Mass Spectrometry Techniques for the Characterization Of

Home , Daly detector

MASS SPECTROMETRY TECHNIQUES FOR THE CHARACTERIZATION OF

SYNTHETIC POLYMERS, BIOPOLYMERS, BIODEGRADATION PRODUCTS

AND THEIR INTERACTIONS

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Madalis Casiano-Maldonado

May, 2012 MASS SPECTROMETRY TECHNIQUES FOR THE CHARACTERIZATION OF

SYNTHETIC POLYMERS, BIOPOLYMERS, BIODEGRADATION PRODUCTS

AND THEIR INTERACTIONS

Madalis Casiano-Maldonado

Dissertation

Approved: Accepted:

______Advisor Department Chair Dr. Chrys Wesdemiotis Dr. Kim C. Calvo

______Committee Member Dean of the College Dr. Abraham Joy Dr. Chand K. Midha

______Committee Member Dean of the Graduate School Dr. David S. Perry Dr. George R. Newkome

______Committee Member Date Dr. Peter Rinaldi

______Committee Member Dr. Michael J. Taschner ii

ABSTRACT

The characterization of synthetic polymers and complex oligomeric mixtures by a combination of different mass spectrometry (MS) techniques is the main topic of this dissertation. Ion mobility (IM) separation and/or tandem mass spectrometry (MS2) were interfaced with single stage mass spectrometry for the fast, complete, precise and accurate characterization of different polymers, biopolymers, and biodegradation products. Additionally, the development of a mass spectrometry protocol for the quantitation of proteins adsorbed on different polymer surfaces was accomplished in this dissertation.

Over the passed years, synthetic routes for the creation of PEGs have been modified by the elimination of organic catalysts and solvents and the application of green chemistry for the generation of new polymers. In Chapter 4 of this dissertation, MS and MS2 were employed, as the main analytical techniques, for the structural elucidation of enzyme- catalyzed, functionalized PEG and tetraethylene glycol (TEG) biomaterials. The samples analyzed were synthesized by two different processes, transesterification and Michael addition reactions, both of them using CALB as the catalyst.

In Chapter 5 of this dissertation, polylactide was degraded using Proteinase K and its biodegradation products were analyzed by MS and MS2. Elucidation of its degradation products is important to the biomedical community; knowledge of these products helps to iii

see any toxicity problems with the use of this polymer. The MS and MS2 results showed that the degradation products are short polylactide chains, down to the monomer. No other, potentially dangerous organics were detected.

In the following chapter of this dissertation (Chapter 6), the adsorption of three model proteins onto two different surfaces was evaluated over a pH range. The results were explained on terms of polymer chemistry, surface morphology, proteins’ isoelectricpoints and molecular dimensions.

Finally, Chapter 7 reports how derivatization, degradation, or chromatographic separation can be avoided for the analysis of complex systems, if multidimensional mass spectrometry methods combining ion mobility separation and different ionization techniques are employed. Using this new approach complex poly(α-peptoid) samples were characterized. The results showed that the main product in the system contained the

N-heterocyclic carbene (NHC) moiety that served as ring-opening polymerization catalyst. Furthermore, it was demonstrated that elimination of the NHC segment leads to cyclic poly(α-peptoid)s that have very similar conformations with comparably sized polypeptides

DEDICATION

To my role model, whom I admire and love, my mom,

Amada Maldonado-Nazario.

To my dad and sisters, for your unconditional support,

Jose Anibal Casiano,

Jeannette M. Casiano,

Lizbeth Casiano,

Roxana Casiano,

and Rosa M. Casiano.

To my husband, who was my first motivator, supporter and my eternal love,

Lyn Gabriel Muñoz-Robledo.

To my latest inspiration, my daughter,

Mikaela Sophia Muñoz-Casiano.

ACKNOWLEDGEMENTS

First, I thank my God, you are The one who was my principal support. Thanks because you help me to keep myself motivated and focused on my dream. I wish to thank

Dr. Chrys Wesdemiotis for his guidance, advice and support throughout this project. It was the most enrichment experience in my academic life and was a pleasure to work for him.

I want to thanks Dr. Xiaopeng Li because I was fortunate to receive his help and advices. My special thanks to Dr. Goy Teck Lim, Dr. Judith E. Puskas and Kwang Su

Seo for all the input, advice and for providing many sample which are the main part of this dissertation. Thanks to Li Guo and Donghui Zhang, who believed in my work and permitted me to collaborate with them.

I would like to thank my committee members for actually take the time to read my dissertation and all their feedback: Dr. Abraham Joy, Dr. Michael J. Taschner, Dr.

Peter Rinaldi, and Dr. David S. Perry.

I would like to express my thankfulness to Nilufer Solak, Danijela Smiljanic-

Jovicic, Gladys Rocio Montenegro Galindo, Tejal Deodhar, Saida Y. Garcia and Aleer

Yol. Thanks for all the wonderful memories we built together during these years. Also, I

wish to thanks Nilufer Solak, David E. Dabney, Wesdemiotis former and current research group for sharing with me their expertise in mass spectrometry.

I would also like to thank the Department of Chemistry and the State of Ohio for providing financial support while conducting my graduate studies at the University of

Akron.

Thanks to all my Akron friends for their company. They made me feel at home.

Finally, thanks to my family who believe in me, who gave me support almost every day, who were capable to hear complaints. Thanks for being the wonderful family ever: Amada Maldonado, Jose Anibal Casiano, Jeannette M. Casiano, Lizbeth Casiano,

Roxana Casiano, Rosa M. Casiano, all my nieces and brothers in law. Thanks to my husband, Lyn G. Muñoz, for his trust, patience, help and love

vii

TABLE OF CONTENTS

Page

LIST OF TABLES.………………………………………………………………… xiii

LIST OF FIGURES..……………………………………………………………….. xiv

LIST OF SCHEMES.……………………………………………………………… xix

LIST OF EQUATIONS…………………………………………………………..... xxi

CHAPTER

I. INTRODUCTION……………………………………………………………. 1

II. MASS SPECTROMETRY BACKGROUND……………………………….. 8

2.1 Mass Spectrometry…………………………………………………. 8

2.2 Ionization Methods………………………………………………… 9

2.2.1 Electrospray Ionization (ESI) …………………………. 10

2.2.2 Matrix Assisted Laser Desorption Ionization (MALDI).. 12

2.3 Mass Analyzers……………………………………………………... 14

2.3.1 Time-of-Flight Mass Analyzer…………………………. 15

2.3.2 Quadrupole Mass Analyzer…………………………….. 19

2.3.3 Quadrupole Ion Trap Mass Analyzer (QIT)……………. 22

2.3.4 Quadrupole Time-of-Flight (Q/ToF) Mass Analyzer…... 25

2.4 Detectors………………………………………………………….… 27

viii

2.5 Ion Mobility Mass Spectrometry (IMMS)………………………….. 28

2.6 Tandem Mass Spectrometry……………………………………….... 34

III. MATERIALS AND INSTRUMENTATION……………………………… 37

3.1 Materials……………………………………………………………. 37 3.2 Instrumentation…………………………………………………….. 39 3.2.1 Esquire-LC ESI-QIT…………………………………… 39 3.2.2 Synapt HDMS Ion Mobility Mass Spectrometer………. 41 3.2.3 MALDI-Q/ToF Mass Spectrometer……………………. 42 3.2.4 Ultraflex III ToF/ToF Mass Spectrometer……………… 43

IV. MASS SPECTROMETRY CHARACTERIZATION OF POLY(ETHYLENE GLYCOL)S SYNTHESIZED BY GREEN CHEMISTRY……………………………………………………………… 45 4.1 Background………………………………………………………… 45 4.2 Sample Preparation and Instruments Used………………………… 48 4.3 Characterization of PEG/TEG Biomaterials……………………….. 49

4.3.1 Characterization of PEG/TEG Diacrylate by ESI-QIT and MALDI ToF/ToF MS……………………………… 50 4.3.2 Characterization of Tetrahydroxy Substituted Dendritic TEG by ESI QIT…………………………….. 59 4.3.3 Characterization of TEG Dimethacrylate by ESI-QIT MS…………………………………………… 64 4.3.4 Characterization of mPEG-VA by MALDI-ToF/ToF MS…………………………………. 66 4.4 Conclusion…………………………………………………..……... 73

V. INVESTIGATION OF THE ENZYMATIC DEGRADATION OF POLY(LACTIDE) BY MASS SPECTROMETRY METHODS………………………………………………………………... 75

5.1 Background……………………………………………………….. 75 5.2 Sample Preparation, Instrument Used and Enzymatic Degradation……………………………………………………….. 76

5.3 ESI-Q/ToF MS Characterization of the Products from the Enzymatic Degradation of PLA……………………………….. 77

5.4 Conclusion…………………………………………………………. 88

VI. PROBING ADSORPTION ON ELECTROSPUN THERMOPLASTIC ELASTOMERIC SURFACES VIA MASS SPECTROMETRY………………………………………………………... 90

6.1 Background………………………………………………………... 90 6.2 Sample Preparation………………………………………………… 93 6.3 Electrospining……………………………………………………… 94 6.4 Compression Molding……………………………………………... 94 6.5 Protein Adsorption………………………………………………… 95 6.6 Calibration Curves………………………………………………… 96 6.7 Matrix-Assisted Laser Desorption/Ionization-Time of Flight Mass Spectrometry (MALDI-ToF MS)……………………. 96

6.8 Scanning Electron Microscopy (SEM) Imaging………………….. 97 6.9 Water Contact Angle (WCA) Measurement……………………… 97

6.10 Reasults and Discussion…………………………………………. 98

6.10.1 Calibration Results……………………………………. 98

6.10.2 Surface Hydrophobicity and Morphology…………….. 101

6.10.3 Protein Adsorption Behavior of D_IBS and PS Surfaces at Different pH Levels………………………. 104

6.11 Conclusion………………………………………………………… 110

VII. TOP-DOWN MULTI-DIMENSIONAL MASS SPECTROMETRY METHODS FOR SYNTHETIC POLYMER ANALYSIS……………….. 113

7.1 Background………………………………………………………... 113

7.2 Experimental……………………………………………………….. 115

7.2.1 Materials………………………………………………... 115

7.2.2 MALDI MS……………………………………………. 115

7.2.3 ESI MS and ESI-TWIM MS…………………...... 116

7.3 Results and Discussion…………………………………………….. 117

7.3.1 MALDI MS Analysis………………………………….. 117

7.3.2 MALDI MS2 Characterization of the NHC-substituted Macrocycle…...... 122

7.3.3 ESI MS Analysis………………………………………. 123

7.3.4 ESI MS2 characterization of the NHC-poly(α-peptoid) and its noncovalent adduct with NHC………………… 125

7.3.5 Separation of the Synthetic Product by ESI-TWIM MS……………………………………... 127

7.3.6 ESI-TWIM MS2 characterization of the higher order noncovalent complexes..……………… 131

7.3.7 Is the Poly(α-peptoid) Formed After NHC Release cyclic?...... 134

7.4 Conclusion………………………………………………...... 136

VIII. SUMMARY…………………………………………………………… 138

REFERENSES………………………………………………………………… 141

APPENDICES………………………………………………………...... 164 xi

APENDIX A: ADDITIONAL DATA………………………...... 165

APENDIX B: COPYRIGHT PERMISSIONS………………………... 171

xii

LIST OF TABLES

Table Page

5.1 PLA degradation products detected after 7 days of Incubation…..………………………………………………………………………. 80

2 . + 5.2 Fragments in the MS (CAD) spectrum of PL6NaOH (NaOH)2Na (m/z 575) formed according to the mechanism in Scheme 5.2. All have NaOH end groups (see Schemes 5.1 and 5.2) and all contain a Na+ charge……..………………………….. 88

6.1 Conditions for compression molding…………………………………………... 94

6.2 Summary of protein adsorption results for the molded and electrospun D_IBS samples……..……………………………………………………………..... 106

6.3 Summary of protein adsorption results for the molded and electrospun PS Samples….…………………………………………………………………………. 106

xiii

LIST OF FIGURES

Figure Page

2.1 Components of a mass spectrometer…………………………………………… 8

2.2 (Top) Droplet formation at the tip of the ESI needle tip and nebulization. (Bottom) Coulomb explosions of charged droplets to produce smaller charged droplets……………………………………………………………………..……… 12

2.3 Schematic representation of ion formation in the MALDI source…………….. 14

2.4 Linear time-of-flight analyzer………………………………………………….. 16

2.5 Reflectron time-of-flight mass spectrometer…………………………………... 18

2.6 Continuous and delayed pulsed extraction mode in a time-of-flight mass spectrometer………………………………………………………………..…….... 19

2.7 Schematic representation of a quadrupole mass analyzer……………………… 20

2.8 A quadrupole ion trap mass analyzer…………………………………………... 23

2.9 Typical stability diagram for a QIT, in which four singly charged ions are identified as m1, m2, m3, and m4. The masses of the ions are m1>m2> m3> m4. Ions will have stable trajectories if qz<0.908…………………………..………………………………….. 25

2.10 Schematic of a Q/ToF mass spectrometer…………………………………...... 26

2.11 Cross-section of a microchannel plate (left) and electron multiplication within a channel (right)……………………………………………………………………….. 27

2.12 Schematic diagram of a Daly detector………………………………………… 28

xiv

2.14 Traveling wave device inside the Synapt HDMS system…………………….... 30

2.15 (Left) Stacked ring electrodes; (right) schematic of the principle of operation of a traveling wave ion guide containing ring electrodes……………………………….. 32 3.1 Bruker Daltonics Esquire-LC ESi -QIT diagram………………………………… 39 4.1 ESI-QIT mass spectrum of TEG diacrylate. TEGAH = monosubstituted TEG molecule (acrylate/hydroxyl end groups, 72 Da)….………………………….. 51

4.2 ESI-CAD mass spectrum of sodiaded TEG diacrylate, acquired with the Bruker Daltonics Esquire-LC ion trap mass spectrometer. The precurson ion was subjected to CAD with helium at an rf excitation amplitude of 0.75 V….…………. 52

4.3 Partial MALDI-ToF mass spectrum of PEG diacrylate, acquired with a Bruker Ultraflex-III ToF/ToF mass spectrometer. The spectrum shows the m/z 1073-1158 region where the 21- and 22 mer were detected. The full spectrum is shown in the insert. Na+ was used for cationization…….………………………………………… 54

4.4 MALDI-CAD mass spectrum of the sodiaded 22-mer from a diacrylate terminated PEG (two acrylate end groups). The spectrum was acquired with a Waters Micromass Q/ToF Ultima mass spectrometer. The insert shows an expanded view of the m/z 340- 420 region. Argon was used as collision gas at a laboratory-frame collision energy of 110 eV….…………………………………………………………………………… 57

4.5 ESI-QIT mass spectrum of TEG 512. F = fragments from sodiaded TEG 512 created inside the mass spectrometer; M = TEG 512……………………………………….. 60

4.6 (a) ESI-CAD mass spectrum of the sodiaded tetrahydroxy substituted dendritic TEG (TEG 512), acquired with an ESI-an rf excitation amplitude of 0.70 V. (b) MS2 fragmentation scheme, rationalizing the fragments observed. All ions contain Na+(not included in the scheme)…………………………………………………………...... 62 4.7 ESI-QIT mass spectrum of TEG dimethacrylate………………………………... 64

4.8 ESI-CAD mass spectrum of the sodiaded TEG dimethacrylate, acquired with an ESI- QIT mass spectrometer. The precursor ion was subjected to CAD with helium at an rf excitation amplitude of 0.75 V. A superscripted + indicates that the ion does not contain Na+ (ionized by H+)………………………………………………………………… 66

4.9 Partial MALDI-ToF mass spectrum of mPEG-VA, acquired with the Bruker Ultraflex-III ToF/ToF mass spectrometer. The spectrum shows the m/z 2190-2272 region where the 45- and 46-mer are detected. The full spectrum is shown in the insert. Na+ was used for cationization………………………………………………………………. 67

4.10 MALDI-CAD mass spectrum of the lithiated 45-mer from mPEG-VA. The spectrum was acquired with a Bruker Ultraflex-III ToF/ToF mass spectrometer. The inserts show expanded traces of a low and an intermediate m/z region. The ions marked by # are due to a trace of protonated polymer co-transmitted with the lithiated product……………………………………………………………………………... 69

5.1 Structure of polylactide polymer with di-decyl end groups. Repeating units (C3H4O2) : 72.021129 Da. Polymers used for the enzymatic degradation studies supplied by Sigma………………………………………………………………………………. 76

5.2 ESI-Q/ToF mass spectrum of the water-soluble degradation products of PLA after 7 days of biodegradation with Proteinase K………………………………………… 78

. + 5.3 Network of hydrogen bonds and salt bridges (ion pairs) in PLnNaOH (NaOH)xNa clusters…………………………………………………………………………….. 82

5.4 ESI-Q/ToF mass spectrum of the water-soluble degradation products of PLA after10 days of biodegradation with Proteinase K……………………………………...... 83

. + 5.5 ESI-CAD mass spectrum of the PL6NaOH (NaOH)2Na cluster ion, acquired with the Waters Q/ToF tandem mass spectrometer. The precursor ion was subjected to CAD with Argon at a collision energy of 55 eV………………………………………… 85 6.1 Mass spectra of PMS-IS mixtures containing 1 pmol/µL (top) or 4 pmol/µL (bottom) PMS, and 0.5 pmol/µL Cytochrome C (IS). The spectra were produced by signal averaging 1000 laser pulses……………………………………………………….. 99 6.2 Normalized peak area of different proteins versus protein concentrations. Each point represents the average of 10 mass spectra, with the error bars representing one standard deviation of the data………………………………………………………………. 100 6.3 Water droplet profiles on (a) D_IBS molded sheet, (b) PS molded sheet, (c) D_IBS electrospun fiber mat, and (d) PS electrospun fiber mat………………………….. 102 6.4 SEM micrographs of (a) D_IBS and (b) PS electrospun fiber mats at different magnifications…………………………………………………………………….. 104

6.5 Positive mode MALDI-ToF mass spectra of insulin, ubiquitin and lysozyme adsorbed on the D_IBS fiber mat at various pH levels. Incubation with the PMS for 24 hr at pH 4.4, 5.3, 6.9 and 8.7. Because of the small intensity of the lysozyme peak, its m/z region (14000-14350) is shown in expanded view on the right………………………….. 105

6.6 Adsorption of (a) insulin and (b) lysozyme on D_IBS molded sheet and fiber mat………………………………………………………………………………… 108

xvi

7.1 MALDI mass spectra of poly(N-butylglycine) 1500, synthesized via NHC-mediated polymerization of N-butyl-N-carboxyanhydride (Scheme 3.1.) using different matrices and cationizing salts; (a) DIT plus NaTFA; (b) DCTB plus NaTFA; (c) DHB plus KTFA. CHCA gives rise to similar spectra if NaTFA or KTFA is added. An increased detector gain was used for spectrum (c), which distorted the isotope pattern of the main ion series. The m/z values marked in the spectra are for the monoisotopic signals…………... 119

7.2 MALDI mass spectra of poly(N-butylglycine) 1500, synthesized via NHC-mediated polymerization of N-butyl-N-carboxyanhydride (Scheme 3.1.) using different matrices but no cationizing salt; (a) CHCA, (b) DHB , and (c) HPA. The molecular weight distribution obtained with HPA most closely agrees with the average molecular weight of ~1500 Da determined by size exclusion chromatography. The m/z values marked in the spectra are for the monoisotopic signals…………………………………………… 120

7.3 MALDI MS2 spectrum of the protonated 7-mer from NHC-poly(N-butylglycine) + 1500 ([37+H] at m/z 1181). The asterisks indicate fragments with more than 7 repeat units.………………………………………………………...... 122 7.4 ESI mass spectra of poly(N-butylglycine) 1100 acquired (a,b) without or (c) with NaTFA cationizing salt. Other varied parameters: ESI source temperature (a,c) 40 oC or (b) 100 oC; desolvation temperature (a,c) 60 oC or (b) 150 oC; trap cell bias (a,c) 2 V or (b) 6 V; transfer cell bias (a,c) 1 V or (b) 4 V. The m/z values marked in the spectra are for the monoisotopic signals……………………………...... 125 7.5 ESI MS2 spectrum of the protonated 7-mer from NHC-poly(N-butylglycine) + 1100 ([37+H] at m/z 1181). The asterisks indicate fragments with more than 7 repeat units………………………………………………………………………………. 126 7.6 ESI MS2 spectrum of the protonated cluster containing the 7-mer from NHC-poly(N- + butylglycine) 1100 and a second NHC unit ([37+NHC+H] at m/z 1569). The asterisks indicate fragments with more than 7 repeat units………………………………… 127 7.7 2-D ESI-TWIM MS plots of (a) all ions and (b) mass-selected m/z 1181 from poly(N-butylglycine) 1100. The sample was dissolved in methanol (no cationization salt added). The source and desolvation temperatures were set at 40 and 60oC, respectively……………………………………………………………………….. 128 7.8 Mass spectrum extracted from the 2+ region in the ESI-TWIM MS plot of poly(N- butylglycine) 1100 (Figure 7.7a). Five doubly charged ion distributions were identified. The inset shows an expanded view of the m/z 640-680 range. Measured monoisotopic m/z ratios are given for one species in each distribution; the corresponding calculated m/z values are 647.49, 658.47, 674.49, 1207.89, and 1854.37, respectively……… 130

xvii

7.9 2-D ESI-TWIM MS plot of poly(N-butylglycine) 1100 (all ions), dissolved in EtOH/ACN (v/v, 50/50). No cationization salt was added. The source and desolvation temperatures were set at 40 and 60oC, respectively………………….. 131 7.10 (a) 2-D ESI-TWIM MS2 plot of the m/z 1854 ion from poly(N-butylglycine) 1100 dissolved in MeOH and (b) ESI-TWIM MS2 mass spectrum of doubly charged m/z 1854 extracted from the encased region of the plot. (c,d) ESI-TWIM MS2 mass spectra of doubly charged (c) m/z 1208 and (d) 674.5 extracted from analogous 2-D TWIM MS2 plots (not shown). Doubly charged fragments are indicated by a 2+ superscript. The + ions at m/z 142.1, 154.1, 542.5, and 580.4 in part (d) agree well with (butyl)2N =CH2, + + + [N-butylglycine+Na] , [31+H2O+Na] , and [45+Me] , respectively………………………………………………………………………... 133 7.11 (a) 2-D ESI-TWIM MS plot of poly(alanine) (all ions formed upon ESI) and (b) 2-D ESI MS2-TWIM plot of the protonated 9-mer from NHC-poly(N-methylglycine) (m/z 1028.6). Both plots were acquired at the same TWIM conditions (traveling wave velocity of 350 m/s and traveling wave height of 8.5 V). The samples were dissolved in methanol (acidified by trifluoroacetic acid); source and desolvation temperatures were set at 40 and 60oC, respectively……………………………………………………. 135

xviii

LIST OF SCHEMES

Scheme Page

3.1 Synthesis of poly(α-peptoid)s by N-heterocyclic carbene (NHC) mediated zwitterionic ring-opening polymerization…………………………………………... 38 4.1 Nomenclature schemes: (a) nomenclature scheme for the MS2 fragments of polymers with different end groups, (b) adjusted scheme for the MS2 fragments of polymers with identical end groups………………………………………………………………… 47

4.2 Enzyme-mediated funtionalization of PEG……………………………...... 50

4.3 Charge-remote 1,5 rearrangement to form bn fragment ions from TEG diacrylate by the expulsion of acrylic acid (72 Da)………………………………………………. 52

+ ” 4.4 Charge-induced fragmentation pathway induced by Na to form cn fragment ions from TEG diacrylate by the eliminaton of vinyl acrylate (98 Da)………………… 53

4.5 Charge-remote 1,5-H rearrangement in vinyl-terminated PEG to yield a truncated vinyl-terminated chain by loss of a repeat unit in the form of acetaldehyde……… 58

4.6 Charge-induced fragmentation in the PEG chain to form truncated chains carrying ” one original end group (acrylate) and either a new hydroxyl (cn ) or a new vinyl (bn) end group……………………………………………………………………………….. 58

4.7 Charge-remote fragmentation in the PEG chain involving homolytic bond cleavages and consecutive β scissions to give truncated chains carrying one original end group (acrylate) and either a new vinyl (bn) or a new carbonyl (cn) end group…………… 59

4.8 Elimination of a 105-Da molecule from cationized TEG 512 as a result of a retro- Michael reaction……………………………………………………………………. 63

4.9 Elimination of a 159-Da molecule from cationized TEG 512 as a result of a intramolecular transesterification reaction…………………………………………. 63

4.10 Charge-remote 1.5-H rearrangement at the adipate groups of mPEG-VA…………………………………………………………………………. 70

xix

4.11 Charge-remote fragmentation of lithiated mPEG-VA via hemolytic C-O bond scissions in the PEG chain, followed by β-H. eliminations, to form terminal fragment . series bn, cn, xn and yn. Homolitic C-C bond scissions followed by β losses CH2=O + H provide an additional route to bn and xn fragments. EG designates the mass of the combined end groups. Note that the subscripts in the fragment nomenclature include monomer portions that have been incorporated in the end group; for example, x = 21 for b22, y = 18 for z19, etc…………………………………………………………… 71

. 4.12 Backbiting in the bn radical ions resulting from hemolytic C – O bond cleavages in the PEG chain, ultimately leading to internal fragments with ethyl/vinyl (72 Da) end EV groups, and shorter cn fragments. Analogous reactions sequences generate Jn + zn . HV . HV fragments from xn , Jn + bn fragments from cn , and Jn + xn fragments from . zn ...... 72

5.1 Plausible structures of the products formed by enzymatic degradation of PLA………………………………………………………………………………… 80

5.2 Fragmentation of polylactide ester groups to which NaOH has been added (nucleophilic addition/fragmentatio). This dissociation yields two truncated polylactide chains with HO- and –Na end groups……………………………………………… 86

. 5.3 Elimination of C3H4O2 (72 Da) from the hydroxyl chain end of PL6NaOH + (NaOH)2Na (m/z 575) to yield the m/z-503 fragment……………………………. 87

+ 7.1 Fragmentation of energetically activated [37+H] ions to yield fragments with or without the NHC moiety. The reaction is exemplified at the C=O group attached to NHC, but can similarly occur at the other C=O groups…………………...... 123

LIST OF EQUATIONS

Equation Page

2.1 The Rayleigh limit……………………………………………………………. 12

2.2 The flight time of the ions……………………………………………………. 17

2.3 Velocity of the ions…………………………………………………………… 17

2.4 Mathieu equation in a quadrupole……………………………………………. 21

2.5 Dimensionless trapping parameter au in a quadrupole………………………. 21

2.6 Dimensionless trapping parameter qu in a quadrupole………………………. 21

2.7 Mathieu equation in a quadrupole ion trap…………………………………... 24

2.8 Mobility of an ion…………………………………………………………….. 33

2.9 Reduced mobility of an ion…………………………………………………... 33

2.10 Collision cross section of an ion……………………………………………. 33

2.11 Internal energy gained by an ion……………………………………………. 35

xxi

CHAPTER I

INTRODUCTION

Mass Spectrometry (MS) is an analytical technique in which the gas-phase ions need to be created in order to obtain a spectrum. This analytical technique measures mass-to-charge ratios with the purpose of determining the composition and identity of the analyte and sometimes its fragments. It has been applied over the years to numerous areas, including polymers and biopolymers due to its high sensitivity, low sample consumption, and speed of analysis.1 Similar to other analytical techniques, MS faces challenges, for example in the separation of complex mixtures which may be difficult or impossible to analyze using only MS. This disadvantage can be overcome if separation techniques are combined with MS analysis.2 In this dissertation, ion mobility spectrometry and tandem mass spectrometry were combined with MS analysis to obtain a complete and precise characterization of the sample’s composition, as showed in

Chapters IV-V and VII.

As mentioned above, MS requires sample ionization. Many ionization methods are available and each has its own advantages and disadvantages. The two used most frequently for polar and/or large molecules today are matrix assisted laser desorption ionization (MALDI) and electrospray ionization (ESI). MALDI is a soft ionization process where the creation of ions involves an energy transfer process. In this process, the

analyte of interest is dissolved in an organic solvent and then mixed with a solution of suitable organic compound called the matrix. The matrix is used to prevent the analyte from forming large aggregates and it is the responsible for the absorption of UV (or IR) light from the laser. The sample-matrix solution is allowed to crystallize on a target and a pulsed laser is focused on it to initialize the desorption/ionization event.3 ESI is a soft ionization technique which minimizes the formation of fragments, similar to MALDI. In

ESI, the analyte is dissolved in a volatile solvent and then introduced into the mass spectrometer using a syringe. The ionization of the analyte is due to direct ionization in solution.1 One of the advantages of this ionization technique is the ability to form multiply charge ions. Both of these techniques are well known for their applications to polymers and biopolymers and were used for the completion of this dissertation.

Once the ions are formed, they are directed into the mass analyzer where they are separated according to their mass- to-charge ratios. Both time of flight (ToF) devices and ion traps are common mass analyzers for MALDI and ESI. Each of them has advantages for specific applications. Unlike ion traps, which distort the molecular weight distribution of polymeric products, true distributions can be obtained by using ToF mass spectrometers.4 In particular MALDI-ToF MS is ideally suitable for the analysis of polymer molecular weight distributions.5-7 In MALDI-ToF/ToF MS, the parent ion can be isolated to obtain fragmentation information. In contrast, multiple fragmentation stages (MSn) can be performed by the use of ion trap analyzers. Ion traps have the ability to accumulate ions for a defined length of time. As a result, a specific fragment may be isolated for further fragmentation. This capability can be used in many applications for in-depth structural studies. 8-12 2

In Chapter VI, the adsorption behavior of insulin, ubiquitin and lysozyme on two different surfaces is evaluated using sensitive MALDI-ToF instrumentation. Protein adsorption on polymers has significant importance in biomedical research. It is a fact that the protein adsorption capability of the surface of an implanted material is the key factor for healthy cell growth and tissue colonization.13-15 Compression molded films and electrospun fiber mats of dendritic poly(isobutylene-b-styrene) (D_IBS) polymer, engineered by Judith E. Puskas et al. (The University of Akron), are the surfaces analyzed. D_IBS polymer has structural similarity with poly(styrene-b-isobutylene-b- styrene) triblock copolymer,16 which was approved in 2004 by the FDA for use as coating in drug delivery systems. For this reason, it is believed that D_IBS is a promising biopolymer with a favorable combination of properties suitable for soft tissue implant devices. For the study described in Chapter VI, the D_IBS fiber mats were created by the electrospinning technique17, 18. Oleschuk and coworker19 presented a qualitative and semi-quantitative analysis of plasma proteins adsorbed onto a polyurethane using

MALDI MS. By using the same analytical method, our study demonstrates that protein adsorption is a function of surface morphology, protein isoelectric point, polymer chemistry, and protein molecular dimensions.

Tandem mass spectrometry, also called MSMS or MS2, is a technique that combines more than one mass analyzer in order to examine an analyte and its fragments that were generated inside a collision cell. In this process, two tapes of mass analysis are coupled; in the first stage, the precursor ion is isolated and then broken into pieces inside the collision cell. The fragments created continue to the second spectrometer for separation.20 One of the most accurate ways for identification of synthetic polymer 3

composition and end groups is to use fragments generated by tandem mass spectrometry analysis. As reported by Wesdemiotis et al.21 and Polce et al.22, tandem MS analysis of synthetic polymers may be applied to determine chain-end or in-chain substituents, distinguish architectures and differentiate isobaric and isomeric species. In Chapters IV and V of this dissertation, such applications of tandem mass spectrometry are discussed.

Chapter IV reports the complete characterization via MS and MS2 of the products formed during the enzyme-catalyzed functionalization of poly(ethylene glycol) (PEG) with various vinyl esters. PEG is an important tool in the medical and pharmaceutical field due to its water solubility. This biocompatible polymer is used in many clinical applications such as the PEGylation of protein therapeutics23 and the preparation of carriers for chemotherapeutic agents24. 25 Due to the wide range of applications of PEG polymers and PEG macromolecules, their synthesis has attracted the attention of many researchers. Recent research has concentrated on the development of new “greener” synthetic routes that avoid the use of organic solvents and catalyst which may be harmful to the environment. The Puskas research group has placed substantial effort in generating dendrimers with enzyme catalysts and eliminating organic solvents in the synthetic process.26 They showed that Candida Antarctica lipase B (CALB) can be employed for the successful functionalization of poly(ethylene glycols)s in solution.27 Additionally, they were able to demostrate that synthesis under solventless conditions was as successful as the same synthesis using organic solvents.26 MALDI MS, MALDI MS2, ESI MS, ESI

MS2 were the spectrometry tools used to complete the characterization of the resulting attractive biomaterials.

Chapter V describes the application of ESI MS and ESI MS2 techniques to the characterization of poly(lactide) (PLA) degradation products. Biodegradable polymers are utilized in many medical applications, for example wound management, orthopedic devices, dental, cardiovascular, intestinal, drug delivery and tissue engineering.

Degradation studies are useful for understanding degradation routes, toxicity of the degradation products, etc. Methods that probe morpholical changes,28 thermal behavior changes29, and molecular weight changes30 have been employed for the study of polymer degradation. Gupta et al.31 studied the thermal degradation of PLA and found that the isothermal degradation of this polymer is mainly due to random chain scissions.

Yamashita and coworkers32 investigated the biodegradation of PLA in the presence of

Proteinase K via quartz crystal microbalance and atomic force microscopy (AFM). They concluded that the erosion rate of the PLA film was proportional to the amount of enzyme (enzyme concentration < 100 µg/ml) in the incubation/degradation medium. The degradation of PLA under the presence of Proteinase K was revised by Numata et al.33

Weight loss and changes in the molecular weight (Mn) were evaluated to draw conclusions. Li and coworkes34 evaluated the same degradation system as well, but in this occasion by nuclear magnetic resonance (NMR), size exclusion chromatography (SEC), environmental scanning electron microscope (ESEM), differential scanning calorimetry

(DSC), etc. A disadvantage of many of these methods is the lack of information about the chemical structure of the degradation products and their relationship to the original polymer’s structure. Chapter V reports the characterization of PLA degradation products after incubation of the polymer with Proteinase K in buffer solution at 37 ˚C for 7 and 10 days. PLA was rapidly biodegraded by Proteinase K. After one week in the enzyme

environment, the film showed signs of erosion and degradation products were detected by mass spectrometry methods, while the film in enzyme-free environment still looked intact.

Ion mobility spectometry (IMS) has the ability to separate ions coming from an ion source according to their gas phase mobilities as they travel against a gas flow

(usually Nitrogen) under an electric field.35 Separation of the ions inside the IMS device is based on size, charge and shape due to their interactions with the drift gas.36 The

Waters Synapt HDMS mass spectrometer available in our laboratory has the IMS feature.

This instrumentation uses a variant of IMS called traveling-wave (T-wave) IMS. This method of ion mobility separation employs a pulsed electric field inside the T-wave chamber that creates waves to drive the ions forward.37 IMS has been applied to the accurate elucidation of isobars, isomers, and conformers from biomolecules, biopolymers and synthetic polymers.36,38-40 MS coupled with IMS provides the potential to obtain shape, composition and structural information about the molecule of interest rather than composition and structural information alone. In the field of biomolecules, Williams and coworkers41 reported the conformational analysis of isomeric oligosaccharides and hydrazine-released N-linked glycans by a home-built nano-IMS-MS instrument. T-Wave ion mobility mass spectrometry (IMMS) studies on the identification of human hemoglobin variants were also described by Williams and coworkers.42 Hilton and coworkers43 and Solak44 utilized IMMS data to discriminate between polyether oligomers with the same nominal mass. Chapter VII of this dissertation will demonstrate the power of combining IMMS with other spectrometry techniques such as MALDI MS, MALDI

MS2, ESI MS, ESI MS2 for the complete elucidation of a complex synthetic system. The 6

system analyzed was cyclic poly(α-peptoid)s prepared via N-heterocyclic carbene (NHC) mediated zwitterionic ring-opening polymerization. In peptoids, the side chain is attached to the nitrogen of the peptide backbone, whereas it is on the α-carbon in peptides.45

Previously, a MALDI MS study on the sequence determination of peptoids was reported by Thakkar et al.46 Morishetti et al.47 reported on the fragmentations of alkali metalated peptoids. In the past, individual mass spectrometry techniques have been employed for the analysis and characterization of peptoids. Chapter VII demonstrates for the first time how combined ion mobility separation and tandem MS can be used for the simultaneous separation and determination of polymer architectures. Additionally, data from both ESI and MALDI analysis are utilized to determine whether the poly(α-peptoid) synthesis created a mixture or only product that underwent modification during the mass spectrometry analysis.

CHAPTER II

MASS SPECTROMETRY BACKGROUND

2.1. Mass Spectrometry

Mass spectrometry (MS) is an analytical technique that is widely applied to establish the elemental composition and structure of molecules from a substance or a complex mixture. The technique is useful in areas such as forensic science, polymers, proteomics, pharmaceuticals, environmental chemistry, metabolomics, natural products, and drug discovery. MS involves ionization of the analyte to create charged molecules that can be separated by a mass analyzer and detected, in order to obtain their mass-to- charge ratios (m/z). A typical mass spectrometer consists of five different components which are the inlet system, ion source, mass analyzer, ion detector and data system (see

Figure 2.1). A mass spectrometer can have different component combinations depending of the type of experiments and the application.1

sample inlet ion source mass analyzer detector system vacuum system

data system

Figure 2.1. Components of a mass spectrometer.

The inlet system is used to introduce the sample into the ion source. Three different methods are used to introduce samples inside the MS instrument; direct insertion, direct infusion, or chromatography.48 Direct insertion involves the use of a probe or target to place the sample into the instrument, as the one used in MALDI. In direct infusion, the delivery of the gas or liquid sample to the ion source is completed using a pump or tubing. This method is employed in ESI. As a second choice for ESI experiments, the introduction of the sample to the mass spectrometer can be achieved using chromatographic methods such as liquid chromatography (LC) and gas chromatography (GC).

2.2. Ionization Methods

In order to perform mass spectrometry experiments, ions need to be created before a mass spectrum can be measured. The ions are formed inside the ion source and are directed to the mass analyzer where they are separated according to their m/z ratios.

One of the processes that can be used to ionize an analyte is by addition or removal of electrons. In this procedure, the sample is bombarded with an electron beam leading to the loss of one electron from the molecule or the addition of one electron (rarely observed). Another alternative route for the formation gas phase ions is by the addition or subtraction of an ion; the species formed is known as quasimolecular ion. Quasimolecular ions can arise by the loss or addition of a proton. Cationizing agents such as sodium, potassium and lithium can be added to the analyte for the creation of gas phase ions inside the ion source.48 The process used for ion formation will depend on the analyte identity and ionization source used. 9

Many ionization sources exist and the sample type should be considered at the time of the ionization source selection in an experiment. The ionization methods can be divided as soft and hard. MALDI, ESI, and atmospheric pressure chemical ionization

(APCI) are classified as soft ionization methods where, in general, the sample is detected intact and analyte fragments are not observed. In hard ionization methods, the sample is destroyed completely or partially and only or mainly fragments are detected. Electron ionization (EI), chemical ionization (CI), field ionization (FI), fast atom bombardment

(FAB), and field desorption (FD) are categorized as hard ionization methods (given in order of decreased hardness). MALDI and ESI were the techniques used in this dissertation and both of them will be discussed in this chapter.

2.2.1. Electrospay Ionization (ESI)

ESI is a soft ionization technique, able to produce intact multiply charged ions from large molecules and complexes. ESI was developed by Dole in the late 1960s.49 The first application of ESI to macromolecules was completed by John B. Fenn. Because of the successful ESI analysis of large molecules, Fenn was awarded the Nobel Prize in

Chemistry (2002).50 ESI is a technique that generates ions from a liquid solution after its exposure to a strong electric field at atmospheric pressure. The ion formation process in

ESI can be divided in three general steps: droplet formation, droplet shrinkage and gaseous ion formation.51 The sample solution is introduced into the ionization chamber using a line that is connected to a capillary held at a high voltage. This voltage usually ranges from +3 to 6 KV. The high voltage is responsible for the formation of highly charged droplets at the end of the capillary. Additionally, the formation of highly charged

droplets is helped by a nebulizing gas (usually nitrogen). This process, known as nebulization, is shown in Figure 2.2. The nebulization process creates tiny charged droplets from a “Taylor cone” at the needle tip. The “Taylor cone” is observed due to the presence of an electric field between the charged solution and the counter electrode. The charged droplets are formed from the “Taylor cone” when the droplets’ electrostatic repulsion exceeds the surface tension of the liquid (Rayleigh limit).

After the droplets are formed, the solvent will evaporate as the droplets pass through a heated curtain (drying gas, ie. nitrogen) or capillary. As the solvent evaporates, the droplets start shrinking until the repulsion forces between the charged analyte molecules exceed the surface tension and the droplet can no longer sustain the charge

(Rayleigh limit) at which point a “Coulombic explosion” happens and the droplets break up forming smaller droplets (see Figure 2.2). This process is repeated until no more explosions can happen, usually when single ions are formed. Alternatively, when the droplets become very small, single ions may desorb (evaporate) off their surface. Finally, the ions are transferred to the mass analyzer for determination of their m/z ratios. The relationship between the electrostatic repulsion of the droplet charges and the surface tension is described by the Rayleigh equation (Equation 2.1).52,53

Figure 2.2. (Top) Droplet formation at the tip of the ESI needle tip and nebulization. (Bottom) Coulomb explosions of charged droplets to produce smaller charged droplets. Reproduced with permission from reference 54.

2 2 3 q = 8π ε0γD (Equation 2.1)

Where: q = charge ε0 = permittivity environment γ = surface tension D = diameter of a spherical droplet

2.2.2. Matrix Assisted Laser Desorption Ionization (MALDI)

MALDI is a laser-based soft ionization technique similar to ESI but it does not generate as many multiply charged ions. It is used to analyze biomolecules, oligonucleotides and large organic molecules such as proteins, polymers, dendrimers and other macromolecules. The beginning of this technique is attributed to Michael Karas and 12

Franz Hillenkamp55,56 in 1985. They discovered that the ionization of the amino acid alanine could be improved if it was blended with the amino acid tryptophan and irradiated with a laser.56 Since then, the combination of laser and matrix for the analysis of large molecules has been a success. MALDI matrices consist of a chromophore that strongly absorbs at the laser wavelength. The laser is the energy source for desorption and ionization of the analyte.

The MALDI process occurs in two steps: desorption and ionization, as illustrated in Figure 2.3. In the first step, the sample is mixed with a molar excess of organic matrix.

The matrix is used to prevent the formation of analyte aggregates and it is responsible for energy absorption from the laser. The sample-matrix solution is deposited on the target and it is allowed to crystallize. Then, a pulsed laser is focused on the sample-matrix crystal and the matrix rather than the analyte absorbs the photon energy from the laser.

The matrix has the key role of protecting the sample molecules from photon-induced decomposition. Absorption of photons promotes desorption of the matrix and of the analyte surrounded in the matrix. A fraction of the matrix is also ionized by the laser.

Once in gas phase, the analyte is ionized by gas phase proton transfer reactions with matrix ions. Sometimes, solutions of salts such as the trifluoroacetates NaTFA, LiTFA and AgTFA are added to help in the ionization of samples that are not easily protonated.

Irradiation

Desolvation & Ionization

Desorption

Matrix Proton exchange Analyte

Figure 2.3. Schematic representation of ion formation in the MALDI source. Reproduced with permission from reference 57.

2.3. Mass Analyzers

The main purpose of a mass spectrometer is to measure the mass-to- charge ratio of the gas-phase ions created in the ionization source. The mass analyzer component performs this particular function. Once the mass of the ions is determined by the mass analyzer, the identity of the ions can be established. The selection of a mass analyzer depends on three main characteristics: upper mass limit, transmission and resolution. The upper mass limit is the maximum mass value that can be measured.

Transmission is the ratio between the amount of ions that reach the detector and the number of ions entering the mass analyzer. Resolution is the ability to distinguish two ions with similar m/z ratio values. These mass analyzer main characteristics determine the type of experiments that can be performed and the analytes that can be studied.1

Time-of-flight (ToF), linear ion trap (LIT), quadrupole ion trap (QIT), and Fourier transform ion cyclotron resonance (FT-ICR) analyzers are the four basic types of mass analyzing devices used in modern mass spectrometry. Some of the mass 14

analyzers scan the ions, allowing only one m/z value or a very narrow m/z range to reach the detector. In other cases, all ions are dispersed in space inside the mass analyzer and transferred to the detector. All mass analyzers function under vacuum conditions to prevent collisions with gas molecules. For specific types of analysis, to improve separation and detection of the ions, two or more analyzer can be combined.

The following sections of this chapter will cover time-of-flight (ToF), quadrupole (Q), quadrupole ion trap (QIT) and quadrupole time-of-flight (Q/ToF) analyzers. These mass analyzers were used for the completion of this dissertation

2.3.1. Time-of-Flight Mass Analyzer

Stephens was the first person to explain the concept of time -of flight analysis in the 1940s58. On the other hand, the first commercially available linear time-of- flight mass spectrometer was designed by Wiley and McLaren.59 It was not until the

1980s, when the original linear time-of-flight mass analyzer was coupled to laser desorption ionization, that ToF devices achieved popularity.60 The time-of-flight analyzer has become one of the most useful and popular analyzers after the invention of matrix assisted laser desorption ionization.

Time-of-flight analyzers separate ions, created in the ionization source, according to time measurements. The time that takes a charged particle to travel from the source to the detector is used to determine its m/z ratio.

Figure 2.4. Linear time-of-flight analyzer. Reproduced with permission from reference 1.

Time-of-flight analyzers are known to work better with pulsed ion formation, as in MALDI. Once the ions are formed, they are accelerated in the ion source region using a potential (V). The ions gain a certain kinetic energy K (K = ½ mυ2) after the potential is applied. After the ions leave the source and enter the analyzer drift tube, their velocity does not change since they travel inside a field-free region. Consequently, the separation of the ions will be according to their velocity inside the flight tube. Ions with smaller m/z ratio (lighter) will travel faster than those with higher m/z ratio

(heavier). The time needed by different ions to reach the detector is determined by using

Equation 2.2 where L is the known drift tube distance and v the constant velocity of each

individual ion. Finally, the m/z ratio will be calculated using Equation 2.3 which shows that the m/z ratios depend on the flight time since all terms inside the parenthesis are constant.

t = L/v (Equation 2.2)

v = (2zeV/m)1/2 t2 = m/z (L2/2eV) (Equation 2.3)

At the beginning, the linear time-of-flight analyzer faced problems of poor mass resolution due to the fact that not all ions with the same m/z ratio are created at the same time or with the same initial kinetic energy. This affects the time needed for ions with the same m/z ratio arrive at the detector, causing wide peaks and poor resolution.

This problem was defeated by the introduction of a reflectron and delayed pulse extraction. Reflectron is an ion optical device attached to the flight tube that reverses the trajectory of the ions inside the mass spectrometer (see Figure 2.5). Ions with the same m/z ratio, formed with different kinetic energies (initial velocity) will spend different time inside the reflectron. Ions with higher kinetic energy will penetrate deeper than ions with smaller kinetic energy. Therefore, faster ions will spend more time inside the reflecting field and “catch up” with the slower ions further down in the flight path. As a result, both will reach the detector at the same time.

Figure 2.5. Reflectron time-of-flight mass spectrometer. Reproduced with permission from reference 1.

Additionally, the effect of forming ions of the same m/z ratio with different initial velocities is minimized by the use of delayed pulse extraction. Ions of the same m/z but different initial kinetic energies will have different velocities, thus reaching the detector at different times. Delayed pulse extraction is when ion acceleration, by application of a voltage pulse, takes place after the ions were allowed to drift inside the source for a short time. After the delay time, the accelerating voltage is applied and the ions enter the drift tube region. Ions travelling faster will be closer to the intermediate grid (Figure 2.6); as a result, will receive less kinetic energy when the acceleration pulse is applied. Ions closer to the pulsed electrode will receive more kinetic energy, allowing them to catch up with the faster ions. A scheme of delayed pulse extraction is illustrated in Figure 2.6.

Figure 2.6. Continuous and delayed pulsed extraction mode in a time-of-flight mass spectrometer. Reproduced with permission from reference 1.

2.3.2. Quadrupole Mass Analyzer

A Quadrupole mass analyzer is the only analyzer that actually measures mass directly; it employs a set of four rods or electrodes arranged 90 degrees to each other to separate the ions using an electrostatic field (see Figure 2.7).

Figure 2.7. Schematic representation of a quadrupole mass analyzer. Reproduced with permission from reference 1.

Once the ions are created inside the ionization source, an electric potential is utilized to accelerate the ions out of the source into the analyzer. A series of lenses, called focusing or accelerating lenses, located between the source and analyzer provide the appropriate electric potentials. As the ions enter the analyzer, they start moving with a zigzag motion in the z-direction. The zigzag motion is due to the presence of a direct component (DC = U) and an alternating component (AC = rf = Vcosωt) in the voltages applied to the rods. The strength of the rf potential establishes whether or not an ion with a specific m/z value is discharged from the quadrupole. The same rf and DC potential is applied to each opposite pair of parallel rods. Two rods have an applied potential of (U-

Vcosωt) and the other two rods have a potential of (-U+Vcosωt). The potential applied to the rods is alternating frequently. In other words, two parallel roads have positive potential while the other two parallel roads have negative potential, and after a certain time the rod’s polarities reverse. Positive ions will be attracted by the rods with negative

potential and repelled from the rods with positive potential. When positive ions get close enough to the negative rods and the sign of the potential changes before the ions have the chance to contact the rods, the direction of the ions will change leading to the zigzag motion. The polarity changes on the rods happen really fast. If the ions strike the rods with opposite charge, they will be discharged from the quadrupole. The same concept is applied to negative charge ions.1,61 By proper selection of U and V, only ions with a specific m/z ratio are permitted to pass the quadrupole in zigzag motion, while all other ions are discharged.

The trajectory of the ions depends on their x/y coordinates which are related with the au and qu dimensionless parameters, also known as trapping or stability parameters. The ions will have a complete trajectory, i.e. they will not be discharged from the quadrupole, if their x/y coordinates remain smaller than ro (see Figure 2.7). If the ions have x/y coordinates > ro, they will touch the rods and consequently be ejected from the quadrupole. The Mathieu equation (Equation 2.4) describes mathematically the motion of the ions in a quadrupole.62

2 d u (Equation 2.4) (au  2q cosω t)u  0 d(ωt )2 u 2

8zeU a a a  (Equation 2.5) u x y 2 2 mω ro

4zeV qu qx qy  (Equation 2.6) 2 2 mω ro

In these equations, u represents the coordinates x or y; au is proportional to

U (the DC potential); and qu is proportional to V (the rf amplitude). For mass analysis, the U and V values are changing while ω and ro are constant. U and V are scanned at a constant ratio to transmit successive m/z values to the detector. The trajectory of a charged particle is determined by a specific U/V combination; therefore, the specific U and V values selected determine which mass is transmitted and detected.

2.3.3. Quadrupole Ion Trap Mass Analyzer

The quadrupole ion trap (QIT) mass analyzer was an ingenious idea of

Wolfgang Paul.63 With ESI and MALDI, ions are created in an external source from where they are introduced into the trap using a hole on the entrance end cap. The analyzer is composed of one ring electrode and two end caps; an rf voltage of appropriate magnitude and frequency is applied on the ring electrode to maintain the ions confined.

The rf voltage creates a hyperbolic field inside the trap, with the force exerted on the ions being proportional to their distance from the center of the device. This quadrupolar electric field is responsible for maintaining ions within a certain m/z range in stable trajectories inside the trap. Once the ions are trapped inside the QIT, an auxiliary AC voltage can be applied to the end caps to isolate, fragment or eject specific ions. The ejection of the ions from the trap is through the exit end cap which contains a small hole.

Figure 2.8. A quadrupole ion trap mass analyzer. Reproduced with permission from

reference 1.

A mass spectrum is recorded by changing the amplitude of the rf potential.

Amplitude increments of this rf voltage will cause the trajectories of ions with increasing m/z ratio to become unstable. Unstable trajectories are also obtained due to repulsion ion- ion. Once the ions reach unstable trajectories, they will be ejected from the trap. Ion repulsions can be eliminated if a buffer gas such as helium is introduced inside the ion

trap. Helium gas helps to kinetically cool the ions due to thermalizing collisions among with the helium atoms. As a result, the ions remain in stable trajectories and are not ejected from the trap. Kinetic cooling means loss of excess kinetic energy from the ions. Ion trajectories in the QIT are governed by the Mathieu equation (Equation 2.7).

2 2 2 qz qr  8zeV/m(ro  2zo )Ω (Equation 2.7) where V is the amplitude of the rf voltage applied to the ring electrode, ze is the charge of the ion, m is the mass of the ion, ro is the internal radius of the ion trap, zo is the axial

2 distance from the center of the trap (ro = 2zo ) (see Figure 2.8). The motion of the ions inside the QIT is described by using three dimensional coordinates x, y and z.

Nevertheless, the Mathieu equation can be expressed in terms of z and r because x2 + y2 = r2 in devices cylindrical symmetry.1

Ions are ejected from the ion trap if the qz and qr potentials are set to a value that exceeds the ro and zo (see Figure 2.8). Ions with different mass-to-charge ratio can be successively expelled from the trap since each ion has a unique qz value.

Figure 2.9 shows the stability diagram for the ions inside the QIT, which depends on the rf voltage and m/z values of the ions. In equation 2.7, the values of Ω, zo, ro and e are constant whereas m/z and V are not. As a result, qz is defined only by V and m/z. Ions with qz values smaller than 0.908 will be trapped. In Figure 2.9, the unstable ion is labeled as m4 which has a qz value bigger than 0.908. As a consequence, ion m4 is ejected from the trap. On the other hand, ions m1, m2, and m3 fall into the stability region and thus are trapped. For a mass spectrum, the instrument is scanned from low to high values of m/z. Ions having progressively higher m/z values develop larger oscillations in the z axis and escape the trap. In other words, V values are changed until ions m1, m2, and m3 reach qz values bigger than 0.908 to be able to escape the trap. Once the ions are out of the trap, they reach the detector to produce a signal.

qz= 0.908 unstable

Figure 2.9. Typical stability diagram for a QIT, in which four singly charged ions are identified as m1, m2, m3, and m4. The masses of the ions are m1>m2> m3> m4. Ions will have stable trajectories if qz<0.908. Reproduced with permission from reference 64.

2.3.4. Quadrupole Time-of-Flight (Q/ToF) Mass Analyzer

A Q/ToF is a hybrid quadrupole time-of-flight mass spectrometer with tandem capability. The MS/MS potential is given by the quadrupole which can be used as a mass filter. In this particular circumstance, the quadrupole is set to transmit a selected ion inside the ToF device by making use of rf and DC voltages. After a particular analyte ion is chosen, it is transfered to a collision cell where collisions with a gas (i.e. Argon) produce fragment ions. Fragments and remaining parent ions are guided to the ToF analyzer where they are separated according to their m/z values. The Q/ToF mass analyzer operates in MS mode if the quadrupole is employed as an ion guide. In this occasion, the quadrupole is set to rf-only mode, in which all the ions coming from the 25

ionization source are transferred to the ToF device. The reflectron time-of-flight analyzer is placed orthogonally with respect to the quadrupole and works as a mass resolving device for both MS and MS/MS modes.

Figure 2.10 shows a scheme of a Q/ToF instrument. The resolution of this mass spectrometer can reach values higher than 10,000 with an accuracy of 10 ppm or better.1

Figure 2.10. Schematic of a Q/ToF mass spectrometer. Reproduced with permission from reference 65.

2.4. Detectors

The fifth component of the mass spectrometer is the detector, whose main purpose is to detect the ions that were separated by the mass analyzers. A beam of ions will strike the detector to produce an electric current that is proportional to the amount of ions reaching the detector. Since not all ions issued from the ion source reach the detector, the ion signal is multiplied to obtain an utilizable intensity. Microchannel plate

(MCP) and photon multiplier devices (Daly) are typical detectors used in mass spectrometry instrumentation and the ones used in this dissertation.

The MCP consists of numerous parallel cylindrical channels drilled on a plate.

The channels have a diameter between 5 – 25 μm and are spaced apart by a distance ranging from 6 to 32 μm. Each channel performs an electron multiplier function. When the ions strike the wall of the channel, the original signal is amplified under the presence of a strong electric field due to a cascade of electrons (see Figure 2.11). By combining multiple MCP plates, the number of electrons can be multiplied by a factor up to 108. 1

Figure 2.11. Cross-section of a microchannel plate (left) and electron multiplication within a channel (right). Reproduced with permission from reference 1.

The Daly detector is classified as an electro-optical ion detector. This device combines ion and photon detection. The process involves converting ions to electrons and electrons to photons. The photons reach photomultiplier where they are detected and the

4 5 1 initial signal is amplified. The initial signal can be multiplied by a factor of 10 -10 .

Figure 2.12. Schematic diagram of a Daly detector. Reproduced with permission from reference 1.

2.5. Ion Mobility Mass Spectrometry (IMMS)

Conventional mass spectrometry is unable to differentiate ions with same m/z values but different shapes. Two dimensional separations based on mass and size can be achieved if the ion mobility capability is coupled with MS. The incorporation of IM measurements in mass spectrometry analyses renders a powerful and high-resolution technique for the rapid separation of complex samples and contributes conformational information regarding the analyte. Complex samples containing isobars, isomers and chiral molecules have been successfully separated by IMMS.

The Synapt HDMS, a Q/ToF tandem mass spectrometer supplied by Waters, combines high-efficiency ion mobility separation and tandem mass spectrometry to provide sample identification with high specificity and accuracy. Figure 2.13 shows a scheme of the Waters Synapt HDMS system which incorporates an ion mobility chamber between the quadrupole and time-of-flight mass analyzers.

The sample can be ionized either by ESI or MALDI to produce ions. Ions are accumulated in the ion guide from where they are sent as packets into a quadrupole before injection into the Triwave device. The quadrupole is set in rf-only mode for the acquisition of convectional mass spectra. The Triwave division incorporates three main

compartments: trap, IMS and transfer cell (see Figure 2.14). Inside the trap, the ions are trapped and then released as packets to the IMS section where they travel under the influence of a traveling wave field against the flow of a buffer gas (N2). Inside the IMS section, the ions are separated according to their mobilities on the traveling waves. An ion’s mobility determines the time that it takes for the ion to travel across the length of the IMS cell. Separation occurs because ions with high mobility (smaller size) move faster with the wave and exit the device earlier than ions with lower mobility (larger size) that roll over the wave top and, thus, exit the device later. Finally, the transfer cell transmits the ions to the orthogonal acceleration ToF section for mass analysis. In addition to the functions mentioned above, the trap and transfer regions can also be used as collision cells for fragmentation analyses.

Figure 2.14. Traveling wave device inside the Synapt HDMS system. Reproduced with permission from reference 37.

The Triwave section possesses stacked-ring rf ion guides accommodated in sequence.37 The ring electrodes of the IMS region receive a repeating sequence of transient voltages. Opposite-phase rf voltages are applied to adjacent electrodes; as a consequence the ions are traveling with an axial movement. In addition to the rf voltage, a transient DC voltage is also applied to each ring electrode. The purpose of the DC voltage is to produce a local change in the electric field, so that the ions will move away from the electrode in both forward and reverse direction. To push the ions along the axis in one direction, the DC voltage is changed to an adjacent ring after a defined time, and so on along the guide (see Figure 2.15). This pattern occurs continuously during operation of the Triwave, creating the travelling waves that carry the ions out of the cell.

Application of the DC voltage reduces the time spent by the ions inside the IMS device.

A schematic representation of the stacked ring electrodes and how they work is shown in

Figure 2.15.

Ring Electrode Pair

Ring RF (+) Electrode

Time

Ions Ions

in out

RF (-)

DC Pulse Ions

Figure 2.15. (Left) Stacked ring electrodes; (right) schematic of the principle of operation of a traveling wave ion guide containing ring electrodes. Reproduced with permission from reference 37.

A weak voltage of about 1 to 2 V and a slow travelling wave (300 m/s) are kept inside the transfer cell in order to maintain the ions mobility separation of the ion exiting the IMS cell and entering the ToF analyzer. Normally, the trap cell does not have a voltage.

The Synapt HDMS instrument has the capability to provide collision cross-section information after calibration using ions with defined cross-section values. The standards should be measured under the same conditions as the analyte. Collision cross-sections provide information about the shape and topology of the ions. Contrary to traveling wave

IMMS that requires a calibration curve to obtain unknown cross-sections, traditional

IMMS can directly provide cross-section values from the measured drift times in the IMS cell and no calibration is needed. Because traditional IMMS employs a constant electric

field, drift times are proportional to the corresponding ion cross-sections. In a uniform electric field drift times are directly proportional to cross-section values. Equations 2.8-

2.10 illustrate the correlation between mass, charge and collision cross section.67

v 2 L Normalization 273.15 P k d  k  K o (Equation 2.8) E (t )V T 760 d 1/2 1/2 3ze  2π   mM  1     ko  (Equation 2.9)     Ω 16No  kT    Mm 

 1/2  1/2    3ze 2π mM 273.15 P (V)t Ω         d        (Equation 2.10) kT (m)(M) T 760 2 16No        L

2 -1 -1 Here k is the ion mobility (cm /V s ), ko the reduced mobility (measured mobility corrected to 273.2 K and 760 Torr), vd the drift velocity (cm/sec), E the electric field strength, td the drift time, m the ion mass of the analyte, N the number density (drift gas), T the gas temperature (K), Ω the collision integral, e the unit charge (Coulomb), M the molecular mass (drift gas), k the Boltzmann-constant (Latm/K), z the number of charge on the ion, V the voltage applied to IMS cell, P the pressure inside IMS cell (atm).

IMMS has expanded the applicability of mass spectrometry technology. The

Triwave instrument has been successfully used to separate isobaric components, derive molecular weight information, remove background and reduce complexity, and separate different compound classes. Also, ion mobility separations can be combined with tandem mass spectrometry to reduce the complexity of MSMS data or to separate species for more conclusive structural elucidation. As mentioned previously, tandem mass spectrometry experiments can be performed either in the trap or transfer cell. In other 33

words, ion mobility separation can take place before or after CAD fragmentation.

Furthermore, time aligned parallel fragmentation (CAD/IMS/CAD) experiments are also possible with Synapt HDMS instruments.

2.6. Tandem Mass Spectrometry (MS/MS)

With modern, soft ionization techniques, such as ESI and MALDI, spectra usually contain the intact molecular ion with very little fragmentation data and, thus, often lack the information needed for an accurate structural characterization. In cases like this, tandem mass spectrometry experiments are essential if additional information about the analyte is required. Tandem mass spectrometry, also known as MS/MS or MS2, couple several (at least two) stages of mass analysis. The process involves multiple steps: (1) ion formation; (2) precursor ion selection; (3) generation of fragment ions (4) mass analysis and fragment ion detection.68 Fragmentation of the precursor ion can be carried out by two methods, either in space or in time. For in-space fragmentation two or more analyzers need to be coupled. In-time fragmentation is performed in an ion storage device. The QIT is a common tandem mass spectrometer for in-time fragmentation. In this process, ions trapped in the same place experience multiple separation steps at different times. Initially, ions of all m/z ratios are trapped in the same space. Then, all ions except the specific precursor ion to be fragmented are ejected from the trap. After that, the trapped precursor ions undergo fragmentation, induced by collisions with a neutral gas, typically helium, during a certain time period. Finally, the fragments are sequentially ejected from the trap for detection. Tandem MS in space is the mode of

MS/MS performed in time -of-flight, Q/ToF and triple quadrupole instruments. For

tandem MS in space, the fragmentation steps are carried out by coupling two different instruments. The first analyzer is used to isolate the precursor ion which is guided to a collision cell. In the collisions cell, the precursor ion is allowed to collide with neutral gas

(usually argon) molecules for the generation of fragments. The resulting fragments, also called product ions, are analyzed by the second mass analyzer. The separation in the second mass analyzer is according to m/z ratios. QIT, Q/ToF and ToF/ToF were the analyzer configurations used to complete this dissertation.

Collisionally activated dissociation (CAD), also known as collision-induced dissociation (CID), is the most common fragmentation method for mass spectrometry techniques. In general, the fragmentation process follows a few sequential steps: (1) the precursor ion is accelerated to undergo multiple collisions with the collision gas; (2) the potential energy in the molecule is increased; (3) the fragmentation threshold is reached; and (4) product ions are created. During the collisions with helium gas, some of the kinetic energy gained from the precursor ion is converted to internal (potential) energy which causes break-up of the molecule’s bonds. The maximum internal energy that can be reached by precursor ions having a defined kinetic energy is given by Equation 2.11.

m E  E t int kin M  m (Equation 2.11) i t

where Mi is the mass of the ion; mt is the mass of the gas (target); and Ekin is the kinetic energy. For example, an ion of m/z 300 that has been accelerated to 20eV collides with helium gas (atomic mass = 4 Da), it will achieve in one collision a maximum internal energy equal to 0.26 eV.

4 E   0.26 eV int 300  4

If the same ion collides with the argon gas (atomic mass= 40 Da) instead with helium under the same conditions, the maximum internal energy gained by the ion will be

2.35 eV. Argon gas particles have a larger mass than helium gas particles and, thus, more internal energy would be transferred with the former. For this reason, fragmentation in the Q/ToF, in which Ar serves as the collision gas, will produce more fragments than fragmentation with He in the QIT.

CHAPTER III

MATERIALS AND INSTRUMENTATION

3.1. Materials

Transesterification and Michael addition reactions were utilized for the synthesis of the functionalized PEGs polymers used in Chapter IV. Transesterification is the process where the R group of an ester is replaced by the alkyl (R) group of an alcohol.

Michael addition happens when a nucleophile is added to an unsaturated carbonyl compound. The syntheses were performed by Kwang Su Seo in the Department of

Polymer Science at The University of Akron under the advice of Professor Judith E.

Puskas. Transesterifications were conducted by introducing vinyl acrylate or vinyl methacrylate in a flask containing PEG (HO-PEG-OH) and CALB enzyme. The reaction mixture was purged with inert gas and stirred for 4 hours at 50 ˚C. After the reaction, the mixture was disolved with THF and filtered. The product was concentrated using a rotary evaporator. Finally, the final product was dried for further analysis. For the functionalization of PEGs with divinyl adipate, transesterification was also used, but in this occasion the reaction took place in a sealed and argon-purged flask. Divinyl adipate was added to the flask containing PEG (HO-PEG-OH) and CALB. The mixture was stirred for 4 hours at 50 ˚C. After the reaction, the mixture was diluted, the enzyme was

removed and the product was precipitated. The final product was dried in a vacuum oven for additional analysis. Michael addition reactions between PEG diacrylate and diethanolamine took place inside a flask in the presence of CALB under nitrogen. The reaction proceeded for 24 hours at 50 ˚C. After the reaction, the enzyme was removed.

The final product was precipitated, washed and dried.

The degradation products of polylactide in the presence of Proteinase K are characterized in Chapter V. The polylactide sample was purchased from Fluka (St. Louis,

MO). Proteinase K from Tritirachium album and sodium azide were supplied by Sigma-

Aldrich (St. Louis, MO).

The adsorption of insulin, ubiquitin and lysozyme onto dendritic poly(isobutylene-co-styrene) (D_IBS) compression molded films, D_IBS electrospun fiber mats, polystyrene (PS) molded film and PS electrospun fiber are examined in

Chapter VI. Polymer synthesis and surfaces were performed by Dr. Goy Teck Lim in the

Department of Polymer Science at the University of Akron under the direction of

Professor Judith E. Puskas. Insulin, ubiquitin and lysozyme were obtained from Sigma-

Aldrich (St. Louis, MO). D_IBS was synthesized using living carbocationic polymerization. PS was acquired from Americas Styrenics LLC (Marrieta, OH). Both polymers were compression molded separately. Polymer solutions in THF and toluene were used to create the electrospun fiber mats. For protein incubation purposes, two different buffers were utilized, ammonium acetate and Trizma base. Both were purchased from Sigma-Aldrich

(St. Louis, MO). Urea obtained from MP Biomedicals, LLC (Solon, Ohio) was utilized to remove the adsorbed proteins from the polymer surfaces.

Chapter VII demonstrates the benefit of combining many mass spectrometry techniques for the characterization of complex polymer mixtures, in order to overcome ionization effects, detect minor products and identify different molecular architectures.

The study was done using poly(α-peptoid)s. The poly(α-peptoid)s, poly(N-butylglycine) and poly(N-methylglycine) were synthesized via N-heterocyclic carbene (NHC) mediated ring-opening polymerization (Scheme 3.1), as described by Guo and Zhang.69 Cyclic poly(alanine), an isomer of poly(N-methylglycine), was prepared by the procedure of

Zhou et al.70 Chemicals and solvents were purchased from Fisher Scientific (Pittsburgh,

PA) or Sigma/Aldrich (St. Louis, MO) and were used as received.

Scheme 3.1. Synthesis of poly(α-peptoid)s by N-heterocyclic carbene (NHC) mediated zwitterionic ring-opening polymerization.

3.2. Intrumentation

In the next few sections, the mass spectrometry settings used for the completion of this dissertation are discussed.

3.2.1. Esquire-LC ESI-QIT

Esquire-LC (Bruker Daltonics, Bilerica, MA) is the electrospray ionization quadrupole ion trap mass spectrometer used to characterize the low molecular weight molecules/polymer discussed in Chapter IV. A scheme of this instrumentation is showed in Figure 3.1.

Figure 3.1. Bruker Daltonics Esquire-LC ESi-QIT diagram. Reproduced with permission from reference 71.

In this instrument, samples are introduced inside the mass spectrometer either by direct infusion or chromatography. For this dissertation, direct infusion was used. In direct infusion mode, liquid solutions of the analyte are pumped into the ion source through a metal needle at a flow rate of 300 µl/h. The flow rate is controlled by a syringe pump. The needle is surrounded by a nebulizer which contains a gas, usually nitrogen.

The nitrogen gas flows to the needle tip where it blends with the analyte liquid solution.

The nebulizer gas pressure can be adjusted between 0 psi to 80 psi. The drying gas removes solvent from the charged droplets produced in the ESI process in order to promote the formation of ions. The drying gas temperature can be set between 35 ˚C and

300 ˚C at a flow rate of 0 L/min to 12 L/min. For all the experiments, the nebulizer gas pressure was maintained at 10 psi, the drying gas temperature was kept at 300 ˚C and the drying gas flow was 8 L/min. All the analyses were recorded in positive mode.

In ESI, the analyte liquid solution is sprayed into the spray chamber at atmospheric pressure. With the Esquire-LC the charged droplets created at the needle tip are driven inside a glass capillary by a voltage difference between the spraying needle and the entrance of the glass capillary. The entrance of the glass capillary was held at - 4 kV relative to the needle. The glass capillary acts as an ion transfer and transmission device between atmospheric pressure and vacuum environment. Once the ions leave the glass capillary, they are focused, transported and injected into the ion trap through the action of a combination of skimmers and octapole lenses. The pressure in the capillary-skimmer is at 1.3 mbar, which facilitates solvent evaporation. The droplet size reduction occurs by two processes. The first process is desolvation, which involves evaporation of the solvent; it starts in the spray chamber and may continue at the skimmers. Since the

skimmers are under vacuum, any drying gas or solvent left will be evaporated at this stage. The second process, droplet fission, is caused by electrostatic repulsion between same charges. Then, the ions are concentrated, transported and focused inside the quadrupole ion trap by the octapoles and two lenses; the pressures at these regions are

0.15 mbar and 0.0016 mbar, respectively. The operation principle of a quadrupole ion trap was discussed in Chapter 2.3.3. Ions ejected from the ion trap reach a Daly detector.

3.2.2. Synapt HDMS Ion Mobility Mass Spectrometer

The Synapt MS system is a quadrupole orthogonal acceleration time-of- flight mass spectrometer commercialized by Waters (Waters, Milford, MA) (see Figure

2.13). This instrument is equipped with an electrospray ionization source and a tri-wave ion mobility region. The operation principle of this instrument and the triwave ion mobility function were explained in Chapter 2.5. The Synapts versatility is documented in Chapters V and VII.

Chapter V describes the characterization of PLA biodegradation products.

Samples of PLA degradation products were introduced to the ESI source by direct infusion. The instrument was operated in positive mode with a capillary voltage of 3.5 kV; cone voltage of 35 kV; sampling cone voltage of 3.2 kV; source temperature of 120

˚C; desolvation temperature of 250 ˚C; and sample flow rate of 10 μL/min. The ion mobility section was turned off. Conventional CAD experiments were performed by isolation of the precursor ion in the trap cell with the IMS section turned off. The

following CAD parameters were selected: trap collision energy (CE), 40 eV; transfer

CE, 4 eV; and trap gas flow of 1.5 mL/min (Ar).

Chapter VII describes the characterization of a poly(α-peptoid)s system. Both ESI

MS and ESI traveling wave ion mobility (TWIM) experiments were performed.72 The following ESI and TWIM parameters were selected: ESI capillary voltage, 3.5 kV; sample cone voltage, 35 V; extraction cone voltage, 3.2 V; desolvation gas flow, 500 L/h

(N2); trap collision energy (CE), 2 or 6 eV; transfer CE, 1 or 4 eV; trap gas flow, 1.5 mL/min (Ar); IM gas flow, 22.7 mL/min (N2); sample flow rate, 5 μL/min; source temperature, 40 or 100 °C; desolvation temperature, 60 or 150 °C; traveling wave velocity, 350 m/s; traveling wave height, 8.5, 9.5, or 11 V, depending on the m/z ratio to be separated. Conventional MS2 (CAD) spectra were acquired in the trap cell with the

TWIM device turned off. MS2 (CAD) experiments combined with TWIM separation were performed in the trap cell (fragmentation before IM separation) or in the transfer cell (fragmentation after IM separation). All MS2 studies employed 60 eV collisions with

Ar targets.

3.2.3. MALDI-Q/ToF Mass Spectrometer

A Waters (Milford, MA) MALDI tandem mass spectrometer (Micromass Ultima) was used in this dissertation for the characterization of poly(ethylene glycol) polymers.

As described in Chapter IV, the Q/Tof instrument was mainly used for MS2 experiments.

The Micromass Q/ToF Ultima mass spectrometer is composed of quadrupole and time- of-flight analyzers. The quadrupole analyzer works as either ion guide or mass selector.

To perform MS2 experiments, the quadrupole is set in mass-selective mode. In this mode,

rf and DC voltages are applied to the quadrupole and only the ion of interest travels through for two-stage mass analysis. To operate the quadrupole as an ion guide, it is set in rf-only mode. The reflectron analyzer is placed orthogonally to the quadrupole and serves as a mass resolving device for both MS and MSMS experiments. The Micromass

Ultima is equipped with a N2 laser that irradiates the analyte sample at a wavelength of

337 nm.

Two-stage mass analysis also employs a collision cell located after the quadrupole device. The collision cell is pressurized with argon gas. Once the desired precussor ion enters the collision cell, the collision energy (CE) is adjusted to observe fragment ions.

Adjustment of the precursor ion CE promotes energetic collisions between this ion and the argon particles. For the MS2 experiments in Chapter IV, the CE was generally set between 80 – 150 eV to obtain the desired fragmentation. The appropriate CE value corresponds to the value where the precursor ion was still visible and the fragment ions were well resolved. All ions are detected by a microchannel plate detector. All the spectra were recorded in positive mode.

3.2.4. Ultraflex III ToF/ToF Mass Spectrometer

Another MALDI mass spectrometer used in this dissertation is the

Ultraflex III ToF/ToF mass spectrometer (Bruker Daltonics, Billerica, MA). It consists of two time-of-flight analyzers, a short linear (ToF-1) interfaced with a reflectron analyzer

(ToF-2). In MS2 mode, the first ToF segment is used for selection of the precursor ion.

The second ToF segment separates the fragment ions produced from the selected precursor by increasing the laser power and by collisions in the region between ToF-1

and ToF-2 (LIFT mode). In MS mode ToF-1 and ToF-2 operate as one combined linear or reflectron ToF analyzer. The smartbeam 200 Hz laser (Nd:YAG), incorporated in

Ultraflex III, emits light at the wavelength of 355 nm. All spectra in Chapter IV and VII were measured in positive reflectron mode. On the other hand, all spectra presented in

Chapter VI were acquired in positive linear mode.

The ions were accelerated inside the drift tube at 25 kV (IS1). Additional important parameters in MS mode were: laser energy was adjusted as needed; IS2 voltage at 21.65, lens voltage at 9.65 kV and delay time 150 ns. In reflectron mode experiments, the reflectron lenses 1 and 2 were set at 26.30 and 13.70 kV, respectively.

For MS2 experiments, the set up for the LIFT parameters was as follows: IS1 at 8 kV, IS2 at 7.15 kV, lens potential at 3.6 kV, reflectron 1 and 2 lenses at 29.50 and 13.85 kV, respectively, and the LIFT 1 and 2 settings at 19.00 and 2.90 kV, respectively.

CHAPTER IV

MASS SPECTROMETRY CHARACTERIZATION OF

POLY(ETHYLENE GLYCOL)S SYNTHESIZED BY GREEN CHEMISTRY

4.1. Background

In recent years, the effort of the polymer synthetic community to work in a

“greener atmosphere” and reduce/eliminate the use of organic solvents and/or organic catalysts has had its reward with the green polymer chemistry. Polymers can now be synthesized without the use of organic catalysts which are replaced with enzymes.73-75

Enzymatic catalysts for organic synthesis have made it possible to perform traditional organic reactions under “green chemistry” conditions, and this has led to the development of environmentally friendly synthetic routes to drug delivery systems. 76-77

Enzymes have many advantages, such as enviromental acceptability,78-79 recyclability80 and high efficiency,81-82 which have made them popular for the synthesis of polymers. Lipases are enzymes that are able to catalyze many reactions, including hydrolysis,77 esterification,83 amidation84 and transesterification.85 The research group of

Judith E. Puskas uses Candida Antarctica lipase B (CALB) enzyme as a catalyst in polymer syntheses. They employ the ability of CALB to perform transesterification86-87

and Michael addition88-89reactions to modify poly(ethylene glycol)s (PEG) and tetra(ethylene glycol)s (TEG) for drug delivery purposes.

PEG is used in various biomedical applications because of its non-toxicity,90 non- immunogenicity90, non-antigenicity,91 and water solubility.92-93 Additionally, the hydrophilic nature of PEG makes it an important tool in the medical/ pharmaceutics field

(for example, for drug delivery systems) because many drugs have poor water solubility and PEG helps to overcome this limitation.94 Consequently, PEG is one of the most frequently used polymers in biomedical applications.

Since not many polymer functionalizations involving enzymes have been reported, the precise characterization of the enzyme-catalyzed functionalized PEG/TEG biomaterials and their conjugates achieved from our mass spectrometry techniques will be valuable. Previously, the characterization of polymers generated by green polymer chemistry has been performed primarily by chromatographic techniques (i.e. SEC/GPC)

95 and spectroscopic methods (i.e. NMR, UV-Vis).96,27 Mass spectrometry (MS) techniques have been used before, but not as the main technique and no tandem mass spectrometry (MS2) studies have been performed. As will be shown here, MS2 experiments help to confirm the polymer composition and provide an accurate identification of its end groups.

In order to describe the fragments obtained in the MS2 studies in a correct way, the detailed polymer nomenclature introduced by Wesdemiotis et al.21 was used.

Collisionally activated dissociation (CAD) fragments were designated according to the nomenclature shown in Scheme 4.1. PEG has three types of backbone bonds and, thus, is

able to create six homologous fragments series by CAD. To name the six fragment ions formed from backbone cleavages, the letters a-c and x-y are used (see Scheme 4.1.(a)). If after the cleavage the charge is retained on the initiating (α) side, an, bn and cn ions are formed. If after the cleavage the charge is retained on the terminating (ω) yside, xn, n and zn ions are formed. Some of the enzyme-catalized funtionalized polymers analyzed and discussed in this chapter have identical end groups. When both of the end groups are the same, the initiating (α) and terminating (ω) sides cannot be distinguised, and only an, bn and cn ions will be created by backbone cleavages. Any other types of fragments will be explained futher in the chapter.

x3 y3 z3 x2 y2 z2 x1 y1 z1 ω H2 H2 H2 H2 H2 H2 a) RO C C O C C O C C O R' α a b 1 1 c1 a2 b2 c2 a3 b3 c3

b3 a3 c2 b2 a2 c1 b1 a1 α H2 H2 H2 H2 H2 H2 b) RO C C O C C O C C O R α a1 b1 c1 a b2 c2 a b c 2 3 3 3

Scheme 4.1. Nomenclature schemes: (a) nomenclature scheme for the MS2 fragments of polymers with different end groups, (b) adjusted scheme for the MS2 fragments of polymers with identical end groups.

In this chapter, the challenge is to develop a protocol for the complete and precise characterization of these attractive biomaterials. The studies performed include the products from the first until the final step of the synthesis; such complete characterization

should provide a better understanding of the molecular structures generated in each synthetic step. Different MS techniques, including electrospray ionization quadrupole ion trap (ESI-QIT) mass spectrometry and matrix-assisted laser desorption ionization time- of-flight (MALDI-ToF/ToF) mass spectrometry were used. MS techniques have the potential to provide structural and molecular weight information in a short time using trace amounts of sample. This project produced detailed information that could not be obtained by other analytical techniques.

4.2. Sample preparation and instruments used

Results were obtained using MALDI-ToF MS (Bruker Ultraflex-III ToF/ToF mass spectrometer), MALDI-Q/ToF MS (Waters Micromass Q/ToF Ultima mass spectrometer) and ESI-QIT MS (Bruker Daltonics Esquire-LC ion trap mass spectrometer). For all analyses, the samples were dissolved in tetrahydrofuran (THF).

Ionization was achieved using two different types of salts: sodium trifluoroactate

(NaTFA) and lithium trifluoroactate (LiTFA). LiTFA was primarily used in MS2 experiments. In the MALDI experiments, trans-2-[3-(4-tert-butylphenyl)-2-methyl-2- propenylidene] malononitrile (DCTB) served as matrix. Solutions of the matrix (20

μg/μL), the sample (10 μg/μL), and salt (10 μg/μL) were mixed in the ratio 10:2:1, respectively. In ESI experiments, sample and salt solutions were prepared at a concentration of 1 μg/μL and mixed in the ratio 100:1 (sample:salt) (v/v). All spectra were collected in positive mode.

4.3. Characterization of PEG/TEG Biomaterials

This chapter reports the procedure developed for the analysis of polymers that were prepared using green processes. A summary of the structures analyzed in this chapter are shown in Scheme 4.2. The Scheme illustrates the enzyme-catalyzed functionalization of PEGs with various vinyl esters (vinyl methacrylate, vinyl acrylate and divinyl adipate). These functionalized materials were synthesized via transesterification reactions. The product of the Michael addition of diethanolamine to

TEG diacrylate was characterized as well.

HO O OH N O O N HO n O OH

H N HO OH O O O O

O O O O O O n n

O O

O O O H RO O n (R = H, CH3)

O O O O

O O O O O n O

Scheme 4.2. Enzyme-mediated funtionalization of PEG.

4.3.1. Characterization of PEG/TEG Diacrylate by ESI-QIT and MALDI ToF/ToF MS

The mass spectrum of TEG diacrylate will be first discussed. TEG diacrylate was prepared by the transesterification of vinyl acrylate (CH2CHCO2CHCH2) with HO-

(CH2CH2O)4-OH in the laboratory of Professor Judith E. Puskas. The ESI-QIT mass spectrum of TEG diacrylate, which has a molecular weight of 302 Da, was acquired with a Bruker Daltonics Esquire-LC ion trap mass spectrometer (Figure 4.1). It shows an intense peak at m/z 325, arising from the TEG diacrylate sodium ion adduct. Na+ originates from the use of a sodium cationizing agent in the sample preparation. The

observed [M+Na]+ ion agrees with the expected TEG diacrylate molecular structure. Two other, minor peaks are produced by ESI of TEG diacrylate, see Figure 4.1. The ion that appears at m/z 341 arises from naturally present potassium in the glass container of the sample. On the other hand, the ion at m/z 271 corresponds to the monosubstituted TEG molecule (only one hydroxy end group in HO-TEG-OH, was replaced by an acrylate group). This monosubstitution is probably due to an incomplete transesterification reaction.

O + O [M+Na] 4 325.1 O

TEGAH [M+K]+ 271.0 341.0 260 280 300 320 340 360 380 m/z

Figure 4.1. ESI-QIT mass spectrum of TEG diacrylate. TEGAH = monosubstituted TEG molecule (acrylate/hydroxyl end groups, 72 Da)

To corroborate the sample composition and to confirm the sample end groups,

MS2 experiments were performed. For this, the sodiaded TEG diacrylate was fragmented by CAD in the QIT mass spectrometer (see Figure 4.2). The central fragmentation of 51

TEG diacrylate proceeds by a 1,5-H rearrangement (see Squeme 4.3), leading to the elimination of acrylic acid (CH2=CHCO2H, 72 Da) and the formation of the b4 ion (m/z

253; the subscript gives the number of repeat units remaining in the fragment).21,22 The

” less abundant fragment corresponds to the c3 ion which originates by the loss of 98 Da

(CH2=CHCO2CHCH2, vinyl acrylate molecule) from the precursor ion. The formation of

” c3 is facilitated by the ability of TEG (and PEG) to form salt bridges which promote charge-induced fragmentation (see Scheme 4.4).21,22

b4 253.0

[M+Na]+ 325.1

” c3 227.0

240 260 280 300 320 m/z

Figure 4.2. ESI-CAD mass spectrum of sodiaded TEG diacrylate, acquired with the Bruker Daltonics Esquire-LC ion trap mass spectrometer. The precurson ion was subjected to CAD with helium at an rf excitation amplitude of 0.75 V.

O O 1,5-H O rearrangement O O O O O H n-1 n-1 bn

Scheme 4.3. Charge-remote 1,5 rearrangement to form bn fragment ions from TEG diacrylate by the expulsion of acrylic acid (72 Da). 52

Na+ O H O O C O O n-2 H O

rH-

+ Na+ Na O + H rH O O O O O C O n-2 H O O - n-2 O O " cn O H3C + ” Scheme 4.4. Charge-induced fragmentation pathway induced by Na to form cn fragment ions from TEG diacrylate by the eliminaton of vinyl acrylate (98 Da).

PEG diacrylate was also examinated. Similar results were obtained when the same processes were applied to HO-PEG-OH (Mn = 1,000 g/mol). Due to the higher molecular weights the mass spectrum was measured using the Bruker Ultraflex-III ToF/ToF mass spectrometer. Figure 4.3 shows the spectrum obtained with a sodium salt as the cationizing agent. Two different distributions containing oligomer ions 44 Da (-

CH2CH2O-) apart from each other were identified. The main distribution corresponds to the sodiated oligomer of PEG diacrylate. Upon closer examination, the main distribution detected ranges from the 13-mer at m/z 721 to the 32-mer at m/z 1557. It is futher observed that the main distribution peaks near 1100 Da, close to the average molecular weight measured by GPC. The minor distribution arises from PEG polymer with acrylate/acetyl end groups (CH2=CHCO2-/-OCCH3, 114 Da), which was labeled as

PEGacryl/acet. CAD studies to corroborate the end groups of the small distribution could not be performed due to the low abundance of its ions.

O O n O [M+Na]+ 1073.6, 21-mer 1073.6

1117.6

13-mer 32-mer

800 900 1000 1100 1200 1300 1400 1500 m/z 44 Da

PEG Acryl/Acet 44 Da 1105.6 1149.7

1070 1080 1090 1100 1110 1120 1130 1140 1150 m/z Figure 4.3. Partial MALDI-ToF mass spectrum of PEG diacrylate, acquired with a Bruker Ultraflex-III ToF/ToF mass spectrometer. The spectrum shows the m/z 1073- 1158 region where the 21- and 22 mer were detected. The full spectrum is shown in the insert. Na+ was used for cationization.

Figure 4.4 shows the MALDI-CAD mass spectrum of the sodiaded 22-mer from diacrylated PEG (two acrylate end groups), measured on the Waters Micromass Q/ToF

Ultima Mass Spectrometer. All fragment series marked in the spectrum contain -

CH2CH2O- (44 Da) repeat units. The product ion at m/z 1073 corresponds to the loss of one monomer unit, ethylene oxide (-C 2H4O-). The spectrum shows two homologous distributions which were labeled as series bn and series cn. The bn series results from the nominal loss of 44n + 72, leading to vinyl/acrylate (98 Da) end groups. Consecutive

charge-remote 1,5-H rearrangements account for the bn ion series formation. The first such rearrangement proceeds as shown in Scheme 4.3 for TEG diacrylate and produces the largest bn fragment, i.e. b21 (m/z 1045.9). Consecutive 1,5-H rearrangements at the vinyl chain end releases CH3CH=O (Scheme 4.5), producing a smaller bn fragment, i.e. b20; continued fragmentation by this mechanism provides a route to the smaller members of this series. Alternatively, consecutive 1,5-H rearrangements can occur at the second acrylate chain end to yield a fragment with two vinyl end groups, i.e. CH2=CHO-

(CH2CH2O)n-CH=CH2. Unlike the terminal fragments, which contain one original end group (Scheme 4.1), this fragment is internal, missing both original end groups. Loss of both acrylate substituents according to Scheme 4.3 creates the largest possible internal

VV fragment with two vinyl end groups, i.e. J22 ; J is used to designate internal fragments,

VV while the superscript designates the end groups (vinyl/vinyl, 70 Da). The J22 fragment can undergo sequential losses of CH3CH=O, via 1,5-H rearrangements, to generate the smaller members of this fragment distribution.

The dissociation channels discussed take place at the end groups. Both cleavages in the PEG chain are, however, also possible via change-induced (Scheme 4.6) or charge- remote (Scheme 4.7) pathways. Both can yield bn fragments with vinyl/acrylate (98 Da) end groups. Charge-induced fragmentation cogenerates the cn” series with hydroxyl/acrylate end groups (combined mass of 72 Da), whereas charge remote fragmentation generates the cn series with acrylate/carbonyl end groups (114 Da).

Consecutive elimination of acrylic acid via 1,5-H rearrangements according to Scheme

VV 4.3 leads to internal fragments, viz. Jn with vinyl/vinyl (70 Da) end groups (from bn),

HV ” CV Jn with hydroxy/vinyl (44 Da) en groups (from cn ), and Jn with carbonyl/vinyl (86 55

Da) end groups (from cn), all of which are present in the CAD spectrum of Figure 4.4.

VV Note that series cn and Jn have the same nominal masses and, thus, overlap.

Precursor ions with sufficient internal energy may also undergo consecutive

CC cleavages in the PEG chain. Such fragmentation can explain the minor Jn series with two carbonyl end groups (end groups mass of 102 Da). The very small relative intensity of this series (Figure 4.4) indicates though, that this process is much less competitive than the alternative channels described.

b5 385.3 c 341.2 5 401.3

” 357.2 c6 * bn = c “

3 359.3 cv J6 + * HV [M+Na] VV ө J7 ө J + c = 271 CC n n J5 # * 1117.9 ^ # 373.3 cv ^ 403.3 417.3 Jn = ө 375.3 345.3 419.3 355.3 389.3 ” cn = * 340 350 360 370 380 390 400 410 420 m/z CC Jn = ^

HV 533 Jn = # 577 621 489 445

* 665 709 401 357 753 313 b22

VV 1045.9 797 ө * 22

ө J + * 22 C ө 841 ө ө 885 ө ө ө ө * * өө ө ө 929

ө 973 ө ө ө

* ө M- EG

50 150 250 350 450 550 650 750 850 950 1050 1150 m/z Figure 4.4. MALDI-CAD mass spectrum of the sodiaded 22-mer from a diacrylate terminated PEG (two acrylate end groups). The spectrum was acquired with a Waters Micromass Q/ToF Ultima mass spectrometer. The insert shows an expanded view of the m/z 340-420 region. Argon was used as collision gas at a laboratory-frame collision energy of 110 eV.

O Na Na

O O H O O n-2 -CH3CH=O n-2 O O bn bn-1

Scheme 4.5 Charge-remote 1,5-H rearrangement in vinyl-terminated PEG to yield a truncated vinyl-terminated chain by loss of a repeat unit in the form of acetaldehyde.21,22

Na O O

O-(CH2CH2O)-CH2-CH2-O-CH2-CH-O-(CH2CH2O) H

(1) C-O bond (2) rH- cleavage

O O Na Na CH =CH-O-(CH CH O) O(CH2CH2O) CH2CH2-OH or 2 2 2 n x-n-2

" cn bn

Scheme 4.6. Charge-induced fragmentation in the PEG chain to form truncated chains ” carrying one original end group (acrylate) and either a new hydroxyl (cn ) or a new vinyl 21,22 (bn) end group. See Scheme 4.4 for a more detailed mechanism.

O Na

O (CH CH O) CH CH O CH CH O (CH CH O) 2 2 n-1 2 2 2 2 2 2 x-n-1

O (1)C C (1)O C (1)C O (2)-O=CH2 (2)-H (2)-H (3)-H

O Na (CH2CH2O) CH2CH=O n-1

O cn O Na (CH2CH2O) CH=CH2 n-1

O bn Scheme 4.7. Charge-remote fragmentation in the PEG chain involving homolytic bond cleavages and consecutive β scissions to give truncated chains carrying one original end 21,22 group (acrylate) and either a new vinyl (bn) or a new carbonyl (cn) end group.

4.3.2. Characterization of tetrahydroxy substituted dendritic TEG by ESI-QIT

Tetrahydroxy substituted dendritic TEG (TEG 512) was created after TEG- diacrylate was successfully synthetized and completely characterized. For this, di-(2- hydroxyethyl) amine ((HOCH2CH2)2NH) was mixed with TEG-diacrylate to obtain TEG

512 (512.6 Da) via enzyme-catalyzed Michael addition. The reaction products were analyzed by ESI MS using the Bruker Daltonics Esquire-LC ion trap mass spectrometer ans Na+ for cationalization. Figure 4.5 shows the resulting mass spectrum which contains peaks at m/z 271.0, 325.1, 376.1, 430.1, 513.3, and 535.2. The peaks at m/z 513.3 and

535.2 Da agree with [M + H]+ and [M + Na]+ of TEG 512, respectively. The other peaks correspond to fragments from these ions. Tandem (MS2) experiments confirmed these assignments (vide infra).

HO O OH N O O 4 N + HO [M+Na] O OH 535.2

[F+Na]+ 430.1 [F+Na]+ [F+Na]+ [F+Na]+ 376.1 [M+H]+ 271.0 325.1 513.2

250 300 350 400 450 500 m/z Figure 4.5. ESI-QIT mass spectrum of TEG 512. F = fragments from sodiaded TEG 512 created inside the mass spectrometer; M = TEG 512.

Two different mechanisms, a retro-Michael reaction97 and an intramolecular transesterification,98 are proposed to explain the ions observed upon ESI-CAD of sodiaded TEG 512, see Figure 4.6.(b). The ions at m/z 430 and 325 can be produced by retro-Michael reactions initialized by the lone pair of electrons at the amine terminal sides of TEG 512 (see Scheme 4.8). These reactions involve abstraction of a proton in α

position to the carbonyl group and C-N bond cleavage to form sodiaded fragments with a molecular weight of 430 Da or 325 Da by release of one or two molecules of di-(2- hydroxyethyl) amine ((HOCH2CH2)2NH), 105 Da), respectively. Retro-Michael reactions on dedrimers have been previously observed, generating asymmetrical products and/or a mixture of products.99

The product ions at m/z 376 and 271, in Figure 4.6.(a), are formed by the loss of

HO N O 159-Da molecules ( O ) from the precursor ion. TEG 512 includes free hydroxyl groups at its termini, which can cause intramolecular transesterifications that degrade the polymer by detaching cyclic monomers units.100 In this process, an ester bond is broken due to intramolecular attack by the hydroxy end group (Scheme 4.9). Concomitant proton transfer leads to the elimination of a 159-Da cyclic moiety and a new truncated molecule, see Scheme 4.9. The losses of 105-Da and 159-Da molecules, provide corroborating evidence that the synthesized dendrimer indeed contained the desired termini, on which attachment of diagnostic and therapeutic agents is planned.

F 325.1

F 430.1

[M+Na]+ 535.2 F 271.0 c4 376.1 250 300 350 400 450 500 m/z

b) O O O 4 O m/z 325 F

OH (retro Michael) - 2 x HN -210 Da OH HO N OH - O HO O O O OH HO O N N O 4 intramolecular O4 N OH HO O m/z 376 transesterification m/z 535 OH c4 -159 Da

OH (retro Michael) - HN OH HO -105 Da N - O HO O O O HO N O O 4 O 4 intramolecular HO O m/z 271 transesterification m/z 430 F -159 Da F Figure 4.6. (a) ESI-CAD mass spectrum of the sodiaded tetrahydroxy substituted dendritic TEG (TEG 512), acquired with an ESI-an rf excitation amplitude of 0.70 V. (b) MS2 fragmentation scheme, rationalizing the fragments observed. All ions contain Na+ (not included in the scheme).

H O + HO Na O H N N O O C C HO 4 H H2 O O H

retro Michael addition

HO +Na HO O N O O C CH2 + NH HO 4 H O HO 105-Da loss Scheme 4.8. Elimination of a 105-Da molecule from cationized TEG 512 as a result of a retro-Michael reaction.

Na+ O C H2 H2 O C C O

N H2 H HO H C C 2 O

intramolecular transesterification HO

N Na+ + O HO O O 159-Da loss Scheme 4.9. Elimination of a 159-Da molecule from cationized TEG 512 as a result of a intramolecular transesterification reaction.

4.3.3. Characterization of TEG Dimethacrylate by ESI-QIT MS

The ESI spectrum shown in Figure 4.7 was obtained from a TEG dimethacrylate solution. TEG dimethacrylate was prepared by transesterification of vinyl methacrylate with HO-TEG-OH in the Puskas research group. This spectrum was collected using ESI on a Bruker Daltonics Esquire-LC ion trap mass spectrometer. It shows a single peak corresponding to the sodiaded adduct of TEG dimethacrylate molecule.

O [M+Na]+ O O 353.2 O 4

225 250 275 300 325 350 375 400 425 450 m/z Figure 4.7. ESI-QIT mass spectrum of TEG dimethacrylate.

An MS2 experiment was performed to verify the structure and end groups of the product. The MS2 spectrum of [TEG dimethacrylate+Na]+ (m/z 353) displays structurally characteristic peaks (Figure 4.8). The intense ion observed at m/z 267 is the b4 fragment with combined end groups of 112 Da (vinyl/methacrylate), arising by the elimination of a

86-Da molecule (methacrylic acid) from the precursor ion via the 1,5-H rearrangement

” mechanism (Scheme 4.3). The second most dominant fragment, which was labeled as c3

( m/z 241) has a combined end group mass of 86 Da (methylacrylate/hydroxyl). It is accounted for by charge-induced fragmentation according to Scheme 4.4. This reaction releases vinyl methacrylate (112 Da) which is also observed, in protonated form, at m/z

+ 113; the later ion corresponds to a b1 fragment (recall that charge-induced dissociation

” generates cn -type and bn-type fragments, cf. Scheme 4.6). Finally, the weaker ion at m/z

283 was assigned to a c4 fragment with methacrylate/carbonyl end groups (128 Da), resulting from charge-remote C-O cleavage in the TEG backbone. Such reactions have high energy requirements and, therefore, are less competitive in CAD experiments with

QITs which deposit lower average internal energies as compared to CAD experiments in

Q/ToF instrumentation.21 Overall, the MS2 results verify the presence of two methacrylate end groups in the m/z-353 precursor ion.

b4 267.0

+ ” [M+Na] c3 + 353.2 b1 241.0 c4 113.2 283.0

125 150 175 200 225 250 275 300 325 350 m/z Figure 4.8. ESI-CAD mass spectrum of the sodiaded TEG dimethacrylate, acquired with an ESI-QIT mass spectrometer. The precursor ion was subjected to CAD with helium at an rf excitation amplitude of 0.75 V. A superscripted + indicates that the ion does not contain Na+ (ionized by H+).

4.3.4. Characterization of mPEG-VA by MALDI-ToF/ToF MS

Finally, mPEG-VA was analyzed using the Bruker Ultraflex-III ToF/ToF mass spectrometer; this polymer was prepared by transesterification of divinyl adipate (DVA) with methoxy-PEG by Puskas et al. using CALB as a biocatalyst. The mass spectrum of this polymer showed a Poisson molecular weight distribution maximizing 2230 Da, close to the average molecular weight measured by GPC. Two different polymer distributions are observed, see Figure 4.9. Both of the distributions show oligomer ion series that are

44 Da mass of PEG repeat unit apart from each other. The more abundant distribution belongs to sodiaded mPEG-VA homopolymer, for which oligomers from the 30-mer at

m/z 1530 to the 63-mer at m/z 3027 are detected. Based on their m/z values, these oligomers have end groups of 186 Da which agrees with the expected CH3O- and

CH2=CHCO2(CH2)4CO- termini. The less abundant distribution is attributed to protonated mPEG-VA homopolymer. This is consistent with the 22-Da mass difference between the two distributions. The protonated series of oligomers arise by proton transfer from the matrix to the sample, while the sodiaded oligomers are due to the use of sodium trifluoroacetate as cationizing agent.

O O O O n O 2190.6, 45-mer [M+Na]+ 2234.7 2190.6

30-mer 63-mer

44 Da 1400 1600 1800 2000 2200 2400 2600 2800 3000 m/z

[M+H]+ 2212.7

44 Da 2256.7

2190 2200 2210 2220 2230 2240 2250 2260 2270 m/z Figure 4.9. Partial MALDI-ToF mass spectrum of mPEG-VA, acquired with the Bruker Ultraflex-III ToF/ToF mass spectrometer. The spectrum shows the m/z 2190-2272 region where the 45- and 46-mer are detected. The full spectrum is shown in the insert. Na+ was used for cationization. 67

For MS2 analysis, the lithiated 45-mer from mPEG-VA was selected and the spectrum obtained is displayed in Figure 4.10. In the upper mass region, one intensive signal next to the parent peak can be observed. This intense signal is assigned to the fragment formed by cleavage of the vinyl ester end group and expulsion of acetaldehyde

(44 Da) via a 1,5-H rearrangement (see Scheme 4.10, route 1) . A 1,5-H rearrangement is also possible at the mPEG ester group (route 2 in Scheme 4.10); it releases vinyl adipate

+ to form the largest possible bn-type fragment, viz. b45 (m/z 2002.7). Both the [M + Li]

+ precursor ion (m/z 2174.6) as well as the major [M + Li – CH3CHO] fragment (m/z

2130.3) dissociate extensively by loss of C4H8O2 (88 Da). Earlier studies provided evidence that the 88-Da molecule is dioxane.101 The elimination of dioxane is also observed from the b45 fragment (cf. Figure 4.10).

Four homologous series of terminal fragment ions (i.e. fragments containing one original end group) are detected with low intensity in the middle m/z region of the MS2 spectrum (m/z 989 – 1070). Two, viz. bn and cn, carry the CH3O chain end, and the other two, viz. xn and zn, carry the VA (vinyl adipate) chain end. These structurally diagnostic fragments are rationalized by charge-remote hemolytic cleavages along the PEG

. . backbone and consecutive β scissions of H (or CH2=O + H ), as illustrated in Scheme

4.11. Interestingly, the largest members of series xn, cn and zn (n = 45).

Loss of dioxane (88 Da) from the precursor ion Loss of dioxane (88 Da)

from the ion at m/z 2130.3 b22

Loss of dioxane (88 Da) c22 b23

989.8 c23 from the ion at m/z 2002.7 (b45) 1005.9 1033.9 1049.9 M-44

z19 x19 z20 2130.3

1013.9 x20 . 997.8 b2 1057.9 HV

1041.8 J23 J HV

154.2 22 1064.0 72.9 . b3 1019.9 . a3 HV EV 198.3 J J4 EV 4 . J3 c2 -2 184.3 255.3 227.3 148.9 2042.7 211.3 990 1000 1010 1020 1030 1040 1050 1060 1070

168.3 m/z . b4 a2 241.3 + . 140.2 c1 -2 HV J3 b3 EV c2 . J2 b4 [M+Li]

124.2 b2 c4 c3 c45 108.4 2174.6 b45 2086.3 2002.7 # 45 80 100 120 140 160 180 200 220 240 260 # m/z x 800 1000 1200 1400 1600 1800 2000 m/z Figure 4.10. MALDI-CAD mass spectrum of the lithiated 45-mer from mPEG-VA. The spectrum was acquired with a Bruker Ultraflex-III ToF/ToF mass spectrometer. The inserts show expanded traces of a low and an intermediate m/z region. The ions marked by # are due to a trace of protonated polymer co-transmitted with the lithiated product.

Li 2 1 H CH3O (CH2CH2O) O H n-1

O O

1 2 O

-CH CH=O 3 1,5-H rearrangements -HO2C(CH2)4CO2CH=CH2

O Li Li

CH3O (CH2CH2O) O CH3O (CH2CH2O) CH=CH2 n n-1 C

m/z 2130.3 (n = 45) m/z 2002.7 (b45)

Scheme 4.10. Charge-remote 1.5-H rearrangement at the adipate groups of mPEG-VA.

In the lower m/z region (insert up to m/z 260), the MS2 spectrum showed small

. . internal and radical fragments. The radical fragments bn and cn originate from C-O bond

. homolysis where as an fragments are due to C-C bond homolysis. Only the small members of these radical ion families survive and are observed. Their larger homologs react consecutively via β scissions, as depicted in Scheme 4.11, generating the closed- shell series dominating the middle range of the MS2 spectrum (vide supra). Consecutive depolymerization of the larger radical ions is also possible, until they reach the small sizes that cannot readily dissociate further. Finally, the larger radical ions may also undergo backbiting reactions, cf. Scheme 4.12, which explain the internal series with

EV HV ethyl/vinyl end groups (Jn ) and hydroxyl/vinyl en groups (Jn ); the internal fragments are most abundant in the low m/z range of the MS2 spectrum, although traces are also detected in the middle region (cf. Figure 4.10).

O Li+ zn xn

CH3O (CH2CH2O) CH2 -CH2 - O - CH2 - CH2 - O - (CH2CH2O) O x y bn cn

(1) C - O (1) O - C . . (2) -H (2) -H

Li+ Li+

CH3O (CH2CH2O) CH=CH2 CH3O (CH2CH2O) CH2CH=O x x

bn cn EG = 58 Da EG = 74 Da

+ + Li+ Li+ O O

O=CHCH2O - (CH2CH2O) O CH2 =CHO - (CH2CH2O) O y y O O

zn xn EG = 214 Da EG = 198 Da

Scheme 4.11. Charge-remote fragmentation of lithiated mPEG-VA via hemolytic C-O bond scissions in the PEG chain, followed by β-H. eliminations, to form terminal

fragment series bn, cn, xn and yn. Homolitic C-C bond scissions followed by β losses . CH2=O + H provide an additional route to bn and xn fragments. EG designates the mass of the combined end groups. Note that the subscripts in the fragment nomenclature include monomer portions that have been incorporated in the end group; for example, x = 21,22 21 for b22, y = 18 for z19, etc.

CH3O - (CH2CH2O) - CH2CH2

bn backbiting

CH O - (CH CH O) - CH - CH - O - CH - CH - O - CH - CH - O - (CH CH O) - CH CH 3 2 2 x 2 2 2 2 2 2 2 y 2 3

βscission

- H CH O - (CH CH O) - CH - CH - O c 3 2 2 x 2 2 CH3O - (CH2CH2O) - CH2 - CH = O n x + 1 + or +

CH2 = CH - O - (CH2CH2O) - CH2CH3 y + 1 CH2 - CH2 - O - (CH2CH2O) - CH2CH3 y

EV Jn

. Scheme 4.12. Backbiting in the bn radical ions resulting from hemolytic C – O bond cleavages in the PEG chain, ultimately leading to internal fragments with ethyl/vinyl (72 EV Da) end groups, and shorter cn fragments. Analogous reactions sequences generate Jn + . HV . HV . zn fragments from xn , Jn + bn fragments from cn , and Jn + xn fragments from zn .

4.4. Conclusion

The combination of MS and MS2 was used to characterize the enzymatic tranesterification of vinyl methacrylate, vinyl acrylate and divinyl adipate with PEGs and the Michael addition reaction of diethanolamine to TEG diacrylate. The PEG polymers from enzyme-catalyzed functionalization could be characterized with high sensitivity using these mass spectrometry techniques. The first molecule analyzed was TEG diacrylate whose electrospray ionization produced an intense ion at m/z 325 corresponding to [TEG + Na]+ adduct. In the ESI mass spectrum of TEG diacrylate, two other minor peaks were also observed, arising from the potassiated adduct and the monosubstituted molecule. On the other hand, when PEG diacrylate was studied two distributions were detected. The main one results from the desired product, observed as

[polymer + Na]+ at m/z 44n + 126 . The minor distribution is due to an acrylate/acetate by product, appering at m/z 44n + 114. CAD experiments confirmed the presence of diacrylate end groups in the main product. The fragmentations of both TEG and PEG materials can be explained by charge-remote and charge-induced processes. Charge- remote free radical cleavages were observed only in MALDI-CAD experiments.

Additionally, MALDI-CAD led to internal fragments due to secondary fragmentations from existent ions.

An ethylene oxide based dendrimer core with a molecular weight of 512 Da was examined in this chapter. The ESI MS experiments demonstrated the successful synthesis of tetrahydroxy substituted TEG by the presence of [M + Na]+ and [M + H]+ quasimolecular ions in its mass spectrum. Four additional ions observed in the ESI mass spectrum could be identified, with the help of ESI-CAD experiments, as fragments of the 73

tetrahydroxy substituted TEG. Retro-Michael addition and intramolecular transesterification reactions were responsible for the fragmentation pattern of the dendritic TEG product.

The analysis of TEG dimethacrylate demonstrated the ability of CALB to catalyze transesterification to a single product, observed as [TEG + Na]+ at m/z 353 in the ESI spectrum. ESI-CAD of TEG dimethacrylate rendered similar results to those obtained for

” PEG/TEG diacrylate. The CAD experiments produced b n, cn, and cn fragments via

HH charge-remote and charge-induced cleavages. Also a small internal fragment (Jn ) was observed due to sequential decomposition of initially formed fragment ions.

. Finally, the MALDI mass spectrum of mPEG-VA displayed two different distributions composed of [mPEG-VA + Na]+ and [mPEG-VA + H]+ adducts. In analogy to PEG/TEG diacrylate and TEG dimethacrylate, the fragmentation of this polymer proceeds by a combination of charge-induced and charge-remote cleavages. Additionally, the MALDI-CAD spectrum showed radicals ions, as a result of C-C and C-O bond cleavages, as well as internal Jn series, due to consecutive fragmentation and backbiting reactions.

In summary, an effective, fast, sensitive and simple procedure has been developed for the precise elucidation of chemical structure and end groups of samples synthesized via enzyme-catalyzed transesterification and Michael addition.

CHAPTER V

INVESTIGATION OF THE ENZYMATIC DEGRADATION OF POLY(LACTIDE) BY MASS SPECTROMETRY METHODS

5.1. Background

Synthetic polymers are constantly used in medical (sutures and implants), pharmaceutical (drug release) and environmental areas.102-103 Consequently, polymer degradation studies are important for the selection of the proper polymers in each particular application. The majority of enzymatic degradation studies so far have been examined by weight loss of the decomposing material,104-105 atomic force microscopy

(AFM),106 X-ray diffraction,105,107 and size exclusion chromatography/gel permeation chromatography.107 A disadvantage to these studies is the lack of information about the chemical structure of the degradation products and their relationship to the original polymer’s structure.108

Poly(lactide) (PLA), see Figure 5.1, is a biodegradable and biocompatible thermoplastic with good mechanical properties for use in drug delivery systems.109 The biodegradation of PLA by various enzymes has been extensively studied,110 in particular the enzymatic degradation by proteinase K.106,111 Proteinase K is a serine protease, obtained from Tritirachium album.112 Previous studies have documented the catalytic activity of proteinase K in the degradation of PLA polymer.113 The enzymatic

degradation studies of PLA showed that proteinase K peferientially degrades L-lactide units and armorphous regions.114 These tests have confirmed that degradation occurs, but they did not provide information about the structures of the degradation products.

H3C(H2C)9O (CH2)9CH3 O n

Figure 5.1. Structure of polylactide polymer with di-decyl end groups. Repeating units (C3H4O2) : 72.021129 Da. Polymers used for the enzymatic degradation studies supplied by Sigma.

Knowing that PLA is degradable by proteinase K, it appears interesting to evaluate the biodegradation products. For this reason, ESI mass spectrometry was used in this dissertation to reveal the identity of the enzymatic degradation products of high molecular weight PLA. The goal was to obtain structural information about the degradation products that is not available in the literature. This chapter reports the first characterization of the enzymatic degradation products of PLA by mass spectrometry techniques.

5.2. Sample Preparation, Instrument Used and Enzymatic Degradation

The PLA biodegradation products were analyzed using the Waters Synapt HDMS mass spectrometer and electrospray ionization. PLA with an average molecular weight of

50,400 was obtained from Fluka. A pellet of PLA (0.0100 g) was dissolved in 200 μL of chloroform and the solution was spread out on a petri dish and dried at 4 ˚C for 24 hours to obtain a film. A film with an approximate thickness of 0.06 mm was placed in a 76

scintillation vial containing 5 mL of water, 1.0 mg of proteinase K (Sigma, from

Tritirachium album) and 1.0 mg of sodium azide (NaN3, Sigma-Aldrich, Reagent Plus ≥

99.5%). Water of Optima LC/MS grade was used for these experiments. This water was also used as the incubation medium to eliminate background due to salts (buffers). Water was a suitable solvent since the protein exhibits maximum activity near neutral pH.115

The film/degradation medium incubation was carried out at 37 ˚C in a rotary shaker (100 rpm) for 7 and 10 days. After each incubation time was completed, the enzyme reaction was stopped by submerging the sample vial in boiling water. When the sample reached room temperature, the degradation medium was analyzed. In order to obtain a control sample, the PLA film was placed inside a scintillation vial with the water and sodium azide, but without the enzyme; otherwise, the control vial was subjected to the same conditions. NaTFA was used as the ionizing agent. In the ESI analyses, the sample solution was mixed with methanol to improve ionization. The sample for analysis was prepared as follows: degradation medium and methanol were mixed in the ratio 1:1 (v/v) and NaTFA was added to this mixture until the concentration of 0.1% was added. For each incubation time, three replicate vials were tested.

5.3. ESI-Q/ToF MS Characterization of Products from the Enzymatic Degradation of

PLA

Enzymatic degradation of PLA, using proteinase K, was performed and the degradation products after 7 days gave the mass spectrum shown in Figure 5.2. Peaks up to about m/z 1495 are clearly discerned. At least, eight series of peaks with a repeat unit of 72 Da are evident.

A 207.1

B 391.1

A 135.0 C C C 503.2 431.1

B 575.2 319.1 D D D E C B E 687.2 E A E 615.2 759.2 F 359.1 799.3 279.1 463.2 C 871.3 F F D 727.2 F 983.3 E 943.3 GG F G G B F D 911.3 G E FHG HH 839.3 1055.4 647.22 H H 543.2 655.2 535.2 767.2 831.3 1025.3 1095.4 1127.4 1167.4 1311.5 951.3 1015.4 1239.4 1207.4 1279.4 1351.5 1423.5 H H 1199..4 1495.5 100 200 400 500 600 800 900 1000 1200 1300 1400 1500 m/z 300 700 1100

Figure 5.2. ESI-Q/ToF mass spectrum of the water-soluble degradation products of PLA after 7 days of biodegradation with Proteinase K

MS and MS2 results support the conclusion that these series of peaks correspond

+ to cluster ions containing polylactide + NaOH (PL nNaOH), oligomers and (NaOH)xNa

+ ions. The nomeclature used for these clusters in this chapter is (PLnNaOH)(NaOH)xNa where x = number of sodium hydroxide (63 Da) units and n = number of PLA repeating units (72 Da). The observed degradation products occur because of the alkaline nature of the degradation medium which contained NaN3. As a weak base, NaN3 produces NaOH and hydrazoic acid (HN3) through the equilibrium NaN3 + H2O ↔ NaOH+ HN3. The pKa values of trifuoroacetic acid (0.23), formic acid (3.77) , hydrozoic acid (4.72) and acetic

116 - - acid (4.76) indicate that the basicity of N3 is comparable to that of CH3COO and

- - much stronger than that of CF3COO (TFA ). The NaOH formed in the degradation

+ 117 medium can form (NaOH)xNa clusters.

The PLA degradation products observed in Figure 5.2 are composed of

+ (PLnNaOH)(NaOH)xNa clusters, as mentioned above (see also Scheme 5.1, top). Eight different PLnNaOH distributions are discerned, depending on their (NaOH)x content, as listed in Table 5.1.

Table 5.1. PLA degradation products detected after 7 days of incubation.

- + + HO O Na . (NaOH)xNa O n - 1 O

Series x a n b m/z c A 0 1 – 3 135, 207, 279 B 1 2 – 6 247, 319, 391, 463, 535 C 2 3 – 7 359, 431, 503, 575, 647 D 3 5 – 9 543, 615, 687, 759, 831 E 4 6 – 11 655, 727, 799, 871, 943, 1015 F 5 7 – 13 767, 839, 911, 983, 1055, 1127, 1199 G 6 9 – 14 951, 1023, 1095, 1167, 1239, 1311 H 7 10 – 16 1063, 1135, 1207, 1279, 1351, 1423, 1495

a Each NaOH unit provides 39.9925 Da. b Each polylactide repeat unit provides 72.0211 Da. c The combined end groups are NaOH. The mass of Na+ is 22.9898 Da. O

- + + HO O Na . (NaOH)xNa O n - 1 O

polylactide + NaOH Na+ - cationized (PLnNaOH) (NaOH)x cluster

PLA

O O- Na+

HO n - 1 O x O- Na+ Na end group O O HO end group

lactide unit lactide + NaOH (72 Da) repeat unit (112 Da)

Scheme 5.1. Plausible structures of the products formed by enzymatic degradation of PLA.

The NaOH units could form a network of hydrogen bonds and O-Na+ salt bridges with the carbonyl groups in PLnNaOH, as rationalized in Figure 5.3. Alternatively, NaOH could react with the polylactide carbonyl group to form adducts with a tetrahedral C atom substituted by OH and O-Na+ groups, as shown in Scheme 5.1, bottom. In this latter case, the degradation products may be viewed as copolymers with lactide (72 Da) and lactide +

NaOH (112 Da) repeat units, with distributions A - H differing in the number of 112-Da repeat unit (0 – 7, respectively). Under either scenario, the results indicate that degradation of the polylatide took place, via enzymatic hydrolysis of its ester groups, which created small linear oligomers with 1-16 lactide repeat units (Table 5.1). Larger oligomers are probably also formed during the degradation, but they are not water soluble and hence not detected in this experiment. Further, prior to 7 days of biodegradation, no detectable signal was observed.

Na+

H O O H +Na O

H Na+

O O O O

O O

. Figure 5.3. Network of hydrogen bonds and salt bridges (ion pairs) in PLnNaOH + (NaOH)xNa clusters.

A similar mass spectrum as in Figure 5.2 was obtained when analyzing the water medium of PLA after 10 days of biodegradation using proteinase K (see Figure 5.4). A longer degradation time increases the relative intensity of the peaks in the lower mass range; Moreover, after ten days, only 5 distributions were clearly discerned, viz. A to E.

Within these five series, new peaks appeared corresponding to longer polylactide oligomers. For example, series B showed clusters with 3 -12 repeating units, extending

. + . + from PL3NaOH (NaOH)Na (m/z 319) to PL12NaOH (NaOH)Na (m/z 967). In contrast, distribution B streched from the 2-mer to the 6-mer after only 7 days of incubation. Condensation/hydrolysis equilibria are probably responsible for the observation of longer PLnNaOH chains and an overall narrower molecular weight distribution of degradation products after 10 days of incubation.

C 575.2

B 503.1

391.1 B C 463.2

647.2 D C B D 759.3 431.1 535.1 D 687.2 615.2 C A B E 207.1 A 711.2 D 603.3 719.3 B

319.1 B E E 799.3 831.3

279.1 B B D 751.2 E 871.3 A C 823.3 A E B 137.0 B 783.3 895.3

C C 943.4 D C 967.3 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 m/z

Figure 5.4. ESI-Q/ToF mass spectrum of the water-soluble degradation products of PLA after10 days of biodegradation with Proteinase K

It is possible, however, that the presence of peaks corresponding to longer polylatide oligomers after 7 days is obscured by the higher ionization efficiency/solubility of the smaller oligomers. Nevertheless, the results of this study clearly indicate that high molecular weight PLA is enzymatically degraded to small oligomers, down to the monomer. No hazardous byproducts that could compromise biomedical PLA applications were detected.

MS2 experiments were performed on several degradation products. Figure 5.5 shows a representative spectrum, obtained by CAD of m/z 575 in distribution C (cf.

. + Figure 5.2 and Table 5.1). This ion has the composition PL6NaOH (NaOH)2Na (Table

5.1) and, as mentioned above, can be viewed as a copolymer of 3 lactide, -

CH(CH3)COO- (72 Da), and 3 lactide + NaOH, -CH(CH3)C(OH)(ONa)O- (112 Da), repeat units with random sequence.

Δ 391.1

Δ * 319.1 207.1 . + PL6NaOH (NaOH)2Na 575.4

Δ 463.2 * Ω * Δ Ω 135.0 279.1 * 412.2 503.2 -72 Da 269.1 341.1 247.1 351.1

140 170 200 230 260 290 320 350 380 410 440 470 500 530 560 m/z

. + Figure 5.5. ESI-CAD mass spectrum of the PL6NaOH (NaOH)2Na cluster ion, acquired with the Waters Q/ToF tandem mass spectrometer. The precursor ion was subjected to CAD with argon at a collision energy of 55 eV.

Most CAD fragments observed in Figure 5.5 can be explained by lactide ester bond cleavages proceeding from the lactide + NaOH unit (i.e. ester bond cleavages promoted by NaOH), cf. Scheme 5.2. This dissociation pathway can account for the fragments labeled by * and Δ in the CAD spectrum of Figure 5.5 (cf. Table 5.2). Within

. + + + the NaPL6 (NaOH)2Na precusor ion, H /Na exchange is possible between OH and

ONa groups; the fragments marked by Ω are attributed to such exchange , as they appear

22 Da higher that fragments marked by Δ (replacement of H by Na increases the mass by

22 Da). The fragment m/z 412.2 is only 21 Da higher in mass than the most abundant

fragment at m/z 391.1. Hence, m/z 412.2 is attributed to OH → ONa conversion plus loss of a hydrogen radical; the reason for this deviation is not understood. Finally, the heaviest fragment, m/z 503.2, arises by loss of 72 Da, viz. loss of the elements of the lactide repeating unit (C3H4O2). The mechanism depicted in Scheme 5.3 provides a reasonable pathway, analogous to that established for the elimination of the N-terminal residue from ionized peptides.118

Na+ OH O

HO CH C O C ONa - + CH3 O Na

Na+ O O H HO CH C O C Na+ ONa O CH3

Na+ Na+ O O- Na+ HO CH C + HO C ONa O CH3 Scheme 5.2. Fragmentation of polylactide ester groups to which NaOH has been added (nucleophilic addition/fragmentatio). This dissociation yields two truncated polylactide chains with HO- and –Na end groups.

O OH + O Na O O- Na+ H O 2 O 2 O- Na+ O O

loss of O=CHCH3 + CO (72 Da)

OH Na+

O O- Na+

HO 2 O 2 O- Na+ O O m/z 503.2

. Scheme 5.3. Elimination of C3H4O2 (72 Da) from the hydroxyl chain end of PL6NaOH + (NaOH)2Na (m/z 575) to yield the m/z-503 fragment.

2 . + Table 5.2. Fragments in the MS (CAD) spectrum of PL6NaOH (NaOH)2Na (m/z 575) formed according to the mechanism in Scheme 5.2. All have NaOH end groups (see Schemes 5.1 and 5.2) and all contain a Na+ charge.

O O- Na+

O O n OH x

n a x b m/z 1 0 135.0 (*) 2 0 207.1 (*) 2 1 247.1 (Δ) 3 0 279.1 (*) 3 1 319.1 (Δ) 4 0 351.1 (*) 4 1 391.1 (Δ) 5 1 463.2 (Δ)

a Number of lactide repeat units (72 Da). b Number of lactide + NaOH repeat units (112 Da). Fragments with 0 and 1 such units have been labeled by * and Δ, respectively, in Figure 5.5.

5.4. Conclusion

PLA films were degraded by proteinase K in water, leading to rapid formation of water soluble degradation products. The enzymatic degradation products of PLA were elucidated by ESI MS and CAD studies. Comparison with the control sample, which produced no detectable PLA oligomers after incubation, confirmed that degradation of

PLA is accelerated significantly by the presence of proteinase K. It is believed that biodegradation of PLA occurred by random ester bond scissions along the polymer

backbone. The biodegradation products were truncated linear oligomers. All have the

. + composition PLnNaOH (NaOH)xNa where n = number of lactide repeating units in the trancated polymer and x = number of NaOH moieties attached to the oligomer. When the incubation time was raised from 7 to 10 days similar biodegradation products were detected, but with lower molecular weights. After a 7-day incubation, eight distributions differing in (NaOH)x content were observed; whereas after 10 days of incubation only five such distributions were discerned. In MS2 studies, most fragments arise by ester bond cleavages analogous to those taking place upon biodegradation. Some OH/ONa exchange was also observed during MS2 fragmentation. Mass spectrometry studies confirmed that the degradation process using proteinase K occurs fast and that the degradation products are shorter chains, down to the monomer. No alternated structures that might interfere with biological processes were detected.

CHAPTER VI

PROBING PROTEIN ADSORPTION ON ELECTROSPUN THERMOPLASTIC ELASTOMERIC SURFACES VIA MASS SPECTROMETRY

6.1. Background

When cells interact with the environment, cell adhesion to a foreign surface marks the first important step to begin the cycle of migration, proliferation, differentiation and apoptosis.119 This cell life cycle is crucial to the success of in vitro tissue growth or tissue engineering, where a scaffold made of a biocompatible material is employed to provide a substrate to initiate cell growth, and serve as the structural support for sufficient cell accumulation and the eventual tissue formation. Cell adhesion to foreign surfaces is understood to be regulated by the adhesion receptor proteins, known as integrins, in the cell’s membrane.120,121 These integrins would specifically bind to adhesion proteins, like fibronectin, vitronectin and fibrinogen.121 The protein adsorption capability of a surface is the key factor to healthy cell growth and tissue colonization,122-125 and needs to be adequately evaluated to assess the surface’s overall performance as a tissue scaffold.

To build successful scaffolds for tissue replacement, the materials of choice should be at least biocompatible to avoid rejection by the human body, possess the right balance of porosity and surface areas for the transfer of nutrients and growth factors to

initiate protein adsorption and cell growth, and finally have the necessary structural properties to withstand the rigors of stresses imposed by the body after implantation.

Focusing on soft tissue replacement, the structural requirement on the scaffolds is less onerous since soft tissues, like skin and cartilage, typically only have an ultimate tension strength of about 3 MPa and an elongation at break of 18 %.126 Motivated by the successful FDA approval of linear poly(styrene-b-isobutylene-b-styrene) triblock polymer127 as a drug-eluting coating on TAXUS Express2TM coronary stents in 2004,128 dendritic poly(isobutylene-b-styrene) (D_IBS)129 block copolymers were developed that have similar chemical structure as the linear triblock polymer and hence bear great potential for tissue engineering applications with their excellent biocompatibility130 and superior profile of physical and mechanical performance.131-133 Synthesized by living carbocationic polymerization, the D_IBS class of polymers have a dendritic polyisobutylene (PIB) core with end-blocks of polystyrene (PS),132,134 PS derivatives135 or copolymers of PIB and PS.136 The excellent biocompatibility of these PIB-based block copolymers is believed to stem from the phase segregation of PIB to form a thin (~10 nm) layer at the surface.134 From a series of comparative studies and depending on the PS content, D_IBS polymers can have an ultimate tensile strength as high as 8.7 MPa, and elongation at break up to more than 1800 % that far exceed the needs of most soft tissues.131,132

To seek increased surface areas and porosity, synthetic micro- and nano-sized polymer fibers produced by electrospinning have been increasingly employed as scaffolds,137,138 which showed improved cell incorporation for tissue engineering.139,140

By the morphological similarity to natural tissues such as collagen fibrils that are the major extra-cellular matrix components, the three-dimensional structure of electrospun fiber mats helped to promote better cellular response and biocompatibility.141,142 Also, it has been demonstrated that the morphological features of electrospun fiber mats can yield super-hydrophobic surfaces.143-145 By definition, a super-hydrophobic surface can bead up water droplets with a static contact angle of more than 150.144 Several experimental techniques had been designed to convert synthetic polymers into superhydrophobic surfaces, including choosing a suitable solvent and controlling processing temperature to create a textured surface,146 electrospinning,17-18,147 and the template-based extrusion method.148

In recent years, matrix-assisted laser desorption/ionization time-of-flight

(MALDI-ToF) mass spectrometry (MS) has become a popular technique for the analysis and quantification of biomolecules, like oligosaccharides or proteins.149 The inherent advantages of MALDI-ToF MS as a quantitative analysis technique are essentially its high sensitivity, femtomole amounts of material required, rapid sample preparation and analysis over a wide mass range.150 It has been found that to accomplish reproducibility, the addition of a properly selected internal standard to each sample, and a well controlled constant laser power and voltage should be used. The introduction of an internal standard is able to diminish the sample-to-sample variability.151

The main aim of this study was to evaluate the performance of protein adsorption on D_IBS compression molded films and electrospun fiber mats over a range of pHs using MALDI-TOF MS. For comparison, PS molded film and electrospun fiber mats

were also considered to assess the influence of polymer chemistry and surface morphology on the resulting protein adsorption. In this work, three model proteins

(insulin, ubiquitin and lysozyme) were selected for in vitro adsorption experiments.

6.2. Sample preparation

Two polymers, viz. D_IBS (05DNX120) and PS, were used to prepare

compression molded sheets and electrospun fiber mats. D_IBS was synthesized using

132 living carbocationic polymerization and had a molecular weight (Mn) of 220, 300

g/mol, a polydispersity index of 1.87 and a PS content of 29.4 wt%. From size

exclusion chromatography analysis, Mn of the PS endblock was determined to be

around 73,600 g/mol. PS (D4030) with Mn of 111,300 g/mol and a polydispersity

index of 2.33 was used as received from Americas Styrenics LLC. The polymers were

separately compression molded into 100-m thick films (see Section 6.4), and

dissolved in a mixture of tetrahydrofuran (THF) and toluene (95:5 w/w) at a polymer

content of 10 wt% for electrospinning (see Section 6.3). THF (Sigma-Aldrich) was

distilled from calcium hydride before use, and toluene (99.8%, Sigma-Aldrich) was

used as received. Insulin (from bovine pancreas; MW = 5800 g/mol; isoelectric pH, pI

= 5.3), ubiquitin (from bovine erythrocytes; MW = 8560 g/mol; pI = 5.2), lysozyme

(from chicken egg white; MW = 14700 g/mol; pI = 11.0), cytochrome C (from equine

heart; MW = 12384 g/mol) and sinapinic acid (3,5 dimethoxy-4-hydroxycinnamic acid;

MW = 225 g/mol) were purchased from Sigma-Aldrich. All proteins and the matrix

were used as received without further purification. Ammonium acetate (97.7%, Sigma-

Aldrich), Trizma base (99.9%, Sigma-Aldrich), trifluoroacetic acid (99.9%, Aldrich)

and urea (MP Biomedicals, LLC) were used as received. The acetonitrile used (Fisher) was of optima grade. Distilled water was used for buffer solution preparation.

6.3. Electrospinning

A positive potential of 20 kV was applied from a power source between the polymer solution in a glass pipette and a grounded metal collector covered with aluminum foil. The spinning distance between the tip of the glass pipette and the metal collector was maintained at 20 cm.

6.4. Compression Molding

Compression molding was used to prepare flat samples of D_IBS and PS for protein adsorption studies and water contact angle measurements. A molding plate of

100-m thick was filled with polymer and covered with Kapton® (polyimide) sheets as substrate and another flat stainless steel plate on both sides for compression molding.

Table 6.1 lists the conditions of the three-step compression molding process. Liquid nitrogen was applied at the end of the cooling step to detach the molded materials from the Kapton® substrate. The final molded sheet had a dimension of 2.5 cm by 2.5 cm by

100 m.

Table 6.1. Conditions for compression molding.

Step Temperature (°C) Force (N) Time (min) Heating 170 ~ 0 8 Molding 170 50000 2 Cooling 23 50000 12 94

6.5 Protein Adsorption

Compression molded sheets and electrospun fiber mats of D_IBS and PS were

studied for their protein adsorption capability at different pH levels. The protein

mixture system (PMS) examined was composed of three different proteins, i.e. insulin,

ubiquitin, and lysozyme. A stock solution of the internal standard, cytochrome C, was

prepared in acetonitrile-water (50:50 v/v) that contained 0.1% trifluoroacetic acid

(TFA). Protein adsorption experiments were conducted in 20 mL vials by adding the

polymer sample (1 × 1 cm2) to 1 mL of PMS solution containing 10 pmol/µl of each

protein in 0.01 M ammonium acetate (pH = 4.4, 5.4 and 6.9) or 0.05 M Tris (pH = 8.7)

buffers. The vials were shaken for 24 h in an incubator shaker at 37 °C and 100 rpm.

Then, the polymer samples were transferred to new vials and gently washed three times

with 1 mL of the same buffer used for the incubation. The washes were intended to

remove those proteins not attached or just adhering to the polymer surface. The

proteins bounded to the polymer surfaces were removed by the addition of 200 µL of

urea (6 M) to the vials containing the samples and a further incubation for 10 minutes

in an oven at 55 °C, which was followed by a desalting process. Millipore Zip Tips C

18 were used for desalting purpose. After the procedure was completed, about 5 – 10

µL of adsorbed protein solution (APS) was obtained. 2 µL of APS were taken and

mixed with 2 µL of IS (1 pmol/µL) to yield a final IS concentration of 0.05 pmol/µL.

This APS-IS solution was then mixed in the ratio 1:1 with a 20-mg/ml solution of

sinapinic acid matrix in 75% acetonitrile that contained 1% TFA. Finally, 1 µL of the

APS-IS-matrix solution was used to coat the target for MALDI-ToF MS analysis. The

singly protonated protein-ion signals were then integrated and considered for

adsorption quantification. The procedure adsorption/removal mentioned above was

carried out twice for each polymer sample and each APS was subjected to four

MALDI-ToF MS analysis.

6.6 Calibration Curves

For a quantitative analysis of protein adsorption, calibration curves were first

constructed using various concentrations of PMS in acetonitrile-water (50:50 v/v) that

contained 1% TFA. The concentration range of the standard PMS solutions was 0.004 –

8 pmol/µL for each protein (insulin, ubiquitin and lysozyme). A 5-µL aliquot was taken

from each PMS solution and mixed with 5 µL of 1-pmol/µL IS (cytochrome C) to

produce a new set of PMS-IS solutions with a fixed concentration of IS (0.5 pmol/µl)

and 0.002 – 4 pmol/µL for each protein. 1 µL of the PMS-IS solutions was then added

to 1 µL of the 20-mg/mL of solution of sinapinic acid matrix in acetonitrile-water

(75:25 v/v)/0.1% TFA to prepare the final set of PMS-IS-matrix solutions. For

MALDI-ToF MS analysis, the stainless steel sample plate was coated with 1 µL from

each of the final PMS-IS-matrix solutions. To build the calibration curves, 10 replicate

mass spectra were acquired at each concentration point. For quantification, only the

singly protonated protein-ion signals were taken in consideration. The intensity of the

IS peak was maintained between 21000 to 25000 a.u.

6.7 Matrix-Assisted Laser Desorption/Ionization – Time of Flight Mass Spectrometry

(MALDI-ToF MS )

MALDI-ToF mass spectra were acquired on a Bruker Ultraflex-III ToF/ToF mass

spectrometer (Bruker Daltonics, Inc., Billerica, MA) equipped with a neodymium-

doped yttrium aluminum garnet (Nd:YAG) laser (355 nm). All spectra were measured

in positive linear mode. 1 µL of individually prepared solutions of APS for

quantification (see Section 6.5 – Protein Adsorption) or PMS of known protein

concentrations (see Section 6.6 – Calibration Curves) were deposited on microtiter

plate wells (MTP 384-well ground steel plate). After the evaporation of the solvent,

the plate was inserted into the MALDI-ToF ion source. The attenuation of the

Nd:YAG laser was adjusted to maximize the sensitivity.

6.8 Scanning Electron Microscopy (SEM) Imaging

SEM imaging was conducted with a field emission scanning microscope (JEOL

JSM-7401F) at an accelerating voltage of 10 kV. The aluminum foil with the collected fibers was attached onto a SEM stub, which was kept in a vacuum oven at room temperature overnight to remove solvent. To improve electron conductivity, SEM stubs with fibers were coated before imaging, with silver in an argon atmosphere for 1 min using an EMITech K575X Peltier Cooled sputter-coater.

6.9 Water Contact Angle (WCA) Measurement

A contact angle gonoimeter, Model 500 of the F1 series from Ramé-Hart

Instrument Co., was employed for the WCA measurement using de-ionized water. A water droplet of about 4 – 6 L was dispensed from a micro-syringe onto the sample surface. The profile of a water droplet on the sample was captured with a high resolution camera as a bitmap image that was subsequently post-processed using imaging software, ImageJ, to determine the contact angle. At least 3 contact angle measurements were made for each specimen.

6.10 Results and Discussion The following sections illustrate the calibration results and explain the adsoption results in terms of surface hydrophobicity, morphology and behavior at different pH levels.

6.10.1 Calibration Results

Figure 6.1 shows two MALDI-ToF mass spectra used for the construction of

calibration curves, obtained from PMS samples containing insulin (I), ubiquitin (U) and

lysozyme (L) at 1 and 4 pmol/L, and the IS – Cytochrome C (CC, 0.5 pmol/L).

Specific peaks corresponding to the ionized proteins and their dimers were indicated in

the spectra. From the figure, the height and area of various protein peaks can be seen to

increase at higher concentration relative to those of IS whose concentration was fixed.

For a consistent quantification, the area of the singly protonated protein peaks

(Areaprotein) was normalized by that of the IS (Areacc) over the considered range of

concentrations (0.002 – 4 pmol/µL) to build the calibration curves presented in Figure

6.2. In Figure 6.2, each point represents the average peak area ratio (Areaprotein/Areacc)

of 10 averaged mass spectra for a specific concentration. A strong linear fit can be

observed for the normalized signal areas versus concentrations of each individual

protein, with correlation coefficients of > 97.8 %. With these calibration curves, the

protein concentrations of the APS solutions from the protein adsorption study (see

Section 6.5) were determined.

L+ U+ CC+ 14214 8466 12264 I+ L2+ U2+ 5635 7052 I2+ 2778 4187 1 pmol/μl

CC2+ 2I+ 6076 11366 4 pmol/μl 4000 6000 8000 10000 12000 14000 m/z Figure 6.1. Mass spectra of PMS-IS mixtures containing 1 pmol/µL (top) or 4 pmol/µL (bottom) PMS, and 0.5 pmol/µL Cytochrome C (IS). The spectra were produced by signal averaging 1000 laser pulses.

2.4 Insulin 2.0

1.6

1.2

0.8

0.4 y = 0.4524x 0.0 R² = 0.9784 3.5

3.0 Ubiquitin

CC 2.5

/Area 2.0

1.5 protein 1.0 Area 0.5 y = 0.6813x 0.0 R² = 0.9834

8.0 7.0 6.0 Lysozyme 5.0 4.0 3.0 2.0 1.0 y = 1.497x R² = 0.9930 0.0 0 1 2 3 4 5 Sample Concentration (pmol/μl)

Figure 6.2. Normalized peak area of different proteins versus protein concentrations. Each point represents the average of 10 mass spectra, with the error bars representing one standard deviation of the data.

100

6.10.2. Surface Hydrophobicity and Morphology

This study varied the morphological features (molding vs. electrospinning) and chemical makeup (D_IBS vs. PS) on a surface to influence its water repellency and ultimately studied how all these factors impacted protein adsorption performance.

Figure 6.3 (a) and (b) show the profile of water droplets on the surface of molded

D_IBS and PS samples, respectively. Based on the WCA data, both molded surfaces show comparable degree of hydrophobicity at 90.2  (± 2.3 ) for D_IBS and 89.8  (±

1.5 ) for PS. Earlier work by Puskas and Kwon134 established that there is a thin layer

(~ 10 nm) of PIB on the surface of D_IBS-type block copolymers. Considering that a

WCA of 96.7  (± 2.3 ) was reported by Orlowski for a spin-coated PIB (Mn = 135,000 g/mol) surface on a silicon wafer,152 the present results indicate that the segregation of the rubbery PIB phase to the molded flat surface of D_IBS did not change much its wettability as compared to a PS bulk surface.

101

(a) WCA = 90.2  ± 2.3  (b) WCA = 89.8  ± 1.5 

Figure 6.3 Water droplet profiles on (a) D_IBS molded sheet, (b) PS molded sheet, (c) D_IBS electrospun fiber mat, and (d) PS electrospun fiber mat.

However with electrospinning, the hydrophobicity of D_IBS and PS fiber mats was significantly enhanced to 121.9  ± 6.2  and 143.0  ± 3.3 , respectively [see Figure

102

6.3 (c) and (d)]. This can be attributed to the pronounced changes in the surface morphology in the D_IBS and PS fiber mats, as shown in Figure 6.4 (a) and (b). With the electrospinning setup used in this study, the D_IBS and PS fibers collected were randomly oriented over the surface. Owing to the difference in the molecular weight of

D_IBS and PS, the D_IBS fibers were considerably thicker (1614 ± 648 nm) than those of PS (225 ± 89 nm). Furthermore, a significant amount of large beads can be found on the PS fibers with an average size of 3003 nm (± 1739 nm). It is evident from Figure 6.4, at higher magnification, that the surface of PS fibers and beads contains nano-sized pores ubiquitously, as compared to the generally smooth surface of D_IBS fibers. This combination of large micro-beads on thin sub-micron fibers with nano-pores on the surface helped to create the appropriate hierarchical levels of surface roughness that yielded a higher WCA for the PS mat than for the D_IBS mat which has smoother and thicker fibers. The effect of surface properties (hydrophobicity and chemistry) on the protein adsorption capability of various D_IBS and PS surfaces is examined in the next sub-section.

103

(a)

(b)

Figure 6.4. SEM micrographs of (a) D_IBS and (b) PS electrospun fiber mats at different magnifications.

6.10.3. Protein Adsorption Behavior of D_IBS and PS Surfaces at Different pH Levels

Figure 6.5 shows a set of MALDI-ToF mass spectra of the insulin, ubiquitin, and lysozyme mixtures adsorbed on a D_IBS fiber mat at the pH of 4.4, 5.3, 6.9, and 8.7.

Peaks assigned to the charged molecular ions (M + H+ and M + 2H+) and the protonated insulin dimer (2M + H+) are marked. From theses spectra and using the calibration curves discussed in Section 6.10.1, the amount of proteins adsorbed on each surface was calculated from the area ratio (Areaprotein/Areacc) of the respective protein peaks; only singly protonated protein ion signals were considered for the quanitifcation (as also done for the calibration curves). Table 6.2 and 6.3 summarize the calculated

104

amount of proteins adsorbed on the molded and electrospun samples of D_IBS and PS,

respectively, at different pH levels.

I+ 5637

CC+ L+ 2+ CC + 14209 U 2I+ 12263 6082 8467 11369 pH 4.4 pH 4.4

pH 5.3 pH 5.3

pH 6.9 pH 6.9

pH 8.7 pH 8.7 6000 7000 8000 9000 10000 11000 12000 13000 m/z 14050 14150 14250 14350m/z

Figure 6.5. Positive mode MALDI-ToF mass spectra of insulin, ubiquitin and lysozyme adsorbed on the D_IBS fiber mat at various pH levels. Incubation with the PMS for 24 hr at pH 4.4, 5.3, 6.9 and 8.7. Because of the small intensity of the lysozyme peak, its m/z region (14000-14350) is shown in expanded view on the right.

105

Table 6.2. Summary of protein adsorption results for the molded and electrospun D_IBS samples.

Adsorption per unit area (pmol/cm2) Insulin Lysozyme Material pH Ubiquitin (pI = 5.3) (pI = 5.2) (pI = 11.0) 4.4 145.12 ± 2.03 10.20 ± 0.52 0.14 ± 0.02

D_IBS molded sheet 5.3 152.74 ± 5.47 13.06 ± 0.83 0.15 ± 0.00 6.9 101.68 ± 1.25 8.37 ± 0.83 0.68 ± 0.03

8.7 48.52 ± 1.09 4.99 ± 0.62 0.94 ± 0.09

4.4 352.45 ± 2.35 39.26 ± 2.80 1.07 ± 0.09 390.92 ± 5.3 40.80 ± 1.45 1.47 ± 0.189 D_IBS fiber mat 36.11 6.9 222.04 ± 3.28 13.21± 0.42 1.67 ± 0.38

8.7 78.25 ± 2.19 6.09 ± 0.73 3.37 ± 0.24

Table 6.3. Summary of protein adsorption results for the molded and electrospun PS samples.

Adsorption per unit area (pmol/cm2) Insulin Lysozyme Material pH Ubiquitin (pI = 5.3) (pI = 5.2) (pI = 11.0) 4.4 117.97 ± 1.11 9.67 ± 0.74 -

PS molded sheet 6.9 88.17 ± 1.58 6.97 ± 0.74 0.52 ± 0.06

8.7 42.97 ± 1.74 3.60 ± 0.21 0.85 ± 0.07

4.4 258.04 ± 8.2 19.64 ± 0.64 0.29 ± 0.03

PS fiber mat 6.9 200.68 ± 1.26 13.09 ± 1.17 1.36 ± 0.23

8.7 73.00 ± 3.79 4.50 ± 0.42 2.75 ± 0.33

106

It is generally accepted that protein adsorption on nonpolar polymer surfaces in aqueous environment is predominantly initiated by hydrophobic interactions153-155 and can be modulated by the presence of immobilized alkyl (e.g. butyl) sites at the surface.151,152 The favorable segregation of the PIB rubbery phase of D_IBS type block copolymers to the surface would provide abundant butyl binding sites for better protein adsorption performance. Indeed, the data in Tables 6.2 and 6.3 confirm that the flat molded surface of D_IBS adsorbed more of the proteins than a PS surface of the same area, although both surfaces showed a similar degree of hydrophobicity as discussed in

Section 6.10.2. For the more hydrophobic electrospun fiber mats, it is not unexpected to see a significant improvement (22 – 750 %) in the protein adsorption performance of both materials from Table 6.2 and 6.3. Beside the higher hydrophobicity other factors, like the higher surface area, micro-porosity and three-dimensional structure of the fiber mats may play important roles in promoting better protein adsorption onto the mat surface and possibly into the fibrous network. It is surprising, however, that the PS fiber mat repelled water better due to its hierarchical surface roughness, but adsorbed much lesser amounts of proteins than the D_IBS counterpart across all pH levels. This can be linked to the thin layer of PIB on the surface of the D_IBS fibers, which accentuates the influence of surface chemistry to modulate protein adsorption.

107

(a) Insulin (pI = 5.3) (b) Lysozyme (pI = 11.0)

600 D_IBS molded sheet ) 2 500 D_IBS fiber mat

400

300

200

Proteinadsorption(pmol/cm 100

0 4.4 5.3 6.9 8.7 pH 4.0 D_IBS molded sheet ) 2 3.5 D_IBS fiber mat

3.0

2.5

2.0

1.5

1.0

Proteinadsorption(pmol/cm 0.5

0.0 4.4 5.3 6.9 8.7 pH Figure 6.6. Adsorption of (a) insulin and (b) lysozyme on D_IBS molded sheet and fiber mat.

Focusing on the D_IBS polymer, the adsorption extents of insulin and lysozyme on the molded sheets and fiber mats (Table 6.2) were separately plotted in Figure 6.6.

The graphs clearly show that protein adsorption on both D_IBS surfaces peaked at pH =

5.3 for insulin, which corresponds to its isoelectric point (pI), while the binding affinity of lysozyme to D_IBS surfaces continued to increase as the pH level became higher.

108

These distinctive protein adsorption trends can be explained by the change in the net accumulated charge in a protein with respect to its pI. When a protein is in a pH environment that is close to its pI, the net charge of the protein is near zero.156 This would minimize the interaction between different protein molecules and maximize the exposure of the hydrophobic regions of the protein so that they can actively participate in hydrophobic interactions with a surface. Hence, the maximum adsorption of a protein on a hydrophobic surface is observed when the environment has the same pH level as the pI of the protein. Conversely, when the environmental pH is lower or higher than the pI of a protein, the protein accumulates a net charge, positive when pH < pI, and negative when pH > pI.153 In these cases with a net charge, the protein molecule can fold non- cooperatively, and/or interact with others to shield their hydrophobic sites thereby obstructing them from developing hydrophobic interactions with a surface.

Consequently, a lower protein adsorption on a surface occurs at pH levels below and above the pI of the protein. This justifies why the adsorption of lysozyme on D_IBS surfaces continued to rise from pH = 4.4 to pH = 8.7, which lies closer to its pI (11.0).

This reasoning of protein-surface interactions is not limited to the results shown in

Figure 6.6; it can also be used to explain the adsorption results of ubiquitin on D_IBS surfaces as well as the adsorption behavior of the PS surfaces (Tables 6.2 and 6.3).

Lastly, the amount of protein adsorbed on these surfaces can be also associated with the degree of hydrophobicity of the three proteins studied. Wang et al.157 used four different scales to calculate the hydrophobicity of proteins, including bovine insulin and ubiquitin. Based on these four scales, insulin was determined to be more hydrophobic than ubiquitin.154 Clarke and Chen158 reported that if the structures of ubiquitin and

109

lysozyme were to be classified by the ratio of their hydrophobic/polar amino acids, lysozyme would be more polar. Therefore, it was expected and so proven from the results presented in Tables 6.2 and 6.3 that the protein quantity adsorbed onto the studied polymer surfaces was highest for insulin, followed by ubiquitin, and lowest for the hydrophilic lysozyme. Moreover, protein adsorption to a surface can be affected by other factors including the size of the protein. Insulin is a wedge-shaped molecule with a size of about 2 × 2.5 × 2 nm3 and comprised of 51 amino acids (mostly hydrophobic). On the other hand, the lysozyme molecule has an approximately ellipsoid shape with larger dimensions, viz. about 4.5 × 3.0 × 3.0 nm3, and 128 amino acids. Smaller proteins, like insulin, can maneuver and orientate easily to interact with a surface for adsorption with better tenacity and in higher amounts. Conversely, the bigger lysozyme molecule would have more difficulty to seek free surface for interaction and would be more likely to interfere with nearby adsorbed molecules to maintain adherence to the surface.

6.11 Conclusion

In this study, a potentially biocompatible polymer, D_IBS, was investigated for its protein adsorption performance. Two polymer processing techniques, electrospinning and compression molding, were employed to create a fiber mat surface and a flat surface, respectively, and PS was included as a reference material to investigate the effect of surface morphology and polymer chemistry on protein adsorption. Three model proteins, insulin, ubiquitin and lysozyme, were incubated with the surfaces at four pH levels, viz.

4.4, 5.3, 6.9 and 8.7, based on an established protocol, and the amount of protein

110

adsorbed to the surfaces was then measured by MALDI-ToF MS using calibration curves and cytochrome C as an internal standard. Based on SEM and WCA characterization, electrospun fiber mats of D_IBS and PS were more hydrophobic and hence showed a significantly higher affinity to all three proteins across all levels of pH than their flat surfaces prepared by compression molding. It should be highlighted that the higher surface area-to-volume ratio and the three-dimensionality of fiber surfaces could also play significant roles in the enhanced protein adsorption. While PS electrospun fiber surfaces had a more complex hierarchical level of surface features that yielded greater water repellency, their protein adsorption performance was significantly lower than that of the D_IBS fiber mat surfaces. This can be explained by the possible segregation of a thin layer (~ 10 nm) of PIB to the surface of D_IBS fibers, which accentuates the importance of surface chemistry in modulating protein adsorption beside surface morphology. The results of this study also confirmed that the pH of the environment can affect the hydrophobic interaction between proteins and a surface. When the pH of the surrounding environment was close to the pI of the proteins, this hydrophobic interaction

(and hence protein adsorption) approached a maximum, since the proteins carried zero charge that reduced interactions between one another and/or had their hydrophobic sites more exposed for surface adsorption. When the pH of the environment became higher or lower than the pI, the proteins accumulated charges and started to fold uncooperatively.

These phenomena tend to induce interactions between proteins and shield off the active hydrophobic sites of the proteins so that they cannot easily participate in the adsorption process. Moreover, the protein size and hydrophobicity also played an important role in the protein adsorption performance of a surface. This study indicated that insulin, being

111

smaller and more hydrophobic than lyzozyme, can maneuver and orientate easily to develop more extensive hydrophobic interactions with a surface for better adherence. In all, this study demonstrated the efficacy of electrospinning in changing surface morphology, the usefulness of mass spectrometry as an evaluation technique for protein adsorption, and the potential of D_IBS with its promising protein adsorption capability for future applications in tissue engineering.

112

CHAPTER VII

TOP-DOWN MULTI-DIMENSIONAL MASS SPECTROMETRY METHODS FOR

SYNTHETIC POLYMER ANALYSIS

7.1. Background

The introduction of matrix-assisted laser desorption ionization (MALDI)159,160 and electrospray ionization (ESI)161 has enabled the formation of gas-phase ions from a wide variety of biomaterials and synthetic polymers, opening a new era for their mass spectrometry (MS) analysis. MS experiments provide the mass-to-charge ratios (m/z) of the constituent n-mers of a polymeric material, from which compositional heterogeneity, molecular weight, and functionality distributions can be deduced. Such information has been essential in the discovery of new polymerization techniques, the elucidation of polymerization mechanisms, and the advancement and commercialization of new products.162-172 Nevertheless, significant challenges still exist. Both MALDI and ESI require the use of solvents and/or matrices that may alter the identity of reactive or labile analytes. Furthermore, polymerizations, in particular newly developed methods, may create complex mixtures that are difficult or impossible to characterize by single-stage

MS because of discrimination effects in the ionization and detection steps and/or because the product contains isobaric components or a mixture of isomeric architectures that

113

cannot be identified by m/z measurement alone, even at high mass resolution. This chapter will demonstrate how such problems can be bypassed by combining MALDI and

ESI with two-dimensional (2-D) top-down methods involving tandem mass spectrometry

(MS2), i.e. two stages of mass analysis with intermediate fragmentation,21, 22, 173, 174 ion mobility mass spectrometry (IMMS), i.e. post-ionization gas-phase separation by mass, charge, and shape coupled with mass analysis,175-187 or a 3-D combination thereof.188,189

The “top-down” characteristic indicates structural identification entirely in the mass spectrometer without prior derivatization, degradation, or chromatographic separation; in contrast, “bottom-up” analysis includes one or more of the latter steps.

The system chosen to illustrate the challenges addressed by the mentioned multidimensional methods is cyclic poly(α-peptoid)s prepared via N-heterocyclic carbene

(NHC) mediated zwitterionic ring-opening polymerization.69,190-192As peptide mimics, poly(α-peptoid)s have shown promise as therapeutic agents or drug delivery carriers because of their enhanced hydrolytic stability compared to conventional polypeptides.

For those examined in the present study, the major polymeric species observed was the

NHC-containing poly(α-peptoid) by ESI, but either the NHC-bound poly(α-peptoid), or the free poly(α-peptoid) without the NHC functionality, was seen by MALDI depending on the conditions used.69 Several questions are raised by this discrepancy. Do the matrix and salt used in MALDI cause the loss of the NHC segment, or is ESI favoring the ionization of NHC-containing species? Did the ring-opening polymerization yield solely

NHC-substituted poly(α-peptoid) or a mixture of this product plus unsubstituted material? Are the NHC-missing poly(α-peptoid) molecules cyclic or ring-opened isomers without end groups? Herein, we will examine this puzzling polymerization 114

process at different levels of top-down mass spectrometry, using MALDI MS, MALDI

MS2, ESI MS, ESI MS2, traveling wave ion mobility (TWIM) MS (a variant of

IMMS),72,183,185,186,193-195 and TWIM MS2.

7.2 Experimental

The following sections describe the running conditions and instrument settings to obtain the data in this chapter.

7.2.1. Materials

The polymer system and other chemicals used for the completion of this Chapter were presented in Chapter 3.1.

7.2.2 MALDI MS

The MALDI MS experiments were carried out on the Bruker Ultraflex III tandem time-of-flight (ToF/ToF) mass spectrometer (Bruker Daltonics, Billerica, MA), equipped with a Nd:YAG laser emitting at a wavelength of 355 nm. All spectra were measured in positive reflector mode. The instrument was calibrated externally with a poly(methyl methacrylate) standard prior to each measurement. Five different matrices were used: dithranol (DIT), 2,5-dihydroxybenzoic acid (DHB), and trans-2- [3-(4-tert-butylphenyl)-

2-methyl-2-propenylidene]malononitrile (DCTB) dissolved in tetrahydrofuran (THF) at

20 mg/mL, 10 mg/mL, and 20 mg/mL, respectively; as well as 3-hydropicolinic acid

(HPA) and α-cyano-4-hydroxycinnamic acid (CHCA) dissolved in acetonitrile

(ACN)/water (v/v, 50/50) at 5 mg/mL. The polymer samples were dissolved in methanol

(MeOH) at 5-10 mg/mL. Sodium trifluoroacetate (NaTFA), sodium iodide (NaI), and 115

potassium trifluoroacetate (KTFA), which served as cationization salts, were dissolved in

THF at 10 mg/mL each. Matrix and cationizing salt solutions were mixed in the ratio

10:1 (v/v). The samples were analyzed with and without the addition of cationization salt. Sample preparation involved depositing 0.5 μL of matrix/salt mixture or just matrix on the wells of a 384-well ground-steel plate, allowing the spots to dry, depositing 0.5 μL of each sample on top of a dry matrix (or matrix/salt) spot, and adding another 0.5 μL of matrix/salt or matrix on top of the dry sample (sandwich method); compared to the commonly used dry droplet method, in which a mixture of matrix and sample is deposited onto the target plate, the sandwich method minimizes the disintegration of labile substances and allows for facile variation of the matrix to maximize the signal intensity. MALDI MS2 experiments were performed using Bruker’s LIFT mode.196 Data analysis was conducted with the flexAnalysis software.

7.2.3. ESI MS and ESI-TWIM MS

Both ESI MS and ESI-TWIM MS experiments were performed with the Waters

Synapt HDMS quadrupole/time-of-flight (Q/ToF) mass spectrometer (Waters, Milford,

MA).72 The triwave region of this instrument, located between the Q and ToF mass analyzers, contains three confined regions in the order trap cell (closest to Q), TWIM cell, and transfer cell (closest to ToF). Either the trap or the transfer cell can be used for

MS2 studies via collisionally activated dissociation (CAD). The following ESI and

TWIM parameters were selected: ESI capillary voltage, 3.5 kV; sample cone voltage, 35

V; extraction cone voltage, 3.2 V; desolvation gas flow, 500 L/h (N2); trap collision energy (CE), 2 or 6 eV; transfer CE, 1 or 4 eV; trap gas flow, 1.5 mL/min (Ar); IM gas

116

flow, 22.7 mL/min (N2); sample flow rate, 5 μL/min; source temperature, 40 or 100 °C; desolvation temperature, 60 or 150 °C; traveling wave velocity, 350 m/s; traveling wave height, 8.5, 9.5, or 11 V, depending on the m/z ratio to be separated. The sprayed solutions were prepared by dissolving 0.3 mg of sample in 1 mL of MeOH or ethanol

(EtOH)/ACN (v/v, 50/50). In select experiments, a few droplets of a 10-mg/mL NaTFA solution in the same solvent, or of trifluoroacetic acid, were added to the sprayed mixture.

Conventional MS2 (CAD) spectra were acquired in the trap cell with the TWIM device turned off. MS2 (CAD) experiments combined with TWIM separation were performed in the trap cell (fragmentation before IM separation) or in the transfer cell (fragmentation after IM separation). All MS2 studies employed 60-eV collisions with Ar targets. A more detailed description of our TWIM experiments can be found elsewhere.185,186,189

Data analysis was conducted with the MassLynx 4.1 and DriftScope 2.1 programs of

Waters.

7.3. Results and Discussion

The following sections present the results and discuss the analyses performed to characterize complex polymer mixtures, in order to overcome ionization effects, detect minor products and identify different molecular architectures

7.3.1. MALDI MS Analysis

Matrices common in polymer analysis, such as DIT, DCTB, DHB, and CHCA, were employed both with and without the addition of NaTFA, NaI, or KTFA cationizing salt. In the presence of an alkali metal salt, the major species detected is the free poly(N-

117

butylglycine) without the NHC substituent (4 in Scheme 3.1), which gives rise to a

+ + narrow distribution of [4n+Na] or [4n+K] ions maximizing near the expected molecular weight (Figure 7.1). Replacing the trifluoroacetate salt with an iodide salt leads to the same result. In either case, if the detector gain is raised to observe minor products with higher sensitivity, a trace of NHC-poly(N-butylglycine), 3, is also observed in the form of

+ [3n+H] ions (cf. Figure 7.1c). Surprisingly, 3 becomes the predominant species

+ detected, as a distribution of [3n+H] ions, if no salt is added to the sample subjected to

MALDI (Figure 7.2). The highest signal/noise ratio and cleanest spectrum were obtained using HPA as matrix (Figure 7.2c), presumably because of the low internal energy transferred during protonation with this basic matrix.197,198

118

+ [413+Na] + 1493.2 [414+Na] (a) 1606.3 DIT/Na+

800 1600 m/z 1594.3

1468.9 1481.2 1582.0 1480 1520 1560 1600 m/z

(b) DCTB/Na+ 800 1600 m/z

1480 1520 1560 1600 m/z

+ + [413+K] [414+K] (c) 1509.1 1622.2 DHB/K+ 800 1600 m/z

+ + [310+H] [311+H] 1520.2 1598.0 1633.3 1500 1540 1580 1620 m/z

Figure 7.1. MALDI mass spectra of poly(N-butylglycine) 1500, synthesized via NHC-mediated polymerization of N-butyl-N-carboxyanhydride (Scheme 3.1.) using different matrices and cationizing salts; (a) DIT plus NaTFA; (b) DCTB plus NaTFA; (c) DHB plus KTFA. CHCA gives rise to similar spectra if NaTFA or KTFA is added. An increased detector gain was used for spectrum (c), which distorted the isotope pattern of the main ion series. The m/z values marked in the spectra are for the monoisotopic signals.

119

+ [35+H] 954.8 + [36+H] (a) 1067.9

CHCA 800 1600 m/z

960 1000 1040 m/z

+ [310+H] 1520.3 [3 +H]+ (b) 11 800 1600 m/z 1633.4 DHB + + [413+Na] [414+Na] 1619.4 1506.3 1493.2 1531.7 1606.3 1644.4 1500 1540 1580 1620 m/z

+ [3 +H]+ [311+H] (c) 10 1520.2 1633.3 HPA 1000 1800 m/z

1503.2 1542.2 1616.3 1500 1540 1580 1620 m/z

Figure 7.2. MALDI mass spectra of poly(N-butylglycine) 1500, synthesized via NHC-mediated polymerization of N-butyl-N-carboxyanhydride (Scheme 3.1.) using different matrices but no cationizing salt; (a) CHCA, (b) DHB , and (c) HPA. The molecular weight distribution obtained with HPA most closely agrees with the average molecular weight of ~1500 Da determined by size exclusion chromatography. The m/z values marked in the spectra are for the monoisotopic signals.

The basicity imparted to 3 by the NHC substituent explains the preference of this product to ionize via proton addition; in contrast, 4 ionizes by metal ion adduction, as generally do all polyamides.199 Based on Figure 7.2, HPA most closely reproduces the average molecular weight of ~1500 determined by NMR or size exclusion chromatography coupled with multi-angle light scattering and differential refractive index detection.

120

The NHC group is a UV chromophore that absorbs at the wavelength of the

MALDI laser (355 nm). Therefore, a mass spectrum can be obtained even without a matrix; however, this direct laser desorption causes photoinduced degradation, as the resulting spectrum (not shown) does not contain intact ions of 3 or 4. Some of the minor, uninterpretable distributions in Figures 7.1 and 7.2 could be due to such direct laser desorption/fragmentation.

NHC-mediated ring-opening polymerization is believed to proceed via the zwitterionic intermediate 2, which is in equilibrium with 3 (Scheme 3.1). The anion of the cationizing salt added to the MALDI sample (trifluoroacetate or iodide) presumably

– catalyzes the conversion 3 → 4 by replacing the NHC chain end of 2 with CF3COO

(TFA–) or I–, which is subsequently displaced, via intramolecular nucleophilic substitution, by the amide chain end of 2.200 A strategy similar to the one used to prepare

3 (Scheme 3.1) can be applied to lactide to synthesize cyclic polyesters.190-192 The

MALDI results discussed indicate that the major product in such syntheses is the NHC- substituted macrocycle, but that this macrocycle may readily lose the NHC group during

MALDI MS analysis, if the analyte sample contains salts, which are often deliberately added to enhance signal intensities. Since NHC-free product is not observed during ESI analysis, even if the sample is doped with an alkali salt (vide infra), the intramolecular nucleophilic substitution must proceed over a considerable barrier that can only be overcome under MALDI, which generally deposits higher average internal energies than the softer ESI method.197-198

121

31 502 + [4n+H] + NHC [3n+H] 30 389 45 4 47 4 566 43 37 793 453

4 340 46 2 3 3 4 * 36 410* 2 3 5 9 1181 33 4 48* 680 227 955 615 842 729 1068 905 1131 1018

200 400 600 800 1000 m/z

Figure 7.3. MALDI MS2 spectrum of the protonated 7-mer from NHC-poly(N- + butylglycine) 1500 ([37+H] at m/z 1181). The asterisks indicate fragments with more than 7 repeat units.

7.3.2. MALDI MS2 Characterization of the NHC-substituted Macrocycle

+ The protonated 7-mer of NHC-poly(N-butylglycine), viz. [37+H] at m/z 1181, was selected for a MALDI MS2 experiment; the resulting spectrum is shown in Figure

+ 7.3. When energetically activated, the [37+H] precursor ion dissociates to yield two fragment distributions, one containing NHC-bound poly(N-butylglycine) and one free

+ poly(N-butylglycine), which have been labeled in Figure 7.3 by ♦ ([3n+H] ) and ▲

+ ([4n+H] ), respectively. Dissociation most likely proceeds through the ring opened isomer (cf. Scheme 7.2) and involves nucleophilic displacements at the carbonyl groups by the terminal amine to generate 3n (n = 0-6) and 47-n fragments, either of which may keep the proton charge. For instance, bond cleavage at the carbonyl group next to the

NHC substituent would yield the free NHC (30) together with 47 (Scheme 7.1), which are observed in protonated form at m/z 389 and 793, respectively. Note that 47 is the largest 122

poly(N-butylglycine) fragment possible from the heptameric precursor ion selected.

+ Surprisingly, traces of [4n+H] fragments with more than 7 repeat units (n = 8-10) are also observed, cf. Figure 7.3; these are attributed to the presence of isobaric components

+ in the mass-selected [37+H] precursor ion beam, as will be corroborated by the ESI and

TWIM data (vide infra).

+ Scheme 7.1. Fragmentation of energetically activated [37+H] ions to yield fragments with or without the NHC moiety. The reaction is exemplified at the C=O group attached to NHC, but can similarly occur at the other C=O groups.

7.3.3. ESI MS Analysis

The ESI experiments were performed with a poly(N-butylglycine) sample with lower average molecular weight (1100) in order to avoid extensive multiple charging of the polymer, which complicates spectral interpretation. The ions generated by ESI reach the mass analyzing device after passing a desolvation zone and several focusing and guiding lenses; for labile or reactive species, the corresponding temperatures and 123

potentials must be kept low to minimize unintended dissociation (cf. Figure 7.4). Under such mild conditions, poly(N-butylglycine) 1100 gives rise to the spectrum depicted in

Figure 7.4a, in which the main distribution arises from the protonated NHC-carrying

+ polymer, viz. [3n+H] . A second distribution, peaking 388 Da higher, corresponds to noncovalent complexes of NHC-poly(N-butylglycine) and unreacted NHC, viz.

+ [3n+NHC+H] ; this distribution disappears at higher temperature and voltage settings

(Figure 7.4b), substantiating that the second NHC unit is attached noncovalently to the protonated oligomers of 3. The minor distribution observed 22 Da above the major

+ + [3n+H] series is due to [3n+Na] ions, originating from adventitious sodium. Supporting evidence for this assignment is provided by the increase in relative intensity of the latter ions when NaTFA is added to the sample (Figure 7.4c). The added NaTFA also causes

+ some H/Na exchange, which gives rise to [3n+2Na−H] species, cf. Figure 7.4c; this reaction may proceed by enolization of one amide group, followed by OH → ONa exchange. Addition of NaTFA also reduces the relative abundance of the noncovalent complex. The sodium ion most likely interacts with multiple amide sites of the poly(α- peptoid), leading to a geometry that cannot efficiently associate with a second NHC unit.

It is noteworthy, that no free poly(N-butylglycine) (4) is observed in the ESI mass spectra, regardless of whether NaTFA has been added or not. Hence, the presence of

NaTFA in the sample is not sufficient to cause the conversion 3 → 4; higher internal energies are also required. More importantly, the ESI spectra confirm that NHC- mediated ring-opening polymerization produces the NHC-substituted poly(α-peptoid) and that auxiliary reagents (e.g., NaTFA) and energetic excitation (as available in MALDI or

MS2) are necessary to obtain the corresponding free poly(α-peptoid).

124

1181 (n = 7) + (a) [3n+H] 113 [3 +Na]+ 1569 (n = 7) n + [3n+NHC+H] 113

800 1000 1200 1400 1600 1800 2000 m/z

1181 (n = 7) (b) + [3n+H] + [3n+Na]

800 1000 1200 1400 1600 1800 m/z

1181 (n = 7) (c) + [3n+H] + [3n+Na] + [3n+2Na−H]

800 1000 1200 1400 1600 1800 m/z

Figure 7.4. ESI mass spectra of poly(N-butylglycine) 1100 acquired (a,b) without or (c) with NaTFA cationizing salt. Other varied parameters: ESI source temperature (a,c) 40 oC or (b) 100 oC; desolvation temperature (a,c) 60 oC or (b) 150 oC; trap cell bias (a,c) 2 V or (b) 6 V; transfer cell bias (a,c) 1 V or (b) 4 V. The m/z values marked in the spectra are for the monoisotopic signals.

7.3.4. ESI MS2 characterization of the NHC-poly(α-peptoid) and its noncovalent adduct with NHC

The ESI MS2 spectrum of the protonated 7-mer from NHC-poly(N-butylglycine),

+ [37+H] (m/z 1181), shows the same fragment distributions that were observed upon

MALDI MS2, cf. Figure 7.5 vs. 7.3. For example, in both cases, cleavage of the NHC

125

group (388 Da) gives rise to major peaks at m/z 793 and 389, corresponding to the

+ + complementary fragments [47+H] (loss of NHC) and [30+H] (protonated NHC), respectively.

38* 1294 39*

4 * 4 * 1068 310*

8 955 9 1407 906 842 1250 + 1018 [4n+H] + 37 [3n+H] 840 940 1040 1280 1380 1480

47 NHC 1181 3 3 0 1 45 793 389 4 502 4 566 43 46 453 3 3 35 36 42 340 2 33 4 680 955 615 842 729 1068 227

200 400 600 800 1000 m/z

Figure 7.5. ESI MS2 spectrum of the protonated 7-mer from NHC-poly(N- + butylglycine) 1100 ([37+H] at m/z 1181). The asterisks indicate fragments with more than 7 repeat units.

A much simpler MS2 spectrum is obtained from the complex of the 7-mer and

+ NHC, [37+NHC+H] (m/z 1569), which mainly dissociates into its constituents, viz.

+ + [37+H] (m/z 1181) and [NHC+H] (m/z 389), cf. Figure 7.6; such fragmentation pattern corroborates that the binding interaction between the NHC-poly(α-peptoid) and the second NHC unit are of noncovalent nature. It is noticed that the ESI MS2 spectra of

+ + both [37+H] (Figure 7.5) as well as [37+NHC+H] (Figure 7.6) contain fragments with more than 7 repeat units, some with m/z values above those of the respective precursor

126

ions. This observation indicates that the mass-selected precursor ions (m/z 1181 or 1569) overlap with species in higher charge states (z >1). More highly charged ions could originate from larger poly(N-butylglycine) macrocycles or from products with different, not yet identified structures. The separation ability of TWIM MS, combined with MS2, will be used to characterize these species.

+ [37+H] [NHC+H]+ 311*

389 1181 + [3n+H] 3 *

1633 12

313* 3 * 36 38* 1746 14 3 39* 310* 34 5 1859 1972 1068 1294 955 1407 842 1520

850 1150 1450 1600 1900

+ [37+NHC+H] 1569

300 600 900 1200 1500 1800 m/z

Figure 7.6. ESI MS2 spectrum of the protonated cluster containing the 7-mer + from NHC-poly(N-butylglycine) 1100 and a second NHC unit ([37+NHC+H] at m/z 1569). The asterisks indicate fragments with more than 7 repeat units.

7.3.5. Separation of the synthetic product by ESI-TWIM MS

IMMS and its TWIM MS variant may be viewed as post-ionization chromatography methods that separate gas-phase ions according to their mass, charge, and shape.175-189 ESI-TWIM MS can be performed on all ions formed upon ESI, or ions of a specific m/z ratio, by setting the quadrupole mass filter preceding the TWIM region

127

to either rf-only mode or mass-selective mode, respectively.72,189,193 Inside the TWIM cell, ions drift under the influence of an electric field against a carrier gas flowing in the opposite direction. Ions of the same mass and charge can be dispersed by shape, because the more compact architecture will have a shorter drift time through the TWIM cell.

Conversely, different charge states of the same m/z ratio can be dispersed by the number of charges, because the more highly charged species will travel generally faster in the

TWIM cell. The m/z values of the separated ions are determined by ToF analysis and

(usually) plotted against the corresponding drift times, as shown in Figure 7.7 for all ions produced by ESI of poly(N-butylglycine) 1100 as well as for mass-selected m/z 1181.

m/z (a) 2+ 2000 1+

1000

2.0 6.0 10.0 14.0 18.0 22.0 ms

(b) m/z 1181 2.0 6.0 10.0 ms Figure 7.7. 2-D ESI-TWIM MS plots of (a) all ions and (b) mass-selected m/z 1181 from poly(N-butylglycine) 1100. The sample was dissolved in methanol (no cationization salt added). The source and desolvation temperatures were set at 40 and 60oC, respectively. Two discrete regions of ions are observed in Figure 7.7a; based on the corresponding isotope patterns, the region with longer drift times (at right) contains singly charged species and the region with shorter drift times (at left) doubly charged species. Each region can be integrated separately to obtain individual mass spectra. The 128

mass spectrum extracted from the 1+ species is similar with that acquired without TWIM

+ (Figure 7.4a), showing NHC-poly(N-butylglycine) ionized by protons, [3n+H] , or

+ + sodium ions, [3n+Na] , and the protonated noncovalent complex [3n+NHC+H] . On the other hand, the mass spectrum obtained from the 2+ species (Figure 7.8), shows five sizable distributions, only two of which are analogous to those detected for +1, viz.

2+ 2+ [3n+2H] and [3n+H+Na] . A third distribution arises from the adduct of 3n and the ESI

+ solvent (methanol), [3n+MeOH+H+Na] , and the remaining two from higher order complexes whose formation unveils important insight into the aggregation/self- assembling properties of NHC-poly(α-peptoid)s (vide infra). Unambiguous detection and identification of the doubly charged products is impossible without prior ion mobility separation due to their small concentration (~5% of total ions).

129

n=8 n=8 674.47 n=8 2+ 658.51 [3n+2H] 647.52 2+ [3n+H+Na] 2+ [3n+MeOH+H+Na] n=8 2+ [(3n+MeOH+H+Na)+3m] 2+ [(3n+MeOH+H+Na)+3m+3k] 674.47 - 36 n+m=14

1207.87 - 38 n+m+k=22 1854.37

700 900 1100 1300 1500 1700 1900 2100 m/z

Figure 7.8. Mass spectrum extracted from the 2+ region in the ESI-TWIM MS plot of poly(N-butylglycine) 1100 (Figure 7.7a). Five doubly charged ion distributions were identified. The inset shows an expanded view of the m/z 640- 680 range. Measured monoisotopic m/z ratios are given for one species in each distribution; the corresponding calculated m/z values are 647.49, 658.47, 674.49, 1207.89, and 1854.37, respectively.

The TWIM MS plot in Figure 7.7b illustrates the overlap of different charge states at a given m/z ratio. For m/z 1181, two distinct components are detected, a major singly charged and a minor doubly charged. The latter constituent reconciles the appearance of

2 + tiny fragments with more than 7 repeat units in the MS spectra of [37+H] (m/z 1181), as was discussed earlier.

130

m/z

1+ 2000

1000

ms 4.0 8.0 12.0 16.0 20.0

Figure 7.9. 2-D ESI-TWIM MS plot of poly(N-butylglycine) 1100 (all ions), dissolved in EtOH/ACN (v/v, 50/50). No cationization salt was added. The source and desolvation temperatures were set at 40 and 60oC, respectively.

The doubly charged distributions of Figure 7.8 were identified based on their exact m/z ratios and, especially, with the help of MS2 spectra, which will be discussed in detail in the following section. It is important to mention at this point that the yield of 2+ ions depends strongly on the solvent used for ESI. Replacing methanol with an ethanol/acetonitrile mixture essentially eliminates the 2+ charge state, cf. Figure 7.9. A reduction in the proportion of higher charge states with the use of ACN as solvent had been observed in the ESI mass spectra of proteins and was attributed to the higher basicity of this solvent;201 such reasoning also reconciles the differences between the ESI data in Figures 7.7a and 7.9.

7.3.6. ESI-TWIM MS2 characterization of the higher order noncovalent complexes

The identity of the higher mass distributions with 2+ charges was interrogated by combining TWIM with MS2. TWIM disperses the singly and doubly charged components of a selected m/z ratio (precursor ion), so that their MS2 fragmentation

131

patterns can be examined individually via CAD in the transfer cell that follows the

TWIM device. Fragments formed this way have the same drift time as their precursor ions. Those arising from the 2+ component of m/z 1854, a species in the high mass distribution of Figure 7.8 ( ) are encased in a rectangle in the 2-D TWIM MS2 plot of

Figure 7.10a; integration of the encased region leads to the ESI-TWIM MS2 spectrum shown in Figure 7.10b. All major fragments observed originate from losses of intact

NHC-poly(N-butylglycine) (3n) n-mers. Repeating the same procedure with doubly charged m/z 1208, a species in the middle distribution of Figure 7.8 ( ) and a fragment from m/z 1854, reveals a very similar fragmentation behavior, viz. mainly losses of intact

2 3n oligomers (cf. Figure 7.10c). A dramatically different TWIM MS spectrum is obtained for doubly charged m/z 674.5 (Figure 7.10d), a species in the low mass distribution of Figure 7.8 ( ) and a major fragment from m/z 1208. There is no loss of intact 3n oligomers; instead, NHC elimination gives rise to the most abundant signal, and most other fragmentation pathways lead to singly or doubly charged 4n (NHC-free poly(N-butylglycine)). These MS2 characteristics provide convincing evidence that the

2+ products of intermediate ( ) or high ( ) mass in Figure 7.8 correspond to noncovalent complexes, assembled from a core unit plus either one or two 3n units, respectively.

132

1854.4 m/z - 3 2+ 1+ n 1854 n = 4-9 1750 2+ 2+ (a) 2+ (b)

1500 2+ 1321.1 2+ 1377.6 1250 1264.5 - NHC 1208.0 8.0 11.0 14.0 ms 1660.4 m/z 1300 1500 1700 m/z

- 3n 1208.0

2+ n = 4-9 2+ 2+ 674.5 2+ (c) 618.0 1013.8 731.0 - NHC 389.3 400 600 800 1000 1200 m/z # 2+ + - NHC § [3n+H] (n = 0-1)

480.4 (d) + ¶ [4n+MeOH+Na] (n = 5-7) 2+ # [4n+MeOH+H+Na] (n = 6-8) § # § # 674.5

2+ ¶ ¶ 2+ ¶ 502.5 142.1 389.3 542.5 580.4 423.8 154.1 367.3 846.7 411.3 620.4 733.6

100 300 500 700 900 m/z

Figure 7.10. (a) 2-D ESI-TWIM MS2 plot of the m/z 1854 ion from poly(N- butylglycine) 1100 dissolved in MeOH and (b) ESI-TWIM MS2 mass spectrum of doubly charged m/z 1854 extracted from the encased region of the plot. (c,d) ESI-TWIM MS2 mass spectra of doubly charged (c) m/z 1208 and (d) 674.5 extracted from analogous 2-D TWIM MS2 plots (not shown). Doubly charged fragments are indicated by a 2+ superscript. The ions at m/z 142.1, 154.1, + 542.5, and 580.4 in part (d) agree well with (butyl)2N =CH2, [N- + + + butylglycine+Na] , [31+H2O+Na] , and [45+Me] , respectively.

+ + The core is composed of 3n + MeOH and is doubly charged by H and Na (labeled by in Figure 7.8). The 2+ distribution without MeOH or that ionized by two protons (↓ and ▼, respectively, in Figure 7.8) form no association products. Evidently, both MeOH as well as sodium are required to promote self-assembly. The exact structure of the core remains puzzling.

133

7.3.7. Is the poly(α-peptoid) formed after NHC release cyclic?

Our study with NHC-containing poly(N-butylglycine) has shown that such poly(α-peptoid)s can readily lose the NHC substituent either unintentionally, due to the sample preparation procedure used, or intentionally, by activation with collisions or laser light during MS2 experiments. Based on their mass, the resulting NHC-free poly(α- peptoid)s carry no end groups and, thus, have been assumed to be macrocyclic; however, isomeric linear structures with one saturated and one unsaturated chain end are also possible. For an unequivocal determination of the correct architecture, the poly(N- methylglycine) generated from the corresponding NHC-containing polymer was compared to an isomer, viz. cyclic poly(alanine), which was prepared by a method known to yield the cyclic polypeptide. ESI-TWIM MS of the poly(alanine) sample led to the 2-

+ D plot depicted in Figure 7.11a, in which protonated oligomers, [(Ala)n+H] , are the dominant ions (Ala abbreviates the repeat unit of alanine). NHC-free poly(N- methylglycine), produced by MS2 (CAD) of the corresponding NHC-containing product in the cell preceding the TWIM device (see Experimental), led to the 2-D plot depicted in

Figure 7.11b, where protonated oligomers also dominate. The drift times of isomers with the same number of repeat units are identical within experimental error. Also the corresponding drift time distributions are indistinguishable, as attested for the 6-, 7-, and

8-mers in the respective figure insets. These results clearly indicate that poly(alanine) and poly(N-methylglycine) n-mers of the same size have essentially the same geometries

(collision cross sections).175-182,185-187,189

134

(a) m/z cyclo-poly(alanine) n = 8 (MS) 7 6 2.76 n = 6 400 5 3.28 4 H3C O n = 7 HN NH 3.92 200 3 n = 8 O CH3 2 H C N O n-2 3 H 2.00 4.00 ms 1.0 2.0 3.0 4.0 5.0 ms (b) NHC-poly(N-methylglycine) 1000 poly(N-methylglycine) (MS2) 2.71 n = 6

m/z 3.28 O n = 7

n = 9 H3C N CH3 8 N 3.98 O n = 8 500 7 O N n-2 6 H3C 2.00 4.00 ms 2.0 4.0 6.0 8.0 10.0 ms

Figure 7.11. (a) 2-D ESI-TWIM MS plot of poly(alanine) (all ions formed upon ESI) and (b) 2-D ESI MS2-TWIM plot of the protonated 9-mer from NHC- poly(N-methylglycine) (m/z 1028.6). Both plots were acquired at the same TWIM conditions (traveling wave velocity of 350 m/s and traveling wave height of 8.5 V). The samples were dissolved in methanol (acidified by trifluoroacetic acid); source and desolvation temperatures were set at 40 and 60oC, respectively.

Linear poly(N-methylglycine) isomers would have had considerably longer drift times, because of their extended geometries (larger collision cross sections).202,203 Our

TWIM MS/MS2 findings confirm that NHC release leads to macrocyclic poly(α- peptoid)s in the gas phase of the mass spectrometer. This dissociation likely proceeds via acid-catalyzed, intramolecular nucleophilic substitution, as rationalized in Scheme 7.1.

135

Solvents that favor such an intramolecular (and, hence, unimolecular) pathway should also yield macrocyclic poly(α-peptoid) architectures in solution.

7.4. Conclusion

Mass spectrometry advancements in the past two decades have substantially facilitated the characterization of synthetic polymers. This chapter illustrated how multiple levels of mass spectrometry can be applied to gain comprehensive insight into the composition, structure, and architecture of a promising class of biomimetic materials, viz. poly(α-peptoid)s.

Without mass spectrometry analysis, it is difficult to conclusively deduce the real compositional heterogeneity of polymers. MALDI still is the most straightforward and widely used tool to obtain this information, because it produces mainly singly charged ions minimizing superimposed charge states, has a high tolerance for impurities and salts, and provides analysis with high sensitivity and speed. Inappropriate matrix and salt selection can, however, introduce erroneous results, especially for labile species. The experiments with poly(N-butylglycine) clearly show that the acquisition of spectra under a variety of experimental conditions is advisable to ensure that the polymer is not changed during the MALDI MS analysis. Compared to MALDI, ESI is softer and more suitable for labile materials, especially for products held together by noncovalent interactions. It also introduces multiple charging which may complicate the resulting spectra, but which also allows for the detection of products of high molecular weight that might escape detection in charge state 1+ (or 1-) due to mass discrimination effects. MS analysis can be further enhanced by ion mobility spectrometry and tandem mass

136

spectrometry. The former enables separation by charge state and/or architecture, making it possible to identify minor or overlapping components invisible in the total spectrum, while the latter confirms primary structure and the presence or absence of noncovalent interactions (which are weak and, hence, break most easily). Combining ion mobility separation with tandem MS further allows for the simultaneous separation and determination of polymer architectures. Here, all these top-down approaches have been employed to prove beyond doubt that NHC-mediated zwitterionic ring-opening polymerization yields products carrying the NHC functionality and that elimination of this NHC substituent opens a route to cyclic poly(α-peptoid)s having conformations that closely mimic those of cyclic polypeptides. At the same time, this study warns that, although mass spectrometry plays a vital role in synthetic polymer analysis, it may alter the analyte, yielding misleading data; the use of more than one approach is recommended to avoid such pitfalls.

137

CHAPTER VIII

SUMMARY

Over the years, mass spectrometry has been applied for the structural elucidation of synthetic and biological polymers. Nevertheless, the characterization of complex mixtures or polymers synthesized by new synthetic routes has remained challenging using single stage mass spectrometry. These analyses can be accomplished, however, if techniques such as ion mobility spectrometry (IMMS) and tandem mass spectrometry

(MS2) are used. Also, mass spectrometry is not a generally established quantitative analytical tool, especially if MALDI is the ionization technique. The addition of an internal standard and proper control of the experimental parameters (voltages, laser power, etc.) can alleviate this problem and increase the reproducibility, allowing quantitations to be performed using MALDI MS. In this dissertation, techniques such as

MS, MS2 and IMMS were combined for the accurate elucidation of complex mixtures and synthetic polymers. Additionally, protein adsorption on different surfaces was quantified.

Chapter IV reported the complete, accurate and successful characterization of the enzyme-catalyzed functionalization of PEGs with various vinyl esters. The functionalized

PEGs were synthesized by Puskas et al. via transesterification and Michael addition

138

reactions using CALB as a catalyst. To obtain a more precise structural analysis, for each sample, the molecule or one oligomer of interest was isolated and fragmented via collisionally activated dissociation tandem mass spectrometry. From PEGs prepared by already known methods, the MS2 fragments observed were those created by known PEG fragmentation pathways. On the other hand, new PEG macromolecules synthesized by

Michael addition were shown for the first time to decompose, under MS2 conditions, via retro-Michael reaction and intramolecular transesterification mechanisms.

The degradation products formed after the exposure of PLA to Proteinase K for a defined length of time are described in Chapter V. The results showed that degradation happens quickly and that the degradation products correspond to short linear polylactide oligomers. Because of the alkaline nature of the degradation medium, the short polylactide chains contained varying amounts of NAOH base. It was observed that degradation reaches quickly on equilibrium with an approximately constant oligomer composition.

The adsoption behavior of proteins on polymer surfaces, designed for use as biomaterials, is extremely important for biocompatibility studies. Chapter VI evaluated the adsorption of three model proteins onto D_IBS compression molded films and electrospun fiber mats. Insulin, ubiquitin and lysozyme were the proteins examined. The study demonstrated that D_IBS fiber mats adsorb more proteins than other evaluated surfaces due to their surface morphology and composition. Additionally, this study revealed that protein adsorption is highly influenced by the isoelectric point and molecular dimensions of the protein.

139

The applicability of combining different mass spectrometry techniques for the structural elucidation of the components in complex mixtures was discussed in Chapter

VII. Cyclic poly(α-peptoid)s prepared via N-heterocyclic carbene (NHC) mediated zwitterionic ring-opening polymerization were the system chosen to address the advantages of the mentioned approach. Interfacing MALDI MS, MALDI MS2, ESI MS,

ESI MS2, IMMS and IMMS2 for the study of poly(α-peptoid)s system, allowed many conclusions to be drawn: 1) NHC-free poly(α-peptoid)s has a cyclic structure; 2) NHC- containing poly(N-butylglycine) is the main product of the reaction; 3) NHC-containing poly(N-butylglycine) easily loses the NHC substituent to form NHC-free cyclic poly(α- peptoid)s; and 4) the formation of NHC-free poly(α-peptoid)s is promoted by high laser power and by nucleophiles present in the solutions used for MALDI sample preparation.

140

REFERENCES

1.De Hoffmann, E.; Stroobant, V. Mass Spectrometry Principles and Applications, 3rd ed.; John Wiley & Sons: Chichester, WestSussex,UK, 2007.

2.Woods, A. S.; Ugarov, M.; Egan, T.; Koomen, J.; Gillig, K. J.; Fuhrer, K.; Gonin, M.; Schultz, J. A. Lipid/Peptide/Nucleotide separation with MALDI-ion mobility- TOF MS. Analytical Chemistry, 2004, 76(8), 287-2195.

3.Macha, F.; Limbach, P. A. Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry of polymers, Current Opinion in Solid State and Materials Science, 2002, 6(3), 213-220.

4.Bu, L.; Nonidez, W. K.; Mays, J. W.; Tan, N. B. MALDI/TOF/MS and SEC study of astromol dendrimers having cyano end groups, Macromolecules, 2000, 33(12), 4445-4452.

5.Liu, J.; Loewe, R. S.; McCullough, R. D. Employing MALDI-MS on poly(alkylthiophenes):analysis of molecular weights, molecular weight distributions, end-group structures, and end-group modifications, Macromolecules, 1999, 32(18), 5777-5785.

6.Dey, M.; Castoro, J. A.; Wilkins, C. L. Determination of molecular weight distributions of polymers by MALDI-FTMS, Analytical Chemistry, 1995, 67(9), 1575-1579.

7.Li, L.; Overview of MS and MALDI MS for Polymer Analysis, in Maldi Mass Spectrometry for Synthetic Polymer Analysis, John Wiley & Sons: Hoboken, NJ, USA, 2009

141

8.Cao, P.; Stults, J. T. Mapping the phosphorylation sites of proteins using on-line immobilized metal affinity chromatography/capillary electrophoresis/electrospray ionization multiple stage tandem mass spectrometry, Rapid Communications in Mass Spectrometry, 2000, 14(17), 1600-1606.

9. Hsu, F. F., Turk, J. Studies on phosphatidylserine by tandem quadrupole and multiple stage quadrupole ion-trap mass spectrometry with electrospray ionization: structural characterization and the fragmentation processes, Journal of the American Society for Mass Spectrometry, 2005, 16(9), 1510-1522.

10.Waridel, P.; Wolfender, J. L.; Ndjoko, K.; Hobby, K. R.; Major, H. J.; Hostettmann, K. Evaluation of quadrupole time-of-flight tandem mass spectrometry and ion-trap multiple-stage mass spectrometry for the differentiation of C-glycosidic flavonoid isomers, Journal of Chromatography A, 2001, 926(1), 29-41.

11.Song, F.; Cui, M.; Liu, Z.; Yu, B.; Liu, S. Multiple-stage tandem mass spectrometry for differentiation of isomeric saponins, Rapid Communications in Mass Spectrometry, 2004, 18(19), 2241-2248.

12.Hsu, F. F; Turk, J. Electrospray ionization multiple-stage linear ion-trap mass spectrometry for structural elucidation of triacylglycerols: assignment of fatty acyl groups on the glycerol backbone and location of double bonds, Journal of the American Society for Mass Spectrometry, 21(4), 657-669.

13.Allen, L.T.; Tosetto, M.; Miller, I. S.; O’Connor, D. P., Penney, S. C.; Lynch, I.; Keenan, A. K.; Pennington, S. R.; Dawson, K. A.; Gallagher, W. M. Surface- induced changes in protein adsorption and implications for cellular phenotypic responses to surface interaction, Biomaterials, 2006, 27(16), 3096-3108.

14.Chen, C.S.; Mrksich, M.; Huang, S.; Whitesides, G. M.; Ingber, D. E. Geometric control of cell life and death, Science, 1997, 276(5317), 1425-1428.

15.Cheng, X.; Canavan, H. E.; Stein, M. J.;Hull, J. R.; Kweskin, S. J.; Wagner, M. S.; Somorjai, G. A.; Castner, D. G.; Ratner, B. D. Surface chemical and mechanical properties of plasma-polymerized N-isopropylacrylamide, Langmuir, 2005, 21(17), 7833-7841.

142

16.El Fray, M.; Prowans, P.; Puskas, J. E.; Altsädt, V. Biocompatibility and fatigue properties of polystyrene-polyisobutylene-polystyrene, an emerging thermoplastic elastomeric biomaterial, Biomacromolecules, 2006, 7(3), 844-850.

17.Jiang, L.; Zhao, Y.; Zhai, J. A lotus-leaf-like superhydrophobic surface: a porous microsphere/nanofiber composite film prepared by electrohydrodynamics, Angewandte Chemie, 2004, 116(33), 4438-4441.

18.Bognitzki, M.; Hou, H.; Ishaque, M.; Frese, T.; Hellwig, M.; Schwarte, C.; Schaper, A.; Wendorff, J. H.; Greiner, A. Polymer, metal, and hybrid nano- and mesotubes by coating degradable polymer template fibers (TUFT Process), Advanced Materials, 2000, 12(9), 637-640.

19.Oleschuk, R.D.; McComb, M. E.; Chow, A.; Ens, W.; Standing, K. G.; Perreault, H.; Marois, Y.; King, M. Characterization of plasma proteins adsorbed onto biomaterials by MALDI-TOFMS, Biomaterials, 2000, 21(16), 1701-1710.

20.McLafferty, F.W. Tandem mass spectrometry, Science, 1981, 214(4518), 280-287.

21.Wesdemiotis, C.; Solak, N.; Polce, M. J.; Dabney, D. E.; Chaicharoen, K.; Katzenmeyer, B. Fragmentation pathways of polymer ions, Mass Spectrometry Reviews, 30(4), 523-559.

22.Polce, M. J.; Wesdemiotis, C. Tandem Mass Spectrometry and Polymer Ion Dissociation, in MALDI Mass Spectrometry for Synthetic Polymer Analysis. Li, L., Ed.; John Wiley & Sons, Hoboken, New Jersey, 2010, Chapter 5, pp. 85-127.

23.Roberts, M. J.; Bentley, M. D.; Harris, J. M. Chemistry for peptide and protein PEGylation, Advanced Drug Delivery Reviews, 2002, 54(4), 459-476.

24.Greenwald, R. B.; Choe, Y. H.; McGuire, J.; Conover, C. D. Effective drug delivery by PEGylated drug conjugates, Advanced Drug Delivery Reviews, 2003, 55(2), 217-250.

25.Hein, C.; Liu, X. M.; Wang, D. Click chemistry, a powerful tool for pharmaceutical sciences, Pharmaceutical Research, 2008, 25(10), 2216-2230.

143

26.Puskas, J. E.; Seo, K. S.; Sen, M. Y. Green polymer chemistry: precision synthesis of novel multifunctional poly(ethylene glycol)s using enzymatic catalysis, European Polymer Journal, 2011, 47, 524-534.

27.Puskas, J.D.; Sen, M.Y.; Kasper, J.R. Green polymer chemistry: telechelic poly(ethylene glycol)s via enzymatic catalysis, Journal of Polymer Science Part A: Polymer Chemistry, 2008, 46, 3024-3028.

28.Li, S. M.; Vert, M. Morphological changes resulting from the hydrolytic degradation of stereocopolymers derived from L- and DL-Lactides, Macromolecules, 1994, 27(11), 3107-3110.

29.Bourbigot, S.; Gilman, J. W.; Wilkie, C. A. Kinetic analysis of the thermal degradation of polystyrene-montmorillonite nanocomposite, Polymer Degradation and Stability, 2004, 84(3), 483-492.

30.Hakkarainen, M.; Albertsson, A. C.; Karlsson, S. Weight losses and molecular weight changes correlated with the evolution of hydroxyacids in simulated in vivo degradation of homo- and copolymers of PLA and PGA, Polymer Degradation and Stability, 1996, 52(3), 283-291.

31.Gupta, M. C.; Deshmukh , V. G. Thermal oxidative degradation of poly-lactic acid, Colloid & Polymer Science, 1982, 260(5), 514-517.

32.Yamashita, K.; Kikkawa, Y.; Kurokawa, K.; Doi, Y. Enzymatic degradation of poly(l-lactide) Film by proteinase K: quartz crystal microbalance and atomic force microscopy study. Biomacromolecules, 2005, 6(2), 850-857.

33.Numata, K.; Srivastava, R. K.; Finne-Wistrand, A.; Albertsson, A. C.; Doi, Y.; Abe, H. Branched poly(lactide) synthesized by enzymatic polymerization: effects of molecular branches and stereochemistry on enzymatic degradation and alkaline hydrolysis, Biomacromolecules, 2007, 8(10), 3115-3125.

34.Li, S.; Girard, A.; Garreau, H.; Vert, M. Enzymatic degradation of polylactide stereocopolymers with predominant d-lactyl contents, Polymer Degradation and Stability, 2000, 71(1), 61-67.

144

35.Baumbach, J. I.; Eiceman, G. A. Ion mobility spectrometry: arriving on site and moving beyond a low profile, Applied Spectroscopy, 1999, 53(9), 338A-355A.

36.Ruotolo, B. T. ; Benesch, J. L.; Sandercock, A. M.; Hyung, S. J.; Robinson, C. V. Ion mobility-mass spectrometry analysis of large protein complexes, Nature Protocols, 2008, 3(7), 1139-1152.

37.Pringle, S. D. An investigation of the mobility separation of some peptide and protein ions using a new hybrid quadrupole/travelling wave IMS/oa-ToF instrument, International Journal of Mass Spectrometry, 2007, 261, 1-12.

38.Chan, Y. T.; Li, X.; Soler, M.; Wang, J. L.; Wesdemiotis, C.; Newkome, G. R. Self-assembly and traveling wave ion mobility mass spectrometry analysis of hexacadmium macrocycles, Journal of the American Chemical Society, 2009, 131(45), 16395-16397.

39.Valentine, S. J. ; Plasencia, M. D.; Liu, X.; Krishnan, M.; Naylor, S.; Udseth, H. R.; Smith, R. D.; Clemmer, D. E. Toward plasma proteome profiling with ion mobility-mass spectrometry, Journal of Proteome Research, 2006, 5(11), 2977- 2984.

40.Trimpin, S.; Plasencia, M. D.; Isailovic, D.; Clemmer, D. E. Resolving oligomers from fully grown polymers with IMS-MS, Analytical Chemistry, 2007, 79(21), 7965-7974.

41.Williams, J. P.; Grabenauer, M.; Holland, R. J.; Carpenter, C. J.; Wormald, M. R.; Giles, K.; Harvey, D. J.; Bateman, R. H.; Scrivens, J. H.; Bowers, M. T. Characterization of simple isomeric oligosaccharides and the rapid separation of glycan mixtures by ion mobility mass spectrometry, International Journal of Mass Spectrometry, 2010, 298(1-3), 119-127.

42.Williams, J.P.; Giles, K.; Green, B. N.; Scrivens, J. H.; Bateman, R. H. Ion mobility augments the utility of mass spectrometry in the identification of human hemoglobin variants, Rapid Communications in Mass Spectrometry, 2008, 22(20), 3179-3186.

145

43.Hilton, G.R.; Jackson, A. T.; Thalassinos, K.; Scrivens, J. H. Structural analysis of synthetic polymer mixtures using ion mobility and tandem mass spectrometer, Analytical Chemistry, 2008, 80(24), 9720-9725.

44.Solak, N. Structural Characterization and Quantitative Analysis by Interfacing Liquid Chromatography and/or Ion Mobility Separation with Multi-dimensional Mass Spectrometry, University of Akron, Department of Chemistry, May, 2010.

45.Fowler, S. A.; Blackwell, H. E. Structure-function relationships in peptoids: Recent advances toward deciphering the structural requirements for biological function, Organic & Biomolecular Chemistry, 2009, 7(8), 1508-1524.

46.Thakkar, A.; Cohen, A. S.; Connolly, M. D.; Zuckermann, R. N.; Pei, D. High- throughput sequencing of peptoids and peptide-peptoid hybrids by partial edman degradation and mass spectrometry, Journal of Combinatorial Chemistry, 2009, 11(2), 294-302.

47.Morishetti, K. K. ; Russell, S. C.; Zhao, X.; Robinson, D. B.; Ren, J. Tandem mass spectrometry studies of protonated and alkali metalated peptoids: enhanced sequence coverage by metal cation addition, International Journal of Mass Spectrometry, 2011, 308(1), 98-108.

48.Siuzdak, G.; The expanding role of mass spectrometry in biotechnology; MCC Press: San Diego: 2006.

49.Dole, M.; Mack, L. L.; Hines, R. L. Mobley, R. C.; Ferguson, L. D.; Alice, M. B. Molecular Beams of Macroions, Journal of Chemical Physics, 1968, 49, 2240- 2249.

50.Grayson, M. John Bennett Fenn: A curious road to the prize, Journal of The American Society for Mass Spectrometry, 2011, 22(8), 1301-1308.

51.Oberacher, H. On the use of different mass spectrometric techniques for characterization of sequence variability in genomic DNA, Analytical and Bioanalytical Chemistry, 2008, 391(1), 135-149.

52.Iribarne, J. V.; Thomson, B. A. On the evaporation of small ions from charged droplets, Journal of Chemical Physics, 1976, 64, 2287-2294. 146

53.Cole, R.B. Electrospray ionization mass spectrometry: fundamentals, instrumentation, and applications; John Wiley & Sons: New York City, New York, USA, 1997

54.Gaskell, S. J. Electrospray: principles and practices, Journal of Mass Spectrometry, 1997, 32, 677-688.

55.Karas, M.; Hillenkamp, F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 daltons, Analytical Chemistry, 1988, 60(20), 2299-2301.

56.Karas, M.; Bachmann, D.; Bahr, U,; Hillenkamp, F. Matrix-assisted ultraviolet laser desorption of non-volatile compounds, International Journal of Mass Spectrometry and Ion Processes, 1987, 78(0), 53-68.

57.Available at http://www.proteomicsnijmegen.nl/Maldi_pages/Maldi-about.htm (MALDI-ToF MS, Nijmegen Proteomics Facility webpage edited by Dr. Huub op den Camp, Dr. Jan Keltjens and Dr. Peter Steenbakkers)

58.Available at http://www.aps.org/publications/apsnews/200104/history.cfm, American Physical Society News, April 2001, 10 (4), 1-8.

59.Cotter, R.J. Time-of-flight mass spectrometry, Encyclopedia of Genetics, Genomics, Proteomics and Bioinformatics; John Wiley & Sons, Ltd, 2005.

60.Sturiale, L.; Naggi, A.; Torri, G. MALDI mass spectrometry as a tool for characterizing glycosaminoglycan oligosaccharides and their interaction with proteins, Semin Thromb Hemost, 2001, 27(05), 465,472.

61.Montaudo, G.; Lattimer, R. P. Mass spectrometry of polymers; CRC Press: Boca Raton, Florida, USA, 2002.

62.Dass, C. Principles and practice of biological mass spectrometry; John Wiley: New York City, New York, USA, 2001.

63.Cooks, R.G.; Kaiser, R. E. Quadrupole ion trap mass spectrometry, Accounts of Chemical Research, 1990, 23(7), 213-219. 147

64.March, R.E. An Introduction to Quadrupole Ion Trap Mass Spectrometry, Journal of Mass Spectrometry, 1997, 32, 351-369.

65.Q-Tof UltimaTM MALDI User’s Guide, Micromass.

66.Waters Synapt HDMS System Operator’s Guide, Waters Corporation, 2006.

67.Kanu, A.B.; Dwivedi, P.; Tam, M.; Matz, L.; Hill, H.H. Jr. Ion Mobility-mass Spectrometry, Journal of Mass Spectrometry, 2008, 43,1-22.

68.Johnson, J. V.; Yost, R. A.; Kelley, P. E.; Bradford, D. C. Tandem-in-space and tandem-in-time mass spectrometry: triple quadrupoles and quadrupole ion traps, Analytical Chemistry, 1990, 62(20), 2162-2172.

69.Guo, L.; Zhang, D. Cyclic poly(α-peptoid)s and their block copolymers from N- heterocyclic carbene-mediated ring-opening polymerizations of N-substituted N- carboxylanhydrides, Journal of the American Chemical Society, 2009, 131, 18072-18074.

70.Zhou, N.; Lu, K.; Liu, Y.; Chen, Y.; Tang, G.; Cao, S.-X.; Qu, L.-B.; Zhao, Y.-F. Electrospray ionization mass spectrometric studies of phosphorus oxychloride directed synthesis of homo-oligopeptide ester libraries, Rapid Communications in Mass Spectrometry, 2002, 16, 919-922.

71.Esquire~LC Operations Manual, Bruker Daltonics-Hewlett Packard.

72. Giles, K.; Pringle, S. D.; Worthington, K. R.; Little, D.; Wildgoose, J. L.; Bateman, R. H. Applications of a traveling wave-based radio-frequency-only stacked ring ion guide, Rapid Communications in Mass Spectrometry, 2004, 18, 2401-2414.

73. Kadokawa, J.; Kobayashi, S. Polymer synthesis by enzymatic catalysis, Current Opinion in Chemical Biology, 2010, 14, 145-153.

148

74. Kobayash, S. Recent developments in lipase-catalyzed synthesis of polyesters, Macromolecules Rapid Communications, 2009, 30, 237-266.

75. Sen, M.Y.; Puskas, J.E.; Ummadisetty, S.; Kennedy, J.P. Green polymer chemistry: II. enzymatic synthesis of methacrylate-terminated polyisobutylenes, Macromolecular Rapid Comunications, 2008, 29, 1598-1602.

76. Dunn, P.J.; Wells, A.S.; Williams, M.T. Chapter 1. Introduction to Green Chemistry, Organic Synthesis and Pharmaceuticals, Green chemistry in the pharmaceutical industry, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2010.

77. Albertsson, A.C.; Srivastava, R.K. Recent developments in enzyme-catalyzed ring-opening polymerization, Advanced Drug Delivery Reviews, 2008, 60, 1077- 1093.

78. Shogren, R.L.; Willett, J.L.; Biswas, A. HRP-mediated synthesis of starch– polyacrylamide graft copolymers, Carbohydrate Polymers, 2009, 75, 189-191.

79. Ochoteco, E.; Mecerreyes, D. Oxireductases in the enzymatic synthesis of water- soluble conducting polymers, Advances in Polymer Science, 2010, 237, 1-19.

80. Puskas, J.E.; Sen, M.Y.; Seo, K.S. Green polymer chemistry using nature’s catalysts-enzymes, Journal of Polymer Science: Part A: Polymer Chemistry, 2009, 47, 2959-2976.

81. Veld, M.A.J.; Palmans, A.R.A. Hydrolases Part I: enzyme mechanism, selectivity and control in the synthesis of well-defined polymers, Advances in Polymer Science, 2011, 237, 55-78.

82. Erdemir, S.; Yilmaz, M. Synthesis of calix[n]arene-based silica polymers for lipase immobilization, Journal of Molecular Catalysis B: Enzymatic, 2009, 58, 29-35.

149

83. Kartal, F.; Janssen, M.H.A.; Hollmann, F.; Sheldon, R.A.; Kilic, A. Improved esterification activity of candida rugosa lipase in organic solvent by immobilization as cross-linked enzyme aggregates (CLEAs), Journal of Molecular Catalysis B: Enzymatic, 2011, 71, 85-89.

84. Palsule, A.S.; Poojari, Y. Enzymatic synthesis of silicone fluorinated aliphatic polyesteramides, Polymer, 2010, 51, 6161-6167.

85. Chen, Y.; Xiao, B.; Chang, J.; Fu, Y.; Lv, P.; Wang, X. Synthesis of biodiesel from waste cooking oil using immobilized lipase in fixed bed reactor, Energy Conversion and Management, 2009, 50, 668-673.

86. Kramer, M.; Cruz, J.C.; Pfromm, P.H.; Rezac, M.E.; Czermak, P. Enantioselective transesteriﬁcation by candida antarctica lipase B immobilized on fumed silica, Journal of Biotechnology, 2010, 150, 80-86.

87. Netto, C.G.C.M.; Andrade, L.H.; Toma, H.E. Enantioselective transesteriﬁcation catalysis by candida antarctica lipase immobilized on superparamagnetic nanoparticles, Tetrahedron: Asymmetrtry, 2009, 20, 2299-2304.

88. Dhake, K.P.; Tambade, P.J.; Singhal, R.S.; Bhanage, B.M. Promiscuous Candida antarctica lipase B-catalyzed synthesis of b-amino esters via aza-Michael addition of amines to acrylates, Tetrahedron Letters, 2010, 51, 4455-4458.

89. Priego, P.; Ortiz-Nava, C.; Carrillo-Morales, M.; Lopez-Munguia, A. Solvent engineering: an effective tool to direct chemoselectivity in a lipase-catalyzed Michael addition, Tetrahedron, 2009, 65, 536-539.

90. Pasut, G.; Veronese, F.M. Improvement of drug therapy by covalent PEG conjugation: an overview from a research laboratory, Israel Journal of Chemistry, 2010, 50, 151-159.

91. Li, J.; Tan, D.; Zhang, X.; Tan, H.; Ding, M.; Wan, C.; Fu, Q. Preparation and characterization of nonfouling polymer brushes on poly(ethylene terephthalate) film surfaces, Colloids and Surfaces B: Biointerfaces, 2010, 78, 343-350.

150

92. Parshetti, G.K.; Doong, R. Dechlorination of trichloroethylene by Ni/Fe nanoparticles immobilized in PEG/PVDF and PEG/nylon 66 membranes, Water Research, 2009, 43, 3086-3094.

93. Mero, A.; Pasut, G.; Veronese, F.M. PEGylated Protein Drugs: Basic Science and Clinical Applications, PEG: a useful technology in anticancer therapy, Birkhauser Verlag, Germany, 2009.

94. Chen, H.; Khemtong, C.; Yang, X.; Chang, X.; Gao, J. Nanonization strategies for poorly water-soluble drug, Drug Discovery Today, 2011, 16, 354-360.

95. Bruno, F.F.; Trotta, A.; Fossey, S.; Nagarajan, S.; Nagarajan, R.; Samuelson, L.; Kumar, J. Enzymatic Synthesis and Characterization of PolyQuercetin, Journal of Macromolecular Science, Part A: Pure and Applied Chemistry, 2010, 47, 1191-1196.

96. Medina-Ramirez, I.; Bashir, S.; Luo, Z.; Liu, J.L. Green synthesis and characterization of polymer-stabilized silver nanoparticle, Colloids and Surfaces B: Biointerfaces, 2009, 73, 185-191.

97. Wang, H.Y.; Zhang, X.; Guo, Y.L.; Lu, L. Mass spectrometric studies of the gas phase retro-Michael type fragmentation reactions of 2-hydroxybenzyl-n- pyrimidinylamine derivatives, Journal of The American Society for Mass Spectrometry, 2005, 16, 1561-1573.

98. Lucas, N.; Bienaime, C.; Belloy, C.; Queneudec, M.; Silvestre, F. Polymer biodegradation: mechanisms and estimation techniques, Chemosphere, 2008, 73, 429-442.

99. Newkome, G.R. Poly(propylene imine) dendrimers, Advances in Dendritic Macromolecules, JAI Press Inc.,Volume 4, Stamford, 1999.

100.Stridsberg, K.M.; Ryner, M.; Albertsson, A.C. Controlled Ring-Opening Polymerization: Polymers with Designed Macromolecular Architecture, Advances in Polymer Science, Springer-Verlag, Berlin Heidelberg, 2002.

151

101.Selby, T. L.; Wesdemiotis, C.; Lattimer, R. P. Dissociation characteristics of [M + H]+ ions (x = H, Li, Na, K) from linear and cyclic polyglycols, Journal of the American Society for Mass Spectrometry, 1994, 5, 1081-1092.

102.Yamamoto, N.; Nakayama, A.; Oshima, M.; Kawasaki, N.; Aiba, S. Enzymatic hydrolysis of lysine diisocyanate based polyurethanes and segmentedpolyurethane ureas by various proteases, Reactive & Functional Polymers, 2007, 67, 1338-1345.

103. Nakayama, A.; Kawasaki, N.; Arvanitoyannis, I.; Iyoda, J.; Yamamoto, N. Synthesis and degradability of a novel aliphatic polyester: poly(β-methyl-δ- valerolactone-co- -lactide), Polymer, 1995, 36(6), 1295-1301.

104. Li, S.; McCarthy, S. Influence of crystallinity and stereochemistry on the enzymatic degradation of poly(lactide)s , Macromolecules, 1999, 32, 4454-4456.

105.Liu, L.; Li, S.; Garreau, H.; Vert, M. Selective enzymatic degradations of poly(L- lactide) and poly(E-caprolactone) blend films, Biomacromolecules, 2000, 1, 350- 359.

106. Numata, K.; Finne-Wistrand, A.; Albertsson, A.C.; Doi, Y.; Abe, H. Enzymatic degradation of monolayer for poly(lactide) revealed by real-time atomic force microscopy: effects of stereochemical structure, molecular weight, and molecular branches on hydrolysis rates, Biomacromolecules, 2008, 9, 2180-2185.

107. Lenglet, S.; Li, S.; Vert, M. Lipase-catalysed degradation of copolymers prepared from 3-caprolactone and DL-lactide, Polymer Degradation and Stability, 2009, 94, 688-692.

108. Sato, H.; Kiyono, Y.; Ohtani, H.; Tsuge, S.; Aoi, H.; Aoi, K. Evaluation of biodegradation behavior of poly(ε-caprolactone) with controlled terminal structure by pyrolysis-gas chromatography and matrix-assisted laser desorption/ionization mass spectrometry, Journal of Analytical and Applied Pyrolysis, 2003, 68-69, 37-49.

109. Middleton, J. C.; Tipton, A. J. Synthetic biodegradable polymers as orthopedic devices, Biomaterials, 2000, 21, 2335-2346.

152

110. Oda, Y.; Yonetsu, A.; Urakami, T.; Tonomura, K. Degradation of polylactide by commercial proteases, Journal of Polymers and the Environment, 2000, 8, 29-32.

111. Tokiwa, Y.; Jarerat, A. Biodegradation of poly (L-lactide), Biotechnology Letters, 2004, 26, 771-777.

112.Ebeling, W.; Hennrich, N.; Klockow, M.; Metz, H.; Orth, H.D.; Lang, H. Proteinase K from Tritirachium album Limber, European Journal of Biochemistry, 1974, 47, 91-97.

113. Li, S.; Girard, A.; Garreau, H.; Vert, M. Enzymatic degradation of polylatide stereocopolymers with predominant D-lactyl contents, Polymer Degradation and Stability, 2001, 71, 61-67.

114. Yamashita, K.; Kikkawa, Y.; Kurokawa, K.; Doi, Y. Enzymatic degradation of poly (L-lactide) film by proteinase K: quartz crystal microbalance and atomic force microscopy study, Biomacromolecules, 2005, 6, 850-857.

115. Reevee, M.S.; McCarthy, S.P.; Downey, M.J.; Gross, R.A. Polylactide stereochemistry: effect on enzymatic degradability, Macromolecules, 1994, 27, 825-831.

116.Available at http://research.chem.psu.edu/brgroup/pKa_compilation.pdf (pKa Data Compiled by R. Williams)

117. Steinbach, C.; Buck, U. Reaction and solvation of sodium in hydrogen bonded solvent clusters, Physical Chemistry Chemical Physics, 2005, 7, 986-990.

118.Paizs, B.; Suhai, S. Fragmentation pathways of protonated peptides, Mass Spectrometry Reviews, 2005, 24, 508-548.

119.Saltzman, W. M.; Kryiakides, T. R. Chapter 20. Cell Interaction with Polymers, In Principles of Tissue Engineering, Elsevier Academic Press, 2007.

153

120. Woo, K. M.; Chen, V. J.; Ma, P. X. Nano-fibrous scaffolding architecture selectively enhances protein adsorption contributing to cell attachment, Journal of Biomedical Materials Research, 2003, 67, 531 – 537.

121. Horbett, T.A. The role of adsorbed proteins in animal cell adhesion, Colloids and Surfaces B Biointerfaces, 1994, 2, 225 – 240.

122. Veiseh, M.; Turley, E. A.; Bisell, M. J. Chapter 2.Top-down analysis of a dynamic environment: extracellular matrix structure and function, Nanotechnology and Tissue Engineering- The Scaffold, CRC Press, Taylor & Francis Group, LLC, 2008.

123. Allen, L. T.; Tosetto, M.; Miller, I. S.; O’Connor, D. P.; Penney, S. C.; Lynch, I. Surface-induced changes in protein adsorption and implications for cellular phenotypic responses to surface interaction, Biomaterials, 2006, 27(16), 3096 – 3108.

124. Chen, C. S.; Mrksich, M.; Huang, S.; Whitesides, G. M.; Ingber, D. E. Geometric control of cell life and death, Science, 1997, 276 (5317), 1425 – 1428.

125. Cheng, X.; Canavan, H. E.; Stein, M. J.; Hull, J. R.; Kwesking, S. J.; Wagner, M. S. Somorjai, G.A.; Castner, D.G.; Ratner, B. D. Surface chemical and mechanical properties of plasma polymerized N-isopropylacrylamide, Langmuir, 2005, 21(17), 7833 – 7841.

126. Duck, F. A. Physical properties of tissue – A comprehensive reference book, Academic Press, 1990, 143-160.

127. Kennedy, J. P.; Puskas, J. E.; Kaszas, G.; Hager, W. G. US Patent 4,946,899, 1990.

128. US FDA, TAXUS Express2TM Paclitaxel-eluting coronary stent system (monorail and over the wire), 2004; PO30025.

129. Puskas, J. E.; Paulo, C.; Antony, P. US Patent 6,747,098, 2004.

154

130. Puskas, J. E.; Foreman-Orlowski, E. A.; Lim, G. T.; Porosky, S. E.; Evancho- Chapman, M. M.; Schmidt, S. P.; El Fray, M.; Piątek, M.; Prowans, P.; Lovejoy, K. A nanostructured carbon-reinforced polyisobutylene-based thermoplastic elastomer, Biomaterials, 2010, 31(9), 2477-2488.

131. Puskas, J. E.; Chen, Y. H. Biomedical application of commercial polymers and novel polyisobutylene-based thermoplastic elastomers for soft tissue replacement, Biomacromolecules, 2004, 5(4), 1141 – 1154.

132. Puskas, J. E.; Kwon, Y. M.; Antony, P.; Bhowmick, A. K. Synthesis and characterization of novel dendritic (arborescent, hyperbranched) polyisobutylene–polystyrene block copolymers, Journal of Polymer Science Part A – Polymer Chemistry, 2005, 43(9), 1811 – 1826.

133. Puskas, J.E.; Fray, M.E.; Tomkins, M.; Dos Santos, L.M.; Fischer, F., Altstädt, V. Dynamic stress relaxation of thermoplastic elastomeric biomaterials, Polymer, 2009, 50, 245–249.

134. Puskas, J. E.; Kwon, Y. M. Biomacromolecular engineering: design, synthesis and characterization. One-pot synthesis of block copolymers of arborescent polyisobutylene and polystyrene, Polymer Advanced Technologies, 2006, 17(9-10), 615 – 620.

135. Puskas, J. E.; Dos Santos, L. M.; Sen, M. Y. Effect of architecture on the properties of polyisobutylene-based TPE materials, Rubber Chemistry and Technology, 2007, 80, 661 – 671.

136. Puskas, J. E.; Dos Santos, L. M.; Kaszas, G.; Kulbaba, K. Novel thermoplastic elastomers based on arborescent (dendritic) polyisobutylene with short copolymer end sequences, Journal of Polymer Science Part A – Polymer Chemistry, 2009, 47(4), 1148 – 1158.

137. Pham, Q.P.; Sharma, U.; Mikos, A.G. Electrospinning of polymeric nanofibers for tissue engineering applications: a review, Tissue Engineering, 2006, 12(5), 1197-1211.

155

138. Sokolsky-Papkov, M.; Agashi, K.; Olaye, A.; Shakesheff, K.; Domb, A. J. Polymer carriers for drug delivery in tissue engineering, Advanced Drug Delivery Reviews, 2007, 59, 187–206.

139. Ito, Y.; Hasuda, H.; Kamitakahara, M.; Ohtsuki, C.; Tanihara, M.; Kang, I.; Kwon, O. H. A composite of hydroxyapatite with electrospun biodegradable nanofibers as a tissue engineering material, Journal of Bioscience and Bioengineering, 2005, 100 (1), 43–49.

140. Li, C.; Vepari, C.; Jin, H. J.; Kim, H. J.; Kaplan, D. L. Electrospun silk- BMP-2 scaffolds for bone tissue engineering, Biomaterials, 2006, 27, 3115– 3124.

141. Zong, X.; Bien, H.; Chung, C.; Yin, L.; Fang, D.; Hsiao, B. S.; Chua, B.; Entcheva, E. Electrospun fine-textured scaffolds for heart tissue constructs, Biomaterials, 2005, 26, 5330–5338.

142. Choi, J. S.; Leong, K. W.; Yoo, H. S. In vivo wound healing of diabetic ulcers using electrospun nanofibers immobilized with human epidermal growth factor (EGF), Biomaterials, 2008, 29, 587– 596.

143.Ma, M. L.; Mao, Y.; Gupta, M.; Gleason, K. K.; Rutledge, G. C. Superhydrophobic fabrics produced by electrospinning and chemical vapor deposition, Macromolecules, 2005, 38(23), 9742-9748.

144.Ma, M.; Hill, R. M.; Rutledge, G. C. A review of recent results on superhydrophobic materials based on micro- and nanofibers, Journal of Adhesion Science and Technology, 2008, 22(15), 1799-1817.

145.Miyauchi, Y.; Ding, B.; Shiratori, S. Fabrication of a silver-ragwort-leaf-like super-hydrophobic micro/nanoporous fibrous mat surface by electrospinning, Nanotechnology, 2006, 17(20), 5151-5156.

146. Erbil, H. Y.; Demirel, A. L.; Avci, Y.; Mert, O. Transformation of a simple plastic into a superhydrophobic surface, Science, 2003, 299(5611), 1377- 1380.

156

147.Reneker, D. H.; Chun, I. Nanometre diameter fibres of polymer, produced by electrospinning, Nanotechnology, 1996, 7(3), 216 – 223.

148.Feng, L.; Song, Y.; Zhai, J.; Liu, B.; Xu, J.; Jiang, L.; Zhu, D. Creation of a superhydrophobic surface from an amphiphilic polymer, Angewandte Chemie International Edition, 2003, 42(7), 800 – 802.

149. Marvin, L. F.; Roberts, M. A.; Fay, L. B. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in clinical chemistry, Clinica Chinmica Acta, 2003, 337, 11-21.

150. Chen, H.; He, M. Quantitation of synthetic polymers using an internal standard by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, Journal of the American Society of Mass Spectrometry, 2005, 16, 100-106.

151. Nelson, R.W.; McLean, M. A. Quantitative determination of proteins by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, Analytical Chemistry, 1994, 66(9), 1408-1415.

152. Orlowski, E. A. Synthesis, Characterization, and Surface Functionalization of Polyisobutylene Based Biomaterials. Ph.D. Thesis, University of Akron, August 2009.

153. Puskas, J. E.; Dahman, Y.; Margaritis, A. Novel thymine-functionalized polystyrenes for applications in biotechnology. 2. Adsorption of model proteins, Biomacromolecules, 2004, 5(4), 1412 – 1421.

154. Lienqueo, M. E.; Mahn, A.; Salgado, J. C.; Asenjo, J. A. Current insights on protein behaviour in hydrophobic interaction chromatography, Journal of Chromatography B. 2007, 849, 53 – 68.

155.Jennissen, H. P. Evidence for negative cooperativity in the adsorption of phosphorylase b on hydrophobic agaroses, Biochemistry, 1976, 15(26), 5683 – 5692.

157

156. Uygun, D. A.; Karagözler, A. A.; Akgöl, S.; Denizli, A. Magnetic hydrophobic affinity nanobeads for lysozyme separation, Material Science and Engineering C, 2009, 29(7), 2165 – 2173.

157. Wang, D.; Douma, M.; Swift, B.; Oleschuk, R. D.; Horton, J. H. The adsorption of globular proteins onto a fluorinated PDMS surface, Journal of Colloid and Interface Science, 2009, 331(1), 90 – 97.

158. Clarke, M. L.; Chen, Z. Polymer surface reorientation after protein adsorption, Langmuir, 2006, 22(21), 8627 – 8630.

159.Karas, M.; Hillenkamp, F. Laser desorption ionization of proteins with molecular masses exceeding 10,000 datons, Analytical Chemistry, 1988, 60, 2299-2301.

160.Tanaka, K; Waki, H.; Ido, S.; Akita, Y.; Yoshida, Y.; Yoshida, T. Protein and polymer analyses up to m/z 100 000 by laser ionization time-of-flight mass spectrometry, Rapid Communications in Mass Spectrometry, 1988, 2, 151-153.

161.Fenn, J. B.; Mann, M.; Meng, C. K.; Wong, S. F.; Whitehouse, C. M. Electrospray ionization for mass spectrometry of large biomolecules, Science, 1989, 246, 64-71.

162.Hanton, S. D. Mass Spectrometry of polymers and polymer surfaces, Chemical Reviews, 2001, 101, 527-569.

163.Montaudo, G.; Lattimer, R. P., Eds. Mass Spectrometry of Polymers, CRC Press, Boca Raton, 2002.

164.McEwen, C. N.; Peacock, P. M. Mass spectrometry of chemical polymers, Analytical Chemistry, 2002, 74, 2743-2748.

165.Murgasova, R.; Hercules, D. M. Polymer characterization by combining liquid chromatography with MALDI and ESI mass spectrometry, Analytical and Bioanalytical Chemistry, 2002, 373, 481-489.

158

166.Pasch, H.; Schrepp, W. MALDI-TOF Mass Spectrometry of Synthetic Polymers, Springer, Berlin, 2003.

167.Peacock, P. M.; McEwen, C. N. Mass spectrometry of synthetic polymers, Analytical Chemistry, 2004, 76, 3417-3428.

168.Arnould, M. A.; Polce, M. J.; Quirk, R. P.; Wesdemiotis, C. Probing chain-end functionalization reactions in living anionic polymerization via matrix-assisted laser desorption ionization time-of-flight mass spectrometry, International Journal of Mass Spectrometry. 2004, 238, 245-255.

169.Peacock, P. M.; McEwen, C. N. Mass spectrometry of synthetic polymers, Analytical Chemistry, 2006, 78, 3957-3964.

170.Weidner, S. M.; Trimpin, S. Mass spectrometry of synthetic polymers, Analytical Chemistry, 2008, 80, 4349-4361.

171.Weidner, S. M.; Trimpin, S. Mass spectrometry of synthetic polymers, Analytical Chemistry, 2010, 82, 4811-4829.

172.Gruendling, T.; Weidner, S.; Falkenhagen, J.; Barner-Kowollik, C. Mass spectrometry in polymer chemistry: a state-of-the-art up-date, Polymer Chemistry, 2010, 1, 599-617.

173.Jackson, A. T.; Bunn, A.; Chisholm, S. Polymer, 2008, 392, 575-583.

174.Crecelius, A. C.; Baumgaertel, A.; Schubert, U. S. Tandem mass spectrometry of synthetic polymers, Journal of Mass Spectrometry, 2009, 44, 1277-1286.

175.Von Helden, G.; Hsu, M.-T.; Kemper, P. R.; Bowers, M. T. Structure of carbon cluster ions from 30 to 60 atoms: linears to rings to fellerenes, Journal of Chemical Physics, 1991, 95, 3835-3837.

159

176.Bowers, M. T.; Kemper, P. R. von Helden, G.; Koppen, P. A. M. Gas phase ion chromatography: transition metal state selection and carbon cluster formation, Science, 1993, 260, 1446-1451.

177.Clemmer, D. E.; Jarrold, M. F. Ion mobility measurements and their applications to clusters and biomolecules, Journal of Mass Spectrometry, 1997, 32, 577-592.

178.Verbeck, G. F.; Ruotolo, B. T.; Sawyer, A. A.; Gillig, K. J.; Russell, D. H. A fundamental introduction to ion mobility mass spectrometry applied to the analysis of biomolecules, Journal of Biomolecular Techniques, 2002, 13, 56-61.

179.Trimpin, S.; Plasencia, M.; Isailovic, D.; Clemmer, D. E. Fractionation and solvent-free MALDI-MS analysis of polymers using liquid adsorption chromatography at critical conditions in combination with a multisample on- target homogenization/transfer sample preparation method, Analytical Chemistry, 2007, 79, 7965-7974.

180.Bohrer, B. C.; Merenbloom, S. I.; Koeniger, S. L.; Hilderbrand, A. E.; Clemmer, D. E. Biomolecule analysis by ion mobility mass spectrometry, Annual Review of Analytical Chemistry, 2008, 1, 293-327.

181.Trimpin, S.; Clemmer, D. E. Ion mobility spectrometry/mass spectrometry snapshots for assesing the molecular compositions of complex polymeric system, Analytical Chemistry, 2008, 80, 9073-9083.

182.Anderson, S. E.; Bodzin, D. J.; Haddad, T. S.; Boatz, J. A.; Mabry, J. M.; Mitchell, C.; Bowers, M. T. Structural investigation of encapsulated fluoride in polyhedral oligomeric silsesquioxane cages using ion mobility mass spectrometry and molecular mechanic, Chemistry of Materials, 2008, 20, 4299-4309.

183.Hilton, G. R.; Jackson, A. T.; Thalassinos, K.; Scrivens, J. H. Structural analysis of synthetic polymer mixtures using ion mobility and tandem mass spectrometry, Analytical Chemistry, 2008, 80, 9720-9725.

160

184.Gies, A. P.; Kliman, M.; McLean, J. A.; Hercules, D. M. Characterization of braching in aramid polymers studied by MALDI-ion mobility/mass spectrometry, Macromolecules, 2008, 41, 8299-9301.

185.Chan, Y.-T.; Li, X.; Soler, M.; Wang, J. L.; Wesdemiotis, C.; Newkome, G. R. Self-assembly and traveling wave ion mobility mass spectrometry analysis of hexacadmium macrocycles, Journal of the American Chemical Society, 2009, 131, 16395-16397.

186.Perera, S.; Li, X.; Soler, M.; Schultz, A.; Wesdemiotis, C.; Moorefield, C. N.; Newkome, G. R. Hexameric palladium (II) terpyridyl metallomacrocycles: assembly with 4,4‘-bipyridine and characterization by TWIM mass mass spectrometry, Angewandte Chemie International Edition, 2010, 49, 6539-6544.

187.Brocker, E. R.; Anderson, S. E.; Northrop, B. H.; Stang, P. J; Bowers, M. T. Structures of metallosupramolecular coordination assemblies can be obtained by ion mobility spectrometry-mass spectrometry, Journal of the Am erican Chemical Society, 2010, 132, 13486-13494.

188.Ren, X.; Sun, B.; Tsai, C.-C.; Tu, Y.; Leng, S.; Li, K.; Kang, Z.; Van Horn, R. M.; Li, X.; Zhu, M.; Wesdemiotis, C.; Zhang, W.-B.; Cheng, S. Z. D. Synthesis, self-assembly, and crystal structure of a shape-persistent polyhedral- oligosilsesquioxane-nanoparticle-tethered perylene diimide, The Journal of Physical Chemistry B, 2010, 114, 4802-4810.

189.Li, X.; Chan, Y.-T.; Newkome, G. R.; Wesdemiotis, C. Gradient tandem mass spectrometry interfaced with ion mobility separation for the characterization of supramolecular architectures, Analytical Chemistry, 2011, 83, 1284-1290.

190.Culkin, D. A.; Jeong, W.; Csihony, S.; Gomez, E. D.; Balsara, N. P.; Hedrick, J. L.; Waymouth, R. M. Zwitterionic polymerization of lactide to cyclic poly(lactide) by using N-heterocyclic carbene organo-catalysts, Angewandte Chemie International Edition, 2007, 46, 2627-2630.

191. Jeong, W.; Hedrick, J. L.; Waymouth, R. M. Organic spirocyclic initiators for the ring-expansion polymerization of β-lactones, Journal of the Am erican Chemical Society, 2007, 129, 8414- 8415. 161

192.Jeong, W., Shin, E. J., Culkin, D. A., Hedrick, J. L.; Waymouth, R. M. Zwitterionic polymerization: a kinetic strategy for the controlled synthesis of cyclic polylactide, Journal of the American Chemical Society, 2009, 131, 4884- 4891.

193.Pringle, S. D.; Giles, K.; Wildgoose, J. L.; Williams, J. P.; Slade, S. E.; Thalassinos, K.; Bateman, R. H.; Bowers, M. T. Scrivens, J. H. An investigation of the ion mobility separation of some peptides and protein ions using a new hybrid quadrupole/traveling wave IMS/oa-tof instrument, International Journal of Mass Spectrometry, 2007, 261, 1-12.

194.Shvartsburg, A. A.; Smith, R. D. Fundamentals of traveling wave ion mobility spectrometry, Analytical Chemistry, 2008, 80, 9689-9699.

195.Hilton, G. R.; Thalassinos, K.; Grabenauer, M.; Sanghera, N.; Slade, S. E.; Wyttenbach, T.; Robinson, P. J.; Pinheiro, T. J. T.; Bowers, M. T.; Scrivens, J. H. Structural analysis of prion proteins by means of drift cell and traveling wave ion mobility mass spectrometry, Journal of the Am erican Society for Mass Spectrometry, 2010, 21, 845-854.

196.Suckau, D.; Resemann, A.; Schuerenberg, M.; Hufnagel, P.; Franzen, J.; Holle, A. A novel MALDI lift-tof/tof mass spectrometer for proteomics, Analytical and Bioanalytical Chemistry, 2003, 376, 952-965.

197.Knochenmuss, R. Chapter 5, pp 149-183, In Electrospray and MALDI Mass Spectrometry: Fundamentals, Instrumentation, Practicalities, and Biological Applications, 2nd ed., Cole, R. B., Ed.; John Wiley & Sons, Hoboken, New Jersey, 2010.

198.Hossain, M.; Limbach, P. A. Chapter 7, pp 215-244, In Electrospray and MALDI Mass Spectrometry: Fundamentals, Instrumentation, Practicalities, and Biological Applications, 2nd ed., Cole, R. B., Ed., John Wiley & Sons, Hoboken, New Jersey, 2010.

162

199.Puglisi, C.; Samperi, F.; Di Giorgi, S.; Montaudo, G. Exchange reactions occurring through active chain ends. MALDI-tof characterization of copolymers from nylon 6,6 and nylon 6,10, Macromolecules, 2003, 36, 1098-1107.

200.Smith, M. B.; March, J. p 427, March’s Advanced Organic Chemistry: Reactions, Mechanisms, and Structure, 5th ed., John Wiley & Sons, New York, 2001.

201.Iavarone, A. T.; Jurchen, J. C.; Williams, E. R. Journal of the American Society for Mass Spectrometry, 2001, 11, 976-985.

202.Ruotolo, B. T.; Tate, C. C.; Russell, D. H. Ion mobility-mass spectrometry applied to cyclic peptide analysis: conformational preferences of gramicidin s and linear analogs in the gas phase, Journal of the American Society for Mass Spectrometry, 2004, 15, 870-878.

203.Riba-Garcia, I.; Giles, K.; Bateman, R. H.; Gaskell, S. J. Evidence for structural variants of a- and b-type peptide fragment ions using combined ion mobility/mass spectrometry, Journal of the American Society for Mass Spectrometry, 2008, 19, 609-613.

163

APPENDICES

164

APPENDIX A

ADDITIONAL DATA

165

Table A.1. Insulin, lysozyme, and ubiquitin peak area ratios (relative to cytochrome C peak area) and their adsorbed amounts per unit area surface.

Area Ratio ptoteina/cytichrome C Average Protein Protein Material Protein pH Run # (Average Area Ratio for Each Area Ratio of pmol/µl pmol/cm2* single Run) Two Runs 1 1.602 4.4 1.595 3.525 352.454 2 1.587 1 1.884 5.3 1.769 3.909 390.915 2 1.653 Insulin 1 0.994 6.9 1.005 2.220 222.038 2 1.015 1 0.347 8.7 0.354 0.782 78.249 2 0.361 1 0.281 0.268 0.393 39.263 4.4 2 0.254 1 0.271 D_IBS fiber mat 5.3 0.262 0.384 40.800 2 0.252 Ubiquitin 1 0.092 6.9 0.090 0.132 13.210 2 0.088 1 0.045 8.7 0.042 0.061 6.091 2 0.038 1 0.015 4.4 0.016 0.011 1.069 2 0.017 1 0.024 Lysozyme 5.3 0.022 0.015 1.470 2 0.020 1 0.021 6.9 0.025 0.017 1.670 2 0.029 1 0.048 8.7 0.051 0.034 3.373 2 0.053 1 0.650 4.4 0.657 1.451 145.115 2 0.663 1 0.691 5.3 0.674 1.527 152.741 2 0.656 Insulin 1 0.456 6.9 0.460 1.017 101.680 2 0.464 1 0.216 8.7 0.220 0.485 48.519 2 0.223 1 0.067 4.4 0.070 0.102 10.201 2 0.072 1 0.089 5.3 0.093 0.131 13.063 D_IBS films 2 0.097 Ubiquitin 1 0.053 6.9 0.057 0.084 8.366 2 0.061 1 0.037 8.7 0.034 0.050 4.990 2 0.031 1 0.0019 4.4 0.002 0.001 0.140 2 0.0023 1 0.001 Lysozyme 5.3 0.001 0.001 0.150 2 0.001 1 0.0098 6.9 0.010 0.007 0.675 2 0.0104 1 0.013 8.7 0.014 0.009 0.935 2 0.015

166

(Continuation from page 166) Table A.1. Insulin, lysozyme, and ubiquitin peak area ratios (relative to cytochrome C peak area) and their adsorbed amounts per unit area surface.

Area Ratio ptoteina/cytichrome C Average Protein Protein Material Protein pH Run # (Average Area Ratio for Each Area Ratio of pmol/µl pmol/cm2* single Run) Two Runs 1 1.182 4.4 1.156 2.555 255.526 2 1.130 1 0.895 Insulin 6.9 0.899 1.987 198.718 2 0.903 1 0.339 8.7 0.327 0.723 72.281 2 0.315 1 0.134 4.4 0.131 0.192 19.228 2 0.128 Polystyrene fiber mat 1 0.081 Ubiquitin 6.9 0.087 0.127 12.696 2 0.092 1 0.028 8.7 0.030 0.044 4.403 2 0.032 1 0.0041 4.4 0.004 0.003 0.294 2 0.0047 Lysozyme 1 0.018 6.9 0.021 0.014 1.369 2 0.023 1 0.038 8.7 0.042 0.028 2.772 2 0.045 1 0.525 4.4 0.529 1.168 116.821 2 0.532 1 0.400 Insulin 6.9 0.395 0.873 87.312 2 0.390 1 0.198 8.7 0.193 0.426 42.551 2 0.187 1 0.061 4.4 0.065 0.095 9.467 2 0.068 Polystyrene films 1 0.050 Ubiquitin 6.9 0.047 0.068 6.825 2 0.043 1 0.023 8.7 0.024 0.035 3.523 2 0.025 1 n/a 4.4 n/a n/a n/a 2 n/a Lysozyme 1 0.0085 6.9 0.008 0.005 0.528 2 0.0073 1 0.009 8.7 0.009 0.006 0.568 2 0.008

2 protein */][ 200 llpmol * pmol / cm  2cm 2

Standard deviation formula used on Tables 6.2 and 6.3.

(x  )x 2 SD = (n 1)

167

Insulin Ubiquitin Lysozyme Concentration Concentration Concentration Area ratio Area ratio Area ratio (pmol/μl) (pmol/μl) (pmol/μl)

0.005 0.022 0.005 0.014 0.002 0.015

0.010 0.038 0.010 0.019 0.010 0.013

0.050 0.131 0.050 0.065 0.050 0.015

0.100 0.173 0.100 0.163 0.100 0.039

0.500 0.374 0.500 0.425 0.500 0.313

1.000 0.409 1.000 0.958 1.000 1.352

2.000 0.935 2.000 1.334 2.000 3.016

4.000 1.782 4.000 2.657 4.000 6.071

Table A.2. Mass spectral data for the construction of calibration curves in Chapter 6 (see Figure 6.2)

168

911.3 677.2 821.3 615.2 799.3 503.2 687.3 575.2 639.2 544.3 709.2 727.3 391.1 592.3 431.1 463.2 565.2 603.4 749.3 759.3 775.2 781.3

350 450 500 550 600 650 700 750 800 850 900 m/z 400

. + Figure A.1. ESI-CAD mass spectrum of the PL9NaOH (NaOH)5Na cluster ion, acquired with the Waters Q/ToF tandem mass spectrometer. The precursor ion was subjected to CAD with Argon at a collision energy of 45 eV.

169

337.1512

261.1540

438.2820 489.4104 553.5760 233.1240

554.5805 207.0659 631.0840 88.0082 696.0989 726.0845

m/z 100 300 500 700 900 1100 1300 1500 1700 1900

Figure A.2. ESI-Q/ToF mass spectrum of the PLA degradation medium after10 days of incubation in a Proteinase K-free environment. No polylactide distribution is detected

170

APPENDIX B

JOHN WILEY AND SONS LICENSE TERMS AND CONDITIONS Nov 01, 2011

This is a License Agreement between Madalis Casiano-Maldonado ("You") and John Wiley and Sons ("John Wiley and Sons") provided by Copyright Clearance Center ("CCC"). The license consists of your order details, the terms and conditions provided by John Wiley and Sons, and the payment terms and conditions.

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

License Number 2777110417876 License date Oct 27, 2011 Licensed content publisher John Wiley and Sons Licensed content Journal of Mass Spectrometry publication Licensed content title Electrospray: Principles and Practice Licensed content author Simon J. Gaskell Licensed content date Jul 1, 1997 Start page 677 End page 688 Type of use Dissertation/Thesis

Requestor type University/Academic Format Print and electronic Portion Figure/table Number of figures/tables 1 Number of extracts Original Wiley figure/table Figure 2. Droplet production in the electrospray interface number(s) Will you be translating? No

171

JOHN WILEY AND SONS LICENSE TERMS AND CONDITIONS Nov 01, 2011

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

License Number 2780231164029 License date Nov 01, 2011 Licensed content publisher John Wiley and Sons Licensed content Wiley Books publication Licensed content title Mass Spectrometry: Principles and Applications, 3rd Edition Licensed content author Edmond de Hoffmann, Vincent Stroobant Licensed content date Nov 1, 2007 Type of use Dissertation/Thesis Requestor type University/Academic Format Print and electronic Portion Figure/table Number of figures/tables 7 Original Wiley figure/table Figure 2.4 Quadrupole instrument made up of the source, the number(s) focusing lenses, the quadrupole cylindrical rods and the detector. Figure 2.12 Top: a complete Paul ion trap mass spectrometer (Finnigan MAT). Bottom: detail of the trap. Right: figures obtained when injecting a fine dust of aluminum in a quadrupolar field such as is encountered in an ion trap. Figure 2.38 Principles of an LTOF instrument tuned to analyse positive ions produced by MALDI. Figure 2.39 Schematic description of a continuous extraction mode and a delayed pulse extraction mode in a linear time-of-flight mass analyzer. Figure 2.40 Schematic description of a TOF instrument equipped with a reflectron. Figure 3.4 Cross-section of a microchannel plate and electron multiplication within a channel. Figure 3.7 Schematic diagram of Daly detector. Will you be translating? No Order reference number Total 0.00 USD

172

JOHN WILEY AND SONS LICENSE TERMS AND CONDITIONS Nov 01, 2011

All payments must be made in full to CCC. For payment instructions, please see information listed at the bottom of this form.

License Number 2780521221205 License date Nov 01, 2011 Licensed content publisher John Wiley and Sons Licensed content Journal of Mass Spectrometry publication Licensed content title An Introduction to Quadrupole Ion Trap Mass Spectrometry Licensed content author Raymond E. March Licensed content date Apr 1, 1997 Start page 351 End page 369 Type of use Dissertation/Thesis

Requestor type University/Academic Format Print and electronic Portion Figure/table Number of figures/tables 1 Number of extracts Original Wiley figure/table Figure 7. Stability diagram in space for the region of simultaneous number(s) stability in both the r- and z-directions near the origin for the three- dimensional quadrupole ion trap. Will you be translating? No Order reference number Total 0.00 USD

173

174

175

176

177

178