The Pennsylvania State University
The Graduate School
College of Engineering
NOVEL TEMPLATE-LESS SYNTHESIS OF POLYCYANOACRYLATE
NANOFIBERS
A Thesis in
Chemical Engineering
by
Pratik Mankidy
2007 Pratik Mankidy
Submitted in Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
December 2007
The thesis of Pratik Mankidy was reviewed and approved* by the following:
Henry C. Foley Professor of Chemical Engineering Thesis Advisor Chair of Committee
Carlo G. Pantano Distinguished Professor of Materials Science and Engineering
Janna K. Maranas Associate Professor of Chemical Engineering
Andrew L. Zydney Walter L. Robb Chair Professor of Chemical Engineering Head of the Department of Chemical Engineering
*Signatures are on file in the Graduate School
iii ABSTRACT
Polymer nanofibers are 1D nanostructures that are gathering significant interest because of their potential applications in a wide variety of fields such as biomedicine, separation, electronics and sensor materials. They are also fascinating structures because in some cases they posses unique properties arising from the preferential arrangement of the polymer chains parallel to the fibers axes. Challenges that synthesis techniques for polymer nanofibers face are scalability and control during fabrication. An alternate to current common approaches such as electrospinning and templated-synthesis of polymer nanofibers, is template-less synthesis of polymer nanofibers during polymerization. This approach offers the advantage of bottom-up fabrication and the potential for large scale synthesis. Herein results from an investigation using this approach for poly (ethyl 2- cyanoacrylate) [PECA] nanofiber formation are reported.
The template-less growth of PECA nanofibers is first demonstrated using the initiators present in human fingerprint residue via vapor phase polymerization of the monomer under conditions of high humidity. Studies showed that anionic initiators like the Cl - ion present in fingerprint residues is responsible for fiber formation whereas other
anions such as OH - results in the formation of a polymer film under the same conditions of polymerization. Insights into individual initiator effects reveal a classification of the initiators to explain the basis for the morphology of the polymer obtained (1D fiber or 2D film). The classification based on the Hard Soft Acid Base principle implies that for faster initiation (achieved by harder anions) a polymer film is obtained versus polymer nanofibers that are obtained for slower initiation (achieved by softer anions). This
iv suggests that for faster initiation two dimensional growth of the polymer occurs versus one dimensional growth for slower initiation rates leading to fiber formation.
This technique is also extended to demonstrate growth of PECA nanofibers on silane-modified glass slides with varying fiber densities depending upon ECA (monomer) wettability of the glass surface. These results showcase the ability of control over placement of the fibers using this template-less synthesis approach.
Finally two examples of exercising control during synthesis of polymer nanofibers are provided. In the first case by controlling conditions during polymerization, termination steps are varied to create polymer nanofibers with different molecular weights. In the second case, random copolymer nanofibers of ethyl 2-cyanoacrylate and methyl 2-cyanoacrylate are formed by introducing both monomers simultaneously during nanofiber synthesis.
v TABLE OF CONTENTS
LIST OF FIGURES ...... ix
LIST OF TABLES...... xiv
ACKNOWLEDGEMENTS...... xv
Chapter 1 Introduction and Background for Polymer Nanofibers...... 1
1.1 Interest in 1D Nanostructures ...... 1 1.2 Why Polymer Nanofibers? ...... 3 1.2.1 Application of Polymer Nanofibers...... 5 1.2.2 Unique properties of Polymer Nanofibers...... 8 1.3 Synthesis techniques for Polymer Nanofibers...... 9 1.3.1 Electrospinning of Polymer Nanofibers ...... 9 1.3.1.1 Details of electrospinning process...... 10 1.3.1.2 Capabilities of electrospinning process...... 11 1.3.2 Templated synthesis of Polymer Nanofibers...... 13 1.3.2.1 Mesoporous channel templates ...... 14 1.3.2.2 Porous membranes as templates...... 15 1.3.3 Template-less techniques for creating Polymer Nanofibers...... 17 1.4 Motivation and Goals of this work ...... 19 1.5 Background information on Ethyl 2-cyanoacrylate...... 20 1.5.1 Polymerization of ECA ...... 21 1.5.2 Uses of cyanoacrylates ...... 23 1.6 Organization of the Thesis...... 25 1.7 References...... 26
Chapter 2 Experimental Section; Methods and Materials ...... 29
2.1 Techniques for Polymer Synthesis ...... 29 2.1.1 Batch Set-up for Polymerization ...... 29 2.1.2 Semi-continuous Flow Set-up for Polymerization ...... 32 2.2 Techniques for Applying Initiators on Substrates ...... 34 2.2.1 Spin-coating initiator solutions...... 35 2.2.2 Silanation...... 35 2.3 Characterization Techniques ...... 36 2.3.1 Scanning Electron Microscopy (SEM)...... 36 2.3.2 Infra-red Spectroscopy (IR)...... 37 2.3.3 Gel Permeation Chromatography (GPC)...... 38 2.3.4 X-ray Photoelectron Spectroscopy (XPS)...... 39 2.3.5 Atomic Force Microscopy (AFM)...... 39 2.4 Materials ...... 40
vi 2.5 References...... 41
Chapter 3 Template-less Growth of Poly(ethyl 2-cyanoacrylate) Nanofibers by Initiators Present in Fingerprints ...... 42
3.1 Introduction...... 42 3.2 Experimental Section...... 43 3.2.1 Materials...... 43 3.2.2 Methods ...... 44 3.3 Results...... 44 3.3.1 Fingerprint fuming...... 44 3.3.2 Investigation of fingerprint residue components for PECA nanofiber formation...... 48 3.4 Discussion...... 54 3.4.1 Formation of nanofibers ...... 54 3.4.2 Initiators for nanofiber formation...... 56 3.5 Conclusions...... 57 3.6 References...... 57
Chapter 4 Influence of Initiators on the Growth of Poly (ethyl 2-cyanoacrylate) Nanofibers...... 59
4.1 Introduction...... 59 4.2 Experimental Section...... 60 4.2.1 Materials...... 60 4.2.2 Methods ...... 60 4.3 Results and Discussion ...... 61 4.3.1 Classification of Initiators ...... 62 4.3.2 IR investigation of PECA film and nanofibers...... 73 4.3.3 Molecular weight estimations of PECA film and nanofiber ...... 79 4.3.4 Mechanism of formation of different polymer morphologies during vapor phase polymerization ...... 82 4.4 Conclusions...... 84 4.5 References...... 85
Chapter 5 Polymerization of Ethyl 2-Cyanoacrylate Nanofibers on Glass Substrates...... 87
5.1 Introduction...... 87 5.2 Experimental Section...... 89 5.2.1 Materials...... 89 5.2.2 Methods ...... 90 5.3 Results...... 92 5.3.1 Long-time fuming on commercial glass slides...... 92 5.3.2 Variable humidity fuming on Superamine glass ...... 95
vii 5.3.3 Water condensation imaging on Superamine glass ...... 100 5.3.4 AFM imaging on commercial glass slides ...... 102 5.3.5 Elemental analysis by XPS of commercial glass slides ...... 106 5.3.6 Long-time fuming on (Lab-prepared) silane modified glass...... 107 5.4 Discussion...... 111 5.4.1 Polymerization on commercial glass slides...... 111 5.4.2 Polymerization on lab-prepared silane modified glass surfaces...... 116 5.5 Conclusions...... 117 5.6 References...... 118
Chapter 6 Controlling Molecular weight of PECA Nanofibers during Polymerization...... 119
6.1 Introduction...... 119 6.2 Experimental...... 122 6.2.1 Materials...... 122 6.2.2 Methods ...... 123 6.3 Results...... 125 6.3.1 SEM investigation of 2-stage PECA nanofibers ...... 125 6.3.2 GPC estimates of Molecular weight of PECA nanofibers ...... 129 6.3.3 IR analysis of different molecular weight PECA nanofibers ...... 132 6.4 Discussion...... 133 6.4.1 Diameter of 2-stage PECA Nanofibers ...... 133 6.4.2 Controlling Molecular weight of PECA nanofibers...... 134 6.4.3 Interpretation of IR spectra...... 137 6.5 Conclusions...... 138 6.6 References...... 139
Chapter 7 Synthesis of Poly (Methyl 2-cyanoacrylate) [PMCA] Nanofibers and Poly (Ethyl 2-cyanoacrylate-co -Methyl 2-cyanoacrylate) [P(ECA-MCA)] Nanofibers...... 140
7.1 Introduction...... 140 7.2 Experimental Section...... 143 7.2.1 Materials...... 143 7.2.2 Methods ...... 144 7.3 Results...... 146 7.3.1 Fuming of MCA vapor in Batch set-up...... 146 7.3.2 Copolymers of ECA and MCA by vapor phase polymerization (Batch Fuming) and Liquid phase polymerization...... 150 7.3.3 Copolymers of ECA and MCA by vapor phase polymerization (Flow setup Fuming) ...... 154 7.3.4 Molecular weight estimations of P(ECA-MCA) copolymer samples by GPC ...... 156
viii 7.4 Discussion...... 158 7.4.1 Synthesis of poly (methyl cyanoacrylate) [PMCA] nanofibers ...... 158 7.4.2 Copolymerization of ECA and MCA monomers ...... 159 7.5 Conclusions...... 161 7.6 References...... 162
Chapter 8 Conclusions and Future directions ...... 163
8.1 Conclusions from this study ...... 163 8.2 Future directions ...... 165
Appendix A Supplemental Information for Chapter 3 ...... 168
Appendix B Supplemental Information for Chapter 5...... 170
ix LIST OF FIGURES
Figure 1.1 : Routes to 1D Nanostructure synthesis; (A) Intrinsic growth; (B) VLS growth; (C) Templated growth; (D) Preferential growth; (E) Self assembly; (F) Size reduction [Adapted from Xia et al. 1]...... 3
Figure 1.2 : Number of publications per year containing the keywords ‘polymer nanofibers’. Results obtained using SciFinder Scholar TM . *2007 results only till June...... 4
Figure 1.3 : Potential applications for Polymer Nanofibers. Adapted from reviews by Huang et al. 5 and Li and Xia 6 ...... 4
Figure 1.4 : Schematic illustration of the electrospinning process. SEM image of PLGA nanofibers created using this technique 11 ...... 11
Figure 1.5 : Schematic illustration of templated synthesis of polymer nanorods...... 13
Figure 1.6 : SEM image of commercial alumina membrane and polypyrrole nanorods 42 ...... 16
Figure 1.7 : Examples of template-less nanofiber formation (a) Intrinsic polyaniline nanofibers 46 ; (b) Electrochemically deposited polyaniline fibers 50 ; (c) Polymer whiskers, helices etc. during heterogeneous polymerization 51 ...... 18
Figure 1.8 : General form of α-alkyl cyanoacrylates...... 21
Figure 1.9 : Initiation & propagation steps of ECA polymerization ;Nu - represents a nucleophile...... 21
Figure 1.10 : Termination by acidic species (Eq.1) and chain transfer by hydride elimination (Eq.2)...... 22
Figure 1.11 : Degradation pathway for polyalkylcyanoacrylate in vivo 65 ...... 24
Figure 2.1 : Batch Set-up for ECA polymerization (cyanoacrylate fuming)...... 30
Figure 2.2 : Chart for relative humidity in equilibrium with different concentrations of sulfuric acid solution...... 31
Figure 2.3 : Semi-continuous flow set-up for ECA polymerization...... 32
Figure 2.4 : Photograph depicting white polymer residue on glass slides in tubular chamber for flow setup...... 34
x Figure 3.1 : Latent fingerprint observable after ECA fuming ...... 45
Figure 3.2 : SEM pictures of nanofibers of PECA grown on fingerprint ridges at room temp. & relative humidity >95% over a period of 12h (a) Low magnification view (b) Close-up view of the ridge pattern. (c) Close-up view of the nanofibers (d) Magnified view of a single fiber...... 46
Figure 3.3 : Snapshot of initial polymer fiber (15 min exposure to monomer and high humidity) growth on fingerprint at room temp. & relative humidity >95% (a) Low magnification view (b) Close-up view of the same (inset showing the top view of fiber)...... 47
Figure 3.4 : Composition of fingerprint residue (a) Eccrine sweat & (b) Sebum constituents [Adapted from Scruton et. al 11 ]...... 48
Figure 3.5 : PECA nanofibers deposited on substrates spin coated with 0.1M (a) NaCl & (b) KCl ...... 50
Figure 3.6 : PECA nanofibers deposited on substrates spin coated with (a) Palmitic & (b) Stearic acid...... 51
Figure 3.7 : PECA nanofiber deposited on substrates spin coated with 1M NaCl...... 53
Figure 3.8 : PECA nanofibers initiated from synthetic mixtures of (a) Squalene and NaCl and (b) Linoleic acid and NaCl ...... 53
Figure 3.9 : PECA involuted film obtained via initiation by ammonium hydroxide ...54
Figure 4.1 : SEM images of PECA morphology obtained from vapor phase polymerization on substrates spin coated with (a) 0.2M Na 2HPO 4 and (b) 0.2M Na 3PO 4. Scale bars are 1 µm...... 63
Figure 4.2 : SEM images of different polymer morphologies obtained by vapor phase polymerization on substrates spin coated with different anionic initiators at the same concentration of 0.2M. All scale bars are 1 µm...... 64
Figure 4.3 : SEM images of polymer morphologies observed for polymerization using (a) 0.2M KOH and (b) 0.2M KCl as initiators. Scale bars shown are 1µm ...... 66
Figure 4.4 : SEM images of spin coated (a) 0.2M NaOH and (b) 0.2M NaCl solutions after drying on Si substrates prior to polymerization...... 67
Figure 4.5 : Complete IR spectra obtained for (a) cured ECA film; (b) Vapor phase polymerized PECA film and (c) PECA nanofibers ...... 74
xi Figure 4.6 : IR spectra for different PECA samples showing Region I: 3300-2700 cm -1 ...... 75
Figure 4.7 : IR spectra for different PECA samples showing Region II: 2300-2200 cm -1 ...... 75
Figure 4.8 : IR spectra for different PECA samples showing Region III: 2000-1500 cm -1 ...... 76
Figure 4.9 : Probable chain transfer route during vapor phase ECA polymerization...78
Figure 4.10 : PECA samples used for molecular weight measurements initiated by (a) 1M NaCl and (b) 1M NaOH ...... 80
Figure 4.11 : GPC traces of (a) PECA nanofibers made via vapor phase polymerization using 1M NaCl; (b) PECA film made via vapor phase polymerization using 1M NaOH ...... 81
Figure 5.1 : SEM Image results of long-time fuming on (a)Superamine (b)Superfrost Plus (c) Corning GAPS II (d) Nexterion A Schott (e) Superclean (f) Superclean*. Scale bars = 1 µm. Inset Scale bars = 100nm...... 94
Figure 5.2 : PECA polymer growth on Superamine glass substrates at 2 hours of polymerization under different relative humidities ...... 96
Figure 5.3 : SEM images taken at a tilt view of 45° of the 68%RH polymerization sample on Superamine glass (a) 20K magnification (b) 40K magnification...... 97
Figure 5.4 : Polymer bud deposition on Superamine glass substrates under 48%RH at various time periods of fuming. Scale bar =100nm...... 98
Figure 5.5 : Polymer nanofiber growth on Superamine glass substrates under 68%RH at various time periods of fuming. Scale bar =100nm...... 100
Figure 5.6 : ESEM images of water condensation on Superamine glass substrate at >95%RHScale bar =5 µm...... 101
Figure 5.7 : AFM height images of (a) as-received Superamine glass & (b) Superamine glass after 2 hours of polymerization under 48%RH ...... 103
Figure 5.8 : AFM height images of (a) as-received Schott glass & (b) Schott glass after 10 hours of polymerization under 48%RH...... 103
Figure 5.9 : AFM height images of (a) as-received Corning glass & (b) Corning glass after 5 hours of polymerization under 48%RH...... 104
xii Figure 5.10 : AFM height image (a) on as received Superclean substrates & (b) SEM image of at the same magnification of short polymer fiber growth on the same substrate...... 105
Figure 5.11 :(a) AFM height image on Superclean* and (b) SEM image at the same magnification of polymer deposited on surface after 10hours of polymerization ...... 106
Figure 5.12 : SEM Image results of long-time fuming on (a)APS (b)AAS (c)DETA (d)PTS (e)MTMS & (f)HDF all on Superclean*. Scale bars = 1 µm. Inset Scale bars = 100nm...... 109
Figure 6.1 : Protocol for 2-Stage polymerization in Flow set-up ...... 123
Figure 6.2 : SEM images of PECA nanofibers after 1 st stage during 2-stage fuming at (a) 10K magnification & (b) 40K magnification ...... 126
Figure 6.3 : PECA nanofibers grown under (a)&(b) 20%RH; (c)&(d) 25%RH; (e)&(f) 30%RH during the 2 nd stage of the fuming process...... 127
Figure 6.4 : PECA nanofibers grown under (a)&(b) 35%RH; (c)&(d) 40%RH; (e)&(f) 45%RH during the 2 nd stage of the fuming process...... 128
Figure 6.5 : PECA nanofibers after (a) 1 hour & (b) 2 hours of 2 nd stage fuming under 45%RH. Scale bars are 100nm in each image...... 129
Figure 6.6 : GPC traces for PECA nanofibers created under (a) 20% RH and (b) 45% RH during the 2 nd stage fuming process...... 130
Figure 6.7 : Variation of molecular weight of PECA nanofibers with relative humidity in the 2 nd stage of fuming ...... 131
Figure 6.8 : Variation of PDI of PECA nanofibers with relative humidity during 2nd Stage of fuming...... 132
Figure 6.9 : ATR-IR spectrum of different molecular weight PECA nanofibers created under different humidities in the 2 nd stage ...... 133
Figure 6.10 : Plot of 1/ Mn versus [ H2O]/[ M] with a linear fit to the data ...... 136
Figure 7.1 : SEM image of PMCA nanofibers after 12 hours of batch fuming; (a) at 5K magnification and (b) at 40K magnification...... 147
Figure 7.2 : ATR-IR spectra obtained from (a) PMCA nanofibers and (b) from PECA nanofibers ...... 148
xiii Figure 7.3 : IR Spectral region between 3100-2800cm -1of PMCA & PECA nanofibers ...... 148
Figure 7.4 : IR Spectral region between 1900-1300cm -1of PMCA & PECA nanofibers ...... 149
Figure 7.5 : SEM image of P(ECA-MCA) nanofibers after 12 hours of batch fuming; (a) at 5K magnification and (b) at 40K magnification...... 150
Figure 7.6 : ATR-IR spectra obtained from (a) P(ECA-MCA) nanofibers and (b) from P(ECA-MCA) liquid phase polymerized bulk copolymer...... 151
Figure 7.7 : IR Spectral region between 3200-2800cm -1of PMCA, PECA & P(ECA-MCA) nanofibers ...... 152
Figure 7.8 : IR Spectral region between 1900-1300cm -1of PMCA, PECA & P(ECA-MCA) nanofibers ...... 153
Figure 7.9 : SEM image of P(ECA-MCA) nanofibers after > 5 days of flow setup fuming; (a) at 10K magnification and (b) at 40K magnification...... 154
Figure 7.10 : IR Spectral region between 3200-2800cm -1of P(ECA-MCA) nanofibers made in the batch setup and flow setup ...... 155
Figure 7.11 : IR Spectral region between 1900-1300cm -1of P(ECA-MCA) nanofibers made in the batch setup and flow setup ...... 156
Figure 7.12 : Milky suspensions of insoluble P(ECA-MCA) copolymers in THF ...... 157
Figure 7.13 : GPC traces of P(ECA-MCA) copolymer samples in THF...... 158
Figure A.1 : Control SEM of palmitic acid coated Si wafer...... 168
Figure A.2 : Control SEM of stearic acid coated Si wafer ...... 169
Figure B.1 : SEM image of the plain untreated Superamine slide...... 170
Figure B.2 : Silane molecules used for modifying the Superclean * glass surfaces....171
Figure B.3 : Typical XPS survey scan used to determine elemental composition of the Superamine glass surface...... 172
xiv LIST OF TABLES
Table 1.1 : Brief list of electrospun polymer nanofibers and their applications...... 12
Table 2.1 : Materials used in this study and their purchase source...... 41
Table 4.1 : Water uptake of different initiators for ECA polymerization and the morphology observed by SEM ...... 68
Table 4.2 : Classification of initiators on the basis of their relative softness/hardness ...... 71
Table 5.1 : Surface Elemental Atomic Compositions of Commercial glass slides by XPS...... 107
Table 5.2 : Surface Elemental Atomic Compositions of silane modified glass substrates...... 110
Table 6.1 : Flow settings to achieve different RH conditions in 2 nd stage of fuming process ...... 124
xv ACKNOWLEDGEMENTS
I would like to sincerely thank my thesis advisor, Dr. Henry Foley. I am deeply grateful for his guidance during this entire work. I am also immensely thankful for his unrelenting patience and encouragement with this work especially during extended periods of time when useful results were hard to come by. I have thoroughly enjoyed our discussions and am very glad that I had this opportunity to work with him.
I would like to give a special thanks to Dr. Ramakrishnan Rajagopalan. Ram assisted me greatly during these last five years. His never ending list of ideas to try and impromptu lectures about theoretical topics has contributed immensely to the completion of this work. His strive to always help not just me but other graduate students in the group; in finding solutions to research problems is testament of his commitment to all our projects. I wish him all the very best as he begins a new phase of his career in academia.
I would also like to thank all my thesis committee members. I am grateful to Dr.
Carlo Pantano, especially for the work regarding polymer nanofiber growth on silane modified glass surfaces. He provided me with valuable insights and suggestions each time we met. I am extremely glad to have had those discussions with him. I thank Dr.
Andrew Zydney, Dr. Janna Maranas, and Dr. Seong Kim for serving on my thesis committee. A special note of thanks goes out to the Late Dr. Larry Duda. I feel extremely fortunate to have had the opportunity to interact with him during the first four years. I loved the stories he would tell us in his Transport classes and the enthusiasm which he would share with me when discussing issues about the work in this project. I shall cherish my memories of him.
xvi I am most definitively thankful of all the support I got from the Penn State’s
Materials Characterization Labs. Bob Hengstebeck, Josh Stapleton, Tad Daniels, Vince
Bojan, Maria Klimkiewicz, Jon Cantolina and Mark Angelone have all helped me learn, analyze and understand various characterization tools used in this work. I thank them for all their efforts and patience.
There is a bunch of people that made working here in Fenske very pleasurable and extremely memorable. They are Anna Merritt, Krishna Dronavajjala, Billy-Paul
Holbrook, Chris Burket, Bo Yi, Magdalena Giraldo and Roman Galdamez. Their willingness to help me with my research efforts coupled with their ever-ready attitude for fun activities was the perfect combination I could have asked for during my stay here. I will miss them all and I wish them the very best for all their future endeavors.
Last but certainly not the least I would like to thank my family. My parents have, as always, been the pillars of my success. I have received my best education from them and I hope to continue to learn from them. I would not be able to do much without their support and love. My brother Rishikesh, sister-in-law Veena and niece Amolika have been a great escape for me to go to frequently on weekends during these last five years.
Having them near during that time has been absolutely fantastic and I owe them a lot for my success.
Chapter 1
Introduction and Background for Polymer Nanofibers
An introduction to the terms nanostructures, nanoscience or nanotechnology in this day and age seems gratuitous given by the fact of the large popularity of the four- letter prefix of these terms; nano . The appreciation for this arena of research ranging from the well known visionary lecture of Richard Feynman ( There’s plenty of room at the
bottom , December 1959) to the ever-expanding volume of scientific literature on these
topics and from the large (generous) amounts of funding by government and other
agencies to the banal usage of these terms in present day TV shows and science fiction
movies, delivers testament to their popularity. The content of this chapter involves the
introduction to 1-dimensional nanostructures, specifically polymer nanofibers, their
applications and methods of fabrication. Subsequently, information about the polymer
used in this study, poly (ethyl 2-cyanoacrylate) [PECA] and the motivation for this work
is provided.
1.1 Interest in 1D Nanostructures
By definition 1D nanostructures are structures with one dimension measuring between 1-100nm. Such materials include nanorods, nanotubules, nanowires, nanofibers and nanobelts. The immense interest in this area of research stems from the quest to answering intriguing fundamental scientific questions about their unique properties and
2 also from their potential applications in a wide variety of fields. 1,2 In contrast to other dimensional structures, (0D or 2D), 1D structures offer a better system for studying the effects of nanoconfinement of materials on their properties such as electrical conductivity and mechanical strength. 1D nanostructures also offer the most potential for practical utilization in nanomaterials-based devices or components. 3
In a recent review Xia et al. 1 summarized the commonly used methods to synthesize 1D nanostructures. They constitute primarily of six routes depicted in Fig. 1.1
;(A) Intrinsic 1D growth of a solid; (B) Vapor-liquid-solid growth 4, commonly known as whisker growth wherein the surface tension of a liquid droplet is utilized for 1D crystal growth from a vapor source; (C) Use of a template to direct 1D growth; (D) Kinetic control over growth rates of specific facets of a nucleated seed using surfactants; (E) Self assembly of smaller dimensional structures and (F) Reduction of a larger 1D structure into a 1D nanostructure. These techniques are employed in the synthesis of various inorganic and organic nanostructures. Specific examples of these strategies pertaining to the synthesis of polymer 1D nanostructures are discussed in the following sections.
3
Figure 1.1: Routes to 1D Nanostructure synthesis; (A) Intrinsic growth; (B) VLS growth; (C) Templated growth; (D) Preferential growth; (E) Self assembly; (F) Size reduction [Adapted from Xia et al. 1]
1.2 Why Polymer Nanofibers?
The attention received by polymer nanofibers is evident in the almost exponential
increase in the number of research publications about them in the past 8-10 years. Fig. 1.2
depicts this increase based on search results for the term ‘polymer nanofibers’ using the
SciFinder Scholar TM database search engine. This reason for this significant interest emerges from their unique properties and the variety of potential applications they offer.
Most of these applications are a result of the high surface areas and high porosity associated with networks of polymer nanofibers. A summary of these applications is presented in the chart depicted in Fig. 1.3 .
4
500
400
300
200 Number Publications of 100
0 99 00 01 02 03 04 05 06 7* 19 20 20 20 20 20 20 20 200 Year Figure 1.2: Number of publications per year containing the keywords ‘polymer nanofibers’. Results obtained using SciFinder Scholar TM . *2007 results only till June.
Figure 1.3: Potential applications for Polyme r Nanofibers. Adapted from reviews by Huang et al. 5 and Li and Xia 6
5 1.2.1 Application of Polymer Nanofibers
Nanofiber reinforced composites : The high surface to volume ratio (also called as
aspect ratio) of a nanofiber enables significant translation of the required (and valuable)
properties of the nanofiber material to a composite matrix with only a small concentration
of the nanofiber component. Reneker and Kim 7 have shown that the incorporation of ultra fine poly(benzimidazole) nanofibers in an epoxy matrix and a rubber matrix improved their mechanical properties considerably. A 15wt. % fraction of nanofiber filler in the epoxy matrix caused an increase in the fracture toughness of the composite and an increase in the Young’s modulus and tear strength was observed for the case of the composite with rubber.
Membrane applications : The use of polymer nanofibers as filtration media for
separation purposes arises from the advantageous fact that the filter efficiency increases
with decrease in the diameter of the fibrous filtration media. A thin layer comprised of
polymer nanofibers hence provides good separation capabilities. Gibson et al. 8 investigated the transport properties of elastomeric nanofibrous membranes. The fibrous polymer mats were deposited using a technique called electrospinning (will be discussed in coming sections). They demonstrated high efficiencies of the membranes in trapping airborne particles and low resistance towards water vapor. Such lightweight nanofibrous membranes are ideal for protective clothing applications. Also, commercial filtration products based on polymer nanofibers for the separation of small particles have already been in use for several years for applications such as dust collection and gas-turbine air filtration. 9
6 Biomedical applications : By virtue of the interconnected, high surface area,
porous network of polymer nanofibers, they are an ideal candidate for use as scaffolds in
tissue engineering. The natural extra-cellular matrix (ECM) on which cells attach and
organize is composed of a three-dimensional network of nanometer-sized protein
fibrils. 10 In order to regenerate tissues and organs, a fibrous architecture similar to ECM that facilitates cell adhesion and growth is required. For that purpose, a nanofiber network of a biocompatible and biodegradable polymer is used to mimic the natural
11 scaffold. Li et al. have demonstrated poly ( D, L -lactide-co -glycolide) (PLGA) nanofiber
networks to successfully sustain cell attachment and proliferation. Other examples
include growth of smooth muscle and endothelial cells on poly( L-lactide-co-ε-
caprolactone) nanofibers 12 and the use of poly( ε-caprolactone) nanofibers for bone tissue
engineering. 13
Polymer nanofibers have also been suggested to be good candidates for wound dressing applications. 14 The airborne-entrapping quality of the nanofiber network
facilitates in keeping harmful bacteria out of the wound while still providing a favorable
framework for the formation of skin. Khil et al. 15 demonstrated this potential, using ultra
thin polyurethane nanofibers as wound dressing material. The fibers showed excellent
oxygen permeability, promoted fluid drainage from the wound and even increased the
rate of epithelialization (skin formation).
The use of polymer nanofibers for drug delivery applications evolves from the
principle that the dissolution rate of a drug particle increases with increasing the surface
area of the drug-carrier matrix. 16 Hence the large surface to volume ratio of polymer nanofibers again proves to be instrumental in its potential as a carrier for drug delivery
7 applications. Luu et al. 17 have demonstrated up to 80% of the initially loaded drug component (DNA in this case) from electrospun nanofibers of PLGA and PLA-PEG.
Such a delivery system is required for site specific and controlled gene delivery (DNA delivery), while the DNA is protected until release.
A concise review of these and more applications of polymer nanofibers in biomedicine and biotechnology is available by Venugopal and Ramakrishna. 18
Sensor applications : The high surface to volume ratio of polymer nanofibers also
makes them an obvious choice for highly sensitive sensor materials. Samuelson et al. 19 fabricated sensors for metal ions (Fe 3+ and Hg 2+ ) and 2,4-dinitrotoluene using electrospun
nanofibers of a fluorescent polymer poly (acrylic acid)-poly(pyrene methanol). The fibers
showed 2 to 3 orders of magnitude greater sensitivities compared to continuous films.
This was caused by the higher surface area of the fibers and easier transfer of the analytes
to the active centers within the nanofiber network. Sadek and co-workers 20 also showed increased responsiveness in the conductivity of doped polyaniline nanofibers towards H 2 gas for 30nm diameter fibers as compared to 50nm fibers.
Use in electronic devices : Polymers with electrical properties are also being investigated for their use in nanoscale electronic devices as nanofibers. MacDiarmid et al. 21 have proven that the conductivity of polyaniline/PEO blend nanofibers is dependent
on their diameters. These same fibers have been used to construct FETs by placement of
22 individual fibers on pre-patterned Si/SiO 2 substrates. Other applications of polymer
nanofibers involve use as electrode materials for accommodating electrolytes in high-
performance lithium batteries. 23 Here poly (vinylidene fluoride) nanofibers exhibited
greater ionic transport than polymer films due to their inherent pore structure and
8 porosity and also better wettability by the electrolyte due to the high surface area of the
fibers
Polymer nanofibers as sacrificial templates : Polymer nanofibers can also be used
to template the formation of hollow 1D nanostructures, i.e. nanotubules. This can be
achieved by deposition of the material of interest (other polymers, metals etc.) onto the
polymer nanofibers followed by selective removal (by dissolution or decomposition) of
the nanofiber template. Poly ( p-xylylene) nanotubules were created in this way using fiber templates of PLA polymer which were removed subsequently by thermal degradation. 24 Polymer nanofibers have also been utilized as templates to create nanofluidic channels made of spin-on glass. 25
Another interesting applications of polymer nanofibers is their use as an adhesive layer by mimicking gecko foot-hair. 26 This adhesive property again emerges from the high surface area of the nanofibers causing an increased surface to surface interactions.
1.2.2 Unique properties of Polymer Nanofibers
The confinement of polymer chains in a polymer nanofiber to a diameter of ~
100nm can cause an alignment or orientation at the molecular level. Kageyama et al. 27 reported the formation of polyethylene fibrils with extended chain crystals because of confinement during polymerization of the fibrils in the nanometer sized pores of a mesoporous silica supported catalyst. Similarly, electrospun poly(ferrocenyldimethylsilane) nanofibers exhibited polymer chains extended and aligned along the fiber axis. 28 The reason for such orientation were strong shear forces
9 experienced by the polymer chains during electrospinning and rapid solidification of the
polymer solution that do not allow the chains to re-conform to their equilibrium state.
Also Martin and Liang 29 showed that template-synthesized polyacetylene fibrils had greater conductivities along the fibril axis as compared to bulk polymer samples. They attributed this enhancement to preferential orientation of the polymer chains parallel to the fibril axis.
This orientation of chains in a polymer nanofiber along with their potential
applications makes these fibers fascinating structures for investigation.
1.3 Synthesis techniques for Polymer Nanofibers
Current fabrication routes for polymer nanofibers primarily involve one of the
following three methods; electrospinning of a polymer solution or melt, use of a
nanometer-sized template or template-less development of nanofibers. These methods
can be considered as examples of category F (size reduction), category C (templating)
and category A (intrinsic 1D growth) from Fig. 1.1, respectively. Of these three,
electrospinning is the most widely used technique followed by templating and template-
less techniques.
1.3.1 Electrospinning of Polymer Nanofibers
Electrospinning is a process of creating nanofibers of polymer from a polymer solution or melt using an external electric field. 5,6,30,31 The formation of fibers is a result
10 of thin viscoelastic jets that are ejected out of the solution or melt on application of the electric field. The process is continuous and consequently leads to the deposition of a non-woven mat of nanofibers on a collector surface (electrode). The process was initially known as electrostatic spinning until the early 1990s with very few publications.
Subsequently, its potential for creating nanofibers was more widely recognized and the current nomenclature of electrospinning was adopted. 6 A more detailed description of the process follows.
1.3.1.1 Details of electrospinning process
Fig. 1.4 is an illustration of the electrospinning process to create polymer
nanofibers. It involves placing a polymer solution in a syringe fitted with a needle at the
end of which a pendant drop of the solution is present at all times. The needle is
connected to a high DC voltage (~20KV) power source and is placed some distance away
from a grounded collector electrode. On application of the electric filed the drop
experiences two electrostatic forces; repulsion due to charge accumulation and Columbic
forces because of the external field. At a certain stage these forces overcome the surface
tension forces holding the drop together and a liquid jet ejects from the pendant drop.
This jet stretches, elongates and undergoes a whipping action that causes it to become a
thin long thread. During this process by evaporation of the solvent its diameter is reduced
further to sub micron sizes. Eventually it is attracted onto the grounded collector surface
where it is deposited as a randomly oriented mat of fibers as shown in Fig. 1.4.
11
Figure 1.4: Schematic illustration of the electrospinning process. SEM image of PLGA nanofibers created using this technique 11
Multiple branching and threading can occur from a single drop and liquid jet, which leads to a complex network of fibers.
1.3.1.2 Capabilities of electrospinning process
The majority of the fifty or so polymer nanofibers reported in the literature to have been created by electrospinning involves using a polymer solution versus a polymer melt. The reason for that being, complex operational requirements such as spinning at high temperatures and in some cases the requirement of vacuum for polymer melt electrospinning. Tab. 1.1 is a brief summary of some of the polymer nanofibers made by electrospinning.
12
Table 1.1: Brief list of electrospun polymer nanofibers and their applications
Polymer Solvent Potential application
Nylon6,6 32 Formic acid Protective clothing
Polyurethanes 32 Dimethyl formamide Protective clothing
Polyacrylonitrile 33 Dimethyl formamide Carbon nanofiber
Polylactic acid 34 Dichloromethane Drug delivery
Polyethylene oxide 35 Distilled water Electrical interconnects Polyaniline/PEO Chloroform Conductive fiber blend 36 Polystyrene 37 Tetrahydrofuran (THF) Catalyst, filter Scaffold for tissue PLGA 11 THF: Dimethyl formamide engineering
The morphology and diameter of the spun fibers depends on a number of processing parameters such as type of polymer, surface tension of the solvent, viscosity
(i.e. concentration) of polymer solutions, electrical conductivity, feed rate of polymer solution, distance between needle and collector and strength of electric field.
Some of the drawbacks of this process include lack of sufficient control over the chaotic nature of the spinning process. In order to effectively create nanofibers reproducibly the above mentioned processing parameters must be intimately controlled and monitored. Bead formation in the fibers i.e. the formation of spherical droplets of polymer along the fiber axis is a common occurrence during electrospinning. 38 Also the
placement of the fibers, as they are spun, lacks control. The fibers are deposited randomly
on the collector surface which would suffice to create a nanofiber mat but not suitable for
precise placement of these fibers, for example in an electronic device. Another limitation
13 of this process is that it is essentially a ‘top-down’ approach to making nanostructures. It begins with an already synthesized polymer in bulk and reduces it to thin fibers.
Although successful in making the nanofibers a large amount of energy (high voltage electric field) is essential in re-organizing the polymer chains in the nanofibers. The counter approach for top-down fabrication is bottom-up fabrication, wherein the polymer chains would be organized into nanofibers during their polymerization itself. The following template-based techniques demonstrate this idea.
1.3.2 Templated synthesis of Polymer Nanofibers
The templated synthesis of 1D polymer nanostructures is based on the simple idea of using the physical dimensions of a mold to create structures of desired shapes and sizes. 39 A schematic illustration of the templating techniques is provided in Fig. 1.5. It involves using the pores of the template to shape the polymer into a nanostructure
(nanorod in this case). The polymer is either synthesized inside the pores or is drawn into the pores by virtue of capillary action of a polymer solution.
Figure 1.5: Schematic illustration of templated synthesis of polymer nanorods
14 Once created the polymer nanorods are released by removal (dissolution) of the template to yield an aggregate of nanorods. The templates commonly used for such work are mesoporous silicates (or aluminosilicates) and membranes with discrete nanometer sized pores.
1.3.2.1 Mesoporous channel templates
Wu and Bein 40 have reported fabricating conducting polyaniline filaments in the
pores of the aluminosilicate MCM-41. The method involved adsorption of aniline vapor
in the dehydrated host template followed by reaction with peroxydisulfate. TEM
micrographs of the host before and after reaction proved the presence of encapsulated
polyaniline filaments. Similar results were observed by Li et al.41 while fabricating polystyrene nanofibers and nanotubes in MCM-41 and A1-SBA-15 aluminosilicates.
Such techniques of synthesis result in the nanofibers being lodged in the pore of the host template which had to be dissolved in Hydrofluoric acid to obtain the fibers separately.
Kageyama et al.27 however, reported a novel technique for reactive extrusion
polymerization of nanofibers. In their work they polymerized ethylene with titanocene
catalyst supported on mesoporous silica in conjunction with methylaluminoxane as the
co-catalyst. Scanning electron microscope images of the resultant material show fibers
with diameters between 30nm to 50 nm. The authors suggest that these fibers are a result
of polyethylene chains growing out of the mesopores of the support. In effect, they were
able to create a ‘nano-extruder’ by anchoring the catalyst to the walls of the nanochannels
in the support.
15 1.3.2.2 Porous membranes as templates
Membranes that are commonly used as templates for creating polymer
nanostructures are either track-etch polymeric membranes or porous anodic alumina
membranes. 39 Both these kinds of membranes are available commercially with a wide variety of pore sizes. However, the alumina membranes possess much more orderliness and have higher porosities. Alumina membranes have cylindrical pores of uniform diameter and therefore, on synthesis, monodisperse nanorods of the desired polymer are obtained in these pores.
Conducting polymers such as polypyrrole can be synthesized in the pores by oxidative polymerization of the corresponding monomer. This has been accomplished either electrochemically 42 or with a chemical oxidizing agent. 43 The electrochemical method involves coating one surface of the alumina membrane with a metal film and using it as an anode to electrochemically deposit nanofibers of polymer in the pores.
Fig. 1.6 depicts SEM images of a typical anodic alumina membrane and nanorods of polypyrrole obtained after electrodeposition and dissolution of the membrane. In the chemical method, polymerization is achieved by using the membrane as a separating wall between two compartments; one containing an aqueous solution of the monomer, pyrrole, and the other containing an aqueous solution of the oxidizing agent. The reagents diffuse through the pores of the membrane and react in them to form polypyrrole nanofibers.
16
Figure 1.6: SEM image of commercial alumina membrane and polypyrrole nanorods 42
Similar chemical synthesis of polypyrrole was carried out by Martin and Cai 44 in
alumina membranes. They observed from SEM images of the membrane during synthesis
that the growth of the polymer started along the walls of the pores and eventually the
polymer completely filled the pores.
Martin et al. 45 also carried out radical polymerization in alumina membranes, where the membrane was immersed in an aqueous solution of acrylonitrile along with species which served as initiators for the polymerization reaction. The solution was raised to a temperature of 40°C at which polyacrylonitrile nanotubules were created.
The use of this templated-synthesis route for 1D polymer nanostructures, although
possessing the benefit of being a bottom-up fabrication approach suffers from certain
inherent shortcomings. Firstly it involves a one-time use of the template. The template is
either destroyed or inseparable from the polymer product. Furthermore, the axial
dimension of the polymer nanostructure obtained is limited by the template height (pore
length of the membrane). As these templates are short, the polymer structure created is in
17 fact not a nanofiber but a nanorod. And lastly this process is not continuous; hence it is unsuitable for bulk or large scale synthesis of polymer nanostructures.
1.3.3 Template-less techniques for creating Polymer Nanofibers
Few reports exist in the current literature concerning template-less fabrication of polymer nanofibers compared to the two approaches discussed above. This fabrication involves the initiation and development of the polymer as a 1D nanofiber intrinsically without a template or use of external forces. Such synthesis falls into category A of the strategies illustrated in Fig. 1.1. Reports of polyaniline nanofiber formation via template- less techniques appear most frequently in this category.
Kaner and co-workers 46-49 have studied the intrinsic formation of polyaniline
nanofibers during chemical oxidative polymerization of the monomer. Their protocol for
nanofiber formation entailed interfacial polymerization or rapidly mixed polymerization
reactions systems, where (in both cases) the polymer nanofibers were created in a non-
solvent environment. An SEM image of these polyaniline fibers is given in Fig. 1.7(a). A
factor that governed the morphology of the polymer formed (fibers or aggregate particles)
was the rate of introduction of monomer during polymerization. Their reasoning for
obtaining this 1D growth of the polymer was that for very early stages of polymerization
the polymer chains were fibrillar in nature and the successful suppression of secondary
growth of this propagating polymer results in nanofibers. This implies that although there
is no physical template there is an inherent ‘templating-effect’ associated with initial
stages of polymerization which if controlled correctly would lead to the formation of
18 polymer nanofibers. Fibers with different diameters were also created using different dopants or different solvents.
Figure 1.7: Examples of template-less nanofiber formation (a) Intrinsi c polyaniline nanofibers 46 ; (b) Electrochemically deposited polyaniline fibers 50 ; (c) Polymer whiskers, helices etc. during heterogeneous polymerization 51
Yu et al. 52 also observed nanofibrous morphology of polyaniline during polymerization without a template. The polymer was a polyaniline-sodium alignate complex and the explanation for nanofiber formation was the templating-effect the sodium alginate offers during the early initiation stages of polymerization.
Electrochemical deposition of polyaniline nanofibers has also been demonstrated by Liu et al. 50 via reducing the current density after the initial nucleation stages of
deposition to preferentially grow 1D nanofibers (Fig. 1.7 (b)). The lower current density
slows polymer deposition and hence retards secondary growth of the polymer.
Early accounts of template-less formation of polymer nanofibers were given by
Blais & Manley 53 and Theodore Davidson 51 , when they observed ‘worms’, helices and other 1D whisker-like structures of polymers, (Fig. 1.7 (c)), during heterogeneous
polymerization using Ziegler-Natta catalysts. They suspected this fiber growth to be
19 caused by peculiar physicochemical processes occurring at the catalyst surface (initiation sites).
The common explanation given by all these reports of template-less fiber formation is the existence of inherent ‘templating’ that occurs for 1D morphology during the initial stages of polymerization which when suitably sustained continues to grow into polymer nanofibers.
1.4 Motivation and Goals of this work
It is evident that polymer nanofibers are an important class of nanostructures to study and a significant amount of effort is required to make the potential applications of these nanofibers a reality. The drive or motivation for this work arises from the need to overcome some of the limitations of popular nanofiber synthetic techniques
(electrospinning and templated synthesis) in order to bring polymer nanofibers closer to a viable technology. This can be achieved by using the template-less techniques mentioned in the previous section, i.e. intrinsic formation of nanofibers during polymerization. The advantages over the other methods are as follows;
• True bottom-up approach : As the polymer is created in the form of a nanofiber,
control while synthesis can be exercised.
• No use of templates or external forces : Fewer processing steps and parameters
imply lower cost of production.
20 • Scalability : Polymerization (i.e. fiber formation) can continue as far as monomer
and suitable conditions exist impling process can be scaled for bulk synthesis of
nanofibers.
• Control over placement of nanofibers : In some instances, polymer (nanofibers)
can be made to occur preferentially only on certain areas where initiator and
conditions for fiber formation are favorable. This suggests the benefit of precise
placement of fibers, i.e. ‘in-place production ’ of fibers.
Taking into account these facts, the specific objectives of the work in this thesis are as follows:
1. Investigation of a system of intrinsic growth of polymer nanofibers during
polymerization to understand why and how it occurs.
2. Develop a strategy for bulk or larger scale synthesis of polymer nanofibers using
this technique
3. Demonstrate ability for control during synthesis of nanofibers (benefit of bottom
up approach)
The system chosen for investigation is the vapor phase polymerization of ethyl 2- cyanoacrylate (ECA) monomer onto suitable substrates. Background information of this monomer and its polymerization follows in the next section.
1.5 Background information on Ethyl 2-cyanoacrylate
Ethyl 2-cyanoacrylate (ECA) often termed as cyanoacrylate, is widely known by its popular name when used as an adhesive, Superglue® or Krazyglue®. However,
21 cyanoacrylates were not created with the purpose of using it as glue. The discovery of its adhesive property was serendipitous when Harry Coover working at Eastman Kodak, for the purpose of making plastic gun-sight lenses found that the liquid compound would rapidly stick to anything it came into contact with. 54,55 The general form of these α-alkyl cyanoacrylates is given in Fig. 1.8.
H C N C C H C O OR Figure 1.8: General form of α-alkyl cyanoacrylates
The adhesive property of cyanoacrylates comes from the rapid polymerization that they undergo upon initiation by commonly occurring initiators.
1.5.1 Polymerization of ECA
ECA monomer undergoes rapid anionic polymerization once initiated by a nucleophile. The reaction is exothermic and occurs at ambient temperatures. 54 The
initiation and propagation mechanisms are illustrated in Fig. 1.9.
C N H C N C N H C N H C N Nu- H H n C C Nu C C C C Nu C C C C H C O H C O H C O H C O H C O
OC2H5 OC2H5 OC2H5 OC2H5 OC2H5 ECA monomer n
INITIATION PROPAGATION Figure 1.9: Initiation & propagation steps of ECA polymerization ;Nu - represents a nucleophile
22 The electron withdrawing effects of the electronegative cyano (-CN) and ester (-
COOR) groups cause the double bond in the monomer to be susceptible to nucleophilic attack (initiation). Once initiated the newly formed carbanion ion attacks the polar double bond of another monomer and propagation continuous further in this fashion. Initiation can be caused by a variety of anions and remarkably also by weak bases such as alcohol and water. This anomalous fact that initiation caused by water is what renders
Superglue® its famous attribute of being able to stick any objects and surfaces. The reason is that most surfaces under ambient conditions possess some absorbed moisture which is usually sufficient to initiate polymerization in a film of cyanoacrylate applied on it. Termination 56 and chain transfer 57 steps for ECA polymerization are illustrated in
Fig. 1.10 . Termination (Eq.1) by an acidic species is the most prominent route of ceasing
ECA polymerization by neutralization of the propagating carbanion and chain transfer can occur via hydride ion elimination (Eq.2) resulting in formation of a double bond.
H C N H C N + H+ Nu C C C CH Eq. (1) H C O H C O OC H OC H H C N H C N 2 5 2 5 n Nu C C C C H C O H C O - OC H OC H -H C N C N 2 5 2 5 H H n Nu C C C C Eq. (2) H C O C O
OC2H5 OC2H5 n Figure 1.10 : Termination by acidic species (Eq.1) and chain transfer by hydride elimination (Eq.2)
23 Pepper and co-workers have extensively studied the solution polymerization
kinetics of cyanoacrylates and the effect of different initiators 58-62 and water 63 on the
kinetics. They studied several different simple anions that were capable of initiating
- - - - - polymerization; CH 3CHOO , CN , I , Br and Cl and also several covalent bases that were
active to initiate ECA polymerization; phosphines, pyridine, polyvinyl pyridine and
amines. Their observations about the effects of water on polymerization concluded that
ECA polymerization continues in the presence of water even though for most anionic
polymerizations water acts as a ‘killer’ of propagation. This peculiar characteristic of
ECA polymerization by which it can be initiated by water and even continue propagation
in the presence of water is explainable by the simultaneous presence of the two
electronegative cyano and ester groups which makes the monomer very reactive.
Although the propagating chain is terminated by the H + ion of water (Eq.1 in Fig. 1.10 ), a new monomer unit can immediately be initiated by the OH - ion of water and the propagation of a new polymer chain begins.
1.5.2 Uses of cyanoacrylates
The use of cyanoacrylates as glue is well known as it bonds to almost any surface.
This same adhesive property is useful for their application as surgical glue. 55 Medical glues such as Dermabond® and Traumaseal® are used to bond body tissues and offer a suture-less option of closing wounds. The ability to use cyanoacrylates for clinical applications comes from the fact that the polymer is a biodegradable and biocompatible polymer. 64 The primary pathway for in vivo degradation of the polymer is suggested to
24 occur via the hydrolysis of the ester bond on the alkyl side chain (Fig. 1.11 ) to yield an alkyl alcohol and poly(cyanoacrylic acid) which are both soluble in water and are removed from the body by the kidneys. 65 This reaction is catalyzed by esterases (serum, lysosomes and pancreatic juices) and the time of degradation depends on the alkyl side chain length.
H CN H CN n OH CC CC n R-OH
H C O H C O
OR O n n Polycyanoacrylate Poly(cyanoacrylic acid) Figure 1.11 : Degradation pathway for polyalkylcyanoacrylate in vivo 65
Similar applications for polyalkylcyanoacrylate include their use as a drug delivery carrier in the form of nanoparticles. 64,66-68
Another interesting application of ECA is in the area of forensic science. Here
fumes of the ECA vapor are used to develop latent (not visible) fingerprints at crime
scenes. 69 The residue left by a fingerprint contains various anionic salts and covalent
bases such as ammonia in addition to moisture, which are sufficient for initiating
polymerization of the monomer vapor to cause a white polymer (PECA) to deposit. 70-72
As this polymer deposits only on the areas where the fingerprint residue was present (i.e. ridges of the fingerprint), the initially latent print now becomes visible by the white polymer deposit. This vapor phase polymerization of ECA monomer is known to get accelerated by first exposing the fingerprint residue to moisture and then fumes of ECA. 73
This process is well known in forensic science as cyanoacrylate fuming . This is achieved
25 in an enclosed chamber where the artifact laden with fingerprints is subjected to controlled fuming conditions. This vapor phase polymerization of ECA is the method of choice adopted for the work in this thesis.
1.6 Organization of the Thesis
The following is a brief outline of the work in this thesis.
Chapter 2: Experimental details of all set-ups and procedures employed in this study
Chapter 3: Demonstration of vapor phase polymerization of ECA monomer to form polymer nanofibers by cyanoacrylate fuming of fingerprint residues
Chapter 4: Insights into the mechanism of polymer nanofiber formation by investigating the effect of different anionic initiators on polymer morphology during fuming.
Chapter 5: Extension of this template-less approach of polymer nanofiber formation to
demonstrate nanofiber growth on modified glass substrates.
Chapter 6: Use of a flow setup for polymerization to synthesize different molecular weight polymer nanofibers by controlling the extent of termination reactions during growth.
Chapter 7: Synthesis of a copolymer nanofiber of ethyl 2-cyanoacrylate and methyl 2-
cyanoacrylate monomers using the template-less vapor phase polymerization approach.
Chapter 8: Conclusions from the study in this work and a summary of future directions
to pursue with regard to this research.
26 1.7 References
(1) Xia, Y. N.; Yang, P. D.; Sun, Y. G.; Wu, Y. Y.; Mayers, B.; Gates, B.; Yin, Y. D.; Kim, F.; Yan, Y. Q. Adv Mater 2003 , 15 , 353-389. (2) Hu, J. T.; Odom, T. W.; Lieber, C. M. Accounts Chem Res 1999 , 32 , 435-445. (3) Xia, Y. N.; Yang, P. D. Adv Mater 2003 , 15 , 351-352. (4) Wagner, R. S.; Ellis, W. C. Appl Phys Lett 1964 , 4, 89-91. (5) Huang, Z. M.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S. Compos Sci Technol 2003 , 63 , 2223-2253. (6) Li, D.; Xia, Y. N. Adv Mater 2004 , 16 , 1151-1170. (7) Kim, J. S.; Reneker, D. H. Polym Composite 1999 , 20 , 124-131. (8) Gibson, P.; Schreuder-Gibson, H.; Rivin, D. Colloid Surface A 2001 , 187 , 469- 481. (9) http://www.donaldson.com/en/filtermedia/nanofibers/index.html . (10) Kadler, K. E.; Holmes, D. F.; Trotter, J. A.; Chapman, J. A. Biochem J 1996 , 316 , 1-11. (11) Li, W. J.; Laurencin, C. T.; Caterson, E. J.; Tuan, R. S.; Ko, F. K. J Biomed Mater Res 2002 , 60 , 613-621. (12) Mo, X. M.; Xu, C. Y.; Kotaki, M.; Ramakrishna, S. Biomaterials 2004 , 25 , 1883-1890. (13) Yoshimoto, H.; Shin, Y. M.; Terai, H.; Vacanti, J. P. Biomaterials 2003 , 24 , 2077-2082. (14) Min, B. M.; Lee, G.; Kim, S. H.; Nam, Y. S.; Lee, T. S.; Park, W. H. Biomaterials 2004 , 25 , 1289-1297. (15) Khil, M. S.; Cha, D. I.; Kim, H. Y.; Kim, I. S.; Bhattarai, N. J Biomed Mater Res B 2003 , 67B , 675-679. (16) Jing, Z.; Xu, X. Y.; Chen, X. S.; Liang, Q. Z.; Bian, X. C.; Yang, L. X.; Jing, X. B. J Control Release 2003 , 92 , 227-231. (17) Luu, Y. K.; Kim, K.; Hsiao, B. S.; Chu, B.; Hadjiargyrou, M. J Control Release 2003 , 89 , 341-353. (18) Venugopal, J.; Ramakrishna, S. Appl Biochem Biotech 2005 , 125 , 147-157. (19) Wang, X. Y.; Drew, C.; Lee, S. H.; Senecal, K. J.; Kumar, J.; Sarnuelson, L. A. Nano Lett 2002 , 2, 1273-1275. (20) Sadek, A. Z.; Baker, C. O.; Powell, D. A.; Wlodarski, W.; Kaner, R. B.; Kalantar-zadeh, K. Ieee Sens J 2007 , 7, 213-218. (21) Zhou, Y. X.; Freitag, M.; Hone, J.; Staii, C.; Johnson, A. T.; Pinto, N. J.; MacDiarmid, A. G. Appl Phys Lett 2003 , 83 , 3800-3802. (22) Pinto, N. J.; Johnson, A. T.; MacDiarmid, A. G.; Mueller, C. H.; Theofylaktos, N.; Robinson, D. C.; Miranda, F. A. Appl Phys Lett 2003 , 83 , 4244-4246. (23) Choi, S. W.; Jo, S. M.; Lee, W. S.; Kim, Y. R. Adv Mater 2003 , 15 , 2027-2032. (24) Hou, H. Q.; Jun, Z.; Reuning, A.; Schaper, A.; Wendorff, J. H.; Greiner, A. Macromolecules 2002 , 35 , 2429-2431. (25) Czaplewski, D. A.; Kameoka, J.; Mathers, R.; Coates, G. W.; Craighead, H. G. Appl Phys Lett 2003 , 83 , 4836-4838.
27 (26) Geim, A. K.; Dubonos, S. V.; Grigorieva, I. V.; Novoselov, K. S.; Zhukov, A. A.; Shapoval, S. Y. Nat Mater 2003 , 2, 461-463. (27) Kageyama, K.; Tamazawa, J.; Aida, T. Science 1999 , 285 , 2113-2115. (28) Chen, Z. H.; Foster, M. D.; Zhou, W. S.; Fong, H.; Reneker, D. H.; Resendes, R.; Manners, I. Macromolecules 2001 , 34 , 6156-6158. (29) Liang, W. B.; Martin, C. R. J Am Chem Soc 1990 , 112 , 9666-9668. (30) Reneker, D. H.; Chun, I. Nanotechnology 1996 , 7, 216-223. (31) Subbiah, T.; Bhat, G. S.; Tock, R. W.; Parameswaran, S.; Ramkumar, S. S. J. Appl. Polym. Sci. 2005 , 96 , 557-569. (32) Schreuder-Gibson, H.; Gibson, P.; Senecal, K.; Sennett, M.; Walker, J.; Yeomans, W.; Ziegler, D.; Tsai, P. P. J Adv Mater-Covina 2002 , 34 , 44-55. (33) Wang, Y.; Serrano, S.; Santiago-Aviles, J. J. J Mater Sci Lett 2002 , 21 , 1055- 1057. (34) Caruso, R. A.; Schattka, J. H.; Greiner, A. Adv Mater 2001 , 13 , 1577-1579. (35) Shin, Y. M.; Hohman, M. M.; Brenner, M. P.; Rutledge, G. C. Polymer 2001 , 42 , 9955-9967. (36) Norris, I. D.; Shaker, M. M.; Ko, F. K.; MacDiarmid, A. G. Synthetic Met 2000 , 114 , 109-114. (37) Megelski, S.; Stephens, J. S.; Chase, D. B.; Rabolt, J. F. Macromolecules 2002 , 35 , 8456-8466. (38) Fong, H.; Chun, I.; Reneker, D. H. Polymer 1999 , 40 , 4585-4592. (39) Martin, C. R. Science 1994 , 266 , 1961-1966. (40) Wu, C. G.; Bein, T. Science 1994 , 264 , 1757-1759. (41) Li, B. S.; Zhu, G. S.; Gao, F. F.; Guo, Z.; Liu, C. Z.; Yin, X. J.; Wang, C. L.; Qiu, S. L. Abstr Pap Am Chem S 2002 , 224 , U451-U451. (42) Vandyke, L. S.; Martin, C. R. Langmuir 1990 , 6, 1118-1123. (43) Mativetsky, J. M.; Datars, W. R. Solid State Commun 2002 , 122 , 151-154. (44) Cai, Z. H.; Martin, C. R. J Am Chem Soc 1989 , 111 , 4138-4139. (45) Parthasarathy, R. V.; Phani, K. L. N.; Martin, C. L. R. Adv Mater 1995, 7, 896- 897. (46) Huang, J. X.; Kaner, R. B. Chem Commun 2006 , 367-376. (47) Li, D.; Kaner, R. B. J Am Chem Soc 2006 , 128 , 968-975. (48) Huang, J. X.; Kaner, R. B. J Am Chem Soc 2004 , 126 , 851-855. (49) Huang, J. X.; Kaner, R. B. Angew Chem Int Edit 2004 , 43 , 5817-5821. (50) Liu, J.; Lin, Y. H.; Liang, L.; Voigt, J. A.; Huber, D. L.; Tian, Z. R.; Coker, E.; Mckenzie, B.; Mcdermott, M. J. Chem-Eur J 2003 , 9, 605-611. (51) Davidson, T. Polymer Letters 1970 , 8, 855-859. (52) Yu, Y. J.; Si, Z. H.; Chen, S. J.; Bian, C. Q.; Chen, W.; Xue, G. Langmuir 2006 , 22 , 3899-3905. (53) Blais, P.; Manley, R. S. J. Science 1966 , 153 , 539-541. (54) Coover, H. W.; Dreifus, D. W.; T., O. C. J. Handbook of Adhesives , 3rd ed.; Van Nostrand Reinhold: New York, 1990. (55) http://en.wikipedia.org/wiki/Ethyl_cyanoacrylate . (56) Costa, G.; Cronin, J. P.; Pepper, D. C.; Loonan, C. Eur Polym J 1983 , 19 , 939- 945.
28 (57) Yang, D. B. J Polym Sci Pol Chem 1993 , 31, 199-208. (58) Donnelly, E. F.; Johnston, D. S.; Pepper, D. C.; Dunn, D. J. J Polym Sci Pol Lett 1977 , 15 , 399-405. (59) Pepper, D. C.; Ryan, B. Makromol Chem 1983 , 184 , 383-394. (60) Johnston, D. S.; Pepper, D. C. Macromol Chem Phys 1981 , 182 , 421-435. (61) Johnston, D. S.; Pepper, D. C. Macromol Chem Phys 1981 , 182 , 407-420. (62) Johnston, D. S.; Pepper, D. C. Macromol Chem Phys 1981 , 182 , 393-406. (63) Eromosele, I. C.; Pepper, D. C.; Ryan, B. Makromol Chem 1989 , 190 , 1613- 1622. (64) Vauthier, C.; Dubernet, C.; Fattal, E.; Pinto-Alphandary, H.; Couvreur, P. Adv Drug Deliver Rev 2003 , 55 , 519-548. (65) Lenaerts, V.; Couvreur, P.; Christiaens-Leyh, D.; Joiris, E.; Roland, M.; Rollman, B.; Speiser, P. Biomaterials 1984 , 5, 65-68. (66) Soppimath, K. S.; Aminabhavi, T. M.; Kulkarni, A. R.; Rudzinski, W. E. J Control Release 2001 , 70 , 1-20. (67) Schmidt, W.; Roessling, G. Chem Eng Sci 2006 , 61 , 4973-4981. (68) Damge, C.; Vranckx, H.; Balschmidt, P.; Couvreur, P. J Pharm Sci 1997 , 86 , 1403-1409. (69) Kendall, F. G.; Rehn, B. W. J Forensic Sci 1983 , 28 , 777-780. (70) Lewis, L. A.; Smithwick, R. W.; Devault, G. L.; Bolinger, B.; Lewis, S. A. J Forensic Sci 2001 , 46 , 241-246. (71) Wargacki, S.; Dadmun, M. D.; Lewis, L. Abstr Pap Am Chem S 2005 , 229 , U981-U982. (72) Edwards, H. G. M.; Day, J. S. J Raman Spectrosc 2004 , 35 , 555-560. (73) Lee, H. C.; Gaensslen, R. E. Advances in fingerprint technology ; Elsevier: New York, 1991.
Chapter 2
Experimental Section; Methods and Materials
This section includes the general equipment set up, protocols for experimental procedure, characterization tools and techniques employed and materials utilized in this work. Specific modifications to procedures for specific experiments are discussed in subsequent chapters.
2.1 Techniques for Polymer Synthesis
Poly (ethyl 2-cyanoacrylate) [PECA] was synthesized via vapor phase
polymerization. ECA monomer introduced in the vapor phase polymerizes on a substrate
containing the appropriate initiators for polymerization. Two experimental set-ups were
utilized for these; (i) Batch Set-up and (ii) Semi-continuous Flow set-up.
2.1.1 Batch Set-up for Polymerization
Fig. 2.1 illustrates the enclosed glass chamber used for batch polymerization. The polymerization was carried out under ambient atmospheric conditions and room temperature with only a control of the relative humidity inside the chamber. The substrates used were either glass slides or Si wafers with the initiator for polymerization applied on the surface. A beaker containing an aqueous solution of sulfuric acid of
30 appropriate concentration in the chamber was used to control the humidity of the chamber air.
Figure 2.1: Batch Set-up for ECA polymerization (cyanoacrylate fuming)
The vapor pressure of water in equilibrium over aqueous solutions of sulfuric acid depends on the concentration of sulfuric acid. 1 Hence for an enclosed chamber containing an aqueous solution of sulfuric acid, providing the temperature remains constant, after sufficient time the relative humidity inside the chamber attains reasonable equilibrium.
For example a relative humidity of 97% in the chamber at room temperature (~20°C), is achieved by an 8 weight% aqueous sulfuric acid solution. Based on this equilibrium data a calibration plot of relative humidity versus volume % of sulfuric acid solution was created for use (shown in Fig. 2.2 ). A humidity meter was placed in the chamber such that it could be monitored from outside the chamber. A small fan was also placed in the chamber to provide adequate mixing in the gas phase. The experiment was carried out in two steps-humidification and polymerization (also called cyanoacrylate fuming). For humidification, the substrate (one or more) with the initiator was placed in the chamber along with the humidity solution. Once closed sufficient time was given for humidity in
31 the chamber to attain equilibrium (~10 hours). After this time liquid ECA monomer was introduced in the chamber through a small opening in the top of the chamber into a container placed directly below the opening. ECA has a vapor pressure of ~0.17mm of
Hg at 20°C, which proved to be adequate to initiate polymerization on the substrate. The polymerization was carried out for different times depending on the experiment objective.
For long times of polymerization the monomer in the container inside the chamber was replenished every 12 hours to maintain a constant concentration of monomer in the vapor phase. Once complete the chamber was opened and the polymer deposited on the substrates was characterized by techniques described later in this section.
100
90
80
70
60 Relative% humidity
50
40 5 10 15 20 25 30 35 40
Sulfuric acid volume % Figure 2.2: Chart for relative humidity in equilibrium with diffe rent concentrations of sulfuric acid solution.
32
2.1.2 Semi-continuous Flow Set-up for Polymerization
Fig. 2.3 illustrates the semi-continuous flow set-up used for vapor phase polymerization of ECA monomer onto substrates containing initiator. The substrates were placed in a tubular glass chamber downstream from which was an oil trap to prevent backflow. Upstream from the chamber were three gas flow lines. Line #1 is an Ar purge line, Line #2 is Ar flow saturated with water vapor and Line #3 is an Ar stream saturated
Figure 2.3: Semi-continuous flow set-up for ECA polymerization with ECA monomer. Mass flow controllers (MFCs) in each line are used to control the flowrate through the lines and hence the eventual composition of the environment inside
33 the chamber. Ultra High Purity (UHP) Ar was used and the MFCs were initially calibrated for Ar gas flow before use.
Prior to polymerization the substrates in the chamber were subjected to only flow from line #2 to humidify the chamber. This was done by maintaining a flow rate of
~50sccm in line #2 for sufficiently long period. The polymerization was carried out by a
2-stage procedure. In the first stage, initiation, a small amount of monomer was introduced into the chamber by maintaining flow from line #3 to enter the system along with flow from line #2. Typically this flowrate in line #3 was kept at 11.79sccm for a period of 1 hour. This accounted for the environment inside the chamber to be at a relative humidity of ~ 80%. In the second stage the flowrates of water vapor and ECA monomer and Ar purge lines to the chamber were set depending on the specific requirements of monomer concentration and humidity of the experiment. The duration of the second stage also depended on the specifications of the experiment being conducted.
The Ar purge from Line #1 was used to control the effective humidity in the chamber and also to purge and remove water vapor and ECA monomer vapor from the lines and chamber before and after polymerization runs. A photograph of white polymer deposit on glass slides in the tubular chamber is depicted in Fig. 2.4.
34
Figure 2.4: Photograph depicting white polymer residue on glass slides in tubular chamber for flow setup.
The advantage offered by using the 2-stage flow set-up approach over the batch set-up is the ability to separate the humidification and polymerization parts of the experiment to enable control over the polymerization process. The batch set-up was therefore used primarily when polymerization was to be carried out on a number of substrates at once under the same conditions and the flow set-up was used for controlled polymerization experiments
2.2 Techniques for Applying Initiators on Substrates
Application of initiators was made onto Si wafers and glass microslides either by spin coating or Silanation. Substrates were cleaned prior to use by immersing for 30min in a freshly prepared Piranha etching solution (3 parts conc. sufuric acid, 1 part hydrogen peroxide). Note: Extreme safety precaution must be exercised while preparing, using and disposing piranha etching solutions . Substrates are then rinsed thoroughly with flowing
DI water followed by ethanol and dried.
35 2.2.1 Spin-coating initiator solutions
Anionic initiators were spin-coated onto substrates, (typically Si wafers) from
aqueous solutions of the initiators Substrates were spun at a maximum speed of 1600rpm
and a total volume of 4ml of solution was deposited in 1minute. The substrate was
allowed to dry while spinning for an additional 1 min.
2.2.2 Silanation
Silanation is a convenient method to chemically alter the surface of a substrate. 2
Silanes are compounds that have a typical structure of X 3SiRY, where Y is the functional group of interest that is to be coupled onto the surface and X are hydrolyzable groups. X for example can be Cl or OR. For all the silanes used in this study X was of the OR form.
Coupling the silane onto a surface essentially involves two steps; (a) hydrolysis which is the formation of corresponding silanol (OH) 3SiRY and (b) condensation which is the reaction between OH groups on the surface and the silanol groups to yield Si-O-surface bonds. Protocol for silanation was adopted from the procedure described to obtain monolayer coverage of silane molecules on soda lime glass. 3 Water soluble silanes were applied from 2% aqueous solution of the silane in DI water. Water insoluble silanes were dissolved in a 95% ethanol, 5%water mixture with a resulting 2% silane concentration.
The silanes solution were prepared at least one hour prior to use to provide sufficient time for hydrolysis of the silane molecules to form silanol groups.
Substrates (Si wafers or glass slides) were cleaned as described above using a piranha etching solution. The substrates were immersed in the silane solution for times of
36 up to 30min after which they were removed and immediately rinsed with ethanol. The substrates were then placed in an oven maintained at 110°C for 2 hours to drive condensation between silanol groups on the silane molecules and the OH groups of the substrate. The silane-modified substrates were used for cyanoacrylate polymerization immediately after their preparation to minimize effects of contamination on the newly prepared surface. Characterization of the modified surfaces was done using X-ray
Photoelectron Spectroscopy described later in this section.
2.3 Characterization Techniques
2.3.1 Scanning Electron Microscopy (SEM)
SEM was used extensively for characterization of PECA nanofibers and other morphologies. A Hitachi S-3000H conventional SEM and a Jeol 6700F Field Emitter
SEM was used in this study .FESEM was better for images requiring higher magnification.( >30000X) The PECA polymer being non-conductive, an accelerating voltage of 3-3.5kV was used for all imaging which only probes the surface of the samples and minimizes charging accumulation effects on the sample .To further reduce charging a thin line of silver paint or conductive carbon tape is applied from the top surface of the sample to the SEM sample stub to make better electrical contact . In addition all samples were sputtered with a thin (~200 Å) conductive gold layer. Tilt views in the FESEM were carried out by using special sample stubs that were machined at angles of 90°, 45° or 60° to the horizontal.
37 2.3.2 Infra-red Spectroscopy (IR)
IR was used to characterize different chemical functional groups present in the
PECA polymer fibers and films. A Bruker IFS 66/S with a FTIR spectrometer was used in this study. An Attenuated Total Internal Reflection (ATR) objective was used to acquire the spectroscopic information from the polymer fibers and films. The polymer samples were either deposited or grown on glass surfaces. To avoid the appearance of the underlying glass in the IR spectra a small amount of polymer was scraped together into a pile and a microscope fitted above the ATR objective was used to position the ATR crystal directly above the accumulated mass of polymer. In this way, when the crystal was brought into contact with the polymer sample the volume of material probed by the
IR beam was consisted only of the polymer and not of the underlying glass substrate. The sampling area probed was a 100 µm X 100 µm area and was kept constant between samples. Spectra were collected in the mid IR range (4000-400 cm -1) by averaging over
100 scans. A spectrum collected with a bare ATR crystal in air prior to each measurement was used for background/reference subtraction.
For the experiment to collect the IR spectrum of ECA monomer, Transmission mode-IR was used. Here a thin film of liquid ECA monomer was applied to one side of two KBr pellets. The pellets were then sandwiched together and held in the path of the IR beam to collect the ECA IR spectra. The spectrum for two bare KBr pellets was used as a background for this case. Scans made on the ECA sample changes with time as the monomer polymerizes rapidly, hence only the scans taken immediately after applying the
38 liquid monomer onto the KRr pellet were indicative of the chemical functionalities in the monomer
2.3.3 Gel Permeation Chromatography (GPC)
GPC was used to determine molecular weights of PECA polymer films and fibers.
A Waters HPLC fitted with a four (0.5, 1, 3, 4) Styragel HRcolumns and a Refractive
Index detector were used for this estimation. Tetrahydrofuran (THF) was used as solvent.
Estimations were made with reference to a Universal calibration curve (log η[M] versus
Elution volume) made using linear Polystyrene standards. The Mark-Houwink
Parameters for the polystyrene standards and PECA polymers were obtained from previous studies 4 to construct the Universal calibration curve. 5
Samples injected into the GPC columns were prepared by dissolution of the polymer deposited on the glass substrates in THF. The weight loss from the glass substrates before and after dissolution was recorded and also corroborated with the weight of the polymer collected after drying the polymer-THF solution. Injection samples were made out to have a concentration of ~1.5mg of polymer./ml. The flow rate of THF through the GPC columns was set at 1ml/min giving a total run time of 50 min for a single injection. Multiple injections from the same solution sample were made to obtain an average value for molecular weight for that sample.
39 2.3.4 X-ray Photoelectron Spectroscopy (XPS)
XPS was used for determining the elemental composition of the different modified and unmodified substrates (silanated and untreated glasses & Si wafers) that were used for polymerization of ECA. A Kratos Analytical Axis Ultra Unit with a monochromatic Al k α X-ray source was used. The settings for the X-ray current and high
tension anode voltage and pass energy were kept constant for all the samples analyzed at
20mA, 14kV and 80eV, respectively.
The elemental composition expressed as atomic percentages was obtained by
carrying out survey scans of the surface ranging from binding energies of 1200eV to
50eV. Calibration of the survey scans was done relative to the position of the C 1s peak
set at 285.0eV. In the case of glass substrates (treated or untreated) the samples were
covered with conductive Al foil with an area of their surface cut out for the X-ray
analysis. This reduces the charging effect associated with the X-ray analysis of insulator
materials such as glasses.
2.3.5 Atomic Force Microscopy (AFM)
AFM was used to investigate the topology of different substrates before and after different treatments. A Nanoscope-Digital Instruments AFM fitted with a Si tip operating in tapping mode was used for this purpose. AFM height and phase images were obtained for the substrates. Scans were taken at resolutions of 256 or 512 once the scan settings
(Drive amplitude and Feedback gain) were optimized. The height image provides insight
40 into the topographical nature of the substrate and the phase images resolves regions in the scan area based on differences of the interaction between the tip and substrate.
2.4 Materials
Most chemical reagents used in this study were ACS grade purchased from Sigma
Aldrich unless other wise mentioned. DI water used for silanation was obtained from a
Barnstead Nanopure TM Water system. Ethanol used for silanation was 200% proof. The
following table (Tab. 2.1) summarizes the other materials used in this study along with their purchase source.
41
Table 2.1: Materials used in this study and their purchase source Material Source Material Source Sirchie Fingerprint Lab Methyl Ethyl 2-cyanoacrylate (ECA) Inc, CVS trimethoxysilane Gelest Pharmacy & (MTMS) Sigma Aldrich Propyl Tetrahydrofuran (Omnisolv VWR trimethoxysilane Gelest HPLC Grade) International (PTS) Aminopropyltrimethoxy silane Gelest Fused SiO slides Dell Optics (APS) 2 (3-trimethoxysilylpropyl) GAPS II Sigma Aldrich Corning diethylenetriamine (DETA) microslides N-(2-aminoethyl)- TeleChem 3- Superamine microslides Gelest International. aminopropyltrimet hoxy silane (AAS) TeleChem Superclean microslides ODAC Gelest International. Superfrost Gold Erie Scientific Nexterion microslides Schott Plus microslides Inc. Methyl 2-cyanoacrylate Polysciences Linoleic acid Sigma Aldrich (MCA) Inc. Palmitic acid Sigma Aldrich Stearic acid Sigma Aldrich
2.5 References
(1) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics , 72 ed.; CRC Press: Boston, 1991-1992. (2) Plueddemann, E. P. Silane Coupling Agents , 2nd ed.; Plenum Press: New York, 1991. (3) Metwalli, E.; Haines, D.; Becker, O.; Conzone, S.; Pantano, C. G. J Colloid Interf Sci 2006 , 298 , 825-831. (4) Donnelly, E. F.; Pepper, D. C. Makromol Chem-Rapid 1981 , 2, 439-442. (5) Painter, P. C.; Coleman, M. M. Fundamentals of polymer science: an introductory text , 2nd ed.; Technomic Pub. Co.: Lancaster, Pa., 1997
Chapter 3
Template-less Growth of Poly(ethyl 2-cyanoacrylate) Nanofibers by Initiators Present in Fingerprints
3.1 Introduction
By the introduction in Chapter 1, it is evident that polymer nanofibers have been a subject of great interest due to their enormous potential application in a variety of fields that include nanofiber reinforced composites, filtration, tissue templating, wound dressing, cosmetics, protective clothing, electrical and optical applications. The most promising techniques currently used to fabricate polymer nanofibers are (a) template- based synthesis 1 and (b) electrospinning. 2 The former method involves synthesizing the polymer chemically or electrochemically inside the pores of a template (e.g. anodized alumina membrane), or drawing a polymer melt into the pores by capillary action. The template then is sacrificed by dissolution to release the nanorods, nanowires or nanofibers. Furthermore the length of the nanostructures is limited by the template thickness. On the other hand electrospinning requires the polymer to be soluble in a suitable solvent or melt processible and also lacks control over the placement of the nanostructures. In this Chapter, novel catalytic growth of long nanofibers (>100 mm) of poly(ethyl 2-cyanoacrylate) is reported without a template via vapor phase polymerization. 3 The results provide insight into this method, which is a versatile route for growth of nanofibers with relative ease .
43 The monomer used in this study, ethyl 2-cyanoacrylate (ECA), commonly known as Superglue®, undergoes rapid anionic polymerization by nucleophilic attack to form a
4 mechanically strong polymer with excellent adhesive properties. This rapid polymerization, as mentioned earlier, has also been used as an effective method for developing latent fingerprints by forensic scientists via a process known as cyanoacrylate fuming, wherein fingerprints left on a surface are subjected to fumes of ECA. 5,6
Polymerization of the cyanoacrylate by initiators in the fingerprint results in the formation of a white polymer residue that makes the ridges of the fingerprint visible.
There are a number of techniques to enhance the development of the fingerprints for better identification. 7 Fuming in the presence of a relative humidity (RH) of 50–80% is known to accelerate the polymerization process. The study in this Chapter shows that at very high RH, not only does rapid polymerization of ECA occur, but also high aspect ratio nanofibers in large quantities are formed and only on the crests of the fingerprint.3
3.2 Experimental Section
3.2.1 Materials
Materials used for single initiator studies were NaCl, KCl, NH 4OH, Palmitic acid
(C 16 H32 O2), Stearic acid (C 18 H36 O2), Linoleic acid (C 18 H32 O2) and Squalene (C 30 H50 ). All were obtained from Sigma Aldrich. Cyanoacrylate used in this study was in the form of
Superglue® adhesive obtained from CVS Pharmacy.
44 3.2.2 Methods
Cyanoacrylate fuming for fingerprints : The batch set-up fuming chamber was used at room temperature with a humidity of > 95%. A silicon substrate was cleaned using a freshly prepared piranha etch solution, dried and then imprinted with a fingerprint. The substrate was then placed in the chamber along and humidified for ~ 10 hours at 95%RH. Subsequently, 1.5g of liquid ECA monomer was placed in a dish inside the chamber. Fans inside the chamber provided sufficient circulation in the gas phase.
Reaction was allowed to proceed for 12 hours. For low humidity (RH~30%) the experiment was carried out without the humidifying solution in the chamber. Humidity was monitored by a hygrometer (Fisher Scientific). Single initiator studies were carried out in the same fuming chamber. Solutions of initiators were spun coated and dried on clean Si wafers before placing them in the chamber.
Characterization of the morphology of the deposited polymer (PECA) was done using the Hitachi S-3000H conventional SEM.
3.3 Results
3.3.1 Fingerprint fuming
The clear indication of polymer depositing on the ridges of the fingerprint is the development of white residue after a sufficient duration of cyanoacrylate fuming as seen in Fig. 3.1 .In this manner the fingerprint that initially was invisible (latent) becomes
observable for identification.
45
Figure 3.1: Latent fingerprint observable after ECA fuming
On observing this white polymer deposit using SEM dense masses of polymer nanofibers are noticeable. Fig. 3.2 shows a typical scanning electron micrograph of
PECA grown on the ridges of the fingerprint on a silicon substrate in the fuming chamber at a RH greater than 95%. As seen in Fig. 3.2 (a) and (b) the fibers do not grow between the ridges of the fingerprint, illustrating the in-place deposition of fibers only on areas where initiator was present. Part (c) depicts an average area of fiber growth showing fibers that have diameters ranging from 100 nm up to 400 nm. However, interestingly, there seems to exist a monodispersity in the diameters of the fibers; since a large population of the fibers is clustered in the 200-250 nm range. A magnified view of such a typical fiber is shown in Fig. 3.2 (d). The length of the fibers was on the order of hundreds of microns implying aspect ratios in excess of 500.
46
Figure 3.2: SEM pictures of nanofibers of PECA grown on fingerprint ridges at room temp. & relative humidity >95% over a period of 12h (a) Low magnification view (b) Close-up view of the ridge pattern. (c) Close-up view of the nanofibers (d) Magnified view of a single fiber
When this same experiment was repeated under low humidity (RH < 30%) with other conditions remaining the same, only film-like polymer or no polymer at all, was generated. This is in accordance with previous studies of this polymer growth on
47 fingerprints that had low moisture content.6 This same study observed ‘noodle-like’
structures on prints that had higher moisture content. Also the PECA fibers here appear
similar to the poly(methyl 2-cyanoacrylate) fibers created by vapors of the monomer on
ice crystals on glass slides observed by Smith-Johannsen. 8,9
This interesting result warranted further investigation of the onset of growth of these nanofibers. Fig. 3.3 shows an SEM picture taken of the substrate surface at the early stages of the reaction (< 15min of exposure to ECA vapor). This set of micrographs clearly shows the development of the nanofibers, indicating that the catalytic growth of the polymer nanofibers has its genesis directly in the deposit of the fingerprint. Although, the fibers appear to be hollow they do have a inner core filling as seen in the top view of a single fiber stub (inset picture).
Figure 3.3: Snapshot of initial polymer fiber (15 min exposure to monomer and high humidity) growth on fingerprint at room temp. & relative humidity >95% (a) Low magnification view (b) Close-up view of the same (inset showing the top view of fiber).
48 3.3.2 Investigation of fingerprint residue components for PECA nanofiber formation
The residue left behind by the fingerprint is a complex composition of amino
acids, urea, lactic acids, fatty acids, glycerides mixed together with inorganic constituents
like chlorides, ammonia, sulfates and phosphates 10,11 . The polymerization of
cyanoacrylate proceeds anionically, so several constituents in the fingerprints such as
chlorides, moisture, carboxylic acid ions and amines can act as nucleophilic initiators to
trigger polymerization.12-14 Fingerprint residues are primarily comprised of secretions from the eccrine sweat glands present on the palms of the hands but also contains components of sebaceous gland (oily) secretions from other parts of the body. An overview of the most probable constituents of a fingerprint residue is given in Fig. 3.4
(adapted from Scruton et al. 11 ).
Figure 3.4: Composition of fingerprint residue (a) Eccrine sweat & (b) Sebum constituents [Adapted from Scruton et. al 11 ].
49 To elucidate the exact initiator in the fingerprints that is responsible for nanofiber formation, single initiators were subjected at a time to similar fuming conditions. From the chart in Fig. 3.4, the major component of eccrine sweat appears to be sodium or potassium chloride, therefore solutions of 0.1M NaCl and KCl were spin coated on separate substrates and subjected to ECA fuming. The SEM images of the polymer deposited on these substrates are shown in Fig. 3.5. It is evident that both NaCl and KCl are capable of initiating nanofiber polymer growth. The polymer fibers are shorter compared to those observed on the fingerprint residue. In some instances as seen for the case of KCl grown fibers there also is present some large pieces of polymer residue.
Major components of sebaceous gland secretions were also investigated for anionic initiation for PECA nanofiber formation. Fig. 3.6 depicts SEM images of fuming results on palmitic and stearic acid, two saturated fatty acids commonly present in these secretions. The substrates in each case did not show any white polymer residue accumulation visible to the naked eye after fuming. This was an indication of very little, if any polymerization occurring on the substrates. On viewing in the SEM very sparse nanofiber deposition is noticeable in both cases. The majority of the substrate appears to be covered with a rippled film (seen in both SEM images). This rippled film is in fact the original palmitic and stearic acid film deposit on the substrates, as confirmed by control
SEMs of only palmitic and stearic acid coated substrates included in Appendix A. Other components such as unsaturated fatty acid, linoleic acid and large hydrocarbon, squalene were found to be entirely non-initiating for ECA polymerization.
50
Figure 3.5: PECA nanofibers deposited on substrates spin coated with 0.1M (a) NaCl & (b) KCl
51
Figure 3.6: PECA nanofibers deposited on substrates spin coated with (a) Palmitic & (b) Stearic acid
52 These attempts to reproduce the neat nanofibers observed on the fingerprint residue only produced sparsely populated short nanofibers. Even increasing the concentration of the spin coating solution of NaCl, now a known initiator for nanofiber formation, to 1M or more only created tufts of fibers, some larger in diameter as seen in
Fig. 3.7 and does not resemble the nanofibers grown of the fingerprint residue (Fig. 3.2).
To investigate if the non-initiating components of the fingerprint at high humidity play a major role in dispersing the initiators that favor the formation of the nanofibers, a mixture of a non-initiating component, (linoleic acid or squalene), and a saturated solution of known initiator, NaCl was prepared to synthetically mimic the fingerprint residue. The mixtures, composed of equal volumes of non-initiating component and 4.5M NaCl, were shaken vigorously before applying on the substrate and dried immediately with air. The
SEM image results are shown in Fig. 3.8 for squalene and linoleic acid with NaCl. The
fibers here are longer than previously observed. In the case of the mixture of linoleic acid
and NaCl, the fibers appear very similar to those observed in the fingerprint residue.
However, in both these synthetic mixtures, the entire area of the substrate coated with
initiator did not yield fibers. In some areas phase separation of the aqueous and oily phase
occurred, causing the oily phase most likely to not initiate any polymerization.
The single initiator studies yielded another interesting observation with regard to the morphology of the polymer. Initiation by ammonium hydroxide resulted in a film with an interesting morphology; the film was seemingly involuted with ‘tortellini-like’ features as seen in Fig. 3.9.
53
Figure 3.7: PECA nanofiber deposited on substrates spin coated with 1M NaCl
Figure 3.8: PECA nanofibers initiated from synthetic mixtures of (a) Squalene and NaCl and (b) Linoleic acid and NaCl
54
Figure 3.9: PECA involuted film obtained via initiation by ammonium hydroxide
3.4 Discussion
3.4.1 Formation of nanofibers
The formation of PECA nanofibers without a template during polymerization of the monomer vapor is similar to a few other systems of biased orientation growth mechanisms observed previously. One is vapor-liquid-solid growth 15,16 of silicon
whiskers wherein a silicon crystal preferentially grows along a single axis upon
condensation from the vapor phase to form a needle-like Si crystal. Another example is
the intrinsic formation of polyaniline nanofibers during polymerization in a non-
solvent. 17 Here again the monomer preferentially associates with the growing tip of the fiber in the non-solvent environment. Surmising from these examples and the
55 observations made in this study it is evident that the PECA nanofiber tips must carry the active ends of the polymer chains in order to continue polymerization preferentially along the axis of the fiber. The intriguing question is what makes the polymer deposit as a 1- dimensional nanofiber? To address this question an analogy between the growth mechanisms during two phase transition processes; crystallization from a melt and heterogeneous polymerization is proposed. The modified Avrami equation has been successfully applied to explain different morphologies such as rods (1-D), discs (2-D) and spheres (3-D) observed during polymer melt crystallization 18-22 . The equation defines
the relationship between growth of polymer to a kinetic parameter such as time for each
dimensional modality. During crystallization, nucleation and the initial growth rate of the
crystals control the morphology of the polymer. This same analysis can be applied to
heterogeneous polymerization such as polymer formation from ECA vapor onto a
substrate. Here initiation denotes nucleation and hence concentration of initiator and rate
of initiation are the relevant parameters that control the morphology of the polymer. This
hypothesis is evident in Iguchi and Murase’s study.23 They had observed variation in
morphologies (plates and rods) of poly(oxymethylene) during its liquid-phase
polymerization using different ratios of catalyst to co-catalyst concentrations. These
variations were attributed to the differences in the rate and sites of nucleation for the
different polymerization conditions. Hence for the case of PECA nanofiber formation the
conditions of high humidity, concentration and type (that determines rate) of initiator
must be favorable for 1-dimensional fiber initiation (nucleation) during polymerization.
Subsequent monomer addition from the vapor phase occurs only at the tip of the fiber as
the sides of the fiber must not contain any active propagating species leading to continued
56 fiber growth. This premise of determination of polymer formation during initiation is re- confirmed by the fact that by using ammonium hydroxide as the initiator, under the same conditions of fuming, a polymer film was deposited. By varying the type of initiator the rate of initiation was altered thereby resulting in the formation of 2-D film morphology.
The formation of these different morphologies imply that 1-D (fiber) or 2-D (film) growth mechanisms during polymerization can occur with different initiators. This advocates the idea that for the appropriate choice of reaction conditions and initiators one can controllably fabricate the required morphology of the polymer
3.4.2 Initiators for nanofiber formation
The single initiator studies revealed the component of fingerprint residue responsible for PECA nanofiber growth. The chloride ion, the most prevalent anion in fingerprint residue, initiated PECA nanofiber formation successfully. The high humidity essential during nanofiber formation was most likely required to solvate the chloride ion and make it available for anionic initiation. The stearate and palmitate anion of the stearic and palmitic acid were also sparsely active towards nanofiber initiation however not as efficient as the chloride ion. The best result obtained in this study was with the attempt to synthetically mimic the composition of the fingerprint residue using mixtures of aqueous sodium chloride and linoleic acid. These results indicated that the growth of high aspect ratio, densely populated nanofibers arises from the complex composition of fingerprint interacting with the monomer vapor at high humidity, however the role played by the presence of the non-initiating components in the mixture is not completely understood.
57 3.5 Conclusions
In conclusion, this study demonstrated a facile, template-less, catalytic route to synthesizing PECA nanofibers. Such fabrication as in the case of other examples of intrinsic formation of nanofibers occurs due to preferential growth at the tip of the fibers.
An analogy between polymerization and crystallization kinetics suggests conditions of initiation were favorable for 1-D growth mechanism. By varying the initiator type (from chloride to hydroxide) a 2-D polymer film morphology was obtained signifying two dimensional growth mechanism. Hence for appropriate initiation and reaction conditions this template-less technique for nanofiber formation can potentially be applied as a model for other polymer systems.
The influence of the type of initiator on the polymer morphology to elucidate the reasons for observing such morphologies is explored in the next chapter.
3.6 References
(1) Martin, C. R. Science 1994 , 266 , 1961-1966. (2) Li, D.; Xia, Y. N. Adv Mater 2004 , 16 , 1151-1170. (3) Mankidy, P. J.; Ramakrishnan, R. B.; Foley, H. C. Chem Commun 2006 , 1139- 1141. (4) Coover, H. W.; Dreifus, D. W.; T., O. C. J. Handbook of Adhesives , 3rd ed.; Van Nostrand Reinhold: New York, 1990. (5) Menzel, E. R.; Burt, J. A.; Sinor, T. W.; Tubachley, W. B.; Jordan, K. J. J Forensic Sci 1983 , 28 , 307-317. (6) Lewis, L. A.; Smithwick, R. W.; Devault, G. L.; Bolinger, B.; Lewis, S. A. J Forensic Sci 2001 , 46 , 241-246. (7) Lee, H. C.; Gaensslen, R. E. Advances in fingerprint technology ; Elsevier: New York, 1991. (8) Smith-Johannsen, R. I. Nature 1965 , 205 , 1204-1205. (9) Smith-Johannsen, R. I. Science 1971 , 171 , 1246-1247. (10) Knowles, A. M. J Phys E Sci Instrum 1978 , 11 , 713-721.
58 (11) Scruton, B.; Robins, B. W.; Blott, B. H. J Phys D Appl Phys 1975 , 8, 714-723. (12) Eromosele, I. C.; Pepper, D. C.; Ryan, B. Makromol Chem 1989 , 190 , 1613- 1622. (13) Pepper, D. C.; Ryan, B. Makromol Chem 1983 , 184 , 383-394. (14) Wargacki, S.; Dadmun, M. D.; Lewis, L. Abstr Pap Am Chem S 2005 , 229 , U981-U982. (15) Wagner, R. S.; Ellis, W. C. Appl Phys Lett 1964 , 4, 89-91. (16) Levitt, A. P. Whisker technology ; Wiley-Interscience: New York, 1970. (17) Huang, J. X.; Kaner, R. B. Chem Commun 2006 , 367-376. (18) Avrami, M. J Chem Phys 1941 , 9, 177-184. (19) Avrami, M. J Chem Phys 1940 , 8, 212-224. (20) Avrami, M. J Chem Phys 1939 , 7, 1103-1112. (21) Hay, J. N. Br. Polym. J. 1971 , 3, 74-82. (22) Meares, P. Polymers: structure and bulk properties ; Van Nostrand: London, New York, 1965. (23) Iguchi, M.; Murase, I. Makromol Chem 1975 , 176 , 2113-2126.
Chapter 4
Influence of Initiators on the Growth of Poly (ethyl 2-cyanoacrylate) Nanofibers
4.1 Introduction
In the last Chapter a novel technique for facile growth of poly (ethyl 2- cyanoacrylate) [PECA] nanofibers was demonstrated by vapor phase polymerization of the ethyl 2-cyanoacrylate [ECA] monomer. This synthesis of polymer nanofibers presents an interesting and truly ‘bottom-up’ route that is an alternative to template based processes or electrospinning. In our previous studies, it was observed that vapor phase polymerization carried out under high relative humidity conditions resulted in different morphologies of the polymer depending upon the type of initiator used. For certain initiators, such as NaCl, a mass of nanofibers were obtained, while for NH 4OH, an involuted polymer film was obtained. The hypothesis suggested was one of different dimensional growths occurring in each case owing to different rates of initiation.
In this chapter that hypothesis is investigated further. The study of polymerization of ECA vapor on substrates coated with initiators resulting in either polymer films or fibers is reported. A classification of these initiators is proposed that explains the basis of polymer film versus fiber formation. The polymer nanofibers and films were characterized by infrared (IR) spectroscopy, scanning electron microscopy (SEM) and molecular weight of the polymer was measured by gel permeation chromatography
60 (GPC). The mechanism of formation of the different morphologies is discussed based on the results
4.2 Experimental Section
General descriptions of the materials and methods employed in this work are
discussed in Chapter 2. Specific details follow
4.2.1 Materials
ECA monomer was used as received (>95% purity, Sirchie Fingerprint
Laboratories Inc.). The initiator compounds, NaI, NaBr, NaCl, NaH 2PO4, Na 2SO 4,
CH 3COONa, Na 2HPO 4, NaOH, Na 3PO 4, Na 2CO 3 and NaHCO 3, were all ACS reagent grade and used as received (Sigma-Aldrich). Water used in this study was Nanopure water obtained from a Barnstead TM purification system
4.2.2 Methods
Fuming : Vapor phase polymerization of ECA was carried out in the batch setup
apparatus. RH was maintained at ~ 95% at room temperature using an 8% by wt. aqueous
sulfuric acid solution in the chamber for all fuming experiments. Each initiator compound
was spin coated onto a clean Si wafer from a 0.2M aqueous solution of the salt.
Approximately 1ml of solution was deposited onto the wafer by spin coating at a speed of
~1600 rpm. The wafer was dried and placed in the chamber at 95% RH for a period of
61 10hrs, after which ~2 gm of liquid ECA monomer was introduced into the chamber. After sufficient time for polymerization (12 hours), the wafers coated with polymer residue, were removed and characterized.
Characterization: Scanning electron microscopy was done using a JEOL 6700F
Field Emission SEM for morphological observations. Infrared spectroscopy was done using attenuated total reflectance IR (ATR-IR) on representative samples of polymer taken from the surface of the Si wafer on which they were prepared . Molecular weight estimations were made using GPC by dissolving the polymer films or fibers in THF. For these samples polymerization was carried out for sufficiently long (> 5days) periods of time to acquire enough quantity of polymer for GPC sample preparation.
4.3 Results and Discussion
As was observed in the initial studies with vapor phase polymerization of ECA,
under the same high relative humidity conditions using different initiators the polymer
morphology obtained was either in the form of nanofibers as shown in Fig. 4.1 (a) or a smooth, textured film depicted in Fig. 4.1 (b). The nanofibers in this case were obtained
using 0.2M Na 2HPO 4 as the initiator and the film was obtained using 0.2M Na 3PO 4 as the initiator. The two kinds of morphologies signifying 1D (fiber) and 2D (film) forms of the polymer morphology represent differences in the initiation and/or subsequent growth during polymerization. To better understand this mechanism of polymerization a classification of the different initiators is required to explain the basis for polymer morphology. As this polymerization occurs via an anionic mechanism, this study looks at
62 the characteristics of the anionic initiator and how that influences the polymer morphology.
4.3.1 Classification of Initiators
Several initiators were examined for the type of polymer morphology that they developed. Fig. 4.2 shows SEM images of the polymer grown using a few of those different initiators. The cation (Na) for the different anionic initiators was kept the same for consistency. It was observed that the chloride, monophosphate, diphosphate and sulfate anions resulted in the formation of polymer nanofibers while the acetate, hydroxide, triphosphate, carbonate and bicarbonate anions all resulted in the formation of the polymer film. Interestingly, iodide and bromide anions were unable to initiate polymerization of the monomer from the vapor phase and resulted in no polymer being deposited at all. The nanofibers obtained were similar to those observed in Chapter 3 and
others. 1 The diameters of the fibers are roughly in the range of 150-200nm. The polymer
film obtained was smooth, continuous but also appears involuted. A noteworthy
observation was that for the initiators examined, a given initiator only resulted in one
kind of polymer morphology – not both, i.e. from an initiator that resulted in polymer
nanofibers, no polymer film was observed and vice versa.
63
Figure 4.1: SEM images of PEC A morphology obtained from vapor phase polymerization on substrates spin coated with (a) 0.2M Na 2HPO 4 and (b) 0.2M Na 3PO 4. Scale bars are 1 µm. 64
Figure 4.2: SEM images of different polymer morphologies obtained by vapor phase polymerization on substrates spin coated with d ifferent anionic initiators at the same concentration of 0.2M. All scale bars are 1 µm.
65 Also, the concentration of the spin coated solution did not seem to affect the polymer morphology. Polymer grown on Si wafers spin coated with 0.2M and 1M NaOH solutions resulted in identical polymer films and polymer grown on wafers spin coated with 0.2M and 1M NaCl solutions gave identical polymer nanofibers. This phenomenon, that a particular anion always initiated polymerization to develop the same exclusive morphology (film or fiber), suggests an inherent growth directing or ‘templating’ effect occurring during the initiation step which was influenced by the type of anionic initiators used for ECA polymerization .
To observe the effect of a different cation, as an example, 0.2M solutions of KOH and KCl were also spin coated on substrates and examined for the type of polymer morphology they developed. Fig. 4.3 (a) and (b) show SEM images of polymer grown on these substrates. The morphologies observed, i.e. film for hydroxide ion and fibers for chloride ion, are in agreement with those observed for initiation using NaOH and NaCl, thereby suggesting that the morphology of the polymer obtained was inherently dependent on the type of anion used for initiation.
The morphology of the spin coated material (initiator) was also observed prior to polymerization in order to examine the influence, if any, of the morphology of the underlying initiator layer on the morphology of the resultant polymeric material. Again as an example, 0.2M solutions of NaOH and NaCl were spin coated onto Si substrates, dried and viewed using SEM (Fig. 4.4 ). The SEM images confirm that the underlying spin coated initiator does not resemble the morphology of the polymer (film or fiber) grown on these substrates. Thus polymer morphology must be a function of the polymerization mechanism itself.
66
Figure 4.3: SEM images of polymer morphologies observed for polymerization using (a) 0.2M KOH and (b) 0.2M KCl as initiators. Scale bars shown are 1 µm
It was observed that the presence of a high relative humidity (95% RH) atmosphere during the polymerization process was essential for polymer nanofiber formation. The high relative humidity is required for solvating the ions of the spin-coated initiator, but water is also known to be an effective initiator itself for the polymerization of ECA. 1,2 Hence it was important to examine the hygroscopicity of the initiator compounds studied as the amount of water taken up by different initiators could be the cause for different polymer morphologies. Tab. 4.1 shows the amount of water taken up by each initiator after being placed in a 95% RH environment for a period of 20 hours on the basis of grams of water uptake per gram of dry solid. The table lists the initiators in the order of increasing water uptake and also depicts the corresponding polymer morphology observed by SEM. There appears to be no correlation between the hygroscopicity of the initiator and the polymer morphology.
67
Figure 4.4: SEM images of spin coated (a) 0.2 M NaOH and (b) 0.2M NaCl solutions after drying on Si substrates prior to polymerization
68
Table 4.1: Water uptake of different initiators for ECA polymerizat ion and the morphology observed by SEM Water uptake Polymer Initiator (gms of water/gm dry Morphology solid)
NaHCO 3 not measurable film
Na 2SO 4 1.61 fiber
Na 3PO 4 2.16 film
NaH 2PO 4 2.95 fiber
Na 2HPO 4 3.07 fiber KCl 3.23 fiber NaI 3.37 no polymer
Na 2CO 3 3.66 film NaBr 4.37 no polymer
CH 3COONa 4.94 film NaCl 5.23 fiber KOH 5.67 film NaOH 7.70 film
For example, Na 2SO 4 which had an uptake of 1.61 gm of water/gm of dry solid and NaCl which had a water uptake of 5.23 gm of water/gm of dry solid, both resulted in polymer nanofibers being formed. But by contrast Na3PO 4 and NaOH that had uptakes of
2.16 and 5.67 gm of water/gm of dry solid, respectively, resulted in polymer films. Also
NaI and NaBr that had water uptake amounts similar to those of KCl and CH 3COONa,
69 respectively caused no polymerization at all. This supports the hypothesis that the factor most responsible for affecting polymerization must be the anion itself
A classification of the initiators on the basis of the relative softness-hardness of the anions proved to be a more viable explanation for the different morphologies observed. The Hard Soft Acid Base (HSAB) theory developed by Pearson 3,4 defines hard acids or bases as small, slightly polarizable species and soft acids and bases as larger, easily polarizable species. The HSAB principle suggests the following general rule of thumb for qualitatively estimating the stability of acid-base complexes ‘hard acids prefer
to associate with hard bases and soft acids prefer to associate with soft bases.’ This
statement implies that acids or bases can be classified as hard or soft by a measure of
their apparent preference to react with other hard or soft species. In order to classify the
anionic initiators in this investigation a comparison was made between the preference of
- + an anionic species [B ] to bind with a soft acid, methylmercury cation CH3Hg (pK s
+ values) and its preference towards the hard acid H (pK a values). Here pK a =
+ - + - log[HB]/[H ][B ] and pK s = log[CH 3HgB]/[CH 3Hg ][B ]. Large pK s values then imply a
strong affinity for the soft acid and, therefore, is an indication of [B -] possessing a soft character. Similarly, higher pK a values represent a greater affinity for the hard acid, and
thus reflect a hard character for [B -]. To obtain a measure of the relative soft-hard
character of the anions it is necessary to consider both pK s and pK a, hence we looked at the difference between the pK s and pK a values. Larger (pK s-pK a) values indicated a softer anion and smaller (pK s-pK a) values represent a harder anion. Tab. 4.2 lists the anionic initiators along their respective pK s and pK a values and the corresponding morphology of the polymer observed by SEM. The pK s and pK a values shown in Tab. 4.2 were obtained
70 5-9 from reports published in the literature. For certain anions whose pK s values were unavailable, estimations were made for their values. The list in Tab. 4.2 is arranged from the top in decreasing order of (pK s-pK a) values. Interestingly, the corresponding observed polymer morphology follows a trend with the decreasing (pK s-pK a) values. For the larger
values of (pK s-pK a), (greater than 12.25) i.e. the softest anions investigated, no polymer formation was observed. For an intermediate range of (pK s-pK a) values (from 12.25 to -
1.76), the anions appear to possess a certain soft-hard character that is favored for polymer nanofiber formation. For values below that range i.e. the harder anions, only the textured film-like polymer morphology was observed. The only exception from the general trend was the acetate ion which had resulted in a polymer film, but had a (pK s- pK a) value of -1.18. This could be attributed to variations of the pK s and pK a values
obtained from the literature. This classification then implies that it is possible to
distinguish between nanofiber-yielding initiators and polymer film-yielding initiators
based on their relative soft-hard character.
71
Table 4.2: Classification of initiators on the basis of their relative softness/hardness
Anion pK s pK a pK s-pK a Morphology I- 8.6 a -9.5 a 18.1 no polymer Br - 6.62 a -9a 15.62 no polymer Cl - 5.25 a -7a 12.25 fiber - * e H2PO 4 >5.03 2.12 >2.91 fiber 2- d e SO 4 2.64 1.92 0.72 fiber - b b CH 3COO 3.36 4.54 -1.18 film 2- a a HPO 4 5.03 6.79 -1.76 fiber OH - 9.37 a 15.7 a -6.33 film 3- * e PO 4 <5.03 12.67 < -7.64 film 2- c e CO 3 1.89 10.25 -8.36 film - * e HCO 3 >1.89 6.37 >-4.48 film a Ref. [5]; b Ref. [6]; c Ref. [7] ; d Ref. [8]; e Ref. [9]; * Qualitative estimates based on relative size and polarizability
The HSAB principle has been widely and successfully applied to explain experimental observations in organic chemistry. 10-12 Qualitative correlations have been
made based on the hard-hard, soft-soft interaction principle to rationalize the
thermodynamic stability of acid-base complexes and the kinetics of their formation. The
general rule with respect to kinetics states that ‘hard electrophiles react quickly with hard
nucleophiles and soft electrophiles react quickly with soft nucleophiles.’ With regard to
ECA polymerization, initiation is the interaction between an electrophile and a nucleophile. Because of the strong electron withdrawing effect of the –CN and –
COOC 2H5 groups, the double bond in the monomer is slightly polarized rendering the β-
carbon atom susceptible for nucleophilic attack by the anionic species 2. This slightly
72 positive center then acts as a hard acid center because of its positive charge and low polarizability. The trend observed in Tab. 4.2 implies that as the soft character (pK s-pK a)
of the initiating anion decreases, or conversely as the hard character increases, the rate of
initiation becomes faster because of an increased hard-hard nature of the interaction
between the anion and the monomer which results in polymer film formation. Whereas,
when the interaction is between a softer anionic initiator and the hard center of the
monomer, the rate of initiation is relatively slower and the resulting polymer morphology
obtained is nanofibrous. Tab. 4.2 also then indicates that for initiators that are even softer,
no initiation takes place at all for the vapor phase monomer as no polymer is deposited
for those initiators (iodide and bromide ions). However, it must be pointed out that in the
13 liquid phase these initiators are capable of initiating polymerization.
Hence the classification of the initiators in terms of their relative soft-hard character in fact points to a distinction based on the corresponding rates of initiation associated with each of those initiators, thereby leading to the prediction that for fast rates of initiation, the result will be a polymer film and for relatively slower rates of initiation, the fibrous polymer morphology will be obtained. This assertion, in fact even agrees with the analogy of polymerization and polymer crystallization kinetics suggested in the last chapter. Comparing initial rates of nucleation and crystal growth, 1-D needles occur at a rate slower relative to 2-D disc growth. 14
73 4.3.2 IR investigation of PECA film and nanofibers
IR and Raman techniques have been used to study real time polymerization kinetics of alkyl cyanoacrylates. 15-18 These studies looked at characteristic differences in the spectral bands as monomer was consumed to make the polymer. In this study IR spectroscopy was used to examine differences in the chemical structure between the different PECA samples. The IR spectra of ECA monomer, vapor phase polymerized
PECA-film and PECA-nanofiber samples and an ECA film that was cured under ambient conditions were compared. In this last sample the top layer of the liquid monomer is first initiated by moisture in the air which leads to subsequent propagation (curing) taking place in the liquid phase of the monomer until the entire film was completely polymerized. Fig. 4.5 depicts the complete IR spectra obtained for the cured ECA film sample and the vapor phase polymerized film and nanofibers samples. The spectra show certain differences among each other that are discussed below.
74
Figure 4.5: Complete IR spectra obtained for (a) cured ECA film; (b) Vapor phase polymerized PECA film and (c) PECA nanofibers
Fig. 4.6 Fig. 4.7 and Fig. 4.8 depict specific regions of interest in the IR spectra obtained for these samples along the ECA monomer spectrum. The ECA monomer IR spectrum was obtained by transmission-IR of a thin film of liquid monomer spread between two KBr pellets. The remaining three PECA samples were analyzed by ATR on polymer residue collected on a Si wafer. Peak assignments in the spectra were made by referring to characteristic IR absorption of the different functional groups. 19 Interestingly, the spectra for the two vapor phase polymerized PECA samples appear identical, suggesting that the chemical structure of the two morphologically different polymers is the same.
75
Figure 4.6: IR spectra for different PECA samples showing Region I: 3300-2700 cm -1
Figure 4.7: IR spectra for different PECA samples showing Region II: 2300-2200 cm -1
76
Figure 4.8: IR spectra for different PECA samples showing Region III: 2000-1500 cm -1
In region I, (Fig. 4.6 , 3300 cm -1-2700 cm -1), the most prominent spectral change between
-1 the ECA monomer and PECA polymer is the disappearance of the peak at 3130 cm in the polymer. The vibration at this frequency arises because of the =CH 2 stretching of the
vinyl group present in the monomer, whereas for the PECA polymer chain this vinyl
functionality does not exist and hence disappears. This is apparent in all three polymer
samples. Peaks between 3000-2800 cm -1 are due to the asymmetric and symmetric stretches of the –CH 3 and –CH 2– groups. A noteworthy fact about the composition of the cured PECA film is that it contains small amounts (up to 10%) of poly(methyl methacrylate) (PMMA). PMMA is present in the as received ECA monomer from Sirchie
77 Fingerprint Lab Inc. to increase the viscosity of the adhesive for better application. The presence of this PMMA is observed as a small peak at 2850 cm -1 because of symmetric
C-H stretches of the CH 3-O group. This peak is not detectable in the two vapor phase
polymerized samples as they do not contain any PMMA. Also the relative ratios of the
methyl and methylene peaks in the cured PECA film are different from this ratio in the
two vapor phase polymerized samples. This again is an effect of the additional PMMA in
the cured PECA. More details of how the presence of a CH 3-O group is manifested in the
IR spectra is discussed in Chapter7.
In region II, (Fig. 4.7 , 2300 cm -1-2200 cm -1), the peak due to –CN stretching is
-1 seen at 2239 cm in the ECA monomer. This peak is completely absent in the bulk liquid
PECA-film and appears at higher wavenumbers and of a lesser intensity in the two vapor phase polymers. The shift of the –CN peak and reduction in its intensity has been observed previously in IR studies of alkylcyanoacrylate polymerization. The shift has been attributed to loss of conjugation between the –CN, C=C and C=O groups and because of the presence of –CN in two different environments, 16 or because of the significant stablilization of the carbanion formed by delocalization of the negative charge, or an effect of inter and intra-molecular hydrogen bonding. 15 However, at this point the cause of the decrease in intensity of the –CN peak is not known.
Region III, (Fig. 4.8 , 2000 cm -1-1500 cm -1) is important for two peaks, one at
-1 -1 ~1730 cm assigned for C=O stretching and one near 1615 cm representing C=C stretching vibrations. The C=O peak shifts to a lower frequency for the bulk phase
PECA-film and shifts to a slightly higher frequency for the vapor phase PECA. The net shift in position of the C=O peak usually is a result of a few factors that influence upward
78 or downward shifts in the frequency of IR absorption. 19 Some of the reasons causing such a decrease in frequency are, liquid monomer converting to solid polymer causing an increase in H bonding interactions and resonance hybrids with the C=O group. The factor that causes an increase in the absorption frequency is the loss of unsaturation in the β-
position because of polymer backbone formation. Hence it appears that for the vapor-
grown PECA, the increase in frequency caused by polymer back bone formation is
greater than the decrease in frequency caused by H-bonding or resonance hybrids. With
regard to the C=C stretch, in going from the ECA monomer to the cured ECA film this
peak disappears as expected with the elimination of the double bond in the monomer as
the polymer back bone is formed. However, in the two vapor-grown PECA polymers a
weak signal for this peak still exists. This can be explained by chain transfer steps
occuring by hydride ion elimination ( Fig. 4.9 ). Such chain transfer steps result in a dead polymer chain having a C=C at the end. These C=C ended chains account for the
-1 intensity of the peak at 1615 cm to be present still in the polymer. Hence it can be suggested that in the vapor phase polymerization significant chain transfer steps occur by
H- elimination.
C N C N C N C N H H -H- H H Nu C C C C Nu C C C C H C O H C O H C O C O
OC2H5 OC2H5 OC2H5 OC2H5 n n Figure 4.9: Probable chain transfer route during vapor phase ECA polymerization
79 To summarize, the IR investigation indicated no structural differences between the PECA-film and PECA-nanofiber sample that were deposited by vapor phase polymerization of the monomer. However, they did differ from a PECA polymer sample that was cured in the liquid phase. The differences suggesting that transfer steps in the vapor-phase polymerization route occur via hydride ion elimination from the growing end of an active polymer chain.
4.3.3 Molecular weight estimations of PECA film and nanofiber
In order to collect sufficient amount of sample for injection into the GPC, fuming was carried on two separate substrates spin coated with 1M NaCl and 1M NaOH. The duration of fuming was in excess of 5 days in the batch setup with periodic replenishment of liquid ECA monomer in the chamber. The SEM pictures of the polymer formed on each substrate are shown in Fig. 4.10 . As expected the NaCl coated substrate yields
nanofibers and the NaOH substrate yields the polymer film.
80
Figure 4.10 : PECA samples used for molecular weight measurements initiated by (a) 1M NaCl and (b) 1M NaOH
81 Molecular weight estimations were obtained using a universal polystyrene calibration curve as the standard. The concentration of all the polymers samples injected in to the GPC was kept at approximately the same value of 1.5mg/ml. The flowrate of solvent was maintained at 1ml/min which gave a retention time of 40min in the GPC columns. Fig. 4.11 presents the trace obtained from the Refractive Index (RI) detector on the GPC as a function of elution time for different polymer samples. The traces are for vapor phase polymerized PECA-nanofibers and PECA-film initiated by 1M NaCl and
1M NaOH, respectively.
Figure 4.11 : GPC traces of (a) PECA nanofibers made via vapor phase polymerization using 1M NaCl; (b) PECA film made via vapor phase polymerization using 1M NaOH
82 Peak molecular weights are indicated for each major peak in all traces.
Comparing the two vapor phase polymers, the PECA-film made by using NaOH as initiator has a higher molecular weight than the PECA-nanofibers made using NaCl as the initiator. As the concentration of initiator used for making these two polymers was the same and the duration of polymerization was the same, the difference is due to the differences in the rates of initiation, propagation or chain transfer (termination) during their polymerization. The IR analysis indicated that transfer or termination reactions are identical in these two PECA polymers implying that the lower molecular weight for
PECA-nanofibers signifies a slower rate for initiation in contrast to the PECA-film polymerization. Correlating this observation with the classification of initiators already established there is an agreement with the observed trend in rates of initiation of polymerization and resulting molecular weights of the polymer. For faster rates of
initiation with harder anions leading to PECA-film formation, the molecular weight of
the polymer obtained is higher than that obtained by slower rates of initiation with softer anions leading to PECA-fiber formation.
4.3.4 Mechanism of formation of different polymer morphologies during vapor phase polymerization
The IR investigation of the different PECA samples revealed that the two vapor phase PECA samples; film and nanofibers, were identical with respect to their chemical structure, hence the explanation of the different morphologies during polymerization must lie in the mechanism of polymer growth. The classification of the different initiators
83 based on the HSAB principle and the molecular weight estimations by GPC that revealed higher molecular weight fractions in the PECA-film and lower molecular weight fractions in the PECA-fiber sample, confirmed that faster rates of initiation favored film formation (2-D growth) and relatively slower rates favored fiber formation (1-D growth).
Based on these observations a suggested mechanism for formation of different polymer morphologies is put forth below.
The rates of initiation and propagation for polymerization can be written as,
− − − I + M → P Initiation; Ri = ki [I ][ M ]
− − − Pn + M → Pn+1 Propagation: RP = k P [M ][ Pn ]
− − Where, I , M, Pn , n, R’s and k’s are anionic initiator, monomer, active polymer chain, number of monomer units in a polymer chain, rates and rate constants, respectively. Since the concentrations of the initiator [ I − ] and monomer [ M ] were kept
constant between the different PECA samples, then ( ki )fiber < ( ki )film or ( Ri )fiber <
( Ri )film . Therefore, at any instant after initiation, the concentration of growing polymer
− chains [ Pn ] will always be lower in the PECA-nanofiber sample than the PECA-film sample. This initial inequality in concentration of active chains caused by the difference in rates of initiation results in fewer centers of growing polymer per unit area (localized sites of initiation) for the PECA-fiber as compared to the PECA-film polymer. As propagation continues, the fewer number of polymer chains growing per unit area have a growth front propagating in one-direction, thus ‘templating’ the formation of 1-D nanofibers whereas relatively larger number of polymer chains growing per unit area have growth fronts in more than one direction, thereby ‘templating’ formation of a 2-D
84 film. Hence the initial ‘templating’ effect that determines the ultimate polymer
morphology is a function only of the rate constant for initiation, ki . Chain transfer
reactions affect both types of the growing polymers in the same manner. But because of
the difference in the concentration of active chains between the two polymers i.e.,
− − [ Pn ]fiber < [ Pn ]film , the overall progress of polymerization are also different, thereby
causing the polymer nanofibers to have a lower average molecular weight than the
polymer film.
4.4 Conclusions
In summary, we set out to answer some of the major questions regarding the mechanism of template-less formation of PECA nanofibers during vapor phase polymerization of ethyl 2-cyanoacrylate. To explain the mechanism the influence of different types of anionic initiators for polymerization that resulted in polymer nanofibers or a textured polymer film were investigated. Anionic initiators were compared on the basis of their hygroscopicity and softness-hardness character. A suitable method for classifying the initiators studied was established by comparing their relative soft-hard character (pK s-pK a). There appeared to be a correlation between (pK s-pK a) of the
initiators and the observed morphology. Typically, a harder anion with a more rapid
interaction with the hard acid center of the monomer would result in PECA-film
formation whereas a softer anion with a slower interaction with the hard acid center of
the monomer would result in PECA-nanofiber formation. This trend was also evident in
the molecular weight analysis where the PECA-film showed a higher molecular weight
85 than the PECA-nanofibers. IR spectra showed that there were no differences between the
two vapor phase polymers with respect to their chemical structure, however it did provide
insight into the possible chain transfer steps that occur during vapor phase polymerization
suggesting that transfer by hydride ion elimination is prominent. Finally, an explanation
is proposed for the mechanism of formation of PECA-nanofiber and PECA-film which
suggests that, the initial discrepancies between the rates of initiation causes localized sites
of initiation in the case of slow rates that promotes 1-D fiber growth versus 2-D film
morphology of the polymer when there is a greater concentration of initiation sites caused
by faster rate of initiation.
4.5 References
(1) Doiphode, S. V.; Reneker, D. H.; Chase, G. G. Polymer 2006 , 47 , 4328-4332. (2) Coover, H. W.; Dreifus, D. W.; T., O. C. J. Handbook of Adhesives , 3rd ed.; Van Nostrand Reinhold: New York, 1990. (3) Pearson, R. G. J Am Chem Soc 1963 , 85 , 3533-3539. (4) Pearson, R. G. Science 1966 , 151 , 172-177. (5) Schwarze.G; Schellen.M. Helv Chim Acta 1965 , 48 , 28-46. (6) Alderighi, L.; Gans, P.; Midollini, S.; Vacca, A. Inorg Chim Acta 2003 , 356 , 8- 18. (7) Sanz, J.; Raposo, J. C.; de Diego, A.; Madariaga, J. M. Appl Organomet Chem 2002 , 16 , 339-346. (8) De Robertis, A.; Foti, C.; Patane, G.; Sammartano, S. J Chem Eng Data 1998 , 43 , 957-960. (9) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics , 72 ed.; CRC Press: Boston, 1991-1992. (10) Pearson, R. G.; Songstad, J. J Am Chem Soc 1967 , 89 , 1827-1836. (11) Pearson, R. G. J Org Chem 1989 , 54 , 1423-1430. (12) Huheey, J. E. Inorganic chemistry; principles of structure and reactivity ; Harper & Row: New York, 1972. (13) Donnelly, E. F.; Johnston, D. S.; Pepper, D. C.; Dunn, D. J. J Polym Sci Pol Lett 1977 , 15 , 399-405. (14) Hay, J. N. Br. Polym. J. 1971 , 3, 74-82.
86 (15) Tomlinson, S. K.; Ghita, O. R.; Hooper, R. M.; Evans, K. E. Vib Spectrosc 2006 , 40 , 133-141. (16) Edwards, H. G. M.; Day, J. S. J Raman Spectrosc 2004 , 35 , 555-560. (17) Urlaub, E.; Popp, J.; Roman, V. E.; Kiefer, W.; Lankers, M.; Rossling, G. Chem Phys Lett 1998 , 298 , 177-182. (18) Yang, D. B. J Polym Sci Pol Chem 1993 , 31 , 199-208. (19) Socrates, G.; Socrates, G. Infrared and Raman characteristic group frequencies: tables and charts , 3rd ed.; Wiley: Chichester; New York, 2001.
87
Chapter 5
Polymerization of Ethyl 2-Cyanoacrylate Nanofibers on Glass Substrates
5.1 Introduction
As mentioned in earlier sections, the potential applications for polymer nanofibers extends over a wide variety of areas such as electronics, protective clothing and medicine. 1 The applicability of polymer nanofibers in some of these areas would depend
on the ability to produce these fibers via a facile route, one which involves few
processing steps, provides control over placement of the nanofibers and can be utilized
for bulk synthesis or larger scale production of nanofibers. The template-less fabrication
approach to making polymer nanofibers demonstrated and discussed in Chapters 3 and 4
is a potentially promising route for meeting those objectives. This inherent growth of
polymer nanofibers does away with the requirement of a template, occurs only on areas
where initiators for nanofiber polymerization are present, 2 and, as will be shown in this chapter and the next, can occur on suitably modified glass substrates for bulk synthesis.
To extend this template-less route of making PECA nanofibers further, a method to grow nanofibers on any substrate is essential. Smith-Johannsen 3 in 1971 had observed the formation of polymer whiskers of methyl 2-cyanoacrylate on ice crystals on glass slides that were exposed to monomer vapor. The explanation proposed was the availability of sufficient initiator in the form of adsorbed water on the glass surface to facilitate the formation of whiskers. Recent studies 4 have demonstrated the growth of
88 PECA nanofibers on electrospun fibers after exposing them to water vapor and then ECA vapor. Water condensed on the surface of electrospun fibers as tiny droplets, act as initiator ‘islands’ that grow PECA nanofibers from them. Analogies are made to explain this phenomenon with vapor-liquid-solid growth in whisker technology. 5 These techniques though successful, are not completely understood and could result in irregularity of the outcome depending on the surrounding environment humidity.
Initiating PECA nanofiber growth on substrates for the studies in Chapter 4 involved introducing ECA monomer vapor to substrates spin coated with appropriate initiators for polymerization. Spin coating can at times result in non-uniform films of initiator being deposited and this technique lacks control over where the initiator is deposited on a substrate. An alternate route to applying the initiator is by surface modification to chemically incorporate the initiator on the surface of the substrate.
Silanation 6 of the surface hence provides an easy procedure to apply the initiator uniformly on a substrate. Controlling the area of application of the silane would then translate to control over where the polymer nanofibers develop. In this chapter, silanes applied on glass substrates is employed to demonstrate a strategy for growth of PECA nanofibers on substrates. Using AFM, XPS and SEM, this study also provides insight into the mechanism of template-less growth of the PECA nanofibers on glass substrates and assists in bringing this technique a step closer to being a viable technology.
The results in this chapter are organized as follows. First, several commercially available silane coated glass slides are subjected to long time ECA fuming to elucidate the slides that were apt for PECA nanofiber formation. Subsequently, one particular glass slide, Superamine glass substrates, is chosen for further investigation under different
89 relative humidities and different times of fuming under similar humidities to understand how PECA nanofiber formation occurs on these substrates. Studies on these slides were also carried out to reveal the method of water condensation on them. The AFM study investigates the topology of the substrates and also compares the mechanism of ECA condensation on Superamine and a few other glass slides under low humidity. Finally, glass slides with different silane coatings were prepared in the laboratory and subjected to
ECA fuming to duplicate nanofiber formation and reconfirm the results from the commercial glass slides to conclude an all encompassing hypothesis for PECA nanofiber formation on glass substrates.
5.2 Experimental Section
General descriptions of the materials and methods employed in this work are discussed in Chapter 2. Specific details follow.
5.2.1 Materials
The following commercially available silane-modified glass slides were used in this work- Superamine (TeleChem Int. Inc.), Schott Nexterion A, Corning GAPS II, and
Superfrost Plus (Erie Scientific Company). In addition, unmodified glass slides also available from TeleChem (called as Superclean glass) were used as substrates to conduct surface modifications by silanes in our laboratory. The silanes used in this study were three aminosilanes (mono-, di- and tri-), 3-aminopropyltriethoxy silane (APS), N-(2-
90 aminoethyl)-3-aminopropyltrimethoxy silane (AAS) and (3-trimethoxysilylpropyl) diethylenetriamine (DETA), two alkylsilanes, propyltrimethoxy silane (PTS) and methyltrimethoxy Silane (MTMS) and one fluorosilane, (heptadecafluoro-1,1,2,2,- tertrahydrodecyl)trimethoxy silane (HDF), all purchased from Gelest Inc. The chemical structures of the silane molecules are included for reference in Appendix B. Ethyl 2- cyanoacrylate (ECA) purchased from Sirchie Fingerprint Lab Inc. was used as the monomer source.
5.2.2 Methods
Long-time Polymerization (Fuming): Polymerization of ECA vapor onto the glass
substrates was carried out at room temperature in the batch set-up chamber as described
in section 2.1.1 . The experiment involved two steps- first a period (10hrs) of high relative humidity (~95%) followed by a period (10hrs) after introducing monomer for the polymerization to take place in the presence of high humidity. The high relative humidity
(RH) was achieved by placing a trough containing an 8 wt% aqueous solution of sulfuric acid in the chamber.
Variable Humidity Fuming: In these experiments, polymerization of the ECA vapor was carried out under different conditions of RH in the chamber on the same type of glass substrate (Superamine, TeleChem) for the same amount of time. The first step of the experiment was subjecting the substrate to the particular RH for 10hours which was followed by 2 hours of polymerization after the introduction of monomer in the chamber.
Polymerization was carried out at five different humidities, 18%, 48%, 68%, 81%, and
91 94% achieved at room temperature using aqueous solutions of sulfuric acid with concentrations of 62 vol.%, 34 vol.%, 24 vol.% 17 vol.% and 8 vol.% respectively.
Polymerization sets were also carried out on the Superamine substrates under
48% RH for different lengths of time (2hrs, 6hrs, 10hrs and 12hrs) using separate substrates and also under 68% RH for different lengths of time (0.5hr, 2hrs, 4hrs, and
9hrs).
Water Condensation Imaging: Water condensation on Superamine substrates was imaged using a FEI Quanta 200 Environmental SEM (ESEM) fitted with a temperature controlled stage set at 5 °C. By controlling the pressure in the ESEM sample chamber, the relative humidity on the surface of the substrate was increased. As the humidity rises above 95%RH saturation in the vapor phase causes condensation of water droplets on the surface which were then imaged as a function of time.
Silanation: Superclean glass cleaned for 1 hour in a freshly prepared piranha etching solution (denoted as Superclean*) was used as the substrate for surface modification by silanes. Silanes were applied from 2% silane solutions in a 95:5, ethanol to water mixture. The silane solution was first allowed to undergo hydrolysis for
1hr30min before introducing the substrate for 30min. Upon removal the substrate was rinsed with ethanol and placed in an oven maintained at 110°C for 2 hours to complete the condensation reaction after which the sample were used immediately for polymerization.
Characterization: Field Emission Scanning Electron Microscopy (FESEM) was used to observe the morphology and characterize the dimensions of the polymer nanofibers. Atomic Force Microscopy (AFM) in tapping mode was used to determine the
92 topographical information about the different glass substrates. X-Ray Photoelectron
Spectroscopy (XPS) was used to determine the elemental composition of the glass substrate surfaces.
5.3 Results
5.3.1 Long-time fuming on commercial glass slides
The commercially available Superamine, Superfrost, Corning, and Schott glasses were used as-received as substrates for the long-time fuming experiments. Additionally, as- received Superclean glass and a piranha etched Superclean* were also tried for the same conditions of polymerization. The development of a white residue on the glass substrates was an indication of PECA polymer deposition. The Superamine, Superfrost and
Superclean glasses all developed a white residue during polymerization while the
Corning, Schott and Superclean*, glasses did not appear to have any polymer deposit as observed by the naked eye. Fig. 5.1 presents the SEM images of the surface of the six glass slides after polymerization. The scale bars on all the images depict 1 µm while for the inset images at higher magnifications the scale bars depict 100 nm. The three slides that visibly had a white polymer deposit, when viewed under the SEM show dense masses of polymer nanofibers. The fibers on the Superamine substrate have diameters ranging from 100nm to 300nm. The fibers grow into an entangled branched network making it difficult to estimate the length of the fibers but they appear to at least be greater than several tens of microns. Nanofibers deposited on the Superfrost Plus glass appear to
93 be more coiled than those on the Superamine glass. The diameters here range from 50nm to 300nm. The fibers on the Superclean glass appear similar to the fibers on the
Superamine glass and have diameters ranging from 50nm to 200nm. The Corning and
Schott glasses that appeared to have no visible polymer residue, when viewed under the
SEM show short polymer nanofibers sparsely dispersed on the substrate surface. Under high magnification the diameters of fibers on the Corning glass appear to be monodisperse centered at ~50nm. The fibers on Schott glass also appear to have monodisperse diameters centered at ~100nm. The fibers here appear to have grown along the surface of the substrate initially before pointing upwards. Comparing the number density (number of fibers per unit area), the Corning glass substrate has a slightly greater density of polymer fibers deposited than the Schott glass substrate. The image of the surface of the Superclean* glass does not show the formation of any polymer nanofibers, however, there does appear some thin polymer deposit on the surface.
94
Figure 5.1: SEM Image results of long-time fuming on (a)Superamine (b)Superfrost Plus (c) Corning GAPS II (d) Nexterion A Schott (e) Superclean (f) Superclean*. Scale bars = 1µm. Inset Scale bars = 100nm.
95 The extent of this film polymer deposition is fairly insignificant compared to the polymer deposition on Superamine, Superfrost and Superclean glasses as this substrate did not show any white deposit visible to the naked eye
An image of the surface of the plain Superamine glass out-of-the-box is included in the Appendix B for comparison.
5.3.2 Variable humidity fuming on Superamine glass
To capture the onset of polymer deposition Superamine glass substrates were subjected to 2 hours of fuming under different humidities. The substrates that were subjected to relative humidities of 18%, 48% and 68% did not develop any white polymer residue as seen by the naked eye, while the 81% and 94% RH samples did develop a faint white coloration. Fig. 5.2 presents the SEM images of the surface of the
Superamine substrates at two hours of polymerization after they were subjected to various humidities. The scale bar for all images in this figure is 100nm. The glass under
18%RH shows only a few nubbins or buds of polymer deposited on the surface. The size of the polymer buds is ~100 nm and appears to be constant across the entire glass surface.
Comparing with the 48%RH sample, the number of the polymer buds deposited on the surface is greater and the size of each bud is ~60nm which again is constant across the entire glass surface. For the 68%RH sample however, very few polymer buds are observed, instead short stubs of polymer nanofibers are present. Fig. 5.3 (a) and (b) are
SEM images of these short polymer stubs taken at a tilt angled view of 45°.The diameter of the stubs is 50-100nm and the length of the polymer stubs extends up to 200-250nm.
96
Figure 5.2: PECA polymer gr owth on Superamine glass substrates at 2 hours of polymerization under different relative humidities
97
Figure 5.3: SEM images taken at a tilt view of 45 ° of the 68%RH polymerization sample on Superamine glass (a) 20K magnification (b) 40K magnification
Here again as in the case of short polymer nanofibers on the Schott glass substrate, the fibers appear to grow parallel along the surface in some instances before pointing upwards and away from the surface. In some cases for the 68%RH sample (Fig. 5.2 (c))
Y-shaped branching of the polymer stubs is also evident. The 81%RH sample (Fig. 5.2
(d)) shows longer nanofibers for the same 2 hrs of polymerization. The fiber diameters
are in the 50-100nm range and more occurrences of branching are evident. Comparing
these fibers with the 94%RH sample (Fig. 5.2 (e)), the fibers grow even longer at the
higher humidity with even more branching but roughly the same 50-100nm diameters.
To investigate the development of the polymer residue at a given relative
humidity the formation of polymer on the Superamine glass substrates was investigated
as a function of time at 48%RH and 68%RH. Fig. 5.4 (a)-(d) are SEM images of the polymer deposited on the Superamine substrates during polymerization under 48%RH for
2, 6, 10 and 12 hours, respectively. The scale bar for all the images (including insets at higher magnification) is 100nm. The noticeable difference between the image at 2 hours
98 and 6 hours is the appearance of more polymer buds at the longer time. Although their number density increases the size of the polymer buds remains ≤ 60nm with some buds
as small as 30nm.The same trend is observed in going from the 6 hour sample to the 10
hour sample image. The density of the polymer buds increases even further as more
number of smaller size buds appear on the surface. In the final image at 12 hours, there is
further increase in the density of the polymer stubs to the extent that they appear to
almost completely cover the glass substrate surface. There also appears formation of
some isolated larger buds (~100nm) of polymer deposit on the surface.
Figure 5.4: Polymer bud deposition on Superamine glass substrat es under 48%RH at various time periods of fuming. Scale bar =100nm.
99 Fig. 5.5 (a)-(d) presents the SEM images of the surface of the Superamine glass after 0.5, 2, 4 and 9 hours of polymerization under 68%RH. In the image at 0.5 hours only short polymer nanofibers are seen. The diameter of the fibers is between 30-50nm and their length is ~200nm. Here again the fibers appear to grow parallel to the surface before pointing upwards. For the 2 hour sample the diameter of the short stubs is 50-
100nm and more fibers appear to be pointing upwards. At 4 hours, the fibers are longer, appear branched. The diameters of some of the fibers appear to be smaller than 80nm. In the final image at 9 hours, there are more of the branched nanofiber structures and the diameters of the fibers vary from 50nm to 100nm.
100
Figure 5.5: Polymer nanofiber growth on Superamine glass substrates under 68%RH at various time periods of fuming. Scale bar =100nm.
5.3.3 Water condensation imaging on Superamine glass
As water is a known initiator for ECA polymerization 7 we investigated the
formation of water droplets during condensation on a Superamine substrate. Fig. 5.6 (a)-
(d) are ESEM images of the surface of the Superamine Substrate taken as water
condenses on it at a relative humidity >95%.
101
Figure 5.6: ESEM images of water condensation on Superamine glass substrate at >95%RHScale bar =5 µm.
The images show a progressive nucleation and growth of water droplets condensing on the surface. In the right hand side portion of the image at 30sec a collection of water droplets having a diameter of ~500 nm appear on nucleation, while at the same time larger droplets (several microns) are already present. Observing this collection of droplets through the next three images taken at times 40, 75 and 110 sec, the drops seem to grow at about the same rate till they eventually pool together with other droplets to form larger water reservoirs on the surface.
102 5.3.4 AFM imaging on commercial glass slides
To capture the onset of polymer deposition on the glass substrates AFM imaging was done on the Superamine, Schott Nexterion and Corning GAPS II glasses before and after polymerization under 48%RH. Fig. 5.7 shows AFM height images of the (a) as- received Superamine glass surface and (b) same surface after 2 hours of polymerization under 48%RH. The as-received Superamine glass is a smooth surface with no major topographical features. After 2 hours of polymerization a large number of 60nm polymer buds are deposited on the surface. This is the AFM image of the same sample depicted in
Fig. 5.4 (a).The polymer buds are mostly the same size except in a few areas where they are larger (~100nm). The 3D view of the height image illustrates that the polymer buds on average measure about 100nm in height.
Fig. 5.8 shows the AFM height images of the (a) as-received Schott Nexterion glass surface and (b) same surface after 10 hours of polymerization under 48%RH. As in the case of the Superamine glass surface the as-received Schott glass is a smooth surface with absolutely no apparent features. The height images after 2 hours and 5 hours of polymerization under 48%RH (not shown) does not show any polymer deposition and appears identical to the as-received surface (Fig. 5.8 (a)). Only after 10 hours of
polymerization (Fig. 5.8 (b)) does any polymer deposition occur on this surface. The
sparse polymer depositions, seen as white dots in the AFM image, are of varying sizes.
103
Figure 5.7: AFM height images of (a) as-received Superamine glass & (b) Superamine glass after 2 hours of polymerization under 48%RH
Figure 5.8: AFM height images of (a) as-received Schott glass & (b) Schott glass after 10 hours of polymerization under 48%RH
104 For the case of the Corning GAPS II glass surface, (Fig. 5.9), the as-received
substrates shows a large population of evenly dispersed “silane islands” 8 on the surface.
The diameter of these islands is 50nm or smaller and their height is roughly 5nm. After 2 hours of polymerization under 48%RH this glass substrate does not show any polymer deposition and the AFM image (not shown) appears identical to the as-received surface image. At 5 hours of polymerization under 48%RH, the surface now has sparse polymer deposition on it that appears as white dots in the AFM height image of the surface
(Fig. 5.9 (b)). These polymer deposits as in the case with those on the Schott glass surface also vary in size.
Figure 5.9: AFM height images of (a) as-received Corning glass & (b) Corning glass after 5 hours of polymerization under 48%RH
Fig. 5.10 (a) shows the AFM height image of the as-received Superclean substrate.
In order to understand the mechanism of polymer deposition on these substrates the AFM
105 image of this surface is compared to a SEM image of short time polymerization on the same substrate (Fig. 5.10 (b)). The SEM image depicts short polymer fibers growing from the surface. The AFM image reveals certain ‘star-shaped’ morphologies on the surface,
(two of them are encircled by white dotted lines). These structures present over the entire surface have diameters of about 500nm. Comparing with the SEM image of the fibers it is observed that the number density and diameter of these star-shaped structures correspond and match up well with the density and diameters of the ‘footprint’ or base of the polymer fibers (also encircled by white dotted lines in the SEM image). An additional observation from the SEM image is the presence of small hair-like polymer fibers sprouting from the base of the fibers.
Figure 5.10 : AFM height image (a) on as received Superclean substrates & (b) SEM image of at the same magnification of short polymer fiber growth on the same substrate.
AFM images were also taken on the Superclean* glass substrate (Fig. 5.11 (a)) that did not result in any polymer nanofiber formation. The height image shows a smooth surface devoid of major features. Comparing with the SEM image of the polymer
106 deposited on this substrate after 10 hours of polymerization (Fig. 5.11 (b), shown
previously in Section 5.3.1 ), there appears to be no structures evident on the AFM
height image that corresponds to the thin polymer deposit found on this substrate.
Figure 5.11 :(a) AFM height image on Superclean* and (b) SEM image at the same magnification of polymer deposited on surface after 10hours of polymerization
5.3.5 Elemental analysis by XPS of commercial glass slides
XPS survey scans were used to estimate the elemental composition of the surface of the glass substrates. Tab. 5.1 shows the atom % compositions of the various commercial glass slides used in this study. The compositions are for the as-received glasses except Superclean* which was piranha etched for 1hour.
107
Table 5.1: Surface Elemental Atomic Compositions of Commercial glass slides by XPS
Atomic % Na Sn Cu O N Ca Mg K C Si Al
Superamine 4.38 - - 55.39 1.75 1.40 1.16 - 13.63 21.11 1.19
Schott 0.50 - - 57.25 2.15 - - 1.08 12.73 23.50 1.27
Corning 0.61 - - 59.91 1.12 1.06 0.19 - 8.24 22.15 5.14
Superfrost 2.89 - - 53.21 1.69 1.55 1.20 0.29 15.22 22.87 1.09
Superclean 2.88 0.44 0.15 53.43 - 1.52 0.97 - 16.52 24.08 -
Superclean* 1.10 - - 64.85 - 1.27 0.89 - 4.55 26.85 0.48
The compositions of all the glasses except the Schott Nexterion glass are indicative of a soda lime glass constitution. The nitrogen content on these silane modified glass substrates, range from 1.12% on the Corning glass to 2.15% on Schott glass. A noteworthy observation is the carbon content for the as-received Superclean slides
(16.5%) which is significantly reduced after 1 hour of soaking in a piranha etching solution to 4.55% (Superclean*).
5.3.6 Long-time fuming on (Lab-prepared) silane modified glass
Using Superclean* slides as substrates, APS, AAS, DETA, PTS, MTMS and HDF were coated on glass and subjected to long time fuming conditions at 95%RH. The results from these polymerization experiments are presented by SEM images representative of the entire glass surface in Fig. 5.12 . The scale bars for all SEM images are 1 µm with inset images scale bars set at 100nm. APS, AAS and DETA on
108 Superclean* all result in similar polymer nanofiber formation. The fibers have diameters of ~50nm and appear to be in the initial stages of formation of the dense mass of polymer nanofiber network. The fibers on the glasses coated with PTS, MTMS and HDF also show nanofibrous polymer, however the number density of fibers is reduced as compared to the aminosilanes, in spite of them being subjected to the same time of polymerization
Even within these non-aminosilanes the number density appears to decrease in the order
PTS > MTMS > HDF. The diameter of the fibers still remains constant at roughly 50nm
109
Figure 5.12 : SEM Image results of long-time fuming on (a)APS (b)AAS (c)DETA (d)PTS (e)MTMS & (f)HDF all on Superclean*. Scale bars = 1 µm. Inset Scale bars = 100nm
110 Tab. 5.2 presents the surface atom % elemental compositions of the lab-prepared
silane modified glass surfaces. The N-content of the three aminosilanes as expected
increases in the order APS < AAS < DETA for the mono-, di and triaminosilanes,
respectively. The N-content for the APS coated Superclean* is 2.3% which is greater
than the value expected for a APS monolayer (~1.5%) as observed by Metwalli et
al. 9Also the fact that the C-content for APS treated glass is greater than AAS and the C-
content for MTMS treated glass is greater than PTS is suggestive of a multilayer silane
formation on the substrates. The presence of Sn on the surface for some of the coated
slides comes from the underlying Superclean substrate that inherently has some Sn
present on it (Tab. 5.1 )
Table 5.2: Surface Elemental Atomic Compositions of silane modified glass substrates
Atomic % Na O Sn F N Ca Mg K C Cl Si Al APS on 1.17 49.42 - - 2.31 0.86 0.76 19.24 0.32 25.39 0.54 Superclean* AAS on 1.34 51.00 - - 2.68 0.86 0.56 - 16.72 0.30 25.89 0.66 Superclean* DETA on 1.19 47.25 - - 3.84 0.77 0.72 - 21.66 0.32 23.83 0.43 Superclean* MTMS on 2.00 55.26 0.3 - - 1.09 0.70 0.18 14.65 - 25.04 0.81 Superclean* PTS on 1.55 55.88 0.28 - - 1.15 0.61 0.23 13.73 - 26.40 - Superclean* HDF on 1.49 54.44 0.80 2.00 1.01 0.92 1.15 0.39 12.63 - 24.70 0.46 Superclean*
111 5.4 Discussion
5.4.1 Polymerization on commercial glass slides
Taking into consideration only the four coated commercial glass slides, the large number density of fibers on the Superamine and Superfrost substrates as compared to the
Schott and Corning glass substrates, (Fig. 5.1 (a)-(d)) for the same time of fuming under
the same conditions suggest that there were fewer sites for initiation for polymer
nanofibers on the Schott and Corning glass. The coating on these slides is known to be an
amine functional moiety as these slides are either used for microarray analysis 10,11 or to
increase tissue adhesion on microscope slides. 12 This coating chemistry is also evident in
the elemental analysis of the surface of the substrates by XPS indicating presence of
nitrogen on all these surfaces (Tab. 5.1 ). However the % of nitrogen does not correspond to the number density of PECA fibers observed after fuming. The Superamine and
Superfrost substrates have nitrogen contents of 1.75 and 1.69 %, respectively has a large density of fibers while Corning and Schott glass substrates with 1.12 and 2.15% nitrogen, respectively yield relatively few fibers. Also for the case of the as-received Superamine glass substrate, with no detectable trace of nitrogen on the surface, yields dense mass of polymer nanofibers on fuming. This suggests that the initiation of polymer nanofibers on these substrates is independent of the nitrogen %. It is significant to take note of the fact that the carbon content of the as-received slides (16.52%) is considerably higher for what is to be expected of a clean glass surface. Hydrocarbon contamination from typical laboratory environments 13 is the most likely source of this carbonaceous deposit as these
112 slides were used a few months after procurement. Once these slides are piranha etched the carbon content drops down to 4.55% (on the Superclean*) indicating removal of the majority of the carbonaceous contamination. Upon fuming now, the Superclean* substrate does not initiate polymerization of PECA nanofibers implying the importance of the contamination for nanofiber formation in this case.
The requirement of high relative humidity for PECA nanofiber formation is known 2 and has also been reconfirmed in this study with the short time fuming
experiments on Superamine glass under different RHs. Water is required for initiation of
polymerization and for the same 2 hours of fuming, the polymer deposited on the
substrates goes from few polymer buds at 18%RH, to larger number of buds at48%RH, to
short fiber stubs at68%, to long fibers at 81%Rh to even longer fibers at 94%RH. This
indicates that initiation of polymer nanofibers occurred at, at least 68%RH but it
progressed even faster at higher relative humidities. Previous studies 4 have suggested that condensation of tiny water droplets on substrates at high relative humidities serve as initiation islands that template the formation of PECA nanofibers when the substrates are exposed to ECA monomer vapor. This scheme of water condensing on the substrate first and providing a locale for ECA initiation does not agree with two observations made in the present study. Firstly, during imaging of water condensation on the Superamine substrate in the ESEM (Fig. 5.6), the nucleation of water droplets on the surface was
observed to be progressive. Such nucleation behavior results in droplets of different sizes
being present on the surface at any time, ranging from few hundred nanometers to greater
than 2 microns. If in fact, ECA initiation were to occur on such a surface with each
droplet or pool of water serving as an initiation island, it would result in fibers with
113 diameters of that same size range sprouting from the surface. This however is not observed. In Fig. 5.2 (c), (d) & (e) and Fig. 5.3, the PECA nanofibers all appear to have
relatively consistent diameters between 100-200nm. Secondly, for the case of fuming on
the Superclean* slide that was etched with piranha solution for 1hour to remove
hydrocarbon contamination 14 , there occurred very small amounts of polymer film deposition upon fuming for long times. The piranha etching step by cleaning the glass surface would have essentially made the substrate hydrophilic 6 and hence very fitting for water adsorption and condensation. Consequently, there should have been large amounts of PECA polymer film deposition on this substrate. However such polymer film deposition did not occur on this substrate. Instead thin polymer film, almost negligible compared to the dense mass on polymer nanofibers deposited on other glass substrates was observed. These specifics suggest a different course for polymer nanofiber initiation on substrates.
A review of the SEM images for fuming on the Superamine substrates under
48%RH and 68%RH for different times, (Fig. 5.4 & Fig. 5.5 ) advocate an alternate hypothesis for polymer nanofiber initiation. The premise is that the adsorption of ECA monomer vapor on the surface occurs first after which it is subsequently initiated by water vapor to deposit the polymer. At 48%RH the concentration of water vapor is low and hence after ECA adsorption on the surface, it is only slowly converted and deposited as polymer buds. The volume of adsorbed ECA monomer that interacts with water must effectively determine the size of the eventual polymer bud (~60nm). As time progresses more ECA is initiated and more polymer buds are deposited, however the polymer buds never propagate into fibers at this humidity. This implies that the initiation of ECA at this
114 humidity occurs slow enough that active living polymer chains are terminated and no significant concentration of living chains exist to propagate polymerization further.
Hence the size of the polymer buds does not change appreciably but the number density changes with time. At 68%RH, on the other hand, initiation occurs at a faster pace as more initiator (water vapor) is present. In this case after initiation the living chains of the deposited polymer buds are able to continue propagating and grow into a fiber that extends even longer as time proceeds. This phenomenon of localized initiation sites leading to the formation of polymer nanofibers conforms to the discussion presented in
Chapter 4 regarding the explanation for nanofiber formation. This hypothesis is also supported by the observations for the 2 hours fuming experiments at 18%RH, 81%RH &
94%RH (Fig. 5.2 (a), (d) & (e)). At the lowest humidity extremely few polymer buds are
deposited due to insufficient initiator being present and at the higher humidities for the
same time of fuming, propagation has ensued to result in long polymer fibers.
In accordance with this hypothesis for the same conditions of relative humidity
and time of fuming the number density of polymer initiation sites on a surface would
depend on the wettability of the ECA monomer on that surface. This conjecture implies
that a surface with low wettability for ECA would yield a lower number density of
polymer initiation sites. This is verified by the results from the Superamine, Schott and
Corning glass substrates. The AFM images of the same moderate humidity (48%RH)
fuming experiments on Superamine, Schott and Corning (Fig. 5.7, Fig. 5.8 & Fig. 5.9 )
revealed that PECA deposition occurred on the Superamine substrate the quickest (within
2hours) followed by deposition on the Corning substrate (within 5 hours) and lastly on
the Schott glass substrate (within 10hours). This observation that the ECA preferentially
115 wets the substrates in the order Superamine > Corning > Schott, in view of the wettability supposition then not only explains the differences in number density of polymer fiber growth evident on these substrates at long times (Fig. 5.1), but also the differences in extent of polymer growth. In other words a substrate with good ECA wettability has a greater amount of polymer nanofiber deposition.
The AFM image of the as-received Superclean substrate and corresponding SEM image of polymer fibers on Superclean (Fig. 5.10 ), also supports the ECA wettability reasoning for fiber growth. The star-shaped structures seen in the AFM image that are removed after piranha etching (Superclean* AFM image Fig. 5.11 (a)), most likely are hydrocarbon contamination patches. These patches are usually hydrophobic and would prefer to adsorb a hydrophobic material, in this case, ECA monomer. 15 This adsorbed
ECA on the hydrophobic patches then serve as initiation sites for PECA fiber formation
as evident in the SEM image. Once cleaned (Superclean*), the substrate is devoid of the
hydrophobic patches and no fiber formation is observed as there occurs no ECA
adsorption on the now hydrophilic surface. The little polymer film formation observed on
the Superclean* substrate is possibly due to superficial polymerization of ECA on top of
an expected condensed water layer on the hydrophilic Superclean* surface.
The occurrence of Y-shaped branching on fibers and the formation of smaller
diameter fibers like the ones seen in the SEM image for polymer nanofibers grown on
Superfrost substrates is not completely understood. A likely explanation of these
structures is that secondary initiation of polymer nanofibers occurs on existing fibers after
termination or transfer of active polymer chains. Also some fibers that have larger
116 diameters (>100nm) might be the result of continued polymerization wherein now the fibers now not only increase in axially but also radially with polymerization time.
5.4.2 Polymerization on lab-prepared silane modified glass surfaces
Gauging from the elemental composition of the lab-prepared silane coated slides the treatment of various silanes exceeded monolayer coverage on the surface, thereby definitively altering their surface properties. This fact is evident in the results obtained from the long-time fuming experiments on these modified substrates (Fig. 5.12 ). For the same conditions and time of fuming the number density of fiber growth on these substrates decreases in the order of Aminosilanes > PTS > MTMS> HDF. This variation can again be explained by the ECA wettability criteria stated above. A review of the expected critical surface tension values ( γc) for soda lime glass treated with these particular silanes obtained from past literature 6,16 reveal that these values decrease in the
same order. The values for γc (in mN/m) for APS, AAS, PTS, MTMS, and HDF are 35,
33.5, 28.5, 22.5 and 14.9, respectively. To obtain good wettability of a liquid on a surface, the surface tension of the liquid must be below the critical surface tension of the surface. Hence comparing these values with the surface tension value for ECA monomer 15 , which is 34.32 mN/m, it is clear that in going from the aminosilanes to alkylsilanes and finally fluorosilane the ECA wettability and therefore adsorption on the surface decreases. This would cause fewer polymer initiation sites on the low γc surfaces, which is the observed case. These results reconfirm the nanofiber formation hypothesis of
ECA wetting the surface first, which is then initiated and grows to form polymer
117 nanofibers under sufficient humidity. The diameter of the fibers however, remains unchanged (~50nm) as it is probably determined by the interaction of water and adsorbed
ECA which remains the same on all the surfaces.
5.5 Conclusions
The study in this chapter successfully demonstrates template-less growth of
PECA nanofibers on silane modified glass surfaces. The premise established for the mechanism by which polymer nanofiber growth initiated on these surfaces entails the surface to be conducive for ECA monomer wettability. Once adsorbed on the surface the
ECA monomer is initiated by water to form polymer deposits. For low concentration of initiator (low to moderate RH) only polymer buds were deposited where as in presence of sufficient initiator concentration (moderate to high RH) polymer nanofibers were observed. When the substrate was unfavorable for ECA wetting, with low γc the rate of initiation of polymerization was relatively slower and the number density for polymer nanofibers was also low. The diameter of the polymer nanofibers however, does not change appreciably owing to the consistent interaction between ECA and water that must occur on any surface for initiation. Based on the findings in this study it is reasonable to envision a protocol of controllably applying silanes on substrates where polymer nanofiber growth is required with the choice of the silane coating determining the number density of fiber growth. Additionally, ECA polymerization can continue as long as there is monomer present (living polymerization discussed in the next chapter), facilitating means for bulk synthesis of these nanofibers. Such control of placement during
118 fabrication and bulk synthesis of nanofibers would prove integral to the extension of this technique further.
5.6 References
(1) Huang, Z. M.; Zhang, Y. Z.; Kotaki, M.; Ramakrishna, S. Compos Sci Technol 2003 , 63 , 2223-2253. (2) Mankidy, P. J.; Ramakrishnan, R. B.; Foley, H. C. Chem Commun 2006 , 1139- 1141. (3) Smith-Johannsen, R. I. Science 1971 , 171 , 1246-1247. (4) Doiphode, S. V.; Reneker, D. H.; Chase, G. G. Polymer 2006 , 47 , 4328-4332. (5) Wagner, R. S.; Ellis, W. C. Appl Phys Lett 1964 , 4, 89-91. (6) Plueddemann, E. P. Silane Coupling Agents , 2nd ed.; Plenum Press: New York, 1991. (7) Eromosele, I. C.; Pepper, D. C.; Ryan, B. Makromol Chem 1989 , 190 , 1613- 1622. (8) Turrion, S. G.; Olmos, D.; Gonzalez-Benito, J. Polym Test 2005 , 24 , 301-308. (9) Metwalli, E.; Haines, D.; Becker, O.; Conzone, S.; Pantano, C. G. J Colloid Interf Sci 2006 , 298 , 825-831. (10) Stears, R. L.; Martinsky, T.; Schena, M. Nature Medicine 2003 , 9, 140-145. (11) Tech.Information. http://www.us.schott.com/nexterion/english/products/coated_substrates/slide_a/te chnical_information.html , Schott North America, 2007. (12) Tech.Information. http://www.eriesci.com/microscope/micro_slides.aspx?id=6 , Erie Scientific Company. (13) Smith, G. C. J Electron Spectrosc 2005 , 148 , 21-28. (14) Shirai, K.; Yoshida, Y.; Nakayama, Y.; Fujitani, M.; Shintani, H.; Wakasa, K.; Okazaki, M.; Snauwaert, J.; Van Meerbeek, B. J Biomed Mater Res 2000 , 53 , 204-210. (15) Leonard, F.; Kulkarni, R. K.; Brandes, G.; Nelson, J.; Cameron, J. J. J Appl Polym Sci 1966 , 10 , 259-272. (16) Kobayashi, H. Makromol Chem 1993 , 194 , 259-267.
Chapter 6
Controlling Molecular weight of PECA Nanofibers during Polymerization
6.1 Introduction
The successful demonstration of template-less growth of PECA nanofibers, the study of their formation and growth on different substrates, illustrated in previous chapters provides the insight to begin exploiting this attractive and novel approach to synthesize fibers to impart user-tuned properties or characteristics to the fibers. In this chapter and in Chapter 7 such manipulation of the nanofibers is demonstrated during polymerization.
The unique benefit this fabrication technique offers is that the polymer is deposited as a nanofiber during vapor phase polymerization of ECA vapor. This added to the fact that ECA polymerization is an anionic polymerization with an inclination for a
‘living’ nature 1,2 facilitates the potential for controlling the molecular weight of polymer nanofibers formed. A living polymer by definition is one where the propagating species does not undergo chain transfer or termination steps 3, hence in the presence of monomer
and with the dearth of external termination agents the polymer chain will continue to
propagate, thereby increasing its molecular weight. In that sense, ECA polymerization is
not a true living polymer as the propagating carbanion does encounter chain transfer steps
as discussed in Chapter 4. Nevertheless, in the absence of termination agents this
polymerization does approach living status. 4 Controlling the molecular weight of the
120 nanofibers would translate into controlling other properties of the fibers that are derived from molecular weight, for example, thermal properties such as glass transition temperature, mechanical properties such as tensile strength 5 and even controlling rates of
de-polymerization (important for drug delivery applications). 6 For a clear understanding of how molecular weight of the ECA nanofibers can be controlled, a brief review of the steps that occur during ECA anionic polymerization follows 7:
Initiation: Eq. (6.1)
− − − I + M → P Ri = ki[I ][ M ] (Eq. 6.1)
Propagation: Eq. (6.2)
− − − Pn + M → Pn+1 RP = kP[M ][ Pn ] (Eq. 6.2)
Termination: Eq. (6.3)
− 0 − − Pn + HA → Pn + A RT = kT [HA ][ Pn ] (Eq. 6.3)
Chain transfer: Eq. (6.4)
− 0 − − Pn + M → Pn + P RCT = kCT [M ][ Pn ] (Eq. 6.4)
− − 0 Here, I , M, Pn , n, Pn , HA , R’s and k’s are anionic initiator, monomer, active polymer chain, number of monomer units in a polymer chain, terminated polymer chain, acidic termination agent, rates and rate constants, respectively. The number average degree of
polymerization, DP n , which by definition is the average number of monomer molecules
contained per polymer molecule is essentially representative of the number average
121 molecular weight, Mn, of the polymer. DP n is given by the ratio of all the chain
propagating steps to the sum of the chain transfer and termination steps, Eq. (6.5).
kP[M ] DP n = (Eq. 6.5) kT [HA ]+ kCT [M ]
Altering DP n involves changing the relative ratio of components in the above
equation. For example, keeping the monomer concentration the same, to increase DP n , the concentration of the acidic species, [ HA ], would have to be reduced or the reaction would have to be carried out a lower temperature to suppress chain transfer. For PECA nanofiber formation via vapor phase polymerization, the monomer concentration is decided by the vapor pressure of ECA at room temperature and without altering the
temperature, the only route to increase DP n is by reducing the termination agents for
ECA polymerization.
Water as shown in previous studies and this work is a known initiator for ECA polymerization, however it is also known as a suitable carbanion killer for the propagating polymer 1,7 and therefore its removal during the propagation steps is warranted. For efficient template-less formation of PECA nanofibers it has been shown to be imperative that the fuming (polymerization) process occur under at least ~70%RH or higher. The high RH is required for initiation of PECA; however its presence during
subsequent polymerization would impede the propagation steps yielding low DP n .values.
By virtue of the protocol for fuming in the batch set-up apparatus high water vapor concentration is present during the entire fuming process. Hence there is a requirement for an alternate set-up and protocol that separates the initiation and propagation steps of
122 PECA nanofiber polymerization. The process would then involve 2 stages; the first under high humidity to cause efficient initiation of PECA fibers and the second, a low humidity and high monomer concentration stage to limit termination steps and promote propagation
In this chapter the semi-continuous flow set-up apparatus described in Chapter 2 is used to perform the 2-stage polymerization of PECA nanofibers to produce fibers with different molecular weights.
6.2 Experimental
General descriptions of the materials and methods employed in this work are discussed in Chapter 2. Specific details follow
6.2.1 Materials
Superfrost Gold Plus microslides were used as substrates for all the experiments in this study. ECA monomer was obtained from Sirchie Fingerprint Lab Inc. and used as received. THF used as a solvent for GPC was HPLC grade obtained from VWR. UHP
Argon was used as the carrier gas in the flow set-up.
123 6.2.2 Methods
Description of flow set-up fuming protocol : The semi-continuous flow set-up shown in Fig. 2.3 was used for the 2-stage polymerization of PECA nanofibers. The general procedure for this is illustrated in the flowchart diagram in Fig. 6.1
Figure 6.1: Protocol for 2-Stage polymerization in Flow set-up
The 1 st stage was consistent for all experiments and involved humidification for a
sufficiently long time to purge all the lines in the manifold with moisture loaded Argon.
This was followed by PECA nanofiber initiation under 80%RH for a short time. The 2 nd stage involved switching the flows so as to reduce the relative humidity inside the
124 chamber to different values (for different experiments) while always maintaining the same monomer concentration (55% relative to ECA saturation pressure) in the vapor phase. That monomer mole fraction was estimated to be ~ 23.1 ×10 −4 based on the saturation vapor pressure data 8 for ECA at room temperature, ~0.17 torr @ 24°C. The details of the flow conditions to achieve six different relative humidities during the 2nd stage are given in Tab. 6.1 .The 2nd stage was continued for 5 days to allow sufficient
PECA nanofiber deposition on the substrate for injection into the GPC column.
Table 6.1: Flow settings to achieve different RH conditions in 2 nd stage of fuming process Relative Line 1 Flow Line 2 Flow Line 3 Flow Total Flow humidity (%) (sccm) (sccm) (sccm) (sccm) 45 - 47.64 57.33 104.97
40 5.22 41.70 57.33 104.25
35 10.49 36.52 57.33 104.34
30 15.66 31.28 57.33 104.27
25 20.85 26.06 57.33 104.24
20 26.07 20.85 57.33 104.25 Line 1 is pure argon; Line 2 & 3 are Ar saturated with water and ECA vapor respectively, For more details see chapter 2.
GPC Sample Preparation : For injecting PECA nanofiber samples into the GPC column, THF was used to dissolve the polymer deposited on the substrates and the resultant washed polymer solution was retained. This solution was then dried under flowing air and the precipitated polymer mass was weighed. The mass loss of the substrates before and after dissolving the PECA nanofibers was also measured to tally with the amount of precipitated polymer. The polymer was then dissolved in a requisite
125 amount of THF to give a concentration of approximately 1.5 mg of polymer/ml that was
injected into the column using the GPC method described in Chapter 2. Three injections
were made for each polymer sample and the average values with their standard deviations
are reported. Molecular weight estimates were made with reference to a universal
calibration curve using Mark-Houwink parameters of K = 0.00039 dl/g and α = 0.59 for
PECA dissolved in THF at 25۠ °C. 9
IR and SEM Characterization : ATR-IR was carried out on the different molecular weight PECA nanofiber samples deposited on the glass substrates to reveal any differences in their chemical signature and SEM was used to investigate their morphology.
6.3 Results
6.3.1 SEM investigation of 2-stage PECA nanofibers
The results after the 1 st stage of the 2-stage fuming process are shown in Fig. 6.2 at two different magnifications. The short time fuming of the substrates under ~80%RH yields short nanofiber stubs similar to those observed previously during short time fuming experiments in Chapter 5. The diameters of the fibers are in the 50-100nm range with lengths of the order of few hundred nanometers.
126
Figure 6.2: SEM images of PECA nanofibers after 1 st stage during 2-stage fuming at (a) 10K magnification & (b) 40K magnification
Fig. 6.3 and Fig. 6.4 are SEM images of PECA nanofibers grown under six different
humidities during the 2 nd stage of the 2-stage fuming process. Each figure depicts the
fibers at two magnifications. Comparing images, the fibers appear similar to each other
irrespective of the relative humidity they are subjected to. The fibers are long (greater
than at least a few tens of microns) and appear to have fewer Y-shaped branching
networks compared to the nanofibers obtained during batch set-up fuming under high
humidity in Chapter 4 (Fig. 5.1 ). In some instances the fibers are packed close together
(20%RH sample Fig. 6.3 (a) & (b) and 45%RH sample Fig. 6.4 (e) & (f)) and appear
almost fused to one another. The diameters of the fibers from all these experiments
appear consistent and fall within the 300-400nm range which is greater than those at the
end of the 1 st stage (Fig. 6.2). Consequently a couple of short-time 2-stage fuming experiments were carried out to study this effect of increase in diameter in the 2 nd stage.
The SEM images of fibers after 1 and 2 hours of fuming in the 2 nd stage under 45%RH are shown in Fig. 6.5
127
Figure 6.3: PECA nanofibers grown under (a)&(b) 20%RH; (c)&(d) 25%RH; (e)&(f) 30%RH during the 2 nd stage of the fuming process
128
Figure 6.4: PECA nanofibers grown under (a)&(b) 35%RH; (c)&(d) 40%RH; (e)&(f) 45%RH during the 2 nd stage of the fuming process
129
Figure 6.5: PECA nanofibers after (a) 1 hour & (b) 2 hours of 2 nd stage fuming under 45%RH. Scale bars are 100nm in each image.
The images are at the same 50K magnification. The fiber diameters after 1 hour of
fuming are roughly 100nm or lesser, whereas after 2 hours the diameters have increased
to lie in the range of 100-150 nm.
6.3.2 GPC estimates of Molecular weight of PECA nanofibers
Fig. 6.6 illustrates two representative traces obtained after injection of the PECA
nanofiber samples into the GPC column. The traces are for the conditions of (a) 20 and
(b) 45 %RH in the 2 nd stage of the fuming process. The peak molecular weight is shown for each case. The molecular weight distributions are unimodal with the distribution for the 45 %RH sample slightly wider than the 20% RH sample’s distribution.
130
Figure 6.6: GPC traces for PECA nanofibers created under (a) 20% RH and (b) 45 % RH during the 2 nd stage fuming process.
The variation of molecular weight of the PECA nanofibers with respect to the
relative humidity in the 2 nd stage of the fuming process is illustrated by the plot in
Fig. 6.7 of the number average molecular weight, Mn, and peak molecular weight, Mp,
versus estimated humidity. Mp values, which represent the molecular weight of the greatest population of polymer chains in the sample, decrease consistently with increase in humidity, ranging from 270K at 20% RH to approximately 209K at 45% RH. Mn values, that represent the average molecular weight of a polymer chain in the sample, also decrease with increasing humidity, varying from 198K to 147K.
131
3.0e+5
2.8e+5 Mn - Number Avg. Mol. Wt. Mp - Peak Mol. wt.
2.6e+5
2.4e+5
2.2e+5
2.0e+5
Molecular Weight Weight Molecular 1.8e+5
1.6e+5
1.4e+5
15 20 25 30 35 40 45 50 nd Relative Humidity (%) during 2 Stage Figure 6.7: Variation of molecular weight of PECA nanofibers with relative humidity in the 2 nd stage of fuming
The error bars in each plot represent standard deviations from the average values of three
injections for each sample.
The variation of the polydispersity index (PDI) of the nanofibers with humidity is
shown in Fig. 6.8. PDI defined as the ratio of the weight average to number average
molecular weights is a measure of the breadth of the molecular weight distribution in the
sample. PDI values for the PECA nanofibers increases constantly from 1.22 to 1.36 in
going from 20% to 45% RH.
132
1.5
1.4
1.3
1.2 Polydispersity Index (PDI) Index Polydispersity 1.1
1.0 20 25 30 35 40 45 nd Relative Humidity (%) during 2 Stage Figure 6.8: Variation of PDI of PECA nanofibers with relative humidity during 2 nd Stage of fuming
6.3.3 IR analysis of different molecular weight PECA nanofibers
Fig. 6.9 depicts the IR spectrum for PECA nanofiber polymer samples created under 20%, 30%, 35% and 40% RH during the 2 nd stage fuming process. The peaks at
1740 & 2250 cm -1due to CN and C=O vibrations and the peaks between 3100 and 2850
cm -1 because of C-H stretching and bending vibration in the methyl and methylene functionalities appear identical in each spectra suggesting no chemical differences between the different PECA molecular weight samples.
133
Figure 6.9: ATR-IR spectrum of different molecular weight PECA nanofibers created under different humidities in the 2 nd stage
These spectra also appear similar to those obtained for different vapor phase polymerized
PECA (film and nanofibers) obtained for Chapter 4.
6.4 Discussion
6.4.1 Diameter of 2-stage PECA Nanofibers
The SEM images of the fibers at the end of the 1 st stage of the fuming process and after the 2 nd stage suggest that the short fiber stubs seen in Fig. 6.2 evidently grow into the long nanofibers seen in Fig. 6.3 & Fig. 6.4. The diameter of the PECA nanofibers changes from ~100nm after 1hour of the 1 st stage to between 300-400nm after 5 days of
134 fuming in the 2 nd stage irrespective of the humidity in the second stage. This change in the diameter of fibers is also witnessed in the short 2 nd stage time fuming experiments
(Fig. 6.5), where the fiber diameter increases to about 150nm after 2 hours of fuming. As the monomer concentration during the 2 nd stage for all the experiments was the same, this change in the diameter of fibers is clearly unrelated to the presence of humidity but is only a function of time of fuming and obviously monomer concentration. The growth of the fibers radially however is rather insignificant compared to the apparent increase along the length. This suggests that the fibers still possess the majority of their propagating polymer chains growing chains at the tip of the fibers. The lack of Y-branching networks on the PECA nanofibers here, compared to those observed in previous chapters can be explained by the reduced concentration of water vapor in the 2 nd stage. The Y-branching
considered to be secondary initiation after termination of an active chain (Eq. (6.3) )
would be more prevalent in a humid atmosphere such as the one in previous studies at
~95% RH versus the 2 nd stage fuming in this study in which case the humidity is < 45%.
This causes the fibers to grow longer without much branching and consequently have higher molecular weight as discussed below.
6.4.2 Controlling Molecular weight of PECA nanofibers
Two aspects of the 2-stage fuming process make the evaluation between the
different molecular weight PECA nanofibers comparable. The first is that for the six
different humidities attained during the 2 nd stage of fuming, the 1 st stage was always the same. This made the starting point for all the experiments the same. The second factor is
135 that the monomer concentration during the 2 nd stage was also kept constant irrespective
of the humidity and the duration of the 2 nd stage fuming was also the same for all experiments. That implies that the propagating PECA chains in all cases were always exposed to the same amount of monomer and for the same duration, leading to the conclusion that the differences in the molecular weight was solely attributable to the different humidities. Hence higher molecular weights ( Mp or Mn) observed for lower
relative humidities was the anticipated result based on the fact that fewer termination
steps due to fewer termination agents (water vapor in this case) would cause longer
polymer chains to form. PDI values decreasing with decreasing humidity is also an
indication of the direct effect of reduced termination during polymerization. A PDI value
of 1.0 corresponds to the theoretical living polymer made with no termination or chain
transfer steps, hence the PDI values reducing from 1.36 to 1.22 signifies approach
towards living polymerization. This is also evident in the narrowing of the unimodal
curve in the GPC trace going from the 45% RH sample to the 20% RH sample seen in
Fig. 6.6.
Applying the insight obtained in the previous section about changes in the
diameter of the fibers with time during the 2 nd stage fuming period, it is reasonable to assume a possibility of controllably creating PECA nanofibers with a stipulated diameter and molecular weight. The predetermined diameter would be achieved by using the appropriate time of fuming in the 2 nd stage and the molecular weight would be achieved
by maintaining the appropriate humidity during that stage.
136 A qualitative analysis of the rates constants of the propagation, termination and
chain transfer steps relative to each other is also possible using the obtained molecular
weight data. Rearranging the form of Eq. (6.5) gives Eq. (6.6)
1 kT [H 2O] kCT = + (Eq. 6.6) DP n k P []M k P
As mentioned earlier the number average molecular weight of the polymer, Mn, is
essentially indicative of the degree of polymerization DP n . Hence /1 DP n ∝ /1 M n .
Representing Eq. (6.6) then, as a plot of /1M n versus [H 2O /[] M ] we obtain Fig. 6.10
7.5e-6
7.0e-6 -1 6.5e-6
6.0e-6
5.5e-6
Equation: Linear 5.0e-6 y = 3.9409E-08x + 3.5209E-06 (Number average wt.) average mol. (Number
4.5e-6
4.0e-6 30 40 50 60 70 80 90
Ratio of water to ECA monomer; [H 2O]/[M]
Figure 6.10 : Plot of 1/ Mn versus [H2O]/[ M] with a linear fit to the data
137 A linear equation fit for the data gives slope and intercept values of 94.3 ×10 −8 and
52.3 ×10 −6 , respectively. From Eq. (6.6) the slope is the ratio of the termination to propagation rate constants ( kT/k P) and the intercept is the ratio of chain transfer and propagation rate constants ( kCT /k P) which yields that kP>> kT or kCT . This is expected for a living anionic polymerization mechanism. Also, taking the ratio of the slope to the intercept gives Eq. (6.7)
k 94.3 ×10 −8 T = = −6 .0 011 (Eq. 6.7) kCT 52.3 ×10
This estimate suggests that the rate of chain transfer is greater that the rate of
termination by about two orders of magnitude. Interestingly, this premise corroborates
the earlier implication from the IR studies of vapor phase polymerized PECA samples in
Chapter 4 about prominent chain transfer steps occurring during polymerization.
6.4.3 Interpretation of IR spectra
Since there are no notable differences in peak intensities and positions in the IR
spectra between the different molecular weight PECA nanofibers and between these and
other previously recorded spectra for vapor phase deposited PECA (Chapter 4), it is
reasonable to assume that the changes in humidity do not affect the expected chain
transfer steps that occur during vapor phase polymerization but only the termination
steps.
138 6.5 Conclusions
The results in this chapter successfully demonstrate tunability of molecular weight of
PECA nanofibers created by the template-less vapor phase polymerization using the semi-continuous flow apparatus. The use of two stages with different conditions in the vapor phase during the fuming process can efficiently separate the initiation and propagation steps of the anionic PECA polymerization so as to exercise control over the extent of termination steps that take place. Using this protocol, PECA nanofibers with a number average molecular weight of ~198K and a PDI of 1.22 were synthesized by maintaining a relative humidity of 20% during the 2nd stage of fuming. At higher
humidities nanofibers with lower molecular weights and higher PDIs resulted, indicating
deviation from a living polymerization mechanism. The consistent change in the
diameters of the fibers from ~100nm to 300-400nm after the course of the 2 nd stage
irrespective of humidity suggests the possibility of creating nanofibers with set diameters
and molecular weights.
The qualitative analysis of the degree of polymerization yielded information on
the relative rates of the polymerization reactions for PECA. As expected the rate of
propagation was several orders of magnitude greater than termination and chain transfer
rates. The rate of chain transfer was at least two orders greater than termination rates
which reconfirms the suggested hypothesis in Chapter 4 based on IR investigation of
vapor phase polymerized PECA nanofibers and film about prominent chain transfer
reactions.
139 6.6 References
(1) Pepper, D. C.; Ryan, B. Makromol Chem 1983 , 184 , 383-394. (2) Coover, H. W.; Dreifus, D. W.; T., O. C. J. Handbook of Adhesives , 3rd ed.; Van Nostrand Reinhold: New York, 1990. (3) Odian, G. G. Principles of polymerization , 3rd ed.; Wiley: New York, 1991. (4) Johnston, D. S.; Pepper, D. C. Macromol Chem Phys 1981 , 182 , 393-406. (5) Sperling, L. H. Introduction to physical polymer science , 3rd ed.; Wiley- Interscience: New York, 2001. (6) Vauthier, C.; Dubernet, C.; Fattal, E.; Pinto-Alphandary, H.; Couvreur, P. Adv Drug Deliver Rev 2003 , 55 , 519-548. (7) Eromosele, I. C.; Pepper, D. C.; Ryan, B. Makromol Chem 1989 , 190 , 1613- 1622. (8) Tech.Information. MSDS: M-Bond 200 Adhesive;Davidson Measurement, 2004. (9) Donnelly, E. F.; Pepper, D. C. Makromol Chem-Rapid 1981 , 2, 439-442.
Chapter 7
Synthesis of Poly (Methyl 2-cyanoacrylate) [PMCA] Nanofibers and Poly (Ethyl 2-cyanoacrylate-co -Methyl 2-cyanoacrylate) [P(ECA- MCA)] Nanofibers
7.1 Introduction
In addition to the impact that polymer nanofibers create by their potential
application in various fields discussed previously, the possibility of modifying already
synthesized polymer nanofibers of one kind with another polymer or moiety would only
broaden their applicability to even more areas. Modification would imply extending the
valuable properties of these nanofibers (high aspect ratio) for creating nanostructures of
materials that would ordinarily not be possible to synthesize using current nanofiber
fabrication approaches.
Recent examples of such modifications include grafting of a polymer on pre-
synthesized polymer nanofibers. Ma et al. 1 demonstrated the attachment of a reactive dye,
Cibacron blue on electrospun polysulfone nanofibers. In order to facilitate this attachment the nanofibers were first grafted with poly (methacrylic acid) by a plasma treatment of the fiber surface followed by exposure to methacrylic acid monomer. Also polypyrrole- poly (ethylene oxide) composite nanofibers were created in two steps by Kim et al. 2 In the first step poly (ethylene oxide) nanofibers implanted with FeCl 3, an oxidant for pyrrole polymerization was produced by electrospinning and in the second step the fibers were exposed to pyrrole vapors to incorporate polypyrrole.
141 Such modifications allow the inclusion of other polymers as a sheath-coating on the core polymer nanofibers, however the template-less approach for fabrication of polymer nanofibers discussed in the previous chapters offers a unique way for incorporating another polymer as a nanofiber. From the results in the previous chapters it is apparent that the growth (i.e. propagation) of the poly (ethyl 2-cyanoacrylate) [PECA] nanofibers occurs preferentially at the tip of the fibers via monomer addition from the vapor phase. That implies that the ‘living’ end of the fibers must have the active ends of the propagating polymer chains, namely the carbanionic ends. This situation of localized growth of the PECA nanofibers can then be exploited to copolymerize other polymers that undergo anionic polymerization by introducing their monomer vapor to the living carbanionic ends of the PECA chains. It could be then envisioned that the new polymer would retain the size of the of the original PECA nanofiber thus creating a polymer nanofiber that is a block copolymer of PECA and the new polymer. Alternatively if the presence of ECA monomer vapor is maintained after introduction of the second monomer vapor the polymer nanofiber would then be a random copolymer of PECA and the new polymer. In this manner the template-less growth of PECA nanofibers can be utilized to generate copolymers combinations with other polymers which can impart interesting properties to the composite nanofibers.
Studies of copolymerization of different polymers with alkyl α-cyanoacrylates
primarily involve radical polymerization mechanism because of low selectivities between
the different monomer types and greater monomer reactivities associated with radical
copolymerization. 3 Oikawa et al. 4 have successfully demonstrated the copolymerization of furan and methyl 2-cyanoacrylate (MCA) using azobisisobutyronitrile (AIBN) as the
142 free radical initiator. Kinsinger et al. 5 concluded that alternating copolymers were formed with MCA and different monomers such as styrene, methyl methacrylate, vinyl acetate and others using AIBN. Also, ethyl 2-cyanoacrylate (ECA) has been copolymerized with ethylene 6, polyethylene glycol 7 and chloroprene 8 via radical copolymerization to impart specific properties such as strength, hydrophilicity and flexibility to the cyanoacrylate polymer.
Ionic (anionic in this case) copolymerizations however are relatively more selective in nature. In general there exist few comonomer pairs for successful anionic copolymerization owing to the wide range of monomer reactivities for anionic copolymerization. 3,9 Maruyama et al. 10 showed the anionic copolymerization of ethyl 2-
cyanoacrylate and methyl vinyl ketone in THF using water as the initiator. The
copolymer formed had increased heat resistance than the PECA homopolymer but
decreased adhesive capabilities. Copolymers of butyl and ethoxyethyl 2-cyanoacrylates
were also prepared by anionic copolymerization to enhance the flexibility of the alkyl
cyanoacrylate. 11 Anionic copolymerization usually proceeds when conditions approach ideal copolymerization, which is when the product of the two monomer reactivities (r 1
3 and r 2) approach unity. In this case the relative reactivities of the monomers are equal towards either carbanionic propagating species. This scenario is facilitated by electron withdrawing substituents on the α-carbon of the monomers involved that would decrease
the electron density on the vinyl bond and resonance stabilize the carbanion. Therefore,
monomers with groups such as –CN, –COOR, and halides would be most suitable for
anionic copolymerization with alkyl 2-cyanoacrylates. This suggests that monomers such
as methyl methacrylate, methyl methacrylic acid, acrylonitrile and even vinylidene
143 chloride could potentially form copolymers with alkyl 2-cyanoacrylates via anionic
mechanism.
In this chapter, as a proof of concept, copolymerization of PECA nanofibers
formed by the template-less vapor phase approach, with poly(methyl 2-cyanoacrylate)
[PMCA] is attempted. ECA and MCA differ by a single –CH 2– in the ester group and
hence should have relatively similar reactivities towards either propagating carbanion.
ATR-IR and GPC are used to determine the formation of a copolymer and SEM is used
for morphological investigation of the copolymer nanofibers formed. In the first section
of the work however, PMCA homopolymer nanofibers formation is demonstrated using
only MCA monomer.
7.2 Experimental Section
General descriptions of the materials and methods employed in this work are
discussed in Chapter 2. Specific details follow
7.2.1 Materials
Ethyl 2-cyanoacrylate monomer used in this study was pure ECA without any
additives and was used as received from Sigma-Aldrich. Pure methyl 2-cyanoacrylate
was obtained and used as received from Polysciences Inc. ( Note : It was essential to use
pure monomer reagents for positive identification of peaks in the IR spectra for
homopolymer and copolymers of ECA and MCA. Common sources for ECA
144 (Superglue®) contain small quantities of poly (methylmethacrylate) to increase viscosity
and other inhibitors for polymerization that are noticeable in the IR spectra of the bulk
(liquid phase polymerized) samples and can account for reasons for misinterpretations.)
For all the experiments conducted in this study the substrates used was the Superfrost
Gold Plus microslides obtained from Erie Scientific Inc.
7.2.2 Methods
Fuming in Batch set-up : To confirm the formation of PMCA nanofibers, vapor phase polymerization was first carried out of only MCA monomer vapor on the
Superfrost substrates in the batch set-up polymerization unit. Fuming was carried out under 95% RH for 12 hours. Copolymerization of ECA and MCA was also carried out in the batch set-up under high humidity for 12 hours. In this case after initial humidification at 95%RH both ECA and MCA monomer were simultaneously introduced into the chamber by dropping both liquid monomers together into the container dish inside the chamber.
Fuming in Flow set-up: In order to collect sufficient quantity of the ECA & MCA
copolymer samples (P(ECA-MCA)) for injection into the GPC, polymerization was
carried out using the flow setup by exposing ECA and MCA vapor simultaneously to the
substrates. This was achieved by an additional Ar line upstream from the tubular
chamber, saturated with MCA vapor by bubbling through a liquid MCA monomer
reservoir. Random copolymerization of ECA and MCA was attempted. Here after
humidification, ECA and MCA vapor were introduced simultaneously into the chamber
145 to initiate polymerization. The ratio of flows of ECA to MCA was set at 1:2 and the effective humidity in the chamber was maintained at 45%RH for 1 hour. Subsequently the moisture laden stream was switched off and polymerization was continued under only monomer vapors (still in the 1:2 ratio) for >5days to collect sufficient sample for GPC.
Copolymerization in the liquid phase : To compare and contrast with the copolymer sample made by vapor phase polymerization, copolymerization was also carried out in tetrahydrofuran (THF) using equal volumes of ECA and MCA monomer
(1ml each in 10ml THF). The monomers were injected simultaneously into a vial containing THF and sonicated for 30 min. As this polymerization was carried out under ambient conditions, trace amounts of water dissolved in the THF was sufficient to initiate polymerization. After 30 minutes of sonication the vial was left at room temperature for
12 hours before injection into the GPC. At that same time a small amount of the copolymer was dried on a clean Si wafer for IR analysis.
Characterization : SEM was used as before for visual investigation of morphology of the polymer nanofibers produced. ATR-IR was used to determine chemical compositions of the polymer samples. For the vapor phase polymerized samples, the white polymer deposit on the glass substrate was scraped together before bringing into contact with the ATR crystal. For the liquid phase polymerized bulk samples, this was not necessary as the polymer film dried on the Si wafer was sufficiently thick for the incident IR beam to sample only the polymer layer and not the underlying silicon. IR spectral interpretation was made with reference to characteristic IR absorption of different functional groups. 12,13 Molecular weight estimations were made by injecting
146 polymer solutions in THF (~1.5 mg/ml) to the GPC with reference to a universal
polystyrene calibration curve.
7.3 Results
7.3.1 Fuming of MCA vapor in Batch set-up
Fig. 7.1 are SEM images of the polymer deposited on the Superfrost substrate after 12 hours of fuming by MCA vapor under 95%RH in the batch chamber. Fig. 7.1 (a) shows PMCA nanofibers on the substrate which are similar to PECA nanofibers described in previous chapters. Fig. 7.1 (b) depicts the same fibers at higher
magnification. The scale bar in this image is 100nm. Fibers of different sizes are visible
with diameters ranging between 50-100nm. The length of the fibers as before is difficult
to estimate due to entanglement of the fibers but appear to at least be greater than tens of
microns.
147
Figure 7.1: SEM image of PMCA nanofibers after 12 hours of batch fuming; (a) at 5K magnification and (b) at 40K magnification.
Fig. 7.2 (a) illustrates the IR spectrum obtained by ATR-IR on the PMCA nanofibers. To distinguish this spectrum, Fig. 7.2 (b) illustrates the IR spectrum obtained
by ATR-IR on PECA nanofibers from previous chapters for comparison. Similarities
between the two spectra include the small peak at 2250cm -1signifying the presence of –
CN functionalities in both and the major peak at 1740cm -1 due to C=O stretching vibrations. Differences between the two owing to the extra –CH 2– in the ester group of the PECA sample are clearly evident in two regions of the spectra, represented in Fig. 7.3 and Fig. 7.4. In the region between 3100-2800cm -1, (Fig. 7.3 ) methyl and methylene asymmetric and symmetric C-H stretches are responsible for the four peaks in the PECA spectrum. In the case of the PMCA only the asymmetric and symmetric C-H stretches from the methyl group are present, however, these peaks are shifted to 2960 and 2850cm -
1 respectively. This shift is due to electronic effects of the methyl group being directly
attached to the oxygen atom in PMCA.
148
Figure 7.2: ATR-IR spectra obtained from (a) PMCA nanofibers and (b) from PECA nanofibers
Figure 7.3: IR Spectral region between 3100-2800cm -1of PMCA & PECA nanofibers 149
Figure 7.4: IR Spectral region between 1900-1300cm -1of PMCA & PECA nanofibers
For the region between 1900-1300cm -1, (Fig. 7.4 ), the shift in the location of the symmetric C-H bending vibration of the methyl group (commonly known as the CH 3 umbrella mode) distinguishes the two spectra. For PECA this vibration occurs at 1375cm -
1 whereas for the case of PMCA, the attachment of methyl directly to the oxygen atom causes this peak to shift to ~1440cm -1 making the peak at 1435cm -1 more pronounced.
Another observation was that the peak at ~1620cm -1 due to C=C stretching was more
pronounced in the PMCA. This peak in the polymer as discussed in Chapter 6 arises as a
result of chain transfer steps via hydride elimination during polymerization in the vapor
phase.
150 7.3.2 Copolymers of ECA and MCA by vapor phase polymerization (Batch Fuming) and Liquid phase polymerization
The SEM images of the copolymer nanofibers made by the batch set-up fuming
on the Superfrost substrate in the presence of both MCA and ECA monomer vapors is
given in Fig. 7.5
Figure 7.5: SEM image of P(ECA-MCA) nanofibers after 12 hours of batch fuming; (a) at 5K magnification and (b) at 40K magnification
As seen in Fig. 7.5 (a), the copolymer nanofibers obtained (i.e. P(ECA-MCA)) appear identical to the PMCA fibers observed earlier or PECA fibers seen previously in earlier chapters. There does not seem to appear any marked differences in the morphology of the P(ECA-MCA) fibers. From the image at higher magnification the diameter of the fibers again appear to range from 50-100nm.
The ATR-IR spectrum for the P(ECA-MCA) nanofibers is depicted in Fig. 7.6
(a). Also shown in Fig. 7.6 (b) is the IR spectrum of the bulk liquid phase polymerized copolymer sample made in THF.
151
Figure 7.6: ATR-IR spectra obtained from (a) P(ECA-MCA) nanofibers and (b) from P(ECA-MCA) liquid phase polymerized bulk copolymer
The two spectra appear exactly identical to each other with regard to the peaks
responsible for vibrations due to the copolymer. There are however, significant
differences between these two copolymer IR spectra and the IR spectra of the PECA and
PMCA homopolymers. These are evident in the specific regions between 3200-2800cm -1 and 1900-1300cm -1 shown in Fig. 7.7 and Fig. 7.8, respectively. The spectra for PMCA
and PECA homopolymer nanofibers in these regions are also shown for comparison.
From these figures, it is evident that the P(ECA-MCA) nanofibers spectrum includes all
the peaks that are present in both the PECA and PMCA homopolymer spectra.
152
Figure 7.7: IR Spectral region between 3200-2800cm -1of PMCA, PECA & P(ECA-MCA) nanofibers
153
Figure 7.8: IR Spectral region between 1900-1300cm -1of PMCA, PECA & P(ECA-MCA) nanofibers
The C-H asymmetric and symmetric stretching vibrations for methyl and methylene
groups (2990, 2944, 2907 & 2875cm -1) from the PECA polymer and the relocated C-H
-1 stretching vibrations for CH 3-O at 2960 and 2850cm from the PMCA polymer are all manifested in the P(ECA-MCA) polymer nanofibers spectra. Also in the second region,
-1 (Fig. 7.8 ), the CH 3 umbrella mode vibrations present in PMCA at 1435cm and in PECA
at 1375cm -1 are both evident in the P(ECA-MCA) spectra. The presence of a slightly
pronounced C=C vibration at 1620cm -1 in P(ECA-MCA) is also representative of PMCA in the copolymer nanofiber.
154 7.3.3 Copolymers of ECA and MCA by vapor phase polymerization (Flow setup Fuming)
The SEM images of the P(ECA-MCA) copolymer nanofibers created in the long
time fuming experiment conducted in the flow setup is shown in Fig. 7.9
Figure 7.9: SEM image of P(ECA-MCA) nanofibers after > 5 days of flow setup fuming; (a) at 10K magnific ation and (b) at 40K magnification
The images depict nanofibers similar to PECA nanofibers obtained after 2-stage fuming in the flow setup described in Chapter 6. The diameter of the fibers range between 100-
150nm and the fibers uniformly cover the substrate.
Fig. 7.10 and Fig. 7.11 represent specific regions of the ATR-IR spectra obtained
from the P(ECA-MCA) nanofibers created in the flow setup. The spectrum of the
P(ECA-MCA) copolymer nanofibers made previously in the batch setup is also included
for comparison. As in the case of the latter, the P(ECA-MCA) nanofibers made in the
flow setup also show peaks representative of both PECA and PMCA homopolymers.
155 However, from Fig. 7.10 , it is evident that the peak at 2960cm -1 due to the C-H asymmetric stretching vibrations of the CH 3-O group is more intense in the nanofibers
-1 made in the flow setup. Also the peak at 2850cm due to symmetric stretches of CH 3-O is more prominent in this spectrum.
Figure 7.10 : IR Spectral region between 3200-2800cm -1of P(ECA-MCA) nanofibers made in the batch setup and flow setup
156
Figure 7.11 : IR Spectral region between 1900-1300cm -1of P(ECA-MCA) nanofibers made in the batch setup and flow setup
Similarly in Fig. 7.11 the peak due to the CH 3 umbrella mode vibration at
1435cm -1 is more pronounced in the flow setup copolymer nanofibers as compared to batch setup copolymer fibers.
7.3.4 Molecular weight estimations of P(ECA-MCA) copolymer samples by GPC
Estimation of molecular weight by GPC required the copolymer samples to be
soluble in THF, the solvent used for GPC. Interestingly, however both copolymer
samples, P(ECA-MCA) nanofibers made via vapor phase polymerization in the flow
157 setup and P(ECA-MCA) copolymer made in bulk liquid phase, were THF insoluble. In
fact for the case of the bulk liquid phase sample, THF was used as solvent to conduct the
liquid phase polymerization and 12 hours after injecting ECA and MCA liquid monomers
into THF, a white polymer residue precipitated out of the solution signifying insolubility
in THF. Fig. 7.12 depicts milky suspensions of insoluble copolymer samples (nanofiber
and bulk liquid phase) in THF.
Figure 7.12 : Milky suspensions of insoluble P(ECA-MCA) copolymers in THF
Fig. 7.13 illustrates the GPC traces obtained after injection of the above copolymer samples. By virtue of a guard column fitted prior to the GPC columns all THF insoluble material is removed and neither sample exhibits any significant polymer peak.
158 Peaks of small intensities observed after ~ 35min are possibly due tiny impurities that get
past the guard column or very low molecular weight polymer.
Figure 7.13 : GPC traces of P(ECA-MCA) copolymer samples in THF.
7.4 Discussion
7.4.1 Synthesis of poly (methyl cyanoacrylate) [PMCA] nanofibers
The resemblance of PMCA nanofibers by batch fuming of MCA vapors under
high humidity to the PECA nanofibers made previously in similar conditions suggest
identical mechanisms of nanofiber formation for each. This is expected considering the
two monomers are not very different. They have similar vapor pressures at room
temperature, 0.17 torr for ECA 14 and ~ 0.2 torr for MCA, 15 which results in similar
deposition of polymer from the vapor phase onto suitable substrates. The PMCA fibers
159 also are 50-100nm in size, similar to PECA nanofibers. This observation is explainable
by the fact that surface tension of MCA is 37.4mN/m, which is similar to ECA’s surface
tension of 34.3mN/m and as suggested in Chapter 5, the size of the fiber is probably
determined by the interaction of water and monomer, the diameter of the fibers then
would remain somewhat consistent between PMCA and PECA nanofibers.
The IR spectrum of the PMCA fibers illustrates marked differences from the
PECA homopolymer IR spectrum namely; the C-H asymmetric and symmetric stretches
-1 of the CH 3-O group at 2960 and 2850cm , respectively and the CH 3 umbrella mode vibration at 1435cm -1. The distinctive locations of these peaks make it possible to
positively identify the presence of PMCA in a copolymer sample.
7.4.2 Copolymerization of ECA and MCA monomers
The SEM images of the copolymer P(ECA-MCA) nanofibers shown in Fig. 7.5 and Fig. 7.9 made by batch setup and flow setup fuming show no marked differences in
morphology compared to homopolymer nanofibers of either monomer. This is suggestive
of the fact that both monomers are incorporated identically into the growing nanofiber
without any change in diameter of the fibers. The IR spectra of the copolymers (Fig. 7.7,
Fig. 7.8, Fig. 7.10 and Fig. 7.11 ) all include representative peaks from both
homopolymers confirming their presence in the copolymer nanofibers. As vapors of both
monomers were present during the fuming process the nanofiber formed is a random
copolymer. The IR spectrum of copolymer samples made in the liquid phase, also
confirm the incorporation of both monomers.
160 Comparing the IR spectra of the P(ECA-MCA) nanofibers made in the batch set up versus those made in the flow setup for fuming, the intensity of the peaks due to
PMCA is stronger in the flow setup copolymer samples. This result is understandable by the higher concentration of MCA monomer during the flow setup fuming experiment. In the batch setup fuming, the monomers were introduced simultaneously into the chamber and because of their similar vapor pressures the nanofibers formed were most likely a 1:1 random mixture copolymer of the two monomers. However in the case of the flow setup, the ratio of ECA to MCA flows was 1:2, thereby resulting in copolymer nanofibers with a greater proportion of MCA than ECA. This confirms that it is possible to control the composition of a nanofiber during its vapor phase template-less growth by controlling the composition of the vapor phase.
An issue that arises when interpreting the information of copolymer samples is the eventuality of having two homopolymers formed independently during attempted copolymerization. One can imagine a situation where monomers with dissimilar reactivities towards either propagating carbanion would result in the formation of two individual homopolymers but not a random copolymer. This mixture of homopolymers would cause an IR spectrum representative of the presence of both monomers and would be difficult to tell apart from that of an actual random copolymer of the two monomers. A second characterization such as molecular weight estimation of the polymer would then be required to tell these two cases apart. A unimodal molecular weight distribution in this case would signify the formation of copolymer whereas any other distribution (e.g. bimodal) would mean individual homopolymers in the sample. All copolymer samples made in this study (nanofibers and bulk liquid phase samples) were THF insoluble, which
161 was the solvent used for GPC. Hence the GPC traces for these copolymer samples
yielded no major polymer peaks making their molecular weight estimations impossible
with the current solvent. However, this absence of peaks in the GPC traces in fact
confirms the formation of a P(ECA-MCA) copolymer. Though the PECA homopolymer
is completely soluble in THF even for high molecular weight polymers (made in Chapter
6), PMCA is THF insoluble. 16 This distinction of PMCA makes it possible to identify its
presence in a polymer sample as any copolymer that includes MCA would also be THF
insoluble. Therefore, the complete insolubility of the copolymer samples in THF
confirms their incorporation of MCA and hence the formation of a random copolymer
7.5 Conclusions
The first section of this study successfully demonstrated the fabrication of PMCA
nanofibers just as in the case of PECA nanofiber formation reported in earlier chapters.
Reasons for such similarity with PECA nanofiber formation were similar mechanisms of
polymerization due to similar surface tension and vapor pressure values for MCA and
ECA monomers
The observations made in the copolymerization studies confirm the concept of
using the template-less growth of PECA nanofibers to create copolymers with other
monomers. The growing tips of the PECA fibers that house the active carbanionic ends of
the polymer chains are utilized for initiating polymerization of a second monomer (MCA)
introduced in the vapor phase. As the PECA polymer was propagating as a nanofiber the
MCA monomer is also incorporated into the copolymer as a nanofiber. The copolymer
162 nanofiber thus formed was a random mixture of PECA and PMCA as confirmed by IR
and GPC studies. The fact that assists such copolymerization is similar reactivities of
both monomers towards either propagating carbanion. This proof of concept then opens
up the potential for synthesizing copolymer nanofibers of different monomers that also
have similar reactivities with ECA to form interesting copolymer nanofibers.
7.6 References
(1) Ma, Z. W.; Masaya, K.; Ramakrishna, S. J Membrane Sci 2006 , 282 , 237-244. (2) Nair, S.; Natarajan, S.; Kim, S. H. Macromol Rapid Comm 2005 , 26 , 1599-1603. (3) Odian, G. G. Principles of polymerization , 3rd ed.; Wiley: New York, 1991. (4) Oikawa, E.; Aoki, F.; Katano, K. Polym J 1979 , 11 , 257-259. (5) Kinsinge.Jb; Panchak, J. R.; Kelso, R. L.; Bartlett, J. S.; Graham, R. K. J Appl Polym Sci 1965 , 9, 429-437. (6) Sperlich, B.; Eisenbach, C. D. Acta Polym 1996 , 47 , 280-284. (7) Deng, L.; Yao, C.; Li, A.; Dong, A. Polymer International 2005 , 54 , 1007-1013. (8) Mashiko, Y.; Yoshida, M.; Koga, M. Polymer 1997 , 38 , 4757-4763. (9) Morton, M. Anionic polymerization: principles and practice ; Academic Press: New York, 1983. (10) Maruyama, K.; Tsushima, Y.; Kuramochi, T.; Ibonai, M.; Nagasawa, K. Int J Adhes Adhes 1989 , 9, 143-144. (11) Stein, M. J Appl Polym Sci 1992 , 46 , 2217-2222. (12) Smith, B. C. Infrared spectral interpretation: a systematic approach ; CRC Press: Boca Raton, 1999. (13) Socrates, G.; Socrates, G. Infrared and Raman characteristic group frequencies: tables and charts , 3rd ed.; Wiley: Chichester; New York, 2001. (14) Tech.Information. MSDS: M-Bond 200 Adhesive;Davidson Measurement, 2004. (15) Woodman, A. L.; Adicoff, A. J Chem Eng Data 1969 , 14 , 479-&. (16) Donnelly, E. F.; Pepper, D. C. Makromol Chem-Rapid 1981 , 2, 439-442.
Chapter 8
Conclusions and Future directions
8.1 Conclusions from this study
The objective of developing a facile template-less technique for creating polymer
nanofibers during polymerization was achieved and demonstrated by the growth of poly
(ethyl 2-cyanoacrylate) [PECA] nanofibers. The PECA fibers were initiated by suitable
anionic initiators present in fingerprint residue during high humidity vapor phase
polymerization (fuming) of the ECA monomer. In Chapter 3, the initiators present in the
fingerprint residue that were responsible for PECA nanofiber growth were identified via
fuming studies of individual components of fingerprint residue under similar conditions.
Chapter 4 provided insights into the mechanism of intrinsic nanofiber growth of
PECA during polymerization. The investigation resolved why polymer fiber growth
occurs when initiation is achieved by anions such as a Cl -, and an involuted polymer film results when initiation is carried out using initiators such as OH - under the exact same conditions of polymerization. The answer lies in differences in the rates of initiation of the different anionic initiators. This detail was revealed when the different anions were classified on their relative softness-hardness character using the Hard Soft Acid Base principle. The classification uncovered that faster initiation (harder anions) meant more initiation sites which resulted in a 2-dimensional film growth and slower initiation (softer
164 anions) meant localized initiation sites that resulted in 1-dimensional fiber growth during polymerization.
In Chapter 5, this route of fabricating PECA nanofibers from certain initiators was extended to demonstrate fiber growth on glass slides modified with silanes. With the insight that localized sites of initiation lead to fiber growth, similar PECA nanofiber growth on glass slides was explained by the scenario of ECA monomer wetting the surface of the glass slides first followed by initiation by sufficient water vapor. Different silanes on glass offering different wettability for the ECA monomer on the surface resulted in differences in number density of fibers formed per unit area, confirming this premise. Such control in placement of the nanofibers demonstrates the distinctive nature of this technique.
Chapter 6 and Chapter 7 showcase the unique advantages of this route of fabricating polymer nanofibers during polymerization by demonstrating control over the composition of the nanofibers during synthesis. In Chapter 6 fibers of different molecular weights are created by controlling the conditions during the propagation period of polymerization. By altering the concentration of termination agents during propagation
(i.e. water vapor), PECA nanofibers of different molecular weights and PDIs were synthesized. Such control over the properties of the polymer nanofibers is an exclusive benefit of this technique of making nanofibers. The work in Chapter 7 further proves the versatility of this approach by creating copolymer nanofibers of methyl and ethyl 2- cyanoacrylate. The ability to use the growing carbanionic tips of the fibers to initiate copolymerization of a second monomer is made possible by similar reactivities of both monomers towards either propagating carbanion. The copolymer nanofibers’ morphology
165 is identical to the homopolymer nanofibers suggesting seamless incorporation of the two
monomers in the nanofiber. This demonstration opens up the potential of creating
copolymer nanofibers of these cyanoacrylates with other interesting polymers such as
methyl methacrylate, vinylidene chloride, etc.
Overall this work is a description of this novel template-less approach for polymer
nanofiber formation and exhibits the unique advantages that it offers compared to current
nanofiber formation methods such as electrospinning and templating. This study creates
some very interesting avenues for future research directions, some of which are discussed
below.
8.2 Future directions
A relatively consistent observation of the PECA nanofiber fabrication has been
the diameter of the fibers. Different relative humidities did not appear to affect the
diameter of the fibers. Even for the case of the PMCA nanofibers the diameter does not
appear to vary. It has been suggested in this work that absorbed ECA or MCA monomer
on a surface initiated by water is what determines the diameter of the fiber. Hence by
varying the surface tension of such monomers the size of the absorbed monomer could be
altered which would lead to different sized fibers. Other higher homologues of α-alkyl
cyanoacrylates such as hexyl and octyl cyanoacrylates have lower surface tension values
(~ 29mN/m) but are initiated during polymerization in the same manner as ECA or MCA.
Intuitively, fuming of these monomers would also yield fibers but the fibers could have
different diameters than those observed for ECA or MCA (surface tension values
166 ~37mN/m). The availability of these monomers is however scant. They are currently not
sold commercially in the United States and would require to be synthesized for this
investigation.
Another useful analysis would be to measure the properties of the polymer
nanofibers made in this way. As the molecular weight of the fibers can be controlled, the
effect of molecular weight on different properties of the polymer fibers such as, glass
transition temperature, and mechanical properties such as tensile strength can be
examined.
With respect to extending this approach of polymer nanofiber fabrication to other polymers, the use of this intrinsic fiber growth for copolymerization with the other polymer is the most encouraging route. Other anionically initiated polymers that are similar to ECA include methyl methacrylate (MMA), methacrylic acid (MA) and vinylidene chloride (VC). Homopolymers of these monomers each have functional properties for example; PMMA is impact resistance but transparent (Plexiglass), PMA is useful for its hydrophilicity (PMA) and PVC (Saran®) is used as plastic wrapping sheets.
These monomers are most alike to ECA and hence would have similar reactivities for copolymerization to occur. Hence an important study to think about in the future would be to attempt vapor phase copolymerization of ECA with these monomers using the template-less approach to create copolymer nanofibers that would possess interesting properties. A vital consideration during these copolymerization attempts would be to remove the presence of all water vapor from the system before introducing the second monomer. The reason being that though water vapor is a termination agent for ECA polymerization, this polymerization continues regardless of water being present owing to
167 the fact that water is strong initiator also for its polymerization, however, for MMA or other monomers that polymerize anionically water is a definite terminating species and copolymerization would not proceed in its presence. The removal of other terminating agents ordinarily present in ambient air might also have to be removed from the system before introduction of the second monomer.
Appendix A
Supplemental Information for Chapter 3
SEM images of only the spin coated palmitic acid and stearic acid on clean Si
wafers are depicted in Fig. A.1 and Fig. A.2.
Figure A.1: Control SEM of palmitic acid coated Si wafer
169
Figure A.2: Control SEM of stearic acid coated Si wafer
Appendix B
Supplemental Information for Chapter 5
This section contains the SEM image of the plain untreated out-of-box
Superamine glass surface ( Fig. B.1 ), molecular structures of the different silane molecules used in the study ( Fig. B.2 ) and a representative XPS survey scan to
determine the surface elemental composition of the Superamine glass slides ( Fig. B.3).
Figure B.1: SEM image of the plain untreated Superamine slide
171
Figure B.2: Silane molecules used for modifying the Superclean * glass surfaces
172
Figure B.3: Typical XPS survey scan used to determine elemental composition of the Superamine glass surface
VITA Pratik Mankidy EDUCATION Ph.D. Chemical Engineering The Pennsylvania State University December 2007
Dissertation Title : Novel template-less synthesis of Polycyanoacrylate Nanofibers
M.S. Chemical Engineering The University of Missouri-Columbia August 2002
Dissertation Title : Catalysis and Microcalorimetry of Alkali Zeolite.
B.E. Chemical Engineering University of Pune-India June 2000
PUBLICATIONS Mankidy, P., Rajagopalan, R., Pantano, C.G., Foley, H.C., ‘Template-less growth of polymer nanofibers on silane-modified glass surfaces’ Mankidy, P., Rajagopalan, R., Foley, H.C., ‘Influence of Initiators on the Morphology of poly (ethyl 2-cyanoacrylate)’ Submitted to Macromolecules (Dec 2006 ). Mankidy, P., Rajagopalan, R., Foley, H. C., ‘Facile catalytic growth of cyanoacrylate nanofibers’, Chemical Communications , 2006 , 1139–1141. Rajagopalan, R., Ayyappan, P., Mankidy, P., Brooks, A., Yi, Bo, Foley, H. C., ‘Molecular sieving platinum nanoparticle catalysts kinetically frozen in nanoporous carbon’, Chemical Communications, 2004 , 21, 2498-9. Suppes, G. J., Dasari, M. A., Doskocil, E. J., Mankidy, P. J., Goff, M. J., ’Transesterification of soybean oil with zeolite and metal catalysts’, Applied Catalysis, A: General, 2004 , 257(2), 213-223. Doskocil, E. J., Mankidy, P., ‘Effects on solid basicity for sodium metal and metal oxide occluded NaX zeolites’ , Applied Catalysis, A: General, 2003 , 252(1), 119-132. PRESENTATIONS Mankidy, P., Rajagopalan, R., Pantano, C., Foley, H. C., ‘A Simple Method to grow polymer Nanofibers from Superglue®’ Upcoming oral presentation at The Materials Research Society Fall Meeting, Fall 2006 , Boston, MA. Mankidy, P., Rajagopalan, R., Foley, H. C., ‘Template-less catalytic growth of poly (ethyl 2-cyanoacrylate) nanofibers’, Oral presentation at The American Chemical Society National Meeting, Spring 2006 , Atlanta, GA. Mankidy, P., Rajagopalan, R., Foley, H. C., ‘Novel Template-less Fabrication of Poly (ethyl 2-cyanoacrylate) Nanofibers’, Poster presentation at The Materials Research Society Fall Meeting, Fall 2005 , Boston, MA. Mankidy, P., Rajagopalan, R., Foley, H. C., ‘Fabrication of a Robust Anodized Aluminum Oxide Membrane on Aluminum’, Oral presentation at The American Institute of Chemical Engineers, Fall 2004 , Austin, TX. AWARDS Teaching Fellowship , Dept. of Chemical Engineering, Penn State University, Spring 2005. Outstanding Teaching Assistant Award, Dept. of Chemical Engineering, Penn State University, Fall 2003.