993). Ion-exchange and molecular adsorption of a cationic surfactant on clay -minerals. Col loids and Surfaces a-Physicochemical and Engineering Aspects, 71(3), 299307. Pitt, J. I. (2000). Toxigenic fungi and mycotoxins. Br Med Bull, 56 (1), 184-192. Praus, P., & Turicova, M. (2007). A physico-chemical study of the cationic surfactants adsorption on montmorillonite. Journal of the Brazilian Chemical Society, 18(2), 378383. Price, K. R., & Fenwick, G. R. (1985). Naturally-occurring estrogens in foods a review. Food Additives and Contaminants, 2(2), 73 106. Rytwo, G., Serban, C., Nir, S., & Margu lies, L. (1991). Use of methylene -blue and crystal violet for determination of exchangeable cations in montmorillonite. Clays and Clay Minerals, 39(5), 551555. Sabater -Vilar, M., Malekinejad, H., Selman, M., van der Doelen, M., & Fink Gremmels, J. (2007). In vitro assessment of adsorbents aiming to prevent deoxynivalenol and zearalenone mycotoxicoses. Mycopathologia, 163(2), 81. Sharmasarkar, S., Jaynes, W. F., & Vance, G. F. (2000). Btex sorption by montmorillonite organo clays: Tmpa, adam, hdtma. Water Air and Soil Pollution, 119(1 4), 257273. Shier, W. T., Shier, A. C., Xie, W., & Mirocha, C. J. (2001). Structure activity relationships for human estrogenic activity in zearalenone mycotoxins. Toxicon, 39(9), 14351438. Tahani, A., Karroua, M., Van Damme, H., Levitz, P., & Bergaya, F. (1999). Adsorption of a cationic surfactant on na -montmorillonite: Inspection of adsorption layer by xray and fluorescence spectroscopies. Journal of Colloid and Interface Science, 216(2), 242249.

PAGE 135

135 Tanaka, T., Hasegaw a, A., Yamamoto, S., Lee, U. S., Sugiura, Y., & Ueno, Y. (1988). Worldwide contamination of cereals by the fusarium mycotoxins nivalenol, deoxynivalenol, and zearalenone.1. Survey of 19 countries. Journal of Agricultural and Food Chemistry, 36(5), 979983. Turcotte, J. C., Hunt, P. J. B., & Blaustein, J. D. (2005). Estrogenic effects of zearalenone on the expression of progestin receptors and sexual behavior in female rats. Hormones and Behavior, 47(2), 178. Urry, W. H., Wehrmeis.Hl, Hodge, E. B., & Hidy, P H. (1966). Structure of zearalenone. Tetrahedron Letters (27), 3109&. Vaia, R. A., Teukolsky, R. K., & Giannelis, E. P. (1994). Interlayer structure and molecular environment of alkylammonium layered silicates. Chemistry of Materials, 6 (7), 10171022. va n Olphen, H. (1976). Clay and colloid chemistry (second edition ed.). New York: WileyInterscience. Watts, C. W., Chen, Y. C., Ledoux, D. R., Broomhead, J. N., Bermudez, A. J., & Rottinghaus, G. (2003). Effects of multiple mycotoxins and a hydrated sodium calcium aluminosilicate in poultry. International Journal of Poultry Science, 2(6), 372378. Zhang, Z. Z., Sparks, D. L., & Scrivner, N. C. (1993). Sorption and desorption of quaternary amine cations on clays. Environmental Science & Technology, 27(8), 1625 1631. Zinedine, A., Soriano, J. M., Molt, J. C., & Maes, J. (2007). Review on the toxicity, occurrence, metabolism, detoxification, regulations and intake of zearalenone: An oestrogenic mycotoxin. Food and Chemical Toxicology, 45(1), 1.

PAGE 136

136 BIOGRAPHICA L SKETCH Kerri -Ann Hue was born in Kingston, Jamaica. She migrated to the Boston Massachusetts in 1998. She received her Bachelor of Science degree in Materials Science and Engineering from the Massachusetts Institute of Technology in Cambridge, MA in 2002. She began graduate school that same year at the University of Florida and earned a Master of Science degree in 2005.