Deposition and Structural Formation of 3D Alginate Tissue Scaffolds. A Thesis. Submitted to the Faculty. Drexel University. Saif El Din Khalil

Transcription

1 Deposition and Structural Formation of 3D Alginate Tissue Scaffolds A Thesis Submitted to the Faculty of Drexel University by Saif El Din Khalil in partial fulfillment of the requirements for the degree of Doctor of Philosophy December 2005

2 Copyright 2005 Saif E. Khalil. All Rights Reserved

3 ii Dedications I dedicate this work to my parents Ossama Khalil and Wafika Sultan for their love, support, and enabling me to become an engineer, my wonderful wife Dalia, for her support, motivation and encouragement throughout my doctorate graduate studies.

4 iii Acknowledgements I thank Dr. Wei Sun for his patience, support, and excellent guidance through my graduate studies, research, and thesis work. Special thanks to the thesis committee members, Dr. Alan Lau, Dr. Christopher Li, Dr. Bradley Layton, Fred Allen, and Dr. Jack Zhou. I thank all the professors that I have taken courses with them and shared their professional knowledge with me. In addition, I thank all my colleagues at the Laboratory for Computer-Aided Tissue Engineering (CATE) for their support and assistance. Many thanks to NSF for their support of the research work (NSF ) and to Therics, Inc. in their kind support of some of the deposition equipment used for this research work. Also, many thanks to the Department of Mechanical Engineering and Mechanics for their TA support and Drexel University College of Engineering for granting me the George Hill Fellowship. I would like to also thank Dr. Anthony Lowman for the mechanical testing equipment and Dr. Giuseppe Palmese for using the rheometry equipment.

5 iv Table of Contents LIST OF TABLES...x LIST OF FIGURES... xi ABSTRACT...xx 1. INTRODUCTION TISSUE ENGINEERING AND TISSUE GUIDING SCAFFOLDS HYDROGELS IN TISSUE ENGINEERING ALGINATE HYDROGEL Alginate as a Biopolymer Alginate in Tissue Engineering Mechanical Properties of Alginate POLYMER DEPOSITION FREEFORM FABRICATION USING POLYMERS FREEFORM FABRICATION IN TISSUE ENGINEERING DEVELOPMENT OF NOVEL HYDROGEL FREEFORM DEPOSITION SYSTEM RESEARCH OBJECTIVES THESIS OUTLINE DEVELOPMENT OF A MULTI-NOZZLE DEPOSITION SYSTEM FOR BIOACTIVE FABRICATION OF TISSUE SUBSTITUTE CONSTRUCTS THE SYSTEM CONFIGURATION THE MOTION AND CONTROL SUB-SYSTEM...52

6 v 2.3. THE DEPOSITION NOZZLE SUB-SYSTEMS Pneumatic Microvalve Solenoid Microvalve Piezoelectric Microvalve DEVELOPMENT OF PROCESS MODELS FOR 3D ALGINATE FABRICATION SODIUM ALGINATE RHEOLOGY PREDICTIVE MODEL FOR FLOW RATE DEVELOPMENT OF A PREDICTIVE MODEL FOR STRUT DIAMETER DEVELOPMENT OF A PREDICTIVE MODEL FOR SCAFFOLD POROSITY BULK ALGINATE ELASTIC MODULUS SCAFFOLD ELASTIC MODULUS DEVELOPMENT OF A PREDICTIVE PROCESS MODEL FOR MAXIMUM SHEAR STRESS DEPOSITION FEASIBILITY STUDY ON SODIUM ALGINATE AQUEOUS SOLUTIONS MATERIALS AND METHODS Sodium Alginate Flow Rate Measurements PNEUMATIC VALVE DEPOSITION % sodium alginate aqueous solution % sodium alginate aqueous solution % sodium alginate aqueous solution...100

7 vi % sodium alginate aqueous solution SOLENOID MICROVALVE DEPOSITION % sodium alginate aqueous solution % sodium alginate aqueous solution % sodium alginate aqueous solution % sodium alginate aqueous solution % sodium alginate aqueous solution PIEZOELECTRIC MICROVALVE DEPOSITION % sodium alginate aqueous solution % sodium alginate aqueous solution % sodium alginate aqueous solution DISCUSSION ON THE DEPOSITION SYSTEMS STRUCTURAL FORMATION OF 3D ALGINATE SCAFFOLD DESIGN AND ARCHITECTURE EARLY DEPOSITION TERMINATION SYSTEM SET-UP AND DUAL DEPOSITION OF SODIUM ALGINATE AND CALCIUM CHLORIDE SOLUTION FOR CROSS-LINKING FLOW RATE CONTROL USING PROCESS MODEL STRUT DIAMETER CONTROL USING PROCESS MODEL PORE SIZE AND POROSITY CONTROL USING PROCESS MODEL DISCUSSION ON 3D ALGINATE SCAFFOLDS BIOACTIVE CELL DEPOSITION OF 3D ALGINATE SCAFFOLD CONSTRUCTS...152

8 vii 6.1. THE STUDY OF ALGINATE CROSSLINK ON CELL VIABILITY Materials and Methods Results of Alginate Crosslink Cell Viability Discussion on Alginate Crosslink Cell Viability STUDY ON THE GELATION OF ALGINATE Materials and Methods Results of Alginate Gelation Discussion on Alginate Gelation STUDY ON THE MECHANICAL PROPERTIES OF THE BULK ALGINATE Materials and Methods Results of Bulk Alginate Mechanical Properties Discussion on Bulk Alginate Mechanical Properties STUDY OF BULK ALGINATE DEGRADATION Materials and Methods Results of Bulk Alginate Degradation Discussion Bulk Alginate Degradation CHARACTERIZATION OF BIOACTIVE ALGINATE TISSUE SCAFFOLDS: SWELLING, DEGRADATION, CELL VIABILITY AND CELL PROLIFERATION Materials and Methods Results of Bioactive Alginate Tissue Scaffolds Discussion of Bioactive Alginate Tissue Scaffold SUMMARY, CONCLUSION AND RECOMMENDATIONS SUMMARY OF THE RESEARCH...193

9 viii 7.2. CONCLUSION AND REMARKS RESEARCH CONTRIBUTIONS FUTURE WORK RECOMMENDATIONS LIST OF REFERENCES APPENDIX A: PNEUMATIC MICROVAVLE FLOW RATE RESULTS APPENDIX B: SOLENOID MICROVALVE FLOW RATE RESULTS APPENDIX C: PIEZOELECTRIC MICROVALE FLOW RATE RESULTS APPENDIX D: PROTOCOL FOR COUNTING CELLS USING HEMOCYTOMETER APPENDIX E: PROTOCOL FOR PREPARING RAT HEART ENDOTHELIAL CELLS (RHEC) CULTURE MEDIUM APPENDIX F: PROTOCOL FOR SPLITING RAT HEART ENDOTHELIAL CELLS (RHEC) APPENDIX G: PROTOCOL FOR COUNTING CELLS IN 3D ALGINATE SCAFFOLDS USING CYTOFLUORIMETRY APPENDIX H: PROTOCOL FOR FREEZING & THAWING RAT HEART ENDOTHELIAL CELLS (RHEC) APPENDIX I: PROTOCOL FOR CALCULATING PERCENTAGE OF DEAD CELLS USING LIVE/DEAD KIT WITH MICROPLATE READER...250

10 VITA ix

11 x List of Tables Table 2.1: Characteristics and comparison of the three nozzle systems...58 Table 4.1: Operating Parameters for 0.1% sodium alginate deposition using the piezoelectric microvalve Characteristics and comparison of the three nozzle systems Table 4.2: Operating Parameters for 0.25% sodium alaginate deposition using the piezoelectric microvalve Table 4.3: : Operating Parameters for 0.4% sodium alaginate deposition using the piezoelectric microvalve Table 5.1: Process Parameters Table 5.2: Results of the nozzle velocity and feasibility study for pneumatic valve using 3% (w/v) sodium alginate aqueous solution at various pressures and nozzle diameters Table 5.3: Results of the process model to predict the alginate strut diameter...142

12 xi List of Figures Figure 1.1: Schematic diagram of β-d- Mannuronic acid (M units) and α- L-Guluronic acid (G units), and sodium alginate...7 Figure 1.2: Schematic diagram of Diffusion Setting and Internal Gelation of alginate...9 Figure 1.3: Biopolymer deposition: (A) extrusion mode, (B) droplet mode Figure 2.1: Schematic diagram of 3D motion set-up for biopolymer Figure 2.2: Deposition system data processing system for converting designed scaffold models into a layered process tool path using multi-nozzles in 3D motion...50 Figure 2.3: Deposition system configuration for nozzles in 3D motion Figure 2.4: Structure of the motion system Figure 2.5: Functionality of the control software Figure 2.6: Object-relationship diagram of the control software Figure 2.7: (a) Schematic diagram of the pneumatic microvalve; (b) Schematic diagram of the air control per cycle...60 Figure 2.8: (a) Schematic diagram of the solenoid microvalve; (b) Schematic diagram of the solenoid microvalve deposition cycle...62 Figure 2.9: (a) Schematic diagram of the piezoelectric microvalve; (b) Schematic diagram of the piezoelectric microvalve deposition cycle...65 Figure 3.1: Viscosity measurement for Newtonian and Non-Newtonian Liquids...67 Figure 3.2: Effect of temperature on viscosity for typical polymer melts Figure 3.3: Viscosity versus temperature for Sodium Alginate aqueous solutions at various concentrations... 72

13 xii Figure 3.4: Viscosity versus concentration for Sodium Alginate aqueous solutions at various temperatures Figure 3.5: Comparison of experimental and analytical viscosity results for sodium alginate aqueous solutions at 25 o C Figure 3.6: Experimental and analytical values of viscosity and different shear rates for 1% (w/v) sodium alginate aqueous solution, where K = 610 and n = Figure 3.7: Experimental and analytical values of viscosity and different shear rates for 1.5% (w/v) sodium alginate aqueous solution, where K = 1259 and n = Figure 3.8: Experimental and analytical values of viscosity and different shear rates for 2% (w/v) sodium alginate aqueous solution, where K = 2010 and n = Figure 3.9: Experimental and analytical values of viscosity and different shear rates for 3% (w/v) sodium alginate aqueous solution, where K = 8587 and n = Figure 3.10: Three scenarios for strut diameter; Case 1 (v 1> v N ), Case N (v N ), and Case 2 (v 2 <v N ) Figure 3.11: (a) Layer stacking 90 degrees orientation of alginate scaffold; (b) Unit cell geometry of alginate scaffold Figure 3.12: Schematic diagram of fluid velocity through the cross section of a capillary or nozzle tip Figure 4.1: Experimental set-up for measurement of flow rate using pneumatic, solenoid, and piezoelectric microvalve systems Figure 4.2: Pressure vs. flow rate for 1 % (w/v) sodium alginate aqueous solution Figure 4.3: Nozzle diameter vs. flow rate for 1 % (w/v) sodium alginate aqueous solution Figure 4.4: Pressure vs. flow rate for 1.5 % (w/v) sodium alginate aqueous solution Figure 4.5: Nozzle diameter vs. flow rate for 1.5 % (w/v) sodium alginate aqueous solution... 99

14 xiii Figure 4.6: Pressure vs. flow rate for 2 % (w/v) sodium alginate aqueous solution Figure 4.7: Nozzle diameter vs. flow rate for 2% (w/v) sodium alginate aqueous solution Figure 4.8: Pressure vs. flow rate for 3% (w/v) sodium alginate aqueous solution Figure 4.9: Nozzle diameter vs. flow rate for 3% (w/v) sodium alginate aqueous solution Figure 4.10: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 2 mills Figure 4.11: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 3 mills Figure 4.12: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 4 mills Figure 4.13: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 5 mills Figure 4.14: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 7.5 mills Figure 4.15: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 2 mills Figure 4.16: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 3 mills Figure 4.17: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 4 mills Figure 4.18: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 5 mills Figure 4.19: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 7.5 mills Figure 4.20: Frequency vs. flow rate for 0.75% (w/v) sodium alginate aqueous solution at 2 mills

15 xiv Figure 4.21: Frequency vs. flow rate for 0.75% (w/v) sodium alginate aqueous solution at 3 mills Figure 4.22: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 4 mills Figure 4.23: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 5 mills Figure 4.24: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 7.5 mills Figure 4.25: Frequency vs. flow rate for 0.85 % (w/v) Sodium Alginate aqueous solution at 3 mills Figure 4.26: Frequency vs. flow rate for 0.85 % (w/v) Sodium Alginate aqueous solution at 4 mills Figure 4.27: Frequency vs. flow rate for 0.85 %(w/v) Sodium Alginate aqueous solution at 5 mills Figure 4.28: Frequency vs. flow rate for 0.85 % (w/v) Sodium Alginate aqueous solution at 7.5 mills Figure 4.29: Frequency vs. flow rate for 1 % (w/v) Sodium Alginate aqueous solution at 3 mills Figure 4.30: Frequency vs. flow rate for 1 % (w/v) Sodium Alginate aqueous solution at 4 mills Figure 4.31: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figure 4.32: Voltage vs. voltage for 0.1 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figure 4.33: Frequency vs. flow rate for 0.25 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figure 4.34: Frequency vs. voltage for 0.25 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figure 4.35: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figure 4.36: Frequency vs. voltage rate for 0.4 % (w/v) sodium alginate

16 xv Figure 5.1: Schematic diagram of case I were the effect of strut distance on the pore size at 250 micron constant strut diameter was studied Figure 5.2: Schematic diagram of case II were the effect of the strut diameter on the pore size at 1000 micron constant strut distance was studied Figure 5.3: (A) Excess material deposited when vale is closed; (B) EDT for a 2mm distance between vale closing and end of nozzle travel Figure 5.4: First Layer Deposition Figure 5.5: Second Layer Deposition Figure 5.6: Results of the average pressure gradients dp/dz of 2%, 3%, and 4% (w/v) Manugel solutions versus pressure for nozzle diameters of 250 μm, 330 μm, and 410 μm with the pneumatic valve Figure 5.7: Comparison of the experimental and analytical flow rate versus pressure using the average pressure gradients dp/dz of 2%, and 3%, and 4% (w/v) Manugel solutions for 3% (w/v) sodium alginate solution with nozzle diameters of 250 μm, 330 μm, and 410 μm using the pneumatic valve Figure 5.8: Results of v N versus the pressure for 3% (w/v) sodium alginate aqueous solution at 25 o C using the pneumatic microvalve system Figure 5.9: Alginate struts images fabricated at the three different velocities; (A) 5 mm/s; (B) 10 mm/s; 15 mm/s (C) Figure 5.10: Comparison of alginate strut diameters at nozzle velocities of 5 mm/s, 10 mm/s, and 15 mm/s with a constant flow rate of 0.51 μl/s for the process model and experiments Figure 5.11: Results of Case I pore sizes and strut diameters for 1000 μm, 670 μm, and 500 μm strut distances Figure 5.12: Results of Case II pore sizes and strut diameters for 750 μm, 670 μm, and 590 μm strut diameters

17 xvi Figure 5.13: (A) Close up image of pore for 3D alginate scaffold; B) Overall image of 3D alginate scaffold Figure 5.14: 40 layer scaffold (A) angle view ; (B) side view; (C) top view Figure 5.15: (A) Excess material deposition during closing of the valve in curved portions of the scaffold at 0 mm EDT; (B) Concave 3D alginate scaffold at 0 mm Figure 5.16: (A) Ideal structure formation around bends at 1 mm EDT; (B) Uniformly shaped 3D alginate scaffold at 1 mm EDT Figure 5.17: (A) Irregular material deposition placement of the scaffold at 2 mm EDT; (B) Irregularly shaped 3D alginate scaffold at 2 mm Figure 5.18: (A) Dynamic nozzle technique; (B) Stationary nozzle technique Figure 6.1: Results of the fluorescent reading versus incubation day for 1%, 1.5%, 2%, and 3% (w/v) sodium alginate solutions under constant 0.5% (w/v) calcium chloride Figure 6.2: Results of the fraction initial fluorescent reading (normalized) versus incubation day for 1%, 1.5%, 2%, and 3% (w/v) sodium alginate solutions under constant 0.5% (w/v) calcium chloride Figure 6.3: Results of the fluorescent reading versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions Figure 6.4: Results of the fraction initial fluorescent reading (normalized) versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions Figure 6.5: Results of the fraction initial fluorescent reading (normalized) versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions Figure 6.6: Optical image of Manugel RHEC non-encapsulated (control) on day 18 of incubation time

18 xvii Figure 6.7: Fluorescent image of Manugel RHEC non-encapsulated (control) on day 18 of incubation time Figure 6.8: Fluorescent image of Manugel RHEC encapsulated on day 18 of incubation time Figure 6.9: Fluorescent image of Manugel RHEC encapsulated on day 18 of incubation time Figure 6.10: Results of the swelling ratio versus Manugel concentration at various gelation times Figure 6.11: Results of the elastic modulus versus the sodium alginate concentration at various gelation times using a constant 0.5% (w/v) calcium chloride solution Figure 6.12: Results of the elastic modulus versus the gelation time at various sodium alginate concentrations using a constant 0.5% (w/v) calcium chloride solution Figure 6.13: Results of the elastic modulus versus the calcium chloride concentration at various gelation times for a constant 1.5% (w/v) Manugel Figure 6.14: Results of the elastic modulus versus the gelation time at various calcium chloride concentrations for a constant 1.5% (w/v) Manugel Figure 6.15: ESEM image of 3% (w/v) Manugel after 10 minutes gelation time using a constant 0.5% calcium chloride Figure 6.16: ESEM image of 3% (w/v) Manugel after 24 hours gelation time using a constant 0.5% calcium chloride Figure 6.17: ESEM image of 3% (w/v) Manugel after 24 hours gelation time Figure 6.18: Results of the elastic modulus versus the degradation time for Manugel at various concentrations using 0.5% (w/v) calcium chloride solution Figure 6.19: Results of the elastic modulus versus the degradation time for 1.5% (w/v) Manugel at various cell densities using 0.5% (w/v) calcium chloride solution

19 xviii Figure 6.20: Results of the elastic modulus versus the degradation time for 1.5% (w/v) Manugel at 500,000 cells/ml using 0.5% (w/v) calcium chloride solutions Figure 6.21: Results of the percentage weight loss versus the degradation time for Manugel at various concentrations using 0.5% (w/v) calcium chloride solution Figure 6.22: Results of the percentage weight loss versus the Manugel concentration at various degradation times using 0.5% (w/v) calcium chloride solution Figure 6.23: Comparison between the swelling ratios of Manugel scaffolds and bulk Manugel. P < Figure 6.24: Comparison between the percentage weight loss of Manugel scaffolds and bulk Manugel over 21 days of degradation time. P < Figure 6.25: Comparison between the elastic modulus of Manugel scaffolds and bulk Manugel over 21 days of degradation time. P < Figure 6.26: Calibration curve for determining the number of cells over the fluorescent reading. P < Figure 6.27: Calibration curve for determining the number of cells over the fluorescent reading Figure 6.28: Calibration curve for determining the number of cells over the fluorescent reading using an approximated linear equation Figure 6.29: Incubation day versus the cell number for 1.5% (w/v) sodium alginate bioactive scaffold using 0.5% (w/v) calcium chloride Figure 6.30: Effect of the flow rate on the maximum shear stress of 1.5% (w/v) of sodium alginate using pneumatic microvalve valve Figure 6.31: Effect of the maximum shear stress on the percentage of dead cells using1.5% (w/v) of sodium alginate with 1,000,000 cells/ml for pneumatic microvalve valve

20 xix Figure 6.32: Effect of the maximum shear stress on the percentage of live cells using1.5% (w/v) of sodium alginate with 1,000,000 cells/ml for pneumatic microvalve valve Figure 6.33: Optical image of a bioactive strut of Manugel RHEC encapsulated on day 14 of incubation time Figure 6.34: Fluorescent LIVE/DEAD assay image of a bioactive strut of Manugel RHEC encapsulated on day 14 of incubation time Figure 6.35: Optical image of a bioactive scaffold of Manugel RHEC encapsulated on day 14 of incubation time Figure 6.36: Fluorescent LIVE/DEAD assay image of a bioactive scaffold of Manugel RHEC encapsulated on day 14 of incubation time Figure 7.1: Alginate struts with RGD modified and unmodified surfaces Figure 7.2: Alginate scaffolds made of RGD modified and unmodified struts stacked next to one another to for a selective cell adherent bioactive scaffold Figure 7.3: Osteoblasts and HA particle encapsulated into alginate struts with RGD modified surface to promote endothelial attachment for vascularization Figure 7.4: Alginate/HA scaffolds fabricated using the current bioactive fabrication system; (A) overview of whole scaffold; (B) Close-up of scaffold showing the HA particles encapsulated within the scaffold

21 xx Abstract Deposition and Structural Formation of 3D Alginate Tissue Scaffolds Saif El Din Khalil Wei Sun, Ph.D. Tissue engineering is considered to be one of the most innovative approach for tackling many diseases and body parts that need to be replaced. Biopolymeric scaffolds have been utilized in tissue engineering as a technique to confide the desired proliferation of seeded cells in vitro and in vivo into its architecturally porous three-dimensional structures. Novel freeform fabrication methods for tissue engineering polymeric scaffolds have been an interest because of its repeatability and capability of high accuracy in fabrication resolution at the macro and micro scales. A novel multi-nozzle biopolymer deposition system which is capable of extruding biopolymer solutions and living cells for bioactive fabrication of 3D alginate tissue scaffolds is presented. The deposition process is biocompatible and occurs at room temperature and low pressures to reduce damage to cells. Sodium alginate aqueous solution is deposited into calcium chloride solution using 3DD to form hydrogel structures. Feasibility studies showed that the system is capable of extruding Manugel alginate between 0.4% and 3% (w/v). The flow rate, nozzle diameter, and nozzle velocity were studied and a model was developed to design 3D scaffolds with controlled strut diameters (D = microns) and pore sizes. In addition, rat heart endothelial cells were deposited through the system with alginate to form gel scaffold structures with encapsulated cells in a bioactive fabricated manor. The study showed that the suitable bioactive parameters

22 xxi preferred 1.5% (w/v) sodium alginate with 0.5% (w/v) calcium chloride. Cell viability studies were conducted on the cell encapsulated scaffolds for validating the bioactive freeform fabrication process that sowed viability up to 85%. The bioactive scaffold supported proliferation up to 21 days of incubation time. The elastic modulus was studied over degradation time that showed that the stiffened during the 24 hours due to crosslinking and degraded then on up to 21 days of incubation at 37 o C. The system showed potential use for accurate cell placement in tissue engineering applications and promote regenerative medicine based on CAD systems.

23

24 2 1. INTRODUCTION 1.1.Tissue Engineering and Tissue Guiding Scaffolds The loss or failure of an organ or tissue is one of the most devastating and costly problems in human health care [1]. The need for substitutes to replace or repair tissues or organs because of disease, trauma, or congenital problems is overwhelming. For example, just in the US alone, as many as twenty million patients per year suffer from various organ and tissue related maladies including burns, skin ulcers, diabetes, bone, cartilage, and connective tissue defects and diseases. More than eight million surgical procedures are performed annually to treat these cases, over 70,000 people are on transplant waiting lists, and an additional 100,000 patients die without even qualifying for the waiting list [1]. The financial cost to care for these patients has been estimated at as much as $400 billion annually. Tissue engineering integrates a variety of science and engineering disciplines to create functional tissues and organs for transplantation, which restore, maintain or improve the function of human tissues, evolves as one of the most promising therapies in regenerative medicine [2, 3]. It is considered to be the most innovative approach for tackling many diseases and body parts that need to be replaced [4, 5]. Cellular tissue engineering involves the in vitro seeding and attachment of cells into scaffolds, which then proliferate, migrate and differentiate into the specific tissue [6]. The strategy of cellular tissue engineering has focused on manipulating the cell environment by the modulation of cell extracellular matrix (ECM) and cell cell interactions. The ECM may influence

25 3 cell behavior through it material properties, surface treatment, degree of porosity, and pore size. Controlling each of these environmental influences has been used to facilitate the development of functional tissue [7]. The selection of biomaterials as the ECM depends on the type of tissue that is to be regenerated [8]. In addition, the size and the geometry of the pores to be designed within the scaffold is also important to the cell behavior and has always been one of the key parameters in the design of the ECM scaffold [9].To date, the ideal approach for perusing engineered tissue substitutes involves three subsequent procedures. First, the biological cells are identified and gathered in sufficient numbers. Second, the suitable biomaterial is identified and designed accordingly to host the gathered cells. Finally, the cells are seeded into the biomaterial for cell culturing in vitro or in vivo. Another approach of engineering tissue is to encourage cells in the host to populate in to the biomaterial structure upon implantation [10]. The biomaterial that is designed for housing the cells is referred to as scaffolds and is usually three-dimensional. Biomaterials are used as scaffolds for their uniqueness of being biocompatible, which mean that they will not be rejected by the body upon implantation. In addition to biodegrabiltiy and bioabsorbability, biomaterials can be fabricated from a wide range of materials that could be inorganic synthetic such as metals, ceramics, and polymers or from organic materials such as proteins, chitosan, and alginate [11-14]. Scaffolds used in tissue engineering are also required to have specific properties that are vital for cell regeneration. The microstructure and internal architecture of scaffolds play an important role in not only the behavior of the

26 4 cultured cells and their health, but also in influencing cell shape modeling and gene expressions that relate to cell growth to specific tissue [15, 16]. The required pore sizes vary from one cell type to another; however scaffold pore sizes have been fabricated in the range of 5 to 500 microns. In addition, scaffolds need to have sufficient mechanical strength to support and maintain the porous matrix that is required for the transportation of nutrient in the stages of incubation or implantation [17-19]. The degradation should match the proliferation of cells that migrate into the spaces that were once occupied by the scaffold. In addition, there should have an advanced fabrication system to manufacture such designed scaffold in order to achieve the consistency and repeatability in accordance to the initial design. Scaffold-guided tissue engineering, i.e., using tissue scaffold with seeded patient s own cells, is the most popular approach used to engineer tissues. In this approach, scaffolds carry cells to the desired site in the patient s body, provide a space for new tissue growth, and control the structure and function of the engineered tissues [20, 21]. In the subsequent process that gives rise to tissue function and morphogenesis, cells are significantly influenced by the microarchitecture of (3D) tissue scaffolds for their proliferation, migration and differentiation [22, 23]. The 3D tissue scaffolds play a critical role as extracellular matrices onto which cells can attach, grow, and form new tissue. These scaffolds often have intricate internal architecture, porosity, pore and connectivity in order to provide the required structural integrity, transport, and microenvironment for cell and tissue growth [24-27]. Among many reported tissue

27 5 scaffolds, biopolymer hydrogel-based scaffold materials, such as using alginate, fibrin, and chitosan, are particularly appealing due to their structural similarities to the macromolecular-based human tissues, and their biocompatibility, low toxicity, relative low cost, and availability. These polymers potentially have numerous applications in tissue engineering as delivery vehicle for cell and drug, as wound dressing, dental impression, and immobilization matrix [28-34]. However, most hydrogel scaffolds reported in the literature exhibit simple geometrical configurations, for example, injectable bulk gel, simple plates, lines and gel beads [35-43]. Although they might be suitable for studying the cell biology behavior in scaffold and hydrogel material formation, research on using such scaffolds in a 3D structural configuration and the importance of these configurations for structure-cell-tissue formation have not been well addressed [44-53]. In particular, the process science and engineering approach for fabricating 3D hydrogel scaffolds are still missing. 1.2.Hydrogels in Tissue Engineering Hydrogels are a class of highly hydrated polymer materials with water contents over 30% by weight [35]. They are composed of hydrophilic polymer chains that could be either synthetic or natural. Examples of synthetic hydrogels include HEMA (2-hydroxyethyl methacrylate), poly (ethylene oxide) and its copolymers, and PVA (poly vinyl alcohol). Natural hydrogels, which are usually a biopolymer, include alginate, collagen, gelatin, fibrin, chitosan, agarose, and hyaluronate [54]. The structural integrity of hydrogels depends on the

28 6 crosslinking of the polymer molecules that could be bonded physically or chemically [55]. Hydrogels that are crosslinked physically may involve molecular entanglements or secondary forces such as ionic, H-bonding, and hydrophobic forces. Chemically crosslinked hydrogels are generally covalently bonded. They may also be formed via methods that make them biodegradable or nonbiodegradable. The mechanical properties of hydrogels can be described by the theories of rubber elasticity and viscoelasticity [56, 57]. The properties of hydrogels have made them attractive in the medical field such as in drug delivery and controlled release [58-62]. Tissue engineering seeks to fabricate and regenerate tissue constructs. Numerous strategies have been developed to successfully engineer functional tissue. As an alternative to hydrophobic scaffold materials that may go under severe conditions, hydrogels offer an entreating opportunity to seed living cells and other biological species during fabrication process of scaffolds. In tissue engineering applications, hydrogels are preferred to be biodegradable so that encapsulated cells may proliferate and form a tissue construct that eventually takes place of the initial hydrogel scaffold [63-65]. 1.3.Alginate Hydrogel Alginate as a Biopolymer Alginates are naturally derived polysaccharides that are derived primarily from seaweed. They are composed of (1-4) linked β-d- mannuronic acid (M units) and α-l-guluronic acid (G units) monomers along the polymer backbone.

29 7 The M and G units may vary in ratio depending on the source of the alginate [14]. The alginate molecule is considered to be a block copolymer with regions of M and G units that both have carboxylic groups. The carboxylic groups are capable of forming salt formations such as sodium alginate, where the sodium monovalent ions are attached ionically to the carboxylic groups as shown in Figure 1.1. Figure 1.1: Schematic diagram of β-d- mannuronic acid (M units) and α-lguluronic acid (G units), and sodium alginate

30 8 Sodium alginate is soluble in water and when dissolved forms a viscous solution that could vary depending on the concentration and molecular weight of the biopolymer. In the presence of divalent calcium ions, sodium alginate solution is crosslinked ionically between chains to form a hydrogel. The calcium ions exchange with sodium ions on the G blocks and binds together adjacent chains that causes the gelation of alginate. It has been reported that ionically crosslinked alginates lose mechanical properties in vitro over time. It is presumed that this phenomenon occurs due to an outward flux of crosslinking ions to the surrounding medium [66]. A study by Wang et al. was conducted to investigate the ability of rat bone marrow cells to proliferate and differentiate on alginates of differing composition and purity. It was found that high purity and high G-type alginate retained 27% of its initial strength after 12 days in culture [67]. The formation of alginate gels may take place in two known methods; diffusion setting and internal gelation. In the first, alginate solution such as sodium alginate is deposited directly in to a reservoir of divalent ions such as calcium chloride. The gelation process takes place as the calcium ions gel the outer most layer of sodium alginate and then diffuse into the core of the material to gel internally. In the later method, the process is vise versa where the calcium chloride solution is introduced into a reservoir of sodium alginate. The gelation in this method begins at the outer surface of the calcium chloride solution and then proceeds into the sodium alginate outer body. Both gelation methods have been used for various applications. Figure 1.2 presents schematic diagrams of the both methods.

31 9 Figure 1.2: Schematic diagram of diffusion setting and internal gelation of alginate Alginate in Tissue Engineering Tissue engineering biomaterials should fulfill several preconditions such as high level of biocompatibility and biodegradability. These are also required to have low degree of cytotoxicity, high affinity to biological surfaces, mechanical

32 10 stability to the construct, and serve as a guide for 3D tissue regeneration. Alginate is a nontoxic immunologically inert hydrogel and therefore has been widely used for scaffold material for immobilization of enzymes or cells for bioreactors, and also for tissue engineering [68]. Alginate has been mainly utilized in tissue engineering as a delivery vehicle or supporting matrix cell via encapsulation techniques [69-75]. Alginate-based microencapsulation is currently a favored approach because animal studies and small scale clinical trials have shown that the requirements for long-term immunoisolation and simultaneous maintenance of transplant function can most likely be fulfilled by this hydrogel [76]. Many groups have demonstrated various techniques for producing alginate beads of controlled dimensions for cell encapsulation [77]. The use of alginate in cell encapsulation requires the gel to be of high purity to exploit cytotoxic and apoptosis-inducing impurities. The purification and sterilization processes are time consuming labor work that is accomplished by filtration techniques [78]. Alginate degradation is not carried out by mammalian cell digestion, instead, divalent calcium ions slowly diffused out of the hydrogel allowing alginate molecules to be excreted in the urine [79]. Alginate has been used extensively in the culturing of chondrocytes, as well as for hepatocytes and Schwann cells for nerve regeneration [50, 80-87]. One of the drawbacks of alginate is that there is no specific interaction between mammalian cells alginate gel. In addition, a negative charge balance exists in alginate gels that inhibit protein absorption due to electrostatic repulsion. For this reason, many studies have modified alginate with peptides for

33 11 cell adhesion. The carboxylic acid on alginate makes it attractive for modification such as in peptide attachments [88]. Genes et al. characterized the attachment of chondrocytes to RGD-functionalized alginate by examining the effect of substrate stiffness on cell attachment and morphology [89]. Their study showed that the rate of cell attachment density increased with crosslink density and substrate stiffness. In another study by Kreeger et al., the roles of peptide density in alginate on murine granulose cell adhesion, morphology, and steroid secretion were conducted [90]. Their results showed that murine granusola cells attached and spread on the modified surfaces with morphologies specific to the peptide identity and density. Alginate has also been mixed with other materials to enhance its biological performance such as with chitosan [91, 92]. Chitosan is a polysaccharide and is prepared by N-deacetylation of chitin that is naturally derived from marine crustaceans [93-95]. The degree of crystallinity of chitosan is a function of the degree of deacetylation [13]. The highest degree of crystallinity for chitin and chitosan are at 0% and 100% deacetylation, respectively. Therefore, minimum degrees of crystallinity are found at intermediate degrees of deacetylation. Chitosan has found many biomedical applications, including tissue engineering approaches, due to its biocompatibility, low toxicity, structural similarity to natural glycosaminoglycans, and degradation by enzymes such as chitosanase and lysozyme. Chitosan is generally insoluble in neutral conditions as well as in most organic solvents due to the existence of amino groups and high crystallinity. Chitosan forms hydrogels by ionic or chemical cross-linking with

34 12 glutaraldehyde and other mediums [96]. Azide-derivatized chitosan was also reported to form gels by UV irradiation [97]. Various derivatives have been developed to alter the biological functions of chitosan, including enhancement of cellular interactions for tissue engineering approaches. Ribeiro et al. presented work on encapsulating proteins into alginate microspheres by internal gelation and were then subjected to chitosan coating [98]. The results suggested an optimization of the coating method to protect proteins and permit a sustained release. Alginate has also been blended with hydroxyapatite to form osteoconductive composites [99-102]. Tampieri et al. performed in vitro tests on different morphology and thermal stability of hydroxyapatite/alginate ratios and were then cultured for seven days with MG63 osteoblast-like cells [103]. Their scaffolds data showed that the tested materials favored cell growth and maintained their osteoblastic functionality. Another research effort by Miyamoto et al. focused on the evaluation of non-decay type fast-setting calcium phosphate cement (nd-fscps) by adding amounts of sodium alginate to the liquid form of bone cement [104]. Cement pastes were immersed in serum at 37 o C prior to mechanical testing and were also implanted into rat subcutaneous tissue for the initial evaluation of biocompatibility. Their results showed that the (nd-fscps) transforms to hydroxyapatite increasing its mechanical properties.

35 Mechanical Properties of Alginate The mechanical properties of alginate can be controlled through the molecular weight, gelation rate, type of crosslink, alginate concentration, crosslink concentration, and Mannuronic acid/guluronic acid ratio [105]. Studies have shown that the mechanical properties of alginate attribute to cell function, locomotion, and morphology [89]. Alginates have been crosslinked with calcium sulfate, barium chloride, calcium chloride, and calcium carbonate with each having a specific gelation rate. As an example, calcium sulfate kinetics is difficult to control resulting non-uniform gel structures. By contrast, calcium chloride provides alginate with a relatively fast gelation rate that result in a crosslink density and a polymer concentration gradient within gel beads [106]. Many researchers have studied the mechanical behavior of alginate under various conditions [ ]. Kuo et al. have reported the use of slow-gelling calcium carbonate-gdl and calcium carbonate-gdl-calcium sulfate systems that form structurally uniform gels for tissue engineering [114]. They have also demonstrated how to control the structural integrity, mechanical properties, and cell incorporation uniformity. Slower gelation systems generate more uniform and mechanically stronger gels than faster gelation systems. Their results also showed that the compressive modulus and strength increased with alginate concentration, total calcium content, molecular weight and Guluronic acid content of the alginate. In some cases it has been observed that the elastic modulus of alginate decreases imidiately after incubation. Awad et al. have conducted

36 14 experiments to study the mechanical properties of alginate seeded with Human adipose derived adult stem (hadas) cells [115]. Their alginate elastic modulus decreased between days 0 and 14 suggesting the hydrogel have weakened due to a loss of calcium ion. However, the elastic modulus increased between days 14 and 28 that is likely due to the increase of cell matrix. However, it should be noted that the elastic modulus of the alginate samples were taken only on three days (0, 14, and 28). One may argue that the alginate may have experienced increased elastic modulus values a few days after incubation since the gelation time for the samples were only 10 minutes and then experienced degradation. The degradation of alginate can be monitored by the mechanical properties. One group investigated whether alginate gel degradation could be controlled by combining partial oxidation of polymer chains prior to gel formation [116]. This was conducted by reacting alginate with sodium periodate that cleaves the carbon-carbon bond of the cis-diol group in the urinate residue and the aldehyde groups form six-membered hemiacetal rings with the closest carboxylic groups on the alginate. Myoblasts were performed to ensure the change did not severely decrease the alginate biocompatibility. Their results showed that the myoblasts adhered, proliferated, and differentiated on the modified gels at a comparable rate to those cultured on unmodified alginate gels. The mechanical properties of alginate are not directly responsible for encapsulated cell proliferation since the internal crosslink structure is greatly involved in the gel structure. Simpson et al. postulated that the mechanical properties of alginate were not responsible for cell metabolic reactions in their

37 15 research but rather, changes in the strength of the alginate gel network caused by changes in the number of alginate strands held together in the egg-box model [117]. Their data show that increasing the concentration of the calcium chloride solution used at the time of gelation, thus increasing the strength of the alginate gel network, impedes the growth characteristics of βtc3 cells encapsulated in a high Guluronic acid content alginate. However, preparations of βtc3 cells encapsulated in an alginate with high Mannuronic acid content are not affected by changes in calcium chloride concentration due to the low percentage of consecutive Guluronic acid residues. 1.4.Polymer Deposition Polymer dispositions have been studied in many industries such as the petroleum, food processing, manufacturing, and pharmaceutical [118, 119]. These studies have been conducted to improve the efficiency and accuracy of processes in a technical and economic manner [ ]. Understanding the deposition of polymers studies through nozzles and syringes may lead to proper fabrication methods and conditions for building tissue scaffolds with specified architecturally designed matrices using the biopolymeric materials. Although a lot of work has been done up to date, many research groups are filling the gaps to better understand the controllability of polymer depositions. Christani et al. studied the general effects of viscoelasticity and the specific effect of added polymers on the break up, which were not well understood [123]. Their work focused on the probing the effects of extensional viscosity on jet break up of polymer solutions

38 16 using a series of model for viscoelastic fluids that have already been developed and found the droplet size is function of molecular weight and the polymer viscosity that the applications for polymer depositions may require individual study for each system in order to have complete control over the product geometry accuracy. Material recirculation in a system that deposits polymer solution has an effect on the droplet size. The polymer chains are subjected to high stresses especially at the middle of the chain when subjected to high shear rates that could cause molecular scissioning. As a consequence, the viscosity of the polymer solution will decrease since molecular weight is a function of the viscosity and the droplet size will decrease. In the field of agriculture, polymer depositions have been used for reducing the off-target drift of pesticides and improve its efficiency and performance on the intended targets. One way of increasing the mean droplet size to reduce spray drift away from the intended target is to introduce drift retardants in spray solutions. Zhu et al. presented work in this area on twelve polymers by testing them to determine shearing effect on physical properties and sprayed droplet size distributions [124]. Some of the polymers tested were polyethylene oxides, polyacrylamides, and a polysaccharide, with a range of molecular weights and concentrations. The group found that as the number of recirculations of the anionic and non-ionic polymer in the system increased, the droplet size decreased. Similarly, the droplet size increased as the molecular weight of the polymers decreased. It was concluded that recirculation of solutions containing water and polymers progressively decreased spray droplet size distribution and low and high

39 17 shear rate viscosities. Such studies of polymer deposition under varying parameters are essential for improving accuracy and process efficiency. The procedures require precise control of particle size and size distribution. Particles in the nano-range are useful in the applications of controlled release formulations as carriers of DNA in gene therapy and ability to deliver vaccines through peroral route of administration [125]. Some of the advanced drugs include protein and peptide therapeutics, which cannot be administered orally. One way of delivering them is by encapsulating them in microspheres made of biopolymeric materials such as PLGA. There are several ways of fabricating microspheres such as spraying and phase separation for the encapsulation of proteins in biodegradable polymeric devices, from which a drug could be delivered locally or systematically for a prolong period of time. Berkland et al. presented work on their developed methodology based on spraying polymer solution through a small orifice for fabricating polymeric microspheres [126]. The group used 50:50 Poly(D,L-lactide-co-glycolide) dissolved in DCM that was pumped through a small-gauge needle ranging from 30 to 100 microns at a flow rate of to 5 ml/min while an ultrasonic transducer controlled by a frequency generator distributed the stream into uniform droplets. The microspheres were then extracted by evaporating the DCM and then centrifuged and placed in purified water. The process was monitored using a camera to analyze the microspheres. The nozzle used was made of a piezoelectric material that vibrates using a transducer that is driven by a wave generator at a frequency to produce the desired flow rate and droplet size. The droplet size could also be controlled

40 18 employing an annular flow of a non-solvent phase around the polymer jet. The velocity of the non-flow solvent is greater to that of the polymer stream, which put a downward force on the deposited polymer solution away from the tip of the nozzle. In general, the droplet size increased with increase in flow rate and decreased with the increase of acoustic frequency. This could be a very useful method of depositing polymer droplets of sizes that are almost equal to the nozzle diameter for applications where nozzle diameters cannot be reduced since the droplets without acoustic frequencies tend to have sizes much greater than the nozzle diameter. The system has many advantages, such as avoided nozzle clogging, unwanted microsphere sizes elimination, broad size range, and precise and instantaneous control of microsphere diameter. However, its disadvantages are increasing polydispersity for microspheres less than 1 µm and more complex apparatus and control system are required for the processing. Supercritical fluid precipitation is a method also known for polymer deposition for drug delivery and controlled release applications. The process involves dissolving a solute in an organic solvent, which is the injected through a supercritical fluid anti solvent. By adjusting the temperature and pressure, the physical properties of the supercritical liquid such as density, viscosity, and diffusivity can be controlled. At specific parameters, the diffusion rate between the solvent and the supercritical liquid becomes very high that supersaturation occurs, which leads to solute precipitation. Carbon dioxide is generally used as the compressed fluid anti solvent for the precipitation process. Jarmer et al. used the method of producing small particle size polymers by using supercritical fluid

41 19 precipitation technologies that uses an antisolvent in the process [119]. The objective was to employ nozzle designs similar to those used in the gas combustion and propulsion fields and applying it to precipitation with a compressed fluid antisolvent for producing nano-scale poly(l-lactic acid) PLLA particles. It was found that the jet mixing length and average PLLA particle diameter were both functions of the power input in the swirl chamber that was designed. Furthermore, the design combined with this method proved to produce nano-scale polymeric particles with a smaller average particle diameter and a sharper size distribution than conventional designs. One of the limitations of this nozzle design is the lack of control over the jets characteristics, which results in non-uniform conditions in the nucleating medium, and therefore little control over the resulting particle size and size distribution. This process is capable of producing particles in the nano-range, however, large powder input is required because mixing configurations have to be designed to generate and compress turbulent length scales to extremely small sizes in order to produce a large, homogeneous supersaturation. Another way of controlling the droplet size of polymer solutions is to apply an electric potential between the polymer solution and the collecting solution during dropwise extrusion from the nozzle. It is known that the droplet size decrease gradually until the applied electric potential reaches a critical level [127]. Any further increase in the droplet size causes unstable jets. Poncelet et al. presented work in this area to evaluate the observed effects of physico-chemical

42 20 parameters on the dispersion process of sodium alginate in to calcium chloride bath under an applied electric potential. For any nozzle, the polymer solution flows out of the tip of the nozzle forming a droplet that continuously grows until it reaches a critical mass where it is then detached from the nozzle tip. In the absence of an electric potential, the critical mass can be defined by the equilibrium between the gravitational force and the polymer solution surface tension as can be observed in Equation (1.1). 3 πdo mg = ρg = πd γ 6 s (1.1) Where m is the mass, g is the gravitational constant, d o is the diameter of the falling droplet, ρ is the density of the polymer solution, d s is the diameter of the droplet neck, and γ is the surface tension of the polymer solution, which is usually 40 mj/m 2 for polymer aqueous solutions. d s is approximately equal to the external diameter of the nozzle tip. Equation (1.1) can be rearranged to show that d o is a function of the nozzle tip diameter, polymer solution surface tension and density as shown in Equation (1.2). d o 6d γ s = 3 (1.2) ρg When an electric potential is added to the system then the mechanical equilibrium of a droplet can be expressed as shown in Equation (1.3).

43 21 πd 6 3 ρg + F e = πd s γ (1.3) Where F e is the electric force that depends on the geometry of the system. In the case of a single nozzle system and a collecting solution bath can be expressed by Equation d Fe = πε o U (1.4) 2 h Where ε o is the electric permitivity of air, h is the distance between the droplet and plane electrode, and U is the applied electric potential. Equation (1.4) is valid for a droplet that is deposited into a bath solution beaker that has its diameter three times the distance between the liquid and the drop h. Poncelet et al. data had a poor fit to Equation (1.4), which directed them into modeling their experimental data accordingly. Their data showed that diameter of the droplet decreases faster than the expected. In addition their data showed that the drop is independent of the distance between the droplet and collecting solution h. For this reason, a model was designed that allowed the simulation of more accurate results from experimental data as shown in Equations (1.5) and (1.6). mg 3 2 πd U = ρg = πd sγ U c (1.5)

44 22 U c d γ c o (1.6) ε o Where U c is the critical electrostatic potential. Other models were also developed that expressed the effect of droplet polarity on the droplet size, the effect of the flow rate on the droplet size, and the effect of viscosity on droplet size. 1.5.Freeform Fabrication using Polymers In the area of Solid Freeform Fabrication or Rapid Prototyping, the fabricated 3-dimensional structures are built by reducing CAD designs of particular prototypes into a group of sliced 2-dimentional layers, to where the prototyping material is deposited to build the final structure in a layer-by-layer process. Rapid prototyping has been used in many research studies, such as in composite materials. In the area of freeform fabrication, polymers have been widely used as the material to fabricate freeformed structures [128, 129]. Lombardi et al. explored methods to extend the materials processing capabilities of RP technology to tough and strong functional prototypes made from linear engineering thermoplastics by direct polymerization of monomer in layers using the 3D plotting method [130]. The work was focused on developing a flexible route to the characterization of polymers and blends made by concurrent polymerization and solidification rather than by conventional polymerization followed by processing. One of the major advantages of freeforming liquid

45 23 monomers is that they have low viscosity, which facilitates accurate dispensing and do not exhibit many of the viscoelastic and rheological problems when compared to a molten engineering thermoplastics. The freeforming procedure was done using an Asymtec Automotive. The results showed that the extrusion from the freeformed bars had only slightly lower strengths and elongation to break values than the commercial Nylon 6 values. The experiments proved that freeform methods could be used for fabricating designed features with various polymer processing techniques. Hydrogels have also been investigated in freeform process for polymeric deposition. Calvert et al. reported the preparation and swelling behavior of multilayer acrylic acid and acrylamide gels [131]. They reported that shapes could readily be freeformed from 2-6% agarose gels by writing a hot solution onto a cold substrate, which rapidly solidifies upon cooling. Similarly, this was done for cross-linked polyacrylamide and polyacrylic acid gels, which were freeformed by writing cool solutions of monomers, cross-linking agent, and catalyst onto a hot plate, which induced the polymerization process. The group also showed that multilayers stacks of cross-linked polyacrylic acid and polyacrylamide hydrogels swell differently from what would be anticipated based on the behavior of the materials taken separately. For instance, it was found that a stack with only one layer of polyacrylic acid and five polyacrylamide swells to half the extent of six layers of polyacrylic acid. This work provides evidence that freeform fabrication of polymer complex structures such as hydrogels is an applicable procedure. This

46 24 fact clearly defines the capabilities of freeform fabrication on producing objects with in a wide range of material selection. 1.6.Freeform Fabrication in Tissue Engineering It has been acknowledged that scaffolds require adequate mechanical strength to support and maintain the porous matrix that is required for the culturing cells in the early stages of incubation or implantation. The mechanical stiffness of the scaffold should be within the range of the implant area [132]. In order to fabricate scaffolds with specific properties, a fabrication method that maintains a high level of accuracy is necessary to maintain the consistency and repeatability in accordance to the intended design. There are several conventional scaffold fabrication techniques that have been introduced in the past few years such as fiber bonding, solvent casting, and particulate leaching, melt molding, gas foaming, and soft lithography [128, 133]. One of the main drawbacks of these techniques is that they hold restrictions on the shape control and the consistency of internal architecture in the designed tissue scaffolds. Unlike the conventional fabrication techniques, solid freeform fabrication (SFF) has no restriction to shape, control, and consistency [134, 135]. SFF are computerized fabrication techniques that can rapidly produce highly complex three-dimensional objects using data from CAD systems or imaging model reconstructed from computer medical imaging data such as MRI and CT. The fabricated three-dimensional structures are built by reducing CAD designs of particular prototypes into a group of sliced 2-dimentional layers, to where the prototyping material is deposited to

47 25 build the final structure in a layer-by-layer process. In applying SFF technique for scaffold fabrication, scaffolds could be designed in CAD systems first and then freeform fabricated using various fabrication methodologies. There are various freeform fabrication techniques such as fused deposition modeling (FDM), laminated object \manufacturing (LOM), three-dimensional printing (3DP) multiphase jet solidification (MJS), stereolithography apparatus (SLA), selective laser sintering (SLS), and 3D deposition (3DD) [ ]. The 3DD technique delivers liquid materials through micro-volve under pressure. The shape of fabricated structure could be controlled through the material properties, process parameters such as pressure and nozzle diameter, and the movement of the nozzle that is controlled via a computer controlled system. Hydrogels based on both natural and synthetic polymers have been of interest for encapsulation of cells and most recently, have become especially attractive to tissue engineering as matrices [54, 55]. They may be chemically stable or they may be also degradable and dissolvable in the body. Hydrogels can be seen as a network of molecular chains that are attached together by covalent bonds, in which it is called a permanent hydrogel. Physical hydrogels are those that have ionic, H-bonding, or molecular entanglement that form the intermolecular network. Hydrogels that are used for tissue engineering may have pores within the inter-molecular network large enough to accommodate or to be designed to release growth factors and nutrients to the surrounding cells. Hydrogels have also been used as a drug delivery medium because their permeability for solutes can be manipulated and incorporated into tissue scaffolds

48 26 [139]. A hydrogel that has an intermolecular network at the microscale may also form into a porous tissue scaffold at the macroscale by aligning the hydrogel components to create voids. To illustrate this further, a hydrogel scaffold that has been fabricated by SFF process may have a designed macro-porous within the scaffold by controlled placement of hydrogel material in the construction of scaffold, it also has an, inherent micro-porous network at the molecular level due to the crosslink bonds between molecules, this mulit-scale porous network is a unique characteristics of hydrogel-based tissue scaffolds. In vitro regeneration of human tissue requires specific aspects to scaffold properties that should be met. These aspects are not only important for cell survival, signaling, and growth, but also for cell shape modeling and gene expressions [140]. A three dimensional scaffold is required to mimic the physiological functions of the native cellular matrix. An ideal scaffold design would promote cellular growth and proliferation that would then take place of the degrading scaffold. There has been a general consensus, that tissue scaffolds should be made from a biodegradable biomaterial that does not cause inflammatory reactions [141], [9]. In general, polymers have been used as biomaterials for tissue engineering applications since many of them are biocompatible and have unique processing techniques. Scaffold based strategies from matrix-producing connective tissue cells such as osteocytes and chondrocytes have been the major approach for successful tissue regeneration. The scaffolds that could be made of natural or synthetic materials serve a purpose of being a temporary surrogate native extracellular matrix [15]. Since complex

49 27 organs such as livers, hearts, and neural tissue, make tissue engineering very difficult because of their specific three-dimensional cell distribution, tissue scaffolds must be constructed with specific pore sizes and internal architecture to accommodate and allow the proliferation of cells. The fabrication of such complex constructions has been tackled using computer-controlled techniques known as SFF or rapid prototyping (RP), which have unique advantages over conventional fabrication methods. Tissue scaffolds can be used to either promote natural regeneration, in vivo, or be used to create a bioartificial organ, in vitro. The makeup of the extra-cellular matrix (ECM), pore size, biodegradability, cell adhesion, growth factors, immunological response, cytotoxicity, and the required vasculature of the tissue and scaffold all play an important role in the design and fabrication of the scaffold [20]. Two types of strategies are utilized in developing scaffolds. In the first strategy, the scaffold has to provide support in vivo. In the second strategy, the scaffold only provides support in vitro until the cells are strong enough to support themselves in vivo [142]. Biomimetic modeling and design can address some of the above concerns. Application of SFF technology in tissue engineering constitutes an important component in Computer-Aided Tissue Engineering (CATE). There are a number of different materials, techniques, and processes that are being explored and combined in the fabrication of scaffolds using SFF techniques [15], [143]. Although many applications of using RP techniques in biomedicine and tissue engineering have been reported, SFF technology for manufacturing tissue engineered constructs still appears to be in a relatively early

50 28 stage. Among available techniques, 3DP and FDM seem to be the most promising process to tissue engineering because of the versatility of using scaffolding materials, possibility to make scaffold feasibility of controlled deposition with high precision and at rates much higher than comparable methods such as indirect mold fabrication [15, 16]. Due to these advantages, new developments on SFF to tissue engineering fabrication are mostly the variations of these two processes, for example, 3DP-based TheriForm fabrication and Ink-Jet 3D-plotting, and FDMbased precision extruding and micro-nozzle/micro-syringe deposition. The TheriForm 3DP-based fabrication process is a technique that has licensed the technology developed at MIT. Therics Incorporated has applied the 3DP TheriForm TM process to tissue engineering. Various experiments have been reported using this process: for example, Zeltinger et al. created poly (L-lactic acid) scaffolds for testing the effects of porosity on cellular growth in vitro [19]. Fibroblasts, smooth muscle cells, and epithelial cells were seeded onto the scaffolds. Their process involved salt leaching to create the pores in the scaffold. Kim, et al. created a scaffold from polylactide-coglycolide (PLGA) that contained interconnected scaffolds to improve circulation [144]. Sachlos et al. used 3D printing for the fabrication of molds that were then filled with collagen for fabricating tissue scaffolds [134]. Lam et al. reported using 3DP with starch polymers for scaffold fabrication [133]. A mixture of cornstarch, dextran, and gelatin were used to create cylindrical scaffolds using a Zcorp 3D printer (Z402). The powder blend consisted of 50 wt.% corn starch, 30 wt.% dextran, and 20 wt.% gelatin. Distilled

51 29 water mixed with blue dye was used as the binder with the intention of not using organic solvents. Scaffolds were designed on CAD software then sliced using a slicing algorithm. Post-processing was done to increase the strength of the scaffolds by drying at 100 o C for 1 hour. Afterwards, tests were conducted to examine their material properties such as shrinkage, water absorption, porosity, and mechanical properties. Porous scaffolds exhibited compression stiffness of to MPa for the various different designs that were reported. The aim of the group was to explore the feasibility of using 3D printing in combination with natural polymers and water-based binders. The water absorption and mechanical tests showed that 3D scaffolds created by a new blend of materials were achievable. Although the study proved the capability of repeating porous unit cells, more study is needed in the future to evaluate the biocompatibility and in vitro capabilities for cell regeneration. FDM technology has been extensively studied by Hutmacher s group with poly(ε-caprolactone) (PCL) filaments to create bioresorbable scaffolds [ ]. Testing was done on its material properties related to anisotropy and porosity. Channel sizes ranged from µm. Compressive stiffness ranged from 4 77 MPa. Different layer patterns were used such as square and honeycombed patterns. The FDM scaffolds were also seeded with fibroblast cells and were successful in creating cell growth and proliferation. Cells initially started at the junction points of the scaffold, grew along the rods, began filling up the pores, and then finally by week 4, the entire scaffold was filled with cells. Seeding with osteoblast-like cells also showed a similar migration pattern. Another FDM

52 30 micro-nozzle based layered manufacturing technique, also referred to as Precise Extrusion Manufacturing (PEM) was developed for tissue scaffold fabrication [148]. Using PEM, thermoplastic material (PLLA) was deposited through a heated, compressed air sprayer. The nozzle diameter of the sprayer was about 0.3 mm. The liquefier was heated and kept at 160 C during the process. They were able to create controlled pore sizes of μm in size. Their machine used a CAD model and used NC technology for manufacturing the scaffold. Similar studies to construct cellular PCL scaffold fabricated by the author s in-house developed multi-nozzle freeform deposition system were also reported. The SEM characterization shows that thus fabricated scaffold micro-architectures could be achieved at about 250 μm scale level with excellent uniformity of the fill gaps, the depositing struts, and the internal pore connectivity. Wang et al. have also produced similar results using PCL [149, 150]. One of the freeform fabrication methods of building tissue scaffolds is the nozzle or syringe based technique known as 3DD, which involves the deposition of extruded parts or micro-droplets of material solutions. 3DD has two clear advantages over conventional SFF methods that make it attractive for tissue engineering applications. First, the deposition technique permits the use of a wide range of biomaterials that have fluid properties, such as thermoplastics, pastes, and hydrogels. Second, the deposition can take place in a sterile environment and at room temperature. This means that it can apply unstable biomaterials as well thermally sensitive biomaterials since it does not require any heating in the process. This also indicates that critical biocomponents such as cell growth factors

53 31 or even cells may be incorporated into the scaffold fabrication process. The parameters that are used in the 3DD processes may vary depending on the system that is utilized. These parameters however influence the droplet or extrusion diameters that are deposited from the nozzle tip. The deposition systems that can be used for 3DD of scaffolds involve pneumatic valves, piezoelectric valves, and solenoid valves. Each system has its advantages and limitations over the others. An ideal selection would depend on the desired thickness of the scaffold strand and biomaterial that is utilized. The 3DD method has been attractive for building tissue scaffolds for two clear reasons. First, this method can apply a wide range of polymeric materials such as thermoplastics, pastes, and hydrogels. This makes the solidification process controllable, which could affect scaffold properties. Second, it can apply reactive materials as well thermally sensitive biomaterial since it does not require any heating in the process. This also means that critical biocomponents such as cell growth factors or even cells may be incorporated into the scaffold fabrication process. Ang et al. presented work on using a 3D plotting technique for the fabrication of chitosan-hydroxiapatite scaffolds [41]. Chitosan is a naturally occurring amino-polysaccharide that is known for it biocompatibility, biodegradability, and non-toxicity. It insoluble in aqueous solutions above a ph of 7 and becomes soluble in dilute acids. The group investigated the design and fabrication of 3D chitosan and chitosan-hydroxiapatite scaffolds and its feasibility in tissue engineering applications. The fabrication process included studying the viscous solutions of chitosan and chitosan- hydroxiapatite extrusion into a bath

54 32 mixture of sodium hydroxide solution and ethanol to form a hydrogel like precipitate. The diameter of the nozzle was 150 μm. The group first started experiments to understand the influence of dispensing pressure (i.e. 44, 58, and 73 psi), initial height of the nozzle 0.1,0.2,0.3, and 0.4 mm), and the speed of the nozzle in the two-dimensional horizontal direction (3, 6, and 9 mm/s). Preliminary in vitro studies were also conducted after the fabrication and sterilization processes. Their results concluded that a dispensing of 44 psi, speed of 6 mm/s, and initial height of 0.2 mm were optimal for fabricating pure chitosan scaffolds. In addition, the optimal sodium hydroxide concentration was found to be between 0.75% and 1.5% v/v. It was also observed that rapid precipitation occurred when the sodium hydroxide concentration was relatively high. At that point, the attachments between layers were unachievable. This suggests that there are critical concentrations for sodium hydroxide that could allow the precipitation process to occur at a relatively high rate but simultaneously relatively slow enough to achieve sufficient adhesion between two-dimensional layers. This phenomenon may be a key issue in fabricating three-dimensional hydrogels by 3D plotting that as a single connected structure. Similar to the chitosan dispensing, Landers et al. reported experiments on fabricating tissue scaffolds form thermoreversible hydrogels using 3D plotting freeform fabrication [40]. Gelatin and agarose were the two hydrogels that were used for their experiments. The phs of the aqueous solutions of both materials were adjusted to 7 by using sodium hydroxide and HEPES buffer solutions. One of the main problems the group encountered was the low viscosity of the aqueous

55 33 polymer solution before gelation. The low viscosity solutions can cause collapse of the three-dimensional structure due to gravitational forces. For this reason, the 3D plotting was performed in a liquid medium. The liquid medium had approximately the same density as the plotting material to compensate the gravitational forces. Bonding of the strands was done by interdiffusion, which means that early gelation prior to contact between the strands must be prevented. To provide better interdiffusion, a double jacket cartridge was used to set the temperature of the plotting medium significantly higher with respect to gelation temperature to delay the gelation process for a few seconds. To adjust this critical balance of strand bonding, the amount of deposited material and nozzle speed should be adjusted. The group concluded that the 3D plotting process depends on various on various parameters such as the ratio of gelation temperature and plotting temperature, nozzle type, densities of the deposited material and liquid medium. The control of the swelling can be controlled by the amount of salts dissolved in the liquid solution. It can be understood that the hydrophilicity of the hydrogel is prime role information of three-dimensional structures, as well as the bonding time and method to maintain a single connected scaffold. In earlier studies, it was appointed that the most ascertained technique for obtaining smooth, even, and continuous deposited polymer solutions was to apply a low and constant pressure and that the polymer be dragged along the surface of the substrate. It was also estimated that the optimal viscosity range for polymer deposition to be with in 100 and 700 cp. At very low viscosities, the polymer could leak out of the tip and at high viscosities, the polymer may require high

56 34 pressure that could damage the syringe tip. Vozzi et al. presented work on a novel method for deposition of biopolymers in high-resolution structures, using a pressure-activated syringe [151]. This system not only has potential applications in tissue engineering, but also to study cell motility, organization, and cell reaction to various topographies. Their work was focused on the deposition system and on the patterns produced or two polymers, PLLA and PCL solutions. Both polymers were dissolved in chloroform to give 1, 2, and 3 % (w/v) PLLA solutions and 10, 15, and 20% (w/v) solutions of PCL. The Syringe that was used was made of stainless steel with a 20-micron glass capillary needle tip and was driven by filtered compressed air at a pressure of about mmhg. As pressure was applied to the syringe, tiny amounts of polymer oozed out through the tip. When the needle was too far from the surface, the polymer solidified at the tip and prevented further deposition. However, when the needle was just above the substrate to a height that did not allow the polymer to break out of the tip, continuous fine lines of polymer were traced on the substrate. The optimum height of the tip was not reported, however, it was estimated that it would be on the order of the tip diameter. The system used by Vozzi et al. included valves, pressure sensors, and position controllers interfaced with appropriate C language software. The system allowed the pressure to be controlled with in the range of 10 to 1000 mmhg and the velocity of the substrate with respect to the syringe varied between 0.5 and 2.5 mm/s. Some models can be used to predict the line width and the height of the deposited polymers. It can be approximated that the deposition occurs as a simple dynamic model that is based

57 35 on the tip of the nozzle. It is assumed that the dimension of the polymer does not change from the evaporation of the solvent and that the forces that play are the driving pressure, surface tension between the polymer solution and air, and the dynamic friction between the fluid and nozzle material that is a function of the polymer solution viscosity. In general, it can be seen that any type of polymer or polymer blends could be used for microsyringe deposition. The concentration of solutions, nozzle diameter, solution viscosity and surface tension, and applied pressure are the parameters to be adjusted to produce deposited polymer lines with the desired geometry. The group presented simple models for calculating the deposited polymer width. Equation (1.7) can express the flow rate exiting the needle tip; dl Q = ah = ahv 0 (1.7) dt where a is the line width and h is the height of the polymer pattern. l is the length of polymer deposited in time t, and hence v 0 is the velocity of the substrate with respect to the syringe. If the polymer is considered to be a Newtonian liquid, the flow in the capillary is given by Poiseulle s equation; 4 πr dp Q = (1.8) 8η dz

58 36 where r is the internal radius of the capillary, η is the solution viscosity, and dp/dz is the applied pressure gradient. Combining Equations (1.7) and (1.8) together, 4 πr dp a = (1.9) 8ηv h dz 0 A certain critical pressure P c must be applied to the before the fluid exits the needle tip proportional to the viscosity of the solution. Below P c deposition cannot occur because the friction forces are greater than the driving pressure. The pressure gradient is negligible in the widest part of the syringe, and is maximum at the tapered region of the tip. In their model, dp/dz has been approximated to (P+P c )/h z, where (P+P c ) is the applied driving pressure and h z is the length of the tapered region of the capillary. Therefore, Equation (1.9) was expressed as; πr a = 8ηv 4 0 h ( P + P ) h z c (1.10) Vozzi et al. also performed some designs using freeform fabrication for fabricating tissue scaffolds [52, 53]. The aim was to explore two techniques for controlling PLGA architecture at the microscale, microsyringe deposition, which is similar to 3D plotting, and soft lithography micromodeling. The microsyringe system incorporated a thermoregulator into the horizontal motors for temperature control to modulate the solvent evaporation. The nozzle diameter was 10 to 20

59 37 μm o.d. capillary and the PLGA was dissolved into chloroform with a concentration of 12% (w/v). A wide range of dimensions from 10 to 300 μm were able to be fabricated by varying the pressure, nozzle motion speed, and solution viscosity. The group reported that solutions of up to 200 cp were necessary in order to achieve high resolution structures. This suggests that relatively high pressures may be needed with small tip nozzles in order to deposit polymers with high viscosities, which can clearly be recognized when looking into the Poiseulle s equation as shown in Equation (1.8). The group also reported that at polymeric solutions of viscosities higher than 400 cp demand high pressures to extrude the liquid that may break the tip.10 μm structures were achievable by depositing 20% (w/v) PLGA/chloroform solution at a pressure of approximately 20 psi and using the 20 μm. The group was also able to fabricate the tissue scaffold with internal microstructure. This was done by adding glucose grains to the PLGA solution and then removed out of the scaffold was fabricated by leaching deionized water. The optimum concentration of the polymer solution for this process was found to be between 5-10%. The solution was then mixed with the glucose particles (20-65 μm) to a weight ratio of 1:1. Yan et al. used a Multi-nozzle Deposition Manufacturing (MDM) as a freeform fabrication technique that is based on an extrusion process for fabricating tissue scaffolds [43]. The work done by the group focused on using different materials for each of the three nozzles that were used in the MDM process of tissue scaffold fabrication. The group built scaffolds by using three

60 38 methods, using a single nozzle, bi-nozzle, and tri nozzle systems. In the single nozzle deposition process, material slurry made of PLLA dissolved in dioxane with suspended tricalcium phosphate (TCP) particles was used as the deposited material. The bi-nozzle deposition used PLLA and TCP for one of the nozzles with lower viscosity in compared with the single nozzle deposition, and a second solenoid valve was used to jet deionized water as a support material to ensure the interconective macroporous structure. The tri-nozzle system, in addition to the binozzle deposition, used a third nozzle to spray BMP particles to recruit stem cells gradually from the surrounding tissue after the scaffold has been implanted. The group conducted in vivo experiments to repair bone in rabbits using scaffolds made from the single nozzle system. Some of the scaffolds were loaded with rhbmp while the scaffolds in the control were not loaded with rhbmp. The results showed that after twelve weeks post surgery, the scaffolds with the rhbmp repaired the bones while the scaffolds without the rhbmp did not. The MDM proves to be a successful technique for fabricating tissue scaffolds. However, this process is more complicated than other freeform fabricating methods since more than one material could be used to fabricate complicated and more sophisticated scaffolds. This may be one of the new leading directions in the field of tissue engineering. However, the group did not mention what were the optimal parameters such as nozzle tip and applied pressure used to fabricate the scaffolds. As mentioned earlier, the 3D plotting method does not require the material to be heated prior to deposition. This allows polymeric solutions, slurries, and pastes to be used while adding so biocomponents such as proteins, growth

61 39 factors, and even cells in to the fabrication process. Xiong et al. conducted experiments on the low-temperature deposition manufacturing (LDM) to fabricate PLLA/TCP composite scaffolds for bone tissue engineering [42]. This group was the same as Yan et al., therefore having similar processing technique for building the scaffold. The nozzle diameter was 0.3 mm, however, the pressure applied was not mentioned. The PLLA was mixed in dioxane and TCP particles were then added to the solution. The polymer deposition occurred in a refrigerator where the temperature was under 0 o C. The material was deposited in adjacent parallel lines for each layer. Every adjacent three layers had the same line directions and then the next three were rotated 90 degrees. This structure was repeated for the rest of the scaffold. After the forming process, the frozen scaffolds were freeze dried to remove the solvent. The fabricated scaffold using LDM has a macro-pore dimension of about 400 μm and a micro-pore dimension of about 5 μm. The scaffold was measured for its mechanical properties using an Instron 4502, which revealed μm compressive yield strength of 4.71 MPa and a compressive modulus of MPa. These results are close to the human lumbar spongy bone (40-49 years old), which have compressive yield strength of 1.86 MPa and a compressive modulus of 88 MPa. However these results are much lower than that of human compact bone. In vivo experiments were conducted on animals to evaluate the scaffold. The PLLA has hydrophobic surface that makes surface cell adhesion very poor. Also, PLLA had a low degradation rate, which was too low to match the cell regeneration of the surrounding tissue. In contrast, the PLLA/TCP composite had a degradation rate that matched the tissue regeneration. However

62 40 one of its drawbacks is that it s mechanical properties were less than the PLLA scaffolds. The overall view, suggests that the PLLA/TCP compressive yield strength of 4.71 MPa and a compressive modulus of MPa cp composite was much more suitable for tissue engineering applications than the PLLA scaffold. The freeze-drying process seemed to be a very important factor in producing the micro-pore structure of the scaffolds although it was not mentioned in the publication. In general, there are two modes of biopolymer deposition; the extrusion mode and the droplet mode. In the extrusion mode, the material is extruded out of the nozzle tip under an applied pressure. This mode can basically lay down the material in the form of line structures to create the desired model by moving the nozzle tip over a substrate in the designed path. This process can be repeated layer by layer to develop a freeform fabricated part. In the droplet mode, the material is deposited in the form of droplets that is controlled by using a frequency function and key parameters in the nozzle system settings. The droplet mode can form a structured layer by depositing multiple droplets at desired locations on a substrate. Similarly, this process can be repeated to fabricate a 3D structure. Figure 1.1 presents a schematic diagram of the extrusion mode and droplet mode for fabricating structures.

63 41 Nozzle Nozzle (A) (B) Figure 1.3: Biopolymer deposition: (A) extrusion mode, (B) droplet mode. 1.7.Development of Novel Hydrogel Freeform Deposition System It has been observed that most SFF manufacturing methods utilize harsh solvents, high pressures or temperatures, or post-processing methods that are not suited for working with cellular and bioactive materials. This research developed a biofabrication system that can simultaneously with the scaffold construction, deposit cells, growth factors, or other bioactive compounds in a controlled manner with precise spatial position to form a designed cell-seeded tissue constructs. This process may solve the problem of cell loading to tissue scaffolds which has been a significant barrier in tissue engineering, and enables the fabrication of larger or thicker tissue constructs with complex layouts, such as vascular networks, thus layout a preliminary foundation for future organ printing. Biopolymer-based hydrogels, such as chitosan and alginate, are appealing scaffold materials because they are structurally similar to the extracellular matrices of many tissues, processed under relatively mild conditions, and

64 42 delivered in a minimally invasive manner. Hydrogels comprised of naturallyderived macromolecules have potential advantages in terms of biocompatibility, cell-controlled degradability, and intrinsic cellular interaction [ ]. These materials, such as chitosan, are useful for many applications in tissue engineering, such as space filling agents, delivery vehicles for bioactive molecules, and threedimensional structures that organize cells and present stimuli to direct the formation of a desired tissue [96, ]. The availability of appropriate materials and the capability of manufacturing them into a desired 3D scaffold structure often determine the success of biopolymeric scaffolds in tissue engineering. Numerous gels have been incorporated as the scaffold material using SFF techniques such as acrylamide gels, chitosan, gelatin, and agarose [164]. Such gels have been of interest for encapsulation of cells and most recently such hydrogels have become especially attractive to the new field of tissue engineering as matrices [165]. The formation of structure with intended design geometry has been introduced by controlling the nozzle velocity, flow rate, and nozzle diameter. 1.8.Research Objectives The objective of this research is to develop a scientific and engineering knowledge required to fabricate 3-dimensional (3D) cell-embedded hydrogelbased tissue scaffolds, to study the effect of the processing, materials and structural configuration on the scaffold structure formation, and to understand cell viability during scaffold construction. The specific aims are:

65 43 1. Develop a viable deposition system for bioactive fabrication of tissue substitute constructs 2. Deposition feasibility study for alginate aqueous solutions 3. Study 3D alginate scaffold structural formation 4. Study cell viability in freeform fabrication of bioactive alginate scaffolds The proposed research and activities will help to develop knowledge and novel solutions in determining optimal design and process parameters for bioactive polymeric scaffolds in tissue engineering applications. Such parameters are critically important when considering, for example, 1) the mechanical properties; 2) the appropriate process parameters for 3D structure formation of biopolymer scaffolds; 3) the scaffold structural uniformity for pore size and distribution, cell distribution and mechanical properties; 4) the scaffold degradation rate; and 5) the desirable cell-scaffold bioactive process parameters that satisfy from manufacturing and biological perspectives for cell survivability and proliferation. In this research, a novel biopolymer deposition system designed for fabricating 3D hydrogel tissue scaffolds is presented. Alginate was selected as the biopolymer gel for the fabrication process, which is capable of being deposited at ambient temperatures in aqueous solution. Flow rate measurements of sodium alginate through three different systems were studied to select a suitable nozzle system. Calcium chloride was deposited in a heterogeneous technique to crosslink

66 44 sodium alginate into a hydrogel. The flow rate determines the amount of material required to build 3D alginate scaffolds. The research is focused on the development of an engineering approach for manufacturing alginate hydrogel 3D tissue scaffolds by using a multi-nozzle biopolymer deposition process. A model was developed for the pneumatic microvalve to determine alginate strut diameter and scaffold porosity and was then compared to the experimental results. In addition, rat heart endothelial cells (RHEC) were mixed with alginate solution to fabricate bioactive scaffolds with embedded cells to develop a Bioactive Freeform Fabrication (BFF) process. Cell viability studies were conducted on the cell seeded structures for validating the bioactive process. The mechanical properties of scaffolds have proven to be a major factor in cell proliferation and differentiation. Models to predict the elastic modulus and encapsulated cell viability through process parameters are part of the on-going research, which could establish design methods on fabricating customized scaffolds for specific functions. Towards the proposed objectives, the major research activities are presented below: Activity 1: Deposition feasibility study of sodium alginate Study the effect of the key system parameters, such as pressure, nozzle diameter, solution viscosity, type of nozzle system, and process environment on the flow rate of sodium alginate;

67 45 Viscosity measurements of sodium alginate at different shear rates and at various concentrations to find the power law index; Mechanical testing of bulk alginate at various concentrations to extract the elastic modulus Characterization of morphological and structural properties of alginate scaffold through optical imaging; Activity 2: Process Parameter Predictive Models Compare a model and experimental results of flow rate for sodium alginate as a function of the process parameters such as pressure, nozzle diameter, solution viscosity, type of nozzle system; Develop a model to predict the diameter of the alginate scaffold struts based on the flow rate nozzle velocity; Develop a model to predict the porosity of the alginate scaffold based on the process parameters such as pressure, nozzle diameter, solution viscosity, type of nozzle system, and scaffold geometrical design and architecture; Activity 3: Study 3D alginate scaffold structural formation Construct 3D of alginate tissue scaffold; the controllability of micronozzles, spatial deposition, and the effect of the key system parameters, such as variation of velocity, flow rate, and process environment;

68 46 Design and fabricate 3D alginate scaffolds with different strut velocities and compare experimental results to the predictive process model for predicting the strut diameter; Design and fabricate 3D alginate scaffolds with different strut distances and compare experimental results to the predictive process model for predicting the pore size; Activity 4: Study cell viability in freeform fabrication of bioactive alginate scaffolds Fabricate 3D alginate scaffolds using a bioactive freeform fabrication technique for fabricating struts in a layer by layer process. The struts will be placed next to each other at a known distance for each fabricated scaffold. Each layer of struts will be oriented 90 degrees from the corresponding layers. The pore size of every scaffold will be controlled via the known distance of the struts from one another in each layer and the diameter of the struts. One pore size will be designed for the scaffolds; Investigate and study the effect of alginate and calcium chloride concentrations on the viability of encapsulated cells in alginate struts; Develop a method to fabricate alginate scaffolds fabricated with cell/alginate using a sterile bioactive process; Develop cell proliferation constraints for alginate and calcium chloride concentrations in the bioactive fabrication process;

69 47 The scaffolds are characterized for cell proliferation and viability. The average count for each scaffold process parameter is determined. Images are taken using the inverted microscope for documentation; The medium will be removed after 2-3 days and fresh medium will be added to the scaffolds; Surface images are taken using the inverted microscope for documentation. The average count for each scaffold pore size is determined; 1.9.Thesis Outline The second chapter of this thesis discusses the development of a multinozzle deposition system for the bioactive fabrication of tissue substitute constructs. The second chapter introduces the system configuration, motion control system, and the type of nozzle being used. The third chapter introduces the development of process models for the fabrication of 3D scaffolds with controlled strut diameters, scaffold pore size, and scaffold porosity. The fourth chapter introduces the deposition feasibility study on sodium alginate aqueous solutions by studying the nozzles operating parameters and their deposition performance using flow rate measurements. The fifth chapter presents the structure formation methods and techniques for fabricating 3D alginate gels. The sixth chapter presents the bioactive cell deposition of 3D scaffolds including cell viability and proliferation studies. Finally, the seventh and last chapter presents the conclusion of this thesis with recommended future work.

70 48 2. DEVELOPMENT OF A MULTI-NOZZLE DEPOSITION SYSTEM FOR BIOACTIVE FABRICATION OF TISSUE SUBSTITUTE CONSTRUCTS 2.1.The System Configuration This research developed a multi-nozzle deposition system (MND) for freeform fabrication of biopolymer based 3-diemnsional (3D) tissue scaffolds and cell-embedded tissue constructs. The system utilizes different types of delivery nozzle systems that can deposit a wide range of solutions at varying viscosities and material properties, including a solenoid-actuated nozzles, piezoelectric glass capillary nozzles, pressure-actuated syringe nozzles, and spray nozzles. The system can deposit different viscosity gel-like solutions in a continue extruding mode, or in a droplet format to deposit individual droplets with pico-liter volumes. The developed system was designed to operate in a biological friendly environment with capability to deposit living cells or other biological compounds at room temperature and at low pressures. A schematic representation of the system and the system configuration is shown in Figure 2.1 and Figure 2.2, respectively.

71 49 Positive Pressure Negative Pressure Moving Directions PC Material Container Motion and Nozzle Controller Material Delivery Tube Nozzle 3D Motion Arm Figure 2.1: Schematic diagram of 3D motion set-up for biopolymer depositions used SFF of tissue engineered scaffold As the system configuration shown in Figure 2.2, the data processing system processes the designed scaffold model and converts it into a layered process toolpath. The motion control system is driven by the layered manufacturing technique; the material delivery system consists of multiple nozzles with different types and sizes, thus enabling the deposition of a wide range of hydrogels with different viscosities for constructing 3D tissue scaffolds. Four types of the nozzles are studied in the system: solenoid-actuated nozzles, piezoelectric glass capillary nozzles, pneumatic syringe nozzles, and spray nozzles, with size ranges varying from 30 μm to 500 μm. The system can continuously extrude hydrogel gels, or form hydrogels in single droplets with

72 50 picoliter volumes. The multiple nozzle capability allows a simultaneous deposition of cells, growth factors, and scaffold materials, thus enabling the construction of heterogeneous scaffolds with living cells or bioactive compounds, or the construction of functional gradient scaffolds with designed mechanical/structural properties in desirable scaffold regions. Figure 2.2: Deposition system data processing system for converting designed scaffold models into a layered process tool path using multi-nozzles in 3D motion

73 51 Figure 2.3 shows the information pipeline flowing through the fabrication process. The designed scaffold CAD model is first converted into STL format, and then sliced with the slicing patterns stored in the pattern library for toolpath generation. Initialized by a parameters file, the in-house developed system control software provides functions for 3D part visualization, machine and process setup, testing and monitoring during the real-time fabrication process. In this part of research, three nozzle systems were investigated to evaluate their performances and ability to deposit polymer solutions that could be utilized as a biomaterial for tissue engineered scaffolds. The three systems were the pneumatic, piezoelectric, and solenoid valves. Sodium alginate was chosen as the biopolymer material to be utilized during the testing of the MND systems. The three nozzle systems were assembled to a 3D motion system (Parker Hannifin Corporation Automation Group, Cleveland, OH) that could move the nozzles in a 20X20X20 cm space. The 3D motion system was placed onto an optical table (Vere Inc., Kensington, PA) for rigid support and decreased external vibrations. A PC was connected to the 3D motion system to control two functions; one, the motion of the nozzles in 3D space and two, activation and deactivation of the nozzles. A material delivery system was assembled to supply the nozzles with the appropriate biopolymer. The system consisted of an air pressure supply both positive and negative, a material container or reservoir, and a material delivery tube. Each nozzle system had its independent parameters adjusted as requires such as the air pressure and biopolymer concentration.

74 52 CAD and Data Processing CAD based scaffold geometry Conversion to STL format Generate slice Layers Patterning for scaffold internal architecture Prepare scaffold System control Define fabrication process parameters 3D visualization of parts with contoured patterns Check fabrication parameters and operations test Setup Machine for fabrication Machine and process Control System Temperature XYZ positioning Material extruder Fabrication Process MND System Scaffold model Figure 2.3: Deposition system configuration for nozzles in 3D motion 2.2.The Motion and Control Sub-system The second major subsystem is the motion control system. Figure 2.4 shows the structure of the positioning system. The motion system consists of XYZ three axes; each axis is actuated by one AC servo motor which is driven by servo driver; the three servo driver is controlled by 6K4 servo/stepper controller which communicates with the computer through RS232 serial communication

75 53 port. The 6k4 servo/stepper controller has the ability to control 4 axes and also provides digital input/output function. Position feedback Travel Limits (POS,NEG,HOM) Travel Limits Command Position Feedback Servo Driver Model #: GV-L3E SN#: Power Servo Motor SM232AE-NMSN SN: coupler Screw-Drive Actuator Model #: A Serial#: A RS232 to computer 6K4 servo controller Travel Limits Command Position Feedback Travel Limits Command Position Feedback Servo Driver Model #: GV-L3E SN#: Servo Driver Model #: GV-L3E SN#: Position feedback Power Position feedback Power Servo Motor SM232AE-NMSN SN: Servo Motor SM232AE-NMSN SN: coupler coupler Travel Limits (POS,NEG,HOM) Screw-Drive Actuator Model #: A Serial#: B Travel Limits (POS,NEG,HOM) Screw-Drive Actuator Model #: A Serial#: C Command Position Feedback Servo Driver Model# :SGDL-03BS SN# Position feedback Power Servo Motor SGML-03BF12 SN # Worm-gear Protective Break Extruder Assembly Model #: Serial# Temperature control system Figure 2.4: Structure of the motion system

76 54 Automatic iterating Control Software File operation (read & verify) Part display Part fabrication Machine operation& tuning First, last, Next, Previous Show specific layer Zoom in, out, extent Pane left, right, up, down Whole part fabrication Depositing specific layer nozzle test Motion system test Parameter Management Figure 2.5: Functionality of the control software The functionality of the software can be classified into four major groups as shown in Figure 2.5; 1) machine component operation and testing functions. This group of function is used to test and tune different machine components to achieve the best performances and monitor the status of the machine while it is running; 2) Part geometric information input and displaying functions. The input data is the tool paths file generated by the data processing software. The tool paths of each layer can be displayed on the screen, and can be manipulated and iterated. These functions will provide the user tools to visually check the correctness and conformity of the tool paths; 3) Part fabrication functions. These functions will translate the tool paths information into NC code controlling the machine to make the desired part; and 4) Parameters management functions. The

77 55 parameters of the machine and fabrication process should be well maintained and managed in a convenient and safety way. The control software will be developed using the object-oriented methodology, and the following picture shows the first degree object-relation diagram. There will be three kinds of file involved in the program. The first one is the tools path file which is generated by the data processing software. The second one is the NC code file which will be generated based on the tool path file and will be downloaded to the 6K4 controller to control the machine. The third one is the parameter file which maintains the system parameters such as velocity, acceleration, working space and so on. The software will provide a user-friendly interface, so people can use the key board, mouse and the monitor to interact with the software. The software includes an object of communication interface through which the software communicates with the communication server. The communication server is an OLE automation server provided by the Parker Automation Corporation. It acts as a driver of the 6K4 controller similar as the drivers for other computer peripheral equipment such as printer. The above discussed is the interfaces of the software with circumstances including disk file, operator and machine. The software itself consists of several objects; the major objects are shown in the first degree object-relationship diagram. The parameter object maintain the parameter file, the parameter can be referenced by other relative objects, but can be modified only the properties manager. The Tools path object read in the tools paths file, check the validation of the file, generate NC code file, reference to the parameter object. The

78 56 communication interface object is the interface between communication server and the software, via which the software send command, get feedback, check status, download file, and download operating system to or from the 6K4 controller. Machine operation tools provide the user the capability to check and tune the machine, it references to the parameter object, send command to and get feed back from the 6K4 controller via the interface communication object. The software will be constructed based on the view-document structure, the document object will maintain the tools path object, and view object will display the part, receive and process the user input command. The object-relationship diagram of the control software is shown in Figure 2.6. Display & Input Monitor Mouse Keyboard Tools Path NC Code Para meters Read Validate Generate Read Write Tools path Parameter Reference View Document Machine operation Tools Properties manager Command Feedback Communication interface Communication server Read/modify/implement Reference file name Figure 2.6: Object-relationship diagram of the control software

79 The Deposition Nozzle Sub-Systems Each deposition nozzle sub-system is unique in the sense of its method of operation and the key operating parameters. This makes each system have its advantages and limitations over the others when applying to biopolymer deposition. All nozzle sub-systems have a material delivery component, however, the detailed set-up of the material delivery component in each sub-system is different and the operating parameters are controlled by an in-house developed control software. Three nozzle sub-systems were studied, including a pneumatic microvalve, a piezoelectric nozzle, and a solenoid valve. Characteristics and comparison of the three nozzle systems are shown in Table 2.1 and detailed descirption of the nozzle sub-system is given in the following sub-sections.

80 58 Table 2.1: Characteristics and comparison of the three nozzle systems Microvalve Nozzle System Features Solenoid Micro-nozzle Piezoelectric Micro-nozzle Pneumatic Micronozzle Deposition Mode Continuous/Droplet Droplet Continuous/Droplet Operation/ Control Frequency pulse of voltage Frequency pulse of voltage Frequency pulse of air pressure Key Process Parameters Pressure Frequency pulse Material Nozzle diameter Deposition speed Pressure Frequency pulse Material Nozzle diameter Deposition speed Pressure Frequency pulse Material Nozzle diameter Deposition speed Operating Range limitations V: 40V(DC) H: ( Hz) D: (30, 50, 70 µm) H: ( Hz) V: ( ) D: (30, 50, 70 µm) H: ( Hz) Fluid P: (0-50 psi) Valve P ( psi) Structure Formation Physical solidification Chemical reaction Physical solidification Chemical reaction Physical solidification Chemical reaction Advantages Room temperature Room temperature Micro-droplet deposition Extrusion and droplet Controlled volume Sterile environment Sterile Environment Room temperature High viscosity Extrusion and droplet Sterile environment Disadvantages Low viscosity Droplet controllability Low viscosity Not continuous deposit Droplet controllability Precision deposition

81 Pneumatic Microvalve The pneumatic microvalve is a typical mechanical valve that opens and closes the valve via an applied air pressure regulated by a controller (EFD Inc., East Providence, RI). Figure 2.7 presents a schematic diagram of the pneumatic microvalve. The air pressure to the controller was adjusted to 70 psi to open and close the valve. The system could work in extrusion or droplet mode. In extrusion mode, the controller applies pressure to the open the valve by lifting the piston against the spring that lifts the needle from the needle seat. The biopolymer material is then extruded out of the nozzle tip under an applied pressure that is adjusted through the material delivery system. The extrusion is ended when the controller closes the valve by placing the needle back to the needle seat. The pneumatic microvalve could perform in droplet mode by repeating the continuous mode in a cyclic manner. Sodium alginate aqueous solution at a concentration of 3% (w/v) was used to investigate the deposition at the extrusion mode. The variable parameters were the nozzle tip diameters at 100, 150, 200, 250, 330, 410 µm and the sodium material pressures at 8, 16, 24, and 32 psi.

82 60 Pressure 70 psi Open valve Closed valve Time 1 Cycle (a) (b) Figure 2.7: (a) Schematic diagram of the pneumatic microvalve; (b) Schematic diagram of the air control per cycle Solenoid Microvalve The solenoid microvalve used in the study was a viton seal solenoid microvalve (The Lee Company, Westbrook, CT). The microvalve operates by opening and closing the valve through a solenoid system. Figure 2.8 shows a schematic diagram of the solenoid microvalve with a cross-sectional image. In the valve is a coil that is wounded around most of the valves length. When an electric current is applied to the microvalve, a magnetic field is induced that forces a piston to open the valve. When the no electrical current is applied to the microvalve, the spring forces the piston onto the valve seat to close the valve. During operation, a control unit sends electrical pulses to the valve at high frequencies that could reach 1200 Hz to shut off and on the microvalve. The

83 61 control unit could also adjust the opening and closing duration of the microvalve for any frequency. The opening and closing duration of the microvalve are referred to as the height time and low time, respectively. This is shown in the schematic diagram in Figure 2.8. At every pulse from the control unit, the microvalve opens and closes to complete one cycle. At a frequency of 800 Hz, the microvalve opens and closes 800 times per second. During the height time, there are two consecutive voltages that are applied to the solenoid microvalve with each voltage having its own time. The first voltage in the height time is the Peak voltage, which is applied to microvalve to initially open the microvalve with a reasonable amount of force. The Peak voltage for the microvalve was set to 40 volts. The second voltage is called the Hold voltage, which is lower than the Peak and was set to 4 volts. It function is to maintains the valve open till no more current is applied. The reason that the Hold voltage is lower than the Peak voltage is because not much force is needed in maintaining the valve open than when opening the valve and to also keep the microvalve from heating. Sodium alginate aqueous solutions were tested for flow rate at concentrations ranging from 0.1 % to 1 % (w/v). The data were for values of flow rate at different pressures and varying frequencies. Five nozzle sizes were used for the measurements; 2, 3, 4, 5, and 7.5 MIL. The height time was set to 250 μs with a peak voltage time of 168 μs. The change of frequency was controlled by only changing the low time.

84 62 Voltage 40 V 4 V Peak voltage time Hold voltage time Time High time Low time (a) 1 Cycle (b) Figure 2.8: (a) Schematic diagram of the solenoid microvalve; (b) Schematic diagram of the solenoid microvalve deposition cycle Piezoelectric Microvalve The Piezoelectric microvalve used was the MJ-SF (Microfab Technologies, Inc., Plano, TX) that had an inner nozzle diameter of 50 microns, total length of 34 mm, and body diameter of 12 mm. The MJ-SF series from Microfab has been developed to dispense single droplets of solvents, polymers, solder, water-based fluids, and inks with in an operating temperature range of 20 and 250 C o. With proper fluid preparation and device maintenance, the jetting device will provide reliable delivery of fluid microdrops. It is known that the performance of the jetting device is directly related to the dispensing fluid properties. The MJ-SF is known to successfully deposit fluids with viscosities less

85 63 than 40 cp and surface tensions in the range of N/m. This device can also generate drops ranging from picoliters in volume. The jetting system operates in positive and negative air pressures depending in the viscosity and contact angle of the fluid. This can be determined by applying the fluid to the material reservoir that is placed directly above the nozzle. If the fluid does not drip from the nozzle tip, then a positive pressure may be required in order to have microdroplets form when the piezo nozzle is activated. If the fluid drips from the nozzle tip with out having the nozzle tip activated, then a negative pressure is required to keep the fluid from dripping and holding it at a level close to the tip. A holding pressure is always needed weather it is positive or negative to keep the fluid held just at tip of the nozzle. Once a holding pressure is applied to a fluid, the piezoelectric nozzle frequency is adjusted to the optimal vibration that allows the fluid to deposit in to microdroplets. The vibration occurs by applying a voltage to the piezoelectric nozzle that is used to change the volume of the fluid reservoir in order to produce the fluid injection and retraction pressure pulse. This pressure pulse causes the fluid to be jetted out as a continuous stream of approximately the diameter of the ejection aperture nozzle [87]. The mechanical vibration induced by the piezoelectric component causes instability that breaks the fluid stream into individual droplets that have an approximate diameter size twice the diameter of the nozzle hole. The droplet size and deposition rate can be controlled by the frequency of the vibrating piezoelectric nozzle and the voltage applied even at a

86 64 constant nozzle diameter and is directly related to the fluid s viscosity and surface tension. Controlling the voltage time intervals, which are illustrated in Figure 2.9, can strictly control the expansion and contraction of the piezoelectric nozzle to adjust the frequency. Five time intervals, Trise, Tdwell, Tfall, Techo, and Tfinal control the pulse width. The T lag is the time between the end of a pulse width and the beginning of the following pulse width. The addition of the pulse width and the T lag is the time for one cycle, which is the inverse of the frequency as shown in Equation (2.1). Sodium alginate aqueous solutions with concentrations of 0.1% and 0.4% (w/v) were prepared using deionized water to be deposited using the MJ-SF nozzle system. The 0.1% and 0.4% (w/v) sodium alginate aqueous solutions had viscosities of 8 and 24 cp, respectively. The sodium alginate solutions were tested at voltages of 55, 60, 65, 70 volts and frequencies of 500, 1000, 1500, 2000 Hz. The frequencies were adjusted by changing the T lag. Frequency = 1 Pulsewidth + Tlag (2.1)

87 65 Trise Tdwell Tfall Techo Tfinal + Voltage 0 Time - Pulse Width Tlag (a) (b) Figure 2.9: (a) Schematic diagram of the piezoelectric microvalve; (b) Schematic diagram of the piezoelectric microvalve deposition cycle

88 66 3. DEVELOPMENT OF PROCESS MODELS FOR 3D ALGINATE FABRICATION 3.1.Sodium Alginate Rheology Understanding the polymer rheology is important in the development of process model of polymer freeform deposition. Among many material and process parameters, polymer viscosity is one of the most important in many manufacturing processes, such as extrusion, direct writing, fiber spinning, and injection molding [166, 167]. The viscosity is equal to the shear stress over the shear rate is given in Equation (3.1). τ η = (3.1). γ Where τ is shear stress, η is the viscosity, and. γ is the shear rate. The units of viscosity can be represented by dynes.s/cm 2, which is also known as Poise. For a Newtonian liquid, viscosity is a constant directly proportional to the shear rate and that the slope is constant [168]. On the other hand, a non-newtonian fluid viscosity is a function of the shear rate. A typical shear stress - shear rate relation for Newtonian and non-newtonian liquids is shown in Figure 3.1.

89 67 Figure 3.1: Viscosity measurement for Newtonian and Non-Newtonian Liquids from a shear stress versus shear rate graph. A viscosity in a non-newtonian fluid is often called as apparent viscosity. For non-newtonian polymer melts, the apparent viscosities are the function of the temperature, pressure, and shear rate that are being applied. In general, the viscosity increases with the increase of pressure, with the drop of temperature and with the decrease of shear rate [169]. It can also be increasing with the increase of molecular weight and molecular architecture. The viscosity decreases as the shear rate increases, known as shear thinning phenomenon. At low shear rate which known as the lower Newtonian region, the viscosity of polymer melts usually has a constant value called limited viscosity η 0 at low shear rates, which known as the lower Newtonian region as can be seen in Figure 3.2. Note that the Newtonian region diminishes as the

90 68 temperature increases and the viscosity almost changes linearly with shear rate at higher temperatures. Figure 3.2: Effect of temperature on viscosity for typical polymer melts In the non-linear range at the low temperature, the shear stress and the shear rate of polymer the power law as shown in Equations (3.2) and (3.3) can represent melts [169];. n τ = K γ (3.2)

91 69. ( n 1) τ η a = = K γ (3.3). γ where the constant K is the consistency index and the constant n is the power law index is equal to one for Newtonian liquids and is less than one for non- Newtonian liquids as polymer melts. For the non-linear curve of the viscosity versus shear rate as shown in Figure 3.2, following Carreau equation as shown in Equation (3.4) can be applied [169]: η η a 0. 2 = 1 + λc γ ( n 1) / 2 (3.4) where the constant η 0 is the limited viscosity as low shear rates and the constant λ c is the point at which shear thinning occurs. Since polymer melts behave with non-newtonian flow property, the shear stress within the flow may take a certain amount of time to reach a static equibrulim state even at constant shear rates. For this reason, the fluid viscosity can also be expressed as function of time at constant deformation rates. When considering the temperature affect on the polymer, its glass transition temperature T g, which is a result of free volume found in polymers, has a direct influence on the viscosity [170]. Free volume increases with temperature and at the glass transition temperature turns a glassy amorphous polymer into a rubbery state until it reaches its melting temperature. The more crystalline a polymer the less free volume it has, which

92 70 indicates that the polymer does not experience significant rubbery state before the melting temperature. Williams, Landel, and Ferry suggested to express the viscosity of most polymers at different temperatures as following in Equation (3.5) [168]: η0 ( T ) log η ( T ) 0 g = 17.44( T T ( T T g g ) ) (3.5) At temperatures above the glass transition temperature or at the melting point of the polymer, following Arrhenius equation [171] could be used for calculating the viscosity: ( E / RT ) η 0 = Ke (3.6) where K is the consistency index of the polymer as defined before, E is the activation energy for the flow process, and R is the universal gas constant. The units for temperature is in degrees Kelvin. Similarly to polymer melts, viscosity of polymer solution is also a function of temperature, pressure, shear rate, molecular weight, molecular architecture, as well as the degree of concentration. For an infinitely dilute polymer solution, the relative viscosity can be related to the polymer volume fraction φ through the Einstein result as shown in Equation (3.7). η ηr = = φ (3.7) η s

93 71 where η and η s are the solution and solvent viscosities, respectively. The polymer volume fraction φ is a function of the polymer single molecule volume v, mass concentration c, Avogadro s number N A, and polymer molecular weight M as shown in Equation (3.8). vcn φ = A 3.8 M The Carreau model defined in Equation (3.4) is usually applied to polymer solutions with low viscosity [169]. For high viscous polymeric solutions, the Carreau equation could be modified as: η η a η η o. 2 = 1 + λc γ ( n 1) / 2 (3.9) where η is the limiting viscosity at high shear rates, which is usually taken as the solvent viscosity. Figures 3.3 and 3.4 present an experimental results of viscosity of Sodium Alginate solutions with different concentrations at three different temperatures; 25 o C, 30 o C, and 35 o C. The tests were conducted using a Brookfield viscometer model HBDT at a constant shear rate of approximately 91 s -1. The Shear rate was calculated through Equation (3.11) following equation [169].

94 ωr R γ () r = (3.10) 2 r ( R R ) 2 1 Where ω is the angular rotation in radians per second, R 1 and R 2 are the inner and outer diameter of the cylinders, respectively, and r is any arbitrary radius between R 1 and R 2. Sodium Alginate Aqueous Solution Viscosity (cp) % 0.50% 0.75% 1% 1.25% 1.50% Temperature (C) Figure 3.3: Viscosity versus temperature for Sodium Alginate aqueous solutions at various concentrations.

95 73 Sodium Alginate Aqueous Solution Vsicosity (cp) T= 25 C T = 30 C T= 35 C (w /v) Concentration (%) Figure 3.4: Viscosity versus concentration for Sodium Alginate aqueous solutions at various temperatures. The behavior of sodium alginate aqueous solutions is similar to polymer solutions. As shown in Figures 3.3 and 3.4, the viscosity of sodium alginate aqueous solution is a non-linear function of temperature and the concentration. In addition, it should also be a function of the shear rate since it is a non-newtonian liquid. The temperature dependency of the viscosity can be expressed using Equation (3.12); B T η = Ae (3.11)

96 74 and where, T1T 2 η2 B = ln (3.12) T1 T2 η 1 B T 1 1 A = η e (3.13) where, T 1 and T 2 are the temperatures at which viscosities η 1 and η 2 were measured, respectively. The viscosity of sodium alginate solution can be determined at any temperature between the two temperatures when taking T 1 and T 2 at 35 o C and 25 o C, respectively. Comparison of the experimental data and the predictions by Equation (3.12) is given in Figure 3.5. Viscosity vs. Concentration for Sodium Alginate Aqueous Solution at T= 30C 2000 Viscosity (cp) Analytical Experimental Concentration %(w/v)

97 75 Figure 3.5: Comparison of experimental and analytical viscosity results for sodium alginate aqueous solutions at 25 o C. The consistency index K and the power law index n in Equation (3.3) for viscosity with the shear rate can be determined using curve fitting method from the experimental data. For example, we used the experimental data of the viscosity versus the shear rate shown in the curves in Figures 3.6 to 3.9 to determine the values of K and n in Equation (3.2) for sodium alginate aqueous solutions with 1%, 1.5%, 2%, and 3% (w/v) concentration at an ambient temperature (T=25 0 C). Comparisons of the experimental curves and the curves predicted by Equation (3.2) with the fitted K and n are given in these figures. 1% (w/v) Na-Alginate Aqueous Solution at T = 25C Viscosity (cp) Experimental Analytical, n= Shear Rate (1/s)

98 76 Figure 3.6: Experimental and analytical values of viscosity and different shear rates for 1% (w/v) sodium alginate aqueous solution, where K = 610 and n = % (w/v) Na-Alginate Aqueous Solution at T = 25C Viscosity (cp) Experimental Analytical, n= Shear Rate (1/s) Figure 3.7: Experimental and analytical values of viscosity and different shear rates for 1.5% (w/v) sodium alginate aqueous solution, where K = 1259 and n = 0.7

99 77 2% (w/v) Sodium Alginate Aqueous Solution at T = 25C Viscosity (cp) Experimental Analytical, n= Shear Rate (1/s) Figure 3.8: Experimental and analytical values of viscosity and different shear rates for 2% (w/v) sodium alginate aqueous solution, where K = 2010 and n = % (w/v) Sodium Alginate Aqueous Solution at T = 25C Viscosity Model, n=0.76 Experiment Shear Rate (s^-1) Figure 3.9: Experimental and analytical values of viscosity and different shear rates for 3% (w/v) sodium alginate aqueous solution, where K = 8587 and n = 0.76

100 Predictive Model for Flow Rate A development of an analytical model for predicting the flow rate of non- Newtonian sodium alginate aqueous solutions is presented in this section. The model is based on the Poiseulle s equation and is implemented on the pneumatic microvalve system and considered the pressure, nozzle diameter, fluid viscosity, and length of the material delivery capillary as key process parameters. For Newtonian liquids, if we consider the forces acting on an element of fluid in a capillary, we can express the shear stress as follows; r P τ = (3.14) 2 z where r is the radius of capillary, P is the pressure, and z is the direction in capillary axis. If we consider the pressure drop uniform, then for a pressure drop P over a length L the shear stress is given by; rp τ = (3.15) 2L and since the shear stress is known as;. v τ = η γ = η r (3.16)

101 79 where v is the velocity in the flow direction and r is the distance form the tube axis then; v r P η = (3.17) r 2 z now integrating this gives; v dv = 0 r R 1 dp 2η dz rdr (3.18) v = 2 1 dp r 2η dz 2 R 2 2 (3.19) At r = 0, v = v 0 so v 0 1 dp 4η dz 2 = R (3.20) 2 r v = v 0 1 (3.21) R The volumetric flow rate can be determined;

102 80 Q R R 2 r π 2 2π rvdr 2πrv = 0 1 = dr = v0r (3.22) R r 0 0 and rearranging gives ; 2 πr dp Q = (3.23) 8ηL dz and is known as the Poiseulle s equation. The power law model can be altered to address the man limitation s by using; 0. n 1 η = η λ c γ (3.24) For the non-linear curve of the viscosity versus shear rate, a common model could be used that is known by the Carreau equation as shown in Equation (1.24). η η a 0. 2 = 1 + λc γ ( n 1) / 2 (3.25) as indicated before, η 0 is the limited viscosity as low shear rates and the constant λ c is the point at which shear thinning occurs. The flow of incompressible

103 81 power law liquid flowing through a uniform circular cross-section can be analyzed starting from the force of balance to give; = = n n r v γ η γ γ η η (3.26) and z P r v r r v n = γ η (3.27) which gives z P r r v n n = η γ (3.28) and can be integrated between limits v = v at radius r and v = 0 at radius R to give + = + + n n n n n n n R r z P n n v η γ (3.29) at r = 0, v = v n n n n n R z P n n v + + = η γ (3.30)

104 82 therefore, the velocity at any radius can be expressed by; = + n n R r v v (3.31) To obtain the volume flow rate, the integral in Equation (1.22) can be modified to; dr R r rv rvdr Q R R n n = = π π (3.32) giving n n n n n R z P n n Q = η πγ (3.33) It can be seen that when n = 1, Equation (1.33) becomes Equation (1.23). Equation is a generalized Poiseulle equation for a non-newtonian fluid behaving with non-linear property. The power-law index n characterizes the nonlinearity of the flow. It can be seen that the conventional Poiseulle Equation 1.23 can be considered as a special application of Equation (1.23) at n = 1 for a linear flow:

105 Development of a Predictive Model for Strut Diameter The microvalve systems deposit alginate solution to form cylindrical gel struts, which would then be oriented to form layers of a tissue scaffold. The geometry and the diameter of the strut are functions of many process parameters including the operating pressure gradient, viscosity of the fluid, the nozzle diameter, and the speed of the nozzle movement. D = f. v, R, n, η 0, γ, P Z (3.34) where D is the strut diameter and is directly proportional to the flow rate of deposition. While at a constant flow rate, D is indirectly proportional to the nozzle velocity. Therefore, D can be expressed in terms of the flow rate Q and the deposition speed v:.. D = f Q, ν = f Q ( R, n, η 0, γ, P ), ν Z (3.35) For a given flow rate Q and a given time t the volume extruded from the nozzle system V can be calculated as; V = Q t (3.36)

106 84 We assume the extruded volume can form a cylindrical strut of a uniform diameter D with a length L. L D V 2 ) 2 ( π = (3.37) Considering that vt L = (3.38) In which v is the nozzle speed. Substituting Equation (3.36) into (3.37) and considering 3.38 yield 2 4 D Q v π = (3.39) Therefore, Equation 3.39 can be explicitly expressed by: ν η γ πν n n n n n R z P n n Q D = = (3.40)

107 85 Equation (3.40) can be used to predict the diameter of a strut if the process parameters are given. For a given constant flow rate, the fabricated strut diameter is generally inversely proportional to the nozzle travel speed. However, for a given nozzle and given flow rate, we can find a nozzle travel speed v N at which the fabricated strut diameter D is equal to the nozzle radius R and nozzle diameter D N, i.e., v N 4Q = (3.41) πd 2 N It is hypothesized that if the nozzle travel v speed is greater than v N then the fabricated strut diameter is usually smaller than the nozzle diameter, and if the nozzle travel speed is smaller than v N the fabricated strut diameter is usually larger than the nozzle diameter, This relationship is illustrated further is Figure The strut diameter is a function of the nozzle velocity, nozzle tip inner radius, power law index of the fluid, limited viscosity at low shear rates, shear rate, the pressure gradient as shown is Equation (3.20).

108 86 Figure 3.10: Three scenarios for strut diameter; Case 1 (v 1> v N ), Case N (v N ), and Case 2 (v 2 <v N ) The strut fabrication process is preferred to operate using v N instead of using any other velocities. The reason for this is because there is no tension or compression applied on the struts while the strut is being formed as sodium alginate is depositing through the nozzle tip, which means that the strut avoids any slippage on the substrate surface when its coefficient of friction is relatively low.

109 Development of a Predictive Model for Scaffold Porosity It is known that the porosity of an object is the ratio between the space volume and total volume. In the case of the fabricated alginate scaffold, the geometry and architecture of the scaffold are function of the porosity as shown in Figure 3.11 and expressed in Equation (3.42). L 2 z 2H L 1 D y x Figure 3.11: (a) Layer stacking 90 degrees orientation of alginate scaffold; (b) Unit cell geometry of alginate scaffold Porosity = = f { L 1, L 2, H f ( L 1, L 2, H, D ), D ( v, R, n, η, γ, P (3.42) )} Where H is Z the distance from one layer to another in the z direction, L 1 and L 2 are the unit cell lengths or distance between struts in the x and y directions, respectively. 0. Volume of the scaffod unit cell element, V uc : V UC = 2HL L (3.43) 1 2

110 88 Volume of the strut in the element, V S : 2 D V S = π ( L1 + L2 ) (3.44) 4 The porosity of scaffolds is known as the void space volume and total volume ratio, which is expressed in Equation (3.26); Porosity V V S = 1 (3.45) UC Substituting Equation (3.40) in Equation (3.45), we can establish the following functional relationship between the scaffold porosity and the key process parameters: Porosity = 1 π ( L + L ) n n γ + 1 2vHL L 1 n 1 n P 3n+ 1 n z n 0 R 2η (3.46) In the event where distance between layers H is equal to the strut diameter D, then Equation (3.45) can be simplified to;

111 89 ( L + L ) πd 1 Porosity = 1 8L L (3.47) In the case where L 1 and L 2 are equal to one another and equal to the distance between adjacent struts d in the x and y directions for each layer, respectively, then Equation (3.47) can be modified to; D Porosity = 1 π (3.48) 4d Therefore the porosity in terms of the process parameters can be expressed by; n 1 n P n n n n γ z 0 R π 3n + 1 2η 0 Porosity = 1 (3.49) 2d ν Bulk Alginate Elastic Modulus The bulk elastic modulus E bulk of alginate is a function of the sodium alginate concentration ρ, sodium alginate molecular weight M n, sodium alginate Guluronic and Mannuronic ratio GM, calcium chloride concentration C CaCl2, and the gelation time t g as expressed in Equation (3.50) along with some other parameters. bulk ( C M, GM, C t ) E = f, (3.50) A lg inate, n CaCl 2 g

112 90 From statistical thermodynamics, the elastic modulus for a gel structure could be expressed by; 2 2 ρ RT r 0 2 M c E bulk = 1 ( 1 + υ ) (3.51) 2 M c r f M n Where R is the universal gas constant, r f 2 is the end-to-end distance in the bulk state for linear chain, r o 2 is the end-to-end distance in a real network, T is the temperature, M c is the number average molecular weight between crosslinks, ρ is the polymer concentration, and υ is the Poisson s ratio. It should be noted that r 2 o is a function of gelation time t g and crosslink concentration C CaCl2 in Equation (3.51). The relationship between the elastic modulus E, shear modulus G, and Poisson s ratio v is expressed by; E = 2 G (1 + υ ) (3.52) 3.6.Scaffold Elastic Modulus The elastic modulus of the scaffold in the x and y directions are equal since the unit cell has equal lengths and widths and the struts are 90 degree from layer to layer stating it as a transversely isotropic structure. Under such condition, the unit cell of the scaffold has six independent constants E z, E x = E y, G z, G x = G y, v z, v x = v y. The stress-strain relationship could be represented as follows;

113 91 σ = E e (3.53) ij ijkl kl And can be expressed by the following matrix; σ 1 σ 2 σ 3 = τ 23 τ 31 τ 12 C C C C C C C C C C C C ε 1 ε 2 ε 3 γ 23 γ 31 γ 21 (3.54) Where C 11 C 12 C 66 = (3.55) Development of a Predictive Process Model for Maximum Shear Stress The bioactive fabrication process involves depositing live cells mixed homogeneously with sodium alginate through the pneumatic system. During the deposition, the cells are exposed to the shear forces with in the cell suspension that could potentially harm live cells and destroy them. A study was conducted to investigate the effect of the shear stress in terms of the process parameters on the percentage of live and dead cells. In other terms, a study was performed to relate the fabrication process parameters on cell viability. The shear stress τ and the apparent

114 92 shear rate γ can be represented by the power law as shown in Equations (3.56) and (3.57);. n τ = K γ (3.56). ( n 1) η = K γ (3.57) Where η is the viscosity,. γ is the shear rate, constant K is the consistency index, and the constant n is the power law index that is equal to unity for Newtonian liquids and is less than one for non-newtonian liquids. Figure 3.12: Schematic diagram of fluid velocity through the cross section of a capillary or nozzle tip The shear rate in a capillary can be represented by Equation (3.58), where R is the nozzle radius;. v N γ = (3.58) R

115 93 By substituting Equation (3.41) into Equation (3.58), we find that the maximum shear stress τ max in the nozzle is represented in terms of the process parameters as shown in Equations (3.59) and (3.60); n Q τ max = K (3.59) 3 πr n P n n 1 τ K z max = γ 0 R n + (3.60) 3 1 2η 0

116 94 4. DEPOSITION FEASIBILITY STUDY ON SODIUM ALGINATE AQUEOUS SOLUTIONS 4.1.Materials and Methods Sodium Alginate Sodium alginate (Manugel DMB) Mw = 150,000, Guluronic acid 63% and Mannuronic acid 37% (ISP, San Diego, CA) aqueous solutions with concentrations ranging from 0.1% to 3.0% (w/v) were prepared using deionized water. The viscosities of the sodium alginate aqueous solutions were measured using a toque viscometer (Brookfield, Middleboro, MA) at a constant shear rate of approximately 91 per second. Sodium alginate solutions were prepared for deposition using the Pneumatic microvalve, Piezoelectric nozzle, and Solenoid microvalve systems. Calcium chloride (Sigma, St. Louis, MO) was used as the crosslink solution to gel sodium alginate aqueous solution Flow Rate Measurements Experiments were conducted to measure the flow rate of the deposited sodium alginate solutions at various parameters to understand the feasibility and controlibity of deposition of sodium alginate aqueous solutions. All experiments that have been conducted have been at a room temperature and humidity range of 21.5 C o C o and 43%-69%, respectively. Measuring the mass of the deposited solution on a balance and then dividing it by the measuring time as showed in Equation (4.1) first calculated the mass flow rate. The flow rate was then calculated by dividing the mass

117 95 flow rate by the solution density as shown in Equation (4.2). The nozzle tips were approximately 3cm above the Petri dish surface.. m mass = grams / sec ond time = (4.1). m Q = = microlitre/second (4.2) ρ Where Q is the flow rate, ṁ is the mass flow rate, and ρ is the density. The density of the solution is calculated by using Equation (4.3) as shown below, keeping in mind that the density of deionized water (DI) is 1 g/ml and x is the concentration percentage of the solution in g/ml (w/v) Density = x Soltion (4.3) 100 The mass flow rate measurements were recorded three times and the average number was taken as the final value. The standard deviation for all the mass flow data was less than 1.9 %. A Petri dish was placed on the surface of the balance for collecting the deposited solution. The sodium alginate solutions were made of Sodium Alginate Manugel DBM powder and DI water in quantities of 75 ml in 100 ml glass containers and were allowed to mix between 2-4 hours. As shown in Equations (4.1) and (4.2), the diameter or pressure increase, the flow rate of the liquid also increases. As the viscosity and length of the liquid delivery

118 96 tube increase, the flow rate of the liquid decreases. However, the equation does not account for the frequency of microvalves. The experimental set-up for the pneumatic, solenoid, and piezoelectric microvalves was schematically shown in Figure 4.1. An air supply that is adjustable to positive and negative pressures was connected to that material delivery system, where sodium alginate aqueous solution was delivered to the microvalve for deposition. The deposited sodium alginate was collected into a Petri dish that was placed on a balance for measuring the mass of the sodium alginate deposited over a specific interval of time to calculate the mass flow rate as expressed by Equation (4.1). The volumetric flow rate was then determined by dividing the mass flow rate by the sodium alginate solution density as described in Equation (4.2). Figure 4.1: Experimental set-up for measurement of flow rate using pneumatic, solenoid, and piezoelectric microvalve systems

119 Pneumatic Valve Deposition The pneumatic microvalve system was used to deposit 1%, 1.5%, 2%, and 3% (w/v) sodium alginate (Manugel) at four different pressures, 8 psi, 16 psi, 24 psi, and 32 psi. The flow rate measurements were conducted using three different nozzle diameters, 250 μm, 330 μm, and 410 μm. Results of the measurement are presented in Figures 4.2 to % sodium alginate aqueous solution 1% (w/v) Manugel Aqueous Solution 1000 Flow Rate (microlitre/second) μm 330 μm 410 μm Pressure (psi) Figure 4.2: Pressure vs. flow rate for 1 % (w/v) sodium alginate aqueous solution

120 98 1% (w/v) Manugel Aqueous Solution 1000 Flow Rate (microlitre/second) psi 16 psi 24 psi 32 psi Nozzle Diameter (microns) Figure 4.3: Nozzle diameter vs. flow rate for 1 % (w/v) sodium alginate aqueous solution Figure 4.2 and Figure 4.3 present a graph of the nozzle diameter versus the flow rate for 1%, (w/v) sodium alginate (Manugel) at four different pressures, 8 psi, 16 psi, 24 psi, and 32 psi. The flow rate measurements were conducted using three different nozzle diameters, 250 μm, 330 μm, and 410 μm. The results show that the flow rate is directly proportional to the pressure and the nozzle diameter.

121 % sodium alginate aqueous solution 1.5% (w/v) Manugel Aqueous Solution Flow Rate (microlitre/second) μm 330 μm 410 μm Pressure (psi) Figure 4.4: Pressure vs. flow rate for 1.5 % (w/v) sodium alginate aqueous solution 1.5% (w/v) Manugel Aqueous Solution Flow Rate (microlitre/second) psi 16 psi 24 psi 32 psi Nozzle Diameter (microns) Figure 4.5: Nozzle diameter vs. flow rate for 1.5 % (w/v) sodium alginate aqueous solution

122 100 Figure 4.5 and Figure 4.5 present a graph of the nozzle diameter versus the flow rate for 1.5%, (w/v) sodium alginate (Manugel) at four different pressures, 8 psi, 16 psi, 24 psi, and 32 psi. The flow rate measurements were conducted using three different nozzle diameters, 250 μm, 330 μm, and 410 μm. The results show that the flow rate is directly proportional to the pressure and the nozzle diameter % sodium alginate aqueous solution 2% (w/v) Manugel Aqueous Solution 320 Flow Rate (microlitre/second) Pressure (psi) 250 μm 330 μm 410 μm Figure 4.6: Pressure vs. flow rate for 2 % (w/v) sodium alginate aqueous solution

123 101 2% (w/v) Manugel Aqueous Solution 320 Flow Rate (microlitre/second) psi 16 psi 24 psi 32 psi Nozzle Diameter (microns) Figure 4.7: Nozzle diameter vs. flow rate for 2% (w/v) sodium alginate aqueous solution Figure 4.6 and Figure 4.7 present a graph of the nozzle diameter versus the flow rate for 2%, (w/v) sodium alginate (Manugel) at four different pressures, 8 psi, 16 psi, 24 psi, and 32 psi. The flow rate measurements were conducted using three different nozzle diameters, 250 μm, 330 μm, and 410 μm. The results show that the flow rate is directly proportional to the pressure and the nozzle diameter.

124 % sodium alginate aqueous solution 3% (w/v) Manugel Aqueous Solution 60 Flow Rate (microlitre/second) μm 330 μm 410 μm Pressure (psi) Figure 4.8: Pressure vs. flow rate for 3% (w/v) sodium alginate aqueous solution 3% (w/v) Manugel Aqueous Solution 60 Flow Rate (microlitre/second) psi 16 psi 24 psi 32 psi Nozzle Diameter (microns) Figure 4.9: Nozzle diameter vs. flow rate for 3% (w/v) sodium alginate aqueous solution

125 103 Figure 4.8 and Figure 4.9 presents a graph of the nozzle diameter versus the flow rate for 3% (w/v) sodium alginate (Manugel) at four different pressures, 8 psi, 16 psi, 24 psi, and 32 psi. The flow rate measurements were conducted using three different nozzle diameters, 250 μm, 330 μm, and 410 μm. The results show that the flow rate is directly proportional to the pressure and the nozzle diameter. It can be seen through the graphs in Figures 4.2 to 4.9 that the flow rate increases with pressure and nozzle diameter. It can also be seen that the flow rate decreases as the sodium alginate concentrations increases when comparing the graphs. 4.3.Solenoid Microvalve Deposition The solenoid microvalve system was used to deposit 0.1%, 0.4%, 0.75%, 0.85%, and 1% (w/v) sodium alginate (Manugel) at different pressures ranging from 8 psi to 20 psi. The flow rate measurements were conducted using three different nozzle diameters at 2, 3, 4, 5, and 7.5 MIL.

126 % sodium alginate aqueous solution 0.1% Sodium Alginate Aqueous Solution with 2 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.10: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 2 MIL 0.1% Sodium AlginateSolution with 3 MIL WC Gaiser 30 Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frquency (Hz) Figure 4.11: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 3 MIL

127 % Sodium Algnate Aqueous Solution with 4 MIL WC Gaiser 60 Flow Rate (micrlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi Frequency (Hz) Figure 4.12: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 4 MIL. 0.1% Sodium Alginate Aqueous Solution with 5 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.13: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 5 MIL.

128 % Sodium Alginate Aqueous Solution with 7.5 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.14: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution at 7.5 MIL. In Figure 4.10 to Figure 4.14, the pressures that were applied were in ascending order to avoid pressure accumulation that could distort the collected data. At each pressure, a frequency range from 1000 Hz to 9.97 Hz was applied. The solenoid microvalve deposited 0.1% (w/v) sodium alginate solution in droplet form. It can be seen from the graphs that the flow rate decreased as the frequency of the microvalve decreased. It can was also observed that the flow rate decreased as the pressure decreased. Some data points were not plotted on the graphs because the deposition did not occur in droplet form and instead was in dripping form. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing

129 107 sodium alginate is a key factor in controlling the size of the scaffold size, strut diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper microdroplet deposition can be determined from the presented data % sodium alginate aqueous solution 0.4% Sodium Alginate Aqueous Solution with 2 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.15: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 2 MIL

130 % Sodium Alginate Aqueous Solution with 3 MIL WC Gaiser 25 Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.16: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 3 MIL 0.4% Sodium Alginate Aqueous Solution with 4 MIL WC Gaiser Flow Rate (microlirte/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.17: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 4 MIL

131 % Sodium Alginate Aqueous Solution with 5 MIL WC Gaiser 60 Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.18: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 5 MIL 0.4% Sodium Alginate Aqueous Solution with 7.5 mills WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.19: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution at 7.5 MIL

132 110 In Figure 4.15 to Figure 4.19, the pressures that were applied were in ascending order to avoid pressure accumulation that could distort the collected data. At each pressure, a frequency range from 1000 Hz to 9.97 Hz was applied. The solenoid microvalve deposited 0.4% (w/v) sodium alginate solution in droplet form. It can be seen from the graphs that the flow rate decreased as the frequency of the microvalve decreased. It can was also observed that the flow rate decreased as the pressure decreased. Some data points were not plotted on the graphs because the deposition did not occur in droplet form and instead was in dripping form. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing sodium alginate is a key factor in controlling the size of the scaffold size, strut diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper microdroplet deposition can be determined from the presented data.

133 % sodium alginate aqueous solution 0.75% Sodium Alginate Aqueous Solution with 2 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.20: Frequency vs. flow rate for 0.75% (w/v) sodium alginate aqueous solution at 2 MIL 0.75% Sodium Alginate Aqueous Solution with 3 MIL WC Gaiser Frequency (Hz) Psi 12 Psi 18 Psi 20 Psi Flow Rate (microlitre/second) Figure 4.21: Frequency vs. flow rate for 0.75% (w/v) sodium alginate aqueous solution at 3 MIL

134 % Sodium Alginate Aqueous Solution with 4 MIL WC Gaiser Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.22: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 4 MIL. 0.75% Sodium Alginate Aqueous Solution with 5 MIL WC Gaiser Flow Rate (microlitre/second) Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.23: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 5 MIL.

135 % Sodium Alginate Aqueous Solution with 7.5 MIL WC Gaiser Frequency (Hz) Psi 18 Psi Flow Rate (microlitre/second) Figure 4.24: Frequency vs. flow rate for 0.75 % (w/v) sodium alginate aqueous solution at 7.5 MIL. In Figure 4.20 to Figure 4.24, the pressures that were applied were in ascending order to avoid pressure accumulation that could distort the collected data. At each pressure, a frequency range from 1000 Hz to 9.97 Hz was applied. The solenoid microvalve deposited 0.75% (w/v) sodium alginate solution in droplet form. It can be seen from the graphs that the flow rate decreased as the frequency of the microvalve decreased. It can was also observed that the flow rate decreased as the pressure decreased. Some data points were not plotted on the graphs because the deposition did not occur in droplet form and instead was in dripping form. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing

136 114 sodium alginate is a key factor in controlling the size of the scaffold size, strut diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper microdroplet deposition can be determined from the presented data % sodium alginate aqueous solution 0.85% Sodium Alginate Aqueous Solution with 3 MIL WC Gaiser Frequency (Hz) Psi 10 Psi Flow Rate (microlitre/second) Figure 4.25: Frequency vs. flow rate for 0.85 % (w/v) sodium alginate aqueous solution at 3 MIL.

137 % Sodium Alginate Aqueous Solution with 4 MIL WC Gaiser 35 Flow Rate (microlitre/second) Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.26: Frequency vs. flow rate for 0.85 % (w/v) Sodium Alginate aqueous solution at 4 MIL. 0.85% Sodium Alginate Aqueous Solution with 5 MIL WC Gaiser Flow Rate (microlitre/second) Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.27: Frequency vs. flow rate for 0.85 %(w/v) Sodium Alginate aqueous solution at 5 MIL.

138 % Sodium Alginate Aqueous Solution with 7.5 MIL WC Gaiser Frequency (Hz) Psi 20 Psi Flow Rate (microlitre/second) Figure 4.28: Frequency vs. flow rate for 0.85 % (w/v) Sodium Alginate aqueous solution at 7.5 MIL. In Figure 4.25 to Figure 4.28, the pressures that were applied were in ascending order to avoid pressure accumulation that could distort the collected data. At each pressure, a frequency range from 1000 Hz to 9.97 Hz was applied. The solenoid microvalve deposited 0.85% (w/v) sodium alginate solution in droplet form. It can be seen from the graphs that the flow rate decreased as the frequency of the microvalve decreased. It can was also observed that the flow rate decreased as the pressure decreased. Some data points were not plotted on the graphs because the deposition did not occur in droplet form and instead was in dripping form. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing sodium alginate is a key factor in controlling the size of the scaffold size, strut

139 117 diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper microdroplet deposition can be determined from the presented data. No data at all was collected for the 2 MIL because the sodium alginate solution was dripping and flow rate calculation was not possible % sodium alginate aqueous solution 1% Sodium Alginate Aqueous Solution with 3 MIL WC Gaiser Flow Rate (microlitre/second) Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.29: Frequency vs. flow rate for 1 % (w/v) Sodium Alginate aqueous solution at 3 MIL.

140 118 1% Sodium Alginate Aqueous Solution with 4 MIL WC Gaiser 35 Flow Rate (microlitre/second) Psi 10 Psi 12 Psi 14 Psi 16 Psi 18 Psi 20 Psi Frequency (Hz) Figure 4.30: Frequency vs. flow rate for 1 % (w/v) Sodium Alginate aqueous solution at 4 MIL. In Figure 4.29 and Figure 4.30, the pressures that were applied were in ascending order to avoid pressure accumulation that could distort the collected data. At each pressure, a frequency range from 1000 Hz to 9.97 Hz was applied. The solenoid microvalve deposited 1% (w/v) sodium alginate solution in droplet form. It can be seen from the graphs that the flow rate decreased as the frequency of the microvalve decreased. It can was also observed that the flow rate decreased as the pressure decreased. Some data points were not plotted on the graphs because the deposition did not occur in droplet form and instead was in dripping form. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing sodium alginate is a key factor in controlling the size of the scaffold size, strut

141 119 diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper microdroplet deposition can be determined from the presented data. No data at all was collected for the 2, 5, and 7.5 MIL because the sodium alginate solution was dripping and flow rate calculation was not possible. 4.4.Piezoelectric Microvalve Deposition % sodium alginate aqueous solution 0.1% Sodium Alginate Aqueous Solution with 50 microns Orifice 1.4 Flow Rate (microlitre/second) V 60 V 65 V 70 V Frequency (Hz) Figure 4.31: Frequency vs. flow rate for 0.1 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter

142 % Sodium Alginate Aqueous Solution with 50 microns Orifice 1.4 Flow Rate (microlitre/second) Hz 1000 Hz 1500 Hz 2000 Hz Voltage (v) Figure 4.32: Voltage vs. voltage for 0.1 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figures 4.31 and 4.32 present the flow rate versus the frequency and the flow rate versus the voltage, respectively, for 0.1% sodium alginate aqueous solution at four positive and negative voltages of 55, 60, 65, and 70 volts and at frequencies of 500, 1000, 1500, and 2000 Hz. It can be seen that the flow rate increases as the frequency of the piezoelectric nozzle increases. It can also be seen from both figures that the applied voltage is directly proportional to the flow rate of the deposited sodium alginate solution. The increase in flow rate due to voltage at every frequency increases at almost relative proportions and can be seen more clear when studying the chart in the appendix. The constant parameters used in Figures 4.31 and 4.32 are shown in Table 4.1.

143 121 Table 4.1: Operating Parameters for 0.1% sodium alginate deposition using the piezoelectric microvalve Characteristics and comparison of the three nozzle systems Nozzle Holding Pressure T rise T dwell T fall T echo T final MJ-SF psi 7 μs 15 μs 12.5 μs 30 μs 7 μs % sodium alginate aqueous solution 0.25% Sodium Alginate Aqueous Solution with 50 microns Orfice 1.4 Flow Rate (microlitre/second) V 60 V 65 V 70 V Frequency (Hz) Figure 4.33: Frequency vs. flow rate for 0.25 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter

144 % Sodium Alginate Aqueous Solution with 50 microns Orifice 1.4 Flow Rate (microlitre/second) Hz 2000 Hz Voltage (v) Figure 4.34: Frequency vs. voltage for 0.25 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figures 4.33 and 4.34 present the flow rate versus the frequency and the flow rate versus the voltage, respectively, for 0.25% sodium alginate aqueous solution at four positive and negative voltages of 55, 60, 65, and 70 volts and at frequencies of 1500 and 2000 Hz. It can be seen that the flow rate increases as the frequency of the piezoelectric nozzle increases. It can also be seen from both figures that the applied voltage is directly proportional to the flow rate of the deposited sodium alginate solution. The increase in flow rate due to voltage at every frequency increases at almost relative proportions and can be seen more clear when studying the chart in the appendix. The constant parameters used in Figures 4.33 and 4.34 are shown in Table 4.2.

145 123 Table 4.2: Operating Parameters for 0.25% sodium alaginate deposition using the piezoelectric microvalve Nozzle Holding Pressure T rise T dwell T fall T echo T final MJ-SF psi 7 μs 25 μs 12.5 μs 50 μs 7 μs % sodium alginate aqueous solution 0.4% Sodium Alginate Aqueous Solution with 50 microns Orifice 1.4 Flow Rate (microlitre/second) V 60 V 65 V 70 V Frequency (Hz)) Figure 4.35: Frequency vs. flow rate for 0.4 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter

146 % Sodium Alginate Aqueous Solution with 50 microns Orifice 1.4 Flow Rate (microlitre/second) Hz 1000 Hz 1500 Hz 2000 Hz Voltage (v) Figure 4.36: Frequency vs. voltage rate for 0.4 % (w/v) sodium alginate aqueous solution with 50 microns nozzle diameter Figures 4.35 and 4.36 present the flow rate versus the frequency and the flow rate versus the voltage, respectively, for 0.4% sodium alginate aqueous solution at four positive and negative voltages of 55, 60, 65, and 70 volts and at frequencies of 500, 1000, 1500, and 2000 Hz. It can be seen that the flow rate increases as the frequency of the piezoelectric nozzle increases. It can also be seen from both figures that the applied voltage is directly proportional to the flow rate of the deposited sodium alginate solution. The increase in flow rate due to voltage at every frequency increases at almost relative proportions and can be seen more clear when studying the chart in the appendix. The constant parameters used in Figures 4.35 and 4.36 are shown in Table 4.3.

147 125 Table 4.3: : Operating Parameters for 0.4% sodium alaginate deposition using the piezoelectric microvalve Nozzle Holding Pressure T rise T dwell T fall T echo T final MJ-SF psi 7 μs 25 μs 12.5 μs 50 μs 7 μs 4.5.Discussion on the Deposition Systems Experiments were conducted to measure the flow rate of the deposited sodium alginate solutions at various parameters to produce a wide feasibility range for sodium alginate aqueous solutions deposition through pneumatic, solenoid, and piezoelectric and the microvalves by revealing working ranges and limitations. In general, the experiments showed that the flow rate was directly proportional to the nozzle diameter and inversely proportional to the viscosity of the sodium alginate aqueous solution. The pneumatic microvalve deposited sodium alginate in extrusion form. The pneumatic microvalve is capable of depositing sodium alginate concentrations up to 3% (w/v) using nozzle diameters of 250 μm, 330 μm, and 410 μm and pressures ranging from 8 psi to 32 psi. The deposition flow rate is directly proportional to the

148 126 operating pressure and the nozzle diameter, however, is indirectly proportional to the sodium alginate concentration. The solenoid microvalve deposited sodium alienate in droplet form. The solenoid microvalve is capable of depositing sodium alginate concentrations up to 1% (w/v). The operating pressure range was 8 psi to 20 psi and the applied frequencies ranged from Hz to 1000 Hz. The deposition flow rate is directly proportional to the operating pressure and the nozzle diameter, however, is indirectly proportional to the sodium alginate concentration. Some of the parameters produced undesired position in the form of large droplets referred to as nozzle dripping. The dripping form of sodium alginate could not be used for the fabrication of scaffolds since there is no controllability on the deposition. The controllability of depositing sodium alginate is a key factor in controlling the size of the scaffold size, strut diameter, and the strut distance from one another. Therefore, only the deposition of alginate in the form of micro-droplets was considered for the deposition feasibility study of the solenoid microvalve. Although nozzle dripping cannot be used for the structure formation of scaffolds, the parameter boundary limits of proper micro-droplet deposition can be determined from the presented data. It is observed that the deposition flow rate is directly proportional to the operating pressure, nozzle diameter, and frequency. However, is indirectly proportional to the sodium alginate concentration. Nozzle dripping occurred more frequently at relatively higher concentrations of sodium alginate, lower operating pressures, and lower frequencies. The MJ-SF piezoelectric nozzle was used to deposit aqueous solutions of sodium alginate at concentrations of 0.1%, 0.25%, and 0.4% (w/v). The flow rates

149 127 were calculated by dividing the mass flow rate by the solution density. Measuring the weight of the deposited solution via a balance and dividing it by the deposition time obtained the mass flow rate. The solutions were tested to find the operating windows in which micro-droplets could be generated. It can be seen from tables 4.1 and 4.2 that the constant parameters were the same except for the T dwell and T echo. This is because each fluid has different properties that make them require different energy pulses at different timings in order to form harmonic distractions at the nozzle tip that could cause the microdroplet deposition. The 0.4% (w/v) concentration of aqueous solution of sodium alginate was unable to form microdroplets within the operating parameters of the 0.1% (w/v) concentration of aqueous solution of sodium alginate. The both solutions had different holding pressures due to the difference in their viscosities and densities. The 0.1%, 0.25%, and 0.4% (w/v) sodium alginate aqueous solutions had their flow rates calculated at frequencies of 500, 1000, 1500, and 2000 Hz at four applied positive and negative voltages of 55, 60, 65, and 70 volts. It was seen that the flow rate increases as the frequency of the piezoelectric nozzle increases. Also, the applied voltage was directly proportional to the flow rate of the deposited sodium alginate solution. The increase in flow rate due to voltage increased at almost relative proportions at every frequency and can be seen more clear when studying the chart in the appendix. In addition, when the two different concentrations were compared at identical frequency and voltage, the 0.4% (w/v) sodium alginate solution showed to have a lower flow rate than the 0.1% sodium alginate solution. This makes practical

150 128 sense since the higher concentration solution has a higher viscosity than the lower concentrations solution. The feasibility study showed that the pneumatic microvalve, solenoid, and piezoelectric microvalves could deposit sodium alginate of different concentrations and modes of deposition. The pneumatic system deposits sodium alginate in extrusion mode and up to relatively high concentrations [3% (w/v)] compared to the other two systems. The solenoid and piezoelectric microvalves deposit sodium alginate in micro-droplet mode and not in extrusion form. The solenoid system can deposit sodium alginate concentrations up to 1% (w/v), while the piezoelectric system can deposit sodium alginate concentrations up to 0.4% (w/v). This means that alginate scaffolds are limited in their mechanical properties according to which system is being used for the fabrication process. Scaffolds made of higher concentrations of alginate can produce alginate scaffolds with the high mechanical properties. In this case, the pneumatic system has the capability of producing scaffolds with the most enhanced mechanical properties. The operating parameters vary on the three deposition systems. The pneumatic valve operates by varying the pressure and nozzle diameter while the solenoid microvalve operates using the same parameters as the pneumatic valve with the addition the solenoid frequency. The piezoelectric microvalve operates on the nozzle diameter, holding pressure, and voltage. This makes the pneumatic valve have one less operating parameter than the solenoid piezoelectric microvalves, which makes the pneumatic system more attractive for controlling the deposition system when fabricating scaffold.

151 129 The valve mechanics have a direct impact on the fluids passing through the internal fluid passages inside the valves. The solenoid microvalve has a valve that opens and closes at high frequency as the fluid is being deposited. This action may impede intense mechanical stresses on any living cell that passes through the valve during a bioactive fabrication process. The piezoelectric microvalve acts as a continuously open tube, however, the motion of the material that narrows and widens the tube vibrates at high frequency that may also impact living cells during deposition. In addition, the piezoelectric microvalve has a maximum diameter of 70 μm, which is very narrow to cells with a size range of 20 to 30 μm especially when the sodium alginate is deposited in the form of micro-droplets. On the other hand, the pneumatic microvalve extrudes sodium alginate without any mechanic stress on the fluid other than the applied pressure and the closing of the valve when the vale closes. This type of situation appears to be ideal for living cells that are to be nozzle system. The feasibility study concluded that the development of a bioactive fabrication system capable of depositing live cells favored the pneumatic microvalve system.

152 STRUCTURAL FORMATION OF 3D ALGINATE 5.1. Scaffold Design and Architecture Two case studies were performed to study the structural formability of scaffolds according to the designed configurations. Case I study was conducted to study the structural formability according to the designed strut spacing varying from 500 μm, 670 μm, 1000 μm, and to fabricate scaffolds with pore sizes of 750 μm, 420 μm, and 250 μm, respectively, at a constant strut diameter of 250 μm as shown in Figure 5.1. Case II was conducted to study the structural formability according to the designed constant strut spacing with varying strut diameters at 250 μm, 330 μm, and 250 μm to fabricate pore sizes of 750 μm, 670 μm, and 590 μm, respectively, at a constant strut distance of 1000 μm as shown in Figure 5.2. The strut diameters were obtained using their optimum velocities (v N ). Figure 5.1: Schematic diagram of case I were the effect of strut distance on the pore size at 250 micron constant strut diameter was studied.

153 131 Figure 5.2: Schematic diagram of case II were the effect of the strut diameter on the pore size at 1000 micron constant strut distance was studied. The scaffolds were designed to have parallel struts at each layer while alternating rotating every layer by 90 degrees as shown in Figure 5.3. The spacing of the struts was 1 mm apart and the size of each layer was 10 mm x 10 mm. The diameter of the nozzle tip was 250 μm and the velocity v N and pressure were 10.5 mm/s and 16 psi, respectively, based on the calculated velocity in Table 5.1. The pneumatic valve is closed every time a straight strut line is completed and opened again when a formation of a new straight strand line begins. The excess material deposited each time the valve closes forms the connection between the straight lines. This excess material deposition is caused by the valve spring that pushes the valve pin onto the valve seat during the closing of the valve. The key process parameters for the experiments were listed in Table 5.1.

154 132 Table 5.1: Process Parameters Parameter Pressure Nozzle Velocity Deposition Surface Strand Length Distance between Strands Sodium Alginate Concentration Calcium Chloride Concentration Nozzle Diameter Value 16 psi 10.5 mm/s Petri Dish 10 mm 1 mm 3% (w/v) 0.5-3% (w/v) 250 μm 5.2. Early Deposition Termination The pneumatic valve experiences some excess material deposition during the closing of the valve. This is caused by the spring that pushes the valve pin onto the valve seat. This phenomenon was studied by introducing an early deposition termination (EDT) parameter, which can be used to close the valve prior to the structure final length. Figure 5.3 shows the EDT at a distance of 2 mm prior to the structure final intended length. The EDT for the scaffolds was tested at 0 mm/s, 1 mm/s, and 2 mm/s.

155 133 Figure 5.3: (A) Excess material deposited when vale is closed; (B) EDT for a 2mm distance between vale closing and end of nozzle travel 5.3. System set-up and Dual Deposition of Sodium Alginate and Calcium Chloride Solution for cross-linking Two pneumatic valves were simultaneously operated for performing heterogeneous deposition in the development of the 3D alginate scaffolds. The first valve was used for the sodium alginate solution, and the second for the calcium chloride solution. The sodium alginate solution was extruded through the pneumatic valve nozzle and is deposited into a calcium chloride reservoir. The calcium chloride reservoir is adjusted to a level that is almost equal to the vertical thickness of the sodium alginate continuous deposition. The calcium chloride deposition is performed using a second pneumatic valve or by the spray valve. The sodium alginate is continuously deposited in the calcium chloride reservoir. Simultaneously, the nozzle tip moves in the desired directions at the same horizontal level depending on the desired pattern. The sodium alginate solution forms into a cylindrical shape

156 134 deposition and is permanently geometrically stable in the crosslinking solution to form a hydrogel. After the first layer has been formed, the level of the calcium chloride solution in the reservoir is increased one by one increment of the sodium alginate deposition vertical thickness. The tip of the syringe system is also raised by the same increment to deposit a second layer of sodium alginate over the previous layer in the calcium chloride solution. The layer of calcium chloride solution and the nozzle tip where programmed to move in the z vertical axis with increments of 240 μm. The deposition technique is illustrated in Figure 5.4 and Figure 5.5. The level of the calcium chloride is controlled by timing the deposition (t). This is accomplished by dividing the intended volume of the solution by the flow rate, which is expressed in Equation (5.1); V t = = Q Volume FlowRate (5.1)

157 135 Figure 5.4: First Layer Deposition Figure 5.5: Second Layer Deposition The two nozzles, one for the sodium alginate deposition and the other for the calcium chloride deposition, were both fixed on to the same moving arm that placed

158 136 the nozzles in 3D according the given toolpath. Two methods of depositing calcium chloride were used to fabricate the alginate structures. The first technique was the dynamic nozzle deposition, which is based on displacing the calcium chloride nozzle vertically above the alginate structure. As this technique is performed, the sodium alginate nozzle is displaced away for its final deposition point by an offset equal to the calcium chloride displacement as the calcium chloride is deposited. Once the chloride has finished its deposition, the two nozzles will move back to their original places to start a new layer deposition of sodium alginate. The second technique does not involve the movement of the calcium chloride nozzle above the alginate structure, but instead, it remains stationary at its place above the Petri dish where the alginate scaffold is being fabricated in to deposit the calcium chloride Flow Rate Control using Process Model An analytical model for expressing flow rate of non-newtonian was used to determine the pressure gradient dp/dz. We first substitute the K and n values for a sodium alginate solution by experimental viscosity measurements from chapter three into Equation (3.2). Since the pressure gradient is independent of the sodium alginate concentration, any concentration of sodium alginate can be used. In this case, sodium alginate with a concentration of 3% (w/v) was chosen for the determination of the pressure gradient of the pneumatic valve. The shear rate γ was arbitrarily chosen as 50 s -1 and inserted in to Equation (3.26) with K and n to find η 0 = The only parameter left for analytically expressing the flow rate for 3% (w/v) sodium alginate

159 137 is the pressure gradient dp/dz in Equation (3.33). The pressure gradient is a unique value for any deposition system and is a function of the nozzle diameter, pressure, and the internal geometry that the biopolymer passes through the whole valve system. The experimental flow rate measurements for 2%, 3%, and 4% (w/v) sodium alginate solutions were inserted in to Equation (3.33) using the determined values for K and n for every concentration to solve for the pressure gradients dp/dz using nozzle diameters of 250 μm, 330 μm, and 410 μm. The average pressure gradients of the three sodium alginate solutions are shown in Figure 5.6. The average pressure gradients were then used in the analytical expression for the flow rate using 3% (w/v) sodium alginate and compared to the experimental results as shown in Figure 5.7.

160 138 Average Na-Alginate Aqueous Solution Concentration (2%, 3%, 4%) (w/v) with Pneumatic Microvalve 1E+11 dp/dz (pa/m) 9E+10 8E+10 7E+10 6E+10 5E+10 4E+10 3E+10 2E+10 1E Pressure (psi) D = 410 microns D = 330 microns D = 250 microns Figure 5.6: Results of the average pressure gradients dp/dz of 2%, 3%, and 4% (w/v) Manugel solutions versus pressure for nozzle diameters of 250 μm, 330 μm, and 410 μm with the pneumatic valve 3% (w/v) Sodium Alginate Aqueous Solution with Pneumatic Microvalve Flow Rate (microlitre/second) D = 410 microns (Experimental) D = 410 microns (Analitical) D = 330 microns (Experimaental) D = 330 microns (Analytical) D = 250 microns (Experimental) D = 250 microns (Analytical) Pressure (psi) Figure 5.7: Comparison of the experimental and analytical flow rate versus pressure using the average pressure gradients dp/dz of 2%, and 3%, and 4% (w/v) Manugel solutions for 3% (w/v) sodium alginate solution with nozzle diameters of 250 μm, 330 μm, and 410 μm using the pneumatic valve

161 Strut Diameter Control using Process Model As mentioned earlier in section 3.3, the nozzle velocity v N is used to fabricate alginate struts to avoid strut misplacement due to slippage. However, one limitation for the multi-nozzle deposition system is to not exceed the nozzle travel speed above 20 mm/s in order to avoid the mechanical vibrations that could interfere with the geometry of the fabricated scaffold. By looking into Equation (3.41), we can see that the operating parameters may be tuned to obtain flow rates that are suitable for depositing sodium alginate at a v N value less than 20 mm/s. This was done for various operating parameters and nozzle diameters to calculate tha v N value for 3% (w/v) sodium alginate. The results are presented in Table 5.2 and a plot of the results is also shown in Figure 5.8, which presents v N versus the pressure for 3% (w/v) sodium alginate aqueous solution at 25 o C using the pneumatic valve. The bold numbers in Table 5.2 are the v N values that satisfy the 20 mm/s criterion. Table 5.2: Results of the nozzle velocity and feasibility study for pneumatic valve using 3% (w/v) sodium alginate aqueous solution at various pressures and nozzle diameters D (μm) Area (m2) 8 psi 16 psi 24 psi 32 psi E E E E E E

162 140 3% (w/v) Sodium Alginate Aqeous Solution 600 Nozzle Velocity (mm/s) μm 150 μm 200 μm 250 μm 330 μm 410 μm Pressure (psi) Figure 5.8: Results of v N versus the pressure for 3% (w/v) sodium alginate aqueous solution at 25 o C using the pneumatic microvalve system The multi-nozzle deposition system using the pneumatic microvalve system was set to fabricate alginate struts to investigate experimentally the validation of the strut diameter process model introduced in chapter three Equation (3.40) by comparing the experimental strut diameters to the computed results. The effect of change in nozzle velocity v on the strut diameter D at a constant volumetric flow rate was selected using 3% (w/v) sodium alginate with a nozzle diameter of 250 μm and an operating pressure of 16 psi to give a constant flow rate of 0.51 μl/s. The v N was also obtained by using Equation (3.41) using these operating parameters and nozzle diameter, which is equal to 10.5 mm/s at these conditions. The three velocities that were set for this experiment were 5, 10, and 15 mm/s.

163 141 The alginate strands were removed form the Petri dish they were fabricated in and put immediately under an optical microscope for image capturing and their images were then measured using Image J 1.30 software. The mean diameter was recorded for each of the three strands, which were μm, μm, and μm for the 5 mm/s, 10 mm/s, and 15 mm/s nozzle velocity, respectively. Figure 5.9 presents images of the alginate struts fabricated at the three different velocities (A) micron (B) micron (C) 500 micron Figure 5.9: Alginate struts images fabricated at the three different velocities; (A) 5 mm/s; (B) 10 mm/s; 15 mm/s (C) Table 5.3 presents the computed strut diameters results. Figure 5.10 presents the results of the computed and experimented strut diameters under the three different velocities. In comparing these experimental results to the computed results in Table 5.2 the error was found to be 1.96%, 3.87%, and 2.49% for the 5 mm/s, 10 mm/s, and 15 mm/s nozzle velocity, respectively.

164 142 Table 5.3: Results of the process model to predict the alginate strut diameter Velocity (mm/s) Flow Rate (μl/s) Diameter (μm) Alginate Strut Diameter vs. Nozzle Velocity at Constant Flow Rate Alginate Strand Diameter (microns) Nozzle Velocity (mm/s) Experimental Model Figure 5.10: Comparison of alginate strut diameters at nozzle velocities of 5 mm/s, 10 mm/s, and 15 mm/s with a constant flow rate of 0.51 μl/s for the process model and experiments

165 Pore Size and Porosity Control using Process Model As mentioned in section 5.1, Case study I was conducted to study the effect of strut distance on the pore size at a constant strut diameter. The chosen strut distances were 1000 μm, 670 μm, and 500 μm to fabricate pore sizes of 750 μm, 420 μm, and 250 μm, respectively, at a constant strut diameter of 250 μm. The results showed that the pore sizes and strut diameters for all strut distances were similar to that estimated as shown in Figure Case II was conducted to study the effect of the strut diameter on the pore size at a constant strut distance. The chosen strut diameters were 250 μm, 330 μm, and 250 μm to fabricate pore sizes of 750 μm, 670 μm, and 590 μm, respectively, at a constant strut distance of 1000 μm. The experimental results showed that the pore sizes and strut diameters were similar to that estimated as shown in Figure Figure 13A and Figure 13B show micro and macro images of a 40 layer Case I 1000 μm strut distance scaffold using a 250 μm nozzle tip diameter using velocity v N and pressure 10 mm/s and 16 psi, respectively. The distance recorded between the struts was found to be 960 μm and the length of the square pore was found to be 720 μm for the scaffolds. The struts showed uniform geometry at a diameter of 250 μm.

166 144 Alginate Scaffold Variable Pore Size at Constant Strut Diameter (250 micron) SIze (Micron) Experimental Pore Applied Pore Experimental Strut Diameter Applied Strut Diameter Strut Distance (Micron) Figure 5.11: Results of Case I pore sizes and strut diameters for 1000 μm, 670 μm, and 500 μm strut distances. Alginate Scaffold Variable Pore Size at Constant Strut Distance (1000 micron) SIze (Micron) Experimental Pore Applied Pore Experimental Strut Diameter Applied Strut Diameter Strut Diameter Diameter (Micron) Figure 5.12: Results of Case II pore sizes and strut diameters for 750 μm, 670 μm, and 590 μm strut diameters

167 145 Figure 5.13: (A) Close up image of pore for 3D alginate scaffold; B) Overall image of 3D alginate scaffold The level of calcium chloride was increased for every alginate scaffold layer by an increment equal to the layer thickness. The level was achieved by using the pneumatic microvalve and spray valve for calcium chloride deposition. The process was timed to adjust the level to the appropriate value. Figure 5.14 presents images of a 40-layer alginate scaffold after the end of fabrication in the Petri dish where the level of calcium chloride is almost equal to the height of the scaffold. Figure 5.14: 40 layer scaffold (A) angle view ; (B) side view; (C) top view

168 Discussion on 3D Alginate Scaffolds For the scaffold fabrication experiments, the nozzle tip diameter was 250 μm and the velocity v N and pressure were 10 mm/s and 16 psi, respectively, based on the calculated velocity in Table 5.2 using Equation (3.41). Experiments were set to investigate the effect of the EDT on the geometrical structure of 3D alginate scaffolds. The first EDT tested was at 0 mm/s, which means that the valve was closed at the very end of the formation of every straight line. The results were analyzed using an optical digital camera that viewed that excess material was deposited at the curved line structures at the edge perimeter of each layer. It was observed that the diameter of the curved sections (310 μm) were much larger than the straight lines (270 μm) shown in Figure 5.15A. As a result, the 3D alginate structure was thicker around the edges and caused concaving of the scaffold after 20 layers as shown in Figure 5.15B. (A) (B) Figure 5.15: (A) Excess material deposition during closing of the valve in curved portions of the scaffold at 0 mm EDT; (B) Concave 3D alginate scaffold at 0 mm

169 147 The second EDT tested was at 1 mm/s so that the valve would close 1 mm prior to the end point of the straight-line formation. The image in Figure 5.16A reveals that the diameters of the curved portion of the alginate strut is almost equal to the straight struts. As a result, the 3D alginate structure is fabricated uniformly up to 40 layers without geometrical defects as shown in Figure 5.16B. The third and last EDT was at 2 mm/s to have the valve close 2 mm prior to the end point of the straight-line formation. It was observed through the images in Figure 5.17A and Figure 5.17B that the lines formed were not placed as was originally designed and as a result constructed an irregularly shaped alginate 3D scaffold. This resulted from the relatively large EDT (2 mm/s) that caused dragging of the alginate strands every time the valve was closed. The EDT at this parameter forced the nozzle tip without any deposition for a particular distance. The alginate strut was attached to the nozzle tip because of the calcium chloride crosslinking solution during the time where the nozzle was traveling without any deposition. The dynamic technique produced inconsistent layers because of the pneumatic valve movement, which drags some residual alginate material and disrupts the layer design. The dragging of the alginate strand occurs primarily because of the calcium chloride solution gels the tip of the pneumatic valve and therefore keeps the alginate strand connected to the nozzle tip when the pneumatic valve is closed. On the other hand, the stationary technique produced uninterrupted layers of alginate that had the strands placed fixed to its original design. The stationary technique was more successful than the dynamic although the strand in connected to the nozzle tip when the pneumatic valve is closed because there is no movement of the nozzle tip while

170 148 calcium chloride solution is deposited, which could lead to the dragging of the strut. The movement of the pneumatic valve plays a major role in the fabrication process while the calcium chloride is being deposited through the spray nozzle. Figure5.18A and Figure 5.18B show results of the dynamic and stationary processes, respectively. (A) (B) Figure 5.16: (A) Ideal structure formation around bends at 1 mm EDT; (B) Uniformly shaped 3D alginate scaffold at 1 mm EDT

171 149 (A) (B) Figure 5.17: (A) Irregular material deposition placement of the scaffold at 2 mm EDT; (B) Irregularly shaped 3D alginate scaffold at 2 mm (A) (B) Figure 5.18: (A) Dynamic nozzle technique; (B) Stationary nozzle technique

172 150 Under the conditions of fabricating the scaffold with EDT 1mm and using stationary mode, unique scaffolds were fabricated. The diameter of the nozzle tip was 250 μm and the velocity v N and pressure were 10 mm/s and 16 psi, respectively. The distance recorded between the struts was found to be 920 μm and the length of the square pore was found to be 650 μm for the scaffolds fabricated with EDT at 1 mm. The pressure gradient of the pneumatic valve was determined for operating pressures of 8 psi, 16psi, 24psi, and 32 psi using nozzle diameters of 250 μm, 330 μm, and 410 μm. The pressure gradient is independent of the sodium alginate concentration and increases with the operating pressure and the nozzle diameter. The pressure gradient was determined and averaged using sodium alginate concentrations of 2%, 3%, and 4% (w/v). The determined pressure gradient for the pneumatic valve system was used for determining the flow rate analytically under some given operating parameters. Subsequently, the process model for determining the alginate strut diameter was tested by comparing the experimental model analytical results. The results proved that the model showed significantly similar results to that of the experimental and can be controlled by tuning the operating parameters and the nozzle speed velocity. The pressure gradient that has been determined for the pneumatic valve can be used for all other sodium alginate concentrations such as the 1%, 1.5% and 2% (w/v). It was also observed that the alginate struts were fabricated with the 1.5%, 2% and 3% (w/v) sodium alginate, however, the 1% (w/v) was unable to fabricate uniform structures of alginate gel scaffolds. It can be concluded that the structure formability

173 151 of Manugel scaffolds can be decreases with the decrease in sodium alginate concentration. In this case, the 1.5% (w/v) was the critical boundary to which uniform structure of alginate scaffolds could be fabricated using calcium chloride concentrations in the range of 0.5%-3%. The distanced alginate molecules from one another can explain this phenomenon since inefficient crosslinking occurs in dilute solutions of sodium alginate. The calcium chloride concentration varied from 0.5% to 3% (w/v) during the fabrication process of the alginate scaffolds. The strut structure formation did not change with the change in the calcium chloride concentrations; however, the strut cut-off from the nozzle tip was greatly affected. The struts occasionally attached to the nozzle tip at relatively high concentrations of calcium chloride due to the high gelation rate of alginate. By contrast, the struts were able to detach from the nozzle tip when the valve closed at relatively low concentrations of calcium chloride because of the low gelation rate. Although the lower concentrations of calcium chloride have a lower gelation rate than the higher concentrations, the struts were sensibly formed in to a cylindrical structure that would eventually crosslink over time to become mechanically enhanced without changing the shape of the strut. As a result, it was concluded that the relatively low concentrations of calcium chloride were more efficient for constructing alginate scaffolds with sodium alginate concentrations varying from 1% - 3% (w/v) sodium alginate.

174 BIOACTIVE CELL DEPOSITION OF 3D ALGINATE SCAFFOLD CONSTRUCTS 6.1. The Study of Alginate Crosslink on Cell Viability Materials and Methods Experiments were conducted to investigate the effect of sodium alginate and calcium chloride concentrations on cell viability during and after the process, as well as, the cell proliferation. Sodium alginate was prepared with concentrations of 1%, 1.5%, 2%, and 3% (w/v). The sodium alginate solutions were sterilized using a series of filters with mesh sizes of 5 μm, 1.2 μm and 0.45 μm. Calcium chloride reagent grade (Sigma, St. Louis, MO) was prepared at concentrations of 0.5%, 1%, and 2% (w/v). Gel samples were prepared in a cell culture 24-wellplate. Each well contained 750 μl of sodium alginate and was crosslinked by adding 1.5 ml to each well. Six samples were tested and their results were averaged to represent the value for each specimen. The proliferation of cells was monitored using fluorescent cytofluoremetry by adding Alamar Blue dye (Biosource International, CA) to the gel samples. The Alamar Blue assay could be repeated to the same samples over a number of times. The Alamar Blue assay incorporates a fluorometric growth indicator based on metabolic activity. As cells are being tested, innate metabolic activity results in a chemical reduction of Alamar Blue. Continued growth maintains a reduced environment while inhibition of growth maintains an oxidized environment. Reduction related to growth causes the indicator to change from oxidized state (non-

175 153 fluorescent, blue) to reduced state (fluorescent, red). Samples were tested by removing medium from the wells and adding 1 ml of fresh medium and 100 μl of Alamar Blue to each well with the alginate sample. The well plate was then incubated for 3 hours and then 400 μl of Alamar Blue/medium was placed in a fresh 24- wellplate prior to fluorescent reading using a Tecan GENios (Tecan U.S, NC). The alarm Blue assay was conducted on day 0, 4, 7, 14, and 21. Two experiments were conducted. The first experiment was to investigate the effect of the sodium alginate concentration on the cell proliferation encapsulated in the alginate samples. This experiment used sodium alginate concentrations of 1%, 1.5%, 2%, and 3% (w/v) at constant crosslinking concentration of 0.5% (w/v) calcium chloride to make gel samples as described above. The second experiment was conducted to investigate the calcium chloride concentration effect on the cell proliferation encapsulated in the alginate samples with the constant alginate concentration. The crosslinking concentration of 0.5%, 1%, and 2%, (w/v) were used and the gelation time was 10 minutes before removing the calcium chloride solution from the wells and adding 1 ml of medium. Rat Heart Endothelial cells (RHEC), passages were used to study cell proliferation in the alginate gel samples. The cell culture medium consisted of low glucose with 2mM l-glutamine DMEM, 10% FBS, and Antibiotic/antimycotic solution (10,000 IU Penicillin, 10,000µg/mL Streptomycin and 25µg/mL Amphotericin B). RHEC with a 500,000 cells/ml concentration were mixed with 0.5% saline and sodium alginate. The cell culture medium protocol is presented in Appendix E. Cytotoxicity assessments were conducted on the gels using

176 154 LIVE/DEAD assay (Invitrogen Corporation, CA) for cell viability study and images were taken under a Leica DMIL fluorescent microscope. Calcein AM (live) and ethidium homodimer (EthD-1) (dead) can be viewed simultaneously with a conventional fluorescein longpass filter. The fluorescence from these dyes may also be observed separately; calcein can be viewed with a standard fluorescein bandpass filter and EthD-1 can be viewed with filters for propidium iodide or Texas Red dye Results of Alginate Crosslink Cell Viability The first experiment was conducted to investigate the effect of the sodium alginate concentration on the cell proliferation encapsulated in the alginate samples. Results of the experiment are presented in Figures 6.1 and 6.2. Controls were used in the experiment by using alginate without cell encapsulation under the same conditions. This was conducted to validate that no contamination occurred during the experiment. It can be seen from the data that the metabolic activity of the cell encapsulated gels increased over 3 weeks of incubation for all sodium alginate concentrations.

177 155 Fluorescent Reading vs. Incubation Day with 0.5% Calcium Chloride Fluorescent Reading Day 1% 1.50% 2% 3% 1% control 1.5% control 2% control 3% control Figure 6.1: Results of the fluorescent reading versus incubation day for 1%, 1.5%, 2%, and 3% (w/v) sodium alginate solutions under constant 0.5% (w/v) calcium chloride. Fluorescent Reading vs. Incubation Day with 0.5% Calcium Chloride Fluorescent Reading (Fraction Initial) Day 1% 1.50% 2% 3% 1% control 1.5% control 2% control 3% control Figure 6.2: Results of the fraction initial fluorescent reading (normalized) versus incubation day for 1%, 1.5%, 2%, and 3% (w/v) sodium alginate solutions under constant 0.5% (w/v) calcium chloride.

178 156 The second experiment was conducted to investigate the calcium chloride concentration effect on cell proliferation encapsulated in the alginate samples and the results of the experiment are presented in Figures 6.3 and 6.4, respectively. Controls were used in the experiment by using alginate without cell encapsulation under the same conditions. This was conducted to validate that no contamination occurred during the experiment. It can be seen from the data that the metabolic activity of the cell encapsulated gels increased over 3 weeks of incubation for sodium alginate in all calcium chloride concentrations. Fluorescent Reading vs. Incubation Day for 1.5% Manugel at various Calcium Chloride Concentrations Fuorescent Reading Incubation Day 0.50% 1% 2% 0.5% control 1% control 2% control Figure 6.3: Results of the fluorescent reading versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions

179 157 Fluorescent Reading vs. Incubation Day for 1.5% Manugel at various Calcium Chloride Concentrations Fuorescent Reading (Fraction Initial) % 1% 2% 0.5% control 1% control 2% control Incubation Day Figure 6.4: Results of the fraction initial fluorescent reading (normalized) versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions Discussion on Alginate Crosslink Cell Viability In the first experiment, the study was aimed at the investigation of the effect of the sodium alginate concentration on cell proliferation encapsulated in the alginate samples. Sodium alginate concentrations of 1%, 1.5%, 2%, and 3% (w/v) and a constant 0.5% (w/v) calcium chloride were used to make gel disk samples with a cell seeding density of 500,000 cell/ml. The control results show that there is no increase in fluorescent reading up to 3 weeks suggesting that the cell-free gels do not change in fluorescence over the duration of the experiment. It also indicated that there was no metabolic activity by other species that out rules the possibility of any contamination. At day 4, there was a slight increase in the fluorescent reading showing that slight

180 158 proliferation has occurred for all sodium alginate concentrations. From day 7 to 14, the fluorescent reading indicates a steady increase in proliferation for all sodium alginate concentrations. The Fluorescent reading also increased from day 14 to day 21; however, the proliferation rate had slightly decreased. There was no significant difference between the sodium alginate concentrations, which may suggest that the RHEC proliferate equally within the 1%- 3% (w/v) range. However, the viscosity of the 3% (w/v) sodium alginate solutions seemed to have changed after 0.45 µm filtration. The observed decrease in viscosity may indicate the concentration may have slightly changed during the filtration process. This phenomenon may change the process parameters for building 3D alginate bioactive scaffolds and cannot be used for cell encapsulation in alginate gel structures. In the second experiment, the study was set to investigate the effect of the calcium chloride concentration on cell proliferation encapsulated in the alginate samples. A constant sodium alginate concentration of 1.5% (w/v) with different crosslinking concentration at 0.5%, 1%, 2% (w/v) calcium chloride solutions were used to make gel disk samples with a cell seeding density of 500,000 cell/ml. The control results show that there is no increase in fluorescent reading up to 3 weeks suggesting that the cell-free gels do not change in fluorescence over the duration of the experiment. It also indicates that there was no metabolic activity by other species that out rules the possibility of any contamination. At day 4, there was a slight increase in the fluorescent reading showing that slight proliferation has occurred for all sodium alginate concentrations. From day 7 to 14, the fluorescent reading

181 159 indicates a steady increase in proliferation for all sodium alginate concentrations. The Fluorescent reading also increased from day 14 to day 21; however, the proliferation rate had slightly decreased. Although the 2% (w/v) calcium chloride solution showed slightly higher proliferation on day 7, there was no significant difference between the calcium chloride concentration samples, which may suggest that the RHEC proliferate equally within the 0.5%-2% (w/v) calcium chloride solution range. Since the calcium chloride concentrations have no significant difference on the proliferation of RHEC, 0.5% (w/v) is preferable because lower concentrations of calcium chloride are more suitable from a bioactive fabrication and 3D alginate scaffold construction perspective. Based on these 2 experiments, a graph was plotted to present potential sodium alginate and calcium chloride concentrations regions to identify the feasible bioactive fabrication range as shown in Figure 6.5. Therefore, 1.5% (w/v) sodium alginate was preferred for 3D alginate bioactive fabrication. Figure 6.6 and Figure 6.7 present optical and fluorescent images of Manugel RHEC non-encapsulated (control), respectively, on day 18 of incubation time. Figure 6.8 and Figure 6.9 present optical and fluorescent LIVE/DEAD images of Manugel RHEC encapsulated, respectively, on day 18 of incubation time.

182 160 Calcium Chloride Concentration %(w/v) Cell Culture Bioactive Fabrication Fabrication Sodium Alginate Concentration %(w/v) Figure 6.5: Results of the fraction initial fluorescent reading (normalized) versus incubation day for constant 1.5%, (w/v) sodium alginate concentration under different crosslinking at 0.5%, 1%, and 2%, (w/v) calcium chloride solutions

183 161 Figure 6.6: Optical image of Manugel RHEC non-encapsulated (control) on day 18 of incubation time Figure 6.7: Fluorescent image of Manugel RHEC non-encapsulated (control) on day 18 of incubation time

184 162 Figure 6.8: Fluorescent image of Manugel RHEC encapsulated on day 18 of incubation time Figure 6.9: Fluorescent image of Manugel RHEC encapsulated on day 18 of incubation time

185 Study on the Gelation of Alginate Materials and Methods Gel samples were prepared in a cell culture 24-wellplate. Each well contained 750 μl of sodium alginate and was crosslinked by adding 1.5 ml to each well. Six samples were tested and their results were averaged to represent the value for each specimen. The samples produced were cylindrical having a diameter of 12 mm and a height of 6 mm. Sodium alginate samples concentrations were prepared at 1%, 1.5%, 2%, and 3% (w/v) with gelation times of 10 minutes, 6 hours, 24 hours, and 48 hours using a constant 0.5% (w/v) calcium chloride. The gelation times were conducted by allowing the calcium chloride solution to interact with sodium alginate for the gelation time period and then removing the gel sample from the well plate and washing it in DI water to terminate the gelation process. The swelling ratio of alginate samples was calculated by measuring the weight of the sample hydrated and dry. The experiments were conducted to study the effect of the sodium alginate concentration and gelation time on the swelling ratio of alginate gels. The swelling ratio was calculated using Equation (6.1) as shown below; SR W W s d = (6.1) W d

186 164 Where SR is the swelling ratio, W s is the weight of hydrated or swollen gel and W d is the weight of dry gel. The samples were weighed immediately after the gelation process to obtain (W s ) and then freeze dried overnight to be reweighed to obtain (W d ) Results of Alginate Gelation The experiments were conducted to study the effect of the sodium alginate concentration and gelation time on the swelling ratio of alginate gel. The results show that the swelling ratio decreases as the sodium alginate concentration and gelation times increases as shown in Figure Swelling Ratio vs. Manugel Concentration at Various Gelation Times Swelling Ratio % 1.0% 1.5% 2.0% 2.5% 3.0% Manugel Concentration (w/v) 10 min 6 hr 24 hr Figure 6.10: Results of the swelling ratio versus Manugel concentration at various gelation times

187 Discussion on Alginate Gelation The swelling ratio of the gel samples decreased with the increase in Manugel (sodium alginate) concentration. This is due to the fact that there are more alginate chains in the gel that occupy space and less room for water to be absorbed as the volume of the sample are equal. The swelling ratio was also decreases as the gelation time increased. At longer gelation the time the more the alginate molecules are arranged into a gel network because of the increased crosslink density Study on the Mechanical Properties of the Bulk Alginate Materials and Methods Uniaxial compression tests were performed to measure the mechanical properties of alginate (Manugel) using a 4442 Instron mechanical tester (Instron Corporation, Canton, MA). Gel samples were prepared in a cell culture 24-wellplate. Each well contained 750 μl of sodium alginate and was crosslinked by adding 1.5 ml to each well. Six samples were tested and their results were averaged to represent the value for each specimen. The samples produced were cylindrical having a diameter of 12 mm and a height of 6 mm. Two experiments were conducted in this study. The first experiment was to investigate the effect of the sodium alginate concentration and gelation time on the elastic modulus of alginate gels. The second experiment was to investigate the calcium chloride concentration and gelation time on the elastic modulus for alginate. The gels were tested in a water bath at 37 o C with a 0.5 s -1 initial strain rate. The elastic modulus of the alginate gels was calculated at 10% strain from the stress strain data.

188 166 In the first experiment, the sodium alginate concentrations were prepared at 1%, 1.5%, 2%, and 3% (w/v) with gelation times of 10 minutes, 6 hours, 24 hours, and 48 hours at the constant crosslinking concentration of 0.5% (w/v) calcium chloride. The gelation times were conducted by allowing the calcium chloride solution to interact with sodium alginate for the gelation time period and then removing the gel sample from the well plate and washing it in DI water to terminate the gelation process. The second experiment was performed using the constant concentration of sodium alginate at 1.5% (w/v) but varying crosslinking concentration at 0.5%, 1%, and 2% (w/v) calcium chloride at gelation times of 10 minutes, 6 hours, 24 hours, and 48 hours. The amount of all calcium chloride solutions used in the experiments satisfied the criteria of allowing the gels to reach saturated crosslink gelation, hence, there were enough calcium ions to crosslink all the Guluronic acid sites in all the sodium alginate solutions Results of Bulk Alginate Mechanical Properties The first experiment was conducted to investigate the effect of various sodium alginate concentrations and gelation times on the elastic modulus of alginate samples, and the results are presented in Figures 6.11 and 6.12, respectively. It can be seen from the data that for a given the elastic modulus increased with increase of sodium alginate concentration for all gelation times. The second experiment was conducted to investigate the effect of various calcium chloride concentrations and gelation times on the elastic modulus of alginate samples, and the results are presented in Figures 6.13 and 6.14, respectively. It can be seen from the data the elastic modulus increased with

189 167 increase of calcium chloride concentration for all gelation times. Some ESEM images were taken for some of the alginate samples at various gelation times presented in Figures 6.15 through Elastic Modulus vs. Manugel Concentration at various Gelation Time Elastic Modulus (KPa) Concentration (%) 10 minutes 6 Hours 24 Hours 48 Hours. Figure 6.11: Results of the elastic modulus versus the sodium alginate concentration at various gelation times using a constant 0.5% (w/v) calcium chloride solution

190 168 Elastic Modulus vs. Gelation Time for Manugel Elastic Modulus (KPa) % 1.50% 2% 3% Time (Hours). Figure 6.12: Results of the elastic modulus versus the gelation time at various sodium alginate concentrations using a constant 0.5% (w/v) calcium chloride solution Elastic Modulus vs. Calcium Chloride Concentration for 1.5% (w/v) Manugel Elastic Modulus (KPa) Minutes 6 Hours 24 Hours Calcium Chloride Concentration (%) Figure 6.13: Results of the elastic modulus versus the calcium chloride concentration at various gelation times for a constant 1.5% (w/v) Manugel.

191 169 Elastic Modulus vs. Gelation Time for 1.5% (w/v) Manugel 70 Elastic Modulus (Kpa) % 1% 2% Time (Hours) Figure 6.14: Results of the elastic modulus versus the gelation time at various calcium chloride concentrations for a constant 1.5% (w/v) Manugel. Figure 6.15: ESEM image of 3% (w/v) Manugel after 10 minutes gelation time using a constant 0.5% calcium chloride.

192 170 Figure 6.16: ESEM image of 3% (w/v) Manugel after 24 hours gelation time using a constant 0.5% calcium chloride. Figure 6.17: ESEM image of 3% (w/v) Manugel after 24 hours gelation time using a constant 0.5% calcium chloride.

193 Discussion on Bulk Alginate Mechanical Properties The sodium alginate concentration has a direct effect on the elastic modulus. The elastic modulus increased at a higher rate for sodium alginate solutions at relatively higher concentrations. This can be seen when comparing the data for 1% and 3% (w/v) sodium alginate as shown is Figure The elastic modulus for 1% (w/v) sodium alginate samples was 5.23 KPa at 10 minutes gelation time and KPa at 48 hours. On the other hand, the elastic modulus for 3% (w/v) sodium alginate samples was KPa at 10 minutes gelation time and KPa at 48 hours. The gelation time has a direct influence on the mechanical properties of alginate. The elastic modulus increases with increasing gelation time. Increasing the gelation time provides higher chances for calcium ions to diffuse through the gel and solution and crosslink more carboxylic sites on Guluronic acid blocks on the alginate backbone. However, the number of carboxylic sites are limed and therefore, the gels become fully crosslinked (saturated) after a period of time for all sodium alginate concentrations as shown in Figure It was observed that the gels reached a crosslink saturation level at various times. Gels of higher sodium alginate concentrations had relatively longer gelation saturation times as compared to gels with lower sodium alginate concentrations. The gelation saturation time for 3% (w/v) sodium alginate was observed to be at 24 hours since the elastic modulus was almost stable when comparing the values form the 24 hours and 48 hours gelation times. It was also observed that the calcium chloride concentration has an effect on the gelation rate. The elastic modulus of the alginate gels increased with the gelation time for the 0.5%, 1%, and 2% (w/v) calcium chloride concentrations with 1.5%

194 172 (w/v) sodium alginate as shown in Figure At 0.5% (w/v) calcium chloride, the elastic modulus achieved at 24 hours was KPa. This value is lower than the 1% and 2% (w/v) calcium chloride solutions at 24 hours, which were KPa and 53.9 KPa, respectively. One may suggest that the 1.5% (w/v) sodium alginate solution should reach the same elastic modulus values at all calcium chloride concentrations since all cases have reached complete gelation. However, the results suggest that the gelation rate may influence the internal crosslink structure and there for produce stiffer structures at higher gelation rates for completely crosslinked alginate. However, this phenomenon tends to diminish as the calcium chloride concentration increases. This can be seen in Figure 6.14, where the difference in elastic modulus is relatively lower for the 1% and 2% (w/v) calcium chloride concentrations when compared to the 0.5% and 1% (w/v) calcium chloride concentrations for all gelation times. The results show that the end-to-end distance in the alginate gel network r 2 o is a function of gelation time and crosslink concentration in Equation (3.51) Study of Bulk Alginate Degradation Materials and Methods Gel samples were prepared in a cell culture 24-wellplate. Each well contained 750 μl of sodium alginate and was crosslinked by adding 1.5 ml to each well. Six samples were tested and their results were averaged to represent the value for each specimen. The samples produced were cylindrical having a diameter of 12 mm and a height of 6 mm. Sodium alginate samples concentrations were prepared at 1%, 1.5%, 2%, and 3% (w/v) with gelation times of 10 minutes, 6 hours, 24 hours, and 48 hours

195 173 using a constant 0.5% (w/v) calcium chloride. The gelation times were conducted by allowing the calcium chloride solution to interact with sodium alginate for the gelation time period and then removing the gel sample from the well plate and washing it in DI water to terminate the gelation process. The experiments were conducted to study the effect of degradation time on the mechanical properties of alginate gels. Two experiments were conducted. The first was to investigate the effect of the degradation time on the elastic modulus of alginate gels of various sodium alginate concentrations. The second experiment was to investigate the degradation time on the elastic modulus for alginate gels encapsulating various cell densities. The gels were tested in a water bath at 37 o C and 7.4 ph with a 0.5 s -1 initial strain rate. The elastic modulus of the alginate gels were calculated at 10% strain from the stress strain data. In the first experiment, sodium alginate concentrations were prepared at 1%, 1.5%, 2%, and 3% (w/v) with a gelation time of 10 minutes using a constant 0.5% (w/v) calcium chloride solution. The data was collected at degradation times of 0, 7, 14, and 21 days. The samples were allowed to degrade in medium (ph 7.4) at 37 o C. The experiment was performed using 1.5% (w/v) sodium alginate with 0.5% (w/v) calcium chloride solution with a gelation time of 10 minutes. The cell encapsulated samples were prepared with 250,000 cells/ml, 500,000 cells/ml, 750,000 cells/ml, and 1,000,000 cells/ml. The degradation was also studied by observing the percentage weight loss of Manugel samples. The percentage weight loss was calculated using Equation (6.2) as shown below;

196 174 Wi WD PWL = 100 (6.2) W i Where PWL is the percentage weight loss, W i is the weight of dry gel before degradation and W D is the weight of dry gel after degradation time. The samples were weighed immediately after the gelation process to obtain (W i ) and then freeze dried overnight to be reweighed to obtain (W D ) Results of Bulk Alginate Degradation The results showed that the elastic modulus for Manugel was increased from day 0 to day 1 for all the sodium alginate concentrations. The elastic modulus then slowly decreased after day 7 as shown in Figure This phenomenon was also observed for 1.5% (w/v) of sodium alginate seeded with different cell densities as shown in Figures 6.19 and The percentage weight loss of alginate increased with the degradation time for all concentrations of Manugel, with the higher rate increase for the higher concentrations of Manugel as shown in Figures 6.21 and 6.22.

197 175 Elastic Modulus vs. Degradation Time for Manugel Elastic Modulus (Kpa) Degradation Time (Days) 1% 1.5% 2% 3% Figure 6.18: Results of the elastic modulus versus the degradation time for Manugel at various concentrations using 0.5% (w/v) calcium chloride solution Elastic Modulus vs. Degradation Time for Encapsulated RHEC in 1.5% Manugel Elastic Modulus (Kpa) ,000 cell/ml 500,000 cells/ml 750,000 cells/ml 1,000,000 cells/ml Control (No Cells) Degradation Time (Days) Figure 6.19: Results of the elastic modulus versus the degradation time for 1.5% (w/v) Manugel at various cell densities using 0.5% (w/v) calcium chloride solution

198 176 Elastic Modulus vs. Degradation Time for Manugel Elastic Modulus (KPa) Degradation Time (Days) 1.50% 1.5% with RHEC Figure 6.20: Results of the elastic modulus versus the degradation time for 1.5% (w/v) Manugel at 500,000 cells/ml using 0.5% (w/v) calcium chloride solutions Percentage Weight Loss vs. Degradation Time for Bulk Manugel Disks 12 Percentage Weight Loss (%) % 1.50% 2% 3% Degradation Time (Days) Figure 6.21: Results of the percentage weight loss versus the degradation time for Manugel at various concentrations using 0.5% (w/v) calcium chloride solution

199 177 Percentage Weight Loss vs. Manugel Concentration for Bulk Disks 12 Percentage Weight Loss (%) Manugel Concentration (%) 1 week 2 weeks 3 weeks Figure 6.22: Results of the percentage weight loss versus the Manugel concentration at various degradation times using 0.5% (w/v) calcium chloride solution Discussion Bulk Alginate Degradation Experiments were conducted to investigate the effect of the degradation time on the elastic modulus of alginate gels of various sodium alginate concentrations. In addition the degradation time on the elastic modulus for alginate gels encapsulating various cell densities were also conducted. The elastic modulus of all Manugel concentrations increases from day 0 to day 1 due to the fact that there are many ions available in the cell culture medium that contributes to gelation and crosslink density of alginate, hence, increasing the mechanical properties. This also occurred for the gels with various cell densities. However from approximately day7, the alginate gels elastic modulus started to decrease due to the loss of crosslink ions in to the medium and the detachment of some alginate molecules from the gel structure. The effect of

200 178 encapsulating cell within the gel structures did not change the behavior of the gel s mechanical properties and shown in Figures 6.19 and Also, the encapsulated cells did not significantly change the degradation process of the alginate samples, which suggests that the mechanical properties of the gel do not change for duration of 3 weeks in a physiological environment using RHEC initial densities up to 1,000,000 cells/ml Characterization of Bioactive Alginate Tissue Scaffolds: Swelling, Degradation, Cell Viability and Cell Proliferation Materials and Methods Alginate scaffolds were fabricated with parallel strands for each layer while alternating rotating every layer by 90 degrees. The spacing of the strands was 1mm apart and the length of each 2D square layer was 10 mm with a strut diameter of 250 µm up to 10 layers. The scaffolds were fabricated using 1.5% (w/v) sodium alginate and were crosslinked using 0.5% (w/v) calcium chloride using 1,000,000 cells/ml. Four samples were tested and their results were averaged to represent the value for each specimen. The gelation times was approximately 10 minutes for each scaffold and was then washed in DI water to terminate the gelation process. The swelling ratio of alginate samples was calculated by measuring the weight of the sample hydrated and dry. The experiments were conducted to study the effect of the sodium alginate concentration and gelation time on the swelling ratio of alginate gels. The swelling ratio was calculated using Equation (6.1) as described earlier in the chapter.

201 179 Degradation studies were conducted on the alginate scaffolds by observing the percentage weight loss of the samples. The percentage weight loss was calculated using Equation (6.2). Also, degradation studies were conducted by investigating the effect of the degradation time on the elastic modulus of alginate scaffolds. The gels were tested in a cell culture medium bath at 37 o C and 7.4 ph with a 0.5 s -1 initial strain rate. The elastic modulus of the alginate gels were calculated at 10% strain from the stress strain data. Bioactive scaffolds were also fabricated with encapsulated RHEC using the multi-nozzle deposition system described in chapter 2. The biomanufactured scaffolds were fabricated under sterile conditions and were placed in 24-well plates for incubation up to 21 days. The proliferation of cells was monitored using fluorescent cytofluoremetry by adding Alamar Blue dye (Biosource International, CA) to the gel samples. The Alamar Blue assay could be repeated to the same samples over a number of times. The Alamar Blue assay incorporates a fluorometric growth indicator based on metabolic activity. As cells are being tested, innate metabolic activity results in a chemical reduction of Alamar Blue. Continued growth maintains a reduced environment while inhibition of growth maintains an oxidized environment. Reduction related to growth causes the indicator to change from oxidized state (nonfluorescent, blue) to reduced state (fluorescent, red). Samples were tested by removing medium from the wells and adding 1 ml of fresh medium with 100 μl of Alamar Blue to each well containing the alginate sample. The well plate was then incubated for 3 hours and then 400 μl of Alamar Blue/medium was placed in a fresh 24-wellplate prior to fluorescent reading using a Tecan GENios (Tecan U.S, NC).

202 180 The alarm Blue assay was conducted on day 0, 4, 7, 14, and 21. Also, a calibration graph was plotted to determine the number of cells contributing to each fluorescent reading. Cytotoxicity assessments were conducted on the gels using LIVE/DEAD assay (Invitrogen Corporation, CA) for cell viability study and images were taken under a Leica DMIL fluorescent microscope. Calcein AM (live) and ethidium homodimer (EthD-1) (dead) can be viewed simultaneously with a conventional fluorescein longpass filter. The fluorescence from these dyes may also be observed separately; calcein can be viewed with a standard fluorescein bandpass filter and EthD-1 can be viewed with filters for propidium iodide or Texas Red dye. In order to obtain the greatest sensitivity using a plate reader with the LIVE/DEAD assay, it is recommend exciting the fluorophores using optical filters optimal for their respective absorbencies. Calcein was excited using a fluorescein optical filter (485 ± 10 nm) whereas EthD-1 was excited using a typical rhodamine optical filter (530 ± 12.5 nm). The fluorescence emissions were acquired separately as well, calcein at 530 ± 12.5 nm, and EthD-1 at 645 ± 20 nm. The protocol for calculating the percentage of dead cells is presented in Appendix I. This method was used to relate the percentage of live and dead cells in terms of the shear rate using the bioactive fabrication process parameters. The bioactive fabrication process involves depositing live cells mixed homogeneously with sodium alginate through the pneumatic system. During the deposition, the cells are exposed to the shear forces with in the cell suspension that could potentially harm live cells and destroy them. A study was performed to relate the fabrication process parameters on cell viability

203 181 using Equations (3.58) and (3.59). Student t- test was used to assess the means of two groups were statistically different from each other Results of Bioactive Alginate Tissue Scaffolds Experiments were conducted to study the difference in swelling ratio of sodium alginate scaffold and bulk alginate using 1.5% (w/v) sodium alginate and 0.5% (w/v) calcium chloride with 10 minutes gelation time. The results showed that the Manugel scaffolds had a lower swelling ratio than the bulk disks that had no porosity as shown in Figure The percentage weight loss of occurred at a higher rate for the alginate scaffolds as compared to the bulk Manugel as shown in Figure The results also showed that the elastic modulus for bulk Manugel was higher than the Manugel scaffold from day 0 to day 21. The elastic modulus for both the scaffold and bulk gel then slowly decreased after day 7 after increasing from day 0 to day 1 as shown in Figure The micro-plate fluorescent readings using Alamar blue results for the bioactive scaffold are shown in Figure 6.26, where the control is the readings of Manugel scaffolds without cell encapsulation. The results of the calibration curve to determine the number of cells for the fluorescent reading are shown in Figures 6.27 and The number of cells over incubation days using the calibration curve is plotted in Figure The bioactive fabrication process involves depositing live cells mixed homogeneously with sodium alginate through the pneumatic system. During the deposition, the cells are exposed to the shear forces with in the cell suspension that could potentially harm live cells and destroy them. Results of the study are presented in Figure 6.30 and 6.31.

204 182 Swelling Ratio of 1.5% Manugel Scaffold Swelling Ratio Scaffold Bulk Disk % Manugel Concentration Figure 6.23: Comparison between the swelling ratios of Manugel scaffolds and bulk Manugel. P < Percentage Weight Loss vs. Degradation Time for 1.5% Manugel % Weight Loss Scaffold Bulk Disk Degradation Time (Days) Figure 6.24: Comparison between the percentage weight loss of Manugel scaffolds and bulk Manugel over 21 days of degradation time. P < 0.05.

205 183 Elastic Modulus vs. Degradation Time for 1.5% Manugel Scaffold Elastic Modulus (KPa) Degradation Time (Days) Scaffold Bulk Disk Figure 6.25: Comparison between the elastic modulus of Manugel scaffolds and bulk Manugel over 21 days of degradation time. P < Fluorescent Reading vs. Incubation Day for 1.5 % Manugel Bioactive Scaffold with 0.5% Calcium Chloride Fluorescent Reading % 1.5% control Day Figure 6.26: Calibration curve for determining the number of cells over the fluorescent reading. P < 0.05.

206 184 Fluorescent Reading vs. Cell Number for 1.5% Manugel with 0.5% Calcium Chloride Flourescent Reading y = x Cell Number (X10^6) Series1 Figure 6.27: Calibration curve for determining the number of cells over the fluorescent reading. Fluorecent Reading vs. Cell Number for 1.5% Manugel with 0.5% Calcium Chloride Flourescent Reading Cell Number (X10^6) Figure 6.28: Calibration curve for determining the number of cells over the fluorescent reading using an approximated linear equation.

207 185 Incubation Day vs. Cell Number for RHEC in 1.5 %Manugel Bioactive Scaffold with 0.5% Calcium Chloride Cell Number (X10^6) Incubation Day Figure 6.29: Incubation day versus the cell number for 1.5% (w/v) sodium alginate bioactive scaffold using 0.5% (w/v) calcium chloride. Shear Stress vs. Flow Rate for 1.5% (w/v) Sodium ALginate Shear Stress (MPa) D = 250 Micron D = 330 Micron D = 410 micron Flow Rate (microliter/second) Figure 6.30: Effect of the flow rate on the maximum shear stress of 1.5% (w/v) of sodium alginate using pneumatic microvalve valve.

208 186 Percentage of Dead Cells versus Maximum Shear Force 1.5% Sodium Alginate % of Dead Cells Shear Stress (KPa) 250 micron 330 micron 410 micron Figure 6.31: Effect of the maximum shear stress on the percentage of dead cells using1.5% (w/v) of sodium alginate with 1,000,000 cells/ml for pneumatic microvalve valve. Percentage of Live Cells versus Maximum Shear Force 1.5% Sodium Alginate % of Live Cells Shear Stress (KPa) 250 micron 330 micron 410 micron Figure 6.32: Effect of the maximum shear stress on the percentage of live cells using1.5% (w/v) of sodium alginate with 1,000,000 cells/ml for pneumatic microvalve valve.

209 Discussion of Bioactive Alginate Tissue Scaffold The percentage weight loss of the scaffold occurred at a higher rate for the alginate scaffolds as compared to the bulk Manugel as shown in Figure This may have occurred because of the porous structure that may have accelerated the degradation process since the inner core of the scaffold is exposed to the culture medium, hence, larger surface area. The results also showed that the elastic modulus for bulk Manugel was higher than the Manugel scaffold from day 0 to day 21. The elastic modulus for both the scaffold and bulk gel increased form day 1 to day 7 and then slowly decreased after day 7 after increasing from day 0 to day 1 as shown in Figure This is because of the alginate crosslinking that occurs for almost a week that is initiated by the ions found the in the cell culture medium. The ions the diffuse and interchange within the medium past a week of incubation, which allows that alginate molecules to detach from the main gel crosslinked structure and diffuse in to the culture medium. Eventually, this activity degrades the structure and causes the gel to loose its mechanical integrality over time. This phenomenon is seen in both the bulk alginate and the alginate scaffold. The fluorescent reading for the bioactive scaffolds with cell encapsulation increased with time for a period over 3 weeks incubation (Figure 6.26). The control showed that the scaffold without cell encapsulation did not have an increase in the fluorescent reading. These results showed that the cells encapsulated in the scaffold struts survived than biomanufacturing process and have increased in their metabolic reaction indicating the cells were proliferating for over 3 weeks. The bioactive fabrication process was conducted using 1.5% (w/v) sodium alginate and 0.5% (w/v)

210 188 calcium chloride with 10 minutes gelation time. Theses parameters were chosen based on the initial results that presented the feasible bioactive fabricate range (Figure 6.6). A calibration curve was conducted to determine the number of cells at a given fluorescent reading. Encapsulating a determined number of cells encapsulated into the alginate samples and then reading their fluorescent reading using Alamar blue was the method used to develop the calibration curve. The experiment used cell numbers of 0.5 x 10 6, 1 x 10 6, 1.5 x 10 6, and 2 x 10 6 with 3 samples of each into a well plate (Figure 6.27). Then calibration curve was then linearised to simplify the cure into an approximate calibration (Figure 6.28). The linear calibration curve was then used to determine the number of cells proliferating over the number of incubation days for the 1.5% (w/v) sodium alginate and 0.5% (w/v) calcium chloride with 10 minutes gelation time bioactive scaffold with an initial cell density of 0.5 x 10 6 /ml (Figure 6.29). The curve shows that the initial cell number was 0.5 x 10 6 at day 1 and then increased to approximately 3 x 10 6 by day 21. The bioactive fabrication process has proven to be a viable process. The process models that have been developed to determine the strut diameter and scaffold porosity were used to deposit living RHEC. The sodium alginate and calcium chloride concentrations were studied to determine the bioactive fabrication range and were used to test the viability of the RHEC. The cells were able to survive for more than 21 days and have proliferated to six folds the initial cell number. In this perspective, bioactive scaffolds can be designed in a fashion that allows cells to proliferate with a specific rate by controlling the cell density prior to culture or by

211 189 controlling the scaffolds porosity. However, it should be noted that controlling the scaffold porosity also affects the mechanical properties, which means that careful design should take place to have a final scaffold with the desired properties and function. Cytotoxicity assessments were conducted on the bioactive scaffolds using LIVE/DEAD assay (Invitrogen Corporation, CA) for cell viability study and images were taken under a Leica DMIL fluorescent microscope. Figures 6.33 through 6.36 resent optical and fluorescent images of RHEC after 14 days of incubation time. The fluorescent Cytotoxicity images show a higher number of live cells (green) than dead cells (red) on day 14. The bioactive scaffolds were fabricated using 1.5% (w/v) Manugel and 0.5% calcium chloride. The bioactive fabrication process involved depositing live cells mixed homogeneously with sodium alginate through the pneumatic system. During the deposition, the cells are exposed to the shear forces with in the cell suspension that affect their viability. A study was performed to relate the fabrication process parameters on cell viability using Equations (3.58) and (3.59). The pressures in the study ranged between 8 psi and 32 psi for nozzle diameters of 250 μm, 330 μm, and 410 μm using 1.5%(w/v) sodium alginate and 1,000,000 cells/ml. The study showed that the maximum shear stress at these operating parameters effect cell viability that ranges from 53% to 83% of live cells. However, the results also showed that the change in the maximum shear stress did not have a statistic significant change in the percentage of live cell. This suggests that the bioactive fabrication process is capable of changing its process parameters within the operating parameters conducted in the study without changing the percentage of viable cells. These results conclude that the

212 190 developed bioactive fabrication process is capable of fabricating scaffold of various designs by changing the process parameters to manufacture scaffolds with specific mechanical properties without changing the ratio of live and dead cells. Figure 6.33: Optical image of a bioactive strut of Manugel RHEC encapsulated on day 14 of incubation time

213 191 Figure 6.34: Fluorescent LIVE/DEAD assay image of a bioactive strut of Manugel RHEC encapsulated on day 14 of incubation time Figure 6.35: Optical image of a bioactive scaffold of Manugel RHEC encapsulated on day 14 of incubation time

214 Figure 6.36: Fluorescent LIVE/DEAD assay image of a bioactive scaffold of Manugel RHEC encapsulated on day 14 of incubation time 192

215 SUMMARY, CONCLUSION AND RECOMENDATIONS 7.1. Summary of the Research In this research, a novel biopolymer deposition system designed for fabricating 3D hydrogel tissue scaffolds was presented. A scientific and engineering knowledge required to fabricate 3-dimensional (3D) cell-embedded hydrogel-based tissue scaffolds was developed in this thesis. Alginate was selected as the biopolymer gel for the fabrication process, which is capable of being deposited at ambient temperatures in aqueous solution. The research was focused on the development of an engineering approach for manufacturing alginate hydrogel 3D tissue scaffolds by using a multi-nozzle biopolymer deposition process. Studying the effect of the processing, materials and structural configuration on the scaffold structure formation, and understanding cell viability during scaffold construction executed the development of a bioactive fabrication system. The first part of the thesis included the development of a viable deposition system for bioactive fabrication of tissue substitute constructs. Three nozzle systems were identified on the fabrication system and were characterized for their operating parameters and functionality. The nozzle systems were then set to a deposition feasibility study for alginate aqueous solutions by measuring flow rates and studying the operating parameter boundaries using various concentrations of sodium alginate. The feasibility study proved the pneumatic valve was a sensible candidate for the fabrication of alginate scaffolds and for biomanufacturing. Following the feasibility on the nozzle deposition systems, another study was performed on 3D alginate scaffold structural

216 194 formation was conducted by introducing the EDT and dual nozzle deposition using sodium alginate and calcium chloride. In addition, predictive process models were developed to determine the strut diameter and porosity based on the Poiseulle s equation for the bioactive scaffolds. Although the system toolpath could be used to fabricate complex anatomy structures form CT-Scans and MRI, the scaffold designs were based on 10mm x 10mm 2D layers of struts stacked over one another. This was done to build a foundation on how to fabricate simple 3D structures such as cubes and to understand the fundamentals of hydrogel scaffold biomanufacturing. The struts were separated from one another in each layer so that the 3D scaffolds are fabricated with pores integrated in to the structure. Process models were also developed and tested for controlling the strut diameter and scaffold porosity. The alginate scaffolds were characterized for their mechanical properties, swelling behavior, and degradation properties. In addition, Rat Heart Endothelial Cells (RHEC) were mixed with alginate solution to fabricate bioactive scaffolds with embedded cells to develop a Bioactive Freeform Fabrication (BFF) process. Cell viability studies were conducted on the cell-seeded structures for validating the bioactive process by monitoring the cell metabolic reaction over a period of time. The study developed an understanding of where the suitable bioactive fabrication parameters are in terms of the sodium alginate and calcium chloride concentrations. The mechanical prosperities of bulk alginate and bioactive scaffolds with encapsulated cells were characterized and the behavior of their degradation and swelling were also determined. Moreover, the

217 195 bioactive scaffolds were built using the developed process models with the intended designs in while satisfying the bioactive fabrication conditions. A study was conducted to develop the relationship between cell viability and bioactive fabrication parameters study using a developed process model Conclusion and Remarks Novel freeform fabrication methods for tissue engineering polymeric scaffolds have promising capabilities in revolutionizing tissue engineering because of its repeatability and capability of high accuracy in fabrication resolution at the macro and micro scales. These methods are unique since traditional SFF manufacturing methods utilize harsh solvents, high pressures or temperatures, or post-processing methods that are not suited for working with bioactive materials. By contrast, threedimensional dispensing (3DD) is a SFF technique that involves the deposition of extruded parts or micro-droplets of polymer solutions without toxic solvents and high temperatures. 3DD is a multi-nozzle system that can, simultaneously with the scaffold construction, deposit controlled amount of cells, growth factors, or other bioactive compounds with precise spatial position to form well-defined cell-seeded tissue constructs. The deposition process has proven its biocompatibility and occurs at room temperature and low pressures to reduce damage to the cells mixed in the sodium alginate solution. The operating parameters are important in terms of depositing the proper amount of sodium alginate to control the volume of the forming alginate gel. Feasibility studies showed that the system is capable of extruding Manugel alginate

218 196 between 0.4% and 3% (w/v). The flow rate, nozzle diameter, and nozzle velocity were studied and a model was developed to design 3D scaffolds with controlled strut diameters (D = microns) and pore sizes. In addition, rat heart endothelial cells were deposited through the system with alginate to form gel scaffold structures with encapsulated cells in a bioactive fabricated manor. The study showed that the suitable bioactive parameters preferred 1.5% (w/v) sodium alginate with 0.5% (w/v) calcium chloride. Cell viability studies were conducted on the cell encapsulated scaffolds for validating the bioactive freeform fabrication process that sowed viability up to 85%. The bioactive scaffold supported proliferation up to 21 days of incubation time. The elastic modulus was studied over degradation time that showed that the stiffened during the 24 hours due to crosslinking and degraded then on up to 21 days of incubation at 37 o C. Cell viability studies have proven that the cell-encapsulated scaffolds are able to maintain the RHEC in a healthy state and may proliferate for several weeks in culture. These results have demonstrated that the system is capable of biomanufacturing tissue constructs. Models to predict the Elastic modulus and encapsulated cell viability through process parameters are part of the on-going research, which could establish design methods on fabricating customized scaffolds for specific functions The bioactive fabrication process involved depositing live cells mixed homogeneously with sodium alginate through the pneumatic system. During the deposition, the cells are exposed to the shear forces with in the cell suspension that affect their viability. A study was performed to relate the fabrication process

219 197 parameters on cell viability. The results showed that the change in the maximum shear stress did not have a statistic significant change in the percentage of live cell. This suggests that the bioactive fabrication process is capable of changing its process parameters within the operating parameters conducted in the study without changing the percentage of viable cells. These results conclude that the developed bioactive fabrication process is capable of fabricating scaffold of various designs by changing the process parameters to manufacture scaffolds with specific mechanical properties without changing the ratio of live and dead cells. The system showed potential use for accurate cell placement in tissue engineering applications and promote regenerative medicine based on CAD systems Research Contributions The proposed research and activities will help to develop knowledge and novel solutions in determining optimal design and process parameters for bioactive polymeric scaffolds in tissue engineering applications. Such parameters are critically important when considering the: 1) mechanical properties; 2) appropriate process parameters for 3D structure formation of biopolymer scaffolds; 3) scaffold structural uniformity for pore size and distribution, cell distribution and mechanical properties; 4) scaffold degradation rate; and

220 198 5) desirable cell-scaffold bioactive process parameters that satisfy from manufacturing and biological perspectives for cell survivability an proliferation. The thesis research and activities will help to develop knowledge and novel solutions in determining optimal design and process parameters for bioactive polymeric scaffolds in tissue engineering applications Future Work Recommendations The system is a capable of fabricating scaffold structures that a relatively simple in terms of geometry. An interesting study would be to fabricate more complex structure such as anatomical parts e.g. a femur made form alginate. Such a refinement would enable bioactive scaffolds to be perfectly implanted into the body with suitable geometry and regenerate tissue constructs at the proper locations. The study may include additional materials and equipments to fine tune the fabrication process. An FEA analysis on the unit scaffold unit cell to compute the effective elastic modulus would be beneficial. A homogenization theory can then be used to predict the effective elastic modulus of the alginate scaffold. The porosity has an effect on the effective mechanical properties of the scaffold and therefore can be a part of a model to predict the effective mechanical properties the process parameters. The process model is based on the process parameters such as pressure, nozzle diameter,

221 199 sodium alginate concentration, type of nozzle system, calcium chloride concentration, and scaffold geometrical design and architecture. The end-to-end distance in the alginate gel network r 2 o shown in Equation (3.51) is a function of gelation time and crosslink concentration. Future work may include a study to develop an equation to determine the end-to-end distance in the alginate gel network r 2 o given known values of gelation time and calcium chloride concentration. The current system uses calcium chloride as the crosslink agent for sodium alginate during the biomanufacturing process. Another alternative to the calcium chloride solution is the cell medium its self that contains ions capable of crosslinking sodium alginate. The advantage of such a crosslink is that the encapsulated cells will be exposed to suitable physiological environment with proper nutrients and necessary supplements. By contrast, the cells are in a relatively uncomfortable environment for a period of time approximately 10 minutes in calcium chloride solution. However, it should be noted that cell mediums vary in ions and nutrients types and concentrations depending on the cells that are to be nutritioned, which would affect the gelation rate, mechanical properties, and degradation rate of bioactive alginate scaffolds. Selective cell adherent scaffold is recommended for future work that involves the modification of alginate with peptides such as RGD that could be conducted using methods such as aqueous carbodiimide chemistry. Since natural alginate prohibits cell adhesion to its surface, a peptide modified alginate surface may A scaffold can be fabricated using two nozzles, one with RGD modified alginate and the other with unmodified RGD modification. As a result, a scaffold could be biomanufactured with

222 200 a specific cell type encapsulated into specific struts within the scaffold and another cell type in other struts as shown in Figure 7.1. An addition, the scaffold may be fabricated with a classified design that places RGD modified alginate struts in desired locations to seed other cell types to the modified surfaces in a post biomanufacturing process. As an example, Fibroblasts could be encapsulated into the struts and have endothelial cells adhered to the alginate surface in a manor that may promote vascularization within tissue constructs as shown in Figure 7.2. Figure 7.1: Alginate struts with RGD modified and unmodified surfaces.

223 201 Figure 7.2: Alginate scaffolds made of RGD modified and unmodified struts stacked next to one another to for a selective cell adherent bioactive scaffold Alginate has been used as a scaffold for bone regeneration despite the relatively low mechanical properties that gels posses. However, future work may include investigating bioactive scaffolds using osteoblasts and Hydroxiapetite (HA) as shown in Figure 7.3. HA has been known as an oseocondutive biomaterial that promotes the growth and proliferation of osteoblasts. An RGD modified alginate scaffold could be utilized to allow endothelial cells to be seeded on to the struts surface. This technique may promote vascularization within scaffolds that supply nutrients to living cells at the core of scaffolds when implanted in vivo. That fact that osteoblasts and HA can be placed in localized areas within a bioactive scaffold may evolve new directions on how bone regeneration may be efficiently engineered.

224 202 Figure 7.3: Osteoblasts and HA particle encapsulated into alginate struts with RGD modified surface to promote endothelial attachment for vascularization The potential of mixing osteoblasts with HA is very promising since the bioactive process is capable of fabricating scaffolds made of an alginate/ha composite as shown in Figure 7.4. Figure 7.4: Alginate/HA scaffolds fabricated using the current bioactive fabrication system; (A) overview of whole scaffold; (B) Close-up of scaffold showing the HA particles encapsulated within the scaffold