Methyl-1-Pentene

Functional Copolymers of Maleic Anhydride - Synthesis and Application

Von der Fakultät für Mathematik, Informatik und Naturwissenschaften der RWTH Aachen University zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften genehmigte Dissertation

Vorgelegt von

Diplom-Chemiker Marian Szkudlarek aus Krappitz

Berichter: Universitätsprofessor Dr. rer. nat. Martin Möller Universitätsprofessor Dr. rer. nat. Andrij Pich

Tag der mündlichen Prüfung: 03.09.2019

Diese Dissertation ist auf den Internetseiten der Universitätsbibliothek verfügbar.

To my family: Sylwia, Marta and Robert

Table of Contents

List of abbreviations v

Summary viii

Chapter 1 Introduction 1

Chapter 2 Literature review: Radical batch copolymerization. Copolymers of homogenous 8 composition. Fluorinated polymers. 10 Maleic anhydride 11 Antimicrobial Quaternary Ammonium Salt Polymers (QAS) 21

Chapter 3 Synthesis of Terpolymers with Homogeneous Composition by 36 Free Radical Copolymerization of Maleic Anhydride, Perfluorooctyl and Butyl or Dodecyl Methacrylates: Application of the Continuous Flow Monomer Addition Technique Appendix Chapter 3 65

Chapter 4 Solubility, Emulsification and Surface Properties of Maleic 79 Anhydride, Perfluorooctyl and Alkyl Meth-Acrylate Terpolymers

Chapter 5 Water soluble perfluorinated terpolymers containing phosphate 97 groups.

Chapter 6 Synthesis, modification and antimicrobial properties of copolymers 118 of maleic anhydride and 4-methyl-1-pentene.

List of Publications 151

Acknowledgements 152

List of abbreviations

List of abbreviations, acronyms and symbols:

δ chemical shift Θ contact angle μL micro liter º degree (angle) °C Celsius degree 1H-NMR proton nuclear magnetic resonance Å angstrom Ac2O acetic anhydride AcOH acetic acid AFM atomic force microscopy AIBN 2,2’-azobisisobutyro nitrile Al anodised aluminium Al2O3 aluminium oxide ATRP atom-transfer radical polymerisation BMA butyl methacrylate BPO benzoyl peroxide C carbon CaH2 calcium hydrate CDCl3 chloroform CFU colony forming unit CH3COONa sodium acetate CHCl3 chloroform d doublet (NMR) D2O deuterium oxide DMA dodecyl methacrylate DMAPA N,N-dimethylamino propyl amine DMF dimethyl formamide DMSO dimethyl sulfoxide DSC differential scanning calorimetry EA elemental analysis EGMP ethylene glycol methacrylate phosphate Eq. equation Et3N triethylamine EtAc ethyl acetate EtOH ethanol F8H2MA 1H,1H,2H2H-perfluorodecyl methacylate fl.p. flesh point Freon-113 1,1,2-Trichlor-1,2,2-trifluorethan FT-IR fourier transform infrared spectroscopy GPC gel permeation chromatography H hydrogen h hour H3PO4 phosphoric acid HCl hydrochloric acid HD hexadecane HEMA 1-hydroxyethyl methacrylate HFX 1,3-Bis(trifluoromethyl)benzene

List of abbreviations

Hg mercury Hz hertz I iodine i-BuAc iso-butyl acetate IR infrared spectroscopy J coupling constant K kelvin KBr potassium bromide L liter m multiplet (NMR) MEHQ hydroquinone monomethyl ether MEK 2-butanone MeOH methanol mg milligram MHz megahertz MIC minimum inhibitory concentration min minute mm millimeter MMA methyl methacrylate Mn number average molecular weight Mw weight average molecular weight MP 4-methyl-1-pentene MSA maleic anhydride Mw/Mn molecular weight distribution (dispersity, Đ) N nitrogen NBu3 tributylamine NH4OH ammonium hydroxide NHex3 trihexylamine nm nanometer NN non-identified NOct3 trioctylamine NPr3 tripropylamine O oxygen PBS phosphate buffered saline Đ dispersity ppm parts per million PVA poly(vinyl alcohol) q quartet QAS quaternary ammonium salt RFMA perfluoalkyl methacrylate RH alkyl moiety RHMA alkyl methacrylate RI refractive index Rp reaction rate constant RT room temperature s singulett (NMR) SDS sodium dodecyl sulfate t triplet (MNR) tBMA tert-butyl methacrylate Tg glass transition temperature

List of abbreviations

TGA thermogravimetric analysis THF tetrahydrofuran TMSC trimethylsilyl chloride UV ultraviolet vol volume W watt wt% weight percent

vii

Summary

This thesis deals with the synthesis, characterization and properties of copolymers containing maleic anhydride and fluorinated building blocks prepared by means of free radical copolymerization. Free radical polymerization in binary or ternary systems usually leads to a blend of polymer chains with different composition as a consequence of different monomer reactivity, hence the most reactive monomers are consumed first, and consequently the polymer is enriched in monomers of lower reactivity. This effect is even stronger, when monomers which cannot undergo homopolymerization are used. The preferred route to overcome this problem is to feed continuously the reaction mixture with monomers at the same rate at which they are consumed. In Chapter 3 copolymerization of maleic anhydride, butyl-methacrylate and 1H,1H,2H,2H-perfluorodecylmethacrylate is described. The kinetics of the copolymerization of F8H2MA/MSA, BMA/MSA, and F8H2MA/BMA have been extensively studied under well-defined reaction conditions: the determined copolymerization parameters were rF8H2MA = 4.9, rMSA = 0, rBMA = 8.2, rMSA = 0, and rF8H2MA = 1.02, rBMA = 0.94. The reaction rates at chosen conditions were between Rp=0.47 wt%/min for a monomer mixture BMA/F8H2MA/MSA = 1.75:0.75:7.5 and Rp=0.73 wt%/min for BMA/F8H2MA/MSA = 1:1:1. The determined reaction rates and the composition of the terpolymers were used to perform successfully continuous addition experiments in order to produce a bigger quantity of homogenous terpolymers. The versatility of the method has been proven by using a different set of monomers namely dodecyl methacrylate (DMA), 1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA) and maleic anhydride (MSA). The copolymers were characterized in term of molecular weight and thermal properties. Fluorinated terpolymers P[RFMA-co-RHMA-co-

MSA] (RH = C4H9-, C12H25-, RF- = C10H4F19-) obtained in continuous addition experiments and containing ca. 20 mol% fluorinated side chains can be dissolved in semi polar solvents like tetrahydrofuran, chloroform or ethyl acetate as well as in fluorinated solvents like HFX and

Freon 113 (Chapter 4). On incorporation of dodecyl-side chains (RH = C12H25-) the polymers become also soluble in alkane solvents. Up to 15-20 mol% MSA content is not sufficient to induce water solubility, even in case of carboxylate formation by hydrolysis of the anhydride units. Emulsification of solutions in organic solvents of the terpolymers showed to be unstable; they demix within days. The P[RFMA-co-RHMA-co-MSA] terpolymers were coated on glass from 1 wt% solution in chloroform. The contact angles of water and hexadecane as wetting liquids were measured at 20°C prior and immediately after an annealing step (12h, 120°C) and were found to be equal: 110 ° against water and above 70 ° for hexadecane. This means that

viii

during the film formation process the mobility of fluorinated chains is sufficient to ensure proper orientation of the fluorinated side chains already at room temperature. Such behavior is possible because of relatively low Tg of these copolymers. The surface properties of the coatings obtained on aluminum substrates are comparable with coatings on glass both for water and hexadecane. The properties of the coating obtained on paper are different and the measured values for both wetting liquids are lower than for the other two substrates. This phenomenon can be explained by the fact that the terpolymer solution penetrates the paper and consequently does not form a closed film. In summary RH, RF, MSA–terpolymers of moderate fluorine content are versatile, flexible to handle materials that offer wide screen of application for surface modification. To further increase the adhesion properties of these copolymers - especially to metallic surfaces - phosphoric acid groups were incorporated using ethylene glycol methacrylate phosphate (EGMP) as comonomers (Chapter 5). An obstacle in performing continuous addition polymerization experiments was the fact that due to overlapping signals in the 1H NMR spectra the composition of these copolymers could not be unequivocally determined. However, these materials showed good solubility in aqueous ammonia solution and appropriate surface properties due to the presence of fluorinated building blocks. The surface properties of different polymer composition were investigated after coating from water based formulations on glass substrates and annealing above the glass transition temperature. Coatings with hydrophobicity up to 120 ° were obtained. The relatively low thermal stability of phosphorus containing copolymers implies limitations in an application. The aim to create a synthetic, amphiphilic structure with strong antimicrobial properties comparable with natural toxins, led to copolymers of maleic anhydride with vinyl monomers (Chapter 6). Vinyl monomers and maleic anhydride yield alternating copolymers. This ensures a constant 1:1 ratio of the hydrophobic and cationic part similar to those in leucine/lysine (1:1) peptides (LK-peptides). The choice of 4-methyl-1-pentene as hydrophobic comonomer is based on the similarity of the structure with leucine, while the choice of maleic anhydride leaves ample of space for further design of the cationic/hydrophilic part by means of chemical modification. An alternating copolymer of maleic anhydride and 4-methyl-1-pentene was synthesized by means of free radical copolymerization in the presence of benzoyl peroxide (BPO) as initiator. Modification of the P[MP-alt-MSA] copolymer with diamine to poly[(4- methyl-1-pentene)-alt-(1-(3-N,N-dimethylaminopropyl)maleimide)] was performed as a one pot synthesis in relatively mild conditions (DMF at 120 °C). Poly[(4-methyl-1-pentene)-alt-(1- (3-N,N-dimethylaminopropyl)maleimide)] was converted into a polycationic polymer by

means of alkyl iodide. The modification with methyl iodide yields a yellow colored powder easily soluble in water and polar solvents like DMF and DMSO but insoluble in lower alcohols and less polar solvents. Sequential quaternization with methyl iodide and dodecyl iodide ensures solubility in a wide variety of polar solvents including alcohols and ketons. Modification with dodecyl iodide only leads to polymers soluble in non-polar solvents. Copolymers with quaternary ammonium groups and long alkyl chains were tested for their antimicrobial properties.

Chapter 1

Introduction

Fluoropolymers and fluorochemicals

Fluoropolymers are a family of polymers produced from molecules where hydrogen atoms are replaced by fluorine [1]. The development of fluoropolymers started in 1938 when Dr. Plunket discovered poly(tetrafluoroethylene) (PTFE) and significantly speeded up after its commercialization in 1949 under the brand name Teflon®. In 1967 Food and Drug Administration (FDA) approved PTFE for cookware application and five years later Zonyl® product was approved for use in food packaging [2].

Fluorinated polymers offer a unique combination of outstanding properties like chemical resistance, even to hot and concentrated organic acids and bases, thermal stability in a wide range of temperature (-200 to 260 degC), low dielectric constants, low friction coefficient, flame retardation as well as excellent outdoor durability, low water absorption and very good resistance to oxidation and aging. They found many industrial applications e.g. as friction modifiers in bearings, anti-fouling coatings, membranes, flame retardants and surfactants in extinguishing foams or insulators [1-9].

In 1953 an accidental discovery of stain repellent properties of an experimental compound at company 3M triggered the development of first fluorochemical cleaning agents. This was first time commercialized in 1956 under the brand name Scotchgard® - the most famous fluorinated stain repellent for fabrics [10].

There are two major synthetic routes to obtain fluorinated compounds both for fluorochemicals as well as for monomers other than tetrafluoroethylene (TFE is produced by catalytic pyrolysis of monochlorodifluoromethane):

- In the electrochemical fluorination of linear hydrocarbons in the presence of HF leading to complete substitution of hydrogen atoms to fluorine [2, 8]. The yield of this process towards the desired linear products is on average 70%. The rest is a variety of branched and cyclic isomers. Historically this method was used to produce among others

Chapter 1

perfluorooctane sulfonyl fluoride a precursor for production of fluorinated coatings for paper and textiles [11,12].

- The second main and most relevant process is telomerization [2] of tetrafluoroethylene

(CF2=CF2) and perfluoroethyl iodide (CF3-CF2I) to produce low molecular weight, linear oligomers of fluorinated iodides with the chain length a multiple of 2. These low molecular weight oligomers are useful in synthesis processes as intermediates to e.g. carboxylic acids, alcohols and olefins [13]. The advantage of the process is absence of side reactions leading to branched or cyclic products.

Environmental and health aspects

Because of the confidentiality reasons, changes in regulation and product portfolios the annual production of fluorochemicals and polymers over the decades cannot be accurately quantified. The estimated, cumulative production of fluorochemicals only in the peak years 1970-2002 is estimated to be nearly 100 000 tons. Since 2004 11 000 - 13 000 tons of polymers and coatings are being produced annually [2].

In 1968 Donald Taves as the first one reported presence of two kinds of fluorine in human serum: the inorganic fluorine ion and the fluorine in organic compounds [14] and in 1976 he tentatively identified perfluorooctanoic acid (PFOA) in blood samples. Form that moment fluorochemicals have attracted attention and the environmental and health aspects are widely investigated.

The long-chain fluorochemicals as used for example in the original Scotchgard® formulation are proven to be very persistent and bio-accumulative. The average serum half-life of perfluorohexane sulfonate (PFOS) is over 8 years [15]. Due to the nature of organic fluorocompounds they tend to accumulate in lipid rich tissues e.g. liver and are suspected to enhance liver cancer and cirrhosis [16]. PFOA and PFOS accumulated in lipid rich tissues of the brain are having potentially negative effects on cognitive and motor functions [17].

Due to the increased awareness of ubiquitous contamination of environment and of human beings with organic fluorocompounds, programs to reduce consumption of those (as well as searching for alternatives) started.

Chapter 1

Because there is no evidence of other than anthropogenic origin of organic fluorochemicals [18] a retrospective monitoring of PFOA and PFOS in human plasma archived by the German Environmetal Specimen Bank was done. For PFOS sharp increase of concentrations in 1980s, stabilization in 1990s and steady decrease of those concentration between 2001 and 2010 was observed [19].

Driven by concerns about the adverse environmental and health effect, since 2000, major manufacturers have voluntarily discontinued or committed to phase out long perfluoroalkyl chemicals [20]. Since 2009 PFOS is being listed by Stockholm Convention as Persistent Organic Pollutants and therefore restricted for production and use. Since 2013 PFOA

(and its derivatives) as well as C11-C14 perfluorocarboxylic acids are included in the Candidate List of Substances of Very High Concern [21]. As a result of industry and regulatory actions, a transition started to replace long chain perfluorochemicals with alternatives which are of less environmental and health concern.

Based on systematic studies on biopersistence of fluorochemicals done by 3M it was found that carboxylic acid and sulfonic acids with four or less perfluorinated carbon atoms do not bioaccumulate and are not biopersistent [22] although they easily spread in the environment [23]. The recent toxicity study of perfluoroalkyl acids towards Daphnia Magna showed that acute toxicity decreased with decreasing carbon chain length. In chronic toxicity test there was not effect of increased exposure time [24]. The reported toxic effect concentrations were at least a factor 106 higher than the highest reported concentrations in riverine samples [25,26].

Also very recently published papers described perfluorinated amphiphilic copolymers as non-cytotoxic and biocompatible materials for protein conjugation [27,28].

Based on the reviewed literature one can conclude that the environmental hazards regarding the florochemicals have been sufficiently proven (mainly due to their biopersistance) however the toxicity and adverse health effects still require thorough investigation.

In case of polymers, which can be a base for new type of stain repellent coatings, alternative monomers to long chain perfluorinated monomers has been patented by Guo et.al. [29]. The invention describes perfluoroalkyl acrylates and methacrylates and their polymers with side chains which can be described by the general formula –(CH2)n-CF2-CF2-CF2-

Chapter 1

CF3where n=2-4. The corresponding precursors for monomer synthesis can be obtained by means of telomerization method [2].

Approach and content of this thesis

In order to remain compliant with more and more strict environmental regulations and utilize outstanding properties of fluorinated coatings, following approach in synthesis of novel fluoropolymers described in this thesis was taken:

- homogenous composition of the polymer chains

- relatively low fluorine content

- presence of adhesion promoting moieties

- self-stratifying properties of the obtained coating due to presence of both perfluorinated and non-fluorinated alkyl chains

Stratification is a spontaneous formation of separate layers of inherently incompatible components (e.g. fluorinated and non-fluorinated) in polymer films when molecular mobility is promoted by a temperature above the glass transition temperature or in the presence of a solvent [30,31].

The goal of this thesis is to demonstrate that terpolymers which contain fluorinated methacrylate, non-fluorinated methacrylate and maleic anhydride as adhesive moiety and which are of homogenuous composition can be synthesized by means of continuous addition copolymerization.

In the following an outline of the contents of this thesis is given:

Chapter 2 reviews the literature concerning the objectives given in this thesis

Chapter 3 reports the synthesis of novel ternary copolymers containing maleic anhydride, alkyl methacrylate and perfluorinated methacrylate with very well defined composition. Such terpolymers were obtained via radical copolymerization with continuous feeding of monomers to the reaction mixture.

Chapter 1

Chapter 4 concerns solubility, emulsification and surface properties of films obtained from poly[RH-co-RF-co-MSA] which are synthesized via free radical copolymerization with continuous feeding of the monomers.

Chapter 5 describes the synthesis, characterization as well as the surface properties of water soluble ternary copolymers containing maleic anhydride, perfluorinated methacrylate and methacrylate with phosphoric acid moieties. These copolymers were obtained via radical copolymerization with continuous feeding of monomers to the reaction mixture.

Chapter 6 presents the synthesis and antimicrobial properties of quaternary ammonium salt containing copolymers based on maleic anhydride and 4-methyl-1-pentene.

References [1] G. Odian, Principles of Polymerization, 4th edition, John Wiley & Sons, Hoboken, New Jersey, 2004 [2] A.B. Lindstrom, M.J. Strynar, E.L. Libelo, Environ. Sci. Technol., 2011, 45, 7954- 7961 [3] E. Kissa; Fluorinated Surfactants and Repellents, 2nd edition, M. Dekker, ed. New York, 2001. [4] R.F. Brady Jr.; In Encyclopedia of Polymer Science and Technology, 2nd ed. H. Mark et all (eds), John Wiley & Sons, New York, 1986 [5] J. Yu*, B. Yi, D. Xing, F. Liu, Z. Shao, Y. Fu, H. Zhan; Phys. Chem. Chem. Phys., 2003, 5, 611-615 [6] V. Arcella, A. Ghielmi, G. Tommasi, Annals of the New York Academy of Sciences, 2003, 984: 226–244 [7] J. Scheirs, Modern Fluoropolymers: High Performance Polymers for Divers Applications, John Wiley & Sons, New York, 1997. [8] G. Hougham, K. Johns, P. E. Cassidy, T. Davidson, Fluoropolymers 1: Synthesis, Plenum Press New York, 1999. [9] G. Hougham, K. Johns, P. E. Cassidy, T. Davidson, Fluoropolymers 2: Properties, Plenum Press, New York, 1999. [10] R. Renner, Environ. Sci. Technol., 2006, 40, 12–13 [11] A.G. Paul, K.C. Jones, A.J. Sweetman, Environ. Sci. Technol., 2009, 43, 386–392 [12] G.W. Olsen, H.Y. Huang, K.J. Helzlsouer, K.J. Hansen, J.L. Butenhoff, J.H. Mandel, Environ. Health Perspect., 2005, 113, 539-545 [13] C.G. Krespan, V.A. Petrov, US Patent 5908966A, 1997 [14] D.R. Taves, Nature, 1968, 217, 1050 – 1051 [15] C. Bjerregaard-Olesen, C.C. Bach, M. Long, M. Ghisari, R. Bossi, B.H. Bech, E.A. Nohr, T.B. Henriksen, J. Olsen, E.C. Bonefeld-Jørgensen, Envirion. Int., 2016, 91, 14- 21 [16] L.W.Y. Yeung, K.S. Guruge, S. Taniyasu, N. Yamashita, P.W. Angus, C.B. Herath, Ecotox. Environ. Safety, 2013, 96, 139-146 [17] K.E. Pedersen, N. Basu, R. Letcher, A.K. Greaves, C. Sonne, R. Dietz, B. Styrishave, Environ. Research., 2015, 138, 22-31 [18] C.A. Moody, J.A. Field, Environ. Sci. Technol., 1999, 33, 2800-2806

Chapter 1

[19] C. Schroeter-Kermani, J. Mueller, H. Juerling, A. Conrad, C. Schulte, International J. Hyg. Environ. Health, 2013, 216, 633– 640 [20] I.T. Cousins, R. Vestergren, Z. Wang, M. Scheringer, M.S. McLachlan, Environ. Int., 2016, 94, 331–340 [21] Z. Wang, I.T. Cousins, M. Scheringer, K. Hungerbuehler, Environ. Int., 2015, 75, 172–179 [22] J. Guo, P. Resnick, K. Efimenko, J. Genzer, J. M. DeSimone, Ind. Eng. Chem. Res. 2008, 47, 502-508 [23] L.Vierke, A. Möller, S. Klitzke, Environmental Pollution, 2014, 186,7-13 [24] S.H. Barmeto, J.M. Stel, M. van Doorn, C. Eschauzier, P. de Voogt, M.H.S. Kraak, Environ. Pollut., 2015, 198, 47-53 [25] M.S. McLachlan, K.E. Holmstrom, M. reth, U. Berger, Environ. Sci. Technol., 2007, 41, 7260-7265 [26] A. Moeller, L. Ahrens, R. Sturm, J. Vesterveld, F. vd Wielen, R. Ebinghaus, P. de Voogt, Environ. Pollut., 2010, 158, 3243-3250 [27] Y. Koda, T. Terashima, M. Sawamoto, H.D. Maynard, Polym. Chem., 2015, 6, 240- 247 [28] Y. Koda, T. Terashima, H.D. Maynard, M. Sawamoto, Polym. Chem., 2016, 7, 6694- 6698 [29] J. Guo, US Patent 20090053462, 2009 [30] V.V. Verkholantsev, Progress in Organic Coatings, 1985, 13, 71-96 [31] Th. Mezger, Progress in Organic Coatings, 1992, 20, 353-36

Chapter 2

Literature Review

1. Radical batch copolymerization. Copolymers of homogenous composition.

The term “copolymerization” means any polymerization of the mixture of at least two different monomers. The composition of the polymer chains is time dependent and a function of the relative rate of incorporation of monomers into the growing chain and on the monomer ratio in the feed. As consequence the monomer ratio changes with time; the concentration of the less reactive monomer increases. When two monomers M1 and M2 are copolymerized, the relationship between the composition of the copolymer and the composition of the monomer in the feed is given by equation 1 [1]: dm M (r M + M ) 1 = 1 1 1 2 (1) dm2 M 2 (r2 M 2 + M1 )

where M1 and M2 are moles of monomer 1 and monomer 2 in the feed, m1 and m2 are moles of monomer 1 and monomer 2 incorporated in the copolymer. The ability of a specific monomer to add to the growing polymeric chain depends on the individual type of radical (identical or different monomer) and is decreased or enhanced by the tendency of the monomer to either homopolymerize or copolymerize. The ratio of homopolymerization to copolymerization ability is called “copolymerization parameter” or

“reactivity ratio” and is commonly known as “r parameter”. So in the equation r1 and r2 are the reactivity ratios of monomer 1 and monomer 2. If two monomers are copolymerized the system is called binary copolymerization. If in a binary system the ability of the monomer to form a copolymer is much higher than the formation of the homopolymer the “r” parameters (r1= k11/k12 and r2 = k22/k21) are r1=r2=0.

(where kij is a reaction rate constant of polymerization of monomer “ï” with monomer “j” ). In this case, both monomers are incorporated in the polymeric chain in strictly alternating manner and the composition of the copolymer is completely independent on the composition of the monomer mixture. Such a behavior is usually shown by pairs of monomers like maleic anhydride and styrene [2] or maleic anhydride and other vinyl monomers [3]. If both r1=1 and

Chapter 2

r2=1 it is called ideal statistical copolymerization and the composition of the copolymer is always identical with the composition of the monomer mixture.

The combinations of r1<1 and r2<1 or r1>1 and r2>1 as for example for styrene (M1) and methyl methacrylate (M2) [1,4] yields copolymers with composition identical with the composition in the feed only for a certain ratio of the monomers and is called azeotropic copolymerization. The point in which the copolymerization line intersects the ideal statistic copolymerization line is called azeotropic point. For any other monomer ratio one of the monomers will be consumed with higher rate and this will cause the change in the actual composition of the monomer mixture. This change will lead to the inhomogeneity of the copolymer composition. The direction of the compositional drift is depicted by arrows in Figure 1 [5].

If r1<1 and r2>1 as for example for maleic anhydride (M1) and methyl methacrylate (M2) or r1>1 and r2<1 [4,6] the copolymerization curve is placed below the ideal statistical line. In this case one of the monomers shows the tendency to form a homopolymer rather than a copolymer while the second one forms preferably copolymers. This means that one of the monomers is consumed with significantly higher rate than the other one. This enrichment of the reaction mixture in less reactive monomer will in consequence lead to the drift in the composition of the copolymer.

1,0 1,0

0,9 0,9

r =1, r =1 0,8 1 2 0,8

0,7 0,7

0,6 r =0, r =0 0,6

1 2 1

F 0,5 0,5

0,4 r <1, r <1 A 0,4 1 2 0,3 0,3

0,2 r <1, r >1 0,2 1 2 0,1 0,1

0,0 0,0 0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

Figure 1. Different types of copolymerization diagrams (alternating copolymerization r1=r2=0, ideal statistic copolymerization r1=r2=1, azeotropic copolymerization r1<1, r2<1 (A

Chapter 2

= azeotropic point) tendency to form homopolymer r1<1, r2>1. Arrows show direction of the compositional drift (r1, r2 – copolymerization parameters, F1 – molar fraction of monomer 1 in the copolymer, f1 – molar fraction of monomer 1 in the feed).

If the uniform composition of the copolymer is of absolute importance one of the following solutions need to be chosen [5]: - Select a combination of monomers that undergo ideal statistical copolymerization - Select a system that undergoes alternating copolymerization. In this case the structure is always limited to the 1:1 ratio of the monomers. - Select an azeotropic system and carry out the copolymerization only at the composition of the azeotropic point. This solution is also limited to only one composition. - Perform the copolymerization at low conversion only, so the compositional drift of the monomer mixture is negligible. This applies to any combination of monomers but results in wasting 90% of the monomer mixture. - Perform continuous addition copolymerization. In this method monomers are added to the reaction mixture with the same rate they are consumed.

The last option offers much wider possibilities to form copolymers with uniform composition. This method has been investigated in binary systems in the past [7].

2. Fluorinated polymers

Fluorinated polymers offer a unique combination of outstanding properties like chemical resistance, even to hot and concentrated organic acids and bases, thermal stability up to 260°C, low dielectric constants, low friction coefficient, flame retardation as well as excellent outdoor durability, low water absorption and very good resistance to oxidation / aging and lower surface tension because of the low polarizability and the strong electronegativity of the fluorine atom, its small van der Waals radius (1.32Å), and the strong C-F bond (485 kJ mol−1). Fluorinated polymers found many industrial applications e.g. as friction modifiers in bearings, anti-fouling coatings, membranes, flame retardants and surfactants in extinguishing foams or insulators [8- 17].

Two main distinct routes are used for the synthesis of fluorinated polymers. In the first route the fluorinated polymer is synthesized by the polymerization of fluorine containing monomers. For example polytetrafluoroethylene (PTFE) is synthesized from tetrafluoroethylene and is well

Chapter 2 known under the trade name Teflon®. Those types of polymers are linear and contain the fluorine directly in the main chain. Fluoropolymers can also be prepared by means of polymer modification (second route) that incorporates the fluorine atoms or moieties into non-fluorinated polymer [18]. This route can lead to polymers with terminal or side chain fluorinated moieties [19]. Heptadecafluorodecane end caped poly(ethylene glycol) (PEG 35000) for example was prepared by reaction of heptadecafluorodecane monoisocyanate with PEG-diol. In aqueous solution the fluorocarbon moieties strongly associate to form micelle like structures bridged by PEG chains resulting in a three dimensional network [20]. Polyethylene synthesized by means of ring-opening metathesis polymerization of cyclododecene in the presence of a linear fluorinated olefin

Rf(CH2)6CH=CH(CH2)6Rf with Rf = C10F21 or C4F9 followed by hydrogenation of double bonds in the backbone gives fluoroalkyl-ended polyethylene [21]. Poly(methylmethacrylate) and poly(acrylic acid) are common polymers which contain reactive groups which are often base polymers for modification by transesterification with perfluoroalcohols or by reaction with sulfur tetrafluoride (SF4) and hydrogen fluoride (HF) yielding polymers with fluorinated side chains [22]. Those types of polymers can be directly synthesized by polymerization of commercially available methacrylates and acrylates [23]. Another method of synthesis is modification of maleic anhydride copolymers by means of fluorinated amines [24] or alcohols [25].

3. Maleic anhydride

Maleic anhydride (MAH) is chemically 2,5-furandione [26] and it is also known by other names like toxilic acid [27]. It is a white hygroscopic solid which forms crystalline needles. It was first time produced by dehydration of maleic acid. The production of maleic anhydride finds many end uses: manufacturing of unsaturated polyesters, coatings, lubricants, reactive plasticizers [28], agricultural chemicals [29] and copolymers. It is also a base for manufacturing of other chemical intermediates like fumaric acid, succinic acid [30] or malic acid [31]. The unsaturation in maleic anhydride is activated due to conjugation with carboxyl groups as a result MAH undergoes not only reactions typical for olefinic unsaturation but enters also in reactions like nucleophilic addition. The electrophilic character of the unsaturation due to the conjugation plays significant role in a number of reactions e.g. Diels-Alder and some photoreactions. MAH and its acidic forms undergoes a number of specific reactions like hydrogenation [32-35] hydroformylation [36-38], hydration and dehydration [31, 39-42],

Chapter 2 addition of alcohols [43-45], addition of amines [46,47], sulfonation [48], halogenation [37,49- 52], oxidation [53-55] and free radial reactions [56]. Maleic anhydride like most of anhydrides undergoes typical reactions of hydrolysis (see Scheme 1) [57], peracid formation [58], esterification [59-61], reaction with amines to yield amic acid and imides [62-65], acylation [66,67], alkylation [68,69], acid chloride formation [70,71], formation of metal compounds [72-74].

ROH HO

O O O O O RO

R- H, alkyl Scheme 1. Alcoholysis (hydrolysis) of maleic anhydride.

It is worth to name the [4+2] cycloaddition reaction (Scheme 2) better known as Diels- Alder reaction reported in 1928 by Otto Diels and Kurt Alder [75].

O O

O + O

O O

Scheme 2. Diels-Alder addition of maleic anhydride and butadiene.

The role played by this reaction in organic chemistry cannot be overestimated. The efforts of Diels and Alder were acknowledged by a Nobel Prize for chemistry in 1950.

Homopolymerization of maleic anhydride In the large family of monomers used to prepare polymeric materials maleic anhydride occupies a unique position. The presence of both reactive unsaturation and the anhydride structure enables the monomer to be used in addition polymerization (chain growth reaction) as well as in polycondensation (step growth) reactions. Until early 1960s it was believed that MAH is unable to form homopolymers [76-78]. It was first time reported in 1961 that maleic anhydride had been homopolymerized [79]. Later studies showed that MAH can be

Chapter 2 homopolymerized using UV and γ radiation, in the presence of free-radical initiators, pyridine bases, electrochemically [80,81] and undergoes spontaneous homopolymerization when heated under high pressure > 20 000 bar [82]. First attempts to polymerize maleic anhydride in the solid state by means of γ irradiation failed [83]. It was discovered that the irradiation of molten MAH or of acetic anhydride solution results in a homopolymer with 87 % yield (see Table 1) [84].

Table 1. Radiation dose and conversion of MAH during polymerization in acetic anhydride [59]. Radiation dose Conversion (Mrads) (%) 1.56 2.69 4.68 15.11 9.36 30.68 18.36 55.32 28.08 61.65 39.00 87.08

UV irradiation of dioxane solutions of MAH gives oligomers which contain four monomer unites and at the end a dioxane ring beside a 2+2-dimer (see Scheme 3) [85,86].

O O O O

UV O + O O O O H

O O O O O O O 4 Scheme 3. Product of UV irradiation of MAH in dioxane.

Based on the observation that MAH solutions in dioxane show an increased electrical conductivity during exposure to UV-light it was assumed that ionic intermediates are formed [87]. It is also remarkable that the UV induced homopolymerization of MAH in aromatic solvents is not inhibited by hydroquinone and t-Bu-hydroquinone [88]. Homopolymerization of MAH with catalytic amounts of tertiary amines or pyridine [89,90] yields dark colour products. The reaction occurs with the evolution of carbon dioxide and the obtained material is claimed to be an acrylic acid polymer.

Chapter 2

More recently homopolymerization of MAH using plasma has been reported [91]. Highly reactive films were obtained which can form polyelectrolyte coatings after swelling in water. MAH can also homopolymerize in the presence of free-radical initiators. In the molten state in the presence of BPO it yields a resinous product with low degree of polymerization [92,93]. The attempt to conduct the reaction in different solvents led to very low conversions and yields mainly dimers [94]. The same low yield has been observed when AIBN or lauryl peroxide was employed even in a melt [79,84]. Good results were obtained using percarbonates in benzene with reported yields of 85% of crude material [95]. The homopolymerization of MAH induced by free-radical initiators is difficult under conditions commonly used for vinyl monomers. High conversions of the monomer can be achieved only with large quantities of the initiator or when the initiator is added slowly to the molten anhydride [27]. Poly(maleic anhydride) prepared by means of above described methods is a low molecular weight, white material soluble in water and polar organic solvents. There are possible applications of this polymer in water treatment, detergents, dispersing agents and many others.

Copolymerization of maleic anhydride Introduction of MAH into a polymeric chain will change the physical and chemical properties of polymers. It provides increased polarity, rigidity, thermal properties and will give additional functionality. The presence of succinic anhydride moieties in the polymer will enhance adhesion, hydrophilicity, dyeability and heat distortion. These properties can additionally be tuned by the choice of other monomers in the composition and result in a material with wide range of applications in adhesives, resins, dispersants and emulsifiers, inks and coatings, paper, textiles and fiber sizing, detergent, lubricant and fuel additives, flocculants, enzyme and drug carriers and many others. Additionally the chemical modification of the anhydride unit gives the possibility of fine tuning of the properties of such copolymers. As mentioned above, maleic anhydride does not easily homopolymerize, however it is known to copolymerize readily with a large variety of monomers. In particular with electron donors or monomers with opposite polarity, maleic anhydride yields alternating copolymers with e.g. alkenes, allyl monomers, vinyl ethers, styrenes etc. as depicted in Scheme 4. In fact many 1,2-substituted monomers like stilbene [96] or cyclic olefins which do not homopolymerize will form copolymers with MAH. The alternating copolymerization of MAH with mentioned monomers is in many cases explained by a formation of charge transfer

Chapter 2 complexes [97,98] (CTC) between the monomers [99,100] or between monomer and the propagating radical [101,102] and the complex plays a role in initiation and propagation reaction.

O *

O O O O n

R * R R * O O O R O O O n

O O O n

Scheme 4. Alternating copolymerization of maleic anhydride.

Generally a large amount of initiator is required to homopolymerize both MAH and allyl compounds. In this process only low molecular weight materials can be obtained. A detailed study on the copolymerization of this kind of monomers initiated by radicals [103] showed that copolymerization of allyl acetate with maleic anhydride is more rapid at 30°C than homopolymerization of these monomers at 80°C. At the initiator concentration of 4.58 wt%, a molecular weight of 40 000 was obtained. With such an initiator concentration only low molecular weight homopolymers are obtained [103]. Allyl substituted aromatics such as allylbenzene or allylphenol (see Table 2) are also known to form alternating copolymers both in bulk and in solution with high yields dependent on the temperature and initiator concentration.

Chapter 2

Table 2. Alternating copolymers of MAH with allyl compounds Monomer Polymerization conditions Ref. Allyl acetate BPO, at 80°C 103 Allyl acetate AIBN, at 60°C in benzene 104, 105 Acetonylallylphosphonate AIBN, at 60°C in benzene 106 Allylbenzene AIBN, at 59°C 107, 108 Allylbenzene AIBN, at 60°C in DMF 109 2-allylphenol AIBN, at 80°C in dichloromethane 110 allyl alcohol AIBN, at 70°C in dioxane 111 allyl glycidyl ether BPO, at 60°C 112,113

Considerable attention was paid to the copolymerization of maleic anhydride with olefins. Lower α-olefins copolymerize more easily with MAH than higher α-olefins to yield copolymers with equimolar composition. Many of such copolymers are commercially available from olefins up to C-18. Ethylene copolymerizes readily with MAH in a variety of solvents and using different initiators [114] or γ-irradiation [115]. For example for copolymerization carried out in benzene with BPO as an initiator, an alternating copolymer was obtained for ethylene concentrations between 0.5 and 60 mol%. The maximum rate of copolymerization was found at equimolar ratio of the monomers [116].The experiments with copolymers prepared via γ-irradiation [117] showed that the structure of the obtained material is independent on the initiation method. In many papers the copolymerization of MAH with higher α-olefins from propylene through different butenes, pentanes to 1-octane are described [114,117-119]. In all these investigations alternating copolymers were obtained. Lower olefins gave good yields and relatively high molecular weight copolymers. The reactivity of the olefins toward maleic anhydride can be put in following order [27]:

isobutene > propylene > ethylene> 1-butene

Higher, C6-C8 olefins gave only low molecular weight materials with poor yield. It is remarkable that there are some exceptions from this rule and 2-methyl-1-pentene is almost as reactive as propylene. The yield and molecular weight of copolymers is strictly dependent on the reaction conditions. The type of initiator, solvent, temperature and in case of lower olefins the pressure is of essential influence on the copolymer formation. As an example the copolymerization of MAH with ethylene can be given [114]. Relatively long polymer chains are quantitatively formed in MEK in the presence of BPO and under the pressure of 10 bars, while under the pressure of 2 bars the yield dropped to only 12 %. The detailed investigation on the pressure effect on the copolymerization of different olefins with maleic anhydride was

Chapter 2 studied by Kellou [115,120,121]. The copolymerization was carried out in chloroform in the presence of lauroyl peroxide in the temperature range of 70-90 °C, the pressure varied from 1 to 3000 bars. It was found that pressure has considerable influence on the rate of copolymerization. In general it is two times higher at 1000 bar and ten times higher at 3000 bar comparing to the rate at 1 bar.The pressure hardly influences the composition of the copolymer that remains alternating.

Very interesting can be the fact that modification of MAH-olefin copolymer can lead to another copolymer but of the structure that cannot be obtain by means of conventional polymerization. Ethylene-alt-MAH copolymer after an esterification with methanol [122, 123] yields head-to-head poly(methyl acrylate) that cannot be prepared by conventional homopolymerization of methyl acrylate (see Scheme 5).

Analogous is the route to poly(methyl crotonate) via modification of 2-butene-alt-MAH with methanol. Also in this case the head-to-head structure has a considerably higher Tg (107 °C) than conventional head-to-tail structure (80 °C) [124].

R copolymerization H H RHC CHR +  C CH CH C  n O O O R C C O O O

R CO2CH3 CO2CH3 esterification H H H H H  C C C C C C C C  H H H

CO2CH3 R R CO2CH3 R

where: R=H head-to-head polymethyl acrylate R=CH3 head-to-head polymethyl crotonate Scheme 5. Head-to-head copolymer obtained via esterification of alternating copolymer of MAH and olefin.

Butadiene, isoprene and other diolefins are known to undergo free-radical copolymerization with MAH [125-127]. In the absence of radical initiators, well known Diels-

Chapter 2

Alder reaction takes place. In all cases the copolymers of MAH with diolefins are equimolar regardless of the monomer feed composition. The yields and glass transition temperature of chosen MAH – diene copolymers are summarized in Table 3.

Table 3. Copolymers of dienes with MAH [125] Diene Yield [%] Tg [°C] Butadiene 9 135-145 Isoprene 42 145-152 1,3-Pentadiene 41 85-90 2,3-Dimethylbutadiene 24 120-125 2,4-Hexadiene 46 175-178 1-Methoxybutadiene 41 168

A variety of cyclic olefins also have been copolymerized with maleic anhydride to yield alternating copolymers. 1-Methylcyclopropene copolymerizes spontaneously with MAH at room temperature [128]. The availability of electrons in the cyclopropane promotes copolymerization with electron acceptors like MAH even without radical initiators. The reaction times, yields and glass transition temperatures of chosen MAH – cyclo-olefin copolymers are summarized in Table 4.

Table 4. Copolymers of cyclic olefins with MAH [129].

Olefin Reaction time [h] Yield [%] Tg [°C] Cyclobutene 0.32 41.8 180-190 Cyclopentene 1.25 21 190-195 Cyclohexene 23 0.8 170-185 Cycloheptene 19 1.0 240-245 Cyclooctene 14 1.4 250-255

High electron deficiency of maleic anhydride makes it a perfect candidate for alternating copolymerization with electron-rich styrene monomers [130,131]. Free radical initiated copolymerization of MAH and styrene is recognized as one of the best studied examples of alternating radical copolymerization. The styrene MAH copolymerization is usually carried out in the range of temperature between 40 and 100 °C in the presence of peroxide initiators. It is known that an equimolar mixture of these two monomers will undergo spontaneous copolymerization at 80 °C yielding an alternating copolymer [132,133]. If the copolymerization is conducted at temperatures above 80 °C the random structure of the copolymer is increasing with increasing temperature [134]. Copolymerization in bulk at 180 °C or in decalin at 130 °C yields only random copolymers. The composition of the copolymer may also depend on the monomer feed and may differ from the alternating. Equimolar ratios or excess of MAH always

Chapter 2 result in alternating copolymers up to the temperature of 120 °C in case AIBN is employed as an initiator [135]. Various substituted styrenes also undergo alternating copolymerization with maleic anhydride. α-Methylstyrene will copolymerize by irradiation or by initiation with typical free- radical initiators. Copolymerization may occur even without added initiator [135,136]. Stilbene and 1,1-diphenylethylene were also found to yield alternating copolymers with MAH [137]. The monomer pair stilbene-MAH was the first described case where two non- homopolymerizable monomers could be combined to form an alternating copolymer [2]. As it was already mentioned MAH can form with styrene not only alternating but also random copolymers. This happens when styrene is in excess in the monomer mixture at the beginning of the copolymerization [138,139] or when certain reaction conditions are chosen [138]. Vinyl chloride [140,141], vinyl esters [139,141], acrylonitrile [142,143], acrylamide [144], acrolein [145,146] and other monomers copolymerize with maleic anhydride giving random copolymers. Reaction scheme of MAH with chosen monomers yielding random copolymers are depicted in Scheme 6. Copolymerization of maleic anhydride with acrylic [147-149] and methacrylic [150,151] esters is usually conducted in solution in the presence of BPO or AIBN as an initiator. UV initiated copolymerization proceed well in solution even without a photoinitiator. Typical copolymerization conditions produce random copolymers. The copolymerization of methyl methacrylate with MAH shows unusual behaviour: the reaction rate measured at 60°C in benzene increased with the increase of the molar fraction of MAH in the monomer mixture [149]. This is completely opposite to the normal copolymerization behaviour and what seems to be related to the termination reaction. It was found that the maximum rate of copolymerization occurred in 1:1 monomer mixtures [138].

Chapter 2

* * O RO m O O O n O

CN OR CN * *

m O O O O O O n

H * * O H m

O O O n Scheme 6. Random copolymerization of maleic anhydride.

As a rule MAH and olefins form only equimolar copolymers, however under certain conditions random copolymers with lower MAH content (< 50 mol%) can be produced. For example such a copolymer can be obtained via radical polymerization of ethylene with 3 % of MAH under high pressure [152,153]. Maleic anhydride has been copolymerized with a number of monomer pairs in ternary systems. The monomer pair styrene-maleic anhydride was combined with a variety of vinyl monomers as divinylbenzene [154], vinyl acetate [155], acrylonitrile [156], methacrylates [157] and acrylates [158].

Maleic anhydride copolymer analysis Maleic anhydride copolymers are characterized by means of standard spectroscopic methods like nuclear magnetic resonance (NMR) or infrared spectroscopy (IR). Characteristic IR absorption bands are two regions at 1920-1770 cm-1 for anhydrides and 1770-1665 cm-1 for esters. The NMR spectra show signals of the anhydride ring protons at 3.25 ppm. The chemical shift changes after ring opening (upon hydrolysis or esterification) to 2.8 ppm [27]. Potentiometric titration techniques provide excellent possibility for determining MAH content both in the polymer as well as of free MAH in the presence of polymers [159].

Chapter 2

4. Antimicrobial Quaternary Ammonium Salt Polymers (QAS)

The threat of infection with pathogenic microorganism is of great concern in the field of health care (hospitals, dental care) due to the possible contamination of medical equipment, furniture and all kind of health care products. The health care is possibly the most exposed area due to the often presence of drug resistant species but it is not the only field of life struggling with the presence of microorganism like bacteria, viruses and fungi. The food industry: food processing, packaging and storing, water treatment, household sanitation and many others [160,161]. Infectious diseases supposed to be the major cause of death worldwide [162]. The most common substances used to fight microorganisms are low molecular weight compounds which are usually applied for water treatment, as food additives and for treatment of surfaces [163]. The low molecular weight substances have very often limited efficiency due to volatility or exhibit residual toxicity if need to be applied at efficient concentrations [164]. One of the possible methods to ensure long time effectiveness and limited release of the low molecular weight biocides to the surrounding is their incorporation into the polymer matrix [162]. The wide utilization of polymers especially for food contact and biomedical applications creates the need to add the antimicrobial functionality to the polymer itself without the risk of the release of toxic substance to the food or body. Establishing the biocidal function as an intrinsic feature of the polymeric material has been widely investigated over past decades. There is a large variety of antimicrobial polymers described in literature. All of them should fulfill some general characteristics: should be easily and inexpensively synthesized, should exhibit stability at storage and application conditions, should not release toxic compounds, should not be harmful or irritating to people (in general non-toxic towards mammalian cells), possibly regenerated upon loss of activity and act rapidly towards a broad spectrum of microorganisms [165]. There are many factors which can influence the activity of antimicrobial polymers like molecular weight, spacer length between active moiety and polymeric chain, hydrophilic/hydrophobic balance, the type of counterion (in case of salts) and the nature of the active group itself. The type of microorganism also plays substantial role, for example the morphological differences between Gram-positive and Gram-negative bacteria (see Figure 2) can already influence the activity of the antimicrobial polymer. In case of Gram-negative bacteria e.g. E. coli, due to the additional protective membrane, higher minimum inhibition concentration (MIC) values are reported. Generally researchers compare the antimicrobial activity of the investigated polymers with antimicrobial peptides which are natural and usually extremely efficient antibacterial polymers.

Chapter 2

Figure 2. Schematic difference between Gram-positive (left) and Gram-negative (right) bacteria.

Cationic, amphiphilic polymers are currently being used as antimicrobial agents that disrupt biomembranes, although their mechanism(s) remain poorly understood [166,167]. The commonly accepted bactericidal mechanism of amphiphilic polycations schematically consist on several steps: (1) adsorption of the polymer onto the bacterial cell surfaces, (2) diffusion through the cell wall, (3) binding to the cytoplasmic membrane, (4) disruption of the cytoplasmic membrane, (5) release of K+ ions and constituents of the cytoplasmic membrane, (6) death of the cell [168,169]. According to Tashiro, two most important factors which determine the antimicrobial efficiency of amphiphilic disinfectants are electrostatic interaction between the polymer and the cell and the hydrophobic moiety. The first feature enables binding to the cell by the electrostatic interaction and the second one (after diffusion through the cell wall) binding and penetrating the cytoplasmic membrane (see Figure 3). Recently reported study on the polymer interaction with the cell membrane suggests that the mechanism is not always related to the total disruption of the membrane. The proposed alternative mechanism consists on formation of pores in the membrane through reorganization of the lipid bilayer and the cell death is caused by the ion gradient rather than by the release of cytoplasmic constituents [167]. The polycations with quaternary nitrogen are the most extensively investigated bactericidal macromolecules. The group of QAS polymers can be divided in different subgroups due to their chemical composition and/or architecture. The antimicrobial activity of quaternary ammonium compounds depends very much on the type of the groups attached to the nitrogen atom and the type of counterion [170,171]. Usually there are four groups linked to nitrogen, where one is linking the nitrogen atom to the polymer. The organic substituents can be of different chemical nature: aliphatic, aromatic or heterocyclic. In some cases one of them is chosen to be hydrophobic and able to penetrate the cell wall [172].

Chapter 2

It has been demonstrated by Abel et al. that modified, DABCO containing polymers, quaternized with alkyl chlorides show a significant effect on bacteria growth when the alkyl chloride contains ten or more carbon atoms [173].

Figure 3. The penetration of alkyl chains into the bacterial cell (S. Jiang et al.[169]).

One of the widely investigated groups of QAS is polymer containing heterocyclic structures. Antimicrobial pyridinium copolymers attracted a lot of attention of many researchers. Most of the reports deal with soluble, random copolymers of 4-vinylpiridine (4VP) with styrene [170] as well as block copolymers [174] and crosslinked polymers with divinylbenzene [175,176]. The aromatic ring nitrogen of 4VP has been quaternized with different alkyl bromides [175] and is supposed to kill bacteria on contact. Quaternization of insoluble (crosslinked) 4VP copolymers with different butyl halogens (Cl, Br, I) has been reported by Li [176]. In this case it has been shown that the antimicrobial activity of these insoluble QAS with chlorine and bromine counterion is characterized by irreversible adsorption of the bacteria in a living state. The iodine containing, insoluble polymer also adsorbs the cell but additionally kills it at the same time. The effect on antimicrobial efficacy of different counterions has been well investigated by Sharma et al. [177] by synthesis of poly(4-vinyl 2- hydroxyethyl pyridinium) chloride and subsequent exchange of chlorine with Br-, OH-, SH-, - - NO3 and BF4 . The hydroxyl modified polymers showed highest biocidal activity towards wide range of microorganisms. Kawabata and co-workers reported extensive work on soluble and insoluble pyridinium polymers coagulating and sedimenting microorganisms [178], capturing and killing bacteria and viruses [179] and removing organic pollutants [180].

Chapter 2

Copolymers of 4VP with other monomers like methacrylates have been synthesized and systematically studied [181] (see Scheme 7).

The authors [181] investigated the influence of spatial positioning of the positive charge and the hydrophobic alkyl tail. They were interested whether the antimicrobial activity will differ when the charge and hydrophobic moiety will be located on the same center or would be spatially separated as in the Scheme 8. This aspect seemed to be unexplored before. The authors concluded that for homologous polymers having similar backbone composition and hydrophobic/hydrophilic ratio the spatially separated centers results in higher membrane disrupting ability.

* * * * n n n n Series A + same center O O O O O O N CH3 CH3 CH3 N N I R Precursor for series A R = CH3 - C10H11

* * * * Series B n n n n same center + O O O O O O N R R R N N I CH3 Precursor for series B R = CH3 - C10H11

* Series C * both centers O O

R N

R R = CH3 - C4H9 Scheme 7. Series of 4VP copolymers with methacrylates investigated by Sambhy and coworkers [181]. Series A: n-iodoalkane CH3I. Series B: n-iodoalkane (RI: R=ethyl, propyl, butyl, hexyl, octyl, decyl). Series C: n-iodoalkane CH3I.

Chapter 2

* * * *

m1 m2 n m n versus O O O O

CH3 Bu N I N I N I

Bu CH 3 CH3 Scheme 8. Two series of pyridinium–methyl acrylate copolymers with different spatial positioning (Sambhy et al. [181]).

A completely new methacrylamide monomer with a pyridine group was synthesized and polymerized by Dizman [182] (see Scheme 9).

O O O CH2Cl2 +H2N N TEA N N O 0 degC, 1 h H 25 degC, 11 h

* * * O O * O 1,4-dioxane z R-Br z

AIBN, N 2 HN CH3CN/MeOH HN HN 70 degC, 24 h 60 degC, 48 h

Br N N N R

* * * O y O O * O y x DMF x O R-Br + CH CN/MeOH HN AIBN, N 3 HN HN 2 HN 60 degC, 48 h 70 degC, 24 h HN HN

N Br N N R

Scheme 9. Synthesis, homopolymerization of 3-(methacrylamidomethyl)-pyridine and its copolymerization with N-isopropylacrylamide (Dizman et al.[182]).

The obtained polymers were quaternized with various bromoalkanes and their antimicrobial activity was tested. The best results were achieved for copolymers containing 10 mol% of 3-(methacrylamidomethyl)-pyridine and quaternized with 1-bromotetradecane.

Chapter 2

A separate class of polymers with heterocyclic substituents is the imidazole derivatives [183] (see Scheme 10). Due to the fact that the imidazole ring is present in many biomolecules (e.g. histamine), it has excellent bio-compatibility and next to it shows very good stability and chemical resistance. According to Soykan et al. [184] already the non-modified imidazole containing polymer exhibits antimicrobial properties but it can be quaternized as well yielding cationic polymers.

* * n

O O O O

N N

Scheme 10. Examples of polymers with imidazole side groups. [183].

According to Bonilla and Garcia [162] acrylic and methacrylic based polyelectrolytes are most widely synthesized and investigated antimicrobial polymers. Wide variety of available (meth)acrylates enables designing polymers with desired ratio of hydrophilic (cationic) to hydrophobic part, type of the hydrophobic fraction, molecular weight and architecture. The commonly used methacrylates methyl methacrylate (MMA) or tert-butyl methacrylate (t- BuMA) are frequently copolymerized with methacrylic acid, 2-(phenoxycarbonyloxy)ethyl methacrylate (PCEMA) [185], 2-(dimethylamino)ethyl methacrylate (DMAEMA) [186] or 2- hydroxyethyl methacrylate (HEMA) [187], modified when necessary with substituted amines (amination) and subsequently quaternized with alkyl or arylalkyl halides.

The inherent drawback of polycations is their bio-incompatibility [188]. This can be overcome be means of introduction of highly biocompatible methacrylates as HEMA and PEGMA [189, 190]. This makes this type of antimicrobial copolymers extremely interesting for biomedical applications. Despite the fact that (meth)acrylates are most frequently used monomers for synthesis of QAS, the copolymers of styrene and styrene derivatives are another important group of polymers. As a basic monomer which can be quaternized 4-(dimethylaminomethyl)-styrene

Chapter 2 was used [167,191]. In this case as well the general rule has been confirmed that the antibacterial efficiency of ammonium groups increases as the substitute alkyl length increases. 4-(Dimethylaminomethyl)-styrene has been also copolymerized with 4-octylstyrene [192]. The presence of the strongly hydrophobic comonomer improved the bactericidal properties only marginally at relatively low molar fractions (not higher than 20 mol%) in the copolymer. Comparison of the activity of the synthesized polymer with [Ala8,3,13]-Magainin II amide (a natural peptide) shows lower antimicrobial properties (higher MIC values) compared to the peptide. What is very important, the authors have demonstrated that backbone preorganization is not essential for antimicrobial activity of polymeric materials. Number of publications describe polysiloxanes with N,N-dialkylimidazolium salts [192- 194] (see Scheme 11). The polysiloxane QAS exhibit high to moderate antibacterial properties (depending on the type of bacteria) whereas no effect of the polymer architecture has been observed. The biocidal power of the imidazolium salts is high and comparable with standard QAS. The efficacy remains constant even if different counterions Cl- or Br- are used. The clear advantage of the imidazolium substituted polysiloxanes is their high thermal stability.

nOctX HO Si O H HO Si O H n n

N N

N N CmH2m+1 X

X = Cl, Br m = 8

Scheme 11. Quaternization of an N-imidazolopropyl substituted polysiloxane with n-octyl halides (Mizerska et al. [194]).

It has been already mentioned that in many cases the antimicrobial action of polymers is either not well understood or not known. Moreover the role of specific polymer backbones is hard to distinguish from the role of functionalities attached to it. In order to explore this influence Waschinski [195] and co-workers proposed to attach the bioactive function to the end of a polymeric chain. The prerequisite for such approach must be water solubility of the polymer and the polymer backbone should not have any interaction with the microbial cell wall. These two requirements are met best by various poly(2-alkyl-1,3-oxazoline) (Scheme 12).

Chapter 2

Scheme 12. Synthesis of poly(2-alkyl-1,3-oxazoline) with telechelic NH2-Group and antimicrobial function (Waschinski et al. [196]).

In this way a new class of antimicrobial polymers was created: Poly(2-alkyl-1,3- oxazoline) with terminal quaternary ammonium groups [194-197]. The authors concluded that the nature of the polymeric backbone influenced the bioactivity of the terminating antimicrobial functions but the length of the polymer chain plays no role.

Antibacterial hyperbranched and dendritic polymers are interesting to mention due to their completely different structure and architecture compared to previously reported linear polymers. Pasquier [198,199] prepared functional branched poly(ethylene imine) (PEI) (depicted in Scheme 14) in which primary amine groups were functionalized with quaternary ammonium groups, alkyl chains of different length, allylic and benzylic groups in a one-step reaction, using a carbonate coupler (Scheme 13). Such polymers exhibit bactericidal properties against both Gram-positive and Gram-negative bacteria however the dispersity of the molecular weight combined with heterogeneity in the functionality of these cationic amphiphilic polymers are still an obstruction for a better understanding of their interaction with the cell membranes

[198].

Chapter 2

OH HO OH

O O O H O N N O O O Y O O O O O O H O O N H O H X (CH ) O N N H 1: Y = OC6H5 2 n (CH2)m 2: Y = Cl O O

n = 6, 8, 10, 12, 14, 16, 18 PEI QmX QI: m = 1, X = I Q8Br: m = 8, X = Br Q12Br: m = 12, X = Br H H N N N N N H H

N O HN NH OH H H (CH2)n N O O N N H H N O O O (CH2)m H OH O O Scheme 13. Synthetic route for the preparation of hydrophobic polycations using functional ethylene carbonates and PEI (Pasquier et al. [198]).

Chen [200,201] synthesized quaternary ammonium functionalized poly(propyleneimine) dendrimers (see Scheme 14).

Scheme 14. Commercially available generation 4 poly(propyleneimine) primary amine terminated dendrimers used by Chen [199].

The antibacterial properties of these quaternary ammonium dendrimers depend on the size of the dendrimer, the length of hydrophobic chains in the quaternary ammonium groups, and the counterion. The biocidal activity of the dendrimers with bromine as counterion was higher than of those with chlorine anions.

Chapter 2

References [1] F.R. Mayo, F.M. Lewis, J . Amer.Chem. Soc., 1944, 66,1594-1601 [2] B.Klumperman, K.F. O'Driseoll, Polymer, 1993, 34, 1032-1037 [3] K. Salimi, Z.M.O. Rzayev, E. Pişkin, Appl. Clay Sci., 2014, 101, 106-118 [4] J.C. Bevington, H.W. Melville, R.P. Taylor, Polym. Chem., 1954,77, 463–476 [5] U. Beginn, e-Polymers, 2005, no. E_003 [6] D. Spridon, L. Panaitescu, D. Ursu, C.V. Uglea, I. Popa, R.M. Ottenbrite, Polym. Int. 1997, 43, 175–181 [7] M. Kraus, Functional Combcopolymer for Superhudrophobic and Superhydrophilic Surfaces, PhD Thesis, Ulm 2002 [8] G. Odian, Principles of Polymerization, 4th edition, John Wiley & Sons, Hoboken, New Jersey, 2004 [9] A.B. Lindstrom, M.J. Strynar, E.L. Libelo, Environ. Sci. Technol., 2011, 45, 7954- 7961 [10] E. Kissa; Fluorinated Surfactants and Repellents, 2nd edition, M. Dekker, ed. New York, 2001. [11] R.F. Brady Jr.; In Encyclopedia of Polymer Science and Technology, 3rd ed. H. Mark et all (eds), John Wiley & Sons, New York, 2007 [12] I.O. Arukalam, E.E. Oguzie, Y. Li, J. Coll. Interface Sci, 484, 2016, 220–228 [13] T. Harit, F. Malek, B. Ameduri, Eur. Polym. J., 79, 2016, 72–81 [14] X. Zhu, X. Jiang, X. Ai, H. Yang, Y. Cao,Electrochimica Acta, 2015,165, 67-71 [15] L. Zhang, B. Geng, Q.L. Anhou, X.-S. Zhang, Journal of Surfactants and Detergents, 2016, 19, 559–565 [16] Y. Kim, S.-H. Kim, , Organic Electronics21, 2015,111–116 [17] Z. Cui, E. Drioli, Y.M. Lee, Progress in Polymer Science, 2014, 39, 164– 198 [18] P. Arab, A. Verlander, H.M. El-Kaderi, J. Phys. Chem. C, 2015,119, 8174–8182 [19] T. Imae, Current Opinion in Colloid and Interface Science, 2003, 8, 307–314 [20] L. Xu, A. Yekta, Z. Masoumi , S. Kanagalingam, M.A. Winnik, Langmuir, 1997, 13, 2447 –2456. [21] S.J. McLain, B.B. Sauer, L.E. Firment, Macromolecules, 1996, 29, 8211-8219 [22] O. Nuyken, W. Dannhorn, W. Obrecht, Macromol. Chem. Phys., 1994, 195, 3531- 3539 [23] X. Yang, L. Zhu, Y. Zhang. Y. Chen, B. Bao, Appl. Surf. Sci., 2014, 295, 44-49 [24] N. Belov, Reactive Fluoropolymers Synthesis and Application, RWTH Aachen 2011 [25] M.R. Giles, S.J. O’Connor, J.N. Hay, R.J. Winder, S.M. Howdle, Macromolecules, 2000, 33, 1996-1999 [26] CAS Registry No. 108-31-6 [27] B. C. Trivedi, B. Culbertson, Maleic Anhydride, Plenum Press N.Y. 1982, ISBN-0- 306-40929-1 [28] J.R. Skeen, Chem. Eng.News, 1948, 26, 3684-3688 [29] I. Kroschwitz (ed.), Encyclopedia of Chemical Technology, 4th edition, 1992, 15, 893 [30] B.B. Allen, B.W. Wyatt, H.R. Henze, J. Am. Chem. Soc., 1939, 61, 843-846 [31] J.M. Weiss, C.R. Downs, J. Am. Chem. Soc., 1922, 44, 1118-1125 [32] D. Freudenberger, F. Wunder, H. Fernholz, US Patent 4096156, 1978 [33] R. Takashi, P.J. Elving, Elecrochim. Acta, 1967, 12, 213-217 [34] S. Unemura, Y. Ikeda, Jpn. Kokkai 75, 1975, 101, 320 [35] H. Adkins, G. Krsek, J. Am. Chem. Soc., 1948, 70, 383-386 [36] S. Unemura, Y. Ikada, Jap. Patent 50101320, 1975 [37] L.H. Flett, W.H. Gardner, Maleic Anhydride Derivatives, John Wiley & Sons, New York 1952 30

Chapter 2

[38] J.M. Rosenfeld, B.B. Murphy, Talnta,, 1967, 14, 91 [39] J.C. Brunett, R.A. Keppel, W.D. Robinson, Catal. Today, 1987, 5, 537 [40] J.E. White, US Patent 4118403, 1978 [41] T. Purdie, J. Chem. Soc., 1881, 39, 334-338 [42] H.S. Rothrock, US Patent 2346612, 1944 [43] L.P. Kyrides, US Patent 2377246, 1945 [44] Y. Tituya, J. Agr. Chem. Soc. Japan, 1941, 17, 706 [45] W. Reppe, H. Ufer, US Patent 2200220, 1941 [46] V. Lamberti, US Patent 3922272, 1975 [47] V.G. Gruzdev, Gen. Chem. USSR, 1979, 49, 725 [48] C.G. Krespan, US Patent 5112993, 1992 [49] A.D. Efendiev, USSR Patent 1616893, 1990 [50] K. Freudenberg, A. Lux, Ber., 1928, 61, 1083-1098 [51] E. Gilbert, Z. Naturforsch., 1977, 32, 1308-1313 [52] J.O. Oludipe, J. Chem. Tech. Biotechnol., 1992, 55, 103-109 [53] H. Weiss, J. Am. Chem. Soc., 1977, 99, 1670-1672 [54] R.W. White, W.D. Emmons, Tetrahedron, 1962, 17, 31-34 [55] K. Petrisch, P. Krol, US Patent 4028407, 1977 [56] A.C.D. Rivetti, N.N. Sidgwick, J. Chem. Soc., 1910, 97, 1677-1686 [57] L. Eberson, L. Landstrom, Acta. Chem. Scand., 1972, 26, 534-542 [58] R.W. White, W.D. Emmons, Tetrahedron, 1962, 17, 31-34 [59] A.G. Gonorskaya, S.A. Marina, Lakokras. Mat. Ikl. Primen., 1984, 1, 18-22 [60] P.D. Thomas, US Patent 3953616, 1976 [61] R. Anshutz, Liebigs Ann. Chem, 1891, 259, 137-148 [62] L.E. Coleman, J.F. Bork, H. Dunn, J. Org. Chem, 1959, 24, 135-136 [63] I.J. Rinkes, Rec. Trav. Chim. 1929, 48, 961-967 [64] F. Bergmann, D. Shapiro, J. Org. Chem., 1942, 6, 774-779 [65] W.N. Marmer, D.E. van Horn, W.M. Linfield, J. Am. Oil Chem. Soc. 1974, 51, 174-177 [66] R.E. Lutz, Org. Synth., 1940, 20, 29-33 [67] K. Alder, F. Pasher, A. Schmitz, Ber., 1943, 76B, 27 [68] C.M. Selwitz, US Patent 3409638, 1968 [69] J.C. Powel, US Patent 4496746, 1985 [70] S.M. Spatz, US Patent 2658168, 1953 [71] W. Ouattlebaum, US Patent 2307157, 1943 [72] F.A. Cotton, G. Wilkonson, Advanced Inorganic Chemistry, 5th Ed. John Wiley Inc. New York 1988, 71, 1152 [73] C.H. LaiC.H. Cheng, F.L. Liao, S.L. Wang, Inorg. Chem., 1993, 32, 5658-5664 [74] A.S. Antsyshkina, L.M. Dikareva, M.A. Porai-Koshits, L.E. Fykin, V.J. Didarev, M.G. Guseinov, Koord. Khim.1982, 8, 1256-1260 [75] W.O. Pillar, US Patent 4113672, 1978 [76] T. Alfrey, E. Lavin, J. Am. Chem. Soc., 1945, 67, 2044-2015 [77] M.C. De Wilde, G. Smets, J. Polym. Sci., 1950, 5, 253-258 [78] J.L. Lang, W.A. Pavelich, H.C. Clarey, J. Polym. Sci., 1961, 55, 531-532 [79] Y. Kubo, A. Yamamoto, S. Ikeda, Bull. Chem. Soc. Jap., 1974, 393-395 [80] S.N. Bhadani, J. Prasad, Makromol. Chem., 1977, 178, 187-194 [81] S.D. Haman, J. Polym. Sci. Polym. Chem. Part A , 1967, 11, 2939–2941 [82] A.J. Restaino, R.B. Mesrobian, H. Morawetz, D.S. Ballantine, G.J. Dienes, D.J. Metz, J. Am. Chem. Soc., 1956, 78, 2939-2943 [83] J.L. Lang, W.A. Pavelich, H.D. Clarey, J. Polym. Sci. Polym. Chem. Part A, 1963, 4 1123–1136

Chapter 2

[84] I. Nagahiro, K, Nishihara, N. Sakota, J. Polym. Sci., 1974, 12¸785-792 [85] K, Nishihara, N. Sakota, J. Polym. Sci., 1974, 12¸57-65 [86] K. Takakura, K. Hayashi, S. Okamura, J. Polym. Sci., 1966, 4¸1747-1757 [87] D. Bryce-Smith, A. Gilbert, B. Vickery, Chem. Ind., 1962 [88] D. Ritenberg, L. Ponticorvo, Proc. Nat. Acad. Sci. US, 1960, 46, 822-824 [89] D. Brown, J. Pomakis, Makromol. Chem., 1974, 175, 1411-1425 [90] S. Schiller, J. Hu, A.T.A. Jenkins, R.B. Timmons, F.S. Sanches-Estrada, W. Knol, R. Foersch, Chem. Mater. 2002, 14, 235-242 [91] P.D. Bartlett, K. Nozaki, J. Am. Chem. Soc. 1946, 68, 1497 [92] M. Ashot, US Patent 3385834, 1968 [93] W.A. Holmes-Walker, K.E. Weale, J. Chem. Soc., 1955, 2295-2301 [94] C. Heuck, M. Lederer, US Patent 3457240, 1969 [95] S.D. Hamann,J. Polym. Sci. Part A1, 1967, 5, 2939-2941 [96] C.I. Simionescu, M. Grigoras, V. Barboiu, Polym. Bull., 1984, 11, 511-515 [97] R.A. Crump, A.H. Price, Trans. Faraday Soc., 1969, 65, 3195-3201 [98] C.C. Price J. Polym. Sci., 1946, 1¸ 83-89 [99] C.C. Price J. Polym. Sci., 1948, 3¸ 772-775 [101] N. G. Gaylord, S. Maiti, J. Macromol, Sci., 1972, 6, 1481-1493 [101] L. Jedlinski, J. Maslinska-Solich, J. Polym. Sci.C, 1973, 42, 411-418 [102] T. Wagner-Jauregg, Ber., 1930, 63B, 3213-3224 [103] F.R. Mayo, F.M. Lewis, C. Walling, J. Am. Chem. Soc., 1948, 70, 1529-1533 [104] S. Robert, US Patent 2308495, 1943 [105] C.S. Rondestvedt, B.H. Wark, J. Org. Chem., 1955, 27, 1201-1208 [106] M.M. Martin, N.P. Jensen, J. Org. Chem. 1962, 27, 1201-1205 [107] R.G. Ismailov, S.M. Aliev, M.R. Bairamov, S.G. Aliev, S.T. Akhmedov, N.R. Amakhanova, Azerb. Neft. Khoz., 1969, 48, 34-37 [108] R.G. Ismailov, S.M. Aliev, M.R. Bairamov, M.N. Maerramov, S.G. Aliev, G.G. Gadzhiev, Azerb. Khim. Zh., 1970, 4, 60 [119] G. Sackman, G. Kolb, Makromol. Chem., 1971, 149, 51-61 [110] G. Sackman, G. Kolb.Angew. Chem. Int. Ed., 1972, 11, 339 [111] R.B. Seymour, F.F. Harris, I. Branum, Ind. Eng. Chem., 1949, 41, 1509-15-13 [112] K. Fujimori, J. Macromol. Sci. Chem. A, 1975, 9, 495-504 [113] H.P. Frank, Makromol. Chem., 1968, 114, 113-117 [114] S. Machi, T. Sakai, M. Gotoda, T. Kogiya, J. Polym. Sci. A, 1966, 4, 821-828 [115] M. Kellou, G. Jenner, Makromol. Chem. 1979, 180, 1687-1696 [116] T. Otsu, A. Shimizu, M. Imoto, J. Polym. Sci. B, 1964, 2, 973-977 [117] R. Bacskai, L.P. Lindeman, D. L. Rabenstein, J. Polym. Sci. A,, 1972, 10, 1297-1310 [118] W.J. Heilman, D.M. Jenkins, US Patent 3461108, 1969 [119] J. H. Johnson, Macromol. Synth., 1963, 1, 42-46 [120] T.A. DuPlessis, A. Lustigant, E. Greling, J. Macromol. Sci. A, 1977, 11, 1015 [121] T. Tanaka, O. Vogl, J. Macromol. Sci. Chem. A, 1974, 8, 1299-1312 [122] P. Kincait, T. Tanaka, O. Vogl, Am. Chem. Soc. Div. Polym. Chem. Prepr., 1974, 15, 222-227 [123] O. Vogl, J. Polym. Sci. Part A; Polym. Chem., 2000, 22, 4013-4022 [124] J.M. Barton, G.B. Butler, E.C.Chapin, J. Polym. Sci. A, 1965, 3¸ 501-511 [125] Y. Yamashita, S. Iwatsuki, T. Kokubo, J. Polym. Sci. C¸1966, 23, 753-759 [126] N.G. Gaylord, M. Stolka, A. Takahashi, S. Maiti, J. Macromol. Sci. A, 1971, 5, 867-881 [127] N.G. Gaylord, A. Takahashi, Macromol. Synth., 1972, 4, 99 [128] N.G. Gaylord, S. Maiti, J. Polym. Sci. B, 1971, 9, 359-366 [129] T. Alfrey, E. Lavin, J. Am. Chem. Soc. 1945, 67, 2044-245

Chapter 2

[130] H. Fritze, Angew. Makromol. Chem., 1973, 33, 101-110 [131] R.B. Seymour, Ind. Eng. Chem., 1970, 62, 47–61 [132] Y. Ogo, T. Imoto, Rev. Phys. Chem. Jap. 1972, 42, 34 [133] R.B. Seymour. D.P. Garner, J. Coat. Technol., 1976, 48, 41-45 [134] R.B. Seymour, Rev. Plast. Mod., 1978, 268, 500 [135] C.C. Price, J.G. Walsh, J. Polym. Sci., 1951, 6, 239-242 [136] C. Walling, F.R. Mayo, Discussions Faraday Soc., 1947, 2, 295 [137] S. Murahashi, S. Nozakura, K. Yasafuku, Bull. Chem. Soc. Jap., 1965, 38, 2082 [138] M. Yokawa, Y. Ogo, T. Imoto. Makromol. Chem., 1973, 163, 135-142 [139] M.G. Roberts, US Patent 4104241, 1978 [140] A.D. McLaren, J. Polym. Sci., 1948, 3, 652-662 [141] E. Dougherty,US Patent 3509089, 1970 [142] R.B. Alan, US Patent 3004958, 1961 [143] H. Dennis, US Patent 2793985, 1957 [144] J.Z. Orszagh, T. Wijns, J. Polym. Sci. C, 1965, 16, 4219-4222 [145] T. William,US Patent 3093506, 1963 [146] R.B. Seymour, I. Baranum, F.W. Hayward, Ind. Eng. Chem. 1949, 41, 1482-1484 [147] M.H. Loucheux, A. Banderet, J. Polym. Sci., 1960, 48, 405-421 [148] M.H. Loucheux, A. Banderet, Bull. Soc. Chim. Fr., 1964, 6, 1216-11219 [149] C. Caze, M.H. Loucheux, J. Macromol. Sci. A, 1978, 12, 1501 [150] J. Tino, J. Placek, Eur. Polym. J. 1971, 7, 1615-1620 [151] J.P. Stallings, J. Poly. Sci. A, 1970, 8, 1557-1570 [152] M. Yokawa, Y. Ogo, T. Imoto, Makromol. Chem., 1973, 163, 135-142 [153] Z. Debowski, Chem. Anal., 1965, 10, 469-480 [154] H. Batzer. H.H. Grunewald, Makromol. Chem., 1953, 9, 116-147 [155] R.B. Seymour. P.D. Kincaid, D.R. Owen, J. Paint Technol., 1973, 45, 33-37 [156] L. Rios, J. Guillot, J. Macromol. Sci. A,1978, 12, 1151-1174 [157] N. Gaylord, US Patent 3765934, 1973 [158] DR Nielsen US Patent 3291632, 1966 [159] M. Raetzsch, M. Arnold, J. Macromol. Sci. Chem., 1987, A24, 507-515 [160] M.B. Patel, S.A. Patel, A. Ra, R.M. Patel, J. Appl. Polym. Sci., 2003, 89, 895 [161] E.-S. Park, H.-J. Lee, H.-Y. Park, M.-N. Kim, Chung, K.-H., J.-S.Yoon, J. Appl. Polym. Sci., 2001, 80, 728 [162] A. Munoz-Bonilla, M. Fernandez-Garcia,Progress in Polym. Sci., 2012, 37, 281 [163] E.-R. Kenawy, F. I. Abdel-Hay, A. E.-R. El-Shanshoury, M. H. J. El-Newehy, Polym. Sci., Part A: Polym. Chem., 2002, 40, 2384 [164] G. Li, J. Shen, J. Appl. Polym. Sci., 2000, 78, 676 [165] E.R. Kenawy, S. D. Worley, R. Broughton, Biomacromolecules, 2007, 8 (5), 1359 [166] E.F. Palermo, D.-K. Lee, A. Ramamoorthy, K. Kuroda, J. Phys. Chem. B., 2011, 115, 366 [167] E.R. Kenawy, F.I. Abdel-Hay, A.R. El-Shanshoury, M.H. El-Newehy, J. Polym. Sci. Part A Polym. Chem., 2002, 40, 238 [168] T. Tashiro, Macromol. Mater. Eng. 2001, 286, 63 [169] S. Jiang, L. Wang, H. Yu, Y. Chen, Reac. Func. Polym., 2005, 62, 209 [170] G. Li, J. Shen, Y. Zhu, J. Appl. Polym. Sci., 1998, 67, 1761 [171] G. Li, J. Shen, Y. Zhu, J. Appl. Polym. Sci., 2000, 78, 668 [172] G. Sauvet, S. Dupond, K. Kazmierski, J. Chojnowski, J. Appl. Polym. Sci., 2000, 75,1005 [173] T. Abel, J.I. Cohen, R. Engel, M. Filshtinskaya, A. Melkonian, K. Melkonian, Carbohyd. Res., 2002, 337, 2495

Chapter 2

[174] E.S. Park, H.S. Kim, M.N. Kim, J.S. Yoon, Eur. Polym. J., 2004, 40, 2819 [175] J.C. Tiller, C.-J. Liao, K. Lewis, A.M. Klibanov, Proc. Nat. Acad. Sci. USA, 2001, 98, 5981 [176] G. Li, J. Shen, J. Appl. Polym. Sci., 2000, 78, 676 [177] S. Sharma, G. Chauhan, R. Gupta, J.H. Ahn, J. Mater. Sci. 2010, 21, 717 [178] N. Kawabata, K. Takagishi, and M. Nishiguchi; React.Polym., 1989, 10, 269 [179] N. Kawabata, T. Hayashi, and M. Nishikawa, Bull.Chem. Soc. Jap.,1986,59,2861 [180] N. Kawabata, E. Natsuhara, I. Higuchi, J. Yoshida, Bull. Chem. Soc. Jap., 1983, 56, 1012 [181] V. Sambhy, B.R. Peterson, A. Sen, Angew. Chem., 2008, 120, 1270 [182] B. Dizman, M.O. Elasri, L.J. Mathias, Macromolecules, 2006, 39, 5738 [183] E.B. Anderson, T.E. Long, Polymer, 2010, 51, 2447 [184] C. Soykan, R. Coskun, A. Delibas, J. Macromol. Sci. Pure. Appl. Chem.; 2005, 42, 1603 [185] R. Adelmann, M. Mennicken, D. Popescu, E. Heine, H. Keul, M. Moeller Eur. Polym. J., 2009, 45, 3093 [186] G. Lu, D. Wu, R. Fu, React. Funct. Polym., 2007, 6, 355 [187] S. Punyani, H. Singh, J. Appl. Polym. Sci., 2008, 107, 2861 [189] P.H. Sellenet, B. Allison, B.M. Applegate, J.P. Youngblood, Biomacromolecules, 2007, 8, 19 [190] G. Cheng, H. Xue, Z. Zhang, S. Chen, S. Jiang, Angew. Chem. Int. Ed., 2008, 47, 8831 [191] S. Venkataraman, Y. Zhang, L.H. Liu, Y.Y. Yang, Biomaterials, 2010, 31, 1751 [192] M.A. Gelman, B. Weisblum, D.M. Lynn, S.H. Gellman, Org. Lett., 2004, 6, 557 [193] G. Sauvet, W. Fortuniak, K. Kazmierski, J. Chojnowski, J. Polym. Sci. Part. A Polym. Chem., 2003, 41, 2939 [194] U. Mizerska, W. Fortuniak, J. Chojnowski, R. Halasa, A. Konopacka, W. Werel, Eur. Polym. J., 2009, 45, 779 [195] C.J. Waschinski, J.C. Tiller, Biomacromolecules, 2005, 6, 235 [196] C.J. Waschinski, V. Herdes, F. Schueler, J.C. Tiller, Macromol. Biosci., 2005, 5, 149 [197] C.J. Waschinski, S. Barnert, A.Theobald, R. Schubert, F. Kleinschmidt, A. Hoffmann, K. Saalwachter, J.C. Tiller, Biomacromolecules, 2008, 9, 1764 [198] N. Pasquier, H. Keul, E. Heine, M. Moeller, Biomacromolecules, 2007, 8, 2874 [199] N. Pasquier, H. Keul, E. Heine, M.Moeller,B. Angelov, S. Linser, R. Willumeit, Macromol. Biosci., 2008, 8, 903 [200] C.Z. Chen, N.C. Beck-Tan, P. D., T.K. van Dyk, R.A. LaRossa, S.L. Cooper, Biomacromolecules, 2000, 1, 473 [201] C.Z. Chen, S.L. Cooper, Biomaterials, 2002, 23, 3359

Chapter 3

Synthesis of Terpolymers with Homogeneous Composition by Free Radical Copolymerization of Maleic Anhydride, Perfluorooctyl and Butyl or Dodecyl Methacrylates: Application of the Continuous Flow Monomer Addition Technique

1. Introduction

Due to their outstanding chemical resistance, thermal stability as well as low friction coefficient and flame retardation [1], fluorinated and partly fluorinated polymers have found many industrial applications, e.g. as friction modifiers [2], anti-fouling coatings [3] or membranes [4,5]. Homopolymerization of fluorinated monomers such as perfluoroolefins, (meth) acrylates with perfluorinated chains or fluorinated styrene derivatives leads to polymers with high fluorine content. The properties of copolymers obtained by employing these monomers might not always be optimal for specific applications. The use of fluorine containing monomers in polymer synthesis is under strong pressure from public organizations pointing out the adverse health effects such as enamel fluorosis due to excessive fluorine exposure [6]. The combination of the need to fine tune the polymer properties and limit the amount of the fluorine compound to the necessary minimum concentration (to still maintain all the benefits of using fluorinated materials) leads to the development of materials that are copolymers which contain highly fluorinated monomers reacted with non-fluorinated constituents. The copolymerization of usually very expensive fluorinated monomers with non-fluorinated monomers has also an economic effect of creating more cost effective materials that can be disposed easier, cheaper and with less impact on people and the planet. Batch free radical polymerization in binary or ternary systems usually leads to products which are blends of polymer chains with different composition. The alteration of the composition during the reaction is caused by different reactivity of the monomers in the mixture. Hence, the most reactive monomers are consumed first, and consequently the polymer is enriched in monomers of lower reactivity. This effect is strongly pronounced when monomers which cannot undergo homopolymerization are used. The preferred route to overcome this problem is to feed continuously the reaction mixture with monomers at the rate they are consumed. The present report focuses on the copolymerization of maleic anhydride with alkyl-methacrylates and 1H,1H,2H,2H-perfluorodecyl methacrylate. To ensure a homogenous composition of the terpolymers, continuous addition of the monomers

Chapter 3 was applied. The rate of addition was adjusted by the kinetics of the copolymerization. The composition of synthesized copolymers was determined by means of 1H NMR (nuclear magnetic resonance) spectroscopy and elemental analysis. The obtained copolymers were characterized by means of GPC, while the thermal behavior of the copolymers was investigated by means of TGA (thermal gravimetric analyzer) and DSC (differential scanning calorimetry).

2. Experimental

Materials

1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA, Aldrich, St. Louis, MO, USA), butyl methacrylate (BMA, Merck, Kenilworth, NJ, USA) and dodecyl methacrylate (DMA,

Merck) were distilled over CaH2 ( Sigma-Aldrich Chemie, Steinheim, Germany) under reduced pressure. Maleic anhydride (MAH, Merck, for synthesis) was sublimed under reduced pressure (50 °C, 2.4 × 10−2 bar) before use. Azobisisobutyronitrile (AIBN, Merck) was recrystallized from methanol at 40 °C. 2-Butanone (MEK, Merck) was dried over CaH2 and distilled before use. 1,3-Bis(trifluoromethyl)benzene (HFX, ABCR, Karlsruhe, Germany), Freon 113 (Fluka) and other solvents were used as received.

Measurements/Apparatus

Size exclusion chromatography (SEC) was performed with a system consisting of a LC 1120 pump (Polymer Laboratories, Church Stretton, UK), a UV detector ERC-7215 and RI detector ERC-7515A (ERMA CR INC, Kawaguchi City, Japan), a precolumn 50 × 8 mm with 50 Å nominal pore size and four columns (300 × 8 mm) filled with MZ-Gel SDplus of nominal pore sizes 50 Å, 100 Å, 1000 Å, and 10,000 Å (MZ-Analysentechnik, Mainz, Germany). The set was calibrated with PMMA (polymethyl methacrylate) and PS (polystyrene) standards from Polymer Laboratories. The sample concentration was 7 mg of polymer in 1 mL of solvent; the injected sample volume was 100 μL. Tetrahydrofuran (THF) used for measurements, and was stabilized with 2,6-di-tert-butyl-4-methylphenol (250 mg/L). Samples of high fluorine content were dissolved and measured in a mixture of THF: Freon 113 1:1 (vol:vol), stabilized with 250 mg/L of 2,6-di-tert-butyl-4-methylphenol. 1H NMR spectra were obtained on a Bruker DPX-300 (Bruker Corporation,

Karlsruhe,Germany) spectrometer in CDCl3, acetone-d6 and mixtures with Freon 113 (1:1, vol:vol) at 300 MHz. MestRe-C 4.9.0.0 (Mestrelab Research, S.L., Santiago de Compostela,

Chapter 3

Spain) was used as evaluation software. Solvent signals were used as internal references. In the spectra description the maximum of the signal is given. Thermogravimetric analysis (TGA) was performed using a Netzsch TG 209c (Netzsch, Selb, Germany) thermo balance under nitrogen atmosphere at a nitrogen flow 15 mL/min. Samples of 9–11 mg were placed in standard Netzsch alumina 85 L crucibles and heated at a rate of 10 K∙min−1. Differential scanning calorimetry (DSC) measurements were performed using a Netzsch DSC 204 unit. Samples (typical weight: ~9 mg) were enclosed in standard Netzsch 25 μL aluminium crucibles. Indium and palmitic acid were used as calibration standards. Heating and cooling rates were 10 K∙min−1. Continuous monomer addition was performed with a Harvard Apparatus 11 Plus (Harvard Apparatus, Holliston, Massachusetts, United States) syringe pump equipped with 50 mL Braun syringes. Elemental analyses were performed by Anastasya Buyanowskaya at the A.N. Nesmeyanov Institute of Organoelement Compounds of the Russian Academy of Science in Moscow.

Low Conversion Experiments

Maleic Anhydride and 1H,1H,2H,2H-Perfluorodecyl Methacrylate

Maleic anhydride (0.65 g, 6.63 mmol) and 1H,1H,2H,2H-perfluorodecyl methacrylate (2.34 g, 4.7 mmol) were dissolved in a mixture of 2-butanone (2.25 g) and 1,3-bis (trifluoromethyl) benzene (2.25 g, 1:1 mass:mass). 2,2’-azo-bis-isobutyronitrile (1 mol %) with respect to the total amount of monomers was added. The mixture was placed in a two-necked round bottom flask, equipped with a rubber septum and a valve. The reaction mixture was degassed three times by freeze–pump–thaw cycles and filled back with nitrogen. The reaction was carried out at 65°C for 40 minutes. During the reaction time three samples were taken, precipitated in methanol and dried in vacuo at 40 °C until a constant weight was obtained. The conversion was determined gravimetrically. The polymer composition was determined by 1H NMR for samples with conversions below 10% after alcoholysis in methanol. The experiment was conducted twice with two different monomer compositions as described in Table 1.

Maleic Anhydride and Butyl Methacrylate

Chapter 3

The copolymerization reactions were performed in analogy to the previously described polymerization of maleic anhydride and 1H,1H,2H,2H-perfluorodecyl methacrylate mixtures. The batch compositions are detailed in Table 1. 1 H NMR (CDCl3, δ in ppm) of P [BMA-co-MAH]: 3.95 (m, 2H, –O–CH2–CH2–CH2–

CH3); 3.66 (m, 3H, –O–CH3); 2.71 (m, 2H, HOOC–CH–CH–COOMe); 1.93 (m, NN); 1.60 (m,

2H, –O–CH2–CH2–CH2–CH3); 1.40 (m, 2H, –O–CH2–CH2–CH2–CH3); 0.93 (m, 3H, –O–CH2–

CH2–CH2–CH3).

1H,1H,2H,2H-Perfluoroodecyl Methacrylate and Butyl Methacrylate

The copolymerization reactions were performed in analogy to the previously described polymerization of maleic anhydride and 1H,1H,2H,2H-perfluorodecyl methacrylate mixtures. The batch compositions are in Table 1. 1 H NMR (CDCl3/Freon-113 (1:1, vol:vol), δ in ppm) P [BMA-co-F8H2MA]: 4.34 (m, 2H,

–O–CH2–CH2–(CF2)8–F); 4.03 (m, 2H, –O–CH2–CH2–CH2–CH3); 2.52 (s, 2H, –O–CH2–CH2–

(CF2)8–F); 1.92 (s, 2H, –C–CH2–C– (BMA backbone) overlapped with NN); 1.66 (m, 2H, –O–

CH2–CH2–CH2–CH3); 1.47 (m, 2H, –O–CH2–CH2–CH2–CH3); 1.00 (m, 3H, –O–CH2–CH2–

CH2–CH3).

1H,1H,2H,2H-Perfluorodecyl Methacrylate, Butyl Methacrylate and Maleic Anhydride

The terpolymerization reactions were performed in analogy to the previously described polymerization of maleic anhydride and 1H,1H,2H,2H-perfluorodecyl methacrylate mixtures. The batch compositions are detailed in Table 1. 1 H NMR (acetone-d6/Freon-113 (1:1, vol:vol), δ in ppm): 4.34 (m, 2H, –O–CH2–CH2–

CF2–); 4.00 (m, 2H, –O–CH2–CH2–CH2–CH3); 3.60 (m, 3H, –COO–CH3); 2.8 (s, Freon-113);

1.86 (m, 2H, –O–CH2–CH2–CF2–); 1.65 (m, 2H, –O–CH2–CH2–CH2–CH3); 1.45 (m, 2H, –O–

CH2–CH2–CH2–CH3); 0.98 (m, 3H, –O–CH2–CH2–CH2–CH3). Signals in the range of 1–2 ppm are overlapped with the signals of backbone protons.

1H,1H,2H,2H-Perfluoroodecyl Methacrylate, Dodecyl Methacrylate and Maleic Anhydride

The terpolymerization reactions were performed in analogy to the previously described polymerization of maleic anhydride and 1H,1H,2H,2H-perfluorodecyl methacrylate mixtures. The batch compositions are detailed in Table 1. 1 H NMR (CDCl3, δ in ppm): 4.19 (m, 2H, –O–CH2–CH2–CF2–); 3.86 (m, 2H, –O–CH2–

CH2– (CH2)9–CH3); 3.60 (m, 3H, –COO–CH3); 2.8 (s, Freon-113); 1.82 (m, 2H, –O–CH2–

Chapter 3

CH2–CF2–); 1.54 (m, 2H, –O–CH2–CH2–(CH2)9–CH3); 1.19 (m, 2H, –O–CH2–CH2–(CH2)9–

CH3); 0.81 (m, 3H, –O–CH2–CH2– (CH2)9–CH3). Signals in the range of 1–2 ppm overlapped with the signals of backbone protons.

Table 1. Composition of the reaction mixtures for low conversion copolymerization of monomer systems: MAH/F8H2MA, MAH/BMA, BMA/F8H2MA, BMA/F8H2MA/MAH, DMA/F8H2MA/MAH (T = 65 °C).

mMAH mBMA mF8H2MA mDMA mAIBN mMEK mHFX CM fMAH fBMA fF8H2MA fDMA (g) (g) (g) (g) (g) (g) (g) (mol/L)

0.650 0.60 - 2.34 0.400 - - 0.018 2.25 2.25 2.50 - 0.220 0.30 - 2.77 0.700 - - 0.013 1.55 1.55 2.50 - 1.530 0.60 1.400 - - - - 0.042 5.15 5.15 2.50 - 1.080 0.45 1.920 - - - - 0.040 4.90 4.90 2.50 0.400 - - 0.190 2.81 0.800 - - 0.011 1.32 1.32 2.50 0.550 - - 0.540 2.46 0.550 - - 0.014 1.70 1.70 2.50 0.200 0.360 0.33 0.525 2.00 0.330 - - 0.018 2.00 2.00 2.77 0.450 0.737 0.50 0.530 2.00 0.250 - - 0.025 3.00 3.00 2.50 0.330 3.680 0.75 1.240 2.00 0.075 - - 0.082 10.00 10.00 2.50 0.250 1.840 0.75 - 1.00 0.075 1.11 0.175 0.164 6.27 6.27 2.00 0.175 1.840 0.75 - 1.00 0.075 1.11 0.175 0.164 12.54 - 2.00 - 1.715 0.21 - 0.50 0.040 1.25 0.210 0.150 5.84 5.84 2.00 - 0.920 0.75 - 1.00 0.150 0.32 0.100 0.082 3.13 3.13 2.00 - 2.760 0.90 - 0.50 0.030 0.56 0.070 0.200 7.83 7.83 2.00 - 0.750 0.55 - 1.00 0.315 1.11 0.315 0.091 3.48 3.48 2.00 -

MAH, maleic anhydride; BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate; DMA, n-dodecyl methacrylate;, AIBN, 2,2’-azo-bis-isobutyronitrile; MEK, methyl ethyl

ketone; HFX, 1,3-bis (trifluoromethyl) benzene; fi, molar fraction of monomer in reaction mixture; CM, total monomer concentration.

Continuous Addition Terpolymerization

For details on calculations, please see the Appendix to Chapter 3 “Theory of continuous addition polymerization”.

1H,1H,2H,2H,-Perfluorodecyl Methacrylate, Butyl Methacrylate and Maleic Anhydride (10- Fold Monomer Excess)

Preparation of the stock solution (Solution A): butyl methacrylate (0.448 g, 3.15 mmol), 1H,1H,2H,2H-perfluorodecyl methacrylate (0.722 g, 1.45 mmol), maleic anhydride (1.33 g, 13.57 mmol) and 2,2’-azo-bis-isobutyronitrile (0.03 g, 0.18 mmol) were dissolved in a mixture of 2-butanon (4.53 g, 3.63 mL) and 1,3-bis (trifluoromethyl) benzene (2.79 g, 3.63 mL). The mixture was placed in a three neck round bottom flask, equipped with a nitrogen inlet, reflux condenser bearing an oil bubbler vent as nitrogen outlet, a magnetic stirring bar and a rubber

Chapter 3 septum. The reaction mixture was degassed three times by freeze–pump–thaw cycles and filled back with nitrogen. Preparation of the monomers feed mixture (Solution B): butyl methacrylate (11.0 g, 77.46 mmol), 1H,1H,2H,2H-perfluorodecyl methacrylate (12.4 g, 23.3 mmol) and maleic anhydride (1.6 g, 16.33 mmol) were dissolved in a mixture of 2-butanon (2.0 g, 2.5 mL) and 3-bis (trifluoromethyl) benzene (2.0 g, 1.45 mL)). In a two neck 100 mL round bottom flask equipped with a rubber septum and valve the solution was degassed three times by freeze–pump–thaw cycles and filled back with nitrogen. Subsequently the mixture was transferred into a 50 mL syringe and mounted on a perfusion pump. Preparation of the initiator feed solution (Solution C): 2,2’-azo-bis-isobutyronitrile (0.066 g, 0,4 mmol) was dissolved in mixture of 2-butanon and 1,3-bis (trifluoromethyl)benzene (1.0 g 1:1 vol:vol). The mixture was degassed in a 10 mL round bottom flask equipped with rubber septum by three freeze–pump–thaw cycles and filled back with nitrogen. The mixture was transferred into a 1 mL syringe and mounted on a perfusion pump. Continuous addition terpolymerization: The stock solution A was placed in an oil bath pre- heated to 65 °C and after 5 minutes the continuous addition of solution B and C was started. Solution B was added at a rate of 0.755 mL/h, solution C with a rate of 0.0205 mL/h by means of separate syringe pumps over a period of 2000 minutes (33 h 20 min) (For details on the addition rate, please see the Appendix to Chapter 3). After the addition period, the reaction was heated for another 7 hours to complete the reaction (post-addition phase). The reaction mixture was cooled to ambient temperature and the product was precipitated in methanol (500 mL). Yield of terpolymer: 23.9 g, 87%; Mn = 50500 1 g/mol, Mw/Mn = 2.15 (THF-SEC). H NMR (acetone-d6/Freon-113 (1:1, vol:vol), δ in ppm):

4.34 (m, 2H, –O–CH2–CH2–CF2–); 4.00 (m, 2H, –O–CH2–CH2–CH2–CH3); 3.60 (m, 3H, –

COO–CH3); 2.8 (s, Freon-113); 1.86 (m, 2H, –O–CH2–CH2–CF2–); 1.65 (m, 2H, –O–CH2–

CH2–CH2–CH3); 1.45 (m, 2H, –O–CH2–CH2–CH2–CH3); 0.98 (m, 3H, –O–CH2–CH2–CH2–

CH3). Signals in the range of 1–2 ppm overlapped with the signals of backbone protons. Elemental analysis-calculated: C 48.4; H 5.3; O 15.3; F 31.0 wt %; found: C 48.26; H 5.21; O 15.72; F 30.81 wt %. Terpolymer composition according to 1H NMR was: BMA 69 mol %; F8H2MA 21 mol %; MAH 10 mol %. According to elemental analysis the tercopolymer composition was: BMA 68.7; F8H2MA 21.4; MAH 9.9 mol %.

Chapter 3

1H,1H,2H,2H,-Perfluorodecyl Methacrylate, Butyl Methacrylate and Maleic Anhydride. (5- Fold Monomer Excess)

The reaction procedure was analogous to the process detailed for the 10-fold excess reaction. The composition of the required solutions is summarized in Table 2. Solution B was fed at a rate of 0.755 mL/h and Solution C was added at a rate of 0.0205 mL/h over a period of 1000 minutes (16 h 40 min). 1 Yield: 10.6 g (85% of theoretical). Mn = 65000, Mw/Mn = 1.8 (THF-GPC). H NMR

(acetone-d6/Freon-113 (1:1, vol:vol) δ in ppm): 4.35 (m, 2H, –O–CH2–CH2–CF2–); 3.99 (m,

2H, –O–CH2–CH2–CH2–CH3); 3.60 (m, 3H, –COO–CH3); 1.86 (m, 2H, –O–CH2–CH2–CF2–);

2.8 (s, Freon-113); 1.65 (m, 2H, –O–CH2–CH2–CH2–CH3); 1.46 (m, 2H, –O–CH2–CH2–CH2–

CH3); 0.97 (m, 3H, –O–CH2–CH2–CH2–CH3). Signals in the range of 1–2 ppm overlapped with the signals of backbone protons. Elemental analysis-calculated: C 48.2; H 5.3; O 15.5; F 31.0 wt %; found: C 48.38; H 5.20; O 16.08; F 30.34 wt %. The terpolymer composition according to 1H-NMR was: BMA 67mol %; F8H2MA 21mol %; MAH 12 mol %. The composition determined by elemental analysis: BMA 64.8; F8H2MA 20.0; MAH 15.2 mol %.

1H,1H,2H,2H,-Perfluorodecyl Methacrylate, n-Dodecyl Methacrylate and Maleic Anhydride. (5-Fold Monomer Excess)

The reaction procedure was analogous to the process detailed for the 10-fold excess terpolymerization of butyl methacrylate, 1H,1H,2H,2H-perfluorodecyl methacrylate and maleic anhydride. The composition of the required solutions is summarized in Table 2. Solution B was fed at a rate of 1.764 mL/h and Solution C was added at a rate of 0.0809 mL/h over a period of 500 minutes (8 h 20 min). 1 Yield: 11.25 g (90% of theoretical). Mn = 69000, Mw/Mn = 1.72. H NMR (CDCl3, δ in ppm): 4.19 (m, 2H, –O–CH2–CH2–CF2–); 3.86 (m, 2H, –O–CH2–CH2– (CH2)9–CH3); 3.60 (m,

3H, –COO–CH3); 2.8 (s, Freon-113); 1.82 (m, 2H, –O–CH2–CH2–CF2–); 1.54 (m, 2H, –O–

CH2–CH2–(CH2)9–CH3); 1.19 (m, 2H, –O–CH2–CH2–(CH2)9–CH3); 0.81 (m, 3H, –O–CH2–

CH2–(CH2)9–CH3). Signals in the range of 1–2 ppm overlapped with the signals of backbone protons. Elemental analysis-calculated: C 58.3; H 7.5; O 13.5; F 20.7 wt %; found: C 57.14; H 7.28; O 12.1; F 23.48 wt %. According to 1H-NMR the terpolymer composition was: DMA–60 mol %; F8H2MA–17 mol %; MAH-23 mol %. The composition as obtained from elemental analysis data: DMA– 57.7; F8H2MA–20.00; MAH–22.3 mol %.

Chapter 3

Table 2. Composition of the Stock- and Feed solutions for continuous addition terpolymerization of BMA/F8H2MA/MAH and DMA/F8H2MA/MAH (T = 65 °C)

mDMA mBMA mF8H2MA mMAH mAIBN mMEK mHFX

(g) (g) (g) (g) (g) (g) (g) Stock Solution (A) - 0.448 0.722 1.33 0.03 4.53 2.79 Monomer Feed Solution (B) - 5.5 6.2 0.8 --- 1.133 1.0 Initiator Feed Solution (C) - - - - 0.033 0.5 0.5 Stock Solution (A) 0.707 - 0.063 1.16 0.105 3.2 5.51 Monomer Feed Solution (B) 6.497 - 4.698 1.1 - 1.6 2.0 Initiator Feed Solution (C) - - - - 0.060 0.5 - BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate; MAH, maleic anhydride; DMA, n-dodecyl methacrylate; AIBN, 2,2’-azo-bisisobutyronitrile; MEK, methyl ethyl ketone; HFX, 1,3-bis (trifluoromethyl) benzene. For details on the calculated amounts, please see the example in Appendix to Chapter 3.

Monomethyl Maleate by Methanolysis of Maleic Anhydride

Maleic anhydride (2 g, 20.4 mmol) was dissolved in methanol (20 mL) and placed in a 50 mL round bottom flask, equipped with reflux condenser. After 2 hours, half of the reaction mixture was transferred into 50 mL round bottom flask, the solvent was evaporated under reduced pressure and the sample was analyzed by means of 1H NMR. The rest of the reaction mixture was refluxed for another 13 hours, dried and subsequently investigated by 1H NMR. 1 H NMR (acetone-d6, δ in ppm): 6.37 (2H, –CH=CH–); 3.71 (3H, –O–CH3).

Methanolysis of MAH Copolymers

Polymer sample (40 mg) was dissolved in either dry chloroform or Freon-113 (5 mL) and dry methanol (3 mL) was added. The mixture was placed in a 25 mL round bottom flask and heated under reflux for 24 h. Subsequently, the product was precipitated in a methanol/diethyl ether 1:2 (vol:vol) mixture (15 mL), filtrated and dried at 40 °C under vacuum for 12 hours.

3. Results and Discussion

Alfrey and Goldfinger [7] published the mathematic description of the kinetics of a terpolymerization reaction. When three monomers copolymerize, there are nine polymer chain growth reactions possible:

kij • • ~ Mi + M j → ~ M j (1) • R ij = kij [~ Mi ][M j ]

Chapter 3

· where ~Mi is the growing polymer chain with monomer i (i = 1,2,3) as an active chain end, Mj is the monomer j (j = 1,2,3), Rij is the reaction rate, kij is the reaction rate constant, and [x] is the concentration of x in the reaction. The consumption of the monomers in time is described by three kinetic equations:

dM  • • • − 1 = k [~ M ][M ] + k [~ M ][M ] + k [~ M ][M ] dt 11 1 1 21 2 1 31 3 1 (2)

dM  • • • − 2 = k [~ M ][M ] + k [~ M ][M ] + k [~ M ][M ] (3) dt 12 1 2 22 2 2 32 3 2

dM  • • • − 3 = k [~ M ][M ]+ k [~ M ][M ]+ k [~ M ][M ] dt 13 1 3 23 2 3 33 3 3 (4)

· where [~Mi ] is the concentration of growing polymer chain with monomer i as an active chain end, [Mi] is the concentration of monomer i, and kij is the reaction rate constant of the reaction

Rij. Usually by the radical polymerization, only the stationary state will be considered where the radical concentration is constant. In case of ternary copolymerization:

• • • • k12 [~ M1 ][M2 ] + k13 [~ M1 ][M3 ] = k21 [~ M2 ][M1] + k31 [~ M3 ][M1] (5)

• • • • k21 [~ M2 ][M1] + k23 [~ M2 ][M3 ] = k12 [~ M1 ][M2 ] + k32 [~ M3 ][M2 ] (6)

• • • • k31 [~ M3 ][M1] + k32 [~ M3 ][M2 ] = k13 [~ M1 ][M3 ] + k23 [~ M2 ][M3 ] (7)

The sequence of repeating units in the terpolymer can be predicted from the knowledge of the terpolymerization equations which require six copolymerization parameters of the three binary co-monomer pairs [7,8]. Based on that model (by inserting Equations (5)–(7) into Equations (2)–(4) and some transformations), the monomer incorporation ratios can be described by following equations (Equations (8)–(10)):

 M1  M2  M3    M2  M3  M1   + +  M1 + +  dM  r21  r31 r21  r32 r23  r31 r12 r13 1 =      (8) dM2   M1  M2  M3   M1  M3  M2   + +   + M2 +  r21  r31 r21  r32 r23  r31   r21 r13 

 M1  M2  M3    M2  M3  M1   + +  M1 + +  dM  r21  r31 r21  r32 r23  r31 r12 r13 1 =      (9) dM3   M1  M2  M3   M1  M2   M3   + +   + + M3  r13  r21 r12  r23 r13  r32   r31 r32 

Chapter 3

 M1  M2  M3   M1  M3  M2   + +   + M2 +  dM  r21  r31 r21  r32 r23  r32 r21 r23 2 =      (10) dM3   M1  M2  M3   M1  M2   M3   + +   + + M3  r13  r21 r12  r23 r13  r23   r31 r32  where [Mi] is the concentration of monomer i, and rij/rji is the copolymerization parameter of monomer pair i/j. One can fully predict the behavior of the terpolymerization of three monomers if the behavior of the three monomers has been determined pairwise. Unfortunately, above presented equations are not valid for all monomer systems.

In the case that one of the monomers—for example, maleic anhydride (M3)—cannot be homopolymerized, two of the copolymerization parameters become zero and a modified equation that contains five copolymerization parameters can be used [9]. The parameter set consists of the “binary” parameters r12, r21, r13 and r23 that are experimentally accessible from binary copolymerization experiments with the three monomer pairs, as well as the parameter  that must be determined from the evaluation of a terpolymerization experiments (Equations (11)–(14)).

 M1  M2  M3   M2  M3  M1 ρ + + ρ  M1 + +  dM  r21 r21 r23 r12 r13 1 =      (11) dM2   M1  M2  M3  M1  M3  M2 ρ + +   + M2 +   r12 r12 r13   r21 r13 

 M1  M2  M3   M2  M3  M1  ρ  + + ρ   M1 + +  dM  r21 r21 r23 r12 r13 1 =      (12) dM3   M1  M2  M3   ρ M1 + M2  M3   + +  r13  r21 r12  r23 r13  r23 

 M1  M2  M3  M1  M3  M2 ρ + +   + M2 +  dM  r12 r12 r13 r21 r23 2 =      (13) dM3   M1  M2  M3   ρM1 + M2  M3  + +  r13 r21 r12 r23 r13 r23 

k ρ = 31 (14) k32 where [Mi] is the concentration of monomer i, rij/rji is the copolymerization parameter of monomer pair i/j,  is the terpolymerization parameter, M1 is RHMA, M2 is RFMA, and M3 is MAH.

Chapter 3

The  parameter, as all the copolymerization parameters, must be a positive number and must be valid for full range of ternary monomer compositions. The simplest way to determine the ρ parameter is to perform a terpolymerization reaction at the equal monomer fractions ([M1]

= [M2] = [M3]). If one determines the composition of such a copolymer, the following is valid:

b  e − q  d ρ = 12 (15) q12  c − a  e 1 1 1 1 1 1 d[M ] F where , , , , 1+ b , and 1 1 and Fi is the fraction of a = + b = c = d = + e = q12 = = r r r r r r 1+ a d[M ] F 12 23 21 12 12 13 2 2 , monomer i incorporated in the copolymer.

Terpolymers with succinic anhydride, alkyl (RH) methacrylate and perfluoroalkyl (RF) methacrylate repeating units (Scheme 1) were designed to combine the outstanding properties of fluorinated polymers with the adhesion promoting properties of polymers containing anhydride and alkyl side groups. Such macromolecules were prepared by free radical terpolymerization of alkyl methacrylates (M1 = RHMA), perfluoroalkyl methacrylate (M2 =

RFMA) and maleic anhydride (M3 = MAH). In this study butyl- (RHMA = BMA) and dodecyl methacrylate (RHMA = DMA) were used as alkyl methacrylates, while 1H,1H,2H,2H- perfluorodecyl methacrylate (RFMA = F8H2MA) served as monomer to introduce perfluoroalkyl side chains.

CH3 CH3

H2 H2 * C C *

O O O O O O O

RH RF x y x n

Scheme 1. Structure of the target P[RHMA-co-RFMA-co-MAH]n terpolymers (x + y +

z = 1, RH = C4H9–, C12H25–, RF- = C10H4F19–).

The preparation of well-defined terpolymers from alkyl methacrylates, perfluoroalkyl methacrylates and maleic anhydride (MAH) with uniform microstructure requires the knowledge of copolymerization parameters: the relation between the ratio of monomers in the feed and the ratio of repeating units in the resulting terpolymer as a function of time (Scheme 2).

Chapter 3

CH3

CH3 AIBN O + * CH2 C CH CH

O O O O O O O O O RH

R H x 1-x b

Scheme 2. Investigated binary copolymers: (a) P[MAH-co-F8H2MA]; (b) P[MAH- co-RHMA]; and (c) P[RHMA-co-F8H2MA] with RH = C4H9– (= BMA), C12H25– (= DMA) and RF– = C10H4F19– (= F8H2MA).

The respective copolymerization parameters to the best of our knowledge are not known. To obtain the required values, a series of low conversion copolymerization experiments were performed to determine the conversion of the monomers with time. To evaluate the microstructure of the copolymers, the fraction of the monomers incorporated was determined by means of 1H NMR spectroscopy at monomer conversions up to 10%. The MAH content of the copolymers could not be determined accurately by measuring the 1H NMR spectrum of the crude product; the MAH-proton signals are broad and overlap with signals of the backbone. However, conversion of the anhydride to a monomethyl ester followed by 1H NMR analysis and integration of the methyl ester signal an accurate determination of MAH repeating units was possible (for details, see Appendix to Chapter 3).

Low Conversion Copolymerization Experiments with BMA/MAH, F8H2MA/MAH, and F8H2MA/BMA Monomer Pairs, Determination of the Copolymerization Parameters

The experimental data for the copolymerization of the binary comonomer pairs BMA/MAH, F8H2MA/MAH, and F8H2MA/BMA (for two different monomer compositions) in all cases show a linear increase of the total monomer conversion with time up to conversions

Chapter 3

of ~12–20%, allowing to determine the initial weight rates of polymerization RPw,0 (Figure 1, Table 3).

The highest rate of polymerization was found with the monomer pair BMA/MAH (fMAH =

0.45: RPw,0 = 0.66 wt %/ min), followed by F8H2MA/BMA (fBMA = 0.45: RPw,0 = 0.5 wt %/min) while F8H2MA/MAH was four times slower (fMAH = 0.3: RPw,0 = 0.17 wt %/min). In F8H2MA/MAH mixtures the rate of polymerization decreased with increasing content of MAH in good accordance to prior observations [10]. However, in BMA/MAH mixtures between fMAH

= 0.45 and fMAH = 0.6, virtually no change in the rate of polymerization was found. This result is not in agreement with the experimental data for acrylates, but was reported in the literature for copolymerization of MAH with methyl methacrylate [11, 12].

5 Total MonomerTotal Conversion wt% /

0 0 10 20 30 40 50 60 70 80 Reaction Time / min Figure 1. Time conversion plots of binary copolymerization experiments with the comonomer pairs BMA/MAH (: fMAH = 0.35, : fMAH = 0.60), F8H2MA/MAH (:fMAH = 0.3, : fMAH = 0.6) and F8H2MA/BMA (:fBMA = 0.2, : fBMA = 0.45).

Table 3. Monomer mixture in the feed, copolymer compositions and initial weight rate of polymerization of the three monomer pairs MAH/F8H2MA, MAH/BMA and BMA/F8H2MA.

−1 Monomer 1 Monomer 2 f1 F1 RPw,0/wt % min MAH F8H2MA 0.30 0.080  0.005 0.168  0.008 MAH F8H2MA 0.60 0.195  0.005 0.133  0.006 MAH BMA 0.45 0.095  0.005 0.661  0.009 MAH BMA 0.60 0.140  0.005 0.651  0.008 BMA F8H2MA 0.20 0.195  0.005 0.589  0.04 BMA F8H2MA 0.45 0.440  0.01 0.499  0.02 MAH, maleic anhydride; BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl

methacrylate; RPw,0, initial weight rates of polymerization. As mentioned, the composition of the copolymers was determined by 1H NMR spectroscopy. In Figure 2, typical 1H NMR spectra of the binary copolymers are shown. In the

Chapter 3

spectrum of P[BMA0.9-co-MAH0.1] (Figure 2a), the peak of the butyl ester –COOCH2– protons is well resolved at a chemical shift of  = 4.05 ppm and a very broad signal between 2.7 and 3.5 ppm is caused by one of the succinic anhydride backbone protons. Subsequent to methanolysis (Figure 2b) a new signal appears at  = 3.7 ppm, caused by the methyl ester group which is well separated from the butyl ester signal. According to Equation (16), 10 mol % MAH units are incorporated into the copolymer. Comparing the spectra of the MAH copolymer and the copolymer after methanolysis it was found that a methyl ester signal of low intensity was already present. Obviously a small fraction (~10%) of the anhydride rings have already been hydrolyzed during the isolation of the polymer by means of precipitation in cold methanol, hence the freshly prepared copolymer is correctly described by the formula P[BMA0.9-co-

MAH0.09-co-monomethylmaleate0.01]. For the further purpose of this work, such a minor degree of methanolysis upon polymer preparation can be tolerated and, in the subsequent text, the copolymers that have not deliberately been boiled with methanol will be treated as pure MAH- copolymers.

A /ν F = i i i n (16) A j/ν j j=1

1 where Ai is the integrated intensity of the H NMR signal of a selected structural element of monomer i, and i is the number of protons in the selected structural element of monomer i. The 1H NMR analysis of P[BMA-co-F8H2MA] (Figure 2c) does not require any pretreatment. The signals at 4.34 ppm and 4.03 ppm are assigned to 2H, –O–CH2–CH2–(CF2)8–F) and 2H, –O–CH2–CH2–CH2–CH3, respectively, are well separated and the composition of the copolymer can easily be calculated. Figure 3 summarizes the copolymerization diagrams of the three comonomer pairs. The copolymerization line of F8H2MA/BMA is close to an ideal random copolymerization and the numerical fit of the integrated Lewis–Mayo equation yielded rF8H2MA = 1.02 and rBMA = 0.94 as most probable values for the copolymerization parameter. Both MAH copolymer systems exhibit non-ideal copolymerization behavior, characterized by a value of zero for the MAH copolymerization parameter in both cases. Not more than 50 mol % of MAH can hence be incorporated in the copolymers under such circumstances. The obtained copolymerization parameters were rF8H2MA = 4.9, rMAH = 0 and rBMA = 8.2, rMAH = 0. These values are in line with the observed sequences, the total rate of polymerization

Chapter 3

F8H2MA/MAH BMA/MAH (RPw,0 > RPw,0 ) and with other published copolymerization data for MAH/comonomer systems [13].

-CF -CH -CH O-CO- 2 2 2 -CH O-CO- 2

(c)

CH O-CO- 3

(b)

(a) 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 Chemical Shift  / ppm 1 Figure 2. H NMR spectra of: (a) P [BMA0.9-co-MAH0.1]; (b) P[BMA0.9-co- monomethylmaleate0.1] obtained by methanolysis of (a); and (c) P[BMA0.44-co- F8H2MA0.56].

1.0 1.0

0.9 0.9

i i 0.8 0.8

0.7 0.7

0.6 0.6

0.5 0.5

0.4 0.4

0.3 0.3

0.2 0.2

copolymer composition F copolymer composition F 0.1 0.1

0.0 0.0 0.0 0.2 0.4 0.6 0.8 1.0 monomer mixture composition f i Figure 3. Copolymerization diagrams of the three binary systems F8H2MA/BMA (, i = F8H2MA), MAH/F8H2MA (, i = MAH) and MAH/BMA (, i = MAH) at 65 °C.

Terpolymerization Experiments of the BMA/F8H2MA/MAH System

As already mentioned before the obtained results of binary copolymerization of each pair of monomers is not sufficient to describe the behavior of the ternary system in which one of the monomers cannot homopolymerize, hence the necessary data for the ρ-parameter need to be generated by measurements from the ternary monomer mixture. The value of ρ calculated according to Equation (15) is 2.87 (see Appendix to Chapter 3, “Calculation of ρ -parameter”). To determine the reaction rates of BMA/F8H2MA/MAH mixtures, terpolymerization experiments were performed at three different monomer compositions (Table 4). The 50

Chapter 3 composition of the obtained terpolymers was determined by 1H NMR spectroscopy. In any case, the monomer conversion was below 15 mol % to ensure a constant monomer composition during the reaction. The reactions were carried out in homogenous solution of HFX:MEK (1:1, vol:vol) with AIBN as the initiator at 65 °C.

Table 4. Composition of monomer mixtures and terpolymers, as well as initial weight rates of polymerizations measured with BMA/F8H2MA/MAH.

fBMA FBMA fMAH FMAH fF8H2MA FF8H2MA Rp (wt %/min)

(1) 0.33 0.45 0.33 0.06 0.33 0.49 0.73 (2) 0.25 0.44 0.50 0.09 0.25 0.47 0.65 (3) 0.175 0.65 0.75 0.14 0.075 0.21 0.47 MAH, maleic anhydride; BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl

methacrylate; Rp, initial weight rates of polymerization; fi, molar fraction of the monomer in the feed;

Fi, molar fraction of the monomer incorporated into polymer.

As depicted in Figure 4a, in all cases, the total monomer conversion increased linearly up to conversion of around 20 wt %. In parallel to the binary copolymerization, the presence of MAH reduced the rate of polymerization also in terpolymerization experiments. As for binary copolymers, the presence of maleic anhydride results in a copolymer composition that is different from the composition of monomers in the feed. The differences are depicted in Figure 4b. The arrows in the diagram show the change of the terpolymer composition against the monomer composition in the feed.

(1) 35 (2) 30

25 (3) 20

15 Conversion% / 10

0 0 10 20 30 40 50 Reaction time / min (a)

Chapter 3

BMA 0,00 1,00

0,25 0,75

f MAH BMA

, F , F 0,50 f MAH 0,50 BMA

0,75 0,25

1,00 MAH 0,00 0,00 0,25 0,50 0,75 1,00 F8H2MA f , F F8H2MA F8H2MA

(b) Figure 4. (a) Conversion time plots with BMA/F8H2MA/MAH mixtures ((1)  = 1:1:1, (2)  = 1:1:2, (3)  = 1.75:0.75:7.5); and (b) the corresponding terpolymerization diagram (the arrows connect monomer and terpolymer composition).

Continuous Addition Copolymerization

For the copolymerization of monomers that cannot undergo homopolymerization or that are consumed with different rates, with increasing conversion, the reaction mixture is enriched in the less reactive monomers. The change in the monomer composition with time leads to changing polymer composition. To avoid this phenomenon, the reaction can be stopped at low conversion (usually lower than 10 mol %). In this case the monomer composition is considered to be constant and the variation in copolymer composition is negligible [14]. The huge disadvantage of such a policy is that approximately 90%–95% of the monomer mixture will be lost. Hence, this method can only be applied to analytic investigations, such as the determination of copolymerization parameters. For preparative purposes—in particular, for large industrial production—the loss of large quantities of expensive educts cannot be tolerated. The alternative to low conversion is to deliver the monomers to the reaction mixture with the rate they are consumed. In this way, it is possible to avoid both non-homogenous composition of the desired polymer, and non-reacted monomer mixture as a waste. The theory of continuous addition polymerization with examples has been described in the literature [15] and is summarized in the Appendix to Chapter 3.

To calculate the addition rate, it is necessary to know the copolymerization parameters ri as well as the rate of reaction. When these data are not available in the literature, their determination requires huge effort as demonstrated above. The determination of the r

Chapter 3 parameters requires series of binary copolymerization and, in the case one of the monomers cannot homopolymerize, additional ternary experiments are necessary to determine the ρ parameter and employ the modified equations. Although the composition of the terpolymer can be calculated, the reaction rate is influenced by many different factors e.g. the type of solvent (presented further on in this text, see Table 8) and need to be determined in a separate ternary copolymerization experiment which is specific for a chosen copolymerization conditions. Due to the complexity of this procedure, in this work, the following simplified procedure was applied: (i) for a chosen monomer composition, the reaction rate is determined from the conversion vs. time plot; and (ii) the composition of the copolymer is determined for a conversion lower than 10%. (The reactions were performed twice to determine the margin of error. It is of paramount importance to mention that the results obtained are specific for the chosen reaction conditions: monomer composition, monomer and initiator concentration, type of initiator and solvent, temperature etc.) Based on these results, the amounts of monomers, the initiator and the rate of addition are determined. To understand principles of this method we have to define term-“stock solution”. A “stock solution” is a reaction mixture at the time t0 = 0, that contains the monomer mixture of the initial concentration C0, and the initial mass of monomers m0. On the base of these characteristics of the stock solution, the reaction rate and the composition of the copolymer, the amount and rate of addition of monomers and of initiator are calculated. To calculate the reaction time, the so-called “excess α”—meaning the total mass of monomers that has to be added (usually as the multiple of the mass m0 in the stock solution)—has to be defined. For continuous terpolymerization with a post addition phase (to complete the conversion of unreacted monomers), 10 times excess (α) (10 × m0) has to be added to reduce its influence on the composition of the copolymer.

Continuous Addition Terpolymerization of 1H,1H,2H,2H-Perfluorodecyl Methacrylate (F8H2MA), Butyl Methacrylate (BMA) and Maleic Anhydride (MAH) (Copolymers C1 and C2)

For the continuous addition experiment the monomer stock solution contained 75 mol % of MAH, 17.5 mol % of BMA and 7.5 mol % of fluorinated monomer (MAH:BMA:F8H2MA = 7.5:1.75:0.75). With such a monomer mixture experiments up to low conversion yielded copolymers (see Scheme 3) with acceptable composition with respect to MAH content and F8H2MA content of 14 and 21 mol % respectively. All parameters necessary to perform the continuous addition experiment were calculated according to the described method (for the details, see Experimental Section and Appendix to Chapter 3).

Chapter 3

CH CH CH3 CH3 3 3

* CH2 C CH2 C CH CH * AIBN O + O + O O O O O O O O O O O O

Bu Rf Bu Rf x y z

Rf = CH2CH2(CF2)7CF3 Scheme 3. Copolymerization of 1H,1H,2H,2H-perfluorodecyl methacrylate with butyl methacrylate and maleic anhydride (x + y + z = 0.21 + 0.65 + 0.14 = 1).

The copolymerization was carried out in a homogeneous mixture of 2-butanone (MEK) with hexafluoroxylene (HFX) 1:1 (vol:vol) and AIBN was used as initiator at 65°C (Figure 5).

SYRINGE PUMP SYRINGE PUMP Monomers Initiator stock solution

65 C

Figure 5. Typical set-up for the continuous addition experiment.

In the first experiment (copolymer C1), a 10-fold excess (α = 10) of monomer with respect to the stock solution was added to decrease the influence of the post addition phase on the homogeneity of the product. The calculated addition time for this experiment was 2000 minutes (33.3 hours). For such a long addition time it is impossible to neglect the change of the initiator concentration in time since at the temperature of the process (65 °C) the half-life time of AIBN is 10 hours. For this reason, the initiator had to be continuously added as well (all calculations are available in the Appendix to Chapter 3). To avoid the polymerization of the feed mixture, the solution of the initiator was added separately. Under the reaction conditions even at high monomer conversions no precipitation of the polymer was observed, however the viscosity increased significantly and caused problems with stirring. Such behavior is reasonable for solution containing more than 60 wt % of polymer. To monitor the progress of the reaction and investigate if the system behaves as initially expected, samples were taken and analyzed during polymerization. The composition of the samples was determined by 1H NMR spectroscopy (see Figures S1 and S2) using the method described for low conversion experiments. It was observed that the incorporation of MAH decreased with increasing conversion. Nevertheless, incorporation of fluorinated methacrylate remains constant up to full monomer conversion (see Table 5). The composition drift is depicted in Figure 6. In the case that the low conversion experiment and calculations were done properly, theoretically, no change of the copolymer composition should occur in time.

Chapter 3

Table 5. Compositions of the copolymer samples taken during the continuous addition experiment (copolymer C1). Polymer composition (1) (mol %) Reaction time (h) Conversion (%) F8H2MA BMA MAH

3 8.2 22 64 14 15 40.9 21 65,5 13,5 27 73.6 21 69 11 33 + 7 (2) 90.9 (3) 21 69 10 Theoretical composition (4) 21 ± 2 65 ± 2 14 ± 2 MAH, maleic anhydride; BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate. (1) Determined by 1H NMR spectroscopy; (2) 7 h after the post-addition phase; (3) Conversion calculated for 33 h; (4) As determined from low conversion experiment.

P / % End of addition 100 P

F / mol% i 80

BMA 60

F8H2MA 20

MAH

0 0

-5 0 5 10 15 20 25 30 35 40 45 t / h

Figure 6. Change of the composition of C1 (P [F8H2MA-co-BMA-co-MAH]) terpolymer in time during the continuous addition experiment. P represents the conversion plot.

The low deviation of incorporated monomers (dF/dt) and its constancy in time seems to be caused by a minimal error in monomer addition rates.

dF F 33h − F 3h 10 −14 dF F 33 h − F 3h MAH = MAH MAH =  −0.12 mol %/h; BMA = BMA BMA  +0.13 mol %/h; dt t 33 dt t

dF F 33h − F 3h F8H 2MA = F8H 2MA F8H 2MA  −0.03 mol %/h dt t

Since the measurement of Rp is prone to an error of ~10–20%, the achieved accuracy of ~0.1%/h is a reasonable result. The influence of different factors on the possible deviations of terpolymer composition will be discussed later. The composition of the final products was determined by proton NMR and confirmed by elemental analysis (Table 6).

Chapter 3

Table 6. Comparison between the elemental compositions calculated based on magnetic resonance (1H-NMR) and measured by elemental analysis (EA). Carbon Hydrogen (wt Oxygen (wt Fluorine (wt Copolymer Excess α (wt %) %) %) %) 1H NMR C1 10 48.44 5.35 15.31 30.90 EA C1 10 48.26 5.22 15.71 30.81 1H NMR C2 5 48.19 5.26 15.51 31.04 EA C2 5 48.38 5.21 16.08 30.33

The slight compositional drift in experiment with 10-fold excess of the monomers could be caused by multiple factors: (i) high viscosity of the reaction mixture and as consequence insufficient stirring; (ii) the so called post-addition phase in which monomers are not supplied to the reaction mixture and the consumption of more reactive monomers; and (iii) incorrect dosage of monomers. To assess the effect of the two first factors, the experiment with five-fold monomer excess was repeated without post-addition phase (copolymer C2). The composition of the polymer samples taken during the addition period are presented in Table S2. The composition of the final product as a function of the excess α was determined by 1H NMR and elemental analysis (Table 7). The amount of incorporated fluorinated methacrylate remains the same while the incorporation of anhydride decreases with the increase of the monomer excess. This clear correlation may be treated as a proof of the reproducibility and robustness of the method but also indicates how important the accuracy of low conversion experiments is.

Table 7. Copolymer composition calculated from 1H NMR and elemental analysis (EA) at different monomer excess and low conversion experiment (α = 0). F8H2MA BMA MAH Copolymer Excess α (mol %) (mol %) (mol %) 1H-NMR C1 10 21.00 69.00 10.00 EA C1 10 21.37 68.68 9.95

1H-NMR C2 5 21.00 67.00 12.00

EA C2 5 20.05 64.77 15,18 1H-NMR - 0* 21.00 65.00 13.00 MAH, maleic anhydride; BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate. * Low conversion experiment.

Continuous Addition Terpolymerization of 1H,1H,2H,2H-Perfluorodecyl Methacrylate (F8H2MA), Dodecyl Methacrylate (DMA) and Maleic Anhydride (MAH) (Copolymer C3)

To obtain desired properties (e.g., good level of water and oil repellence) of the terpolymer that contains both maleic anhydride and perfluorinated methacrylate, butyl methacrylate was replaced by dodecyl methacrylate (Scheme 4).

Chapter 3

CH CH CH3 CH3 3 3

* CH2 C CH2 C CH CH * AIBN O + O + O O O O O O O O O O O O

R Rf R Rf x y z

R = CH CH (CF ) CF f 2 2 2 7 3 R=(CH2)11CH3 Scheme 4. Copolymerization of 1H,1H,2H,2H-perfluorodecyl methacrylate with dodecyl methacrylate and maleic anhydride (x + y + z = 0.20 + 0.57 + 0.22 = 1).

This way, one could additionally prove the principles of continuous addition copolymerization working also when other monomers are used. In the first step, low conversion experiments were performed according to the previously described method (see experimental part). Two different monomer compositions were investigated (see Table 8), whereas the initiator concentration was kept constant at 4 mol%.

Table 8. P(DMA-co-F8H2MA-co-MAH). Monomer composition in the feed and copolymer composition and the initial reaction rates at low conversion.

Rp Solvent fF8H2MA fDMA fMAH FF8H2MA FDMA FMAH (%/min)

MEK/HFX 0.075 0.175 0.75 0.21 0.55 0.24 1.0 MEK/HFX 0.04 0.21 0.75 0.13 0.62 0.25 0.82 MEK 0.075 0.175 0.75 0.21 0.54 0.25 0.6

MAH, maleic anhydride; DMA, dodecyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl

methacrylate; MEK, methyl ethyl ketone; HFX, 1,3-bis(trifluoromethyl) benzene; RP, weight rate of

polymerization; fi, molar fraction of the monomer in the feed; Fi, molar fraction of the monomer incorporated into polymer.

Because of the low content of the fluorinated monomer in the copolymer the solvent mixture of HFX and MEK was replaced with pure MEK as a solvent to eliminate problematic and expensive fluorinated solvent. The composition of the obtained terpolymer was identical with the one originating from the mixture of solvents. However, the reaction rate was about two times lower in pure MEK than in the mixture of solvents (Figure 7). In conclusion, the composition of the obtained copolymers is identical within the error range for both type of solvents, the reaction rate in pure MEK, however is 40% lower than in MEK/HFX mixture.

Chapter 3

Conversion% / 10

0 0 5 10 15 20 25 30 Time / min

Figure 7. Terpolymerization of DMA, F8H2MA and MAH: time vs. conversion plot for the reaction in MEK (▪) and MEK:HFX 1:1 vol:vol (•).

For the continuous addition experiment the composition with higher amount of fluorinated monomer was chosen because of higher reaction rate in MEK:HFX 1:1. The determined reaction rate and the composition of copolymer of low conversion experiment were used as base for calculation of the addition rates and composition of the monomer mixture in continuous addition experiment.

Figure 8. Change of the composition of C3 (P[F8H2MA-co-DMA-co-MAH]) terpolymer in time in continuous addition experiment. P, conversion; F copolymer composition; MAH, maleic anhydride; DMA, dodecyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate.

The feed rates were calculated from theory, based on experimentally measured rates. Due to unavoidable experimental errors, it is hence to be expected that the real feed-rates will slightly deviate from the calculated ones. These mismatches do not play a significant role in

Chapter 3 short-time experiments, but can become visible in long-time, high conversion reactions. In the present experiment, it was found that certain composition drift occurred during the reaction (Figure 8). Based on it feeding errors have been calculated on the base of NMR data and appear to be higher than in case of BMA copolymer. In addition, the correlation of the composition obtained by NMR spectroscopy and elemental analysis is lower than in the BMA case (see Tables A2 and A3). It appears that the monomers were misdosed as calculate below:

dF F 8h − F 0h 24 − 23 dF F 8h − F 0h MAH = MAH MAH =  −0.2 mol %/h; DMA = BMA BMA  +0.3 mol %/h; dt t 8.33 dt t dF F 8h − F 0h F8H 2MA = F8H 2MA F8H 2MA  −0.1 mol %/h dt t Based on the NMR analysis, only the content of DMA has been overestimated; simultaneously, the content of fluorinated methacrylate is higher than determined by magnetic resonance. Comparison between low conversion and continuous addition experiments showed that incorporation of maleic anhydride slowed down slightly during reaction. The copolymer composition calculated from elemental analysis (EA) and 1H NMR spectroscopy is presented in Table 9.

Table 9. Comparison of the copolymer C2 composition calculated from elemental analysis (EA) and 1H-NMR data for continuous addition and low conversion (α = 0).

Excess F8H2MA DMA MAH

α (mol %) (mol %) (mol %) 1H NMR 5 17.00 60.00 23.00

EA 5 20.00 57.73 22.27

1H NMR 0 21.00 55.00 24.00 MAH, maleic anhydride; DMA, dodecyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate. Because in the case of low conversion experiment, the composition of polymers was determined only by NMR and has not been checked by elemental analysis, one can assume that compositions of these polymers are comparable to that obtained in continuous addition experiments.

Characterization of the Terpolymers

The molecular weights of synthesized terpolymers were determined by means of GPC in THF as solvent using PMMA calibration standards (Table 10).

Chapter 3

Table 10. Molecular weights of the copolymers for different value of excess α.

Copolymer α Mn Mw Đ

BMA/F8H2MA/MAH 0* 68 000 108 000 1.59 C1 5 63 000 117 000 1.85 C2 10 50 500 120 000 2.38 DMA/F8H2MA/MAH 0* 71 500 134 000 1.88 C3 5 69 000 118 000 1.71

BMA, butyl methacrylate; F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate; MAH, maleic

anhydride; DMA, dodecyl methacrylate; * α = 0; low conversion experiment; C1 (poly[BMA0.69-co- F8H2MA0.21-co-MSA0.1]); C2 (poly[BMA0.67-co-F8H2MA0.21-co-MSA0.12]); C3 (poly[DMA0.57-co-

F8H2MA0.2-co-MSA0.22]).

In the GPC elugram of the BMA, copolymer a shoulder towards low molecular weight appeared. It is explicitly visible in the elugram of copolymer obtained in 10-fold excess (α = 10) experiment (Figure A5). In this particular case, the viscosity of the reaction mixture became relatively high, and significantly higher than in α = 5 experiment where the GPC shoulder is hardly visible. In this experiment, as it was assumed, there was no problem with stirring caused by high viscosity, and the GPC elugram also shows almost symmetric distribution (see Figure A5). Similar effect on the molecular weight distribution has been observed in the case of DMA copolymer. Despite the fact that the experiment was performed with excess of five-fold of monomer, the influence of the viscosity on molecular weight is even more pronounced than in case of BMA (see Figure S6).

Thermal properties

The thermal behavior of synthesized terpolymers was investigated by means of therogravimetric analysis (TGA) and differential scanning calorimetry (DSC). Thermogravimetic curves of investigated terpolymers show degradation in one step. The thermal stability of the terpolymers is relatively high. A weight loss of 5% has been observed at temperatures of around 300 °C. No significant difference can be noticed with respect to the length of the aliphatic chain in the methacrylate (see Figure S7). The DSC measurements of ternary copolymers (see Figure A8) show only glass transition temperatures and no melting points (Table 11). The single transition is visible, both in the first as well as in the second heating run. The lack of crystallinity in ternary copolymers comparing to binary copolymers with high content of fluorinated methacrylate was also reported by Kraus [16]. The relatively

Chapter 3 low content of rigid perfluorinated chains which are prone to form ordered structures combined with the presence of “soft” alkyl chains can perfectly explain the experimental data. An increase of the maleic anhydride content in both butyl methacrylate copolymers increases the Tg value what is in unison with literature data [17–20] and is explained by the lower flexibility of the polymer chain. The presence of long alkyl side chains in the copolymer of DMA results in glass transition temperature lower by 20 °C. The lower glass transition is observed despite the fact that relative amount of maleic anhydride and fluorinated moieties compared to alkyl methacrylate is higher than in butyl methacrylate copolymers. The values of Tg are collected in Table 11.

Table 11. Composition and glass transition temperatures (second heating) of ternary copolymers obtained in continuous addition experiment. Composition according to elemental analysis.

FF8H2MA FAlkyl-MA FMAH Tg C1 21 69 10 56 C2 20 65 15 58 C3 20 58 22 29

MAH, maleic anhydride; Alkyl-MA, butyl methacrylate (C1 (poly[BMA0.69-co-F8H2MA0.21-co-

MSA0.1]), C2 (poly[BMA0.67-co-F8H2MA0.21-co-MSA0.12]); dodecyl methacrylate (C3 (poly[DMA0.57- co-F8H2MA0.2-co-MSA0.22]); F8H2MA, 1H,1H,2H,2H-perfluorodecyl methacrylate; Fi, molar fraction of the monomer incorporated into polymer.

4. Conclusions

To prepare larger quantities of homogenous terpolymers, the desired composition of the terpolymer can be calculated by means of terpolymerization equations. In the case of terpolymer consisting of butyl methacrylate (BMA), 1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA) and maleic anhydride (MAH), the copolymerization parameters could not be found in the literature and had to be determined by investigating copolymerization of pairs of monomers. All the experiments were carried out under homogenous conditions in 1:1 (vol:vol) mixture of 2-butanone (MEK) and hexafluoroxylene (HFX) with 2 mol % AIBN as initiator at the temperature of 65 °C. The temperature has been arbitrary chosen as the half-life time of

AIBN at this temperature t1/2 = 10 hours. The total monomer concentration was 2.5 mol/L. The determined copolymerization parameters were: rF8H2MA = 4.9 and rMAH = 0 with the

F8H2MA/MAH system; rBMA = 8.2 and rMAH = 0 upon polymerization of BMA/MAH mixtures; and rF8H2MA = 1.02 and rBMA = 0.94 of F8H2MA/BMA. It has been demonstrated that all the necessary data for continuous addition terpolymerization can be extracted from low conversion ternary experiment but one needs to

Chapter 3 keep in mind that they are valid only for specific reaction conditions in term of monomer mixture composition, monomer concentrations, concentration and type of radical initiator and reaction temperature. The influence of the solvent on the reaction rate also needs to be considered. Under the chosen reaction conditions, the rate for monomer mixture 1.75:0.75:7.5 BMA/F8H2MA/MAH was Rp = 0.47 wt %/min and for 1:1:1 BMA/F8H2MA/MAH Rp = 0.73 wt %/min. The composition of the reaction mixture has clear influence on the reaction rate. The determined reaction rates and the composition of the terpolymers were used to perform successfully continuous addition experiments to produce larger quantities of homogenous terpolymers. The versatility of the method has also been proven for a different set of monomers namely dodecyl methacrylate (DMA), 1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA) and maleic anhydride (MAH). In this case, terpolymer of uniform composition has also been prepared. All the necessary parameters for the continuous addition experiment of DMA/F8H2MA/MAH system were determined in low conversion terpolymerization experiments. Synthesized copolymers were characterized in terms of molecular weight and thermal properties. DSC measurements of all terpolymers showed glass transition temperatures and absence of melting temperatures. In the case of DMA copolymer, the measured Tg value was lower than in the case of shorter side alkyl chain (BMA), even though double the amount of anhydride moiety was incorporated into polymer chain. The developed method of synthesis of large quantity of homogenous terpolymers with relatively low fluorine content opens multiple opportunities for utilization of new coating materials and additives for surface treatment.

References [1] E. Kissa, Fluorinated Surfactants, Synthesis-Properties Application, ed.; M. Dekker: New York, NY, USA, 1984; pp. 82-84, 362-364, ISBN: 0824790111. [2] T.F. DeRossa, B.J. Kaufman, R.L.-D. Sung, J.M. Russo, Polym. Prepr, 1994, 35, 718. [3] A. Chen, D. Liu, Q. Deng, X. He, X. Wang, J. Polym. Sci. A, Polym. Chem. 2006, 44, 3434 [4] J. Yu, B. Yi, D. Xing, F. Liu, Z. Shao, Y. Fu, H. Zhan, Phys. Chem. Chem. Phys. 2003, 5, 611 [5] V. Arcella, A. Ghielmi, G. Tommasi, Ann. N. Y. Acad. Sci. 2003, 984, 226–244. [6] WHO. Fluorides. In: Air quality guidelines for Europe, 2nd ed. Copenhagen, Regional: Office for Europe, 2000, pp. 143, ISBN: 92 890 1358 3. [7] T. Alfrey, G. Goldfinger, J. Chem. Phys., 1944, 12, 205 [8] U. Beginn, e-Polymers, 2005, 5, 073

Chapter 3

[9] G. Odian Experimental evaluation of monomer reactivity ratios. In Principles of polymerization, 6th ed.; Editor 4, Wiley Interscience: New York ,USA,2004. [10] M. Raetzsch, M. Arnold, J. Macromol. Sci. A, 1987, A24, 507 [11] C. Caze, C. Loucheux, J. Macromol. Sci A, 1978, A12, 1501 [12] D.C. Blackley, H.W. Melville, Macromol. Chem, 1956, 18, 16 [13] Brandrup J.; Immergutt E. H. Free radical copolymerization reactivity ratios (page II/153). In Polymer Handbook, Wiley, New York, 1999. [14] I. Skeist, J. Am. Chem. Soc., 1946, 68, 1781 [15] U. Beginn, e-Polymers 2005, 5. [16] Kraus, M. Functional combcopolymer for superhydrophobic and superhydrophilic surfaces. PhD Thesis, Ulm University, Ulm, Germany, 2002. [18] H. Yokohama, E.J. Kramer, D.A. Hajduk, F.S. Bates, Macromolecules, 1999, 32, 3353 [19] H. Yokohama, E.J. Kramer, Macromolecules, 2000, 33, 187 [20] H. Yokohama, E.J. Kramer, G.H. Fredrickson, Macromolecules, 2000, 33, 2249

Appendix to Chapter 3

Synthesis of Terpolymers with Homogeneous Composition by Free Radical Copolymerization of Maleic Anhydride, Perfluorooctyl- and Butyl- or Dodecyl-Methacrylates; Application of the Continuous Flow Monomer Addition Technique.

Determination of MSA units in copolymers MSA repeating units were converted in a functionality that allows quantitative determinations. In a model experiment maleic anhydride was reacted with an excess of methanol (Scheme1) and the reaction product was analysed by 1H NMR spectroscopy. The 1H NMR spectrum shows signals centred at 6.37 ppm and a singlet at 3.71 ppm. From the coupling pattern and the ratio of the signal intensity (2 : 3, respectively) the product was identified as monomethyl maleate. The absence of a singlet at 6.2 ppm and 7.05 ppm demonstrated the absence of dimethyl maleate.

reflux + CH3OH O O O O O OH O

CH3 Scheme A1: Methanolysis of maleic anhydride as model reaction

To measure the maleic anhydride content of a copolymer, the polymer sample was heated in solution in the presence of an excess of methanol (cf. experimental part) to open the anhydride ring in analogy to the reaction depicted in Scheme A1. Subsequently, the relative MSA content 1 was obtained by comparing the H NMR signal intensity of the -CO-OCH3 group at 3.6 – 3.7 ppm to that of the –CO-OCH2- signals of either F8H2MA, BMA or DMA. Table A1 summarises the chemical shifts and the number of protons per functional group of the four monomer units used in this study. The content of monomer i in the copolymer as expressed by its molar fraction Fi has been calculated according to Eq. A1.

Appendix to Chapter 3

Ai /ν i Fi = n (A1) A j/ν j j=1 Where: 1 (Ai = integrated intensity of the H NMR signal of a selected structural element of monomer i,

i = number of protons in the selected structural element of monomer i)

Table A1. Functional groups, 1H NMR chemical shifts and number of protons of these groups used in Eq. A1 to determine the copolymer compositions (reference signal: CHCl3, 7.26 ppm) Monomer Group  σ in ppm σ in ppm (solvent =

(solvent = CDCl3) Freon 113/CDCl3) 1) F8H2MA -CO-O-CH2-CH2-CF2- 2 - 4.20 – 4.40 BMA -CO-O-CH2- 2 3.90 – 4.10 3.95 – 4.15 DMA -CO-O-CH2- 2 3.75 – 4.00 3.80 – 4.10 MSA -CO-O-CH3 3 3.55 – 3.75 3.60 – 3.80 MSA = maleic anhydride, BMA = butyl methacrylate, F8H2MA = 1H,1H,2H,2H-perfluorodecyl methacrylate 1 1) H-NMR spectra of F8H2MA copolymers have not been measured in pure CDCl3

The binary copolymerisation parameter rij and rji of a monomer pair i/j were subsequently calculated from a series of copolymer compositions Fi arising from experiments with varying comonomer compositions fi. The r-values were obtained by numerical fit of the integrated Lewis-Mayo equation (terminal model, 2) to the data set by means of the program COPOINT [1].

2 ri  fi + fi  f j Fi = 2 2 (A2) ri  fi + 2 fi  f j + rj  f j

(Fi – molar fraction of monomer i in the copolymer, fi – molar fraction of monomer i in the reaction mixture, fj = 1 – fi)

Theory of continuous addition polymerization

The change of the monomer fraction fi with the conversion: df f − F i = i i for p → 0 (A3) dp 1− p

(fi – molar fraction of monomer “i” in the reaction mixture; Fi – molar fraction of monomer “i” incorporated in polymer; p – conversion)

Appendix to Chapter 3

df i  f − F (A4) dp i i

fi  ( fi − Fi )p (A5) for p → 0

fi → 0 it means that if the conversion change Δp is close to 0 also the change of the monomer fraction

Δfi in the mixture is close to 0.

With monomer addition: df f − F  − f dq i = i i + i i  = 0 (A6) dp q − p q − p dp

add n0 + n dqi / dp q = in ; i = ne dq / dp

(fi – fraction of monomer “i” in the feed; Fi – fraction of monomer “i” incorporated in the polymer; p – conversion; α – addition factor (excess); q – number of moles of monomers in the reaction mixture )

To calculate the addition rate it is necessary to know the copolymerisation parameters ri as well as the rate of reaction. When these data are not available in the literature, their determination requires huge effort. The advantage of the method described in this chapter consists on following, simple procedure. For the chosen monomer composition copolymerization reaction has been performed: - the reaction rate has been determined from conversion vs. time plot. - the composition of the copolymer was determined for conversion lower than 10 %. The reaction has been performed twice to eliminate the error. It is of paramount importance to mention that determined parameters are specific for chosen conditions: monomer’s composition, concentration of monomers and the initiator, type of the initiator, temperature and solvents. To understand principles of this method we have to define term- “stock solution”. A “stock solution” is a reaction mixture at the time t0=0, that contains initial the monomer mixture of concentration C0, and the initial mass of monomers m0. On the base of “parameters” of the stock

Appendix to Chapter 3 solution, reaction rate and the composition of the copolymer, one can calculate the amounts of monomers, the initiator and rate of its addition.

Determined reaction rate is equal to the rate of addition: dm Rp = (A7) dt

n Rp =  Rpmi (A8) i=1 For ternary system:

Rp = Rpmi + Rpmj + Rpmk (A9) for

dmi Rpmi = (A10) dt and

mi = Mini (A11)

(mi – mass of the monomer “i”; Mi – molecular weight of monomer “i”; ni – number of moles of the monomer “i”)

After the transformation:

dm dmi  Mj dnj Mk dnk  = 1+  +   (A12) dt dt  Mi dni Mi dni 

dnj Fj dnj dt = = (A13) dni Fi dni dt

(Fi,, Fj – denote the content of monomer “i” and “j” in the copolymer)

After the transformation of equation A12 one can calculate the mass of monomer “i” which has to be added in time unit: dm dm i = dt (A14) dt M  F M  F 1+ j j + k k M i  Fi M i  Fi

Appendix to Chapter 3

For other two monomers:

dm F  M dm j = j j  i (A15) dt Fi  M i dt

dm F  M dm k = k k  i (A16) dt Fi  M i dt In order to calculate the addition time one need up front to assume so-called “excess α” – it means the total mass of the monomers which has to be added (usually as the multiple of the mass m0 in the stock solution). For continuous terpolymerization with post addition phase (to complete the conversion of unreacted monomers) 10 times excess (α) (10 x m0) has to be added to reduce its influence on the composition of the copolymer.  t = (A17) dm dt t – addition time α – total mass of the added monomer mixture (“excess”) dm Multiplying the mass of the monomer which has to be added in time unit i by the addition dt time t, one obtains the mass of the monomer. dm m = i t The sum of calculated monomer mass has to be equal to the excess: i dt

mi =  (A18) It is very important to remember that pumps dose volume and do not dose mass. It is necessary to recalculate masses to volumes using densities.

mi Vi = (A19) di

(Vi – volume of monomer “i”; mi – mass of monomer ”i”; di – density of monomer “i”)

In case one of the monomers is not soluble in others it is necessary to use solvent. This should be also considered in calculations of the addition rates. Addition rates are calculated as follow: V R =  i (A20) add t where Vi consists of volume of each component of the monomer mixture.

Appendix to Chapter 3

The required amount of the initiator can be calculated: dm ini = k m (A21) dt ini ini0 where: ln 2 kini = (A22) t 1 2

(mini0 – initial mass of the initiator in the stock solution; kini – decomposition rate constant of the initiator; t1/2 – half life time of the initiator at given temperature)

Calculation of ρ – parameter: b  e − q  d ρ = 12 q  c − a  e 12

1 1 1 1 1 1 1+ b d[M1 ] F1 Where: a = + , b = , c = , d = + , e = , q12 = = r12 r23 r21 r12 r12 r13 1+ a d[M 2 ] F2

r12=0.94, r21=1.02, r31=0, r13=4.9, r32=0, r23=8

F1=0.45, F2=0.49 1 1 a = + =1.185 0.94 8.2 1 b = = 0.980 1.02 1 c = =1.064 0.94 1 1 d = + =1.268 0.94 4.9

1+ 0.980 e = = 0.906 1+1.185 0.45 q = = 0.918 12 0.49

ρ = 2.87

Example of the calculation: Targeted terpolymer composition:

Appendix to Chapter 3

FF8H2MA= 0.20 MF8H2MA= 532 g/mol

FBMA= 0.66 MBMA= 142 g/mol

FMAH= 0.14 MMAH= 98 g/mol

Reaction rate: Rp= 0.5 %/min A) Stock solution

m0= 2.5 g (arbitrary)

C0= 2,5 mol/L Stock solution composition: -3 mBMA = 0,448 g (3.15 x 10 mol) -3 mF8H2MA= 0,722 g (1.45 x 10 mol) -3 mMSA = 1.33 g (13.57 x 10 mol)

mMEK = 4,53 g (3.63 mL)

mHFX = 2,79 g (3.63 mL) -3 mAIBN = 0,03 g (0.18 x 10 mol)

B) Addition rate dm = m  R = 2.5 0.005 = 0.0125 g dt 0 p

dm dm 0.0125 F8H 2MA = dt = = 0.0062g / min M  F M  F 142  0.66 98  0.14 dt 1+ BMA BMA + MAH MAH 1+ + M F8H 2MA  FF8H 2MA M F8H 2MA  FF8H 2MA 532  0.2 532  0.2

dm F  M dm 0.66 142 BMA = BMA BMA  F 8H 2MA =  0.0062 = 0.0055g / min dt FF8H 2MA  M F8H 2MA dt 532  0.2 dm F  M dm 0.14 98 MAH = MAH MAH  F8H 2MA =  0.0062 = 0.0008g / min dt FF8H 2MA  M F8H 2MA dt 532  0.2

C) Addition time

α=10 (10x m0 - arbitrary)

  10 t = = = = 33.33h dm Rp 0.0125 dt

Appendix to Chapter 3

D) Initiator

t1/2= 10 h (AIBN at 65 °C)

ln 2 −1 -1 k AIBN = = 0,012 min (0,0693 h ) t 1 2 dm AIBN = k  m = 0.0693 0.03 = 0.002g / h dt AIBN AIBN 0 dm m = t  AIBN = 33.33 0.002 = 0.066g AIBN dt E) Densities necessary to calculate volumes 3 dF8H2MA – 1,596 g/cm 3 dBMA – 0.90 g/cm 3 dMAH – 1.32 g/cm

3 dMEK – 0.80 g/cm 3 dHFX – 1.378 g/cm 3 dAIBN – 1.11 g/cm

F) Monomer mixture composition for addition dm F8H 2MA = 0.0062g / min = 0.372g / h dt dm m = F8H 2MA t = 0.372 33.33 =12.4g F8H 2MA dt

mF 8H 2MA 12.4 3 VF 8H 2MA = = = 7.77cm d F 8H 2MA 1.596

dm BMA = 0.0055g / min = 0.33g / h dt dm m = BMA t = 0.3333.33 =11.0g BMA dt

mBMA 11 3 VBMA = = = 12.22cm d BMA 0.9

dm MAH = 0.0008g / min = 0.048g / h dt

Appendix to Chapter 3

dm m = MAH t = 0.04833.33 =1.6g MAH dt

mMAH 1.6 3 VMAH = = = 1.21cm dMAH 1.32 MAH is not soluble in the methacrylates used, additional solven was used:

mMEK 2 3 VMEK = = = 2.5cm dMEK 0.8

mHFX 2 3 VHFX = = = 1.45cm d HFX 1.387

3 Total volume to be added: Vm = 25.15cm G) Addition rates (volume) V Monomers: m = 0.755mL / h t Initiator: 0.66 g of AIBN in 0.62 mL of MEK (minimum amount to obtain stable solution)

3 VAIBN = 0.685cm V V +V 0.685 + 0.62 ini = AIBN MEK = = 0.0205mL / h t t 33.33

F8H2MA n-BuMA -O-CH -CH -R 2 2 f -O-CH2-R

FREON-113

4,6 4,5 4,4 4,3 4,2 4,1 4,0 3,9 3,8 3,7

CHCl 3

6 3 0

 ppm

Figure A1. 1H-NMR spectra of P[F8H2MA-co-BMA-co-MSA]

Appendix to Chapter 3

FREON-113

F8H2MA n-BuMA -O-CH -CH -R MSA 2 2 f -O-CH2-R -O-CH3

Acetone-d6

4,6 4,4 4,2 4,0 3,8 3,6 3,4

6 4 2 0  ppm

Figure A2. 1H-NMR spectra of P[F8H2MA-co-BMA-co-MSA] after alcoholysis of succinic anhydride ring with methanol. The spectra were measured in acetone-d6 as co solvent because of relatively high polarity of hydrolysed copolymer.

The continuous addition polymerization reaction was carried out under homogenous conditions with AIBN as initiator. During the reaction samples were taken and their composition was determined by means of 1H-NMR spectroscopy. In order to avoid problems with stirring and influence of post-addition phase on composition as well as on the molecular weight distribution a 5-fold excess of monomers was used and reaction was interrupted after the addition phase. The product was isolated by precipitation in methanol and dried in vacuo at 40 ºC to the constant weight. The 1H-NMR of the polymer is depicted in Figure A3. In order to determine the MSA content the sample was refluxed in the presence of methanol.

The signal from -CO-OCH3 group at 3.6- ppm originated from the metahnolized succinic anhydride ring incorporated in the polymer (see Figure A4) was used to calculate the copolymer composition.

Appendix to Chapter 3

F8H2MA DMA -O-CH -CH -R 2 2 f -O-CH2-R

DMA -O-CH -(CH ) -CH 2 2 10 3 DMA -O-(CH ) -CH 2 11 3

4,6 4,4 4,2 4,0 3,8

CHCl 3

8 7 6 5 4 3 2 1 0

 ppm

Figure A3. 1H-NMR spectra of poly[F8H2MA-DMA-MSA] before alcoholysis in methanol.

F8H2MA DMA MSA -O-CH -CH -R -O-CH -R 2 2 f 2 -OCH3

DMA -O-CH -(CH ) -CH 2 2 10 3

DMA O-(CH ) -CH - 2 11 3 4,6 4,4 4,2 4,0 3,8 3,6 3,4

CHCl 3

8 7 6 5 4 3 2 1 0

 ppm

Figure A4. 1H-NMR spectra of poly[F8H2MA-DMA-MSA] (final product) after alcoholysis in methanol.

Table A2. Changes in the composition of the copolymer during the continuous addition experiment determined by 1H-NMR. Polymer composition 1) [mol %] Reaction time [min] Conversion [%] F8H2MA DMA MSA 75 13.5 18 57 25 210 37.4 18 57 25 390 70.2 19 60 21 500 90 17 60 23

Theoretical composition 2) 21±2 55±2 24±2 MSA = maleic anhydride, DMA = dodecyl methacrylate, F8H2MA = 1H,1H,2H,2H-perfluorodecyl methacrylate 1) Determined by 1H-NMR on sample taken after the indicated reaction time 2) As determined from low conversion experiment

Appendix to Chapter 3

Table A3: Comparison between the elemental compositions calculated on the basis of proton magnetic resonance and measured by elemental analysis (EA). Carbon [wt%] Hydrogen [wt%] Oxygen [wt%] Fluorine [wt%]

1H-NMR 58.33 7.53 13.44 20.70 EA 57.14 7.29 12.09 23.48

0,0055

0,0053 Voltage / mV Voltage /

=10

=5 0,0050 23 25 28 30 Elution volume / mL

Figure A5: Molecular weight distribution of the BMA copolymers obtained in continuous addition experiment with the monomer excess α=5 ( ___ ) and α=10 (---)

0,0056

0,0054 Voltage / mV Voltage /

0,0052

0,0050 22 24 26 28 Elution volume / mL

Figure A6: Molecular weight distribution of poly[DMA-co-F8H2MA-co-MSA] obtained in continuous addition experiment (α=5).

Appendix to Chapter 3

100

Weightin% 40

0 100 200 300 400 500 600 Temperature / degC

Figure A7: TGA measurement of fluorinated ternary copolymers under nitrogen. ( ___ ) C1 (poly[BMA0.69-co-F8H2MA0.21-co-MSA0.1]), (---) C3 (poly[DMA0.57-co- F8H2MA0.2-co-MSA0.22]).

C2 endo C1

50 100 Temperature / degC

Figure A8: DSC thermogram of C1 (poly[BMA0.69-co-F8H2MA0.21-co-MSA0.1]) and C2 (poly[BMA0.67-co-F8H2MA0.21-co-MSA0.12]) C3 (poly[DMA0.57-co-F8H2MA0.2- co-MSA0.22]). The second heating at 10 K/min.

References: [1] U. Beginn, e-Polymers; 2005; no. 073

Chapter 4

Solubility, Emulsification and Surface Properties of Maleic Anhydride, Perfluorooctyl and Alkyl Meth-Acrylate Terpolymers

1. Introduction

Copolymers with perfluorinated building blocks are desired materials because of their unique properties. The introduction of fluorinated moieties into polymeric materials improves their thermal stability, fastness to outdoor conditions and chemical resistance. One of the most important properties of fluorinated materials is their low surface energy, which allows employing them as water, oil and soil repelling coatings [1–8]. Perfluorinated coatings can be prepared by casting of (i) solutions in organic solvents, (ii) of aqueous polymer emulsions and (iii) of water based solutions/dispersions. The most preferred situation would be the application of a perfluorinated copolymer dissolved or dispersed in pure water, which is the most convenient solution from an economic and environmental point of view. However, the basic idea of forming amphiphilic terpolymer consisting of maleic anhydride and long aliphatic side chain methacrylate has been reported before [9], and a very limited number of publications are found reporting successful self-emulsification of fluorinated polymers in water [1] The creation of water-soluble or water dispersible perfluoroalkyl side chain copolymers is supported by the use of maleic anhydride as a comonomer. Hydrolysis of the succinic anhydride rings in the polymer backbone yields hydrophilic carboxylates [10] that additionally improve the adhesion to any substrate that may interact with carboxylic acid or carboxylate groups [11]. This manuscript presents investigations on the solubility, emulsification and the surface properties of coatings prepared from perfluorinated copolymers of the general structure presented in Scheme 1.

CH3 CH3

H2 H2 * C C *

O O O O O O O

RH RF x y x n

Scheme 1. Investigated P[RHMA-co-RFMA-co-MAH]n terpolymers (x + y + z = 1, RH = C4H9-, C12H25-, RF- = C10H4F17-).

Chapter 4

These copolymers were applied on different materials e.g., glass, paper and anodised aluminium. Perfluorinated copolymers that were employed in these investigations were obtained by means of continuous addition terpolymerization of 1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA), alkyl methacrylates (butyl methacrylate(BMA) or dodecyl methacrylate (DMA)) and maleic anhydride (MAH).

2. Experimental

Measurements/Apparatus

A Branson Sonifier S-450A ultrasonic processor (Emerson Electric Co., St. Louis, United States) was used to homogenize polymer/water mixtures and prepare emulsions in water. Typical sonication parameters were: 300 W, 90% amplitude. Contact angle measurements were performed by means of the sessile drop method with a Krüss G-2 instrument (manufacturer, city, country), using water and hexadecane as wetting liquids. Samples were prepared by spin-coating of dispersions/emulsions on glass microscope slides (76 × 26 × 1 mm) at 2500 rpm for 20 s. Coatings on anodised aluminium (row offset printing plate provided by Agfa Gevaert NV, Antwerpen, Belgium) were prepared in the same way.

Polymer samples

P[F8H2MA0.2-co-BMA0.7-co-MAH0.1] (C1), P[F8H2MA0.2-co-BMA0.65-co-MAH0.15]

(C2) and P[F8H2MA0.2-co-DMA0.58-co-MAH0.22] (C3) were synthesized according to the following method described in a detailed way elsewhere [2]. Preparation of the solution A: butyl methacrylate (0.448 g, 3.15 mmol), 1H,1H,2H,2H- perfluorodecyl methacrylate (0.722 g, 1.45 mmol), maleic anhydride (1.33 g, 13.57 mmol) and 2,2′-azo-bis-isobutyronitrile (0.03 g, 0.18 mmol) were dissolved in a mixture of 2-butanone (4.53 g, 3.63 mL) and 1,3-bis (trifluoromethyl) benzene (2.79 g, 3.63 mL). The mixture was placed in a round bottom flask, equipped with a nitrogen inlet, reflux condenser, a magnetic stirring bar and a rubber septum. The reaction mixture was degassed three times by freeze– pump–thaw cycles. Preparation of the solution B: butyl methacrylate (11.0 g, 77.46 mmol), 1H,1H,2H,2H- perfluorodecyl methacrylate (12.4 g, 23.3 mmol) and maleic anhydride (1.6 g, 16.33 mmol) were dissolved in a mixture of 2-butanone (2.0 g, 2.5 mL) and 3-bis (trifluoromethyl) benzene (2.0 g, 1.45 mL). The solution was degassed three times by freeze–pump–thaw cycles.

Chapter 4

Preparation of the initiator solution C: 2,2′-azo-bis-isobutyronitrile (0.066 g, 0,4 mmol) was dissolved in mixture of 2-butanon and 1,3-bis (trifluoromethyl)benzene (1.0 g, 1:1 vol:vol). The mixture was degassed by three freeze–pump–thaw cycles. Continuous addition terpolymerization: The solution A was placed in an oil bath at 65 °C and after 5 min the addition of solutions B and C was started. Solution B was added at a rate of 0.755 mL/h, solution C with a rate of 0.0205 mL/h by means of separate syringe pumps over a period of 33.33 h. After the addition period, the reaction was heated for another 7 h. The reaction mixture was cooled to room temperature and the product was precipitated in methanol (500 mL). Characteristics of the copolymers are summarized in Table 1.

Table 1. Characteristics of the investigated terpolymers. F8H2MA R -MA MAH Mn T Sample R H Mw/Mn g [mol %] H [mol %] [mol %] [kg/mol] [°C] C1 21 -C4H9 69 10 23.5 2.15 56 C2 20 -C4H9 65 15 36.1 1.80 58 C3 20 -C12H25 58 22 40.1 1.72 29

F8H2MA = 1H,1H,2H,2H-perfluorodecyl methacrylate, MAH = maleic anhydride, RH-MA = alkyl

methacrylate, Mn = poly(methyl methacrylate) equivalent number average molar mass, Mw/Mn =

polydispersity, Tg = glass transition temperature.

Particle size measurements were performed using a Malvern Instruments Zetasizer Nano-ZS particle sizer (Malvern Instruments, Malvern, UK). In addition, 2 mL of 2 wt % water solution of the terpolymer C2 was placed in polystyrene (PS) cuvete without filtration.

Solubility tests were performed using the solvents listed in Table 2.

Table 2. Solvents used for solubility and emulsifying investigations. Solvent Provider Grade Remarks Chloroform - technical distilled THF - technical distilled Heptane - technical distilled Dodecane Fluka p.a. - Hexadecane Fluka p.a. - Butylacetate Fluka p.a. - MEK Merck for analysis distilled HFX ABCR - - Toluene - technical distilled Mesitylene Merck for synthesis - Cyclohexane Merck for synthesis distilled 2-Octyl-1-dodecanol Merck for synthesis - Dipropylene glycolmonomethyl ether Aldrich 97% - Freon-113® Merck for synthesis - White mineral oil Aldrich - fl.p.>110 °C Mineral oil Acros - fl.p.>135 °C Mineral oil for IR Acros - fl.p.>195 °C

Chapter 4

Silicon oil Fluka - fl.p.>300 °C NH4OH 25% Merck pure - Ethyl acetate - technical distilled Water - - bi-distilled THF = tetrahydrofuran, MEK = 2-butanone, HFX = 1,3-bis(trifluoromethyl) benzene.

Emulsification of terpolymers The terpolymer (200 mg) was dissolved in an organic solvent (2 mL). Solutions were emulsified in water (20 mL) by means of ultrasonication without or in the presence of sodium dodecyl sulfate (SDS) (50 mg). The final overall concentration of the copolymer was 1 wt %. Table 3 summarises the compositions of the sonicated mixtures.

Table 3. Emulsions of terpolymers (20 mL H2O, room temperature, sonication for 10 min). Polymer Solvent SDS Sample Stability (200 mg) (2 mL) (50 mg) E 1 C1 chloroform no separation and coalescence within 2 days E 2 C1 chloroform yes stable E 3 C1 HFX yes stable E 4 C2 chloroform yes stable E 5 C2 chloroform no separation and coalescence within 2 days E 6 C2 HFX yes stable E 7 C2 i-butyl acetate yes precipitation of polymer within 2 weeks E 8 C2 i-butyl acetate no precipitation of polymer within one week E 9 C3 i-butyl acetate yes precipitation of polymer within one week E 10 C3 i-butyl acetate no precipitation of polymer within one week E 11 C3 n-heptane yes stable E 12 C3 n-heptane no separation after 3 days but without coalescence E 13 C3 n-hexadecane yes stable E 14 C3 n-hexadecane no separation after 3 days but without coalescence E 15 C3 chloroform yes stable E 16 C3 chloroform no separation and coalescence within 2 days HFX = 1,3-bis(trifluoromethyl) benzene.

Preparation of aqueous solution of C1, C2 and C3 The copolymer (500 mg) was dissolved in i-BuAc (10 g) and added to aqueous ammonia (50 mL, 1%). The mixture was emulsified by means of ultrasonication for 5 min and left for seven days. The spontaneously precipitated copolymer was separated by filtration and transferred to freshly prepared aqueous ammonia (25 mL, 1%). The mixture was homogenized by means of ultrasonication for 10 min followed by stirring at 60 °C for 30 min.

Pretreatment of glass substrates To obtain hydrophilic surfaces of the glass microscope slides (76 × 26 × 1 mm), the substrate was treated for 2 min with a mixture of concentrated sulfuric acid (3 parts) and 30% hydrogen peroxide (1 part). Afterwards, the glass slides were washed with bi-distilled water to remove the acid, and transferred into a glass jar filled with bi-distilled water.

Chapter 4

3. Results and Discussion

Three terpolymers, based on 1H,1H,2H,2H-perfluorodecylmethacrylate (F8H2MA) butyl methacrylate (BMA) or dodecyl methacrylate (DMA) and maleic anhydride were investigated:

P[F8H2MA0.2-co-BMA0.7-co-MAH0.1] (C1), P[F8H2MA0.2-co-BMA0.65-co-MAH0.15] (C2) and

P[F8H2MA0.2-co-DMA0.58-co-MAH0.22] (C3). These copolymers with uniform composition were prepared by means of free radical copolymerization using the continuous flow monomer addition technique [2]. It is expected that the presence of 10–22% of succinic anhydride rings will enhance compatibility with water upon hydrolysis to carboxylate groups. Due to the combination of hydrophilic moieties with hydrophobic alkyl and perfluorinated side chains, these terpolymers should act as surfactants and may exhibit self-emulsifying properties, making them suitable for waterborne applications. The coatings prepared from terpolymers C1, C2, and C3 are expected to exhibit hydrophobic and oleophobic properties due to the presence of 20% of perfluorinated side chains. The perfluorinated side chains should orient towards the air interphase of the polymeric layer and thus strongly influencing the surface properties. This orientation should be supported by the presence of alkyl side chains due to the so-called stratification effect [12,13]. This term denotes a spontaneous formation of separate layers of inherently incompatible components (e.g., fluorinated and non-fluorinated) in polymer films whenever the molecular mobility is promoted by a temperature exceeding the glass transition temperature or in the presence of a solvent. The effect of the length-ratio of alkyl- to perfluoroalkyl side chains on the stratification process should become visible by comparison of C1 and C2 (butyl) with C3 (n-dodecyl side chains). The Solubility of C1–C3 in different solvents was tested prior to any further experiments (Table 4). Polymer samples were added in small portions to 2 mL of solvent while stirring. In cases where the copolymers were highly soluble, the experiment was ended at a concentration of ~20 wt %, and the upper limit of solubility was not further checked except from the case of C1/MEK where the saturated solution contained 68 wt % of C1. In case of experiments where the upper solubility limit was exceeded, the insoluble material was separated by filtration, the filtrate was dried, and the mass of the residue was used to calculate the solubility.

Chapter 4

Table 4. Solubility of the terpolymers in different solvents. Solubility [wt %] Sample Solvent C1 C2 C3 S1 Chloroform >20% >20% >20% S2 THF >20% >20% >20% S3 i-Butylacetate >20% >20% >20% S4 MEK >20% >20% >20% S5 Ethyl acetate >20% >20% >20% S6 HFX <14% <13% <12% S7 Freon-113® <10% <10% <8% S8 Toluene <1.8% <1.9% <19% S9 Mesitylene <1.7% <1.7% <20% S10 Heptane <0.01% <0.01% >20% S11 Dodecane <0.01% <0.01% <15% S12 Hexadecane <0.01% <0.01% <15% S13 Cyclohexane <0.01% <0.01% <12% S14 White mineral oil <0.01% <0.01% <0.01% S15 Mineral oil <0.01% <0.01% <0.01% S16 Mineral oil for IR <0.01% <0.01% <0.01% S17 Silicon oil <0.01% <0.01% <0.01% S18 NH4OH 25% <0.01% <0.01% <0.01% S19 Water <0.01% <0.01% <0.01% S20 2-Octyl-1-dodecanol swollen swollen swollen S21 Dipropylene glycol swollen swollen swollen monomethyl ether S22 1% aqueous ammonia - 2% * -

C1, C2: RH = butyl, C3: RH = dodecyl, THF = tetrahydrofuran, MEK = 2-butanone, HFX = 1,3- bis(trifluoromethyl) benzene, * only one concentration was tested.

The highest solubilities (>20 wt %) of the investigated polymers was observed in chloroform (S1), THF (S2), esters (S3, S5), MEK (S4) as well as perfluorinated solvents like hexafluoroxylene (HFX) (S6) or Freon-113® (S7). In case of insolubility of the polymer in the used solvent, the residual mass is given according to the accuracy of the balance. The butyl methacrylate containing copolymers swell without dissolution in 2-octyl-1-dodecanol (S20) and dipropylene glycol monomethyl ether (S21), while, in non-polar alkanes (S8–S13), the polymers C1 and C2 are practically insoluble (<0.01 wt %). None of the copolymers is soluble in mineral (S14–S16) or silicon (S17) oils as well as in water (S19) and aqueous ammonia (S18). The solubility properties of the dodecyl methacrylate containing polymer C3 are quite similar to the butyl methacrylate copolymers C1 and C2 with the important difference that C3 is well soluble in toluene (S8), mesitylene (S9), cyclohexane (S13) and n-alkanes (S10–S12). This can be explained by the relatively high content of long alkyl chains, which make the copolymer of dodecyl methacrylate more compatible with hydrocarbons. An aqueous solution of the terpolymer, which contains perfluorinated and alkyl chains, may be obtained by hydrolysis of the succinic anhydride repeating units. The succinic anhydride

Chapter 4 ring was expected to hydrolyze in water (S19) or aqueous ammonia (S18) to give hydrophilic carboxylate moieties (Scheme 2).

NH4 O O CH3 CH3 CH3 CH3

* CH2 C CH2 C CH CH * * CH2 C CH2 C CH CH * NH 4OH O O O O O O O O O O O O O NH4 R Bu f z R Bu x y f x y z

Scheme 2. Hydrolysis of the succinic anhydride ring with ammonia solution yields ammonium carboxylate groups.

None of the three investigated terpolymers spontaneously dissolved in water or in aqueous ammonia. Most probably, the MAH content is too low; alternatively, the accessibility of the anhydride moiety might not be sufficiently high to allow for their hydrolysis. In summary, one can state that a moderate content of fluorinated comonomer units results in good solubility of the terpolymer in a wide range of semi-polar solvents, mainly non- halogenated solvents. It has also been demonstrated that the choice of the non-fluorinated comonomer enables enhanced solubility in a specific type of solvent e.g., long alkyl chains enhance the solubility in hydrocarbons. This leaves a certain freedom for designing polymers with better solubility in a solvent of choice.

Polymer coatings by solution casting In a first series of coating experiments, the C1 and C2 terpolymers were coated from 1 wt % organic solution on glass substrates. The advancing contact angles of water and hexadecane as wetting liquids were measured at 20 °C by means of the sessile drop technique with the dried films prior and subsequent to an annealing step. The annealing was performed on dry coating in an oven at 120 °C for 12 h. The contact angles of the coatings obtained from chloroform solutions, as summarized in Table 5, did not change upon annealing (cf. S1-C1, S1-C1 * and S1-C2, S1-C2 *).

Table 5. Contact angle Θ values measured on different substrates covered with C1, C2, and C3 copolymers coated from solutions and emulsions.

ΘH2O ΘHD Sample Polymer Solvent System Glass Glass Paper Al Paper Al S1-C1 C1 CHCl3 109° ± 2 - - 70° ± 1 - - S1-C1 * C1 CHCl3 108° ± 2 - - 71° ± 1 - - S1-C2 C2 CHCl3 110° ± 2 - - 73° ± 1 - - S1-C2 * C2 CHCl3 110° ± 2 - - 73° ± 1 - - S1-C3 C3 CHCl3 108° ± 2 - - 67° ± 1 - - S1-C3 * C3 CHCl3 108° ± 2 - - 67° ± 1 - - S6-C1 C1 HFX 108° ± 2 - - 70° ± 1 - -

Chapter 4

S6-C1 * C1 HFX 108° ± 2 - - 70° ± 1 - - E2 C1 CHCl3/H2O 105° ± 2 - - 67° ± 1 - - E4 C2 CHCl3/H2O 103° ± 2 - - 70° ± 1 - - E15 C3 CHCl3 in H2O 105° ± 2 - - 66° ± 1 - - E13 C3 C16H34 in H2O 108.5° ± 1 - - 65° ± 2 - - E2 C1 CHCl3/H2O - 101° ± 1 105° ± 1 - 59° ± 1 68° ± 1 E4 C2 CHCl3/H2O - 102° ± 1 104° ± 1 - 60° ± 1 70° ± 1 E7 C2 i-BuAc/H2O - 100° ± 1 110° ± 2 - 62° ± 1 69° ± 1 E9 C3 i-BuAc/H2O - 103° ± 1 108.5° ± 1 - 61° ± 2 65° ± 2

E11 C3 n-C7H16/H2O 102° ± 1 108.5° ± 1 - 64° ± 1 65° ± 2

C16H34 = hexadecane (HD), Al = anodised aluminium, i-BuAc = isobutyl acetate, * = no annealing.

Hence, during the film formation process, the mobility of the side groups was sufficient to ensure proper orientation of the fluorinated side chains towards the polymer/air interface even at ambient temperature. Since in any case the glass transition temperature Tg of the pure polymers (29–58 °C, see Table 1) were higher than the used coating temperature, the presence of a solvent induced “plasticizer-effect” was concluded. Although an annealing step was not necessary to obtain large contact angles, it was of crucial importance to ensure a permanent adhesion of the polymers to the glass surface. Since it is known that anhydride groups can form silanol-esters with –Si–OH– groups present at the glass surface at elevated temperatures [14,15], it is assumed that polymers C1–C3 covalently link to the substrate by a similar mechanism (see Figure 1).

Figure 1. The chemical bounding of the terpolymer coated from aqueous ammonia solution with the glass surface.

Note that the contact angles of the dodecyl methacrylate (DMA) based copolymer coatings C3 (S1-C3 and S1-C3 *, Table 5) were also independent of annealing and only for 2–3 degrees lower than that of the samples C2. Both BMA and DMA copolymers contain the same fraction of fluorinated building blocks, hence their minor difference in surface properties is determined by the alkyl-methacrylate comonomer. The hydrophobicity is in principle equal (in the range of measurement error), but the higher oleophilicity of C3 seems to be a result of the presence of long alkyl chains as stated by the good solubility of C3 in hydrocarbons (see Table 4).

Chapter 4

Emulsion of organic polymer solution in water. A possible way to prepare waterborne systems from the investigated fluoropolymers is to prepare water based emulsions of polymer solutions in organic solvents [16]. Organic solvents can only be industrially applied if they are non-miscible with water, exhibit low volatility (i.e., high boiling points) and are non-hazardous with respect to human health and environment. However, for scientific purposes, other solvents like chloroform and HFX can also be employed. The formation of emulsions of C1, C2 and C3 was checked with chloroform (E1, E2, E4, E5, E15, E16), HFX (E3, E6), ester- (E7–E10), and hydrocarbon-solvents (E11–E14) according to Table 3. About 10 wt % of a copolymer was dissolved in an organic solvent that was emulsified either in pure water or in an SDS solution by means of an ultrasonic homogenizer. Emulsions obtained without SDS were found to be stable only for several days. Since detergent- free emulsions of chloroform in water break within seconds, this observation indicates that the terpolymers act as weak surfactants, which are, however, unable to stabilize oil droplets in water permanently. If i-butyl acetate was used as the auxiliary solvent (E7–E10, Table 3), the polymeric material precipitated within one week, most probably because of the slight solubility in water (6.7 g in 1000 g of water). Most probably, the solvent was taken up by the aqueous major phase, leaving back glassy/sticky polymer droplets that easily coagulate. An alternative explanation might be the partial hydrolysis of the maleic anhydride moieties due to water, becoming dissolved in the organic phase. This reaction could have created a polymer that was too hydrophilic to be soluble in the organic phase but still remained too hydrophobic to be soluble in water. This phenomenon was not observed with water immiscible auxiliary solvents. In case of more hydrophobic organic solvents like hexadecane, the dispersion is additionally stabilized by building up a counteracting osmotic pressure. Employment of a so-called “ultrahydrophobe”, a substance which is not able to migrate through the water phase is known to prevent Ostwald ripening [17]. This can be an explanation for the fact that emulsions E12 and E14 showed a liquid/liquid phase segregation, but no coalescence (see Table 3). Introducing 0.25 wt % SDS as the surfactant leads to stable emulsions with the exception of i-BuAc emulsions (E7, E9) where a partial precipitation of the polymer was still observed.

Chapter 4

Coating from emulsions. In subsequent experiments, coatings have been prepared from water-emulsified chloroform-solutions of the P[RF-co-RH-co-MAH]. Emulsions E2 and E4 of the BMA based copolymers C1 and C2 in chloroform gave less hydrophobic or oleophobic surfaces than those obtained after spin-coating of polymers from chloroform solution (S1-C1 and S1-C2, Table 5).

Comparative measurements of coatings obtained from emulsions E15 ([C3/CHCl3]/H2O) and

E13 ([C3/C16H34]/H2O) revealed the latter to exhibit a larger contact angle against water (E15:

ΘH2O = 105°, E13: ΘH2O = 108.5°). The angle was close to that of coating S1-C3 (ΘH2O = 108°), obtained from a water-free chloroform solution. This comparison could not be made for BMA copolymers C1 and C2 due to their insolubility in hexadecane (see Table 4). As was mentioned before, the organic solvent seems to plasticize the polymer during the deposition process, allowing the perfluoroalkyl-groups to move towards the air/film interface before the solvent evaporates [16]. When coating emulsions, the low-boiling solvent obviously evaporates first, leaving back a water rich polymer containing phase. Due to its incompatibility to water, the polymer partially vitrifies, hence the side chains become unable to orient properly and the measured contact angles are lower by 4–7 degrees (cf. S1-C1 vs. E2 and S1-C2 vs. E4 in Table 5). Note that this problem could not be solved by annealing. Although the contact angles of hexadecane do not exceed 65–67°, the oil did not adhere to the coatings. Upon tilting the glass plate, the droplets of hexadecane slid from the glass plates without leaving any oil trace on the surface. This is an important observation that shows the effectiveness of the coating as oil repellent. The above-mentioned explanation of the deposition mechanism is a hypothesis that needs to be further investigated. Such an investigation is not in the scope of the presented work. Not only glass but also paper and anodised aluminium were covered with the emulsion and annealed at 120 °C for 12 h. Coatings prepared from E2 on aluminium substrates showed contact angles of water and hexadecane very close (±2°) to the values measured on coated glass (cf. Table 5, E2). On coating the emulsions E4, E7, E9, and E11 on aluminum, oil and water repelling surfaces were also obtained with ΘH2O = 105–110° and ΘHD = 65–70°. The properties of the coating obtained on paper are different and the measured contact angles for both wetting liquids were lower (ΘH2O = 100–103° and ΘHD = 59–64° cf. Table 5) than in case of glass and aluminum. This phenomenon can be caused by penetrability of the paper. Probably, a large part of the copolymer penetrates the substrate together with the solvent, becomes absorbed on the cellulose fibers inside the structures of the paper and was not available

Chapter 4 to form a closed film. Tests to explore further the structure of the coatings (e.g., X-ray Photoelectron Spectroscopy or Scanning Electron Microscopy) were not performed.

Preparation of aqueous solutions Since environmental regulations impede the use of Freon-113® and other halogenated solvents in industrial applications and since organic solvents are no longer accepted for household users, aqueous formulations of surface modifying polymers are mandatory. As already shown by means of solubility tests, a spontaneous dissolution of the investigated copolymers in water and aqueous ammonia was not observed. In order to facilitate the process, isobutyl acetate (i-BuAc) was used as “auxiliary solvent” to swell the polymeric material and depress its glass transition temperature [18], prior to contact with aqueous ammonia solution. Hence, mixtures of terpolymers C1, C2 and C3 with i-BuAc (terpolymer: i-BuAc = 1:1, mass:mass) were suspended in 1% ammonia solution and stirred for at least 12 h. This treatment did not change the appearance of the copolymer. The application of ultrasound mechanically disrupted the gel-particles, but did not cause emulsification. The terpolymer could not be detected in the liquid phase, neither direct by evaporation of the filtrate, nor indirect by casting a drop of the liquid phase on a glass plate, followed by annealing and subsequent contact angle measurement. Any tested combination of heating, stirring as well as ultrasonication led in the best case to the dispersion of small particles with a strong tendency to agglomeration and sedimentation. The issue was settled by dissolving terpolymers in i-BuAc (0.5 g polymer in 10 g i-BuAc), followed by ultrasonication and subsequent standing of the mixture for seven days at room temperature. The spontaneously precipitated copolymer was separated by filtration and transferred to 1% aqueous ammonia. The mixture was homogenized by means of ultrasonication and stirred at elevated temperature for 30 min to yield a clear aqueous solution with polymer C2, however not with polymer C1 and C3. A picture of the clear solution of 2 wt % of C2 in aqueous ammonia (1%) is presented in Figure 2A. Obviously, the composition of terpolymers C1 and C3 is not inside the “emulsification window”: it is assumed that the MAH- content of C1 (10 mol %) is too small to enable water solubility, while the long alkyl side chains of C3 impede the emulsion formation despite the large MAH content of 22%. The average diameter of the particles present in the aqueous dispersion as measured by means of a particle sizer is 50 nm and the particle size distribution is skewed towards larger diameters (see Figure 2C).

Chapter 4

Figure 2 A: Solution (S 22) of polymer C2 2 wt% in aqueous ammonia (1%). B: Proposed micellar structure of the fluorinated copolymer in water. Hydrophobic side chains form the core that is shielded by hydrophilic moieties. C: Particle size (diameter) distribution of the aqueous dispersion of 2 wt% of fluorinated terpolymer C2 .

A sample of the aqueous solution of C2 was further investigated by means of atomic force microscopy (AFM) after spin coating on mica (see Figure 3). The size difference registered by means AFM and particle sizer methods can be explained by the accuracy of these methods and the size of measured particle population (entire in the case of particle sizer and partial for AFM). It is known that bimodal distributions obtained by AFM are registered as skewed monomodal distributions by a particle sizer [19], which might be valid in this case.

Figure 3. Solution (S 22) of polymer C2 2 wt % in aqueous ammonia (1%): AFM image.

Chapter 4

The size of the spherical structures indicates that the particles are formed by more than one macromolecule. The number of the molecules in such particles was estimated assuming the density of the copolymer to be 1 g/cm3. A particle of 12 nm in diameter should hence consist of about seven polymer molecules, while a 20 nm particle should host 30 macromolecules. The structure of such a particle is schematically depicted as in Figure 2B. The pretreatment process described above generated associated structures and fluorinated and alkane side chains (thin black lines) are proposed to form a hydrophobic core that is shielded by the hydrophilic ammonium succinate groups (spheres) against the aqueous environment. The polymer backbone is depicted as a thick, black line.

Film formation and contact angles

The aqueous “micellar solution” of the P[F8H2MA0.2-co-BMA0.65-co-MAH0.15] (C2) was used to prepare coatings. The contact angle against water and HD of such coating measured after annealing is ΘH2O = 110° and ΘHD = 71° (see S22-1, copolymer C2 in Table 6) and is in principle the same as in the case of the coating obtained from organic solution ΘH2O = 110°,

ΘHD = 73° (see S1-1 for copolymer C2 in Table 6). The annealing at elevated temperature is especially important in the case of the coating obtained from aqueous ammonia solution where the anhydride ring was converted to carboxylic acid ammonium salt as depicted in Scheme 3.

Table 6. Surface properties of the coatings on glass, before and after washing with ethyl acetate. Rinsed C1 C2 C3 Sample Solvent Annealed EtAc ΘH2O ΘHD ΘH2O ΘHD ΘH2O ΘHD S1-1 CHCl3 Yes No 109° ± 2 70° ± 1 110° ± 2 73° ± 1 108° ± 2 67° ± 1 S1-2 CHCl3 Yes Yes 140° ± 4 70° ± 1 141° ± 4 73° ± 2 134° ± 4 68° ± 3 S22-1 Water Yes No - - 110° ± 2 71° ± 1 - - S22-2 Water Yes Yes - - 140° ± 5 70° ± 1 - - S22-3 Water No Yes - - 107° ± 2 - - - EtAc = ethyl acetate, HD = hexadecane.

NH4 OH O CH3 CH3 CH3 CH3

* CH2 C CH2 C CH CH * NH4 OH * CH2 C CH2 C CH CH * + RfOH H O O O 2 O O O O O O O O O O O NH4 NH R Bu 4 f y z Bu x x y z Scheme 3. Hydrolysis of perfluorinated side chains in terpolymer (there is no direct evidence for this assumption!).

It might be possible that the terpolymer became soluble because of partial hydrolysis of perfluorinated side chains during the long-term procedure of preparation of the aqueous solution. The decrease of the number of perfluorinated moieties and generation of additional

Chapter 4 hydrophilic carboxylic groups could enhance dissolution of the terpolymer in water. Scheme 3 depicts the hydrolysis of fluorinated methacrylate and salt formation of the remaining carboxylic acid as well as the salt formation upon hydrolysis of the succinic anhydride moiety. Any long-chain perfluorinated alcohol possibly generated by hydrolysis could form a physisorbed hydrophobic layer on the surface of the coating causing oil- and water repellency. However, because of its physisorbed nature, this layer should be removable by rinsing the sample with a large excess of suitable solvent. Ethyl acetate was chosen as the most suitable because it is a good solvent for fluoro-alcohols but should not destroy the polymer coating. The washing after annealing gave unexpected results. The contact angle against water of the coating washed with ethyl acetate was 140° (see S22-2, Table 6) and was 30° higher than one measured for non-washed surface (S22-1). The contact angle measured against hexadecane did not change. This procedure was repeated for coating obtained from an organic solution for comparison purposes as no presence of free perfluorinated alcohol can be expected. The results obtained for coating prepared from solution were identical: the value against water for surface washed with ethyl acetate was 141° (see S1-2, C2), which was 31° higher than without washing (see S1-1, C2). The comparison between contact angle values for different coatings is summarized in Table 6. The remarkable difference between wettability by water of the coating made from water on the glass slide is very well depicted in Figure 4 before washing 110° (Figure 4A) and after washing with ethyl acetate 141° (Figure 4B).

Figure 4. (A) water droplet on the glass coated with S 22 (Θ = 110°) and (B) on the glass coated with S 22 after rinsing with ethyl acetate (Θ = 140°) (both after annealing). The “ball jump motion” phenomenon—advancing (C) and receding (D,E) on anodised aluminium coated with S 22 after annealing and rinsing steps. The Arrows in Figure 4(C,D) indicate the motion of the cannula.

Chapter 4

Our hypothesis on the occurrence of this very high hydrophobicity of the surface that occurs despite previous hydrolysis can be described in the following manner: (I) the low molecular weight chain aligns perfectly at the surface; (II) heat-treatment causes chemisorption due to the ester formation with surface hydroxyl groups; and (III) the subsequent rinsing removes non-reacted perfluorinated alcohol leaving a perfect surface layer of covalently bound

RF chains. To confirm or negate this possibility, the sample was rinsed with ethyl acetate prior to annealing. The contact angle of the coating obtained from aqueous dispersion and rinsed with ethyl acetate without annealing was measured against water, and it is significantly lower than this of annealed coating. The measured contact angle of 107° (see S22-3, Table 6) is close to the value obtained after annealing but without rinsing with ethyl acetate. This indicates that the assumed hydrolysis did not take place or is not significant. The obtained, lower value of the contact angle is most probably caused by insufficient reorganization of the terpolymer coated form aqueous dispersion at room temperature. Possibly, the behaviour after rinsing with suitable solvent is caused by removing layers of impurities from the coating by the solvent, as well as by giving the coating a second chance for orientation due to the liquids’ plastifying effect. Measurement of contact angles of coating made from the aqueous solution of C2 on the anodized aluminium substrate faced unexpected problems. The water droplet that had to be settled on the coated substrate exhibits behavior similar to so-called “ball jump motion”. Because of the very low adhesion to the surface, the droplet receded together with the needle instead of settling on the substrate. Some adhesion of the droplet to the coating still can be observed; however, it is too weak to keep the droplet on the surface. The observed effect is therefore less pronounced than the “ball jump motion” reported by McCarthy et al. for super hydrophobic surfaces with contact angles > 170° [20,21]. The above described observation is depicted in Figure 4C–E. Another possible explanation could be the roughness of the anodised aluminium. Unusual wettability involving very high contact angles of 150–170° against water has been reported and explained as the influence of the surface roughness on the micro or nanometer scale [22–24]. It is important to remember that the measured contact angles in all cases are sessile drop measurements (in equilibrium with surface) and are neither advancing nor receding measurements and, due to this fact, the hysteresis has not been determined. Measurement of hysteresis is especially important in case the surface topology can play a significant role in the wettability of the coating [24]. This can be a subject of further investigation.

Chapter 4

4. Conclusions

Fluorinated terpolymers of the composition P[RFMA-co-RHMA-co-MAH] (RH = C4H9-,

C12H25-, RF = C10H4F19-) containing ca. 20 mol % of fluorinated monomer units can be dissolved in semi polar solvents like tetrahydrofuran, chloroform or ethyl acetate to give more than 20 wt % solutions. They exhibit good solubility in fluorinated solvents like HFX and ® Freon-113 . Upon incorporation of dodecyl-side chains (RH = C12H25-), the polymer also becomes soluble in hydrocarbons. Coatings on glass obtained from polymer solutions show good water repellence (Θ = 108– 110°) and sufficient oil repellence (Θ = 67–73°). The oleophobicity of the C3 terpolymer based on DMA is clearly lower (by 6°) than that based on BMA. Surprisingly annealing (a well- known method to enhance surface properties) has no effect on the wettability of thin films formed by the investigated copolymers. This is most probably caused by combination of the relatively low Tg of the copolymers and the volatility of solvent enabling proper orientation of fluorinated chains to the surface. Emulsification of organic solutions of the terpolymers resulted in unstable mixtures that demix within days, showing that self-emulsifying properties of investigated polymers (under testing conditions) are poor. Stable emulsions were obtained only in the presence of additional surfactant. Water and hexadecane contact angles of coatings prepared from emulsions of organic solutions (CHCl3/H2O) are slightly worse, ranging from 103° to 105°, than those obtained from solutions (CHCl3) 108–110°.

A stable aqueous solution was obtained only from C2 copolymer (P[F8H2MA0.2-co-

BMA0.65-co-MAH0.15]). The amount of hydrophilic building blocks necessary for water solubility must be in balance with the perfluorinated building block and the right type of third alkyl monomer. Thin films coated from this solution yielded a water (Θ = 110°) and oil-repelling surface (Θ = 71°) and the water repellence was considerably enhanced by washing of the annealed coating with ethyl acetate (Θ = 140°) while the contact angle against hexadecane remained unchanged. This was observed also for coatings made from CHCl3 (141°). The observed effect is most probably caused by removing of an impurity layer. The influence of the alkyl side chain on the orientation of fluorinated side chains (stratification) is not well pronounced and cannot be univocally concluded. However, the contact angles measured for C3 that contains dodecyl side chains are slightly lower for both

Chapter 4

wetting liquids. RH, RF, and MAH–Terpolymers of moderate fluorine content are versatile, flexible to handle materials that offer a wide range of applications for surface modification.

References [1] M.S. Cha, J.W. Kim, J.-W. Ha, I.J. Park, S.-B. Lee, T. Yang, S. Cheng, J. Polym. Sci. A Polym. Chem., 2010, 48, 4574–4582 [2] M. Szkudlarek, U. Beginn, H. Keul, M. Möller, Polymers, 2017, 9, 610 [3] B. Jiang, L. Zhang, B. Liao, H. Pang, Polymer, 2014, 55, 5350–5357 [4] M. Sha, D. Zhang, R. Pan, P. Xing, B. Jiang, Appl. Surf. Sci., 2015, 349, 496–502 [5] D. Kronlunda, M. Lindénb, J.-H. Småtta, Prog. Org. Coat., 2016, 101, 359–366 [6] N.W. Kim, C.Y. Ahn, K.C. Song, Korean Chem. Eng. Res., 2016, 54, 387–393 [7] A. Asakawa, M. Unoki, T. Hirono, T. Takayanagi, J. Fluor. Chem., 2000, 104, 47–51 [8] L. Yu, H. Xu, X. Liu, G.Y. Chen, ACS Nano, 2016, 10, 1076–1085 [9] R.A. El-Ghazawy, R.K. Farag, J. Appl. Polym. Sci., 2009, 115, 72–78 [10] C. Tang, S. Ye, H. Liu, Polymer, 2007, 48, 4482-4491 [11] N. Tirelli, O. Ahumada, U.W. Suter, H. Menzel, V. Castelvetro, Macromol. Chem. Phys., 1998, 199, 2425–2431 [12] V.V. Verkholantsev, Prog. Org. Coat., 1992, 20, 353–368 [13] A. Kulak, Y.-J. Lee, Y.S. Park, K.B. Yoon, Angew. Chem. Int. Ed. 2000, 39, 950– 953 [14] M.N. Tahir, Y. Lee, Food Chem., 2013, 139, 475–481 [15] S. Sheiko, E. Lermann, M. Moeller, Langmuir, 1996, 12, 4015–4024 [16] M. Postel, J.G. Riess, J.G. Weers, Artif. Cells Blood Substit. Biotechnol., 1994, 22, 991–1005 [17] T.S. Chow, Macromolecules, 1980, 13, 362-364 [18] C.M. Hoo, N. Starostin, P. West, M.L. Mecartney, J. Nanoparticle Res., 2008, 10, 89–96 [19] L. Gao, T.J. McCarthy, J. Am. Chem. Soc., 2006, 128, 9052–9053 [20] L. Gao, T.J. McCarthy, Langmuir, 2006, 22, 2966–2967 [21] S. Veeramasuneni, J. Drelich, J.D. Miller, Y. Yamauchi, Prog. Org. Coat., 1997, 31, 265–270 [22] Y. Liao, R. Wang, A.G. Fane, J. Membr. Sci., 2013, 440, 77–87 [23] J.D. Miller, S. Veeramasuneni, J. Drelich, M.R. Yalamanchili, Y. Yamauchi, Polym. Eng. Sci., 1996, 36, 1849–1855 [24] W. Chen, A.Y. Fadeev, M.C. Hsieh, D. Oener, J. Youngblood, T.J. McCarthy, Langmuir, 1999, 15, 3395–3399

Chapter 5

Water soluble perfluorinated terpolymers containing phosphoric acid groups.

1. Introduction

The advantage of fluorinated copolymers as its special characteristics such as water, oil and soil repellence, thermal and chemical resistance are widely described in literature [1,2]. As it is shown in this thesis, the content of succinic anhydride rings in copolymers allows on one hand further modification of the copolymer in order to adjust its properties and on the other hand addition of desired anchoring group for surface modification. The design of environmentally friendly, water soluble, perfluorinated copolymers with good adhesion to different materials and desired surface properties has been explored in this work. The adhesion to a wide range of materials, particularly containing hydroxyl or amine groups, can be ensured by presence of anhydride moieties [3]. Although, mentioned succinic anhydride moiety determines a good adhesion to metallic surfaces [4] its content in copolymer can never exceed 50 % due to the fact that MSA is not homopolymerizing under standard radical polymerization conditions [5]. In case of terpolymers usually the succinic anhydride content is much lower therefore the water solubility of such terpolymer is strongly limited. Adhesion to metals as well as the solubility in polar solvents can be enhanced by introduction of a second or third monomer containing a polar anchoring group such as phosphoric acid moiety. Phosphoric acid and phosphonic acid derivatives are well known and commonly used agents in metal processing against corrosion [6,7]. This is an additional and convincing argument for incorporation of such a functional group into the fluorinated terpolymer. One of the methods of incorporation of phosphoric acid groups into terpolymers beside succinic anhydride rings is copolymerization of maleic anhydride with ethylene glycol methacrylate phosphate (EGMP) and 1H,1H,2H,2H-perfluorodecyl methacrylate. Such a material is expected to show good solubility in water, very good adhesion to different materials (particularly to metals) and appropriate protective properties. The possibility to employ ethylene glycol methacrylate phosphate in synthesis of water soluble perfluorinated terpolymers as well as determination of the properties of such designed polymers and coatings based on them is investigated and described in this chapter as a feasibility study for potential continuous addition synthesis.

Chapter 5

2. Experimental Part

Materials Ethylene glycol methacrylate phosphate (EGMP, Aldrich) was purified before use (see later). 1H,1H,2H,2H-Perfluorodecylmethacrylate (F8H2MA, Aldrich), was distilled over

CaH2 (Fluka) under reduced pressure. Maleic anhydride ( MSA, Merck, for synthesis) was sublimed under reduced pressure (50 °C, 2.4x10-2 bar). 2,2’-Azo-bis-isobutyronitrile (AIBN, Merck) was recrystallized from methanol at 40°C. 2-Butanone (MEK, Merck) was dried over

CaH2 and distilled before use. Trimethylsilyl chloride (TMSC, Fluka), 1,3-bis(trifluoromethyl)benzene (HFX, ABCR), Freon 113 (Fluka), 1,2-dichloroethane (EtCl2, Riedel-de Haën) and other solvents were used as received. Tertiary amines (triethyl- NEt3, tripropyl- NPr3, tributyl- NBu3, trihexyl- NHex3 and trioctyl- NOct3, Merck) were used as received.

Methods Size exclusion chromatography was performed with a system consisting of a LC 1120 pump (Polymer Laboratories), a UV detector ERC-7215 and RI detector ERC-7515A (ERMA CR INC.), a precolumn 50x8 mm, 50Å of nominal pore size, and four columns (300x8 mm) filled with MZ-Gel SDplus of nominal pore size 50Å, 100Å, 1000Å, 10000Å (MZ- Analysentechnik, Mainz, Germany). The set was calibrated with PMMA and PS standards from Polymer Laboratories. The sample concentration was 7 mg of polymer in 1 mL of solvent, the injected volume of the sample: 100 μL. Tetrahydrofuran used for measurements was stabilized with 2,6-di-tert-butyl-4-methylphenol (250 mg/L). 1 H-NMR spectra were obtained from a Brucker DPX-300 spectrometer in acetone-d6 and D2O at 300 MHz. MestRe-C 4.9.0.0 was used as evaluation software. As reference signal the solvent peak was used. Thermogravimetric measurements (TGA) were performed using a NETZSCH TG 209c thermo balance under nitrogen atmosphere at the flow 15 mL/min. Samples (~9 mg) were placed in standard NETZSCH alumina 85 L crucibles. Heating rate was 10 K min-1. Differential scanning calorimetry (DSC) measurements were performed using a Netzsch DSC 204 unit. Samples (typical weight: ~9 mg) were enclosed in standard Netzsch 25 μL aluminium crucibles. Indium and palmitic acid were used as calibration standards. Heating and cooling rates were 10 °C min-1.

Chapter 5

Contact angle measurements were proceed using the sessile drop method with Krüss G- 2 instrument using water and hexadecane as wetting liquids. Samples were prepared by spin- casting on glass microscope slides 76x26x1 mm at 2500 rpm for 20 seconds.

Elemental analysis was performed by Dr. Anastasya Buyanowskaya in A.N. Nesmeyanov Institute of Organoelement Compounds of the Russian Academy of Science in Moscow.

Polymer syntheses

Monomer purification (pre-polymerization) of ethylene glycol methacrylate phosphate EGMP (20 g) inhibited with 1000 ppm MEHQ was dissolved in distilled ethyl methyl ketone (MEK, 100 mL) and placed in a 250 mL two necks round bottomed flask equipped with reflux condenser, nitrogen inlet and magnetic stirring bar. The flask was placed in an oil bath at 60°C for 90 minutes and nitrogen was passed through the mixture. During that time 800 mg of the homopolymer of EGMP was precipitated. The homopolymer was removed by filtration and the solvent was removed under reduced pressure at 25°C. During evaporation air was passed through the liquid to inhibit the polymerization reaction.

Protection of acidic groups of ethylene glycol methacrylate phosphate EGMP with tertiary alkyl amines EGMP after pre-polymerization (5g, 23.8 mmol) was dissolved in MEK (30 mL) and a tertiary amine NR3 (47.6 mmols, where R=Et, Pr, Bu, Hex, Oct) was added. The mixture was stirred at 0°C (ice bath) for 2 hours. During that time small amount of white solid precipitated and was removed by filtration. Spectroscopic analysis of the precipitate showed that

H3PO4•(NR3)3 has been obtained. The filtrate was evaporated under reduced pressure according to the method described above to give EGMP(NR3)2 as a viscous, liquid product.

Silylation of EGMP with TMSC

EGMP (5g, 23.8 mmol) was dissolved in EtCl2 (20g) and TMCS (10g) was added. The mixture was placed in 100 mL round bottomed flask equipped with reflux condenser and drying tube filled with anhydrous calcium chloride. The reaction mixture was refluxed for 5 hours and after that time unreacted TMSC and the solvent were removed by distillation under reduced pressure at 40º. The reaction proceeded quantitatively.

Chapter 5

Synthesis of poly[1H,1H,2H,2H-perfluorodecyl methacrylate-co-maleic anhydride-co- ethylene glycol methacrylate phosphate] EGMP after pre-polymerization (0.6g, 2.82 mmol), F8H2MA (1.5g, 2.82 mmol) and MSA (0.28g, 2.82 mmol) were dissolved in a 1:1 (vol:vol) mixture of MEK and HFX (6 mL) and AIBN (0.028g, 0.17 mmol, 2 mol%) was added. The reaction mixture was placed in a 25 mL round bottomed flask equipped with a valve and degassed 3 times by freeze-pump-thaw cycles. The reaction mixture was placed in oil bath at 65ºC. After 4 minutes the product started to precipitate. The reaction was carried out for additional 10 minutes. The solid product was separated by filtration. The product cannot be dissolved in any solvent.

Example of the copolymerization reaction of 1H,1H,2H,2H-perfluorodecyl methacrylate (F8H2MA) with maleic anhydride(MSA) and ethylene glycol methacrylate tripropylammonium phosphate(EGMP). (Table 1, Experiment 4.)

EGMP(NPr3) (0.7 g, 2 mmol) MSA (0.2 g, 2.04 mmol) and AIBN (2 mol%) were dissolved in DMF (17 mL) and placed in a 50 mL round bottomed flask equipped with valve and degassed 3 times by freeze-pump-thaw cycles. The flask was placed in an oil bath at 65ºC. The reaction was carried out for 30 minutes. In case of the presence of MSA the reaction mixture changed the color to dark brown. In each case a crosslinked product was obtained. All the reactions were carried out in the same manner, the amounts of monomers and solvents are summarized in Table 1.

Table 1. Compositions of reaction mixtures of EGMP(NR3) Exp. EGMP(NR3) Mon. 2 / [g] Solvent / [g] No Mon 3 / [g]

1 EGMP(NEt3) MSA / 2,33 MEK / 5 2 EGMP(NEt3) MSA / 2.0 MEK / 10 3 EGMP(NEt3) MSA / 1.0 MEK/HFX / 10 - F8H2MA / 1.0 - 4 EGMP(NPr3) MSA / 0.2 DMF/17 5 EGMP(NPr3) MSA / 1.0 MEK/HFX / 10 - F8H2MA / 1.0 - 6 EGMP(NBu3) MSA / 1.4 MEK / 10 7 EGMP(NHex3) MSA / 1.4 MEK / 10 8 EGMP(NOct3) MSA / 1.4 MEK / 10 MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H-perfluorodecyl methacrylate, MEK = methyl ethyl ketone, HFX = 1,3-bis(trifluoromethyl) benzene, EGMP = ethylene glycol methacrylate phosphate, NEt3 = triethylamine, NPr3 = tripropylamine, NBu3 = tributylamine, NHex3 = trihexylamine, NOct3 = trioctylamine

100

Chapter 5

Example of the synthesis of poly[1H,1H,2H,2H-perfluorodecyl methacrylate-co-maleic anhydride-co-ethylene glycol methacrylate trimethylsilyl phosphate] ((TMS)2EGMP) and removal of the TMS protecting group. EGMP-TMS (0.66 g 1.88 mmol), F8H2MA (1.0 g, 1.88 mmol), MSA (0.15 g, 1.88 mmol) and AIBN (0.02 g, 0.056 mmol) were dissolved in a mixture of MEK (1.5 g, 1.88 mL) and HFX (2.6 g, 1.88 mL). The reaction mixture was transfer into round bottomed flask, degassed by repeating freeze-pump-thaw cycles and filled with nitrogen. The reaction mixture was placed in oil bath at 65ºC.The reaction was stopped at conversion below 10 % and the product was precipitated in diethyl ether, separated by filtration and dissolved again in acetone. The acetone solution was mixed with 1% water solution of hydrochloride (1:1 vol:vol with respect to acetone) to deprotect the polymer and stirred vigorously for two minutes. Such a mixture was additionally diluted with water to precipitate the copolymer. The solid was separated by filtration, washed with dry diethyl ether to remove water and dissolved in anhydrous acetone and 10 ppm of p-methoxyphenol was added. After drying the copolymer cannot be dissolved in any solvent. 1 H-NMR (CDCl3, δ in ppm) of poly[ethylene glycol methacrylate phosphate-co

1H,1H,2H,2H-perfluorodecyl methacrylate-co-maleic anhydride]: 4.34(s, 2H, -O-CH2-CH2-O-

PO3H2); 4.24 (s, 2H, -O-CH2-CH2-(CF2)8-F); 3.65 (s, 2H, 2H, -O-CH2-CH2-O-PO3H2); 3.59

(s, 3H, -O-CH3); 1.23 (s, -CH2- backbone); 0.85 (s, NN).

Table 2. Compositions of reaction mixtures of (TMS)2EGMP / MSA / F8H2MA – kinetic experiments (for non-kinetic experiments - amount x5)

Exp. (TMS)2 fEGMP F8H2MA fF8H2MA MSA fMSA AIBN MEK HFX CMonomer EGMP [g] [g] [g] [g] [g] [mol/L] [g] A 0.66 0.15 1.0 0.15 0.86 0.70 0.082) 2.5 4.31 2.0 B 0.66 0.33 1.0 0.33 0.18 0.33 0.021) 1.5 2.6 2.0 C 0.6 0.25 1.0 0.25 0.36 0.50 0.0251) 2.0 3.5 1.5 D 1.55 0.175 1.0 0.075 1.84 0.75 0.1642) 10.0 - 2.0 E, E’ 1.55 0.07 1.0 0.03 5.53 0.90 0.412) 25 - 2.0 F 0.665 0.05 1.0 0.05 3.31 0.90 0.2462) 15.04 - 2.0 EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate, AIBN = 2,2’-azo-bis-isobutyronitrile, MEK = methyl ethyl ketone, HFX = 1,3-bis(trifluoromethyl) benzene, CMonomer = total monomer concentration, TMS = trimethylsilyl chloride 1) 2 mol%, 2) 4 mol%

Preparation of NMR samples Small amount of copolymer was refluxed in 10 mL methanol for 12 hours, evaporated to very small volume and 1 mL of deuterium oxide was added. The obtained mixture was kept

101

Chapter 5 on rotary evaporator under conditions that allowed to remove the rest of methanol but the copolymer could not be completely dried.

Preparation of GPC samples Deprotected terpolymer (30 mg) was dissolved in methanol (5 mL) and reflux for 10 hours. Afterwards the product was precipitated in chloroform, filtrated and dissolved in small amount of the mixture of acetone and 1,2-dichloroethane. Into that solution TMSC (1g) was added and the reaction mixture was refluxed for 5 hours. Solvents were evaporated under reduced pressure to very small volume and mixed with stabilized THF for GPC.

Preparation of glass substrate To obtain a hydrophilic surface of glass microscope slides (76x26x1 mm), the substrate has been treated for 2 minutes with a mixture of 3 parts of concentrated sulphuric acid with 1 part of 30% hydrogen peroxide. Afterwards the glass slides were washed with bidest water to remove the acid, and transferred into a glass jar filled with bidest water.

3. Results and Discussion

Commercially available ethylene glycol methacrylate phosphate (EGMP) is a viscous, colorless liquid that contains 1000 ppm of the inhibitor (p-metoxy phenol). In the available literature descriptions [8-15] EGMP was always used as received. It means no effort was made to remove the inhibitor and side products formed during storage. Due to the high boiling point (> 300ºC/760 mmHg), high viscosity, presence of acidic groups and relatively high water solubility one cannot use standard purification procedures as distillation, purification on basic

Al2O3 as well as aqueous extraction. As only manner to remove the initiator in relatively uncomplicated way, was to generate radicals in the solution of the monomer until the monomer start to homopolymerize. In order to avoid any difficulties caused by non-decomposed initiator radicals the monomer solution in MEK was heated to 60ºC under protective atmosphere. The homopolymer which was formed precipitated during the process and the final yield of homopolymer was 5% determined gravimetrically. Any trial of removing residual solvent under reduced pressure in the range of temperature 0-40ºC led to the homopolymerization of the monomer. In order to avoid the polymerization a two-neck flask equipped with rubber septum and needle was used during the distillation. The reduced pressure caused air bubbling trough

102

Chapter 5 the monomer solution and oxygen was acting as an inhibitor. In this way solvent can be evaporated at 40ºC without any problems. Fluorinated terpolymers that contain phosphoric acid moieties were synthesized under homogenous conditions in the presence of AIBN as an initiator. In the standard procedure that was worked out for synthesis of copolymers of fluorinated and alkyl methacrylates (see Chapter 4) a mixture of 2-butanone (MEK) and 1,3-bis(trifluoromethyl)xylene (HFX) was used. A schematic representation of the copolymerization is illustrated in Scheme 1.

CH3 CH3 CH3 CH3

+ * CH2 C CH2 C CH CH  H2C C + H2C C x y z C O C O C C O O O O O O O O O O Rf O

Rf CH2 CH2 H2C

CH2 O HO O O O P

P OH OH HO Rf=CH2CH2(CF2)7CF3 Scheme 1. Synthesis of P[F8H2MA-co-EGMP-co-MSA]

The copolymerization of EGMP with other monomers, yields a crosslinked product, what has been also reported in the literature [16] however authors did not explain the mechanism. As 2-hydroxyethyl methacrylate (HEMA) gives non-crosslinked polymer there is a possibility that the phosphoric acid group of EGMP (derivative of HEMA) causes the crosslinking. One possibility to avoid crosslinking was to protect the phosphoric acid groups with tertiary amines. A series of tertiary amines was checked as protecting agents (see Table 1). Into the solution of the monomer in MEK a double molar excess of the amine was added and the mixture was stirred in an ice bath for two hours. During this process a white solid precipitated which was identified as H3PO4·(NR3)3 complex. The amount of obtained solid corresponds 3% of pure phosphoric acid in the monomer.

Copolymerization of the protected monomer (EGMP·NR3) yielded crosslinked polymers, too. The presence of tertiary amine in the reaction mixture which contains maleic anhydride caused also discoloration of the reaction mixture as well as of the final product. It is known that even traces of tertiary amines during the polymerization of MSA are yielding dark colored products [17] which supposed to be a product of decarboxylation of maleic anhydride [18].

103

Chapter 5

Silylation of ethylene glycol methacrylate phosphate (EGMP) The commonly used protective agent for the protection of strong acids [19] is trimethylsilyl chloride (TMSC). This was also seen as a chance to eliminate the problem with complicated inhibitor removal (the inhibitor used to stabilize EGMP is p-methoxyphenol). However the reaction of TMCS with phenols is slow and the yields are low [20] the process can be enhanced by using 1,2-dichloroethane as a solvent. The silylation reaction in 1,2- dichloroethane is about 100 times faster than in benzene or tetrachloromethane [21]. The method recommended in case of phenols, alcohols etc. consists on a reaction with TMSC in the presence of Li2S in acetonitrile. This method enables to silylate hydroxyl groups under mild conditions with high yield [22]. Because of relatively low content of phenolic inhibitor the reaction was carried out in ethylene chloride. The solution of the monomer in 1,2- dichloroethane was mixed with an excess of TMSC and refluxed. The evolution of HCl is observable only up to one hour however, in order to complete the reaction the process was carried out for five hours. Both the solvent and unreacted silylation agent were removed under reduced pressure at 40ºC. The process yielded a colorless liquid with significantly lower viscosity than the starting feedstock. The monomer prepared according to the described procedure was stored in refrigerator at -18 ºC in order to avoid spontaneous homopolymerization. Scheme 2 illustrates silylation reaction of EGMP with TMSC.

CH3 CH3 Cl CH3 -HCl H2C C CH2 CH2 H2C C + Si CH2 CH2 O O CH3 O O CH3 CH OH 3 CH O CH O 3 3 P O Si P Si HO O H3C O O CH3 CH3

Scheme 2. Silylation reaction of ethylene glycol methacrylate phosphate.

Copolymerization of silylated ethylene glycol methacrylate phosphate ((TMS)2EGMP) with 1H,1H,2H,2H-perfluorodecyl methacrylate and maleic anhydride

The copolymerization of (TMS)2EGMP with F8H2MA and MSA was carried out under homogenous conditions with AIBN as an initiator. Initially a mixture of MEK/HFX (1:1, vol:vol) was used and it was reasonable for monomer compositions with high content of fluorinated methacrylate in order to ensure good solubility of the product. In case of low content of the fluorinated methacrylate in the reaction mixture the presence of expensive and rather

104

Chapter 5 environmental unfriendly fluorinated co-solvent is not necessary. As it has been described in Chapter 4 of this thesis the solvent has no influence on the final terpolymer composition. It means that the properties of obtained polymer do not depend on the composition of the solvent used in the reaction. Scheme 3 describes the copolymerization of silylated EGMP with MSA and F8H2MA.

CH3 CH3 CH3 CH3

+  CH2 C CH2 C CH CH  H2C C + H2C C x y z C O C O C C O O O O O O O O O O O O

Rf CH2 Rf CH2 H2C CH2 Me O Me O O O O Si P Me Me Me P O Me Me O Me Si O Si Me Si

Me Me Me Rf=CH2CH2(CF2)7CF3

Scheme 3. Copolymerization of perfluorinated methycrylate with (TMS)2EGMP and maleic anhydride.

The silylation of the commercially available EGMP enables obtaining non-crosslinked terpolymer in a conventional radical copolymerization reaction. The copolymer in its protected (silylated) form is soluble in MEK, chloroform and similar solvents. The product has been precipitated in diethyl ether, separated by filtration and immediately redissolved in acetone. In order to remove the protective trimethylsilyl groups 1% HCl water solution was added. During this operation the copolymer started to precipitate. Additional portion of water was added to facilitate precipitation and separation of the solid material. Separated copolymer was washed with dry diethyl ether and immediately redissolved in dry acetone. The copolymer after drying could not be dissolved in any solvent. In order to avoid the crosslinking the EGMP terpolymer must be kept in contact to solvent (in solution). In acetone solution this copolymer may be stored for undefined period without any negative consequences. Among the large number of approaches for the deprotection of TMS esters [23,24] the treatment with acidic water was found as the most suitable . The deprotection with aqueous ammonia solution, for example, did not proceed at all. All the polymerization experiments were performed twice. In the first experiment the reaction rates were determined by means of gravimetric method. In the second series all the experiments were interrupted at low conversions in order to avoid the

105

Chapter 5 influence of compositional drift on the polymer properties. The reaction rates and conversions are summarized in Table 3.

Table 3. Determined reaction rates of terpolymerization of EGMP, F8H2MA and MSA. Experiment fEGMP fF8H2MA fMSA AIBN CMonomer Rp Conversio [mol%] [mol/L] [%/min] n [%] A 0.15 0.15 0.70 4.0 2.0 0.65 8.2 B 0.33 0.33 0.33 2.0 2.0 0.64 9.06 C 0.25 0.25 0.50 2.0 1.5 0.51 6.7 D 0.175 0.075 0.75 4.0 2.0 0.64 9.5 E, E’ 0.07 0.03 0.90 4.0 2.0 0.32 8.2 F 0.05 0.05 0.90 4.0 2.0 0.68 8.8 EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate, AIBN = 2,2’-azo-bis-isobutyronitrile, CMonomer = total monomer concentration, Rp = weight rates of polymerization

The expected tendency in the reaction rates according to the literature data as well as to the experimental data described in Chapter 3 indicate a decreasing reaction rate with increase of the MSA fraction in the monomer mixture. The obtained values are difficult to compare with each other due to lower initiator and/or monomer concentration in case of experiments B and C. The reaction rates are comparable to those with 70-90% of MSA in monomer mixture what is reasonable when one considers two times higher concentration of initiator in case of the other experiments. Comparison of experiments A, D, E and F (the same monomer and initiator concentration) shows comparable reaction rates with the exception of experiment E where strong inhibition has been observed. The influence of the concentration of two other monomers on the polymerization rate is difficult to determine on the base of obtained limited experimental data.

Molecular weights Determination of the molecular weight of the copolymers that contain two kind of adhesive groups required a special method for sample preparation. As it was already described in other chapters which deal with maleic anhydride copolymers, the gel permeation chromatography measurement cannot be done by simple dissolving the sample in a suitable solvent. The presence of the in-line filters in GPC setups causes strong adsorption of the copolymer and enormous, rapid increase of the pressure in the system which leads to automatic switch off of the equipment. As a consequence the measurement could not be performed. This problem can be solved by alcoholysis of succinic anhydride rings with methanol. Although, this

106

Chapter 5 kind of treatment functions perfectly with other MSA copolymers, the presence of additional strong anchoring group resulted in above mentioned difficulties. In the process of sample preparation an additional treatment in order to protect the phosphoric acid groups was necessary. The proved method of protection by means of TMSC has been successfully applied. The methanolysed terpolymer was dissolved in a mixture of acetone and dichloroethylene and refluxed in the presence of silylation agent. The non-reacted TMCS as well as solvents were distilled off under reduced pressure to a very small volume and mixed with stabilized THF. In this procedure it is of paramount importance to avoid drying of the terpolymer. Samples prepared according to the described procedure did not cause any difficulties for GPC measurement. The obtained values of molecular weights confirmed the rule that increase of the MSA content in the feed composition results in decrease of molecular weight of the copolymer. The determined molecular weights are summarized in Table 4.

Table 4. Molecular weights (Mn) and molecular weight distributions (Mw/Mn) of P[F8H2MA-co-EGMP-co-MSA].

# Mn Mw/Mn

A 36 800 1.6 B 55 500 1.4 C 37 600 1.6 D 33 300 1.7 E 24 200 1.9 E’ 26 400 1.7 F 26 900 1.8

Polymer composition The standard method of determination of the polymer composition in case of fluorinated terpolymers described in Chapter 4 consists on 1H-NMR spectroscopy. In that case the alcoholysis of succinic anhydride ring with methanol allows precise determination of the incorporated maleic anhydride. Due to the good separation of the characteristic signals in case of mentioned copolymers this method was chosen also in the case of P[F8H2MA-co-EGMP- co-MSA]. It was expected to be the best tool for fast and uncomplicated determination of the terpolymer composition. Figure 1 depicts the 1H-NMR spectrum of P[F8H2MA-co-EGMP-co-MSA] after methanolysis of succinic anhydride rings. The signals 1’a and 1’b around 6 ppm and 2’at 2 ppm are the proton signals next to the double bond belonging to unreacted EGMP. The partly overlapped broad singulets at 4.34 ppm (3) and 4.24 ppm (1) are belonging to EGMP

(-O-CH2-CH2-O-PO3H2) and F8H2MA (-O-CH2-CH2-(CF2)8-F) built in the copolymer.

107

Chapter 5

Unfortunately the signal of -OCH3 group obtained during methanolysis at 3.6 ppm (5) is overlapped completely with the signal from EGMP (-O-CH2-CH2-O-PO3H2). Due to these facts the calculation of the copolymer composition is loaded with significant error.

HO O 2' CH3 CH3 CH3 OH 6 6 O 4' * CH2 C CH2 C CH CH  1'a H O P x y z C O OH C 3' O O O O O OCH3 H O 1 3 5 CH CH 1'b 2 2 4 2 H2C H2C

Rf O OH O P

CHCl 3

6 2 2' 1'a,b 4,5 3 1

8 6 4 2 0

 ppm

Figure 1. 1H-NMR spectrum of P[F8H2MA-co-EGMP-co-MSA]. Product contains non-polymerized EGMP.

The composition of the terpolymer calculated on the base of the 1H-NMR method always led to confusing results. The calculated amount of the MSA in the copolymer exceeded significantly 50 mol% (with the exception of copolymer B) what is in contradiction with the available knowledge about the copolymerization of maleic anhydride. Even the calculations that considered influence of overlapping signals gave unexpected results. The possible compositions calculated on the base of proton NMR are summarized in Table 5. Because the nuclear magnetic resonance did not give clear answer regarding the terpolymers’ composition the elemental analysis was considered as the method which could give unequivocal answer to the still open question: what is the composition of the obtained copolymers? The standard CHN

108

Chapter 5 analysis plus additional determination of fluorine and phosphorus content seemed to be an ideal way of solving the problem.

Table 5. Compositions of monomer mixtures and terpolymers calculated on the base of 1H-NMR data Monomer in the feed Polymer composition # fEGMP(Si) fF8H2MA fMSA FEGMP FF8H2MA FMSA A 0.15 0.15 0.70 0.14 0.21 0.65 B 0.33 0.33 0.33 0.24 0.31 0.45 C 0.25 0.25 0.50 0.21 0.24 0.55 D 0.175 0.075 0.75 0.24 0.11 0.65 E 0.07 0.03 0.90 0.18 0.10 0.72 E’ 0.07 0.03 0.90 0.19 0.11 0.74 F 0.05 0.05 0.90 0.11 0.14 0.75 EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate

The EA has been performed in A.N. Nesmeyanov Institute of Organoelement Compounds of the Russian Academy of Science in Moscow by Dr. Anastasya Buyanowskaya. On the base of EA data, calculations according to the following equation have been done:

EX =  f X ieX i (1) i=1 Where:

EX - the content of the element “X” in the terpolymer (known from elemental analysis)

eXi - the content of the element “X” in monomer “i” (calculated based on chemical structure)

f Xi - the fraction of the element “X” in monomer “i”

The fact that phosphorus and fluorine belong to only one of the monomers makes the calculations even simpler. For example for F8H2MA and EGMP it can be simplified to:

EF = f F ieF i (2) and

EP = f PiePi (3)

The content of the third monomer can be calculated in relation to carbon or hydrogen content. The calculated terpolymer compositions are summarized in Table 6.

109

Chapter 5

Table 6. Elemental composition of water soluble P[F8H2MA-co-EGMP-co-MSA]. # Carbon Hydrogen [wt%] Oxygen Fluorine Phosphorus [wt%] [wt%] [wt%] [wt%] D 35.12 3.45 39.14 18.54 3.67 E 39.29 3.80 41.91 10.65 3.10 E’ 40.52 5.12 40.68 10.46 3.22 F 37.60 3.19 23.03 34.18 1.40

In this case, as well as in case of spectroscopic analysis, the calculated values are different than expected; particularly the content of maleic anhydride in the copolymers is significantly higher than 50 % which is the theoretical limit in binary copolymerization of MSA with methacrylates [5]. Tables 7 and 8 present the calculated polymer composition.

Table 7. Compositions of monomer mixtures and water soluble terpolymers based on the content of FCH, determined by elemental analysis. Monomer composition Polymer composition (FCH) # fEGMP fF8H2MA fMSA FEGMP FF8H2MA FMSA D 0.175 0.075 0.75 0.165 0.805 0.754 E 0.07 0.03 0.90 0.173 0.538 0.773 F 0.05 0.05 0.90 0.920 0.219 0.688 EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate

Table 8. Compositions of water soluble terpolymers based on the content of FPC and FPO, determined by elemental analysis. Polymer composition (FPC) Polymer composition (FPO) # FEGMP FF8H2MA FMSA FEGMP FF8H2MA FMSA D 0.216 0.267 0.523 0.190 0.238 0.572 E 0.213 0.179 0.607 0.194 0.163 0.643 F 0.087 0.538 0.375 0.091 0.561 0.348 EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate

In order to confirm the analytical data a new synthesis of copolymer E has been performed. The elemental analysis gave identical results. The estimated composition depends on the element that has been used for calculation. The possibility that EGMP monomer can contain free HEMA was also considered. Such assumption could explain the “excess” of carbon and hydrogen in the composition of copolymer. The presence of HEMA can be a result of its initial presence in the monomer (which could not be confirmed by analysis) or later hydrolysis of EGMP during deprotection procedure. This could also explain the tendency of the material to undergo crosslinking. The calculations which assume composition of four different monomers, including HEMA, on the base of described above model always gave negative

110

Chapter 5 results. There are many other possible explanations for the “excessive” amount of carbon and hydrogen as ene-reactions, esterification of double bonds of MSA or grafting reactions or presence of impurities e.g. non-reacted monomers or solvents.

Thermal analysis The thermal degradation of obtained copolymers in an oxidizing atmosphere proceeds in three steps. Depending on the composition the first step starts between 120 ºC (copolymer E) and 160 ºC (copolymer C) and ends at 170 ºC and 210 ºC respectively and a mass loss of 55% is observed as depicted in Figure 2. In the second step the weight loss occurs immediately after the previous and ended around 250°C and is of 5 % and is followed by the third degradation and a mass loss of 25 % which ends at 500 °C. The shift of thermogravimetric curves toward higher temperatures has been observed with increasing of the content of perfluorinated methacrylate what is reasonable due to higher thermal stability of fluorinated materials [1]. There were hardly any differences observed between thermograms in the sense of number of degradation steps and in principle no difference in the weight loss in separate curves. The thermogravimetric curves are only shifted toward higher or lower temperature depending on the composition. The remaining weight of about 15-18% at the temperatures above 400 ºC can be assigned to the formation of a carbon char which is usually stable up to 650 ºC in an oxidizing atmosphere. This behavior is possible due to flame retardant properties of phosphorus [15]. The presented thermal degradation behavior of copolymers C, D and E is representative for all the synthesized copolymers A to F. Figure 2 depicts the TGA curves of copolymers C, D and E.

C 100 D E

60 Weight / % / Weight 40

0 0 100 200 300 400 500 600 Temperature / deg C Figure 2. Examples of TGA curves obtained with copolymers C, D and E.

111

Chapter 5

The determined thermal stability of phosphorus containing perfluorinated terpolymers is surprising low particularly in comparison with poly[BMA-co-F8H2MA-co-MSA] and poly[DMA-co-F8H2MA-co-MSA] which showed first signs of thermal degradation at nearly 300 °C. In those cases the thermal degradation is a one step process with a complete weight loss above 400 ºC. The relatively low thermal stability of phosphorus containing copolymers implies limitations in an application of the copolymer as for example release agent in processes carried out at high temperatures. Due to a low thermal resistance of investigated polymers the DSC measurements of P[F8H2MA-co-EGMP-co-MSA] were performed twice. In the first procedure (Figure 3) the first heating was interrupted at 120 °C – the temperature that marked the beginning of thermal decomposition. The second heating proceeded up to 200 °C. The first heating curve shows the beginning of the thermal transition at 100 °C, the second heating depicts thermal transition with the peak maximum at 120-140 °C, and two minor changes at higher temperatures of around 140-150 °C and 160-170 °C respectively. The second DSC measurement (Figure4) was performed with first heating up to 200 °C and exhibits identical thermal transitions as second heating in the first measurement. Since the copolymers undergo decomposition in this range of temperatures the second heating does not reproduce the curves. All the transition observed by means of DSC can be assign, to the thermal decomposition of the copolymers rather than to any other type of thermal transition.

(1)

(2) endo

(3)

-50 0 50 100 150 200 Temperature / deg C

Figure 3. DSC measurement of P[F8H2MA-co-EGMP-co-MSA] (copolymer E) with first heating to 120 °C. (1) – 1st heating, (2) – cooling, (3) 2nd heating. Peak maximum at 128 °C.

112

Chapter 5 endo

(1) (2) (3)

-50 0 50 100 150 200 Temperature /deg C

Figure 4. DSC measurement of P[F8H2MA-co-EGMP-co-MSA] (copolymer E) with first heating to 200 °C. (1) – 1st heating, (2) – cooling, (3) 2nd heating.

In Table 9 thermal transitions of the synthesized copolymers are summarized.

Table 9. Thermal transitions of different P[F8H2MA-co-EGMP-co-MSA].

# 1st transition 2nd transition 3rd transition

A 131 145 164 B 140 152 171 C 140 150 169 D 130 145 163 E 128 145 162 F 130 144 165

Summarizing: both TGA and DSC measurements of investigated series of poly[1H,1H,2H,2H-perfluorodecyl methacrylate-co-maleic anhydride-co- ethylene glycol methacrylate phosphate] show identical thermal behavior in a sense of the number of thermal degradation steps and thermal transitions. The difference that has been observed consists on slightly different temperatures in which the transitions or degradations occurred. The differences can be assigned to the composition of the copolymer but are smaller than one would expect. Due to the unclear composition of the copolymers it is difficult to explain the nature of the transitions in other way than to be caused by thermal decomposition.

113

Chapter 5

Solubility and surface properties The TMCS protected form of copolymer with phosphoric acid moieties shows very good solubility in non-polar solvents such as HFX and its mixtures with MEK. The deprotection procedure changes dramatically the polarity of the copolymer. The solubility after deprotection of different synthesized P[F8H2MA-co-EGMP-co-MSA] in chosen solvents has been investigated. The solubility was assessed as “well soluble” when 1 wt% polymer solution could be obtained by stirring the sample at room temperature for no longer than 1 hour. The exception was made for water solubility test. In this case a period of 24 hours at room temperature was considered. The deprotected terpolymers are well soluble in polar and semipolar solvents. In whole investigated range of compositions the terpolymers are soluble in semipolar solvents as THF as well as polar solvents as acetone and lower alcohols despite the fact that in some cases applying of elevated temperature and strong agitation are necessary in order to obtain clear solutions. The solubility in 1 % ammonia in water occurred in case of materials with low content of perfluorinated monomer in the feed mixture (< 10 %) and high content of maleic anhydride what seems to be reasonable. Solubility in non-polar but halogenated solvents as chloroform occurred only in case of higher fluorine content. All the terpolymers are completely insoluble in hydrocarbons. Table 10 shows the results of the solubility study.

Table 10. Solubility of P[F8H2MA-co-EGMP-co-MSA] in selected solvents. Monomers composition Solubility 1) # fEGMP(Si) fF8H2MA fMSA Acetone MeOH Water THF CHCl3 Hexane EtOH A 0.15 0.15 0.70 + + - + + - B 0.33 0.33 0.33 +/- +/- - + + - C 0.25 0.25 0.50 + +/- - + + - D 0.175 0.075 0.75 + + + 2) + - - E 0.07 0.03 0.90 + + + 2) +/- - - F 0.05 0.05 0.90 + + + +/- - - EGMP = ethylene glycol methacrylate phosphate, MSA = maleic anhydride, F8H2MA = 1H,1H,2H,2H- perfluorodecyl methacrylate, MeOH = methanol, EtOH = ethanol, THF = tetrahydrofurane

1) 1 wt% solution of the copolymer in 1% NH4OH water solution 2) soluble after 24 hours at room temperature “+” - well soluble “-“ - insoluble “+/-“ – soluble at 60 °C after strong agitation

114

Chapter 5

The surface properties of P[F8H2MA-co-EGMP-co-MSA] coatings prepared on glass have been investigated. The samples on glass slides were prepared by means of spin coating at 2500 rpm for 20 seconds. The glass substrate was prepared as described in the Experimental Part of this chapter. The coatings of terpolymers prepared from organic solvents were usually annealed at 100 °C for 2 hours. The measured contact angle values are summarized in Table 11.

Table 11. Contact angle values measured for P[EGMP-co-F8H2MA-co-MSA] coatings measured against water and hexadecane. # Solvent Annealing ΘH2O ΘHD A Acetone 100 °C / 2 h 103±1 62±1 B CHCl3 100 °C / 2 h 120±2 68±1 C Acetone 100 °C / 2 h 104±2 65±1 D Water 100 °C / 2 h 95±1 64±1 D Acetone 100 °C / 2 h 96±1 64±1 E Water 100 °C / 2 h 99±1 65±1 E Acetone 100 °C / 2 h 100±2 66±1 E Acetone 120 °C / 10 min 100±2 66±1 F Water 100 °C / 2 h 101±2 65±1 F Acetone 100 °C / 2 h 101±2 65±1 F Acetone 120 °C / 10 min 103±2 65±1 ΘH2O – contact angle against water, ΘHD – contact angle against hexadecane

The obtained copolymers show relatively good hydrophobicity of 100±5 °. The exception here is the copolymer B for which the measured contact angle against water is 120 °. This is related to the relative high amount of incorporated fluoro-monomer in the copolymer. Despite the fact that the composition could not be well defined based on performed analyses one can assume the relative high fluorine content in this copolymer based on the composition of monomers in the feed as well as to low solubility in polar media. The type of solvent seems to have no effect on the surface properties after annealing. Annealing at higher temperatures and shorter time is possible without measurable influence on the surface properties. The same effects of the solvent and annealing can be observed in the measured contact angle values against hexadecane (Table 11). The effect caused by the copolymer composition seems to be only marginal in case of polymers A, C, D, E and F despite differences in solubility in media of different polarity.

4. Conclusions

The incorporation of phosphoric acid groups into fluorinated terpolymers by means of copolymerization of ethylene glycol methacrylate phosphate leads to a crosslinked product. The crosslinking during polymerization can be prevented by protection of phosphate group with

115

Chapter 5

TMSC. Removal of the protection groups from the terpolymer in acidic media yields the copolymer with mono phosphate acidic moieties. The terpolymer obtained according to this procedure should be kept in contact to solvent since drying leads to crosslinking. The difficulty in determination of the composition of synthesized terpolymers is a serious obstacle to perform continuous addition synthesis. A solution of the copolymer in acetone or water-ammonia can be stored for undefined period of time without any negative influence on the properties. Neither precipitation nor separation has been observed. The films coated from water based solution resulted in water and oil repelling surfaces however, the low thermal stability of EGMP terpolymers limits their application as mold release agents in plastics processing (e.g. injection molding).

References

[1] E. Kissa, Fluorinated Surfactants, Synthesis-Properties-Application, M. Dekker, ed. New York,1984. [2] T.F. DeRossa, Kaufman, B.J.; Sung, R.L.-D.; Russo, J.M.; Polym. Prepr., 1994, 35(1), 718 [3] D. Bikiaris, P. Matzinos, A. Larena, V. Flaris, C. Panayiotou, J. Appl. Polym. Sci., 2001, 81, 701 [4] B.H. Mahlman, et all US Patent 3483276, 1969 [5] B.C. Trivedi, B.M. Culbertson, Maleic Anhydride, Plenum Press, New York, 1982 [6] I. Maege, E. Jaehne, A. Henke, H.-J.P. Adler, C. Bram, C. Jung, M. Stratmann, Progress in Organic Coatings; 1997, 34, 1 [7] T.A. Truc, N. Pebere, T.T.X. Hang, Y. Hervaud, B. Boutevin, Corrosion Science; 2002, 44, 2055 [8] H. Sawada, D. Tamada, T. Kawase, Y. Hayakawa, K. Lee, J. Kyokane, M. Baba, J. Appl. Polym. Sci., 2001, 79, 228 [9] O.N. Tretinnikov, Y. Ikada, Macromolecules, 1997, 30, 1086 [10] H. Ando, M. Nakahara, M. Yamamoto, K. Itoh, Langmuir, 1996, 12, 6399 [11] B.-O. Jung, C.-H. Kim, K.-S. Choi, Y.M. Lee, J.-J. Kim, J. Appl. Polym. Sci., 1999, 72, 1713 [12] K. Kato, Y. Eika, Y. Ikada, J. Biomed. Mater. Research, 1996, 32, 687 [13] T. Jimbo, A. Tanioka, N. Minoura, J. Colloid Interface Sci., 1998, 204, 336 [14] A. Jimbo, M. Higa, N. Minoura, A. Tanioka, Macrmolecules, 1998, 31, 1277 [15] T. Miyata, K. Nakamae, Macromol. Chem. Phys., 1994, 195, 1111 [16] C.I. Lindsay, S.B. Hill, M. Hearn, G. Manton, N. Everall, A. Bunn, J. Heron, I. Fletcher, Polym. Int., 2000, 49, 1183, [17] D. Rittenberg, L. Ponticorvo, Proc. Natl. Acad. Sci. USA., 1960, 46, 822 [18] H. Sack et al., FR Patent 1400556, 1966 [19] H.H. Marsmann, Z. Horn, Naturforsch., 1972, 27 B, 1448 [20] S.H. Langer, S. Connel, I. Wender, J. Org. Chem., 1958, 23 (1), 50 [21] H.H. Hergott, G. Simchen, Synthesis, 1980, 8, 626 [22] G.A. Olah, B.G.B. Gupta, S.C. Narang, R. Malhotra, J. Org. Chem., 1979, 44 (24), 4272 [23] T. Ishizone, G. Uehara, A. Hirao, S. Nakahama, K. Tsuda, Macrom. Chem. Phys.

116

Chapter 5

1998, 199,1827 [24] J.-B. Kim, H. Kim, Polymer, 1999, 40, 405

117

118

Chapter 6

Synthesis, characterization and antimicrobial properties of peptides mimicking copolymers of maleic anhydride and 4- methyl-1-pentene.

1. Introduction

Since in 1960s Cornell and Donaruma [1] have described 2-methacryloxytropones based polymers that exhibited antibacterial activity, antimicrobial copolymers have gained significant interest. Over the last decades a large number of publications and multiple reviews exploring the knowledge on antimicrobial polymers have been published. In the 90s and at the beginning of XXI century the main focus was on the synthesis and the chemical nature of the polymers [2]. For example Kenawy et al. [3] have classified antimicrobial copolymers regarding their structure as quaternary ammonium salts (QAS) [4-20], phosphonium salts [20-27], sulfonic acid derivatives (salts, sulfonamides) and N-halamines. In most recent reviews the main attention has been paid to the bactericidal mechanism and influence of relevant parameters such as molecular weight and charge distribution [28]. The antimicrobial properties of maleic anhydride copolymers with olefins [29] and styrene [30-32] have been investigated starting from 1960s. In all cases the succinic anhydride functionality was used as a base for further modification in order to introduce an antibacterial moiety (diamine, aminophenol etc.). From this perspective the modified maleic anhydride copolymers belong to one of the groups mentioned above. These polymers are usually amphiphilic and act as surfactants. At the same time, antimicrobial peptides represent a large group of natural compounds with a broad spectrum of antimicrobial activity [33-36]. The antimicrobial activity of these molecules also comes from their amphiphilic structure [34]. In the past a lot of effort has been made in order to understand the mechanism of cytotoxicity of amphiphilic peptides. One of the examples is Melittin from bee venom, with the highest hemolytic activity. Compared to Magainin from Xenopus frog skin it has a high content of hydrophobic residues). It has a 26-amino acid residue-long sequence with a characteristic cluster of lysine and arginine residues. Magainins, 23-residue-long peptides, are essentially toxic for bacterial strains while being poorly hemolytic [37]. Synthetic peptides of different composition and structure have been widely investigated for their antimicrobial properties. Particular attention has been paid to peptides which are only constituted by nonpolar leucine

119 Chapter 6 and charged lysine, so-called LK-pepties [37, 38]. These molecules are ideally amphiphilic and proper choice of the Lys:Leu ratio as well as the chain length enables designing structures most similar to natural toxins. The aim to create a new synthetic, amphiphilic structure with strong antimicrobial properties comparable to natural antimicrobial peptides led to the choice of an alkene and maleic anhydride copolymer. Lower alkenes and maleic anhydride yield alternating copolymers. This ensures constant 1:1 ratio of the hydrophobic and cationic part similar to those in Leu:Lys 1:1 LK-peptides. The choice of 4-methyl-1-pentene as hydrophobic co-monomer is based on the similarity of its structure with leucine (cf. Scheme 1) while the choice of maleic anhydride leaves ample space for further design of the hydrophilic part by means of chemical modification.

H2N COOH

leucine 4-methyl-1-pentene Scheme 1. Comparison of the structure of leucine with 4-methyl-1-pentene

This work covers polymeric quaternary ammonium salts obtained by chemical modification of maleic anhydride 4-methyl-1-pentene copolymers. The free radical copolymerization of maleic anhydride with 4-methyl-1-pentene leads to alternating copolymers, which were grafted using N,N-substituted diamines. Although the antimicrobial activity of non-quaternized N,N-substituted derivatives has been reported [1], biocidal activity should be enhanced by quaternization of the tertiary amine groups with different alkyl halides. The higher biological activity of positively charged moieties can be explained by electrostatic interaction with negatively charged cell surfaces. The advantage of polycations is based on the higher charge density that allows highly extended adsorption on the bacterial cell. The expected penetration of the cell wall and disruption of the membrane by the lipophilic alkane chain releases cytoplasmic constituents and leads to the death of the cell.

120 Chapter 6

2. Experimental

Maleic anhydride (MSA, Merck, for synthesis) was sublimed under reduced pressure (50 °C, 2.4x10-2 bar), benzoyl peroxide (BPO, Merck, for synthesis) was recrystallized from a chloroform : methanol mixture (1:5 vol:vol), 2-butanone (MEK, Merck, p.a.) and other solvents (technical grade) were dried over calcium hydride (Fluka) and distilled before use. 4-methyl-1- pentene (Aldrich, for synthesis), methyl iodide (ABCR), benzyl alcohol (Merck, for synthesis), dimethyl sulfoxide (Fluka, p.a.), acetic anhydride (Merck, for synthesis), 3- dimethylaminopropylamine (Aldrich), succinic anhydride (Merck, p.a.), triethylamine (Fluka, p.a.) and anhydrous sodium acetate (Merck, p.a.) were used without additional purification. The inorganic salts sodium chloride, sodium dihydrogen phosphate monohydrate, disodium hydrogen phosphate dehydrate (all Merck) were used as received.

Bacteria have been provided by the German Resource Centre for Biological Material (Leibniz‐Institut DSMZ‐Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig) and LGC Standards GmbH, Wesel).Gram-negative bacteria: Escherichia coli (DSM 498, ATCC 23716), Pseudomonas aeruginosa (DSM 1117, ATCC 27853) Gram- positive bacteria: ; Staphylococcus aureus (ATCC 6538), Staphylococcus epidermidis (DSM 1798, ATCC 27853)

Nutrient solutions: (1) NL1: 0.5 g of meat extract (Merck) and 0.3 g of peptone obtained from casein (Merck) were dissolved in 99.2 g of sterile water. (2) Mueller-Hinton-Broth (MHB, Roth) per L: 2.0 g Beef Extract Powder, 17.5 g Acid Digest of Casein, 1.5 g Starch pH 7,4 ±0,2

Phosphate buffered saline (PBS) 100 mL of phosphate buffer and 1.8 g of sodium chloride were placed in 200 mL measuring flask and filled with distilled water ad 200 mL.

Measurements/Apparatus

Size exclusion chromatography was performed using a system consisting of a LC 1120 pump (Polymer Laboratories), a UV detector ERC-7215 and RI detector ERC-7515A (ERMA CR INC.), a precolumn (50x8 mm) of nominal pore size 50Å, and four columns (300x8 mm) filled with MZ-Gel SDplus of nominal pore size 50Å, 100Å, 1000Å, 10000Å (MZ- Analysentechnik, Mainz, Germany). The set was calibrated with PMMA and PS standards from Polymer Laboratories. The sample concentration was 7 mg of polymer in 1 mL of

121 Chapter 6 tetrahydrofuran, the injected volume of the sample was 100 μL. The tetrahydrofuran was stabilized with 2,6-di-tert-butyl-4-methylphenol (250 mg/L). 1 H-NMR spectra were obtained on a Bruker DPX-300 spectrometer in acetone-d6 and

D2O at 300 MHz. MestRe-C 4.9.0.0 was used as evaluation software. The solvent peak as the reference signal was used. IR spectra were measured with on FT-IR Thermo Nicolet Nexus spectrometer in KBr- pellets with a resolution of 4 cm-1. Thermogravimetric analysis (TGA) was performed by means of NETZSCH TG 209c thermo balance under nitrogen atmosphere at a nitrogen flow of 15 mL/min. Samples of 9-11 mg were placed in a standard NETZSCH alumina 85 L crucible. The heating rate was 10 Kmin-1. Differential scanning calorimetry (DSC) measurements were performed by means of Netzsch DSC 204 unit. Samples (typical weight: ~9 mg) were enclosed in standard Netzsch 25 μL aluminium crucibles. Indium and palmitic acid were used as calibration standards. Heating and cooling rates were 10 °C min-1. CHN elemental analysis was performed by means of Carlo Erba MOD-1106 elemental analysis apparatus in the Institute of Organic Chemistry of the RWTH Aachen. Each measurement was performed twice. The optical density measurements were performed by means of Tecan Genios PRO Multiwell plate Reader/Incubator at wave length 612 nm in cycles of 30 minutes for 20 h.

Syntheses

Poly[(4-methyl-1-pentene)-alt-maleic anhydride] (copolymer 1 (C1)) . In a 100 mL round bottomed flask equipped with a reflux condenser and a valve, maleic anhydride (7g, 71 mmol, 55 mol-%) and of 4-methyl-1-pentene (5.04g 60 mmol) were dissolved in dry MEK (35 mL). To this mixture BPO (0.3g, 1.23 mmol, 0.93 mol-%) was added and the mixture was degassed for three times by freeze-pump-thaw cycles and filled with nitrogen. The reaction mixture was placed in an oil bath at 80°C for 8 hours. After the given time the reaction mixture was cooled to ambient temperature and the polymer was precipitated in a mixture of diethylether : methanol (4:1 vol:vol), separated by filtration and dried under vacuum at 40 °C for 8 hours. The copolymer (9.3g, 77 % yield) was obtained as a white solid. The molecular weight of the obtained copolymer was determined by THF-GPC using PMMA standard: Mn=5800, Đ=1.67. The characteristic absorption peaks of IR spectra are: 1852, 1777(C=O); 927 (C-O-C) cm-1.

122 Chapter 6

Elemental analysis for C10H14O3; calculated: C 65.9 ; H 7.7; O 26.4 wt%, found: C 65.54; H 1 7.85; O 26.61 wt%. H-NMR (acetone-d6, δ in ppm): 0.71-4.03; (m, backbone and side chains)

Determination of the maleic anhydride content of C1 by 1H-NMR. C1 (100 mg) was dissolved in 2-butanone (5 mL) and benzyl alcohol (0.5g) was added. The mixture was heated at 50 °C under reflux for 18 hours. The product was precipitated in a diethylether/methanol (1:1) mixture 1 and dried under vacuum at 40°C for 8 hours. H-NMR (CDCl3, δ in ppm): 7.16-7.49 (m, 5H, aromatic); 5.11 (2H, -O-CH2-); 0.71-4.03 (m, 14H, backbone and side chains)

Amidoacidification of C1 to poly[(4-methyl-1-pentene)-alt-(1-(3-N,N- dimethylaminopropyl)malemic acid)] (copolymer 2 (C2) . C1 (2.0 g) was dissolved in dry DMF (45 mL) and DMAPA (1.19 g, 11.7 mmol) was added dropwise during 2 hours to the stirred solution using a syringe pump at room temperature. During the addition phase precipitation of the product occurred. The mixture was stirred for additional 24 hours, the precipitated product, a cream-colored solid, was filtered off and dried at 40°C under reduced pressure for 8 hours. The characteristic absorption peaks of IR spectra are: 1714; 1657; 1567cm-1.

Thermal imidization of C2 to poly[(4-methyl-1-pentene)-alt-(1-(3-N,N- dimethylaminopropyl)maleimide)] (copolymer 3a (C3a)). Amidoacidified C2 (1.0 g) was dissolved in DMF (20 mL) and heated under reflux at 120°C for 24 hours. The product was precipitated from diethylether and dried under vacuum at 40°C for 5 hours. The molecular weight of the obtained copolymer was determined by THF-GPC using PMMA standard: -1 Mn=8600, Đ=1.46. The characteristic absorption peaks of IR spectra are: 1772; 1697 cm .

Elemental analysis for C15H26O2N2; calculated: C 67.7; H 9.8; N 10.5; O 12.0 wt%; found: C 67.09; H 9.70; N 10.11; O 11.10 wt%.

Chemical imidization of amidoacidified C2 (copolymer 3b (C3b)). The amidoacidified C2 (0.5 g) was dissolved in a mixture of DMF, acetic anhydride, 2-butanone triethylamine or sodium acetate (see Table 1 for detailed composition). The reaction mixture was either heated under reflux or stirred at RT and under a protective atmosphere of nitrogen for 12-18 hours. The product was precipitated in diethylether as a dark brown solid and dried under vacuum at 40° for 24 hours. The repeated precipitation had no influence on the appearance. The characteristic absorption peaks of IR spectra are: 1770; 1697 cm-1.

123 Chapter 6

Table 1. Composition of the reaction mixtures, reaction conditions for the chemical imidization of C2, yield and appearance of the product. # DMF Ac2O MEK Et3N CH3COONa T t Yield Appearance [g] [g] [g] [g] [g] [°C] [h] % 1 15 15 - 0.1 - RT 12 88 brown powder 2 15 15 - 0.1 - 80 18 70 brown powder 3 20 1.07 - 1.41 - 80 24 62 brown, tarry 4 - 1.0 15 0.8 - 90 18 87 grey powder 5 - 0.55 15 0.73 - 90 12 90 grey powder 6 - 7 - 1 - 80 20 78 grey powder 7 - 10 - - 0,1 80 18 69 brown powder DMF = dimethylformamid, Ac2O = acetic anhydride, MEK = 2-butanone, Et3N = triethylamine, CH3COONa = sodium acetate, RT = room temperature

Preparation of imide C3 in a one pot reaction (copolymer 3c (C3c)). C1 (3.5 g) was dissolved in dry DMF (20 mL) and placed in a 250 mL round bottomed flask. Within 2 hours DMAPA (2.5 g) dissolved in dry DMF (40 mL) was added by means of a syringe pump at room temperature. The dispersion of the precipitated product was stirred additionally for 12 hours and the mixture was placed in an oil bath heated to 120°C for another 24 hours. The product was purified by precipitation in 400 mL of diethylether. After drying the precipitate under vacuum at 40°C for 8 hours a cream-coloured powder was obtained (3.2g, 97.5%). The molecular weight of the obtained copolymer was determined by THF-GPC using PMMA standard: Mn=9350, Đ=1.40. The characteristic absorption peaks of IR spectra are: 1772; 1697 -1 cm . Elemental analysis for C15H26O2N2; calculated: C 67.7; H 9.8; N 10.5; O 12.0 wt%; found: C 67.16; H 9.76; N 10.21; O 12.87 wt%.

Synthesis of poly[(4-methyl-1-pentene)-alt-(1-(3-N,N,N-trimethylammonium-propyl)- maleimidoiodide)] by quaternization of imidized C3 with methyl iodide (copolymer 4 (C4)). Thermally imidized C3a (0.1g) was dissolved in DMSO (4 mL) and placed in a 25 mL round bottomed flask. Methyl iodide (0.1 mL) was added and the mixture was stirred at room temperature for 18 hours. The product was precipitated in THF and dried under vacuum at 40°C for 12 hours. The product (0.147g, 96 %) was obtained as a lemon-yellowish powder.

Elemental analysis for C16H29O2N2I; calculated: C 47.1; H 7.2; N 6.9; O 7.6; I 31.2 wt%; found: C 46.63; H 7.0; N 6.97; O 7.87 wt%.

Sequential quaternization of imidized C3 with methyl and dodecyl iodide (copolymer 5 (C5)). Thermally imidized C3a (0.3 g) was placed in a 25 mL round bottomed flask and dissolved in a mixture of acetone (10 mL) and DMSO (5 mL). Dodecyl iodide (0.164 g, 0.564 mmol, 50 % with respect to the N,N-dimethylammonium groups) of) was added and the mixture was stirred

124 Chapter 6 for 48 h. In a subsequent step methyl iodide (0.2 g, 1.4 mmol) dissolved in DMSO (5 mL) was added and the reaction mixture was stirred for another 24 hours. The product was precipitated in a THF/hexane (4:1) mixture, separated and dried under vacuum at 40ºC for 8 hours.

Elemental analysis for C43H80O4N4I2; calculated: C 53.2; H8.2; N5.8; O 6.6; I 26.2 wt%; found: C 53.0; H 8.17; N 5.84; O 6.9 wt%.

Synthesis of poly[(4-methyl-1-pentene)-alt-(1-(3-N,N-dimethyl-N-dodecylammoniumpropyl)- maleimidoiodide)] (copolymer 6 (C6)). Thermally imidized C3a (0.1g) was dissolved in acetone (4 mL) and placed in a 25 mL round bottomed flask and dodecyl iodide (0.1 mL) was added. The flask was closed with a glass stopper and the mixture was stirred for 48 hours. The product was precipitated in hexane, filtered and dried under vacuum at 40ºC for 8 hours.

Elemental analysis for C27H51O2N2I; calculated: C 57.6; H 9.0; N 5.0; O 5.7; I 22.7 wt%; found: C 57.56; H 9.06; N 5.01; O 5.88 wt%.

Synthesis of 3-(N,N-dimethylamino)propyl maleic acid amide (M1). Maleic anhydride (7.0 g, 71 mmol) was dissolved in dry chloroform (100 mL) and placed in a 250 mL flask. A solution of DMAPA (6.8 g, 67 mmol) in chloroform (20 mL) was added dropwise and stirred for 1 hour at 60°C. After the addition period the reaction mixture was heated at 60°C for one hour. The product was precipitated in an 5-fold excess of acetone (with respect to the volume of the reaction mixture), separated and dried under vacuum at 40 °C for 8 hours. The product (12.2 g, 88%) was obtained in as a white powder. The characteristic absorption peaks of IR spectra are: -1 1 3 1710, 1650, 1588 cm . H-NMR (D2O, δ in ppm): 6.33 (dd, 1H, C=CH-CON-, J=12 Hz); 5.96 3 3 (dd, 1H, C=CH-COOH, J=12Hz); 3.33 (t, 2H, -NH-CH2-CH2-CH2-N(CH3)2, J=6 Hz); 3,17 (t, 3 2H, -NH-CH2-CH2-CH2-N(CH3)2, J=9 Hz); 2,88 (s, 6H, -NH-CH2-CH2-CH2N-(CH3)2); 1,95 ( 3 q, 2H, -NH-CH2-CH2-CH2N-(CH3)2) J= 6 Hz).

Chemical imidization (Z)-4-(N,N-dimethylamino)propylamino-4-oxobut-2-enoic acid (M2) Maleamic acid (1.0 g, 5 mmol) and sodium acetate (0.2 g, 2 mmol) were dissolved in acetic anhydride (30 mL) and heated in a 50 mL round bottomed flask equipped with a reflux condenser for 4 hours at 90°C. After cooling to ambient temperature the reaction mixture was filtered to remove sodium acetate, and acetic anhydride was distilled off under reduced pressure. A brown oily product (0.8 g, 72 %) was obtained. The characteristic -1 1 absorption peaks of the IR spectra are: 1773; 1697 cm . H-NMR (D2O, δ in ppm): 6.86 (s, 2H, 3 -OC-CH=CH-CO-); 3.62 (t, 2H, -NH-CH2-CH2-CH2-N(CH3)2, J=6 Hz); 3.14 (t, 2H, -NH-

125 Chapter 6

3 CH2-CH2-CH2-N(CH3)2, J=9 Hz); 2.86 (s, 6H, -NH-CH2-CH2-CH2N-(CH3)2); 1.98 ( q, 2H, - 3 NH-CH2-CH2-CH2N-(CH3)2) J= 6 Hz).

Synthesis of 3-(N,N-dimethylamino)propyl succinamic acid (M3). DMAPA (3.6 g, 35 mmol) was added slowly to a solution of succinic anhydride 3.5 g (35 mmol) in dry acetone (20 mL). The mixture was stirred for 2 hours at room temperature. The precipitated product was filtered and immediately used for the thermal imidization reaction without further characterization.

Thermal imidization of (Z)-4-(N,N-dimethylamino)propylamino-4-oxobutanoic acid (M4) 3- (N,N-dimethylamino)propyl succinamic acid obtained from the reaction of succinic anhydride and DMAPA was heated under flowing nitrogen at 170°C for 2 hours. A brownish oily liquid (6.3 g, 98 %) was obtained. The characteristic absorption peaks of IR spectra are: 1770; 1697 -1 1 3 cm . H-NMR (D2O, δ in ppm): 3.23 (t, 2H, , -NH-CH2-CH2-CH2-N(CH3)2, J=6 Hz); 2.97 (t, 3 2H, -NH-CH2-CH2-CH2-N(CH3)2, J=6 Hz); 2.73 (s, 6H, -NH-CH2-CH2-CH2N-(CH3)2); 2.41

(s, 4H, -OC-CH2-CH2-CO-); 1.85 (q, 2H, -NH-CH2-CH2-CH2N-(CH3)2

Antimicrobial tests

The antibacterial activity of the amphiphilic polymers in solution was determined by measuring the minimum inhibitory concentration (MIC) using the test bacteria mentioned above. Suspensions of strains with known colony forming units (CFU; 2x106 CFU/mL) were incubated at 37°C in nutrient solution (Mueller-Hinton Broth, MHB) with different concentrations of the polymer samples. The polymer samples were solubilized in bidistilled water and added to the nutrient solution at a constant ratio of 1:10. The growth of the bacteria was followed during the incubation over 20 h by measuring the optical density at 612 nm every 30 min (with 1000 s of shaking at 100 rpm per 30 min cycle by using a micro well plate reader/incubator (TECAN Infinite 200 Pro, Tecan Trading AG, Männedorf, Switzerland). The testing is performed with defined concentrations specifically for each polymer until, within the monitoring time of 20 h, no bacterial growth curve is recorded. All experiments were performed in triplicate duplicates and MIC determination was repeated on three different days. The polymers were not sterilized. Sterile controls (defined polymer concentrations in nutrient solution without bacteria) were assessed in every growth curve monitoring testing series. MICs were determined according to broth microdilution in 96-well microtitre plates [Deutsches Institut für Normung. Medical Microbiology—Susceptibility Testing of Microbial Pathogens to

126 Chapter 6

Antimicrobial Agents—Part 7: Determination of the Minimum Bactericidal Concentration (MBC) with the Method of Microbouillondilution; Deutsches Institut für Normung: Berlin, Germany, 2009]. The minimum inhibitory concentration (MIC) corresponds to the concentration of the test substance at which a complete inhibition of the growth of the inoculated bacteria was observed by comparison with control samples without test substance.

Hemolytic activity

Hemolytic activity was assessed according to literature [40]. Human erythrocytes (from healthy donors, red blood cells (RBC), 0, Rh positive, citrate-phosphate-dextrose-adenine- stabilized; CPDA1 Sarstedt Germany) were obtained by centrifugation (3500 rpm, 12 min) to remove plasma, washed three times in PBS (0.01 M phosphate buffered saline, Sigma Aldrich Chemie GmbH, Steinheim, Germany), and diluted in PBS to obtain a stock solution of 2.5 × 108–3.0 × 108/mL RBC. Solutions of defined polymer concentration (250 µL) were pipetted into 250 μL of the stock solution; the final amount of RBC being 1.2 × 108 –1.5 × 108 RBC/mL. The RBC were exposed for 60 min at 37 °C under 3D-shaking; centrifuged thereafter (4000 rpm, 12 min) and the absorption of the supernatant (diluted 10-fold in PBS) was determined at 414 nm in a microplate reader. As reference solutions, (i) PBS for determining spontaneous hemolysis and (ii) 1% Triton X-100 for 100% hemolysis (positive control) were used. Hemolysis was plotted as a function of polymer concentration and the hemolytic activity was defined as the polymer concentration that causes 50% hemolysis of human RBC relative to the positive control (HC50).

3. Results and discussion

Copolymerization of maleic anhydride with 4-methyl-1-pentene

Alternating copolymers of maleic anhydride with alkenes were obtained by free radical copolymerization in solution in the presence of radical initiators [39-43]. However copolymers which contain an excess of the olefin have also been described in the literature [44,45], but as a general rule alternating copolymers are formed, in particular when an excess of anhydride was used. The alternating copolymer of 4-methyl-1-pentene with maleic anhydride (MSA) was synthesized accordingly via free radical polymerization. The copolymerization was carried out under homogenous conditions in anhydrous 2-butanone (MEK) at 80°C in the presence of

127 Chapter 6 benzoyl peroxide (BPO) as an initiator. Scheme 2 depicts the copolymerization reaction of 4- methyl-1-pentene with maleic anhydride:

O O O

MEK, BPO * CH CH CH CH * + 80 C 2 O O 8 h, 77% n O

Scheme 2. Copolymerization of MSA with 4-methyl-1-pentene

An excess of maleic anhydride was used to ensure equimolar composition of the resulting copolymer. The product was separated from the reaction mixture by precipitation in a

Et2O:MeOH (4:1, vol:vol) mixture. For its molecular weight determination by gel permeation chromatography (GPC), the succinic anhydride units of the copolymer were methanolized at room temperature. The methanolysis is necessary because maleic anhydride copolymer adsorbed on the inline filter, impeding any measurement. GPC measurement of the methanolyzed copolymer was performed in THF with PMMA standards and the determined values were: Mn=5800 and Mw/Mn=1.67. The composition of the obtained polymer could not be determined by 1H-NMR because of overlapping signals of both monomers. Figure 1a depicts a typical 1H-NMR-spectrum of a P[MP-alt-MSA] copolymer. The range of overlapping signals between 1 and 4 ppm precludes the calculation of the polymer composition in case of the non-methanolyzed copolymers (MSA protons 3.25 ppm) as well as methanolyzed products (MSA protons 2.8 ppm) [45]. Treatment of the copolymer with benzyl alcohol in MEK results in the formation of monobenzyl esters. This method allows determining the content of protons of succinic acid benzyl ester since the benzyl groups NMR signals were well separated from the other polymer peaks. Figure 1b depicts the 1H-NMR spectrum of a P[MP-alt-MSA] copolymer after reaction with benzyl alcohol. The aromatic protons, as well as the benzylic protons are well separated from the signals of the polymer backbone and the alkyl side groups. The monomer composition was calculated from the integrated signal intensity of the benzylic protons and that of the alkyl signals between 0.8 and 5.2 ppm according to equation 1. A F = m (Eq. 1) MSA A B + m n A- integration value of benzylic proton signal (σ= 4.90 – 5.30 ppm) m- number of benzylic protons = 2 128 Chapter 6

B- integration value of polymer backbone and side chains proton signal (σ=0.70 – 4.2) n-number of protons in signal mentioned in B = 14

(1')

CH3 (1) acetone CH (2) H3C (3) CH2 (6) (6') * CH2 CH CH CH n* (4) (5) C C (4) (2) O O O (6,6') (3)

(1,1')

10 8 6 4 2 0

 ppm

B CH3

CH H3C CH2

* CH2 CH CH CH n*

C C O O HDO OH O (1)

H C (2) 2

aromatic protons (1) Benzylic protons (2)

polymer backbone and side chains

10 8 6 4 2 0

 ppm

1 1 Figure 1. (A) H-NMR spectrum of P[MP-alt-MSA] measured in acetone d6; (B) H- NMR spectrum of P[MP-alt-MSA]esterified with benzyl alcohol measured in D2O.

129 Chapter 6

This method was developed for determination of the MSA content in terpolymers, and it was proved that without a catalyst only one carboxyl group was esterified [46]. It is very useful and precise in the range of error of the used spectroscopic method. The 46 ± 5 % of calculated content of succinic anhydride in the copolymer is reasonable and allowed to assume that an alternating copolymer was obtained. The composition of the copolymer was also determined by elemental analysis. The obtained results correspond well to the composition calculated for alternating copolymer (see Table 2).

Table 2. Comparison of the experimental and calculated elemental composition of an alternating copolymer poly[(4-methyl-1-pentene)-alt-maleic anhydride] (C10H14O3). Carbon Hydrogen Oxygen Calculated for C,H,O 65.90 7.70 26.40 Found 65.54 7.85 26.61

Grafting of DMAPA onto P[MP-alt-MSA] copolymer

Model reaction of amidoacidification

Maleic anhydride can easily react with 3-(dimethylamino)-1-propylamine (DMAPA) to yield 3-(N,N-dimethylamino)propyl maleamic acid (M1). The reaction was carried out in chloroform by addition of the amine to MSA at room temperature and subsequent heating to 60ºC. The reaction is exothermic and proceeds easily, hence the heating step was performed to ensure complete reaction. The product was precipitated in a 5-fold excess of acetone and dried. The desired compound has been obtained as a white powder. Scheme 3 depicts the reaction of maleic anhydride with DMAPA:

H3C HO + N H2H O O CH3 O O O HN

(1) (2) se Ba + O Ac 2 r T o (M1) N

O O N

(M2) N

130 Chapter 6

Scheme 3. Model reaction of amidoacidification and imidization of maleic anhydride with DMAPA

Because of its molecular asymmetry the product can be easily identified and distinguished from non-reacted MSA by means of 1H-NMR. In the course of the reaction the single peak of MSA around 7 ppm (double bond protons) disappeared and two doublets appeared at 6.33 and 5.96 ppm respectively (see Figure 2).

(1) (2) (6,6') H H

O N H

OH H2C (3)

CH2 (4)

H2C (5) HDO N CH3 (6) H3C (6') (5)

(1) (2) (3) (4)

6 4 2 0

 ppm

Figure 2. 1H-NMR spectra of cis-3-(N,N-dimethylamino)propyl maleamic acid.

Amidoacidification and imidization of poly[(4-methyl-1-pentene)-alt-maleic anhydride

Scheme 4 depicts the reaction of the anhydride moiety in the polymeric chain (C1) with DMAPA yielding amidacid (C2) and followed by thermal imidization which yields the cyclic N-substituted imide (C3):

131 Chapter 6

CH CH CH CH2 + H2N N CH CH CH CH2

O O O O O OH NH

C1 C2

-H2O N

CH CH CH CH2

O O N

N C3

Scheme 4. Amidoacidification and imidization of MSA copolymer.

The amidoacidification of maleic anhydride copolymer with N,N-dimethylamino-1- propyl amine (DMAPA) was performed in analogy to the model reaction [18,42,47-58]. The amine was added dropwise to a DMF solution of the copolymer. The product precipitated from the reaction mixture and could be easily obtained in its pure form. Because of the heterogeneity of the reaction mixture an excess of amine was used and the mixture was stirred for about 20h to obtain an almost quantitative conversion. This intermediate product was separated by filtration and analysed. The prepared macromolecular amic acid could not be characterized well by means of NMR spectroscopy, because of the occurrence of broad overlapping signals. The IR spectra of the amidoacidified copolymer C2 showed in Figure 3 corresponded to the comparative spectrum of the model product (Figure 3 M1) from the reaction of maleic anhydride with DMAPA.

132 Chapter 6

3,0

-1 2,5 1697 cm

2,0

1,5 M2

Absorbance 1,0 C3

0,5 M1

C2 0,0

500 1000 1500 2000 2500 3000 3500 Wavenumbers / cm-1

Figure 3. IR spectra of amidoacidified C2, thermally imidized C3 and model compounds M1 and M2 (3-(N,N-dimethylamino)propyl maleamic acid and 3-(N,N- dimethylaminopropyl)maleimide). The characteristic absorption band of cyclic imides in C3 ans M2 at 1697 cm-1 are shown.

To determine the content of modified succinic anhydride moieties, elemental analysis was employed. The standard measurement for carbon, hydrogen and nitrogen determination showed a strong correlation to the calculated values for the hydrogen and the nitrogen content while measured content of carbon strongly differs from the calculated value (Table 3). However the product was found to be hygroscopic [48] and the standard analytical treatment did not include drying step. When one recalculates the elemental composition of the copolymer assuming that each amic acid unit binds additionally two molecules of water, the determined elemental composition becomes reasonable (see Table 3, #3).

Table 3. Comparison of the CHN elemental analysis results of the amidoacidified copolymer with calculated values. # Carbon Hydrogen Nitrogen # 1 Calculated for C15H28N2O3 64.30 9.80 9.80 # 2 Calculated for C15H28N2O3·2H2O 56.23 10.07 8.74 # 3 Found 56.18 9.73 9.78

The above mentioned hypothesis was verified in a simple experiment. A small sample of the C2 was dried in vacuo to a constant weight and placed in an open vessel on a very precise

133 Chapter 6 balance to absorb the humidity from air. A rapid increase in weight of the sample was observed. The calculated difference between initial mass of the sample and the mass after 12 hours of exposure to humid air showed that the composition of the copolymer was C15H28N2O3·1.86H2O. The difference between expected and obtained value can be explained as a result of insufficient drying, absorption of water during transfer from the drying oven to balance (the initial period of the experiment showed rapid increase of the weight), degree of modification lower than assumed 100% or impurities. Most probably it should be treated as the combination of several reasons however the measured ~1.9 moles H2O per amic acid unit are very close to the postulated stoichiometric 1:2 composition.

Chemical imidization

The imidization reaction of amic acid in the presence of a dehydration agent and a base has been described in many publications [49,50,56-58]. The most common reagents are acetic anhydride in the presence of triethylamine or sodium acetate. This reaction is usually carried out under mild conditions (temperature 80-90 °C) and is useful for synthesis of a wide range of N-substituted imides. The reaction was first tested by means of a model reaction between cis-3-(3- dimethylaminopropyl carbamoyl) acrylic acid and acetic anhydride. As depicted in Scheme 3 the substrate (M1) was heated at 90 °C for 4 hours in the presence of acetic anhydride (dehydrating agent) and sodium acetate (catalyst). At the end of the reaction time the mixture was cooled down and sodium acetate was separated by filtration. Acetic anhydride, as well as acetic acid formed during the reaction, were distilled off under reduced pressure. The product was obtained as a brown oily liquid. 1H-NMR analysis of the crude 3-(3-dimethylaminopropyl)maleimide (see Figure 4) shows trace signals of double bond protons (between 6 and 6.5 ppm) from unreacted maleic acid amide which is about 2 % in respect to maleimide.

134 Chapter 6

(1) (1') H H

O O N HDO

CH2 (5') (2) (5,5',6) H2C (3) (6) CH3 H CH2 N (4) double bond CH3COO CH protons (opened ring) (7) 3 (5) acetic acid (1,1') (7) ?

-CH - amine chain (3,2,4) 2 9 8 7 6 5 4 3 2 1  ppm

Figure 4. 1H-NMR spectra of crude 3-(3-dimethylaminopropyl)maleimide. Characteristic peak at ~7 ppm confirms the ring formation.

However, the singlet at ~7 ppm shows that the 5-membered ring has been successfully formed. The comparison between the signal at σ=7 ppm and aliphatic protons (σ=2-4 ppm) which belong to the amine chain shows that less than 50% of the material that has been obtained was converted to the required compound. The excess of aliphatic amine protons cannot be explained by incomplete conversion of the amic acid because of the very low signal intensity of its double bond protons. Also a very strong signal of acetic acid has been registered, although the sample was kept under vacuum for long time. The IR spectrum (Figure 3) shows the characteristic absorption bands of cyclic imides at 1697 cm-1 and 1714 cm-1 that confirm the imide formation.

The reaction of amidoacidified C2 in the presence of Ac2O and both mentioned bases, always yielded dark brown products of sometimes even tarry consistency. Although characteristic absorption bands of imides were observed, no way was found to purify the macromolecular product or at least to remove the dark colour.

Thermal imidization

Because of the poor quality of obtained products, other possibilities of imidization were explored. Thermal imidization of amic acids, as a main synthesis method, has been described in many organic chemistry handbooks and publications [18,51-56].

135 Chapter 6

TGA analysis of copolymer 2 was performed in order to determine suitable conditions for the reaction. The calculated mass difference between amic acid form of the copolymer and its imidized form was 6.5 %. The thermogravimetric analysis showed a weight loss of approximately 7 % in the temperatures range between 80 and 130°C this was attributed to the formation of the cyclic imide by elimination of water (see Figure 5).

100 ~7 %

60 Weight in% Weight 40

0 0 100 200 300 400 500 600 Temperature / deg C

Figure 5. TGA measurement of amic acid form C2.

The imidization of C2 was carried out in DMF at 120°C. The imidized copolymer C3 was obtained in form of a pale cream-colored powder which was soluble in acetone. For comparison the amidoacidified C2 was insoluble in acetone. The infrared spectroscopy (see Figure 3) showed characteristic absorption band of imides at 1567 cm-1.

The advantage of thermal imidization is its simplicity and the improved product quality as well as the absence of other reagents. It is of paramount importance for antibacterial test to avoid the presence of low molecular weight toxic agents, which can cause false positive results. The molecular weight of the imidized copolymer was measured by THF-GPC. The measurement showed an increase of the molecular weight of the copolymer after modification from Mn=5800 to Mn=8800. This increase in Mn is partially caused by the modification as well as by removal of the low molecular weight fraction during purification. This is indicated also by a change of the polydisperisity of the sample from Mw/Mn=1.67 to Mw/Mn=1.47.

Quaternization of C3

136 Chapter 6

The quaternization reaction was carried out in solution according to the method described in literature [18,48,51-53,59-61]. The C3 copolymer was dissolved in DMSO and reacted with methyl iodide to give copolymer C4. With a mixture of methyl iodide and dodecyl iodide (1:1, mol:mol) copolymer C5 was obtained and applying dodecyl iodide alone yielded copolymer C6. The reaction was performed at room temperature and in the presence of an excess of alkyl iodide (with respect to amine groups) yielding quantitative conversion of the tertiary amine. According to literature, a long alkyl chain within the ammonium group shows better biocidal properties [62-64]. For this reason three different quaternized polymers have been prepared bearing trimethyl ammonium, dimethyl-dodecyl ammonium groups as well as a 1:1 (mol:mol) mixture of both ammonium moieties as the quaternary ammonium side groups. The IR spectra of quaternized copolymers showed no significant changes because the absorption bands characteristic for ammonium salts overlap with absorption bands of non- quaternized copolymer (~1500 and 3000 cm-1)

Solubility in selected solvents

The different polarity of the copolymers C3, C4, C5, and C6 is also reflected in the different solubility (see Table 4). The introduction of cationic moieties into the polymeric chains strongly increases the polarity of the copolymer with huge influence on the solubility. While non-quaternized copolymer C3 was well soluble in all of the chosen solvents, the methyl iodide quaternized copolymer C4 was soluble only in very polar solvents such as water and DMSO but insoluble even in lower alcohols like methanol and ethanol. The introduction of long alkyl chains by quaternization with dodecyl iodide makes copolymer C6 insoluble in water, DMSO, DMF and alcohols but well soluble in ketones, THF and chloroform. The simultaneous quaternization with both methyl iodide and dodecyl iodide (C5) ensure good solubility in very polar solvents (water, DMSO) as well as less polar as 2-butanone. The solubility of the quaternized copolymers in water is of utmost importance because of antimicrobial investigations.

Table 4. Solubility of imidized C3 and quaternized C4, C5 and C6 in chosen solvents (10 mg of copolymer / mL). (+) good soluble (>10 wt%); (-) non-soluble (<1 wt%). Solvent C3 C4 C5 C6 Water + + + - Methanol + - + - 137 Chapter 6

Ethanol + - + - Acetone + - + + 2-Butanone + - + + Ethyl acetate + - - - THF + - + + DMSO + + + - DMF + + + - Chloroform + - - + THF = tetrahydro furane, DMSO = dimethyl sulfoxide, DMF = dimethylformamide

Thermal properties

Figure 6 depicts the TGA thermograms of copolymer C3 and its quaternized derivatives C4, C5 and C6. The non-quaternized copolymer C3 (curve 1) is the thermally most stable polymer and shows only slight weight loss below 200 °C. The highest weight loss is observed above 300 °C. Any modification of C3 by quaternization causes a decrease in thermal stability of the polymer due to Hofmann elimination of the ammonium groups [59]. The Hofmann elimination occurs when quaternary ammonium salts are exposed to high temperatures and the reaction is yielding an alkene and a tertiary amine and a low molecular weight compound specific for the counterion (e.g. water, HCl, HI etc.). The quaternization with methyl iodide causes the C4 (curve 2) to decompose above 150 °C via a three-stage thermal degradation. The first stage starts at 150 °C and ends at 200 °C with a weight loss of 32% corresponding to the loss of HI (31wt%). Even partial quaternization with long alkyl-chain iodide versus methyl iodide increased the thermal stability as observed in curves 3 and 4 (C5 and C6) in Figure 6 However, such derivatives also show faster thermal degradation than non-quaternized C3. In all cases the investigated copolymers showed a certain weight loss at relatively low temperatures around 100 °C, most probably caused by a loss of adsorbed water because of the hygroscopic nature of salts.

138 Chapter 6

100

Weightin% C6 40 C5 C4 C3

0 0 100 200 300 400 500 600 Temperature / deg C

Figure 6. TGA measurement of non-quaternized C3, and quaternized copolymers: C4 + + + + (-N(CH3)3 ), C5 ((-N(CH3)3 + -N(CH3)2 C12H25 ) and C6 (-N(CH3)2 C12H25 ).

DSC measurement of dodecyl iodide quaternized copolymer is typical for all ammonium copolymers based on the C3 copolymer in the temperature range of -50 and 200 °C. Each copolymer showed two thermal transitions: one at 17 °C and a second one in the region of 80-105°C (see Table 5). The temperature of the second transition seems to depend on type of quaternizing agent and is about 20 °C higher for copolymers quaternized with dodecyl iodide. There is a correlation with the TGA data where the presence of dodecyl iodide increased the thermal stability. Since the copolymers undergo the decomposition in this range of temperatures the reverse heating does not reproduce the curves. Only the first transition is fully reproducible.

The fact that the presence of longer alkyl chain gives higher Tg value is not in line with the expectations. It is known that polymers which contain longer, alkyl side chains exhibit lower glass transition temperatures than the shorter ones due to the plastifying effect. Although there has been no melting temperature observed this phenomenon can be assigned to the formation of ordered structures on micro or even nano scale. These types of crystalline structures usually do not give any measurable thermal response and further investigation is required. Detailed values of the thermal transition of the investigated copolymers are summarized in Table 5.

139 Chapter 6

Table 5. Transition temperatures for non-quaternized C3 and its cationic modifications. # Transition 1 [°C] Transition 2 [°C] C3 18 88 -N(CH3)2 + C4 17 85 -N(CH3)3 + + C5 17 104 -N(CH3)3 + -N(CH3)2 C12H25 + C6 18 101 -N(CH3)2 C12H25

Investigation of antimicrobial properties of the cationic copolymer.

Amphiphilic polymers with quaternary ammonium groups are known to have antimicrobial properties [20]. Thus the water soluble C3, C4 and C5 were tested for their antimicrobial efficacy. In order to find the Minimum Inhibitory Concentration (MIC) of the copolymers bacterial growth was monitored in the presence of all the copolymers. The test was performed in microwell plates and the proliferation potential of the bacteria was monitored at 37 °C by measuring the optical density at 612 nm for 20 h using a microwell plate incubator/reader in comparison to a reference without the respective polymer The MIC values are summarized in Table 6.

Table 6. Minimum inhibitory concentration (MIC) of the polymeric materials in 6 bidistilled water and MHB against bacteria (1-2x10 cfu/mL) (MIC100: complete growth inhibition during the monitoring time of 20 h) compared to references Nisin and ε- polylysine, and hemolytic effect (HC50). MIC100 in µg/mL HC50 in µg/mL E. coli P. aeruginosa S. aureus ATCC S. epidermidis Polymer ATCC 23716 ATCC27853 6538 ATCC 12228 C3 20 100 >1000 20 100* C4 200 >>1000 1000 20 >>1000** C5 10 200 100 20 60# NISIN 100 >200 3 3 >>340+ *Agglutination of RBC at all concentrations tested 10 – 1000 µg/mL; ** agglutination of RBC at 20 – 1000 µg/mL; # no agglutination of RBC, + at 340 µg/mL = 100 mM Nisin only 10.8 % of the RBC are lysed. Nisin (Handary 97.9%)

The polymers are more effective by a factor of 5 to 10 against the Gram negative E. coli than against the Gram positive S. aureus. Compared to polymer C4 polymers C3 and C5 are more active against the Gram negative bacteria E. coli and P. aeruginosa whereas, in the case of S. aureus C5 was the most active polymer compared to C3 and C4. The non-quaternized C3 being the least active in the latter case. Previous studies in our group, although with a different polymer backbone, have shown that best results against S. aureus were obtained with the cationic residue directly linked to the aliphatic residue without a spacer in between [65], and 140 Chapter 6 that longer alkyl chains showed best efficacy [66] and, thus, confirm these results. Regarding the solubility properties (Table 4) C3 and C5 show similar solubilities in the different solvents compared to C4. The limited solubility of the latter seems to restrict the effect against E. coli, P. aeruginosa and S. aureus since the hydrophilic-lipophilic balance is decisive for the efficacy of the respective polymer. Against S. epidermidis all 3 polymers are equally active in a comparable range like C3 and C5 against E. coli. Comparing the MIC for the Gram negative bacteria gave 5 (C3) to 20 fold (C5) higher values for P. aeruginosa than for E. coli, meaning that E. coli is 5 to 20 times better inhibited compared to P. aeruginosa. Gram negative bacteria are known to actively secrete outer membrane vesicle (OMV) from the outer membrane (OM) [67]. OMV production is correlated with an increased rate of survival upon antimicrobial peptide treatment [68] In P. putida OMV are generated, e.g., as a response to stress caused by cationic surfactants which can contribute to OMV biogenesis, through a physical mechanism, by induction of the curvature of the membrane [69]. Although OMV production is common in many bacteria the extent and mechanism of OMV production is species specific and thus, the higher MIC values for P. aeruginosa might be due to the level of OMV production since environmental stresses result in increased OMV formation by P. aeruginosa [70]. For the Gram-negative bacteria E. coli and P. aeruginosa and for the Gram-positive S. aureus C5 quaternized with the long alkyl chain, i.e., the repeat unit structure with the hydrophobic moiety being directly accompanied by the charged moiety, exhibits a higher efficacy compared to C4 quaternized with methyl iodide [71].

Whereas, polymers C3 and C4 led to an agglutination of human red blood cells (RBC) at all concentrations tested 10 – 1000 µg/mL, C5 did not agglutinate RBC, however showed lysis of 50 % of the RBC relative to the positive control (HC50) at a concentration of 60 µg/mL.

The higher value of HC50 compared to the MIC100 against E. coli (10 μg/mL) proved that polymer C5 has a selectivity to differentiate between mammalian cells and bacterial cell walls. However, since the values are overall in the same order of magnitude, the selectivity is low.

Since the investigated antimicrobial copolymers were designed to mimic peptides a comparison with a reference compound is needed. The type A lantibiotics e.g. Pep5 or nisin are in general of linear conformation and all the nisin type peptides are positively charged [33]. Combination of the cationic nature and the presence of leucine make nisin a good reference. Nisin is a 34-residue-long peptide which is predominantly active against Gram-positive bacteria. It is generally accepted that the bacterial plasma membrane is the target for nisin, and

141 Chapter 6 that nisin kills the cells by pore formation and inhibition of peptidoglycan synthesis. The pore formation causes collapse of vital ion gradients resulting in cell death [72]. In this study it was shown that nisin compared to the copolymer modified with both iodides is highly active against the Gram positive bacteria, but 30 to more than 60 times lower active against the Gram negative bacteria. Moreover, comparing the MIC of nisin against the Gram-positive bacteria on the molar basis instead on the weight basis, the difference between the effect of nisin and the copolymer is significantly lower.

This gap indicates that the activity of the peptides is determined not only by the amphiphilic nature but most probably the secondary peptide structure plays also a substantial role.

4. Conclusions

Maleic anhydride copolymers are versatile, easy for modification materials which can be used as a base for a wide range of antimicrobial copolymers. The modification of P[MP-alt-MSA] copolymer with diamine to poly[(4-methyl-1- pentene)-alt-(1-(3-N,N-dimethylaminopropyl)maleimide)] can be performed as a one pot synthesis without using any additives in relatively mild conditions. Poly[(4-methyl-1-pentene)- alt-(1-(3-N,N-dimethylaminopropyl)maleimide)] can be easily converted into polycationic material by means of any alkyl iodide. Sequential quaternization with methyl iodide and dodecyl iodide which introduce hydrophobic long alkyl chain moiety but thank to methyl iodide ensure solubility in polar solvents shows the best properties in sense of antimicrobial activity. The antimicrobial properties of poly[(4-methyl-1-pentene)-alt-(1-(3-N,N,N- trimethylammoniumpropyl)-maleimidoiodide)] are even lower than the properties of the non- quaternized copolymer. The polymers are more effective by a factor of 5 to 10 against the Gram negative E: coli than against the Gram positive S. aureus. Compared to poly[(4-methyl-1-pentene)-alt-(1-(3- N,N,N-trimethylammoniumpropyl)-maleimidoiodide)] polymers poly[(4-methyl-1-pentene)- alt-(1-(3-N,N-dimethylaminopropyl)maleimide)] and poly[(1-(3-N,N,N-trimethylammoniumpropyl)-maleimidoiodide)-co-(1-(3-N,N-dimethyl-N-dodecylammoniumpropyl) maleimidoiodide)-alt-(4-methyl-1-penten)] are more active against the Gram negative bacteria E. coli and P aeruginosa whereas, in the case of S. aureus only poly[(1-(3-N,N,N- trimethylammonium-propyl)-maleimidoiodide)-co-(1-(3-N,N-dimethyl-N- dodecylammoniumpropyl) maleimidoiodide)-alt-(4-methyl-1-penten)] was the most active

142 Chapter 6 polymer compared to the non-quaternized polymer and the polymer quaternized with methyl iodide. The non-quaternized copolymer being the least active in the latter case. The MIC of the synthetic copolymers is higher than of the natural peptide nisin. However, the freedom in designing the basic polymer, molecular weight as well as the way of modification and, in particular, the much lower price of the synthetic compounds shows the potential of the presented strategy to develop new “surface protection”.

References [1] R.J. Cornell, L.G. Donaruma, J. Med. Chem. 1965, 8, 388–390 [2] T. Tashiro, Macromol. Mater. Eng., 2001, 286, 63 [3] E.-R. Kenawy, Y. Mahmoud, Macromol. Biosc., 2003, 3, 107 [4] F. Sanda, T. Endo, Macromol. Chem. Phys., 1999, 200, 2651 [5] M. Ilker, K. Nüsslein, G. Tew, E. Coughlin, J. Am. Chem. Soc., 2005, 126, 1587 [6] B.L. Rivas, E.D. Pereira, M.A. Mondaca, Polymer Bulletin, 2003, 50, 327 [7] T. Tashiro, J. Appl. Polym. Sci., 1992, 46, 899 [8] N. Nurdin., G. Helary, G. Sauvet, J. Appl. Polym. Sci. 1993, 50, 651 [9] N. Nurdin., G. Helary, G. Sauvet, J. Appl. Polym. Sci. 1993, 50, 663 [10] N. Nurdin., G. Helary, G. Sauvet, J. Appl. Polym. Sci. 1993, 50, 671 [11] N. Nurdin., G. Helary, G. Sauvet, J. Appl. Polym. Sci. 1995, 58, 77 [12] G. Sauvet, W. Fortuniak, K. Kazimierski, J. Chojnowski, J. Appl. Polym. Sci. 2003, 41, 2939 [13] T. Ikeda, H. Hirayama, K. Suzuki, H. Yamaguchi, S. Tazuke, Macromol. Chem., 1986, 187, 333 [14] G. Li, J. Shen, Y. Zhu, J. Appl. Polym. Sci., 1998, 67, 1761 [15] G. Sauvet, S. Dupond, K. Kazimierski, J. Chojnowski, J. Appl. Polym. Sci., 2000, 75, 1005 [16] T. Tashiro, J. Appl. Polym. Sci., 1991, 43, 1369 [17] S. Augusta, H.F. Gruber, F. Streichsbier, J. Appl. Polym. Sci., 1994, 53, 1149 [18] J. E. Fields, J. H. Johnson, U.S. Patent 3,340,680 (1967) [19] N. Destais, D. Ades, G. Sauvet, Polymer Bulletin, 2000, 44, 401 [20] E.-R. Kenawy, F.I. Abdel-Hay, Abd El-Rahem R. El-Shanshoury, M. H. El- Newehy,Journal of Controlled Release, 1998, 50, 145 [21] E.-R. Kenawy, F.I. Abdel-Hay, Abd El-Rahem R. El-Shanshoury, M. H. El-Newehy, J. Polym. Sci. Part A: Polym. Chem., 2002, 40, 2384 [22] A. Kanazawa, T. Ikeda, T. Endo, J. Appl. Polym. Sci., 1994, 53, 1245 [23] A. Kanazawa, T. Ikeda, T. Endo, J. Polym. Sci. Part A: Polym. Chem, 1993, 31, 335 [24] A. Kanazawa, T. Ikeda, T. Endo, J. Polym. Sci. Part A: Polym. Chem, 1993, 31, 1441 [25] A. Kanazawa, T. Ikeda, T. Endo, J. Polym. Sci. Part A: Polym. Chem, 1994,32, 1997 [26] A. Kanazawa, T. Ikeda, T. Endo, J. Appl. Polym. Sci., 1994, 54, 1305 [27] A. Kanazawa, T. Ikeda, T. Endo, J. Appl. Polym. Sci., 1994, 53, 1237 [28] L. Timofeeva, N. Kleshcheva, Appl. Microbiol. Biotechnol., 2011, 89, 475 [29] Monsanto Company, GB Patent 1,260,451 (1972) [30] H. Patel, D. A. RAval, D. Madamwar, S. R. Patel, Angew. Macrom. Chem., 1998, 263,25 [31] J.-H. Jeong, Y.-S. Byoun, S.-B Ko, Y.-S. Lee, J. Ind. Eng. Chem. 2001, 7, 310 [32] J.-H. Jeong, Y.-S. Byoun, S.-B Ko, Y.-S. Lee, Reactive & Funcctional. Polym., 2002, 50, 257 [33] G. Bierbaum, Chemotherapie Journal, 1999, 6, 204 [34] M. Zasloff, Nature, 2002, 415, 389

143 Chapter 6

[35] M. Zasloff, Proc. Natl. Acad. Sci. USA, 1987, 84, 5449 [36] W.F. Broekaert, F.R.G. Terras, B.P.A. Cammue, R.W. Osborn, Plant Physiol., 1995, 108, 1353 [37] I. CornuP, K. Biittnerb, J.-L. Dasseuxb, J. Dufourcq, FEBS Letters, 1994, 349, 29 [38] L. Beven, S. Castano, J. Dufourcq, A. Wieslander, H. Wroblewski, Eur. J. Biochem., 2003, 270, 2207 [40] M. Arnold, M. Rätzsch, Macromol. Chem., 2003, 187, 1593 [41] J. F. Nash, T.-M. Lin, U.S. Patent 3,224,941 (1965) [42] H. Komber, Macromol. Chem. Phys., 1996, 197, 343 [43] E. Bortel. M. Styslo, Macromol. Chem., 1990, 191, 2653 [44] E. Bortel. M. Styslo, Macromol. Chem., 1990, 189, 1155 [45] M.L. Tamayo, Rev. R. Acad. Ciene. Exact. Fis. Nat. Madrid., 1945, 39, 185 [Chem. Abstr., 1948, 42, 8175a] [46] S. Iwabuchi, Y. Watanabe, T. Nakahira, K. Kojima, J. Polym. Sci. Polym. Chem. Ed., 1979, 17, 1721 [47] M. Szkudlarek , U. Beginn, H. Keurl M. Möller, Polymers, 2017, 11, 610 [48] E.-R. Kenawy, M. A. Sakran, J. Appl. Polym. Sci., 2001, 80, 415 [49] W.-F. Lee, Y.-M. Chen, J. Appl. Polym. Sci., 2000, 80, 1619 [50] Y.-H. Seo, C. Lee, Bull. Korean Chem. Soc., 2000, 21, 241 [51] S.S. Lee, T. O. Ahn, J. Appl. Polym. Sci., 1998, 71, 1187 [52] W.-F. Lee, Y.-M. Chen, J. Appl. Polym. Sci., 1996, 59, 599 [53] W.-F. Lee, Y.-M. Chen, J. Appl. Polym. Sci., 1996, 60, 187 [54] W.-F. Lee, Y.-M. Chen, J. Appl. Polym. Sci., 1997, 66, 95 [55] L. M. Rice, C. H. Grogan, E. E. Reid, J. Am. Chem. Soc., 1953, 75, 2261 [56] Y. Z. Meng, S. C. Tjong, A. S. Hay, S. J. Wang, J. Polym. Sci. Part A: Polym. Chem., 2001, 39, 3218 [57] P.-H. Sung, C.-Y. Chen, S.-Y. Wu, J. Y. Huang, J. Polym. Sci. Part A: Polym. Chem., 1996, 34, 2189 [58] Y. L. Liu, R. J. Yeng, Y.-S. Chiu, J. Polym. Sci. Part A: Polym. Chem., 2001, 39, 1716 [59] C.-S. Wang, H.-J. Hwang, J. Polym. Sci. Part A: Polym. Chem., 1996, 34, 1493 [60] A. J. I. Alfaia, A. R. T. Calado, J. C. R. Reis, Eur. J. Org. Chem., 2000, 3627 [61] T. W. Johnson, I. M. Klotz, Macromolecules, 1974, 7, 149 [62] V. Bütün, S. P. Armes, N. C. Billingham, Macromolecules, 2001, 34, 1148 [63] T. Tashiro, Macromol. Mater. Eng., 2001, 286, 63 [64] O. Vogl, D. Tirrell, J. Macromol. Sci. Chem., 1979, A13, 415 [65] S.D. Worley, G. Sun, TRIP, 1996, 4, 364 [66] He, Y.; Heine, E.; Keusgen, N.; Keul, H.; Moeller, M., Biomacromolecules, 2012, 13, 612 [67] E. Kiss, E.T. Heine, K. Hill, Y.C. He, N. Keusgen, C.B. Penzes, D. Schnoller, G. Gyulai, A. Mendrek, H. Keul, M. Moeller, Macromol. Biosci., 2012, 12, 1181 [68] S. Roier, F. G. Zingl, F. Cakar, S. Schild, Microbial Cell, 2016, 3, 257 [69] Manning, A. J., and Kuehn, M. J., BMC Microbiol., 2011, 11, 258 [70] R. M. Heredia, P. S. Boeris, A. S. Liffourrena, M. F. Bergero, G. A. López, G. I. Lucchesi, Microbiology, 2016, 162, 813 [71] I. A. MacDonald, M. J. Kuehn, J. Bacteriol., 2013, 195, 2971 [72] R. Gartzen, E. Heine, Polymers, 2018, 10(1), 96 [73] E. Breukink, B. de Kruij, Biochimica et Biophysica Acta, 1999, 1462, 223 [74] I.E. Pol. E.J. Smid, Lett. Appl. Microbiol., 1999, 29, 166

144 Chapter 6

145

146 List of Publications

List of Publications

Parts of this thesis are published or submitted for publication.

Szkudlarek, M., Beginn, U., Keul, H., Möller, M. Synthesis of Terpolymers with Homogeneous Composition by Free Radical Copolymerization of Maleic Anhydride, Perfluorooctyl and Butyl or Dodecyl Methacrylates: Application of the Continuous Flow Monomer Addition Technique. Polymers, 9(11), 2017, 610.

Szkudlarek, M., Beginn, U., Keul, H., Möller, M. Solubility, Emulsification and Surface Properties of Maleic Anhydride, Perfluorooctyl and Alkyl Meth-Acrylate Terpolymers. Polymers, 10(1), 2018, 37.

Szkudlarek, M., Beginn, U., Heine E., Keul, H., Möller, M. Synthesis, characterization and antimicrobial properties of peptides mimicking copolymers of maleic anhydride and 4- methyl-1-pentene. Polymers, Manuscript submitted.

147 Acknowledgements

Acknowledgements

The work described in this thesis was carried out at the Institute of Technical and Macromolecular Chemistry and DWI at RWTH Aachen at Department of Organic Chemistry III/Macromolecular Chemistry, University of Ulm, as well as, under the guidance of Prof. Dr. Martin Moeller. Many people gave their time and expertise to help me during my Ph.D. work. Therefore, I would like to express my deepest appreciation to all of them.

I am grateful to Prof. Dr. Martin Moeller for the interesting topic and the possibility of working in an enthusiastic and creative group. I appreciate all the discussions we had as well as his ideas and work guidelines, which frame the chapters of this book. I am also grateful for his continuous support and strong encouragement.

I thank my direct supervisor Professor Dr. Uwe Beginn for his comprehensive discussions, scientific advices and his enormous patience in correcting the PhD thesis.

I gratefully acknowledge the support of Dr. Helmut Keul in critical revision of this thesis and enormous support in preparing the publications.

I would like to thank Dr. Elisabeth Heine for her crucial contribution to the work presented in Chapter 6 of this thesis as well as for correcting this part of the manuscript. I am very grateful to: Sylwia Szkudlarek and Dr. Janis Lejnieks for performing TGA and DSC measurements, Dr. Walter Tilmann for IR, Dr. Xiaomin Zhu for help in performing contact angle measurements, Dr. Ahmed Mourran for AFM and Rita Gartzen for conducting the antimicrobial tests. I wish to thank Dr. Krystyna Albrecht, Dr. Blazej Gorzolnik for their help and support in solving the “greenhorn issues”. I would like to cordially thank Dr. Dragos Popescu for his support, inexhaustible optimism and friendship. I am grateful to all my colleagues and team members for the open and enthusiastic atmosphere. Many thanks in particular to: Yvonne, Katarzyna, Irene, Ling-Long, Heidrun, Christoph, Reza, and Sherif. All co-workers at DWI, ITMC and OC III department for the inspiring collaboration and the non-scientific activities we did together, have my appreciation.

148 Acknowledgements

I would like to dedicate many thanks to my parents for their support, encouragement and interest in my work Finally, I would like to thank my wife Sylwia for her support, patience and love during the entire Ph.D. period. I am very grateful to my daughter Marta and son Robert for motivation and patience during writing of this thesis. No words can express my gratefulness. However, many have not been mentioned, none is forgotten

149

150

151

152