Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Development of hollow out-of-plane polymer microneedles using solvent casting Mansoor, Iman 2009

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

Item Metadata

Download

Media
24-ubc_2009_fall_mansoor_iman.pdf [ 7.29MB ]
Metadata
JSON: 24-1.0067796.json
JSON-LD: 24-1.0067796-ld.json
RDF/XML (Pretty): 24-1.0067796-rdf.xml
RDF/JSON: 24-1.0067796-rdf.json
Turtle: 24-1.0067796-turtle.txt
N-Triples: 24-1.0067796-rdf-ntriples.txt
Original Record: 24-1.0067796-source.json
Full Text
24-1.0067796-fulltext.txt
Citation
24-1.0067796.ris

Full Text

DEVELOPMENT OF HOLLOW OUT-OF-PLANE POLYMER MICRONEEDLES USING SOLVENT CASTING by Iman Mansoor BASc., The University of British Columbia, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Electrical and Computer Engineering)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) October 2009 ©Iman Mansoor, 2009  ABSTRACT In recent years, extensive research has been done to find innovative ways of drug delivery to replace traditional injection using hypodermic needles. Although microneedles are proposed to provide one of the most effective and convenient transdermal drug delivery methods, their expensive fabrication techniques have created a barrier for their mass fabrication and as a result, their entry to the commercial market. A novel method, based on solvent casting, is presented for inexpensive fabrication of hollow out-of-plane polymer microneedles. Microneedles are formed during a solvent evaporation process which leaves a thick polymer layer around pillars in a pre-fabricated mold. This process is fast and allows fabrication of microneedles in variety of shapes and dimensions. The effectiveness of the microneedle arrays fabricated using this process has been demonstrated through in vivo and in vitro experiments. In order to further optimize the microneedle design, a novel experimental method based on confocal microscopy and particle image velocimetry (PIV) is presented for characterizing the flow in a thin film during solvent casting. Using this method, the impact of temperature on polymer film formation, on a vertical profile in a mold, is investigated and discussed. This method also allowed observing some important phenomena during solvent casting such as a surface counter flow. The PIV measurements show significant differences in the flow velocity fields at different temperatures that correlate with different final polymer thicknesses on the vertical wall of the mold.  ii  TABLE OF CONTENTS ABSTRACT ....................................................................................................................... ii TABLE OF CONTENTS ................................................................................................ iii LIST OF TABLES ........................................................................................................... vi LIST OF FIGURES ........................................................................................................ vii ACKNOWLEDGEMENTS ............................................................................................ xi DEDICATION................................................................................................................. xii CO-AUTHORSHIP STATEMENT ............................................................................. xiii 1  2  INTRODUCTION..................................................................................................... 1 1.1  Microneedles for Transdermal Drug Delivery ..................................................... 1  1.2  Types of Microneedles ......................................................................................... 4  1.3  Research Objectives ............................................................................................. 7  1.4  Thesis Organization.............................................................................................. 8  1.5  References ............................................................................................................ 9  Fabrication of Hollow Out-of-plane Polymer Microneedles for Transdermal  Drug Delivery .................................................................................................................. 13 2.1  Introduction ........................................................................................................ 13  2.2  Fabrication Process for Hollow Out-of-plane Polymer Microneedles ............... 16  2.3  Test Procedures .................................................................................................. 25  iii  2.3.1  Microneedle Robustness Tests .................................................................... 25  2.3.2  Microneedle Insertion Tests ........................................................................ 26  2.3.3  Microneedle Injection Tests ........................................................................ 27  2.4  3  Results and Discussion ....................................................................................... 28  2.4.1  Microneedles Robustness............................................................................ 28  2.4.2  Results of Insertion Tests ............................................................................ 30  2.4.3  Results of Injection Tests ............................................................................ 33  2.5  Conclusions ........................................................................................................ 34  2.6  Acknowledgment ............................................................................................... 35  2.7  References .......................................................................................................... 36  PIV Measurements of Flow in Thin Drying Films during Solvent Casting ...... 40 3.1  Introduction ........................................................................................................ 40  3.2  Materials and Methods ....................................................................................... 44  3.2.1  Polymer Solution ........................................................................................ 44  3.2.2  Experimental Setup ..................................................................................... 44  3.2.3  Experimental Procedure .............................................................................. 46  3.3  Results and discussion........................................................................................ 49  3.3.1  Flow Fields at Different Temperatures from Confocal/PIV Measurements49  3.3.2  Verification of the Confocal/PIV Results ................................................... 55  3.4  Conclusions ........................................................................................................ 57 iv  3.5  Acknowledgements ............................................................................................ 58  3.6  References .......................................................................................................... 59  4  Conclusions and Future Work............................................................................... 62 4.1  Conclusions ........................................................................................................ 62  4.2  Recommendations for Future Work ................................................................... 64  4.3  References .......................................................................................................... 66  Appendices ....................................................................................................................... 67 A.  Mold Fabrication and Photolithography Parameters .......................................... 67  B.  Photolithography Mask ...................................................................................... 71  C.  More Needle Images .......................................................................................... 73  D.  Temperature Chamber ........................................................................................ 75  E.  Polymer Solution Flow during Solvent Casting at 25°C ................................... 77  v  LIST OF TABLES Table 2.1 Vertical and horizontal plasma etch rates, measured for different combinations of RF power and gas chamber pressure, for the Trion RIE/PECVD tool. A higher ratio of the horizontal etch rate over the vertical etch rate indicates a more isotropic etching process........................................................................................................... 22 Table 2.2 Relative radioactivity biodistribution, measured 1 hour after transdermal microneedle injection of 160 µCi of 99mTc radiolabeled human serum albumin. ..... 33 Table A.1 Recipes used for fabrication of 250 µm and 400 µm thick molds. .................. 69  vi  LIST OF FIGURES Figure 1.1 Conceptual sketch of a microneedle inserted into human skin. ........................ 3 Figure 2.1 Fabrication process using solvent casting for hollow polymer out-of-plane microneedles. (a) & (b) fabrication of pillars from SU-8 (c) Parylene C deposition (d) & (e) PMMA deposition & solvent evaporation (f) & (g) SU-8 + cyclopentanone deposition & evaporation of cyclopentanone (h) UV exposure of remaining SU-8 layer on pillars (i) removing of SU-8 layer from the top of the pillars and sharpening of the needles using plasma etching (j) dissolving PMMA layer in chloroform which releases the needle structure. ..................................................................................... 17 Figure 2.2 SEM image of an array of tapered pillars fabricated using backside exposure of SU-8. The vertical ridges along the pillars are created due to the poor resolution of the photomask. ........................................................................................................... 19 Figure 2.3 Microneedle mold consisting of an array of pillars surrounded by square vertical walls. ............................................................................................................. 19 Figure 2.4 SEM image of PMMA solution dried at 70°C on an array of pillars .............. 20 Figure 2.5 SEM image of PMMA solution dried at 120°C on an array of pillars ............ 21 Figure 2.6 Image of the microneedle array destroyed and stuck in mold after a failed liftoff ............................................................................................................................... 23 Figure 2.7 SEM image of microneedles fabricated using the process illustrated in Figure 2.1. The distance between needles in the arrays is 300 µm (a) an array of microneedles 110 µm long with approximately 45 µm tip diameter (b) an array of microneedles 180 µm long with approximately 30 µm tip diameter (c) a single  vii  180 µm microneedle from the array shown in b (d) microneedles lumen openings on the backside of the array. ........................................................................................... 24 Figure 2.8 Schematics of the experimental setup used for testing microneedle robustness under different compressive loads. ............................................................................ 25 Figure 2.9 An array of 42 microneedles bonded to the tip of a 1 ml syringe, used for the injection experiment. ................................................................................................. 28 Figure 2.10 Histology image of potato skin showing a hole created by a microneedle. .. 30 Figure 2.11 (a) SEM image of potato skin after application of an array of microneedles (b) image of the microneedles after the insertion experiment. .................................. 31 Figure 2.12 Food color application on potato skin (a) application site with (left) and without (right) applying microneedles beforehand (b) cross section A:A. ............... 32 Figure 3.1 The experimental setup used to characterize flow during solvent evaporation in solvent casting process. ......................................................................................... 45 Figure 3.2 Mold used for characterizing flow during solvent evaporation. Each mold is made by bonding four 200 µm thick coverslips to a microscope slide base to create an enclosed rectangular region. The scan volume is chosen near the vertical wall of one of the coverslips, far from the corners of the rectangular cavity. ....................... 46 Figure 3.3 Recording of fluorescent particles in the scan volume and the 2D side projection. .................................................................................................................. 47 Figure 3.4 Vector fields for the initial phase of solvent evaporation at (a) 25ºC, (b) 35ºC, and (c) 45ºC. .............................................................................................................. 49 Figure 3.5 Detail of the surface counter flow at 25 ºC. .................................................... 50  viii  Figure 3.6 Streamlines for the initial phase of solvent evaporation at (a) 25ºC, (b) 35ºC, and (c) 45ºC. .............................................................................................................. 51 Figure 3.7 Average of the horizontal component of velocity over time from the PIV data. ................................................................................................................................... 52 Figure 3.8 Vector fields and streamlines for the second phase of solvent evaporation at (a) & (b) 25ºC, (c) & (d) 35ºC, and (e) & (f) 45ºC. ................................................... 53 Figure 3.9 Vector fields and streamlines for the final phase of solvent evaporation at (a) & (b) 25ºC, (c) & (d) 35ºC, and (e) & (f) 45ºC. ........................................................ 54 Figure 3.10 Volume of the polymer solution in the scan volume over time. ................... 55 Figure 3.11 Solvent evaporation rate over time. ............................................................... 56 Figure A.1 The third mask used for fabrication of microneedle arrays. The mask contains arrays of 40 µm circular dots with 300 µm spacing. ................................................. 72 Figure A.2 Microneedle arrays fabricated using the technique shown in chapter 2. Different shapes of microneedles are achieved by varying some of the process parameters (a) an array of an array of microneedles 110 µm long fabricated using the recipe presented in chapter 2 (b) an array of microneedles 200 µm long fabricated with high solvent casting temperature (120°C) (c) an array of microneedles made with a weaker plasma etching step (150 s, power 100W, 90% O2 and 10% CF4) and d) microneedles fabricated with a stronger plasma etching step(400 s, 200W power, 90% O2 and 10% CF4). .............................................................................................. 73 Figure A.3 Some examples of microneedles (a) a microneedle created using the process shown in chapter 2 with a stronger plasma etching step (400 s, 200W power, 90% O2 and 10% CF4) (b) and (c) less sharper microneedles fabricated using a similar  ix  process presented in chapter 2; tip openings in these devices are created by using fine sanding papers instead of plasma etching. ................................................................. 74 Figure A.4 Temperature chamber working principle. ...................................................... 75 Figure A.5 Temeparture chamber installed on an inverted Nikon microscope. ............... 76 Figure A.6 Slides from the supplementary video file submitted with the manuscript used in chapter 3. The slides correspond to polymer solution flow during solvent evaporation at 25°C. The time elapsed between each consecutive slide pair is 231.6 s.. ...................................................................................................................... 78  x  ACKNOWLEDGEMENTS I would like to express my genuine gratitude to my supervisor Dr. Boris Stoeber for his extensive guidance and support during the last two years. I am thankful to him for giving me the opportunity to work on this project, and also sharing his knowledge throughout the project. I would like to thank Dr. Urs O. Häfeli for his valuable inputs and also for helping me with the mouse experiment as well as the microneedle insertion trials. I also would like to thank Dr. Karen Cheung and Dr. Mu Chiao for teaching me the important basics of MEMS, and providing intellectual contributions during my MASc. years. I also thank them for sharing their lab equipment without which I wouldn’t be able to work on the project. I’d like to acknowledge the undergraduate students, David Hung and Michael Young, for designing and building the temperature control chamber. I would like to appreciate my brother Hadi for his endless support and helpful suggestions; and I thank my parents (Abdolsamad and Nasrin), my brother (Mehdi), and my sister (Soudeh) for always being supportive. I also thank my friends (Ramin, Navid, Saeed, Amir, Mahmoud, Hamed, Sina, Monsieur Mahyar, Ali, Masoud, and the Talebbeydokhtis), my colleagues (Vahid, Reynald, Allison, Nazly, Reza, Farid, Jonas, and other MEMS group members) and the rest of my family (Shima, Firoozeh, and Mahan) for their helpful support and contributions. Finally, I’m grateful to my girlfriend Soudeh for her unconditional moral support and encouragements.  xi  DEDICATION  This work is dedicated to my parents  xii  CO-AUTHORSHIP STATEMENT My contributions in identification and design of the research program include the design of the general microneedle fabrication process based on the ideas that were discussed with my supervisor in the beginning of my master’s project. I have prepared both of the manuscripts presented in chapters 2 and 3 of this report. I have performed all the research, experiments, and data analysis for the manuscript presented in chapter 2, excluding the sections related to the in vivo injection experiment (sections 2.3.3 and 2.4.3). For the injection experiment, I was partly involved in the experimental procedure as well as data collection. For the manuscript presented in chapter 3, I have performed all the research as well as the experimental procedures. My contributions for data analysis include preparation and manipulation of the graphs and figures, and also measurement and comparison of the final polymer profiles.  xiii  1 INTRODUCTION 1.1 Microneedles for Transdermal Drug Delivery Drugs can be administered using different body routes depending on the type of drug being delivered as well as its treatment mechanism. The most common drug delivery methods are oral and parenteral. In oral administration, pills, drops, capsules, and tablets are used as the most common delivery mechanisms. In parenteral administration, the drug is delivered through skin by injection or infusion mostly using hypodermic needles as the main delivery mechanism. Although drug administration using the oral route is painless and easy, this technique is associated with some potential hazards to the digestive system. In addition, in many cases the rate of drug absorption is slow. Drug injection using traditional hypodermic needles is much faster than oral administration; however, it is painful and can cause infections. Normal flora of the skin can penetrate into the blood stream upon insertion of the needle and cause infection [1]. Traditional hypodermic needles also represent a high risk of vein rupture especially in the case of small children who cannot tolerate pain and move during puncture. There is also a high contamination risk for medical staff when handling needles, and there is also an injury risk and cost associated with disposing of large volumes of syringes with hypodermic needles [2, 3]. In recent years, researchers have been investigating various methods of drug delivery in order to overcome the limitations and potential risks associated with the current common methods and also to provide novel methods for controlled drug release as well as targeted 1  treatment. For instance one potential method for controlled and targeted drug delivery is using implantable biocompatible MEMS-based devices [4-6] which require surgical procedures for installation and removal. Another potential method for controlled drug release is using biodegradable micro/nanospheres [7-9], whose administration mostly relies on oral delivery or direct injection. A relatively new drug administration approach without the potential risks and pain of traditional needles is transdermal drug delivery [10]. This technique has yet to achieve its full potential as a substitute for oral delivery and hypodermic injections. In this technique, the drug passes through the skin’s outer most layer (stratum corneum) into the epidermis and is then diffused into the blood stream. One of the various methods for transdermal drug delivery is using adhesive skin patches [11]. The first commercially available adhesive skin patch was introduced in 1979 which administered scopolamine to treat motion sickness [12]. Since then, the number of drugs approved for transdermal administration using adhesive patches has been on the rise, leading to availability of 19 different drugs on commercial patches in 2007 in the United States, including pain relief patches [12]. Although skin patches provide a very easy and convenient transdermal drug delivery method, they are also associated with certain limitations. For instance, due to impermeability of the stratum corneum to many hydrophilic agents, only lipophilic agents can be administered using this technique [13]. In addition, in some cases there have been reports of skin reactions upon application of the patches [14]. Another potential approach for transdermal drug delivery is using microneedles. Microneedles are micron scale solid or hollow mechanical structures designed to penetrate through the stratum corneum and deliver an agent to epidermis. Microneedles 2  have been an active area of micoelectromechanical system (MEMS) research since the early 1990s. Because of their very small size, microneedles introduce major advantages over the conventional hypodermic needles for many applications. They can be designed to penetrate into the epidermis layer and release an agent, without being in contact with the blood stream or the nerve endings; therefore, they can be painless and less associated with potential infections. In addition, they bear a low risk of device contamination through blood. Figure 1.1 shows a conceptual demonstration of a hollow microneedle penetrated into the skin. Drug injection through microneedle lumen Stratum Corneum ~ 20 µm Microneedle  Epidermis ~ 100 µm  Capillary Nerve  Dermis ~ 600 µm – 3 mm  Figure 1.1 Conceptual sketch of a microneedle inserted into human skin.  In vivo and in vitro evaluations of microneedles have shown successful and promising results. The reports indicate painless and fast drug delivery into the skin of volunteers 3  [15-17]. All these studies suggest that microneedles can provide a powerful new approach to transdermal drug delivery. In addition to transdermal drug delivery, microneedles are promising devices for biosensing applications such as for continuous glucose monitoring [18, 19].  1.2 Types of Microneedles Various concepts for microneedles fabrication have been demonstrated in the past. Inplane and out-of-plane microneedles have been reported with solid or hollow structures in a variety of shapes and dimensions [20-45]. Solid microneedles are typically designed to improve skin permeability to agents, while devices with hollow structures are designed to puncture the skin and release the agent under the stratum corneum. Out-of-plane microneedles with solid and hollow structures have been made from silicon [20, 22-25, 27-30], silicon/silicon dioxide [32], metal shells [33, 34, 37, 39], glass [40], and polymers [41-45] in form of two dimensional arrays. Studies have shown that hollow microneedles allow faster drug delivery into the body [16]. Previously, MEMS-based hollow out-of-plane microneedles have been fabricated in a variety of lengths ranging from 30 µm to 600 µm made using different fabrication techniques. The lumen diameters in these devices range from 3 µm up to 70 µm. Arrays of single-crystal silicon microneedles have been developed by Stoeber and Liepmann [20], Griss and Stemme [25], and Gardeniers et al. [28]. Fabrication of 200 µm long silicon microneedles presented by Stoeber and Liepmann [20] is mainly based on DRIE of silicon, corresponding to microneedle channel, followed by isotropic etching for  4  formation of the needle structure. The authors have also investigated the performance of these devices by clinical injection of methyl nicotinate into human skin [16]. The results showed successful injection of agent using the microneedle devices and the volunteers reported feeling pressure but no pain. In the work presented by Griss and Stemme [25], the authors have developed side-opened silicon 210-µm long microneedle arrays mainly using silicon DRIE and two photolithography steps. Gardeniers et al. [28] presented silicon microneedles fabricated using mainly DRIE of silicon, for microneedle channels, and anisotropic wet etching for the formation of the outer needle shape. The needles produced in this work were 400 µm long. The performance of these devices has also been demonstrated by painless blood withdrawal from human subjects. Although the reported silicon microneedles are sharp and have robust structures, their fabrication process is expensive and slow. Davis et al. [36], McAllister et al. [37], Kobayashi and Suzuki [38], and Kim et al. [39] have developed thin-walled metallic microneedles by electro-plating Ni, NiFe, or platinum on mold substrate. The needles created in these works are between 400 µm to 500 µm tall; however, they are less sharp than silicon microneedles and are more prone to failure during needle insertion into human skin. In a separate work by Chun et al. [32], 30 µm-long SiO2 microcapillaries have been developed for injection of agents into individual cells. These needles are very sharp; however, their length is limited by the aspect ratio obtained after a silicon DRIE step. In another work, hollow glass microneedle used for cell physiology have been reported by Brown and Flaming [40]. The 900 µm long microneedle device reported in this work has  5  not been fabricated using MEMS standard techniques. Instead, it has been made using a conventional micropipette puller. Previously, arrays of out-of-plane microneedles made from polymer materials have been presented by Kuo and Chou [43], Huang and Fu [44], and Moon et al. [45]. Kuo and Chou have used PDMS molding technique as well as several photolithography steps to fabricate 600 µm long microneedles. Huang and Fu [44] have used a back-side exposure technique for fabrication of 210 µm long hollow polymer structures on a glass substrate, followed by a DRIE step for microneedle openings. Moon et al. [45] have produced 600 µm long microneedles by using two masking steps including an inclined LIGA process. The microneedles presented in these works are not as sharp as silicon and metal devices, and require multiple photolithography steps and are not suitable for mass fabrication.  6  1.3 Research Objectives The broad adoption of microneedles for transdermal drug delivery and biosensing to date has mainly been impeded by the expensive and/or sequential fabrication steps currently used for microneedles. The aim of this project is to develop a new fabrication technique applicable for mass production of inexpensive hollow out-of-plane microneedles from polymeric materials. The availability of these devices would represent a major contribution to the biotech industry as well as the public health and safety in Canada and the world. The goal of this research is to develop a fabrication process that uses a single photolithography step for fabrication of a reusable mold that can be used in a solvent casting process to form microneedle arrays. After development of the fabrication process, sample microneedle prototypes will be fabricated and their usefulness will be demonstrated through proof-of-principle applications. In addition, injection of a fluid through the needles into an animal subject will be demonstrated. Furthermore, for characterizing the microneedle formation during solvent casting and controlling the needle tip shape, an experimental technique will be developed based on particle image velocimetry (PIV) and confocal microscopy that investigates polymer film flow during solvent evaporation. Poly(vinyl alcohol) (PVA) will be used as the test polymer. This technique can be used to characterize the influence of process parameters such as temperature and humidity on the profile of the deposited polymer. For this purpose, the drying process of a PVA solution in a mold will be investigated under different temperature conditions. 7  1.4 Thesis Organization In the first chapter of this thesis, the basic concept of microneedles for transdermal drug delivery has been introduced, followed by a discussion on the different types of previously fabricated microneedles. Finally, the objectives of this project were presented. The second chapter presents a manuscript to be submitted to a journal. In this manuscript the detailed process for the fabrication of hollow out-of-plane polymer microneedles using a solvent casting method is presented. Fabricated prototypes are shown, and the results of mechanical tests as well as insertion tests performed on them are discussed. Finally, result of an in vivo injection trial on an animal subject is presented. The third chapter of the thesis is a version of a submitted manuscript. In this manuscript, a new technique, utilizing a confocal/PIV system, is presented for flow visualization during solvent casting. This technique is then used to investigate the impact of temperature on polymer film flow during solvent evaporation, and the corresponding velocity fields, drawn from the PIV measurements, are analyzed. Finally, the fourth chapter presents conclusive remarks drawn from the investigations and experiments, as well as some future recommendations for further development of the microneedle fabrication process.  8  1.5 References [1] T. E. Andreoli, C. C. J. Carpenter, R. C. Griggs, and J. Loscalzo, CECIL Essentials of Medicine 6th ed. Elsevier, Canada, 2004. [2] K. Krasinski, R. LaCouture, and R. S. Holzman, “Effect of changing needle disposal systems on needle puncture injuries,” Infection Control, vol. 8, pp. 59-62, 1987. [3] D. Adams D and T. S. Elliott, “Impact of safety needle devices on occupationally acquired needlestick injuries: a four-year prospective study,” The Journal of Hospital Infection, vol. 64, pp. 50-55, 2006. [4] S. Z. Razzackia, P. K. Thwara, M. Yanga, V. M. Ugazb, and M. A. Burnsa, “Integrated Microsystems for controlled drug delivery,” Advanced Drug Delivery Reviews, vol. 56, pp. 185-198, 2004. [5] G. Voskericiana, M. S. Shivea, R. S. Shawgoc, H. von Recumd, J. M. Anderson, M. J. Cimac, and R. Langere, “Biocompatibility and biofouling of MEMS drug delivery devices,” Biomaterials, vol. 24, pp. 1959-1967, 2003. [6] Y. Lia, R. S. Shawgoa, B. Tylerb, P. T. Hendersonc, J. S. Vogelc, A. Rosenberga, P. B. Stormb, R. Langerd, H. Bremb, and M. J. Cima, “In vivo release from a drug delivery MEMS device,” Journal of Controlled Release, vol. 100, pp. 211-219, 2004. [7] R. Gref, A. Domb, P. Quellec, T. Blunk, R. H. Müller, J. M. Verbavatz, and R. Langer, “The controlled intravenous delivery of drugs using PEG-coated sterically stabilized nanospheres,” Advanced Drug Delivery Reviews, vol. 16, pp. 215-233, 1995. [8] M. N. V. R. Kumar, “Nano and Microparticles as Controlled Drug Delivery Devices,” Journal of Pharmaceutical Sciences, vol. 3, pp. 234-258, 2000. [9] K. S. Soppimatha, T. M. Aminabhavi, A. R. Kulkarnia, and W. E. Rudzinskib, “Biodgradable polymeric nanoparticles as drug delivery devices,” Journal of Controlled Release, vol. 70, pp. 1-20, 2001. [10] M. R. Prausnitz, “A practical assessment of transdermal drug delivery by skin electroporation,” Advanced Drug Delivery Reviews, vol. 35, pp. 61-76, 1999. [11] S. Venkatraman and R. Gale, “Skin adhesives and skin adhesion: 1. Transdermal drug delivery systems,” Biomaterials, vol. 19, pp. 1119-1136, 1998. [12] M. R. Prausnitz and R. Langer, “Transdermal drug delivery,” Nature Biotechnology, vol. 26, pp. 1261-1268, 2008.  9  [13] A. Naik, Y. N. Kalia, and R. H. Guy, “Transdermal drug delivery: overcoming the skin barrier’s function,” Pharmaceutical Science & Technology Today, vol. 3, pp. 318-326, 2000. [14] M. Murphy and A. J. Carmichael, “Transdermal Drug Delivery Systems and Skin Sensitivity Reactions: Incidence and Management,” American Journal of Clinical Dermatology, vol. 1, pp. 361-368, 2000. [15] M. R. Prausnitz, “Microneedles for transdermal drug delivery,” Advanced Drug Delivery Reviews, vol. 56, pp. 581-587, 2004. [16] R. K. Sivamani, B. Stoeber, G. C. Wu, H. Zhai, D. Liepmann, and H. Maibach, “Clinical microneedle injection of methyl nicotinate: stratum corneum penetration,” Skin Research and Technology, vol. 11, pp. 152-156, 2005. [17] S. Kaushik, A. H. Hord, D. D. Denson, D. V. McAllister, S. Smitra, M. G. Allen, and M. R. Prausnitz, “Lack of Pain Associated with Microfabricated Microneedles,” Anesthesia & Analgesia, vol. 92, pp. 502-504, 2001. [18] W. H. Smart and K. Subramanian, “The Use of Silicon Microfabrication Technology in Painless Blood Glucose Monitoring,” Diabetes Technology & Therapeutics, vol. 2, pp. 549-559, 2000. [19] S. Zimmermann, D. Fienbork, B. Stoeber, A. W. Flounders, and D. Liepmann, “A Microneedle-based Glucose Monitor Fabricated On A Wafer-Level Using In-device Enzyme Immobilization,” The 12lh International Conference on Solid State Sensors, Actuators and Microsystems, Boston, pp. 99-102, 2003. [20] B, Stoeber and D. Liepmann, “Arrays of hollow out-of-plane microneedles for drug delivery,” Journal of MEMS, vol. 14, pp. 472-479, 2005. [21] S. J. Paik, S. Byun, J. M. Lim, Y. Park, A. Lee, S. Chung, J. Chang, K. Chun, and D. Cho, “In-plane single-crystal-silicon microneedles for minimally invasive microfluid systems,” Sensors and Actuators A: Physical, vol. 114, pp. 276-284, 2004. [22] S. Henry, D. V. Mcallister, M. G. Allen, and M. R. Prausnitz, “Microfabricated microneedles: A novel approach to transdermal drug delivery,” Journal of Pharmaceutical Sciences, vol. 87, pp. 922-925, 1998. [23] W. Martanto, S. Davis, N. Holiday, J. Wang, H. Gill, and M. Prausnitz, “Transdermal delivery of insulin using microneedles in vivo,” Pharmaceutical Research, vol. 21, pp. 947-952, 2004. [24] J.A. Mikszta, J.B. Alarcon, J.M. Brittingham, D.E. Sutter, R.J. Pettis, and N.G. Harvey, “Improved genetic immunization via micromechanical disruption of skin-barrier function and targeted epidermal delivery,” Nature Medicine, vol. 8, pp. 415-419, 2002.  10  [25] P. Griss and G. Stemme, “Side-opened out-of-plane microneedles for microfluidic transdermal liquid transfer,” Journal of MEMS, vol. 12, pp. 296-301, 2003. [26] J. Brazzle, D. Bartholomeusz, R. Davies, and J. Andrade, “Active microneedles with integrated functionality,” Proceeding of Solid State Sensors and Actuators Workshop, Hilton Head, pp. 199-202, 2000. [27] J. Ji, F. E.H. Tay, and J. Miao, “Microfabricated Hollow Microneedle Array Using ICP Etcher,” Journal of Physics: Conference Series, vol. 34, pp. 1132-1136, 2006. [28] H. J. G. E. Gardeniers, R. Luttge, E. J. W. Berenschot, M. J. de Boer, S. Y. Yeshurun, M. Hefetz, R.van’t Oever, and A. van den Berg, “Silicon micromachined hollow microneedles for transdermal liquid transport,” Journal of MEMS, vol. 12, pp. 855-862, 2003. [29] E.V. Mukerjee, S.D. Collins, R.R. Isseroff, and R.L. Smith, “Microneedle array for transdermal biological fluid extraction and in situ analysis,” Sensors and Actuators A, vol. 114, pp. 267-275, 2004. [30] M. Shikida, M. Ando, Y. Ishihara, T. Ando, K. Sato, and K. Asaumi, “Nonphotolithographic pattern transfer for fabricating pen-shaped microneedle structures,” Journal of Micromechanics and Microengineering, vol. 14, pp. 1462-1467, 2004. [31] J. D. Zahn, N. H. Talbot, D. Liepmann, and A. P. Pisano, “Microfabricated Polysilicon Microneedles for Minimally Invasive Biomedical Devices,” Biomedical Microdevices, vol 2, pp. 295-303, 2000. [32] K. Chun, G. Hashiguchi, H. Toshiyoshi, and H. Fujita, “Fabrication of array of hollow microcapillaries used for injection of genetic materials into animal/plant cells,” Japanese Journal of Applied Physics, vol. 38, pp. L279-L281, 1999. [33] J.A. Matriano, M. Cormier, J. Johnson, W.A. Young, M. Buttery, K. Nyam, and P.E. Daddona, “Macroflux microprojection array patch technology: a new and efficient approach for intracutaneous immunization,” Pharmaceutical Research, vol. 19, pp. 6370, 2002. [34] S. Chandrasekaran and A. B. Frazier, “Characterization of Surface Micromachined Metallic Microneedles,” Journal of Microelectromechanical Systems, Vol. 12, pp. 289295, 2003. [35] J. D. Brazzle, I. Papautsky, and A. B. Frazier, “Hollow Metallic Micromachined Needle Arrays,” Biomedical Microdevices, vol. 2, pp. 197-205, 2000. [36] S. P. Davis, M. R. Prausnitz, and M. G. Allen, “Fabrication and characterization of laser micromachined hollow microneedles,” Proceedings of the 12th International Conference on Solid-State Sensors and Actuators, Boston, MA, pp. 1435-1438, 2003.  11  [37] D. V. McAllister, P. M.Wang, S. P. Davis, J.-H. Park, P. J. Canatella, M. G. Allen, and M. R. Prausnitz, “Microfabricated needles for transdermal delivery of macromolecules and nanoparticles: fabrication methods and transport studies,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, pp. 1375513760, 2003. [38] K. Kobayashi and H. Sizuki, “A sampling mechanism employing the phase transition of a gel and its application to a micro analysis system imitating a mosquito,” Sensors and Actuators B, Chemical, vol. 80, pp. 1-8, 2001. [39] K. Kim, D. S. Park, H. M. Lu,W. Che, K. Kim, J. Lee, and C. H. Ahn, “A tapered hollow metallic microneedle array using backside exposure of SU-8,” Journal of Micromechanics and Microengineering, vol. 14, pp. 597-603, 2004. [40] K. T. Brown and D. G. Flaming, Advanced Micropipette Techniques for Cell Physiology, Wiley, New York, 1986. [41] J. Park, M. G. Allen, M. R. Prausnitz, “Biodegradable polymer microneedles: Fabrication, mechanics and transdermal drug delivery,” Journal of Controlled Release, vol. 104, pp. 51-66, 2005. [42] M. Han, D. Hyun, H. Park, S. S Lee, C. Kim, and C. Kim, “A novel fabrication process for out-of-plane microneedle sheets of biocompatible polymer,” Journal of Micromechanics and Microengineering, vol. 17, pp. 1184-1191, 2007. [43] S. Kuo and Y. Chou, “A Novel Polymer Microneedle Arrays and PDMS Micromolding Technique,” Tamkang Journal of Science and Engineering, vol. 7, pp. 95-98, 2004. [44] H. Huang and C. Fu, “Different fabrication methods of out-of-plane polymer hollow needle arrays and their variations,” Journal of Micromechanics and Microengineering, vol. 17, pp. 393-402, 2007. [45] S. J. Moon, S. S. Lee, H. S. Lee, and T. H. Kwon, “Fabrication of microneedle array using LIGA and hot embossing process,” Microsystem Technologies, vol. 11, pp. 311318, 2005.  12  2 Fabrication  of  Hollow  Out-of-plane  Polymer Microneedles for Transdermal Drug Delivery1 2.1 Introduction Transdermal drug delivery is becoming increasingly popular because it is not associated with the potential risks and pain of drug administration through traditional hypodermic needles. One method for transdermal drug delivery uses adhesive skin patches. Various types of pressure sensitive adhesive patches (PSA) made from polyisobutylene, acrylic, and silicone are commercially available for use on skin [1] for applications such as the quitting of smoking (nicotine delivery) [2], pain relief (fentanyl delivery) [3, 4], and high blood pressure treatment (glyceryl trinitrate delivery) [5]. It is estimated that currently more than one billion transdermal patches are being manufactured each year [6]. However, adhesive skin patches have limitations and are therefore not the ultimate solution for delivering any kind of drug. The limitations include that the outer-most layer of the skin, i.e. the stratum corneum, is impermeable to most hydrophilic drugs and therapeutic agents and only some compounds of lipophilic nature can be effectively administered at useful rates using these patches [7]. In addition, adhesive skin patches potentially induce skin reactions upon application [8].  1  A version of this chapter will be submitted for publication. Mansoor, I., Häfeli, U.O., and Stoeber, B. Fabrication of Hollow Out-of-plane Polymer Microneedles for Transdermal Drug Delivery.  13  A transdermal drug delivery approach that potentially circumvents these limitations is the use of microneedles. Microneedles are sub-millimeter pointed structures, designed to pierce through the stratum corneum and deliver both lipophilic and hydrophilic compounds into the skin. In contrast to adhesive patches, microneedles are not limited by skin permeability; and unlike hypodermic needles, well-designed microneedles are painless [9, 10] and have a low risk of infection [11]. In addition, they require relatively low loads for application on skin [12]. Since the early 1990s, various concepts for fabrication of microneedles have been developed. In-plane and out-of-plane microneedles have been fabricated with solid and hollow structures in a variety of dimensions and geometries [13-38]. These devices have been fabricated out of silicon [13-23], polysilicon [24, 25], metals [26-32], glass [33], and polymers [34-38]. The lengths of the needle shafts in these works range from 30 µm up to 1 mm. Out-of-plane microneedles with solid structures are designed to increase skin permeability, while hollow microneedles are typically designed to penetrate though the stratum corneum and release agents though microneedle channels into the epidermis. The released agents then diffuse toward the capillary bed of the dermis where they are absorbed by the blood stream. Since the needle tips are not required to be in direct contact with the blood vessels and the nerve endings, these devices are painless if they do not penetrate deeper than the epidermis [39] and bear a low risk of hazardous device contamination through blood. Transdermal drug delivery through hollow out-of-plane microneedles has been demonstrated in clinical trials, where methyl nicotinate was injected using pointed hollow microneedles [40]. During this study, the volunteers confirmed that the method was painless and they only felt a feeling of pressure during the  14  injection. In a different study, an array of 16 500 µ-tall single hollow metal microneedles were used to deliver insulin to diabetic hairless rats [41]; injection of 2 ml of insulin at 30 psi resulted in 47% reduction in the blood glucose level over a 4-hr period. Expensive fabrication techniques currently used to form microneedles such as deep reactive ion etching of silicon needles [13, 18, 20], sequential formation of disposable polymer molds for electroplating [26, 32] or multiple UV exposure or mold transfer and assembly to form polymer needles [34-36] have caused a barrier for microneedles entry into the commercial market. Here, we present a simple solvent casting fabrication method for  hollow  polymer  out-of-plane  microneedles  requiring  only  one  step  of  photolithography. Our fast and inexpensive fabrication process uses a single mold for sequential solvent casting of polymer films, eliminating the need of mask alignment. This process is very versatile in terms of the possible dimensions of the microneedle array as well as the needle material. In this work, first the detailed fabrication process of microneedles is presented and discussed. This is followed by a study of the mechanical strength as well as the penetration capability of the needles, and then followed by the demonstration of fluid injection into the skin of a live mouse.  15  2.2 Fabrication  Process  for  Hollow  Out-of-plane  Polymer Microneedles Microneedles were formed though a solvent casting process shown in Fig 2.1. Fabrication of microneedles started with fabrication of a mold containing an array of pillars, followed by a sacrificial layer deposition. A second polymer solution, corresponding to the needle structural material, was then deposited on top of the sacrificial layer and then let dry to form thorn-shaped structures around the pillars. After curing the structural polymer layer, a plasma etching step was used to sharpen the needle tips and at the same time provide access to the sacrificial layer. Finally, the microneedle array was separated from the mold by removing the sacrificial layer.  16  Figure 2.1 Fabrication process using solvent casting for hollow polymer out-of-plane microneedles. (a) & (b) fabrication of pillars from SU-8 (c) Parylene C deposition (d) & (e) PMMA deposition & solvent evaporation (f) & (g) SU-8 + cyclopentanone deposition & evaporation of cyclopentanone (h) UV exposure of remaining SU-8 layer on pillars (i) removing of SU-8 layer from the top of the pillars and sharpening of the needles using plasma etching (j) dissolving PMMA layer in chloroform which releases the needle structure.  17  The epoxy-type negative photoresist SU-8 (Microchem) was used for the mold as well as the structural material for the needles, while PMMA (Microchem) was used as a sacrificial release layer. SU-8 cross-links upon exposure to UV light during photolithography, forming a stiff material. A 350 µm layer of SU-8 2150 (Microchem) was spin-coated on a 500 µm thick Pyrex glass substrate, and then soft-baked at 65°C for 10 min and at 95°C for 1 hr and 40 min on a hotplate (See Appendix A for detailed photolithography recipes). The SU-8 was then exposed to 2484 mJ cm-2 UV light through a dark field mask that contained arrays of circular transparent regions with diameters of 40 µm (see Appendix B). The UV light power density was 6.9 mW cm-2 and the exposure was done in two 3 min steps with a 10 s rest in between. The exposure was also performed through the base glass substrate, as shown in Figure 2.1a, in order to take advantage of light diffraction caused by the gap between the mask and the photoresist [42, 43]. This method of exposure results in tapered pillar structures with the wider bases attached to the Pyrex base plate (Figure 2.2). After exposure, the sample was baked on a hotplate at 65°C for 5 min and at 95°C for 30 min. Next, the sample was immersed in SU-8 developer for approximately 80 min and then washed for 5 min with fresh developer and isopropanol. The resulting structure was an array of slightly tapered cylindrical pillars (Figure 2.1b), with a base diameter of 60 µm and tip diameter of 30 µm. These pillars constitute the mold and they will eventually form the lumens of the needles.  18  Figure 2.2 SEM image of an array of tapered pillars fabricated using backside exposure of SU-8. The vertical ridges along the pillars are created due to the poor resolution of the photomask.  The complete mold structure contained an array of pillars surrounded by a square (8 mm × 8 mm) of vertical walls, as shown in Figure 2.3.  1 mm  Figure 2.3 Microneedle mold consisting of an array of pillars surrounded by square vertical walls.  19  The entire mold structure was then coated with a 3 µm layer of Parylene C (Figure 2.1c). This layer improved adhesion of the pillars to the Pyrex substrate and provided a protective surface on the mold for consecutive fabrication steps. 20 µl of PMMA A4 (4 wt% PMMA in anisole, Microchem) was then deposited onto the pillar array using an adjustable pipette (Figure 2.1d). The sample was then placed on a hotplate at 70°C for 15 min. After evaporation of the solvent, an approximately 5 µm thick PMMA layer remained coating the pillars (Figure 2.1e), providing a sacrificial layer to be removed after needle formation for separating the microneedle array from the mold. Figure 2.4 shows SEM image of an array of pillars (on silicon substrate) coated with a uniform layer of PMMA that was dried at 70°C.  Figure 2.4 SEM image of PMMA solution dried at 70°C on an array of pillars.  A series of experiments showed that the temperature of the hotplate is an important process parameter; choosing a temperature too high (more than 100°C) would lead to non-uniform accumulation of the PMMA layer on top of the pillars, as shown in Figure 2.5. 20  Figure 2.5 SEM image of PMMA solution dried at 120°C on an array of pillars.  A 25 µl mixture of SU-8:SU-8 solvent (cyclopentanone) at 1:3 by volume was then deposited into the mold structure (Figure 2.1f). The mold was then placed on hotplate for 1 hr at 45°C. This temperature is a critical process factor since it impacts the final shape of the needle structure (the impact of temperature on polymer film flow is discussed in Chapter 3). After complete evaporation of the cyclopentanone, the polymer formed the microneedles with wide bases and a 200 µm thick backing plate, as shown in Figure 2.1g. The entire sample was then exposed to 1000 mJ cm-2 of UV light (with power density of 6.9 mW cm-2 for 2 min and 25 s) which causes the structural SU-8 layer to cross link. Concurrently, the covalent bonds of the positive photoresist PMMA were broken to facilitate subsequent removal of the sacrificial layer (Figure 2.1h). From the SU-8 solvent casting step, a thin layer of SU-8 resided on the top surface of the pillars. This layer was removed by a 300 s plasma etching step (100 W power, 90% O2 and 10% CF4), using a Trion RIE/PECVD tool (Trion Technology), which simultaneously sharpened the needles (Figure 2.1i). For this purpose, the chamber pressure and RF power settings were 21  optimized to achieve a more isotropic etch process. Having a more isotropic etching step would lead to a faster etching rate in the horizontal direction and therefore results in sharper devices without losing their height. Table 2.1 summarizes the etch rates in horizontal and in vertical direction for four different combinations of RF power and gas chamber pressure. Table 2.1 Vertical and horizontal plasma etch rates, measured for different combinations of RF power and gas chamber pressure, for the Trion RIE/PECVD tool. A higher ratio of the horizontal etch rate over the vertical etch rate indicates a more isotropic etching process.  Chamber Pressure (Torr)  RF Power (W)  Vertical Etch Rate (µm/min)  Horizontal Etch Rate (µm/min)  Horizontal Etch Rate / Vertical Etch Rate  0.55  100  0.73  0.13  0.17  0.55  200  0.98  0.26  0.27  1.6  100  1.02  0.43  0.42  1.6  200  1.25  0.48  0.38  It was observed that higher pressure (1.6 Torr) and lower RF power (100 W) resulted in a relatively faster etch rate in the horizontal direction with respect to the vertical direction (0.42:1), and therefore, a more isotropic plasma etching. To guarantee these optimal conditions, a chamber pressure of 1.6 Torr and an RF power of 100 W were chosen for the SU-8 etching process. Lifting off the microneedle array was finally achieved by  22  dissolving the sacrificial PMMA layer with chloroform at 60°C for a period of 30 min (Figure 2.1j). The presented fabrication process has a yield of 60-70% in terms of the percentage of molds leading to intact microneedle arrays. Failures in the process occur in the microneedle lift-off step as shown in Figure 2.6. The process therefore requires additional characterization. This is discussed in more detail in chapter 4.  1 mm  Figure 2.6 Image of the microneedle array destroyed and stuck in mold after a failed lift-off.  Figure 2.7 shows SEM images of the microneedles fabricated through this process. Figure 2.7a shows an array of 110 µm long microneedles with 45 µm tip diameter. The dimensions of the microneedles can be adjusted by changing the dimensions of the mold structure during the photolithography steps. For instance, longer pillars with smaller diameter result in longer needles with smaller channel diameters. Figure 2.7b shows an array of 180 µm long microneedles with 30 µm tip diameter, and Figure 2.7c shows a single 180 µm long microneedle. Wider bases of pillars, due to the backside exposure of the SU-8, result in larger channel openings on the backside of the backing plate as shown in Figure 2.7d. The horizontal and the vertical pitch between needles in both designs is  23  300 µm. See Appendix C for more SEM images of microneedles fabricated using this process.  Figure 2.7 SEM image of microneedles fabricated using the process illustrated in Figure 2.1. The distance between needles in the arrays is 300 µm (a) an array of microneedles 110 µm long with approximately 45 µm tip diameter (b) an array of microneedles 180 µm long with approximately 30 µm tip diameter (c) a single 180 µm microneedle from the array shown in b (d) microneedles lumen openings on the backside of the array.  24  2.3 Test Procedures 2.3.1 Microneedle Robustness Tests Microneedles were tested for robustness under compressive loading conditions. Vertical in-plane loads of up to 4 N were applied to arrays of 40 microneedles of 110 µm length with a tip diameter of 35 µm. For this purpose, a test rig was constructed with a light weight flat aluminum arm that was secured on a rigid support and allowed to pivot freely as shown in Figure 2.8.  Figure 2.8 Schematics of the experimental setup used for testing microneedle robustness under different compressive loads.  The microneedle arrays were placed on a 1 mm thick slab of silicone rubber located on an adjustable vertical stage. The stage was placed under the arm and was adjusted so that the arm would be in horizontal position in order to distribute the load on all needles evenly. Subsequently various loads were applied to the free end of the arm. The load applied to the microneedle array  25  FN =  WL + FL L ' d  (1)  is found from the load applied to the end of the arm F L , the weight of the arm W , the distance between the fixture on the support and the applied load L' , the distance between the fixture on the support and the middle of the arm L , and the distance between the arm’s center of rotation and the reaction load d , as shown in the schematic in Figure 2.8. The microneedles were inspected for any sign of failure after application of each load using an Olympus SZ61stereo microscope.  2.3.2 Microneedle Insertion Tests In order to investigate the suitability of microneedles for skin penetration, preliminary insertion trials were performed on potato skin. For this purpose, an array of 40 microneedles containing 110 µm long needles was pressed against fresh potato skin with an approximate load of 3.5 N. After pressing the microneedle array on the skin, a section of the potato containing the puncture site was cut and then frozen using dry ice. Histology sections of the frozen sample with thickness of 10 µm were then obtained (Wax-it Histology) and inspected using a PSM-1000 microscope (Signatone). For further investigating the suitability of the microneedles for skin puncture, additional experiments were performed to test the impact of microneedle insertion on skin permeability. For this purpose, similar to the previous experiment, a microneedle array, containing 42 110 µm long needles, was pressed against potato skin with a load of 3.5 N. After the needles were removed from the application site, a 100 µl drop of blue food color was placed on the site using a syringe. The food ink was then allowed to be 26  absorbed into the potato for 20 s, through the holes created by the microneedle array. Similarly, another 100 µl drop of blue food color was placed on a non-damaged site of the potato skin for comparison.  2.3.3 Microneedle Injection Tests The epidermal delivery capability of the microneedles was investigated by injection of human serum albumin (HSA, 67 kDa protein) into the body of a live mouse. This molecule is relatively large and is intended to stay in the epidermis after injection. HSA was radiolabeled with  99m  Tc, by reducing  99m  Tc-pertechnetate using Sn(II), so that it can  bind in Tc(+V) form to HSA. For this purpose, a freshly prepared solution of 0.1 mg of SnCl2, 1 µl of sodium potassium tartrate, and 0.5 ml of 0.01 N HCl (all from Sigma Aldrich) was added to 0.5 ml of an aqueous solution containing 0.1 mg of HSA. 370 MBq of sodium pertechnetate Na99mTcO4¯ was added in 250 µl of saline to the HSA and incubated for 10 min at 37°C. Sodium pertechnetate Na99mTcO4¯ was kindly provided by Vancouver General Hospital. The radiolabeled HSA was concentrated using an Ultracel YM-30 column (Millipore Corp.) by centrifugation at 12,000 g for 10 min. The labeling efficiency was 89.2%. The solution was then diluted with saline. A 2.2 mm × 1.9 mm array of microneedles, containing 42 needles of 110 µm length, was attached to a syringe using a fast curing 2-component epoxy adhesive as shown in Figure 2.9. The syringe was then filled with the HSA solution and pressed for 10 sec against a shaved skin section of a C57BL/6 mouse. After pressing for 10 s, a portion of the HSA solution in the syringe was injected through the microneedle array. The amount of radioactivity before and after injection was measured to be 1200 µCi and 1040 µCi, respectively; therefore, the net  27  injected activity was 160 µCi which was delivered through 20 µl of saline. After 40 min, the mouse was sacrificed and different organs were analyzed for activity using a CRC-15 dose calibrator (Capintec) and a Cobra II gamma counter (Packard).  Array of 42 microneedles with 300 µm spacing bonded to the end of syringe  0.5 cm Figure 2.9 An array of 42 microneedles bonded to the tip of a 1 ml syringe, used for the injection experiment.  2.4 Results and Discussion 2.4.1 Microneedles Robustness In a simple strength analysis, the needles were modeled as straight tubes with 25 µm inner diameter and 5 µm wall thickness. This model represents the worst possible case in terms of the needle strength since the actual needles have much larger wall thickness at the bottom. The maximum load such tubes can sustain  FMax = σ Max × π (ro − ri ) 2  (2)  28  can be calculated from the compressive yield stress σ Max of SU-8, and the cross sectional area of the tube wall. In this case for σ Max= 34 Mpa, ro = 17.5 µm, and ri = 12.5 µm, the maximum calculated sustainable load was 0.016 N for each tube. After applying a wide range of loads to the fabricated microneedle arrays using the apparatus shown in Figure 2.8, it was observed that each individual needle is capable of sustaining inward compressive loads of up to 0.1 N without exhibiting any sign of bending or fracture. Loads beyond this amount caused the rim of the needles to bend inwards, blocking the needle lumens. As expected, this value is well above the maximum sustainable load of the worst case model. This is due to the thicker wall thickness of the fabricated devices, which increases gradually from 5 µm at the needle tip to 80 µm at the needle base. Davis et al. investigated the relationship between the cross sectional area of a microneedle tip (interfacial area) and the force required for penetration of the microneedle into skin [12]. In this work, the authors have derived a linear relationship between insertion force and the interfacial area from experimental data. This study only investigates microneedles with diameters ranging from 60 µm to 160 µm. This range does not include the tip diameter of devices reported in this work (35 µm) but it has been used here as the only available study for comparison. According to this study, the insertion force required for skin puncture Fi [ N ] = 0 .00019 A f [ µ m 2 ] − 0 . 66  (3)  29  is dependent on the full cross-sectional area of the needle tip A f . Plugging in an insertion force of 0.1 N in equation (3) and calculating the corresponding interfacial area (= 4000 µm2), it can be concluded that the needles with tip diameters of smaller than 71.4 µm require insertion forces of less than 0.1 N for penetration into human skin; however, since not enough experimental data is available, it is inadequate to draw convinced conclusions for the designed devices, and therefore further investigation is necessary to predict the appropriate insertion force.  2.4.2 Results of Insertion Tests The histology sections, obtained after application of a microneedle array on potato skin, indicate successful penetration of the needles into the skin. The histology image in Figure 2.10 shows a hole created by a needle of the microneedle array.  40 µm  Hole created by a microneedle  Potato skin Surface  Figure 2.10 Histology image of potato skin showing a hole created by a microneedle.  30  Figure 2.11 shows a SEM image of potato skin after microneedle array application as well as an image of the needles after insertion into the potato skin. The holes created on the skin surface by the microneedles can be observed in Figure 2.11a. Figures 2.10 and 2.11 suggest that the fabricated devices were strong enough to puncture the potato skin and penetrate well below the surface, since there is no damage observed on the skin surface around the holes and the microneedle array after application. (a)  Holes created on the surface by the microneedle array  (b)  300 µm  Figure 2.11 (a) SEM image of potato skin after application of an array of microneedles (b) image of the microneedles after the insertion experiment.  31  Figure 2.12 shows the application of food ink on the surface of potato skin onto a microneedle insertion site as well as onto a non-damaged site for comparison. Penetration of food ink through the holes created on the skin by the microneedle array indicates that the microneedles have substantially improved the potato skin permeability (Figure 2.11b left); whereas without the holes the ink was not able to penetrate through the skin (Figure 2.11b right).  Application of food ink after pressing microneedles on the surface  Application of food ink on the surface without applying microneedles  (a)  A  A  (b)  Figure 2.12 Food color application on potato skin (a) application site with (left) and without (right) applying microneedles beforehand (b) cross section A:A.  32  2.4.3 Results of Injection Tests Table 2.2 summarizes the percentage of radioactivity that was detected in different organs of the animal subject as well as the amount of activity that was not recovered or effectively delivered. Table 2.2 Relative radioactivity biodistribution, measured 1 hour after transdermal microneedle injection of 160 µCi of 99mTc radiolabeled human serum albumin.  Recovery Site  Percentage of Activity  Blood  0.02%  Heart  0.00%  Liver  0.02%  Skin  6.91%  Wipe 1  10.69%  Wipe 2  4.43%  Wipe 3  3.19%  Carcass  52.87%  Not Recovered  21.86%  Of the 78.1% of the activity recovered outside the syringe, 18.3% was bound to the surface of the skin around the injection site and could be removed using wet tissue 33  papers. Most of the recovered radioactivity, 52.9%, was measured in the carcass of the animal excluding the blood, heart, liver, and the skin around injection site. About 6.9% was detected in the skin around the injection site. This corresponds to the radiolabeled protein that is delivered to its target site and therefore demonstrates the successful penetration of some of the needles in to the skin of the animal. This also suggests that many needles were most likely bent or broken during the application of the array on the skin. From the recovered activity, only 0.02% was measured in the blood of the animal, which is probably due to the impermeability of the blood vessel membrane to large HSA proteins which prevents diffusion of the proteins from skin tissue across the wall of the blood vessels.  2.5 Conclusions A new fabrication technique, based on solvent casting, has been presented for fabrication of hollow out-of-plane microneedles. This fabrication procedure is flexible in terms of needle dimensions and the choice of polymer material. In addition, the process is inexpensive and is therefore applicable for mass production of polymer microneedles. Sample microneedle arrays using this process have been fabricated. The robustness of the fabricated microneedles was measured and compared with the required insertion force into human skin presented in a previous study. It was observed that the measured sustainable force of the microneedles might be enough for human skin puncture; however, due to the lack of experimental data in the study, additional experiments have to be performed to validate the fabricated microneedles’ strength.  34  After performing some insertion trials, the needles were found to be strong enough for penetration into potato skin. Furthermore, radioactive protein injection through the fabricated microneedles into a live mouse demonstrated partial efficacy of the fabricated devices for in vivo drug delivery. Overall, the presented fabrication process requires additional work in order to change it to a fully repeatable process. The test results suggest that further investigation on the choice of polymer material is necessary in order to produce stronger needles for more effective drug delivery.  2.6 Acknowledgment The authors acknowledge contribution from Vancouver General Hospital for providing sodium pertechnetate used in the in vivo experiment, and financial support from the UBC Faculty of Applied Science and British Columbia Innovation Council for funding through the BCIC Innovation Scholarship program.  35  2.7 References [1] S. Venkatraman and R. Gale, “Skin adhesives and skin adhesion: 1. Transdermal drug delivery systems,” Biomaterials, vol. 19, pp. 1119-1136, 1998. [2] S. K. Govil and P. Kohiman, “Transdermal Delivery of Nictoine,” US Patent: 4908213, 1990. [3] K. J. Miller, S. K. Govil, and K. S. Bhatia, “Fentanyl Suspension-based Silicone Adhesive Formulations and Devices for Transdermal Delivery of Fentanyl,” US Patent: 7556823, 2009. [4] I. Takeshi, T. Tetsuro, and H. Naruhito, “Transdermal Patch for External Use Compromising Fentanyl,” International Patent: WO/2004/035054, 2004. [5] S. K. Govil, E. M. Rudnic, and D. G. Sterner, “Transdermal nitroglycerin patch with penetration enhancers,” US Patent: 5262165, 1993. [6] M. R. Prausnitz and R. Langer, “Transdermal drug delivery,” Nature Biotechnology, vol. 26, pp. 1261-1268, 2008. [7] A. Naik, Y. N. Kalia, and R. H. Guy, “Transdermal drug delivery: overcoming the skin barrier’s function,” Pharmaceutical Science & Technology Today, vol. 3, pp. 318-326, 2000. [8] M. Murphy and A. J. Carmichael, “Transdermal Drug Delivery Systems and Skin Sensitivity Reactions: Incidence and Management,” American Journal of Clinical Dermatology, vol. 1, pp. 361-368, 2000. [9] S. Kaushik, A. H. Hord, D. D. Denson, D. V. McAllister, S. Smitra, M. G. Allen, and M. R. Prausnitz, “Lack of Pain Associated with Microfabricated Microneedles,” Anesthesia & Analgesia, vol. 92, pp. 502-504, 2001. [10] H. S. Gill, D. D. Denson, B. A. Burris, and M. R. Prausnitz, “Effect of Microneedle Design on Pain in Human Volunteers,” The Clinical Journal of Pain, vol. 24, pp. 585-594, 2008. [11] P. Zhang and G. A. Jullien, “Microneedle arrays for drug delivery and fluid extraction,” Proceedings of 2005 international conference on MEMS, NANO, and Smart Systems, pp. 392-395, 2005.  36  [12] S. P. Davis, B. J. Landis, Z. H. Adams, M. G Allen, and M. R. Prausnitz, “Insertion of microneedles into skin: measurement and prediction of insertion force and needle fracture force,” Journal of Biomechanics, vol. 37, pp. 1155-1163, 2004. [13] B, Stoeber and D. Liepmann, “Arrays of hollow out-of-plane microneedles for drug delivery,” Journal of MEMS, vol. 14, pp. 472-479, 2005. [14] S. J. Paik, S. Byun, J. M. Lim, Y. Park, A. Lee, S. Chung, J. Chang, K. Chun, and D. Cho, “In-plane single-crystal-silicon microneedles for minimally invasive microfluid systems,” Sensors and Actuators A: Physical, vol. 114, pp. 276-284, 2004. [15] S. Henry, D. V. Mcallister, M. G. Allen, and M. R. Prausnitz, “Microfabricated microneedles: A novel approach to transdermal drug delivery,” Journal of Pharmaceutical Sciences, vol. 87, pp. 922-925, 1998. [16] W. Martanto, S. Davis, N. Holiday, J. Wang, H. Gill, and M. Prausnitz, “Transdermal delivery of insulin using microneedles in vivo,” Pharmaceutical Research, vol. 21, pp. 947-952, 2004. [17] J.A. Mikszta, J.B. Alarcon, J.M. Brittingham, D.E. Sutter, R.J. Pettis, and N.G. Harvey, “Improved genetic immunization via micromechanical disruption of skinbarrier function and targeted epidermal delivery,” Nature Medicine, vol. 8, pp. 415419, 2002. [18] H. J. G. E. Gardeniers, R. Luttge, E. J. W. Berenschot, M. J. de Boer, S. Y. Yeshurun, M. Hefetz, R.van’t Oever, and A. van den Berg, “Silicon micromachined hollow microneedles for transdermal liquid transport,” Journal of MEMS, vol. 12, pp. 855-862, 2003. [19] J. Brazzle, D. Bartholomeusz, R. Davies, and J. Andrade, “Active microneedles with integrated functionality,” Proceeding of Solid State Sensors and Actuators Workshop, Hilton Head, pp. 199-202, 2000. [20] J. Ji, F. E.H. Tay, and J. Miao, “Microfabricated Hollow Microneedle Array Using ICP Etcher,” Journal of Physics: Conference Series, vol. 34, pp. 1132-1136, 2006. [21] P. Griss and G. Stemme, “Side-opened out-of-plane microneedles for microfluidic transdermal liquid transfer,” Journal of MEMS, vol. 12, pp. 296-301, 2003. [22] E.V. Mukerjee, S.D. Collins, R.R. Isseroff, and R.L. Smith, “Microneedle array for transdermal biological fluid extraction and in situ analysis,” Sensors and Actuators A, vol. 114, pp. 267-275, 2004. [23] M. Shikida, M. Ando, Y. Ishihara, T. Ando, K. Sato, and K. Asaumi, “Nonphotolithographic pattern transfer for fabricating pen-shaped microneedle structures,” Journal of Micromechanics and Microengineering, vol. 14, pp. 1462-1467, 2004. 37  [24] J. D. Zahn, N. H. Talbot, D. Liepmann, and A. P. Pisano, “Microfabricated Polysilicon Microneedles for Minimally Invasive Biomedical Devices,” Biomedical Microdevices, vol 2, pp. 295-303, 2000. [25] K. Chun, G. Hashiguchi, H. Toshiyoshi, and H. Fujita, “Fabrication of array of hollow microcapillaries used for injection of genetic materials into animal/plant cells,” Japanese Journal of Applied Physics, vol. 38, pp. L279-L281, 1999. [26] K. Kim, D. S. Park, H. M. Lu,W. Che, K. Kim, J. Lee, and C. H. Ahn, “A tapered hollow metallic microneedle array using backside exposure of SU-8,” Journal of Micromechanics and Microengineering, vol. 14, pp. 597-603, 2004. [27] J.A. Matriano, M. Cormier, J. Johnson, W.A. Young, M. Buttery, K. Nyam, and P.E. Daddona, “Macroflux microprojection array patch technology: a new and efficient approach for intracutaneous immunization,” Pharmaceutical Research, vol. 19, pp. 63-70, 2002. [28] S. Chandrasekaran and A. B. Frazier, “Characterization of Surface Micromachined Metallic Microneedles,” Journal of Microelectromechanical Systems, Vol. 12, pp. 289-295, 2003. [29] J. D. Brazzle, I. Papautsky, and A. B. Frazier, “Hollow Metallic Micromachined Needle Arrays,” Biomedical Microdevices, vol. 2, pp. 197-205, 2000. [30] S. P. Davis, M. R. Prausnitz, and M. G. Allen, “Fabrication and characterization of laser micromachined hollow microneedles,” Proceedings of the 12th International Conference on Solid-State Sensors and Actuators, Boston, MA, pp. 1435-1438, 2003. [31] K. Kobayashi and H. Sizuki, “A sampling mechanism employing the phase transition of a gel and its application to a micro analysis system imitating a mosquito,” Sensors and Actuators B, Chemical, vol. 80, pp. 1-8, 2001. [32] D. V. McAllister, P. M.Wang, S. P. Davis, J.-H. Park, P. J. Canatella, M. G. Allen, and M. R. Prausnitz, “Microfabricated needles for transdermal delivery of macromolecules and nanoparticles: fabrication methods and transport studies,” Proceedings of the National Academy of Sciences of the United States of America, vol. 100, pp. 13755-13760, 2003. [33] K. T. Brown and D. G. Flaming, Advanced Micropipette Techniques for Cell Physiology, Wiley, New York, 1986. [34] S. J. Moon, S. S. Lee, H. S. Lee, and T. H. Kwon, “Fabrication of microneedle array using LIGA and hot embossing process,” Microsystem Technologies, vol. 11, pp. 311-318, 2005. [35] J. Park, M. G. Allen, M. R. Prausnitz, “Biodegradable polymer microneedles: Fabrication, mechanics and transdermal drug delivery,” Journal of Controlled Release, vol. 104, pp. 51-66, 2005. 38  [36] M. Han, D. Hyun, H. Park, S. S Lee1, C. Kim, and C. Kim, “A novel fabrication process for out-of-plane microneedle sheets of biocompatible polymer,” Journal of Micromechanics and Microengineering, vol. 17, pp. 1184-1191, 2007. [37] H. Huang and C. Fu, “Different fabrication methods of out-of-plane polymer hollow needle arrays and their variations,” Journal of Micromechanics and Microengineering, vol. 17, pp. 393-402, 2007. [38] S. Kuo and Y. Chou, “A Novel Polymer Microneedle Arrays and PDMS Micromolding Technique,” Tamkang Journal of Science and Engineering, vol. 7, pp. 95-98, 2004. [39] R. K Sivamani, D. Liepmann, and H. I Maibach, “Microneedles and transdermal applications,” Expert Opinion on Drug Delivery, vol. 4, pp. 19-25, 2007. [40] R. K. Sivamani, B. Stoeber, G. C. Wu, H. Zhai, D. Liepmann, and H. Maibach, “Clinical microneedle injection of methyl nicotinate: stratum corneum penetration,” Skin Research and Technology, vol. 11, pp. 152-156, 2005. [41] S. P. Davis, W. Martanto, M. G. Allen, and M. R. Prausnitz, “Hollow Metal Microneedles for Insulin Delivery to Diabetic Rats,” IEEE Transactions on Biomedical Engineering, vol. 52, pp. 909-915, 2005. [42] W. Kang, E. Rabe, S. Kopetz, and A. Neyer, “Novel exposure methods based on reflection and refraction effects in the field of SU-8 lithography,” Journal of Micromechanics and Microengineering, vol. 16, pp. 821-831, 2006. [43] Y. J. Chuang, F.G. Tseng, and W.K. Lin, “Reduction of diffraction effect of UV exposure on SU-8 negative thick photoresist by air gap elimination,” Microsystem Technologies, vol. 8, pp. 308-313, 2002.  39  3 PIV Measurements of Flow in Thin Drying Films during Solvent Casting2 3.1 Introduction Solvent casting is a common method for forming polymer articles. It has been used for large scale thin film manufacturing for more than a century [1]. In the solvent casting process, a polymer solution is applied to a mold; after solvent evaporation, a layer of polymer remains on the cast which is then separated by physical or chemical methods. The versatile nature of this process allows rapid fabrication of polymer devices in a variety of shapes. For instance, solvent casting is used as the most common method for fabrication of scaffolds for tissue engineering [2-4]. Solvent casting can also be used for the fabrication of polymer-based micro/nano-scale devices. While the microelectromechanical systems (MEMS) and the nanofabrication industry are seeking cheaper alternatives to some of the current expensive fabrication methods, solvent casting offers an inexpensive alternative for the formation of structures and coatings made from polymers. A variety of polymers and polymer blends such as polyacrylamide, poly(ethylene oxide), poly(N-hydroxyethylacrylamide), and poly(vinyl alcohol) (PVA) have been used to coat microchannels in microfluidic systems for DNA and proteins separation, and also to modify surface properties such as hydrophilicity [5]. Solvent casting also has the potential to be used in a similar way as conventional thin film 2  A version of this chapter has been submitted for publication. Mansoor, I. and Stoeber, B. PIV Measurements of Flow in Thin Drying Films during Solvent Casting.  40  technology where the structural layers are deposited on top of sacrificial layers that are later removed. Similar to conventional thin film technology, fine tuning the process conditions will allow achieving more or less conformal layers. Previous works have investigated the surface pattern [6-8], optical properties [9, 10], rheology [11], internal strain [12], and residual internal stresses [13] of solvent cast polymer and polymer composite films in order to optimize the process conditions. In one of the studies performed on surface characteristics of solvent cast polymer films [7], the authors have used confocal laser scanning microscope (CLSM) and atomic force microscope (AFM) to compared the surface and bulk morphology of solvent cast poly(methyl methacrylate) (PMMA)/polystyrene (PS) blend films under different drying conditions with toluene as the solvent. According to this study, the mean toluene evaporation rate strongly affects the surface and bulk morphology of solvent cast PMMA/PS films. By sufficiently suppressing the toluene evaporation rate, a highly monodisperse distribution of PMMA-rich regions on the surface can be achieved. In another work, confocal microscopy is used to investigate polymer film formation kinetics in a process similar to solvent casting [14]. This study reveals that the final thickness of the polymer film and its shape varies considerably with the initial polymer concentration and deposition volume. Optimizing the process conditions of solvent casting is crucial for the fabrication of MEMS or nanodevices where tight tolerances of the profile dimensions need to be achieved. Therefore, it is useful to characterize the flow of polymer solutions during solvent evaporation in order to characterize the drying process and establish models to predict the final shape of the polymer profile. As an example, the polymer thickness 41  along a vertical wall in a mold from top to bottom is an important device dimension, and it can be adjusted by controlling the temperature during the solvent evaporation phase, which affects the flow in the polymer solution as will be shown. Until now, no study has been carried out to investigate the effect of temperature on microscopic flow during the solvent casting process and on the resulting structure. Instantaneous velocity measurement of the flow can be achieved using particle image velocimetry (PIV). PIV, derived from laser speckle velocimetry (LSV) in the early 1980s [14] and introduced for digital imaging in the early 1990s [16], is an increasingly common method used to characterize flow both at the macroscopic scale [17] and at the microscopic scale [18]. In this technique, the velocity field is obtained from the displacement of seed particles that are transported with the flow under investigation. The emitted light from fluorescent seed particles is recorded under flow conditions, where their images are captured in well-defined time intervals. The displacement of the particles is obtained by cross-correlating corresponding regions from pairs of subsequent images, and the velocity field is calculated based on the time delay between two images of one pair. There are many variations of this technique such as stereoscopic PIV, micro-PIV, or tomographic PIV [17]. Micro-PIV was first demonstrated in 1998 [18] as a PIV system with sufficiently high spatial resolution to measure velocity fields in microscopic systems. Since then, numerous works have used micro PIV to analyze flow in microfluidic systems [19-22]. A typical micro-PIV system contains a standard epifluorescence microscope for acquiring 2D images, equipped with a fluorescence detector.  42  Instead of standard epifluorescence microscopes, CLSM can be used to acquire the spatial distribution of tracer particles in a volume. Equipped with the necessary data processing tools, such systems have been used by several research groups instead of the conventional micro-PIV for acquiring 3D flow profiles in microfluidic channels and this method is referred to as a confocal micro-PIV [23-26]. In typical confocal micro-PIV systems, a special type of CLSM (Nipkow Disk CLSM) is used for real time image acquisition with scanning rates as fast as 2000 frames per second (fps) [23]. Although using a Nipkow Disk CLSM provides a faster scanning rate, regular CLSM offer better spatial resolution for thick scan volumes [27], which is necessary to detect the important details in the flow during solvent casting such as the low-velocity surface counter flow as will be shown below. The aim of this study is to demonstrate a confocal micro-PIV system based on a conventional CLSM as an experimental method that is adequate to characterize the flow of a polymer solution during solvent evaporation in a solvent casting process. Threedimensional images of the fluorescent seed particles in the solution are captured by a CLSM, processed, and then fed into a PIV analysis tool. We apply this method to the investigation of the effect of temperature on the flow of the polymer solution during solvent casting and observe the flow behavior near a vertical wall as the solvent evaporates. Different flow phenomena inside the polymer film and at its surface are revealed with this method for different phases of the evaporation process. The flow observations are then compared with the polymer structures formed through this process.  43  3.2 Materials and Methods 3.2.1 Polymer Solution An aqueous solution of PVA was used as the sample polymer solution for these experiments. An 11 wt% solution of PVA was prepared by dissolving 2.2 g of dry PVA (typical MW 85,000 – 124,000, Sigma Aldrich) into 20 ml of distilled water at 55ºC. Polystyrene fluorescent particles (diameter 2.28 µm, Bangs Laboratories) with an excitation peak at 540 nm and an emission peak at 600 nm were used as tracer particles for the PIV measurement. The original 0.1 wt% suspension of the fluorescent particles was diluted with water to 0.01 wt%. About 0.88 g of the diluted suspension was then mixed with 3 g of the 11 wt% PVA solution using a Branson 1510 ultra sonic bath (Crystal Electronics) for 10 min at room temperature. The mixture was then left under vacuum for 5 min to remove air bubbles. The seed particle and PVA concentrations were 0.0023 wt% and 8.5 wt%, respectively, in the final mixture.  3.2.2 Experimental Setup The experimental setup, shown in Figure 3.1, mainly consists of an inverted stage Eclipse TE2000-U microscope (Nikon), a D-Eclipse C1 confocal laser scanning unit (Nikon), a green (543.5 nm) HeNe laser source (Melles Griot), a Retiga EXi CCD digital camera (QImaging ), and a temperature control chamber for the microscope stage.  44  Figure 3.1 The experimental setup used to characterize flow during solvent evaporation in solvent casting process.  The temperature control chamber is a custom design. It uses a Peltier device for heating and cooling the inside of the chamber, and an RTD for temperature feedback measurement. The air temperature inside the chamber is controlled and maintained using a PID controller implemented in LabVIEW software. The average heating and cooling rates were +2.5ºC/min and -0.5ºC/min, respectively. The temperature inside the chamber was measured to be accurate within +/- 0.2ºC. For more information on the temperature chamber refer to Appendix D. Each experiment was performed with a new glass mold. For fabricating each mold, 200 µm thick glass coverslips (Thomas Scientific) were fixed to a microscope slide with  45  two part epoxy adhesive in a way that their edges enclosed a 5 mm × 10 mm rectangular region to accommodate the sample solution (Figure 3.2).  Figure 3.2 Mold used for characterizing flow during solvent evaporation. Each mold is made by bonding four 200 µm thick coverslips to a microscope slide base to create an enclosed rectangular region. The scan volume is chosen near the vertical wall of one of the coverslips, far from the corners of the rectangular cavity.  The glass bottom of the molds gave optical access to the polymer solution from the bottom using the inverted stage Nikon TE2000-U microscope.  3.2.3 Experimental Procedure For each experiment, the mold was placed inside the temperature control chamber. After deposition of 20 µl of the 8.5 wt% PVA solution inside the rectangular cavity of the mold, the chamber was sealed and placed on the microscope stage. The confocal scan field was set to cover an area of 636.5 µm × 636.5 µm that included the vertical edge of a coverslip, far from the corners of the rectangular cavity. The scanning depth was adjusted 46  to 300 µm (captured in 10 µm-thick slices). A 20.0X objective lens with 2.1 WD and 0.5 NA was used to collect the emission from the fluorescent particles. Each image pixel in the horizontal plane corresponded to a physical size of 1.066 µm × 1.066 µm. A 570 nm high pass filter was used for fluorescence detection. Setup for each experiment including the deposition of the polymer solution, placement of the mold inside the chamber and sealing of the chamber, adjustment of the chamber temperature, focusing the microscope, and setting the scan parameters took approximately 5 min. After the desired chamber temperature was set and reached, the observation volume was scanned in 77.2 s intervals until complete evaporation of water from the mold. The scanned horizontal slices for each time step were then combined to create a 3D volume using the EZ-C1 software (Nikon) supplied with the confocal system. For each rendered volume, a 2D side view projection (with a physical pixel size of 1.066 µm × 1.066 µm) of the particles was constructed for later PIV analysis (see Figure 3.3).  Figure 3.3 Recording of fluorescent particles in the scan volume and the 2D side projection.  47  This procedure was carried out for the different chamber temperatures 25ºC, 35ºC, and 45ºC. The supplementary video file (see Appendix E.) shows a movie (frame rate: 30 fps) made of the stack of 2D images for the polymer solution at 25ºC. The actual time elapsed between each frame of this video is 77.2 s. For PIV measurements, stacks of the 2D images for each temperature were imported into the PIV software Davis 7.2 (LaVision) and processed for each experiment. For the PIV analysis, multi-pass processing with 8 passes was used with decreasing interrogation window size. The first 4 passes were set to use 128 × 128 pixels window size with 75% overlap. The final 4 passes were adjusted to use 64 × 64 pixels window size with 75% overlap. Since the flow movement is dominant in the horizontal direction, a horizontal elliptical Gaussian weighing function has been selected for the interrogation window in all passes.  48  3.3 Results and discussion 3.3.1 Flow Fields at Different Temperatures from Confocal/PIV Measurements Figure 3.4 shows the velocity field from PIV for all three temperatures at the start of the experiments. The velocity fields indicate a rapid movement of flow in the horizontal direction from right to left, towards the vertical wall of the mold, which indicates a higher evaporation rate near that wall.  µ  µ  µ  Figure 3.4 Vector fields for the initial phase of solvent evaporation at (a) 25ºC, (b) 35ºC, and (c) 45ºC.  49  The magnitude of velocity gradually decreases as the flow approaches the wall. The vector fields also reveal reverse flow at the surface of the fluid. Figure 3.5 provides a closer view of the vector field for the flow at the surface at 25ºC.  n  Figure 3.5 Detail of the surface counter flow at 25 ºC.  This opposite flow movement is seen for all the investigated temperatures at the beginning of the evaporation process. This surface flow is likely to be a Marangoni flow driven by a PVA concentration gradient along the fluid. The higher thermal conductivity of glass, with respect to the polymer solution, causes a higher temperature and therefore a high evaporation rate near the vertical wall of the mold at the start of the evaporation process, where heating occurs through the metal wall of the temperature chamber. This leads to an increased polymer concentration near the wall which is associated with a lower surface tension [28] driving the fluid at the surface away from the wall to a region of higher surface tension. Figure 3.6 shows the streamlines for the flow fields in Figure 3.4 which also feature the surface flow and the recirculation in the film near the surface.  50  Figure 3.6 Streamlines for the initial phase of solvent evaporation at (a) 25ºC, (b) 35ºC, and (c) 45ºC.  From the computed velocity fields in Figure 3.4, the average magnitude of the horizontal component of velocity is calculated and plotted over time for all investigated temperatures (Figure 3.7). For each investigated temperature, the average velocity is initially high until it suddenly decays to near zero velocity. At high temperatures, this velocity is much higher but over a shorter period of time. We will refer to this period as the initial phase of solvent evaporation. The velocity fields in Figure 3.4 therefore correspond to the initial solvent evaporation phase for all temperatures.  51  Figure 3.7 Average of the horizontal component of velocity over time from the PIV data.  After the initial phase of solvent evaporation with high horizontal velocity, Figure 3.7 shows a sudden decrease in velocity. Figure 3.8 shows the corresponding velocity fields and the streamlines during this second phase. Similar to the initial phase, there is a fast flow towards the left, and the flow velocity is higher at higher temperature. The second phase is shorter for high temperatures, whereas the velocity change is larger. In addition, the general flow direction also has a downward component, indicating that the fluid volume shrinks rapidly, and the surface flow is not noticeable for any of the experiments during this phase. At this stage the polymer concentration has most likely increased to a point where the surface tension is virtually independent of concentration. In addition, the shear viscosity of the fluid has increased significantly, suppressing potential recirculating flow.  52  µ  µ  µ  Figure 3.8 Vector fields and streamlines for the second phase of solvent evaporation at (a) & (b) 25ºC, (c) & (d) 35ºC, and (e) & (f) 45ºC.  The final phase of solvent evaporation is defined as the period in which most of the polymer is settled on the surface of the mold and there is very slow movement of polymer near the wall. The vector fields and the streamlines for this phase are shown in Figure 3.9. The flow direction is mainly towards the wall of the mold and downward, resulting from volume shrinkage due to the evaporation of the remaining solvent from the material.  53  µ  µ  µ  Figure 3.9 Vector fields and streamlines for the final phase of solvent evaporation at (a) & (b) 25ºC, (c) & (d) 35ºC, and (e) & (f) 45ºC.  For all phases of the evaporation, the flow has an upward component near the wall of the mold for the experiment at 45ºC, indicating a high evaporation rate at the rim of the mold. For the other temperatures, the flow direction near the wall is rather horizontal or even downward suggesting that diffusive transport of solvent to the surface is sufficiently fast to sustain the lower evaporation rate.  54  3.3.2 Verification of the Confocal/PIV Results In order to validate the general observations from PIV, the volume of the polymer solution in the test volume is evaluated over time from the projected confocal images. Using MATLAB image processing tools, the area corresponding to the polymer solution is calculated for all individual 2D frames. The area is then multiplied by the depth of the observation volume (636.5 µm) to calculate the volume of the polymer solution. Figure 3.10 shows the resulting volume over time.  Figure 3.10 Volume of the polymer solution in the scan volume over time.  The volume decrease shows timescales similar to the ones for the horizontal velocity in Figure 3.7 indicating, as expected, that the evaporation driven flow inside the film is related to the solvent evaporation and therefore, the decrease in volume of the polymer solution. Figure 3.11 shows the evaporation rate of the polymer solvent, calculated as the change in volume per time interval from the data in Figure 3.10.  55  Figure 3.11 Solvent evaporation rate over time.  The evaporation rate in Figure 3.11 peaks for all temperatures just before the decrease in flow velocity in the second evaporation phase from Figure 3.7. This is consistent with the more pronounced vertical velocity components during this second phase shown in Figure 3.8, indicating a decreasing volume as the main mechanism driving the flow. Higher and sharper peaks in evaporation rate can be observed for evaporation at higher temperatures. Finally, the profiles of the polymer after solvent evaporation, as observed in the processed images, show that higher temperatures cause more accumulation of the polymer on the vertical wall and thus create a thicker vertical polymer layer with 78.8, 82.1, and 86.4 µm for 25ºC, 35ºC, and 45ºC, respectively. This again is consistent with the higher horizontal velocities toward the wall at higher temperature shown in Figure 3.7, together with the significant upward flow at this temperature near the wall.  56  3.4 Conclusions In this study, a method was demonstrated to measure flow in thin films of a polymer solution during drying, that allows characterizing this drying process. This characterization technique can be used to control solvent casting parameters to form microstructures with desired shapes. In this method, CLSM is used to trace seed particles in a PVA solution during solvent evaporation at three different temperatures. The scanned 3D volumes acquired during solvent evaporation were transformed into 2D side projections for PIV analysis. The 2D images were processed to achieve velocity vector fields for the flow at the different temperatures. This method allows capturing important details in the flow field during all phases of the evaporation process and therefore can be considered as a useful tool for solvent casting process development. The flow fields indicate fast evaporation-driven flow inside the film towards the wall of the mold as well as a surface counter flow due to the Marangoni effect from a PVA concentration gradient across the film in the initial phase of solvent evaporation. The horizontal flow velocity towards the wall of the mold over time correlates well with the changing total volume of polymer solution over time as well as the solvent evaporation rate. In addition, the higher horizontal velocity as well as the significant upward flow, measured for the flow at higher temperature, can also be related to the thicker polymer layer on the vertical wall after solvent evaporation. This shows that the drying temperature is an important factor in process development in solvent casting.  57  3.5 Acknowledgements The authors acknowledge Thomas Scientific (Swedesboro, NJ, U.S.A.) for supplying the coverslips. Iman Mansoor thanks the British Columbia Innovation Council for funding through the BCIC Innovation Scholarship program.  58  3.6 References [1] U. Siemann “Solvent cast technology–a versatile tool for thin film production,” Progress in Colloid and Polymer Science, vol. 130, pp. 1-14, 2005. [2] G Vozzi, C. J. Flaim, F. Bianchi, A. Ahluwalia, and S. Bhatia “Microfabricated PLGA scaffolds: A comparative study for application to tissue engineering,” Materials Science and Engineering C, vol. 20, pp. 43-47, 2002. [3] J. R. Morgan and M. L. Yarmush, Tissue engineering methods and protocols, Springer, New York, 1999. [4] A. G. Mikos, A. J. Thorsen, L. A. Czerwonka, Y. Bao, R. Langer, D. N. Winslow, and J. P. Vacanti, “Preparation and characterization of poly (L-lactic acid) foams,” Polymer, vol. 35, pp. 1068-1077, 1994. [5] E. A. S. Doherty, R. J. Meagher, M. N. Albarghouthi, and A. E. Barron, “Microchannel wall coatings for protein separations by capillary and chip electrophoresis,” Electrophoresis, vol. 24, pp 34-54, 2003. [6] J. Raczkowska, J. Rysz, A. Budkowski, J. Lekki, M. Lekka, A. Bernasik, K. Kowalski, and P. Czuba, “Surface patterns in solvent-cast polymer blend films analyzed with an integral-geometry approach,” Macromolecules, vol. 36, pp. 2419-2427, 2003. [7] E. Kumacheva, L. Li, M. A. Winnik, D. M. Shinozaki, and P. C. Chengs, “Direct imaging of surface and bulk structures in solvent cast polymer blend films,” Langmuir, vol. 13, pp. 2483-2489, 1997. [8] H. P. Schreiber and M. D. Croucher MD, “Surface characteristics of solvent-cast polymers,” Journal of Applied Polymer Science, vol. 25, pp. 1961-1968, 1980. [9] J. S. Machell, J. Greener, and B. A. Contestable, “Optical properties of solvent-cast polymer films,” Macromolecules, vol. 23, pp. 186-194, 1990. [10] W. M. Prest and D. J. Luca, “The origin of the optical anisotropy of solvent cast polymeric films,” Journal of Applied Physics, vol. 50, pp. 6067-6071, 1979. [11] H. J. Choi, S. G. Kim, Y. H. Hyun, and M.S. Jhon, “Preparation and rheological characteristics of solvent-cast poly (ethylene oxide)/montmorillonite nanocomposites,” Macromolecular Rapid Communications, vol . 22, pp. 320-325, 2001. [12] S. G. Croll, “Internal strain in solvent-cast coatings,” Journal of Coatings Technology, vol. 51, pp. 64-68, 1979. [13] S. G. Croll, “The origin of residual internal stress in solvent-cast thermoplastic coatings,” Journal of Applied Polymer Science, vol. 23, pp. 847-858, 1979. 59  [14] Y. Jung, T. Kajiya, T. Yamaue, and M. Doi, “Film formation kinetics in the drying process of polymer solution enclosed by bank,” Japanese Journal of Applied Physics, vol. 48, pp. 031502-031506, 2009. [15] R. J. Adrian, “Scattering particle characteristics and their effect on pulsed laser measurements of fluid flow: speckle velocimetry vs. particle image velocimetry,” Applied Optics, vol. 23, pp. 1690-1691, 1984. [16] C. E. Willert and M. Gharib “Digital particle image velocimetry,” Experiments in Fluids, vol. 10, pp. 181-193, 1991. [17] M. Raffel, C. Willert, and J. Kompenhans, Particle image velocimetry, a practical guide, Springer, Berlin Heidelberg New York, 1998. [18] J. G. Santiago, S. T. Wereley, C. D. Meinhart, D. J. Beebe, and R. J. Adrian, “A particle image velocimetry system for microfluidics,” Experiments in Fluids, vol. 25, pp. 316-319, 1998. [19] M. Hoffmann , M. Schlüter, and N. Räbiger, “Experimental investigation of liquid– liquid mixing in T-shaped micro-mixers using µ-LIF and µ-PIV,” Chemical Engineering Science, vol. 61, pp. 2968-2976, 2006. [20] K. Shinohara, Y. Sugii, A. Aota, A. Hibara, M. Tokeshi, T. Kitamori, and K. Okamoto, “High-speed micro-PIV measurements of transient flow in microfluidic devices,” Measurement Science and Technology, vol. 15, pp. 1965-1970, 2004. [21] H. Klank, G. Goranovic, J. P. Kutter, H. Gjelstrup, J. Michelsen, and C. H. Westergaard, “PIV measurements in a microfluidic 3D-sheathing structure with threedimensional flow behavior,” Journal of Micromechanics and Microengineering, vol. 12, pp. 862-869, 2002. [22] C. D. Meinhart, S. T. Wereley, and J. G. Santiago, “PIV measurements of a microchannel flow,” Experiments in Fluids, vol. 27, pp. 414-419, 1999. [23] K. Kikuchi and O. Mochizuki, “Micro-PIV measurements in micro-tubes and proboscis of mosquito,” Journal of Fluid Science and Technology, vol. 3, pp. 975-986, 2008. [24] R. Lima, S. Wada, K. I. Tsubota, and T. Yamaguchi, “Confocal micro-PIV measurements of three-dimensional profiles of cell suspension flow in a square microchannel,” Measurement Science and Technology, vol. 17, pp. 797-808, 2006. [25] H. Kinoshita, M. Oshima, S Kaneda, and T. Fujii, “Three-dimensional measurement for internal flow of a micro-droplet using confocal micro-PIV,” Nihon Kikai Gakkai Ryutai Kogaku Bumon Koenkai Koen Ronbunshu (CD-ROM), vol. 84, 2006.  60  [26] J. S. Park, C. K. Choi, and K. D. Kihm “Optically sliced micro-PIV using confocal laser scanning microscopy (CLSM),” Experiments in Fluids, vol. 37, pp. 105-119, 2004. [27] E. Wang, C. Babbey, and K. Dunn “Performance comparison between the highspeed yokogawa spinning disc confocal system and single-point scanning confocal systems,” Journal of Microscopy, vol. 218, pp. 148-159, 2005. [28] A. Bhattacharya and P. Ray, “Studies on surface tension of poly (vinyl alcohol): Effect of concentration, temperature, and addition of chaotropic agents,” Journal of Applied Polymer Science, vol. 93, pp. 122-130, 2004.  61  4 Conclusions and Future Work 4.1 Conclusions In the first part of this study, a novel process for mass fabrication of hollow out-of-plane microneedles from polymeric materials was presented. The presented process uses MEMS photolithography techniques for fabrication of a mold including an array of cone shaped pillars. After deposition of a sacrificial payer, microneedles are formed on the mold through a solvent casting technique, and their tips are sharpened through a plasma etching step. The plasma etching step also provides openings to the sacrificial layer around the pillars in the mold. The microneedle array is then separated from the mold by dissolving the sacrificial layer in its solvent. Using this process, microneedle prototypes were fabricated with different dimensions. These prototypes were made out of SU-8 and PMMA was used as the sacrificial layer. The current fabrication process has yield of 6070%. In some cases the lift-off process (Figure 2.1j) is not successful and results in partial separation of the microneedle array from the device. This may be due to the plasma etching step which may not provide openings to sacrificial layer in some cases. Another possible reason could be the impact of temperature caused by UV exposure (used for curing the SU-8 layer) on solubility of the PMMA layer. In order to measure the robustness of the prototypes, mechanical tests were performed on them to determine the maximum vertical load that the needles can withstand without breaking. The results of these tests were compared with a previous study that investigated the insertion force necessary for skin puncture. For further validation, in vitro insertion 62  trials on potato skin as well as an in vivo injection experiment were carried out using sample devices. The results of these experiments suggest that there is a need for a stronger material choice for microneedle fabrication. The presented process allows fabrication of microneedle arrays with a variety of polymer materials, by selecting the proper polymer pair for the sacrificial layer as well as the microneedle array. Choosing a suitable polymer pair and also further characterization of the process can lead to a fully repeatable process allowing microneedle fabrication in a wide range of dimensions and shapes simply by changing the photolithography mask, adjusting the photolithography parameters for mold fabrication, adjusting the microneedle material solution deposition volume, and controlling process parameters in the solvent casting step. Since the mold is reusable for consecutive fabrication, this process is very cheap and the only fabrication cost corresponds to the cost of the sacrificial polymer solution, the structural microneedle polymer solution, and plasma etching. Due to flexibility, in the choice of material and needle dimensions, and low cost a tuned process would be very suitable for mass production of microneedle arrays. The second part (chapter 3) of this thesis focused on characterization of the microneedle formation during solvent casting. In order to be able to predict and control the final shape of the needles, a new technique based on particle image velocimetry and confocal microscopy was developed. This technique allows characterizing solvent casting parameters by observing the flow during the solvent evaporation and obtaining the corresponding velocity vector field. Using this technique, the impact of three temperature conditions was investigated on needle formation during solvent casting. It was observed that higher temperature results in thicker vertical profile and therefore it was concluded 63  that the final shape of a polymer profile formed by the solvent casting process can be controlled by changing the temperature. This technique is applicable for characterization of other solvent casting process parameters such as humidity, initial solution concentration, and solvent volatility, which can ultimately lead to designing sharper microneedles.  4.2 Recommendations for Future Work The current microneedle fabrication process needs to be further optimized in order to achieve a fully reliable and repeatable process. A possible approach for verifying the potential impact of plasma etching (Figure 2.1i) on the lift-off failure would be to add two different fluorescent labels to both the sacrificial polymer solution as well as the microneedle polymer solution; after forming the microneedle arrays on the mold with the fluorescently labeled solutions and performing plasma etching on them, they can be inspected by a confocal microscope. This technique helps to see the effect of plasma etching on the structural polymer layer and also see whether the plasma etching have provided openings to the sacrificial layer. The injection test indicated that the fabricated microneedle devices were not effective to their full potential; partial delivery of the drug in the experiment suggests that some needles have failed during their application on the skin. In order to achieve stronger needles, further research has to be performed to select a stronger material for needle fabrication. Potential choices include using Polyimide as the structural material and PVA as the sacrificial layer. Polyimide has a higher Young’s modulus than SU-8 and is a common material for protective coatings [1]. Some polyimide microneedle prototypes 64  have been previously fabricated using the process presented here, but no tests have been performed on them to determine their strength and validate their skin puncture ability due to time constraints. Additionally, using polymer composite blends such as Polyimide/Nanoclay [2-4] or SU-8/Carbon nanotube [5, 6] blends may improve device strength. For using new polymers in the fabrication process, some experiments must be performed to optimize PECVD parameters as well as solvent casting drying conditions for the new materials. Other possible solutions for improving the needles’ strength could be coating the SU-8 microneedle arrays with materials with high strength such as metals. Additional in vivo and in vitro trials should be performed on any new prototypes for verification of their strength. A technique was demonstrated in chapter 3 for characterization of solvent casting process parameters. The same technique should be used to investigate the effect of humidity, initial solution concentration, and solvent volatility on needle formation during solvent casting. Additional important details in polymer film flow may be discovered when investigating these factors, which may help in designing and fabricating sharper and stronger devices.  65  4.3 References [1] M. K. Ghosh and K.L. Mittal, Polyimides: Fundamentals & Applications, Marcel Dekker, New York, 1996. [2] H. A. Patel, R. S. Somani, H. C. Bajaj, and R. V. Jasra, “Nanoclays for polymer nanocomposites, paints, inks, greases and cosmetics formulations, drug delivery vehicle and waste water treatment,” Bulletin of Materials Science, vol. 29, pp. 133145, 2006. [3] T. Agag, T. Koga, and T. Takeichi, “Studies on thermal and mechanical properties of polyimide–clay nanocomposites,” Polymer, vol. 42, pp. 3399-3408, 2001. [4] A. Perica-Tripalo and D.W. Radford, “Montmorillonite nanoclays as reinforcements in thermoplastic polyimide matrix,” 36th International SAMPE Technical Conference, 2004. [5] X. Xu, M. M. Thwe, C. Shearwood, and K. Liao, “Mechanical properties and interfacial characteristics of carbon-nanotube-reinforced epoxy thin films,” Applied Physics Letters, vol. 18, pp. 2833-2835, 2002. [6] H.C. Chiamori, J.W. Brown, E.V. Adhiprakash, E.T. Hantsoo, J.B. Straalsund, N.A. Melosh, and B.L. Pruitt, “Suspension of nanoparticles in SU-8: Processing and characterization of nanocomposite polymers,” Microelectronics Journal, vol. 39, pp. 228-236, 2008.  66  Appendices A. Mold  Fabrication  and  Photolithography  Parameters For fabricating the molds used in microneedle fabrication, MEMS standard photolithography was used. This section describes the photolithography process and the specific parameters used for fabrication of 250 µm thick and 400 µm thick molds. The fabrication process was performed under fume hood in a class 10 cleanroom. SU-8 2150 was chosen as the material for fabrication of mold. SU-8 is a negative photoresist used for fabrication of thick polymer MEMS structures. The first stage of the photolithography process starts with cleaning the base Pyrex wafer. Both sides of the wafer are washed with acetone for 2 min followed by isopropanol for another 2 min. The wafer is then rinsed with DI water and dried with high pressure N2 gas. The wafer is then placed on a hot plate, set at 120°C, for 5 min to remove all the water molecules from its surface; this improves the adhesion of the photoresist to the base substrate. After cleaning and drying the wafer, the SU-8 is spin coated on the wafer using a spinner. For doing so, the wafer is first placed on a spinner with an appropriate chuck, and then fixed in place using vacuum provided to the spinner. SU-8 2150 is then poured on the wafer so that about 50% of the wafer is covered. The spinner is then set to rotate and its rotation speed depends on the desired thickness of the photoresist in the final mold structure. After spinning, the wafer is placed on hotplate for soft baking. In order to assure a uniformly patterned structure at the end, the hotplate must be adjusted to 67  be perfectly flat and horizontal, because otherwise, the photoresist will flow to one side of the wafer during soft baking which will lead to thicker structures on some areas of the mold. The next step after soft baking is UV exposure through a photolithography mask. Appendix B shows the used mask. Since SU-8 is a negative photoresist, the areas which are exposed to UV are cross-linked and cured, but the unexposed areas will later be removed. For fabrication of cone-shaped pillars, the exposure has to be performed from backside of the wafer. For this purpose, the backside is gently cleaned with acetone and a cleanroom cloth, and then the wafer is flipped and placed on the mask aligner with the photoresist layer facing down. The mask is then positioned on the backside surface and a transparent glass plate is placed on the mask to minimize the gap between the mask and the wafer. The exposure dose varies depending on the thickness of the photoresist. After exposure, the next step is post exposure baking. The wafers are placed on hotplate for about 2 hours to remove any excess solvent from the photoresist. After post exposure bake, the final step is photoresist development. For this purpose, the wafer is placed in a SU-8 developer bath. The developer is an organic solvent that dissolves the unexposed SU-8. After development, the wafer is rinsed with some clean developer for about 10 s followed by isopropanol for about 30 s. Table A.1 summarizes the temperatures, spinner speeds, and exposure doses used for fabrication of 250 µm thick and 400 µm thick molds.  68  Table A.1 Recipes used for fabrication of 250 µm and 400 µm thick molds.  Photolithography 250 µm Thick Mold  400 µm Thick Mold  2 min wash with acetone  2 min wash with acetone  2 min wash with  2 min wash with  isopropanol  isopropanol  Hotplate at 120°C for 5  Hotplate at 120°C for 5  min  min  500 rpm ramped at  500 rpm ramped at 110  110 rpm/s  rpm/s  2000 rpm ramped at  1300 rpm ramped at  330 rpm/s  330 rpm/s  10 min at 65°C  10 min at 65°C  100 min at 95°C  120 min at 95°C  Parameter  Wafer preparation  Spinning  Soft bake  6 min with 6.9  Exposure  mW cm 2  exposure power  11 min and 40 s with 6.9  mW exposure power cm 2  Done in two 3 min steps  Done in 3 min steps with  with 10 s rest in between  10 s rest in between steps  69  Photolithography 250 µm Thick Mold  400 µm Thick Mold  Post exposure  5 min at 65°C  5 min at 65°C  bake  30 min at 95°C  30 min at 95°C  Photoresist  35 min in SU-8 2000  50 min in SU-8 2000  Development  developer  developer  Fresh SU-8 2000  Fresh SU-8 2000  developer for 10 s  developer for 10 s  Isopropanol for 30 s  Isopropanol for 30 s  Parameter  Final Rinse  70  B. Photolithography Mask The following figure shows the darkfield mask used for fabrication of microneedle molds. The mask contains many arrays of circular dots enclosed by transparent square or circular regions. The + sign marks between regions are used as guidelines to cut wafer in 1 cm × 1 cm pieces after fabrication. The mask was designed using CleWin Layout Editor v. 4.0.2 and printed by CAD/Art Services (Bandon, OR).  71  Figure A.1 The mask used for fabrication of microneedle arrays. The mask contains arrays of 40 µm circular dots with 300 µm spacing.  72  C. More Needle Images The microneedle fabrication technique presented here can be used to create microneedle with different lengths, channel diameters, and tip shapes. The following figure shows some of the microneedles made with different fabrication parameters.  a)  b)  c)  d)  Figure A.2 Microneedle arrays fabricated using the technique shown in chapter 2. Different shapes of microneedles are achieved by varying some of the process parameters (a) an array of an array of microneedles 110 µm long fabricated using the recipe presented in chapter 2 (b) an array of microneedles 200 µm long fabricated with high solvent casting temperature (120°C) (c) an array of microneedles made with a weaker plasma etching step (150 s, power 100W, 90% O2 and 10% CF4) and d) microneedles fabricated with a stronger plasma etching step(400 s, 200W power, 90% O2 and 10% CF4).  73  Figure A.3 Some examples of microneedles (a) a microneedle created using the process shown in chapter 2 with a stronger plasma etching step (400 s, 200W power, 90% O2 and 10% CF4) (b) and (c) less sharper microneedles fabricated using a similar process presented in chapter 2; tip openings in these devices are created by using fine sanding papers instead of plasma etching.  74  D. Temperature Chamber The temperature chamber is custom made for use in solvent casting flow characterization experiments (Chapter 3); and its task is to maintain a certain temperature, set by user, during the experiments. It was designed and built by undergraduate students (David Hung and Michael Young) as a part of their undergraduate course work. The following Figure demonstrates the working principle of temperature control system.  Figure A.4 Temperature chamber working principle.  The temperature chamber system hardware mainly consists of an aluminum chamber, an aluminum heatsink, a cooling fan, a custom made power supply, a peltier device, an RTD sensor, a DAQ, RTD sensor circuit, and an H-bridge circuit. The chamber is designed to fit on inverted Nikon microscopes, as shown in Figure A.5.  75  Cooling Fan Heatsink  Temperature Chamber  Figure A.5 Temeparture chamber installed on an inverted Nikon microscope.  The temperature chamber uses the 100W peltier device for heating and cooling the chamber, and the RTD sensor for temperature feedback. The average heating and cooling rates of the system were +2.5ºC/min and -0.5ºC/min, respectively. The temperature inside the chamber was measured to be accurate within +/- 0.2ºC for temperature range of 17 to 50°C.  76  E. Polymer Solution Flow during Solvent Casting at 25°C The following figures show some of the slides from the supplementary video file supplied with the second manuscript. These slides correspond to polymer solution flow during solvent evaporation at 25ºC; the video was created by piling 2D side projection images of the 3D volumes scanned in 77.2 s intervals. The time elapsed between the slides shown in the Figure A.6 is 231.6 s.  77  a)  b)  c)  d)  e)  f)  g)  h)  i)  j)  k)  l)  Figure A.6 continued on next page.  78  m)  n)  o)  p)  q)  r)  s)  t)  u)  v)  w)  x)  Continue of Figure A.6 Slides from the supplementary video file submitted with the manuscript used in chapter 3. The slides correspond to polymer solution flow during solvent evaporation at 25°C. The time elapsed between each consecutive slide pair is 231.6 s.  79  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0067796/manifest

Comment

Related Items