Electrospun Nafion/Polyacrylonitirile Nanofibers As Ultrathin Layer Chromatography Stationary Phase THESIS Presented in Partial

Electrospun Nafion/Polyacrylonitirile Nanofibers as Ultrathin Layer Chromatography Stationary Phase

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

Yanhui Wang

Graduate Program in Chemistry

The Ohio State University

2016

Master's Examination Committee:

Dr. Susan Olesik, Advisor

Dr. Philip Grandinetti

Yanhui Wang

2016

Abstract

A method to separate charged molecules on ultrathin layer chromatographic

(UTLC) plates using electrospun Nafion/Polyacrylonitrile (PAN) nanofibers as the stationary phase is described in this work. Sulfonate groups and the hydrophobic tetrafluoroethylene backbone on Nafion enable the UTLC plates to separate analytes based on charge and hydrophobicity. The addition of PAN with large molecular weight makes Nafion solution more compatible with the electrospinning process. Various compositions of Nafion and PAN in N, N-dimethylformamide were studied to create a bead-free nanofiber network. Nafion-PAN nanofibers as a stationary phase for UTLC were evaluated using the separations of amino acids and proteins, followed by visualizations using ninhydrin and fluorescamine, respectively. The electrospun Nafion-

PAN plates showed high chemical stability in a wide range of buffer mobile phases with various organic modifiers. Mobile phase velocity decreased with the addition of Nafion into the electrospinning solutions. Contributions to band broadening of the spots were also investigated. The efficiency of separating amino acids was greatly improved compared to that determined on commercial ion exchange TLC. The separation of four proteins demonstrated the feasibility of Nafion-PAN UTLC for separating large biomolecules.

Dedication

This document is dedicated to my family.

iii

Acknowledgments

First and foremost, I would like to express my sincere appreciation to my research advisor, Professor Susan Olesik for her encouragement and guidance throughout the past two years. I’m grateful to her for her great motivation and patience during all the difficult times in my research. Without her valuable ideas and expertise, the completion of this work would have not been possible.

My thanks also go out to all of the members in Olesik group, especially Michael

Beilke, Martin Beres, Jiayi Liu, Raffeal Bennett and Juan Bian, for providing a sounding board of ideas and the most pleasurable research environment. I may not have been able to solve problems over the duration of my project without their generous counsel and help. Also, I must recognize the National Science Foundation for providing the funding for this work.

Additionally, I would like to express my gratitude to my beloved parents for their endless and unconditional support and love. No matter how hard life can get, they are always there like a safe harbor to get me recharged and keep me going. Without their moral and financial support, I would certainly not be here today.

Last but certainly not least, I would like to give my thanks to my boyfriend Wey

Jian Tan, who has always been there for me through my ups and downs for the past four

iv years. He always encourages me to be patient and persistent on what I am doing. Thanks for all the positive thoughts, for reminding me not to give up and for having faith in me.

Vita

2013...... B.S. Chemistry, SUNY-Buffalo

2013 to present ...... Graduate Teaching Associate, Department

of Chemistry and Biochemistry, The Ohio

State University

Publications

Deuro, R. E.; Leiker, K. M.; Wang, Y.; Deuro, N. J.; Milillo, T. M; Bright, F.V. “Rapid,

Nondestructive Denim Fiber Bundle Characterization Using Luminescence Hyperspectral

Image Analysis,” Appl. Spectrosc. 2015, 69, 103-114.

Fields of Study

Major Field: Chemistry

Table of Contents

Abstract ...... ii

Dedication ...... iii

Acknowledgments...... iv

Vita ...... vi

Publications ...... vi

Fields of Study ...... vi

Table of Contents ...... vii

List of Tables ...... x

List of Figures ...... xi

Chapter 1: Introduction ...... 1

1.1 Thin Layer Chromatography ...... 1

1.2 Electrospinning...... 2

1.3 Ion Exchange Chromatography ...... 7

1.4 Nafion ...... 8

1.5 Nafion-Polyacrylonitrile Electrospun Ultrathin Layer Chromatography ...... 11

1.6 Research Focus ...... 12

vii

Chapter 2: Experimental ...... 13

2.1 Materials ...... 13

2.2 Instrumentation...... 14

2.3 Preparation of Nafion/PAN solution ...... 14

2.4 Electrospinning...... 15

2.5 Nanofiber and Mobile Phase Compatibility ...... 15

2.6 Ultrathin Layer Chromatography ...... 15

2.7 Visualization...... 16

Chapter 3: Results and Discussion ...... 19

3.1 Optimization of electrospun Nafion/PAN nanofiber stationary phase ...... 19

3.1.1 Effect of polymer compositions on nanofiber structure ...... 19

3.1.2 Effect of solution flow rate on nanofiber structure ...... 22

3.1.3 Effect of voltage on nanofiber structure ...... 25

3.1.4 Effect of distance on nanofiber structure ...... 28

3.1.5 Electrospun Nafion/PAN nanofibers ...... 31

3.2 Nanofiber stability in mobile phases ...... 34

3.3 Separation of amino acids ...... 36

3.3.1 Optimization of mobile phase...... 36

3.3.2 Comparison of Nafion/PAN UTLC and commercial ion exchange TLC ...... 41

viii

3.3.3 Mobile phase velocity ...... 46

3.3.4 Band broadening ...... 49

3.4 Separation of proteins...... 52

3.4.1 Visualization Reagent ...... 53

3.4.2 Buffer selection ...... 54

3.4.3 Effect of buffer pH ...... 60

3.4.4 Effect of salt concentration ...... 62

3.4.5 Effect of organic modifier concentration ...... 65

Chapter 4 Conclusion ...... 69

References ...... 71

List of Tables

Table 1. Fluorescamine spray procedure...... 18

Table 2. Structures, pI values and charges in pH 4.0 of selected amino acids...... 43

Table 3. Retardation factors of amino acids separated on Nafion/PAN UTLC plate and on commercial IE-TLC reported in literature ...... 44

Table 4. Abbreviations, molecular weights and pI values of the proteins used in this work

...... 57

List of Figures

Figure 1. Basic electrospinning setup ...... 5

Figure 2. SEM image of PVA electrospun nanofibers...... 6

Figure 3. Molecular structure of Nafion ...... 10

Figure 4. SEM images of electrospun nanofibers using Nafion/ PAN/ DMF at different compositions: A) 4: 6: 90, B) 3: 7: 90, C) 2: 8: 90, D) 9: 6: 85, E) 7.5: 7.5: 85, F) 6: 9: 85

...... 21

Figure 5. SEM images of electrospun Nafion/PAN nanofibers at solution flow rate: A)

1.5 mL/h, B) 0.5 mL/h ...... 23

Figure 6. Distribution of Nafion/PAN fiber diameters at flow rate: A) 1.5 mL/h, B) 0.5 mL/h ...... 24

Figure 7. Effect of voltage on diameter of electrospun Nafion/PAN nanofibers...... 26

Figure 8. SEM images of electrospun Nafion/PAN nanofibers at A) 10 kV, B) 15 kV. .. 27

Figure 9. SEM images of electrospun Nafion/PAN nanofibers at distance: A) 3 cm, B) 10 cm, C) 15 cm...... 29

Figure 10. Effect of distance on diameter of electrospun Nafion/PAN nanofibers ...... 30

Figure 11. SEM image of electrospun Nafion/ PAN (40: 60, w/w) under optimized condition: flow rate= 0.5 mL/h, voltage= 12 kV, distance= 15 cm, RH= 15%...... 32

Figure 12. SEM image of stationary phase thickness electrospun for 10 min...... 33

Figure 13. Nafion/ PAN nanofibers soaked in mobile phases A) before soaking, B) in acetate buffer: butanol: methanol (6:2:2 v/v/v) after 60 min, C) in bicine buffer: acetonitirle (6:4 v/v) after 60 min...... 35

Figure 14. Retardation factors of Arg (▲), Lys (■), Pro(◆) and Val (●) as a function of different % acetate buffer in the mobile phase. The volume ratio of butanol and methanol was 1:1...... 39

Figure 15. Comparison of plate number, N of Arg (■), Lys (■), Pro (■) and Val (■) using different % acetate buffer in the mobile phase. The volume ratio of butanol and methanol was 1:1...... 40

Figure 16. Comparison of the plate height on the Nafion/PAN UTLC plate (■) and the commercial IE-TLC plate (■)...... Error! Bookmark not defined.

Figure 17. Comparison of mobile phase migration rate on the pure PAN (●) and the

Nafion/PAN (▲) UTLC plates...... 48

Figure 18. Change in plate height H of phenylalanine with increasing solvent migration distance on Nafion/PAN UTLC plate...... 51

Figure 19. The calculated relationship between protein net charge and the buffer pH of

BSA (◆), INS (■), LYZ (✕) and CHY (▲)...... 58

Figure 20. Chromatogram for separation of 1) LYZ; 2) BSA; 3) INS; 4) CHY on electrospun Nafion/ PAN UTLC using with 0.4 M NaCl in bicine buffer (pH=9.0, 50 mM) with 40% acetonitrile...... 59

xii

Figure 21. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of buffer pH with 0.4 M NaCl and 40% acetonitrile...... 61

Figure 22. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of salt concentration in bicine buffer with pH at 8.5 and 40% acetonitrile...... 64

Figure 23. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of the organic modifier percentage in bicine buffer with pH at 9.0 and 0.4 M

NaCl...... 67

Figure 24. Kyte-Doolittle hydrophobicity plots of A) LYZ; B) BSA; C)CHY; D) INS. . 68

xiii

Chapter 1: Introduction

1.1 Thin Layer Chromatography

Thin layer chromatography (TLC) is the simplest chromatographic method which involves the separation of multiple samples and standards on an open layer developed by

1 a mobile phase. Basic TLC is carried out by applying sample spots near the bottom edge of the stationary phase, which is placed in a closed chamber containing a mobile phase, and analytes will move through the layer along with the solvent by capillary forces.2 TLC is a widely used chromatographic method in many areas including synthetic chemistry,3 food science,4 pharmaceutical industry5 as well as clinical research6 due to its simplicity, rapidness and low cost. The history of TLC dates back to 1938 when Izmailov and

Schraiber introduced drop chromatography by applying drops of solvents to glass plates containing unbound alumina or other absorbents to separate medical compounds.1

Modern TLC was not established until in 1950s, Kirchner et al. introduced quantitative

TLC, and Stahl established the term “thin-layer chromatography” and standardized the separation technique.1, 7 High performance thin layer chromatography (HPTLC), an extension of TLC, was developed in 1970s. Owing to the smaller particles in the stationary phase, HPTLC offers better chromatographic efficiency, faster separation,

8 smaller analytes and mobile phase consumption and lower detection limit. In order to further improve the chromatographic performance and reduce analysis time and the amount of consumables, ultrathin layer chromatography (UTLC) was introduced in 2001 1 using monolithic silica as the stationary phase. Compared to classic TLC with thicknesses of 100-400 µm, UTLC utilizes sorbent layers as thin as 5-25 µm, approximately.9

There are many materials that have been developed as stationary phases for TLC, including the most commonly used silica gel and modified silica gel, and less frequently used aluminum oxide, cellulose, polyamides and ion exchange resins.10 Recently, our group has reported electrospun nanofibers as the stationary phases for UTLC without a

11 binder. Various polymers can be used to fabricate electrospun nanofibers based on desired interactions. Polyacrylonitrile (PAN), glassy carbon, polyvinyl alcohol (PVA), silica-based nanofiber, carbon nanotube and carbon nanorod-filled polyacrylonitrile were successfully fabricated as stationary phases for UTLC using the electrospinning method.

All of these electrospun UTLC plates showed highly enhanced separation efficiency compared to commercial HPTLC plates.11,12,13,14,15

1.2 Electrospinning

Electrospinning is a technique that utilizes electrical charge to produce continuous fibers from polymer solution, with nanometer sized diameters.16 It is a simple, inexpensive and robust process that is applicable to a wide range of polymers. The basic setup of electrospinning consists of a high voltage power supply, a syringe pump, a syringe containing polymer solution and a collector at a defined distance, as shown in figure 1. Resulted nanofibers offer several advantages, including tunable diameters, high surface-to-volume ratio and adjustable porosity. Due to these advantages, electrospun nanofibers have attracted increasing attention in many different fields for a wide range of

2 applications, such as electrode materials17, optical and chemical sensor18, filtration19 and biological scaffolds.20

In the electrospinning process, the polymer solution is expelled from the syringe at a constant and controllable rate using a syringe pump. One electrode from the power supply is connected to the syringe needle while the other electrode is connected to the collector. The liquid droplet at the syringe tip gets electrostatically charged when a high voltage is applied. When the voltage reaches a critical value, electrostatic repulsion forces overcome the surface tension of the solution and a liquid jet erupts from the droplet to form a Taylor cone. The liquid jet continues to be elongated by a whipping process and the solvent evaporates before reaching the grounded collector. Finally, a nonwoven mat of random oriented nanofibers is deposited on the collector.21,22,23,24 Electrospun nanofibers as sorbents for UTLC not only provide tunable diameter and mat thickness, but also require no binder between the stationary phase and the solid support, which eliminate band dispersion resulting from interactions with different materials. In order to minimize band broadening of UTLC, electrospun nanofibers must have homogenous morphology and uniform diameter. There are many parameters that affect the fiber morphology, including the properties of polymer solution (viscosity, molecular weight and conductivity), applied voltage, syringe pump flow rate, distance between syringe tip and collector, and ambient conditions such as relative humidity and temperature.21 All of these parameters have to be optimized for different polymers to produce uniform and fine fibers. Figure 2 shows the SEM image of PVA electrospun nanofibers under optimum

3 conditions.13 Optimization of these parameters for the polymer solution used in this work will be discussed in detail in Chapter 3.

Figure 1. Basic electrospinning setup.

Figure 2. SEM image of PVA electrospun nanofibers.13

1.3 Ion Exchange Chromatography

Ion exchange chromatography (IEC) is a process that separates charged molecules based on their affinities to the ionic functional groups immobilized on the stationary phase. These ionic functional groups on the stationary phase are called ion exchangers and they are classified into two types: i) cation exchangers, which carry negative charge, are used to separate cations; ii) anion exchangers, which are positively charged, are used to separate anions. Both ion exchangers can be further classified into two types based on the degree of ionization: i) strong exchangers, which are fully ionized over a broad pH range; ii) weak exchanger, which are partially ionized in a narrow pH range.25,26 The mobile phase in IEC is most commonly composed of a buffer solution with a suitable salt and a certain percentage of organic modifier.27 The separation and elution in IEC can be realized via two approaches. Since the net charge of all molecules is pH dependent, changing the pH of the mobile phase makes analytes carry the same charge as the ion exchanger. Sorbed analytes can be displaced and eluted by increasing the concentration of competing ions in the mobile phase.

IEC is probably the most popular chromatographic technique for separating biological molecules, such as amino acids, peptides and proteins, owing to its simplicity, high capacity, high resolving power and easy controllability of the separation process.28

Separation of these biological molecules is essential in food science, agricultural science, pharmaceutical industry and medicine.29,30 In practice, this technique is more often used in column chromatography,31 which exhibits disadvantages in costly equipment, time

7 consuming experiments and difficulties in detection. Employing the ion exchange technique on TLC eliminates these requirements for time and expensive instruments.

Also, the detection in TLC can be realized by a simple spray reaction, by absorption of ultraviolet light, by using fluorescent labeling, or even direct observation in the case of colored analytes.32

1.4 Nafion

Nafion is a synthetic perfluorinated anonic polymer developed by Dupont in the

1970s.33 As shown in figure 3, a hydrophobic tetrafluoroethylene (Teflon) backbone makes Nafion highly chemical resistant, while sulfonate groups on the side chains give

Nafion superior proton conductivity. Scaling down Nafion into the nanometer scale via electrospinning can further improve its proton conductivity. During the electrospinning process, the shear force elongates fibers and orients ionic domains along the fiber axis direction, and the aligned ionic structures result in higher conductivity.34,35 Therefore, electrospun Nafion nanofibers are desirable for a variety of applications including fuel cells,36 sensors37 and ionic conducting nanofibers.34

Conventional ion exchange chromatographic sorbents have drawbacks. For example, styrene-divinylbenzene, the most commonly used ion exchange resin,2 is a highly cross-linked polymer, which only permits the separation of small molecules; on the other hand, cellulose ion exchangers with low cross-linking content, which are designed for large molecules, suffer from the poor mechanical stability resulting in swelling and contraction and consequently reduced resin life. In addition, conventional

8 resins have ion exchange sites on every monomer unit leading to a poor selectivity which is solely based on charge difference.38 The combination of hydrophobic backbone and sulfonate groups make Nafion a great candidate as a UTLC stationary phase with mixed- mode retention mechanism. The triflic acid-resembling sulfonate group on Nafion is superacidic with a pKa of -6.34 In this respect, Nafion can be used as a strong cation exchanger to separate charged analytes over a broad pH range. Moreover, Nafion has one sulfonate group every eight monomer units giving rise to large hydrophobic segments, thus analytes can be separated based on their hydrophobicities.38 Hence, as a cationic exchange/ reverse phase UTLC stationary phase, Nafion nanofibers are expected to provide better selectivity towards analytes in comparison with single mode sorbents.

Besides, non-cross-linked Nafion is permeable to both small and large molecules,38 and its high chemical resistance can prevent the fiber mat from swelling during mobile phase migration.

Figure 3. Molecular structure of Nafion.

1.5 Nafion-Polyacrylonitrile Electrospun Ultrathin Layer Chromatography

As mentioned in section 1.2, electrospinning is highly dependent on the polymer solution properties. For instance, when the solution viscosity is too high, it is difficult to pump the polymer solution out of the syringe due to the high cohesiveness, and the solution may also get dried up at the needle tip before electrospinning occurs; on the other hand, at low solution viscosities, the lack of polymer chain entanglements results in the formation of beads along the fibers, and if the viscosity gets extremely low, electrospraying occurs to yield only droplets instead of fibers.21 Unfortunately, the pure

Nafion solution is found to have a low shear viscosity that results in electrosprays.39

Therefore, another polymer is needed to blend with Nafion for electrospinning to take place. Several researchers have successfully blended Nafion with carrier polymers, such as poly (acrylic acid) (PAA),39, 40 poly (vinylalcohol) (PVA),41,42 poly (ethylene oxide)

(PEO)34, 43,44,45 and polyacrylonitrile (PAN).46,47 In our group, PAN and multi-walled carbon nanotubes (MWNT) filled PAN have been previously electrospun as UTLC stationary phases and showed substantially improved chromatographic performance.

Moreover, unlike PVA or PEO, which interacts with Nafion through hydrogen bonding,48,49 PAN is not known to have any affinity to Nafion.45 Thus, in this work, PAN was chosen as the carrier polymer to electrospin with Nafion. Different fiber morphologies were obtained from previous work by varying the compositions of Nafion/

PAN in the polymer solution.45 As previously stated, bead-free and uniform nanofibers are desired for a UTLC sorbent. In addition, the fiber dimension is crucial for separation as the smaller the particle diameter, the higher the separation efficiency.11 This relation

11 can be realized by examining the van Deemter equation in Chapter 3. Optimum compositions of Nafion/ PAN for fabricating ideal UTLC nanofibers will be explored in this study.

1.6 Research Focus

The thesis describes Nafion/PAN electrospun nanofibers as a UTLC stationary phase, which provides an ion exchange/ reverse phase mixed-mode mechanism, and the evaluation of its chromatographic performance by separating amino acids and proteins.

The impact of electrospinning parameters on the fiber morphology were explored.

Chemical stability of the resulting fiber mat was also investigated. Due to the unique chemical properties of Nafion, this UTLC stationary phase was expected to have enhanced separation efficiency. Different selectivities for small and large molecules, and contributors to band broadening were examined in detail.

Chapter 2: Experimental

2.1 Materials

Nafion containing solution LIQUION 1115 (15% by weight NAFION®, 1100 equivalent weight) was purchased from Ion Power Inc. (New Castle, DE).

-1 Polyacrylonitrile (PAN), average Mw 150,000 g mol , was purchased from Sigma-

Aldrich (St. Louis, MO). Electrospinning solvent N, N-Dimethylformamide (DMF)

(99.8%), HPLC grade methanol (99.9%), acetonitrile (99.9%), 2-propanol (99.9%) and methylene chloride (99.9%) were acquired from Fisher Scientific (Fair Lawn, NJ).

Ethanol (91%) was purchased from Decon Labs Inc. (King of Prussia, PA). Buffer reagents MES hydrate (minimum 99.5%), bicine (≥ 99%) were purchased from Sigma-

Aldrich; sodium chloride (100%), glacial acetic acid (100%) and ammonium acetate

(98.2%) were purchased from Fisher Scientific. Water used for preparing buffers was purified at 18.1 MΩ by a Barnstead Nanopure Infinity System from Thermal Scientific

Inc. (Odessa, TX).

The amino acids including arginine (Arg), lysine (Lys), histidine (His), proline

(Pro), valine (Val), phenylalanine (Phe) and aspartic acid (Asp); Arg, Lys and Asp were purchased from Sigma-Aldrich, Pro and Val were purchased from Amresco (Solon, OH),

His was purchased from Matheson Coleman & Bell (Gardena, CA), and Phe was purchased from Eastman Chemical Company ( Kingsport, TN). Protein analytes included lysozyme from chicken egg, insulin from bovine pancreas, bovine serum albumin (BSA)

13 and alpha-chymotrypsin from bovine pancreas. Lysozyme, insulin and BSA were obtained from Sigma-Aldrich, and alpha-chymotrypsin was purchased from Worthington

Biochemical Corporation (Lakewood, NJ). Visualization spray reagents ninhydrin (≥

98%), fluorescamine (≥ 98%) and triethylamine (≥ 99.5%) were purchased from Sigma-

Aldrich.

2.2 Instrumentation

The morphology of electrospun nanofiber mats was characterized using a Hitachi

S-4300 (Hitachi High Technologies, America, Inc., Pleasanton, CA) scanning electron microscope (SEM). Each sample was sputter coated with gold for 2 min at 10 µA to create a conductive surface before SEM analysis. Fiber diameters were measured on

SEM images using ImageJ software (Available from the National Institute of Health at http://www.rsbweb.nih.gov/ij/index.html).

2.3 Preparation of Nafion/PAN solution

Pure Nafion was obtained following the drying/curing procedure provided by Ion

Power Inc. Dried Nafion and PAN were dissolved in DMF and stirred for at least 24 h using a magnetic stirrer under gentle heating. Electrospinnability of Nafion/PAN as a function of blend composition and total solid concentration in DMF was previously studied by Tran, et al.46 The optimum composition of Nafion/PAN in DMF for a bead- free and uniform UTLC sorbent with sufficient ionic capacity was further explored in this work.

2.4 Electrospinning

A syringe pump (Harvard Apparatus, Holliston, MA), a high voltage power supply (Spellman, Hauppauge, NY), a stainless steel collector covered with aluminum foil (Reynold Super Strength) and a Plexiglas enclosure were used as the electrospinning apparatus. Solution compositions, feeding rate, voltage, distance between syringe needle tip and collector were varied in order to examine the effects of each parameter on the nanofiber morphology. The relative humidity was controlled to 20% or below using a

® nitrogen purge and monitored by a VWR Traceable humidity sensor while electrospinning. Each nanofiber mat was electrospun for 10 min.

2.5 Nanofiber and Mobile Phase Compatibility

In order to investigate the mechanical properties of Nafion/PAN nanofibers during the development of mobile phases, UTLC plates were soaked in mobile phases for different periods of time (15 min, 30 min, 45 min and 60 min). For amino acids, the optimized mobile phase was ammonium acetate buffer (pH=4.0, 10 mM)/ methanol/ butanol (60:20:20, v/v/v); for proteins, the optimized mobile phase was bicine buffer

(pH= 9.0, 50 mM)/ acetonitrile (60:40 v/v). After soaking, each plate was allowed to dry and SEM images were taken for the measurements of fiber diameters.

2.6 Ultrathin Layer Chromatography

After electrospun nanofiber mats were prepared, they were cut into 3 cm × 6 cm

UTLC plates. Fused silica capillary tubes with an inner diameter of 250 µm were used for

15 spotting analytes onto the origin line drawn at 1 cm from the bottom of the plate. The volume of analytes spotted, which was approximately 50 nL, was calculated as the volume difference in the capillary tube before and after spotting. Mixtures of organic solvents and aqueous buffers were required to prepare analyte solutions instead of using

100% aqueous buffers or water, otherwise analytes could not be spotted onto the plates due the hydrophobic nature of Nafion/PAN nanofibers. Amino acids (30 mM) were prepared in ammonium acetate buffer (pH=4.0, 10mM)/ methanol (60:40, v/v). Proteins were prepared in MES buffer (pH=6.5, 10 mM)/ acetonitrile (80:20, v/v) with the concentrations as follows: insulin (0.5 mg/ mL), chymotrypsin, lysozyme and BSA (1 mg/mL for each).UTLC plates were then placed in a 250 mL developing chamber containing 5 mL of mobile phase. Prior to each development, the mobile phase was allowed to equilibrate for 10 min. Each UTLC plate was removed from the developing chamber till the mobile phase migrated 3cm above the origin.

2.7 Visualization

After development, UTLC plates were dried at room temperature. For separations of amino acids, plates were sprayed with freshly prepared ninhydrin solution (0.3 g of ninhydrin dissolved in 100 mL of n-butanol with 3 mL of HOAc) using a TLC reagent nebulizer (Kimble-Chase Vineland, NJ), and then heated in the oven at 110 ºC for 10 min. After the ninhydrin treatment, amino acids showed purple violet or pink colors

(except proline with yellow color) immediately under ambient light. For separations of proteins, plates were treated according to the fluorescamine spray procedure (Table 1).50

Protein analytes were then visualized under UV radiation at λ= 365 nm. A Canon A650IS

12.1 MP digital camera was utilized to take photographs after separations, and the digital images were then analyzed using ImageJ, PeakFitTM (version 4, SPSS Inc.) and TLC analyzer.

Step Treatment

Spray with a solution of 10% triethylamine in methylene chloride; air 1 dry for several seconds.

Spray with a solution of 0.05% fluorescamine in acetone; air dry for 2 several seconds.

3 Re-spray with a solution of 10% triethylamine in methylene chloride.

Table 1. Fluorescamine spray procedure. 40

Chapter 3: Results and Discussion

3.1 Optimization of electrospun Nafion/PAN nanofiber stationary phase

3.1.1 Effect of polymer compositions on nanofiber structure

The properties of the polymer solution play the most significant role in the electrospinning process and the fiber morphology.21 In this study, PAN was selected as a carrier polymer to allow the electrospinning of Nafion to occur. DMF was chosen as the solvent because both Nafion and PAN are soluble in it. In order to obtain bead-free nanofibers with uniform fiber diameters, Nafion and PAN were blended at various ratios

(Nafion: PAN) in DMF. The total polymer concentration in DMF was started with 10 wt% (Nafion + PAN: DMF). As shown in figure 4A, when the concentration of Nafion was 40 wt% out of the Nafion/PAN blend, spherical droplets were formed, which indicated that the Nafion concentration was too large. This phenomenon was attributed to the low viscosity of Nafion. Due to ionic interactions, Nafion aggregates in the solution resulting in its low viscosity.51 The initiation of electrospinning requires the electrostatic repulsion to overcome the surface tension. Electrostatic repulsion tends to stretch the solution and increase the surface area; in contrast, surface tension tends to minimize the surface area per unit mass of the solution. The opposing forces on the fluid jet causes the jet to break into droplets, and this effect is known as Rayleigh instability.52 When high content of low viscous liquid is used, it has low resistance against Rayleigh instability and the molecules have a greater tendency to congregate and form spherical droplets with

19 minimal surface area. Therefore, the Nafion content was lowered to 30 wt%, the SEM image of which is shown in figure 4B. Unfortunately, bead-free fibers were not obtained until the Nafion content was reduced to 20 wt%, as shown in figure 4C. Since the goal of this work is to utilize the unique chemical properties of Nafion as the UTLC stationary phase, too much of the carrier polymer would suppress the performance of Nafion.

Therefore, a solution with higher Nafion content is desired. Increasing the total polymer concentration in the solution can compensate for the increased Nafion content.46 Figure 4

D-F show nanofibers electrospun using 15 wt% total polymer concentration (Nafion+

PAN) in DMF. Bead-free nanofibers with considerable Nafion content (figure 4F) were successfully fabricated with 6 wt% Nafion, 9 wt% PAN and 85 wt% DMF composition

(40 wt% Nafion out of Nafion/PAN blend). To further increase the Nafion content, the total polymer concentration has to be increased as well; however, with DMF solvent content lower than 85%, the viscosity of the solution became too high that the resulting deposition area on the collector was too small to cut into UTLC plates. Since the bending instability which determines the total path length of the fluid jet from the syringe needle tip to collector would be reduced by a high enough viscosity, the charged jet only spreads over a smaller area.53 In addition, the high viscosity counters the coulombic stretching of the charged jet and results in fibers with larger diameters, which are undesirable as UTLC stationary phases compared to those with smaller diameters.

Figure 4. SEM images of electrospun nanofibers using Nafion/ PAN/ DMF at different compositions: A) 4: 6: 90, B) 3: 7: 90, C) 2: 8: 90, D) 9: 6: 85, E) 7.5: 7.5: 85, F) 6: 9: 85.

3.1.2 Effect of solution flow rate on nanofiber structure

Different fiber diameters can be obtained by changing the solution flow rate at a given voltage. Using the electrospinning solution optimized above, while other parameters were held constant (voltage= 10 kV based on previous studies,46 distance= 15 cm, RH= 15%), the solution flow rate was initially set as 1.5 mL/h, which was the same as the flow rate for pure PAN in previous work. As shown in figure 5A, even though no beads were observed, the nanofibers obtained using 1.5 mL/h showed non-uniform diameters and a high chance of formation of larger fibers. Since the flow rate determines the volume of solution available for electrospinning, a higher flow rate will draw a larger amount of solution from the syringe, which requires a longer time to dry.21 With a higher flow rate, the resulting fibers might not have enough time for the solvent to evaporate completely and the residual solvent can cause the fibers to fuse together and form webs.54

Therefore, a lower flow rate at 0.5 mL/h was used. Under this condition, bead-free and uniform fibers were obtained (figure 5B). Also, fibers made with the lower flow rate show smaller average fiber diameters in a narrower distribution compared to those made with the higher flow rate (figure 6). As for a UTLC sorbent, a lower solution flow rate is more desirable for fabricating bead-free and fine fibers. Therefore, the flow rate of 0.5 mL/ h was selected as the standard parameter for electrospinning Nafion/PAN in the following studies.

Figure 5. SEM images of electrospun Nafion/PAN nanofibers at solution flow rate: A)

1.5 mL/h, B) 0.5 mL/h.

Figure 6. Distribution of Nafion/PAN fiber diameters at flow rate: A) 1.5 mL/h, B) 0.5 mL/h.

3.1.3 Effect of voltage on nanofiber structure

The electrical voltage is a crucial parameter that affects the morphology of electrospun nanofibers. Using the solution composed of Nafion/ PAN/ DMF (6: 9: 85, w/w/w), the critical voltage at which electrostatic repulsion overcomes surface tension was found to be 8 kV. While other electrospinning conditions were held constant

(distance = 15cm, RH= 15%, flow rate = 0.5 mL/ h), the applied voltage was varied from

8 to 20 kV. Figure 7 shows that the fiber diameters decrease drastically with increasing voltage from 8 to 12 kV, and slightly decrease in the range of 12-20 kV. Owing to the stronger electric field and greater electrostatic forces in the solution when high voltage was applied, the solution experienced greater stretching as well as faster solvent evaporation which led to reduced fiber diameter.55, 56, 57 As mentioned above, finer fibers are favored in the application of UTLC. However, when voltage was raised above 15 kV, more beads were observed compared to those using lower voltages (figure 8). As shown in figure 8B, electrospinning with 15 kV also deposited nanofibers onto the collector in a more chaotic manner. The tendency for bead formation and chaotic orientation of resulting nanofibers may be due to the increased jet instability as the Taylor cone recedes into the syringe needle when higher voltage is applied.58, 59 Moreover, Krishnappa et. al. reported that at very high voltage, as the beads intensity increases, the beads will join together to form thicker fibers.60 This also explains the observation of fiber growth in air from collector when voltage was raised above 20 kV. Based on these observations, 12 kV was selected as the applied voltage for later experiments.

Figure 7. Effect of voltage on diameter of electrospun Nafion/PAN nanofibers.

Figure 8. SEM images of electrospun Nafion/PAN nanofibers at A) 10 kV, B) 15 kV.

3.1.4 Effect of distance on nanofiber structure

Varying the distance between the syringe needle and the grounded collector can directly affect the jet flight time and the electrical field strength, which in turn have an impact on the morphology of resultant fibers.21 By holding other electrospinning parameters constant (flow rate= 0.5 mL/h, voltage= 12 kV, RH= 15%), the distance was varied from 3 cm to 20 cm. When the distance was as small as 3 cm, a large amount of beads were collected (figure 9A). At this point, the significant electrical field strength increases the chance of beads formation, as discussed above. Meanwhile, the extremely short distance does not allow the sufficient evaporation of solvent and also results in less stretching of the liquid jet. As shown in figure 9, when the distance was increased to 10 cm, beads-on-string morphology was observed, and with even longer distance, smooth fibers were obtained. Average fiber diameters were reduced as the distance increased

(figure 10); additionally, the resulting fiber mat area increased with increasing distance.

However, when the distance was larger than 15 cm, a higher applied voltage was required, otherwise fibers could not reach the collector. Thus, 15 cm was selected as the optimal distance for electrospinning in subsequent experiments.

Figure 9. SEM images of electrospun Nafion/PAN nanofibers at distance: A) 3 cm, B) 10 cm, C) 15 cm. 29

Figure 10. Effect of distance on diameter of electrospun Nafion/PAN nanofibers.

3.1.5 Electrospun Nafion/PAN nanofibers

Using the electrospinning parameters optimized above, Nafion/ PAN (40: 60 w/w) nanofibers as the UTLC stationary phase were successfully fabricated as shown in figure 11. Compared to the fiber diameter of pure PAN (400 ± 50 nm), Nafion/ PAN nanofibers have average diameter of 246 ± 19 nm. The reduced diameter is mainly due to the addition of Nafion in the electrospinning solution owing to its ionic conductivity. The thickness of the stationary phase can be controlled by simply varying the electrospinning time. When Nafion/ PAN solution was electrospun for 10 min, a film thickness of 10 µm was produced (figure 12), which falls within the range of the typical thickness of UTLC plates.

Figure 11. SEM image of electrospun Nafion/ PAN (40: 60, w/w) under optimized condition: flow rate= 0.5 mL/h, voltage= 12 kV, distance= 15 cm, RH= 15%.

Figure 12. SEM image of stationary phase thickness electrospun for 10 min.

3.2 Nanofiber stability in mobile phases

In order to confirm the stability of the stationary phase for use in UTLC,

Nafion/PAN fiber mats were immersed in the optimized mobile phase mixtures for 15 min, 30 min, 45 min and 60 min. After each period, no change in fiber diameters was observed. Generally, the migration time of each mobile phase on Nafion/ PAN plates was within 1 h. Figure 13 shows there is no swelling or fiber dissolution after soaking Nafion/

PAN (40:60 w/w) fiber mats in the optimized mobile phase mixtures for 60 min. This is ascribed to the superb chemical resistance of the Teflon backbone of Nafion. In addition, compared to previously reported PVA UTLC for amino acids, which required crosslinking treatment to avoid mat fusion and mass loss when it was exposed to mobile phases, PAN is insoluble and infusible in water and common organic solvents expect

DMF, dimethyl acetamide (DMAA), dimethyl sulfoxide (DMSO), etc.13,61 The excellent chemical stability of Nafion/ PAN make separations even more reliable and effective.

3.3 Separation of amino acids

3.3.1 Optimization of mobile phase

The separation of amino acids plays an important role in biological research. By determining the amino acid sequence, the composition and structure of a protein can be investigated. Analyses of amino acids in physical fluids and tissues are also frequently applied in the biological field.62 Moreover, to confirm food values, it is essential to determine the free amino acids in the food. As small biological molecules, amino acids were selected to ascertain the feasibility of electrospun Nafion/ PAN nanofibers as the

UTLC stationary phase.

Since the sulfonate groups on Nafion work as strong cation exchangers, buffer was used for the mobile phase. As the majority of the amino acids have an isoelectric point (pI) value larger than 5 except aspartic acid and glutamic acid, in ammonium acetate buffer with pH of 4, they would carry positive charges and be readily adsorbed onto the negatively charged UTLC surface. Due to the hydrophobic property of the

Nafion/PAN plate, an aqueous buffer cannot migrate on the plate without a less polar organic modifier. Organic solvents including acetonitrile, tetrahydrofuran, methanol, ethanol, 1-propanol, 2-propanol were evaluated as modifiers in the aqueous buffer mobile phase for the separation of amino acids. All of these binary mobile phases exhibited tailings and very poor selectivity among amino acids. Due to the limited miscibility of n- butanol in aqueous buffer, the equal volume of methanol and butanol was added to the buffer to create a fully miscible mobile phase. This ternary solvent mixture containing acetate buffer, butanol and methanol was found to provide relatively better selectivity for

36 amino acids. In order to further optimize the mobile phase composition, the proportion of acetate buffer (10 mM) in the ternary mobile phase was varied in the separation of amino acids including arginine (Arg), lysine (Lys), proline (Pro) and valine (Val) (figure 14).

Meanwhile,. The retardation factor Rf was calculated using equation 1:

풁풔 푹풇 = (1) 풁풇 where Zs is the distance traveled by the analytes, and Zf is the distance traveled by the solvent front.2

With only 20% acetate buffer in the mobile phase, Pro and Val only migrated very short distance while Arg and Lys barely moved up from the origin. There’s no improvement in migration distance for Pro and Val, but Arg and Lys started to move up when acetate buffer content was raised up to 40%. Nevertheless, the selectivities between

Pro and Val, Arg and Lys were still poor. Good selectivity among all of these amino acids was observed when 60% or more acetate buffer was used. As shown in figure 14, the retention order and migration distance for each amino acid was almost same for 60% and 80% acetate buffer content. However, 60% acetate buffer provided more reproducible Rf values with smaller % RSD. In addition, due to the hydrophobic nature of the Nafion/ PAN plate, with 80% acetate buffer, where only 20% organic modifier was used, the migration time of mobile phase mixture was very long (~1 h). The slow mobile phase velocity contributed to band broadening, therefore a poorer separation efficiency was observed (figure 15). Separation efficiency is described by theoretical plate number

N which was calculated by equation 2:

풁 ퟐ 푵 = ퟏퟔ ( 푺) (2) 풘

2 where, Zs is the distance migrated by the analyte, and w is the developed spot width. A mixture composed of 60:20:20 ammonium acetate buffer/ butanol/ methanol (v/v/v) was determined as the optimum mobile phase as it offered the best separation efficiency among all the compositions, as shown in figure 15, and it was used for all other studies with amino acids in this work.

Figure 14. Retardation factors of Arg (▲), Lys (■), Pro(◆) and Val (●) as a function of different % acetate buffer in the mobile phase. The volume ratio of butanol and methanol was 1:1.

Figure 15. Comparison of plate number, N of Arg (■), Lys (■), Pro (■) and Val (■) using different % acetate buffer in the mobile phase. The volume ratio of butanol and methanol was 1:1.

3.3.2 Comparison of Nafion/PAN UTLC and commercial ion exchange TLC

As amino acids can be separated on an ion exchange plate based on the difference in their charges, using the optimized mobile phase with a pH of 4.0, amino acids with different pI values were expected to migrate different distances. Therefore, several basic, neutral and acidic amino acids were selected to test the performance of the Nafion/PAN stationary phase and to compare with the separation that had been realized on the commercial ion exchange TLC. The structures, pI values and their charges in the acetate mobile phase are listed in table 2. The acidic amino acid aspartic acid (Asp) would readily exist as an anion in the mobile phase, and repulsion from the negatively charged stationary phase was expected. Neutral amino acids valine (Val) and phenylalanine (Phe), which carried one positive charge in the mobile phase, would be retained by sulfonate groups on the stationary phase. Three basic amino acids arginine (Arg), lysine (Lys) and histidine (His) with doubly positive charges, were expected to be mostly retained on the

Nafion/PAN plate. The retardation factors of these amino acids separated on the

Nafion/PAN plate were calculated and compared with the results generated on commercially available ion exchange TLC (IE-TLC) plates produced from the Dowex

32 50X8 type resin in table 3. The Rf values of amino acids on commercial IE-TLC were calculated by measuring the migration distance of each amino acid and solvent front on the photograph provided in literature,32 therefore no relative standard deviation was available.

From table 3, as expected, Asp has the largest retardation factor, followed by Val and Phe, and lastly by the basic amino acids Lys, His, Arg. It is worth noting that good

41 selectivity among the amino acids with the same amount of charges was also realized on the Nafion/PAN plate, and this is owing to the additional retention mechanism provided by Nafion where hydrophobic character of the stationary phase interacts with the side chain of the amino acids. Overall, the separation of amino acids is predominated by ion exchange mechanism, and hydrophobic interaction allows further separation among equally charged amino acids. With respectively optimized mobile phase, the selectivity on the Nafion/PAN UTLC plate is comparable with the one on commercial IE-TLC; however, the Nafion/PAN plate presents much better resolving power as the solvent front migration distance was only 3 cm for the Nafion/ PAN plate, whereas the commercial plate required 18 cm of solvent migration for the comparable resolution to occur. Here, the separation efficiency can be illustrated by using plate height, H, which is obtained from equation 3:

풁 푯 = 풔 (3) 푵 where Zs is the distance of the analyte migration and N is the plate number. Plate height measures the efficiency per unit length of the migration distance for a given chromatographic system.63 Therefore, a smaller H value means a more efficient separation system with a larger plate number. Even though the retardation factors are similar for amino acids on both the Nafion/PAN plate and the commercial IE-TLC, the calculated H values on the Nafion/PAN plate are remarkably lower than those on the commercial IE-TLC (figure 16). Error bars were not provided in the literature for the commercial plate.

Amino acid Structure pI Charges in pH 4.0

Arginine (Arg) 10.76 2+

Lysine (Lys) 9.74 2+

Histidine (His) 7.59 2+

Valine (Val) 5.96 1+

Phenylalanine (Phe) 5.48 1+

Aspartic acid (Asp) 2.77 1-

Table 2. Structures, pI values and charges in pH 4.0 of selected amino acids.

Amino acids Retardation factor, Rf (% RSD)

Nafion/PAN UTLC Commercial IE-TLC32

Arg 0.15 (3.0) 0.04

His 0.19 (0.6) 0.14

Lys 0.24 (2.9) 0.08

Phe 0.38 (2.4) 0.52

Val 0.53 (0.6) 0.77

Asp 0.90 (0.3) 0.91

Table 3. Retardation factors of amino acids separated on the Nafion/PAN UTLC plate and on the commercial IE-TLC reported in literature.

Figure 16. Comparison of the plate height on the Nafion/PAN UTLC plate (■) and the commercial IE-TLC plate (■).

3.3.3 Mobile phase velocity

Electrospinning produces nonwoven nanofibers with pores formed by randomly oriented fibers lying loosely on top of each other.64 Porous stationary phase can be modelled as capillaries with sufficiently small diameter, where the liquid is lifted against gravity due to the pressure difference of cohesion and adhesion forces between the liquid and the capillary wall. This kind of solvent migration which does not require assistance of external forces is known as capillary flow, a classical TLC method.27 The mobile phase transport through the porous layer by capillary action can be described by the following relationship:

풁풇 = √к풕 (4) where, Zf is the total distance moved by the solvent front from the origin, к is the flow constant and t is the development time. Therefore, the mobile phase velocity can be determined by plotting migration distance versus the square root of time. The velocity constant к is also related to the properties of the stationary and mobile phases:

к = ퟐ푲ퟎ풅풑(휸⁄ƞ) 풄풐풔 휽 (5) where K0 is the permeability constant, dp is the average particle diameter of the stationary phase, Ƴ is the surface tension of the mobile phase, ƞ is the viscosity of the mobile phase and Ɵ is the contact angle between mobile and stationary phases. In this study, the optimized mobile phase for amino acids was used on both the pure PAN and Nafion/

PAN plates to allow the comparison under the same condition. Figure 17 shows the variation of the migration distance as a function of the square root of time for the pure

PAN and the Nafion/ PAN plate, and the slope is к0.5 based on equation 4. By comparing 46 the slopes, the Nafion/PAN plate exhibited slightly slower mobile phase transport than the pure PAN plate. Equation 5 indicates that the mobile phase velocity is directly proportional to the particle diameter. The fiber diameters of Nafion/ PAN and pure PAN used in this work were 246 ± 19 nm and 400 ± 50 nm, respectively; meanwhile, the layer with thinner fibers was less permeable to solvents, hence the K0 term was smaller for

Nafion/PAN plate. The decrease in fiber diameter was unavoidable by adding a conductive material into the electrospun solution.65 When the mobile phase contained a considerable amount of aqueous buffer (60%), the contact angle between mobile phase and stationary phase became larger with the increasing hydrophobicity of the stationary phase; therefore, cosƟ term was also smaller for the more hydrophobic Nafion/PAN plate compared to pure PAN. The decreased mobile phase velocity was observed as a result of decreases in fiber diameter and permeability, and an increase in contact angle between the mobile phase and the Nafion/PAN stationary phase. Nevertheless, figure 17 shows a

2 0.5 good linearity (R > 0.99) between Zf and t , which verifies the applicability of equation

4, and also demonstrated the homogeneity of both pure PAN and Nafion/PAN stationary phases.

Figure 17. Comparison of mobile phase migration rate on the pure PAN (●) and the

Nafion/PAN (▲) UTLC plates.

3.3.4 Band broadening

As previously stated, the plate number N is inversely proportional to the square of developed spot width. Therefore, the band broadening in TLC can be expressed by plate number N or plate height H. The goal of efforts to improve the performance of TLC is to obtain small H values and maximum N values.1 Phenylalanine was selected to examine the changes in plate height as functions of solvent migration distance on the Nafion/PAN

UTLC plate. As shown in figure 18, the plate height H exhibited a decreasing trend in 5-

15 mm, remained approximately the same from 15 to 35 mm, and drastically increased at

40 mm. The large values of H in the range of 5-15 mm are due to the small migration distance of the analyte (Zs), and the increases of H after 35 mm is the result of band broadening which is mainly caused by the decreased solvent migration rate as the solvent moves up. Consequently, the optimum migration distance on the Nafion/PAN plates is determined to be 35 mm or less.

One of the most widely accepted chromatographic relationships, the van Deemter equation, describes the dependence of plate height H on the individual factors that lead to band broadening.7,66 The van Deemter equation is expressed as:

푯 = 푨 + 푩⁄풖 + 푪풖 (7) where H is the plate height, u is the mobile phase velocity. In the above equation, the A term is resulted from the multipath flow taken by sample solutes when they travel through the sorbent layer; the B term is the contribution to the broadening from diffusion along the flow direction, i.e. the longitudinal diffusion in mobile phase; and the C term accounts for the broadening resulting from the resistant to mass transfer in the stationary

49 phase and mobile phase. The A term is highly dependent on the layer homogeneity which is determined by the distribution of particle sizes and shapes, the packing density as well as the presence of additive in the layer such as visualization indicators or binders.7 The effect of multipath diffusion is directly proportional to fiber diameters. Smaller diameters promoting smoother flow result in less band broadening. For fine particles with diameters smaller than 10 µm, the A term is fairly small and it can be negligible.67 The B term is inversely proportional to mobile phase velocity as shown in the van Deemter equation.

As a result, the B term increases with decreasing mobile phase velocity. In TLC, under capillary forces, the mobile phase velocity gradually slows down as the solvent moves up. Therefore, the contribution of the B term becomes most significant at longer migration distance. The C term directly depends on the mobile phase velocity. Due to the rather slow mobile phase velocity, the C term is also often neglected in TLC.

Furthermore, the C term is directly dependent on the squared particle diameter,7 so this term becomes even more insignificant with the nanoscale Nafion/PAN fibers. As noticed in section 3.3.3, the Nafion/PAN plate exhibits slower mobile phase velocity compared to pure PAN; therefore, the longitudinal diffusion was considered to be the predominant contributor to band broadening. A relatively fast mobile phase velocity is desired in order to reduce the longitudinal diffusion, and this can be realized by properly adjusting the fiber size.

Figure 18. Change in plate height H of phenylalanine with increasing solvent migration distance on Nafion/PAN UTLC plate.

3.4 Separation of proteins

Protein separations are most frequently conducted by using high-resolution techniques such as SDS-PAGE, capillary electrophoresis (CE), high performance liquid chromatography (HPLC) and column liquid chromatography with different chromatographic modes including size-exclusion, ion exchange, hydrophobic interaction, reversed phase and isoelectrofocusing.1 Even though TLC has been developed with most of the chromatographic modes that are used in column liquid chromatography, the applications of TLC are limited to separations of small molecules. Other than size- exclusion mode, only a very small number of studies have reported the separations of proteins using TLC.68 Ion exchange chromatography is the most popular method for protein separations in columns; however, there is scant literature in protein separations using ion exchange TLC (IE-TLC). Human hemoglobins were initially tested on IE-TLC made of carboxymethylcellulose (CM-cellulose) in 1970s, and later in 1990s, another paper studied protein separation on IE-TLC with diethyl aminoethyl (DEAE) cellulose as

69,70 the sorbent. Several researchers have reported the application of electrospun nanofibers with immobilized functionalities as protein affinity membranes.71,72,73,74,75

Electrospun nanofibers have particularly attractive properties as the TLC sorbent including the high surface area to volume ratio, high porosity and easy accessibility of surface active sites which allow sufficient interaction between analytes and stationary phase.

Herein, protein separations on ion exchange and reversed-phase mixed-mode

UTLC using electrospun Nafion/PAN nanofibers are described. Efforts were made to

52 understand the retention mechanisms of each protein in relation to its size, charge and structure. The influence of the buffer pH, salt concentration and the percentage of organic modifier on the chromatographic behaviors of proteins were also explored.

3.4.1 Visualization Reagent

Amino acids were easily visualized using ninhydrin spray and drying procedure; unfortunately, the sensitivity of ninhydrin for protein detection was found to be extremely low. No spots were observed with protein concentration even up to 5 mg/mL. This is due to the fact that the colors are generated by reactions between ninhyrin and the free amino groups. There is at least one free amine on each amino acid, so abundant free amines will be available in a certain amount of an amino acid sample. Nonetheless, for a protein that consists of one or more long chains of amino acids, only limited free amino groups present in the same given mass of a protein sample. Previous work has included fluorescence indicator green 254nm in electrospinning solution and produced fiber mat with fluorescent background and dark analyte spots under UV light.76 This visualization method was found to have interference with ion exhangers on the Nafion/PAN fibers as this fluorescent agent is manganese-doped zinc silicate which would bring in metal cations and silicates as competing ions. Fluorescein isothiocyanate (FITC) labeling was also attempted to visualize the proteins; however, the attachment of FITC to proteins via amino terminals and primary amines can prevent protonation of amino groups in buffer and block the binding to ion exchangers. Morin hydrate was evaluated as visualization reagent; however, the fluorescent spots had very similar yellowish green color with the

53 green fluorescent background, and this made the visualization very irreproducible.

Finally, fluoresscamine was found to give the best fluorescent results using the minimum amount of proteins among all the visualization reagents tested in our lab. Even though fluorescamine is a similar reagent like ninhydrin that also reacts with amino groups, the sensitivity of fluorescamine is much greater than ninhydrin, thus much smaller quantities of proteins can be detected using this reagent.77 The fluoresamine procedure was described in section 2.7. Green fluorescent spots on a dark background were observed under UV.

3.4.2 Buffer selection

Lysozyme, bovine serum albumin, insulin and α-chymotrypsin were used as model proteins to test the chromatographic performance of the Nafion/PAN UTLC. The different molecular weights and pI values of these proteins (Table 4) would make them have very distinct retention behaviors on this mixed-mode stationary phase. In consideration of ion exchange mode, buffer with appropriate concentration and pH values is essential for separation of charged molecules. The net charges for each protein under different pH values were calculated and plotted in figure 20. For BSA, its net charge changes drastically in the pH range of 2-12, whereas the net charges of LYZ, INS and

CHY only have very slight changes at various pH conditions, and the net charge values of these three proteins are very close to each other in the evaluated pH range. By keeping the percentage of organic modifier (acetonitrile) constant (25%), acidic buffers with pH values of 3.0-5.7 such as ammonium acetate buffer and citrate buffer were tested because

54 in this pH range, these four proteins carry different amounts of positive charges, especially for BSA. All of the proteins were strongly retained under these pH conditions even with salt concentration up to more than 1M, where crystallization was observed on the plate as a consequence of high salt concentration. Since the desorption of proteins could not be realized by further increasing the salt concentration in order to avoid the salting out of proteins, attempts to increase the buffer pH values which in turn reduce the net charges on proteins were made. With pH values of 6.0-7.0, MES buffer and sodium phosphate buffer were used. In this pH range, LYZ and CHY would carry smaller net positive charges, meanwhile, the net charge of BSA and INS would become negative; in this manner, negatively charged proteins were expected to migrate farther than the positive ones, and all of the proteins were supposed to get less retained than in acidic buffers. Surprisingly, all of the four proteins with very elongated spots (streaking) which started from the origin of application were obtained. Compact spots for BSA, INS and

CHY were not observed until the buffer pH was adjusted to above 7.5, while LYZ still did not get rid of elongated spots when the pH increased even up to 9.0. Typically, the elongated streaks are the result of sample overloading. Taking that in consideration, the extent of streaking was evaluated as a function of the concentration of LYZ from 0.5 mg/ml to 5mg/ml in the same solvent system. The degree of streaking was not reduced by decreasing the concentration of LYZ because the length of the elongated spot remained the same, and the only difference observed with the decreased concentration was the diminished color intensity of fluorescent spot. When the LYZ concentration was lower than 1 mg/ml, no fluorescent spot could be detected using fluorescamine spray procedure.

Therefore, in this case, streaking problem cannot be solved by minimizing the sample loads only if a visualization reagent with an increased detection limit was used to investigate whether reduced streaking could be obtained with the sample concentration lower than 1 mg/ml. The elongated spots were also observed by Lepri et al. on silanized silica gel TLC impregnated with ion exchangers for the separation of polypeptides. They concluded that with insufficient salt concentration, hydrogen bonding and/or columbic interactions are responsible for streaking.78 As a very basic protein, the net charge of

LYZ remains positive until the buffer pH becomes extremely basic (> 10), in which the hydrolysis of protein can take place and give rise to several spots on the layer. Figure 21 shows the chromatogram of four proteins separated using bicine buffer (50 mM, pH=9.0) with 0.4 M NaCl and 40% acetonitrile. LYZ was the most retained protein with an elongated spot under this condition. Unexpectedly, the least retained protein was CHY rather than BSA or INS, which carried more negative charges, and the retention order of

BSA and INS was also unanticipated based on the number of charges. Even though the separation of amino acids was predominated by ion exchange mechanism, the retention of proteins on this mixed-mode stationary phase cannot be explained solely on the basis of ion exchange since it also depends on the position of charged sites on the side chains and the hydrophobicity of the proteins. Consequently, the retention behavior of each protein was explored in details by varying the buffer pH, salt concentration and the percentage of organic modifier.

Protein Abbreviation Molecular Weight (kDa) pI

Bovine serum albumin BSA 66 4.7

Insulin INS 5.73 5.3

α-Chymotrypsin CHY 25 8.75

Lysozyme LYZ 14.3 11.35

Table 4. Abbreviations, molecular weights and pI values of the proteins used in this work

Figure 19. The calculated relationship between protein net charge and the buffer pH of

BSA (◆), INS (■), LYZ (✕) and CHY (▲).

Figure 20. Chromatogram for separation of 1) LYZ; 2) BSA; 3) INS; 4) CHY on electrospun Nafion/ PAN UTLC using with 0.4 M NaCl in bicine buffer (pH=9.0, 50 mM) with 40% acetonitrile.

3.4.3 Effect of buffer pH

Retention of the proteins at various pH values was studied keeping the sodium chloride concentration constant at 0.4 M and acetonitrile at 40%. As stated above, when the pH was lower than 7.5, all of the proteins exhibited as elongated spots thus the retardation factors could not be measured precisely. In addition, alkaline eluents with pH values above 10 could not be used due to the unwanted hydrolysis of proteins. This study was conducted within the working pH range of bicine buffer. From figure 22, an increasing trend in retardation factor of each protein was observed as the pH increased while the retention order of four proteins remained the same. In this pH range, BSA and

INS became negatively charged, which would get repelled by the negative sulfonate groups on the sorbent surface; and the number of negative charges increased at a higher pH, hence the repulsive force became greater resulting in less retention of these proteins.

However, CHY, which bore relatively more positive charges than BSA and INS, was least retained in this pH range. LYZ, which carried the most positive charges among all the proteins, was most retained and elongated spots were obtained throughout the pH range. Based on these observations, assumptions were made that the retention of CHY and LYZ were highly dependent on their charges, and interactions other than electrostatic attraction and repulsion played a more important role on the retentions of BSA and INS.

Figure 21. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of buffer pH with 0.4 M NaCl and 40% acetonitrile.

3.4.4 Effect of salt concentration

The influence of salt concentration was evaluated with 40% acetonitrile concentration in bicine buffer with pH at 8.5. The retardation factor of each protein was plotted as a function of salt concentration in figure 22. Proteins CHY and LYZ got less retained as more and more counter ions were introduced. The effect of salt concentration was more obvious on LYZ than on CHY because the net charge on LYZ was more than that on CHY. This observation can be validated by the relationship between net charge and protein migration velocity. Distribution coefficient K, which is the ratio of the sample concentration in stationary phase to the sample concentration in mobile phase, can be used to study the migration velocity of a protein based on the relation shown in equation 8:

풅풛 풖 풑 = (8) 풅풕 ퟏ+푯풄푲

Where dzp /dt is the migration velocity of a sample zone, u is the linear mobile phase velocity, Hc is a constant and K is the distribution coefficient. K is highly dependent on the net charge of a protein. When the number of net charge is relatively large, a slight increase in salt concentration can result in a great reduction in distribution coefficient K, in other words, more proteins desorbed from the stationary phase, therefore a sharp increase in migration velocity of a protein.79 On the other hand, INS presented a decreasing trend as the salt concentration increased. Bearing the previous assumption in mind, the retention of INS was most likely dominated by hydrophobic interaction on this plate. The decreased retention of INS was the result of a stronger hydrophobic interaction between INS and the stationary phase promoted by increasing salt concentration, because 62 at a high salt concentration the solvation of protein molecule got reduced thus the hydrophobic regions were exposed to the stationary phase.80 Salt concentration had minimal impact on the retention behavior of BSA as very slight change in retardation factor was observed. This might be due to the fact that the retention of BSA relied on hydrogen bonding, ionic interactions and hydrophobic interactions, and the strengthened hydrophobic interactions between BSA and the stationary phase was counteracted by the reduced hydrogen bonding and electrostatic interactions as the salt concentration increased.78

Figure 22. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of salt concentration in bicine buffer with pH at 8.5 and 40% acetonitrile.

3.4.5 Effect of organic modifier concentration

The study of the effect of organic modifier concentration was conducted using mobile phase containing 0.4 M NaCl in bicine buffer with pH 9.0. The retardation factors of four proteins with various acetonitrile concentration are shown in figure 24. Due to the super hydrophobicity of the Teflon backbone on Nafion, proteins were expected to unfold during the interaction with the stationary phase. Therefore, each protein was studied in its unfolded form. Figure 25 shows the Kyte-Doolittle hydrophobicity plot of the four proteins, and the values define the hydrophobicity of each amino acid residue on a protein. The more positive value on the plot indicates the more hydrophobic the region where the amino acid located in the protein. The hydrophobicity scales of amino acids were calculated by Kyte-Doolittle using both the water vapor transfer free energies and the interior-exterior distribution of amino acid side-chains.81 As shown in figure 25, LYZ,

BSA and CHY are comprised of less hydrophobic amino acids, in contrast, INS has very significant hydrophobicity. This explained the observations and assumptions on the retention mechanism of INS made previously. The retention order of INS and CHY altered when the percentage of acetonitrile became greater as seen in figure 24, and this is another evidence of the fact that hydrophobic interaction is the dominant mechanism for the retention of INS. Noteworthy, in the acetonitrile assay study, elongated spots were observed for BSA at concentration of acetonitrile higher than 40%. The increase in hydrogen bonding and electrostatic repulsion between BSA and the stationary phase in the mobile phase with less buffer content was responsible for the elongated spots, even though the hydrophobic interaction was lessened. Moreover, the greater retention of BSA

65 compared with CHY and INS was attributed to its large molecular size. As an unfolded protein, the large molecule BSA has more chance to form extremely strong binding through multipoint attachments (hydrogen bonds) with the sorbent layer than the smaller molecules, even though CHY and INS have more basic pI values which were expected to be retained more than the less basic BSA on the cation exchange TLC.82

Figure 23. Retardation factors of LYZ (▲), BSA (■), INS (◆) and CHY (●) as a function of the organic modifier percentage in bicine buffer with pH at 9.0 and 0.4 M

NaCl.

Figure 24. Kyte-Doolittle hydrophobicity plots of A) LYZ; B) BSA; C)CHY; D) INS.

Chapter 4 Conclusion

In this study, Nafion/PAN nanofibers were successfully fabricated via the electrospinning method. To the best of our knowledge, it is the first work that

Nafion/PAN nanofibers have been used as the stationary phase of UTLC which provides mixed-mode of ion exchange and reversed phase retention mechanisms. The Nafion/PAN fiber morphology was highly dependent on the properties of electrospinning solution, solution flow rate, applied voltage and the distance between syringe and collector. Bead- free and uniform nanofibers obtained using optimized electrospinning parameters exhibited excellent mechanical stability and solvent compatibility. The separation of amino acids confirmed the feasibility of Nafion/PAN nanofibers as the UTLC stationary phase. The ion exchange mode predominated for small molecules like amino acids.

Combined with the hydrophobic interactions, high selectivity were obtained among amino acids. The Nafion/PAN UTLC demonstrated better resolution and substantially enhanced efficiency for amino acids compared with commercial IE-TLC. The relatively slower mobile phase velocity compared to the pure PAN was resulted from the reduced fiber diameter by adding conductive Nafion into PAN solution. The slow mobile phase velocity on the Nafion/PAN plate gave rise to the longitudinal diffusion, which was the major contributor to band broadening. The separation of proteins has proved that

Nafion/PAN nanofibers are also suitable for separation of large biomolecules. The retentions of proteins on the Nafion/PAN UTLC largely depend on the properties of

69 proteins including the net charge, hydrophobicity, molecular size and structure. The

Nafion/PAN plates displayed high affinity towards positively charged proteins. Elongated spots observed were ascribed to the strong hydrogen bonding and electrostatic interactions. The streaking problem can be solved by using an appropriate solvent system and possibly by decreasing the sample loads with a highly sensitive visualization reagent.

References

1. Sherma, J.; Fried, B. Handbook of Thin-Layer Chromatography, 3rd ed.; Mercel

Dekker: New York, U.S.A., 2013.

2. Poole, C. F. The Essence of Chromatography; Elsevier Science: Amsterdam,

Netherlands, 2003.

3. Zhao, C.; Xiong, Y.; Yu, L.; Li, S.; Wu, Z. J. Chromatogr. Sci. 2010, 48, 685-689.

4. de Andrade, F. I.; Florindo Guedes, M. I.; Pinto Vieira, I. G.; Pereira Mendes, F.

N.; Salmito Rodrigues, P. A.; Costa Maia, C. S.; Marques Avila, M. M.; de Matos

Ribeiro, L. Food. Chem. 2014, 157, 193-198.

5. Shewiyo, D. H.; Kaale, E.; Risha, P. G.; Dejaegher, B.; Smeyers-Verbeke, J.;

Vander Heyden, Y. J. Pharm. Biomed. Anal. 2012, 66, 11-23.

6. Le Roux, A. M.; Wium, C. A.; Jouber, J. R.; Van Jaarsveld, P. P. J. Chromatogr.

1992, 581, 306-309.

7. Spagenberg, B. Quantitative Thin-Layer Chromatography; Springer: New York,

U.S.A., 2011.

8. Sherma, J. Anal. Chem. 1988, 60, 74 R-86 R.

9. Bezuidenhout, L. W.; Brett, M. J. J. Chromatogr. A 2008, 1183, 179-185.

10. Fried, B.; Sherma, J. Thin-Layer Chromatography, 4th ed.; Marcel Dekker: New

York, U.S.A., 1999.

11. Clark, J. E.; Olesik, S. V. Anal. Chem. 2009, 81, 4121-4129.

12. Clark, J. E.; Olesik, S. V. J. Chromatogr. A 2010, 1217, 4655-4662.

13. Lu, T.; Olesik, S. V. J. Chromatogr. B 2013, 913, 98-104.

14. Newsome, T. E.; Olesik, S. V. Anal. Chem. 2014, 86, 10961-10969.

15. Fang, X.; Olesik, S. V. Anal. Chim. Acta 2014, 830, 1-10.

16. Ahn, Y. C.; Park, S. K.; Kim, G. T.; Hwang, Y. J.; Lee, C. G.; Shin, H. S. Curr.

Appl. Phys. 2006, 6, 1030–1035.

17. Wei, G.; Liu, J.; Zhao, H.; Yan, C. J. Power Sources 2013, 241, 709-717.

18. Ding, B.; Wang, M.; Yu, J.; Sun, G. Sensors 2009, 9, 1609-1624.

19. Gopal, R.; Kaur, S.; Ma, Z.; Chan, C.; Ramakrishna, S.; Matsuura, T. J. Membr.

Sci. 2006, 281, 581-586.

20. Liang, D.; Hsiao, B. S.; Chu, B. Adv. Drug Deliv. Rev. 2007, 59, 1392–412.

21. Ramakrishna, S.; Fujihara, K.; Teo, W.; Lim, T. C.; Ma, Z. An Introduction to

Electrospinning and Nanofibers; World Scientific Publishing: Toh Tuck Link,

Singapore, 2005.

22. Li, D.; Xia, Y. Adv. Matter. (Weinheim, Ger.) 2004, 16, 1151-1170.

23. Huang, Z. M.; Zhang, Y-Z.; Kotai, M.; Ramakrishna, S. Compos. Sci. Technol.

2003, 63, 2233-2253.

24. Burger, C.; Hsiao, B. S.; Chu, B. Annu. Rev. Mater. Sci. 2006, 36, 333-368.

25. Haddad, P. R.; Jackson, P. E. Ion Chromatography-Principles and Applications;

Elsevier Science: Amsterdam, Netherlands, 1990.

26. Cummins, P. M.; Dowling, O.; O’Connor, B. F. Protein Chromatography

Methods and Protocols; Springer: New York, U.S.A., 2011; pp215-228.

27. Acikara, O. B. Ion-Exchange Chromatography and Its Applications, Column

Chromatography, Dr. Dean Martin ed.; InTech, DOI: 10.5772/55744, 2013.

28. Bonnerjea, J.; Oh, S.; Hoare, M.; Dunnill, P. Biotechnol. J. 1986, 4, 954-958.

29. Velickovic, T. C.; Ognjenovic, J; Mihajlovic L. Ion Exchange Technology II;

Springer: Dordrecht, Netherlands, 2012; pp1-34.

30. Kisler, J. M.; Dahler, A.; Stevens, G. W.; O’Connor, A. J. Microporous

Mesoporous Mater. 2001, 44-45, 769-774.

31. Touchstone, J. C.; Dobbins, M. F. Practice of thin layer chromatography, 2nd ed.;

Wiley: New York, U.S.A., 1978.

32. Devenyi, T. Hung. Sci. Instrum. 1974, 30, 13-22.

33. Grot, W. G. U.S. Patent 3, 770, 567, 1973.

34. Dong, B.; Gwee, L.; Cruz, D.; Winey, K.; Elabd, Y. Nano Lett. 2010, 10, 3785-

3790.

35. Mauritz, K. A.; Moore, R. B. Chem. Rev. 2004, 104, 4535–4585.

36. Molla, S.; Compan, V. J. Membr. Sci. 2011, 372, 191-200.

37. Lei, S.; Chen, D.; Chen, Y. Nanotechnology. 2011, 22, 1-7.

38. Madou, M. J.; Morrison, S. R. Chemical Sensing with Solid State Devices;

Academic Press: London, U.K., 1989.

39. Zhou, C.; Liu, Z.; Dai, J.; Xiao, D. Analyst 2010, 135, 1004-1009.

40. Chen, H.; Snyder, J.; Elabd, Y. Macromolecules 2008, 41, 128–135. 73

41. Zhao, J.; Yuan, W. Z.; Xu, A.; Ai, F.; Lu, Y.; Zhang, Y. React. Funct. Polym.

2011, 71, 1102-1109.

42. Sharma, D. K.; Li, F.; Wu, Y. Colloids Surf. A 2014, 457, 236-243.

43. Ballengee, J.; Pintauro, P. J. Electrochem. Soc. 2011, 158, B568-B572.

44. Laforgue, A.; Robitaille, L.; Mokrini, A.; Ajji, A. Macromol.Mater. Eng. 2007,

292, 1229-1236.

45. Nah, C.; Lee, Y-S.; Cho, B. H.; Yu, H. C.; Akle, B.; Leo, D. J. Compos. Sci.

Technol. 2008, 68, 2960-2964.

46. Tran, C.; Kalra, V. Soft Matter 2013, 9, 846-852.

47. Sharma, D. K.; Shen, J.; Li, F. RSC Adv. 2014, 4, 39110-39117.

48. Hu, J.; Baglio, V.; Tricoli, V.; Arico, A.; Antonucci, V. J. Appl.Electrochem.

2008, 38, 543-550.

49. Deluca, N.; Elabd, Y. J. Membr. Sci. 2006, 282, 217-224.

50. Felix, A. M.; Jimenez, M. H. J. Chromatogr. 1974, 89, 361-364.

51. Ma, C.; Yu, T.; Lin, H.; Huang, Y.; Chen, Y.; Jeng, U.; Lai, Y.; Sun, Y. Polymer

2009, 50, 1764-1777.

52. Rayleigh, L. Philos. Mag. 1882, 14, 184-186.

53. Mit-uppatham, C.; Nithitanakul, M.; Supaphol, P. Macromol. Chem. Physic.

2004, 205, 2327-2338.

54. Yuan, X.; Zhang, Y.; Dong, C.; Sheng, J. Polym. Int. 2004, 53, 1704-1710.

55. Lee, J. S.; Choi, K. H.; Ghim, H. D.; Kim, S. S.; Chun, D. H.; Kim, H. Y.; Lyoo,

W. S. J. Appl. Polym. Sci. 2004, 93, 1638-1646. 74

56. Buchko, C. J.; Chen, L. C.; Shen, Y.; Martin, D. C. Polymer 1999, 40, 7397-7407.

57. Megelski, S.; Stephens, J. S.; Chase, D. B.; Rabolt, J. F. Macromolecules 2002,

35, 8456-8466.

58. Deitzel, J. M.; Kosik, W.; McKnight, S. H.; Tan, N. C. B.; DeSimone, J. M.;

Crette, S. Polymer 2002, 43, 1025-1029.

59. Zhong, X. H.; Kim, K. S.; Fang, D. F.; Ran, S. F.; Hsiao, B. S.; Chu, B. Polymer

2002, 43, 4403-4412.

60. Krishnappa, R. V. N.; Sung, C. M.; Schreuder-Gibson, H. Mat. Res. Soc. Symp.

Proc. 2002, 702, 671-676.

61. Iovleva, M. M.; Smirnova, V. N.; Budnitskii, G. A. Fibre Chem. 2001, 33, 262-

264.

62. Khan, A. S.; Faiz, F. CODEN. JNSMAC. 2008, 48, 1-17.

63. Srivastava, MM. High-Performance Thin-Layer Chromatography (HPTLC);

Springer: Verlag, Heidelberg, 2011.

64. Li, W.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. J. Biomed. Mater.

Res. 2002, 60, 613-621.

65. Laurencin, C. T.; Nair, L. S. Nanotechnology and Regenerative Engineering: The

Scaffold, 2nd Ed.; Taylor and Francis Group: Boca Raton, U.S.A., 2014.

66. Carr, P. W.; Sun, L. J. Microcolumn Sep. 1998, 10, 149-152.

67. Geiss, F. Fundamentals of Thin Layer Chromatography; Huethig: Heidelberg,

Germany, 1987.

68. Bhushan, R.; Machesh, V. K.; Mallikharjun, P. V. Biomed. Chromatogr. 1989, 3,

95-104.

69. Schroeder, W. A.; Nelson, N. C. J. Chromatogr. A 1975, 115, 527-533.

70. Luo, Q.; Andrade, J. D.; Caldwell, K. D. C A 1998, 816, 97-105.

71. Schneiderman, S.; Zhang, L.; Fong, H.; Menkhaus, T. J. J. Chromatogr. A 2011,

1218, 8989-8995.

72. Ma, Z. W.; Kotaki, M.; Ramakrishna, S. J. Membr. Sci. 2005, 265, 115−123.

73. Ma, Z. W.; Kotaki, M.; Ramakrishna, S. J. Membr. Sci. 2006, 272, 179-187.

74. Ma, Z. W.; Ramakrishna, S. J. Membr. Sci. 2008, 219, 23-28.

75. Zhu, J.; Sun, G. Appl. Mater. Interfaces 2014, 6, 925−932.

76. Owens, C. N. Investigation into polymer and carbon nanomaterial separations.

Ph.D. Dissertation, The Ohio State University, Columbus, OH, August 2012.

77. Creighton, T. E. Proteins: Structures and Molecular Properties, 2nd Ed.; W. H.

Freeman and Company: New York, U.S.A., 1993.

78. Lepri. L.; Desideri, P. G.; Heimler, D. J. Chromatogr. 1981, 211, 29-37.

79. Yamamoto, S.; Nakanishi, K.; Matsuno, R. Ion Exchange Chromatography of

Proteins; Marcel Dekker: New York, U.S.A., 1988.

80. Thoffeq, M.; Fatima, R.; Reddy, B. C. V. Int. J. LifeSc. Bt & Pharm. Sci. 2015, 1,

86-97.

81. Kyte, J.; Doolittle, R. F. J. Mol. Biol. 1982, 157, 105-132.

82. Cole, R. D. J. Bio. Chem. 1961, 236, 1369-1371.