RING-OPENING APPROACHES TO FUNCTIONAL RENEWABLE POLYMERS
Geng Hua
Doctoral Thesis, 2018 KTH Royal Institute of Technology Department of Fibre and Polymer Technology Polymer Technology SE-100 44 Stockholm, Sweden
Copyright © 2018 Geng Hua All rights reserved
Paper II © 2018 American Chemical Society Paper III © 2016 American Chemical Society Paper IV © 2017 Wiley-VCH
TRITA-CBH-FOU-2018:19 ISSN 1654-1081 ISBN 978-91-7729-781-9
Akademisk avhandling som med tillstånd av KTH i Stockholm framlägges till offentlig granskning för avläggande av teknisk doktorsexamen tisdagen den 5 juni 2018, kl. 10:00 i sal F3, KTH, Lindstedtsvägen 26, Stockholm. Avhandlingen försvaras på engelska. Fakultetsopponent: Professor Sophie M. Guillaume (Université de Rennes 1, FR)
To my dearest family
斧斤以时入山林，材木不可胜用也。
《孟子·梁惠王上》 约公元前三四〇年
Harvesting trees in a sustainable way, timbers will never run out.
Meng Zi · Liang Hui Wang Shang c.a. 340 B.C.
ABSTRACT
The design of renewable substitutes to today’s polymers is a necessary technical solution to the growing environmental concerns connected to polymeric materials. This thesis offers a vision of how polymeric materials could be designed to favor environmentally benign solutions in the different phases of a polymer’s lifespan.
The choice of monomer is the initial phase in the lifespan of a polymer, and the monomers used here were either based on renewable resources or could be created by chemical recycling, i.e., biobased lactones or chemically recycled cyclic carbonates. The reaction conditions during polymerization, as the second phase of a polymer’s lifespan, were carefully optimized to be as sustainable as possible. This optimization included using safer chemicals, reducing the solvent consumption and lowering the reaction temperature to limit energy consumption. Atom-economic ring-opening reactions, both ring-opening polymerization and ring-opening aminolysis, were applied to the cyclic monomers to form polyamides and polycarbonates with tunable linear structures and polyester and polyurethane hydrogels. Applications of these polymers were exemplified by fiber- and film- forming biobased polyamides to potentially replace certain current petroleum-based products; utilizing polyester hydrogel for heavy metal ion wastewater treatment with high adsorption capacity; and polyurethane hydrogels loaded with quaternary ammonium compounds and tertiary sulfonium compounds for antibacterial applications with inhibition ratios over 99% against both Staphylococcus aureus and Escherichia coli. Reconverting the polymer back to the monomer is the last phase of the lifespan and closes the lifecycle loop. This process was realized via chemical recycling of the polycarbonate through ring-closing depolymerization, which resulted in cyclic carbonate monomeric building blocks. One of the cyclic carbonates (1-MeTMC) was polymerized again using an organic base catalyst to form a regioregular polycarbonate with Xreg~0.90.
The choice of the renewable monomers, the versatility of the facile ring-opening reactions and the optimized reaction conditions together realized the construction of structurally diverse renewable polymers. These polymers are well suited to be used in more sustainable material designs, which further facilitate a “greener” future.
Keywords: cyclic carbonate, ring-opening polymerization, ring-opening aminolysis, biobased, lactone, polycarbonate, polyester, polyamide, polyurethane, hydrogel, adsorbent, antibacterial
SAMMANFATTNING
Att designa förnybara alternativ till dagens plaster är en nödvändig teknisk lösning på de ökande utmaningar polymera material ställer på miljön. Denna avhandling erbjuder en vision över hur polymera material kan skapas på sätt som minskar miljöbelastningen, med fokus på de olika delarna av en polymers livscykel.
Valet av monomer, representerar det initiala skedet av en polymers livscykel. De monomerer som har valts i denna studie har ursprung i en förnybar källa eller kan skapas genom återvinning av polymert avfall, t ex biobaserade laktoner eller kemiskt återvunna cykliska karbonater. Reaktionsbetingelserna under polymerisationen, den andra delen av en polymers livscykel, optimerades noggrant för att vara så hållbara som möjligt. Detta inkluderar att använda icketoxiska kemikalier, minska användningen av lösningsmedel och sänka reaktions- temperaturen för att minska energiåtgången. Atom-ekonomiska ringöppnings- reaktioner, både via ringöppningspolymerisation och ringöppningsaminolys, applicerades på de cykliska monomererna för att bilda en rad olika polymertyper, vilka inkluderade linjära polyamider och polykarbonater och nätverk av polyestrar och polyuretaner i form av hydrogeler. Användningsområden för materialen exemplifierades genom att skapa biobaserade polyamidfibrer och filmer som i en framtid skulle kunna ersätta våra oljebaserade produkter, polyesterbaserade hydrogeler för att avlägsna tungmetall från avloppsvatten, liksom polyuretan- baserade hydrogeler med många positiva laddade grupper, vilket ger antibakteriella egenskaper. För att undersöka ett polymert materials sista led i livscykeln återvinningen, användes kemisk återvinning genom ringslutningsdepolymerisation av polykarbonater för att skapa cykliska karbonater, och därigenom nya monomerer. En av dessa cykliska karbonatmonomerer, med ett steriskt center, polymeriserades därefter kontrollerat till en regioregulär polykarbonat.
Valet av förnybara monomerer, mångsidigheten av dessa okomplicerade ring- öppningsreaktioner och de optimerade reaktionsbetingelserna möjliggjorde bildandet av en grupp förnybara och återvinningsbara polymerer med stor strukturell variation. Dessa polymerer kan anses passa väl in i strävan efter mer hållbara materialval genom hela materialens livscykel, och härigenom bidra till tekniska lösningar för en mer hållbar framtid.
Nyckelord: cykliskt karbonat, ringöppningspolymerisation, ringöppningsaminolys, biobaserad, lakton, polykarbonat, polyester, polyamid, polyuretan, hydrogel, adsorbent, antibakteriell
LIST OF APPENDED PAPERS
This thesis is a summary of the following papers:
Paper (I) Hua, G.; Franzén, J.; Odelius, K. Phosphazene-catalyzed Regioselective Ring- opening Polymerization of rac-1-Methyl Trimethylene Carbonate: Colder and Less Catalyst is Better, Manuscript
Paper (II) Hua, G.; Odelius, K. Exploiting Ring-Opening Aminolysis–Condensation as a Polymerization Pathway to Structurally Diverse Biobased Polyamides. Biomacromolecules 2018, acs.biomac.8b00322.
Paper (III) Hua, G.; Odelius, K. From Food Additive to High-Performance Heavy Metal Adsorbent: A Versatile and Well-Tuned Design. ACS Sustain. Chem. Eng. 2016, 4 (9), 4831–4841.
Paper (IV) Hua, G.; Odelius, K. Isocyanate-Free, UV-Crosslinked Poly(Hydroxyurethane) Networks: A Sustainable Approach toward Highly Functional Antibacterial Gels. Macromol. Biosci. 2017, 17 (11), 1700190.
Author contributions to the appended papers:
Paper (I) All of the experimental work and data evaluation; writing a major part of the manuscript.
Paper (II) All of the experimental work and data evaluation; writing a major part of the manuscript.
Paper (III) All of the experimental work and data evaluation; writing a major part of the manuscript.
Paper (IV) All of the experimental work and data evaluation; writing a major part of the manuscript.
Scientific contributions not included in the thesis:
Hua, G.; Franzén, J.; Odelius, K. One-Pot Inimer Promoted ROCP Synthesis of Branched Copolyesters Using α-Hydroxy-γ-Butyrolactone as the Branching Reagent. J. Polym. Sci. Part A Polym. Chem. 2016, 54 (13), 1908–1918.
Xu, H.; Hua, G.; Odelius, K.; Hakkarainen, M. Stereocontrolled Entanglement- Directed Self-Alignment of Poly(lactic Acid) Cylindrites. Macromol. Chem. Phys. 2016, 217 (23), 2567–2575.
ABBREVIATIONS
AOMEC 2-Allyloxymethyl-2-ethyl-trimethylene carbonate BnOH Benzyl alcohol CFU Colony-forming unit
Ð Dispersity (Mw/Mn) DBU 1,8-Diazabicyclo[5.4.0]undec-7-ene DI water Deionized water diamine-B 1,3-Bis(aminomethyl)benzene diamine-C 1,3-Byclohexanebis(methyl-amine) diamine-E 2,2’-(Ethylenedioxy)-bis(ethylamine) E. coli Escherichia coli EB Ethylene brassylate N,N′-((Ethane-1,2-diylbis(oxy))bis(ethane-2,1-diyl))- EBB bis(2,3,4,5,6-pentahydroxyhexanamide) EDS Energy-dispersive X-ray spectroscopy EDTA Ethylenediaminetetraacetic acid EDTAD Ethylenediaminetetraacetic dianhydride
EG-(SH)2 2,2′-(Ethylenedioxy) diethanethiol FTIR Fourier transform infrared spectroscopy FT-Raman Fourier transform Raman spectroscopy GDL δ-Gluconolactone H-bonding Hydrogen bonding HFIP 1,1,1,3,3,3-Hexafluoro-2-propanol HH Head-to-head HT Head-to-tail HUM Hydroxyl urethane macromer MALDI-ToF Matrix-assisted laser desorption/ionization time-of-flight mass MS spectrometry MeCN Acetonitrile MeI Iodomethane
Mn Number-average molecular weight
Mw Weight-average molecular weight
Mη Viscosity-average molecular weight NMR Nuclear magnetic resonance
OD625nm Optic density at 625 nm wavelength PA Polyamide N,N′-(1,3-Phenylenebis(methylene))bis- PBP (2,3,4,5,6-pentahydroxyhexanamide) PhMe Toluene PHU Poly(hydroxyl urethane) QAC Quaternary ammonium compound qm Maximum adsorption capacity R.T. Room temperature rac-1-MeTMC rac-4-Methyl-1,3-dioxan-2-one RCD Ring-closing depolymerization ROA Ring-opening aminolysis ROAC Ring-opening aminolysis-condensation ROP Ring-opening polymerization RTIR Real-time infrared spectroscopy S. aureus Staphylococcus aureus SEC Size-exclusion chromatography SEM Scanning electron microscope TAEA Tris(2-aminoethyl)amine TBD 1,5,7-Triazabicyclo[4.4.0]dec-5-ene 1-Tert-butyl-2,2,4,4,4-pentakis-(dimethylamino)-2λ5,4λ5- t-BuP 2 catenadi(phosphazene)
Tc Ceiling temperature TEA Triethylamine TGA Thermogravimetric analysis THF Tetrahydrofuran TMC Trimethylene carbonate TSC Tertiary sulfonium compound TT Tail-to-tail UV Ultraviolet
Xn Number-average degree of polymerization
Xreg Regioregularity
SAMPLE NAMING SUMMARY
PA-Y Polyamide prepared with EB and the corresponding diamine-Y PA-DS Polyamide prepared with DS and diamine-C B-type gel Polyester hydrogel prepared with PBP and EDTAD B-type gel prepared with PBP (100 eq -OH) and EDTAD (70 eq 70B anhydride) B-type gel prepared with PBP (100 eq -OH) and EDTAD (100 eq 100B anhydride) B-type gel prepared with PBP (100 eq -OH) and EDTAD (140 eq 140B anhydride) E-type gel Polyester hydrogel prepared with EBB and EDTAD E-type gel prepared with EBB (100 eq -OH) and EDTAD (70 eq 70E anhydride) E-type gel prepared with EBB (100 eq -OH) and EDTAD (100 eq 100E anhydride) E-type gel prepared with EBB (100 eq -OH) and EDTAD (140 eq 140E anhydride) Unmodified polyurethane hydrogel prepared with stoichiometric PHU-neat HUM and EG-(SH)2 PHU-5% PHU-neat treated with 5% (v:v) of MeI/THF solution PHU-20% PHU-neat treated with 20% (v:v) of MeI/THF solution
TABLE OF CONTENTS
1 PURPOSE OF THE STUDY ...... 1 2 INTRODUCTION ...... 2 2.1 POLYMERIC MATERIALS IN SUSTAINABLE DEVELOPMENT ...... 2 2.2 CYCLIC MONOMERS AND RING-OPENING REACTIONS ...... 3 2.2.1 Renewable and recyclable monomers ...... 3 2.2.2 Ring-opening polymerization ...... 3 2.2.3 Ring-opening aminolysis ...... 5 2.3 MICROSTRUCTURE AND MACROMOLECULAR STRUCTURE ...... 6 2.3.1 Regioregular polycarbonate ...... 7 2.3.2 Densely functionalized hydrogels ...... 8 3 EXPERIMENTAL ...... 9 3.1 CHEMICALS ...... 9 3.2 SYNTHESIS PROCEDURES ...... 10 3.2.1 Synthesis of cyclic carbonates through depolymerization ...... 10 3.2.2 General ROP procedure for rac-1-MeTMC ...... 11 3.2.3 Synthesis of PA through ROAC ...... 11 3.2.4 Synthesis of gelation precursors ...... 11 3.2.5 Gelation through esterification ...... 12 3.2.6 Gelation through thiol-ene UV radical addition ...... 13 3.2.7 Modification of gels ...... 13 3.3 CHARACTERIZATION METHODS ...... 14 3.3.1 Nuclear magnetic resonance (NMR) ...... 14 3.3.2 Fourier transform infrared spectroscopy (FTIR) ...... 14 3.3.3 Real-time infrared spectroscopy (RTIR) ...... 14 3.3.4 Fourier transform Raman spectroscopy (FT-Raman) ...... 14 3.3.5 Scanning electron microscope (SEM) ...... 14 3.3.6 Energy-dispersive X-ray spectroscopy (EDS) ...... 14 3.3.7 Size-exclusion chromatography (SEC) ...... 15 3.3.8 Thermogravimetric analysis (TGA) ...... 15
3.3.9 Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) ...... 15 3.3.10 Solution viscometry ...... 15 3.3.11 pH measurement ...... 16 3.3.12 Complexometric titration ...... 16 3.3.13 Optic density (OD) ...... 16 3.3.14 Antibacterial assay ...... 16 4 RESULTS AND DISCUSSION ...... 17 4.1 REGIOREGULAR RING-OPENING POLYMERIZATION ...... 17 4.2 RING-OPENING AMINOLYSIS ...... 23 4.2.1 Ring-opening aminolysis (ROA) ...... 23 4.2.2 Ring-opening aminolysis condensation (ROAC) ...... 24 4.3 DENSELY FUNCTIONALIZED HYDROGELS ...... 30 4.3.1 Polyester hydrogels as heavy metal ion adsorbent ...... 31 4.3.2 Antibacterial polyurethane hydrogels ...... 36 5 CONCLUSIONS ...... 45 6 FUTURE WORK ...... 46 7 ACKNOWLEDGEMENTS ...... 47 8 REFERENCES ...... 48 9 APPENDIX ...... 58
PURPOSE OF THE STUDY
1 PURPOSE OF THE STUDY
With the pursuit of building a future sustainable society, polymers as a ubiquitous part of our daily life are at the center of the discussions. Although they provide convenience in many aspects of life, the concerns connected to their petroleum- based origin, often poor recyclability and accumulation in the natural environment are intensifying.
The purpose of this study is to create a wide range of polymeric materials using renewable cyclic monomers as the staring material, with the focus on optimizing and expanding the current polymer design and synthesis in favor of sustainability. Biobased lactones are chosen as the building blocks for the polymers, due to their economic feasibility, renewability, and mass-production in the chemical, food and flavoring industry. Ring-opening reactions are explored to make polymers with varied macromolecular structures, from linear chains to networks. The optimization of the synthetic pathways of ring-opening reactions are targeted under the principles of green chemistry, which includes solvent and waste prevention, use of catalysts, atom economy, less hazardous chemical syntheses and energy efficient design. The synthesized polymer should have the potential of replacing some of the current petroleum-based products, and the intended application should cover the concerns and improvements over environmental well-being.
Chemical recycling of polymers back to cyclic monomers is an attractive way of handling used and end-of-life products. Attempts are made on oligocarbonates and polycarbonates, under-going chemical recycling (ring-closing depolymerization) to form cyclic carbonates. These cyclic carbonates are produced in large-scale and are used as building blocks for polycarbonates and polyurethanes. The full material value is hence retained in this circular pathway, facilitating sustainable development.
1 INTRODUCTION
2 INTRODUCTION
2.1 POLYMERIC MATERIALS IN SUSTAINABLE DEVELOPMENT Polymeric materials play an integral role in the world’s economy and our daily life. Some polymers are of low weight and high strength and are used as automobile and aircraft parts, reducing fuel consumption; some have good barrier properties, so they are used as food packaging to enable a longer shelf life; and some show good insulation properties, which reduces energy consumption in households. As they generally do not degrade and are often used in everyday products with a short usage time, their environmental impact is long lasting and ever growing.
The concept of the Circular Economy provides the context for sustainable management throughout the life span of polymers.1 The principles are to reduce, reuse and recycle. “Reduce” can take place in various forms, such as only using disposable polymeric products when necessary; reducing the use of petroleum- based polymers and using biobased polymers instead; creating new functional polymers to reduce the existing pollutants in the environment; and reduce the energy consumption during commodity polymer production. “Reuse” and “recycle" are more focused on the end-of-life handling of polymers; the polymer products should be designed to be recyclable, and the recycling efforts should not demand more than the effort to produce new products. This requires structural optimization of the existing polymers, development of new polymers and the advancement in recycling technologies.
The 17 Sustainable Development Goals adopted in September 2015 depicted a future in 2030 where poverty is eradicated, the planet is protected and everyone lives in prosperity.2 It is believed that innovative ways of creating and recycling sustainable polymers will assist in reaching many of these goals.
2 INTRODUCTION
2.2 CYCLIC MONOMERS AND RING-OPENING REACTIONS
2.2.1 Renewable and recyclable monomers The definition and scope of renewable monomers and polymers often varies among scientists. One common understanding is that if a monomer is naturally occurring or derived from plant-based products, then the monomer can be considered as renewable.3–5 Lactones are cyclic esters that are well represented within this definition, with a very broad family of commercially available chemicals. Some lactones are naturally occurring, such as δ-gluconolactone and L-galactono-1,4- lactone, while others are derived from mono- and polysaccharides and plant seed oil, (e.g., α-hydroxyl-γ-butyrolactone, ethylene brassylate, ω-pentadecalactone and ω-6-hexadecenlactone). A number of the lactones can be used directly to make polymers, such as polyesters and poly(ester-amide)s,6–12 while others need some modification before polymerization.13–16 The uses of these renewable lactones are already hugely successful in the food and fragrance industry as flavoring and scent additives.17–19 Although these monomers are highly attractive, the synthesis of renewable polymers using these monomers is still underexploited.
The end-of-life management of products is another important aspect to consider.4,20– 22 Cyclic monomers can be synthesized from raw materials, which undergo ring- opening polymerization into a polymer product. In the reverse perspective, this polymer product can itself be a raw material to be converted back to monomers or to other useful building blocks.23–26 This closed-loop methodology is termed chemical recycling or ring-closing depolymerization in some studies, and it is theoretically applicable to all polymers that are depolymerizable,27–30 such as poly(tetrahydrofuran),31 poly(α-methylstyrene)32 and poly(propylene carbonate).17 Aliphatic polycarbonates are a large group of degradable polymers with a variety of structures, among which some have the ability to be chemically recycled under highly controlled conditions.33–36 From a sustainability standpoint, this reversed reaction from polymer to monomer is of great interest, as it closes the life cycle of the product, retaining the full material value, thus giving the resulting cyclic carbonate a sustainable label.
2.2.2 Ring-opening polymerization Ring-opening polymerization (ROP) describes a type of polymerization where the ring-opening of a cyclic monomer takes place from an initiator or a growing chain- end, and the ring-opened monomer subsequently adds to the growing chain-end. For
3 INTRODUCTION
ROP of lactones and cyclic carbonates, a broad range of catalysts from metal complexes to nonmetal organics are available.37–40 The use of a nonmetal organic base as the ROP catalyst is welcomed by microelectronic and biomedical applications where the existence of heavy metals can be a hazard.41,42 These nonmetal organic base catalysts are usually soluble in a wide range of solvents and sometimes in the monomer, meaning that it is easier to obtain a homogeneous catalytic system. They are also very reactive at room temperature, which means that no heating is needed to reach high catalytic activity compared to many metal catalysts, hence promoting the energy efficiency during synthesis. Some of the most frequently reported systems are phosphazenes,43 N-heterocyclic carbenes (NHC),44 and amidines and guanidines45 in combination with the model monomers such as ε- caprolactone,10,46–48 lactide,49–52 and trimethylene carbonate.52–54 These ROP systems generally show moderate to good control over both the molecular weight distribution and chain-end fidelity.38
Thermodynamic feasibility is a major consideration for ROP. Based on the Gibbs free energy equation, ΔGp=ΔHp-TΔSp, ROP takes place when ΔGp is negative. Both ΔHp and ΔSp are influenced by the inherent structure of the cyclic monomer and the applied reaction conditions, which together determine if a cyclic monomer can 55 undergo ROP in a set system. The influence on ΔHp and ΔSp from the inherent structure of the monomer includes the size of the ring (5-, 6-, or 7-membered), the substituents on the ring (methyl- or benzyl-substituted) and the linkages that form the ring (ether, ester or carbonate). The reaction conditions such as the solvation and monomer concentration alter the ΔHp and ΔSp values; and depending on the choice of solvent used, very different ROP behaviors can be observed for the same monomer. The reaction temperature by itself can greatly influence the polymerization behavior, as the ΔGp value can be altered from negative to positive just by changing the temperature alone.55 This in turn offers a facile method of making advanced macromolecular structures in situ merely by controlling the reaction temperature. Some of the recent advances in this area include making multiple block copolymers via a temperature switch,56 producing stereoregular block polylactide from rac-lactide at low temperature,57 and creating macrocyclic poly(γ-butyrolactone) at low temperature with enhanced thermal stability.58
Cyclic carbonates can be polymerized using a variety of nonmetal organic base catalysts, such as TBD,59 DBU,60 and phosphazene base.61 The large variety of polycarbonate homo- and copolymers have been reported over the years showing the versatility of the ROP method.62 When using a nonmetal organic base catalyst, an alcohol is usually required to act as the initiator. Depending on the catalyst, the activation mechanism can be initiator/chain-end activation or nucleophilic
4 INTRODUCTION monomer activation,63 Scheme 1.64 In the initiator/chain end mechanism, the alcohol initiator first forms an H-bond with the base catalyst, enhancing the nucleophilicity of the oxygen atom. A nucleophilic attack from the oxygen atom of the initiator on the acyl carbon of the cyclic monomer then occurs, forming the corresponding ring-opened carbonate with the release of the ring-strain and generation of an alcohol chain-end. This new chain-end is now H-bonded with the catalyst and acts as the activation site for the enchainment of the next monomer. In the nucleophilic monomer activation mechanism, the nucleophilic attack on the acyl carbon from the base catalyst first appears, forming an alkoxide intermediate. The proton exchange between intermediate and the alcohol initiator then occurs, and the resulting alkoxide from the initiator undergoes an acyl transfer reaction, forming carbonate with the release of the catalyst.65 In the case of the base-catalyzed ROP of cyclic carbonate to form linear polycarbonates, the activation mechanism often follows the initiator/chain-end activation mechanism.38,64
Scheme 1. a) Chain-end activation mechanism, and b) nucleophilic monomer activation mechanism of ROP for 6-membered cyclic carbonates.
2.2.3 Ring-opening aminolysis Aminolysis of esters and carbonates is a nucleophilic addition-elimination reaction at the acyl carbon, which forms the corresponding amide and urethane bond. Ring- opening aminolysis (ROA) is a specific case of aminolysis where the starting materials are cyclic monomers, such as lactones or cyclic carbonates, Scheme 2. Compared to their linear analogs, the inherent ring strain of the cyclic monomers can provide an additional driving force for the reactions to reach higher conversions.66 One of the major foci of ROA is to study the reactivity of different amines under different strengths of base catalysis and their temperature dependence.67–70 Some reported studies have systemically tuned these three factors for different cyclic monomers, including 5- and 6-membered lactones,69,71,72 thiol- lactones,73,74 and 5- and 6-membered cyclic carbonates.75 The general conclusion of these studies is that these three factors positively influence the ROA reactions, i.e.,
5 INTRODUCTION increasing the basicity of the amine compound, the strength of the base catalyst and the temperature will increase the conversion and the reaction rate. Alongside these three factors, the concentration of the reactants, the use and type of solvents and the solubility/miscibility of the reactants also influence the reaction kinetics. The reaction conditions can be pushed to favor the kinetics and the conversion; however, this has a limit. For instance, too high a temperature in the ROA of cyclic carbonates will cause the formation of by-products that have urea bonds other than urethane bonds.76
The ROA of lactones provides an alternative route for the application of some less studied renewable lactones. Many of these lactones are either 5-membered with long alkyl substituents or are macrocyclics, both of which are known to be only slightly ring-strained, hence challenging for ROP.55 The ROA route circumvents this challenge through the formation of an amide bond; and instead of forming a polymer directly, it transforms the lactone into amide-containing small molecules. These amide-containing small molecules have found a use as drug intermediates and functionalized monomers.67,77 The attractiveness of ROA for cyclic carbonates is the formation of the urethane-containing building blocks or polyurethanes while avoiding the hazardous isocyanates that are used in the conventional routes.69, 70 The corresponding products termed isocyanate-free polyurethanes or non-isocyanate polyurethanes.
Scheme 2. ROA reactions between lactones, cyclic carbonates and amine, forming products with an amide bond and a urethane bond, respectively.
2.3 MICROSTRUCTURE AND MACROMOLECULAR STRUCTURE The structure of a polymer directly determines its physical and chemical properties, as well as the end use of the material. The macromolecular structure of a polymer is often used to describe it and includes linear chains, branched structures or networks. When further focusing within each macromolecular structure to the scale of the repeating unit, it becomes the study of microstructures, which consists of more structural considerations. For linear polymers, the microstructure considerations include, for instance, the repeating unit’s chemical structure, the length of the
6 INTRODUCTION repeating units, side chain structures, and regio- and stereo-regularity. For networks, the considerations are, for instance, crosslinking density, structures of the building blocks, linkages, and the amount of available functional groups. The changing and tuning of the microstructures greatly influences the material properties. Hence, functional polymers can be tailor-made to meet the performance demand.
2.3.1 Regioregular polycarbonate Due to the possible substituents on the ring structure of a cyclic monomer, stereocenters can sometimes occur. This is exemplified by monomers such as ε- decalactone, lactide and propylene oxide. Because of the stereocenters, the control over the regio- and stereo-selectivity during synthesis is often in focus. A regio- and stereo-regular structure usually gives rise to increase the thermal stability, crystallinity and conductivity of the polymer.80–82 For the enchainment of monomers with stereocenters, there are three possible connectivity orders depending on the locations of the stereocenters along the polymer chain: head-to-tail, head-to-head and tail-to-tail; and for each stereocenter, the spatial configuration of the substituent can either be in the S form or R form. When the head-to-tail connection dominates the chain, the polymer is regioregular; and when either the S or the R form dominates the chain in long continuous sequences, the polymer is stereoregular. Take the three mentioned monomers for instance: ε-decalactone and lactide can only form the head-to-tail connection so the consideration for regularity control will only be stereoregularity; propylene oxide can be either opened from the C-O bond with the methyl substituent or the one without. Controlling the opening sites selectively as one of the two bonds will achieve a regioregular structure; and selectively ring-opening the S or R form of the monomer without racemization will result in a stereoregular structure. Although there is no absolute causation between the two regularities, regioregularity is often a perquisite for further consideration of stereoregularity for many polymers.83 Regioregular polycarbonates can be synthesized through the ring-opening copolymerization between carbon dioxide and propylene oxides84–87 or cyclohexene oxide,88–92 or it can be synthesized through the ROP of asymmetrically substituted cyclic carbonates.93–95 The catalysts used in these systems with high regioregularity control mostly are metal-based such as zinc, magnesium, cobalt, and aluminum. The use of organic catalysts to achieve the same regioselectivity is yet to be explored.
7 INTRODUCTION
2.3.2 Densely functionalized hydrogels Chemical hydrogels are a group of hydrophilic, water absorbing and elastic networks with covalent bonds connecting the building blocks. For specialty purposes, such as in tissue engineering and for use as absorbents or adsorbents, the hydrogels are often incorporated with reactive functional groups.96–98 These functionalities can be an inherent part of the precursors, or they can be post-added to a formed hydrogel. The amount of available functional groups greatly influences the hydrogel’s performance in the application. For instance, the fillers in baby diapers are composed of crosslinked sodium polyacrylates, which are called “superabsorbent”.96 The large number of available hydrophilic functional groups and the porous network structure promotes not only the absorption but also retention of the water. Other examples include ion-exchange resins, and dye and metal ion adsorbents.99–101 When designing a functional hydrogel, factors such as mechanical strength, shape deformation and network stability should also be considered. These factors can be tuned during precursor design by introducing hard/soft or rigid/labile segments into the chemical structure; they can also be tuned during the gel formation period by altering the geometry of the gelation mold or the crosslinking density.102–105
The removal of heavy metal ions from wastewater can be achieved through adsorption using an adsorbent, which is essentially a highly functionalized hydrogel. The network structure characteristics of the adsorbent such as porosity, hydrophilicity and the available amount of metal binding sites together determine the performance of the adsorbent.106 The hydrophilic porous structure enables the water uptake, and physical adsorption of some metal ions. The functional groups chemically fix the ions to the network. Some biobased adsorbent examples include crosslinked cellulose,107 chitosan108 and starch,109 which exhibit moderate to good heavy metal ion adsorption abilities.
The bactericidal capabilities of antibacterial hydrogels can be characterized as contact killing,110,111 diffusion killing112,113 or a combination of both mechanisms. The contact killing mechanism is from the positively charged groups on the hydrogel interacting with the negatively charged bacterial cell membrane, which subsequently causes cell lysis and the death of the bacterium.114 Some bactericidal chemicals, such as antibiotics and silver ions, can be released from the hydrogel upon conditioning with the bacterial solution, and these chemicals account for the diffusion killing mechanism.113,115 In all cases, the total amount and the reactivity of the available bactericidal sites in the network determine the effectiveness of the antibacterial hydrogel.
8 EXPERIMENTAL
3 EXPERIMENTAL
3.1 CHEMICALS
All chemicals were used as received unless described otherwise.
Anhydrous solvents for ring-opening polymerization: Tetrahydrofuran (THF, ≥99.9%), toluene (PhMe, 99.8%), acetonitrile (MeCN, 99.8%), and N,N-dimethylformamide (DMF, 99.8%) were purchased from Sigma- Aldrich and used as received.
Solvents for general reactions, dissolving and purification: Pyridine (anhydrous, 99.8%), methanol (≥99.8%), diethyl ether (anhydrous, ≥ 99.0%), iodomethane (MeI, 99%), m-cresol (99%), chloroform (CHCl3, 99.0%), and THF (99.0%) were purchased from Sigma-Aldrich. Acetone (technical grade), ethyl acetate (EtOAc, 99%), hexane (99%), and heptane (99%) were purchased from VWR. 1,1,1,3,3,3-Hexafluoro-2-propanol (HFIP, 98%) was purchased from Apollo Scientific, UK.
Base catalysts and comonomers for ring-opening reactions: Triethylamine (TEA, ≥99%), 1,5,7-triazabicyclo[4.4.0]dec-5-ene (TBD, 98%), 1,8- diazabicyclo[5.4.0]undec-7-ene (DBU, 99%), 1-tert-butyl-2,2,4,4,4-pentakis- 5 5 (dimethylamino)-2λ ,4λ -catenadi(phosphazene), (t-BuP2, ~2.0 M in THF), 2,2′- (ethylenedioxy)bis(ethylamine) (98%), sodium hydride (NaH, 60% dispersion in mineral oil), tris(2-aminoethyl)amine (TAEA, 96%), 1,4-diazacyclohexane (diamine-A, 99%), 1,3-bis(aminomethyl)benzene (diamine-B, 99%), 1,3- byclohexanebis(methyl-amine) (diamine-C, 99%), 1,10-diaminodecane (diamine-D, 97%), 2,2’-(ethylenedioxy)-bis(ethylamine) (diamine-E, 98%), 1,6-hexanediamine (diamine-H, 98%) and 1,3-pentanediamine (diamine-P, 98%) were purchased from Sigma-Aldrich. Bis(aminomethyl)-norbornane (diamine-N, 98 %) was purchased from TCI, Japan.
Other chemicals used in the studies:
9 EXPERIMENTAL
Hydrochloric acid (HCl, ~37%), acetic anhydride (≥99%), acetic acid (99%), ethylenediaminetetraacetic acid (EDTA, ≥99%) ethylenediaminetetraacetic acid disodium salt dihydrate (≥98.5%), δ-gluconolactone (GDL, ≥99.0%), ammonium hydroxide solution (NH3·H2O, 28.0 - 30.0%), ammonium chloride (NH4Cl, ≥99.5%), citric acid (99%), disodium hydrogen phosphate (≥99.0%), copper (II) sulfate pentahydrate (≥98%), nickel (II) nitrate hexahydrate (99%), cobalt (II) acetate (99.995%), calcium hydride (CaH2, 99%), 2-(allyloxymethyl)-2-ethyl-1,3- propanediol (98%), 2,2′-(ethylenedioxy)diethanethiol (EG-(SH)2, 95%), benzophenone (99%), potassium trifluoroacetate (KTFA, 98%), 1,8,9- anthracenetriol (dithranol, 98%), (±)-1,3-butanediol (99.5%), benzyl alcohol (BnOH, anhydrous, 99.8%), benzoic acid (PhCOOH, ≥99.5%), Escherichia coli (E. coli ATCC® 25922), Staphylococcus aureus (S. aureus ATCC® 6538), LB broth (Miller) and agar were purchased from Sigma-Aldrich. Diethyl carbonate (DEC, 99%) and murexide were purchased from Alfa Aesar. Chloroform-d (CDCl3, 99.8%), deuterium oxide (D2O, 99.9%) and dimethyl sulfoxide-d6 (DMSO-d6, 99.9%) were purchased from Cambridge Isotope Laboratories.
3.2 SYNTHESIS PROCEDURES
3.2.1 Synthesis of cyclic carbonates through depolymerization Both rac-4-methyl-1,3-dioxan-2-one (rac-1-MeTMC) and 5-((allyloxy)-methyl)-5- ethyl-1,3-dioxan-2-one (AOMEC) were synthesized following a ring-closing depolymerization procedure. For rac-1-MeTMC, DEC (80 g, 0.67 mol, 1.4 eq) was added to a 250 mL round-bottom flask, followed by the addition of 1 g NaH (5 mol% to hydroxyl groups) under nitrogen at room temperature. Upon fine dispersion with magnetic stirring, 43 g of (±)-1,3-butanediol (0.48 mol, 1 eq) was gradually added to the mixture. The reaction temperature was then elevated to 120 °C with the constant removal of ethanol condensate overnight. After cooling down and further removing the residual ethanol and unreacted DEC at 60 °C, the reaction temperature was gradually elevated to ~200 °C where the ring-closing depolymerization and subsequent distillation of the rac-1-MeTMC took place. This same procedure applied to the synthesis of 2-allyloxymethyl-2-ethyl-trimethylene carbonate (AOMEC) where the starting materials were DEC and 2- (allyloxymethyl)-2-ethyl-1,3-propanediol. For the following ring-opening polymerization (ROP) reactions, the rac-1-MeTMC was further purified through automated flash chromatography using an Isolera 4 unit (Biotage, Sweden) with hexane and EtOAc as the eluents (1:3, v:v, hexane:EtOAc). The overall yield of rac-1-MeTMC was ~63%, and ~72% for AOMEC.
10 EXPERIMENTAL
3.2.2 General ROP procedure for rac-1-MeTMC ROP of rac-1-MeTMC was carried out under the catalysis of various different base catalysts. For a t-BuP2-catalyzed ROP, approximately 464 mg (4 mmol) of rac-1- MeTMC was measured in a 1 mL disposable syringe inside a glove box. Approximately 1.1 mmol of BnOH (stock solution, 1 M in anhydrous THF) was measured into a 20 mL headspace vial (with a magnetic stir bar) followed by the addition of 50 μL (1 mmol) of t-BuP2 (~2 M THF solution). The headspace vial was then sealed and removed from the glove box together with the monomer-containing syringe. The vial was conditioned in a -40 °C cooling bath for 10 min prior to the quick injection of the monomer under vigorous stirring. The reactions were continued at -40 °C and quenched by adding an excess amount of PhCOOH (THF solution) to the reaction mixture after the desired reaction time. The quenched mixtures were air dried and stored at 4 °C.
3.2.3 Synthesis of PA through ROAC The PA samples were prepared according to a proposed ROAC procedure. A desired amount of diamine-C (3.32 g, 23.1 mmol) was weighed into a 100 mL round-bottom flask together with the catalyst TBD (0.54 g, 3.87 mmol, ~8 mol% to –NH2). The mixture was heated with an oil bath at 80 °C. After 30 min, ethylene brassylate (6.31 g, 22.1 mmol) was added to the mixture. After 4 h of reaction, the temperature was elevated to 100°C and retained for a total reaction time of 5 – 24 h. The raw product was transferred into glass containers, sealed and stored in a -20 °C freezer for future use. The samples used for analysis were purified by precipitation twice from HFIP solution to acetone. The purified PAs were vacuum dried and kept inside a desiccator before use.
3.2.4 Synthesis of gelation precursors Synthesis of N,N′-(1,3-Phenylenebis(methylene))bis- (2,3,4,5,6-pentahydroxyhexanamide) (PBP) PBP was synthesized according to an ROA procedure. δ-Gluconolactone (4.27 g, 24 mmol) was suspended in methanol (100 mL) under vigorous stirring at room temperature. Diamine-B (1.36 g, 10 mmol) was subsequently added to the mixture and the ROA reaction of δ-gluconolactone was initiated upon the addition of TEA as a catalyst. The reaction was left for 4 h to complete, and the crude product was filtrated and washed with cold methanol twice and cold dry diethyl ether once. The obtained white powder was further dried under reduced pressure to remove the residue solvent, yielding PBP as the product (4.63 g, 94% yield). The product was stored in a desiccator at room temperature until further use.
11 EXPERIMENTAL
Synthesis of N,N′-((Ethane-1,2-diylbis(oxy))bis(ethane-2,1-diyl))-bis(2,3,4,5,6- pentahydroxyhexanamide) (EBB) EBB was synthesized following the same ROA procedure as PBP, with the change from diamine-B to diamine-E. EBB was obtained as a white powder (4.58 g, 91% yield). The product was stored in a desiccator at room temperature until further use.
Synthesis of ethylenediaminetetraacetic dianhydride (EDTAD) EDTAD was synthesized according to a ring-closing procedure. EDTA (36 g, 0.12 mol) was suspended in 60 mL of anhydrous pyridine under vigorous stirring in an inert atmosphere. Acetic anhydride (50 mL) was then added to the suspension, and the reaction was continued for 24 h at 65 °C. The crude product was purified by filtration and the solid residue was washed with an excess amount of acetic anhydride and diethyl ether. Residual solvent was removed under vacuum, and the EDTAD was stored in a desiccator for future use.
Synthesis of the unsaturated hydroxyl urethane macromer (HUM) The HUM precursor was synthesized according to an ROA procedure. TAEA (6 g, 40 mmol) was mixed with AOMEC (24 g, 120 mmol) at room temperature under vigorous stirring in a round-bottom flask. The flask was then lowered into a preheated oil bath kept at 60 °C for 6 h. The obtained product was sealed and stored at room temperature for future use.
Synthesis of the bis-carbonate precursor The bis-carbonate precursor was synthesized according to a thiol-ene UV radical addition procedure. AOMEC (24 g, 120 mmol), EG-(SH)2 (11.5 g, 60 mmol) and benzophenone (1.7 g, 5 wt%) were mixed together in a round-bottom flask under vigorous stirring in an inert atmosphere. The mixture was then subjected to UV (365 nm) irradiation for 30 min, and the obtained product was sealed and stored for future use.
3.2.5 Gelation through esterification PBP, EBB and EDTAD were used as the building blocks for hydrogels with ester linkages. The esterification took place between the hydroxyl group on the PBP or EBB and the anhydride groups on EDTAD. The degree of crosslinking for the hydrogels was tuned by varying the amount of EDTAD. The total amount of hydroxyl groups from PBP and EBB are considered as 100 eq, and the total amount of anhydride groups from EDTAD are tuned from 70 eq, 100 eq to 140 eq. The hydrogels formed with PBP are termed as B-type and the gels formed with EBB are
12 EXPERIMENTAL termed as E-type. For a 140E hydrogel (anhydride:hydroxyl, 140:100 eq), EBB (252.2 mg, 0.5 mmol) and EDTAD (897 mg, 3.5 mmol) were dissolved in two separate vessels at 70 °C in a total of 5 mL DMF. The solutions were mixed under vigorous stirring. A trace amount of TEA was added as the catalyst to initiate the ring-opening esterification, leading to an immediate viscosity increase of the mixture. The mixture was then quickly transferred to a glass Petri dish in which the gel was formed. The total gelation time was approximately 5 min upon the addition of TEA. The gel was left still at room temperature for 24 h to enhance crosslinking. After removing the gel from the Petri dish, solvent exchange was performed by immersing the gel in an excess amount of methanol. The gel was then washed twice in diluted HCl solution (pH = 3) and once in deionized water. Finally, the gel was slowly dried inside a fume hood and stored in a desiccator for future use.
3.2.6 Gelation through thiol-ene UV radical addition
HUM precursor and EG-(SH)2 were used as the building blocks for the PHU gels through thiol-ene UV radical addition. HUM precursor (1.4 g, 1 eq), EG-(SH)2 (0.51 g, 1.5 eq) and benzophenone (20 mg, ~3 wt%) were dissolved together in a minimal amount of dry diethyl ether. The solution was cast onto a polytetrafluoroethylene (PTFE) evaporation dish in an N2 atmosphere and slowly dried. The dish was then transferred to a desiccator where reduced pressure was applied for debubbling. The gelation took place by irradiating the mixture under UV (365 nm) for 20 min where the thiol-ene UV radical addition between the double bond of the HU precursor and thiol from EG-(SH)2 occurred. The formed gel was immersed in dry THF to remove the initiator and possible unreacted precursors. The washed gel, namely, PHU-neat, was dried under vacuum and stored in a desiccator for future use.
3.2.7 Modification of gels The modification of the PHU-neat gels took place by immersing the gels into a MeI/THF solution for 28 h at 38 °C, where two reactions happened simultaneously: the quaternization of the tertiary amine from the TAEA building block, and the alkylation of the thiol ether from the EG-(SH)2 building block. The degree of modification was tuned by changing the concentration of the MeI/THF solution. PHU-5% means that the PHU-neat gel was treated with a MeI:THF = 1:19 (v:v) solution and PHU-20% means that the PHU-neat gel was treated with a MeI:THF = 1:4 (v:v) solution. After modification, the gels were washed repeatedly with dry THF, slowly dried inside a fume hood, then vacuum dried and stored in a desiccator for future use.
13 EXPERIMENTAL
3.3 CHARACTERIZATION METHODS
3.3.1 Nuclear magnetic resonance (NMR) The 1H NMR (400.13 MHz) and 13C NMR (100.62 MHz) spectra were recorded using an Avance 400 (Bruker, USA) spectrometer at 298 K. The samples were dissolved in deuterium solvents (CDCl3, DMSO-d6, D2O) using either the residual solvent (CDCl3, δ 7.26 ppm; DMSO-d6, 2.50 ppm; D2O, 4.79 ppm) or TMS (0.00 ppm) as the internal reference. The sample amounts for 1H NMR and 13C NMR were ~10 mg/mL and ~60 mg/mL, respectively.
3.3.2 Fourier transform infrared spectroscopy (FTIR) FTIR spectra were obtained using a Perkin-Elmer Spectrum 2000 equipped with a single reflection attenuated total reflectance (ATR) accessory (Golden Gate). The spectra were recorded for 16 scans at a resolution of 2 cm−1 from 4000 to 600 cm−1. The background was corrected for atmospheric moisture and CO2.
3.3.3 Real-time infrared spectroscopy (RTIR) RTIR was used to monitor the UV crosslinking kinetics of the PHU-neat gel. The apparatus was the same as FTIR, and the UV source (~8 mW/cm2) was from a Hamamatsu L5662 unit. Scans were taken within the range of 4000–600 cm−1 at a resolution of 4 cm-1 every 6 s for a total 20 min.
3.3.4 Fourier transform Raman spectroscopy (FT-Raman) FT-Raman was used to examine the structural change of the PHU gels after modification. The apparatus was the same as FTIR. The spectra were recorded within the range of 4000 ̶ 500 cm-1 at a laser power of 600 mW for 32 scans.
3.3.5 Scanning electron microscope (SEM) SEM images were obtained using a field emission SEM (Hitachi S4800, Japan). The samples were sputter-coated with Ag−Pd by a Cressington 208HR unit. The accelerating voltage varied from 1 kV to 5 kV.
3.3.6 Energy-dispersive X-ray spectroscopy (EDS) EDS images were obtained from the same SEM instrument as above with an 80 mm2 X-Max Large Area Silicon Drift Detector sensor (Oxford Instruments Nanotechnology). The accelerating voltage was 15 kV and the data were analyzed using AZtec INCA software.
14 EXPERIMENTAL
3.3.7 Size-exclusion chromatography (SEC)
SEC was used to obtain the elugram, Mn, Mw and Ð of the raw product and polymers during and after polymerization. A Verotech PL-GPC 50 Plus equipped with a PL-RI detector and two PLgel 5 mm MIXED-D columns (300×7.5 mm, Varian, Santa Clara) was used to analyze the samples. Chloroform was used as the eluent (1 mL/min, 30 °C) with toluene as the internal standard. Polystyrene standards (narrow) were used to create a twelve-point calibration curve with molecular weights ranging from 160 to 371,000 g/mol.
3.3.8 Thermogravimetric analysis (TGA) The thermal stability and thermal degradation profiles of the samples were studied though TGA, either on a TGA/SDTA 851e or a DSC/TGA1 unit (Mettler Toledo, USA). The samples (~4 ̶ 7 mg) were measured into a 70 µL alumina crucible with or without lids, and the experiments were carried out under an N2 flow (50 mL/min) at a heating rate of 10 K/min. The temperature range was from 50 to 600 °C.
3.3.9 Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS) MALDI-ToF was used to examine the repeating units and chain-ends of the PA samples on a Bruker UltraFlex L (Bruker Daltonics, Germany) equipped with a 337 nm N2 laser and a SCOUT-MTP ion source. For sample preparation, PA solution (~3 mg/mL), dithranol solution (~5 mg/mL) and KTFA solution (~8 mg/mL) were prepared in HFIP separately and were mixed at a 10: 30: 2 µL ratio. The mixture was vortexed, and 0.5 µL of the mixture was spotted onto a 384 MTP plate. The mixture was left inside a fume hood for cocrystallization before analysis.
3.3.10 Solution viscometry The viscosities of the PA were measured through solution viscometry. m-Cresol was used as the solvent. Four solutions with different concentrations (~0.6, 0.7, 0.8 and 0.9 g/dL) were made for each sample. The measurements were carried out at 25 ± 0.1 °C in a water bath through a LAUDA iVisc capillary viscometer equipped with an Ubbelhode viscometer (type II). For each sample, three time intervals within a ± 0.02 s range were chosen, and the average value was recorded as the flow-through time.
15 EXPERIMENTAL
3.3.11 pH measurement The pH measurements of buffers and solutions were carried out on a sympHony SB70P unit (VWR) equipped with a BioTrode pH sensor (Hamilton, Switzerland). The pH meter was calibrated at pH = 4.0, 7.0 and 10.0 at room temperature prior to use, and all measurements were conducted at room temperature.
3.3.12 Complexometric titration The concentration of the heavy metal solution was determined by direct complexometric titration. The heavy metal solution was masked with NH3·H2O/NH4Cl buffer, and the titration was carried out at pH = 10 using 0.01 M EDTA disodium solution as the titrant and Murexide as the indicator.
3.3.13 Optic density (OD)
The OD625nm value of the culture media was obtained with a Multiskan FC plate reader (Thermal Scientific). Samples of the culture media (100 µL) were taken at the prescribed times and placed into a 96-well plate without dilution. The average values and standard deviations for each sample were calculated on n = 3 with three readings. The inhibition ratio (IR%) after 24 h of incubation was calculated using the following equation (eq. 6, see Appendix). Negative control was the media conditioned with only bacteria (no PHU samples), and the positive control was media conditioned with only PHU samples (no bacteria).
3.3.14 Antibacterial assay The antibacterial properties of the PHU hydrogels were assessed by two methods: an agar diffusion method and a tube shaking method. The bacteria used for both methods were freshly cultured in LB media for 18 h at 37 °C before use. For the agar diffusion method, ~3 × 106 CFU of bacteria was seeded onto an LB agar plate. The PHU hydrogels were deposited onto the plate 10 min after the bacteria seeding. The agar plate was incubated at 37 °C for 24 h, and the zone of inhibition was visually detected afterwards. In the tube shaking method, square shaped PHU samples (0.5 cm × 0.5 cm) were submerged in 1.5 mL of LB broth containing ~3 × 106 CFU of bacteria inside a 2 mL vortex tube. The tubes were conditioned for 24 h at 37 °C and 100 rpm on a tilting plate. The evaluation was carried out by monitoring the OD625nm values of the media (100 µL) after conditioning with the PHU samples.
16 RESULTS AND DISCUSSION
4 RESULTS AND DISCUSSION
Biobased lactones and recyclable cyclic carbonates were used as the starting materials for the synthesis of functional polymers, ranging from linear polycarbonates and polyamides to crosslinked polyesters and polyurethanes. The principle synthetic routes used were ring-opening polymerizations and ring-opening aminolysis combined with esterification reactions and thiol-ene addition reactions. The focus here was on optimizing the existing routes and designing new routes for polymer preparation with benign conditions. The purpose is to contribute to the development of a future sustainable society and to create economic advantages.
4.1 REGIOREGULAR RING-OPENING POLYMERIZATION Aliphatic polycarbonates have found research applications as solid electrolytes in lithium batteries and as degradable polymeric scaffolds.116 One of the most common methods of synthesizing aliphatic polycarbonates is through the ROP of cyclic carbonates, alongside copolymerization between epoxides and carbon dioxide, and 117 polycondensation. The Gibbs free energy change during ROP is ΔGp= ΔHp - TΔSp, and the ROP occurs only when ΔGp is negative. For most small ring-size cyclic carbonates (e.g., 5- and 6-membered), the ΔHp and ΔSp terms are both negative. Increasing the reaction temperature of such a polymerization system can result in a positive ΔGp, pushing the reaction towards the opposite direction of ROP, i.e., ring-closing depolymerization (RCD), Scheme 3.35,62 The RCD behavior is more obvious and common to see in 5- and 6-membered cyclic carbonate systems than other ring-sizes due to the thermodynamic stability of their cyclic form. Both the ΔHp and ΔSp values can be influenced by changing the reaction solvent, which is due to the different interactions between the cyclic carbonate monomers and the solvent molecules.55 The concentration of the monomer also has an influence on the equilibrium, and diluting the reaction system favors the RCD direction due to entropic increase.62,118–123 The temperature of a polymerization system at and above which polymers cannot be formed is called the ceiling temperature (Tc), which is also influenced by the enthalpic and entropic factors of the system.124 As shown previously for a 6-membered cyclic carbonate (AOMEC), the Tc dropped from 247
17 RESULTS AND DISCUSSION
°C to 137 °C and 77 °C when diluting the system from bulk to 0.25 M with toluene and the very polar MeCN, respectively.36 In this sense, when raising the reaction temperature close to Tc or applying a reaction condition that results in a low Tc, the ring-chain equilibrium for the cyclic carbonate/polycarbonate system will be greatly favored the RCD direction. This is a valuable tool in the synthesis of 5- and 6- membered cyclic carbonates, because the polycarbonates themselves can be used as a starting material to be chemically recycled back into cyclic monomer form. 36,62,125,126 To achieve a solvent-free and large-scale production of the cyclic monomers, the employed RCD was carried out in bulk, and the condition was to raise the reaction temperature close to Tc. The reactive hydroxyl chain-end of the oligocarbonates or polycarbonates underwent continuous ring-closing reactions at the applied high temperature (>200 °C), liberating the 6-membered cyclic monomer one at a time. To facilitate the formation of the cyclic carbonate, the ring-chain equilibrium was constantly disturbed by the removal of the formed cyclic monomers (through distillation under low pressure),127 which continuously pushed the reaction towards the monomer formation side. The two 6-membered cyclic monomers synthesized through this method were rac-1-MeTMC and AOMEC, and they were used as building blocks for regioregular polycarbonates and the starting material of a precursor for a polyurethane hydrogel, respectively.
Scheme 3. Equilibrium polymerization of 6-membered cyclic carbonates using TMC and poly(TMC) as an example, and the monomers produced through RCD in this study, rac-1Me-TMC and AOMEC.
For asymmetrically substituted 6-membered cyclic carbonates, there are two possible acyl-oxygen cleavage sites during ROP, as is shown in Scheme 4a. This yields three possible configurations between the two connecting repeating units, i.e., head-to-head (HH), head-to-tail (HT) and tail-to-tail (TT). When the HT configuration dominates, the polycarbonate has a regioregular structure, which often is a prerequisite to further consider stereoregularity.83 In the case of making regioregular and stereoregular polypropylene oxide and polypropylene carbonate through ROP, these two regularities are often achieved simultaneously. A previous study has shown the possibility of making polylactide with long stereoblock chains 57 from rac-lactide using phosphazene base t-BuP2 as a catalyst. Since t-BuP2 is an achiral catalyst, the proposed mechanism was that the H-bonding interaction activated the initiator/t-BuP2 pair (Scheme 4b), which randomly ring-opened the first monomer. Due to the chiral structure of the first opened monomer, with the
18 RESULTS AND DISCUSSION sterically hindered alcohol chain-end/phosphazene ion pair, it then became a chiral activating site, and the next enchained monomer was dominated by the chirality of the previous monomer. Hence, it is reasonable to assume that t-BuP2 can be used to make regioregular poly(rac-1-MeTMC) following the same mechanism.
Scheme 4. ROP of rac-1-MeTMC catalyzed by t-BuP2. a) The two possible opening sites (i and ii) and three different configurations of the two neighboring monomers; b) H-bonding activation of the added alcohol initiator from t-BuP2.
The model carbonate compound, rac-1-MeTMC, was polymerized under similar conditions as in the aforementioned study with t-BuP2 as the catalyst. The carbonyl region of the crude samples were subsequently analyzed and compared through 13 94 their relative intensities in C NMR, Xreg = 1 - (HH + TT). The carbonyl region of 118,128 the product suggested moderately high regioregularity (Xreg = 0.86). This means that t-BuP2 induced the regioselective ROP behavior of rac-1-MeTMC. The acyl-oxygen cleavage of the rac-1-MeTMC can occur either at the unhindered bond (Scheme 4a, i) or at the hindered bond (Scheme 4a, ii), resulting in a primary alcohol chain-end or a secondary alcohol chain-end, respectively. The preferred cleavage site has been examined in previous studies through the analysis of the product’s chain-end, and a preference of either bond was shown depending on the catalytic systems.93,95,129 The chain-end of the poly(rac-1-MeTMC) in this study was analyzed through a combination of several NMR techniques, and only a slight preference of secondary alcohol (~63%) over primary alcohol (~37%) was seen (for detailed interpretation, see the appended Paper (I)), Figure 1. This not-so-significant preference over chain-end structure but with moderately high Xreg can be explained by the chain-end control pathway during ROP. It could be assumed that only the first enchained carbonate monomer was opened randomly, and the monomer being enchained later was following the same opening site as the previous one. The
19 RESULTS AND DISCUSSION regioregularity along the polymer chain is thus high, but the chain end structure is random. Another explanation is that since 6-membered cyclic carbonates/polycarbonate has an inherent ring-chain equilibrium polymerization behavior, the ring-opening/ring-closing reaction constantly occurs at the chain-end at high monomer conversion. This also results in a product with no significant preference for chain-end structure.
Figure 1. Compiled NMR spectra for chain-end analysis. a) Poly(rac-1-MeTMC) with primary alcohol as the chain-end; b) poly(rac-1-MeTMC) with secondary alcohol as the chain-end; c) 1H NMR before and after chain-end modification; d) 1H-13C HSQC of purified product; e) 13C DEPT of purified product; f) 1H-13C HMBC of purified product.
To examine the dependence of Xreg on reaction conditions, the four variables of solvent, reaction temperature, reaction time and reactant ratios were tuned and studied, Table 1. Under most of the studied conditions, the reaction equilibrium was reached within 180 min, exhibiting the high catalytic reactivity of t-BuP2. The kinetics of these entries was followed by 1H NMR (Figure 2a-b), and the time in Table 1 was recorded as the time closest to reaching equilibrium, (detailed kinetics graphs can be found in the appended Paper (I)). The Xreg of the products at the noted time were subsequently analyzed through 13C NMR, and the results are summarized in Figure 2c-f. In general, to achieve a higher Xreg for the t-BuP2 catalyzed ROP
20 RESULTS AND DISCUSSION system, toluene is a better solvent than THF, and lower temperature and lower catalyst loading are better. Quenching the reaction at the right time upon reaching equilibrium also plays a role in obtaining a higher Xreg. As exemplified by extending the reaction time to 20 h (equilibrium reached at 1 h), and the Xreg of the product dropped to 0, Figure 2f. This is because of the intensive transesterification reaction that occurred during the extended time at a higher temperature (60 °C).
Figure 2. a) 1H NMR showing peak shifts from monomer to polymer; b) compiled 1H NMR showing the region for conversion calculation; c) comparison between the different Xreg exhibited by the 13 carbonyl region in C NMR; d) Xreg dependence over time; e) Xreg dependence on solvent; f) Xreg dependence on temperature and solvent.
When comparing the different reaction conditions, a correlation between Xreg and the rate of polymerization was found; entries with faster polymerization rates usually had a higher Xreg (Table 1, entries 1-5). It appears that a faster ROP reaction resulted in less regioerror and that the regularity was kinetically controlled. The reaction conditions were, therefore, systemically optimized to favor a fast polymerization rate, and the correlation between polymerization rate and regioregularity was further explored. Toluene as a solvent has a lower dielectric constant (ε~2.4, less polar) than THF (ε~7.5),130 and the ether moiety in THF has a slight basicity which toluene does not.131 As the ROP follows an H-bonding and chain-end control mechanism, the competing H-bonding and complexation with THF at the chain-end can influence the enchainment of the monomers. This, in turn, may directly influences the Xreg of the polymer chain. This influence is also shown in the ROP kinetics, where a slower polymerization rate occurs when using
21 RESULTS AND DISCUSSION
THF as a solvent than when using toluene. A decreased polymerization rate and final monomer conversion was found at elevated temperatures (c.f., entries 2, 4, 5). As mentioned previously that the ring-chain equilibrium ROP is a significant behavior of 6-membered cyclic carbonates, the RCD direction in this equilibrium increases at an elevated reaction temperature. Since the ROP is already very fast at - 74 °C, increasing the reaction temperature changed the equilibrium conversion. As ring-opening and ring-closing appeared constantly at the reactive chain-end, the occurrence of regioerror greatly increased. This is supported by the lower Xreg of the formed polycarbonates, where the values decreased from 0.82 (entry 8, -74 °C) to 0.30 (entry 5, 60 °C) upon reaching equilibrium of the reaction. The initial activation site for the ROP was the H-bonding activated alcohol/t-BuP2 pair. When increasing the amount of monomer to activation site, a faster reaction was observed as expected. However, a slightly lower Xreg was also determined for the faster reaction (c.f., entries 1, 6). This can be explained by the number of formed chains in the polymerization system, where in theory entry 1 ([I]:[M]0=1:40) has five times the propagating chains as entry 6 ([I]:[M]0=1:200). Assuming the random cleavage site of the last enchained monomer at the polymer’s chain-end, a lower Xreg is expected for entry 1, which has more polymer chains (more activation sites). This slight decrease in Xreg can also be explained by the transesterification reactions, which are intensified at higher a catalyst loading. This is proved by the increased divergence at the same extended reaction time after reaching equilibrium, where the decrease of Xreg is more significant at higher catalyst (1:1:40) loading than at lower catalyst loading (1:1:200), Figure 2d.
Table 1. Summary of ROP conditions for all studied entries. b c d e a Temp. Ratio Conv. Time Entry Solvent Xreg (°C) [I]:[cat.]:[M]0 (%) (min) 1 PhMe -40 1:1:40 95 1 0.85 2 THF -40 1:1:40 95 30 0.82 3 PhMe R.T. 1:1:40 91 30 0.58
4 THF R.T. 1:1:40 86 30 0.54 5 THF 60 1:1:40 72 60 0.30 6 PhMe -40 1:1:200 94 10 0.90 7 THF -40 1:1:200 88 180 0.83
8 THF -74 1:1:40 94 60 0.82 9 PhMe R.T. 1:1:200 91 30 0.64 aAll reactions were carried out in 1 M solution; ba fluctuation of ±2 °C existed for reactions at -74 and -40 °C, R.T. was approximately 22 °C; cdesigned ratios, refer to the appended Paper (I) for the actual measured values; dcalculated from raw sample 1H NMR; erecorded as time point closest to the equilibrium time.
22 RESULTS AND DISCUSSION
4.2 RING-OPENING AMINOLYSIS Aminolysis of lactones and cyclic carbonates are nucleophilic addition-elimination reactions at the acyl carbon, resulting in a product with a hydroxyl group and an amide or urethane linkage. The ROA route expands the use of lactones and cyclic carbonates from making polyester and polycarbonates to polyamide and polyurethanes. Here, the focus is on the application of ROA reactions to make hydrogel precursors and polyamides, with the study of their respective mechanisms and material properties.
4.2.1 Ring-opening aminolysis (ROA) Two amide-containing polyol small molecules were made through ring-opening aminolysis (ROA) of GDL under the catalysis of the mild base TEA. The reaction takes place through the nucleophilic attack from the diamines on the carbonyl carbon of GDL, forming an amide bond, releasing the ring-strain of the cyclic monomer and resulting in a di-substituted structure, Scheme 5a-b. This reaction to make the precursor used methanol as a solvent, which is generally considered a “green solvent”. The starting material was GDL, a renewable lactone that is industrially mass-produced. No heating throughout the aminolysis reaction was needed, which reduced the energy consumption; and the purification steps were carried out by filtration, which saved additional work-up with the corresponding solvent consumption. No significant changes regarding the yield (~91 – 94%) or the purity of the product were seen when the diamine was changed from diamine-B to diamine-E, suggesting the versatility of the method employed, Scheme 5.
An allyl-functionalized hydroxyl urethane macromer (HUM) was synthesized through the ROA of AOMEC by TAEA, Scheme 5c. The reaction was carried out in bulk at 60 °C for 6 h. As seen from the structure of TAEA, the three primary amine functionalities are bonded to a center nitrogen atom, with TAEA intrinsically carrying a tertiary amine moiety. This tertiary amine has structural resemblance to TEA, indicating that the aminolysis between AOMEC and TAEA is self-catalyzed and no external catalyst is needed for the aminolysis to proceed. In addition, the tertiary amine moieties can easily undergo alkylation reactions with alkyl halides and protonation with acids, which allows for further facile modifications of the resulting macromers.
23 RESULTS AND DISCUSSION
Scheme 5. ROA of cyclic monomers to make hydrogel precursors. a) ROA of GDL with diamine-B to PBP; b) ROA of GDL with diamine-E to EBB; c) ROA of AOMEC with TAEA to HUM.
4.2.2 Ring-opening aminolysis condensation (ROAC) The formation of a polymer through the ROA route described above is a two-step procedure, where the functionalized monomers are made first via aminolysis, which is followed by a second step of polymerization. To make polymers in a one-pot and one-step aminolysis reaction directly, some structural requirements on the monomers are needed; either the monomers need to be an α,ω-amino acid ester,132 or the amine and the cyclic carbonate/lactone both need to be di- or bis- functionalized.79,133
To achieve a one-pot and one-step route to polyamides, ethylene brassylate (EB) was chosen as the di-lactone model compound, and the overall route is described as a ROA-condensation (ROAC) reaction, Figure 3a. Since the one-pot one-step ROAC is a polycondensation reaction, very high conversions of the reactants are crucial because the product from polycondensation will still be a low molecular weight oligomer when the conversion reaches above 90%.134 To reach higher aminolysis conversion, one can increase the amine equivalent, increase the reaction temperature, or use a stronger base catalyst.135 In the case of ROAC, increasing the amine to a slightly off-stoichiometric balance is acceptable and is even sometimes desirable in the sense of preventing the molecular weight from becoming too high. Excess amine will facilitate the aminolysis reaction, but at the same time, there will be an overloading of the amine chain-ends, which directly decreases the degree of polymerization. Hence, the ratio of amine to ester was here fixed at 1.05 for the
24 RESULTS AND DISCUSSION model studies, and the optimization of the conversion was done through optimizing the reaction temperature and base catalysts, Figure 3b.
Diamine-C was chosen as the model diamine compound due to its aliphatic structure, being liquid at room temperature (good miscibility with EB) and having a high boiling point (no loss of reactant at a higher reaction temperature). The reaction conditions for the optimization are summarized in the chart, Figure 3c. When examining the ester aminolysis that occurred between diamine-C and EB by FTIR, the ester carbonyl band intensity (C=O stretch, 1733 cm-1) decreased with the generation of two new amide bands at 1548 cm-1 (N-H binding, C-N stretch) and 1638 cm-1 (C=O stretch, amide), Figure 3c. Such changes were observed only for the TBD-catalyzed reaction (Figure 3c, entry 4) during the first half hour of mixing, meaning that the aminolysis did not occur spontaneously (Figure 3c, entry 1) and that TEA and DBU did not show any catalytic activities (Figure 3c, entry 2, 3) for the model compounds at room temperature. The ester peak continued to decrease after prolonged reaction times (Figure 3c, entry 5) and decreased further to unobservable levels at elevated reaction temperatures (Figure 3c, entry 6-8). From these trial reactions, it is clear that TBD is a principle catalyst for the aminolysis reactions, and a prolonged reaction time with elevated reaction temperatures facilitates the ROAC process.
Figure 3. Summary of the temperature and catalyst optimization reactions. a) Structures of the studied base catalysts, model diamine (diamine-C) and monomer EB; b) general route of ROAC between diamine-C and EB releasing ethylene glycol as a condensate; c) chart summary of reaction conditions and the corresponding carbonyl region of the IR spectra; the dashed line from left to right: 1733 cm-1 (C=O stretch, ester), 1638 cm-1 (C=O stretch, amide) and 1548 cm-1 (N-H binding, C-N stretch).
The model PA-C compound was prepared under the conditions in Figure 4a (1 eq EB, 1.05 eq diamine-C, 0.08 eq TBD, 100 °C, 24 h) and was purified by
25 RESULTS AND DISCUSSION precipitation from HFIP to acetone twice for the accuracy of characterization. The molecular weight of the purified PA-C was measured through solution viscometry, Figure 4b entry 9. The viscosity of polyamide melts and solutions is broadly used and reported in industry and in the literature, which gives more references for comparison.136–138 A good linear fit was shown from the collected viscosities at different solution concentrations (Figure 4c), and a calculated Mη of approximately 6 kDa was found using eq. 1 (see Appendix). Carrying out the ROAC at a higher temperature (170 °C) did not push the conversion further, as exemplified by the nearly identical viscosity, Figure 4b entry 10. This was slightly unexpected, since ester aminolysis was shown in the temperature optimization experiments as an endothermic process. One explanation for this lack of increase can be a maintained catalytic reactivity of TBD above 100 °C, where increasing the temperature will not further promote the conversion.
Another explanation is that the off-stoichiometric balanced addition of the EB and diamine-C system defined the molecular weight. Considering that the ROAC was essentially an off-stoichiometric balanced polycondensation system, the theoretical number-average degree of polymerization (Xn) was calculated using eq. 2 (see Appendix). An Xn of 41 was obtained under the assumption that no side reactions occurred and the polycondensation reached 100% conversion. From the Mη obtained from the solution viscometry, the calculated Xn was approximately 36, which was close to the theoretical value. Assuming that the Xn value from the solution viscometry was accurate, and applying this value (Xn = 36) for the reversed calculation using eq. 2, the conversation of the ROAC reaction was 94.8%. The detailed calculations regarding Mη, Xn and the conversion can be found in the Appendix. Although not an ideal 100%, the conversion is still very high and is comparable to or even higher than the small molecule syntheses discussed in the previous sections.67,69,139 These calculations clearly showed the influence on the molecular weight from an off-stoichiometric addition of the reactants, and this influence was further confirmed by the reaction with stoichiometric balanced reactants, where a higher viscosity of the product was seen, Figure 4b entry 11. It is believed that the cyclic structure of the EB also strongly contributed to this high conversion, as exemplified by the much lower viscosity of the sample PA-DS (Figure 4b entry 12) prepared with diamine-C and a linear diester DS, Figure 4a. Inherent ring-strain of the EB monomer exists regardless of its comparably large ring-size, although it is less significant than, for instance, GDL and AOMEC. The ester bonds on the di-lactone structure are, nonetheless, more activated than the ethyl esters of the linear analog, which makes it easier to reach higher conversion.
26 RESULTS AND DISCUSSION
Figure 4. Influences from monomer structures and stoichiometric balanced factors on the synthesis of PA-C. a) Optimized conditions between diamine-C and EB, chemical structure of DS; b) summary chart of altering monomer structure, stoichiometric balanced factors and reaction temperature; c) solution viscometry of purified PA-C and PA-DS samples using m-cresol as a solvent; the square dots connected by the linear-fitted line represented entry 9, and the other three entries are marked with the corresponding entry numbers.
The repeating unit and chain-end structure of purified PA-C was analyzed with MALDI-ToF, Figure 5. Based on the calculation of the molecular ion peaks, the major population represented PA-C with amine terminal groups at both ends. This is expected since the diamine-C was added at 1.05 eq to the ester amount. Two small populations of other molecular ions were also seen, which represented amine- ethylene glycol end-capped and amine-acid end-capped chains. Detailed calculations for the corresponding peaks can be found in the appended Paper (II).
Figure 5. MALDI-ToF analysis of purified PA-C. The population with the strongest intensity represents NH2-PA-NH2 molecular ions, the weaker populations are NH2-PA-COOH (left) and NH2- PA-OH (right).
27 RESULTS AND DISCUSSION
From the calculation of the molecular ions (see appended Paper (II)), PA-C product did not exhibit any ester bonds in the middle of the polymer chain, which was ideal since the ester bonds were significantly less thermally stable than the amide bonds when heated to a high temperature.6,140,141 However, a possible ROP side reaction should not be overlooked. When the aminolysis takes place between an amine group and a lactone or cyclic carbonate, a hydroxyl group is subsequently generated. This hydroxyl group can act as an initiator to homopolymerize the dilactone. This is not found in the case of GDL and AOMEC aminolysis because the base strength of TEA is not strong enough to promote homopolymerization of either GDL or AOMEC. In contrast, TBD is a much stronger base, which has been shown to be an efficient catalyst for the ROP of EB in recent studies.10,11,142
A model reaction to examine this possible competing pathway was carried out using hexylamine as the amine compound, Figure 6. SEC was applied to examine the elugram of the crude product from the model reaction between hexylamine and EB. The elugram showed peaks at several retention volumes, meaning that there was more than one product generated during the model reaction. Based on the molecular weight analysis of the corresponding peaks, the strongest peak with the highest retention volume represents the mono-substituted ring-opened EB, and the other lower retention volume peaks represent oligomers with different chain lengths, Figure 6. Due to the mono-functionality and slightly overloaded amount of hexylamine (1.05 eq to EB), the ROAC pathway to polyamide would not appear. This observation suggested that the ROP of EB did occur as a side reaction during the ROA of EB under the catalysis of TBD even when the reaction condition favored ROA. A second portion (0.95 eq to EB) of amine was added to the oligomeric mixture to determine if these formed oligomers could be aminolyzed completely. The reaction was followed by SEC and drastic drops in the peak intensities of the oligomers were found in the elugrams, Figure 6. The decrease for oligomers continued until they were unobservable from the elugram and the peak with the highest intensity shifted towards a slightly lower retention volume. The disappearance of the oligomer peaks suggested the oligoesters were aminolyzed from the addition of the second portion of the hexylamine. The shifted unimodal peak confirmed a slightly higher molecular weight product, which in this case was the di-substituted EB, Figure 6. Based on these results, it was clear that TBD- catalyzed ROP of EB occurred as a side reaction during the ROA of EB. The oligoester byproducts, however, were fully converted to amides through aminolysis.
28 RESULTS AND DISCUSSION
Figure 6. Model reactions between hexylamine (blue circle) and EB (red circle). Upon mixing, an oligomeric mixture was formed which was subsequently aminolyzed by the addition of a second portion of hexylamine. The elugram traces show the aminolysis of the oligoesters, forming a di- substituted EB as a product. The disappearance of the oligomeric compound (red dot line) and the formation of the di-substituted EB (black solid line) was compiled and compared.
The versatility of the ROA for small molecules synthesis was shown by changing the structures of amines in the previous section. It is of great interest to check if this versatility can be maintained and extended in the ROAC route. A group of commercially available diamines with systemic structure tuning was chosen based on the following considerations: secondary amine (diamine-A), aromatic (diamine- B), alkyl chain length (diamine-D (C10), H (C6)), heteroatoms (diamine-E), bulky structure (diamine-N) and varied reactivity (diamine-P), Figure 7a. The ROAC reactions between the diamines and EB were carried out under the model ROAC conditions as before (1.05 eq diamine, TBD, 100 °C, 24 h), and the viscosities of the purified samples were analyzed. Similar intrinsic viscosities between all but one (PA-P) of the samples were determined. This, again, supported the polycondensation feature of the system, where the molecular weight of the studied PAs were largely influenced and controlled by the slightly off-stoichiometric addition of the diamine. The much lower viscosity of PA-P is mainly caused by the low reactivity of the hindered primary amine, which has been shown as having one- hundredth the reactivity of the unhindered primary amine on the same molecule, Figure 7b.143 This means that the structure of the amine also influences the ROAC process, and amines with low reactivity cause a low molecular weight product. A possible pathway for the ROAC route is hence proposed based on the model reactions and characterization results discussed above. Upon mixing the diamine with EB under the catalysis of TBD at 100 °C, one of the amine groups of the diamine ring opens the cyclic structure of EB resulting in an amine-ethylene glycol
29 RESULTS AND DISCUSSION intermediate. This ring-opening step was fast and consumed most of the cyclic EB structure. Some ROP of EB may occur at this stage from the ethylene glycol end, forming small amounts of oligoesters. Intermolecular aminolysis-condensation of the intermediate mixture then reoccurred and reached high conversion over a prolonged time, which built the polymeric structure, Figure 7c. Changing the cyclic EB to a linear analog will remove the first step, which results in slower kinetics to the overall ROAC reaction, and the low reactivity of the amine will retard the condensation step.
Figure 7. a) Structures of the studied diamines; b) compiled intrinsic viscosity of the purified polymers, the viscosity of PA-P was taken from one point at 0.95 g/dL; c) proposed ROAC mechanism with two steps in one pot, the first step being the ROA reaction forming an amine-ethylene glycol ester intermediate, and the second step being the self-condensation of the intermediate forming the PAs.
4.3 DENSELY FUNCTIONALIZED HYDROGELS The microstructure of a network, such as the amount of functional groups and crosslink density, typically has a large influence on the end-use performance of a functional hydrogel. Depending on the purpose of the application, sometimes this density should be as high as possible, such as for superabsorbent in baby diapers; other times the amount of functional groups needs to be more precise, such as in peptide-functionalized hydrogels for tissue engineering applications.144,145 There are two ways to tune the amount of functional groups in a hydrogel: during crosslinking by changing the amount of functional crosslinkers or building blocks, or during the functionalization of a formed network.139,146,147 These two approaches were used
30 RESULTS AND DISCUSSION separately for the preparation of two different types of hydrogels with different end- use purposes: polyester hydrogels for heavy metal ion removal and polyurethane hydrogels for antibacterial studies.
4.3.1 Polyester hydrogels as heavy metal ion adsorbent Tuning the microstructure of the polyester hydrogels was performed during crosslinking by varying the amount of the crosslinkers. The preparation of the hydrogel is shown below, where the crosslinking reaction occurred as a ring- opening esterification reaction between the hydroxyl groups of the GDL-derived polyol and the anhydride groups from the crosslinker EDTAD, Figure 8a. It can be seen from the structure that after the anhydride moiety was ring-opened, both an ester group and a pendant carboxylic acid group were generated, Figure 8c. The ratios of the crosslinkers were increased from 70% (anhydride : hydroxyl) to 100% and then to 140%. Depending on the structure of the polyol, the corresponding hydrogels are termed as B-type (70B, 100B, and 140B) and E-type (70E, 100E and 140E), Figure 8b. The B-type precursor has a rigid hydrophobic benzyl ring structure in the middle, while the E-type has a flexible and more hydrophilic ethylene glycol structure. In the ideal case of the 100B and 100E gelation, all the hydroxyl groups from the polyol were stoichiometrically reacted with the anhydride moiety and replaced by the formed ester groups and pendant carboxylic groups. Tuning the amount of the crosslinker to either 70% or 140% will result in two changes to the network: the crosslinking density and the hydrophilicity. The combination of both will have an influence on the swelling behavior and the adsorption performance of heavy metal ions.
Figure 8. General gelation outline. a) Crosslinking through esterification between anhydrides and polyols; b) structures of the precursors, gelation systems and compositions; c) chemical structures at the crosslinking site.
31 RESULTS AND DISCUSSION
An isothermal swelling kinetics study was first carried out to determine the approximate time to reach the swelling equilibrium. To comply with existing reported studies,148,149 a buffer solution with pH 5.4 was chosen as a standard swelling condition to examine the swelling ratio differences between the different types of gels. The swelling ratio was calculated according to eq. 3 (see Appendix). As is shown in Figure 9a, a drastic increase in the swelling ratio appeared within the first hour upon conditioning in the buffer, and a slow increase or decrease in the swelling then occurred over time, with little changes occurring after 6 - 8 h. For ease of operation and to facilitate comparisons, from here on all the hydrogels were conditioned in buffers (citric acid/disodium hydrogen phosphate) for 24 h before characterization. Since the functionality of the pendant groups are carboxylic acids, there will be responsiveness towards the swelling medium when there are counter- ions and when there is a change in pH.150–152 A series of citric acid (0.1 M)/ disodium hydrogen phosphate (0.2 M) buffers with pH 2.6, 3.5, 4.5, 5.4, 6.4 and 7.0 was prepared and used as the swelling medium. The sodium ions in the buffer could undergo ion-exchange reactions with the pendant carboxylic acid groups in the hydrogels. Buffers with higher pH had a higher sodium ion concentration, which could facilitate this exchange reaction by the osmosis and water uptake between the hydrogel and the aqueous environment.148,153,154 This was verified by the gradual increase in the swelling ratio observed for both the 100B and 100E gels when the buffer pH was increased, Figure 9b. The swelling ratio difference between the 100B and 100E gels under the same buffer condition was only marginal, meaning that the structure of the two precursors did not impose a significant influence on the swelling behavior in a buffer medium.
Figure 9. Swelling comparisons of the polyester hydrogels. a) Isothermal swelling kinetics of the B- type gels; b) swelling ratios of both 100B gels and 100E gels under different buffer conditions.
32 RESULTS AND DISCUSSION
The crosslinking density was expected to decrease for the 70B, 70E, 140B and 140E gels, compared to the 100B and 100E gels if the crosslinking of 100B and 100E occurred ideally, where each of the hydroxyl groups were covalently reacted with an anhydride moiety, Figure 10a-c. Under this assumption, the 70B and 70E gels had more hydroxyl groups than the 100B and 100E gels, and the 140B and 140E gels had more carboxylic groups than the 100B and 100E gels. Both of these off-stoichiometric conditions will directly influence the hydrophilicity and the swelling ratio of the networks. This was verified by the swelling ratio differences at pH 5.4, where 70B and 70E swelled much more than the 100B, 100E, 140B and 140E gels, Figure 10d. When conditioning under the same pH in buffers, the interplay of the available hydroxyl groups and the lower crosslinking density made the gels more hydrophilic and swelling than the gels with excess amount of carboxylic groups and higher crosslinking. When changing the medium to a more protic one (pH = 3.5), the gels all swelled less (decrease of 40 – 50%), and the swelling ratio differences among all groups decreased. Hence, the overall swelling behavior was dictated by the pH of the swelling medium, Figure 10d.
Figure 10. Network functionality density illustrations and comparisons. a) Hydroxyl group excess; b) stoichiometrically balanced; c) carboxylic acid group excess; d) swelling ratio comparisons between the different gels in different buffers.
The crosslinker for the polyester hydrogels was EDTA dianhydride. Upon esterification between the GDL-derived polyol precursors (PBP and EBB), the EDTA moieties were immobilized and became a covalently bonded part of the network. EDTA is known as a strong ion chelator that forms stable metal complexes through coordination with the two tertiary amine groups and four carboxylic acid groups, Figure 11a.107,155–158 For each covalently bonded EDTA, at
33 RESULTS AND DISCUSSION least one, but often two, of the carboxylic acid groups were converted to ester groups, and this could be a factor that hinders the formation of metal ion complexes in space. An initial assessment was performed to check if the covalently immobilized EDTA moieties possess binding ability. Disk-shaped dried gels were immersed in a copper sulfate solution (Cu2+, 5 mM) for 24 h and subsequently taken out, rinsed and dried. In the image taken below, the change of the gel’s color from white to blue diagnostically indicates the bonded Cu2+ on the surface of the gel, Figure 11b. A decrease in pH was noticed during the adsorption process which was expected, as the cations in the solution were being replaced from Cu2+ to H+. As was discussed previously, a lower pH of the aqueous environment does not favor the swelling of the studied hydrogels, and causes the hydrogels to shrink during the adsorption. This will directly affect the adsorption capacity of the heavy metal ions, as a thorough and sustained penetration of the solution across the network is essential for ion binding. A decreased amount of Cu2+ per unit area across the thickness dimension of the disk-shaped gel was noticed through SEM-EDS elemental analysis, Figure 11c. To facilitate the metal ion adsorption, the surface area of the gels should be as large as possible while still maintaining a nondissolvable network structure. Hence, the dried gels were ground into gel particles for the following adsorption studies. In the following text, the term gel refers to the disk-shaped samples and the adsorbent refers to the gel particles.
Figure 11. EDTA as the metal-binding moiety in the hydrogel. a) Structure of EDTA, EDTA-metal complex and EDTA immobilized by the polyol precursor inside the gel network; b) color changes after metal binding both in disk shape and in particle form; c) vertical cross-sectional area of disk-shaped gels and EDS comparisons of differences in the amount of the immobilized Cu2+.
34 RESULTS AND DISCUSSION
The isothermal adsorption kinetics was carried out first to check the time for the adsorbent to reach the equilibrium adsorption point. Approximately 200 mg of the adsorbent was suspended in 130 mL of Cu2+ solution and placed on a shaker at 160 rpm at room temperature. After the desired time intervals, the suspension was filtered, and the Cu2+ remaining in the filtrate was determined by direct complexometric titration. As is shown in Figure 12a, the adsorption equilibrium for all studied samples was reached after 8 h from the adsorbent suspension. The isothermal adsorption kinetics was fitted into both Lagergren pseudo-second-order (Figure 12b) and Lagergren pseudo-first-order (Figure 12c) models to examine if the overall adsorption behaved more as physisorption or chemisorption. Similar to many other crosslinked biobased adsorbents, a good linearity was shown in the pseudo-second-order model, indicating a chemisorption-like adsorption behavior.149,159,160 The detailed calculations and equations for the Lagergren pseudo- first-order (eq. 4) and Lagergren pseudo-second-order (eq. 5) modeling are available in the Appendix.
Figure 12. Cu2+ adsorption assessment. a) Isothermal adsorption kinetics of 100B, 100E, 140B and 140E; b) Lagergren pseudo-second-order model fitting; c) Lagergren pseudo-first-order model fitting.
The maximum adsorption capacity (qm) was normalized to a standard unit, which represents the amount of heavy metal ions (mg) that could be adsorbed per gram of -1 adsorbent, Figure 13a-b. The studied adsorbent exhibited qms ~ 70 to 120 mg*g , which is among the higher ranges or is even higher than that of the majority of the reported biobased adsorbents.148,149,156,161,162 Another way to verify the adsorption ability is through a thermal analysis of the adsorbents after adsorption. The polyester network and organic part of the Cu-EDTA complex will undergo thermal degradation to gas when heated above 450 - 500 °C under a nitrogen atmosphere.163 As is shown in Figure 13c, the residue at above 500 °C did not increase or decrease in mass when heated further to 600 °C. These residues were determined as the metals that were adsorbed by the adsorbent. When normalized to qm and compared to the results obtained through titration, very similar values between the two
35 RESULTS AND DISCUSSION characterization methods for all of the studied adsorbents were obtained. The 140E 2+ 2+ 2+ adsorbent exhibited a similar qm as Cu for two other metal ions (Ni and Co ), as determined by TGA, Figure 13c. When comparing the titration results between the qm among all the six adsorbents, some gradual changes within each type were seen. For either the B-type or E-type, the 140B, 140E adsorbents generally adsorbed the highest amount of Cu2+, followed by 100B, 100E, and then 70B and 70E, Figure 13a-b. This was expected, as the active metal binding sites were EDTA, and 140B and 140E had the highest amount of EDTA moieties per unit gram. However, a remarkable difference was noticed when comparing between the 70B and 70E adsorbent (also seen for 100B and 100E), where 70E adsorbed significantly more Cu2+ than 70B. When looking back at the structural differences between the polyol precursors (Figure 8b), the flexible ethylene glycol moiety of the E-type precursor was more hydrophilic than the benzyl moiety in the B-type precursor. The higher hydrophilicity caused by the building blocks could facilitate more ion-exchange in an aqueous medium and further dominate the adsorption capacity, resulting in much more efficient E-type adsorbents than their counterpart B-type adsorbents.
Figure 13. Adsorption assessments through titration and TGA. a) Compiled qm values of different adsorbents from titration; b) comparisons between the qm values from TGA and titration; c) TGA trace comparisons of 140E adsorbed with Ni2+ and Co2+.
4.3.2 Antibacterial polyurethane hydrogels The amide containing polyol precursors was prepared through the ROA between amines and lactones. When using ROA chemistry on cyclic carbonates, the bond formed is hydroxyl urethane. The hydroxyl urethane macromer (HUM) was prepared and used as a precursor for the fabrication of an antibacterial polyurethane hydrogel using a combination of UV crosslinking and alkylation chemistry, Figure 14. Polyurethane gels were prepared through a thiol-ene UV addition103,164–166 between the –ene groups from the HUM precursors and the thiol groups from the crosslinkers, Figure 14. The thiol-ene addition in the preparation of a network is highly attractive because the reaction is very fast with close to quantitative
36 RESULTS AND DISCUSSION conversion at room temperature, can be readily triggered by UV light, and is possible to perform in solvent-free conditions.103,164,167–169
Figure 14. Illustration of network formation between the HUM and EG-(SH)2.
The thiol-ene addition between EG-(SH)2 and AOMEC was first carried out to examine the conversion of the reactant. Stoichiometrically balanced reactants were mixed with 5 wt% of UV initiator (benzophenone) and subjected to a UV365nm light source. The crude product was analyzed by 1H NMR after 15 min of light exposure, where the allylic double bond peaks had completely disappeared and the cyclic structure of the carbonate remained intact, Figure 15a-b.27 The thiol-ene addition was used to crosslink the film, and the kinetics during the crosslinking between the HUM and EG-(SH)2 were further analyzed through real-time FTIR, monitoring the disappearance of the allyl group (C-H absorption band at 918 cm-1) over time.166,170,171 As is shown in Figure 15c, a sharp decrease of the unreacted –ene group was found within the first half minute upon the application of UV light. A 90% conversion was reached after three and a half minutes, and the conversion at 20 min was above 95%. This further proved that the thiol-ene reaction was indeed a fast process in a solvent-free crosslinking system. The high final conversion and the high gel content (~90%) suggested that the fast crosslinking process did not cause any premature crosslinking, which could be a diminishing factor towards the incorporation of the building blocks. This fast and complete crosslinking is directly contributed by the thiol-ene addition, which by nature is a free radical reaction, but it has superior oxygen resistance compared to that of the oxygen or carbon radicals, Figure 15d.169,172–174 This property is especially desirable in applications where an oxygen-free atmosphere cannot be guaranteed.
37 RESULTS AND DISCUSSION
Figure 15. Thiol-ene UV addition reaction assessments. a) Thiol-ene UV radical addition reaction 1 between EG-(SH)2 and AOMEC; b) H NMR of thiol-ene UV addition reaction between EG-(SH)2 and AOMEC showing the disappearance of double bonds (in the dashed box) from the AOMEC reactant; c) crosslinking kinetics monitored through RTIR showing the remaining double bonds in the crosslinking system over time; d) catalytic cycle of thiol-ene UV radical reactions.
The microstructure tuning of the polyurethane hydrogel was performed by postmodification of the formed networks. After the crosslinking reaction, there are two types of linkages within the network possible for modification: the thiol ether linkage and the tertiary amine linkage. Both can be easily modified to the alkylated product through reaction with an alkyl halide, Figure 16a.175–177 The formed hydrogels were treated with different concentrations of methyl iodide (MeI), forming the corresponding methylated thiol ether and ammonium with iodide as the counter ion, Figure 16a. The two modified gels can be visually distinguished by their color, where the gel treated with 5% MeI has a bright yellow color and the gel treated with 20% MeI is brownish, Figure 16b. This difference in color could be directly attributed to the ionic moieties, as the methylated dialkyl thiol ether generally has a yellowish color.178 The increased degree of thiol ether alkylation was further confirmed by Raman spectroscopy, where the peaks at approximately -1 917 cm (S-CH3 rocking vibration) intensified from PHU-neat to PHU-5% then to PHU-20%, Figure 16c.179 The ionic moieties exhibited higher hydrophilicity than the nonmodified counterpart, and hence both the density of functionality and the hydrophilicity were increased simultaneously.
38 RESULTS AND DISCUSSION
Figure 16. Modification of PHU-neat gels. a) Methylation reaction of the PHU-neat network forming methylated moieties with iodides as counter-ions; b) photograph of the PHU-neat, PHU-5% and PHU- 20% gels; c) Raman spectroscopy of the three gels, showing increased intensity of the methylated sulfur ether group and maintained intensity of hydroxyl group with PHU-5% and PHU-20%.
The swelling behavior of the unmodified and modified polyurethane hydrogel was studied using LB-Miller broth as the medium. The isothermal swelling kinetic parameter showed that the swelling equilibrium for the three types of gels was reached after 14 h upon immersion, Figure 17a. Since the modified gels are intended for antibacterial applications, the swelling ratios were tested both in DI water and in LB-Miller broth. It can be seen within each swelling medium that the PHU-20% gels swelled more than the PHU-5% gels and much more than PHU-neat gel. This was directly caused by the hydrophilicity of the network where the modified gels were more hydrophilic than the neat gel, Figure 17b. The PHU-5% and PHU-20% gels also both swelled more in the LB broth than their counterpart in DI water. This was caused by the ions presented in the LB-Miller broth, which facilitated the swelling behavior in a similar manner as that of the buffers for the carboxylic acid polyester hydrogels.
39 RESULTS AND DISCUSSION
Figure 17. Swelling ratio comparisons. a) Isothermal swelling kinetics of the three gels in LB-Miller broth; b) swelling comparisons between the three gels in DI water and in LB broth at equilibrium.
The elemental compositions of the polyurethane gels were analyzed by SEM-EDS. The surface and cross-sectional areas of the modified gels (PHU-5% and PHU- 20%) were compared to the PHU-neat gels, Table 2. Notably, the SEM-EDS result is a good indication of large elemental composition changes, but a reliance on the quantitative analysis should be carefully considered. As is concluded in Table 2, the elemental composition of the gels after modification exhibited high iodide content (~20 wt%) and the iodide content differences between the surface and the cross- sectional areas were marginal. This suggested that the methylation process proceeded thoroughly throughout the network. The elemental composition did not change after conditioning in the DI water for 24 h for the modified gels, meaning that the iodide species within the network didn’t react with water during the conditioning time. However, the iodide species almost completely disappeared (less than 1 wt% remaining) after conditioning in the LB-Miller broth for 24 h. This suggested that the ions and chemicals from the LB-Miller broth reacted and substituted the iodide to another type of counter-ion. Based on the existing studies of similar systems and the increased weight of oxygen in this system, the counter- ions were assumed to have switched from iodides to hydroxides, Figure 18a.178,180– 182 These resulting QAC and TSC with hydroxides as the counter-ions were highly basic, which was believed to have switched to another type of oxygen-containing counter-ion. Since LB-Miller broth is a mixture of tryptone, yeast extract and sodium chloride, further studied are needed to specify what the exact counter-ion becomes. This counter-ion switch was further examined by TGA of the unconditioned and LB-Miller broth-conditioned gels, Figure 18b. PHU-neat gels showed no differences in thermal degradation behavior before and after LB- conditioning, meaning that the unmodified tertiary amine and thiol ether moieties were inert towards the LB-Miller broth. The unconditioned PHU-5% and PHU-20%
40 RESULTS AND DISCUSSION gels exhibited a ~20% mass loss at approximately 110 °C, representing the loss of the iodide species; the first maximum degradation rate occurred at approximately 200 °C. The LB-Miller broth-conditioned PHU-5% and PHU-20% exhibited an increased thermal stability with a small weight loss at approximately 110 °C, which could be attributed to the loss of the residue iodide species and some bonded water within the network. The maximum degradation rate occurred at approximately 300 °C, which was nearly one hundred degrees higher than the unconditioned counterpart. The altered thermal degradation profiles are in good agreement with similar reported studies.180,181
Table 2. SEM-EDS elemental analysis of modified gels before and after conditioning, numbers are in weight percentage.
PHU- PHU-5% PHU-20% Neat cross DI LB cross DI LB surface surface surface sec. cond. cond. sec. cond. cond. C 63.2 55.1 52.9 52.9 61.0 51.2 52.6 51.3 58.5 O 23.4 16.9 22.2 16.8 25.6 17.5 19.0 17.8 26.3 I - 20.5 16.8 20.4 0.1 23.1 21.2 22.1 0.9
Figure 18. a) Proposed counter-ion exchange of QAC iodide, TSC iodide within the PHU network into hydroxides counter-ion as an intermediate; b) comparison of thermal stabilities between unmodified PHU gels, modified PHU gels, unconditioned PHU gels and conditioned PHU gels.
The antibacterial properties of the PHU-5% and PHU-20% gels were assessed both in liquid cultures (tube shaking method) and solid cultures (agar plate diffusion). The active bactericidal sites of the network are the cation species generated through the alkylation of the tertiary amine and thiol ether moieties; the alkylated species are termed as QAC (quaternary ammonium compound) and TSC (tertiary sulfonium compound) in the following discussion. The target of the cationic QAC/TSC during
41 RESULTS AND DISCUSSION the bactericidal activity is the bacterial cell membrane. The electrostatic interactions between the negatively charged bacterial cellular membrane and the positively charged QAC/TSC take place when in contact. The alkyl chains from the QAC/TSC then permeate into the intramembrane area, resulting in the leakage of cytoplasmic material and eventually cell lysis.183–186
The antibacterial properties of the neat (PHU-neat) and modified (PHU-5% and PHU-20%) gels were evaluated against two types of commonly studied bacteria: Escherichia coli (E. coli, gram-negative) and Staphylococcus aureus (S. aureus, gram-positive). The antibacterial assay was first carried out in the liquid culture (LB-Miller broth). The square-shaped samples were immersed in the medium containing ~3 × 106 CFU bacteria inside a vortex tube. After the desired conditioning time, the optic density of the liquid culture at 625 nm (OD625nm) was recorded. The OD625nm value is directly related to the amount of viable bacteria, and a higher OD625nm value means that there are more living bacteria in the culture.185,187,188 A sample containing only bacteria and LB-broth was used as the negative control, and a sample containing only gel and LB-broth was used as the positive control. The clearness of the liquid culture after conditioning was a direct indication of the bactericidal activity of the modified gels, since a culture containing a high amount of bacteria is often turbid, Figure 19a. Judging from the contrast of the black lines in the background, the negative control and the PHU-neat samples were completely opaque, while the liquid culture containing the PHU-5% and PHU- 20% gels were transparent.
Quantitative analysis was carried out by comparing the OD625nm values of each medium and the inhibition ratios calculated based on the OD625nm values from eq. 6 (see Appendix). The histogram below shows that the OD625nm values of the liquid culture for PHU-5% and PHU-20% are nearly identical to their positive controls, and most of the inhibition ratios for both types of bacteria are close to or above 99.9%, Figure 19b-d. To further examine if there were any live bacteria in the PHU- 5% and PHU-20%-treated cultures, 100 µL of the undiluted cultures were plated on agar plates and incubated for 24 h.189 Compared to the PHU-neat-treated samples, there were markedly fewer bacterial colonies on the PHU-5%-treated sample plates, Figure 19c. Significantly, no colonies were formed for the samples treated with PHU-20%, meaning that the seeded bacteria (~3 × 106 CFU) were completely killed by the PHU-20% gels. These results strongly support the high bactericidal capability of the QAC- and TSC-loaded networks.
42 RESULTS AND DISCUSSION
Figure 19. Antibacterial assay in solution. a) Solution opacity after 24 h of incubation with different PHU samples; b) calculated inhibition ratio; c) CFU images after 24 h of incubation; d) optical density comparisons between the different conditions.
As a counter-ion exchange from iodide to hydroxides was observed during the conditioning in LB-Miller broth for the PHU-5% and PHU-20% gels, these exchanged iodides could be seen as a diffusive substance from the gels that also contributed to the bactericide activities.70–72 An antibacterial assay was next carried out on solid culture (LB-Miller broth agar plates) to verify the diffusion killing mechanism. The PHU-5% and PHU-20% samples were preconditioned in LB-broth for 24 h to complete the counter-ion exchange; the samples after the exchange were termed as PHU-5%-OH and PHU-20%-OH. The four different samples, i.e., PHU- 5%, PHU-20%, PHU-5%-OH and PHU-20%-OH, were then placed on an agar plate with freshly seeded bacteria (~3 × 106 CFU) and incubated for 24 h. Due to the hydroscopic properties and water absorption of the PHU-5% and PHU-20% networks, a dimension loss was observed very soon after the deposition of the dried PHU-5% and PHU-20% gels on the agar plate. The actual contact areas with the agar plate for these two dried gels were much smaller than the size of the dried gels themselves, Figure 20. This was not the case for the PHU-5%-OH and PHU-20%- OH gels, since they were applied directly in the wet form after the preconditioning step, Figure 20a. The actual contact areas of all four gels are circled in Figure 20b- c. Zones of inhibition with different areas were clearly shown for the PHU-5% and PHU-20% gels, suggesting that the iodides exchanged between the dried gels and
43 RESULTS AND DISCUSSION the agar plate during incubation acted as the diffusion killing species. The zones of inhibition were also larger in area for the PHU-20% gels than the PHU-5% gel, which was expected since the PHU-20% gels had denser exchangeable iodide counter-ions within the network. Regarding the PHU-5%-OH and PHU-20%-OH gels, there were no bacteria growing under the covered areas, and there were no zones of inhibition showing. This suggested that the contact killing mechanism was still active after the counter-ion exchange, which was also expected since the contact-killing property was provided by the positively charged QAC/TSC moieties. From the above studies, it can be concluded that the alkylation of the PHU-neat network with MeI successfully generated different degrees of cationic- loaded networks with iodide as the counter-ion. The modified networks, namely, PHU-5% and PHU-20%, showed exceptional bactericidal performance against both gram-negative (E. coli) and gram-positive (S. aureus) bacteria, with inhibition ratios above 99.9%. The bactericidal mechanism was proven by a combination of contact killing by the QAC/TSC moieties and diffusion killing through the exchanged iodides.115,192,193
Figure 20. a) Illustration showing the dimension loss of the unconditioned hydroscopic PHU-5% and PHU-20% gels upon contacting the agar plate; PHU gel conditioning prior to disposition on the agar plate to maintain the dimension; b) agar plate diffusion against E. coli, 1, PHU-20% unconditioned, 2, PHU-5% unconditioned, 3, PHU-20% LB media 24 h swelling and 4, PHU-5% LB media 24 h swelling; c) agar plate diffusion against S. aureus, 1, PHU-20% unconditioned, 2, PHU-5% unconditioned, 3, PHU-20% LB media 24 h swelling and 4, PHU-5% LB media 24 h swelling.
44 CONCLUSIONS
5 CONCLUSIONS
The utilization of renewable lactones and cyclic carbonates to make a wide range of polymeric materials, such as polycarbonate, polyesters, polyamides and polyurethane with fine-tuned yet diverse structural differences (i.e., linear and network) has been explored in this study. The following efforts are believed to contribute to the construction of more sustainable polymers for the future:
(I) Cyclic carbonates were synthesized through depolymerization and then polymerized into regioregular polycarbonates. Low temperature and less
catalyst (t-BuP2) facilitated a higher degree of regioregularity. (II) Structurally diverse biobased polyamides were synthesized through ROAC using the rapeseed oil-derived EB as the monomer. The processing window was comparable to that of previous studies; the solvent use was eliminated during synthesis, and the reaction temperature was greatly reduced. (III) A biobased polyester hydrogel was synthesized and used as a heavy metal ion adsorbent with superior removal ability. The precursors were synthesized from sugar-derived lactone (GDL) and crosslinked with the metal chelator EDTA dianhydride. The solvent and energy consumption during the synthesis was greatly reduced compared to the preparation of similar materials. (IV) Polyurethane hydrogels were synthesized through photo crosslinking between an isocyanate-free hydroxyl urethane macromer and a dithiol. A transparent thin film was formed without solvent or high-toxicity chemical consumption during the precursor syntheses. After methylation, the network was loaded with positively charged QACs and TSCs exhibiting high bactericidal performance against both gram-positive and gram- negative bacteria. The bactericidal property was exhibited as a dual- mechanism: contact killing and diffusion killing.
45 FUTURE WORK
6 FUTURE WORK
As this thesis presented a broad range of applications using biobased lactones and cyclic carbonate as the starting materials, more in-depth work can be continued for the described systems for both the ROP and ROA studies.
Since the regioregular ROP of rac-1-MeTMC was achieved by simply lowering the reaction temperature when using phosphazene base (t-BuP2) as the catalyst, it would be very interesting to examine if stereoregular ROP can be achieved at the same time. This approach will serve both as a proof-of-concept for the use of a nonchiral nonmetal organic catalyst to make chiral polymers from racemic mixtures and as a means to facilitate the understanding of phosphazene-catalyzed ROP of cyclic carbonates in general. Linear polyamides were made through the ROAC between dilactone EB and diamines; a polyamide network can be directly obtained when replacing the diamine in the ROAC route with a tri- or polyamine. This ring- opening aminolysis-crosslinking (ROAX) pathway would further exploit the high reactivity at room temperature feature of the TBD-catalyzed ROAC system and the high stability of the amide linkages, with potential application as high-temperature stable adhesives. The polyester hydrogel precursor was synthesized using petroleum-based diamine and biobased GDL. To make a fully biobased precursor with inherent charged groups, α-amino acid with di-amine substituents such as lysine would be a good candidate to replace the petroleum-based diamine. This substitution would further embrace the renewable features of the adsorbents.
46 ACKNOWLEDGEMENTS
7 ACKNOWLEDGEMENTS
Dear Karin Odelius, thank you for offering me a chance to work with you. You inspire me with your sharp scientific instincts, your strict research ethics, and your meticulous working style. Your guidance leaves me enough freedom to explore my research interest, and at the same time pull me back to sanity when I run wild. Your valuable life experiences make me realize my strong and weak points, and lead me to understand myself better. I enjoyed the honor and pleasure of working with you every day of the last four years.
Johan Franzén, thank you for all the scientific inputs from a real organic chemist. You constantly widen my knowledge of chemistry in inspiring ways. The balance you have between life and work is yet something I am eager to learn and acquire.
The harmonious environment at Polymer Technology is built together by our helpful staff, caring seniors and friendly and hardworking colleagues. For sure it offers the strongest support for my future success as a polymer chemist, and leaves me irreplaceable memories in Teknikringen 56-58.
Finally, I would like to acknowledge the financial support from Swedish Research Council, VR (Grant ID: 621201356 25) and Formas (Grant ID: 2016-00700_3) to make this project possible and to welcome me in this beautiful and prosperous country of Sweden.
爸妈，粗算下来我已离家十载，不知道我是否独立成了你们期望的那个模样。 感谢这么多年的无处不在的爱，支持与理解，你们持灯的手永远那么执着， 在灰暗的日子里给我照亮了路，温暖了家。
47 REFERENCES
8 REFERENCES
1 http://ec.europa.eu/environment/circular-economy/index_en.htm (Accessed on 2018-04-26) 2 https://sustainabledevelopment.un.org/ (Accessed on 2018-04-26) 3 K. Yao and C. Tang, Macromolecules, 2013, 46, 1689–1712. 4 A. Gandini, T. M. Lacerda, A. J. F. Carvalho and E. Trovatti, Chem. Rev., 2015. 5 Y. Zhu, C. Romain and C. K. Williams, Nature, 2016, 540, 354–362. 6 L. Ragupathy, U. Ziener, R. Dyllick-Brenzinger, B. von Vacano and K. Landfester, J. Mol. Catal. B Enzym., 2012, 76, 94–105. 7 D. Myers, T. Witt, A. Cyriac, M. Bown, S. Mecking and C. K. Williams, Polym. Chem., 2017, 8, 5780–5785. 8 M. Bouyahyi, M. P. F. Pepels, A. Heise and R. Duchateau, Macromolecules, 2012, 45, 3356–3366. 9 R. Todd, S. Tempelaar, G. Lo Re, S. Spinella, S. A. McCallum, R. A. Gross, J.-M. Raquez and P. Dubois, ACS Macro Lett., 2015, 4, 408–411. 10 A. Pascual, H. Sardón, F. Ruipérez, R. Gracia, P. Sudam, A. Veloso and D. Mecerreyes, J. Polym. Sci. Part A Polym. Chem., 2015, 53, 552–561. 11 J. Fernández, A. Larrañaga, A. Etxeberria and J.-R. Sarasua, RSC Adv., 2016, 6, 22121–22136. 12 M. P. F. Pepels, I. Hermsen, G. J. Noordzij and R. Duchateau, Macromolecules, 2016, 49, 796–806. 13 J. Kollár, M. Mrlík, D. Moravčíková, Z. Kroneková, T. Liptaj, I. Lacík and J. Mosnáček, Macromolecules, 2016, 49, 4047–4056. 14 M. A. Hillmyer and W. B. Tolman, Acc. Chem. Res., 2014, 47. 15 L. Su, S. Khan, J. Fan, Y.-N. Lin, H. Wang, T. P. Gustafson, F. Zhang and K. L. Wooley, Polym. Chem., 2017, 8, 1699–1707. 16 J. A. Galbis, M. de G. García-Martín, M. V. de Paz and E. Galbis, Chem. Rev., 2016, 116, 1600–1636. 17 D. J. Darensbourg and S. H. Wei, Macromolecules, 2012, 45, 5916–5922. 18 D. McGinty, C. S. Letizia and A. M. Api, Food Chem. Toxicol., 2011, 49, S174–S182.
REFERENCES
19 C. Sell, in The Chemistry of Fragrances, Royal Society of Chemistry, Cambridge, 2006, vol. 6, pp. 52–131. 20 M. Hong and E. Y.-X. Chen, Green Chem., 2017, 19, 3692–3706. 21 X. Zhang, M. Fevre, G. O. Jones and R. M. Waymouth, Chem. Rev., 2018, 118, 839–885. 22 D. K. Schneiderman and M. A. Hillmyer, Macromolecules, 2017, 50, 3733– 3749. 23 F. Sasse and G. Emig, Chem. Eng. Technol., 1998, 21, 777–789. 24 K. H. Adolfsson, S. Hassanzadeh and M. Hakkarainen, RSC Adv., 2015, 5, 26550–26558. 25 E. Bäckström, K. Odelius and M. Hakkarainen, Ind. Eng. Chem. Res., 2017, 56, 14814–14821. 26 D. Wu and M. Hakkarainen, Eur. Polym. J., 2015, 64, 126–137. 27 P. Olsén, K. Odelius and A.-C. Albertsson, Macromolecules, 2014, 47, 6189–6195. 28 X. Michel, S. Fouquay, G. Michaud, F. Simon, J.-M. Brusson, J.-F. Carpentier and S. M. Guillaume, Eur. Polym. J., 2017, 96, 403–413. 29 S. Enthaler and A. Trautner, ChemSusChem, 2013, 6, 1334–1336. 30 P. Hodge, Chem. Rev., 2014, 114, 2278–2312. 31 S. Enthaler and A. Trautner, ChemSusChem, 2013, 6, 1334–1336. 32 H. H. G. Jellinek and H. Kachi, J. Polym. Sci. Part C Polym. Symp., 2007, 23, 97–108. 33 A. Sudo, K. Kataoka, F. Sanda and T. Endo, Macromolecules, 2004, 37, 251–253. 34 T. Endo, K. Kakimoto, B. Ochiai and D. Nagai, Macromolecules, 2005, 38, 8177–8182. 35 H. Höcker and H. Keul, Adv. Mater., 1994, 6, 21–36. 36 P. Olsén, J. Undin, K. Odelius, H. Keul and A.-C. Albertsson, Biomacromolecules, 2016, 17, 3995–4002. 37 A. C. Albertsson and I. K. Varma, Biomacromolecules, 2003, 4, 1466–1486. 38 W. N. Ottou, H. Sardon, D. Mecerreyes, J. Vignolle and D. Taton, Prog. Polym. Sci., 2016, 56, 64–115. 39 J. Wu, T. L. Yu, C. T. Chen and C. C. Lin, Coord. Chem. Rev., 2006, 250, 602–626. 40 N. E. Kamber, W. Jeong, R. M. Waymouth, R. C. Pratt, B. G. G. Lohmeijer and J. L. Hedrick, Chem. Rev., 2007, 107, 5813–5840. 41 J. Lagowski, P. Edelman, M. Dexter and W. Henley, Semicond. Sci. Technol., 1992, 7, A185–A192. 42 M. R. Gongora-Rubio, M. B. A. Fontes, Z. M. Da Rocha, E. M. Richter and L. Angnes, in Sensors and Actuators, B: Chemical, Elsevier, 2004, vol. 103,
REFERENCES
pp. 468–473. 43 L. Zhang, F. Nederberg, R. C. Pratt, R. M. Waymouth, J. L. Hedrick and C. G. Wade, Macromolecules, 2007, 40, 4154–4158. 44 E. F. Connor, G. W. Nyce, M. Myers, A. Möck and J. L. Hedrick, J. Am. Chem. Soc., 2002, 124, 914–915. 45 R. C. Pratt, B. G. G. Lohmeijer, D. A. Long, R. M. Waymouth and J. L. Hedrick, J. Am. Chem. Soc., 2006, 128, 4556–4557. 46 N. E. Kamber, W. Jeong, S. Gonzalez, J. L. Hedrick and R. M. Waymouth, Macromolecules, 2009, 42, 1634–1639. 47 A. Sanchez-Sanchez, A. Basterretxea, D. Mantione, A. Etxeberria, C. Elizetxea, A. de la Calle, S. García-Arrieta, H. Sardon and D. Mecerreyes, J. Polym. Sci. Part A Polym. Chem., 2016, n/a-n/a. 48 H. Kim, J. V Olsson, J. L. Hedrick and R. M. Waymouth, ACS Macro Lett., 2012, 1, 845–847. 49 A. Sanchez-Sanchez, I. Rivilla, M. Agirre, A. Basterretxea, A. Etxeberria, A. Veloso, H. Sardon, D. Mecerreyes and F. P. Coss??o, J. Am. Chem. Soc., 2017, 139, 4805–4814. 50 B. G. G. Lohmeijer, R. C. Pratt, F. Leibfarth, J. W. Logan, D. A. Long, A. P. Dove, F. Nederberg, J. Choi, C. Wade, R. M. Waymouth and J. L. Hedrick, Macromolecules, 2006, 39, 8574–8583. 51 N. J. Sherck, H. C. Kim and Y.-Y. Won, Macromolecules, 2016, 49, 4699– 4713. 52 M. Helou, O. Miserque, J.-M. Brusson, J.-F. J.-F. Carpentier and S. M. Guillaume, Chem. – A Eur. J., 2010, 16, 13805–13813. 53 S. M. Guillaume and J.-F. Carpentier, Catal. Sci. Technol., 2012, 2, 898– 906. 54 S. M. Guillaume and L. Mespouille, J. Appl. Polym. Sci., 2014, 131, 40081/1. 55 A. Duda and A. Kowalski, in Handbook of Ring-Opening Polymerization, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Germany, 2009, pp. 1– 51. 56 P. Olsén, K. Odelius, H. Keul and A.-C. Albertsson, Macromolecules, 2015, 48, 1703–1710. 57 L. Zhang, F. Nederberg, J. M. Messman, R. C. Pratt, J. L. Hedrick and C. G. Wade, J. Am. Chem. Soc., 2007, 129, 12610–12611. 58 M. Hong and E. Y.-X. X. Chen, Nat. Chem., 2015, 8, 42–49. 59 S. A. Pendergraph, G. Klein, M. K. G. Johansson and A. Carlmark, RSC Adv., 2014, 4, 20737. 60 J. M. W. Chan, X. Zhang, M. K. Brennan, H. Sardon, A. C. Engler, C. H. Fox, C. W. Frank, R. M. Waymouth and J. L. Hedrick, J. Chem. Educ.,
REFERENCES
2015, 92, 708–713. 61 W. Guerin, M. Helou, M. Slawinski, J.-M. Brusson, J.-F. Carpentier and S. M. Guillaume, Polym. Chem., 2014, 5, 1229–1240. 62 S. Tempelaar, L. Mespouille, O. Coulembier, P. Dubois and A. P. Dove, Chem. Soc. Rev., 2013, 42, 1312–1336. 63 Y. A. Chang, A. E. Rudenko and R. M. Waymouth, ACS Macro Lett., 2016, 5, 1162–1166. 64 B. Bibal, C. Thomas and B. Bibal, Green Chem., 2014, 16. 65 C. Thomas and B. Bibal, Green Chem., 2014, 16, 1687–1699. 66 Y. Pocker and E. Green, J. Am. Chem. Soc., 1976, 98, 6197–6202. 67 W. Guo, J. E. Gómez, L. Martínez-Rodríguez, N. A. G. Bandeira, C. Bo and A. W. Kleij, ChemSusChem, 2017, 10, 1969–1975. 68 S. Kobayashi, Y. Narukawa, S. Mori, T. Hashimoto and T. Saegusa, Macromolecules, 1983, 16, 858–863. 69 W. Liu, D. D. Xu, O. Repič and T. J. Blacklock, Tetrahedron Lett., 2001, 42, 2439–2441. 70 A. S. Goje, S. A. Thakur, V. R. Diware, Y. P. Chauhan and S. Mishra, Polym. Plast. Technol. Eng., 2004, 43, 407–426. 71 A. Viswanathan and D. E. Kiely, J. Carbohydr. Chem., 2003, 22, 903–918. 72 M. A. Foley and T. F. Jamison, Org. Process Res. Dev., 2010, 14, 1177– 1181. 73 N. U. Kaya, F. E. Du Prez and N. Badi, Macromol. Chem. Phys., 2017, 218, n/a. 74 P. Espeel, F. Goethals, F. Driessen, L.-T. T. Nguyen and F. E. Du Prez, Polym. Chem., 2013, 4, 2449. 75 G. Rokicki, P. G. Parzuchowski and M. Mazurek, Polym. Adv. Technol., 2015, 26, 707–761. 76 V. Besse, F. Camara, F. Méchin, E. Fleury, S. Caillol, J. P. Pascault and B. Boutevin, Eur. Polym. J., 2015, 71, 1–11. 77 K. E. Price, C. Larrivée-Aboussafy, B. M. Lillie, R. W. McLaughlin, J. Mustakis, K. W. Hettenbach, J. M. Hawkins and R. Vaidyanathan, Org. Lett., 2009, 11, 2003–2006. 78 X. J. Loh, W. Guerin and S. M. Guillaume, J. Mater. Chem., 2012, 22, 21249–21256. 79 M. Helou, J.-F. Carpentier and S. M. Guillaume, Green Chem., 2011, 13, 266–271. 80 G. Zhang, R. C. Huber, A. S. Ferreira, S. D. Boyd, C. K. Luscombe, S. H. Tolbert and B. J. Schwartz, J. Phys. Chem. C, 2014, 118, 18424–18435. 81 T. M. Ovitt and G. W. Coates, J. Am. Chem. Soc., 2002, 124, 1316–1326. 82 G. W. Coates and R. M. Waymouth, Science (80-. )., 1995, 267, 217 LP-
REFERENCES
219. 83 M. I. Childers, J. M. Longo, N. J. Van Zee, A. M. LaPointe and G. W. Coates, Chem. Rev., 2014, 114, 8129–8152. 84 X. B. Lu and Y. Wang, Angew. Chemie - Int. Ed., 2004, 43, 3574–3577. 85 C. T. Cohen, T. Chu and G. W. Coates, J. Am. Chem. Soc., 2005, 127, 10869–10878. 86 Z. Qin, C. M. Thomas, S. Lee and G. W. Coates, Angew. Chemie - Int. Ed., 2003, 42, 5484–5487. 87 S. Inoue, H. Koinuma and T. Tsuruta, J. Polym. Sci. Part B Polym. Lett., 1969, 7, 287–292. 88 M. Cheng, N. A. Darling, E. B. Lobkovsky and G. W. Coates, Chem. Commun., 2000, 0, 2007–2008. 89 K. Nakano, K. Nozaki and T. Hiyama, J. Am. Chem. Soc., 2003, 125, 5501– 5510. 90 K. Nozaki, K. Nakano and T. Hiyama, J. Am. Chem. Soc., 1999, 121, 11008–11009. 91 K. Nakano, K. Nozaki and T. Hiyama, Macromolecules, 2001, 34, 6325– 6332. 92 W. Guerin, A. K. Diallo, E. Kirilov, M. Helou, M. Slawinski, J.-M. Brusson, J.-F. Carpentier and S. M. Guillaume, Macromolecules, 2014, 47, 4230– 4235. 93 P. Brignou, J.-F. J.-F. Carpentier and S. M. Guillaume, Macromolecules, 2011, 44, 5127–5135. 94 P. Brignou, M. Priebe Gil, O. Casagrande, J.-F. J.-F. Carpentier and S. M. Guillaume, Macromolecules, 2010, 43, 8007–8017. 95 M. Huang, C. Pan and H. Ma, Dalt. Trans., 2015, 44, 12420–12431. 96 Y. Zheng and A. Wang, Eur. Polym. J., 2015, 72, 661–686. 97 K. T. Nguyen and J. L. West, Biomaterials, 2002, 23, 4307–4314. 98 S. E. Stabenfeldt, A. J. Garcia and M. C. LaPlaca, J. Biomed. Mater. Res. - Part A, 2006, 77, 718–725. 99 W. . Wan Ngah, C. . Endud and R. Mayanar, React. Funct. Polym., 2002, 50, 181–190. 100 G. Crini, Prog. Polym. Sci., 2005, 30, 38–70. 101 V. K. Mourya and N. N. Inamdar, React. Funct. Polym., 2008, 68, 1013– 1051. 102 Z. K. Zander, G. Hua, C. G. Wiener, B. D. Vogt and M. L. Becker, Adv. Mater., 2015, 27, 6283–6288. 103 P. M. Kharkar, M. S. Rehmann, K. M. Skeens, E. Maverakis and A. M. Kloxin, ACS Biomater. Sci. Eng., 2016, 2, 165–179. 104 F. Lin, J. Yu, W. Tang, J. Zheng, A. Defante, K. Guo, C. Wesdemiotis and
REFERENCES
M. L. Becker, Biomacromolecules, 2013, 14, 3749–3758. 105 W. Zhao, L. Glavas, K. Odelius, U. Edlund and A. C. Albertsson, Chem. Mater., 2014, 26, 4265–4273. 106 W. S. Wan Ngah and M. A. K. M. Hanafiah, Bioresour. Technol., 2008, 99, 3935–3948. 107 O. K. Júnior, L. V. A. Gurgel, R. P. de Freitas and L. F. Gil, Carbohydr. Polym., 2009, 77, 643–650. 108 Y.-H. Chang, C.-F. Huang, W.-J. Hsu and F.-C. Chang, J. Appl. Polym. Sci., 2007, 104, 2896–2905. 109 L. Huang, C. Xiao and B. Chen, J. Hazard. Mater., 2011, 192, 832–6. 110 T. J. Cuthbert, R. Guterman, P. J. Ragogna and E. R. Gillies, J. Mater. Chem. B, 2015, 3, 1474–1478. 111 W. He, Y. Zhang, F. Luo, J. H. Li, K. Wang, H. Tan and Q. Fu, RSC Adv., 2015, 5, 89763–89770. 112 F. Aubert-Viard, A. Martin, F. Chai, C. Neut, N. Tabary, B. Martel and N. Blanchemain, Biomed. Mater., 2015, 10, 15023. 113 D. E. Fullenkamp, J. G. Rivera, Y. Gong, K. H. A. Lau, L. He, R. Varshney and P. B. Messersmith, Biomaterials, 2012, 33, 3783–3791. 114 K. C. Rice and K. W. Bayles, Microbiol. Mol. Biol. Rev., 2008, 72, 85–109. 115 R. K. Mishra, K. Ramasamy and A. B. A. Majeed, J. Appl. Polym. Sci., 2012, 126, E98–E107. 116 J. Xu, E. Feng and J. Song, J. Appl. Polym. Sci., 2014, 131, n/a-n/a. 117 G. Rokicki and P. G. Parzuchowski, in Polymer Science: A Comprehensive Reference, 10 Volume Set, Elsevier, 2012, vol. 4, pp. 247–308. 118 J. Matsuo, F. Sanda and T. Endo, Macromol. Chem. Phys., 2000, 201, 585– 596. 119 S. Kiihling, H. Keul and H. Hocker, Macromol. Chem., 1992, 1217, 1207– 1217. 120 S. Kühling, H. Keul and H. Höcker, Die Makromol. Chemie, 1990, 191, 1611–1622. 121 O. Haba, H. Tomizuka and T. Endo, Macromolecules, 2005, 38, 3562–3563. 122 D. Takeuchi, T. Aida and T. Endo, Macromol. Chem. Phys., 2000, 201, 2267–2275. 123 T. Ariga, T. Takata and T. Endo, Macromolecules, 1997, 30, 737–744. 124 F. S. Dainton and K. J. Ivin, Q. Rev., 1958, 12, 61. 125 P. Olsén, K. Odelius and A.-C. Albertsson, Biomacromolecules, 2016, 17, 699–709. 126 G. Rokicki, Prog. Polym. Sci., 2000, 25, 259–342. 127 J. W. Hill and W. H. Carothers, J. Am. Chem. Soc., 1933, 55, 5031–5039. 128 M. H. Chisholm, D. Navarro-Llobet and Z. Zhou, Macromolecules, 2002,
REFERENCES
35, 6494–6504. 129 J. Ling, Y. Dai, Y. Zhu, W. Sun and Z. Shen, J. Polym. Sci. Part A Polym. Chem., 2010, 48, 3807–3815. 130 J. R. Rumble, Ed., CRC Handbook of Chemistry and Physics, CRC Press/Taylor & Francis, Boca Raton, FL., 98th Editi., 2018. 131 H. Alamri, J. Zhao, D. Pahovnik, N. Hadjichristidis, J. L. Hedrick, C. G. Wade, T. Kakuchi, Y. Y. Yang, J. L. Hedrick, R. M. Waymouth and J. L. Hedrick, Polym. Chem., 2014, 5, 5471–5478. 132 N. Kolb, M. Winkler, C. Syldatk and M. A. R. Meier, Eur. Polym. J., 2014, 51, 159–166. 133 A. A. Wróblewska, S. Lingier, J. Noordijk, F. E. Du Prez, S. M. A. A. De Wildeman, K. V. Bernaerts, A. A. Wroblewska, S. Lingier, J. Noordijk, F. E. Du Prez, S. M. A. A. De Wildeman and K. V. Bernaerts, Eur. Polym. J., 2017, 96, 221–231. 134 G. Odian, Principles of Polymerization, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2004, vol. 58. 135 M. Blain, L. Jean-Gérard, R. Auvergne, D. Benazet, S. Caillol and B. Andrioletti, Green Chem., 2014, 16, 4286–4291. 136 D. R. Vardon, N. A. Rorrer, D. Salvachúa, A. E. Settle, C. W. Johnson, M. J. Menart, N. S. Cleveland, P. N. Ciesielski, K. X. Steirer, J. R. Dorgan and G. T. Beckham, Green Chem., 2016, 18, 3397–3413. 137 L.-H. Wang, F. J. B. Calleja, T. Kanamoto and R. S. Porter, Polymer (Guildf)., 1993, 34, 4688–4691. 138 S. Samanta, J. He, S. Selvakumar, J. Lattimer, C. Ulven, M. Sibi, J. Bahr and B. J. Chisholm, Polymer (Guildf)., 2013, 54, 1141–1149. 139 P. Espeel, F. Goethals and F. E. Du Prez, J. Am. Chem. Soc., 2011, 133, 1678–1681. 140 A. C. Fonseca, M. H. Gil and P. N. Simões, Prog. Polym. Sci., 2014, 39, 1291–1311. 141 A. Basterretxea, E. Gabirondo, A. Sanchez-Sanchez, A. Etxeberria, O. Coulembier, D. Mecerreyes and H. Sardon, Eur. Polym. J., 2017, 95, 650– 659. 142 A. Pascual, H. Sardon, A. Veloso, F. Ruipérez and D. Mecerreyes, ACS Macro Lett., 2014, 3, 849–853. 143 K. Marchildon, Macromol. React. Eng., 2011, 5, 22–54. 144 J. Su, B. H. Hu, W. L. Lowe, D. B. Kaufman and P. B. Messersmith, Biomaterials, 2010, 31, 308–314. 145 Z. Ma, Q. Li, Q. Yue, B. Gao, X. Xu and Q. Zhong, Bioresour. Technol., 2011, 102, 2853–2858. 146 W. Zhao, L. Glavas, K. Odelius, U. Edlund and A.-C. Albertsson, Chem.
REFERENCES
Mater., 2014, 26, 4265–4273. 147 F. Lin, J. Yu, W. Tang, J. Zheng, A. Defante, K. Guo, C. Wesdemiotis and M. L. Becker, Biomacromolecules, 2013, 14, 3749–3758. 148 L. V. A. Gurgel, R. P. de Freitas and L. F. Gil, Carbohydr. Polym., 2008, 74, 922–929. 149 E. S. Dragan, D. F. Apopei Loghin and A. I. Cocarta, ACS Appl. Mater. Interfaces, 2014, 6, 16577–16592. 150 C. K. S. Pillai, W. Paul and C. P. Sharma, Prog. Polym. Sci., 2009, 34, 641– 678. 151 N. G. Kandile and A. S. Nasr, Int. J. Biol. Macromol., 2014, 64, 328–33. 152 J. Ruiz, A. Mantecón and V. Cádiz, Polymer (Guildf)., 2001, 42, 6347– 6354. 153 A. Ayoub, R. A. Venditti, J. J. Pawlak, A. Salam and M. A. Hubbe, ACS Sustain. Chem. Eng., 2013, 1, 1102–1109. 154 J. He, Y. Lu and G. Luo, Chem. Eng. J., 2014, 244, 202–208. 155 A. M. Senna, K. M. Novack and V. R. Botaro, Carbohydr. Polym., 2014, 114, 260–268. 156 F. Zhao, E. Repo, D. Yin, Y. Meng, S. Jafari and M. Sillanpää, Environ. Sci. Technol., 2015, 49, 10570–10580. 157 E. Repo, J. K. Warchol, T. A. Kurniawan and M. E. T. Sillanpää, Chem. Eng. J., 2010, 161, 73–82. 158 E. Repo, T. A. Kurniawan, J. K. Warchol and M. E. T. Sillanpää, J. Hazard. Mater., 2009, 171, 1071–1080. 159 R. K. Gautam, A. Mudhoo, G. Lofrano and M. C. Chattopadhyaya, J. Environ. Chem. Eng., 2014, 2, 239–259. 160 J. Tao, J. Xiong, C. Jiao, D. Zhang, H. Lin and Y. Chen, ACS Sustain. Chem. Eng., 2016, 4, 60–68. 161 S. Pandey and S. Tiwari, Carbohydr. Polym., 2015, 134, 646–56. 162 S. Pal, S. Ghorai, C. Das, S. Samrat, A. Ghosh and A. B. Panda, Ind. Eng. Chem. Res., 2012, 51, 15546–15556. 163 T. R. Bhat and R. Krishna Iyer, J. Inorg. Nucl. Chem., 1967, 29, 179–185. 164 M. J. Kade, D. J. Burke and C. J. Hawker, J. Polym. Sci. Part A Polym. Chem., 2010, 48, 743–750. 165 M. Uygun, M. A. Tasdelen and Y. Yagci, Macromol. Chem. Phys., 2010, 211, 103–110. 166 D. B. Larsen, R. Sønderbæk-Jørgensen, J. Duus and A. E. Daugaard, Eur. Polym. J., 2018, 102, 1–8. 167 S. C. Ligon, B. Husár, H. Wutzel, R. Holman and R. Liska, Chem. Rev., 2014, 114, 577–589. 168 L. He, D. Szopinski, Y. Wu, G. A. Luinstra and P. Theato, ACS Macro Lett.,
REFERENCES
2015, 4, 673–678. 169 S. R. Zavada, N. R. McHardy and T. F. Scott, J. Mater. Chem. B, 2014, 2, 2598–2605. 170 T. Yoshimura, T. Shimasaki, N. Teramoto and M. Shibata, Eur. Polym. J., 2015, 67, 397–408. 171 C. Nilsson, N. Simpson, M. Malkoch, M. Johansson and E. Malmström, J. Polym. Sci. Part A Polym. Chem., 2008, 46, 1339–1348. 172 C. E. Hoyle, T. Y. Lee and T. Roper, J. Polym. Sci. Part A Polym. Chem., 2004, 42, 5301–5338. 173 J. D. McCall and K. S. Anseth, Biomacromolecules, 2012, 13, 2410–2417. 174 L. Kwisnek, S. Nazarenko and C. E. Hoyle, Macromolecules, 2009, 42, 7031–7041. 175 M. Hirayama, Biocontrol Sci., 2011, 16, 23–31. 176 J. R. Kramer and T. J. Deming, Biomacromolecules, 2012, 13, 1719–1723. 177 T. J. Deming, Bioconjug. Chem., 2017, acs.bioconjchem.6b00696. 178 H. Paulsson, A. Hagfeldt and L. Kloo, J. Phys. Chem. B, 2003, 107, 13665– 13670. 179 G. Socrates, Infrared and Raman Characteristic Group Frequencies: Tables and Charts, John Wiley& Sons Inc., Chichester, 3rd Editio., 2001. 180 H.-S. Dang and P. Jannasch, J. Mater. Chem. A, 2016, 4, 11924–11938. 181 M. A. Hossain, H. Jang, S. C. Sutradhar, J. Ha, J. Yoo, C. Lee, S. Lee and W. Kim, in International Journal of Hydrogen Energy, Pergamon, 2016, vol. 41, pp. 10458–10465. 182 E. Alcalde, I. Dinarès, A. Ibáñez and N. Mesquida, Molecules, 2012, 17, 4007–4027. 183 M. C. Jennings, K. P. C. Minbiole and W. M. Wuest, ACS Infect. Dis., 2016, 1, 288–303. 184 A. Kanazawa, T. Ikeda and T. Endo, J. Polym. Sci. Part A Polym. Chem., 1993, 31, 2873–2876. 185 L. Huang, G. Szewczyk, T. Sarna and M. R. Hamblin, ACS Infect. Dis., 2017, 3, 320–328. 186 J. Haldar, P. Kondaiah and S. Bhattacharya, J. Med. Chem., 2005, 48, 3823– 3831. 187 N. Beyth, I. Yudovin-Farber, R. Bahir, A. J. Domb and E. I. Weiss, Biomaterials, 2006, 27, 3995–4002. 188 Y. Huang, X. Ding, Y. Qi, B. Yu and F.-J. Xu, Biomaterials, 2016, 106, 134–143. 189 W. Z. Xu, L. Yang and P. A. Charpentier, ACS Sustain. Chem. Eng., 2016, 4, 1551–1561. 190 C. Cheng, A. He, C. X. Nie, Y. Xia, C. He, L. Ma and C. S. Zhao, J. Mater.
REFERENCES
Chem. B, 2015, 3, 4170–4180. 191 H. Bakhshi, H. Yeganeh, S. Mehdipour-Ataei, M. A. Shokrgozar, A. Yari and S. N. Saeedi-Eslami, Mater. Sci. Eng. C, 2013, 33, 153–164. 192 D. Macocinschi, D. Filip, S. Vlad, C. G. Tuchilus, A. F. Cristian and M. Barboiu, J. Mater. Chem. B, 2014, 2, 681–690. 193 F. Zheng, S. Wang, S. Wen, M. Shen, M. Zhu and X. Shi, Biomaterials, 2013, 34, 1402–1412.
APPENDIX
9 APPENDIX
0.792 [η] = 0.5 + 0.0353푀휂 (eq. 1)
[η] is the intrinsic viscosity of the polyamide in m-cresol, and Mη is the calculated viscosity-average molecular weight.
1+푟 X̅̅̅̅ = (eq. 2) n 1+푟−2푟푝
Xn is the number-average degree of polymerization, r is the ratio between the two off-stoichiometric balanced reactants (r is always no larger than 1), and p is conversion of the reaction (p is always no larger than 1).
푚 −푚 푆푅 = 푡 0 (eq. 3) 푚0
SR is the swelling ratio of the hydrogel, mt is the weight of the swollen hydrogel at time t, m0 is the dry weight of the gel. ln(푄1푡ℎ푒표 − 푄푡) = 푙푛푄1푡ℎ푒표 − 푘1푡 (eq. 4) 푡 푡 1 = + 2 (eq. 5) 푄푡 푄2푡ℎ푒표 푘2푄2푡ℎ푒표
-1 -1 Qt is the adsorption amount (mg*g ) at time t (min); Q1theo (mg*g ) and Q2theo (mg*g-1) are the theoretical adsorption capacity at equilibrium; the rate constants in -1 the pseudo-first-order k1 (min ) and the rate constant of pseudo-second-order -1 -1 models k2 (g*mg *min ).
푂퐷 −푂퐷 퐼푅% = (1 − 625푛푚,푒푥푝푒푟𝑖푚푒푛푡푎푙 625푛푚,푝표푠𝑖푡𝑖푣푒 푐표푛푡푟표푙 )*100% (eq. 6) 푂퐷625푛푚,푛푒𝑔푎푡𝑖푣푒 푐표푛푡푟표푙−푂퐷625푛푚,푝표푠𝑖푡𝑖푣푒 푐표푛푡푟표푙
IR% is the calculated inhibition ratio.