Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Experimental dental composites with electrospun nanofibers and nanofibrous composites. Urbanetto Peres, Bernardo 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2016_september_urbanettoperes_bernardo.pdf [ 4.06MB ]
Metadata
JSON: 24-1.0300457.json
JSON-LD: 24-1.0300457-ld.json
RDF/XML (Pretty): 24-1.0300457-rdf.xml
RDF/JSON: 24-1.0300457-rdf.json
Turtle: 24-1.0300457-turtle.txt
N-Triples: 24-1.0300457-rdf-ntriples.txt
Original Record: 24-1.0300457-source.json
Full Text
24-1.0300457-fulltext.txt
Citation
24-1.0300457.ris

Full Text

 EXPERIMENTAL DENTAL COMPOSITES WITH ELECTROSPUN NANOFIBERS AND NANOFIBROUS COMPOSITES. by  Bernardo Urbanetto Peres  DDS, The University of British Columbia, 2016  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies  (CRANIOFACIAL SCIENCES)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)   May 2016  © Bernardo Urbanetto Peres, 2016 ii  Abstract  Electrospun nanofibers with and without nanoparticles are poorly explored in dental research. Nanocrystalline cellulose is a nanoparticle with distinguished properties that has already been associated with nanofibers, but yet not applied to any dental aspect. The objective of this work was to investigate the use of polyacrylonitrile (PAN) nanofibers containing nanocrystalline cellulose (NCC) in the light of the mechanical behavior of fibrous mat and experimental dental composites. Three experiments were performed to answer the following research questions: 1) Can nanocrystalline cellulose improve mechanical properties of polyacrylonitrile nanofiber meshes? 2) Does the method of dispersion (simple mixture vs. with solvent exchange) of NCC in PAN solution affect fiber formation and the respective properties of the meshes? and 3) Can NCC-containing PAN electrospun nanofibers affect flexural properties of experimental dental composites? Results showed that nanocrystalline cellulose, at low concentrations, significantly increases PAN nanofibers tensile properties (chapter 2). Dispersion methods affected both the morphology and mechanical properties of the fibers (chapter 3). Finally, when NCC-containing PAN nanofibers were used to produce experimental dental composites, there was a significant improvement in flexural strength and work of fracture (chapter 4).  In conclusion, the findings indicated that the use of electrospun nanofibers and nanofibres containing nanoparticles is a promising approach to reinforce dental composites.     iii  Preface  This dissertation is a continuation of the previous collaboration between Dr. Frank Ko and Dr. Ricardo Carvalho. It is an original work, unpublished and an independent work by the author, Bernardo Urbanetto Peres. iv  Table of Contents  Abstract .......................................................................................................................................... ii	Table of Contents ......................................................................................................................... iv	List of Tables .............................................................................................................................. viii	List of Figures ............................................................................................................................... ix	List of Symbols ............................................................................................................................ xii	List of Equations ........................................................................................................................ xiii	List of Abbreviations ................................................................................................................. xiv	Acknowledgements .................................................................................................................... xvi	Dedication ................................................................................................................................. xviii	Chapter 1: Introduction ................................................................................................................1	1.1	 Nanofiber technology ...................................................................................................... 2	1.1.1	 Electrospinning ....................................................................................................... 2	1.1.1.1	 Parameters affecting fiber morphology .............................................................. 4	1.1.1.1.1	 Viscosity, molecular weight and temperature of spinning dope ................... 4	1.1.1.1.2	 Applied voltage, feeding rate, orifice size, dielectric constant and electrical conductivity ..................................................................................................................... 5	1.1.1.1.3	 Surface tension, humidity and solvent boiling point .................................... 6	1.1.1.1.4	 Distance between syringe tip and collector plate .......................................... 7	1.2	 Multifunctional nanocomposites ..................................................................................... 7	1.3	 Nanocrystalline cellulose (NCC) .................................................................................... 8	1.4	 Nanofibers and nanofibrous composites as reinforcing agents for dental composites . 11	v  1.5	 Closing remarks ............................................................................................................ 12	Chapter 2: Nanocrystalline cellulose as a reinforcing agent for electrospun PAN nanofibers14	2.1	 Introduction ................................................................................................................... 14	2.2	 Materials and methods .................................................................................................. 15	2.2.1	 Production of the nanofibers and incorporation of NCC. ..................................... 15	2.2.1.1	 Electrospinning of PAN nanofibers .................................................................. 15	2.2.1.2	 Incorporation of NCC ....................................................................................... 16	2.2.2	 Characterization of nanofibers and nanofiber mats .............................................. 17	2.2.2.1	 Tensile properties of nanofibers mats ............................................................... 17	2.2.2.2	 Morphological characterization of nanofibers mats and NCC ......................... 18	2.2.2.3	 Crystallinity of NCC ......................................................................................... 19	2.2.2.4	 Fourier transform infrared spectroscopy ........................................................... 20	2.2.3	 Statistical analysis ................................................................................................. 20	2.3	 Results ........................................................................................................................... 20	2.3.1	 Tensile properties of nanofibers mats ................................................................... 20	2.3.2	 Morphological characterization of nanofibers mats and NCC ............................. 22	2.3.3	 FTIR ...................................................................................................................... 24	2.3.4	 X-ray diffraction ................................................................................................... 25	2.4	 Discussion ..................................................................................................................... 25	2.5	 Conclusions ................................................................................................................... 29	Chapter 3: Effect of dispersion mode on physical properties of NCC-reinforced PAN nanofibers .....................................................................................................................................31	3.1	 Introduction ................................................................................................................... 31	vi  3.2	 Materials and methods .................................................................................................. 32	3.2.1	 Electrospinning and NCC addition by different methods ..................................... 32	3.2.1.1	 Production of nanofibers ................................................................................... 32	3.2.1.2	 Addition of NCC by mixing and solvent exchange .......................................... 33	3.2.1.3	 Viscosity measurements and mechanical testing .............................................. 35	3.2.1.4	 Confirmation of stability ................................................................................... 35	3.2.1.5	 Scanning electron microscopy .......................................................................... 35	3.2.1.6	 Other characterization methods ........................................................................ 36	3.2.2	 Statistical analysis ................................................................................................. 36	3.3	 Results ........................................................................................................................... 36	3.3.1	 Tensile properties .................................................................................................. 36	3.3.2	 Viscosity ............................................................................................................... 37	3.3.3	 Confirmation of stability ....................................................................................... 38	3.3.4	 Scanning electron microscopy .............................................................................. 38	3.3.5	 Other characterization methods ............................................................................ 40	3.4	 Discussion ..................................................................................................................... 40	3.5	 Conclusions ................................................................................................................... 45	Chapter 4: Effects of NCC-containing PAN-electrospun nanofibers in flexural properties of experimental dental composites. .................................................................................................47	4.1	 Introduction ................................................................................................................... 47	4.2	 Materials and methods .................................................................................................. 48	4.2.1	 Electrospinning NCC-containing nanofibers and characterization ....................... 48	vii  4.2.2	 Creation and characterization of nanofiber reinforced experimental dental composites ............................................................................................................................. 48	4.2.2.1	 Fabrication of BisGMA/TEGDMA monomer blends ...................................... 48	4.2.2.2	 Production of nanofiber reinforced composite ................................................. 49	4.2.2.3	 Flexural properties testing ................................................................................. 50	4.2.2.4	 Fractographic analysis ...................................................................................... 51	4.2.2.5	 Areal density of fibers in flexural beams .......................................................... 52	4.2.3	 Statistical analysis ................................................................................................. 52	4.3	 Results ........................................................................................................................... 52	4.3.1	 Characterization of electrospun nanofiber mats ................................................... 52	4.3.2	 Flexural properties of PAN with and without NCC reinforced composites ......... 53	4.3.3	 Fractographic analysis .......................................................................................... 53	4.3.4	 Areal density of fibers on flexural beams and mass ratio for fiber in composite blocks...... .............................................................................................................................. 56	4.4	 Discussion ..................................................................................................................... 57	4.5	 Conclusions ................................................................................................................... 60	Chapter 5: Conclusion .................................................................................................................61	Bibliography .................................................................................................................................63	 viii  List of Tables  Table 2.1 Formulations of experimental solutions. ...................................................................... 16	Table 2.2 Effects of NCC on PAN nanofiber tensile properties ................................................... 21	Table 2.3   Average diameter of the nanofibers ............................................................................ 23	Table 3.1 PAN polymer solutions (without NCC) and suspensions of PAN with NCC dispersed by simple mixing and by solvent exchange. ................................................................................. 34	Table 3.2 Effects of dispersion method of NCC on tensile properties of nanofiber mats ............ 37	Table 3.3 Viscosity of solutions tested. Capital letters indicate comparisons among solutions and lower case among batches. ............................................................................................................ 37	Table 3.4 Fiber diameter according to groups. ............................................................................. 40	Table 4.1 Experimental dental resin and viscosity ....................................................................... 49	Table 4.2 Effects of the presence of NCC in PAN nanofiber reinforced dental composites. ....... 53	Table 4.3 Measurements of areal density of fiber on flexural beams. .......................................... 57	 ix  List of Figures  Figure 1.1 Electrospinning schematic representation. .................................................................... 3	Figure 2.1 Representative stress/strain plot. Elastic Modulus (E) was obtained from the steepest part of the low strain section of the plot by dividing the variation in stress by variation in strain. Yield Strength (YS) was determined by constructing a line parallel to the linear portion of the graph and off set at 0.3% from the origin in the horizontal axis. Ultimate Tensile Strength (UTS) was the set as the highest value on the vertical axis of the graph. Elongation at Maximum Stress (EMS) was obtained in percentage by multiplying by 100 the strain value (horizontal axis) at the point of UTS. ................................................................................................................................ 18	Figure 2.2 Representative stress/strain plots for 0, 1, 2, and 3% NCC. ........................................ 21	Figure 2.3 Representative SEM images of the nanofiber mats. A: Group 0% NCC with randomly distributed  fibers with regular diameter and a slightly rough surface; B: Group 1% NCC presented randomly distributed fibers with presence of well defined beads that looked rougher on the surface; C:  Group 2% NCC  where the merging of two fibers can be observed. Beads were also observed. Fibers appeared smoother in this group.; D: Group 3% NCC showed randomly oriented fibers, with presence of beads that appeared rough on the surface. ................................ 22	Figure 2.4 TEM of NCC particles. NCC showed a rod-like appearance with a length close to 200 nm and a diameter around 12 nm. ................................................................................................. 23	Figure 2.5 FTIR Readings. A: Green reading represents pure PAN polymer spectrum as provided by the manufacturer; Red reading shows a PAN fiber with 3% NCC where differences can be seen. A slight peak can be observed in the 1600-1700 range and a significant peak at the range of 2200-2400; Yellow spectra establishes a similar peak, but with higher intensity, in the 1600-1700 x  range as the red reading, and do not show a negative peak in the 2400 range. B: Grey reading represents NCC powder, green spectra shows FTIR database for cellulose. ................................ 24	Figure 2.6 X-Ray Diffraction. The diffraction pattern shows a constructive interference which satisfies Bragg`s law. Sharp peaks indicate that NCC had its atoms organized to form crystalline structures. ...................................................................................................................................... 25	Figure 3.1 Suspensions of PAN and NCC after 120 days. A: Precipitation was observed for simple mixture; B: no visible precipitation for solvent exchange. ............................................... 38	Figure 3.2 SEM of 0% NCC (PAN 11% Control) group. A: Batch 1; B: Batch 2; C: Batch 3. ... 39	Figure 3.3 SEM of 3% NCC Simple Mixture. A: batch 1; B: batch 2; C: batch 3. ...................... 39	Figure 3.4 SEM of 3% NCC with solvent exchange. A: batch 1; B: batch 2, similar to batch 1; C: batch 3, fibers with variation in diameter were observed more frequently in this batch (Table 3.4). ............................................................................................................................................... 40	Figure 4.1 SEM images of fractured surfaces of tested beams. All beams have the load side oriented up. A: Control (no fiber) group with typical fractographic characteristics. Mirror (black arrow) followed by mist (black asterisk). Long hackle lines (white arrows) are well connected to the mist and compression curl (white asterisk); B: PAN 11% beams, features are present but less prominent. Dotted square shows a change in the pattern of fracture with a rough aspect. Hackle lines are less in number and shorter. End of mist area is not as evident as in control group. C: PAN 11% + 3% NCC group presented similar appearance as in PAN 11% group. Mirror (black arrow) is smaller, while mist area (black asterisk) is larger. Hackle lines (white arrows) are shorter, shifted towards compression curl (white asterisk) and can not be traced back to the mist area. ............................................................................................................................................... 55	xi  Figure 4.2 Nanofiber dental composite features. A: PAN 11% group presents fibers randomly oriented; B: PAN 11% + NCC 3%, squares show that areas with less fibers (right square) present more evident crack propagation (hackle lines). In areas with denser fiber concentration (left square) this characteristic is not so prominent; C: PAN 11% + NCC 3% the presence of beads appeared to deviate propagation of fracture lines (arrows); D: PAN 11% + NCC 3%, crack propagation across beads seemed to follow the weaker path at the interface between the bead and the matrix, thus deviating the fracture plane; E: PAN 11% + NCC 3% a detached bead leaves behind the perfect copy of its surface on the resin matrix. The fracture plane does not propagate across the bead (arrows); F: PAN 11% + NCC 3%, cross-sections of the fibers show a bundle-type appearance (a nanofiber seems to be composed of several, smaller, fibrils). The surface of the fibers is not smooth, confirmed by the impression left on the matrix (arrows). ..................... 55	Figure 4.3 Findings on nanofiber reinforced composites. A: Sectioned bead, showing the bundle-type appearance commonly found in cross-sections of the fibers. Nanofibers seems to be formed from multiple fibrils; B: Arrows point the resin flowing over a fiber, with intimate contact; C: Complete pull out of the fiber was not observed in most of the fibers. Fibers fractured cohesively either slightly above or below the fracture plane; D: Sectioned nanofibers from PAN 11% groups features the appearance of bundle-type (right arrow) and also the surface is rough and leaves an impression on the matrix (left arrow). .......................................................................................... 56	 xii  List of Symbols  η - intrinsic viscosity (mPa.s) C – polymer concentration (g/dL) l – support spam on flexural test (mm) b – width of flexural specimen (mm) h – height of flexural specimen (mm) m – slope of the straight line portion of the load-deflection curve (MPa) A – area under stress-strain curve on flexural test (J) xiii  List of Equations  Equation 1 – Berry number Equation 2 – Specific stress of nanofiber mats (g/Tex) Equation 3 – Specific stress of nanofiber mats (MPa) Equation 4 – Work of fracture (KJ/m2) Equation 5 – Flexural strength (MPa) Equation 6 – Flexural modulus (GPa) xiv  List of Abbreviations  ANOVA – Analysis of variance BisGMA – Bisphenol-A-glicidyl methacrylate CQ – Camphorquinone  DMF - Dimethylformamide E – Elastic Modulus EDMAB – Ethyl-4-dimethylamino benzoate EMS – Elongation at Maximum Stress F – Force FM – Flexural Modulus FS – Flexural Strength g/dL – Grams / deciliter g/mol – Grams / mol GPa – Gigapascal ISO – International Organization for Standardization KJ – Kilojoule kV – Kilo voltage m2 – Meter squared ml - Mililiter mm2 – Millimeter squared MPa -  Megapascal mPa.s – Milipascal-second xv  N – Newton NCC – Nanocrystalline cellulose NEU – Nanofiber Electrospinning Unit nm - Nanometer PAN – Polyacrylonitrile PAR – Parallel PER – Perpendicular PVDF-HF - Polyvinylidenefluoride-co-hexafluoropropylene PVS – Polyvinyl siloxane rpm – Rotations per minute SEM – Scanning electron microscopy TEGDMA – Triethylene glycol dimethacrylate UTS – Ultimate Tensile Strength w/w – weight/weight WOF – Work of Fracture YS – Yield Strength µL – Microliter µm – Micrometer   xvi  Acknowledgements  I would like start offering my biggest gratitude to my parents, Paulo, and Rosanara, and to my brother and nephew, Raphael and Pedro. You all showed me that true love does not always come as hugs and physical presence.  This master’s program would never be possible without the great support and love from all of you.  I thank my supervisor, and friend, Dr. Ricardo Marins de Carvalho. Thanks for teaching me the real meaning of the line “people that deserve respect, do not demand it”. Thanks for being patient and thanks for your guidance throughout all the steps of my program. I would like to offer my gratitude to my research committee, Dr. Tom Troczynski and Dr. Frank Ko. Your questions and comments helped me go deeper in science. Also, special gratitude to Dr. Hugo Vidotti. Thanks for training me and sharing with me your knowledge and passion about materials sciences. I would like to appreciate all the help and support from Dr. Adriana Manso. Thanks for your kind words not only regarding my research but also with all the bad times that I`ve been through here. To Brian, Marilia and Anna Guze. Your support made this dream possible. You were part of my history since 2012, and I`ll never forget what you have done for me. To my friends Dr. Luana Carvalho and Dr. Sergio Santiago, your friendship and advice made this journey lighter and funnier. Thanks for teaching me that life is not always about work, work and work. To my “family” here in Canada, Samuel, Evelyn, and Leonardo. I honestly don`t know how I would make it through this program without you. You are the best friends anyone could xvii  ever ask for. Thanks for opening your lives and your house and letting me in. You have no idea how I value every single one of you. Words are, definitely, powerless to express my gratitude. To my friend Clarissa, that even thousands of miles away was more present than never. Thanks for being there for me, as you always been my whole life. To my friends Guilherme, Eduardo, Tiago and Mateus. There is not a single day that I don’t think about you guys. Your friendship made me who I`m today, and it is needless to say that I miss you a lot. Thanks for all these years of friendship and for sharing your lives with me in the past two years. Finally, I`d like to express my gratitude to all the faculty, staff and graduate students from Materials Engineering, Chemistry, and Dentistry. The collaboration and help of every single of you added up to make this project possible. Thank you. xviii  Dedication  This thesis is dedicated to all 242 young people that died in the night of January 27th, 2013 in Santa Maria. I dedicate this work to them, their beloved friends and family. The memories of that weekend are the reasons why I started fighting for my dreams and began to live life to the fullest. Let us never forget. Let`s fight for a better world. A world where people take full responsibility for their lives and acts.1  Chapter 1: Introduction  Missing teeth by decay, periodontal disease, trauma and malformations still are objects of study in all dental fields. Despite the decrease on prevalence of dental caries by preventive measures, other conditions such as trauma, malformations, iatrogenic causes can lead to premature loss of teeth [1][2]. When it comes to the replacement of these missing elements, several treatment approaches can be made to correct such problem. When the best treatment option relies on restorative dentistry, several materials can be used. Dental composites were first introduced more than 50 years ago and despite the technology improvements over the last recent years, they are still composed by organic matrix, coupling agent and inorganic fillers[3][4]. Resin-based composite restorations have proved to be effective on a long term, but bulk fractures are among the major reasons for replacement[4][5][6][7]. One of the best features of composites is their capacity of absorbing higher energy prior to fracture, reproducing some of the toughness behavior of dentin[8]. Conversely, ceramics are known for its powerful resistance to fracture, but present the unwanted brittleness[9][10]. When the challenge consists in replacing cusps, enhanced mechanical properties are expected from dental composites. Different approaches are being tested in that direction. Among them, the use of reinforcement based on electrospun nanofibers is of interest because nanofibres offer increased surface area for interaction with the resin matrix and the possibility of matching different fiber features with other components of dental composites. In general, the addition of fibers combined or not with other fillers has been shown to improve the mechanical behavior of dental composites. However, enhancements can be better achieved with a well-dispersed mesh, a high potential to interact with polymer chains and a filler able to ease the stress applied to the matrix[11][12] [13][14]. 2   1.1 Nanofiber technology Nanofiber technology gained a lot of attention and had been appointed as a very promising method to reinforce composite structures [15][16]. At the nanometer scale (10-9 m), wonderful properties can be found [17][18]. The increase of specific surface area affects chemical, physical, and biological features. As the diameter of fibers reduce, most of the molecules will be present on the outmost layer. The availability of functional groups, ions, molecules and so can grant high reactivity to nanofibers that are not found in their traditional bulk counterparts. Nanofibers are being used in structural composites, biomedical engineering, smart clothing, high-performance electronics, catalytic systems, and drug delivery. As the technology and science evolve, the estimation is of significant further growth [16][18]. Regarding mechanical properties, the concept that tensile strength of fibers increases exponentially as diameter decreases is recognized for a long time. However, manufacturing fibers in the nanometer scale in a controlled and reproducible manner only became possible after several advances in electrospinning technique [19].  1.1.1 Electrospinning Electrospinning is one of the many methods to produce nanofibers. Other processes include chemical vapor deposition, conjugate spinning, sol-gel process, drawing, template synthesis, self-assembly, and melt blown technology. None of these methods outrun the simplicity of electrospinning [18]. Electrospinning consists on drawing fibers out of polymer solutions through electrostatic forces (Figure 1.1). Essentially, a syringe containing a polymeric solution will dispense a dope at the tip of a metallic needle. As a result of the surface tension of the fluid, the dope will be held 3  and slowly dragged by the electric field. First, the dope will assume a cone shape, known as Taylor-cone. At the point where the electrical force surpasses the surface tension, the viscous liquid is stretched through different stages. The initial stage (or straight jet zone), the solution will conform as a straight line. After a certain moment, depending on the viscosity of the solution and distance to the grounded collector plate, the once straight jet starts an ultra-fast whipping movement. During this third and last stage (spiral jet or unstable zone) the solvent used to produce the polymer solution will evaporate and nano-size polymeric fibers will be deposited in a metallic grounded collector plate [16].   Figure 1.1 Electrospinning schematic representation. It is important to emphasize that electrospinning technique, although simpler than others, encloses considerably comprehensive concepts. To fully understand the process, an exhaustive review of all scientific aspects involved would be required. Since it is not the primary focus of this work, this chapter will not scrutinize the details involved in modeling and testing electrospinning. This information is readily available in text-books and peer-reviewed publications [18][16]. 4  Fiber diameter has direct impact on fiber properties. A clear understanding of all parameters involved in fiber formation becomes crucial. The task of transforming polymeric solutions into nanofibers is governed by factors (a) from the process itself, such as applied voltage and feeding rate; (b) from the environment, such as air humidity and temperature; and finally (c) factors from the polymeric solution, such as viscosity and polymer concentration. Below, each parameter will be explained in further detail.  1.1.1.1 Parameters affecting fiber morphology 1.1.1.1.1 Viscosity, molecular weight and temperature of spinning dope Viscosity is fundamental on the electrospinning technique. High concentrations of polymers in a solution will yield higher viscosities. The higher the viscosity, the higher the fiber diameter. There is a minimum combination between polymer concentration and intrinsic viscosity so that solution can be electrospun into nanofibers. This concept, known as Berry number (equation 1), established a range where fibers will start to form free of defects or beads. It is a representative number of the entanglement between polymeric chains. 𝑩𝒆 = 𝜼 𝑪 Equation 1 Where η is intrinsic viscosity and C is polymer concentration. Berry number below 1 will form only droplets of beaded fibers. Above 1 and below 2.7 fibers in the range of 100 and 500 nm in diameter can be expected to form. Berry number between 2.7 and 3.6, fibers are expected to be between 1700 and 2800 nm in diameter, and finally, Be>3.6 will estimate fibers over 3000nm in diameter and up [20][16][21][22]. The molecular weight also affects fiber formation and diameter. It has been proposed that 12.000g/mol is the minimum molecular weight of a polymer to be used for electrospinning. To 5  produce nanofibers, a minimum entanglement between the polymeric chains is required. The concepts of viscosity, berry number, and molecular weight are intimate connected. Polymer solutions at the same concentrations using solutes of different molecular weights will have different viscosity. The higher the molecular weight, the higher the entanglement between chains, the viscosity, and the fiber diameter [20][16][18]. Also, the temperature plays a key role on fiber diameter. Again related to viscosity, as the temperature rises, the excitement between the molecules increases (physical meaning of temperature hike), bringing them looser and far apart from each other, which logically reduces its entanglement. By reducing the entanglement, viscosity also reduces and by consequence smaller fibers can be expected. Interestingly, increased temperatures can  turn highly concentrated solutions that were conceptully not spinnable into spinnable ones . There are also reports of more uniform fiber obtained at higher temperatures [18].  1.1.1.1.2 Applied voltage, feeding rate, orifice size, dielectric constant and electrical conductivity Fiber morphology is controlled by the voltage applied to the needle during electrospinning. The variations will directly affect the electrical field created between the collector plate and the tip of the syringe. Essentially, the applied voltage needs to be sufficient to overcome the surface tension of the polymeric solution. After this tipping point, the solution will be stretched and submitted to gravitational forces and several variations in the path to the collector ground. If too high, the droplet on the tip of the syringe will be dragged into the electrical field in a greater proportion than the syringe speed feeds the dope. High voltages lead to depletion of the dope at the tip of the needle, and fiber formation is discontinued. If too low voltages are applied, there can be interruption in the spinning process by spraying of polymer 6  solution on the plate. Also, the jet will tend to fall on the path to the grounded plate due to larger proportions between gravitational forces and electrical forces.  Just like viscosity, there is an ideal range that involves surpassing surface tension (lower threshold) and process interruption by extinguishing the dope (upper threshold). Within the range of applied voltage that forms continuous fibers, the greater the applied voltage, the larger the amount of spinning dope available in the electrical field, which causes fiber diameter to be larger. So, the higher the applied voltage, the thicker the fiber will be[23][20]. Feeding rate also affects fiber diameter. The greater the feeding rate, the higher the volume of solution that will be available to be draw into fiber by the electrical field. With an increased amount of solution, larger diameter of fibers is expected. Also, if needles of smaller diameters are used, smaller fibers can be expected [20][18].  Dielectric constant of solvents and electrical conductivity of solutions also play a significant role. Solvents present different solubility parameters and should be selected to each polymer according to its solubility parameters. The dielectric constant, which is the ability of a solvent to store charges, will dictate if the solvent can carry charges during the electrospinning process. It was found that the greater the dielectric constant, the smaller the fiber diameter. Also, the electrical conductivity of the solution can be improved by addition of electrical conductors, such as salts and other chemicals. It was found that the larger the charge carried by the jet, more elongation can be expected and by consequence thinner fibers [21][18]. 1.1.1.1.3 Surface tension, humidity and solvent boiling point The surface tension will directly affect the ability to form fibers. In solutions with lower surface tension, the easier will be to the applied voltage to overcome it and form fibers. Once the jet started its path to the collector plate, it goes through the straight and spiral jet. Meanwhile, the 7  solvent used to dissolve the polymer need to evaporate in the air. The presence of moisture in the air may reduce solvent evaporation from the solution and result in thicker fibers. Solvent residues in the fibers later evaporate and render pores in the fibers. Number and size of pores increase as humidity increases. Solvents with lower boiling points are expected to evaporate easier, resulting in smaller fiber diameter. Too low boiling points can impair fiber formation due to clogging on the tip of the needle. Porosity and pore size are necessary for filtration applications, also tissue templates and protective clothing[18][21]. 1.1.1.1.4 Distance between syringe tip and collector plate The distance the jet will travel in the electrical field is of particular importance. Larger distances allow time for the jet to stretch and spin, evaporating the solvent and creating thin fibers on the collector plate. However, increased distances can cause the fibers to be lost before reaching the plate. At the same time, too short distances do not allow the jet to go through the physical changes to produce nanometric fibers. The principal elongation happens in the unstable zone (spiral jet), allowing enough room to this stage is essential to electrospinning[18]. 1.2 Multifunctional nanocomposites Another approach in electrospinning technology consists in the addition of nanoparticles in the polymeric solution prior fiber formation, creating the so-called, multifunctional nanocomposites. Particles of various geometries and formats can be incorporated into the fiber. Examples of such can be found in nature, where fibrous structures are intimately connected and combined with particles to create fundamental building blocks for structural tissues and systems. Several areas of interest can benefit directly from this. For example, PAN polymeric fiber, when combined with carbon nanotubes, showed remarkable increase in tensile properties[23]. When Fe3O4 particles were added, a significant increase in magnetic properties was noted. Also, an 8  electrospun fiber mat that is essentially inert can become antibacterial by the simple addition of silver particles in the polymer solution. New properties can be directly applied to create supercapacitors, drug delivery systems, magnetic wires, biomedical scaffolds and so on[24][25][26]. Desirable properties of such nanofibrous composites rely on a good dispersion of the particles within the fiber (polymeric matrix). Such dispersion is challenging because nanoparticles present a significant surface energy and tend to agglomerate. Electrospinning is regarded as an advantageous method because the electrical forces dragging the polymer jet would assist in the dispersion of particle. Moreover, the solvent used in the polymeric solution can help separating particles equally[27].  1.3 Nanocrystalline cellulose (NCC) Among the nanoparticles used to produce multifunctional composites, a rod-like particle from one of the most abundant polymers in the world gained a lot of attention. Nanocrystalline cellulose is obtained from cellulose. There are multiple sources of cellulose and depending on the source the particle can have slightly different morphological features. Common sources of cellulose include wood, plants (cotton, sisal, potato, soy beam), tunicate, algae, and bacterial. The process of bringing raw cellulose all the way to its crystalline structure is complex and requires comprehensive understanding of all the steps. First, a homogenization and purification process is conducted to remove the matrix (hemicellulose and lignin) that holds the cellulose fibers together. Once the fibers are isolated, they are submitted to a mechanical process several times where the raw fibers will go through high pressure homogenizers, grinders, refiners, cryocrushing, high intensity ultrasonic and microfluidization. This last method exposes the fibers to a high shear stress and by transverse cleavage along the axis of the macrofibrilar structure, 9  microfibrilar cellulose (MFC) is obtained. In this last stage, amorphous and crystalline regions of cellulose will form the microfibers. A chemical step takes place to hydrolyse the crystalline part away from the amorphous. Usually sulfuric acid at 64% is applied for 30 minutes. After that, the process is cooled by distilled water and is taken to centrifugation. At the end of the process NCC is obtained in aqueous suspensions at different concentrations[28][29][30]. There are other methods to prepare NCC as well[31]. However, they all require complex knowledge about all the chemical steps involved in the process. That is one of the reasons scales up, and quality control are appointed as future challenges in using this particle[29]. It is important for readers to make a clear distinction between which type of cellulose is being used in a particular experiment. As pointed by Moon et al. (2011), there still not a standardized manner to name NCC. One can get lost in inconsistencies among different reports. As a matter of fact, there are nine types of cellulose particles, and NCC is only one of them. For instance, NCC can also be called as cellulose nanocrystals, cellulose whiskers, cellulose nano whiskers and cellulose microcrystals[29]. NCC is largely available in nature, presents high surface area, is non-toxic, biocompatible, with low density, sustainable, biodegradable, renewable, eco-friendly, and with extremely high mechanical properties. All these properties remarkably increased the interest in NCC in recent years[30]. The presence of several hydroxyl groups on the surface of NCC gives great opportunity to load or modify it with different functional groups. Such surface chemistry can be of great importance for biological and structural purposes. In the biomedical field, NCC was used to bind drugs and proven to be a potential carrier for drugs in wound dressings and oral cavities [32]. Thanks to the high crystallinity of NCC, its mechanical properties are considerably high. NCC 10  has been reported to have an elastic modulus of up to 220 GPa an a tensile strength of 7.5GPa[29]. Such figures are comparable to or even higher than steel wires. Because of such properties, NCC has been used to reinforce polymers and electrospun nanofibers. The mechanism of reinforcement is attributed to the formation of a percolation network that connects NCC and matrix through hydrogen bonds [33][34]. The hydrophilic nature of the particle can lead to increased water sorption, poor stability when used in hydrophobic polymer matrices, and poor dispersion in non-polar solvents[35]. Attempts to couple silane agents on the surface of NCC have been proposed, however it seems to be a complex task that involves remnants of water molecules and it can be challenging to achieve 100% rate of functionalization[36]. One common concern is that by adding surfactants, the desired properties can be lost if the crystal arrangement changes[37]. Moreover, because drying is fundamental for transportation and commercialization of NCC, techniques of re-dispersion in different media gain particular attention[38]. The use of NCC as a nanoparticle in electrospinning is common for hydrophilic polymers such as nanofibers that will usually be used for biological purposes, like scaffolds[39]. Many authors reported increased mechanical properties of polymer nanofibers and films by adding different amounts of NCC. Some found that high concentrations of NCC can yield the best reinforcement effect[31], while some found that very low concentrations [40][41] render the best properties and a plateau is reached without any further improvement as concentrations of NCC increase. There seems to be no consensus regarding the ideal amount to be used. Also, while NCC is stable in water, mixing with solvents and other organic media require different strategies over which no consensus seem to exist [31][40]. 11  1.4 Nanofibers and nanofibrous composites as reinforcing agents for dental composites The significant reduction in diameter and highly oriented molecules within the fiber structure renders nanofibers remarkably superior mechanical performance. Recently, it has been reported that polyacrylonitrile (PAN) nanofiber presented increased modulus of elasticity combined with significant toughness[42]. Such finding contradicts the normal paradigm of material sciences, that high modulus of elasticity is usually combined with a trade-off of toughness[42]. When it comes to dental composites and nanofiber technology, very little have been investigated, and results are not all in agreement. Flexural properties of dental resin blends have been shown to improve when combined with Nylon 6 nanofiber [13]. Conversely, when the same polymeric fiber is broken into particles and used to reinforce dental sealants, no significant increase in flexural properties was observed[43]. Also, PAN and polymethylmethacrylate (PMMA) nanofibers reduced flexural properties of dental composites. However, when used as core-shell nanofibers (PAN core and PMMA shell) positive reinforcement on flexural properties were found[44]. Cellulose acetate nanofibers were also used in attempt to reinforce dental composites. However, a reduction in flexural properties was found. Lack of adequate wetting was pointed out as a common problem that can explain negative results[45]. More recently, it has been shown that the addition of continuous PAN nanofibers to experimental dental composites increased the work of fracture significantly, without compromising any other mechanical property[46]. Nanofibers reinforced with nanoparticles also have been used to strengthen dental resins, but literature remains controversial and limited. Random and aligned Nylon 6 nanofibers containing carbon nanotubes showed limited improvement in flexural properties of commercially 12  available dental composites[47], while when combined with low concentrations of silicate crystals the properties of experimental resin blends seemed to increase [48]. A particular aspect of interest in dentistry is color. Carbon nanotubes usually give black appearance to the fiber and strategies to cover such feature can remarkably increase costs of production. NCC can be advantageous once it is a white particle when spray-dried. 1.5 Closing remarks Polyacrylonitrile is a very desirable polymer due to its high carbon content and flexibility. It is one of the materials of choice for the production of carbon fibers due to its considerable intermolecular forces between the adjacent nitrile groups, also due to a limited solubility in common solvents. Electrospinning of PAN is also largely studied. Pure PAN nanofibers and PAN-based multifunctional composites find applications from engineering to medicine. PAN fibers combined with titanium dioxide are great for catalysis and sensors. Also, when PAN nanofibers contain carbon nanotubes a remarkable increase in mechanical properties and electrical conductivity can be expected. PAN used with silver nitrate can be used to produce filters to prevent contamination by microorganism. The list of possible combinations and applications is long and seems to be in full expansion [49]. The combination of PAN with NCC seems, however, to be poorly explored. A first report used NCC suspensions to coat PAN nanofiber mats in single and double layers for filtration purposes. When coated twice, tensile properties were three times higher than fiber mats with no NCC[50]. A second report, demonstrated more than double increase in elastic modulus of single PAN nanofiber when NCC was added to the solution at very low concentration [51]. Following the successful improvement of properties of single PAN nanofiber with NCC, this study will address the following research questions: 1) Can NNC improve mechanical properties 13  of PAN nanofiber meshes? 2) Does the method of dispersion (simple mixture vs. with solvent exchange) of NCC in PAN solution affect fiber formation and the respective properties of the meshes? And 3) Can NCC-containing PAN electrospun nanofibers affect flexural properties of experimental dental composites? 14  Chapter 2: Nanocrystalline cellulose as a reinforcing agent for electrospun PAN nanofibers  2.1 Introduction The association between nanofibers and nanoparticles is not new[23][24]. It has been shown as a promising technique to enlarge the applications and potentialities of electrospun nanofibers[25][47]. Nanocrystalline cellulose is a rod-like particle that presents high surface area, distinguish mechanical properties, is renewable and abundant in nature[29][35]. Several polymers made effective use of NCC as a mechanical reinforce, mostly in film or bulk formats[29]. Alginate based films had its tensile modulus and strength increased by addition of NCC[52]. Also, polyurethane films had its elastic modulus, tensile strength and work of fracture remarkably increased by incorporating low concentrations of NCC[53]. More recently, NCC has been successfully incorporated in electrospun polymeric nanofibers with promising results for increase tensile strength and elastic modulus[31][39][33][54][55]. For instance, polyvinylidenefluoride-co-hexafluoropropylene(PVDF-HF) nanofibers were reinforced with dried NCC for applications in batteries and membrane distillation. The addition of low concentrations of NCC showed increased outcomes, but after a certain point, the properties were jeopardized[40][56]. NCC powder was used to reinforce polyacrylic acid(PAA) nanofibers. A 35-fold increase in modulus and a 16-fold increase in strength were noticed. Furthermore, when PAA nanofibers were treated by heat induced esterification, a 77-fold increase in tensile modulus was reported[57].   15  Electrospun polyacrylonitrile (PAN) nanofibers are being widely used in several areas thanks to its low weight associated with great mechanical strength and toughness, high heat and solvent resistance and high carbon content[49]. Association of PAN nanofibers with NCC was made indicating the possibility to further increase the mechanical properties of PAN nanofibers. When PAN nanofiber mats were coated with a single and a doubled layer of NCC suspension to produce stronger filtration membranes, the ultimate tensile strength increased from about 5 MPa to almost 10 MPa with one layer and almost 14 MPa with two layers of NCC[50]. Also, NCC was added at 1 and 3% of PAN nanofibers and tested as a single fiber. The addition of 3% resulted in an increase of 148% in tensile strength and an increase of 111% in elastic modulus of PAN with NCC single fiber[51].  Based on the facts that literature is scarce regarding NCC as a reinforcing agent for PAN nanofibers, that NCC was able reinforce PAN single fibers and has been associated with reinforcement of PAN nanofiber mats, this experiment aims to investigate the effect of NCC in PAN nanofibers mats by testing the following null hypothesis: The addition of NCC in 1, 2 and 3% will not increase tensile properties of PAN nanofiber meshes. 2.2 Materials and methods 2.2.1 Production of the nanofibers and incorporation of NCC. 2.2.1.1 Electrospinning of PAN nanofibers To produce polymer nanofibers, Polyacrylonitrile (PAN) powder (Mw= 150,000 g/mol, Scientific Polymer Products, Ontario, NY, USA) was dissolved in N, N-dimethylformamide (DMF) (Fisher Scientific, Waltham, MA, USA). 2.2 grams of PAN and 17.8 grams of DMF were weighed in an analytical balance (Shimadzu, ATX 124, Kyoto, Japan) yielding a final concentration of 11w/w%. The solution was stirred overnight at room temperature, and the 16  viscosity was measured by a viscometer (Viscomate Model VM-10A, CBC Co. Ltd., Tokyo, Japan) prior electrospinning (viscosity range: 430-445 mPa.s). The solution was then immediately transferred to a plastic syringe and fit to a Nanofiber Electrospinning Unit (NEU – Kato Tech, Japan). Electrospinning parameters were set as: applied voltage at 14.6kV; distance between the needle tip and the collector plate at 20cm; target speed at 2m/min; and transverse speed at 1cm/min. Syringe pump speed varied between 0.05 to 0.1mm/min to keep a clear and visible Taylor cone. Temperature and humidity in the chamber were monitored. Air humidity was kept between 30 and 50% and temperature varied between 24 and 27 degrees Celsius. The collector plate was covered with aluminum foil, and the fibers were produced over a period of 48 hours. Fibers were collected in a random mode and no attempt to align them was made. The formed fiber mat was gently removed from the collector plate and stored in a sealed plastic bag away from heat and humidity until use. 2.2.1.2 Incorporation of NCC NCC was added to the 11% PAN-DMF solution in 1, 2 and 3 w/w% (Table 2.1). Groups 0% (Control) 1% NCC 2% NCC 3% NCC PAN  2.2 g 2.2 g 2.2 g 2.2 g NCC  0g 0.022 g 0.044 g 0.066 g DMF  17.8 g 17.778 g 17.756 g 17.734 g Total weight 20 g 20 g 20 g 20 g Table 2.1 Formulations of experimental solutions. All the weighing measurements were made on an analytical balance (Shimadzu, ATX 124, Kyoto, Japan). After the addition of NCC to PAN/DMF in the vial, the suspension was submitted to vortex mixing (Digital Vortex Mixer, Fisher Scientific, Waltham MA, USA) for 5 17  min at 3000 rpm. The vials containing the suspensions were attached to a holder and sonicated for 2 hours in the maximum output setting (Misonix Sonicator 3000, QSonica Llc, Newtown, CT, USA). The cup of the sonicator was filled with water and ice that were replenished every 15 minutes to avoid overheating the polymer suspension during sonication.  After sonication, the vials were left for 15 minutes on the bench top at room temperature, the viscosity of the suspension was measured and carried to electrospinning following the procedures and parameters described above. Four groups were tested and characterized (Table 2.1).  2.2.2 Characterization of nanofibers and nanofiber mats 2.2.2.1 Tensile properties of nanofibers mats The tensile properties of the nanofiber mats were determined by testing machine (KES-G1 Kawabata, Kato Tech Co, Kyoto, Japan) with an elongation rate of 0.2mm/s. As-spun mats were tested in two orientations: parallel(PAR) and perpendicular(PER) to the rotational direction of the collector drum. Strips (5 cm x 0.5 cm) were cut from the mat following the orientations above and mounted in a custom jig that yields a gauge length of 4mm. Five samples from each orientation from each group were tested. Displacement (mm) was calculated by dividing the time (sec) required to tear the strip by the elongation rate (0.2mm/s). The strain was obtained by dividing the displacement by the gauge length. The load was registered in gram force. The areal density was calculated from the division of the weigh of the nanofibers(g) strips by its area in m2.  The specific stress in g/Tex was given by the following equation: 𝑺𝒕𝒓𝒆𝒔𝒔	 𝒈𝑻𝒆𝒙 = 𝑭𝒐𝒓𝒄𝒆 𝒈 ÷ 𝒔𝒂𝒎𝒑𝒍𝒆	𝒘𝒊𝒅𝒕𝒉	 𝒎𝒎[𝑨𝒓𝒆𝒂𝒍	𝒅𝒆𝒏𝒔𝒊𝒕𝒚	(𝒈/𝒎𝟐)  Equation 2 The stress in MPa was calculated from the following equation: 18  𝑆𝑡𝑟𝑒𝑠𝑠	 𝑀𝑃𝑎 = 	9.8	×	𝑆𝑡𝑟𝑒𝑠𝑠	 𝑔/𝑇𝑒𝑥 	×	𝑑𝑒𝑛𝑠𝑖𝑡𝑦	𝑜𝑓	𝑚𝑎𝑡𝑒𝑟𝑖𝑎𝑙	 𝑔/𝑐𝑐  Equation 3 Ultimate tensile strength (UTS), Elastic Modulus (E), Yield Strength (YS) and Elongation at Maximum Stress (EMS) were calculated from the stress/strain plots (Figure 2.1).  Figure 2.1 Representative stress/strain plot. Elastic Modulus (E) was obtained from the steepest part of the low strain section of the plot by dividing the variation in stress by variation in strain. Yield Strength (YS) was determined by constructing a line parallel to the linear portion of the graph and off set at 0.3% from the origin in the horizontal axis. Ultimate Tensile Strength (UTS) was the set as the highest value on the vertical axis of the graph. Elongation at Maximum Stress (EMS) was obtained in percentage by multiplying by 100 the strain value (horizontal axis) at the point of UTS.  2.2.2.2 Morphological characterization of nanofibers mats and NCC The characteristics of PAN nanofibers and PAN with NCC nanofibers were observed first under an optical microscope (Nikon Eclipse LV100 Optical Microscope, Tokyo, Japan). Fibers were collected at the beginning of electrospinning process by manually placing a glass slide in front of the collector plate. After fibers have been visually detected on the surface of the glass, 19  the lamina was taken to the optical microscope and evaluated under 5, 20 and 50X magnifications. Scanning electron microscope (SEM S-238ON, Hitachi, Tokyo, Japan) was used to evaluate the electrospun mats with an acceleration voltage of 10kV. Randomly selected areas of each mat were cut in 5x5mm squares and mounted on a stub with carbon tape (n = 3). The stubs were then coated with platinum/palladium (Pt/Pd) with an ion sputter coater (Hitachi E-1030 Ion Sputter Coater, Hitachi, Tokyo, Japan). Random images were taken from the selected pieces of each mat. Average fiber diameter was calculated based on 15 random measurements from pictures taken at the same magnification with an image software (ImageJ, NIH, USA). Transmission electron microscope (TEM, JEOL, JEM-1400, Tokyo, Japan) was used to morphologically characterize the NCC particles based on similar studies[31]. An aqueous suspension of NCC powder at a concentration of 0.001% in weigth was stirred overnight at room temperature. One ml of the NCC suspension was mixed for 30 seconds with 1ml Uranyl acetate solution (5% w/v) in an Eppendorf vial. A droplet (2µL) of the mixed suspension was directly applied to a TEM grid with ultra-thin carbon tape. The grids were left to dry in a dust free environment. The equipment was operated with accelerating voltage of 80kV, and NCC was identified at 10.000X magnification. 2.2.2.3 Crystallinity of NCC The crystallinity of NCC was analyzed by powder X-Ray diffraction (Bruker D8-Advance X-ray diffractometer, D8 Advance, Bruker, MA, USA) utilizing NCC powder packed in standard Bruker sample holder and rotated. The generator was set at 40kV and 40mA. The machine was operated under Bragg-Brentano configuration, with Copper Kα1 and Kα2, Nickel filter and a LynxEye silicon strip as detector. 20  2.2.2.4 Fourier transform infrared spectroscopy PAN raw polymer (Scientific Polymer Products, Ontario, NY, USA), NCC (Alberta Innovates, Alberta, Canada), nanofiber mats of pure PAN and mats containing NCC were analyzed by a Fourier transform infrared (IRPrestige-21 FTIR-8400S, Shimadzu, Kyoto, Japan) spectrometer equipped with attenuated total reflectance (ATR). The samples were compressed against the ATR crystal and scanned 36 times at a definition of 4cm-1 within the spectra of 400 and 4000cm-1. The equipment was operated in absorbance mode with Happ-Genzel function.  2.2.3 Statistical analysis Multiple Student t-tests were used to compare the tensile properties of nanofiber mats in the parallel and perpendicular orientation for each concentration of NCC. One-way analysis of variance (1-way ANOVA) was used to evaluate the effect of increasing the concentration of NCC for each mechanical property of the nanofiber mats.  Differences in average fiber diameter were investigated by one-way ANOVA. In cases where the group did not follow a normal distribution, Mann-Whitney U test was used instead of t-test. A level of significance of α=5% was set for all statistical analysis. Sigma Plot software (Systat Software, San Jose, USA) was used for the statistical analysis.  2.3 Results 2.3.1 Tensile properties of nanofibers mats Generally, the incorporation of NCC resulted in significant increases in all mechanical properties(Figure 2.2). PAN nanofibers with and without NCC showed anisotropic behavior with a prevalence of superior mechanical properties in the perpendicular direction (Table 2.2). The addition of NCC at 3% increased the UTS in 80% on the parallel and 62% in the perpendicular orientation. When 3% NCC was added to the nanofibers, an increase of 114% in the modulus of 21  elasticity was observed in the parallel orientation and 80% in the perpendicular. Yield strength followed a linear increase as the NCC was added in greater concentrations. When 3% NCC was added, an increase of 216% for parallel and 148% for the perpendicular. Elongation at maximum stress was significantly higher on the parallel orientation, which reflects its lower elastic modulus when compared to the testing in the perpendicular direction.   UTS - MPa(SD) E - GPa(SD) YS - MPa(SD)  EMS - %(SD) (n=5) PAR PER t-value PAR PER t-value PAR PER t-value PAR PER t-value 0% NCC 10.68 (1.96)C 14.60 (0.99)C* 3.99 0.14 (0.05)B 0.34 (0.11)B* 3.72 1.74 (0.23)C 3.88 (0.49)B* 8.86 39 (10)B* 16 (1)B + 1% NCC 17.47 (1.40)A 19.39 (3.32)B + 0.38 (0.18)A 0.72 (0.18)A + 5.72 (0.99)A 8.65 (1.64)A* -3.42 54 (4)A* 23 (2)A 14.307 2% NCC 14.84 (0.87)B 21.02 (2.53)AB* -5.16 0.25 (0.02)AB 0.52 (0.21)AB* -2.87 3.56 (0.16)B 8.97 (0.74)A* -15.89 60 (4)A* 26 (4)A 13.552 3% NCC 19.30 (0.97)A 23.62 (1.36)A* -5.79 0.30 (0.08)AB 0.61 (0.09)AB* -5.54 5.50 (0.23)A 9.65 (1.14)A* -7.99 56 (3)A* 26 (2)A 15.616 Different letters (one-way ANOVA) indicate significant difference between the values for each property and orientation (p<0.05). * (t-test) Indicate differences between PAR and PER for each NCC concentration for each property. Abbreviations: PAR: Parallel to rotatory drum axis; PER: Perpendicular; UTS: Ultimate tensile strength; E: Elastic modulus; YS: Yield strength; EMS: Elongation at Maximum Stress. n=5 for all groups. +t-value not available (Mann-Whitney performed). Table 2.2 Effects of NCC on PAN nanofiber tensile properties   Figure 2.2 Representative stress/strain plots for 0, 1, 2, and 3% NCC. 22  2.3.2 Morphological characterization of nanofibers mats and NCC SEM images showed randomly oriented fibers in all groups. Occasional beads were observed in groups with the addition of NCC (Figure 2.3). Some fibers appeared to merge in double formation in Group 2% NCC. The presence of NCC particles in the fibers could not be readily detected at the magnifications used for the SEM analysis.   Figure 2.3 Representative SEM images of the nanofiber mats. A: Group 0% NCC with randomly distributed  fibers with regular diameter and a slightly rough surface; B: Group 1% NCC presented randomly distributed fibers with presence of well defined beads that looked rougher on the surface; C:  Group 2% NCC  where the merging of two fibers can be observed. Beads were also observed. Fibers appeared smoother in this group.; D: Group 3% NCC showed randomly oriented fibers, with presence of beads that appeared rough on the surface. 23  The mean fiber diameter (standard deviation) of all groups is presented in Table 2.3. The one-way ANOVA showed that the addition of NCC did not result in significant increase in fiber diameter, except for group 2% NCC. Groups 0% NCC  1% NCC 2% NCC 3% NCC Mean (SD) 400 (47) nmB 422 (88) nmB 596 (199) nmA 388 (41) nmB Range 327 – 502 nm 307 – 560 nm 261 – 828 nm 288 – 450 nm Different letters (one-way ANOVA) indicate significant difference between the values for each fiber diameter(p<0.05). n=15. Table 2.3   Average diameter of the nanofibers Transmission electron microscopy showed NCC particles (Figure 2.4) with an average length of 173 nm (range: 113 to 246 nm) and average diameter of 12 nm (range: 8 to 15 nm).   Figure 2.4 TEM of NCC particles. NCC showed a rod-like appearance with a length close to 200 nm and a diameter around 12 nm.  24  2.3.3 FTIR   The FTIR (Figure 2.5) showed similar spectra for the pure PAN polymer, PAN nanofiber and PAN with NCC at 3%. The FTIR spectrum of pure NCC matched the cellulose spectrum from the database.   Figure 2.5 FTIR Readings. A: Green reading represents pure PAN polymer spectrum as provided by the manufacturer; Red reading shows a PAN fiber with 3% NCC where differences can be seen. A slight peak can be observed in the 1600-1700 range and a significant peak at the range of 2200-2400; Yellow spectra establishes a similar peak, but with higher intensity, in the 1600-1700 range as the red reading, and do not show a negative peak in the 2400 range. B: Grey reading represents NCC powder, green spectra shows FTIR database for cellulose. 25  2.3.4 X-ray diffraction The powder X-ray diffraction interference pattern (Figure 2.6) showed peaks on the 20-30 and 30-40 range of the theta scale. It indicates the crystalline structure of the NCC powder used in this study.  Figure 2.6 X-Ray Diffraction. The diffraction pattern shows a constructive interference which satisfies Bragg`s law. Sharp peaks indicate that NCC had its atoms organized to form crystalline structures. 2.4 Discussion Based on the findings of the present study, the null hypothesis must be rejected.  Data available to compare NCC incorporation to PAN is limited. In our study, spray dried NCC was directly mixed in PAN solutions, followed by a rigorous protocol of sonication prior electrospinning. Fibers were expected to contain estimated and known amounts of NCC (1, 2 and 3% of the weight of PAN used to produce the solutions). In previous study, PAN nanofibers were coated post-spinning with aqueous suspensions of NCC presented an increase of 100% in tensile strength with one layer of NCC and a further 40% when two layers were applied[50]. CNCFile: CNC.raw - Start: 5.000 ° - End: 90.012 ° - Step: 0.040 ° - Step time: 109.2 s - Anode: Cu - WL1: 1.5406 - WL2: 1.54439 - kA2 Ratio: 0.5 - Generator kV: 40 kV - Generator mA: 40 mA - Creation: 26-Lin (Cps)01020304050607080901001101201301401501601701801902002102202302402502602702-Theta - Scale6 10 20 30 40 50 60 70 80 9026  Such results must be cautiously interpreted since the amount of NCC in a single or double layer could not be precisely determined. It was evident from the study, however, that further increase of tensile strength would be expected with increased amount of NCC added to the fibers. In our study, we investigated the addition of known amounts of NCC in the polymeric solution, which could provide a better understanding of the relationship between NCC concentration and the respective changes in the tensile properties. It was evident from our findings that increased amounts of NCC resulted in overall increases in tensile properties, albeit not apparently in a linear relation. In our control group (0% NCC), we tested electrospun nanofibers produced from 11% in weight PAN solutions. Fiber diameter was, in average, 400 nm. Our fibers showed, at the perpendicular orientation, an UTS of 14 MPa, an E of 0.3 GPa, an YS of 3.8 MPa. Other studies that also tested PAN nanofibers using exactly the same methods here reported found superior values (UTS: 24 MPa, E: 1 GPa, YS: 19 MPa)[46]. Such differences can be explained by the polymer concentration used in each experiment. Higher values were found probably because a 7% w/w PAN solution was used to produce the nanofibers. Fiber diameter was mostly around 170 nm, half of the diameter of our fibers (400 nm). It is expected, by electrospinning theory, that reduced polymer concentration will produce less viscous solutions. Therefore, producing fibers with smaller diameter, which are expected to yield higher mechanical properties[18].  Our findings indicated the anisotropic behavior of the randomly oriented fiber mats, which has also been reported by others [46]. Although the reasons for such anisotropy are not fully understood, it is of value to realize that orientation plays a role in the properties and the knowledge is useful when designing applications. A difference between parallel and perpendicular orientation can be explained by the nature of the whipping movement from the 27  electrospinning process[20]. As greater strength is noticed when fiber are tested under tensile stress and unidirectionally[58], it is likely that the fibers are deposited in a way that most of them are displaced longitudinally to the long axis of the collector drum. When, tested perpendicularly to the rotational axis of the drum, fibers are being pulled along their long axis and therefore showing increased strength. Yield strength, tensile modulus and tensile strength were all increased by the addition of NCC at all different concentrations. Groups 1 and 3% NCC showed similar fiber diameter to control group. It emphasizes that increased mechanical properties were possibly due to the presence of NCC and not because of reduced fiber diameter. Reinforcement mechanisms of NCC in polymeric systems can be explained by different models that suggest that at low volume fractions exponential increases in mechanical properties can be found[29]. Percolation effect is one of the models that is used to explain how NCC reinforces composites structures[34]. In this model the formation of a highly connected rigid network is crucial so the nano particle can exceed mechanical properties of conventional composites. To present such effect, there is a minimum amount of NCC that, depending on shape, size, aspect ratio, orientation and interactions between particles, will achieve a percolation threshold. To provide conditions to form the percolation network is necessary to allow time and space so the particles can have an interfacial interaction[30]. Electrical measurements confirmed the presence of such network in composites containing NCC [59]. Considering that in this study ultimate tensile strength, elastic modulus and yield strength were higher in all NCC containing fibers it can be considered that the necessary threshold to form the percolation effect was achieved. In our study, fiber mats were tested as-spun without any further fiber treatment or fiber alignment. It has been shown that alignment of NCC within the fiber [31] [47] and post-28  treatment heat [57] can increase the crosslinking between the hydroxyl groups of NCC with the polymeric matrix and enhance mechanical properties. Although we did not explore these approaches, it is of value that we were able to demonstrate significant improvements in tensile properties of the fibers by simple addition of NCC at low concentrations.  It would be interesting to expand the study to incorporate post-spinning treatments and validate potential additional increases in the properties of PAN-NCC nanofibers. Care was taken to investigate and characterize the NCC used in the study and the presence of NCC in the fibers. The TEM images were only possible after several attempts. The protocol here described was based on previous studies [31]. However, it was only possible to identify the particle when Uranyl acetate[60] was mixed in the aqueous suspension. Since agglomeration was identified in our TEM images, a reasonable suggestion for future work is to submit the suspension of NCC to sonication before applying to the TEM grid. Although demanding, TEM analysis was successful to illustrate the particles used in the study and to confirm their morphological features as expected. The presence of NCC in the fibers could not be confirmed from the SEM images. As the fiber diameters far exceed the size of the particles, it is possible that particles were fully embedded in the fibers and were not visible at the surface. The presence of NCC in the fibers, and more importantly, the effects of their presence could, however, be demonstrated by the marked differences in the tensile properties and the FTIR spectra. SEM findings showed the presence of beads in the fibers when NCC was added. It is generally accepted that fewer beads are expected as viscosity of the mix increases [21]. That did not seem to be case in our study since beads were observed in the NCC groups without significant changes in viscosity. The presence of beads has been anecdotally associated with poor fiber production and contrary to ideal targets in electrospinning [16], that dictates that fibers 29  surface should be defect-free, and beads are considered defects. Contrary to that principle, our study showed the presence of beads did not compromise mechanical properties. In fact, beads were found in higher number and size (data not shown) in the group with 3% of NCC, which resulted in the highest properties. It thus appears that the suggested negative aspect of the presence of beads is only cosmetic but not functional. We were not able to find any study that has demonstrated any functional compromise of nanofibers due to the presence of beads. This fact claims for future research to clarify the role of beads in nanofibers. The reinforcement effect of NCC in PAN can be useful for future research in fiber-reinforced composites for structural purposes[49]. Considering that PAN is one of the main polymers used as precursor of carbon fibers, having a raw material that can be stronger using an eco-friendly filler could bring carbon fibers to a new level. Also, NCC has been proved to be a carrier of drugs[32]. Future research could be focused on bioactivity of composite structures with fibers containing loaded NCC. In dentistry, for example, nanofibers reinforced with loaded NCC could be used to produce temporary crowns to enhance healing activity in damaged gingiva. Additionally, the hydrophilic nature of NCC can be advantageous to produce nanofiber-based filters. The fact that NCC can provide a structural reinforcement can make the filter withstand superior loads and last longer. Also, it can make the filter more effective to particles and impurities that have hydrophilic components[50]. 2.5 Conclusions The addition of NCC resulted in increased Ultimate Tensile Strength, Elastic Modulus and Yield Strength of PAN nanofibers. 30  Fibers showed anisotropic behavior, with higher strength and elastic modulus in the perpendicular orientation. Direct mixing of NCC in polymeric PAN solutions before electrospinning is a feasible method to reinforce electrospun nanofibers. 31  Chapter 3: Effect of dispersion mode on physical properties of NCC-reinforced PAN nanofibers  3.1 Introduction The popularity of Nanocrystalline cellulose increased remarkably in last few years. A future scale-up and mass production appears to be a logical next step[29][30]. In this context, dried forms of NCC are crucial for commercialization and transportation[61][38]. Distribution of particles within polymeric matrix is essential to get the most effective reinforcement effect[62][63][40]. Not only is NCC obtained in several different ways, but also its incorporation into electrospun nanofibers is done by various methods. Some obtained NCC in aqueous forms and solvent-exchanged for DMF using a rotary evaporator[31][55]. Others used dried forms (powder or freeze-dried flakes) and mixed directly in the solvent prior producing electrospun nanofibers[56][64]. Also, NCC-containing poly(lactic acid) nanofibers were generated by creating an emulsion of water-dispersed NCC in chloroform/toluene[62].  Nanocrystalline cellulose is hydrophilic by nature. Abundant hydroxyl groups on the crystal surface make the dispersion in non-polar solvents challenging[65][37]. Albeit being highly hydrophilic and, by principle, not adequate for hydrophobic applications, the renewability and eco-friendly characteristic makes NCC a valuable filler and efforts to incorporate into hydrophobic polymers is justified. In that direction, attempts to graft silane coupling agents to NCC have been reported to be successful[36]. The possibility of direct mixing NCC to solvents and polymers without prior functionalization or re-dispersion in water is desirable to reduce steps and costs for fiber 32  production. In a previous study (Chapter 2), we were able to add 1-3% concentrations of powder NCC to a mix of PAN polymer and DMF solvent by simple mixture. The mix was then sonicated and immediately used to produce electrospun nanofibers, which showed increased tensile properties when compared to PAN nanofibers without NCC. The limitation of that simple mix approach was that the suspended mixture was short-lived and had to be electrospun immediately after sonication to prevent precipitation of NCC. There seems to be no standard method to add NCC to polymer mixtures as reported methods vary widely with no evidence of superiority of one over the other. Ideally, NCC should be suspended in the solution indefinitely without precipitation in the short-term. Proper dispersion of the particle into the polymer/solvent mixture is essential for reinforcement purposes [62][63][40]. It is surprising that no study has investigated how dispersion methods of NCC could affect the properties of nanofibers produced from the mixtures. In this study we investigated the tensile properties and morphological characteristics of electrospun PAN nanofibers produced with the addition of NCC by either a direct mixing or solvent exchange methods. The null hypothesis was that the method of dispersing NCC (simple mixture vs. solvent exchange) will not affect the physical properties of electrospun PAN nanofibers. 3.2 Materials and methods 3.2.1 Electrospinning and NCC addition by different methods 3.2.1.1 Production of nanofibers Polymer solutions of PAN were produced similarly as described in the previous chapter. Polyacrylonitrile powder (Mw= 150,000 g/mol, Scientific Polymer Products, Ontario, NY, USA) was dissolved in N,N-dymethylformamide (Fisher Scientific, Waltham, MA, USA) to produce a solution of 11% PAN by weigh. Viscosity was measured (Viscomate Model VM-10A, CBC Co. 33  Ltd., Tokyo, Japan) at room temperature right before electrospinning. Fiber production was conducted in a Nanofiber Electrospinning Unit (NEU – Kato tech, Japan), with an applied voltage of 17.2kV, at 20 cm distance for all groups. The target and transverse speed were set to 2m/min and 1cm/min, respectively. All the precautions necessary to keep a constant production of the fibers were taken, which includes adjusting syringe pump speed, controlling air humidity and temperature. Fiber production period was 48 hours, the aluminum foils containing the fibers collected on it were stored in plastic sealed bags until testing. 3.2.1.2 Addition of NCC by mixing and solvent exchange Three percent NCC (Alberta Innovates, Alberta, Canada) was added to 11% PAN polymer solution by two methods: (1) simple mixture, meaning direct mixing of NCC powder to PAN polymer solution; or (2) with solvent exchange that included re-dispersing NCC powder in water followed by solvent exchange prior to adding to PAN polymer solution. The group “3% NCC simple mixture” was done by weighing 0.066 grams of NCC powder in an analytical balance (Shimadzu, ATX 124, Kyoto, Japan). The powder was mixed with 20 grams of 11% PAN polymer solution by vortex mixing (Digital Vortex Mixer, Fisher Scientific, Waltham MA, USA) for 5 min at 3000 rpm. After vortex mixing, the vial was sealed and sonicated in ice-cold water bath (Whaledent Biosonic U100, Coltene, Ohio, USA) for 2 hours. Every 15 minutes the bath was filled with small amounts of ice to avoid overheating the mixture during sonication. The suspension was then placed to rest at room temperature for 15 minutes and the viscosity was measured. The suspension was then transferred to the electrospinning unit and nanofibers were produced with the parameters described above. Group with solvent exchange was produced by dispersion of NCC in water then solvent-exchanged into a DMF suspension by vacuum assisted rotary evaporation (Buchi R-124, Flawil, 34  Switzerland).  One hundred and fifty mL of 5% NCC suspension (w/w) in water was dispensed into a round bottom flask, and 154.5 mL of DMF was added under gentle agitation. The flask was attached to the rotary evaporator, and partially immersed in a water bath (Buchi R-481, Flawil, Switzerland). The temperature of the water in the water bath was kept at 39°C. Water from the round flask was evaporated under vacuum by rotation of the flask in 50 rpm. When no more droplets were observed in the collector flask, the process was assumed completed. The weight percentage of NCC in DMF was obtained based on the dried weight of NCC after evaporating the solvent for two days. In detail, an arbitrary amount of the prepared NCC in DMF suspension was weighted in an analytical balance. After that, the vial containing the suspension of NCC in DMF was taken into a fume hood and left to dry under vacuum for 48 hours.  The final concentration of the suspension of NCC in DMF was calculated as 4.28% weight percentage. This way, every 100 g of the suspension of NCC in DMF had 4.28 grams of NCC. PAN polymer solutions and suspensions (Table 3.1) were prepared by stirring at 60 °C for 1 hour and subsequently left stirring overnight at room temperature. Groups 0% NCC (PAN 11% Control) 3% NCC Simple Mixture 3% NCC With Solvent Exchange PAN  2.2 g 2.2 g 2.2 g NCC  - 0.066 g - NCC in DMF  - - 1.55 (contain 0.066g of NCC) DMF  17.8 g 17.734 g 16.25 g Total weight 20 g 20 g 20 g Table 3.1 PAN polymer solutions (without NCC) and suspensions of PAN with NCC dispersed by simple mixing and by solvent exchange. 35  Three groups of fibers were tested and characterized: PAN nanofiber with 0% NCC (control); PAN nanofiber with 3% NCC added by simple mixture; and PAN nanofiber with 3% NCC added after solvent exchange. 3.2.1.3 Viscosity measurements and mechanical testing All viscosity measurements were done using a viscometer (Viscomate Model VM-10A, CBC Co. Ltd., Tokyo, Japan) at room temperature and immediately prior electrospinning. Tensile properties of nanofibers mats were calculated following the protocol from chapter 2. The nanofiber mats were produced in triplicates. The central area (5 cm away from the edges of the aluminum foil) of the mats was divided into 4 equally sized rectangles. From each rectangle, 2 strips were cut from each orientation (parallel and perpendicular) as described in Chapter 2. A total of 8 strips were obtained from each mat for each orientation (parallel and perpendicular) to compose the testing sample. Three fiber mats were produced from three different batches of the solutions. This resulted in 24 strips per group and orientation (n=24).  3.2.1.4 Confirmation of stability Additional vials of suspensions of each group (simple mixture vs. solvent exchange) were prepared to observe stability over time. Vials were left in an enclosed cabinet free to avoid any disturbance that could mix the suspension. Vials were visually observed.  3.2.1.5 Scanning electron microscopy Scanning electron microscopy images were produced by cutting squares of 5x5mm from random areas of the remaining pieces of the nanofiber mats. The samples were mounted on a carbon-taped stub and coated with platinum/palladium (Pt/Pd) with an ion-sputter coater (Hitachi E-1030 Ion Sputter Coater, Hitachi, Tokyo, Japan). The SEM (SEM S-238ON, Hitachi, Tokyo, Japan) was operated with an acceleration voltage of 10kV. Fiber diameter was determined with 36  imaging software (ImageJ, NIH, USA) based on 15 random measurements of each sample at the same magnification. A total of 45 measurements were obtained from each group. 3.2.1.6 Other characterization methods Characterization of NCC was performed by transmission electron microscopy (TEM), X-ray diffraction and Fourier transformed infrared spectroscopy. The NCC particles used to produce the fibers in this experiment were the same used and described in chapter 2. 3.2.2 Statistical analysis Comparison between the mechanical properties of parallel and perpendicular orientation from the nanofibers mats were performed by multiple Student t-test. One-way ANOVA was used to evaluate the effect of different dispersion methods on the tensile properties. Two-way ANOVA tested for differences in viscosity among groups and batches within each group. All data analysis was conducted with Sigma Plot software (Systat Software, San Jose, USA) at α=5%. Groups that failed the normality test had their statistical tests changed accordingly. Mann-Whitney U test was used instead of Student t-test. 3.3 Results 3.3.1 Tensile properties Overall, the addition of NCC by simple mixture increased most mechanical properties, except elastic modulus that was statistically significantly lower at perpendicular orientation (Table 3.2). When the particles were dispersed by solvent-exchange method, there was a significant increase in elastic modulus on both directions (parallel and perpendicular) and a significant decrease in ultimate tensile strength (Table 3.2). Generally, the mats showed anisotropic behaviour, with significantly higher properties at the perpendicular orientation, and significantly superior elongation in the parallel orientation (Table 3.2). Strips that were 37  mechanically stressed by the handling prior testing were not included in the sample (not tested), resulting in n<24 in some groups.  	 UTS (MPa) E (GPa) EMS (%)  PAR  PER  t-value PAR PER t-value PAR PER t-value 0% NCC (PAN 11% Control) 10.55 (2.75)A (n=23)     17.65 (4.25)A* (n=21) -6.62 0.14 (0.04)B (n=23) 0.43 (0.06)B* (n=21) -18.17 37 (6)A* (n=23) 19 (7)A (n=21) 9.27 3% NCC Simple Mixture  11.98 (3.08)A (n=22)  18.63 (5.20)A* (n=23)  + 0.19 (0.07)B (n=22)  0.34 (0.06)C* (n=23)  -7.72 41 (9)A* (n=22)  20 (7)A (n=23)  8.87 3% NCC With Solvent Exchange 7.53 (2.43)B (n=24)  13.52 (3.45)B* (n=21)  + 0.26 (0.14)A (n=22)  0.64 (0.22)A* (n=21)  + 29 (9)B* (n=24)  16 (9)A (n=21)  4.99 Different letters (one-way ANOVA) indicate significant difference between the values for each property and orientation (p<0.05); * (t-test) indicate significant difference between PAR and PER for each NCC dispersion method (p<0.05). Abbreviations: PAR: Parallel to rotatory drum axis; PER: Perpendicular to rotatory drum axis; UTS: Ultimate tensile strength; E: Elastic modulus; EMS: Elongation at Maximum Stress. +t-value not available (Mann-Whitney performed). Table 3.2 Effects of dispersion method of NCC on tensile properties of nanofiber mats 3.3.2 Viscosity  Significantly higher viscosity was observed with 3% NCC with solvent exchange. No significant differences were observed between 3% NCC simple mixture and 0% NCC (Control). There were also no significant differences in viscosity among the three batches within each group (Table 3.3).  Viscosity (mPa.s) Groups (Solution/Suspension) Batch #1  Batch #2 Batch #3 0% NCC (PAN 11% Control) n=3 436 (2.1)Aa 440 (4.7)Aa 440 (2)Aa 3% NCC Simple Mixture n=3 439 (1)Aa 440 (1)Aa 443 (3.1)Aa 3% NCC With Solvent Exchange n=3 663 (1)Ba 662 (0.6)Ba 663 (1)Ba Table 3.3 Viscosity of solutions tested. Capital letters indicate comparisons among solutions and lower case among batches. 38  3.3.3 Confirmation of stability Vial containing suspension produced by simple mixture presented clear precipitation over 120 days, while the one produced by solvent exchange provided a stable suspension, without visual precipitation over the same period of time (Figure 3.1).  Figure 3.1 Suspensions of PAN and NCC after 120 days. A: Precipitation was observed for simple mixture; B: no visible precipitation for solvent exchange. 3.3.4 Scanning electron microscopy Scanning electron microscopy (SEM) images of 0% NCC (Control) group showed fibers with random distribution with uniform diameter (Figure 3.2). Group 3% NCC Simple Mixture (Figure 3.3) and Group 3% NCC With Solvent Exchange (Figure 3.4) also showed randomly distributed fibers with a regular diameter. There was a significant difference among in fiber diameter among the groups (Table 3.4) Group 3% NCC With Solvent Exchange had the greatest fiber diameter, followed by 3% NCC Simple Mixture and Control group. The presence of NCC increased fiber diameter regardless of the method of dispersion. 39   Figure 3.2 SEM of 0% NCC (PAN 11% Control) group. A: Batch 1; B: Batch 2; C: Batch 3.  Figure 3.3 SEM of 3% NCC Simple Mixture. A: batch 1; B: batch 2; C: batch 3. 40   Figure 3.4 SEM of 3% NCC with solvent exchange. A: batch 1; B: batch 2, similar to batch 1; C: batch 3, fibers with variation in diameter were observed more frequently in this batch (Table 3.4).  Group Average diameter (SD) 0% NCC (PAN 11% Control) n=45 570 (130) nmC  Range: 358 – 933 nm 3% NCC Simple mixture n=45 692 (159) nmB Range: 392 – 693 nm 3% NCC With Solvent Exchange n=45 793 (217) nmA Range: 392 – 1742 nm Different letters (one-way ANOVA) indicate significant difference between the values (p<0.05). Table 3.4 Fiber diameter according to groups. 3.3.5 Other characterization methods Characterization of NCC was performed as described in Chapter 2, under 2.2.2.2, 2.2.2.3 and 2.2.2.4 headings.  3.4 Discussion Our findings indicate that the null hypothesis must be rejected. The methods of dispersion of NCC in PAN solutions (dissolved in DMF) affected the physical properties of electrospun 41  nanofibrous composites. To the best of our knowledge, this is the first report on a direct comparison in methods to disperse NCC to produce nanofibrous composites, which makes challenging to stablish connections with current literature. For the group 3% NCC with solvent exchange, a significant increase (50% over control) in viscosity was found. This increase in viscosity was accompanied by a significantly higher fiber diameter (Table 3.4). We believe that the reduced UTS found in our study was a consequence of the greater fiber diameter observed in the NCC-solvent exchange group rather then due to other factors[18]. In fact, it is usually expected that increasing fiber diameter results in lower mechanical properties[21]. Our findings did not fully satisfy that concept. While it was true that UTS was significantly lower for the solvent exchange group the opposite occurred with elastic modulus that showed the highest values (Table 3.2). One possible explanation can rely on a major role played by the NCC in the fiber matrix. Solvent-exchange method seems to provide a better distribution of the particles (Figure 3.1) which made the NCC characteristics (high modulus) prevail over the low-modulus high UTS of the fiber(tough). In the simple-mixture group, the features of the PAN fibers seemed to play a major role considering the higher UTS with lower modulus, which could be used to speculate a worse distribution of NCC within the matrix. Regarding the reasons why an increased viscosity was encountered in the solvent exchange group, it can be speculated that water provided a good dispersion of the particles, increasing dramatically the area occupied by NCC in the suspension. Because NCC is a nano-sized particle, it has indeed a huge surface area[35]. Also its hydrophilic nature[29] leads to a good dispersion in polar solvents, such as water. The step of putting NCC in water, prior to change by DMF, seems to order NCC occupying an expressive space within the suspension. A 42  restricted mobility between molecular chains of PAN due to the presence of well-dispersed NCC in the suspension would logically lead to a greater entanglement and by consequence significantly higher viscosity.  In this study, spray dried NCC from Alberta Innovates was used. The degree of re-dispersability of dried forms of NCC seems to be related to PH, and the presence of remnant water in the moment of drying[37][65] [38]. Dried forms of NCC deserve attention following the rationale of future scale up and industrial applications of NCC. Our experiment can provide evidence for future research regarding the impact of dried forms of NCC in the production of nanofibers reinforced with NCC. As mentioned, to the best of our knowledge there is no literature that compared how different methods of dispersion of NCC would affect the formation of electrospun nanofibers. Comparisons should be limited and done with caution. Authors have used solvent-exchange prior producing polymethymethacrylate (PMMA) nanofibers with NCC at 0, 5, 9, 17, 23, 33 and 41% in weight[31]. Fiber diameter was reported in nanometers as 459, 474, 450, 431, 280, 269, and 182 respectively. As noticed, as the amount of NCC increased, a reduction in fiber diameter was encountered. Our findings oppose from those, once inclusion of NCC in PAN nanofibers resulted in increased fiber diameter. A possible explanation can be given by the fact that the authors[31] adjusted the amount of PMMA in order to add NCC. Group without NCC had 100mg/ml of PMMA, while 41% had only 40mg/ml. It is important to emphasize that there is no reference to viscosity values of PMMA/NCC suspensions as well, which could help to explain different outcomes in fiber diameter. Also, tensile properties were not tested, so comparisons regarding mechanical enhancement given by NCC should be carefully conducted. Conversely, another 43  study using electrospun cellulose fibers with NCC (added by solvent exchange)[55] showed no differences in fiber diameter regardless of NCC concentration. The group 3% NCC Simple Mixture resulted in viscosity values that were no different from control group (Table 3.3), while fiber diameter was significantly higher (Table 3.4). Such increased diameter seemed not to impact mechanical performance of simple mixture group, differently from solvent-exchange group (Table 3.2). Reduction of UTS in the NCC-solvent exchange seems to be an artificial outcome caused by the greater fiber diameter and not by a direct consequence of NCC. An explanation for no alterations in viscosity, but superior diameter might rely on poor dispersion of NCC in the polymeric solution and consequently in the produced fibers. Ultrasound energy has been reported to significantly affect NCC particle size[38]. In general, higher energy output results in improved dispersion of the particles. Energy output can vary dramatically depending on the ultrasound equipment used. Being 15 times higher when delivered from a cell sonicator versus a typical ultrasound bath [62]. In this study, a bath sonication was used only in the 3% Simple Mixture group. Based on our findings from previous experiments (chapter 2), it can be speculated that higher output ultrasound can produce a better dispersion of directly mixed NCC. In chapter 2, similar viscosity and fiber diameter were found for the group containing 3% NCC. Authors[56] that tested the addition of dried NCC at 1, 2, 3, 4 and 5% in 15% poly(vinylidene fluoride-co-hexafluoropropylene) electrospun nanofibers by direct mixing found increased tensile strength up to 3%, and after that a reduction in such mechanical property was reported. Also, it is mentioned that viscosity and fiber diameter increased as NCC was added, but such values are not presented[56]. The use of powder NCC was reported in the production of 44  electrospun polyacrylic acid nanocomposites[57]. The addition of NCC at 5% was responsible for reduction of suspension`s viscosities for polymer concentrations higher than 4%. Authors[57] claim that the presence of NCC hinders polymer intermolecular interactions. In their study, the presence of NCC reduced viscosity and logically fiber diameter. This latter finding disagrees from our findings where viscosity (Table 3.3) was not changed, but fiber diameter was (Table 3.4). A possible explanation for such outcome can be related to the interactions of different polymers with dried forms of NCC in different solvents. In their study[57] NCC was dispersed in alcohol, while our used DMF. The hydroxyls available from the alcoholic solvent might interact easier with NCC(that contains a several hydroxyl groups on its surface), burdening the number of molecules available in the solvent to dissolve the polymer. DMF is a solvent with a completely different chemical characteristic that would hardly interact spontaneously with NCC[37]. Such difficult interaction could allow NCC to be present in the suspension without compromising or increasing its viscosity. Simple mix can be used to add NCC to PAN and produce nanofibers with resultant changes in properties. It can be seen as alternative methods to use NCC in hydrophobic polymers, similar to attempts to graft silane coupling agents [36] or subject the particles to steric stabilization[65].  Anisotropic behaviour was found and corroborates our previous study (chapter 2) and others that used similar methods[46]. In the present study, electrospun mats were produced in triplicates for each experimental groups. This was done to incorporate the expected variability[20] that incur during the electrospinning of nanofibers and our results reflect that reality. It is unfortunate, however, that most, if not all other studies that have investigated nanofibers never report on this aspect. 45  Both dispersion methods resulted in significant differences of properties and morphology, but the diversity of the trends rendered it impossible to claim superiority of one method over the other. Considering that solvent exchange requires specific equipment and significantly more time to accomplish with no striking advantages over simple mixture of NCC to the solvent, it seems reasonable to admit that the later is an acceptable way to prepare NCC-containing polymers for electrospinning. In this study we used only one concentration of NCC (3% in weight relative to PAN concentration). It is unknown if either of the methods could be preferred or better should other concentrations have been used. Theoretically, higher concentrations of NCC would be desirable for improved properties. If that is true, then it appears that solvent exchange would be the method of choice because it apparently resulted in a more homogeneous dispersion that was more stable with no precipitation over the course of 120 days (Figure 3.1). The possible remnant water in fibers produced by solvent-exchange method can be useful for filtration purposes, where a higher affinity for hydrophilic compounds is expected. However, if structural composites were to be made from such fibers, long-term degradation by hydrolysis would need to be prevented with other measures. 3.5 Conclusions Electrospun nanofibers with NCC were successfully produced using two different methods of dispersion of NCC: direct mixing and solvent exchange. The method of dispersion affected physical properties of the fibers. Fibers showed anisotropic behavior, with perpendicular orientation resulting in superior tensile property values. Considering the limitations of this experiment, it is difficult to determine if re-dispersing NCC in water (solvent-exchange) should be preferred over direct mixing. One should consider that simple mixture in polymer solution prior electrospinning is a straight-forward method and a feasible manner to combine NCC with 46  hydrophobic polymers, but precipitation can be expected. For purposes that require long-lasting and stable suspensions (such as shipping NCC containing suspensions overseas, for example), solvent-exchange method seems to provide a better option.  47  Chapter 4: Effects of NCC-containing PAN-electrospun nanofibers in flexural properties of experimental dental composites.  4.1 Introduction Di-methacrylate dental composites are widely and successfully used in clinical dentistry as the material of choice for most of the procedures requiring reconstruction of missing tooth structure[5]. Over longer-term, bulk fracture of the restorative material has been regarded as one of the major cause of failure that requires replacement[3][4]. To overcome the problem, reinforcement of the material to make it more durable is desirable. Among several approaches to address that issue, the reinforcement of dental composites with electrospun nanofibers has been recently the focus of some research.  Fibers have been incorporated into resin matrices either as particles reconditioned from fibers[43][48] or as continuous fibers[46][66][13]. Although some disagreement exists from the reports [45][44], it appears that the general conclusions indicate that the presence of nanofibers increases the energy necessary to break dental composites. Electrospinning technology made possible the combination of nanoparticles with nanofibers, creating the so-called multifunctional composite nanofibers[25]. Such technique can be used to increase the mechanical properties of regular polymeric nanofibers significantly[23][24]. Up to date, few studies have tried to investigate the effects of nanofibrous composites as reinforcing agents for dental composites[63][47]. Nanocrystalline cellulose (NCC) is a renewable, biocompatible, and extremely strong rod-like particle[29]. It has been shown that low concentrations of NCC can remarkably increase mechanical properties of electrospun nanofibers [50][40], and dental glass-ionomer cements[67]. 48  Polyacrilontrile nanofibers are known for its high toughness and high strength at low weight ratio[42][49], and have already been demonstrated to positively reinforce dental composites[46]. However, to the best of our knowledge, there is no report combining NCC and PAN to reinforce dental composites. Expectedly, the use of a particle-reinforced nanofiber to reinforce a resin matrix would combine the effects to result in a fiber-reinforced composite with superior properties. We have previously demonstrated that the tensile properties of PAN nanofibers could be significantly improved by the addition of 1-3% NCC to the electrospun polymer (see chapter 2). In this study, we tested the hypothesis that the addition of NCC-reinforced PAN nanofibers to methacrylate resin blends could improve the flexural properties of the resultant composite. The null hypothesis tested was that there would be no effect on flexural properties of experimental dental composites when reinforced with PAN nanofibers and NCC-containing PAN nanofibers. 4.2 Materials and methods 4.2.1 Electrospinning NCC-containing nanofibers and characterization Production and characterization of electrospun nanofibers are described in chapter 3. Groups 0% NCC (PAN 11% Control) and 3% NCC Simple mixture were used to produce experimental dental composites.  4.2.2 Creation and characterization of nanofiber reinforced experimental dental composites 4.2.2.1 Fabrication of BisGMA/TEGDMA monomer blends Resin monomers (Table 4.1) were provided by Esstech (Essington, PA, USA). Bisphenol A-diglycidyl ether dimethacrylate (BisGMA), triethylene-glycol dimethacrylate (TEGDMA), camphorquinone (CQ) and ethyl N,N-dimethyl-4-aminobenzoate were weighed in an analytical 49  balance (Shimadzu, ATX 124, Kyoto, Japan). Monomers were manually mixed with a spatula at room temperature in an Ambar glass vial, protected with aluminum foil. After initial mix, the vial was left on a stirring plate overnight in a yellow room. Proportion Components Weight % Viscosity (mPa.s) 50:50 BisGMA 49.375 267 TEGDMA 49.375 CQ* 0.25 EDMAB* 1 Abbreviations: BisGMA: bisphenol A-diglycidyl ether dimethacrylate; TEGDMA: triethylene-glycol dimethacrylate; CQ: camphoroquinone; EDMAB: ethyl N,N-dimethyl-4-aminobenzoate. * required for photoactivation and polymerization of the resin Table 4.1 Experimental dental resin and viscosity  A total of 120 ml of experimental resin was produced. This amount was sufficient to produce all specimens, therefore avoiding variations between batches during the manufacturing process of the composites. 4.2.2.2 Production of nanofiber reinforced composite Samples of the nanofiber mat were cut from random areas of the mat in the perpendicular orientation (higher mechanical properties as described in chapters 2 and 3) with a 14.5 x 24.5 mm dimension. The cut samples were stacked dry in a mold (2 x 15 x 25 mm) of polyvinyl siloxane (PVS) to determine the maximum number of sheets (nanofiber mats) that would fit in the mold. It was determined that 5 sheets of cut samples would fill the mold. For each available mold, 5 sheets of mat were cut, separated and weighed in an analytical balance to later determine the mass ratio of fiber to resin of the final, cured block of composite. Blocks of pure resin blend 50  were produced as control group (no fiber), and blocks containing PAN nanofibers with and without the addition of 3% NCC were produced as experimental groups.  To produce the composite blocks, the experimental resin blend was first placed in vacuum overnight (-28 mm Hg) to remove air bubbles from the handling and mixing. Following, separate 180 µL aliquots of the resin were dispensed with a pipette on an aluminum foil. Individual mats were placed on each drop of resin to allow wetting and embedding and then taken to vacuum overnight (- 28 mm Hg).  Subsequently, embedded mats were stacked one by one in the PVS mold. Each layer was meticulously placed and all visible air bubbles were manually removed with the aid of a spatula. Once the mold was filled with the embedded mats, a thin sheet of transparent cellulose acetate was placed on the top and covered with a glass slide. Pressure clamps were placed at each end of the glass slide to hold constant pressure to the block. After that, the samples were light cured (Bluephase 2.0i; Ivoclar Vivadent, Liechstein, Germany) by sequential exposures of 30 seconds in 9 different spots to cover the entire block surface. The energy output of the curing light was measured and kept in the range between 1380 – 1420 mW/cm2. After curing, the clamped block was left in a dark room overnight. Next day, the block was removed from the mold and the other side was light-activated using the same protocol. Once both sides were cured, the block was removed and hand polished with a #320 grit sandpaper. The final block was weighed to determine the fiber/resin mass ratio using the respective, previously measured weight of the fiber mats that composed the block. The cured block was kept in water at 37°C for 24 hours until further processing. 4.2.2.3 Flexural properties testing Testing was performed according to ISO 10477 standard. Beams measuring 25 x 2 x 2 mm were produced by cutting the composites blocks with a precision cutting machine (SYJ-150 51  Low Speed Diamond Saw, MTI Corporation, USA). Specimens were stored in water at 37°C for 5 days before testing. Three-point bend test was performed in a universal testing machine (Shimadzu Corp., Kyoto, Japan) with a 20-mm span and cross-head speed of 1mm/min. All beams were tested with load applied to the surface that was first exposed to light-curing. Work of fracture (WF), flexural strength (FS) and flexural modulus (FM) were calculated by a software (Trapezium X, Shimadzu Corp., Kyoto, Japan), using the following equations: 𝑊𝐹 =	 𝐴𝑏ℎ Equation 4 𝐹𝑆 = 	 3𝐹𝐿2𝑏ℎb Equation 5 𝐹𝑀 =	 𝐿c𝑚4𝑏ℎc Equation 6 Work of fracture was expressed in kJ/m2, flexural strength was given in MPa and flexural modulus in GPa. “A” stands for the area under the stress-strain curve, “b” is the width of the sample, “h” is the height of the beam (both in mm). “F” is the maximum applied load in Newtons, “L” is the support spam and “m” is the slope of the initial straight portion of the load-deflection curve. 4.2.2.4 Fractographic analysis Failure features were analyzed at the fractured surfaces using fractography principles. Tested beams were analyzed in two different steps. First, all samples were examined under an optical microscope. Second, randomly selected samples were further examined under a scanning 52  electron microscope (SEM S-238ON, Hitachi, Tokyo, Japan) operated at an acceleration voltage of 10kV.  4.2.2.5 Areal density of fibers in flexural beams Estimation of fiber density was performed on fractured beams under SEM. Images at the same magnification (1.5K) were taken from the fractured sites. The images were transferred to Image J Software and a standard area (417 µm2) was determined on the images. The number of fibers within the determined area was manually counted. The number of fibers was counted from three separate areas of the examined surface and then reported as the average number of fibers per area. 4.2.3 Statistical analysis Nanofiber mats properties were compared as described in chapter 2. To determine the effect of adding fibers with and without NCC to resin blocks, a one-way analysis of variance (1-way ANOVA) and a Holm-Sidak post hoc test were used. Level of significance was set at α=5% and Sigma Plot software (Systat Software, San Jose, USA) package was used to run all statistical analysis. 4.3 Results 4.3.1 Characterization of electrospun nanofiber mats Characterization of electrospun nanofiber mats used to produce dental composites are presented in chapter 3. Based on the findings, group PAN 11% Control (Figure 3.2) and 3% NCC Simple Mixture (Figure 3.3) from chapter 3 were used in the perpendicular orientation as PAN fibers without and with NCC respectively. 53  4.3.2 Flexural properties of PAN with and without NCC reinforced composites Flexural strength was significantly higher in composites produced with NCC-containing fibers. Flexural modulus was statistically reduced when composites had fibers, regardless of the presence or not of NCC in the fibers. The addition of nanofibers alone, followed by the addition of the nanofibers containing NCC resulted in gradual increases in the work of fracture. The addition of PAN nanofibers resulted in significantly higher energy necessary to break the specimens; and when fiber containing NCC were added, further significant increases were observed in the work of fracture (Table 4.2). Groups Flexural Strength MPa (SD) Flexural Modulus GPa (SD) Work of Fracture KJ/m2 (SD) Resin blend (No Fiber) (n=18)  94.91 (11.43)B 2.85 (0.21)A 5.46 (1.88)C 0% NCC - PAN (n=47) 99.03 (8.80)AB 2.62 (0.21)B 7.11 (2.09)B 3% NCC - PAN (n=42)  101.39 (5.92)A 2.60 (0.17)B 8.05 (1.62)A Different letters (one-way ANOVA) indicate significant difference between the values for each property (p<0.05).  Table 4.2 Effects of the presence of NCC in PAN nanofiber reinforced dental composites. 4.3.3 Fractographic analysis Fractrographic analysis of fractured beams under optical microscope showed that samples from control group (no fiber) had characteristic features of a brittle-like material. Mirror, mist, hackles and compression curl were all detected in the samples examined. Groups containing fibers (with and without NCC) did not exhibit these characteristics under light microscopy, except for compression curl. Under scanning electron microscopy, a more evident difference between the fracture pattern of non-reinforced and reinforced beams was detected (Figure 4.1). The fibers were readily distinguished from the resin matrix, randomly distributed within the resin matrix (Figure 54  4.2 A, B). When beads were present, they were associated with a perceived deviation of the fracture propagation planes and direction (Figure 4.2 C, D, E). There were virtually no voids detected between the fibers and the matrix in the examined samples (Figure 4.2 F, 4.3 D). Generally, fibers were not damaged or altered by the manufacturing process of the composites. Interestingly, individual fibers in both groups (with and without NCC) seemed to be formed by several distinct fibrils bound together to form the reinforcing fiber (Figure 4.3 A).  The fibers seemed to integrate well with the resin matrix, indicating a good wetting of the fibers by the resin, despite no attempts to functionalize the fibers in this study (Figure 4.2 F). In some areas, the resin matrix appeared to intimately cover the fiber (Figure 4.3 B). No complete fiber pull out could be observed, thus corroborating the intimate contact between fibers and matrix (Figure 4.3 C).    55  Figure 4.1 SEM images of fractured surfaces of tested beams. All beams have the load side oriented up. A: Control (no fiber) group with typical fractographic characteristics. Mirror (black arrow) followed by mist (black asterisk). Long hackle lines (white arrows) are well connected to the mist and compression curl (white asterisk); B: PAN 11% beams, features are present but less prominent. Dotted square shows a change in the pattern of fracture with a rough aspect. Hackle lines are less in number and shorter. End of mist area is not as evident as in control group. C: PAN 11% + 3% NCC group presented similar appearance as in PAN 11% group. Mirror (black arrow) is smaller, while mist area (black asterisk) is larger. Hackle lines (white arrows) are shorter, shifted towards compression curl (white asterisk) and can not be traced back to the mist area.   Figure 4.2 Nanofiber dental composite features. A: PAN 11% group presents fibers randomly oriented; B: PAN 11% + NCC 3%, squares show that areas with less fibers (right square) present more evident crack propagation (hackle lines). In areas with denser fiber concentration (left square) this characteristic is not so prominent; C: PAN 11% + NCC 3% the presence of beads appeared to deviate propagation of fracture lines (arrows); D: PAN 11% + NCC 3%, crack propagation across beads seemed to follow the weaker path at the interface between the bead and the matrix, thus deviating the fracture plane; E: PAN 11% + NCC 3% a detached bead leaves behind the perfect copy of its surface on the resin matrix. The fracture plane does not 56  propagate across the bead (arrows); F: PAN 11% + NCC 3%, cross-sections of the fibers show a bundle-type appearance (a nanofiber seems to be composed of several, smaller, fibrils). The surface of the fibers is not smooth, confirmed by the impression left on the matrix (arrows).    Figure 4.3 Findings on nanofiber reinforced composites. A: Sectioned bead, showing the bundle-type appearance commonly found in cross-sections of the fibers. Nanofibers seems to be formed from multiple fibrils; B: Arrows point the resin flowing over a fiber, with intimate contact; C: Complete pull out of the fiber was not observed in most of the fibers. Fibers fractured cohesively either slightly above or below the fracture plane; D: Sectioned nanofibers from PAN 11% groups features the appearance of bundle-type (right arrow) and also the surface is rough and leaves an impression on the matrix (left arrow). 4.3.4 Areal density of fibers on flexural beams and mass ratio for fiber in composite blocks The amount of fibers per mm2 (Table 4.3) was very similar for both groups (with and without NCC). Overall, around 120,000 fibers were present per mm2. The average mass ratio for PAN fibers without NCC was 4.7% and with NCC was 5.8%. 57  PAN 11% PAN 11% + 3% NCC Batch #1 Batch #2 Batch #3 Batch #1 Batch #2 Batch #3 32 40 62 37 59 64 45 51 61 38 58 58 50 48 64 31 60 49 Average: 42.3 Average: 46.3 Average: 62.3 Average: 35.3 Average: 59 Average: 57 Average: 50.3 per 417µm2 Average: 50.4 per 417µm2 Average: 119,904/mm2 Average: 120,863/mm2 Table 4.3 Measurements of areal density of fiber on flexural beams. 4.4 Discussion Based on the results of the present study, the null hypothesis was rejected. It was found that the stacking of 11% PAN nanofibers to built an experimental dental composite increased significantly the work of fracture and resulted in higher flexural strength, albeit not significantly. Furthermore, when 3% NCC was incorporated in the PAN fibers and then inserted in our resin-based composites, the increases were statistically significant for both work of fracture and flexural strength.  The addition of random nanofiber mats, regardless of the presence of NCC, reduced significantly the flexural modulus. Our findings mostly agree with previous works using analogous methodology[46]. In their work, 7% PAN nanofibers were produced and stacked in different concentrations, using different viscosity resin-blends. It was found that 5.4% weight fraction of PAN nanofibers mat produced the highest and most significant increase of work of fracture. Such finding was also present in our study. Differently from Vidotti et al [45], our study 58  showed a significant decrease in flexural modulus. The increase in flexural properties herein reported was also found in studies[13] that used different polymer. Nylon 6 in similar mass concentrations (5 – 7.5%) increased flexural properties of 50/50 BisGMA/TEGDMA blends[13].  When compared to studies[47] that also used nanoparticles in their fibers to reinforce dental composites, our findings are partially in agreement. Borges et al found that random fibers with 0.5% of carbon nanotubes significantly increased flexural strength[47]. The presence of carbon nanotubes in random nylon 6 fibers was only effective as a reinforcing agent at 0.5%. When 1.5% was used, there was no difference from control group[47]. However, when NaF crystals were used, no significant role on flexural properties of BisGMA/TEGDMA blends was observed[63].  In the present experiment, a fixed concentration of NCC was used. NCC is known to reinforce nanofibers by a percolation network formed between the hydroxyl groups available on its surface and the radicals from the polymeric matrix[34]. This close interaction can explain why NCC-containing PAN nanofibers resulted in the highest and statistically superior flexural strength and work of fracture. To our knowledge, this is the third report using composite nanofibers to mechanically improve dental resins, and the first using NCC. In fact, the combination of nanofibers with nanoparticles is a complex matter. Interactions between particles, polymers, solvents and parameters involved in electrospinning can directly affect the dispersion of the fillers in the nanofiber, and by consequence completely change its characteristics[62][38][25]. In the past, nanoparticles were used to produce commercially available dental composites. Their primary issue was regarding the distribution of the particles within the matrix. The high surface energy caused particles to agglomerate and loose the advantage of the nano 59  scale[68]. Findings in our study can serve as a fundamental study to proof the principle of using nanoparticles distributed by nanofibers to reinforce dental composites. Such combination could bring together the positive reinforcement of the nano-particles (dispersed in the fibrous matrix) with the advantageous reinforcement by fibers.  Also, alignment and post-spinning treatments are associated with better mechanical results[66][47]. It is important to stress that in this study fibers were used and tested as-spun.  The increased mechanical properties using PAN and NCC-containing PAN are converse to some literature. Inferior performance of only PAN fibers had been attributed to poor interfacial bonding between fiber and matrix[44]. Such discrepancy can be further explained by our SEM findings.  Scanning electron microscopy images of fibers embedded in dental resin showed a good wetting by the matrix and no pull-outs were observed on the fractured surface, despite the fact that no attempts were made to functionalize PAN and NCC-containing PAN nanofibers. The viscosity (50/50 BisGMA/ TEGDMA) of the monomer blend used might have facilitated infiltration and wetting process. Poor wetting was considered crucial to explain the fact that incorporation of cellulose acetate significantly reduced mechanical properties of dental and epoxy resins[45]. Additionally, care was taken to properly characterize fiber density in flexural beams (Table 4.3). Since the fiber density was not different among groups, we credited the differences in properties to the potential reinforcing effect of NCC. Additionally, the mass fraction (PAN 11% 4.7% and PAN 11% + 3% NCC 5.8%) can be associated with heavier fibers when NCC was added and not due to a higher number of fibers. 60  Another aspect that conflicts nanofiber technology with the data here presented is the presence of beads. The presence of beads in the fibers has long been regarded as not desirable in nanofiber production, although no clear explanation is found to justify the apparent negative effect of the presence of beads [69][21]. Beads were found in our study and, contrary to odds, they seemed to provide a positive effect in deflecting the crack propagation and, along with the fibers themselves, resulted in improved work of fracture of the composites. 4.5 Conclusions The presence of NCC in the PAN nanofibers resulted in significant improvement of work of fracture and flexural strength of experimental composite beams. This fundamental study warrants future investigation in the use of electrospun nanofibers with nanoparticles. NCC was found to be a suitable nanoparticle to reinforce experimental dental composites by incorporation via nanofiber. 61  Chapter 5: Conclusion This thesis had the goal to answer the following research questions:1) Can Nanocrystalline cellulose improve mechanical properties of Polyacrylonitrile nanofiber meshes? 2) Does the method of dispersion (Simple Mixture vs. Solvent Exchange) of NCC in PAN solution affect fiber formation and the respective properties of the meshes? And 3) Can NCC-containing PAN electrospun nanofibers affect flexural properties of experimental dental composites? Nanocrystalline cellulose, at low concentrations, was able to reinforce electrospun PAN nanofibers when directly mixed in the polymeric solution with high output sonication (chapter 2). Also, the dispersion method was shown to play a significant role in PAN nanofibers characteristics and by that its mechanical properties. The re-dispersion of the particles in water first, and then substituting it by DMF was shown to produce fibers of higher diameter and with lower UTS. At the same time, fibers produced by direct mixing at low output sonication showed no significant changes, increasing fiber diameter with no changes in mechanical properties (chapter 3). Finally, PAN fibers that contained NCC were able to improve flexural strength and work of fracture of resin composite beams made from common methacrylate dental polymers (chapter 4). Generally, the use of electrospun nanofibers and, more importantly, nanofibres containing nanoparticles is not widely explored in current dental literature. Despite the limitations of these studies, the findings point to a promising technology that has been proven to be a great asset in the future. Fundamental studies as the ones presented in here are crucial to surround researchers with evidence further investigate in the field.  62  Although NCC presents the drawback of hydrophilicity, the use of strong eco-friendly particles has a lot of potential in the future. Its incorporation in hydrophobic matrices by adding the particle in hydrophobic nanofibers might be a great asset to reinforce dental composites. Future investigations could be focused on nanofiber assemblies into yarns, functionalization of NCC and electrospun nanofibers to better match the di-methacrylate matrix characteristics of dental composites, methods of manufacturing (vacuum assisted infusion, high-pressure conditions, polymerization methods), and explore the ability that composite nanofibers can have to produce bioactive materials.  63  Bibliography [1] Friedman PK, Kaufman LB, Karpas SL. Oral Health Disparity in Older Adults. Dent Clin North Am 2014;58:757–70. doi:10.1016/j.cden.2014.06.004. [2] Zhu Y, Hollis JH. Tooth loss and its association with dietary intake and diet quality in American adults. J Dent 2014. doi:10.1016/j.jdent.2014.08.012. [3] Ferracane JL. Resin-based composite performance: Are there some things we can’t predict? Dent Mater 2013;29:51–8. doi:10.1016/j.dental.2012.06.013. [4] Ferracane JL. Resin composite—State of the art. Dent Mater 2011;27:29–38. doi:10.1016/j.dental.2010.10.020. [5] Opdam NJM, van de Sande FH, Bronkhorst E, Cenci MS, Bottenberg P, Pallesen U, et al. Longevity of Posterior Composite Restorations: A Systematic Review and Meta-analysis. J Dent Res 2014;93:943–9. doi:10.1177/0022034514544217. [6] da Rosa Rodolpho PA, Cenci MS, Donassollo TA, Loguércio AD, Demarco FF. A clinical evaluation of posterior composite restorations: 17-year findings. J Dent 2006;34:427–35. doi:10.1016/j.jdent.2005.09.006. [7] Lynch CD, Opdam NJ, Hickel R, Brunton PA, Gurgan S, Kakaboura A, et al. Guidance on posterior resin composites: Academy of Operative Dentistry - European Section. J Dent 2014;42:377–83. doi:10.1016/j.jdent.2014.01.009. [8] Bertassoni LE, Kury M, Rathsam C, Little CB, Swain MV. The role of proteoglycans in the nanoindentation creep behavior of human dentin. J Mech Behav Biomed Mater 2016;55:264–70. doi:10.1016/j.jmbbm.2015.10.018. [9] He L-H, Swain M. A novel polymer infiltrated ceramic dental material. Dent Mater 2011;27:527–34. doi:10.1016/j.dental.2011.02.002. [10] Chen C, Trindade FZ, de Jager N, Kleverlaan CJ, Feilzer AJ. The fracture resistance of a CAD/CAM Resin Nano Ceramic (RNC) and a CAD ceramic at different thicknesses. Dent Mater 2014;30:954–62. doi:10.1016/j.dental.2014.05.018. [11] Cheng L, Zhou X, Zhong H, Deng X, Cai Q, Yang X. NaF-loaded core–shell PAN–PMMA nanofibers as reinforcements for Bis-GMA/TEGDMA restorative resins. Mater Sci Eng C 2014;34:262–9. doi:10.1016/j.msec.2013.09.020. [12] Guo G, Fan Y, Zhang J-F, Hagan JL, Xu X. Novel dental composites reinforced with zirconia–silica ceramic nanofibers. Dent Mater 2012;28:360–8. doi:10.1016/j.dental.2011.11.006. [13] Fong H. Electrospun nylon 6 nanofiber reinforced BIS-GMA/TEGDMA dental restorative composite resins. Polymer 2004;45:2427–32. doi:10.1016/j.polymer.2004.01.067. [14] Sun W, Cai Q, Li P, Deng X, Wei Y, Xu M, et al. Post-draw PAN–PMMA nanofiber reinforced and toughened Bis-GMA dental restorative composite. Dent Mater 2010;26:873–80. doi:10.1016/j.dental.2010.03.022. [15] Kim J, Reneker DH. Mechanical properties of composites using ultrafine electrospun fibers. Polym Compos 1999;20:124–31. [16] Huang Z-M, Zhang Y-Z, Kotaki M, Ramakrishna S. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos Sci Technol 2003;63:2223–53. doi:10.1016/S0266-3538(03)00178-7. [17] Roco MC. National nanotechnology initiative-past, present, future. Handb Nanosci Eng Technol 2007;2. 64  [18] Ko FK, Wan Y. Introduction to nanofiber materials. Cambridge, United Kingdom ; New York: Cambridge University Press; 2014. [19] Doshi J, Reneker DH. Electrospinning process and Applications of Electrospun nanofibers. J Electrost 1995;35:151–60. [20] Deitzel JM, Kleinmeyer J, Harris DEA, Tan NB. The effect of processing variables on the morphology of electrospun nanofibers and textiles. Polymer 2001;42:261–72. [21] Fong H, Chun I, Reneker DH. Beaded nanofibers formed during electrospinning. Polymer 1999;40:4585–92. [22] Ko FK. Nanofiber Technology. In: Gogotsi IG, editor. Nanotub. Nanofibers, Boca Raton, FL: CRC Taylor & Francis; 2006, p. 233–44. [23] Ko F, Gogotsi Y, Ali A, Naguib N, Ye H, Yang GL, et al. Electrospinning of Continuous Carbon Nanotube-Filled Nanofiber Yarns. Adv Mater 2003;15:1161–5. doi:10.1002/adma.200304955. [24] Ayutsede J, Gandhi M, Sukigara S, Ye H, Hsu C, Gogotsi Y, et al. Carbon Nanotube Reinforced Bombyx mori Silk Nanofibers by the Electrospinning Process. Biomacromolecules 2006;7:208–14. doi:10.1021/bm0505888. [25] Ko FK, Li Y, Lin L, Yang H. Multifunctional Composite Nanofibers. J Fiber Bioeng Inform 2013;6:129–38. doi:10.3993/jfbi06201302. [26] Chronakis IS. Novel nanocomposites and nanoceramics based on polymer nanofibers using electrospinning process—A review. J Mater Process Technol 2005;167:283–93. doi:10.1016/j.jmatprotec.2005.06.053. [27] Zhu J, Cao W, Yue M, Hou Y, Han J, Yang M. Strong and Stiff Aramid Nanofiber/Carbon Nanotube Nanocomposites. ACS Nano 2015;9:2489–501. doi:10.1021/nn504927e. [28] Pirani S, Hashaikeh R. Nanocrystalline cellulose extraction process and utilization of the byproduct for biofuels production. Carbohydr Polym 2013;93:357–63. doi:10.1016/j.carbpol.2012.06.063. [29] Moon RJ, Martini A, Nairn J, Simonsen J, Youngblood J. Cellulose nanomaterials review: structure, properties and nanocomposites. Chem Soc Rev 2011;40:3941. doi:10.1039/c0cs00108b. [30] Habibi Y, Lucia LA, Rojas OJ. Cellulose Nanocrystals: Chemistry, Self-Assembly, and Applications. Chem Rev 2010;110:3479–500. doi:10.1021/cr900339w. [31] Dong H, Strawhecker KE, Snyder JF, Orlicki JA, Reiner RS, Rudie AW. Cellulose nanocrystals as a reinforcing material for electrospun poly(methyl methacrylate) fibers: Formation, properties and nanomechanical characterization. Carbohydr Polym 2012;87:2488–95. doi:10.1016/j.carbpol.2011.11.015. [32] Jackson JK, Letchford K, Wasserman BZ, Ye L, Hamad WY, Burt HM. The use of nanocrystalline cellulose for the binding and controlled release of drugs. Int J Nanomedicine 2011;6:321–30. doi:10.2147/IJN.S16749. [33] Zhou C, Chu R, Wu R, Wu Q. Electrospun Polyethylene Oxide/Cellulose Nanocrystal Composite Nanofibrous Mats with Homogeneous and Heterogeneous Microstructures. Biomacromolecules 2011;12:2617–25. doi:10.1021/bm200401p. [34] Favier V, Chanzy H, Cavaille JY. Polymer nanocomposites reinforced by cellulose whiskers. Macromolecules 1995;28:6365–7. 65  [35] Brinchi L, Cotana F, Fortunati E, Kenny JM. Production of nanocrystalline cellulose from lignocellulosic biomass: Technology and applications. Carbohydr Polym 2013;94:154–69. doi:10.1016/j.carbpol.2013.01.033. [36] de Oliveira Taipina M, Ferrarezi MMF, Yoshida IVP, Gonçalves M do C. Surface modification of cotton nanocrystals with a silane agent. Cellulose 2013;20:217–26. doi:10.1007/s10570-012-9820-3. [37] Viet D, Beck-Candanedo S, Gray DG. Dispersion of cellulose nanocrystals in polar organic solvents. Cellulose 2007;14:109–13. doi:10.1007/s10570-006-9093-9. [38] Beck S, Bouchard J, Berry R. Dispersibility in Water of Dried Nanocrystalline Cellulose. Biomacromolecules 2012;13:1486–94. doi:10.1021/bm300191k. [39] Zoppe JO, Peresin MS, Habibi Y, Venditti RA, Rojas OJ. Reinforcing Poly(ε-caprolactone) Nanofibers with Cellulose Nanocrystals. ACS Appl Mater Interfaces 2009;1:1996–2004. doi:10.1021/am9003705. [40] Lalia BS, Guillen E, Arafat HA, Hashaikeh R. Nanocrystalline cellulose reinforced PVDF-HFP membranes for membrane distillation application. Desalination 2014;332:134–41. doi:10.1016/j.desal.2013.10.030. [41] Fortunati E, Luzi F, Puglia D, Dominici F, Santulli C, Kenny JM, et al. Investigation of thermo-mechanical, chemical and degradative properties of PLA-limonene films reinforced with cellulose nanocrystals extracted from Phormium tenax leaves. Eur Polym J 2014;56:77–91. doi:10.1016/j.eurpolymj.2014.03.030. [42] Papkov D, Zou Y, Andalib MN, Goponenko A, Cheng SZD, Dzenis YA. Simultaneously Strong and Tough Ultrafine Continuous Nanofibers. ACS Nano 2013;7:3324–31. doi:10.1021/nn400028p. [43] Hamilton MF, Otte AD, Gregory RL, Pinal R, Ferreira-Zandoná A, Bottino MC. Physicomechanical and antibacterial properties of experimental resin-based dental sealants modified with nylon-6 and chitosan nanofibers. J Biomed Mater Res B Appl Biomater 2015;103:1560–8. doi:10.1002/jbm.b.33342. [44] Lin S, Cai Q, Ji J, Sui G, Yu Y, Yang X, et al. Electrospun nanofiber reinforced and toughened composites through in situ nano-interface formation. Compos Sci Technol 2008;68:3322–9. doi:10.1016/j.compscitech.2008.08.033. [45] Boyd SA, Su B, Sandy JR, Ireland AJ. Cellulose Nanofibre Mesh for Use in Dental Materials. Coatings 2012;2:120–37. doi:10.3390/coatings2030120. [46] Vidotti HA, Manso AP, Leung V, do Valle AL, Ko F, Carvalho RM. Flexural properties of experimental nanofiber reinforced composite are affected by resin composition and nanofiber/resin ratio. Dent Mater 2015;31:1132–41. doi:10.1016/j.dental.2015.06.018. [47] Borges ALS, Münchow EA, de Oliveira Souza AC, Yoshida T, Vallittu PK, Bottino MC. Effect of random/aligned nylon-6/MWCNT fibers on dental resin composite reinforcement. J Mech Behav Biomed Mater 2015;48:134–44. doi:10.1016/j.jmbbm.2015.03.019. [48] Tian M, Gao Y, Liu Y, Liao Y, Xu R, Hedin NE, et al. Bis-GMA/TEGDMA dental composites reinforced with electrospun nylon 6 nanocomposite nanofibers containing highly aligned fibrillar silicate single crystals. Polymer 2007;48:2720–8. doi:10.1016/j.polymer.2007.03.032. [49] Nataraj SK, Yang KS, Aminabhavi TM. Polyacrylonitrile-based nanofibers—A state-of-the-art review. Prog Polym Sci 2012;37:487–513. doi:10.1016/j.progpolymsci.2011.07.001. 66  [50] Cao X, Huang M, Ding B, Yu J, Sun G. Robust polyacrylonitrile nanofibrous membrane reinforced with jute cellulose nanowhiskers for water purification. Desalination 2013;316:120–6. doi:10.1016/j.desal.2013.01.031. [51] Yang H, Ayranci C, Cho M, Yu Y, Chan S, Park BD, et al. Process and Characterization of Cellulose Nanocrystal Reinforced Polyacrylonitrile Composite Nanofibers from Electrospinning. 27th Annu. Tech. Conf. Am. Soc. Compos. Held Jointly 15th Jt. US-Jpn. Conf. Compos. Mater. ASTM-D30 Meet., American Society for Composites; 2012, p. 213. [52] Huq T, Salmieri S, Khan A, Khan RA, Le Tien C, Riedl B, et al. Nanocrystalline cellulose (NCC) reinforced alginate based biodegradable nanocomposite film. Carbohydr Polym 2012;90:1757–63. doi:10.1016/j.carbpol.2012.07.065. [53] Pei A, Malho J-M, Ruokolainen J, Zhou Q, Berglund LA. Strong Nanocomposite Reinforcement Effects in Polyurethane Elastomer with Low Volume Fraction of Cellulose Nanocrystals. Macromolecules 2011;44:4422–7. doi:10.1021/ma200318k. [54] Peresin MS, Habibi Y, Zoppe JO, Pawlak JJ, Rojas OJ. Nanofiber Composites of Polyvinyl Alcohol and Cellulose Nanocrystals: Manufacture and Characterization. Biomacromolecules 2010;11:674–81. doi:10.1021/bm901254n. [55] He X, Xiao Q, Lu C, Wang Y, Zhang X, Zhao J, et al. Uniaxially Aligned Electrospun All-Cellulose Nanocomposite Nanofibers Reinforced with Cellulose Nanocrystals: Scaffold for Tissue Engineering. Biomacromolecules 2014;15:618–27. doi:10.1021/bm401656a. [56] Lalia BS, Samad YA, Hashaikeh R. Nanocrystalline-cellulose-reinforced poly(vinylidenefluoride- co -hexafluoropropylene) nanocomposite films as a separator for lithium ion batteries. J Appl Polym Sci 2012;126:E442–8. doi:10.1002/app.36783. [57] Lu P, Hsieh Y-L. Cellulose nanocrystal-filled poly(acrylic acid) nanocomposite fibrous membranes. Nanotechnology 2009;20:415604. doi:10.1088/0957-4484/20/41/415604. [58] Vallittu PK. High-aspect ratio fillers: Fiber-reinforced composites and their anisotropic properties. Dent Mater 2015;31:1–7. doi:10.1016/j.dental.2014.07.009. [59] Flandin L, Cavaillé JY, Bidan G, Brechet Y. New nanocomposite materials made of an insulating matrix and conducting fillers: processing and properties. Polym Compos 2000;21:165–74. [60] Kaushik M, Fraschini C, Chauve G, Putaux J-L, Moores A. Transmission Electron Microscopy for the Characterization of Cellulose Nanocrystals. In: Maaz K, editor. Transm. Electron Microsc. - Theory Appl., InTech; 2015. [61] Lee J, Cheng Y. Critical freezing rate in freeze drying nanocrystal dispersions. J Controlled Release 2006;111:185–92. doi:10.1016/j.jconrel.2005.12.003. [62] Li Y, Ko FK, Hamad WY. Effects of Emulsion Droplet Size on the Structure of Electrospun Ultrafine Biocomposite Fibers with Cellulose Nanocrystals. Biomacromolecules 2013;14:3801–7. doi:10.1021/bm400540v. [63] Cheng L, Zhou X, Zhong H, Deng X, Cai Q, Yang X. NaF-loaded core–shell PAN–PMMA nanofibers as reinforcements for Bis-GMA/TEGDMA restorative resins. Mater Sci Eng C 2014;34:262–9. doi:10.1016/j.msec.2013.09.020. [64] Lu P, Hsieh Y-L. Cellulose nanocrystal-filled poly(acrylic acid) nanocomposite fibrous membranes. Nanotechnology 2009;20:415604. doi:10.1088/0957-4484/20/41/415604. [65] Heux L, Chauve G, Bonini C. Nonflocculating and Chiral-Nematic Self-ordering of Cellulose Microcrystals Suspensions in Nonpolar Solvents. Langmuir 2000;16:8210–2. doi:10.1021/la9913957. 67  [66] Sun W, Cai Q, Li P, Deng X, Wei Y, Xu M, et al. Post-draw PAN–PMMA nanofiber reinforced and toughened Bis-GMA dental restorative composite. Dent Mater 2010;26:873–80. doi:10.1016/j.dental.2010.03.022. [67] Silva RM, Pereira FV, Mota FAP, Watanabe E, Soares SMCS, Santos MH. Dental glass ionomer cement reinforced by cellulose microfibers and cellulose nanocrystals. Mater Sci Eng C 2016;58:389–95. doi:10.1016/j.msec.2015.08.041. [68] Jandt KD, Sigusch BW. Future perspectives of resin-based dental materials. Dent Mater 2009;25:1001–6. doi:10.1016/j.dental.2009.02.009. [69] Huang Z-M, Zhang Y-Z, Kotaki M, Ramakrishna S. A review on polymer nanofibers by electrospinning and their applications in nanocomposites. Compos Sci Technol 2003;63:2223–53. doi:10.1016/S0266-3538(03)00178-7.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0300457/manifest

Comment

Related Items