Enzyme- and Metal-Catalyzed Synthesis of a New Biobased Polyester

Jul 26, 2017 - Rheological measurements of the polyesters in melt revealed a Newtonian-like behavior at 80 °C. The dynamic viscosities fulfill the Co...
5 downloads 10 Views 2MB Size
Subscriber access provided by UNIV OF NEWCASTLE

Full Paper

Enzyme and metal catalyzed synthesis of a new biobased polyester Jakob Gebhard, Björn Neuer, Gerrit Albert Luinstra, and Andreas Liese Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.6b00418 • Publication Date (Web): 26 Jul 2017 Downloaded from http://pubs.acs.org on July 27, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Organic Process Research & Development is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Enzyme and metal catalyzed synthesis of a new biobased polyester Jakob Gebhard†,║, Björn Neuer‡,║, Gerrit A. Luinstra‡,*, Andreas Liese†,* †

Hamburg University of Technology, Institute for Technical Biocatalysis, Denickekestraße 15,

21073 Hamburg, Germany. ‡

University of Hamburg, Institute for Technical and Macromolecular Chemistry, Bundesstraße

45, 20146 Hamburg, Germany.

ACS Paragon Plus Environment

1

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Table of Contents. Enzyme and metal catalyzed synthesis of a new biobased polyester starting from PripolTM 1012 and 1,3-propanediol.

ACS Paragon Plus Environment

2

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

KEYWORDS: Polyester, polycondensation, metal-catalysis, enzyme-catalysis, 1,3-propanediol, polyurethane ABSTRACT: Linear aliphatic polyesters were prepared from PripolTM 1012, a diacidic C18 fatty acid dimer, and 1,3-propanediol employing a lipase or a titanium tetrabutanoate. Metal-based catalysis (route M) was carried out with a precondensation at 180 °C and 600 mbar followed by a final condensation at 220 °C and 0.3 to 0.6 mbar. Enzyme catalysis was carried out with an immobilized Candida antarctica lipase B after either a precondensation step at 180 °C and 600 mbar (route E1) or 80 °C and 100 mbar (route E2) and a final condensation at 80 °C and 0.3 to 0.6 mbar. Polyesters were obtained along route M, E1 and E2 with weight average molecular weights Mw at final conversion of 84.6, 26.7 and 15.6 kg mol-1, respectively. The final molecular weight via route E2 was most probably constrained by depletion of 1,3-propanediol during precondensation. Rheological measurements of the polyesters in melt revealed a Newtonian-like behavior at 80°C. The dynamic viscosities fulfill the Cox-Merz rule. The power law for the viscosity as a function of Mw possesses an exponent of 3.7 ±0.2. A polyesterdiol of Mn ≈ 6 kDa prepared along route M was used in the synthesis of a polyurethane elastomer with a Young modulus of 2.5 MPa, an elongation at break of 554%, an ultimate tensile strength of 3.5 MPa and a Shore A hardness of 38.

ACS Paragon Plus Environment

3

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

INTRODUCTION Synthetic polyesters are standard materials in today’s society. They are applied as packing materials, e.g. as fibers for apparel, medical surge threads and as precursors in the production of polyurethanes.1,2 Polyesters are routinely prepared in a polycondensation reaction with formation of water as side product. The removal of water is necessary to drive the equilibrium towards the products, as the equilibrium constants in case of aliphatic polyester are in the range of about 110.3 The reaction proceeds under the action of a catalyst in form of protonic and/or Lewis acids, like titanium alkoxides, antimony oxides and acetates of manganese, calcium, magnesium or zinc and at elevated temperatures ranging from 180 to 280 °C.4 A high reaction temperature can put some constraints on the choice of starting materials as it can lead to undesired side reactions like alcohol dehydration5, double bond isomerization or Ordelt saturation of a double bond.6 Especially the synthesis of polyesters with thermally sensitive or chemically reactive sites comprising siloxane, epoxy entities or double bonds can end up challenging. Enzyme catalysis at lower temperatures may be an alternative in such cases.7–10 The green synthesis of polyesters should encompass production processes and starting materials.11,12 One possibility for their utilization is the preparation of polyurethanes from short polyester chains containing fatty or dimerized fatty acids. In recent years, the synthesis of such polyurethanes, as well as the assessment of their possible implementations, have been a topic of ongoing research.13–18 The establishment of green alternatives to existing technologies, in particular for new products, seems generally attractive19, provided that costs and the volume of auxiliaries are in an economically viable range. Environmentally friendly polyester production

ACS Paragon Plus Environment

4

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

should be based on renewables and use non-toxic catalysts as opposed to toxic metals as catalysts.4 Reaction temperatures should be low to conserve energy. Highly selective catalysts would comprise the option of tolerating the presence of reactive entities in monomers. Until now polyesters are mainly synthesized from monomers produced from fossil resources, which are being consumed faster than generated, leading to a potential lack of these resources in the future. Green precursors for linear polyesters are available; two prominent examples are used in this study. The dimerized fatty acid PripolTM 1012 (Pripol) produced from unsaturated C18 vegetable oil fatty acids, was chosen as dicarboxylic acid. Pripol is a mixture of diacids containing 36 carbon atoms in the backbone. It is produced by a dimerization of unsaturated C18 fatty acids of vegetable origin and is claimed to be partly unsaturated (0-1 double bonds per molecule). A structural proposal is given in Scheme 1.20 The diol of choice, 1,3-propanediol (PDO), can be obtained by fermentation of glycerin.21–23 Enzymes can be used as catalysts to enhance the polycondensation reaction and generally show a high activity at mild reaction conditions (temperature, pH, pressure etc.) next to a high chemo-, enantio- and regioselectivity.24–28 Isolated enzymes are used in this regard for organic synthesis and have replaced traditional chemical catalysts e.g. in the production of high-fructose corn syrup (HFCS)29, of L-tert-leucine30 or of surfactants.31,32

Scheme 1. Structure of PripolTM 1012 (Pripol)20.

ACS Paragon Plus Environment

5

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

The field of enzymatic polycondensation has developed in the last decades. Peroxidases, cellulases, cutinases and lipases are the most prominent enzymes used as catalysts for polycondensation..28,33–36 Lipase B from Candida antarctica shows the broadest substrate spectrum and highest activity for polycondensation reactions.37,38 So far, no industrial processes have been established for the production of polymeric polyester with lipase catalysts. A restriction of enzyme applications in polyester synthesis may be, their deactivation at elevated temperatures. These are required for reducing viscosity in polycondensation processes. Enzymes immobilized on an acrylic resin, e.g. available as CalB immo by c-LEcta or Novozyme 435 by Novozymes, have a substantially improved thermal stability. Enzymes in this form remain active longer and can be reused several times.39 Here we compare the polycondensation of PDO and Pripol using the immobilized lipase CalB immo. The objective was to determine, which maximum molecular weight could be achieved using immobilized CalB for the chosen polycondensation reaction at 80°C. A further objective of this study was to investigate the constitution of the polyester products, in particular whether side reaction occur in dependence of the catalytic system. Therefore the polycondensation of the starting materials was also carried out using a standard titanium based catalyst, titanium(IV) tetra-n-butanolate (TiOnBu4 ) at 220 °C.40 Titanium alkoxide catalysts are routinely used for the production of various polyesters like poly(butylene terephthalate) or poly(trimethylene terephthalate).41 The obtained polyester was further used in the synthesis of an exemplary polyurethane elastomer.

RESULTS AND DISCUSSION The formation of ester bonds from starting materials containing alcohol and acid functionalities is thermodynamically not very favorable, and it is usually necessary to

ACS Paragon Plus Environment

6

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

dynamically remove the side product water.3 The preparation of high molecular weight, linear aliphatic polyester of the AA/BB type with differing volatility of AA and BB can be achieved by starting from a mixture of monomers containing a molar excess of the most volatile component, here of the PDO over Pripol (Scheme 2).42 The exact structure of Pripol is not given by the manufacturer. Analysis by nuclear magnetic resonance (NMR) indicates a mixture of various compounds (see Supplementary Information). However, the composition is not easily elucidated from these spectra. Short oligomers are formed in the beginning of the polycondensation. The major amount of water that may form is separated from the reaction mixture, together with (some of) the surplus of PDO. The reaction may predominantly be autocatalyzed by the diacid in this phase. As the concentration of diacid is decreasing, the rate becomes more and more dependent on the action of a catalyst. The molecular weight is consecutively increased mostly by transesterification reactions and removal of PDO (and some water), that is formed in this second phase.43–45

ACS Paragon Plus Environment

7

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Scheme 2. Phases of polycondensation towards polyester from 1,3-propanediol and a diacid with precondensation and transesterification phase with a titanium alkoxylate. The predominant reactions are displayed.

The polycondensation was carried out in bulk, comprising Pripol and PDO in a molar ratio of 1:1.4, and optionally a catalyst. Two protocols were used to perform the polycondensation: i) a two stage process consisting of a precondensation at 180 °C and 600 mbar without catalyst for a time of 125 min, followed by addition of catalyst: A: no catalyst, M: titanium(IV) tetra-nbutanolate, E1: CalB immo, raising the temperature to 220 °C and reducing the pressure to below 1 mbar, and, ii) a two stage process with an enzyme catalyzed precondensation (E2) at 80 °C and 100 mbar (150 min), followed by a period of condensation at 1 mbar at 80 °C (Table 1). The first phase of both protocols was terminated at acid conversions in the range of 75-80%. All reactions were carried out, either in air (index a), or in an argon atmosphere, to investigate the role of oxygen. No obvious difference was recognized, thus only the results of the experiments under argon are depicted. The data for the reactions in air are given in the Supplementary Information.

ACS Paragon Plus Environment

8

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Table 1. Protocols of metal- (M), enzyme- (E) and non-catalyzed (autocatalyzed) (A) polycondensations. Reaction Atmosphere M

Argon

A

Argon

E1

Argon

E2

Argon

First stage conditions

Second stage conditions 220 °C, ≤1 mbar

180 °C, 600mbar 125 min

No catalyst

365 min – 24h

TiOnBu4 No catalyst

80 °C, ≤ 1 mbar 80 °C, 100 mbar

CalB immo

48-55 h

CalB immo

150 min

The progress of ester formation was monitored by sampling the reaction mixture and determining the acid concentration in the aliquots by titration with potassium hydroxide (Figure 1). The conversion follows the expected course for a polycondensation with dynamic removal of water and diol.42 The progression of acid conversion with time in the first stage (about 80 %) of protocol M, A and E1 is similar, showing that the autocatalytic ester formation is quite reproducible in the equipment. A comparable conversion of about 75% was chosen at the end of the first phase of the second protocol (E2), i.e. precondensation at a lower temperature and lower pressure in the presence of CalB immo. The rate of conversion in the latter experiment at 80 °C is just somewhat lower than in the autocatalyzed route at 180 °C, indicating the high catalytic activity of the enzyme at temperatures below 100 °C. The lower conversion is most likely caused by the limitations of physically removing diol and water from the reaction vessel. An increase of temperature in enzyme-catalysis is not possible, due to inactivation of the enzyme.46

ACS Paragon Plus Environment

9

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

Figure 1. Conversion of acid entities vs. time of metal- (M), enzyme- (E) and non-catalyzed (autocatalyzed) (A) polycondensation.

The elugrams of SEC measurements of samples after the precondensation show four maxima between 23 and 28 minutes (selected samples in Figure 2). They correspond to oligomeric ester with a number average molecular weight of 4200 g mol-1 at maximum (relative to PS standards). Such products are potentially useful precursors for polyurethanes, provided that the end groups carry a hydroxyl functionality.

ACS Paragon Plus Environment

10

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Figure 2. Elugrams from SEC-measurements for selected samples of the metal- (M) and enzyme- (E) polycondensation reactions.

Table 2. Conversion, molecular weight, PDI (including low molecular weight fractions) and zero shear viscosity of end products. Index a indicates polycondensations in air. PDI

|η*|

Time

T

X

Mw

[h]

[°C]

[%]

[kg mol-1]

Ma

6

220

99

84.6

4.4

4370

M

6

220

97

49.3

3.0

320

A

24

220

95

28.4

2.6

45

E1a

55

80

98

25.5

2.3

56

E1

52

80

98

26.7

2.3

38

E2a

48

80

93

15.1

2.0

12

E2

50

80

93

15.6

2.0

7

[Pa·s]

ACS Paragon Plus Environment

11

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

The autocatalytic polycondensation of the starting materials along protocol i) finally leads to a product with Mw = 28.4 kg mol-1 and a conversion of 95% in the equipment used after 24h (Table 2). The addition of the titanium catalyst after the first phase and increasing the temperature to 220 °C (protocol M), leads to a rapid formation of high molecular weight polyester (up to Mw = 84.6 kg mol-1). The fraction of the initially formed oligomers is transformed into polymer in the period between 125 to 185 min of the reaction (Figure 2). The conversion of acid after 365 min is 97-99%. The variation in conversion between experiments may be due to small variations in pressure. The removal of water and PDO by vacuum shifts the equilibrium towards the products, and hence is directly linked to conversion and molecular weight. The pump created vacuum in the reactor between 0.3 and 0.6 mbar at the end of the polycondensation. A comparatively high PDI (Table 2) was found for the metal-catalyzed reactions. It should be noted, that the PDI was calculated over the whole range of molecular weight, excluding the monomers as these numbers have the highest significance for application. Most likely, the higher viscosity leads to the broadening of the molecular weight distribution as the reactor setup was not optimized for ideal mixing at those conditions. The difference in PDI between catalysis by enzyme and metal catalysts in the studied polymerization cannot a priori be assigned to a difference in catalytic action as the obtained molecular weights and hence the viscosities are different. Small signals for cyclic structures are tentatively observed in MALDI-TOF spectra (see Supplementary Information) in reactions catalyzed by metal and by enzyme; but the identification could not unequivocally be reached on account of the complexity of the spectra and the fact that Pripol is a mixture of compounds. The determination of the ring-chain

ACS Paragon Plus Environment

12

Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

equilibrium and of the distribution of cycles possibly present exceeds the scope of this paper as it is a field of extensive studies.47–50 The dominant formation of Pripol containing rings is not expected, as large membered rings (>20 atoms) are thermodynamically and kinetically not favorably formed. The maximal molecular weight of the polyester obtained by enzyme catalysis at 80 °C in 50 h (E1, Mw = 26.7 kDa) is in the same order of magnitude as of the non-catalyzed reaction at 220 °C after 25 h (A, Mw = 28.4 kDa Figure 3). This shows both the restriction in rate and the possibilities of enzymatic catalysis at lower temperature. The rate of condensation becomes more and more limited by the vapor pressure of the side products (water and PDO) in the polymeric matrix with increasing conversion. The higher performance of the metal-catalyzed reaction is probably related to the reaction temperature.51 The higher temperature and the concomitant lower viscosity in the titanium-based process are favorable for removing the condensation products. The concentration of active sites of 0.005 mol% in route M and 0.003 mol% in route E2 are comparable (with an estimated enzyme loading 5.6 wt% CalB immo).

ACS Paragon Plus Environment

13

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

Figure 3. Weight average molecular weight as function of conversion. A higher final acid conversion of 98% was found in the protocol E1, compared to E2 with 93% (Figure 1; Table 2). The molecular weight reached in E2 is about half of that reached in E1 (Figure 3; Table 2). This seems majorly related to the stoichiometry of diacid to diol that exists after the precondensation phase (Figure 1). The higher temperature during precondensation in E1 was more effective for ester formation, than the lower pressure and lower temperature in E2. The molecular weight or viscosity does not seem to fundamentally limit the conversion in E2, as higher molecular weights at higher viscosities were reached with CalB immo under the same conditions in phase 2 of E1 (up to 26.7 kg mol-1). The reaction conditions in the precondensation phase of E2 lead to a less than stoichiometric ratio of diol to diacid. Consequently, the molecular weight and the acid conversion are limited. The molecular weight in the second phase of protocol E1 is momentarily lower than that in E2 at similar acid conversion. This is due to the higher content of PDO in form of end groups. The final achievable molecular weight is higher in E1 on account of the better stoichiometry of Pripol to PDO.

ACS Paragon Plus Environment

14

Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

The products of the polycondensations were isolated as yellow, highly viscous oils. Molecular, thermal and rheological characterization of the crude products show no obvious differences in dependency of the various protocols used. No evidence for side reactions, e.g. cross-linking can be found by 1H NMR spectroscopy or by rheology. The proton spectra are comparable (Figure 4). Signals in the area around 4.16 ppm (4.3 - 3.6 ppm) belong to the terminal methylene groups of the PDO moiety, at 1.97 ppm to its intern methylene group and signals at 2.30, 1.62, 1.27, 0.89 ppm to the methylene and methyl groups of Pripol. The methylene groups adjacent to an oxygen atom create the signals in the area of 4.5 to 3.6 ppm in the 1H NMR spectra (polymer chain, hydroxyl terminus of polymer chain, PDO), which show up with different intensities in the different products. The relative amounts of the diol to diacid thus document the different conversions.

E2 after 3120 min E1 after 3000 min A after 1440 min M after 365 min

Figure 4. 1H NMR (CDCl3) of products, synthesized by different protocols The thermal behavior, as determined in DSC and TGA measurements, was similar for all of the products, and independent of the route of preparation (Figure 5). Glass transition

ACS Paragon Plus Environment

15

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

temperatures of -55 °C were observed for all products, within the uncertainty of standard DSC methods. No noteworthy melting occurred during heating of the second cycle in DSC measurement, as expected for polyesters from dimerized fatty acids.52,53 Two merging decomposition processes were observed by TGA, with inflection points at about 427 and 464 °C for all products, the second step being less intense. The values are similar to comparable polyesters and have previously been ascribed to ester cleavage, dehydrogenation and alkyl group decomposition. 14,53,54

Figure 5. Analysis by a) DSC and b) TGA of the final products obtained by the different protocols. The dynamic viscosity of the products determined by rheology show the polyester products to be rheologically simple fluids (Figure 6).55 The shear rate in shear experiments is limited by the usual sample discharge of low viscous samples. Data obtained in oscillatory mode extend the accessible frequency domain. The flow curves in shear and oscillation experiments at 80 °C in the linear viscoelastic regime show, that the samples follow the Cox-Merz rule, indicating that non-branched polymers have formed. The zero-shear viscosity increases with the molecular

ACS Paragon Plus Environment

16

Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

weight along the usual power law with an exponent of 3.7 ±0.0 (3.7±0.2 including polycondensations under argon as well as in air, see Supplementary Information). This value is typical of most non-branched polymer systems above the critical entanglement molecular weight (including linear polyester chains as well as traces of cycles that may be present).56 No systematic difference from this power law was noticed for either the enzymatic or the chemically catalyzed obtained products, what could have been an indication of different polymer structures. Samples show terminal flowing in oscillatory measurements. The slope of the storage moduli G’ curves are approximately two, and for those of the loss modulus G’’ curves one. The absolute values of G’ and G’’ depend on the molecular weight (Figure 7).

Figure 6. Viscosity by steady state shear (η) and oscillatory (|η∗ |) experiments at 80 °C (l) and the dependency of zero shear viscosity on the molecular weight (r). a) Steady state viscosity was not possible to determine because of early sample discharge.

ACS Paragon Plus Environment

17

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

Figure 7. Elastic and viscous moduli vs. angular velocity of oscillatory measurements at 80 °C.

A polyurethane elastomer (Index 100) based on 4,4'-methylene diphenyl diisocyanate (MDI) was prepared to evaluate possible further application of the polyester in polyurethanes. It was prepared from a polyesterdiol prepolymer of Mn = 5 980 g mol-1 as soft segment component and 1,4-butanediol as chain extender. The polyesterdiol was synthesized along protocol M with two intermediate additions of PDO during the transesterification to lower the molecular weight and to ensure that the end groups are hydroxyl functionalities. The obtained opaque polyurethane had a light yellow color similar to previous reports on similar products.14 A glass transition temperature of -53 °C was observed resulting from the Pripol based soft phase.57 These are close to the literature values of polyurethanes from natural or dimerized rapeseed oil with MDI and 1,4-butanediol (-47 to -51 °C).15,58 Three degradation processes were determined by differentiating the TGA curve with temperatures at the inflection point of 311, 430 and 465 °C, which fit well with values from literature for comparable polyurethanes.14,53,54 The first inflection point is reported to be the breaking of the urethane bonds and the latter two correspond well with the degradation of the polyester. A Young modulus of 2.5 MPa, an elongation at break of 554%,

ACS Paragon Plus Environment

18

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

an ultimate tensile strength of 3.5 MPa and a shore a hardness of 38 was determined. Young modulus and elongation at break are in the range reported for similar polyurethanes with polyesters basing on fatty acids.15,59 A maximal elongation was observed in the range of 100 to 600 % for polyurethanes with various hard segment contents, with widely used polyols or polyesterols and with diacids from renewable resources.15,59 The found tensile strength is at the lower edge of the range of 1.3 to up to 32 MPa reported for various polyurethanes from different dimeric fatty acids with MDI and with different chain extender as 1,2-ethandiol or 1,4butanediol. These values demonstrate the comparability of the derived polyurethane with previously reported polyurethanes from bio-derived precursors despite its potential for optimization.

CONCLUSION Polyesters from Pripol and PDO were obtained along various protocols and using different catalysts (metal and enzymatic). The monomers and products seem reasonably stable at temperatures up to 220 °C. The presence of oxygen does not change the course of the reaction nor the products. SEC-, NMR-spectroscopic and rheological measurements underline, that structurally identical linear polyesters were obtained in metal-, enzymatic- and auto-catalyzed reactions. High molecular weights were only obtained in the metal-catalyzed polycondensation, due to the higher temperature. It is of importance for a high final molecular weight to keep the stoichiometry balanced, in particular during the precondensation step. In this study, the maximal achievable molecular weight Mw using the enzyme CalB immo is 26.7 kDa (relative to PS). In order to obtain polyesters of Pripol and PDO of a molecular weight higher than 26.7 kDa in a reasonable time, a higher temperature than applicable for enzyme catalysis or an advanced set up

ACS Paragon Plus Environment

19

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

for removing the byproduct water would be needed. A polyurethane could easily be synthesized out of a polyester made from Pripol and PDO, if prepared in a modified protocol in order to obtain two hydroxyl termini.

EXPERIMENTAL SECTION Materials. All chemicals were used as received, if not stated otherwise. TiOnBu4 , 0.1 N potassium hydroxide solution and PDO were purchased from Sigma Aldrich. Pripol was a gift from Croda chemicals, the immobilized enzyme CalB immo from c-LEcta. 1,4-Butanediol, an anti-foaming agent on silicon basis, the catalyst (a mixture of phenylmercury(II) 2-(dodec-1-en-1-yl)succinate and 2-ethylhexanoic acid), Lupranat MM103 and IsoMMDI (all by BASF) were used for synthesis of a polyurethane elastomer. 1,4-Butanediol was dried at 80 °C at a rotational evaporator prior to use. Synthesis of polyesters. All polyesters were prepared in bulk polymerization of 27.77 g PDO (365.1 mmol) and 149.96 g Pripol (258.0 mmol) in a 250 ml glass reactor. The mixture was stirred with an anchor stirrer connected to an overhead agitator. The temperature was controlled with a heating jacket linked to a PT-100 resistance thermometer. Water, as well as the excess of PDO, were removed by vacuum. A distillation bridge, connected to the reactor, along with a cooling trap, both cooled with liquid nitrogen, was used to collect the distillate. At defined time points, samples were withdrawn and analyzed by titration and SEC. The final product was additionally characterized by 1H NMR and its shear dependent viscosity.

ACS Paragon Plus Environment

20

Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

Metal- and auto-catalyzed polycondensation. A non-catalyzed polycondensation was carried out in order to form oligomers prior to adding of the catalyst TiOnBu4 to the reaction mixture. The mixture was heated to 180 °C at a reduced pressure of 600 mbar and kept at these conditions for 100 min. Subsequently, the pressure was reduced to 20 mbar within 18 min and temperature was raised to 220 °C within 7 min. The pressure was reduced to less than 1 mbar by a rotary vane pump after adding 0.005 mol% of TiOnBu4 as solution in toluene (10 wt-%). The reaction was stopped after 365 min and dihydroxybenzene was added to the melt to inhibit oxygene-mediated crosslinking. A blank reaction without any catalyst was conducted equivalently for 1440 min. Enzyme-catalyzed polycondensation. Two different types of precondensation were carried out and compared to each other for enzyme-catalyzed polycondensations. Either an uncatalyzed precondensation (E1), identical to that in the metal-catalyzed route (180 °C, 600 mbar), or a precondensation at 80 °C and 100 mbar (E2) with enzyme initially added, were performed. Final reaction conditions were adjusted (80 °C and 1 mbar) after precondensation and enzyme was added (E1). In both reactions 5 wt% of CalB immo (aprox. 0.003 mol% active sites) were used as catalyst.

Preparation of a polyurethane elastomer. The polyesterdiol was prepared according to the metal-catalyzed route (M) described above with 0.08 mL Ti(OnBu)4. Three cycles of polymer build-up and chain scission by additional dosing of PDO (16 and 14 mL) were carried out to ensure hydroxyl end groups. The hydroxide number of the polyester was determined by a titrator TA 20 and the software TitriSift 2.6 (both

ACS Paragon Plus Environment

21

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 29

by Schott Geraete) according to DIN 53240-2 (half scale, double determination) with 30 mL methyl-ethyl ketone as solvation help. For polyurethane synthesis, the polyesterdiol, isocyanates and chain extender 1,4-butanediol were heated to 100 °C before use. Every component or mixture was extensively mixed by a SpeedMixer DAC400 FV at 2500 rpm for 30 s prior to further use. The polyol component (component A) consisted of 91.6 wt% polyesterdiol, 8.0 wt% 1,4-butanediol, 0.3 wt% antifoaming agent und 0.1 wt% catalyst. In the isocyanate component (component B), equal amounts by weight were present of Lupranat MM103 and IsoMMDI (both by BASF). After regaining a temperature of 100 °C in 30 min of both components, the A component was mixed again before the B component was added with an index of 100. After mixing for 30 s, the reaction mixture was spread with a layer thickness of 2 mm on a preheated aluminum plate (70 °C), stayed there for another 30 min and was annealed for 4 h at 80 °C. The PU sheet was conditioned at about 20 °C and 60% relative air humidity for 14 d before tensile testing by a ZwickRoell machine with videoXtens (in accordance of DIN 53504) and determining of Shore A hardness (according to DIN 53505) were carried out. Measurements. Molecular weight distribution was determined by size exclusion chromatography (SEC) at room temperature using a HPLC pump (Flom AI-12), a degasser (PLDG 802 by Polymer Laboratories, now Varian), a differential refractometer (RI 101 by Shodex) and an interface (hs 2600 by hs). A precolumn (MZ Gel SDplus, 5 µm, 100 Å by MZ-Analysentechnik) and a series of three columns were used (a Polypore linear, 1000 Å, 5 µm by Polymer Laboratories (now Agilent), a MZ-Gel SDplus linear, 1000 Å, 5 µm and a MZ-Gel SDplus 100 Å, 3 µm both by MZ-Analysentechnik). Distilled tetrahydrofuran (THF) was used as eluent at a flow rate of 1 ml

ACS Paragon Plus Environment

22

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

min-1. The system was conventionally calibrated with seven narrow distributed polystyrene standards (980, 2000 11870, 22480, 46790, 119900, 222100, 630000 and 1000000 g mol-1). 5 mg ml-1 polyester samples were solved in THF, filtered with a 0.45 µm PTFE filter and, after adding 3 µl of toluene as internal standard, injected in the SEC. Calculations were done with the software NTeQGPC V 6.4 (Version V 1.0.25 by hs). NMR spectra of samples after precondensation and of the final products were recorded on a Bruker Fourier 300, an Avance 400 or a DRX500 MHz spectrometer with CDCl3 or acetone-d6 (for characterization of Pripol) as solvent, dependent of the utilized experiment and TMS as internal standard. DSC measurements were carried out on a DSC 1 equipped with an autosampler (Mettler Toledo) under nitrogen and calibrated in temperature and energy with high-purity standards (indium and zinc). TGA measurements were conducted on a TGA 1 also equipped with an autosampler (Mettler Toledo). The data was evaluated in both cases by the software STARe Evaluation Software, Version 11.00 (Mettler Toledo). For DSC measurements, about 8 mg of sample were placed in an aluminum pan and subjected to heating from -100 °C to 150 °C followed by cooling to -100 °C and another heating to 230 °C with a rate of 10 K·min-1 in all steps. TGA curves were recorded of 10 to 20 mg sample by heating from 25 to 550 °C with 10 K·min-1 in a nitrogen atmosphere. MALDI-TOF spectra were obtained by a Bruker UltrafleXtreme equipped with a Smartbeam II Laser in linear and repulsion mode with Dithranol as matrix and silver(I) trifluoroacetate as ionization agent. Rheological measurements were performed on a AR 2000ex controlled stress rheometer (TA Instruments, New Castle, USA) with a plate-plate geometry (diameter = 25 mm). SAOS experiments were carried out at a frequency range of 628 to 0.1 rad s-1 at a temperature of 80 °C.

ACS Paragon Plus Environment

23

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 29

The strain (8 to 40 %) was kept within the limits of the linear viscoelastic regime. A steady state shear experiment has been carried out after each SAOS experiment in a frequency range of 0.1 to 1000 Hz. Time sweeps of all samples were carried out at constant oscillatory frequency at 80 °C. No significant changes of G’, G’’, δ and |η*| of the samples were detected. The model of CROSS resulted in the best fits for obtaining the zero shear viscosity, except for the measurements of the polymer of Ma which was fitted according to MIN. Acid value (AV) titration was performed for determination of the acid conversion. Withdrawn samples were dissolved in 40 ml THF and titrated with 0.1 M potassium hydroxide in ethanol using a Brand Titrette on a 1 L Schott bottle. Phenolphthalein was used as indicator. The acid value (AV) was calculated according to eq 1. AV =

 V∙c∙56.1 

m



 

  !"

(1)

with V being the volume of potassium hydroxide solution in milliliters, c the concentration of potassium hydroxide solution in mol L-1 and m the sample mass in gram. The conversion was calculated according to eq 2. X=

AV0 #AV AV0

∙ 100 $%%

(2)

with the acid value AV of the initial reactants’ mixture and at a reaction time t. For reactions with initially present immobilized enzyme, AV0 is calculated using the total mass of reaction mixture and enzyme, elsewise with just the mass of the mixture. The dilution by adding immobilized catalyst during the course of reaction is considered by adjusting the measured AV upwards by the factor 1.05. The loss of mass by removing water as well as dilution by adding titanium catalyst in the chemical route is not taken into account.

ACS Paragon Plus Environment

24

Page 25 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

ASSOCIATED CONTENT Supporting Information. The following files are available free of charge. 13

C NMR spectra of Pripol, results of corresponding experiments in air atmosphere, MALDI-

TOF spectra of final products (exemplary) for identification of cyclic products. (pdf file)

AUTHOR INFORMATION Corresponding Author *E-Mail: [email protected]. *E-Mail: [email protected].

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ║These authors contributed equally. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT The authors would like to thank the BASF for donating the chemicals for polyurethane synthesis. They also gratefully acknowledge the contribution of Dr. F. Scheliga and D. Szopinski in form of

ACS Paragon Plus Environment

25

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 29

many in depth discussions, of I. Fink for assistance in PU synthesis and of K. Rehmke, who carried out the rheological measurements. ABBREVIATIONS NMR, nuclear magnetic resonance; PDO, 1,3-propanediol; Pripol, PripolTM 1012; SEC, size exclusion chromatography; THF, tetrahydrofuran

REFERENCES (1) Loos, K. Biocatalysis in polymer chemistry; Wiley: Weinheim, 2011. (2) Köpnick, H.; Schmidt, M.; Brügging, W.; Rüter, J.; Kaminsky, W. Ullmann's Encyclopedia of Industrial Chemistry; Wiley: Weinheim, 2000. (3) Elias, H.-G. J. Macromol. Sci. 1978, 12, 183. (4) Yu, Y.; Di Wu; Liu, C.; Zhao, Z.; Yang, Y.; Li, Q. Process Biochem. 2012, 47, 1027. (5) Szmercsányi, I. V.; Maros, L. K.; Zahran, A. A. J. Appl. Polym. Sci. 1966, 10, 513. (6) Ordelt, Z.; Novák, V.; Krátký, B. Collect. Czech. Chem. Commun. 1968, 33, 405. (7) Gross, R. A.; Ganesh, M.; Lu, W. Trends in Biotechnol. 2010, 28, 435. (8) Jiang, Y.; Woortman, Albert J. J.; Alberda van Ekenstein, Gert O. R.; Loos, K. Polym. Chem. 2015, 6, 5451. (9) Scherkus, C.; Schmidt, S.; Bornscheuer, U. T.; Gröger, H.; Kara, S.; Liese, A. ChemCatChem. 2016, 8, 3446. (10) Wedde, S.; Rommelmann, P.; Scherkus, C.; Schmidt, S.; Bornscheuer, U. T.; Liese, A.; Green Chem. 2017, 19, 1286. (11) Yang, Y.; Lu, W.; Zhang, X.; Xie, W.; Cai, M.; Gross, R. A. Biomacromolecules. 2010, 11, 259.

ACS Paragon Plus Environment

26

Page 27 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

(12) Vilela, C.; Sousa, A. F.; Fonseca, A. C.; Serra, A. C.; Coelho, Jorge F. J.; Freire, Carmen S. R.; Silvestre, Armando J. D. Polym. Chem. 2014, 5, 3119. (13) Hablot, E.; Zheng, D.; Bouquey, M.; Avérous, L. Macromol. Mater. Eng. 2008, 293, 922. (14) Bueno-Ferrer, C.; Hablot, E.; Garrigós, María del Carmen; Bocchini, S.; Avérous, L.; Jiménez, Polym. Degrad. Stab. 2012, 97, 196.4. (15) Bueno-Ferrer, C.; Hablot, E.; Perrin-Sarazin, F.; Garrigós, M. C.; Jiménez, A.; Avérous, L. Macromol. Mater. Eng. 2012, 297, 777. (16) Reulier, M.; Avérous, L. Eur. Polym. J. 2015, 67, 418. (17) Poussard, L.; Lazko, J.; Mariage, J.; Raquez, J.-M.; Dubois, P. Prog. Org. Coat. 2016, 97, 175. (18) Petrović, Z. S.; Milić, J.; Zhang, F.; Ilavsky, J. Polymer 2017, 121, 26. (19) Gandini, A.; Lacerda, T. M. Prog. Polym. Sci. 2015, 48, 1. (20) Kim, D.; Kim, J.-H.; Kwon, S. H.; Lee, S. O.; Seo, B.; Lim, C.-S. Polym. Bull. 2017, 130, 1421. (21) Liu, H.; Xu, Y.; Zheng, Z.; Liu, D. Biotechnol. J. 2010, 5, 1137. (22) Biebl, H.; Menzel, K.; Zeng, A.-P.; Deckwer, W.-D. Appl. Microbiol. Biotechnol. 1999, 52, 289. (23) Leja K. Afr. J. Microbiol. Res. 2011, 5, 4652. (24) Liese, A.; Seelbach, K.; Wandrey, C. Industrial Biotransformations; Wiley: Hoboken, 2006. (25) Varma, I. K.; Albertsson, A.-C.; Rajkhowa, R.; Srivastava, R. K. Prog. Polym. Sci. 2005, 30, 949. (26) Uyama, H., Kobayashi, S. Enzyme-Catalyzed Synthesis of Polymers; Springer-Verlag: Berlin/Heidelberg, 2006.

ACS Paragon Plus Environment

27

Organic Process Research & Development

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 29

(27) Kobayashi, S. Polym. Adv. Technol. 2015, 26, 677. (28) Gustini, L.; Lavilla, C.; Janssen, William W T J; Martínez de Ilarduya, Antxon; MuñozGuerra, S.; Koning, C. E. ChemSusChem. 2016, 9, 2250. (29) Ulber, R.; Sell, D. White biotechnology; Springer: Berlin, New York, 2007. (30) Bommarius, A. S.; Schwarm, M.; Stingl, K.; Kottenhahn, M.; Huthmacher, K.; Drauz, K. Tetrahedron: Asymmetry 1995, 6, 2851. (31) Hilterhaus, L.; Thum, O.; Liese, A. Org. Process Res. Dev. 2008, 12, 618 (32) Thum, O. TSD 2004, 41 , 287. (33) Hollmann, F.; Gumulya, Y.; Tölle, C.; Liese, A.; Thum, O. Macromolecules 2008, 41, 8520. (34) Gross, R. A.; Kalra, B.; Kumar, A. Appl. Microbiol. Biotechnol. 2001, 55, 655. (35) Gross, R. A.; Kumar, A.; Kalra, B. Chem. Rev. 2001, 101, 2097. (36) Ferrario, V.; Pellis, A.; Cespugli, M.; Guebitz, G.; Gardossi, L. Catalysts 2016, 6, 205. (37) Kobayashi, S.; Makino, A. Chem. Rev 2009, 109, 5288. (38) Nguyen, H.; Löf, D.; Hvilsted, S.; Daugaard, A. Polymers 2016, 8, 363. (39) Liese, A.; Hilterhaus, L. Chem. Soc. Rev. 2013, 42, 6236. (40) Isaacs, E.; Standring, N.; Fletcher, N. DE1420515, 1969. (41) Moorhouse, D. JSDC 1993, 109 ,255. (42) Carothers, W. H. Trans. Faraday Soc. 1936, 32, 39. (43) Ouhadi, T.; Stevens, C.; Teyssié, P. Makromol. Chem. 1975, 1, 191. (44) Kricheldorf, H. R.; Berl, M.; Scharnagl, N. Macromolecules 1988, 21, 286. (45) Kricheldorf, H. R.; Mang, T.; Jonte, J. M. Macromolecules 1984, 17, 2173. (46) Poojari, Y.; Clarson, S. J. Biocatal. Agric. Biotechnol. 2013, 2, 7.

ACS Paragon Plus Environment

28

Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Organic Process Research & Development

(47) Berkane, C.; Mezoul, G.; Lalot, T.; Brigodiot, M.; Maréchal, E. Macromolecules 1997, 30, 7729. (48) Wallace, J. S.; Morrow, C. J. J. Polym. Sci. A Polym. Chem. 1989, 27, 3271. (49) Linko, Y.-Y.; Wang, Z.-L.; Seppälä, J. Enzyme Microb. Technol. 1995, 17, 506. (50) Kricheldorf, H. R. Macromol. Symp. 2003, 199, 15. (51) Banach, T. E.; Colonna, M. Polymer 2001, 42, 7517. (52) Kozlowska, A.; Ukielski, R. Eur. Polym. J. 2004, 40, 2767. (53) Liu, X.; Xu, K.; Liu, H.; Cai, H.; Su, J.; Fu, Z.; Guo, Y.; Chen, M. Prog. Org. Coat. 2011, 72, 612. (54) Yu, F.; Saha, P.; Suh, P. W.; Kim, J. K. J. Appl. Polym. Sci. 2015, 132, 1. (55) Cox, W. P.; Merz, E. H. J. Polym. Sci. 1958, 28, 619. (56) Bueche, F. J. Polym. Sci. 1960, 43, 527. (57) Zhang, S.-L.; Wang, G.-B.; Jiang, Z.-H.; Wang, D.; Ma, R.-T.; Wu, Z.-W. J. Polym. Sci. B Polym. Phys. 2005, 43, 1177. (58) Reulier, M.; Avérous, L. Eur. Polym. J. 2015, 67, 418. (59) Bueno-Ferrer, C.; Hablot, E.; Garrigós, María del Carmen; Bocchini, S.; Avérous, L.; Jiménez, A. Polym. Degrad. Stab. 2012, 97, 1964.

ACS Paragon Plus Environment

29