Preparation and Characterization of Polymer Nanocomposites with New Techniques Valentina Romeo

Unione Europea

UNIVERSIT DEGLI STUDI DI SALERNO

Fondo sociale europeo

PROGRAMMA OPERATIVO NAZIONALE 2000/2006 Ricerca Scientifica, Sviluppo Tecnologico, Alta Formazione Regioni dellObiettivo 1 Misura III.4 Formazione superiore ed universitaria

Department of Chemical and Food Engineering Ph.D. Course in Chemical Engineering (VI Cycle-New Series)
Preparation and Characterization of Polymer Nanocomposites with New Techniques

Supervisor Prof. Vittoria Vittoria Scientific Referees Prof. Vincenzo Tucci Dott. Angelo Ferraro Ph.D. Course Coordinator Prof. Paolo Ciambelli

Ph.D. student Valentina Romeo

To my family

Publications List
G.Gorrasi, V.Romeo, D.Sannino, M.Sarno, B. De Vivo, P.Ciambelli*, V. Vittoria*, V.Tucci*. Carbon nanotube induced structural and physical property transitions of syndiotactic polypropylene. Nanotechnology 18 (2007) 275703 (11pp).

-V.Romeo, G.Gorrasi, I.S. Chronakis, V.Vittoria*. Encapsulation and

Exfoliation of inorganic lamellar fillers into Polycaprolactone by electrospinning.Biomacromolecules 8 (2007) 3147-3152. V. Romeo, G.Gorrasi, I.Chronakis, V.Vittoria*. Encapsulation of nanohybrids into Polycaprolactone by electrospinning. Atti del I Congresso Nazionale di Electrospinning (CNES 2007) tenutosi a Milano il 4-5 giugno 2007. V.Romeo, A.Sorrentino, V. Vittoria. Development of nanostructured thermoregulating textile materials. Atti del Congresso Nazionale (NanoItalTex 2007: Nanotecnologie e altre tecnologie emergenti, per il tessile abbigliamento) tenutosi a Milano il 21-22 novembre 2007. A.Sorrentino, V.Romeo, V. Vittoria. Development of nanostructured thermoregulating textile materials. Journal of Nanoscience and Nanotechnology (2008) In Press. A.Sorrentino, V.Romeo, V. Vittoria. Efficient Orientation of Carbon Nanotubes into Polcaprolactone Fibers. Advanced Functional Materials (2008) Submitted.

Table of Contents
Abstract. Objective......................... Introduction.......................................................................
VII IX

1 The Electrospinning process. 6 Electrospinning processing variables 9 Remarkable features of electrospun nanofibers 12 Application of electrospun nanofibers.. 13

Chapter I
Influence of the main process parameters on the final fibers.......... 21 I.1 Introduction. 21 I.2 Process Parameters: Control of Morphology and Diameter 24

Chapter II
Microsphere for nanostructured thermoregulating textile materials 31 II.1 Materials 32 II.2 Analytical techniques 32 II.3 Methods 33 II.4 Nanofibers Characterization 33 II.5 Conclusions 39

Chapter III
Electrospinning carbon nanotube polymer composite nanofibers III.1 Introduction III.2 Experimental Results. III.3 Nanofibers Characterization III.4 Experimental evaluation of the electric conductibility III.5 Conclusion 41 41 49 50 54 57

Chapter IV
Nanocomposites 59 IV.1 Materials.... 61 IV.2 Sample Preparation 62

Table of Contents

IV.3 Methods IV.4 Analysis of Nanocomposites IV.5 Conclusions

62 63
68

Chapter V
Encapsulation and exfoliation of inorganic lamellar fillers into polycaprolactone. V.1 Experimental V.2 Methods.. V.3 Analysis of Nanocomposites V.4 Conclusion 69 70 71 72 81

Chapter VI
Electrospinning of nanohybrids for controlled release in biomedical usage 83 VI.1 Materials................. 85 VI.2 Methods. 87 VI.3 Characterization of nano-hybrid....... 88 VI.4 Antimicrobial Release 95 VI.5 Conclusions.... 98

Conclusions. References..

101 105

II

List of figures
Figure 1 Phenomena of Electrospraying Figure 2 Schematic of Formhals electrospinning setup Figure 3 Schematic of Guignards electrospinning setup... Figure 4 Schematics of electrospinning process.. Figure I.1 SEM images of PCL fibers. Figure I.2 SEM photos Figure I.3 20% PCL in acetone at different distances Figure I.4 SEM images and diameter distributions Figure I.5 SEM images of PCL fibers electrospun. Figure I.6 Graphs of the main parameters... Figure II.1 SEM micrographs of PCM microsphere... Figure II.2 SEM images of PCL fibers... Figure II.3 SEM images of composite PCL fibers-PCM Figure II.4 DSC thermogram of microencapsulated PCM Figure II.5 DSC thermogram of composites .. Figure III.1 Schematic illustration of nanotubes. Figure III.2 Schematic illustration of various carbon nanotubes Figure III.3 (TEM) images of carbon nanotubes. Figure III.4 (TEM) images of carbon nanotube Figure III.5 SEM images of composites with carbon nanotubes.. Figure III.6 SEM micrographs of PCL with MWNT Figure III.7 Circuit for the measurement of tension resistance.. Figure III.8 Course of in operation conductivity... Figure III.9 Course of in operation conductivity. Figure III.10 Comparison of the two techniques. Figure IV.1 Three types of nanocomposite structures.. Figure IV.2 XRD diagram ... Figure IV.3 XRD diagram .. Figure IV.4 SEM micrographs of electrospun fibers. Figure IV.5 Thermal degradation of composites Figure IV.6 Thermal degradation of composites Figure IV.7 DSC of composite samples by electrospinning . Figure IV.8 DCS of composite samples by solvent casting Figure V.1 SEM micrographs of electrospun fibers ..... Figure V.2 Electrospun nanofibers of composites Figure V.3 FESEM equipped with (EDX)... Figure V.4 FESEM equipped with (EDX) 2 3 5 7 23 25 26 27 28 29 34 35 36 38 39 42 43 48 48 51 53 54 55 56 56 60 63 64 65 66 67 67 68 72 74 75 75

List of Figures

Figure V.5 FESEM equipped with (EDX) 76 Figure V.6 X-Ray diffractograms .... 77 Figure V.7 X-Ray diffractograms .... 78 Figure V.8 Thermal degradation.. 79 Figure V.9 Calorimetric curves (DSC) for all the samples.. 80 Figure VI.1 SEM images of electrospun fibers. 89 Figure VI.2 FESEM equipped with an (EDX).. 90 Figure VI.3 X-ray diffraction 91 Figure VI.4 X-ray diffraction patterns 92 Figure VI.5 X-ray diffraction 93 Figure VI.6 Thermal degradation 94 Figure VI.7 Thermal degradation 94 Figure VI.8 Thermal degradation 95 Figure VI.9 The release of composites PCL +3% LDHBz 96 Figure VI.10 The release of composites PCL +3% LDH-pOHBz. 96 Figure VI.11 The release of composites PCL +3% LDHDCB 97 Figure VI.12 The patterns of the releases of composites 98

IV

List of Tables
Table I.1 Electrospinning conditions in different solvents Table III.1 Summary of various nanocomposite fibers. Table IV.1 Organoclays properties. Table V.1 Calorimetric parameters ... 22 46 61 81

Abstract
Nanotechnology is a rapidly growing multidisciplinary field of research incorporating various divisions. Among them nanofiber technology, wherein fibrous materials are fabricated at nano scale, is an emerging field. Materials in nanofiber form not only lead to superior functions but also provide a method to deliver new functions to higher order structures. Electrospinning is an attractive process capable of producing polymeric fibers having diameters ranging from nano to sub micron level. Nanofibers produced by electrospinning can be reinforced by particles to fabricate a composite with unique and tailor made properties. In this thesis, an attempt was made to use polymeric nanofibers as a medium to transfer properties of nanoparticles to polymeric matrices. Several nanoparticles were selected to generate new multifunctional nanocomposites. In particular, the effects of particles on fiber formation were investigated. In general, it was found that the presence of nanoparticles, in addition to molecular weight and solvent concentration, play an important role in governing fiber formation during electrospinning of polymer solutions. Electrospinning shows a high efficiency in the dispersion of the micro and nano filler into a polymer matrix, whereas, differently from other techniques, electrospinning works at room temperature and in normal air conditions, and these conditions may help to protect from harmful treatments, avoiding filler damage. In summary, electrospinning have proved to be a very versatile technique and allows the use of a variety of polymers, blends of different polymers as well as additives and filler, for tailoring different application requirements. This capability opens the door to a practical means of connecting nanostructured materials to macroscopic structures.

Objective
Project objectives are aimed at understanding the effect of nanoparticles addition on the nanofibers obtained with electrospinning technique. To that purpose an experimental setup was designed and some aspects related to the process were investigated with the aim to create nanometer-scale polymer fibers. Processing paprameters, such as concentration, screen distance, needle and type of solvent was evaluated. In this regard we have conducted some experiments on solutions of poly(-caprolactone) varying only one parameter at a time, while maintaining the others. The second is the formation of composites and, therefore, the introduction of micro particles inside the fibers. Accordingly we have directly dispersed into the polymer solution, microsphere like phase-change materials in order to produce thin fibers with diameters ranging from microns down to a few nanometers of great use in many applications, such as in the textiles. Lastly, several poly(-caprolactone)-based nanofibers containing different fillers such as: carbon nanotubes; (layered double hydroxides in carbonate form or LDH organically modified with 12-hydroxydodecanoic acid (LDHHA); different clay such as closite 10A and closite 30B; nano-hybrids- a) benzoate (Bz); b) 2,4 dichloro benzoate (DCB); c) p-hydroxybenzoate (pOHBz) were produced and submitted to chemical-physical characterization, in order to investigate the relationships between system composition, nanoscale dispersion level of the filler and final properties of the material.

INTRODUCTION

Electrostatic spinning, or electrospinning is a process capable of producing polymer fibers in the nanometer diameter range. It is a novel and efficient fabrication process that can be utilised to assemble fibrous polymer mats composed of fiber diameters ranging from several microns down to fibers with diameter lower than 100 nm. The electrospinning technique may be considered as a variant of the electrostatic spraying process "Bailey (1998)". In order to understand the fundamental principle underlying the process of electrospinning, consider a spherically charged droplet of a low molecular weight conducting liquid that is held in vacuum. As shown in Figure 1a, the droplet is under the influence of two forces, 1) the disintegrative electrostatic repulsive force and 2) the surface tension that strives to hold the droplet within a spherical shape. At equilibrium, the two forces completely balance each other, as is depicted by:

(1 where Q is the electrostatic charge on the surface of the droplet, R is the radius of the droplet, o is the dielectric permeability of vacuum and s is the surface tension coefficient. With increasing electric field strength, the charge on the surface of the droplet increases until it reaches a critical point when the electrostatic repulsive force overcomes the surface tension. When this happens the droplet disintegrates and leads to the formation of smaller droplets [Fig.1b]. This process is termed as electrospraying and has been utilized extensively in automotive spray painting.

Introduction

Figure 1 Phenomena of Electrospraying: when electrostatic repulsive forces overcome the surface tension of the liquid, the droplet (1a) disintegrates into smaller droplets (1b). If this concept is extended to a high molecular weight that has sufficient chain entanglements, instead of the formation of the droplets, a steady jet is formed that later solidifies in a polymer filament. Thus, the fundamental principle underlying fiber formation by electrospinning can be stated as follows: a high electric potential is applied to a polymer solution (or melt) suspended from the end of the spinneret that imparts an electrostatic charge on the polymer solution. At low electric potentials, the disintegrative electrostatic repulsive forces that primarily reside on the liquid surface are balanced by the surface tension. At high electric potentials, the electrostatic repulsive force on the surface of the fluid overcomes the surface tension that results in the ejection of a charged jet. The jet extends in a straight line for a certain distance, and then bends and follows a looping and spiraling path. The electrostatic repulsion forces can elongate the jet by several thousand times leading to the formation of a very thin jet. When the solvent evaporates, solidified polymer filaments are collected on a grounded target (or a winder) in the form of a non-woven fabric. The small diameter provides a large surface to volume ratio that makes these electrospun fabrics interesting candidates for a number of applications including filtration, 2

Introduction

protective clothing, biomedical applications including wound dressings, drug delivery systems, the design of solar sails, light sails, and mirrors for use in space, the application of pesticides to plants, as structural elements in artificial organs, and in reinforced composites. Although the process of electrospinning has been known for almost 70 years, polymeric nanofibers produced by electrospinning have become a topic of great interest in the past few years. The first patent that described the operation of electrospinning appeared in 1934, when Formhals disclosed an apparatus for producing polymer filaments by taking advantage of electrostatic repulsions between surface charges "Formhals (1934)". As shown in Figure 2, the spinning solution was discharged using a high electric field from an electrode of negative polarity. The fine filaments were attracted towards the movable electrode that was at positive polarity. A stripping device facilitated the removal of fibers in the form of a fibrous sliver. A major drawback of such a setup was the difficulty of removing the fine fibers that adhered to the moving parts of the collecting belts, drums, wheels etc. Formhals formed fibers from a solution of cellulose acetate in ethylene glycol at a potential difference of 57 kV.

Figure 2 Schematic of Formhals electrospinning setup to produce cellulose acetate fibers. In the 1960s, Taylor analyzed the conditions at the point of a droplet that was deformed by an electric field in a conical geometry. He identified the critical electrical potential for an electrostatic formation of a cone of liquid (now known as a Taylor cone) at the end of a capillary tube. The conducting drop was supposed to be surrounded by air and suspended in a stable position from the capillary tube. By examining a range of low molecular 3

Introduction

weight fluids, Taylor concluded that the conical interface between air and the fluid was stable at the semi-angle conical angle of 49.3o. However, it was shown that conductivity and viscosity both play an important role in the electrostatic atomization process which then influence the equilibrium angle and other aspects of the process. Taylor cones are considered to be important to electrospinning as they define the onset of extensional velocity gradients in the fiber formation process. In 1966, Simons "Simons (1966)" invented a process to produce a patterned non-woven fabric by electrospinning that was collected on a segmented receiver. The segmented receiver had two sets of segments such that one of the sets was at a different potential than the other causing the preferential deposition of fibers leading to a patterned fabric that had regions of low and high fiber density. Filaments were electrospun from a solution of polyurethane in methyl ethyl ketone having a viscosity in the range of 100 to 3000 centipoise. In a different process, Isakoff "Isakoff (1971)" devised a process to prepare fibrous sheets of organic synthetic polymers in which a filamentary web of fibers were electrostatically charged before being collected on a grounded movable surface. Fibers were electrically charged after they were formed by conventional solution spinning to facilitate their passage in a controlled trajectory. Utilizing a different geometry of the electrospinning apparatus, Fine et. al. "Fine (1980)" electrospun a thermoplastic elastomer where a cup-like apparatus contained the charged polymer solution that rotated about its vertical axis. The centrifugal forces pushed the polymer solution to the edge of the cup and into the ambient air. The presence of an electrical field caused the formation of a jet that was attracted to a grounded movable aluminum screen that was driven slowly around the rollers in the form of a belt. The combined action of electrostatic and centrifugal forces led to the formation of high strength fibrous sheets. Instead of placing the polymer solution in a cup, Guignard "Guignard (1980), Guignard (1981)" utilized another movable belt that carried the charged molten polymer. Here, two movable belts were used, as shown in Figure 3. As one of the belts, carrying the charged molten polymer, approached the grounded belt, the formation of several jets from the surface of the exposed melt took place. These jets traveled towards the grounded collector belt and were deposited in the form of short filaments on the grounded belt. In a different study, in 1977 Martin et al. "Martin (1977a), Martin (1977b)" extended the concepts of single component electrospinning to electrospin a mat of organic fibers from a liquid containing a plurality of components from a single nozzle/syringe. They were also able to electrospin polymeric solutions having different compositions (blends) simultaneously from multiple nozzles/syringes. Collectors were either a stationary metal screen or a movable belt. The patent claimed the mat was composed of fibers of a high molecular weight thermoplastic polymer based either on a fluorinated hydrocarbon, a silicone or an urea/formaldehyde. The 4

Introduction

resulting mats were flexible, non-absorbent, porous, and hydrophobic. It was suggested they be used as a wound dressing and as a lining for a prosthetic device.

Figure 3 Schematic of Guignards electrospinning setup. In 1982, Bornat "Bornat (1982)" conducted simultaneous electrospinning by using several nozzles/syringes to form fibers. As shown in Figure 4, the syringes were filled with the polymer solution and placed a certain distance from the collector that was in the form of a long metallic cylinder. The syringes were kept at a ground potential whereas an electrical potential of 50 kV was applied to the collector. The electrostatic forces caused the formation of jets that were attracted towards the rotating collector. Solidified polymer filaments of poly(tetrafluoroethylene) and poly(ethylene oxide) (electrospun simultaneously) were wound on the rotating collector in this fashion to form a tubular product. The tubular products were suggested to find applications as synthetic blood vessels and urinary ducts. Utilizing a slightly different approach, Simm et al. "Simm (1978)" were able to fabricate a composite filter based on electrospun fibers. They invented a process for the production of fiber filters in which a solution of polystyrene in methylene chloride was electrospun from an annular electrode onto two equidistant collector electrodes in the form of movable belts. The collector electrodes were already covered with a layer of cellulose fibers produced by conventional spinning methods. Thus, a composite filter that had electrospun polystyrene fibers on top of a layer of cellulose fleece was fabricated. More recently, in 2000, Scardino et al "Scardino (2000)" patented a process to make a hybrid/composite yarn. The electrospun fibers having diameters ranging from 0.4 nm to 1 nm were suctioned into an air vortex and then combined with a carrier filament(s) (obtained from conventional spinning techniques) to form nonwoven or linear (yarns) assemblies. The yarns were later utilized for weaving, braiding or knitting fabrics. Utilizing a similar approach to make composite fibrous mats, Zarkoob et al. "Zarkoob (2000)" 5

Introduction

patented a process to produce silk nanofibers and silk nanofiber composites utilizing electrospinning. The electrospinning principles used were very similar to those mentioned above with some differences in the engineering details of the apparatus. In a detailed study in 1971, Baumgarten "Baumgarten (1971)" studied the relationships between fiber diameter, jet length, solution viscosity, flow rate of the solution and the composition of the surrounding gas. He investigated the electrospinning of acrylic nanofibers where fibers less than a micron in diameter were reported. In 1981, Lorrondo and Manley "Larrondo (1981)" not only studied the effect of some of the parameters on fiber formation but also conducted some characterization studies on fibers of poly(ethylene) and poly(propylene) that were electrospun from the melt. The electrospun fibers were characterized by x-ray diffraction and mechanical testing. When the applied electric field was increased, the diffraction rings became arcs, thereby indicating the increasing alignment of the crystalline phase along the fiber axis. The fiber diameter was observed to decrease with increasing melt temperature and a qualitative correlation between the fiber diameter and viscosity was established. After a brief gap of a decade and a half, Reneker and his coworkers revived interest in electrospinning. In 1995, Doshi and Reneker "Doshi (1995)" studied the effect of several process parameters on fiber formation of poly(ethylene oxide) (PEO) fibers. Ever since then, many polymers have been electrospun and various aspects that better characterize the electrospinning process have been studied in great detail. In the following sections, several of these aspects will be described. The current emphasis of research is to exploit such properties and focus on determining appropriate conditions for electrospinning various polymers and biopolymers for future applications including multifunctional membranes, biomedical structural elements (scaffolding used in tissue engineering, wound dressing, drug delivery, artificial organs, vascular grafts), protective shields in speciality fabrics, filter media for submicron particles in separation industry, composite reinforcement, and structures for nano-electronic machines among others.

In.1 The Electrospinning process

The basic setup for electrospinning consists of three major components: a high-voltage power supply, a spinneret (a metallic needle) and a collector (a grounded conductor) (fig.4). The spinneret is connected to a syringe in which the polymer solution (or melt) is hosted. With the use of a pump, the solution can be fed through the spinneret at a constant and controllable rate. When a high voltage is applied, the pendent drop of polymer solution at the nozzle of the spinneret will become highly electrified and the induced charges are evenly distributed over the surface. 6

Introduction

Consequently the drop will experience two major types of electrostatic forces: electrostatic repulsion between the surface charges and Coulombic force exerted by the external electric field. Under the action of these electrostatic interactions, the liquid drop will be distorted into a conical object commonly kown as the Taylor cone "Huang (2003)" .When the strength of the electric field has surpassed a threshold value, the electrostatic forces can overcome the surface tension of the polymer solution and thus force the ejection of a liquid jet from the nozzle.

Figure 4 Schematics of electrospinning process As the liquid jet is continuously elongated and the solvent is evaporated, its diameter can be greatly reduced from hundreds of micrometers to as small as a few nanometers. Attracted by the grounded collector placed under the spinneret, the charged fiber is often deposited as a randomly oriented, non woven mat. The setup for electrospinning is extremely simple, but the spinning mechanism is rather complicated. Recent experimental observations demonstrate that the thinning of a jet during electrospinning is mainly caused by the bending instability associated with the electrified jet "Yarin (2001)". Theoretical studies clearly showed that the spinning process only involves whipping of a liquid jet. The whipping instability is mainly caused by electrostatic interactions between the external electric field and the surface charges on the jet. The formation of fibers of tiny diameters is mainly

Introduction

achieved by the stretching and acceleration of the fluid filament in the instability region "Fridrikh (2003)". In general, electrospinning is a process with low productivity because the polymer solution has to be fed at relatively slow rates in order to obtain ultrathin fibers. This problem was solved in different ways by several authors. Chu and co-workers used an array of multiple needles as the spinneret. In this case, electric field interference between adjacent nozzles may arise, and the configuration for the arrayed needle needs to be carefully designed to ensure that the electric field strength at the tip of each needle is identical in order to produce samples containing uniform fibers "Fang (2003)". Another method is to take a piece of tin foil, spray one side with gold, and gluf the other side to the tube of a syringe. An array of nozzles with diameters varying in the range of hundreds of nanometers to tens of micrometers. For this system, it is necessary to derivatize the gold surface of the membrane with an alkanethiolate monolayer terminated in CH3 or CF3 group to confine the liquid within each individual pore and thus to eliminate the coalescence of individual droplets "Whitesides (1990)". In a third approach, rigid needles made of stainless steel can be replaced with flexible, polyimide-coated glass capillaries commercially available for use in capillary electrophoresis. They can also be easily assembled in an array that can contain dozens of capillaries. By connecting each capillary to a different solution, it will be possible to simultaneously electrospin nanofibers in a range of different compositions and to obtain assemblies of nanofibers. In addition to the use of capillaries with hollow interiors, spinnerets can also be fabricated with solid tips. As demonstrated by Craighead and co-workers,a microfabricated, pyramidal silicon tip could serve as the spinneret for electrospinning when it is integrated with a microfluidic device; in this case they can be used to process functional material on miniaturized chips "Kameoka (2003)" . In an analogous study, Kessick and Tepper placed a drop of polymer solution on the surface of an interdigitated array of metal electrodes and found that fibers could be formed between drops sitting on adjacent electrodes if a potential difference was applied between the two sets of electrodes or if the drops were charged in advance via electrospray "Kessick (2003)" . These two demonstrations imply that electrospinning can probably be integrated with conventional lithographic techniques to form a new fabrication platform for generating patterned microstructures from various materials and in a broad range of length scales.

Introduction

In.2 Electrospinning processing variables

The following system and process parameters have been known to affect fiber diameter "Fong (1999), Deitzel (2000), Buer (2001), Casper (2003), Zhao (2004)": System Parameters: Viscosity Concentration Net charge density (conductivity) Surface tension of the polymer fluid. Molecular weight Molecular weight distribution Topology (branched, linear etc.) of the polymer. Process Parameters: Electric potential Flow rate of the polymer solution Distance between the capillary-end and target/collection screen Ambient parameters (temperature, humidity and air velocity in the chamber) Motion of target screen Internal diameter of the nozzle/capillary As discussed earlier, an electrically driven jet of liquid of low molecular weight would form droplets, a phenomenon more commonly referred to as electrospraying. The formation of such droplets was reported to be due to the capillary breakup of the jet by surface tension "Khil (2004)". In the case of polymer solutions, the pattern of capillary breakup is quite different. Instead of a complete breakup of the jet in response to the capillary instability, the jet between the droplets forms a nanofiber and the contraction of the radius of the jet, which is driven by surface tension, causes the remaining solution to form a bead. Fong et al. "Fong (1999)" observed the formation of beaded nanofibers that were electrospun from aqueous PEO (900,000 g/mol) solution at concentrations of 1-3 wt%, 21.5 cm and 15 kV. The corresponding zero shear rate viscosities of these solutions ranged from 13 527 centipoise respectively. As the concentration (and consequently the viscosity) was increased, the beads were observed to be larger, the average distances between beads were longer, the fiber diameter was larger and the shape of the beads changed from spherical to spindle-like. At concentrations higher than 4wt % (1250 centipoise), uniform electrospun fibers were observed. In a similar study, Jaeger et al. "Jaeger (1996)" observed the surface of beads on PEO nanofibers by atomic force microscopy. The beads were reported to possess a highly ordered surface at the molecular level. Similar results, 9

Introduction

where systematic increases in concentration (and consequently viscosity) lead to the formation of uniform fibers, have been observed by other researchers as well "Kattamuri (2003)". Recently, McKee et al, investigated the effect of solution rheological behavior on fiber morphology of linear and branched copolymers of poly(ethylene terephthalate-co-ethylene isopthalate) (PET-co-PEI) that were electrospun from a good solvent. In that study, investigation of the dependence of electrospun fiber diameter on concentration in addition to correlations between viscosity and concentration in different concentration regimes, "Colby (1991), Colby (1992)" viz. dilute, semidilute unentangled and semidilute entangled, were investigated. A systematic approach to investigate electrospun fiber formation in different concentration regimes (dilute, semidilute unentangled and semidilute entangled) is very relevant to understand the process of fiber formation from a solution rheological standpoint and the effect it has on electrospinning. While some of these issues were addressed in the study by McKee "McKee (2004)" , the authors recent work investigates and further tests the effects of scaling relationships between viscosity and concentration in different concentration regimes on fiber formation and morphological features of electrospun systems. To study the effect of surface tension on the formation of electrospun fibers, Fong et al. added a co-solvent (ethanol) to the aqueous PEO solution to change the surface tension of the polymer solution. The surface tension was observed to decrease with the addition of ethanol. PEO solutions of 3wt% were prepared in varying weight ratios of ethanol/water (from 0 to 0.702, by weight). The surface tension of a 3wt% aqueous PEO solution was 76 mN/m and that of a 3wt% PEO solution in ethanol/water:0.702 was 51 mN/m. When these solutions were electrospun at 0.5 kV/cm, beaded fibers were observed at higher values of surface tension (>59 mN/m), whereas uniform fibers were observed to form at lower values of surface tension (<59 mN/m). It has now been established, that increasing the concentration (and consequently viscosity) and lowering the surface tension favors the formation of uniform fibers "Liu (2002)".The effect of net charge density has also been studied. By adding a salt to the polymer solution, the conductivity of the solution was observed to increase. As a result the net charge density of the polymer solution increased. With increasing concentrations of NaCl in aqueous 3 wt% PEO solution electrospun at 21.5 cm and 15 kV, uniform fibers were observed to form at a net charge density of 3 Coulomb/liter of higher. The formation of beaded fibers was also studied by neutralizing the charges on the jet with ions of opposite polarity generated by a corona discharge. These ions were carried to the jet by flowing air. It was found that by increasing the number of neutralizing ions that intercepted the electrospun jet, more beaded fibers were formed. The electrical forces on the jet were reduced by a neutralizing charge, so that the jet behavior gradually became that of an uncharged jet, and as a result beads were observed to 10

Introduction

form. The effects of neutralizing ions were consistent with the results produced by the addition of a salt to the aqueous PEO solution. Both show that the net charge density carried by the moving jet tends to resist the formation of beads. Thus, the higher the net charge density carried by the jet, the more likely that a smooth fiber will be formed. Hsieh et al. studied the effect of the nature of the collecting target on the fiber morphology and fiber packing. Fibers collected on paper were found to have smooth surfaces and more uniform sizes than those collected on a water surface. However, the fibers collected on the water surface were found to be densely packed. Electrically conductive targets, such as tin foil and water, favor a tightly packed and thick membrane structure, whereas, a loosely packed fibrous network structure was observed to form on a nonconductive target, like paper. The overall pore volume was higher in electrospun mats collected on paper than those collected on water. Also, the pores in the mats collected on Al foil and paper were better interconnected in the planar directions than those collected on the water surface. By utilizing a novel approach, Kameoka and coworkers "Kameoka (2003), Czaplewski (2003)" devised a method to achieve a controlled deposition of oriented polymeric nanofibers. This method used a metallic micron tip that was dipped in the polymer solution to gather a droplet as a source material. Subsequent application of the voltage to this tip led to the formation of jet that was deposited on a rotating counter electrode. This resulted in the collection of oriented nanofibers that were suggested to be integrated with microfabricated structures. Zong et al. "Zong (2002)" studied the effect of feed rate on fiber diameter, where the polymer solution was contained in a syringe such that the plunger was pushed by a metering pump. Observations of fibers electrospun from 25wt% poly(D,L- lactic acid) (PDLA) solution in DMF at 0.8kV/cm indicated that the fiber diameter increased with increasing flow rate as might be expected. Potassium dihydrogen phosphate (KH2PO4) at a concentration of 1wt % was added to the solution to increase the conductivity. With increasing electric potential, the jet velocity was observed to increase significantly and the solution was removed from the capillary-end more quickly. As a result the volume of the Taylor cone at the capillary end was observed to become smaller. At higher voltages (<25kV), the Taylor cone disappeared completely and the jet was found to initiate directly from a region inside the capillary. In the case where polymer solution was kept in a glass capillary (or in other words, the polymer solution was not forced through the capillary-end by a plunger) Larrondo and Manley "Larrondo (1981)" determined that doubling the applied electric field decreased the fiber diameter by roughly 50%. However, Baumgarten showed that the diameter of the jet reached a minimum after an initial increase in field strength and then became much larger with increasing electric fields. This was caused by the feed rate of the polymer solution through the capillary, 11

Introduction

and illustrated one of the complexities of the electrospinning process, i.e., increasing the electric field not only increased the electrostatic stress, thereby creating smaller fibers, but it also drew more material out of the capillary. Utilizing a similar glass capillary apparatus instead of the syringe, Doshi "Doshi (1995)" conducted some investigations where he plotted the jet diameter at different positions along the jet to verify both of these relationships. It was shown that increasing the electric field resulted in an increased spinning rate. At the capillary orifice, there was a threefold increase in the jet diameter with less than a twofold increase in applied voltage. Demir et al. found a power-law dependence between the flow rate and electric voltage based on their measurements conducted on polyurethaneurea fibers that were electrospun from a solution of 21 wt% in dimethylformamide (DMF). The effect of another key process parameter, namely the distance between the capillary-end and the target has also been studied extensively by several workers. At low separation distances between the capillary-end and the target, wet fibers were collected, primarily due to the presence of significant amounts of residual solvent in the fibers. As the separation distance is increased, the time for the solvent to evaporate increased, and as a result, dry solid fibers were collected at the target. Also, with increasing distance between the capillary-end and the target, the jet underwent a larger amount of electrically driven bending or whipping instability. Consequently, the amount of stretching or elongation of the jet increased and as a result the fiber diameter was observed to decrease. At a given capillary-end to target distance, polymer filaments electrospun from solutions in high vapor pressure solvents tend to dry faster than those electrospun from low vapor pressure solvents because of an increased evaporation rate. Lastly, another parameter that affects the fiber diameter is the temperature of the polymeric fluid. This is especially critical for electrospinning from polymer melts "Lyons (2003)". Larrondo and Manley observed that the fiber diameter deceased with increasing melt temperature. Given the inverse relationship between melt temperature and viscosity, they were able to draw a qualitative correlation between diameter and viscosity, i.e., fiber diameter increased with increasing viscosity of the melt. However, no quantitative relationships between fiber diameter and viscosity were reported.

In.3 Remarkable features of electrospun nanofibers

Nanofibers exhibit a range of unique features and properties that distinguish themselves from one-dimensional nanostructures fabricated using other techniques. For example: a) Extremely long length. Compared with 1D nanostrucutres synthesized or fabricated using other chemical or physical methods, electrospun nanofibers 12

Introduction

are extremely long. Because electrospinning is a continuous process, the fibers could be as long as several kilometres "Frenot (2003)". b) High surface area and complex pore structure. Compared with the fibers fabricated using a conventional mechanical extrusion or spinning process, electrospun fibers are much thinner in diameter and thus higher in surface-to volume ratio. A high density of pores can also be formed as a result of entanglment of nanofibers. c) Alignment on the molecular level. Electrospinning involves the rapid stretching of an electrified jet and rapid evaporation of the solvent. The polymer chains are expected to experience an extremely strong shear force during the electrospinning process "Pedicini (2003)". This shear force and rapid solidification may prevent polymer chains from relaxing back to their equilibrium conformations. As a result, the chain conformation and crystallinity of the resultant polymer nanofibers should be different from products obtained by solution casting or conventional spinning processes. In particular the orientation of chains can be achieved in the electrospun fibers. In the study of a liquid-crystalline polymer, it was established that the orientation development in the fibers was dependent on the diameter of the fiber "Lin (2003)".

In.4 Application of electrospun nanofibers

The simplicity of the fabrication scheme, the diversity of materials suitable for use with electrospinning, as well as the unique and interesting features associated with elecetrospun naofibers, all make this technique and resultant structures attractive in a number of applications, some of which are reported below. -Nanofiber reinforced composites Fiber-based reinforcement represents one of the most effective strategies for enhancing the strength and other performance of a composite material. Because of the high surface-to volume ratio, use of nanofibers as the components may significantly increase the interaction between the fibers and the matrix material, leading to better reinforcement than conventional fibers. Because the cross-sections of electrospun fibers are small, composites fabricated using this method could maintain their optical transparency "Bergshoef (1999)". Electrospun poly(benzimidazole) (PBI) nanofibers of diameter ca. 300 nm "Kim (1999)" were used as reinforcing fibers in polymer matrix composites. The epoxy matrix composites reinforced with PBI nanofibers had higher tensile modulus, fracture toughness and fracture energy. The fracture toughness and the fracture energy of the composites depended on the relative direction of the majority of nanofibers in the fabric (the winding direction) and the crack. If the crack was transverse to the winding direction of the 13

Introduction

fabric in the composites, fracture toughness and the fracture energy were found to be higher than for those samples in which the crack was along the winding direction. Dzenis and Reneker have mathematically modeled the micromechanics of electrospun nanofibers in hybrid polymer composites on the principle of effective homogeneity. The effective thermoviscoelastic characteristics of nano and microfiber composite as well as nanofiber and microparticulate composites were calculated. Analysis showed that a small amount of nanofibers substituted for larger conventional microfiber and microparticulate reinforcements can significantly change the effective behavior. At higher total volume fractions (0.4 and above) of nanofibers, the hybrid composites exhibit properties that were not displayed by two component materials. It was concluded that a better packing of reinforcing elements of different sizes holds potential for improved fracture toughness. In composite applications, if there is a refractive index difference between the reinforcing fibers and the matrix, then the resulting composite becomes opaque or non-transparent due to light scattering. One possible way of circumventing this limitation is to use fibers with a diameter significantly lower than the wavelength of visible light (400-700nm). In view of this consideration, Bergshoef and Vancso "Bergshoef (1999)" electrospun smooth fibers of nylon-4,6 with diameters in the range of 30-200 nm from solutions in formic acid. The fibers were then incorporated in a matrix of commercially available two-component phenolic epoxy resin (Epo-plast). The composite films obtained by fiber reinforcement were fully transparent. Visible light was found to transmit over the complete range of wavelengths in the visible spectrum (400 700 nm), as was observed by spectroscopy and visible inspection. In a different study, Fong et al. "Fong (2001)" were able to reformulate exfoliated montmorillonite-nylon 6 (NLS) nanocomposites by electrospinning. Fibers and nanofibers of NLS nanocomposite (diameters between 100 and 500nm) were electrospun from solution and, collected as non-woven fabrics, or as aligned yarns. The electrospinning process resulted in highly aligned montmorillonite layers (normal to fiber axis) and nylon 6 crystallites (parallel to fiber axis). Electrospun fibers with engineered surface chemistry that were suggested to find applications in composites have been produced by Deitzel et al. "Deitzel (2002)". They conducted electrospinning on a series of random copolymers of poly(methyl meth acrylate) with tetrahydroperfluorooctyl acrylate (PMMA-r-TAN) from a mixed solvent of toluene and dimethylformamide to produce fibers that had a surface rich in fluorine. In fact, x-ray photoelectron spectroscopy (XPS) analysis showed that the atomic percentage of fluorine in the near surface region of the electrospun fibers was about double the atomic percentage of fluorine found in a bulk sample of the random copolymer. These electrospun fibers with an engineered surface chemistry were suggested to find potential applications in the biomedical and composite reinforcements. More recently, composite nanofibers were 14

Introduction

processed that contained single and multi-walled carbon nanotubes, "Seoul (2003), Wang (2003), Lam (2004)". A key challenge is the uniform dispersion and alignment of the carbon tubes along the fiber axis. Due to the high stress-fields involved during electrospinning, small bundles of carbon nanotubes have been observed to be oriented parallel to the fiber axis, although there were some instances where these appeared to be twisted, bent or with other irregularities. These nanocomposite fibers were suggested to serve as precursors for linear, planar and 3D assemblies for macrocomposites, conductive polymeric nanofibers and as supercapacitors. -Nanofiber-based membranes and smart cloths The electrospun fiber mats had a higher convective resistance to air flow compared to normal clothing materials while the resistance to the transport of water was much lower than commercial membrane laminates "Gibson (2001)". In particular, the electrospun membranes exhibited the excellent ability to capture aerosol particles. The filtration efficiency was also high even though an extremely thin layer of electrospun fibers was used. The light-weight, breathable mats made of electrospun nanofibers are promising candidates for applications such as protective clothing and in the field of filtration/separation. With the incorporation of magnetic components, it is expected that the separation efficiency for magnetic-active particles can be further enhanced through the application of an external magnetic field. In principle, it is possible to incorporate many types of active components and species into electrospun nanofibers to fabricate smart cloths that are responsive to all types of changes in the surroundings. Gibson et al. "Gibson (2000)" have studied the transport properties of elastomeric electrospun mats and it was suggested they be utilized in protective clothing applications. Performance measurements on electrospun mats compared favorably well with the transport properties of textiles and membranes currently used in protective clothing systems. The electrospun layers presented minimal impedance to moisture vapor diffusion required for evaporative cooling. When stretched in biaxial tension to strain levels of 100%, porous elastomeric nanofiber membranes were significantly affected in terms of their convective gas flow properties because of the opening of the interfiber pores, as might be anticipated. However, the water vapor diffusion properties remained unchanged. Experimental measurements and calculations showed that electrospun fiber mats were extremely efficient at trapping aerosol particles. The high filtration density was concluded to be a direct result of the sub-micron size high specific surface fibers generated by electrospinning process. Electrospun nanofiber coatings were applied directly to a polyurethane foam containing activated carbon, which is used as a component in military chemical protective clothing systems.The airflow resistance and aerosol filtration properties correlated with the electrospun 15

Introduction

coating add on weight. Aerosol particle penetration through the foam layer was eliminated by minuscule levels of nylon electrospun nanofibers sprayed on to the surface of the foam. It was suggested that the electrospun fibers be sprayed directly onto a 3D screen form obtained by 3D body scanning to form custom-fit seamless clothing "Gibson (1998)". -Biomedical usage Because of their peculiar structures, electrospun nanofibers have received much attention for use in tissue engineering. Natural scaffolds for tissue growth are three-dimensional networks of nanometer-sized fibers made of various proteins. Successful tissue engineering requires man-made scaffolds bearing similarity to their natural counterparts in terms of chemical composition, morphology and surface functional groups. Non-woven mats of electrospun nanofibers are well known for their interconnected, threedimensional porous structures and relatively large surface areas. They provide a class of ideal material to mimic the natural extra-cellular matrix required for tissue engineering. To this end a number of biodegradable polymers have been electrospun as non woven membranes and their use as scaffolds for tissue engineering have been demonstred in recent years " Li (2002)". Several research groups are also actively exploring other biomedical applications for electrospun nanofibers. For example, electrospun fibers have been applied to wound dressing because porous mats made of fibrous structures can protect the wound from bacterial penetration via the aerosol particle capturing mechanism while providing a good path for the transport of vapor "Khil (2003)". In addition to the protection function, electrospun non woven mats are also potentially useful as the supports or carriers for drug delivery because of their high specific surface areas. The application of electrospun fibers for biomaterial applications was reported as early as 1977, when Martin et al. utilized fibrillar mats for wound dressings and vascular prosthetics. It was noted that polymer concentration required for electrospinning depended upon the molecular weight of the polymer, with lower molecular weights requiring higher concentrations. Sub-micron electrospun fibers of poly (glycolic acid)143, poly(lactic-coglycolic acid) (PLA/PGA), poly(lactic acid) (PLA) and poly(ethylene-covinyl acetate) (PEVA)144 have been electrospun from solutions in HFIP, methylene chloride or chloroform. As PLA, PGA, PEVA and PLA/PGA are biodegradable materials, the electrospun mats from these polymers have been suggested in applications as tissue scaffolds. Very recently Badami et al, utilized the authors electrospinning apparatus to electrospin mats of PLA that were subsequently tested for tissue scaffolding applications. Collagen nanofibers were electrospun from a solution in HFIP " Matthews (2002)" to form fibers with a diameter of 100nm. Aortic muscle cells were then plated out in-vitro onto different formulations of electrospun 16

Introduction

collagen mats. Microscopic examination of the cultures revealed that the scaffolds were densely populated with smooth muscle cells, within 7 days. Cross sectional analysis indicated that electrospun collagen promoted extensive cellular infiltration into the fibrillar network. Smooth muscle cells were observed deep within the matrix and fully enmeshed within the fibrils of the electrospun collagen. In another study, Li et al." Li (2002)", seeded BALB/c C7 mouse fibroblast cells on electrospun fibrous mats of poly(D,Llactide-co-glycolide) (PLGA) with fibers ranging from 500 800 nm in diameter. The in-vitro studies indicated that the mouse fibroblasts adhered and spread on the surface of the PLGA fiber network and had started to migrate through the pores and to grow under layers of the fiber in network at day 3. These fibroblasts interacted and integrated well with the surrounding fibers. Scanning Electron Microscopy (SEM) showed that the development of cell growth was guided by the fiber architecture. Cells grew in the direction of fiber orientation, forming a three dimensional and multi-cellular network according to the architecture of the nanofibrous structure. It was concluded that the electrospun structure of PLGA featured a morphological similarity to the extra cellular matrix (ECM) of natural tissue, which is characterized by a wide range of pore diameter distribution, high porosity, and effective mechanical properties. Yoshimoto et al. "Yoshimoto (2003)" also produced microporous, non-woven poly(-caprolactone) (PCL) scaffolds by electrospinning that were seeded with Mesenchymal stem cells (MSCs) derived from the bone marrow of neonatal rats. The cell-polymer constructs, which were cultured with osteogenic supplements under dynamic culture conditions for up to 4 weeks, maintained the size and shape of the original scaffolds. Penetration of the cells and abundant extra cellular matrix was observed in the cell-polymer constructs after 1 week. SEM studies indicated that the surfaces of the cell-polymer constructs were covered with cell multilayers after 4 weeks. In addition, mineralization and type I collagen were observed after 4 weeks suggesting that electrospun PCL could serve as a potential candidate for bone tissue engineering. Xu and coworkers "Mo (2003), Xu (2004), Mo (2004)", observed good biocompatibility of electrospun scaffolds of poly(L-lactide-co-e-caprolactone) (PLLA-CL) when in vitro cultures of muscle, endothelial and human coronary artery smooth muscle cells proliferated and migrated along the axis of nanofibers. Kenawy et al."Kenawy (2002)" explored the potential of PLA and PEVA electrospun fiber mats as drug delivery vehicles. Tetracycline hydrochloride was used as a model drug. The fibers were electrospun from chloroform solutions containing a small amount of methanol to stabilize the drug. In vitro release of tetracycline hydrochloride was followed by UV-Vis spectroscopy. It was found that electrospun PEVA and 50/50 blend of PLA/PEVA mats gave relatively smooth release of the drug for 5 days. When compared to the drug release from a cast film, it was found that the total percentage of the drug released from the electrospun mats was greater. This was attributed to the 17

Introduction

much lower surface area of the cast films. Zong et al. also found similar results regarding the total percentage in-vitro release of Mefoxin from porous PDLA electrospun membranes. In addition, the loading efficiency, defined as the released amount by the initial loading amount, of Mefoxin in PDLA was found to be around 90%. By utilizing a novel approach, Sanders et al "Sanders (2003)" conducted electrospinning of a biocompatible polymer, poly(ethylene-co-vinyl acetate) (EVA) from a solution containing dispersed aqueous domains to produce EVA fibers containing aqueous micro reservoirs. The osmotic swelling of these aqueous reservoirs in the electrospun fibers lead to an eventual bursting that could potentially be utilized for drug delivery applications. In another in-vitro study, Kenawy et al." Kenawy (2002)" investigated the antimicrobial properties of electrospun mats and compared those to bulk materials. It was found that the anti microbial properties, assessed by measuring the inhibition zone diameter against the testing strains of B. subtilis, S. choleraesuis, S. aureas and E. coli, of the electrospun fibers of poly(vinyl phenol) PVP were enhanced in comparison to the bulk materials. Poly(vinyl phenol) (PVP) was also sulfonated to enhance its antimicrobial characteristics. Jia et al. "Jia (2002)" demonstrated a unique approach to the construction of nanosized bioactive materials for use as highly active and stable biocatalysts. Functionalized polystyrene (PS) nanofibers were electrospun, that had a typical diameter of 120 nm, followed by the chemical attachment of a model enzyme, chymotrypsin. The enzyme loading, as determined by active site titration was up to 1.4 wt%. The apparent hydrolytic activity of the nanofibrous enzyme in aqueous solutions was over 65% of that of the native enzyme, indicating a high catalytic efficiency as compared to other forms of immobilized enzymes. In addition, the nanofibrous -chymotrypsin exhibited a much-improved nonaqueous activity that was over 3 orders of magnitude higher than that of its native counterpart suspended inorganic solvents including hexane and isooctane. In a different study, Huang et al. "Huang (2000)" obtained an 81 kDa recombinant protein based upon the repeating elastomeric peptide sequence of elastin [(Val-Pro-Gly-Val-Gly)4 (Val-Pro-Gly-Lys-Gly)] using genetic engineering and microbial protein expression. Elastin is a structural component of arterial blood vessels and is responsible for providing fatigue resistance and resilience to the arterial wall. The recombinant protein was electrospun to produce 200 nm 3m diameter fibers laid in a random fashion in the non-woven mat. The ultimate tensile strength and modulus of these nonwoven structures was 35 MPa and 1.8 GPa respectively. These electrospun mats were suggested to find applications in human tissue constructs and synthetic artificial organs. Nagapudi et al." Nagapudi (2002)" were able to modify the elastin-like recombinant protein by incorporating an acrylic moiety. Electrospun fibers of this modified protein, in the diameter range of 300 nm 1.5 m, were later cross-linked by visible-light-mediated photo irradiation. This caused the 18

Introduction

mats to have a higher modulus and tensile strength and insoluble in water. The fabric networks were found to have porosities in the same range as those exhibited by other porous tissue scaffolds. Buchko et al., also referred earlier, have electrospun a Silk-like polymer with Fibronectin functionality (SLPF) to make biocompatible thin films for use in implantable devices. The electrospinning process was shown to be a means of creating porous thin films with structural gradients and controlled morphology that could enhance biocompatibility. Fang and Reneker " Fang (1997)" have electrospun DNA fibers from aqueous solutions that had a diameter around 50 to 80 nm. The electrospun DNA fibers were reported to have a diameter in the range that was about 10 to 1000 times smaller than those produced by conventional techniques. In another study, M13 viruses suspended in 1,1,1,3,3,3hexafluoro-2-propanol were blended with a highly water soluble polymer, polyvinyl pyrrolidone (PVP) and later electrospun into continuous uniform virus-blended PVP nanofibers. The resulting electrospun mats of virus-PVP nanofibers maintained their ability to infect bacterial hosts after being resuspended in a buffer solution." Lee (2004)" -Nanofiber-based sensor The fabrication of fluorescence-based sensors with electrospun nanofibers containing a fluorescent indicator as the active component was achieved. The sensitivities of such devices to metal ions and 2,4-dinitrotoluene were two to three orders of magnitude greater than those based on continuous thin films. The improvement was mainly achieved by the availability of a larger surface area and the relatively easy transport of analytes to the active sites on each nanofiber. Gas sensors based on individual polyaniline nanofibers were recently fabricated " Liu (2004)". -Nanofiber-based electrode materials Jo and co-workers have demonstrated that porous membranes made of electrospun fibers were good matrices for holding polymer electrolytes and might find immediate use in the fabrication of high-performance lithium batteries "Choi (2003)". They have also found that the porous membrane (when soaked with an electrolyte solution) exhibit an enhanced ionic conductivity, this is due to the peculiar pore structure and the high porosity associated with membranes composed of electrospun nanofibers. The high surface area also improves the wettability of fibers by the electrolyte solution, as well as the affinity between these two heterogeneous components. Yang and Kim have also evaluated the performance of carbon nanofibers as an electrode material for making supercapacitor "Wang (2002)". By controlling the activation temperature, the pore density and structure within

19

Introduction

the nanofibers could be varied to further improve the performance of such a supercapacitor at higher current densities.

20

Chapter I Evaluation of the influence that the main process parameters may have on the final fibers

I.1 Introduction
To investigate the effects of some of the parameters on electrospinning, such as spinning voltage, solution flow rate, the distance between the needle tip and the ground electrode (working distance) and concentration of the solution, were conducted some preliminary experiments. These experiments were conducted strictly to validate the proper functioning of the electrospinning apparatus that was constructed and installed from the concepts described in the literature. In addition, the experiments described in the following sections were conducted solely for the purposes of supporting the concepts already elucidated in the electrospinning literature while at the same time obtaining a better practical understanding of the effect of some of these parameters on fiber formation and diameter. PCL solutions were electrospun in order to investigate the effects of process parameters. It was found that solution properties are the main factors influencing the transformation of polymer solution into ultrafine fibers Deitzel (2001). As a first parameter was investigated the type of solvent, choosing the solvents according to the solubility parameters that are very useful to understand the interactions between polymer and solvent. Among the solvents used (dichloromethane-methanol, chloroform-methanol, tetrahydrofuran and acetone) the best was found to be acetone, that is a medium hydrogen bonded solvent with a solubility parameter of 10 (cal/cm3)1/2(Tab.I.1).

Chapter I

Table I.1 The electrospinning conditions in different solvents

Solvents dichloromethanemethanol dichloromethanemethanol dichloromethanemethanol chloroform-methanol chloroform-methanol chloroform-methanol tetrahydrofuran tetrahydrofuran tetrahydrofuran Acetone Acetone Acetone Acetone A 10% 15% 20% 10% 15% 20% 10% 15% 20% 10% 12.5% 15% 20% B 20 20 20 20 20 20 20 20 20 20 20 20 20 C 3 3 3 3 3 3 3 3 3 3 3 3 3 D 20 20 20 10 10 10 20 20 20 20 20 20 20 Formation Beads Drops Fibers Beads Stick fibers No homogeneous fibers Beads Beads No homogeneous fibers Beads Beads Thin fibers Fibers with drops

A. Concentration (wt%), B. Potential (KV), C. Flow rate (ml/h), D. Distance (cm)

22

Process Parameters

Figure I.1a SEM images of PCL fibers electrospun from a solution of 3:1 dichloromethane-methanol (10%).

Figure I.1b SEM images of PCL fibers electrospun from a solution of 3:1 chloroform-methanol(15%).

Figure I.1c SEM images of PCL fibers electrospun from a solution of tetrahydrofuran (10%). 23

Chapter I

Figure I.1d SEM images of PCL fibers electrospun from a solution of acetone (15%). When dissolving polycaprolactone in this solvent, we tried different concentrations, with a distance of the needle from the screen alternatively of 10 or 20 cm. The concentration of the PCL solutions influences the spinning of fibers and control the morphology. Formation of beads along fibers took place at low PCL concentration. As the concentration of PCL was increased continuous nanofibers without beads formation were obtained by electrospinning from 15% wt solutions. I.2 Process Parameters: Control of Morphology and Diameter The morphology and diameter of electrospun fibers are dependent on a number of processing parameters that include: a) the intrinsic properties of the solution such as the type of polymer, the conformation of polymer chain, viscosity (or concentration), elasticity, electrical conductivity, and the polarity and surface tension of the solvent; b) the operational conditions such as the strength of applied electric field, the distance between spinneret and collector, and the feeding rate for the polymer solution. To understand how to control electrospun nanofibers I studied the effect of the variation of some of these parameters on the PCL fibers produced and analysed with SEM. First of all the distance between the capillary and collector was increased at low concentrations (fig.I.2).

24

Process Parameters

Figure I.2a SEM photos of 10% of PCL fibers at distance of 10 cm.

Figure I.2b SEM photos of 10% of PCL fibers at distance of 20 cm. Figure I.2 shows that the bead structure is predominant at low concentrations; the average diameter of the fibers and the size of the beads decrease if the capillary-screen distance is increased. At higher concentrations (20%) we observe a blob configuration; so if the polymer fraction is increased and other parameters are kept constant, the diameters keep growing with a high dispersivity (fig.I.3) known as the Taylor cone "Huang (2003)".When the strength of electric field has surpassed a threshold value, the electrostatic forces can overcome the surface tension of the polymer solution and thus force the ejection of a liquid jet from the nozzle. 25

Chapter I

Figure I.3a 20% PCL in acetone at a distance of 10cm.

Figure I.3b 20% PCL in acetone at a distance of 20cm. Figure I.3a shows another characteristic: the point of rupture. This phenomenon is due to the excessive viscosity of the solution that creates deformation stress during the whipping motion and breaks the jet instead of drawing and stretching. If the potential difference is increased, and the other parameters are kept constant, the fiber diameters decrease (fig.I.4b). The nanofibrous structure of the sample is evident, and is composed of individual, uniform and randomly oriented fibers with an average diameter of 600 50 nm.

26

Process Parameters

frequency
0

500

1000

1500

2000

Fibers Diameter (nm)

Figure I.4a SEM images and diameter distributions of ultrafine PCL fibers electrospun from 15% wt PCL in acetone at 15KV.

frequency
0

200

400

600

800

1000

Fiber Diameter (nm)

Figure I.4b SEM images and diameter distributions of ultrafine PCL fibers electrospun from 15% wt PCL in acetone at 20KV.

If the voltage is increased, the polymer jet is discharged with great electrostatic repulsion, which leads to a great traction effort of the jet. The result is an increase in fineness. With high potential differences, the diameters of the fibers grow and control of the process at high voltages, deliveries and concentrations gradually becomes more difficult.

27

Chapter I

Figure I.5a SEM images of electrospun PCL fibers ( applied voltage 20KV, solution concentration 15wt%, distance 20cm) and needle of 0.7mm.

Figure I.5b SEM images of electrospun PCL fibers ( applied voltage 20KV, solution concentration 15wt%, distance 20cm) and needle of 0.8mm.

A comparison of the two figures (fig.I.5a and b) shows that the blob configuration is predominant in fibers obtained by using a short needle (0.7mm), the other parameters are kept constant, in particular we can see stick fibers, and their size distribution is quite limited. The distance from the screen must be increased and the solution delivery and concentration must be decreased in order to obtain fibers with small diameters. The increase in electric potential initially has a beneficial effect, but after certain limit it no longer affects the fineness of the fibers (fig.I.6).

28

Process Parameters

400

Fiber (nm)

At constant electric potential, screen distance and flow rate

300

500

At constant electric potential, flow rate and concentration.

400

Fiber (nm)
5 10 15 20 25 30

300

200

200

100

100

0 0 2 4 6 8 10 12 14 16 18 20 22

Concentration (wt%)

Screen distance (cm)

500 450 400

At constant electric potential,screen distance, concentration and flow rate.

1000

At constant flow rate,screen distance, and concentration

Fiber (nm)

300 250 200 150 100 50 0,6 0,8 1,0 1,2

Fiber (nm)

350

800

600

400

200 0 5 10 15 20 25 30

needle (mm)

Electric Potential (KV)

Figure I.6 Graphs that relate the trend of the main parameters with respect to the fiber diameters.

29

Chapter I

30

Chapter II Encapsulation of microsphere for nanostructured thermoregulating textile materials

Microencapsulation is a versatile technology that offers many opportunities for a wide range of applications in numerous areas of daily life Hong (1999). Early commercial applications were in food, pharmaceuticals, cosmetics and agrochemicals, but in recent years the textile industry is also revealing a growing interest towards the possibilities offered by microencapsulation Nelson (2002).In this field, some of the major interests are currently in developing innovative textiles treated with durable fragrances, antimicrobial agents, insect repellents, flame retardants and phase-change materials (PCMs). In particular, the use of microencapsulated PCMs appears particularly interesting for thermal regulation functions in clothing for sports and extreme weather. In fact, the application of a PCM to a garment provides an active thermal insulation effect that controls and adjusts the heat flux through the garment layers according to thermal conditions. The intensity and duration of the PCM active thermal insulation effect depend mainly on the heat storage capacity of the PCM and their applied quantity. In addition, performance tests carried out on textiles with incorporated microcapsules have shown that the thickness and density of its component fabrics also influence the efficiency of the active thermal insulation effect of the composite. High thickness, low density or air gaps between the garment layers make insulation better. Up to now, an extensive use of this technology has been restricted by problems encountered in the even distribution of PCM in a polymer matrix. Different ways were explored in order to incorporate PCM microcapsules into textile products: melt spinning Colvin (2002) or wet spinning

Chapter II

Charunyakorn (1991) of polymer/MPCM mixture, coating on fibers and fabrics with a binder Shin (2005) dispersion into polymer foam or in multilayer composites Colvin (1996). At present no one of these processes is capable of making high quality thermal regulated textile products with the desired substrate structure. In particular, it was difficult to incorporate enough PCM in the fiber structure to get a thermal result while keeping the fibers mechanical characteristics. Processes that produce high quality and inexpensive fibers generally require severe heat treatments. In this case, the PCM microcapsule, are required to have very good thermal and mechanical stability to ensure intact structure during manufacturing processes Zhang (2004). In detailed studies on electrospinning process Fenn (1989), Reneker (2000), Frenot (2003) it was found that the fiber diameter can be controlled within a broad range down to just a small number of nanometers by proper selection of the processing parameters Burger (2006). Nanofibers generated by this technique have shown wide application in areas where their unique properties contribute to product functionality. Those properties include nanofiber-reinforced composites, nanofiber-based membranes, biomedical usage, nanofiber-based sensors, nanofiber-based electrode materials and nanofiber-based electronic and optical devices Fong (2002), Ma (2004), Lee (2005).

II.1 Materials
Poly(-caprolactone) (PCL) was kindly supplied by Solvay (CAPA 6501) (Mn = 50 000 g/mol, Mw/Mn = 2) and was used without further purification. Microencapsulated phase change material (octadecane) PCM was furnished by Frisby Technologies (Thermasorb 95), having a peak transition temperature of about 35C (95F), diameter in the range of 15-50 micrometers and a latent heat capacity of about 180 J/g. PCL was dissolved in acetone at 40C under mechanical stirring for 15 min. After that, PCM microcapsule was added to the solution under continuous stirring for another 10 min.

II.2 Analytical techniques

The preparation method can be summarized in the following major steps: 1) dissolution of fiber-forming polymer in the solvent; 2) dispersion of microencapsulated PCM to the mixture; 3) electrospinning process of the resulting system. One of the principal advances in using electrospinning is that the process takes place in gentle conditions, avoiding capsule damage. This was verified by testing the microcapsules in different electrospinning 32

Thermoregulating Textile Materials conditions. To that purpose, a solution of acetone with 4% (wt/wt) of microcapsules stored in a reservoir was pumped through a metal capillary connected to a voltage supply using a controlled pump. Several processing parameters, namely flow rate, applied voltage, distance between tip of orifice and target were varied in an ample range of values. The electrospinning process of pure PCL, microcapsule and composite fibers were carried out in a vertical spinning configuration. The process was carried out at room temperature at a high voltage of 20 kV (HV Power Supply, Gamma High Voltage Research, Ormond, FL). The spinneret used in the experiments had an inner diameter of 1.1 mm. A copper wire was mounted in the spinneret and used as the positive electrode. Grounded tin foil was used as the counter electrode and mounted at a distance of 20 cm from the spinneret. Continuous PCL fibers were collected on the aluminium foil in the form of a fibrous mat.

II.3 Methods
Scanning Electron Microscopy (SEM). Pictures were obtained with a LEO 420 microscope. The samples were covered with a 250 thick gold film using a sputter coater (Agar mod. 108 A). Differential Scanning Calorimetry (DSC). DSC measurements were carried out using a thermal analyzer Mettler DSC 822/400 at a heating rate of 10C/min between -50 and 200C.

II.4 Nanofibers Characterization

There are several parameters that affect size and defects associated with nanofibers produced by electrospinning. The parameters that were studied include concentration and electrode spacing. The main defect that is common and difficult to eliminate is called beads. In order to observe the morphology and diameter size of the fibers, Scanning Electronic Microscopy (SEM) was utilized. Figure II.1a shows an example of a SEM photograph of the microcapsules collected in the powder state after the electrospinning test. The majority of the capsules retain a sphere shape with a smooth surface morphology. Agglomerations were not observed, whereas the particle size distribution (Fig.II.1a) show a mean capsule diameter of 20 m and a size ranging between 10 and 40 m.

33

Chapter II

Figure II.1a SEM micrographs of PCM microspheres.

Figure II.1b SEM micrographs of PCM microspheres at different magnification. Electrospinning conditions were optimized to produce neat PCL nanofibrous mats composed of individual fibrils without beads formation. In particular, good results were obtained with concentration of PCL in acetone ranging between 15 and 20% (wt/wt), a flow rate between 0.3-3 L/min, a voltage of 30 kV and a distance between the grounded collector and the needle of 10-30 cm. Figure II.2a shows the fibers obtained with 15% (wt/wt) of PCL in solution, a distance between the target and the needle of 10 cm and a flow rate of about 1 L/min. The fibrous structure of the sample is evident, and is composed of individual, uniform and randomly oriented fibers. The diameter distribution of the fibers is relatively wide (ranging between 0.2 and 15 m).

34

Thermoregulating Textile Materials

Figure II.2a SEM images of PCL fibers treated with electrospinning.

Figure II.2b SEM images of PCL fibers treated with electrospinning at different magnification. The mats collected on the ground are white and opaque due to light scattering by the fine fibers, yet retain the general mechanical characteristics of the base polymer, such as elasticity. In literature it is well described that the concentration of the polymer solutions influences the spinning of fibers and controls the morphology. A decrease in concentration of PCL leads to the formation of beads and the disappearance of fibers morphology (for concentrations lower than 10% wt/wt). In a similar way, the addition of microcapsules to the polymer solution influences the morphology of the composites fibers. The PCL-PCM system microspheres seem to have limitations in loading microcapsules. High levels of microcapsules, in fact, increase solution viscosity and lead to the disappearance of fiber morphology (for concentrations of microcapsules 35

Chapter II

greater than about 30% wt/wt with respect to the PCL). Probably, the critical concentration of microcapsules (i.e. the concentration at which fibers morphology is lost) depends on various factors such as the polymer matrix, microcapsules diameter, the interaction matrix-microcapsules shell material. Figure II.3 shows how the addition of 26% in PCL weight of microcapsules to the solution resulted in fibers with a completely different morphology. In particular, the fibers present a significantly lower diameter (in the range of 100 - 500 nm, with a maximum at about 250 nm) in comparison with neat PCL fibers. Probably, this is due to the increase in solution viscosity that improves the spinnability of the mixture. Another reason can be found in the elongational force produced by the microcapsule on the polymer emulsion during its flight in the air. The resulting mats spun from PCL-MPCM blend are stiffer than mats of PCL alone, whereas the appearance in terms of softness and tensile strength is almost the same as neat PCL mats.

Figure II.3a SEM images of composite PCL fibers-PCM.treated with electrospinning.

36

Thermoregulating Textile Materials

Figure II.3b SEM images of composite PCL fibers-PCM.treated with electrospinning at different magnification. Differential scanning calrimetry (DSC): In textile applications, fabric containing MPCM can be used to absorb excessive heat released by the human body or act as a barrier against cold Colvin (1998). In heating, the phase transition temperature of MPCM is above the bodys normal skin temperature, whereas in cooling it has a transition phase below the bodys normal skin temperature (fig.II.4). The resulting garments can regulate the body temperature by absorbing or generating heat when the temperature is too high or too low, respectively Colvin (1998), LennoxKerr (1998), Shim (2001). This leads to an overall enhancement of thermophysiological wearing comfort and a prevention of heat stress situations as a serious health risk. Possible applications include jacket liners, long underwear, boot liners, glove liners, sleeping bags and socks Tao (2001).

37

Chapter II

Figure II.4 DSC thermogram of microencapsulated PCM

Figure II.5 shows the DSC thermogram of the PCL fibersMPCM composite reported in figure 4. An endothermic peak located between 15 and 42C is observable during the heating stage at 10C/min. The enthalpy released in this case is about 40J/g. This value is approximately proportional to the concentration of MPCM in the fiber composites (180 J/g * 0.26 = 46.8 J/g). In general, the heat storage capacity of the fibers composites samples increased as the addition of microcapsules increased up to the critical concentration. However, the increase was not proportional to the number of microcapsules, indicating that for concentrations close to the critical value a large quantity of microcapsules were lost in the three-dimensional fibers network.

38

Thermoregulating Textile Materials

Figure II.5 DSC thermogram of PCL fiber-MPCM composites (26% wt/wt in microencapsulated PCM). II.5 Conclusions Electrospinning represents the system with highest efficiency in the dispersion of microcapsules. By appropriate selection of MPCM, polymer matrix and processing conditions very high concentrations of microcapsules can be reached without particular detriment to the resulting polymer fibers. Differently from other techniques applied, electrospinning works at room temperature and in normal air conditions, which may protect from harsh treatments and enhance thermal efficiency by avoiding capsule damage. The electrospinning process is capable of producing extremely thin fibers with diameters ranging from microns down to a few nanometers (typically hundreds of nanometers in diameter). It is a very versatile technique and allows the use of a variety of polymers, blends of different polymers as well as the integration of additives, in order to tailor different application requirements.

39

Chapter II

40

Chapter III Electrospinning carbon nanotube polymer composite nanofibers

The unique and exceptional physical properties of carbon nanotubes have inspired their use as a filler within a polymeric matrix to produce carbon nanotube polymer composites with enhanced mechanical, thermal and electrical properties. A powerful method of synthesising nanofibers comprising these polymer composites is electrospinning, which utilises an applied electric stress to draw out a thin nanometer-dimension fiber from the tip of a sharp conical meniscus. The focussing of the flow due to converging streamlines at the cone vertex then ensures alignment of the carbon nanotubes along the fiber axis, thus enabling the anisotropic properties of the nanotubes to be exploited. We consider the work that has been carried out to date on various aspects encompassing preprocessing, synthesis and characterisation of these electrospun polymer composite nanofibers as well as the governing mechanisms and associated properties of such fibers.

III.1 Introduction A carbon nanotube (CNT) is a seamless hollow cylindrical structure with a diameter of 12 nanometers, comprising a hexagonal network of carbon atoms, one or both ends of the nanotube being capped with a hemispherical fullerene molecule Iijima (1991). In general, CNTs take two forms, namely single-walled nanotubes (SWNTs) and multi-walled nanotubes (MWNTs). SWNTs are single cylindrical layers of this carbon atom network, as shown in figure III.1a, whereas MWNTs consist of several layers in the form of concentric cylinders bounded by weak van der Waals forces, as shown in figure III.1b.

Chapter III

Figure III.1 Schematic illustration of (a) a single-walled nanotube (SWNT), and (b) a multi-walled nanotube (MWNT).

Due to the covalent bonding between the carbon atoms arising from sp2 hybridisation, similar to that of graphite, CNTs exhibit a high surface area to volume ratio and hence very high tensile strength (typically 100600 GPa), elastic modulus (typically 15 TPa) and compressive strength to weight ratios Pan (1999), Li (2000). They are extremely flexible due to their ability to reversibly absorb large stresses by deforming through twisting and buckling motions without fracturing Iijima (1996), Lourie (1998). In addition, the structure of the molecular sites facilitates the possibility of tailoring its surface functionality to specific interfacial requirements for various applications Kim (2005). The electrical and thermal properties of CNTs (SWNTs in particular since, unlike MWNTs, their atoms are bound in a single covalent network) are extremely anisotropic. The electrical conductivity depends on the chiral angle, which is the angle at which the atoms of the CNT are twisted about the main axis of the CNT. Bachilo et al Bachino (2002) classified these as armchair, zigzag and intermediate (or chiral) structures according to the chiral angle and diameter, as shown in figure III. 2.

42

Polymer Composite Nanofibers

Figure III.2 Schematic illustration of various carbon nanotube structural forms. (a) Armchair structure. (b) Zigzag structure. (c) Intermediate or chiral structure. Armchair forms exhibit metallic properties and hence have extremely high conductivities and current densities whereas the zigzag and intermediate forms are either semi-conducting or semimetallic. Both the thermal and electrical conductivities are observed to increase with alignment along the tube axis compared to the unaligned material Hone (2000). These unique size, structural, and topological characteristics endow CNTs with excellent electromechanical, thermal, chemical and optical properties, demonstrating the potential of CNTs to revolutionise the development of nanocomposite materials and nanodevices Dresselhaus (1996), Ajayan (1999), Baughman (2002). There are several methods currently used for the synthesis of CNTs. Among these are arc-discharge, laser ablation Guo (1995), vapour growth Tibbetts (1983) and chemical vapour deposition Ren (1998). These methods can produce aligned SWNTs and MWNTs but at low yields, and so remain prohibitively expensive. At present, no inexpensive high-yield production methods are known, though novel methods show promise, including fluidised bed reactors Sinnott (2001). It should be noted that the vapour growth method produces vapour-grown carbon fibers (VGCF) which are similar in structure to MWNTs but with larger diameters, of the order of 43

Chapter III

100 nm. CNTs tend to aggregate parallel to each other in tight bundles due to van der Waals forces Tucknott (2002), a problem in fabrication that prevents achievement of their superior physical properties, especially through exploitation of their inherent anisotropy. These bundles consist of a dense entangled network with a typical characteristic length scale of 10 100nm and are prone to breakage when sheared at stresses significantly below that required to break a single nanotube along the CNT bundle axis Lefrant (2001), Dror (2003). Several techniques have therefore been developed to attempt orientation of individual CNTs along a single axis. In preparation for processing, CNT bundles are deagglomerated using ultrasonication Biercuk (2002) and ballmilling Kim (2002). During solution processing, CNTs can also be aligned by directly introducing the aqueous CNT dispersion into a rotating poly(vinyl alcohol) solution, the induced shear resulting in CNT dispersion Vigolo (2000). Other techniques that have been applied are the addition of surfactants to stabilise against aggregation, use of magnetic field alignment Smith (2000), spinning of extruded polymer composite melts Kumar (2002), and, mechanical stretching Jin (1998), Bower (1999) ,and shearing Tang (1999), Hobbie (2003) of polymer/nanotube composite films. Schlittler et al. Schlittler (2001), on the other hand, considered the self-assembly of perfectly ordered carbon nanotube crystals deposited onto nanopatterned chemically modified substrates using thermolysis. The other difficulty in exploiting the exceptional properties offered by CNTs for common applications is due to their very small sizes. Thus, methods are required to incorporate these nanoscale structures at the macroscopic level. One area in which active research is being pursued is the embedding of CNTs within a polymer matrix such that conventional polymeric fibers are reinforced to produce an ultra-light composite material with enhanced physical properties Breuer (2004). These reinforced polymer nanocomposite fibers with micron and nanometer diameters are extremely versatile materials which have a wide variety of applications in catalysis Dersch (2005), electronic, optical or optoelectronic devices e.g. photocells, LEDs, highly conducting nanowires, etc. Norris (2000), Bradley (2003), Kymakis (2004), filtration Tsai (2002), lightweight and breathable textile materials for protective military clothing Gibson (1998), Gibson (2001), smart materials for aerospace vehicles Park (2002), Smith (2004), and, piezo/ pyro/ferro-electric sensor and actuator technology Steinhart (2003), Tahhan (2003). A wide variety of other possible applications are compiled in Breuer and Sundararaj and Huang et al. Specific polymeric properties such as biodegradability and biocompatibility Huang (2003) have also been exploited such that polymer fibers are currently being used in biopharmaceutical and biotechnological applications, e.g. biosensors, wound care therapy, drug delivery and encapsulation, medical prostheses, and 44

Polymer Composite Nanofibers bioscaffolds for tissue and orthopaedic engineering Zong (2002), Wnek (2003), Joshi (2005), Yeo (2005), although biodegradable polymer nanocomposite fibers have yet to be fully explored. Polymer/CNT composites have been synthesized through a host of fabrication methods, although there remains no high-yield commercial process for fabricating such composites. Among these are melt-mixing techniques such as extrusion, internal mixing, injection moulding and blow moulding (and references therein), combinations of solvent casting and melt mixing Haggenmueller (2000), film and spin casting Safadi (2002), in-situ polymerization Jia (1999), shear mixing Andrews (2002), matrixassisted pulsed laser evaporation, chemical vapour deposition growth of CNTs and deposition onto the surface of carbon fibers Thostenson (2002). A detailed discussion of the different processing techniques is given here. Nevertheless, these widely different techniques, unite in the quest to limit filler (CNT) dispersion to avoid stress concentration localisation, maximise filler alignment whilst maintaining structural integrity of individual CNTs, and increase interfacial bonding between the filler and polymer matrix to enhance load transfer across the interface. In the absence of these requirements, the polymer/CNT composite is ineffective and the enhancement of its properties compromised. A very powerful synthesis technique for fabricating polymer/CNT composite fibers is electrospinning. Of late, there has been an explosive growth in the use of electrospinning for fiber synthesis, and more recently, for the production of polymer nanocomposite fibers. Electrospinning provides a safe, low-cost and rapid one-step integrated processing assembly to produce continuous nanofibers, an advantage over other multi-step traditional discontinuous fiber template manufacturing methods. In addition, the technique can also be massively parallelised for mass fiber production without significantly affecting production costs. The use of electrospinning to synthesise polymer/CNT nanocomposite fibers offers a possible way to simultaneously align CNTs along a single axis during processing without compromising the structural integrity of the individual CNTs. The focussing of the meniscus into a sharp conical shape gives rise to a converging flow and hence concentration of the streamlines in the extremely narrow region at the tip of the cone from which the jet and hence, a fiber emanates. As a result the CNTs dispersed in solution are oriented along the streamlines and then along the fiber axis. The extremely large longitudinal strain rate of the emanating jet also ensues in the CNT alignment within the fiber. In contrast to agglomeration of embedded CNTs in the fiber, aligned CNTs within the fiber increases the effective contact area for interfacial bonding with the polymer matrix and enhances the effective aspect ratio for reinforcement. For these reasons and because of the low cost and relatively uncomplicated process involved, electrospinning is quickly becoming an attractive and accepted technique for embedding CNTs 45

Chapter III

within micron and nanometer sized polymeric fibers. A summary of the various polymer/CNT nanocomposite fibers generated by electrospinning is summarised in table III.1. Table III.1 Summary of the various polymer/CNT nanocomposite fibers

While it is not possible to provide a comprehensive list here, it suffices in demonstrating the viability and increasing popularity of using electrospinning to generate these polymer nanocomposites for various applications. In order to obtain stable aqueous CNT dispersions, surfactants or amphiphilic polymers such as sodium dodecyl sulphate, poly(vinyl pyrrolidone) and natural polysaccharide Gum Arabic, which absorb onto the hydrophobic CNTs, are added . The mixture is then sonicated for several hours after the addition of the dispersing agents thus creating a stable homogeneous dispersion. Subsequently, the dispersion is mixed with the polymer/solvent mixture and passed into a micro-syringe (typically several hundred microns in diameter) by means of a syringe pump. As with the electrospinning of conventional polymeric fibers, voltages of several kV (typically 1030 kV) are applied across the syringe and the ground electrode, which consisted of a grounded disk rotating at several hundred rpm. Nevertheless, CNT dispersion by sonication possibly introduces defects in the CNTs themselves Kearns (2002). Kearns and Shambaugh suggest an optimum sonication period, below which inadequate dispersion is achieved and above which CNT damage or breakage resulted. Other means of CNT dispersion that could possibly circumvent this problem are the use of ball milling, AC electrophoresis and polymer wrapping to overcome the van der Waals forces between the CNTs. The latter employs reversible thermodynamic gradients such that the addition of polymer forms a tight association with the sides of the CNTs thereby disrupting the van der Waals interaction between CNTs and also the hydrophobic CNT interface with 46

Polymer Composite Nanofibers water Kearns (2002) . The CNT surface can also be functionally modified to provide specificity for improved interactions between the CNT and its matrix, thus allowing the enhancement of processability and polymer composite properties, and permitting its coupling with other materials in its final integration into products Yamamoto (1998). A summary of different approaches to chemical functionalisation is given in Breuer and Sundararaj, namely, covalent functionalisation, polymer/surfactant supramolecular adduct noncovalent exohedral functionalisation, chemical vapour deposition surface modification, plasma surface treatment and gamma radiation functionalisation. Electrospun polymer nanocomposite fibers can be characterised through visual inspection of the CNTs, the polymer matrix and the CNT/polymer matrix interface in order to determine the CNT alignment in the fibers such that the fiber properties can be optimised. Dror et al. examined electrospun polymer nanocomposite fibers with low electron dose transmission electron microscopy (TEM) with acceleration voltages of 120 kV by direct deposition of the fibers onto a perforated carbon-film-coated copper grid. In addition to TEM, Kim et al. also utilised field emission gunenvironmental scanning electron microscopy (FEG-ESEM). Nevertheless, optical and electron microscopy characterisation is not without its difficulties, in particular, due to the small length scales involved and hence the inability to distinguish between single CNTs. Moreover, Breuer and Sundararaj report difficulty in implementing electron microscopy due to issues with regard to sample preparation and the lack of polymer/ CNT contrast. They list other methods, such as magnetic force microscopy (MFM) which exploits the interaction between the CNTs and the applied magnetic field, thus allowing direct visualisation of the polymer embedded CNTs Breuer (2004) and scanning Kelvin microscopy which employs an oscillating probe to determine the conductivity distribution in heterogeneous materials such as the conducting CNTs in the insulating polymer matrix OConnell (2001). Yet another characterisation method that has been used is X-ray diffraction (XRD), as employed by Dror et al. Dror (2003). In XRD, long exposure times, typically 1 day for wide-angle diffraction and 1 3 days for small-angle diffraction, are required due to the small CNT quantities. FigureIII. 3a and b show TEM images of MWCNT embedded electrospun poly(ethylene oxide) and polycarbonate fiber morphologies indicating some success of orienting individual CNTs along the fiber axis Kim (2005).

47

Chapter III

Figure III.3 Transmission electron microscopy (TEM) images of carbon nanotube embedded polymer composite nanofibers showing alignment of the nanotubes along the fiber axis. (a) MWNT embedded poly(ethylene oxide) nanofiber. (b) MWNT embedded polycarbonate nanofiber. Nevertheless, some electrospinning carbon nanotube polymer composite nanofibers of convolution in CNT orientation are entangled and protrude from the fiber, resulting in poor alignment of the CNTs along the fiber axis, as shown in figure III.4.

Figure III.4 Transmission electron microscopy (TEM) images of carbon nanotube embedded polymer composite nanofibers showing irregularities in the alignment of the nanotubes within the fiber. (a) Protrusion of the nanotube from the fiber. (b) Entanglement of the nanotubes within the fiber. Such irregularities were observed to be closely correlated with irregularities occurring in fiber diameter. The source of the irregularities in the CNT orientation within the fiber is as yet unclear, although Dror et al. Dror (2003) give possible interpretations, i.e. the result of topological defects in the CNTs or the failure to completely stretch the CNTs as the fiber is ejected from the meniscus during the electrospinning process. By comparing cryoTEM images of vitrified CNT dispersions with those of raw nanotubes, Dror et al. suggest that irregularities could have possibly been introduced during the sonication process, consistent with the findings of Kearns and Shambaugh, reported above. Kim et al. Kim (2005) also suggest that the smaller the nanocomposite fiber diameter, the greater the tendency for curves to be introduced in the fiber. They attribute this to the curvature of the MWCNTs themselves, the thin fibers being unable to entirely accommodate or encapsulate those MWCNTs with large curvatures. In addition, they observe that nanopore formation on the fiber surface are suppressed due to insufficient space for phase separation to occur. To study the in situ mechanical deformation of the electrospun nanocomposite fibers, Kim et al. employed a novel method through the use of thermal stresses induced by a TEM electron beam. In this method, the impingement of the electron beam on the fiber results in local thermal expansion and hence tensile deformation, 48

Polymer Composite Nanofibers the strain rate being controlled by the amount of beam flux applied Velasco-Santos (2002). This tensile strain results in fiber elongation and subsequent necking, the necking occurring specifically at the end of an embedded MWCNT. This is due to some slippage of the MWCNTs in the direction of the applied tensile stress occurring beyond a threshold stress value. Kim et al. suggest that this CNT slip mechanism, coupled with the concentration of stress in the nanopore region (the nanopores form effective sites for which the stress concentration results in plastic deformation and hence necking) and the transfer of the mechanical load from the polymer matrix to the MWCNTs, are responsible for the enhancement of the critical fracture strain and also the increased toughness reinforcement of the fiber. III. 2 Experimental Results -Materials. Poly(-caprolactone) in pellets was purchased from Aldrich (Mw=80,000). Carbon nanotubes were synthesized from Sigma-Aldrich, they are multi walled (MWNT) (OD=20-30 nm; ID=5-10 nm, length=0.5-200 micron, 95%). -Incorporation of Carbon Nanotubes in PCL. The preparation of the solution consists in three steps: 1) Firstly, carbon nanotubes (1% and 3% of the total weight) are added to acetone, plus 0.5% of surfactant polyoxyethylene sorbitan mono-stearate (Tween 60). 2) The beaker is placed in a sonication bath for 20 minutes, in order to have maximum dispersion. 3) Finally, the polymer is added (12.5%,13.5% and 15%) and is dissolved by the solvent. This operation takes a few hours. After that the solution (about 3 ml) was spun. -Processing technique. All the process was performed at room temperature.Fresher solutions (about 3 ml) were put inside a syringe and were driven by a syringe pump. This allowed control of the flow rate and the volume of polymer ejected during each experiment. The syringe that was used had a capacity of 5 ml and had a thin needle (0.8 mm OD). An electrical potential difference, which was around 30kV, was established between the surface of the liquid drop and the collector. The distance was around 0.3m. The nanofibers formed a mat on the collector. -Method. 49

Chapter III

SEM pictures were obtained with a LEO 420 microscope. The samples were covered with a 250 thick gold film using a sputter coater (Agar mod. 108 A). D.c. electrical conductivity was measured by using an HV generator (FUG HCN35-6500) and a picoammeter (Keithley 6514). III.3 Nanofibers Characterization Scanning electron microscopy (SEM) There are several parameters that affect size and defects associated with nanofibers produced by electrospinning. The parameters that were studied include concentration and electrode spacing. The main defect that is common and problematic to eliminate is called beads. In order to observe the morphology and diameter size of the fibers Scanning Electronic Microscopy (SEM) was utilized. Different percentages of PCL with 1% and 3% of multi-walled carbon nanotubes (MWNTs) were spun. If we compare fig.III.5a (12,5% of PCL with 1% of MWNT) with fig.III.5b (13,5% of PCL with 1% of MWNT) and fig.III.5c (15% of PCL with 1% of MWNT) we can see an entangled structure, and their size distribution is quite narrow. Moreover, the better fibers, with diameters of about 400nm, are electrospun fibers with a lower percentage of PCL.

50

Polymer Composite Nanofibers

Figure III.5 a.SEM images of 12,5% of PCL with 1% of MWNT.

Figure III.5 b SEM images of 13,5% of PCL with 1% of MWNT.

51

Chapter III

Figure III.5 c SEM images of 15% of PCL with 1% of MWNT. Nanofibers of 3% of MWNT in :12,5% PCL (fig.III.6a); 13,5% PCL (fig.III.6b) and 15% PCL (fig.III.6c) show an increase of the dispersion of CNT in the polymer matrix with the increase in percentage of carbon nanotubes. The diameter of the fiber is around 400nm for the sample with 12,5% PCL.

52

Polymer Composite Nanofibers

Figure III.6.a SEM micrographs of 12,5% PCL+ 3% MWNT.

Figure III.6.b SEM micrographs of 13,5% PCL + 3% MWNT.

Figure III.6.c SEM micrographs of 15% PCL + 3% MWNT. 53

Chapter III

III.4 Experimental evaluation of electric conductibility This analysis is devoted to the study of the electric conductibility of fibers obtained through electrospinning and therefore the comparison with nanocomposites obtained by cold drawing. Fibers of rectangular form were prepared with the purpose to carry out measures of conductibility in the longitudinal direction of the sample. The tension of measure is applied through metallic contacts to the extremities of the sample. However, it is fundamental to make sure there is perfect adherence of the sample to the electrodes, in order to guarantee the most uniform distribution possible of the electric field within the material. With this configuration measurements will be taken to determine the percolation threshold. Electric greatness, DUT, (device under test) settled on the sample and in order to measure the other greatness that appeared on the sample, resistance was finally calculated according to Ohms law R = V/I. For polymeric samples the measurements of tension constant resistance is more effective.

Figure III.7 Circuit for the measurement of tension constant resistance.

This analysis is devoted to the study of the electric conductibility of fibers obtained through electrospinning and therefore the comparison with nanocomposites obtained by cold drawing. Solutions of 12,5% wt/wt of pure PCL and of 12,5% wt/wt of PCL with 2% and 3% wt/wt of carbon nanotubes were prepared. The solutions were submitted to electrospinning and to cold drawing. Measures of conductibility were carried out exploiting the tension constant method, the various samples were submitted to tensions in the [0,5 4] kV range, with step of tension of 0,5 kV (in terms of electric field between 1,67*104 V/m e 1,33*105 V/m). 54

Polymer Composite Nanofibers All the measures are effected in equitable time, with an interval of 2 minutes after every phase of discharge. The results are reported in the graph in figure III.8. As can be seen in the graph, for concentrations up to 2% there is no substantial improvement in conductivity, while a considerable increase (around 6 orders of greatness) is recorded for concentrations in the range of 2%-3%.

1,E-06 1,E-07 Conductivity [s/m] 1,E-08 1,E-09 1,E-10 1,E-11 1,E-12 1,E-13

Electrospinning

0%

2% % wt/wt MWNT

3%

Figure III.8 Course of the in-operation conductivity of the percentage of nanotubes inserted in the composite, obtained by electrospinning technique. From this it can be deduced that the percolation threshold is contained in this range, in that the strong increase of conductibility is justified by the formation of the conductive channel within the matrix. Figure III.9 shows the course of the samples submitted to cold drawing. In this case no percolation phenomenon is observed. Or, the carbon nanotubes do not succeed in creating a percolation grate to allow a passage of flow.Various samples were submitted to tensions in the range [0,5 4] kV with step of tension of 0,5 kV (in terms of electric field between 1,67*104 V/m e 1,33*105 V/m).

55

Chapter III

1,E-09

Cold drawing
Conductivity [S/m] 1,E-10 1,E-11 1,E-12 1,E-13 0% 2% % wt/wt MWNT 3%

Figure III.9 Course of the in-operation conductivity of the percentage of nanotubes inserted in the composite, obtained by cold drawing technique.

The same behaviour occurs in the samples obtained by casting process.

1,E-06 1,E-07 Conductivity[S/m] 1,E-08 1,E-09 1,E-10 1,E-11 1,E-12 1,E-13 0% 2% % wt/wt MWNT 3%

Casting Cold drawing Electrospinning

Figure III.10 Comparison among the two techniques and with the reference.

56

Polymer Composite Nanofibers III.5 Conclusion Electrospinning represents a technique that is able to produce fibers with diameters of nanometers with organic fillers such as carbon nanotubes. The obtained nanocomposites constitute a matrix of PCL containing well dispersed filler. The composites show some important electric ownerships. In the presence of the electric field, the fibers obtained, through this technique, show a percolation network able to increase the conductivity of around six orders of greatness in comparison to the insulator. In order to confirm this ownership further, we have performed comparative tests between two techniques: electrospinning and cold drawing. The fibers obtained through cold drawing show the same course of the insulator, or any percolation phenomenon was observed in the passage since 2 to the 3% of filler inside the matrix of PCL. This result is a clear demonstration that through this technique we have obtained fibers in which the filler is well dispersed and oriented inside the polymeric matrix.

57

Chapter III

58

Chapter IV Nanocomposites

Polymer nanocomposites are mixtures of polymer and nanometer length scale particles, whereas, conventional polymer composites contain micrometer scale particles. The advantages of nanocomposites over macrocomposites include reduced filler amount and better properties than obtained for the conventional composites. Among the nanoscale materials used to form nanocomposite are clays, which can be swollen. Clay particles have high aspect ratios (ratio of length to width) and are composed of layered structures, so that the degrees of enhancement in properties obtained can be varied according to the types of nanocomposite structures. The efficiency of montmorillonite to modify the properties of the polymer is primarily determined by the degree of its dispersion in the polymeric matrix. The hydrophilic nature of the mineral hinders homogeneous dispersion of the montmorillonite in the polymer; however, due to their rich intercalation chemistry, the clays can be organically modified and made compatible with the organic matrix. By replacing the hydrophilic Na+ cations with a more hydrophobic onium ion, such as ammonium cations with long alkyl chains, the miscibility between the silicate layers and the polymer matrix can be enhanced, and a concomitant increase of the interlayer distance is observed. In this way, nanocomposites are obtained in which the nanoparticles dispersed in the organic matrix are either intercalated by the polymer or exfoliated into individual silicate layers of 1 nm thick Lagaly (1999), Le Baron (1999), Carrado (2000). Because of their nanometer size features, these nanocomposites possess unique properties typically not shared by the conventional microcomposites. There are three types Carrado (2003), Ray (2003)of polymer-clay nanocomposites. a) Tactoid structures are obtained when the interlayer space of the clay does not expand, due to its poor affinity with the polymer. As a consequence, clay in this form plays little role in property enhancement and is used only for cost reduction. Property improvement, in which tactoid structures are

Chapter IV

present, requires a loss of other properties as in the case of conventional fillers. b) Intercalated structures are obtained in the case of a little interlayer expansion of clay. In this case there is a well-ordered stacked multilayer structure as a result of some affinity between the polymer and the clay. The interlayer spaces of the clay are expanded as a few polymer chains penetrate the basal spacing of clay, but the shape of layered stack is maintained. c) In the case of exfoliated structures, host clay layers have lost their registry and are well separated into single layers within a continuous polymer matrix. This is due to very good affinity between the polymer and clay (Figure IV.1). The three nanocomposite structures shown in Figure IV.1 could be observed separately or jointly in one sample. .

(a)

(b)

(c)

Figure IV.1 Three types of nanocomposite structures: a) tactoid (conventional composite), b) intercalation, and c) exfoliation structure. In order to make the nancomposite with the most improved in properties, an exfoliated structure is required. The greatest dispersity of clay and interfacial interaction of clay platelets with polymer can be obtained in the exfoliated structure. Proper hydrophobicity of clay and compatibility between organically modified clay and polymer are critical factors in nanocomposite formation Li (2001). Additionally, large interlayer spacing in clay and interaction ability between clay and polymer are the other factors which are considered in the development of exfoliated nanostructures.

60

Nanocomposites Properties that can be improved with the addition of clay include: Physical properties; modulus, dimensional stability, etc. Barrier properties; reduced gas permeability to oxygen and carbon dioxide. Chemical resistance. Flame retardance and slow smoke emission. Optical clarity increased over that of conventional filler. Increased heat distortion temperature. IV.1 Materials Poly(-caprolactone): (PCL) was kindly supplied by Solvay (CAPA 6501) (Mn = 50 000 g/mol, Mw/Mn = 2) and was used without further purification. Organoclays: All of the organoclays used are modified montmorillonite clays. It has been furnished from Southern Clay Products (Texas, USA); namely Cloisite Na+, Cloisite 10A, Cloisite 30B. The properties of each organoclay are shown in Tab. IV 1. Table IV 1.Organoclays properties.
Properties / Treatment Organic Modifier Modifier Concentration Loss on Ignition (% Weight) X-Ray Results (d 001) [Angstrom]

Cloisite Na+ Cloisite 10A Cloisite 30B

None

None

11.7

2MBHT

125meq/ 100g clay 90meq/ 100g clay

39

19.2

MT2EtOT

30

18.5

61

Chapter IV

IV.2 Sample Preparation a)Preparation of nanohybrid by solvent casting of clays and PCL solution: An acetone suspension (10ml) containing 105 mg of clays was stirred for 3 hrs at room temperature and afterwards 1.5g of PCL was added. The mixture was stirred for 3 hrs at room temperature. The solvent was then slowly evaporated in a Petri dish, obtaining a PCL sample with 7% of clay. Films were obtained by molding the samples in a Carter laboratory press, at the temperature of 65C, followed by a quick quenching in a bath at 25C. b)Preparation of nanohybrid by electrospinning process: These solutions were prepared as mentioned above- for the composites obtained by solvent casting. After that the mixture was stirred for 3 hrs at room temperature, the fresher solutions were put inside a syringe and were driven by a syringe pump. The syringe that was used had a capacity of 5ml and leads to a thin needle (0.8mm OD). An electrical potential difference, which was around 20KV, was established between the surface of the liquid drop and the collector. The distance was around 0.2 m. The final composites formed a mat on the collector. IV.3 Methods Scanning Electron Microscopy (SEM). Pictures were obtained with a LEO 420 microscope. The samples were covered with a 250 thick gold film using a sputter coater (Agar mod. 108 A). X-ray Diffraction measurements (XRD). (XRD) measurements were performed with a Brucker diffractometer (equipped with a continuous scan attachment and a proportional counter) with Ni-filtered CuK radiation (=1.54050). Thermo Gravimetric Analysis (TGA). Measurements were carried out with a Mettler TC-10 thermobalance from 25 to 900C at a heating rate of 10C/min under an air flow. Differential Scanning Calorimetry (DSC). DSC measurements were carried out using a thermal analyzer Mettler DSC 822/400 at a heating rate of 10C/min between -50 and 200C.

62

Nanocomposites IV.4 Analysis of Nanocomposites -X-Ray Diffraction (XRD) In nanocomposite research, X-ray diffraction is used to determine the degree of exfoliation of organoclays in the polymer matrix. As the organoclay is dispersed through the polymer, it is forced to delaminate, increasing the distance between the silicate layers. This process is termed intercalation and results in greater distance between the silicate layers. The distance of separation is represented as a change in positioning of the peak from that of the pure organoclay on an XRD diagram. Braggs Law is commonly used to determine the spacing of the organoclay. Braggs Law equation is given as: n = 2dsin where n is an integer indication of the peak number. is the wavelength of the x-rays. d is the spacing between the clay layers. is the angle of incidence of the x-ray beam. Figure IV.2 shows the patterns of 15% PCL with 7% Na+ , 15% PCL with 7% 10A and 15% PCL with 7% 30B. The diffractogram shows an intense peak at 2=21 and 23, corresponding to the PCL crystallization. The peaks are sharp, and the polymer crystallinity is not influenced by the presence of the inorganic component. The first peak observed in all samples are about 3.3nm indicating the presence of intercalation of the materials.

Intensity

15% PCL +7 % 30B 15% PCL +7 % 10A

15% PCL +7 % Na

10

15

20

25

30

35

40

45

2
Figure IV.2 XRD diagram of 15% PCL + 7% organoclays obtained by casting solvent.

63

Chapter IV

15% PCL + 7% 30B

Intensity

15% PCL + 7% 10A

15% PCL + 7% Na
0 5 10 15 20 25 30 35 40 45

2
Figure IV.3 XRD XRD diagram of 15% PCL + 7% organoclays obtained by spinning process. Figure IV.3 shows the presence of inorganic component respectively at 2 values of ca. 4.88, 5.3 and 4.3. The PCL of the composites is crystalline and shows the usual peaks of its structure at 2=21.3 and 23.7, although of different intensity. The pattern of the composites obtained by electrospinning process with nanhybrid obtained by casting solution shows how this new technique allows a partial exfoliation of the inorganic lamellae for samples of 15% PCL + 7% 30B and 15% PCL + 7%Na+. -Scanning electron microscopy (SEM): In order to observe the morphology and diameter size of the fibers, scanning electron microscopy (SEM) was utilized.

64

Nanocomposites

Frequency
0

200

400

600

800

1000

F ib er d iam eter

Figure IV.4 a SEM micrographs of electrospun fibers and diameter distribution of 15%PCL + 7% Na+.

Frequency
0

200

400

600

800

1000

1200

1400

Fiber Diameter

Figure IV.4 b SEM micrographs of electrospun fibers and diameter distribution of 15%PCL + 7% 10A.

Frequency
0

200

400

600

800

1000

Fiber Diameter

Figure IV.4 c SEM micrographs of electrospun fibers and diameter distribution of 15%PCL + 7% 30B.

65

Chapter IV

Figure IV.4 shows that the fibers of composite (a) are homogeneous with no beads, the diameter distribution of the fibers is relatively wide (ranging between 0.3 and 0.7m). For samples with clay minerals modified by organic cations as fillers, we can distinguish two different situations. The composite, (2MBHT as organic modifier) (b), shows fibers with some beads and drops, with an average diameter of 1000 50 nm. However, a higher bead density generally indicates a system with a viscosity too low to offset the surface tension. While composite, (MT2EtOT as organic modifier)(c), shows fibers which are much more individual, uniform, and randomly oriented with an average diameter of 50050 nm. -Thermo Gravimetric Analysis (TGA): All the samples were analyzed by thermogravimetric analysis to determine the degradation temperature and confirm the content of the inorganic component in the nanohybrid materials after the thermal decomposition of the organic part in air up to 800C. Figure IV.5 shows the TGA curves of samples obtained with the electrospinning process. All samples thermally degrade to volatile products above 300C and end above 500C. It appears that the introduction of inorganic material anticipates the onset degradation. No substantial changes are observed for 7% of the organic molecules. The other composite samples, obtained with solvent casting (fig. IV.6), showed a very similar behaviour among them. Thermograms were used to confirm the composition.
100 80
15% PCL + 7% Na 15% PCL + 7%10A 15% PCL + 7% 30B
+

Weight(%)

60 40 20 0 200 400 600 800 1000

Temperature (C)
Figure IV.5 Thermal degradation of composites with different amounts of organoclays, obtained with electrospinning process.

66

Nanocomposites

100 80
15% PCL + 7% Na 15% PCL + 7%10A 15% PCL + 7% 30B
+

Weight(%)

60 40 20 0 200 400 600 800 1000

Temperature (C)
Figure IV.6 Thermal degradation of composites with different amounts of organoclays, obtained with solvent casting. -Differential Scanning Calorimetry (DSC): Figure IV.7 reports the calorimetric curves for all samples obtained with the electrospinning process. We observe the melting range around 60C, where the peak for pure PCL is centered. No relevant changes are observed either in samples obtained by solvent casting (fig IV.8).
15% PCL + 7% 30B 15% PCL + 7% 10A 15% PCL + 7% Na Heat Flow
+

20

40

60

80

100

120

Temperature (C)

Figure IV.7 Calorimetric curves for composite samples by electrospinning process. 67

Chapter IV
15% PCL + 7% 30B 15% PCL + 7% 10A 15% PCL + 7% Na Heat Flow
+

20

40

60

80

100

120

Temperature (C)

Figure IV.8 Calorimetric curves for composite samples by solvent casting. IV.5 Conclusion The successful fabrication of montmorillonite and organic clay chemically modified reinforced PCL nanofibers by electrospinning was realized. In addition we prepared the same samples by solvent casting and afterwards we compared the final composites obtained with two techniques. The morphological and X-rays analyses showed a significant difference between the two techniques. A sample with 7% Cloisite 30B by electrospinning showed the best interaction with the polymer matrix, probably due to the structure of the clay. In fact, Cloisite 30B materials possessed two hydroxyl groups in their modifier, which make it possible to interact strongly with the carboxyl group in PCL molecules, promoting the intercalation of PCL molecules into clay gallery. However, these results mean a partial exfoliation structure, producing fibers with an average diameter in the range of 50050 nm. Samples obtained by solvent casting showed an intercalation structure, whereas sample with 7% Cloisite 10A showed fibers with the inorganic component residing outside the fibers and not exfoliated, probably due to lower compatibility with the polymer matrix. Besides, in this case we did not have homogeneous dispersion of the inorganic component inside the polymer matrix which caused unsatisfactory electrospinnability producing fibers with beads and drops.

68

Chapter V Encapsulation and exfoliation of inorganic lamellar fillers into polycaprolactone

In the past decade, biodegradable and biocompatible polymers have received significant attention because they are environmentally friendly and are extensively used in biomedical applications. In this direction, poly(caprolactone) (PCL) is one of the most promising polymers, and it can be used in many fields, such as drug delivery, scaffolds, and guided bone regeneration Kim (2000), Ma (2004). On the other hand, there are few studies on electrospinning of polymer/ layered inorganic nanocomposites. Poly-(lactic acid) Lee (2005), Nylon 6 Fong (2002) and poly(methyl methacrylate-comethacrylic acid) Wang (2005) are some of the polymers that have been used for the preparation of nanocomposite fibrous membranes. Membranes composed of nanosized fibers, either of neat PCL, or PCL in which inorganic and organomodified layered double hydroxides (LDHs) are encapsulated by electrospinning. The structure of LDHs originates from the packing of layers built up in a manner similar to that found in brucite, the naturally occurring Mg(OH)2 Cavani (1991), Rives (2001). In this mineral, the Mg atoms are octahedrally coordinated by six oxygen atoms belonging to six OH groups; each OH group is, in turn, shared by three octahedral cations and points the hydrogen atom to the interlayer space. When some of the Mg(II) cations are isomorphously replaced by Al(III) cations, the substitution creates positive charges balanced by the presence of counter anions located in the interlamellar region Costantino (1998), Suryanarayana (2001) The ability to replace these anions by simple ion-exchange procedures makes LDHs a unique class of layered solids to be used as a host of active molecules. We used both inorganic and organically modified LDH to be introduced into PCL by electrospinning.

Chapter V

The ability to introduce LDH into drug molecules makes these composites very attractive in the biomedical field for the production of many items such as membranes and suture threads with controlled release. The fibrous structure, containing the inorganic samples, was investigated as a function of the solution concentration, the applied voltage, and the inorganic content. V.1 Experimental Materials The PCL was kindly supplied by Solvay (CAPA 6501; Mn = 50 000), the commercial hydrotalcite (carbonate form) (LDH) magnesium aluminum hydroxycarbonate Mg6Al2(CO3)(OH)16. The 4H2O (basal spacing 3.83 ) was purchased from Aldrich, and the hydrotalcite modified with the organic residue (12-hydroxydodecanoic acid) (LDH-HA) (basal spacing 22.7 ) was obtained according to a previously reported procedure Suryanarayana (2001). The Mg0.66Al0.34(OH)2(CO3)0.17 precursor material was obtained using a procedure based on the in situ formation of ammonium carbonate from urea Costantino (1998). The precursor was converted in nitrate form, of formula Mg0.66Al0.34(OH)2 (NO3)0.34 0.62H2O, by titrating the carbonate form, previously suspended in 1M NaNO3 aqueous solution, with a 0.100 M HNO3 solution operating at pH-stat mode and pH = 4.5. 12-Hydroxydodecanoic acid (97% Aldrich), hexane (Lab Scan), tetrahydrofuran (THF, Lab Scan) were used as received. A solution of deprotonated 12-hydroxydodecanoic acid was prepared by adding 28.3 mL of a 1 M KOH water solution to 100 mL of a 0.283 M of 12hydroxydodecanoic acid solution in THF. A 55.2 mmol (5.05 g) portion of LDH in nitrate form was added to the 12- hydroxydodecanoic acid solution. The mixture reaction was then stirred at reflux for 12 hrs. The pale yellow solid obtained was isolated by centrifugation, suspended in THF, and centrifuged two more times. The product was finally filtered and dried at 80 C. High-Energy Ball Milling (HEBM) Experiments. The powder mixtures were obtained using HEBM technology. Such a process has been revealed as an alternative and innovative strategy that involves an efficient mixing of two or more species by mechanical milling, avoiding high temperatures and solvents Sorrentino (2005), Gorrasi (2007). In every single experiment, hydrotalcite powder (1%, 5% wt/wt) and PCL were milled in the solid state in a Retsch (Germany) centrifugal ball mill (model S 100). Mass samples were milled in a 50 cm3 cylindrical steel jar 70

Encapsulation of LDH-LDH HA with five steel balls of 10 mm diameter. The rotation speed used was 580 rpm, and the milling time was fixed to 60 min. Electrospinning Procedure. Electrospinning of PCL fibers was carried out at room temperature at a high voltage of 20 kV (HV Power Supply, Gamma High Voltage Research, Ormond, FL). The spinneret used in the experiments had an inner diameter of 0.8 mm. A copper wire was mounted in the spinneret and used as the positive electrode. Grounded aluminum foil was used as the counter electrode and mounted at a distance of 20 cm from the spinneret. Continuous PCL fibers were collected on the aluminum foil in the form of a fibrous mat. Electrospinning conditions were optimized to produce PCL nanofibrous mats composed of individual fibrils less than 1000 nm in diameter and without bead formation. Also, 15% (wt/wt) of pure PCL, 15% (wt/wt) of composites formed by PCL, and 1% and 5% (wt/wt) of inorganics were stirred vigorously for 1 h until complete dissolution in acetone and then spun. V.2 Methods X-ray Diffraction (XRD) Measurements. XRD measurements were carried out on the spun samples with a Brucker diffractometer (equipped with a continuous scan attachment and a proportional counter) with Ni-filtered Cu K radiation (= 1.54050 ). Thermogravimetric Analysis. Thermogravimetric analysis was performed with a thermogravimetric analyzer SDTQ TA Mettler with a 5 Kmin-1 heating rate, under flowing air between 25 and 900 C. Differential Scanning Calorimetry (DSC). DSC measurements were carried out using a thermal analyzer Mettler DSC 822/400 at a heating rate of 10 C/min between -50 and 200 C. Scanning Electron Microscopy (SEM). The morphology and diameter of the electrospun nanofibers were determined with a scanning electron microscope (SEM; JEOL JSM-T300). A small section of the fiber mat was placed on the SEM sample holder and sputter-coated with gold prior to the analysis. Energy DispersiVe X-ray Spectroscopy (EDX). Some samples were also characterized by microanalysis to investigate their chemical structure. Elemental analysis and element mapping were conducted with a fieldemission scanning electron microscope (FESEM, model LEO 1525, Carl Zeiss SMT AG, Oberkochen, Germany) equipped with an EDX spectroscope (INCA Energy 350, Oxford Instruments, Witney, United Kingdom).

71

Chapter V

V.3 Analysis of Nanocomposites Scanning Electronic Microscopy (SEM). We chose the concentration of 15%, with a distance of 20 cm, to spin either the pure PCL or the composites, and in Figure V.1(a) we show the obtained fibers. The nanofibrous structure of the sample is evident, which is composed of individual, uniform and randomly oriented fibers with an average diameter of 600 50 nm. Then LDH layers were dispersed into PCL solution, which was subsequently electrospun. Successful PCL in solution was spun with 1%wt and 5%wt of LDH as shown in Figure V.1(b,c). The presence of the inorganic component caused no noticeable change in morphology and average fiber diameter of the nanofibers.

Figure V.1a SEM micrographs of electrospun fibers, applied voltage 20kV, solution concentration of 15%wt PCL (magnification bar:10m).

Figure V.1b SEM micrographs of electrospun fibers, applied voltage 20kV, solution concentration of PCL + 1% LDH (magnification bar:10m).

72

Encapsulation of LDH-LDH HA

Figure V.1c SEM micrographs of electrospun fibers, applied voltage 20kV, solution concentration of PCL + 5% LDH (magnification bar:10m). A series of PCL nanofibers with the addition of 1-5 wt% acid modified filler (LDH-HA) were also fabricated by electrospinning processing. Figure 2(a) and (b) shows the electrospun nanofibers of composite PCL + LDH-HA with 1%wt of LDH-HA with different magnitudes. Figure 2(c) and (d) show the PCL + LDH-HA composite with 5%wt of LDH-HA with different magnitudes. The addition of the LDH-HA intercalation compound, resulted in electrospun nanofibers with significantly lower average diameter, in comparison with PCL or PCL-LDH nanofibers, in the range of 350 50 nm.

73

Chapter V

(a)

(b)

(c)

(d)

Figure V.2 Electrospun nanofibers of composite PCL + LDH-HA with 1%wt of LDH-HA with different magnitudes (a), (b), (c) and (d) shows the PCL + LDH-HA composite with 5%wt of LDH-HA with different magnitudes (magnification bar:10m). Thus, it is most probable that the dispersion of LDH-HA structures improved the electrospinnability of the mixture, as denoted by the lower average diameter of nanofibers obtained. LDH-HA is predominantly exfoliated and well distributed within the fiber and very likely oriented. To determine the structural organization of the inorganic component of either LDH or LDH-HA in the PCL fibers, we determined the chemical structures on the surface of the composite fibers with FESEM-EDX. In Figure V.3 we show the EDX micrographs of electrospun fibers PCL+5% LDH. Besides the elements always appearing in the micrographs (such as Fe and Cr due to the iron support of the samples and the coating of the samples with Cr), we observe the peaks of Mg and Al, indicating the presence of the inorganic component on the surface of the fibers. This means that LDH with carbonate does not interact with PCL in solution and is simply deposited on the surface of the fibers during PCL spinning. 74

Encapsulation of LDH-LDH HA At variance in Figure V.4, where the fibers of PCL+5% of LDH-HA are shown, we observe the absence of the peaks (Mg, Al) on the surface of the fibers. In this case, to ascertain whether the inorganic hybrid is located inside the fibers, we submitted the sample to thermal oxidation up to 450C, followed by an EDX observation. In Figure V.5 we show the result, pointing out the presence of Mg and Al. This is a clear indication that the hybrid LDH-HA is located inside the fibers. Therefore when the inorganic component is modified with an organic molecule and able to interact with the polymeric chains in solution, it can be incorporated into the fibers, also changing the fiber dimensions. On the other hand, when the inorganic is not able to interact with the polymer chains in solution, the polymer is spun without changing the fiber dimensions and the inorganic is simply deposited on the surface. Either way, we cannot exclude the presence of some microcrystals inside the fibers.

Figure V.3 FESEM equipped with energy dispersive X-ray spectroscopy (EDX) micrographs of electrospun fibers PCL + 5% LDH.

Figure V.4 FESEM equipped with energy dispersive X-ray spectroscopy (EDX) micrographs of electrospun fibers of the surface of PCL + 5% LDHHA.

75

Chapter V

Figure V.5 FESEM equipped with energy dispersive X-ray spectroscopy (EDX) micrographs of composites after thermal oxidation up to 450C of PCL + 5% LDH-HA.

X-ray diffractograms Figure V.6 shows the X-Ray diffractograms of pure PCL, PCL+1%LDH, PCL+5%LDH, and pristine LDH. The pure PCL, spun from the 15% solution in acetone, shows the well developed crystalline structure with the main peaks at 21.4 of 2 and 23.8 of 2 and a reduced amorphous fraction. The diffractograms of the composite samples show the peak of the LDH basal spacing at 11.6 of 2 (as evident when observing the diffractogram of pure LDH), with intensity increasing as inorganic content increase. This is an indication that the inorganic lamellae are neither exfoliated nor intercalated, but rather formed conventional microcomposites. We have previously shown that the inorganic component, which is not exfoliated, is simply deposited on the surface of the fibers, and X-rays confirm this result. In Figure V.7 the diffractograms of PCL+1%LDH-HA, and PCL+5%LDHHA, and LDH-HA are shown. The LDH-HA shows the peak of the basal spacing at a lower angle with respect to the LDH, that is, 3.9 of 2 (d= 22.67) due to the enlargement of the inorganic lamellae to host the bigger organic molecular anion. However, this peak is not present in the composite samples and this is a strong indication that the lamellae are exfoliated in the PCL matrix. Also in this case the previous analysis showing that the inorganic hybrid is located inside the fibers, is supported by the X-ray results. We have thus obtained nanocomposites by electrospinning the polymer with the organically modified hydrotalcite. PCL is crystalline, showing peaks at 21.4 and 23.8 of 2, although it seems that there is a higher amorphous fraction.

76

Encapsulation of LDH-LDH HA

LDH

Intensity

PCL + 5% LDH

PCL + 1% LDH

PCL
5 10 15 20 25 30

Figure V.6 X-Ray diffractograms of pure PCL, PCL+1% LDH, PCL+5%LDH, and pristine LDH.

77

Chapter V

LDH-HA

Intensity

PCL + 5% LDH-HA

PCL + 1% LDH-HA

10

15

20

25

30

Figure V.7 The diffractograms of PCL+1%LDH-HA, and PCL+5%LDHHA, and LDH-HA.

Thermal Properties Thermogravimetric analysis was performed in order to investigate both the degradation temperature and the inorganic content in the fibers. The thermal degradation of pristine PCL and its hybrids with different amount of LDH and LDH-HA loadings in air atmosphere are presented in Figure V.8 (a and b rispectively). PCL thermally degrades to volatile products above 300oC, and ends above 500C. It appears that the introduction of even very small amounts (about 1 wt %) of inorganic material anticipates either the onset degradation or the inflection point. Upon increasing the inorganic content no further decrease in the degradation temperature is observed. The effect of decreased thermal stability is due to the presence of the inorganic component, which decomposes at lower temperatures. On the other hand, in the nanocomposites, in which the inorganic lamellae were modified with 5% of the organic molecule, we observe a decrease in thermal stability either at the onset or at the inflection point. However, this decrease is much lower than that of the previous case. No substantial 78

Encapsulation of LDH-LDH HA changes are observed for 1% of the organic molecule, except for a fluctuation in the inflection points.

120
Weight (%)

100

PCL + 1% LDH

PCL + 5% LDH

PCL

80
PCL
600 700 Temperature (C) 800 900

Weight (%)

60

40

20
PCL + 5% LDH

0 0 200

PCL + 1% LDH

400

600

800

1000

Temperature(C)

(a)

120
Weight (%) PCL + 5% LDH-HA PCL + 1% LDH-HA

100

PCL + 1% LDH-HA
80

PCL

600

800

PCL

Temperature (C)

Weight (%)

60

PCL + 5% LDH-HA
40

20

0 0 200 400 600 800 1000

Temperature (C)

(b) Figure V.8 The thermal degradation of pristine PCL and its hybrids with different amounts of LDH (a) and LDH-HA (b) loadings in air atmosphere.

We observe the main melting range between 30 and 70C, with a peak centered at 60C for the pure PCL. No relevant changes are observed either 79

Chapter V

in the LDH or in the LDH-HA samples. In Table V.1 the calorimetric parameters and the results for the degradation temperature and the melting enthalpy are reported for all the samples. We observe a decrease in both parameters in the PCL-LDH microcomposites, whereas the nanocomposites show a lower decrease of the crystallinity and a lower decrease in the degradation temperature.

PCL + 5% LDH-HA

PCL + 1% LDH-HA

Heat Flow

PCL + 5% LDH

PCL + 1% LDH

PCL

50

100

150

Temperature (C)

Figure V.9 The calorimetric curves (DSC) for all the samples.

80

Encapsulation of LDH-LDH HA Table V. 1: The calorimetric parameters and the overall results for the degradation temperature and the melting enthalpy for all the samples
Sample 15% pure PCL Tf (C)1) 60,33 H (J/g) 1) 73,19 Td=Tf -To (C) 1) 5,34 Onset T (C) 2) 294,50 Inflect point Endset T (C) 2) T (C) 2) 408,00 560,07

15% + 1% LDH

60,33

64,10

5,68

279,10

380,17

569,19

15% + 5% LDH

60,67

66,22

6,00

273,05

378,83

552,41

15% + 1%LDH/HA

59,00

76,45

5,69

257,30

337,83

538,53

15% + 5%LDH/HA
1) 2)

59,17

87,23

5,24

253,15

331,67

529,40

Evaluated from DSC Analysis Evaluated from TGA Analysis

V.4 Conclusion We have obtained the successful fabrication of Layered Double Hydroxide (LDH) reinforced polycaprolactone (PCL) nanofibers by electrospinning. Both a completely inorganic LDH carbonate and one organically modified with 12-hydroxydodecanoic acid (LDH-HA) were incorporated into PCL using High Energy Ball Milling (HEBM), which has been proven to be an efficient and unconventional technique to produce composites and nanocomposites avoiding high temperatures and solvents. The organically modified LDH-HA was prepared by an ionic exchange reaction from pristine LDH and encapsulated into PCL from acetone solution at 15 wt%. The morphological analysis showed, in the case of pure PCL, fibers with an average diameter of 600 50 nm, and this dimension was maintained in the fibers with LDH carbonate, with the inorganic component residing outside the fibers and not exfoliated. While the fibers, with the organically modified LDH-HA showed significantly lower average diameter in the range of 300 50 nm, indicating improved electrospinnability of PCL. Moreover the 81

Chapter V

inorganic lamellae were exfoliated and residing inside the nanofibers, as shown by X-rays and Energy Dispersive X-ray Spectroscopy analysis (EDX).

82

Chapter VI Electrospinning of nanohybrids for controlled release in biomedical usage

In recent years there has been a great deal of interest in the field of drugloaded medical devices. Research has been concentrated on the formulation and preparation of drug-loaded medical devices, or engineering scaffolds, to be used for the local delivery of drugs Langer (1990),Gander (2001),Sam (2003),Sinha (2003),Lu (2004),Siha (2004). Local drug release has the advantage of avoiding systemic side effects, assuring the proper level of the drug in the specific site where it is needed. One of the simplest methods to obtain controlled delivery of an active molecule is to physically mix or blend it with a polymer, making it not as readily available to the biological system as when it is used alone. Therefore, great effort is currently being devoted to obtaining drug release at a slow rate, avoiding a too rapid delivery out of the polymeric matrix. The possibility of reducing, or even eliminating, oral administration of possibly dangerous drugs ( such as anti-inflammatories, antibiotics, antihaemorrhage, etc.) using new biocompatible polymers in which drugs are directly incorporated, is currently under investigation. In the medical field, in particular in surgery, it may be of great clinical utility to be able to modulate the activation of tissue repair and regeneration processes, which are at the base of the healing process. In this context, some surgical devices (e.g. sutures, membranes, osteosynthesis plaques) made of reabsorbable materials, capable of favouring the completion of the various reparatory phases and that do not necessitate further surgical intervention for their removal, are extremely useful. These devices can be advantageously loaded with active pharmacological substances. In fact, the possibility of local diffusion of a drug through biocompatible and reabsorbable matrixes offers the advantage of a drastic reduction of the systemically administered dose and, as a consequence, of related side effects. A limitation of these systems is the

Chapter VI

reduced capacity of guaranteeing an adequate and slow local release of the drug. A more remarkable innovation in this field is currently coming from nanoscience and nanotechnologies, which are opening new opportunities for producing materials with surprisingly unusual properties. The possibility of manipulating at atomic or molecular level induces structures with unique characteristics and completely new functionalities to be used not only as structural materials but also as smart structures for many fields such as drug delivery. Nanoscale reinforcement has found widespread interest in the macromolecular field, since polymer properties are remarkably improved compared with virgin polymer or conventional microcomposites. In this viewpoint, non-woven mats of electrospun nanofibers that are well known for their interconnected, three-dimensional porous structures and relatively large surface areas, provide a class of ideal material to mimic the natural extra-cellular matrix required for biomedical application. On the other hand, a wide range of additives, including biologically active molecules, can be immobilised on the inorganic lamellae with ionic bonds. The inorganic component is generally a lamellar solid, such as cationic or anionic clays, in which isomorphous substitution creates charges on the lamellae, counter-balanced by either cations or anions residing in the interlayer galleries space. The inorganic ions can be replaced by organic ions to improve the dispersion of the inorganic component into the polymer. So is possible to fix by ionic bonds, on the inorganic lamellae, active molecules (antimicrobial). These molecules not only improve the compatibility but can also be released with controlled kinetics in particular environments. Layered Double Hydroxides (LDHs or hydrotalcite like compounds) are a class of lamellar hosts consisting of positively charged layers balanced by exchangeable anions. Thus, functional organic species with anionic groups, which in general are much more numerous than those with cationic groups, can be intercalated, via ion exchange processes in this class of layered hosts. Furthermore, LDHs are biocompatible, easily obtainable and cheap. In this chapter there is a presentation of the preparation and characterization of polymeric membrane obtained by electrospining process, formed by LDH intercalated with three antimicrobial molecular anions ,and the dispersion of the intercalation compounds in a polymeric matrix. Polycaprolactone (PCL) was used as polymeric matrix, Zn/Al-LDH as layered system and benzoate (Bz), 2,4 dichloro benzoate (DCB), p-hydroxybenzoate (p-OHBz) as active molecules, having different substituents on the aromatic ring. Moreover we compared the release of these molecules from electrospun samples, with compact membrane.

84

Nanohybrids for Biomedical Usage VI.1 Materials Poly(-caprolactone) (PCL) was kindly supplied by Solvay (CAPA 6501) (Mn = 50 000 g/mol, Mw/Mn = 2) and was used without further purification. Benzoic, 2,4-dichlorobenzoic, p-hydroxybenzoic acids were purchased from Sigma-Aldrich. The hydrotalcite with formula [Zn0.65Al0.35(OH)2] (CO3)0.175 0.5H2O, was synthesized using the urea method Sam (2003). The corresponding chloride form was obtained by titrating the carbonate form, dispersed in a 0.1 mol/dm3 NaCl solution, with a 0.1 mol/dm3 HCl by means of Radiometer automatic titrator operating at pH stat mode, and pH value of 5. The composition of host structure was as following: [Zn0.65Al0.35(OH)2] (Cl)0.35 0.6H2O. In order to obtain the nitrate form the ZnAl-LDH in chloride form was suspended in a CO2-free aqueous solution 0.5 mol/dm3 of NaNO3 for 24 hours (1g/46 ml of solution). The recovered solid was washed three times with CO2-free de-ionized water and finally dried over a saturated NaCl solution (relative humidity, R.H., of 75%). It has formula [Zn0.65Al0.35(OH)2] (NO3)0.35 0.6H2O (LDH-NO3) and ion exchange capacity (IEC) equal to 2.95 meq/g. -Preparation of ZnAl-benzoate: The intercalation of benzoate anion was achieved by equilibrating the nitrate form of hydrotalcite with an aqueous solution of the anion 0.5 mol/dm3 (molar ratio organic anions /NO3-=3) for 24 hours at room temperature. The recovered solids were washed three times with CO2-free de-ionized water and dried at R.H.=75%. -Preparation of ZnAl-2,4 dichlorobenzoate and ZnAl-p hydroxybenzoate: The intercalation of the benzoate derivatives were performed equilibrating 1g of ZnAl-NO3 in 5 cm3 of solution water/acetone (1/1,v/v) 0.5 M of the anions obtained by titrating the corresponding acid forms with NaOH 1M until the hydrolysis pH 7. The suspensions were stirred for 3 days. The obtained intercalation compounds were washed with CO2-free de-ionized water and dried at R.H.=75%. -Intercalation of the nano-hybrids into polycaprolactone by High Energy Ball Milling (HEBM): Powders composed of PCL and LDHs (3% wt/wt) modified with the different organic molecules were milled in a cylindrical steel jar of 50 cm3 85

Chapter VI

with 5 steel balls of 10 mm of diameter. The rotation speed used was 580 r.p.m. and the milling time was 1 hour. -Preparation of compact membranes: The powders obtained by milling of PCL and the three modified LDHs, were melted in a Carver laboratory press between two teflon sheets, at the temperature of 80C, followed by a quick quenching in an ice-water bath (T=0C). Films 100 m thick were obtained and analyzed. The films appeared homogeneous, white and opaque, without visible clusters of inorganic material under an optical microscope. -Preparation of resorbable antimicrobial membrane for periodontal usage: Insertion of implants into thin or anatomically compromised alveolar bone is often followed by more or less extensive exposure of the implants threads. This exposure, which can be defined as an actual bone defect, generally requires regenerative treatment to promote bone formation above the exposed implant component. Until now, membranes are used to create a barrier for the cells that are not competent for regeneration, while favoring osteogenetic cells. The possibility of regenerating bone around the implant depends on several factors: the morphology of the defect type of membrane applied (resorbable or nonresorbable) type of filling material (autologous bone or replacement bone) the surgical technique used. Nowadays, the membranes, used as a white matrix, are compact, not friable in collagene obtained from bovine Achilles tendon, and resorbable. However, they can cause some some problems such as: - contra-indications in patients that are allergic to bovine products; - patients have to use an antimicrobial solution after their operation, several times a day over a long period of time (approximately a month). Consequently, the innovative idea ,after exploring many fields, is based on the fact that it is possible to fix by ionic bonds, on the inorganic lamellae, active antimicrobial molecules and these molecules not only improve compatibility with the polymer matrix, but they can be released with controlled kinetics in particular environments, too. Because of their peculiar structures, electrospun nanofibers have received a great deal of attention for uses in topical biomedical applications, in particular in the production of periodontal membranes. The electrospinning process of composites of PCL with the three modified LDHs (after milling), were carried out in a vertical spinning configuration. 86

Nanohybrids for Biomedical Usage The process was carried out at room temperature at a high voltage of 20 kV (HV Power Supply, Gamma High Voltage Research, Ormond, FL). The spinneret used in the experiments had an inner diameter of 0.8 mm. A copper wire was mounted in the spinneret and used as the positive electrode. Grounded aluminium foil was used as the counter electrode and mounted at a distance of 20 cm from the spinneret. Continuous fibers were collected on the aluminium foil in the form of a fibrous mat VI.2 Methods Scanning Electron Microscopy (SEM). Pictures were obtained with a LEO 420 microscope. The samples were covered with a 250 thick gold film using a sputter coater (Agar mod. 108 A). Energy Dispersive X-ray Spectroscopy (EDX). Some samples were also characterized by microanalysis to investigate their chemical structure. Elemental analysis and element mapping were conducted by field emissionscanning electron microscope ( FESEM, mod. LEO 1525, Carl Zeiss SMT AG, Oberkochen Germany) equipped with an energy dispersive X-ray spectroscopy (EDX, INCA Energy 350, Oxford Instruments, Witney, United Kingdom). X-ray Diffraction measurements (XRD). (XRD) measurements were performed with a Brucker diffractometer (equipped with a continuous scan attachment and a proportional counter) with Ni-filtered CuK radiation (=1.54050). Thermo Gravimetric Analysis (TGA). Measurements were carried out with a Mettler TC-10 thermobalance from 25 to 800C at a heating rate of 10C/min under an air flow. The Zn and Al content of the LDH, was obtained with standard EDTA titration after having dissolved a weighed amount of the sample (~100mg) in a few drops of concentrated HCl and diluted with water to 50ml. The amount of NO3- and organic in solution, before and after equilibration, was determined by Ion Chromatography using a Dionex 2000 Ion chromatograph equipped with an ionic conductivity detector. UV spectroscopy, set at different wavelengths,with respect to the active molecules, was used to quantify the amount of these molecules released into the solution in each time interval. A calibration curve was used to determine the molar extinction coefficient in the investigated concentration range. Measurements were made in triplicate and the values averaged. A circular sample of each PCL-LDH sample, with a surface of 1 cm diameter, was cut from the film (thickness about 150 micron) and placed into 25 ml of a physiological saline solution at room temperature and, kept in the dark. After specific intervals, the solution was removed and replaced with a fresh one. No hydrolytic degradation of PCL occurred in the time we performed the 87

Chapter VI

release measurements and therefore active molecule release is not due to polymer degradation and erosion but only to diffusive movement into the matrix. VI.3 Characterization of nano-hybrid The intercalation of 3% of the nano-hybrids (LDHBz; LDHDCB; LDHpOHBz) into polycaprolactone by electrospinning process as well as the dispersion of the inorganic lamellar solid into the polymeric matrix were investigated. Scanning Electronic Microscopy (SEM). Figure VI.1a and 1b show images of electrospun nanofibers of respectively PCL with 3% LDH-Bz and PCL with 3% LDH-p-OHBz . The fibers for both of samples are very similar, with a uniform structure and a homogeneous distribution. The average diameter was approximately 250 nanometers for both. Figure VI.1c shows fibers with a greater occurrence of beads and drops. There is a quite distribution of stick fibers and the average diameter is not calculable. To determine the structural organization of the inorganic component of LDHs in the PCL fibers, we determined the chemical structures of the composite fibers with FESEM-EDX. To ascertain if the inorganic hybrid is located inside the fibers we submitted the sample to thermal oxidation up to 450C, followed by EDX observation. In Figure VI.2 we show the EDX micrographs of electrospun fibers PCL+3% LDHBz. Besides the elements that are always present in the micrographs (other peaks that are characteristic of elements like Fe and Cr due to the iron support of the samples and the coating of the samples with Cr), we observe the peaks of Zn and Al, this is a clear indication that the hybrid LDH-Bz is located inside the fibers. In addition we report the qualitative results, too.

88

Nanohybrids for Biomedical Usage

(a)

(b)

(c) Figure VI.1 SEM images of electrospun fibers ( applied voltage 20KV, needle 0.8mm,screen distance 20cm): (a) PCL + 3% LDH-Bz; (b) PCL + 3% LDH-p-OHBz;(c) PCL + 3% LDH-DCB.

89

Chapter VI

Quantitative results

Weight%

Figure VI.2. FESEM equipped with energy dispersive X-ray spectroscopy (EDX) micrographs of composites after thermal oxidation up to 450C of PCL + 3% LDH-Bz.

1.5

1.0

0.5

0.0 Al Zn

X-ray Diffraction measurements (XRD). X-ray analysis was used to study the effect of the intercalation of the three anions into the interlayer region of LDH and to investigate the dispersion degree of the nano-hybrids into the polymer. Figure VI.3 shows, in the 2-30 interval of 2, the X-ray diffractograms of the nano-hybrid LDHBz (a), and composites of PCL with 3% LDHBz, obtained by an intercalation with HEBM(b) (mill.), and PCL with 3% LDHBz, obtained by electrospinning process (c) (spin.). We observe that the peak at 2= 5.69 corresponding to the interlayer distance of the pristine hybrid shifted at 2=4.55 for the milling composite, 90

Nanohybrids for Biomedical Usage indicating an increase of the distance. The peaks are very broad, and so we can deduce that, in this case, the inorganic lamellae are partially intercalated with an expansion of the basal distance corresponding to 1.940nm. Polycaprolactone in both composites is crystalline, showing the peaks of its well developed crystalline structure at 2=21.3 and 2=23.7. In the electrospun fibers (c) the basal peak of the pristine nano-hybrid is almost absent indicating a possible exfoliation of the inorganic lamellae.

Intensity

(c) (b) (a)

2 4 6 8 10 12 14 16 18 20 22 24 26 28 30

Figure VI.3 X-ray diffraction patterns of: (a)LDHBz; (b) PCL + 3% LDHBz (mill.);(c) PCL + 3% LDHBz (spin.) Figure VI.4. shows, in the 2-40 interval of 2, the X-ray diffractograms of the nano-hybrid LDHp-OHBz (a), and composites PCL with 3% LDHpOHBz, obtained by an intercalation with HEBM(b) (mill.), and PCL with 3% LDHp-OHBz, obtained by electrospinning process (c) (spin.). The peaks of the nano-hybrid are unchanged in the composite obtained by milling process (b); their intensity is very high and they are even sharper that in the pristine sample. Therefore in this sample the inorganic lamellae neither exfoliated nor intercalated the polymer. Also in the composite obtained by electrospinning process (c) we have very sharp peaks, indicating an ordered accommodation of the drug anions within the interlayer region. In addition, we can observe the absence of the peak in the basal spacing and this is a strong indication that the lamellae are exfoliated in the PCL matrix.

91

Chapter VI

Intensity

(c)

(b)

(a)
5 10 15

20

25

30

35

40

Figure VI.4 X-ray diffraction patterns of: (a) LDHp-OHBz; (b) PCL with 3% LDHp-OHBz (mill.);(c) PCL with 3% LDHp-OHBz (spin.)

Figure VI.5. shows, in the 2-40 interval of 2, the X-ray diffractograms of the nano-hybrid LDH-DCB (a), and composites PCL with 3% LDH-DCB, obtained by an intercalation with HEBM(b) (mill.), and PCL with 3% LDHDCB, obtained by electrospinning process (c) (spin.). In the last case, we observe that the basal peak of the pristine nano-hybrid is absent both in the composite PCL with 3% LDH-DCB (mill.);and in PCL with 3% LDH-DCB (spin.); so we can therefore safely assume that the inorganic with 2,4-dichlorobenzoate anion mixed by milling and processed by electrospinning is exfoliated into polycaprolactone

92

Nanohybrids for Biomedical Usage

Intensity

(c) (b) (a)

5 10 15 20 25 30 35 40

Figure VI.5 X-ray diffraction patterns of: (a) LDH-DCB; (b) PCL with 3% LDH-DCB (mill.);(c) PCL with 3% LDH-DCB (spin.)

Thermo Gravimetric Analysis (TGA). All the samples were analyzed by thermogravimetric analysis to determine the degradation temperatures and confirm the content of the inorganic component in the polymeric nano-hybrid materials after thermal decomposition of the organic part in air up to 800C. Figure VI.6 shows the TGA curves of sample: (a) PCL obtained by HEBM (mill.) ;(b) LDHBz; (c) composites PCL with 3% LDHBz, obtained by an intercalation with HEBM (mill.); (d) PCL by electrospinning process (spin.);(e) PCL with 3% LDHBz, obtained by electrospinning process (spin.). The thermogravimetric curve of PCL (a) displays one main degradation step with a Td midpoint value of 402C, followed by a small tail at about 450C. Incorporation of the nano-hybrid LDHBz within PCL in both processes anticipates the midpoint of thermal degradation with respect to the corresponding PCL), due to the lower degradation point of the LDHBz. In figures VI.7 and VI.8 we show the curves for the other composites. In each case we observe that the degradation midpoint is anticipated because of the presence of the organic molecule in the nano-hybrid, that drives the polymer degradation.

93

Chapter VI

100 80

(d) (e)

Weight (%)

60 40 20 0 100 200

(b)

(c)
300 400 500 600 700

(a)
800

Temperature (C)

Figure VI.6 Thermal degradation of: a) PCL (mill.); b) LDHBz; c) PCL with 3% LDHBz (mill); d) PCL (spin); e) PCL with 3% LDHBz (spin.).
100 80

(d) (e)

Weight (%)

60 40 20 0 100 200 300 400 500 600 700

(b)

(c)

(a)
800

Temperature (C)

Figure VI.7 Thermal degradation of: a) PCL (mill.); b) LDHp-OHBz; c) PCL with 3% LDHp-OHBz; (mill); d) PCL (spin); e) PCL with 3% LDHpOHBz (spin.).

94

Nanohybrids for Biomedical Usage

100 80

(d) (e)

Weight (%)

60

(b)
40 20

(c)
0 100 200 300 400 500 600 700

(a)
800

Temperature (C)

Figure VI.8 Thermal degradation of: a) PCL (mill.); b) LDHDCBBz; c) PCL with 3% LDHDCBBz; (mill); d) PCL (spin); e) PCL with 3% LDHDCBBz (spin.). VI.4 Antimicrobial Release The release of benzoate from composites of PCL +3% LDHBz of both processes , is presented in Figures VI.9. We observe that the two samples show the same qualitative release behaviour. But the release of the electrospinning composite is slower, probably due to release of the organic molecule penetrates the inorganic lamellae. This is compatible with the microscopy analyses. The typical time-dependent profile of each sample is very fast release in an early period, followed by two slower stages characterized by reduced release. The first stage lasts a few hours and it is completed in an increasing time on increasing the drug loading in the sample. We suggest that the first stage is due to the exchange of the Benzoate ions anchored to the inorganic lamellae on the surface of the sample. This phenomenon is evident because we have in both cases samples of about 130m, in fact the surface is proportionally higher in these thin samples. The releasing period was extended from days to several weeks up about two months.

95

Chapter VI

90 80 70 60 50 40 30 20 10 0 0 200 400 600 800 1000 1200 1400 1600 1800

% release

PCL +3%LDH Bz (mill) PCL +3% LDHBz (spin)

Hours

Figure VI.9.The release of composites PCL +3% LDHBz of both processes at absorbance at =223 nm, is reported as a function of time (hours). In figure VI.10, the sample with LDH-pOHBz dispersed into the polymer PCL using HEBM, with a concentration of 3%, is compared with the same achieved by electrospinning. In this case we have almost the same qualitative release behaviour. The typical time-dependent profile of each sample shows fast release in the early period, followed by a reduced release. The first stage lasts ca. 4 days. In the second stage, drug release increases linearly with time. The same behaviour of both samples, indicate that the different dispersion of the inorganic sample, in terms of distortion of the structure, as shown by X-ray, appears not to have a strong influence on the release time.
60 50

% Release

40 30 20 10 0 0 150 300 450 600 750 900 1050 1200 PCL +3% LDH-pOHBz (mill) PCL +3% LDH-pOHBz (spin)

Hours

Figure VI.10 The release of composites PCL +3% LDH-pOHBz of both processes at absorbance at =243 nm, is reported as a function of time (hours). 96

Nanohybrids for Biomedical Usage The release of 3% LDH-DCB from PCL is presented in figure VI.11. A fast release is observed in the early period, for which completion time is dependent on the hybrid concentration in the polymeric matrix. The second stage, which is much slower, and extends for a long time, also depends on the hybrid concentration. We suggest that the first stage is due to an exchange of organic molecule ions anchored to the inorganic lamellae on the surface of the sample. Therefore, the amount of antimicrobial molecules and completion time for this stage can be varied by varying the composition, the ionic force of the outside solution, and the surface area. In particular in the nanofibers, as we seen in microscopy analyses, we have a structure that is not homogeneous, so this suggests that the organic molecule is dispersed at random in the fiber. In this case, release is strongly dependent on the structure of the fiber, so the first step is very rapid in a burst, as is the second stage. Probably there are much more molecules on the surface that inside the fiber. In figure VI.12 we report the pattern between the three electrospinning composites.

80 70 60

% Release

50 40 30 20 10 0 0 500 1000 1500 2000

PCL +3% LDHDCB (mill) PCL +3%LDHDCB (spin)

Hours

Figure VI.11 The release of composites PCL +3% LDHDCB of both processes at absorbance at =210 nm, is reported as a function of time (hours).

97

Chapter VI

80 70 60

% Release

50 40 30 20 10 0 0 500 1000 1500 2000

PCL + 3% LDHDCB PCL + 3% LDH Bz PCL + 3%LDHp-OHBz

Hours

Figure VI.12 The patterns of the releases of composites by electrospinning process, are reported as a function of time (hours). VI.5 Conclusions We prepared and characterized composites of Polycaprolatone and double layer hydroxides in which we had previously intercalated three molecular anions based on benzoic acid, having different substituents on the aromatic ring. These anions showed antimicrobial properties and their incorporation into the polymeric matrix can constitute a model system for biomedical application, such as periodontal membranes. We used two different processes: 1) Intercalation of the nano-hybrids into polycaprolactone by High Energy Ball Milling (HEBM); 2) Preparation of resorbable antimicrobial membrane by Electrospinning, and later comparison between final samples of the two techniques. X-ray diffraction shows that in all samples obtained by electrospinning the peak of the basal spacing is absent, and this is a strong indication that the lamellae are exfoliated in the PCL matrix. SEM analysis shows that the inorganic hybrid is located inside the fibers (this is also supported by the Xray results). However, the most interesting property for biomedical topical applications is release behaviour. As a matter of fact, in this membrane the release of antimicrobial molecules, anchored on the lamellae of an inorganic compound, is considered very interesting and promising for tuneable drug delivery. Release consists of two stages: an initial very rapid burst and a second stage that is much slower, extending over a long period of time. 98

Nanohybrids for Biomedical Usage We suggest that the release is strongly dependent on the surface of the membrane: in fact in samples with structures that not are homogeneous, the organic molecule results anchored to the inorganic lamellae mostly on the surface rather than inside. Probably the presence of beads and drops cause an interference with the ionic force of the surrounding solution.

99

Chapter VI

100

Conclusions

The research discussed in Chapter 1 of this dissertation relates to fundamental as well as more applied investigations involving electrospinning. These include investigating the effects of particles on fiber formation and developing novel methodologies to fabricate polymeric mats comprising of high specific surface submicron fibers. More specifically, it was found that the presence of nanoparticles, in addition to molecular weight, plays an important role in governing fiber formation during electrospinning of polymer solutions. Chapter 2 of this dissertation explores the efficiency in the dispersion of microcapsules in a polymer matrix showed by electrospinning. By appropriate selection of MPCM, polymer matrix and processing conditions, very high concentrations of microcapsules can be reached without particular detriment in the resulting polymer fibers. Differently from other techniques, electrospinning works at room temperature and in normal air conditions, which may protect from harsh treatments and enhance thermal efficiency by avoiding capsule damage. The electrospinning process is capable of producing extremely thin fibres with diameters ranging from microns down to a few nanometers (typically hundreds of nanometers in diameter). It is a very versatile technique and allows the use of a variety of polymers, blends of different polymers as well as integration of additives, allowing tailoring to different application requirements. In Chapter 3 of this dissertation the use of electrospinning to produce polymer/CNT nanocomposite fibers was proposed in order to simultaneously align CNTs along a single axis without compromising the structural integrity of the individual CNTs. In contrast to agglomeration of embedded CNTs in the common melt processing, aligned CNTs within the nano fiber increases the effective contact area for interfacial bonding with the polymer matrix and enhances the effective aspect ratio for reinforcement. For these reasons and because of the low cost and relatively uncomplicated process involved, electrospinning is an attractive technique for embedding CNTs within micron and nanometer sized polymeric fibers.

Conclusions

Chapter 4 of this dissertation relates to the fabrication of montmorillonite and organic clay chemically modified reinforced PCL nanofibers by electrospinning. In addition a comparison between samples obtained by solvent casting and samples obtained by electrospinning was carried out. The morphological and X-rays analyses showed significant difference between the two techniques. The sample with 7% Cloisite 30 B by electrospinning showed the best interaction with the polymer matrix, probably due to the structure of the clay. In fact Cloisite 30B materials possess two hydroxyl groups in the organic modifier, which make it possible to interact strongly with carboxyl groups in PCL molecules, promoting the intercalation of PCL molecules into the clay galleries. However, a partially exfoliated structure was obtained, producing fibers with an average diameter in the range of 50050 nm. On the other hand, samples obtained by solvent casting showed an intercalated structure. Samples with 7% Cloisite 10A by electrospinning showed fibers with the inorganic component residing outside the fibers and not exfoliated, probably due to moderate compatibility and hydrophobicity due to the benzene ring structure. In this case we also had unsatisfactory dispersion of the inorganic component inside the polymer matrix that caused disappointing electrospinnability producing fibers with beads and drops. Finally, composites of Polycaprolatone and double layered hydroxides in which there was intercalation of three molecular anions based on benzoic acid, with different substituents on the aromatic ring were investigated in Chapter 6. The intercalated anions showed antimicrobial properties and their incorporation into the polymeric matrix could constitute a model system for biomedical applications, such as periodontal membranes. Two different processes were followed: 1) Intercalation of the nanohybrids into polycaprolactone by High Energy Ball Milling (HEBM) and 2) Preparation of resorbable antimicrobial membrane by Electrospinning, followed later by comparison between final samples from the two techniques. The X-ray diffraction shows that in all samples obtained by electrospinning the peak of the basal spacing of the clay is absent, and this is a strong indication that the lamellae are exfoliated in the PCL matrix. SEM analysis shows that the inorganic hybrid is located inside the fibers (also supported by X-ray results). However, the most interesting property for biomedical topical applications is release behavior. As a matter of fact, in this membrane the release of antimicrobial molecules, anchored on the lamellae of an inorganic compound, is considered to be very interesting and promising for tunable drug delivery. Release consists of two stages: an initial very rapid burst and a second stage that is much slower, extending over a long period of time. The release is strongly dependent on the surface of the membrane: in fact in samples with a structure that is not homogeneous, the organic molecule is anchored to the inorganic lamellae mostly on the surface rather than inside. 102

Conclusions Probably the presence of beads and drops causes an interference with the ionic force of the surrounding solution. Success in the development of composite nanofibers by electrospinning creates a new pathway to connect nanoscale effect to macro structure performance. This will lead to the creation of a new family of advanced products of nanoparticle reinforced functional composite nanofibers for various applications such as sensors, biological labelling, tissue engineering and multi-functional textiles.

103

Conclusions

104

References

Ajayan, P.M. Nanotubes from carbon. (1999).Chem. Rev., 99, 1787. Andrews, R. Jacques, D. Minot, M. Rantell, T. Fabrication of carbon multiwall nanotube/polymer composites by shear mixing. (2002). Macromol. Mater. Eng., 287, 395. Bachilo, S.M. Strano, M.S. Kittrell, C. Hauge, R.H. Smalley, R.E. Weisman, R.B. Structure-assigned optical spectra of single-walled carbon nanotubes. (2002). Science, 298, 2361. Bailey, A.G. (1998) Electrostatic Spraying of Liquids, Wiley, New York. Baughman, R.H. Zhakidov, A.A. De Heer, W.A. Carbon nanotubes the route toward applications. (2002) Science, 297, 787. Baumgarten, P. K. (1971)J. Colloid Interface Sci, 36, 71-79. Bergshoef, M. M. Vancso, G. (1999) J. Advanced Materials, 11, 1362-1370. Biercuk, M.J. Llaguno, M.C. Radosavljevic, M. Hyun, J.K. Johnson, A.T. Carbon nanotube composites for thermal management. (2002) Appl. Phys. Lett., 80, 2767. Bornat, A. (1982) US, 4,323,525. Bower, C. Rosen, R. Jin, L. Han, J. Zhou, O. Deformation of carbon nanotubes in nanotube-polymer composites. (1999). Appl. Phys. Lett., 74, 3317. Bradley, K. Gabriel, J.-C.P.Flexible nanotube electronics. (2003).Nano Lett., 3, 1353. Breuer, O. Sundararaj, U. Big returns from small fibers: a review of polymer/carbon nanotube composites. (2004). Polym. Comp., 25, 630. Buer, A. Ugbolue, S. C. Warner, S. B. (2001)Textile Research Journal, 71, 323-328. Burger, C. Hsiao, B. S. Chu, B. (2006), Annu. Rev. Mater. Res., 36, 333 68.

References

Carrado, A. (2000), Appl Clay Sci ,17,1-5. Carrado, K.A. Shonaike, G.O. Advani, S.G. (2003), Advanced Polymeric Materials, CRC Press, Chapter 10. Casper, C. L. Stephens, J. S. Chase, D. B. Rabolt, J. F. (2003) Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), 44, 81. Cavani, F. Trifiro`, F. Vaccari, A. (1991), Catal. Today, 11, 173-301. Charunyakorn, P. Sengupta, S. S.K. Roy, (1991) International Journal of Heat and Mass Transfer, 34,819. Choi, S.W. Jo, S.M. Lee, W.S. (2003), Adv.Mater. 15,2027. Colby, R. H. Fetters, L. J. Funk, W. G. Graessley, W. W. (1991) Macromolecules, 24, 3873-3882. Colby, R. H.Rubinstein, M. Viovy, J. L. (1992) Macromolecules, 25, 996998. Colvin, D. (1989) EP patent 0306202A2, (1997) US patent 5637389, (2002) US patent 6418832. Colvin, D. (1996) US patent 5499460. Colvin, D. P. Bryant. Ye G. (1998) Protective clothing containing encapsulated phase change materials. Advances in Heat and Mass Transfer in Biotechnology, Vol. 40: 123-132. Costantino, U. Marmottini, F. Nocchetti, M. .(1998), Eur. J. Inorg. Chem,1439-1446. Czaplewski, D. A. Kameoka, J. Mathers, R. Coates, G. W. Craighead, H. G. (2003) Applied Physics Letters, 83, 4836-4838. Deitzel, J. M. Kleinmeyer, J. Harris, D. Beck Tan, N. C. (2000) Polymer, 42, 261-272. Deitzel, J. M. Kosik, W. McKnight, S. H. Tan, N. C. B. DeSimone, J. M. Crette, S. (2002) Polymer, 43, 1025-1029. Dersch, R. Steinhart, M. Boudriot, U. Greiner, A. Wendorff, J.H. Nanoprocessing of polymers: applications in medicine, sensors, catalysis, photonics. (2005).Polym. Adv. Technol., 16, 276. Doshi, J. Reneker, D. H. (1995) Journal of Electrostatics, 35, 151-160. Dresselhaus, M.S. Dresselhaus, G. Ecklund, P.C. (1996) Science of Fullerenes and Carbon Nanotubes, Academic, New York.

106

References Dror, Y. Salalha, W. Khalfin, R.L. Cohen, Y. Yarin, A.L. Zussman, E. Carbon nanotubes embedded in oriented polymer nanofibers by electrospinning. (2003).Langmuir, 19, 7012. electrospun nanofibers. Colloid Surf. A: (2001). Phys. Eng. Aspects, 469,187188. Fang, D. Hsiao, B.S Chu, (Am.Chem.Soc.,Div.Polym.Chem.),44,59. B. (2003) Polym.Prep.

Fang, X.; Reneker, D. H. (1997), Journal of Macromolecular Science, Physics, B36, 169-173. Fenn, J.B. Mann, M. (1989), Science, 246, 64. Fine, J. Passaic, N. J. Sigismondo, A (1980). T.US, 4,223. Fong, H. Chun, I. Reneker, D. H. (1999) Polymer , 40, 4585-4592. Fong, H. Liu, W. Wang, C. S. (2002), Polymer, 43,775-780. Fong, H. Liu, W. Wang, C.S. Vaia, R. A. (2001) Polymer, 43, 775-780. Formhals, A. (1934) US patent 1,975,504. Frenot, A. Chronakis, I. S. (2003), Curr. Opin. Colloid Interface Sci. 8, 6475. Fridrikh, S.V. Yu, J.H. (2003),Phys.Rev.Lett, 90,144-502. Gander, B. Meinel, L. Walter, E. Merkle, H.P. Polymers as platform for drug delivery: Reviewing our current portfolio on PLGA microspere. (2001). Chimia 55.212. Gibson, P. Schreuder-Gibson, H. (2000) American Society of Mechanical Engineers, 91, 45-61. Gibson, P. Schreuder-Gibson, H. Pentheny, C. (1998) Journal of Coated Fabrics, 28, 63-72. Gibson, P. Schreuder-Gibson, H. Rivin, D. (2001),Colloids Surf.A, 469,187188. Gibson, P.W. Schreuder-Gibson, H.L. Pentheny, C. Electrospinning technology: direct application of tailorable ultrathin membranes. J. (1998) Coated Fabrics, 28, 63. Gorrasi, G. Sarno, M. Di Bartolomeo, A. Sannino, D. Ciambelli, P. Vittoria, V.(2007), J. Polym. Sci., Part B: Polym. Phys, 45, 597-606. Guignard, C. (1980)US, 4,230,650. Guignard, C. (1981)US, 4,287,139.

107

References

Guo, T. Nikolaev, P. Rinzler, A.G. Toma nek, D. Colbert, D.T. Smalley. R.E. Self-assembly of tubular fullerenes. (1995). J. Phys. Chem., 99, 10694. Haggenmueller, R. Gommans, H.H. Rinzler, A.G. Fischer, J.E. Winey, K.E. Aligned single-wall carbon nanotubes in composites by melt processing methods. (2000). Chem. Phys. Lett., 330, 219. Hobbie, E.K. Wang, H. Kim, H. Lin-Gibson, S. Grulke, E.A. Orientation of carbon nanotubes in a sheared polymer melt. (2003). Phys. Fluids, 15, 1196. Hone, J. Laguno, M.C. Nemes, N.M. Johnson, A.T. Electrical and thermal transport properties of magnetically aligned single wall carbon nanotube films. (2000) Appl. Phys. Lett., 77, 666. Hong, K. Park, S. (1999) Materials Research bulletin,34, 963-969. Huang, L. McMillan, R. A. Apkarian, R. P. Pourdeyhimi, B.; Conticello, V. P.; Chaikof, E. L. (2000) Macromolecules, 33, 2989-2997. Huang, Z.-M. Zhang, Y.-Z. Kotaki, M. Ramakrishna, S. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. (2003).Comp. Sci. Tech., 63, 2223. Iijima, S. Helical microtubules of graphitic carbon. (1991), Nature, 56, 354. Iijima, S. Brabec, C. Maiti, A. Bernholc, J. Structural flexibility of carbon nanotubes. (1996) J. Chem. Phys., 104, 2089. Isakoff, L. (1971) US, 3, 593,074. Jaeger, R. Schonherr, H. Vansco, G. J. (1996) Macromolecules, 29, 76347636. Jia, H. Zhu, G. Vugrinovich, B. Kataphinan, W. Reneker, D. H. Wang, P. (2002) Biotechnology Progress, 18, 1027-1032. Jia, Z. Wang, Z. Xu, C. Liang, J. Wei, B. Wu, D. Zhu, S. Study on poly(methyl methacrylate)/carbon nanotube composites. (1999).Mater. Sci. Eng. A, 271, 395. Jin, L. Bower, C. Zhou, O. Alignment of carbon nanotubes in a polymer matrix by mechanical Joshi, P.P. Merchant, S.A. Wang, Y. Schmidtke, D.W. Amperometric biosensors based on redox polymercarbon nanotubeenzyme composites. (2005). Anal. Chem., 77, 3183. Kameoka, J. Orth, R. Yang, Y. Czaplewski, D. Mathers, R. Coates, G. W. Craighead, H. G.(2003) Nanotechnology, 14, 1124-1129. Kameoka, J.Craighead, H.G, (2003), Appl.Phys.Lett, 83,371.

108

References Kattamuri, N. Shawon, J. Sung, C. (2003) Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), 44, 763-764. Kearns, J.C. Shambaugh, R.L. Polypropylene fibers reinforced with carbon nanotubes. (2002). Kenawy, E.R. Abdel-Fattah, Y. R. (2002) Macromolecular Bioscience, 2, 261-266. Kenawy, E.R. Bowlin, G. L. Mansfield, K. Layman, J. Sanders, E. Simpson, D. G. Wnek, G. E. (2002) Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), 43, 457-458. Kessick, R. Tepper, G. (2003), Appl.Phys.Lett, 83,557. Khil, M. Cha, D. Kim, H. (2003),J.Biomed.Mater.Res.Part B, 67B,675-680. Khil, M. S. Kim, H. Y. Kim, M. S. Park, S. Y. Lee, D.R. (2004) Polymer, 45, 295-301. Kim, B. S. Baez, C. E. Atala, A. (2000), World J. Urol. 18, 2-9. Kim, G.-M. Michler, G.H. Deformation processes of ultrahigh porous multiwalled carbon nanotubes/polycarbonate composite fibers prepared by electrospinning. (2005).Polymer, 46, 7346. Kim, J.S. Reneker, D. H. (1999) Polymer Engineering and Science, 39, 849854. Kim, Y.A. Hayashi, T. Fukai, Y. Endo, M. Yanagisawa, T. Dresselhaus, M.S. Effect of ball milling on morphology of cup-stacked carbon nanotubes. (2002) Chem. Phys. Lett., 355, 279. Kumar, S. Doshi, H. Srinivasarao, M. Park, J.O. Schiraldi, D.A. Fibers from polypropylene/nano carbon fiber composites. (2002) Polymer, 43, 1701. Kymakis, E. Amaratunga, G.A.J. Optical properties of polymer-nanotube composites. (2004).Synth. Met., 142, 161. Lagaly, G. (1999).Appl Clay Sci 15,1-6. Lam, H. Titchenal, N. Naguib, N. Ye, H. Gogotsi, Y. Ko, F. (2004) Materials Research Society Symposium Proceedings, 791, 353-358. Langer, R. New methods of drug delivery. (1990),Science, 249.1527. Larrondo, L. Manley, R. S. (1981) J. Journal of Polymer Science: Polymer Physics, 19, 909-940. Le Baron, PC. Wang, Z. Pinnavaia, TJ. (1999), Appl Clay Sci,15,11-17. Lee, S.W. Belcher, A. M. (2004) Nano Letters, 4, 387-390. 109

References

Lee, Y. H. An, I. G. Kim, C. Lee, D. S. (2005), Biomaterials, 26,31653172. Lefrant, S. Buisson, J.-P. Chauvet, O. Benoit, J.-M. Baibarac, M. Baltog, I. Bernier,P.Interactions in carbon nanotubes and polymer/nanotubes composites as evidenced by Raman spectroscopy. (2001).Mater. Res. Soc. Proc., Vol. 746, Paper Z 7.2. Lennox-Kerr, P. (1998) Techn.Text.Inter. ,7, 25. Li, F. Cheng, H.M. Bai, S. Su, G. Dresselhaus, M.S. Tensile strength of single-walled carbon nanotubes directly measured from their macroscopic ropes. (2000).Appl. Phys. Lett., 77, 3161. Li, W.J. Laurencin Cato, T. Caterson Edward, J. Tuan Rocky, S. K. Frank, K. (2002) Journal of biomedical materials research, 60, 613-621. Li, X. Kang, T.K Cho, W.J. Lee, J.K. Ha, C.S. (2001), Macromol. Rapid Commun. 22, 1306-1312. Lin, D.Y. Martin, D.C. Div.Polym.Chem.) 44,70-76. (2003), Polym.Prep. (Am.Chem.Soc.,

Liu, H. Hsieh, Y.L. (2002) Journal of Polymer Science, Part B: Polymer Physics, 40, 2119-2129. Liu, H. Kameoka, J. Czaplewski, D.A.(2004) Nano Lett,4,671. Liu, L. Qi, Z.Zhu, X. (1999), J Appl Polym Sci 71,1133. Lourie, O. Cox, D.M. Wagner, H.D. Buckling and collapse of embedded carbon nanotubes. (1998) Phys. Rev. Lett., 81, 1638. Lu, Y. Chen, S.C. Micro and nano-fabrication of biodegradable polymers for drug delivery. (2004). Advanced Drug Delivery review 56,1621. Lyons, J.Pastore, C. Ko, F. (2003) Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), 44, 122-123. Ma, X. P. (2004), Mater. Today, 7, 30-40. Martin, G. H. Derek, I. (1977)US, 4,043,331. Matthews, J. A. Wnek, G. E. Simpson, D. G. Bowlin, G. L. (2002) Biomacromolecules, 3, 232-238. McKee, M. G. Wilkes, G. L.Colby, R. H. Long, T. E. (2004) Macromolecules, 37, 1760-1767. Mo, X. M. Xu, C. Y. Kotaki, M. Ramakrishna, S. (2004) Biomaterials, 25, 1883-1890.

110

References Mo, X. Xu, C. Kotak, M. Ramakrishna, S. Weber, H. (2003) J. Polymer Preprints (American Chemical Society, Division of Polymer Chemistry), 44, 128-129. Nagapudi, K. Brinkman, W. T. Leisen, J. E. Huang, L. McMillan, R. A. Apkarian, R. P. Conticello, V. P. Chaikof, E. L. (2002) Macromolecules, 35, 1730-1737. Nelson, G. (2002) Int. J. Pharmac., 242,55-62. Norris, I.D. Shaker, M.M. Ko, F.K. MacDiarmid, A.G. Electrostatic fabrication of ultrafine conducting fibers: polyaniline/polyethylene oxide blends. (2000).Synth. Met., 114, 109. OConnell, M.J. Boul, P. Ericson, L.M.Huffman, C.Wang, Y. Haroz, E. Kuper, C.Reversible water-solubilization of single-walled nanotubes by polymer wrapping. (2001).Chem. Phys. Lett., 342, 265. Pan, Z.W. Xie, S.S. Lu, L. Chang, B.H. Sun, L.F. Zhou, W.Y. Wang, G. Zhang, D.L. Tensile tests of ropes of very long aligned multiwall carbon nanotubes. (1999) Appl. Phys. Lett., 74, 3152. Park, C. Ounaies, Z. Watson, K.A. Pawlowski, K. Lowther, S.E. Connell, J.W. Siochi, E. (2002).Polymer-single wall carbon nanotube composites for potential spacecraft applications. NASA Center for AeroSpace Information ICASE Report No. 2002-36, NASA/CR-2002-211940. Pedicini, A. Farris, R.J. (2003) Polymer, ,44,6857-6866. Ray, S.S. Okamoto, M. ,(2003), Progress in Polymer Science,28, 1539-1641. Ren, Z.F. Huang, Z.P. Xu, J.W. Wang, J.H. Bush, P. Siegal, M.P. Provencio, P.N. Synthesis of large arrays of well-aligned carbon nanotubes on glass. (1998). Science, 282, 1105. Reneker, D. H. Yarin, A. L. Fong, H. Koombhongse, S. (2000) J. Appl. Phys., 90, 4836-4846. Rives, V. Ed. Layered Double Hydroxides, Present and Future, (2001).Nova Science: New York. Safadi, B. Andrews, R. Grulke, E.A. Multiwalled carbon nanotube polymer composites: synthesis and characterization of thin films. (2002). J. Appl. Polym. Sci., 84, 2660. Sam, T. Optimising the therapeutic trinity of active ingredient, delivery system and functional packaging, (2003)J. of Controlled Release, 87,153. Sanders, E. H. Kloefkorn, R. Bowlin, G. L. Simpson, D. G.Wnek, G. E. (2003) Macromolecules, 36, 3803-3805. Scardino, F. L. Balonis, R. J. (2000)US, 6,106,913. 111

References

Schlittler, R.R. Seo, J.W. Gimzewski, J.K. Durkan, C. Saifullah, M.S.M. Welland, M.E. Single crystals of single-walled carbon nanotubes formed by self-assembly. (2001). Science, 292, 1136. Seoul, C. Kim, Y.T. Baek, C.K. (2003) Journal of Polymer Science, Part B: Polymer Physics, 41, 1572-1577. Shim, H. Mc Culloughand, E. A. Jones, B. W. (2001), Text.Res.J., 7, 1495. Shin, Y. Yoo, D. Son, K. (2005), Journal of Applied Polymer Science, 97, 910915. Siha,V.R. Bansal, K. Kanshik, R. Kumria, R. Trehan, A.Poly -caprolactone microspheres and nanospheres: an overview, (2004). Int.J.Pharm., 278,1. Simm, W. L.; Claus, G. O.; Bonart, R. B.; Bela, V. F. G. (1978) US, 4,069,026. Simons, H. L. (1966) US, 3,280,229. Sinha,V.R. Trehan,A, Biodegradable microspheres for protein delivery. (2003).J. Control Release 90.261. Sinnott, S.B. Andrews, R. Carbon nanotubes: synthesis, properties and applications. Crit. (2001) Rev. Solid State Mat. Sci., 26, 145. Smith, B.W. Benes, Z. Luzzi, D.E. Fischer, J.E. Walters, D.A. Casavant, M.J. Schmidt, J. Smalley, R.E. Structural anisotropy of magnetically aligned single wall carbon nanotube films. (2000).Appl. Phys. Lett., 77, 663. Smith, J.G. Connell, J.W. Delozier, D.M. Lillehei, P.T. Watson, K.A. Lin, Y.Zhou, B.Space durable polymer/carbon nanotube films for electrostatic charge mitigation. (2004). Polymer, 45, 825. Sorrentino, A. Gorrasi, G. Tortora, M. Vittoria, V. Costantino,U. Marmottini, F. Padella, F. (2005), Polymer, 46, 1601-1608. Steinhart, M. Senz, S. Wehrspohn, R.B.Curvature-directed crystallization of poly(vinylidene difluoride) in nanotube walls. (2003). Macromolecules, 36, 3646. stretching. (1998).Appl. Phys. Lett., 73, 1997. Suryanarayana, C. (2001), Prog. Mater. Sci. 46 (1-2), 1-184. Tahhan, M. Truong, V.-T. Spinks, G.M. Wallace, G.G. Carbon nanotube and polyaniline composite actuators. (2003). Smart Mater. Struc., 12, 626. Tang, B.Z. Xu, H. Preparation, alignment, and optical properties of soluble poly(phenylacetylene)-wrapped carbon nanotubes. (1999). Macromolecules, 32, 2569.

112

References Tao, X. M. (2001) Smart Fibers, Fabrics and Clothing (Woodhead Publishing Ltd., Cambridge, England) p.31. Thostenson, E.T. Li, W.Z. Wang, D.Z. Ren, Z.F. Chou, T.W. Carbon nanotube/carbon fiber hybrid multiscale composites. (2002). J. Appl. Phys., 91, 6034. Tibbetts, G.G. Carbon fibers produced by pyrolysis of natural gas in stainless steel tubes. (1983). Appl. Phys. Lett., 42, 666. Tsai, P.P. Schreuder-Gibson, H. Gibson, P. Different electrostatic methods for making electret filters. J. (2002).Electrostatics, 54, 333. Tucknott, R.Yaliraki, S.N. Aggregation properties of carbon nanotubes at interfaces. (2002) Chem. Phys., 281, 455. Velasco-Santos, C. Mart nez-Herna ndez, A.L. Lozada-Cassou, M. Alvarez-Castillo, A. Chemical functionalization of carbon nanotubes through an organosilane. (2002).Nanotechnology, 13, 495. Vigolo, B. Pe nicaud, A. Coulon, C. Sauder, C. Pailler, R. Journet, C. Bernier, P. Poulin, P. Macroscopic fibers and ribbons of oriented carbon nanotubes. (2000) Science, 290, 1331. Wang, M. Hsieh, A. J. Rutledge, G. C. (2005), Polymer, 46, 3407-3418. Wang, X. Drew, C..Lee, S.H (2002), Nano Lett, 2,1273. Wang, Y. Santiago-Aviles, J. J. (2003) Journal of Applied Physics, 94, 17211727. Whitesides, G.M. Laibinis, P.E. (1990), Langmuir ,6,87. Wnek, G. Carr, M.E. Simpson, D.G. Bowlin, G.L. Electrospinning of nanofiber fibrinogen structures. (2003). Nano Lett., 3, 213. Xu, C. Y. Inai, R. Kotaki, M. Ramakrishna, S. (2004) Biomaterials, 25, 877886. Yamamoto, K. Akita, S. Nakayama, Y. Orientation and purification of carbon nanotubes using ac electrophoresis. (1998).J. Phys. D: Appl. Phys., 31, L34. Yarin, A.L. Koombhongse, J.Appl.Phys.Fluidics,90,4836. S..Reneker, D.H (2001),

Yeo, L.Y. Gagnon, Z. Chang, H.-C. (2005).AC Electrospray biomaterials synthesis. Biomaterials, 26, 6122. Yoshimoto, H. Shin, Y. M. Terai, H. Vacanti, J. P. (2003) Biomaterials, 24, 2077-2082.

113

References

Zarkoob, S. Reneker, D. H. Ertley, D. Eby, R. K. Hudson, S. D. (2000) US, 6,110, 590. Zhang, X. X. Tao, X. Yick, M. K. L. Wang, X. C. (2004), Colloid Polym. Sci., 88, 330. Zhao, S.Wu, X. Wang, L. Huang, Y.(2004) Journal of Applied Polymer Science, 91, 242-246. Zong, X.Kim, K. Fang, D. Ran, S. Hsiao, B. S. Chu, B. (2002) Polymer, 43, 4403-4412.

114