Investigation of non-isocyanate urethane dimethacrylate reactive diluents for UV-curable polyurethane coatings

Progress in Organic Coatings 76 (2013) 1057–1067

Contents lists available at SciVerse ScienceDirect

Progress in Organic Coatings journal homepage: www.elsevier.com/locate/porgcoat

Investigation of non-isocyanate urethane dimethacrylate reactive diluents for UV-curable polyurethane coatings Xiaojiang Wang, Mark D. Soucek ∗ Department of Polymer Engineering, University of Akron, Akron, OH 44325, USA

a r t i c l e

i n f o

Article history: Received 11 January 2013 Received in revised form 28 February 2013 Accepted 1 March 2013 Available online 26 March 2013 Keywords: UV-curable Non-isocyanate Reactive diluent Acrylated polyester Polyurethane

a b s t r a c t Three non-isocyanate urethane dimethacrylate reactive diluents 2-(methacryloyloxy)ethyl 2-(methacryloyloxy)ethylcarbamate (EOAED), 2-(methacryloyloxy)ethyl 3-(methacryloyloxy) propylcarbamate (POAED), and 1-(methacryloyloxy)propan-2-yl 3-(methacryloyloxy)propylcarbamate (POAPD) were synthesized by the reaction of a cyclic carbonate with an amino alcohol followed by a second reaction with the methacrylic anhydride. These reactive diluents were formulated with an acrylated polyester (APE) oligomer and free radical photoinitiator to prepare UV-curable polyurethane coatings. For comparison with urethane dimethacrylate reactive diluents, ethylene glycol dimethacrylate (EGDMA) was also used. The effect of reactive diluent type and content on the viscosity of the APE oligomer was measured. After UV curing, the viscoelastic, tensile, and thermal properties of the cured films were evaluated as a function of the reactive diluent using dynamic mechanical thermal analysis (DMTA), tensile, differential scanning calorimeter (DSC), and thermal gravimetric analysis (TGA). In addition, coating properties such as pencil hardness, chemical resistance, impact resistance, and gloss were also investigated. It was found that crosslink density, storage and tensile modulus, pencil hardness, chemical resistance, gel content, total water absorption, and glass transition temperature (Tg ) were directly proportional to the amount of the reactive diluent. The urethane dimethacrylate reactive diluents show significant improvements in impact resistance and elongation-at-break properties compared to the EGDMA. It was found that the optimum level of the urethane dimethacrylate reactive diluents concentration is between 10 and 20 wt%. © 2013 Elsevier B.V. All rights reserved.

1. Introduction UV-curable coatings have received considerable increasing attention on account of fast curing and low energy consumption [1]. Formulations for UV-curable coatings usually consist of three major components: photoinitiator, oligomer, and reactive diluent [2]. The oligomer is functionalized with reactive end groups that participate in the film formation process, and the structure of the oligomer governs the viscoelastic properties of the final cured film. The reactive diluent lowers the viscosity of the resin, and copolymerizes with oligomer to form the crosslinked film [3]. There are three major types of oligomers widely used in free radical UV-curable coatings: epoxy acrylate, polyester acrylate (acrylated polyester, APE), and urethane acrylate oligomer. Urethane acrylate oligomers are usually prepared by the reaction of a polyol with a diisocyanate to yield an isocyanate terminated oligomer. This oligomer is then reacted with a hydroxyethyl acrylate monomer [4]. Urethane acrylates are widely used as oligomers

∗ Corresponding author. E-mail address: [email protected] (M.D. Soucek). 0300-9440/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.porgcoat.2013.03.001

for UV-curable coatings, as a consequence of excellent mechanical properties, combined with excellent chemical resistance [1,4–8]. A drawback of the urethane acrylate is the toxicity of the isocyanate starting material [9]. Reactive diluents for free radical based UV-curing systems are usually acrylic or methacrylic monomers which are added to reduce the viscosity of precured liquid oligomer and modify the property of final cured solid film [10]. Generally, mono-functional reactive diluents lead to decreased modulus and increased ductility, whereas diand multi-functional reactive diluents lead to the opposite effect [11]. It was well established that a high degree of functionality leads to high reaction rate and high degree of crosslink density [12]. The high degree of functionality also can lead to low final degree of conversion, because the early gelation of the irradiated sample restricts the mobility of the reactive sites. In addition, the high degree of functionality also can lead to glassy polymer materials, which are harder and less flexible than monofunctional type systems [13,14]. In previous study, the effect of the functionality and chemical structure of the reactive diluents (ethylhexylacrylate, hexanediol diacrylate, and isobornyl acrylate as reactive diluents) on the thermal and mechanical properties of the UV-curable urethane acrylate oligomer were investigated [15]. The functionality of the reactive

1058

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

diluent did not affect the onset value of the glass transition temperature. However, the amount of the diacrylate reactive diluent did have a proportional effect on the equilibrium and storage modulus in the rubbery state [15]. The structure of the reactive diluent has a effect on oxygen inhibition of (meth)acrylates photopolymerization. Compared to acrylates, methacrylates are much less sensitive to oxygen [16]. More recently, urethane dimethacrylate monomers prepared via the non-isocyanate route from the reaction of a urethane diol with methacrylic anhydride have been reported [17–19]. Photopolymerization of the urethane dimethacrylate monomer was investigated with respect to polymerization rates and conversions using photoinitiated differential scanning calorimetry (photo-DSC) [17]. Although the photopolymerization of monomers were reported, the monomers were not used as reactive dilute that is with a reactive oligomer. In this paper, UV-curable polyurethane (PU) coatings were prepared using an acrylated polyester (APE) reactive oligomer with three urethane dimethacrylate reactive diluents, respectively. Three reactive diluents 2-(methacryloyloxy)ethyl 2(methacryloyloxy)ethylcarbamate (EOAED), 2-(methacryloyloxy) ethyl 3-(methacryloyloxy)propylcarbamate (POAED), and 1(methacryloyloxy)propan-2-yl 3-(methacryloyloxy)propylcarbamate (POAPD) were synthesized by the reaction of the cyclic carbonate with the amino alcohol followed by the reaction with the methacrylic anhydride. For comparison, ethylene glycol dimethacrylate (EGDMA) was also used as reactive diluent. The viscoelastic, tensile and thermal properties of the cured films were evaluated as a function of the reactive diluent. In addition, coating properties such as pencil hardness, chemical resistance, impact resistance, and gloss were also investigated. 2. Experimental 2.1. Materials Adipic acid (AA, 99%), isophthalic acid (IPA, 99%), 1,6 hexane diol (HD, 97%), trimethylolpropane (TMP, 98%), p-xylene (99%), dibutyl tin oxide (DBTO, 98%), ethylene carbonate (EC, 99%), propylene carbonate (PC, 99%), 2-aminoethanol (EOA, 99%), 3-aminopropanol (POA, 99%), methacrylic anhydride (MAA, 94%), dichloromethane (99%), hydroquinone (99%), triethyleneamine (TEA, 99%), potassium hydroxide (KOH, 85%), chloroform-d (CDCl3 , 100%), dimethyl sulfoxide-d6 (DMSO-d6 , 100%), anhydrous magnesium sulphate (99%) all Aldrich products were used as received. 4-(dimethylamino)pyridine (DMAP, 99%) was obtained from Acros Organics. Photoinitiator Darocur 4265 was obtained from Ciba Specialty Chemical, NY. Aluminum panels (type A with number A-36, bare surface, smooth mill finish, 3 × 6 in.) were obtained from Q-panel Lab Products. 2.2. Instrumentation The nuclear magnetic resonance (NMR) spectra were taken in a Varian Mercury 300 MHz spectrometer. Molecular weight and its distribution were determined by gel permeation chromatography (GPC) (Waters Corporation). The coating was cured by a Fusion UV-curing chamber (F300SQ Series) having a mercury arc bulb (150 mW cm−2 , UVB, 257 nm). Tensile tests were performed on an Instron 5567 (Instron Corp., Grove City, PA). The viscoelastic properties were measured on a dynamic mechanical thermal analyzer (DMTA, Q800, TA Instruments). The thermal properties were characterized by differential scanning calorimeter (DSC, Q200, TA Instruments) and thermogravimetric analysis (TGA, Q50, TA Instruments).

2.3. Preparation of the acrylated polyester (APE) oligomer In a typical procedure, APE oligomer was prepared via a two-step reaction. The hydroxyl terminated polyester was synthesized in the first step. The molar ratio of alcohol to acid was 11:8. Adipic acid (AA) (100 g, 0.68 mol), isophthalic acid (IPA) (113.67 g, 0.68 mol), 1,6 hexane diol (HD) (161.71 g, 1.37 mol) and trimethylolpropane (TMP) (45.90 g, 0.34 mol) were charged in a 500 mL four-neck round bottom flask which was equipped with a mechanical stirrer, a gas inlet, a temperature controller, a Dean-Stark trap, and a condenser. The reaction proceeded under argon to minimize oxidative degradation. To accelerate the reaction, 0.4 wt% (1.69 g) of DBTO, a transesterification catalyst, was used. Then, 3 wt% (12.64 g) of pxylene was also added into the reaction mixture to remove water from the resin as azeotropic mixture. The reaction temperature was carefully controlled using a temperature controller and a thermocouple in order to minimize evaporation of diol. The temperature of the mixture was increased from 20 to 150 ◦ C at a rate of 4.3 ◦ C/min, and then from 150 to 210 ◦ C at a rate of 0.25 ◦ C/min. The final temperature was held until the resin had an acid number measured by titration less than or equal to 10 mg KOH/g resin (ASTM D 163990). The hydroxyl number of polyesters (160.4 mg KOH/g resin) was determined by ASTM method (ASTM D 1957-86). Acrylate terminated polyester was synthesized in the second step, the temperature was held to 120 ◦ C. Acrylic acid (AA, the mole ratio of the acid group in the AA to the hydroxyl group in the polyester polyol kept at 1.05), 1 wt% (3.72 g, 21.6 mmol) of p-toluene sulphonic acid (p-TSA) catalyst and 0.05 wt% (186 mg, 1.7 mmol) of hydroquinone inhibitor were added, the temperature was maintained at 120 ◦ C for 14 h. Then, 1.5 wt% (5.58 g, 54.6 mmol) of 3-methyl-3-hydroxymethyl-oxetane was added, and maintained at 120 ◦ C for another 20 min. Then, solvent and volatile residuals were removed in vacuo resulting in a light yellow acrylated polyester oligomer: 1 H NMR (300 MHz, CDCl3 ) ı (ppm) 0.84–0.98 (m, 2 H) 1.33–1.47 (m, 8 H) 1.50–1.53 (m, 1 H) 1.56–1.69 (m, 11 H) 1.73–1.83 (m, 4 H) 2.22–2.39 (m, 6 H) 3.57–3.69 (m, 1 H) 3.99–4.16 (m, 6 H) 4.27–4.44 (m, 5 H) 5.75–5.89 (m, 1 H) 6.03–6.16 (m, 1 H) 6.32–6.44 (m, 1 H) 7.45–7.56 (m, 1 H) 8.11–8.28 (m, 2 H). The average molecular weight was found to be Mn = 3500, with a PDI = 1.8 via GPC. 2.4. Preparation of non-isocyanate urethane dimethacrylate reactive diluents The synthesis process consists of two steps. In a typical procedure to prepare EOAED non-isocyanate urethane dimethacrylate reactive diluent, ethylene carbonate (88.06 g, 1.00 mol) was dissolved in 100 mL dichloromethane, and such solution was drop added into the 2-aminoethanol (61.08 g, 1.00 mol) and dichloromethane (200 mL) mixture at 0 ◦ C. After finish adding, the mixture was stirred at room temperature for 24 h. The slightly yellow liquid (yield: 98%) was obtained after rotary evaporation of the dichloromethane. In the second step, urethane diol EOA-EC (14.9 g, 0.1 mol) was dissolved in 300 mL dichloromethane at 0 ◦ C, 4(dimethyl-amino) pyridine (DMAP) catalyst (122 mg, 1 mmol), and hydroquinone inhibitor (28 mg, 0.25 mmol) were added, followed by drop addition of triethylene diamine (TEA) (28.3 g, 0.28 mol), and then drop addition of methacrylic anhydride (39.4 g, 0.24 mol) under the N2 atmosphere. The reaction mixture was stirred at 0 ◦ C for 24 h. Then saturated sodium bicarbonate solution (294.7 g) was drop added to get two phase separated mixture. The product (top layer) was collected, washed with brine (300 mL 3×) and distilled water (300 mL 2×), and dried in anhydrous magnesium sulphate. After the dichloromethane was evaporated, a pale yellow liquid product was obtained (yield: 42%). 1 H NMR (300 MHz, DMSOd6 ) ı (ppm) = 1.86 (m, 6 H, CH3 ), 3.27 (qd, J = 5.67, 1.39 Hz, 2 H,

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

CH2 NH ), 4.07 (td, J = 5.57, 1.63 Hz, 2 H, OCH2 CH2 O ), 4.12 (m, 2 H, OCH2 CH2 NH ), 4.14–4.30 (m, 2 H, OCH2 CH2 O ), 5.62–5.78 (m, 2 H, trans C CH2 ) 5.98–6.12 (m, 2 H, cis C CH2 ), 7.42(t, 1 H, NHCOO ). 13 C NMR (300 MHz, DMSO-d6 ) ı (ppm) 18.11 ( CH3 ), 39.79 ( CH2 NH ), 62.11 ( NHCOOCH2 ), 63.32 ( CH2 CH2 O ), 63.49 ( COOCH2 ), 126.18 (C CH2 ), 136.10 (C CH2 ), 156.46 (C O, carbamate), 166.67 (C O, ester).

2.5. Coating formulation and evaluation Both aluminum and glass panels were used for film preparation. The substrates were cleaned with acetone and dried. The formulations were made by mixing amount (6.0, 7.0, 8.0, 9.0, and 10.0 g) of APE oligomer with corresponding amount (4.0, 3.0, 2.0, 1.0, and 0 g) of reactive diluent (EGDMA, EOAED, POAED, or POAPD), and 0.2 g of photoinitator D1173 in a glass vial at room temperature thoroughly for 20–30 min. For each formulation, it is named by “the name of the reactive diluent” followed by “the weight percent of this reactive diluent based on the total resin”. For example, “EGDMA-10” denotes the formulation which contains 10 wt% EGDMA reactive diluent with 90 wt% APE oligomer, and 2 wt% photoinitator D1173 (based on the weight of total resin). “APE” denotes the formulation which contains the APE oligomer only and 2 wt% photoinitator D1173 (based on the weight of total resin). The films were cast on the substrate with a thickness of 150 ␮m (6 mil) by a drawdown bar. The films were UV-cured using

a Fusion UV-system processor (P300) at the belt speed of 5 fpm (feet per minute) and stored in a dust free cabinet for 3 days before testing. Coating properties were evaluated according to ASTM methods including pencil hardness (D3363-05), impact resistance (G14), gloss rating (D523-89), and MEK double rubs (D4752-06). The gel content of the cured coating films was determined by Soxhlet extraction using acetone for 48 h [20,21]. The insoluble gel fraction was dried under vacuum for 24 h at 60 ◦ C and weighed to determine the gel content. To measure the water absorption of a crosslinked coating, ∼0.2 g coating sample was dried in vacuum for 24 h at 100 ◦ C, and then dipped in water at room temperature for 72 h. Afterwards, the wet sample was dried with paper towel and weighed. The water absorption was then calculated [22] from the difference in the weights of the sample before and after soaking in the water according to Eq. (1)

3. Results The overall objective of this paper was to evaluate urethane dimethacrylate reactive diluents in a UV-curable coating formulation consisting of an acrylated polyester as the reactive diluent

COOH COOH

+ OH

HO

COOH

DBTO

H

O

O

O

O

O

n

120oC/14hr p-TSA

O

4

O

O

OH

O

O

OH 6

O

6

O

O

OH

O

210 oC

O

4

6

O

O

O 6

O

(1)

where wtafter is the weight of the sample after dipping in the water and wtbefore is the weight of the sample before dipping in the water.

HOOC

OH

wtafter − wtbefore × 100 wtbefore

Water absorption (%) =

OH HO

1059

O

O

O

O Fig. 1. The synthesis of acrylated polyester (APE) oligomer.

n

O

1060

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

R'

R HO

+

NH2

O

O

O R' H N

HO

O

+

OH

R

H N

HO

O OH

R

O

O

R'

O O

O

R' H N

O

O

O

R

O

O

O

+ H N

O

O O

R O

O O

R'

Fig. 2. Non-isocyanate urethane dimethacrylate reactive diluents synthesis scheme.

O

oligomer. A series of urethane dimethacrylates were prepared via a non-isocyanate route and used as reactive diluents. Darocur 4265 was chosen as free radical photoinitiator. To investigate the effect of reactive diluents on the coating properties, and compare the urethane dimethacrylates with EGDMA reactive diluent, coatings containing different reactive diluents and concentration were prepared. 3.1. Preparation of reactive oligomers and reactive diluents The acrylated polyester (APE) oligomer was synthesized via a two-step reaction illustrated in Fig. 1. In the first step, hydroxyl terminated polyester was prepared via a typical poly-condensation technique with an excess of diols. In the second step, the hydroxyl terminated polyester was transferred to acrylated polyester by the reaction with acrylic acid. The non-isocyanate urethane dimethacrylate reactive diluents were synthesized via a two-step reaction illustrated in Fig. 2. In the first step, urethane diol was prepared through the reaction of amino alcohol with ethylene carbonate by modification of the Rokicki’s procedure [23]. In the second step, urethane diol was transferred to urethane dimethacrylate by the reaction with was prepared methacrylic anhydride, this step is the modification of Assumption’s procedure [17]. Fig. 3 summarizes the urethane dimethacrylate prepared by this method.

O O O

EGDMA

O O

O O

N H O

O

O

O

O POAED

O

O

O

H N

O

O

O

O O

O

O

H N

O

EOAED

O

H N

O

O

O Fig. 3. Reactive diluents chemical structure.

POAPD

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

a.

4

1061

b.

O

O

10

N

O N

O

Viscosity (mPas)

H H

O O 3

N

10

O O

POAPD EOAED POAED EGDMA

H Fig. 5. Viscosity increase due to the intermolecular hydrogen bonds formation via (a) NHurethane · · ·O Curethane interaction; and (b) NHurethane · · ·O Cester interaction.

2

10

0

10

20

30

40

was chosen to compare with the urethane dimethacrylate reactive diluents. The viscosity of the APE oligomer used was found to be 10840 mPa s. As expected, all three urethane dimethacrylate reactive diluents effectively reduced the viscosity of the APE oligomer. Although the urethane dimethacrylate reactive diluents were effective, the formulations containing urethane dimethacrylate reactive diluents and APE always showed a higher viscosity, this was attributed to intermolecular hydrogen bonds formation in this system as shown in Fig. 5. The intermolecular hydrogen bonds can be formed via either NHurethane · · ·O Curethane or NHurethane · · ·O Cester interactions [24]. Where NHurethane stands for the amine group from the urethane, O Curethane stands for the carbonyl group from the urethane, and O Cester stands for the carbonyl group from the ester.

Reactive Diluents Concentration (Wt %) Fig. 4. Viscosity comparison of reactive diluents with APE oligomer.

3.2. Viscosity of reactive diluents with acrylated polyester (APE) First, to be a reactive diluent, the urethane dimethacrylates should reduce the viscosity of the APE oligomer. The viscosities of the formulations were investigated and compared to that of the APE oligomer as shown in Fig. 4. The viscosity was studied as a function of amount and type of reactive diluent. A commonly used reactive diluent ethylene glycol dimethacrylate (EGDMA)

40 0 10 20 30 40

a.

30 Elongation at Break (%)

Tensile Stress (MPa)

30

35

20

10

25

0 10 20 30 40

b.

20 15 10 5

0

0 APE

EGDMA

0 10 20 30 40

c.

APE

EGDMA

EOAED

POAED

POAPD

--

APE

EGDMA

EOAED

POAED

POAPD

--

1400 1200

Modulus (MPa)

1000 800 600 400 200 0

EOAED

POAED

POAPD

Fig. 6. Tensile properties of UV cured coating as a function of reactive diluents: (a) tensile strength; (b) elongation-at-break; (c) tensile modulus.

1062

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

3.3. Tensile properties The tensile strength, elongation-at-break, and tensile modulus as a function of reactive diluents content are shown in Fig. 6. The tensile strength increased with the addition of any each of these four reactive diluents. When EGDMA was used as the reactive diluent, the tensile strength showed a initially large increase with 10 wt% and 20 wt% reactive diluents content, to maximum of 14.5 MPa with 30 wt% reactive diluents content, and a final decrease with 40 wt% reactive diluents content. When POAED was used as the reactive diluent, it showed the similar trend as EGDMA. The tensile strength showed increase with the amount of reactive diluent initially, and reached the maximum at 30 wt% reactive diluents content. The maximum tensile strength of POAED system was 21.96 MPa, which was 1.5 times higher than the EGDMA system. When EOAED or POAPD was used as the reactive diluent, the tensile strength showed a continuous increase with the amount of reactive diluent, and reached the maximum at 40 wt% reactive diluents content. The maximum tensile strength of EOAED system was 29.51 MPa, which was 2.1 times higher than the EGDMA system; and the maximum tensile strength of POAPD system was 26.30 MPa, which was 1.8 times higher than the EGDMA system. The elongation-at-break showed a continuous decrease with the addition of reactive diluent and reached the minimum of 1.26% at 40 wt% reactive diluents content for EGDMA system. While for all three urethane dimethacrylate systems, the elongation-atbreak increased with the addition of 10 wt% reactive diluents, and remained constant as the reactive diluents increased to 20 wt% (within experimental error). After which, the elongation-at-break

decreased with increased reactive diluent content. In comparison, the elongation-at-break of all three urethane dimethacrylate systems was always higher than that of the EGDMA system at the same percent loading in the range of 10–40 wt%. The elongationat-break of all three urethane dimethacrylate systems was 1.6–1.7 times higher than that of the EGDMA system at 10 wt% loading, ∼3 times higher at 20 wt% loading, and up to 5 times higher at 40 wt% loading. The tensile modulus showed a continuous increase with the addition of any each of these four reactive diluents in the percent loading range of 10–40 wt%. The tensile modules of the EGDMA systems was ∼2 times higher than that of the urethane dimethacrylate system in the percent loading range of 10–20 wt%, 1.5 times higher at 30 wt% loading, and almost equal at 40 wt% loading. This might be due to the fact that EGDMA system had a higher crosslink density than urethane dimethacrylate system. 3.4. Viscoelastic properties The viscoelastic properties of the UV cured coating films were investigated using DMTA. The storage modulus E as a function of temperature is shown in Fig. 7. The E in the rubbery plateau was used to calculate the crosslink density as described in Eq. (2) [25]. E  = 3 e RT

(2)

where e is the number of moles of elastically effective network chains per cubic meter of film and E is the storage modulus. This relationship is effective when T  Tg , and for low crosslink density elastomers. For the highly crosslinked system, it can only be

4

4

10

10

a.

EGDMA-40 EGDMA-30 EGDMA-20 EGDMA-10

b.

3

EOAED-40 EOAED-30 EOAED-20 EOAED-10

3

10

E' (MPa)

E' (MPa)

10

2

10

2

10

1

1

10

-80

(T  Tg )

10

-40

0

40 80 120 o Temperature ( C)

160

200

240

-80

4

-40

0

40 80 120 o Temperature ( C)

160

200

240

4

10

10

c.

POAED-40 POAED-30 POAED-20 POAED-10

POAPD-40 POAPD-30 POAPD-20 POAPD-10

d.

3

3

10

E' (MPa)

E' (MPa)

10

2

2

10

1

10

10

1

10

-80

-40

0

40 80 120 o Temperature ( C)

160

200

240

-80

-40

0

40 80 120 o Temperature ( C)

160

200

240

Fig. 7. Storage modulus (E ) as a function of temperature of the UV curable coating with different reactive diluents of (a) EGDMA; (b) EOAED; (c) POAED; and (d) POAPD.

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

used to evaluate the relative density of crosslinking. Crosslink density was calculated by this equation with the E at corresponding temperature T (T > Tg + 50). The minimum storage modulus E (min) and the crosslink density (e ) of the cured films increased with addition of the all four reactive diluents. At 10 wt% reactive diluents load, the storage modulus was doubled with the addition of the each of the four reactive diluents. In comparison, the minimum storage modulus of these three urethane dimethacrylate systems was always lower than that of the EGDMA system at the same percent loading in the range of 20–40 wt%. Crosslink density (e ) of these four cured films exhibited the same trends as storage modulus. The tan ı of the cured films as a function of reactive diluent content is shown in Fig. 8. The Tg and breadth of ␣-transition increased with the addition of reactive diluent in general. The maximum of ␣-transition decreased with addition of reactive diluent as well. Throughout the loading range, the Tg and ␣-transition maximum of all the urethane dimethacrylate system was higher than the EGDMA system. However, the breadth of ␣-transition of was similar for all the reactive diluents. The Tg , minimum storage modulus, E (min), crosslink density (e ), breadth of tan ı transition, and maximum tan ı transition as a function of reactive diluents are summarized in Table 1. 3.5. Thermal properties The thermal properties of the UV cured coatings were further investigated using DSC and thermal gravimetric analysis (TGA). The corresponding Tg and Cp (change in heat capacity) of the cured coatings derived from the DSC data are summarized in the Table 2.

It has been reported that the Tg increases with the increase of crosslinking density of polymers [26]. Cross-linking reduces the number of chains thermally activated and the chain mobility, and thus raises the Tg and reduces Cp [27]. For EGDMA system, the Tg increased with initial addition of 10 wt% reactive diluent, and kept at the same level with further increase of reactive diluents to the load range of 20–40 wt%. For the three urethane dimethacrylate systems, the Tg was proportional to the reactive diluents content. For all the four systems, the Cp decreased with the addition of the reactive diluents. As a comparison, the Tg of the urethane dimethacrylate system was always higher than that of the EGDMA system at the same reactive diluents content. Thermal gravimetric analysis (TGA) is summarized in Table 3. For the most part, reactive diluent concentration had no major impact on the films thermal stability until higher temperatures (>300 ◦ C) were achieved [28]. For EGDMA systems, it was found that the sample degradation of 5 wt% took place at 307 ◦ C for polyester coating, between 313 and 319 ◦ C for coating containing 10–40 wt% reactive diluent. This indicated that the thermal stability of the UV-cured polyester coatings slightly increased with the addition of the EGDMA reactive diluent. For EOAED systems, it was found that there was a shift to slightly lower temperatures as the reactive diluent loading increased. The sample degradation of 5 wt% took place at 291 ◦ C with 10 wt% reactive diluent content, at 279 ◦ C with 20 wt% reactive diluent, at 270 ◦ C with 30 wt% reactive diluent, and at 256 ◦ C with 40 wt% reactive diluent. This indicated that the thermal stability of the UV cured polyester coatings was slightly decreased with the addition of the EOAED reactive diluent. The POAED and POAPD system exhibited similar trends as EOAED system. The shapes of the weight loss curves of three

0.6

0.5

0.4

1063

EGDMA-40 EGDMA-30 EGDMA-20 EGDMA-10

a.

0.5

EOAED-40 EOAED-30 EOAED-20 EOAED-10

b.

0.4

Tan Delta

Tan Delta

0.3

0.2

0.3

0.2 0.1

0.0 -80

0.1

-40

0

40 80 o 120 Temperature ( C)

160

200

0.0 -80

240

0

40 80 o 120 Temperature ( C)

160

200

240

0.7

0.6

0.5

-40

POAED-40 POAED-30 POAED-20 POAED-10

c.

0.6

POAPD-40 POAPD-30 POAPD-20 POAPD-10

d.

0.5

Tan Delta

Tan Delta

0.4

0.3

0.2

0.3 0.2

0.1

0.0 -80

0.4

0.1

-40

0

40 80 o 120 Temperature ( C)

160

200

240

0.0 -80

-40

0

40 80 o 120 Temperature ( C)

160

200

240

Fig. 8. tan ı as a function of temperature of the UV curable coating with different reactive diluents of (a) EGDMA; (b) EOAED; (c) POAED; and (d) POAPD.

1064

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

Table 1 Viscoelastic properties of UV cured coating as a function of reactive diluent type and content.

APE EGDMA-10 EGDMA-20 EGDMA-30 EGDMA-40 EOAED-10 EOAED-20 EOAED-30 EOAED-40 POAED-10 POAED-20 POAED-30 POAED-40 POAPD-10 POAPD-20 POAPD-30 POAPD-40

 T for Emin (◦ C)

 Emin (MPa)

91.17 179.48 227.53 239.61 240.00 115.03 214.49 220.42 227.42 134.53 190.27 232.40 230.40 122.94 160.75 179.33 196.58

4.92 12.33 42.32 63.26 167.20 11.49 13.48 15.82 22.83 12.33 14.54 25.09 29.21 12.30 17.84 25.02 25.40

e (mol/m3 ) 542 1092 3389 4946 13,064 1187 1108 1285 1829 1213 1258 1990 2327 1245 1648 2217 2168

Table 2 DSC data of UV cured coating as a function of reactive diluents type and content. Samples

Tg (◦ C)

Cp (J/(g ◦ C))

APE EGDMA-10 EGDMA-20 EGDMA-30 EGDMA-40 EOAED-10 EOAED-20 EOAED-30 EOAED-40 POAED-10 POAED-20 POAED-30 POAED-40 POAPD-10 POAPD-20 POAPD-30 POAPD-40

−18.01 −12.40 −13.06 −13.15 −13.05 −9.88 −4.22 1.09 3.43 −10.63 −9.22 −3.80 −2.87 −10.33 −9.42 −6.88 2.33

0.53 0.43 0.40 0.36 0.29 0.51 0.42 0.40 0.27 0.49 0.45 0.30 0.25 0.54 0.51 0.51 0.34

tan ı breadth (◦ C)

Max. tan ı

43.0 47.52 52.21 59.88 61.59 55.41 61.80 75.45 127.83 54.34 61.16 72.68 97.41 53.91 69.69 79.92 103.81

24.99 48.61 71.66 98.95 123.19 45.28 68.38 101.64 131.21 49.89 74.85 98.86 123.82 39.74 64.68 86.86 116.43

0.91 0.39 0.25 0.18 0.13 0.48 0.36 0.27 0.23 0.45 0.32 0.24 0.20 0.59 0.38 0.29 0.24

degradation started initially at the urethane bond, and then at the ester bond in the non-isocyanate urethane dimethacrylate system. 3.6. Gel content properties

urethane dimethacrylate system were almost identical, and overall differences in thermal stability appeared to be small. The urethane bond was known to be relatively thermally unstable, and started to decompose at about 150–200 ◦ C [29]. This suggested that

The gel content is proportional to the crosslink density [22]. In order to further investigate the crosslink properties of the cured coating films, the gel content measurements were performed. The gel contents of the cured films increased with addition of all the four reactive diluents, and showed similar values at the same percent loading in the range of 10–40 wt% (as shown in Fig. 9). For all of these four cured coating films, the gel contents was increased from 88% initiately to 90% at 10 wt%, to 91% at 20 wt%, to 92% at 30 wt%, and to 93% at 40 wt% reactive diluents load. This indicated that the crosslink density increased with the addition of the reactive diluents. However, the gel content values did not reach over 93%; this indicated that at least 7% uncross-linked material was left at the coating system. This may suggest that this coating system was highly, but not completely UV cured. The limitation of double bond conversion is mainly due to the restricted mobility of polymer chain and the vitrification of the system [30]. In general, the gel content is used to measure the degree of curing. Meanwhile, it is generally accepted that a gel content of about 95% indicates

94

Table 3 Thermogravimetric analysis (TGA) data of UV cured coating as a function of reactive diluents type and content.

93

Temperature (◦ C) required to produce % weight loss 5%

10%

25%

50%

APE

307

348

374

401

EGDMA-10 EGDMA-20 EGDMA-30 EGDMA-40

319 314 313 316

351 347 345 347

382 378 378 379

406 403 404 406

EOAED-10 EOAED-20 EOAED-30 EOAED-40

291 279 270 256

330 319 313 303

371 370 367 357

398 399 399 396

POAED-10 POAED-20 POAED-30 POAED-40

298 295 290 279

336 331 326 319

375 376 374 367

402 404 405 402

POAPD-10 POAPD-20 POAPD-30 POAPD-40

291 279 262 251

334 323 313 305

375 371 365 358

402 400 398 396

92 Gel Content (Wt %)

Sample

Tg (◦ C)

91 90 EGDMA EOAED POAED POAPD

89 88 87 0

10

20

30

40

Reactive Diluents Weight Percentage (Wt %) Fig. 9. Effect of the reactive diluent type and content on the gel content properties of UV cured coatings.

Water Absorption Weight Percentage (Wt %)

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

1065

reactive diluent leads to the increase of the ester/urethane polar groups in the APE polymer matrix, increasing the interaction of the APE polymer and water, resulting in the increase in the total water absorption. It is known that the urethane group shows more polarity than the ester group. This results in the higher total water absorption of the urethane dimethacrylate systems comparing with the EGDMA system.

3.5

3.0

2.5

2.0

3.8. General coating properties 1.5

complete curing of the coatings [31]. Therefore, all the films with gel content from 88 to 93% demonstrated the general formation of a highly cross-linked network in the coating system.

A summary of the cured coating properties is found in Table 4. After complete UV curing, several film properties were evaluated: pencil hardness, impact resistance, gloss, and solvent resistance. It was found that the pencil hardness did not change with the addition of 10 wt% reactive diluents, and increased from 2B to HB with the addition of 20–40 wt% reactive diluents for all systems. The impact resistance of the EGDMA system decreased with the addition of reactive diluents in the load range of 10–40 wt%. The impact resistance of the three non-isocyanate urethane dimethacrylate systems initially increased with addition of 10–20 wt% reactive diluents, and then decreased with further addition of 30–40 wt% reactive diluents. The methyl ethyl ketone (MEK) solvent resistance increased with the addition of reactive diluents in the load range of 10–40 wt% for all systems. It was found that there was no effect on the gloss with the incorporation of the reactive diluents for all systems.

3.7. Water absorption properties

4. Discussion

The properties of water absorption by polymers have received considerable attention, due to the importance of these phenomena in the case of environmental aging of polymers. Studies have found that the water absorption of polymers is directly related with their free volume fraction [32], while other studies have found that it is linked to the presence of polar groups capable to form hydrogen bonds with water molecules, and there is no apparent influence of the large-scale structure (crosslinking) of the polymer on water absorption [32,33]. The effect of the reactive diluents on cured film was further verified in terms of water adsorption [34]. In Fig. 10, it is shown that the water absorption of the cured films increased with addition of all the four reactive diluents, and the water absorption of EGDMA system was found to be slightly lower than other three urethane dimethacrylate systems. The water absorption of the cured films increased with the content of reactive diluents. It could be attributed to the fact that the increase of the

Generally, UV-curable PU coatings are formulated using the urethane acrylate oligomers with (meth) acrylate reactive diluents. In this study, UV-curable PU coatings were formulated by using APE as oligomer, and non-isocyanate urethane dimethacrylate as reactive diluents. The UV-curable PU coatings formulated by this new method show more environmental friendly compared with conventional UV-curable PU coatings. As to our knowledge, it is the first time non-isocyanate urethane dimethacrylate reactive diluents were formulated into UV-curable coatings. The tensile modulus of UV-cured coating increased with the addition of the reactive content. This result is attributed to intra/inter molecular reactions between APE oligomer and reactive diluent, and the increase in crosslink density with the addition of reactive diluents. The tensile modulus of the urethane dimethacrylate systems were lower than that of the EGDMA system with the same content of reactive diluent, this may due to the fact that

1.0

EGDMA EOAED POAED POAPD

0.5

0

10

20

30

40

Reactive Diluents Weight Percentage (Wt %) Fig. 10. Effect of the reactive diluent type and content on the water absorption properties of UV cured coatings.

Table 4 General coating properties as a function of reactive diluents type and content. MEK double rubs

60◦ gloss

20◦ gloss

181 ± 3

142 ± 2

Samples

Direct impact (kg/cm)

Reverse impact (kg/cm)

Pencil hardness

APE

64

56

2B

16

EGDMA-10 EGDMA-20 EGDMA-30 EGDMA-40

55 50 44 26

52 40 42 7

2B HB HB HB

20 60 >200 >200

187 185 184 183

± ± ± ±

4 2 4 2

148 146 144 131

± ± ± ±

7 7 7 18

EOAED-10 EOAED-20 EOAED-30 EOAED-40

92 90 56 30

87 71 43 8

2B HB HB HB

36 44 >200 >200

189 182 180 183

± ± ± ±

2 2 5 4

149 140 138 145

± ± ± ±

2 6 6 5

POAED-10 POAED-20 POAED-30 POAED-40

85 77 47 31

71 71 31 10

2B HB HB HB

22 67 >200 >200

185 185 182 186

± ± ± ±

1 3 1 3

147 149 140 147

± ± ± ±

5 4 4 3

POAPD-10 POAPD-20 POAPD-30 POAPD-40

88 86 61 39

82 84 58 11

2B HB HB HB

18 44 >200 >200

189 190 185 186

± ± ± ±

2 2 1 2

154 145 146 143

± ± ± ±

8 6 7 5

1066

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067

urethane dimethacrylate systems provided lower crosslinking sites for the coatings. The urethane dimethacrylate reactive diluents have higher molecular weight than the EGDMA reactive diluent, thus the same weight percentage reactive diluent loading results in less urethane dimethacrylates and as a consequence a lower number of crosslinking sites. The trend in the storage modulus mirrored the trend in tensile strength. The increase in tensile and storage modulus as the concentration of reactive diluents increased corroborated the crosslink density, Tg , impact resistance and gel content data. The flexibility of the coating films was investigated by measuring the elongation at the moment of breakage. For EGDMA system, the elongation-at-break decreased with the addition of the reactive content. Such reduction in elongation was expected due to higher concentration of joint site between polymer chains, which restrict the polymer movement upon stretching [35]. More interestingly, for non-isocyanate urethane dimethacrylate system, the elongation-at-break increased with addition of the 10–20 wt% reactive diluent, and decreased with further increasing reactive diluent to 30–40 wt%. Such behavior might be the results of increasing physical interaction forces from the hydrogen bonding among urethane groups. The effect of the reactive diluent type and content on the impact properties exhibited the similar trends as elongationat-break properties. That is another evidence of improvement of the flexibility achieved through physical interaction forces via the hydrogen bonding in the non-isocyanate urethane dimethacrylate system. The crosslink density has a direct impact on the tan ı such that it reduces long range segmental motion required for viscous flow behavior [28]. As a result, the maximum tan ı can be used to compare the extent of cure. The maximum tan ı decrease with increasing reactive diluent content. This indicated that the crosslink density of the UV cured films increased with reactive diluent content. The maximum tan ı of the EGDMA system was lower than that of the non-isocyanate urethane dimethacrylate system at the same reactive diluent content. For the EGDMA, the tan ı max data was corroborated by the higher crosslink density, and higher tensile and storage modulus. The Tg was obtained as the maximum of ␣-transition. It was found the Tg increased with the content of the reactive diluent, and the Tg of non-isocyanate urethane dimethacrylate system is higher than that of the EGDMA system. As a comparison, the Tg measured by DSC exhibited the same trends as the Tg measured by DMTA [36]. The formation of the crosslinking networks between the reactive diluent and acrylated polyester oligomers would hinder the movement of the polymer chains, resulting in the increase of the Tg [37]. The Tg of urethane dimethacrylate system is higher than that of the EGDMA system. This can be attributed to the additional hydrogen bonding of the urethane groups in the urethane dimethacrylate system as shown in Fig. 5 [24]. It has been reported that incorporation of hydrogen bonds could improve the Tg of polymers to a great extent for a variety of polymer systems [38]. The hydrogen bonds can be considered as effective physical crosslinks [39,40], which reduce the polymer segmental and chain mobility. More specifically, the increase in Tg has been known to be proportional to the number of hydrogen bond crosslinks when the crosslink density is small [41]. 5–10% weight loss values, which corresponds the initial degradation temperature in TGA, showed two big differences between urethane dimethacrylate system and EGDMA system. One difference is that, with the same reactive diluents content, the degradation temperature always showed lower values for urethane dimethacrylate systems. Another difference is that, with the increase of the reactive diluents content, the degradation temperature decreased for urethane dimethacrylate systems, but slightly increased for EGDMA system. This indicated that the existence of

urethane group had great effect on the thermal stability of the cured coatings. For urethane dimethacrylate systems, the initial degradation occurred due to the rupture of C NH bond of urethane linkage (the thermally weakest link) [42,43]. The ruptured urethane linkage further depolymerize into isocyanate, alcohol, primary or secondary amine, olefin, and carbon dioxide [42], resulting the continuous decrease in mass of the cured coatings. For EGDMA system, the initial degradation occurred due to the random chain scission of alkyl-oxygen (C O) bond of the ester linkage [44], resulting the formation of vinyl ester and carboxyl end groups [45]. During the random chain scission the molecular weight of the polymer continuously decreases but with negligible change in polymer mass [46]. Moreover, it is known that the C NH bond of urethane linkage is weaker than the C O bond of ester linkage, and starts to decompose at a relatively lower temperature [44]. Thus, coating system with a higher content of weak urethane group has lower thermal stability [42,47], and shows lower initial degradation temperature. The decrease in impact resistance with an increase in the concentration of EGDMA can be attributed to the increase in crosslink density. But for non-isocyanate urethane dimethacrylate system, the increase in impact resistance with the addition of 10–20 wt% reactive diluent, and decreased with further increasing reactive diluent to 30–40 wt%. This phenomenon can be attributed to combined effects of two factors: the physical interaction force via hydrogen bonds in the urethane groups and crosslink density. At the low reactive diluent load, the physical interaction force via hydrogen bonds plays the dominant role in the effect of the impact resistance which favors the improvement of the impact resistance; while at the high reactive diluent load, the crosslink density factor plays the dominant role in the effect of the impact resistance which leads to the decrease in the impact resistance. 5. Conclusion More environmental friendly UV curable polyurethane coatings were prepared using the acrylated polyester (APE) as oligomer and non-isocyanate urethane dimethacrylate as reactive diluents. The addition of the non-isocyanate urethane dimethacrylate could reduce the viscosity of the APE oligomer. The UV-cured coatings properties were dependent on the reactive diluents type and content. Crosslink density, storage and tensile modulus, pencil hardness, chemical resistance, gel content, total water absorption, and Tg increased with increasing reactive diluent content. Comparing with traditionally used ethylene glycol dimethacrylate (EGDMA) reactive diluent, these new non-isocyanate urethane dimethacrylate reactive diluents show significant improvements in impact resistance and elongation-at-break properties. This improvement in toughness is presumably the result of hydrogen bonds formation in the urethane groups. It was found that the optimum level of the non-isocyanate urethane dimethacrylate reactive diluents concentration was between 10 and 20 wt%. References [1] J. Seo, E.S. Jang, J.H. Song, S. Choi, S.B. Khan, H. Han, Preparation and properties of poly(urethane acrylate) films for ultraviolet-curable coatings, J. Appl. Polym. Sci. 118 (4) (2010) 2454–2460. [2] J. He, L. Zhou, M.D. Soucek, K.M. Wollyung, C. Wesdemiotis, UV-curable hybrid coatings based on vinylfunctionlized siloxane oligomer and acrylated polyester, J. Appl. Polym. Sci. 105 (4) (2007) 2376–2386. [3] J.P. Fouassier, X. Allonas, J. Lalevée, C. Dietlin, Photoinitiators for Free Radical Polymerization Reactions, John Wiley & Sons, Inc., Hoboken, NJ, USA, 2010, pp. 351–419. [4] I.V. Khudyakov, K.W. Swiderski, R.W. Greer, Structure–property relations in UVcurable urethane acrylate oligomers, J. Appl. Polym. Sci. 99 (2) (2006) 489–494. [5] H.V. Patel, J.P. Raval, P.S. Patel, Preparation and performance of UV curable polyurethane coating for metal surfaces, Arch. Appl. Sci. Res. 1 (2009) 294–305. [6] B.-H. Lee, H.-J. Kim, Influence of isocyanate type of acrylated urethane oligomer and of additives on weathering of UV-cured films, Polym. Degrad. Stab. 91 (5) (2006) 1025–1035.

X. Wang, M.D. Soucek / Progress in Organic Coatings 76 (2013) 1057–1067 [7] T.Y. I˙ nan, E. Ekinci, E. Yıldız, A. Kuyulu, A. Güngör, Preparation and characterization of novel UV-curable urethane methacrylate difunctional monomers and their structure–property relationships, 1, Macromol. Chem. Phys. 202 (4) (2001) 532–540. [8] J.-Y. Kwon, H.-J. Yoo, H.-D. Kim, Effect of chemical structure on the properties of UV-cured polyurethane acrylates films, Fibers and Polymers 2 (3) (2001) 141–147. [9] W.J. Blank, Novel polyurethane polyols for waterborne and high solids coatings, Prog. Org. Coat. 20 (3–4) (1992) 235–259. [10] A. Srivastava, D. Agarwal, S. Mistry, J. Singh, UV curable polyurethane acrylate coatings for metal surfaces, Pigment Resin Technol 37 (4) (2008) 217–223. [11] A. Srivastava, A. Devendra, S. Mistry, J. Singh, Analyzing monomer performance: the effect of difunctional reactive diluents on the properties of UV-curable polyurethane acrylates, Eur. Coat. J. 4 (2008) 28–30, 32–33. [12] J.S. Young, A.R. Kannurpatti, C.N. Bowman, Effect of comonomer concentration and functionality on photopolymerization rates, mechanical properties and heterogeneity of the polymer, Macromol. Chem. Phys. 199 (6) (1998) 1043–1049. [13] C. Decker, Photoinitiated curing of multifunctional monomers, Acta Polym. 45 (5) (1994) 333–347. [14] E. Andrzejewska, Photopolymerization kinetics of multifunctional monomers, Prog. Polym. Sci. 26 (4) (2001) 605–665. [15] B. Nabeth, J.F. Gerard, J.P. Pascault, Dynamic mechanical properties of UVcurable polyurethane acrylate with various reactive diluents, J. Appl. Polym. Sci. 60 (12) (1996) 2113–2123. [16] T.Y. Lee, C.A. Guymon, E.S. Jonsson, C.E. Hoyle, The effect of monomer structure on oxygen inhibition of (meth)acrylates photopolymerization, Polymer 45 (18) (2004) 6155–6162. [17] H.J. Assumption, L.J. Mathias, Photopolymerization of urethane dimethacrylates synthesized via a non-isocyanate route, Polymer 44 (18) (2003) 5131–5136. [18] J. van Holen, V.W. Stone, H. van den Bergen, New routes to urethane acrylates, in: RadTech Europe 2005 Conference & Exhibition, RadTech Europe Association, 2005, pp. 237–242. [19] J. van Holen, V.W. Stone, H. van den Bergen, Non-NCO urethane acrylates: new process leads to new urethane(meth)acrylates which cannot be produced with isocyanate chemistry, Eur. Coat. J. (2006) 22–24, 26–27. [20] M. Koshiba, K.K.S. Hwang, S.K. Foley, D.J. Yarusso, S.L. Cooper, Properties of ultra-violet curable polyurethane acrylates, J. Mater. Sci. 17 (5) (1982) 1447–1458. [21] T.Y. Inan, E. Ekinci, E. Yildiz, A. Kuyulu, A. Gungor, Preparation and characterization of novel UV-curable ester methacrylate crosslinking-agents and their structure–property relationships, 2, Macromol. Chem. Phys. 202 (4) (2001) 541–545. [22] S. Wu, J.D. Jorgensen, A.D. Skaja, J.P. Williams, M.D. Soucek, Effects of sulphonic and phosphonic acrylic monomers on the crosslinking of acrylic latexes with cycloaliphatic epoxide, Prog. Org. Coat. 36 (1–2) (1999) 21–33. [23] G. Rokicki, A. Piotrowska, A new route to polyurethanes from ethylene carbonate, diamines and diols, Polymer 43 (10) (2002) 2927–2935. [24] M.T. Lemon, M.S. Jones, J.W. Stansbury, Hydrogen bonding interactions in methacrylate monomers and polymers, J. Biomed. Mater. Res. A 83A (3) (2007) 734–746. [25] H. Ni, W.J. Simonsick, A.D. Skaja, J.P. Williams, M.D. Soucek, Polyurea/ polysiloxane ceramer coatings, Prog. Org. Coat. 38 (2) (2000) 97–110. [26] G.t. Brinke, F.E. Karasz, T.S. Ellis, Depression of glass transition temperatures of polymer networks by diluents, Macromolecules 16 (1983) 244–249. [27] H.-J. Chung, K.-S. Woo, S.-T. Lim, Glass transition and enthalpy relaxation of cross-linked corn starches, Carbohydr. Polym. 55 (1) (2004) 9–15.

1067

[28] C.R. Wold, M.D. Soucek, Viscoelastic and thermal properties of linseed oil-based ceramer coatings, Macromol. Chem. Phys. 201 (3) (2000) 382–392. ´ A. Guo, R. Fuller, Thermal stability of polyurethanes based [29] I. Javni, Z.S. Petrovic, on vegetable oils, J. Appl. Polym. Sci. 77 (8) (2000) 1723–1734. [30] R. Schwalm, UV coaTings: Basics, Recent Developments and New Applications, Elsevier, Amsterdam/Oxford/Boston, 2007. [31] J.-S. Choi, J. Seo, S.B. Khan, E.S. Jang, H. Han, Effect of acrylic acid on the physical properties of UV-cured poly(urethane acrylate-co-acrylic acid) films for metal coating, Prog. Org. Coat. 71 (2011) 110–116. [32] L.B. Manfredi, A.N. Fraga, A. Vazquez, Influence of the network structure and void content on hygrothermal stability of resol resin modified with epoxyamine, J. Appl. Polym. Sci. 102 (1) (2006) 588–597. [33] I. Merdas, A. Tcharkhtchi, F. Thominette, J. Verdu, K. Dean, W. Cook, Water absorption by uncrosslinked polymers, networks and IPNs having medium to high polarity, Polymer 43 (17) (2002) 4619–4625. [34] G. Teng, M.D. Soucek, X.F. Yang, D.E. Tallman, Effect of the addition mode of cycloaliphatic diepoxide on the morphology and film properties of crosslinkable core–shell latex, J. Appl. Polym. Sci. 88 (2) (2003) 245–257. [35] E.P. Pedraza, M.D. Soucek, Effect of functional monomer on the stability and film properties of thermosetting core–shell latexes, Polymer 46 (24) (2005) 11174–11185. [36] S. Wu, J.D. Jorgensen, M.D. Soucek, Synthesis of model acrylic latexes for crosslinking with cycloaliphatic diepoxides, Polymer 41 (1) (2000) 81–92. [37] C.S. Wang, X.Y. Chen, J.Q. Chen, C.G. Liu, H.F. Xie, R.S. Cheng, Synthesis and characterization of novel polyurethane acrylates based on soy polyols, J. Appl. Polym. Sci. 122 (4) (2011) 2449–2455. [38] Q.-H. Zhou, M. Li, P. Yang, Y. Gu, Effect of hydrogen bonds on structures and glass transition temperatures of maleimide-isobutene alternating copolymers: molecular dynamics simulation study, Macromol. Theory Simul. 22 (2013) 107–114. [39] A. Plante, S. Palato, O. Lebel, A. Soldera, Functionalization of molecular glasses: effect on the glass transition temperature, J. Mater. Chem. C 1 (2013) 1037–1042. [40] T.K. Kwei, The effect of hydrogen bonding on the glass transition temperatures of polymer mixtures, J. Polym. Sci. Polym. Lett. Ed. 22 (1984) 307–313. [41] E.P. Otocka, T.K. Kwei, Properties of ethylene–acrylic acid copolymers, Macromolecules 1 (3) (1968) 244–249. [42] D.K. Chattopadhyay, D.C. Webster, Thermal stability and flame retardancy of polyurethanes, Prog. Polym. Sci. 34 (2009) 1068–1133. [43] T.M. Madkour, R.A. Azzam, Non-Gaussian behavior of self-assembled thermoplastic polyurethane elastomers synthesized using two-step polymerization and investigated using constant-strain stress relaxation and molecular modeling techniques, Eur. Polym. J. 49 (2) (2013) 439–451. [44] L. Hojabri, J. Jose, A.L. Leao, L. Bouzidi, S.S. Narine, Synthesis and physical properties of lipid-based poly(ester-urethane)s, I: Effect of varying polyester segment length, Polymer 53 (2012) 3762–3771. [45] J.V. Dzunuzovic, M.V. Pergal, R. Poreba, V.V. Vodnik, B.R. Simonovic, M. Spirkova, S. Jovanovic, Analysis of dynamic mechanical, thermal and surface properties of poly(urethane-ester-siloxane) networks based on hyperbranched polyester, J. Non-Cryst. Solids 358 (23) (2012) 3161–3169. ˜ [46] A. Larranaga, J.-R. Sarasua, Effect of bioactive glass particles on the thermal degradation behaviour of medical polyesters, Polym. Degrad. Stab. 98 (3) (2013) 751–758. [47] J.V. Dzunuzovic, M.V. Pergal, R. Poreba, S. Ostojic, N. Lazic, M. Spirkova, S. Jovanovic, Studies of the thermal and mechanical properties of poly(urethanesiloxane)s cross-linked by hyperbranched polyesters, Ind. Eng. Chem. Res. 51 (2012) 10824–10832.