Revealing the Mechanical Properties of Emulsion Polymer ... - MDPI

15 downloads 0 Views 6MB Size Report
Jun 11, 2018 - Polymer Isocyanate Film in Humid Environments ..... of xPVAc-based EPI films with and without the thermal treatment, and explored the ...
polymers Article

Revealing the Mechanical Properties of Emulsion Polymer Isocyanate Film in Humid Environments Jing Guo 1,2 , Hongjiu Hu 1,2, * 1

2

*

ID

, Kefeng Zhang 1,2 , Yaolong He 1,2 and Xingming Guo 1,2

Shanghai Institute of Applied Mathematics and Mechanics, Shanghai University, Shanghai 200072, China; [email protected] (J.G.); [email protected] (K.Z.); [email protected] (Y.H.); [email protected] (X.G.) Shanghai Key Laboratory of Mechanics in Energy Engineering, Shanghai 200072, China Correspondence: [email protected]; Tel./Fax: +86-21-56-338-345

Received: 6 May 2018; Accepted: 7 June 2018; Published: 11 June 2018

 

Abstract: Knowledge of the mechanical behaviors of polymer film in humid environments is of great significance for predicting the long-term performance of emulsion polymer isocyanate (EPI) as a high-performance wood adhesive. A tri-copolymer latex was cross-linked by the general polymeric methylene diisocyanate (p-MDI) and aqueous emulsified isocyanate (EMDI) at different loadings for preparing EPI. Furthermore, a series of uniaxial tension tests under different relative humidity (RH) were carried out on cured EPI samples before and after post-curing treatment, and the corresponding chemical structure, as well as the microstructure of polymers, was investigated in detail. In addition, a constitutive equation was formulated to calculate the viscoelastic characteristics of the adhesive layer. The results indicate that the EPI films reveal various kinds of intrinsic deformation as RH increases, and the tensile rupture stress and stiffness would obviously decrease, even at cross-linker weight ratios of up to 20%. Furthermore, the moisture resistance could be markedly improved by increasing the isocyanate content and post-cure. Importantly, EMDI-cross-linked film not only exhibits much better mechanical properties than that containing p-MDI at 0–80% RH, but is also more sensitive to post-cure. Finally, the derived viscoelastic model could efficiently track moisture-dependent stress-strain curves of EPI films, and the obtained relaxation time further reveals the influence mechanism of isocyanate and post-cure on the mechanical response of the cured polymer under moist conditions. Keywords: emulsion polymer isocyanate; tensile properties; post-cure treatment; relative humidity; viscoelastic constitutive equation

1. Introduction Emulsion polymer isocyanates (EPI) are composed of a water-based polymer emulsion and an isocyanate cross-linker. There is a remarkably strong interest in the application of EPI adhesives in solid wood processing and wood-based panel production for furniture manufacturing. The outstanding characteristics (high adhesion strength, good heat resistance, and environmental- soundness) make EPI one of the most promising choices for adhesives. It should also be noted that EPI-bonded wood products comprised of structural-composite lumber and cross-laminated timber are commonly used for buildings and bridges in Europe and North America [1]. As such, during their service life, the bond quality is influenced by the humid outdoor climate. Therefore, the moisture resistance of EPI has been extensively investigated in the past few decades. To improve the bonding performance of EPI in different humid conditions, considerable efforts have been made to optimize the adhesive formulation. For example, Hu et al. investigated the warm and boiling water resistances of EPI adhesives for bonding rose wood, in which vinyl acetate Polymers 2018, 10, 652; doi:10.3390/polym10060652

www.mdpi.com/journal/polymers

Polymers 2018, 10, 652

2 of 18

homopolymer and copolymer emulsion were cross-linked by three types of polymeric methylene diisocyanate (p-MDI) respectively. It was found that aqueous emulsified isocyanate (EMDI) exhibits better cross-linking performance, followed by p-MDI mixed with organic solvent. The compression shear strength and percent wood failure are improved as the polyvinyl alcohol (PVOH) content increases, but excess PVOH impairs resistance to warm/boiling water [2]. The amount of isocyanates in the EPI will also notably affect bond performances. Krystiofiak et al. showed that the moisture resistance of the glue line increases with increasing loading of cross-linker [3]. The choice of emulsion will also influence the mechanical properties of EPI joints under both dry and wet states. With the addition of PVOH and nano-CaCO3 to a whey protein isolate solution, Guo et al. developed a novel EPI adhesive which has wet compression shear strength which surpass the JIS K6806 standards [4]. The preparation of wood surfaces is also of high importance regarding the moisture-related durability of EPI-glued timber products. When evaluating the water resistance of EPI for bonding yellow poplar wood acetylated to different weight percent gains, Vick and Rowell found that wet shear strength values significantly increase at higher levels of acetylation when compared to the untreated one, while wet wood failures decrease markedly as acetylation levels increase [5]. In another study, Knorz et al. carried out tensile shear and delamination tests for EPI bonded ash assemblies after exposure to moisture, in which the wood components were prepared using peripheral planing, sanding and face milling respectively. The different surfacing techniques led to markedly different bonding surfaces, notably with regard to cell damage and the extent of fibrillation. The face-milled surfaces provided the highest moisture resistance due to a more homogenous stress distribution in the bond line [6]. Compared to the great deal of attention given to the bonding performance of EPI adhesives [2,4,7–10], only a few studies exist in which mechanical response of the cured polymer films is analyzed. One is a pioneering investigation made by Taki et al. [11,12]. They reported the relationship between the mechanical properties of cured adhesives and bond strength of EPI with different cross-linking densities over a wide temperature range. In another experiment, Umemura et al. performed dynamic mechanical analysis (DMA) measurements on cured EPI films, and observed a very small loss in storage modulus from 20 to 70 ◦ C [13]. DMA revealed that the post curing resulted in wide variations in the viscoelastic behaviors of the polymer. Although the residual isocyanate would have been reduced by post-cure treatments, a complete consumption of isocyanate could not be achieved [14]. That it is to say, the concentration of cross-links in cured adhesives and the corresponding joint durability may further evolve during the servicing environment. However, it is still not clear how environmental moisture impacts the mechanical stability of cured EPI film. Knowledge of the deformation and failure behavior of the adhesive thin film under humid conditions not only allows for a better understanding of the bonding strength of EPI, but is also the basis of adhesive design and application for outdoor timber structures. Up to now, no previous studies have paid close attention to this issue. As an essential mechanism in the deformation of polymer, yielding behavior may control fracture toughness by its effects on the crazing and micro-crack propagation. It is important to understand, on a molecular scale, how the mechanical response of polymeric materials is dependent upon chemical structure, strain rate, and temperature. Therefore, considerable efforts have been made to study the connections between the yielding and nature of the polymer chain motions induced by temperature [15–18]. The similar effects of temperature and relative humidity on glassy stability [19–21] and durability regarding hydrothermal degradation [22] are well recognized. However, for various reasons, few studies in the literature exist on the molecular analysis of yielding behavior and the relaxation mechanism of a glassy polymer in a mechanical way which takes RH variation into account. In this study, we focus on exploring the role of moisture on the mechanical response of EPI film, aiming to provide a simple and accurate basis for theoretical modeling and applications of EPI-bonded wood products. A tri-copolymer latex was cross-linked by the polyisocyanate at different types and loadings for preparing EPI. In addition, a series of uniaxial tension tests under different RH were

Polymers 2018, 10, 652

3 of 18

performed to determine the mechanical properties of EPI cured samples before and after post-cure treatment, and the glass transition relative humidity (RHg ), chemical structure, and microstructure of polymers were checked using DMA RH scans, Fourier transform infrared spectroscopy (FTIR) and optical microscopy respectively. Moreover, a constitutive equation was formulated to quantitatively describe the moisture-effected stress-strain curves of EPI films, and the obtained relaxation time was used to unveil the effect mechanism of cross-linker and post-cure on the mechanical properties of the polymer in humid environments. 2. Experiment 2.1. Materials (1)

(2)

Main component: vinyl acetate/butyl acrylate/hydroxypropyl methacrylate tri-copolymer latex (xPVAc, GCLOCK2000, Gclock Adhesive Company, Ganzhou, China), containing approximately 30% (wt.) of CaCO3 , as well as 10% (wt.) of PVOH as a protective colloid and the main provider of hydroxyl groups. The aqueous vinyl latex was synthesized by Gclock Adhesive Company (Ganzhou, China) and was used as received. It had a solid content of 52.5%, viscosity of 7500 mPa·s (at 25 ◦ C), and pH of 6.3. Cross-linkers: EMDI (Rubinate 9259) and p-MDI (Rubinate 5005), supplied by Huntsman Polyurethanes (China) Ltd. (Shanghai, China), were employed as received. The characteristics of both cross-linkers are shown in Table 1. Table 1. Characteristics of polymer isocyanates. Cross-Linker

Solid Content, %

NCO, %

Functionality

Modification Method

Rubinate 9259

100

30.6

2.7

Aqueous emulsifiable

Rubinate 5005

100

30.5–32.5

2.6–2.7

Standard industrial

2.2. Preparation of Samples EPI adhesives were prepared by adding the cross-linker to the main component at cross-linker weight ratios of 0%, 5%, 10%, 15%, and 20%, respectively. The two components, after metering, were mixed for 30 min by an electric agitator, and degassed for 30 s. The films were prepared on Teflon-coated glass plates and dried in a constant temperature and humidity chamber at 40 ◦ C for 48 h to evaporate the water. Subsequently, some of the samples were placed in the chamber at 140 ◦ C for 24 h to further remove residual isocyanate groups in EPI. Finally, all the cured films, with dimensions of 20 mm × 5 mm × 0.04 mm, were kept in a desiccator with freshly dried silica gel at ambient temperature as samples for testing. 2.3. Test Methods The mechanical measurements were carried out by a film/fiber tension clamp on a DMA (TA Q800, TA Instruments-Waters LLC, New Castle, DE, USA) with a DMA-RH Accessory, where the RH regulation of the sample chamber was realized by controlling pure N2 and a saturated water vapor mixture of different proportions. The humidity accuracy was ±3% at 0–90% RH and ±5% RH above 90% RH. 2.3.1. Uniaxial Tension The mechanical tests were conducted in different RHs at ambient temperature. The samples were first equilibrated at 0% RH for 60 min, and then the relative humidity in the testing chamber was adjusted to fixed values of 0%, 20%, 40%, 60%, and 80% RH. After 90 min equilibration, uniaxial tension was carried out at a strain rate of 2% under these various constant RHs.

Polymers 2018, 10, 652

4 of 18

2.3.2. DMA RH Scans DMA tests were carried out under strain mode at a frequency of 1 Hz and amplitude of 20 m. For the RH scans under a controlled temperature, specimens were quickly dried in a measuring chamber at 0% RH, 25 ◦ C for about 60 min, after which RH scans were performed isothermally at the rate of 1.0% RH/min, with the humidity ranging from 0% RH to 90% RH. 2.3.3. FTIR FTIR spectra were measured with an Avatar Nicolet 380 spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) using the KBr tablet method. Each sample was scanned 16 times with a resolution of 4 cm−1 between 400 and 4000 cm−1 . The obtained spectra were normalized, with the peak height of the C–H stretching band at approximately 2922 cm−1 . 2.3.4. Optical Microscopy The surface morphology of EPI films was observed on an optical microscope (Zeiss Smart zoom 5, Carl Zeiss, Jena, Germany), and the eyepiece used for image acquisition tasks was PlanApo D 5x/0.3 FWD 30 mm. These image acquisition tasks were carried out by graphic stitching under free examination modes. Mixed illumination was also used in the tests and the magnification 200×. The obtained optical microscopy images were then transformed into binary images using a MATLAB program based on the region growing method. 3. Results and Discussion 3.1. Effects of Types and Loadings of Polyisocyanate The mechanical behaviors of the bonding layer are crucial to the comprehensive understanding of the complicated deformation response of EPI adhesive in a moist environment. Therefore, a series of uniaxial tension tests under different RHs (0%, 20%, 40%, 60% and 80%) were carried out on the EPI cured samples, which were cross-linked by EMDI or p-MDI with 0%, 5%, 10%, 15% and 20% of loadings, respectively. Figure 1a shows the mechanical behavior for dry EPI specimens. As the isocyanate content increases, the engineering stress-strain curves of polymer film have noticeably changed. For the low cross-linker loadings (0–5%), the raised strain causes the stress to quickly enlarge, followed by a relatively long period of minimal stress increase until rupture. With the addition of more isocyanate, the level of stress increases, and then plateau gradually shortens and disappears. This indicates that the cured films of aqueous vinyl emulsion are gradually transformed from a thermoplastic polymer to thermoset resin owing to the cross-link reaction between the hydroxyl groups (–OH) in polymer and isocyanate groups (–NCO). Moreover, compared to the standard version, the aqueous emulsifiable polymeric isocyanate exhibits better cross-linking effects, where the EPI films have relatively higher strength and stiffness. Under high humidity conditions, due to the penetration of water molecules, scissioning of chain segments and the destruction of cross-links occur, resulting in the deterioration of the mechanical properties of the polymer. Thus, the tensile strength of the EPI films is greatly weakened, and the fracture strain increases considerably, as shown in Figure 1b. Moreover, it can be observed that the aqueous polymer would undergo various kinds of intrinsic deformation as the amount of isocyanate increases. Similar to the main component (xPVAc), the EPI film with 5% cross-linker displays the typical behavior of a soft elastomer, of which the fracture happens at the point of very high strain value. The values of ε f for xPVAc/EMDI and xPVAc/p-MDI are 240% and 350% at 80% RH, respectively, which are separately 12 times and 15 times of those when the compounds are in a dry state. However, within the range of higher cross-linker loading (15–20%), the polymers are neither rubber-like nor glassy, where the stress steeply goes up to the fracture point corresponding to a strain of 18–34%. It is much less than the value (240–500%) at low isocyanate addition levels (0–5%). At an intermediate

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW Polymers 2018, 10, x FOR PEER REVIEW

5 of 18 5 of 18 5 of 18

hyperelastic material with which is by farp-MDI larger bears than athat of EMDI. The influences cross-linker loading (10%), the εspecimen cross-linked notably hyperelastic material of f of 125%, hyperelastic material with of than 125%, which is farThe larger than that of EMDI. The influences of ε f with ε of 125%, which is far larger that of EMDI. influences of cross-linker type and loading f cross-linker type and loading on the moisture resistance of cured film are further compared in on the moisture resistance of cured film are further compared in Figure 2. cross-linker type and loading on the moisture resistance of cured film are further compared in Figure 2. Figure 2.

18 16

(a) (a)

20% 20%

15% 15%

16 14

Stress(MPa) Stress(MPa)

6 4 4 2

15% 15%

5% 0% 0%

10% 10%

12 10

5%

Untreated xPVAc/EMDI Untreated xPVAc/EMDI

20% 20%

14 12

10%

12 10 10 8 8 6

18 16 16 14

10%

14 12

20 20 18

Stress(MPa) Stress(MPa)

20 20 18

5% 5%

10 8 8 6 6 4 4 2

0%

0%

Untreated xPVAc/p-MDI

Stress(MPa) Stress(MPa)

Stress(MPa) Stress(MPa)

Untreated xPVAc/p-MDI 2 0 2 0 0 1 2 3 4 5 6 7 8 9 10 15 20 25 30 0 1 2 3 4 5 6 7 8 9 10 15 20 25 30 0 0 0 1 2 3 4 5 6 Strain(%) 7 8 9 10 15 20 25 30 0 1 2 3 4 5 6 Strain(%) 7 8 9 10 15 20 25 30 8 8 Strain(%) Strain(%) 20% Untreated xPVAc/p-MDI Untreated xPVAc/EMDI (b) 8 8 7 7 20% Untreated xPVAc/p-MDI Untreated xPVAc/EMDI (b) 7 7 20% 6 6 15% 20% 6 6 15% 15% 5 5 10% 10% 15% 5 5 10% 10% 4 4 5% 5% 4 4 3 5% 3 5% 3 3 2 2 2 2 0% 1 1 0% 0% 1 1 0 0 0% 500 0 5 10 15 20 25 30 35 250 0 10 20 30 40 50 60 250 500 0 0 0 5 10 15 20 25 30 35 250 500 Strain(%) 0 10 20 30 40 50 60 250 500 Strain(%)

Strain(%)

Strain(%)

Figure 1. Effect of cross-linker on the tensile curves of cured EPI films at (a) 0% RH and (b) 80% RH Figure 1. Effect of cross-linker on the tensile curves of cured EPI films at (a) 0% RH and (b) 80% RH Figure Effect of cross-linker thecalculated tensile curves of cured (2)). EPI films at (a) 0% RH and (b) 80% RH (dot: 1. experimental data, solidonline: by Equation (dot: experimental data, solid line: calculated by Equation (2)). (dot: experimental data, solid line: calculated by Equation (2)).

Figure 2. Effect of EMDI (red line) and p-MDI (magenta line) on the rupture stress of EPI film at Figure 2. Effect of EMDI (red line) and p-MDI (magenta line) on the rupture stress of EPI film at different relative humidity. Figure 2. Effect of EMDI (red line) and p-MDI (magenta line) on the rupture stress of EPI film at different relative humidity. different relative humidity.

After EPI is applied on the substrate, as H2O is evaporated into the environment and reacts with After EPI is applied the substrate, as H2O(1), is evaporated into theinenvironment reacts with the isocyanate groups on as shown in Equation the water content the adhesiveand layer decreases, After EPI isgroups applied on the substrate, as H(1), evaporated into the environment and reacts with 2 O is theand isocyanate as shown in Equation the water content in the adhesive layer decreases, tiny particles of polymer gradually form a film; while –OH cross-links with –NCO, the result the isocyanate groups as shown in Equation the water content in the adhesive layer decreases, and tinythe particles of polymer form(1), aand film; whilechemical –OH cross-links with –NCO, the result forms urethane bond [2].gradually These reactions related structures of the EPI films during forms the urethane bond [2]. These reactions and related chemical structures of the EPI films during curing the process are illustrated in Equations (2) and (3) respectively, by which not only xPVAc curing the process are illustrated in Equations (2) and (3) respectively, by which not only xPVAc

Polymers 2018, 10, 652

6 of 18

and tiny particles of polymer gradually form a film; while –OH cross-links with –NCO, the result forms the urethane bond [2]. These reactions and related chemical structures of the EPI films during curing the process are illustrated in Equations (2) and (3) respectively, by which not only xPVAc Polymers 2018, 10, x FOR PEER REVIEW 6 of 18 (vinylPolymers acetate/butyl acrylate/hydroxypropyl methacrylate tri-copolymer latex), but also6 PVOH in 2018, 10, x FOR PEER REVIEW of 18 the main component, can be cross-linked by the polyisocyanate to increase their molecular weights (vinyl acetate/butyl acrylate/hydroxypropyl methacrylate tri-copolymer latex), but also PVOH in the (vinylcomponent, acetate/butyl acrylate/hydroxypropyl tri-copolymer latex), but in also PVOH the the and establish the stable crosslinking after fabrication. As indicated Figure 2,inand with main can be cross-linkednetworks by the methacrylate polyisocyanate to increase their molecular weights main component, can be cross-linked by the polyisocyanate to increase their molecular weights and aid ofestablish polymeric isocyanate, the rupture (σb ) remarkably improves the adhesive the stable crosslinking networksstress after fabrication. As indicated in Figure 2, with the film aid ofunder establish the stable crosslinking networks after fabrication. As indicated inthe Figure 2, with the under of polymeric isocyanate, the rupture stress ( ) remarkably improves adhesive film σ different relative humidity conditions. The magnitude of σb at 0% RH is 46.4% higher aid in the EPI b polymeric isocyanate, the rupture stress ( σ b ) remarkably improves the adhesive film under film with 5.0 relative wt % p-MDI compared to the xPVAcofsample. AsRH theisweight ratio of cross-linker different humidity conditions. The neat magnitude 46.4% higher in the EPI σ b at 0% different relative humiditydifference conditions. magnitude at main 0% RH is 46.4% higher in thetoEPI σ b its increases to 20%, relative inThe σb between EPIofsample. and 122.2%. film with 5.0 the wt % p-MDI compared to the neat xPVAc As thecomponent weight ratioamplifies of cross-linker film with 5.0 wt % p-MDI compared to the neat xPVAc sample. As the weight ratio of cross-linker Noticeably, with increasing EMDI or p-MDI the strength of main dry film continuously increases. increases to 20%, the relative difference in content, EPI and its component amplifies to σ b between increases to to 20%, the relative differenceisocyanate in σ b between EPI and its performance main component amplifies to EPI This is different the effect of polymeric on the bonding of xPVAc-based 122.2%. Noticeably, with increasing EMDI or p-MDI content, the strength of dry film continuously 122.2%.temperature, Noticeably, with increasing EMDI p-MDI content, theweight strength of dry film continuously at ambient which then descends as the of cross-linker increases increases. This is different tofirst the rises effect ofor polymeric isocyanate on theratio bonding performance of increases. This is different to the effect of polymeric isocyanate on the bonding performance of the EPI This at ambient temperature, firstthat risesmore then residual descends stress as the has weight ratiofrom of fromxPVAc-based 0 to 20.0% [2]. may be ascribed towhich the fact arisen xPVAc-based EPI at ambient temperature, which first rises then descends as the weight ratio of cross-linker from 0 todue 20.0% may behigher ascribed to the rate fact that more residual stress shrinkage of theincreases adhesive layer to [2]. theThis relatively curing as the EPI is coated on the cross-linker increases from 0 to 20.0% [2]. This maydue be to ascribed to the fact thatcuring more residual stress has arisen from the shrinkage of the adhesive layer the relatively higher rate as the EPI surface of the dry wood specimen, where a great deal of water molecules are rapidly absorbed by the has arisenon from the shrinkage of the adhesive layerwhere due toathe relatively curing rateare as rapidly the EPI is coated the surface the dry wood specimen, great deal of higher water molecules adherend. Meanwhile, notof only noncovalent interactions, such as electrostatic interaction, hydrogen is coated on the surface of the dry wood specimen, where a great deal of water molecules are rapidly absorbed by the adherend. Meanwhile, not only noncovalent interactions, such as electrostatic bonding, hydrophobic attraction, and van der Waals interaction [23], but also chemical crosslinking absorbed byhydrogen the adherend. Meanwhile, notattraction, only noncovalent interactions, such as[23], electrostatic interaction, bonding, hydrophobic and van der Waals interaction but also occurchemical on the interfaces between wood and EPI film. However, if the evaporation rate of water in the interaction, hydrogen bonding, hydrophobic attraction, and van der Waals interaction [23], crosslinking occur on the interfaces between wood and EPI film. However,but if also the chemical crosslinking occur on the interfaces between wood and EPI film. However, if the EPI adhesive is too low, the coalescence of particles in the emulsion would occur after the chemical evaporation rate of water in the EPI adhesive is too low, the coalescence of particles in the emulsion evaporation ofthe water in thecross-link EPI adhesive is toothe low, the coalescence of particles in theConsequently, emulsion cross-link between the isocyanate group and active hydrogen in the main component. would occur rate after chemical between isocyanate group and active hydrogen in the would occur after the chemical cross-link between the isocyanate group and active hydrogen inmay the the crosslinking reactions on the surface of the particles mayon impede the formation of a continuous main component. Consequently, the crosslinking reactions the surface of the particles main component. Consequently, the crosslinking reactions on the surface of the particles may the formation of a continuous film.ofTherefore, it is important in the design of the robust EPIspeed film. impede Therefore, it is important in the design the robust EPI bond-line to control the water-loss impede the formation ofwater-loss a continuous film. Therefore, it is important in the design of the robust and EPI bond-line to control the speed and the crosslinking reactions well with the diffusion and the crosslinking reactions well with the diffusion and coalescence of tiny particles, film formation, bond-line to control the water-loss speed and the crosslinking reactions well with the diffusion and coalescence of tiny particles, film formation, andsurface emulsion into the porosity of the surface and emulsion migration into the porosity of the of migration the adherend. coalescence of tiny particles, film formation, and emulsion migration into the porosity of the surface of the adherend. of the adherend.

(1) (1)

(2) (2)

(1)

(2)

Polymers 2018, 10, 652 Polymers x FOR PEER REVIEW Polymers 2018,2018, 10, x10, FOR PEER REVIEW

7 of 18 7 of 187 of 18

(3)

(3) (3)

Crosslink density,ν(x10-2,mol/cm3-2)

Crosslink density,ν(x10 ,mol/cm3)

When water-based polymer film is subjected to moisture, because of the strong attraction of the When water-based polymer film subjected to because of attraction of the Whenmolecules water-based polymer filmisischains subjected to moisture, moisture, because of the the strong strong attraction water to the amorphous of aqueous polymer, the electrostatic forces create aof the water molecules to the amorphous chains of aqueous polymer, the electrostatic forces create a mutually water molecules to the amorphous chains of chains aqueous polymer, the electrostatic forces create a mutually exclusive system, forcing molecular to become a looser framework, and leaving exclusive system, forcing molecular chains to become a looser framework, and leaving extra free mutually exclusive system, forcing molecular chains to become a looser framework, and leaving extra free volume. Therefore, the small molecules can penetrate into the non-crystalline zone and volume. Therefore, smallof molecules canmolecules penetrate into the zone and the local the localthe density the which results inpenetrate the non-crystalline strength of the the cured film reduce decreasing extrareduce free volume. Therefore, thesystem, small can into non-crystalline zone and density of the system, which results in the strength of the cured film decreasing through an increase in through an increase in the relative humidity at isothermal (seeof Figure 2). As anticipated, reduce the local density of the system, which results in conditions the strength the cured film decreasing the relative humidity at isothermal conditions (see Figure 2). As anticipated, polymeric isocyanate is polymeric isocyanate is relative helpful in alleviating plasticization. Furthermore, was found the through an increase in the humidity at isothermal conditions (see itFigure 2). Asthat anticipated, moisture resistance of film isin strongly dependent upon the cross-linker content. That theof more helpful in alleviating plasticization. Furthermore, it was found that the moisture EPI polymeric isocyanate is EPI helpful alleviating plasticization. Furthermore, itresistance was is,found thatfilm the isocyanate that is added, the higher the rupture stress of the cured filmisocyanate in humid conditions. To the is strongly dependent upon the cross-linker content. That is, the more that is added, moisture resistance of EPI film is strongly dependent upon the cross-linker content. That is, the more explore the effect of the content on the moisture resistance of thethe EPI films, of the cross-link higher the rupture ofcross-linker the cured humid conditions. explore cross-linker isocyanate that is stress added, the higherfilm theinrupture stress of theTocured film ineffect humidthe conditions. To density was used to describe the degree of cross-linking between the isocyanate groups and content on the moisture resistance of the EPI films, the cross-link density was used to describe the explore the effect of the cross-linker content on the moisture resistance EPI films,bonds the cross-link functional groups in xPVAc. The cross-link density refers to the numberof ofthe cross-linked per degree ofwas cross-linking between the isocyanate groups and functional groups in xPVAc. The cross-link density used to describe the degree of cross-linking between the isocyanate groups unit volume, and can be calculated using the Young’s modulus from the stress-strain test [24]. The and density refers to the thein number ofThe cross-linked bonds per unit to volume, and can using the functional groups xPVAc.density cross-link density refers the number of be cross-linked per evolution of crosslink as a function of the isocyanate content is shown incalculated Figure 3. bonds Young’s modulus from the stress-strain test [24]. The evolution of the crosslink density as a function of unit volume, and can be calculated using the Young’s modulus from the stress-strain test [24]. The 3.0 the isocyanate content is shown in as Figure 3. evolution of the crosslink density a function of the isocyanate content is shown in Figure 3. 2.5

3.0 2.0

2.5

xPVAc/EMDI

1.5

2.0

xPVAc/p-MDI

1.0

xPVAc/EMDI

1.5

0.5

1.0 0.0

xPVAc/p-MDI

5

0.5

10 15 Cross-linker loading (%)

20

Figure 3. Crosslink density of EPI films at 80% RH with respect to crosslinker loading. 0.0

5

10

15

20

As seen in Figure 3, as the isocyanate content increases, the crosslink density rises at an almost Cross-linker loading (%) uniform rate. Importantly, the cross-linked structure is a very powerful barrier to the penetration of Figure 3. Crosslink density of EPI films at 80% RH with respect to crosslinker loading. water molecules: the greater the crosslink density, theRH greater the difficult of penetration of water Figure 3. Crosslink density of EPI films at 80% with respect to crosslinker loading. molecules. A highly cross-linked structure can also cause massive restrictions on chain movement, As seen in Figure 3, relative as the isocyanate content increases, density risesreduced, at an almost which means that the content of amorphous chains inthe the crosslink EPI samples is greatly As inImportantly, Figure 3, as the the isocyanate content increases, crosslink density at water an almost uniform rate.the cross-linked structure is a very powerful barrier to rises the penetration of andseen that free volume among the macromolecular chains isthe also decreased. Because fewer uniform rate. Importantly, the cross-linked structure is a very powerful barrier to the penetration molecules can penetrate into samples with a higher cross-linker content, their strength in a humid water molecules: the greater the crosslink density, the greater the difficult of penetration of water of water molecules: greater theat crosslink density, the greater the restrictions difficult of penetration of water environment can the be maintained a higher can value. molecules. A highly cross-linked structure also cause massive on chain movement, Figure 2 shows that xPVAc/EMDI films have much better moisture resistance than molecules. A highly cross-linked structure can also cause massive onischain movement, which means that the relative content of amorphous chains in therestrictions EPI samples greatly reduced, xPVAc/p-MDI films with the content same isocyanate, and that chains the difference tends to becomeismore obvious which means that the relative of amorphous in the EPI samples greatly reduced, and that the free volume among the macromolecular chains is also decreased. Because fewer water as the cross-linker amount increases. This result may be interpreted asdecreased. follows: OnBecause the one hand, aswater and that the free volume among the macromolecular chains is also molecules can penetrate into samples with a higher cross-linker content, their strengthfewer in a humid shown in Figure 3, the crosslink density of xPVAc/EMDI films is greater than that of xPVAc/p-MDI molecules cancan penetrate into samples with a higher cross-linker content, their strength in a humid environment be maintained at a higher value. films throughout the cross-linker content range tested. Moreover, as the cross-linker addition environment be maintained at a higher value. Figure can 2theshows that xPVAc/EMDI films have much better moisture increases, gap in cross-link density between the two types of films gradually widens.resistance Therefore, than Figure 2 shows that xPVAc/EMDI films have much better moisture resistance than xPVAc/p-MDI xPVAc/p-MDI films with the same isocyanate, and that the difference tends to become more obvious the water resistance of xPVAc/EMDI films is better. On the other hand, as the more non-polar films with the same isocyanate, and that the difference tends to become more obvious as the cross-linker as the cross-linker amount This result may interpreted follows: On the one hand, as cross-linker, p-MDI in theincreases. form of small particles is notbe easy to disperseas into the polar aqueous main amount increases. This result may be interpreted as follows: On the one hand, as shown Figure 3, polymer component, and this situation exacerbated at higher isocyanate ratios. In in EMDI, shown in Figure 3, the crosslink density isoffurther xPVAc/EMDI films is greater than that of xPVAc/p-MDI the crosslink xPVAc/EMDI films is greater than Moreover, that of xPVAc/p-MDI films throughout where thedensity polymeric isocyanate is content modified with non-ionic surfactants, the the interface between the films throughout theof cross-linker range tested. as cross-linker addition hydrophilic and lyophilic groups in the EPI adhesive is optimized. Thus, EMDI will facilitate the the cross-linker content range tested. Moreover, as the cross-linker addition increases, the gap in increases, the gap in cross-link density between the two types of films gradually widens. Therefore, dispersion of the isocyanate into types xPVAcofand PVOH, and thus, enhance the contact and cross-link density between the two films gradually widens. Therefore, theprobability water resistance of

the water resistance of xPVAc/EMDI films is better. On the other hand, as the more non-polar xPVAc/EMDI films isinbetter. On of thesmall otherparticles hand, asisthe more cross-linker, p-MDI in the form cross-linker, p-MDI the form not easynon-polar to disperse into the polar aqueous main of small particles is not easy to disperse into the polar aqueous main polymer component, and this polymer component, and this situation is further exacerbated at higher isocyanate ratios. In EMDI, situation is further exacerbated at higher isocyanate ratios. In EMDI, wherethe the interface polymericbetween isocyanate where the polymeric isocyanate is modified with non-ionic surfactants, the hydrophilic and lyophilic groups in the EPI adhesive is optimized. Thus, EMDI will facilitate the dispersion of the isocyanate into xPVAc and PVOH, and thus, enhance the contact probability and

Polymers 2018, 10, 652

8 of 18

is modified with non-ionic surfactants, the interface between the hydrophilic and lyophilic groups in the EPI 2018, adhesive is optimized. Thus, EMDI will facilitate the dispersion of the isocyanate into xPVAc Polymers 10, x FOR PEER REVIEW 8 of 18 and PVOH, and thus, enhance the contact probability and cross-linking reaction between the –OH of the aqueous polymer –NCOthe in the cross-linker. Consequently, mechanical of the cross-linking reactionand between –OH of the aqueous polymerthe and –NCO in properties the cross-linker. cured film are improved by the higher crosslink density through its effect on the molecular mobility Consequently, the mechanical properties of the cured film are improved by the higher crosslink and the glass transition relative density through its effect on thehumidity. molecular mobility and the glass transition relative humidity. Analogous to the glass transition Analogous to the glass transition induced induced by by temperature temperature change, change, external external relative relative humidity humidity variation variation and and the the resulting resulting change change in in water water content content can can also also lead lead to to aa remarked remarked viscoelastic viscoelastic state state change change of of the the EPI EPI film, film, as as seen seen in in Figure Figure 4. 4. With With the the loading loading of of cross-linker cross-linker varying from 5 to 20%, the the introduction introduction of of isocyanate isocyanate group group crosslinks crosslinks increased increased the the width width of of the the RH-dependent RH-dependent tanδ, tanδ, and and decreased decreasedthe thepeak peak height, height, suggesting suggesting that that the the molecular molecular weight weight increases increases and and the the molecular molecular weight weight distribution cancan be clearly observed that there critical denoted distributionbroadens. broadens.It It be clearly observed that exists there the exists the relative critical humidity relative humidity as RH , corresponding to the peak of loss tangent. Admittedly, the addition of polyisocyanate denoted as RHg, corresponding to the peak of loss tangent. Admittedly, the additionwas of g helpful in increasing the glass humidity, greater constraint chain polyisocyanate was helpful in transition increasing relative the glass transitioncausing relativea humidity, causing on a greater movement andchain a higher crosslink degree. Accordingly, moreAccordingly, isocyanate added, the higher the constraint on movement and a higher crosslink the degree. the more isocyanate RH anticipated, theRH RHg.g As value of xPVAc/EMDI higherofthan that of xPVAc/p-MDI at thethat same added, higher the anticipated, the RHis g value xPVAc/EMDI is higher than of g . Asthe cross-linker content. xPVAc/p-MDI at the same cross-linker content. 0.45

0.45

0.40 0.35

0.35

0.30 0.25

0.30 0.25

0.20

0.20

0.15

0.15

0.10

0.10

0

10

20

30 40 50 60 70 Relative humidity(%)

80

90

(b)

5% 10% 15% 20%

0.40

Tanδ

Tanδ

(a)

5% 10% 15% 20%

0

10

20

30

40

50

60

70

80

90

Relative humidity(%)

Figure 4. Effect of crosslinker content on RH-dependent loss tangent (tanδ) of (a) xPVAc/EMDI and Figure 4. Effect of crosslinker content on RH-dependent loss tangent (tanδ) of (a) xPVAc/EMDI and (b) xPVAc/p-MDI. (b) xPVAc/p-MDI.

As stated above, experimental results show that an increase in relative humidity would lead to As above,ofexperimental results show thatasanyield increase in relative humidity would lead to not onlystated a decrease mechanical properties, such and rupture stress, but also of the glass not only a decrease of mechanical properties, such as yield and rupture stress, but also of the glass transition of adhesive film. One may wonder whether there exists some connection between plastic transition of adhesive film. One may wonder exists connection between plastic behavior and the segmental mobility process.whether Inspiredthere by the worksome of the Halary group [15–18], we behavior and the segmental mobility process. Inspired by the work of the Halary group [15–18], have plotted the yield stress normalized by elastic modulus ( σ y / E ) of cured film versus departure we have plotted the yield stress normalized by elastic modulus (σ y /E) of cured film versus departure RH −− RH RHgg) )ininFigure Figure55for forxPVAc/EMDI xPVAc/EMDI and and from main main mechanical mechanical relaxation relative humidity ((RH from xPVAc/p-MDI. defined as the point where the stress passed at the obvious ‘knee’ xPVAc/p-MDI. The The value valueofofσyσwas y was defined as the point where the stress passed at the obvious in the stress–strain curve (because a maximum in stress was not observed in Figure 1). It can be seen ‘knee’ in the stress–strain curve (because a maximum in stress was not observed in Figure 1). It can that almost all the plots of σy /E against RH − RHg lie on a unique master curve for xPVAc/EMDI / E against RH − RH a unique curve for be seen almost all the plots of σ yxPVAc/p-MDI g lie on film withthat different isocyanate loadings; shows a similar trend. All themaster EPI cured films xPVAc/EMDI film withdensities differentexhibit isocyanate loadings; xPVAc/p-MDI shows a σsimilar All the with various cross-link almost the same profile of evolution a function of y /E as trend. EPI − cured various densities exhibit almostbehavior the same of evolution (RH RHg ),films whichwith implies there cross-link is a correlation between the yielding andprofile segmental mobility in β-relaxation processes induced by moisture. σ the / E as a function of ( RH − RH ), which implies there is a correlation between the yielding y g

behavior and segmental mobility in the β-relaxation processes induced by moisture.

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW

9 of 18 9 of 18 2.4

2.2

2.2

2.0

2.0

1.8 5% 10% 15% 20%

1.6 1.4 1.2 -90

(b)

(a)

σ y/E (x10 -2)

σ y/E (x10 -2)

2.4

-80

-70

1.8

5% 10% 15% 20%

1.6 1.4

-60

-50

-40

RH-RHg (%)

-30

-20

-10

1.2 -90

-80

-70

-60

-50

-40

-30

-20

-10

RH-RHg (%)

Figure 5. Evolution of yield stress normalized by the elastic modulus (σy/E) versus RH-RHg for cured Figure 5. Evolution of yield stress normalized by the elastic modulus (σy /E) versus RH-RHg for cured film at different cross-linker content (a) xPVAc/EMDI and (b) xPVAc/p-MDI. film at different cross-linker content (a) xPVAc/EMDI and (b) xPVAc/p-MDI.

3.2. Effect of Post-Cure Treatment 3.2. Effect of Post-Cure Treatment Although the NCO groups can be so reactive that they react with almost all polar substances in Although the NCO groups can be so reactive that they react with almost all polar substances the wood and aqueous polymers such as water and hydroxyl groups, the isocyanate absorption in the wood and−1aqueous polymers such as water and hydroxyl groups, the isocyanate absorption band at 2270 cm −1was still detected in the spectrum of an EPI film stored for 60 days at ambient band at 2270 cm was still detected in the spectrum of an EPI film stored for 60 days at ambient temperature [14]. As a result, the residual NCO groups may further react with environmental temperature [14]. As a result, the residual NCO groups may further react with environmental moisture, moisture, leading to the presence of minor defects in the adhesive layer due to carbon dioxide (CO2) leading to the presence of minor defects in the adhesive layer due to carbon dioxide (CO2 ) gas arising gas arising from this reaction, as shown in Equation (1). Therefore, to improve the durability from this reaction, as shown in Equation (1). Therefore, to improve the durability performance of performance of EPI-bonded structures under operating conditions, reducing the amount of EPI-bonded structures under operating conditions, reducing the amount of unreacted isocyanate unreacted isocyanate in the glue-line is essential. Importantly, the post-curing treatment, i.e., ◦by in the glue-line is essential. Importantly, the post-curing treatment, i.e., by heating at 120–180 C, heating at 120–180 °C, could decrease the content of residual NCO groups in an EPI film sample, could decrease the content of residual NCO groups in an EPI film sample, thereby contributing to thereby contributing to the higher dynamic storage modulus [14]. How the moisture resistance of the higher dynamic storage modulus [14]. How the moisture resistance of EPI films with different EPI films with different cross-linker types and loadings would be affected, however, if this cross-linker types and loadings would be affected, however, if this additional curing procedure is additional curing procedure is added, remains unclear. To answer this question, we compared the added, remains unclear. To answer this question, we compared the moisture-dependent rupture stress moisture-dependent rupture stress of xPVAc-based EPI films with and without the thermal of xPVAc-based EPI films with and without the thermal treatment, and explored the evolution of the treatment, and explored the evolution of the porosity on the film surface, the cross-link density, and porosity on the film surface, the cross-link density, and residual isocyanate groups in the cured samples residual isocyanate groups in the cured samples with different cross-linkers. with different cross-linkers. To decrease the content of residual NCO groups in polymer film, EPI specimens were heated in To decrease the content of residual NCO groups in polymer film, EPI specimens were heated in an an oven at 140 °C for 24 h, and then subjected to the same uniaxial tensile test. The representative oven at 140 ◦ C for 24 h, and then subjected to the same uniaxial tensile test. The representative engineering stress-strain curves are shown in Figure 6. Furthermore, to compare the effect of engineering stress-strain curves are shown in Figure 6. Furthermore, to compare the effect of post-cure on the moisture resistance of EPI specimens with different cross-linker, the rupture post-cure on the moisture resistance of EPI specimens with different cross-linker, the rupture stresses stresses of xPVAc/EMDI and xPVAc/p-MDI films before and after the treatment were determined, as of xPVAc/EMDI and xPVAc/p-MDI films before and after the treatment were determined, as shown shown in Figure 7a,b, respectively. in Figure 7a,b, respectively.

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW

10 of 18 10 of 18

Figure Figure 6. 6. Representative Representativestress-strain stress-straincurves curvesofofpost-cured post-curedEPI EPIfilm filmatat(a) (a)0% 0%RH RHand and(b) (b) 80% 80% RH RH (dot: experimental data, solid line: calculated by Equation (2)). (dot: experimental data, solid line: calculated by Equation (2)).

As shown in Figure 7, in either a dry atmosphere or a humid environment, the rupture stress As shown in Figure 7, in either a dry atmosphere or a humid environment, the rupture stress ((σ EPIfilms filmscross-linked cross-linked by by both both EMDI EMDI and and p-MDI p-MDI after after post-cure post-cure is σ b) )ofofEPI is significantly significantly higher higher than than b that of untreated specimens, which indicates that the treatment can generally enhance the that of untreated specimens, which indicates that the treatment can generally enhance the mechanical mechanical properties of EPI films. Afterσheating, dry 5% filmEMD with 5% EMD increases As the properties of EPI films. After heating, film with increases 38%. As38%. the weight b of b of dry σ ratio ofratio EMDI the main increases to 20%,toit20%, would boost by approximately 51%, which weight of to EMDI to thecomponent main component increases it would boost by approximately 51%, is considerably greater greater than that of the specimen. specimen. The EPI films which is considerably than that5%ofEMD-cross-linked the 5% EMD-cross-linked The containing EPI films p-MDI alsop-MDI show the but relatively lessrelatively improvement. This observation that containing alsosimilar show trend, the similar trend, but less improvement. Thisindicates observation the treatment effectively promotes thepromotes cross-linking reaction between thebetween –OH and functional indicates that the treatment effectively the cross-linking reaction thethe –OH and the groups in groups the main and may obtain additional intramolecular crosslinking, as indicated functional incomponent, the main component, and may obtain additional intramolecular crosslinking, as in Equation (4). As stated by Wei and Haag [23], the intra-polymer interactions such as physical and indicated in Equation (4). As stated by Wei and Haag [23], the intra-polymer interactions such as chemicaland crosslinking also substantially improve the stability thestability coating.ofItthe is worth noting physical chemical can crosslinking can also substantially improveof the coating. It is that allnoting of the that treated specimens in aspecimens dry environment a monotonic as cross-linker worth all of the treated in a dry show environment showstrengthening a monotonic strengthening loading increases, indicating that indicating the post-cure not result in over-crosslinking of EPI specimens, as cross-linker loading increases, thatwould the post-cure would not result in over-crosslinking of even with 20% even polymeric isocyanate. EPI specimens, with 20% polymeric isocyanate.

Polymers 2018, 10, 652

11 of 18

Polymers 2018, 10, x FOR PEER REVIEW

11 of 18

Polymers 2018, 10, x FOR PEER REVIEW

11 of 18

Figure 7. 7. Effect Effect of of post-cure post-cure treatment treatment on on moisture-dependent moisture-dependent rupture rupture stress stressof of(a) (a)xPVAc/EMDI xPVAc/EMDI and and Figure (b) xPVAc/p-MDI. Figure 7. Effect of post-cure treatment on moisture-dependent rupture stress of (a) xPVAc/EMDI and (b) xPVAc/p-MDI. (b) xPVAc/p-MDI.

(4) (4) (4) Due to plasticization effect of moisture, the rupture stress of the post-cured films also decreases as the relative humidityeffect increases. As anticipated, heating treatment can films significantly increase Due to effect ofmoisture, moisture, therupture rupture stressofof thepost-cured post-cured films also decreases Due to plasticization plasticization of the stress the also decreases as moisture resistance of EPI specimens. Certainly, the degree of improvement depends upon the water as the relative humidity increases. As anticipated, heating treatment can significantly increase the relative humidity increases. As anticipated, heating treatment can significantly increase moisture absorption thespecimens. polymer films. At 80% RH, the withupon 5–10% show moisture resistance of EPI specimens. Certainly, the degree specimens of improvement depends uponabsorption the water resistance ofinEPI Certainly, the degree of treated improvement depends thecross-linker water increases byin20–30% in . However, with the aidtreated of a higher cross-linker content (20%), there isbya absorption the polymer RH, the specimens with 5–10% cross-linker show b films. in the polymer films. Atσ80% RH,At the80% treated specimens with 5–10% cross-linker show increases increases in with theof aid of a higher cross-linker content (20%), there isare a σwet dramatic increase. The strength value post-cured xPVAc/EMDI and 20–30% inby σb .20–30% However, the aid of awith higher cross-linker content (20%), there is axPVAc/p-MDI dramatic increase. b . However, increased by approximately 80% and 60% compared toare their untreated specimens. It The wet strength value post-cured xPVAc/EMDI and xPVAc/p-MDI increased by approximately dramatic increase. Theof wet strength valuerespectively, of post-cured xPVAc/EMDI and xPVAc/p-MDI are was and also 60% found that the treatment is to more forspecimens. the film cross-linked by EMDI p-MDI It at 80% respectively, compared theireffective untreated It was also untreated found thatthan the treatment increased by approximately 80% and 60% respectively, compared to their specimens. other loadings (5%, 10% and 15%). is more effective for the film cross-linked by EMDI than p-MDI at other loadings (5%, 10% and 15%). was also found that treatment is more effective for the film cross-linked by EMDI than p-MDI at To determine the effect mechanismofofthe thepost-cure post-cureononthe themechanical mechanical behavior EPI films, it determine effect behavior ofof EPI films, it is otherTo loadings (5%,the 10% andmechanism 15%). is necessary to systematically analyze theofchanges the on structure and composition of necessary to systematically analyze the changes theinstructure and composition of the polymer. As the To determine the effect mechanism the in post-cure the mechanical behavior ofthe EPIpolymer. films, it Asnecessary the cross-linked structure is an effective to water molecule infiltration, the cross-linked structure is an effective barrier tobarrier water molecule infiltration, the higher thehigher crosslink is to systematically analyze the changes in the structure and composition ofthe the polymer. crosslink density, thestructure more difficult theby diffusion by water molecules. The aim ofthe thehigher following density, the more difficult the diffusion water molecules. Themolecule first aim offirst the following discussion As the cross-linked is an effective barrier to water infiltration, the discussion was thethe clarification of thethe in the density ( υ ) first of xPVAc was the clarification of the changes inchanges the cross-link density (υ) of xPVAc cross-linked byfollowing EMDI by or crosslink density, more difficult diffusion by cross-link water molecules. The aim ofcross-linked the EMDI or p-MDI, before and after post-curing treatment at various isocyanate loadings. p-MDI, before post-curing at the various isocyanate loadings. discussion wasand the after clarification of thetreatment changes in cross-link density ( υ ) of xPVAc cross-linked by the effect of on density ) of of two two kinds kinds of of EPI 8 illustrates theafter effect of post-curing post-curing on the the crosslink density( υ (υ) EMDIFigure or p-MDI, before and post-curing treatment atcrosslink various isocyanate loadings. ◦ υ specimen at 80% RH. Upon 24 h of heating at 140 °C, the value of increases substantially as υ 24 h of post-curing heating at 140 C, crosslink the value density of υ increases Figure 8 illustratesUpon the effect on the ( ) of substantially two kinds of as EPIa consequence post-curing and the formation of°C, new andeffect curedsharply effect consequence ofofpost-curing the formation new crosslinking and cured specimen at 80% RH. Uponand 24 h of heating atof140 thecrosslinking value structures, of υ structures, increases substantially as a sharply enlarges as more cross-linker isImportantly, added. Importantly, the of cross-link post-cured enlarges as more cross-linker is added. the degree of degree cross-link of post-cured filmeffect with consequence of post-curing and the formation of new crosslinking structures, and ofcured film cross-linker with 15% cross-linker is noticeably than of thethe untreated one with 20%ofcross-linker. 15% is higher than that of the that untreated one with 20% cross-linker. The result sharply enlarges asnoticeably more cross-linker is higher added. Importantly, degree of cross-link post-cured The with result may be very helpful in cost-effective developing cost-effective EPI-bonded wood products, especially film 15% cross-linker is noticeably higher than that of the untreated one with 20%ascross-linker. may be very helpful in developing EPI-bonded wood products, especially the price of as the pricemay of polyisocyanate has to cost-effective approximately USD per ton products, at present.especially The result be very helpful in increased developing wood polyisocyanate has increased to approximately USD 5000 perEPI-bonded ton at 5000 present. as the price of polyisocyanate has increased to approximately USD 5000 per ton at present.

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW

12 of 18 12 of 18

untreated xPVAc/EMDI untreated xPVAc/p-MDI post-cured xPVAc/EMDI post-cured xPVAc/p-MDI

4

2

5

10

15

Cross-linker loading (%)

20

300

ν post-cured /ν untreated (%)

6

After post-cure

8

Before post-cure

Cross-link density, ν (x10 -2 mol/cm 3)

320 (a)

(b)

280

xPVAc/EMDI

260 240 220

xPVAc/p-MDI

200 180 160 140 120

5

10

15

20

Cross-linker loading (%)

Figure 8. Effect on the (a) cross-link density and (b) ratio cross-link density of EPI film before and Figure 8. Effect on the (a) cross-link density and (b) ratio cross-link density of EPI film before and after after post-curing at 80% RH. post-curing at 80% RH.

Good dispersion of EMDI in vinyl acetate-based tri-copolymer latex facilitates not only the Good dispersion EMDIwith in vinyl latexenvironment, facilitates notbut only chemical reaction ofof–NCO activeacetate-based hydrogen in tri-copolymer EPI in a humidity alsothe chemical reaction of –NCO with active hydrogen EPI in a humidity environment, also and potential potential intra-crosslinking of the polymer film, in subsequently expediting the curingbut process an intra-crosslinking of the polymer film, subsequently the curing process and increase to in υ. increase in υ . As seen in Figure 8, it is reasonableexpediting to conclude that xPVAc/EMDI filmansubjected As moisture seen in Figure reasonable to conclude xPVAc/EMDI film subjected to moisture results in results8,initaishigher crosslink degree, asthat compared with xPVAc/p-MDI at the same isocyanate content. After being thermally treatedwith for axPVAc/p-MDI day, it is observed that the isocyanate ratio of thecontent. cross-link density a higher crosslink degree, as compared at the same After being of xPVAc/EMDI also than of xPVAc/p-MDI, and the difference (ν post − cured thermally treated for a )day, it is observed is that thelarger ratio of thethat cross-link density (νpost /ν ν untreated −cured untreated ) of xPVAc/EMDI is also larger than that of xPVAc/p-MDI, and the difference enlarges as the cross-linker enlarges as the cross-linker loading increases. The reason for this is that within the post-curing period, loading increases. reasonreaction for thistake is that within post-curing period, a much that stronger a much stronger The cross-link places in thethe EPI film with EMDI, indicating therecross-link may be reaction places in the EPI with EMDI, indicating there may more residual isocyanate more take residual isocyanate in film the untreated polymer. For that the purpose of be probing the difference in in the untreated the purpose of probing thewas difference –NCO residue thevalue EPI films, –NCO residuepolymer. of the EPIFor films, the chemical structure checkedinusing FTIR. The of peak of −1 [25] theCaCO chemical structure was checked FTIR. Thefor peak value of CaCO3 at 875and cmthe was used 3 at 875 cm−1 [25] was used asusing the benchmark spectrum normalization, normalized curves are represented in Figure 9a. Seemingly, fromcurves the isocyanate can be in perceived as the benchmark for spectrum normalization, andthe the reaction normalized are represented Figure 9a. −1 and due to the presence of absorption bands in the C=O stretching in the range of 1740–1620 cm Seemingly, the reaction from the isocyanate can be perceived due to the presence of absorption bands −1 − 1 [14]. However, theseN-H characteristic bands could not−1 be N-HC=O stretching around cmof 1740–1620 in the stretching in the3300 range cm and stretching around 3300 cm [14]. distinguished the C=O stretching covered with a huge the bandC=O due stretching to CaCO3 being However, these because characteristic bands couldregion not bewas distinguished because region −1, and the N-H stretching bands overlapped with the−broad centered 1400 acm assignable to was coveredatwith huge band due to CaCO3 being centered at 1400 cm 1 , andband the N-H stretching water [26]. The following therefore focuses on the evolution of the relative intensity of the peak bands overlapped with the broad band assignable to water [26]. The following therefore focusesaton cm−1 assigned to an asymmetric isocyanate relative absorption ratio of the peaks, −1 assigned the2270 evolution of the relative intensity of the peak stretching; at 2270 cmthe to an asymmetric isocyanate   −1 −1 −1 /A 1 , interpreted as peaks, the amount groups in − cured EPI film R = A 2270 A absorption 875cm , interpreted stretching; thecm relative ratio of the R = of A residual 2270 cmNCO 875 cm as before the amount of post-cured residual NCO groups inplotted cured as EPI film in before and and after treatment, was shown Figure 9b. after post-cured treatment, was plotted shown9b, in under Figurethe 9b. same cross-linker loading, xPVAc/p-MDI films reveal markedly FromasFigure lower absorbance ( R ) than xPVAc/EMDI before post-curing, indicating that lower the Fromrelative Figure infrared 9b, under the same cross-linker loading, xPVAc/p-MDI films reveal markedly –NCOinfrared depletedabsorbance much faster the xPVAc/EMDI p-MDI based EPI filmpost-curing, during the preparing its relative (R)forthan before indicatingprocess. that theAs –NCO cross-link density is evidently smaller compared to that of the film with EMDI, it may be deduced depleted much faster for the p-MDI based EPI film during the preparing process. As its cross-link that more of the cross-linker of xPVAc/p-MDI 2O in this stage. inference be of density is evidently smaller compared to that ofreacted the filmwith withHEMDI, it may beThe deduced thatcan more further verified by the higher porosity in this type of specimen (see Figure 10). After post-cure, the the cross-linker of xPVAc/p-MDI reacted with H2 O in this stage. The inference can be further verified R drops while the ratio of relative infrared absorbance ( R postthe Runtreated ) of by magnitude the higherofporosity in sharply, this type of specimen (see Figure 10). After post-cure, magnitude − cured

(

) (

)

R drops sharply, the ratioand of relative infrared absorbance ) forat–NCO for –NCO in thewhile xPVAc/EMDI xPVAc/p-MDI films after and (R before treatment 24 h is in post−thermal cured /Runtreated 37% and 39%, respectively, and it gradually as the cross-linker loading theabout xPVAc/EMDI and xPVAc/p-MDI films after anddecreases before thermal treatment at 24 h isincreases. about 37% post-curing treatment is necessary forasEPI, with high increases. isocyanate Therefore, content. andTherefore, 39%, respectively, and it gradually decreases the especially cross-linker loading Furthermore, the treated xPVAc/EMDI stillespecially has relatively residual isocyanate. it post-curing treatment is necessary for EPI, withhigher high isocyanate content. However, Furthermore, than the EPIbefilm with be seen from Figurestill 9c has thatrelatively the valuehigher of R post thecan treated xPVAc/EMDI residual However, it can seen from − cured Risocyanate. untreated is lower Figure 9c that the value of R /R is lower than the EPI film with p-MDI, demonstrating postmore −cured isocyanate untreated reacted in post-cured xPVAc/EMDI films than in p-MDI, demonstrating that that more isocyanate reacted in post-cured xPVAc/EMDI films than in xPVAc/p-MDI. This may be

x FOR PEER REVIEW Polymers 2018, 10, 652

13 13 of of 18

xPVAc/p-MDI. This may be attributed to the fact the water vapor from the environment can also attributed the fact the water from environment can also expediently penetrate in into the expedientlytopenetrate into the vapor relaxed EPIthe with EMDI in treatment at 140 °C, resulting there ◦ C, resulting in there being almost no difference in the relaxed EPI with EMDI in treatment at 140 being almost no difference in the reactivity of water molecules with the two kinds of reactivity of water molecules due withto thethe two kinds of polyisocyanate. Meanwhile, due concentration to the uniformly polyisocyanate. Meanwhile, uniformly distributed particles and higher of distributed particles and higher concentration of residual isocyanate, more –NCO in xPVAc/EMDI residual isocyanate, more –NCO in xPVAc/EMDI was consumed by the functional group in the was consumed the functional Hence, group in the polymer than in on xPVAc/p-MDI. Hence,films the post-curing polymer than inbyxPVAc/p-MDI. the post-curing effect EMDI-crosslinked is stronger effect onthat EMDI-crosslinked films is stronger than onisthat containing p-MDI, and the treated film than on containing p-MDI, and the treated film more densely cross-linked, contributing to isa more densely cross-linked, contributing to a more obvious increase in moisture resistance. more obvious increase in moisture resistance. (a)

Absorbance

postcured xPVAc/pMDI

postcured xPVAc/EMDI

untreated xPVAc/pMDI

untreated xPVAc/EMDI

4000

3500

3000

2500

2000

1500

1000

500

Wavenumber (cm-1) 2.4 untreated xPVAc/EMDI untreated xPVAc/p-MDI post-cured xPVAc/EMDI post-cured xPVAc/p-MDI

1.6 1.2 0.8 0.4 0.0

5

10

15

Cross-linker loading (%)

20

(c)

39

R post-cured /R untreated (%)

Before post-cure

2.0

After post-cure

Relative infrared absorbance, R (%)

(b)

xPVAc/p-MDI

36 33 30

xPVAc/EMDI

27 24 21 5

10

15

20

Cross-linker loading (%)

Figure 9. 9. (a) (a) FTIR FTIRspectra spectraof; of;(b) (b)relative relativeinfrared infrared absorbance ) and (c)ratio its ratio for NCO EPI Figure absorbance (R)( R and (c) its for NCO in EPIinfilms films and afterbefore and before post-cure. after post-cure.

As the post-cured cross-linking of EPI film is accompanied by the reaction between isocyanate As the post-cured cross-linking of EPI film is accompanied by the reaction between isocyanate and water as described in Equation (1), the micropore magnitude on the specimen surface will be and water as described in Equation (1), the micropore magnitude on the specimen surface will be considerably changed due to the released carbon dioxide gas. Therefore, the change in micropore considerably changed due to the released carbon dioxide gas. Therefore, the change in micropore distribution may reflect the extent of this reaction. Based on this hypothesis, we used a 3D optical distribution may reflect the extent of this reaction. Based on this hypothesis, we used a 3D optical microscope to observe the morphologies of the micropores on the surface of the EPI specimens with microscope to observe the morphologies of the micropores on the surface of the EPI specimens with different cross-linkers before and after heat treatment, as shown in Figure 10. Distinct micropore different cross-linkers before and after heat treatment, as shown in Figure 10. Distinct micropore distribution and morphology were observed on the film surface. Before post-curing, the micropores distribution and morphology were observed on the film surface. Before post-curing, the micropores are very small and sparsely distributed, and the specimens with EMDI show smaller micropores are very small and sparsely distributed, and the specimens with EMDI show smaller micropores than than p-MDI, indicating that less isocyanate was consumed by water molecules during EPI film p-MDI, indicating that less isocyanate was consumed by water molecules during EPI film preparation. preparation. After treatment, some larger pores appear, and the number of micropores increases After treatment, some larger pores appear, and the number of micropores increases rapidly due to rapidly due to the reaction of residual NCO and moisture of the surroundings; there is also a the reaction of residual NCO and moisture of the surroundings; there is also a comparative increase comparative increase of micropores on the film surface for both types of EPI. This observation may of micropores on the film surface for both types of EPI. This observation may confirm that post-cure confirm that post-cure facilitates residual isocyanate in xPVAc/EMDI preferentially reacting with the facilitates residual isocyanate in xPVAc/EMDI preferentially reacting with the available group in the available group in the polymer, resulting in a higher degree of chemical cross-linking. polymer, resulting in a higher degree of chemical cross-linking.

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW Polymers 2018, 10, x FOR PEER REVIEW

14 of 18 14 of 18 14 of 18

Figure10.10. Optical Optical photographs xPVAc/EMDI; (b) untreated xPVAc/p-MDI; (c) Figure photographsofof(a)(a)untreated untreated xPVAc/EMDI; (b) untreated xPVAc/p-MDI; post-cured xPVAc/EMDI and (d) post-cured xPVAc/p-MDI (black dots: micro-pores). (c) post-cured and (d) post-cured xPVAc/p-MDI (black(b) dots: micro-pores). Figure 10.xPVAc/EMDI Optical photographs of (a) untreated xPVAc/EMDI; untreated xPVAc/p-MDI; (c) post-cured xPVAc/EMDI and (d) post-cured xPVAc/p-MDI (black dots: micro-pores).

3.3. Model Analysis on the Moisture-Dependent Mechanical Response 3.3. Model Analysis on the Moisture-Dependent Mechanical Response 3.3. Analysissection, on the Moisture-Dependent Mechanical Response were systematically subjected to InModel the previous EPI films with different cross-linkers In the previous section, EPI films with different cross-linkers were systematically subjected to quasi-static tensile tests in a variety of humid environments, and the corresponding stress-strain In the previous section, EPI films with different cross-linkers were systematically stress-strain subjected to quasi-static tensile testsHowever, in a variety of humid environments, and the curves were obtained. the experiments could not encompass all corresponding relative humidity, making quasi-static tensileHowever, tests in a the variety of humidcould environments, and the corresponding stress-strain curves were obtained. experiments not encompass all relative humidity, making it necessary to establish a theoretical model that can predict the mechanical behavior of adhesive curves were obtained. However, the experiments could not encompass all relative humidity, making it necessary to establish a theoretical model that can predict the mechanical behavior of adhesive films. This section offers a constitutive model to quantitatively describe the moisture-dependent it necessary to establish a theoretical model that can predict the mechanical behavior of adhesive films. This section offers a constitutive model torelaxation quantitatively moisture-dependent stress-strain curves of cured films; calculated time isdescribe used tothe determine the effect films. This section offers a constitutive model to quantitatively describe the moisture-dependent stress-strain of cured films; calculatedon relaxation time is usedof toEPI. determine the effect mechanism mechanismcurves of cross-linker and post-curing the water resistance stress-strain curves of cured films; calculated relaxation time is used to determine the effect of cross-linker andin post-curing water resistance ofEPI EPI.samples display typical behavior of As shown Figure 1, on thethemoisture-absorbed mechanism of cross-linker and post-curing on the water resistance of EPI. As shown materials. in FigureTherefore, 1, the moisture-absorbed samples display typical behavior thermoplastic it is feasible to deriveEPI a viscoelastic constitutive equation based of As shown in Figure 1, the moisture-absorbed EPI samples display typical behavior of on the rheological modelTherefore, composeditofistwo Maxwell elements in parallel (see Figure 11), and we can thermoplastic materials. feasible to derive a viscoelastic constitutive equation thermoplastic materials. Therefore, it is feasible to derive a viscoelastic constitutive equation based based σ ) and the relationship between stress strainelements ( ε ) in theinequation belowFigure (2). onobtain the model composed of (two Maxwell onrheological the rheological model composed of two Maxwell elements inparallel parallel(see (see Figure11), 11), and and we we can can obtain the relationship between stress (σ) (and strain (ε) in below (2). (2). σ ) and obtain the relationship between stress strain ( εthe ) inequation the equation below

Figure 11. Rheological model for EPI film composed of two Maxwell elements in parallel. Figure 11. Rheological model for EPI film composed of two Maxwell elements in parallel. Figure 11. Rheological model for EPI film composed of two Maxwell elements in parallel.

Polymers 2018, 10, 652

15 of 18

" σ = 2c10 (1 + ε) −

1

(1 + ε )2

#

.

+ εη1 (1 − e



ε η m 1 E1

)

(5)

where c10 is the stiffness of a non-linear spring, E1 is the stiffness of a linear spring, η1 is the viscosity . of an ideal dashpot, and ε is the strain rate. Equation (5) was used to describe all the uniaxial tensile test curves of the dry and wet specimen cross-linked by EMDI or p-MDI. The representative results are shown in Figures 1 and 6, where the calculated values (solid lines) were in agreement with the experimental data (dots). The correlation coefficient close to 1 indicated that the constitutive model was accurate for calculating the mechanical behavior of EPI films in humid surroundings. As elaborated in the published works [27–30], relaxation time is an intrinsic property describing the hysteresis of chain segments of polymer or its composites under external stress. In order to investigate the moisture influence on the mechanical behaviors of EPI film in greater depth, we attempted to determine the relaxation time of a cured sample at 0–80% RH as follows. The relaxation time (τ) of a viscoelastic solid is defined as the ratio of the system viscosity to the system stiffness. Therefore, we can use the material parameters E1 , η1 and c10 obtained by the fitted experimental stress-strain curves to determine the τ value of the EPI films in various conditions according to the Equation (6): η ( E + c10 ) τ= 1 1 (6) E1 c10 The change in relaxation time (τ) against RH for EPI film with different crosslinkers before and after the post-curing treatment is shown in Figure 12. The water molecules penetrating into EPI films form hydrogen bonds with the hydroxyl groups in the main component [31], by which the interchain and intra-chain bonding among amorphous chains in the polymer may be broken [32], and the τ magnitude descends by about 30–50% as RH increases from 0 to 80%. Furthermore, it was found that the relaxation time of EPI film lengthens with the increased polyisocyanate content, also verifying the effective cross-linking between NCO groups and active hydrogen in EPI. Adding more cross-linker or post-curing would cause a stronger interaction force due to the remarkable substantial increase in the number of cross-link bonds among molecular chains, meaning that the film would shrink further and the space around the network chains would be compressed. Under such a steric hindrance effect, the conformational transition of molecular chains is inhibited, and the motion of cross-links is more strictly constrained [33]. As anticipated, xPVAc/EMDI exhibits a slower relaxation process than xPVAc/p-MDI of the same cross-linker ratio, owing to the higher crosslinking capacity of emulsified polyisocyanate. Meanwhile, the relaxation times of all EPI specimens can be also greatly increased by post-curing. In order to further compare the responses of different EPI specimens to post-cured treatment, we calculated the relaxation time and cross-link density of a post-cured film divided by those of an untreated one, as shown in Figure 13. With the increased loading of polyisocyanate, both the ratio of relaxation time (τpost−cured /τuntreated ) and the ratio of crosslink density (νpost−cured /νuntreated ), as shown in Figure 8b, of the polymer films is noticeably enlarged. That is to say, the higher the cross-linker loading, the better the treated effects. Moreover, the relative increment of both material properties for xPVAc/EMDI is higher compared to xPVAc/p-MDI at the same cross-linker loading, also validating the fact that the EMDI-crosslinked films is more sensitive to the post-curing process in terms of chain segments relaxation. Notably, due to coupling the variation of stiffness and viscosity of cured polymer film, the value of τpost−cured /τuntreated is significantly larger than νpost−cured /νuntreated for the identical EPI. Therefore, the relaxation time may be the best intrinsic parameter for characterizing the influence of isocyanate, of post-curing treatment, and of environmental conditions on the EPI polymer.

Polymers 2018, 10, 652 Polymers 2018, 10, x FOR PEER REVIEW Polymers 2018, 10, x FOR PEER REVIEW

16 of 18 16 of 18 16 of 18

Figure 12. 12. Changes in relaxation time against RH and crosslinker content for (a) Figure in relaxation time against differentdifferent RH and crosslinker for (a) xPVAc/EMDI Figure 12.Changes Changes in relaxation time against different RH and content crosslinker content for (a) xPVAc/EMDI and (b) xPVAc/p-MDI films before and after post-curing treatment. and (b) xPVAc/p-MDI films before and after post-curing treatment. xPVAc/EMDI and (b) xPVAc/p-MDI films before and after post-curing treatment.

Figure 13. Effect of cross-linker loading on ratio of relaxation time for post-cured EPI film after and Figure 13. Effect of cross-linker loading on ratio of relaxation time for post-cured EPI film after and Figure 13. Effect of cross-linker loading on ratio of relaxation time for post-cured EPI film after and before post-cure. beforepost-cure. post-cure. before

4. Conclusions 4. Conclusions 4. Conclusions (1) With increasing relative humidity, typical plasticization still occurs for EPI with high isocyanate (1) With increasing relative humidity, typical plasticization still occurs for EPI with high isocyanate (1) content, With increasing humidity, still occurs EPI with high isocyanate leading relative to a clear decreasetypical in the plasticization tensile rupture stress andforYoung’s modulus of cured content, leading to a clear decrease in the tensile rupture stress and Young’s modulus of cured content, leading a cleartemperature. decrease in the tensile rupture stress and Young’s modulus of cured adhesive layer at to ambient adhesive layer at ambient temperature. adhesive layerresistance at ambientoftemperature. (2) The moisture EPI film can be significantly enhanced not only by increasing the (2) The moisture resistance of EPI film can be significantly enhanced not only by increasing the content, but also by the treatment. enhanced Although not xPVAc/EMDI has a higher (2) isocyanate The moisture resistance of EPI filmpost-curing can be significantly only by increasing the isocyanate content, but also by the post-curing treatment. Although xPVAc/EMDI has a higher amount of residual NCO groups the cured film than xPVAc/p-MDI, it exhibits has much better isocyanate content, but also by theinpost-curing treatment. Although xPVAc/EMDI a higher amount of residual NCO groups in the cured film than xPVAc/p-MDI, it exhibits much better mechanical properties andgroups cross-linking for the whole range, because aqueousmuch emulsified amount of residual NCO in the cured film thanRH xPVAc/p-MDI, it exhibits better mechanical properties and cross-linking for the whole RH range, because aqueous emulsified isocyanate led to aand more efficient chemical reaction with the because functional groups available mechanicalhas properties cross-linking for the whole RH range, aqueous emulsified isocyanate has led to a more efficient chemical reaction with the functional groups available in/on the emulsion the general polyisocyanate cross-linker. isocyanate has led polymer to a morecompared efficient to chemical reaction with the functional groups available in/on the emulsion polymer compared to the general polyisocyanate cross-linker. (3) The introduction of polymer isocyanate groups contributes to polyisocyanate an increase in the glass transition relative in/on the emulsion compared to the general cross-linker. (3) The introduction of isocyanate groups contributes to an increase in the glass transition relative of EPI film, and the RH g value of xPVAc/EMDI is higherinthan that oftransition xPVAc/p-MDI at (3) humidity The introduction of isocyanate groups contributes to an increase the glass relative humidity of EPI film, and the RHg value of xPVAc/EMDI is higher than that of xPVAc/p-MDI at the same crosslinker due to a greater constraint on chain movement. at a humidity of EPI film, loading and the RH value of xPVAc/EMDI is higher than that of Moreover, xPVAc/p-MDI the same crosslinker loading dueg to a greater constraint on chain movement. Moreover, at a at the same a greater constraint on chain a RH −crosslinker RH g , the loading ratio of due yieldtostress to elastic modulus doesmovement. not dependMoreover, markedlyaton given given RH − RH g , the ratio of yield stress to elastic modulus does not depend markedly on given RH − RHtype ratio of yield stress elastic modulus not depend the g , the the crosslinker and loading. This alsotosupports the idea does that mobility andmarkedly cohesion on effects the crosslinker type and loading. This also supportsthe theidea ideathat thatmobility mobilityand andcohesion cohesion effects effects crosslinker type and loading. This also supports mainly govern yield behavior of polymeric materials from another viewpoint. mainly govern yield yield behavior of of polymeric materials materials from from another another viewpoint. viewpoint. mainly govern (4) Post-cure is found tobehavior efficiently polymeric reduce the residual NCO groups in cured film, resulting in a (4) Post-cure is found to efficiently reduce the residual NCO groups in cured film, resulting in a significant increase in the crosslink density and also the mechanical behavior of EPI. Notably, the significant increase in the crosslink density and also the mechanical behavior of EPI. Notably, the post-curing effect on EMDI cross-linked films is stronger than on those containing p-MDI. post-curing effect on EMDI cross-linked films is stronger than on those containing p-MDI.

Polymers 2018, 10, 652

(4)

(5)

17 of 18

Post-cure is found to efficiently reduce the residual NCO groups in cured film, resulting in a significant increase in the crosslink density and also the mechanical behavior of EPI. Notably, the post-curing effect on EMDI cross-linked films is stronger than on those containing p-MDI. The derived viscoelastic constitutive equation has quantitatively characterized the tensile stress-strain curves of EPI films under variable relative humidity conditions, and the related relaxation time can clarify the influence mechanism of the isocyanate type and loading, as well as post-curing treatment on the moisture resistance of the cured polymer.

These results may shed light on the complicated mechanical behaviors of emulsion polymer isocyanate film in humid environments, which is of great significance for predicting the long-term performance of EPI-bonded structures. This study also provides insights into the role of EMDI and post-curing treatment, facilitating proper engineering designs and applications for high-performance adhesives. Author Contributions: H.H. conceived and designed the EPI experiments and model; J.G. and K.Z. performed the EPI experiments and simulation; Y.H. analyzed the data; H.H. and J.G. wrote the paper; X.G. supervised the paper. Funding: This research was funded by the National Science Foundation of China under grant numbers 11472164, 11072137 and 11332005, and the National Basic Research Program of China (2014CB046203). Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2. 3. 4. 5. 6.

7.

8. 9. 10. 11.

12.

Wang, X.D.; Hagman, O.; Sundqvist, B.; Ormarsson, S.; Wan, H.; Niemz, P. Impact of cold temperatures on the shear strength of Norway spruce joints glued with different adhesives. Eur. J. Wood Wood Prod. 2015, 73, 225–233. [CrossRef] Hu, H.J.; Liu, H.; Zhao, J.J.; Li, J. Investigation of adhesion performance of aqueous polymer latex modified by polymeric methylene diisocyanate. J. Adhes. 2006, 82, 93–114. [CrossRef] Barrere, M.; Landfester, K. High molecular weight polyurethane and polymer hybrid particles in aqueous miniemulsion. Macromolecules 2003, 36, 5119–5125. [CrossRef] Gao, Z.H.; Wang, W.B.; Zhao, Z.Y.; Guo, M.R. Novel Whey Protein-Based Aqueous Polymer-Isocyanate Adhesive for Glulam. J. Appl. Polym. Sci. 2011, 120, 220–225. [CrossRef] Vick, C.B.; Rowell, R.M. Adhesive bonding of acetylated wood. Int. J. Adhes. Adhes. 1990, 10, 263–272. [CrossRef] Knorz, M.; Neuhaeuser, E.; Torno, S.; van de Kuilen, J.W. Influence of surface preparation methods on moisture-related performance of structural hardwood-adhesive bonds. Int. J. Adhes. Adhes. 2015, 57, 40–48. [CrossRef] Sakurada, S.; Miyazaki, H.; Hattori, T.; Shiraishi, M.; Inoue, T. Adhesive Composition Consisting of Polyvinylalcohol Solution or Polyvinylacetate Latex Modified with Hydrophobic Solution of Isocyanate Compound. US Patent 3,931,088, 6 January 1976. Qiao, L.; Easteal, A.J.; Bolt, C.J.; Coveny, P.K.; Franich, R.A. Thermomechanical analysis and performance tests of some EPI wood adhesives. Pigm. Resin Technol. 2000, 29, 229–237. [CrossRef] Hu, H.J.; Zhu, W.P.; Liu, H.; Zhao, J.J. Investigation on fracture performance of aqueous polymer isocyanates in terms of energy release rate. J. Adhes. Sci. Technol. 2006, 20, 161–174. [CrossRef] Krystofiak, T.; Proszyk, S.; Jozwiak, J. Studies of some properties of EPI adhesives. For. Wood Technol. 2003, 53, 214–217. Taki, K.; Tomita, B.; Mizumachi, H. Studies on aqueous vinyl polymer solution-isocyanate adhesives. Part I. Mechanical properties of base polymers and bond strength over a wide temperature range. Mokuzai Gakkaishi 1982, 28, 143–149. Taki, K.; Tomita, B.; Mizumachi, H. Studies on aqueous vinyl polymer solution-isocyanate adhesives. Part II. Dependence of mechanical properties and bond strength on the concentration of crosslinks in cured adhesives over a wide temperature range. Mokuzai Gakkaishi 1982, 28, 150–155.

Polymers 2018, 10, 652

13. 14. 15. 16. 17. 18. 19. 20.

21.

22. 23. 24. 25. 26.

27. 28. 29. 30.

31.

32. 33.

18 of 18

Umemura, K.; Takahashi, A.; Kawai, S. Durability of isocyanate resin adhesives for wood-I: Thermal properties of isocyanate resin cured with water. J. Wood Sci. 1998, 44, 204–210. [CrossRef] Ling, N.; Hori, N.; Takemura, A. Effect of postcure conditions on the dynamic mechanical behavior of water-based polymer-isocyanate adhesive for wood. J. Wood Sci. 2008, 54, 377–382. [CrossRef] Rana, D.; Sauvant, V.; Halary, J.L. Molecular analysis of yielding in pure and antiplasticized epoxy-amine thermosets. J. Mater. Sci. 2002, 37, 5267–5274. [CrossRef] Rana, D.; Mounach, H.; Halary, J.L.; Monnerie, L. Differences in mechanical behavior between alternating and random styrene-methyl methacrylate copolymers. J. Mater. Sci. 2005, 40, 943–953. [CrossRef] Brule, B.; Halary, J.L.; Monnerie, L. Molecular analysis of the plastic deformation of amorphous semi-aromatic polyamides. Polymer 2001, 42, 9073–9083. [CrossRef] Teze, L.; Halary, J.L.; Monnerie, L.; Canova, L. On the viscoelastic and plastic behaviour of methylmethacrylate-co-N-methylglutarimide copolymers. Polymer 1999, 40, 971–981. [CrossRef] Greco, S.; Authelin, J.R.; Leveder, C.; Segalini, A. A practical method to predict physical stability of amorphous solid dispersions. Pharm. Res. 2012, 29, 2792–2805. [CrossRef] [PubMed] Konidari, M.V.; Papadokostaki, K.G.; Sanopoulou, M. Moisture-induced effects on the tensile mechanical properties and glass-transition temperature of poly(vinyl alcohol) films. J. Appl. Polym. Sci. 2011, 120, 3381–3386. [CrossRef] Hu, H.J.; Zhang, X.L.; He, Y.L.; Guo, Z.S.; Zhang, J.Q.; Song, Y.C. Combined effect of relative humidity and temperature on dynamic viscoelastic properties and glass transition of poly(vinyl alcohol). J. Appl. Polym. Sci. 2013, 130, 3161–3167. [CrossRef] Suvegh, K.; Zelko, R. Physical aging of poly(vinylpyrrolidone) under different humidity conditions. Macromolecules 2002, 35, 795–800. [CrossRef] Wei, Q.; Haag, R. Universal polymer coatings and their representative biomedical applications. Mater. Horiz. 2015, 2, 567–577. [CrossRef] Eroglu, M.S. Characterization of the network structure of hydroxyl terminated poly(butadiene) elastomers prepared by different reactive systems. J. Appl. Polym. Sci. 1998, 70, 1129–1135. [CrossRef] Reig, F. FTIR quantitative analysis of calcium carbonate (calcite) and silica (quartz) mixtures using the constant ratio method. Application to geological samples. Talanta 2002, 58, 811–821. [CrossRef] Hori, N.; Asai, K.; Takemura, A. Effect of the ethylene/vinyl acetate ratio of ethylene-vinyl acetate emulsion on the curing behavior of an emulsion polymer isocyanate adhesive for wood. J. Wood Sci. 2008, 54, 294–299. [CrossRef] Chen, J.K.; Yuan, M. Decoupling of viscous dissipation and damage dissipation in particulate reinforced polymeric materials. Comp. Mater. Sci. 2007, 40, 267–274. [CrossRef] Chen, J.H.; Hu, H.J.; Li, S.; Zhang, K.F. Quantitative relation between the relaxation time and the strain rate for polymeric solids under quasi-static conditions. J. Appl. Polym. Sci. 2016, 133. [CrossRef] Drozdov, A.D. Mechanics of Viscoelastic Solids; John Wiley and Sons: New York, NY, USA, 1998. Mathiesen, D.; Vogtmann, D.; Dupaix, R.B. Characterization and constitutive modeling of stress relaxation behavior of Poly(methyl methacrylate) (PMMA) across the glass transition temperature. Mech. Mater. 2014, 71, 74–84. [CrossRef] Gonzalez-Campos, J.B.; Garcia-Carvajal, Z.Y.; Prokhorov, E.; Luna-Barcenas, J.G.; Mendoza-Duarte, M.E.; Lara-Romero, J.; del Rio, R.E.; Sanchez, I.C. Revisiting the thermal relaxations of poly(vinyl alcohol). J. Appl. Polym. Sci. 2012, 125, 4082–4090. [CrossRef] Hodge, R.M.; Bastow, T.J.; Edward, G.H.; Simon, G.P.; Hill, A.J. Free volume and the mechanism of plasticization in water-swollen poly(vinyl alcohol). Macromolecules 1996, 29, 8137–8143. [CrossRef] Aoki, H.; Horinaka, J.; Ito, S.; Yamamoto, M. Local motion of crosslinks for poly(methyl methacrylate) gels by the fluorescence depolarization method. Polym. Bull. 1997, 39, 109–116. [CrossRef] © 2018 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).