Improvement in Wood Bonding Strength of Poly (Vinyl Acetate-Butyl Acrylate) Emulsion by Controlling the Amount of Redox Initiator

Polyvinyl acetate emulsion adhesive has been widely used due to its good bonding performance and environmentally friendly properties. Indeed, the bonding performance can be further improved by copolymerizing with other monomers. In this study, the effect of the adjunction of redox initiator (hydrogen peroxide–tartaric acid, H2O2–TA) on the properties of the poly (vinyl acetate-butyl acrylate) (P (VAc–BA)) emulsion adhesive was investigated. With increasing dosage, the reaction became more complete and the obtained film was more compact, as identified via SEM. The core-shell structure of the emulsion particles was confirmed via TEM. Results indicate that while the initiator content increased from 0.5 to 1.0%, a clearer core-shell structure was obtained and the bonding strength of the plywood improved from 2.34 to 2.97 MPa. With the further incorporation of H2O2–TA (i.e., 1.5%), the bonding performance deteriorated. The optimum wood bonding strength (2.97 MPa) of the prepared P (VAc-BA) emulsion adhesive was even better than that (2.55 MPa) of a commercial PVAc emulsion adhesive, suggesting its potential application for the wood industry.


Introduction
With improvements in living standards, further application of green products is advocated, which certainly includes adhesives. The wood-adhesive industry is currently dominated by petrochemical-derived resins, such as urea formaldehyde (UF), phenol formaldehyde (PF), and melamine urea formaldehyde (MUF) [1][2][3], which are all harmful due to the presence of formaldehyde. Considering the formaldehyde pollution that originates from such materials, the development of environmentally benign materials as alternative wood adhesives is both reasonable and promising [4]. Polyvinyl acetate (PVAc) emulsion is a type of water-based environment-friendly adhesive. Owing to the fascinating properties of polyvinyl alcohol (PVA), it was used as a protective colloid to provide the initial stage of the emulsion polymerization and could be grafted by other monomers [5]. In addition, due to the milky white color of the liquid, it is also called white latex and it is currently widely used in coatings, fireproof coatings, adhesives, and other industries [6][7][8].

Effect of the Amount of Initiator on the Bonding Strength
Both the content of the PVA and the proportion of the monomer were optimized and the effect of the initiator on the adhesive was studied. First, we studied the effect of different amounts of PVA (5%, 10%, 15%, 20% and 25% of the total monomer mass) on the performance of the P (VAc-BA) emulsion adhesive (VAc:BA = 2:1, initiator of 0.5% of the monomer). The solid content, pH of the adhesives, and tensile strength of the corresponding plywood are shown in Table 1. With increasing PVA dosage, the tensile strength of the prepared plywood gradually enhanced but decreased beyond the dosage of 20%. With a PVA content beyond 20%, the viscosity of the emulsion was seriously increased, which affected the reaction. Therefore, 20% PVA was selected to analyze the effect of different monomer amounts (1:1, 2:1, 3:1, 4:1, and 5:1) on the properties of the emulsion adhesive (Table 2). With an increasing ratio of VAc:BA, the bonding strength and the solid content increased obviously. When the ratio was greater than 4:1, the strength and the solid content led to a worse performance. Based on these data, the optimal properties of the adhesive were obtained at a VAc to BA ratio of 4:1. Hence, the ratio of 4:1 was chosen to study the effect of different amounts of the initiator (0.5, 0.6, 0.7, 0.8, 0.9, 1.0, and 1.5% of the total Materials 2018, 11, 89 3 of 12 monomer mass) on the performance of the P (VAc-BA) emulsion. The results are shown in Table 3 (20% PVA content, VAc:BA = 4:1). No obvious differences were found in the basic performance of the tested materials. However, as shown in Table 3, with an increase in the initiator dosage, the bonding strength of the plywood gradually increased and maximized at 1.0%. When the initiator dosage was increased, the polymerization degree increased and the resulting bonds among the polymers were more compact. The reaction was almost complete when the initiator dosage reached 1.0%, and the bonding strength of the fabricated plywood was maximal (2.97 MPa). However, when the initiator dosage was above 1.0% (such as 1.5%), the bonding strength was reduced (2.78 MPa). It was hypothesized that when more initiator was added, the polymerization reaction was too fast and some unreacted monomers were wrapped by the initially generated polymer. This conjecture needs to be confirmed in future work. However, the strength of the latex increased with the increasing dosage of the initiator when the dosage of the initiator was below 1.0%. Therefore, with a PVA content of 20%, a 4:1 ratio of VAc:BA, and an initiator dosage of 1.0%, the maximum strength of the adhesive was obtained. As can be seen in Table 3, the optimum wood bonding strength (2.97 MPa) of the prepared P (VAc-BA) emulsion adhesive was even better than that (2.55 MPa) of a commercial PVAc emulsion adhesive, suggesting its potential application in wood industry.

Thermal Gravimetric Analysis (TGA)
The thermal stability of the polymeric material is extremely important, especially with respect to an adhesive used to construct plywood. In this study, P (VAc-BA) emulsion samples with varying contents of initiators were analyzed. The DTG (Derivative Thermogravimetric) curves of the products prepared with different amounts of initiator are shown in Figure 1 and the results are listed in Table 4. Two degradation peaks were found in the DTG curves, indicating two degradation processes of the P (VAc-BA) emulsion films, labeled as 1 and 2 . The thermal degradation of each film occurred similarly, revealing that the amount of the initiator had a negligible influence on the thermal degradation mechanism of the P (VAc-BA) emulsion films. Subsequently, the initial degradation temperature (T 0 ), the maximum degradation rate temperature (T p ), and the terminated temperature (T f ) were obtained using the double tangent method. Table 4 shows that for an initiator dosage of 1.0%, the T p of the film was the highest. This indicates that the film with the highest thermal stability was fabricated with an initiator amount of 1.0%. emulsion adhesive was even better than that (2.55 MPa) of a commercial PVAc emulsion adhesive, suggesting its potential application in wood industry.

Thermal Gravimetric Analysis (TGA)
The thermal stability of the polymeric material is extremely important, especially with respect to an adhesive used to construct plywood. In this study, P (VAc-BA) emulsion samples with varying contents of initiators were analyzed. The DTG (Derivative Thermogravimetric) curves of the products prepared with different amounts of initiator are shown in Figure 1 and the results are listed in Table  4. Two degradation peaks were found in the DTG curves, indicating two degradation processes of the P (VAc-BA) emulsion films, labeled as ① and ②. The thermal degradation of each film occurred similarly, revealing that the amount of the initiator had a negligible influence on the thermal degradation mechanism of the P (VAc-BA) emulsion films. Subsequently, the initial degradation temperature (T0), the maximum degradation rate temperature (Tp), and the terminated temperature (Tf) were obtained using the double tangent method. Table 4 shows that for an initiator dosage of 1.0%, the Tp of the film was the highest. This indicates that the film with the highest thermal stability was fabricated with an initiator amount of 1.0%.   Table 4. Thermal degradation temperature of films.

Initiator Ratio
First Degradation Temperature ( • C) Secondary Degradation Temperature ( • C)

Scanning Electron Microscopy (SEM) Analysis
As a precise microstructure analysis tool, scanning electron microscopy (SEM) is often used to analyze characteristics with high resolution, large depth of field, and natural imaging. To investigate the effect of the amount of initiator on the surface structure of the emulsion, the morphology of the P (VAc-BA) matrix was examined via SEM and the images are shown in Figure 2. With the increasing initiator dosage, the structure of the film became increasingly compact, reflecting higher degrees of polymerization of the samples. Furthermore, the porous structure (Figure 2a-c) also provided evidence for the monomer crosslinking reaction.

Scanning Electron Microscopy (SEM) Analysis
As a precise microstructure analysis tool, scanning electron microscopy (SEM) is often used to analyze characteristics with high resolution, large depth of field, and natural imaging. To investigate the effect of the amount of initiator on the surface structure of the emulsion, the morphology of the P (VAc-BA) matrix was examined via SEM and the images are shown in Figure 2. With the increasing initiator dosage, the structure of the film became increasingly compact, reflecting higher degrees of polymerization of the samples. Furthermore, the porous structure (Figure 2a-c) also provided evidence for the monomer crosslinking reaction.

Transmission Electron Microscopy (TEM) Analysis
Transmission electron microscopy (TEM) was used to characterize the internal structure of the emulsion particles. According to a previous study, it was found that the PVAc homopolymer, but not the PBA homopolymer, could be colored by phosphotungstic acid (HPWA) [39]. In the present study, the core of the particles was constituted by a PVAc homopolymer and was covered with a P (VAc-BA) copolymer shell. In addition, the emulsion particles were dyed via HPWA. The TEM images revealed that the cores were completely colored by HPWA, while the shells showed a lighter color due to the presence of the BA monomer ( Figure 3). Hence, an inconspicuous but still distinguishable separation of bright and shaded layers was observed, representing the formation of the shell and the core, respectively.

Transmission Electron Microscopy (TEM) Analysis
Transmission electron microscopy (TEM) was used to characterize the internal structure of the emulsion particles. According to a previous study, it was found that the PVAc homopolymer, but not the PBA homopolymer, could be colored by phosphotungstic acid (HPWA) [39]. In the present study, the core of the particles was constituted by a PVAc homopolymer and was covered with a P (VAc-BA) copolymer shell. In addition, the emulsion particles were dyed via HPWA. The TEM images revealed that the cores were completely colored by HPWA, while the shells showed a lighter color due to the presence of the BA monomer ( Figure 3). Hence, an inconspicuous but still distinguishable separation of bright and shaded layers was observed, representing the formation of the shell and the core, respectively.  When the initiator amount was 0.5-0.8%, the viscosity of the emulsion was not stable, indicating that the reaction was not complete and monomer agglomeration might have occurred. When this value reached 1.0%, the reaction and the viscosity (Table 3) tended to be stable, implying that the reaction degree gradually increased. For an initiator amount above 1.0% of the total monomer content, a clear core-shell structure could be found in the latex particles, which was confirmed via color variation. This suggests that an appropriate amount of initiator was beneficial for the reaction by improving the degree of crosslinking of the emulsion particles, binding the molecule chains, and reducing the cluster migration behavior to prevent the core layer from being intruded upon by the BA monomer or the small molecular chain segments. Hence, a core-shell structure was successfully formed in the latex particles.
It was also found that when the amount of initiator was below 1.0%, a core-shell structure was not apparent. The reason for this is that molecular chains occur as a "phase inversion" or a "cluster migration" [40]. The particle shape changed significantly, which can be seen in Figure 3a,b. There might have been migration, disorder crosslinking, blending, and grafting among the molecular chains of PVAc and P (VAc-BA), causing the highly grey layers of the latex particles [41]. In addition, the poor homogeneity of the reaction system could have also contributed to this.

Differential Scanning Calorimetry (DSC) and Dynamic Mechanical Thermal Analysis (DMA)
Differential scanning calorimetry (DSC) and dynamic mechanical thermal analysis (DMA) were used to measure the glass transition temperature (Tg) of the sample, which could be guidance for the practical application of the synthesis polymer. These tests reflect both the compatibility and thermal stability of each component of the core-shell particles. The DSC and DMA curves of the latex films are shown in Figures 4 and 5, respectively. Figure 4 shows that with the addition of the initiator into the polymer matrix, an increased Tg could be observed. When the initiator amount was 0.5-0.8%, the viscosity of the emulsion was not stable, indicating that the reaction was not complete and monomer agglomeration might have occurred. When this value reached 1.0%, the reaction and the viscosity (Table 3) tended to be stable, implying that the reaction degree gradually increased. For an initiator amount above 1.0% of the total monomer content, a clear core-shell structure could be found in the latex particles, which was confirmed via color variation. This suggests that an appropriate amount of initiator was beneficial for the reaction by improving the degree of crosslinking of the emulsion particles, binding the molecule chains, and reducing the cluster migration behavior to prevent the core layer from being intruded upon by the BA monomer or the small molecular chain segments. Hence, a core-shell structure was successfully formed in the latex particles.
It was also found that when the amount of initiator was below 1.0%, a core-shell structure was not apparent. The reason for this is that molecular chains occur as a "phase inversion" or a "cluster migration" [40]. The particle shape changed significantly, which can be seen in Figure 3a,b. There might have been migration, disorder crosslinking, blending, and grafting among the molecular chains of PVAc and P (VAc-BA), causing the highly grey layers of the latex particles [41]. In addition, the poor homogeneity of the reaction system could have also contributed to this.

Differential Scanning Calorimetry (DSC) and Dynamic Mechanical Thermal Analysis (DMA)
Differential scanning calorimetry (DSC) and dynamic mechanical thermal analysis (DMA) were used to measure the glass transition temperature (T g ) of the sample, which could be guidance for the practical application of the synthesis polymer. These tests reflect both the compatibility and thermal stability of each component of the core-shell particles. The DSC and DMA curves of the latex films are shown in Figures 4 and 5, respectively. Figure 4 shows that with the addition of the initiator into the polymer matrix, an increased T g could be observed.  The arrow of each DSC curve appeared at the first Tg inflection point. The second inflection point occurred at approximately 47 °C and displayed a single cure peak. The exothermic peak in the diagram was observed. In Figure 5, the loss modulus peaks of the P (VAc-BA) copolymer and PVAc homopolymer can be clearly seen. The results suggest that two-phase states were formed in the film particles, which confirms the existence of the core-shell structure. The PVAc core and the P (VAc-BA) shell were originally two incompatible components of the emulsion particles; however, during the emulsion preparation process, the BA monomer had more active free radicals than the VAc monomer [42]. Thus, some BA monomers might have entered the core of the emulsion particles, participating in the copolymer reaction. It should be noted that only 20% of the BA monomer (in the total monomers) was added in the shell reaction. The shell of latex particles might be modified via the inclusion of fewer P (VAc-BA) molecular chains, indicating that there could be a partial loss of the peaks associated with the PVAc copolymer. Furthermore, the Tg of the PVAc homopolymer was approximately 47 °C, while the loss peak of the DMA appeared at above 48 °C.

Materials
Poly (vinyl alcohol) (PVA-1788), VAc, and BA were obtained from the Aladdin Industrial Corporation (Shanghai, China). Hydrogen peroxide solution (H2O2) was supplied by the Beijing Chemical Works (Beijing, China). Tartaric acid (TA) was obtained from the Fine Chemical Research  The arrow of each DSC curve appeared at the first Tg inflection point. The second inflection point occurred at approximately 47 °C and displayed a single cure peak. The exothermic peak in the diagram was observed. In Figure 5, the loss modulus peaks of the P (VAc-BA) copolymer and PVAc homopolymer can be clearly seen. The results suggest that two-phase states were formed in the film particles, which confirms the existence of the core-shell structure. The PVAc core and the P (VAc-BA) shell were originally two incompatible components of the emulsion particles; however, during the emulsion preparation process, the BA monomer had more active free radicals than the VAc monomer [42]. Thus, some BA monomers might have entered the core of the emulsion particles, participating in the copolymer reaction. It should be noted that only 20% of the BA monomer (in the total monomers) was added in the shell reaction. The shell of latex particles might be modified via the inclusion of fewer P (VAc-BA) molecular chains, indicating that there could be a partial loss of the peaks associated with the PVAc copolymer. Furthermore, the Tg of the PVAc homopolymer was approximately 47 °C, while the loss peak of the DMA appeared at above 48 °C.

Materials
Poly (vinyl alcohol) (PVA-1788), VAc, and BA were obtained from the Aladdin Industrial Corporation (Shanghai, China). Hydrogen peroxide solution (H2O2) was supplied by the Beijing Chemical Works (Beijing, China). Tartaric acid (TA) was obtained from the Fine Chemical Research The arrow of each DSC curve appeared at the first T g inflection point. The second inflection point occurred at approximately 47 • C and displayed a single cure peak. The exothermic peak in the diagram was observed. In Figure 5, the loss modulus peaks of the P (VAc-BA) copolymer and PVAc homopolymer can be clearly seen. The results suggest that two-phase states were formed in the film particles, which confirms the existence of the core-shell structure. The PVAc core and the P (VAc-BA) shell were originally two incompatible components of the emulsion particles; however, during the emulsion preparation process, the BA monomer had more active free radicals than the VAc monomer [42]. Thus, some BA monomers might have entered the core of the emulsion particles, participating in the copolymer reaction. It should be noted that only 20% of the BA monomer (in the total monomers) was added in the shell reaction. The shell of latex particles might be modified via the inclusion of fewer P (VAc-BA) molecular chains, indicating that there could be a partial loss of the peaks associated with the PVAc copolymer. Furthermore, the T g of the PVAc homopolymer was approximately 47 • C, while the loss peak of the DMA appeared at above 48 • C.

Preparation of Core-Seed Emulsion
A 500 mL four-flask was placed in a water bath, equipped with a mechanical electric mixer, a thermometer, a reflux condenser, and a constant pressure funnel. Then, 80 mL deionized water and 12 g PVA were added to the flask and the mixture was stirred at room temperature for a few minutes. Subsequently, the temperature of the mixture was increased to 90 • C within 2 h with vigorous stirring until a transparent solution was obtained, which was then cooled to 60 • C. After the temperature was stable, the buffers (0.1 g sodium acetate and 0.2 g ferrous sulfate, respectively), 0.8 g VAc core-monomer, and 1/3 of the total initiator of the formulation (different ratios of the total monomer ranged from 0.5% to 1.5%) were added to the flask. The remaining VAc monomer (19.2 g) was added after the polymerization begun. When the yellow color of the emulsion became lighter and the reflux of the monomer stopped, the core-seed emulsion was prepared.

Shell Copolymerization Reaction
For the second step, 1.6 g VAc/BA monomers (of different ratios) and 1/3 of the total initiator of the formulation were added to the flask. When the polymerization began, the remaining 38.4 g VAc/BA monomers and another 1/3 of the total initiator of the formulation were slowly dripped into the mixture at 60 • C within 4 h. Finally, the emulsions were naturally cooled to 40 • C. The pH value was adjusted to 3-4 using 30 wt % NaAc solution.

Preparation of Core-Seed Emulsion
A 500 mL four-flask was placed in a water bath, equipped with a mechanical electric mixer, a thermometer, a reflux condenser, and a constant pressure funnel. Then, 80 mL deionized water and 12 g PVA were added to the flask and the mixture was stirred at room temperature for a few minutes. Subsequently, the temperature of the mixture was increased to 90 °C within 2 h with vigorous stirring until a transparent solution was obtained, which was then cooled to 60 °C. After the temperature was stable, the buffers (0.1 g sodium acetate and 0.2 g ferrous sulfate, respectively), 0.8 g VAc coremonomer, and 1/3 of the total initiator of the formulation (different ratios of the total monomer ranged from 0.5% to 1.5%) were added to the flask. The remaining VAc monomer (19.2 g) was added after the polymerization begun. When the yellow color of the emulsion became lighter and the reflux of the monomer stopped, the core-seed emulsion was prepared.

Shell Copolymerization Reaction
For the second step, 1.6 g VAc/BA monomers (of different ratios) and 1/3 of the total initiator of the formulation were added to the flask. When the polymerization began, the remaining 38.4 g VAc/BA monomers and another 1/3 of the total initiator of the formulation were slowly dripped into the mixture at 60 °C within 4 h. Finally, the emulsions were naturally cooled to 40 °C. The pH value was adjusted to 3-4 using 30 wt % NaAc solution.

Mechanism of Initiator
H2O2-TA was applied in this study as an efficient redox initiator, while ferrous sulfate was used as the reductant. The respective reaction steps are shown in Scheme 1. When the pH was approximately 3-4, the initial rate of the polymerization reaction was the fastest [43]. The concentration of H + in the emulsion system could be controlled at the beginning of the reaction, which was achieved via addition of a certain amount of tartaric acid and Fe 2+ solution, combined with the appropriate amount of NaAc as the buffer. During the polymerization, the stirring rate was continuously adjusted to ensure that uniform mixture and the monomers could take part in When the pH was approximately 3-4, the initial rate of the polymerization reaction was the fastest [43]. The concentration of H + in the emulsion system could be controlled at the beginning of the reaction, which was achieved via addition of a certain amount of tartaric acid and Fe 2+ solution, combined with the appropriate amount of NaAc as the buffer. During the polymerization, the stirring rate was continuously adjusted to ensure that uniform mixture and the monomers could take part in the reaction at the appropriate time.

Preparation and Characterization of Plywood
Three-layered poplar plywood (400 × 400 × 4.5 mm) was prepared according to a previous study [44]. The core poplar veneer was painted with 180 g/m 2 glue on each side. Subsequently, the plywood was cool-pressed at room temperature under 2.0 MPa for 24 h. Then, the pressed plywood was fabricated and tested for bonding strength, according to the Chinese National Standards (GB/T 17657-2013) [45]. Each prepared plywood section was cut into 12 specimens (25 × 100 mm, "A-type" specimen), which were dried at room temperature for 24 h for the bonding strength test. The universal mechanical testing machine WDS-50KN (SUNS Technology Stock Co., Ltd, Shenzhen, China) was used to measure the bonding strength (CN) of the treated specimens. Reported results were averages of the 12 specimens.

Thermal Gravimetric Analysis (TGA)
The thermal properties of the samples were conducted via thermal gravimetric analysis (TGA) (DTG-60, Shimadzu, Japan). Prior to thermal analysis, all samples were subjected to freeze-drying. The freeze-dried samples (approximately 3-5 mg) were heated in an aluminum crucible from room temperature to 600 • C at a heating rate of 10 • C/min, while the apparatus was continually flushed with a nitrogen flow of 50 mL/min. The results in Table 4 were measured in triplicate.

Scanning Electron Microscopy (SEM)
To observe both the surface roughness and the morphology of the films, SEM (HitachiS-5500, Hitachi Limited, Tokyo, Japan) was used. Fine gold films were sputtered on the otherwise insulating samples and the test was conducted with a voltage of 10 kV.

Transmission Electron Microscopy (TEM)
Japan Optical Electronics (JOEL)'s 200 kV high-resolution transmission electron microscope (JEM 2100, JEOL Ltd., Tokyo, Japan) was used to study the structure of the emulsion particles. The emulsion was slightly diluted to turbidity (about 300 times). The film was molded with a special copper mesh and dried and then the membrane on the copper mesh was stained with 2% phosphotungstic acid (HPWA) and dried again.

Differential Scanning Calorimetry (DSC)
The glass transition temperature (T g ) of the film prepared with the synthesized adhesive was determined using a DSC (Q2000, TA Instrument, New Castle, DE, USA) in the nitrogen atmosphere. Then, 6-8 mg freeze-dried sample was used per measurement. The heating rate was 10 • C/min and the temperature ranged from −60 to 60 • C.

Dynamic Mechanical Thermal Analysis (DMA)
The thermal mechanical properties of the emulsion film were characterized using dynamic mechanical thermal analysis (DMA) (Q800, TA Instrument, New Castle, DE, USA) in a nitrogen atmosphere, with a scanning frequency of 1 Hz. The heating rate was 5 • C/min and the temperature ranged from −60 to 75 • C.

Conclusions
In this study, the influence of different dosages of redox initiator (H 2 O 2 -TA) in semi-continuous emulsion polymerization on the properties of polymer emulsions was studied. The optimum wood bonding strength (2.97 MPa) of the P (VAc-BA) emulsion adhesive was prepared when the conditions included a PVA content of 20%, a ratio of VAc:BA of 4:1, and an initiator dosage of 1.0%. SEM analyses indicate that a polymerization reaction occurred between both monomers (VAc and BA) and that the trend of polymerization was gradually completed. A film with the highest thermal stability was fabricated when the amount of initiator was 1.0%. The core-shell structure became apparent when the amount of initiator surpassed 1.0%. Probing the influence of the redox initiator (H 2 O 2 -TA) on the P (VAc-BA) emulsion structure and the film performance form the subject of our ongoing work.