Abstract
Oxidized starch (OS) adhesives with a balance between their adhesion and disintegration properties were prepared by controlling the degree of oxidation and modifying the cross-linker type and level to replace urea-formaldehyde (UF) resins for easy recycling of medium density fiberboard (MDF). Four molar ratios of H2O2/starch, two types of cross-linker, i.e., blocked-pMDI (B-pMDI) and citric acid (CA), and three levels of the cross-linkers were employed to tailor the performance of the OS adhesives. The OS reacted with the isocyanate groups from the B-pMDI to form amide linkages, while it formed ester linkages by reacting with the CA. The resulting B-pMDI/OS-bonded MDF had better physical and mechanical properties than the CA/OS-bonded MDF, with comparable adhesion (0.34 MPa) to UF resins and ten times greater degree of fiber disintegration than UF resins. The combination of a 0.5 molar ratio OS with 7.5 wt% of B-pMDI produced MDF exhibiting an optimal balance between adhesion and disintegration, suggesting that such OS adhesives could someday replace UF resins in manufacturing and recycling of MDF without formaldehyde emission.
Download PDF
Full Article
Tuning of Adhesion and Disintegration of Oxidized Starch Adhesives for the Recycling of Medium Density Fiberboard
Muhammad Adly Rahandi Lubis,a Byung-Dae Park,b,* and Min-Kug Hong c
Oxidized starch (OS) adhesives with a balance between their adhesion and disintegration properties were prepared by controlling the degree of oxidation and modifying the cross-linker type and level to replace urea-formaldehyde (UF) resins for easy recycling of medium density fiberboard (MDF). Four molar ratios of H2O2/starch, two types of cross-linker, i.e., blocked-pMDI (B-pMDI) and citric acid (CA), and three levels of the cross-linkers were employed to tailor the performance of the OS adhesives. The OS reacted with the isocyanate groups from the B-pMDI to form amide linkages, while it formed ester linkages by reacting with the CA. The resulting B-pMDI/OS-bonded MDF had better physical and mechanical properties than the CA/OS-bonded MDF, with comparable adhesion (0.34 MPa) to UF resins and ten times greater degree of fiber disintegration than UF resins. The combination of a 0.5 molar ratio OS with 7.5 wt% of B-pMDI produced MDF exhibiting an optimal balance between adhesion and disintegration, suggesting that such OS adhesives could someday replace UF resins in manufacturing and recycling of MDF without formaldehyde emission.
Keywords: Blocked pMDI; Citric acid; Cross-linking; Oxidized starch adhesives; Medium density fiberboard; Recycling
Contact information: a: Research Center for Biomaterials, Indonesian Institute of Sciences, Cibinong, Bogor, 16911, Indonesia; b: Department of Wood and Paper Science, Kyungpook National University, Daegu, 41566, Republic of Korea; c: Department of Research and Development, Sunchang Corporation, Incheon, 22302, Republic of Korea; *Corresponding author: byungdae@knu.ac.kr
INTRODUCTION
As medium density fiberboard (MDF) production is growing rapidly, waste MDFs are often disposed of using traditional methods such as burning or landfilling, which can harm the environment (Tatàno et al. 2009). Recently, advances have been made in the recycling of waste MDFs by converting them into cellulose nanocrystals (Couret et al. 2017; Gu et al. 2017; Irle et al. 2019). However, the process requires several chemical treatments to remove the extractives, hemicelluloses, and lignin before acid hydrolysis, which can increase the cost of recycling. Another means of recycling waste MDFs is to re-manufacture them, which mitigates the impacts on climate change, resources, and human health by ensuring that the carbon remains in the products (Oliveux et al. 2015; Morris 2017). However, it is difficult to remove cured urea-formaldehyde (UF) resins from waste MDFs (Grigsby et al. 2014; Lubis et al. 2018b). The remaining cured resins affect the properties of recycled fibers and recycled MDF (Roffael and Hüster 2012; Hong et al. 2018). Moreover, recycling of waste MDFs bonded with UF resins releases formaldehyde, which is toxic. Therefore, the search for alternative adhesives for the manufacture of MDF has become a topic of intensive research. For example, lignin, chitosan, and starch have been developed as alternative adhesives for MDF (Hoareau et al. 2006; Abbott et al. 2012; Hubbe et al. 2018). Because starch is generally cheaper and easier to isolate than other renewable adhesives, it offers greater potential as an alternative adhesive in such contexts.
Wood adhesives are integral to improving the efficient use of raw materials in the wood-based panel industry and to recycling the panels at the end of their service life (Frihart 2015). Researchers have focused on the development of bio-based adhesives because the most commonly used wood adhesives are synthesized from non-renewable resources and toxic, non-recyclable chemicals (He 2017). Among the renewable adhesives, starch has been used for bonding wood (Onusseit 1992), but the high water-affinity of the hydroxyl (–OH) groups of starch results in poor water resistance, slow setting, and high viscosity of starch-based adhesives (Imam et al. 1999; Glavas 2011). Esterification and grafting modifications can increase the water resistance of starch-based adhesives (Wang et al. 2011; Zia-ud-Din et al. 2017). However, the high viscosities of such modified starch-based adhesives limit their use, particularly in the spraying-blending techniques used in MDF manufacturing. Acceptable viscosities for adhesives used in MDF manufacturing are considered to be between 80 and 300 mPa·s (Park et al. 2001; Ji et al. 2018). A potential method for decreasing the viscosity of starch to a workable range is oxidation, which de-polymerizes starch by breaking α-(1→4)-glucosidic linkages and converts –OH groups into aldehyde (–CHO) and carboxyl (–COOH) groups (Rutenberg and Solarek 1984; Yu et al. 2017). Two oxidants are commonly used in the preparation of oxidized starch (OS): sodium hypochlorite and hydrogen peroxide (H2O2). Due to environmental issues related to the use of hypochlorite, peroxide oxidation is preferable (Tolvanen et al. 2009).
A cross-linker is often used to build a network structure and to accelerate the curing process of OS. Cross-linking can reinforce hydrogen bonds between starch molecules and eventually improve the mechanical strength and thermal stability of the OS (Imam et al. 1999; Sridach et al. 2013). Common cross-linkers used for OS include malonic acid, citric acid (CA), and other poly-carboxylic acids (Ghosh Dastidar and Netravali 2012; Shen et al. 2015). Among these acids, CA is preferable because it has three –COOH groups and could also promote adhesion by forming ester linkages with –OH groups of wood (Umemura and Kawai 2015; Del Menezzi et al. 2018). Moreover, a high functionality of polymeric diphenylmethane diisocyanate (pMDI) resin, i.e., blocked pMDI (B-pMDI), facilitates its use as a cross-linker for OS. The B-pMDI releases free –NCO groups after a de-blocking reaction, and then it can react with –CHO, –COOH, or –OH groups of the OS (Wicks and Wicks, Jr. 2001). The B-pMDI should be de-blocked at temperatures less than 100 °C, because the average core temperature in the preparation of wood-based panels during hot-pressing is in the range of 100 °C to 120 °C (Lubis et al. 2017). The free –NCO groups react with the OS to form amide or urethane linkages (Blagbrough et al. 1986; Stankovich et al. 2006). Therefore, OS cross-linked with B-pMDI could potentially be used as a wood adhesive in the manufacture of MDF.
An alternative adhesive should allow both good adhesion and easy disintegration of MDFs for their eventual recycling. Hence, this study aimed to optimize the preparation of OS adhesives with two cross-linkers (i.e., B-pMDI and CA) to achieve a balance between adhesion and disintegration properties to ensure recyclability of OS-bonded MDFs. Four degrees of oxidation (using different H2O2/starch molar ratios) and three levels of cross-linker were employed to tailor the performance of the OS adhesives. Properties of the OS adhesives were characterized by several techniques. The adhesion and recyclability of the OS-bonded MDFs were evaluated in detail. The optimum degree of oxidation, cross-linker type, and cross-linker content were obtained using statistical analysis.
EXPERIMENTAL
Materials
Corn starch (28.5% amylose, 18% moisture), hydrogen peroxide (30 wt%) as an oxidizer, copper sulfate (99 wt%) as a catalyst, sodium dodecyl sulfate (98 wt%) as a surfactant, and ammonium persulfate (99 wt%) as an anti-retrogradation agent were used to prepare the OS. Polyvinyl alcohol (PVA), molecular weight: 22,000 g/mol) was used as an adhesion promoter. Liquid pMDI (31% NCO), sodium bisulfate (58.5 wt%), and acetone (99.8 wt%) were used to prepare the B-pMDI (Lubis et al. 2017). The B-pMDI and CA (99.9 wt%) were used as cross-linkers. Urea granules (99 wt%), formalin (37 wt%), formic acid (20 wt%), and sodium hydroxide (20 wt%) were used to prepare UF resins as a control adhesive, and ammonium chloride (NH4Cl, 20 wt%) was used as a hardener for the UF resins. These chemicals were purchased from Duksan Pure Chemicals (Ansan, Korea) and Daejung Chemicals & Metals Co., Ltd. (Siheung, Korea). Virgin wood fibers made of red pine (Pinus densiflora Siebold & Zucc., 3% moisture content) were supplied from a commercial MDF mill (Sunchang, Incheon, Korea).
Synthesis of Oxidized Starch
Oxidized starches with different H2O2/starch molar ratios were prepared according to the previously published methods with some modifications (Zhang et al. 2015b; Lubis and Park 2018a). Briefly, a starch solution was prepared by mixing 200 g of corn starch (1.2345 mol) and 250 mL of distilled water. The solution was sonicated for 30 min at 25 °C and 20 kHz. The solution was then gelatinized at 80 °C for 1 h under continuous stirring at 100 rpm. The temperature was adjusted to 50 °C, after which 0.5% of copper sulfate, 0.5% of ammonium persulfate, and 0.25% of sodium dodecyl sulfate (based on the mass of the starch) were added to the gelatinized starch. Finally, 70.5 g of H2O2 (0.6221 mol) was added to obtain OS with an H2O2/starch molar ratio of 0.5. The oxidation reaction was performed by maintaining the temperature at 50 °C and a pH of 3 for 5 h under continuous stirring at 500 rpm. Oxidized starches with H2O2/starch molar ratios of 1.0, 1.5, and 2.0 were prepared using the same method as OS-0.5.
Preparation of OS Adhesives
The OS was first mixed with 15% PVA (based on the OS solids content) as an adhesion promoter, and the mixture was stirred for 2 h at 80 °C and 500 rpm. Further, two types of OS adhesives were prepared for the MDF panel manufacturing and recycling. The first adhesive was prepared by mixing four different H2O2/starch molar ratios of OS and B-pMDI (at levels of 5 wt%, 7.5 wt%, and 10 wt% based on the OS solids content), and the mixture was stirred for 10 min at 80 °C and 300 rpm. The second series of OS adhesives was prepared by adding CA at different levels (5 wt%, 7.5 wt%, and 10 wt% based on the OS solids content) to OS with four different H2O2/starch molar ratios, which were also stirred for 10 min at 80 °C and 300 rpm. As a control, UF resin adhesive with a final F/U molar ratio of 1.0 (61.7% solids content, 252 mPa·s viscosity, and 187 s gelation time) was synthesized using an alkaline-acid two-step procedure involving addition and condensation reactions, as reported in the literature (Park and Jeong 2011; Lubis and Park 2018b).
Characterization of OS Adhesives
The aldehyde and carboxyl contents of the OS were measured according to published methods (Smith 1967; Yu et al. 2017). To determine the carboxyl content, approximately 5 g of OS was mixed with 25 mL of 0.1 M HCl for 30 min. The sample was filtered and washed with distilled water. The filtrate was then mixed with 300 mL of distilled water, and the mixture was heated to 90 °C in a water bath for 15 min. The sample was titrated directly with 0.025 M NaOH until a pH of 8.3 was reached. For blank determination, 5 g of starch was evaluated using the same method. The measurements were conducted in triplicate.
The aldehyde content of the OS was used to determine the degree of oxidation (DO) of the OS. Approximately 2 g of the OS was mixed with 100 mL of distilled water and 15 mL of hydroxylamine reagent. The sample was heated to 40 °C in a water bath for 4 h and was then titrated directly to a pH of 3.2 with 0.1 M HCl. Furthermore, approximately 2 g of starch was evaluated using the same method for blank determination. The measurements were conducted in triplicate. The DO of the OS was calculated according to Eqs. 1 and 2 (Zhang et al. 2012),
where CNaOH and CHCl are the concentrations (mol/L) of NaOH and HCl used for titration, respectively; VNaOH and VHCl are the volumes (L) of NaOH and HCl used for titration, respectively; m is the weight (g) of the OS; 162 is the molar mass (g/mol) of the starch monomer (C6H10O5); DOs is the DO of a sample; and DOB is the DO of the blank sample.
Fourier-transform infrared (FTIR) spectroscopy (Frontier, PerkinElmer, Waltham, MA, USA) was performed to qualitatively investigate the functional groups of the OS with different H2O2/starch molar ratios. The OS powder was scanned in the range 4000 to 400 cm-1 in KBr pellets with a resolution of 4 cm-1 and 32 scans per sample. Furthermore, the functional groups of B-pMDI/OS and CA/OS after heating at 100 °C and 160 °C for 30 min were also investigated using FTIR spectroscopy to understand the curing reaction. The spectrum of each adhesive was normalized using the minimum–maximum normalization method (Wilcox 2014).
To ascertain the formation of aldehyde (–CHO) and carboxyl (–COOH) in the OS, the samples were analyzed using a liquid 13C nuclear magnetic resonance (NMR) spectrometer (Avance III 500, Billerica, MA, USA). The OS powder was dissolved in dimethyl sulfoxide (DMSO)-d6 at a concentration of 50 mg/mL and heated at 50 °C to form a clear solution. The sample was scanned 4,000 times at 75.45 MHz with 1 s of relaxation time and a 45° pulse, according to the published method (Lian et al. 2014).
The molecular weight distribution of the OS samples with different H2O2/starch molar ratios was determined using a gel permeation chromatography (GPC) system (GPC Breeze system, Waters, Milford, MA, USA) with a refractive index detector. The GPC samples were prepared by dissolving 50 mg of OS powder in high-purity distilled water. Prior to injection, samples were filtered through a 0.22-μm membrane filter. Universal calibration was performed using pullulan standards with molecular weights of 6,100 kDa to 642,000 kDa. The analytical condition was set at 30 °C, with a 0.8 mL/min flow rate of sodium nitrate (0.02 N NaNO3) as the mobile phase.
An X-ray diffractometer (XRD, D/Max-2500, Rigaku Miniflex II, Rigaku, Tokyo, Japan) with a CuKα radiation source (λ = 0.15406 nm) was used to analyze the crystal structure of the starch before and after oxidation. Prior to analysis, samples were dried in an oven at 105 °C for 3 h and powdered using a grinder. The XRD patterns of native starch and the OS were obtained by scanning the samples at room temperature in the range of 0° to 50°, with a step of 0.02°/min.
A differential scanning calorimetry (DSC) system (Discovery 25, TA Instruments, New Castle, DE, USA) was used to investigate the peak temperatures of the OS adhesives. Approximately 5 mg samples of liquid OS mixed with different contents of cross-linker were sealed in high-pressure capsule pans and then heated from 30 °C to 200 °C. Differential scanning calorimetry was performed at three different heating rates (2.5 °C/min, 5 °C/min, and 10 °C/min) under a nitrogen flow of 50 mL/min.
Properties of OS Adhesives
The solids contents of the OS adhesives were obtained by drying 2 g of a liquid sample in an oven at 105 °C for 3 h and dividing the oven-dried weight by the initial weight. The viscosity of the OS adhesives was determined using a cone-plate viscometer (DV-II+, Brookfield, Middleboro, MA, USA) with a No. 2 spindle at 60 rpm and 27 °C ± 2 °C. The gelation time of the OS adhesives was measured in boiling water using a gelation time meter (Davis Inotek Instrument, Charlotte, NC, USA). The OS was mixed with 5 wt% B-pMDI based on the OS solids content prior to gelation time measurement. Each experiment was repeated three times.
Preliminary Application in Plywood
Plywood was produced using radiata pine (Pinus radiata D. Don) veneers and two sets of OS adhesives. A calculated amount of glue (170 g/m2) was used to produce the three-ply plywood with a target size of 300 mm × 300 mm × 6 mm. The plywood was cold-pressed at 8.8 kPa for 20 min and subsequently hot-pressed at 120 °C and 0.8 MPa for 10 min. The plywood samples were conditioned at 20 °C for 2 d before cutting and testing.
Tensile shear strength (TSS) of the plywood was determined according to a standard procedure (KS F 3101 2016). The TSS values of ten specimens (80 mm × 25 mm × 6 mm) were evaluated using a universal testing machine (H50KS, Hounsfield, Redhill, UK) at a cross-head speed of 2 mm/min and 44,482 N of cell load. Multivariate statistical analysis was conducted to compare the plywood properties, and a Duncan multiple range test at α = 0.05 was applied to determine the optimum level of OS molar ratio, which was then used for MDF manufacturing. Statistical analysis was performed using SPPS 17 statistical software (SPSS Inc., Chicago, IL, USA).
Preparation of MDF panel
Laboratory MDFs were prepared using B-pMDI/OS, CA/OS, and UF resins. As a control, two MDF panels with 700 kg/m3 density were manufactured using UF resins according to published methods (Hong et al. 2017; Lubis et al. 2018a). Briefly, a mixture of 12% UF resin and 1% wax (based on the dried fiber mass) was mixed with 3% of hardener (based on the UF resin solids content). The adhesive was sprayed onto the fibers in a drum-type mechanical blender using an atomization nozzle with a diameter of 1.4 mm. The blended fibers were then used to prepare the fiber mat using a centrifugal air blower. The fiber mat was pre-pressed for 30 s and then hot-pressed at 180 °C using a single-phase pressing schedule of 7.85 MPa for 240 s (20 s/mm) to produce MDF panels (300 mm × 400 mm × 12 mm).
For the evaluation of the OS adhesives, six MDF panels with 700 kg/m3 density were prepared using B-pMDI/OS adhesive with various B-pMDI contents (5 wt%, 7.5 wt%, and 10 wt% based on the OS solids content). Approximately 12% of B-pMDI/OS based on the fiber dry mass was sprayed onto the fibers in a drum-type mechanical blender using an atomization nozzle. The fibers coated with the OS adhesives were then used to prepare the fiber mat using a centrifugal air blower. The fiber mat was pre-pressed for 30 s and then hot-pressed at 180 °C and 7.85 MPa for 600 s (50 s/mm) using a pre-programmed pressing schedule. By using the same method as that for the B-pMDI/OS samples, six MDF panels were prepared using the CA/OS adhesive with different contents of CA (5 wt%, 7.5 wt%, and 10 wt% based on the OS solids content). The resulting MDFs were conditioned for 2 d at 20 °C and 65% relative humidity prior to cutting and testing.
Determination of MDF Properties
Densities of the MDF panels were 731 kg/m3, 736 kg/m3, and 735 kg/m3 for the 5 wt%, 7.5 wt%, and 10 wt% B-pMDI/OS-bonded MDF. Quite similar densities, 730 kg/m3, 734 kg/m3, and 732 kg/m3, were obtained for MDF panels bonded with 5 wt%, 7.5 wt%, and 10 wt% CA/OS. Physical and mechanical properties of the MDF panels were determined using a standard procedure (KS F 3200 2016). For mechanical properties, ten samples (50 mm × 50 mm × 12 mm) were prepared to determine the internal bonding (IB) strength, and four samples (200 mm × 50 mm × 12 mm) were used to obtain the modulus of rupture (MOR) and modulus of elasticity (MOE). The IB strength, MOR, and MOE of the MDF panels were evaluated using the universal testing machine at cross-head speeds of 2 mm/min, 10 mm/min, and 10 mm/min, respectively. Further, ten samples (50 mm × 50 mm × 12 mm) were soaked for 24 h in water, and then their thicknesses and weights were measured after removing the surface water for the thickness swelling (TS) and water absorption (WA) tests. According to the desiccator method, ten MDF samples (150 mm × 50 mm × 12 mm) were placed in a desiccator for 24 h, and then UV–visible spectrophotometry (Optizen 3220 UV, Mecasys, Daejeon, Korea) was used at a wavelength of 412 nm to determine the formaldehyde emission (FE) values.
Evaluation of MDF Recyclability
MDF panels were cut into block specimens (20 mm × 20 mm × 12 mm). A total of forty-two MDF cubes were prepared from seven different MDF samples using UF resins, B-pMDI/OS (5 wt%, 7.5 wt%, and 10 wt% of B-pMDI based on the OS solids content), and CA/OS (5 wt%, 7.5 wt%, and 10 wt% of CA based on the OS solids content) as the adhesives. These blocks were hydrolyzed under two conditions, at 25 °C for 6 h and 80 °C for 1 h, in triplicate for each type of cross-linker. Information about the mass loss, degree of fiber disintegration, weight gain during hydrolysis, pH of the extract solution, and functional groups of the MDFs after hydrolysis was determined to evaluate the recyclability of the MDFs. The mass loss was calculated by dividing the weight of the MDF block after the hydrolysis by the weight before hydrolysis, while the degree of fiber disintegration was calculated by dividing the dry weight of the MDF block after hydrolysis by the dry weight of the MDF block before hydrolysis. The weight gain was calculated by weighing the MDF block during hydrolysis and dividing the mass of MDF block every 1 h for the hydrolysis at 25 °C and every 10 min for the hydrolysis at 80 °C. The functional groups of the solid residues and the hydrolyzates of the MDF blocks after hydrolysis were detected using FTIR spectroscopy from 400 cm-1 to 4000 cm-1 with a resolution of 4 cm-1 and 32 scans per sample. Hydrolyzate samples were prepared by freeze-drying the extract solution as described previously (Lubis and Park 2018b).
Statistical Analysis
Multivariate analysis was conducted to compare the mean values of the MDF properties at α = 0.05. When statistical differences were found, the Duncan multiple range test was applied to determine the optimum levels of B-pMDI and CA in the OS adhesives.
RESULTS AND DISCUSSION
Characteristics of OS
Oxidation causes de-polymerization of starch and introduces aldehyde and carboxyl groups into the OS. The FTIR spectra indicated the presence of C=O of –CHO at 1720 cm-1 and –COOH groups at 1600 cm-1 only in the OS (Fig. 1a) (Ye et al. 2018). This indicated that starch was successfully oxidized by converting the -OH groups on the C-2, C-3, and C-6 positions of the glucose unit to –CHO and –COOH groups using H2O2. The peak at 1640 cm-1 in both native starch and OS samples was attributed to the vibration of hydrogen bonded with C=O bonds (Ye et al. 2018). Further, the normalized intensity of –CHO and –COOH groups increased as a function of the H2O2/starch molar ratio. This implied that a greater DO was obtained for greater H2O2/starch molar ratios, as shown in Fig. 1b. In peroxide oxidation, the H2O2 was split into two OH radicals by Cu2+, which acted as a catalyst. The radical OH groups and catalyst then oxidized the OH of the starch into aldehyde groups and released H2O. Further, the remaining H2O2 converted the aldehyde groups into carboxyl groups (Parovuori et al. 1995). Therefore, it was clear that greater H2O2/starch molar ratios led to a greater DO due to the greater amount of H2O2 present in the system. The range of DO in this study was in agreement with the literature (Zhang et al. 2012).
Fig. 1. Characteristics of OS at different H2O2/starch molar ratios: (a) FTIR spectra and (b) DO of the OS
Aldehyde (–CHO) and carboxyl (–COOH) groups in the OS were also detected in the 13C-NMR spectra (Fig. 2a). The chemical shifts of the carbons of native starch were identified at δ 100.7 ppm for C-1; δ 79.2 ppm for C-4; δ 73.8 ppm for C-3, δ 72.4 ppm for C-2, δ 70.0 ppm for C-5; and then δ 60.4 ppm for C-6 (Guo et al. 2016; Yu et al. 2017). The peak intensities of C-1 and C-4 in the OS decreased when increasing the H2O2/starch molar ratio. This was due to de-polymerization of starch mainly occurring at α-(1→4)-glucosidic linkages. Furthermore, the peaks of C-2, C-3, and C-6 notably decreased at greater H2O2/starch molar ratios, implying oxidation at these positions. Moreover, new peaks of –CHO groups appeared at δ 162.6 ppm and δ 164.5 ppm. However, owing to a high signal-to-noise ratio, a small peak of -COOH groups at δ 175.2 ppm was detected for the OS. This peak only appeared at the greater H2O2/starch molar ratio, indicating that –COOH groups were mainly formed via conversion of –CHO groups to –COOH groups by the excess H2O2 during oxidation (Parovuori et al. 1995).
Fig. 2. (a) 13C-NMR spectra of OS at different molar ratios and (b) the XRD patterns of OS at different H2O2/starch molar ratios
The XRD patterns of the native starch and the OS are presented in Fig. 2b at H2O2/starch molar ratios. The X-ray diffractrogram of the native corn starch showed peaks at 2θ values of 15.0°, 17.0°, 17.9°, 19.9°, and 22.8°, which indicated that the native corn starch had an A-type cryutsalline structure (Zhang et al. 2009, 2012). However, the oxidation chnaged the XRD patterns. The crystalline structure became amorphous even at lower H2O2/starch molar ratios. The oxidation destroyed the crystalline region of the starch by breaking down the α-(1→4)-glucosidic linkages, which eventually produced smaller starch molecules. Further, oxidation at greater H2O2/starch molar ratios changed the XRD pattern to that of a more amorphous phase.
Functional groups of the B-pMDI/OS and CA/OS adhesives were also evaluated using FTIR spectroscopy after heating at 100 °C for 30 min and at 160 °C for 30 min. As displayed in Fig. 3a, the reaction of OS with B-pMDI or CA at 100 °C for 30 min resulted in a new peak at 1640 cm-1. This peak corresponded to the vibration of hydrogen bonded with C=O of –NCO groups of B-pMDI or –COOH of CA due to the diffusion of molecules toward each other and a phase change of the adhesive from liquid to gel (Upadhyay 2006; Zhang et al. 2010). The presence of B-pMDI in the OS was detected by the formation of C=C of aromatic pMDI at 1560 cm-1. In addition, the normalized intensity of C=O of –CHO and –COOH groups at 1720 cm-1 changed with the addition of cross-linkers, but no chemical reaction was observed at this stage. Peaks of around 2924 cm-1 and 3300 cm-1 were detected for C-H and N-H for OS/B-pMDI, and C-H and O-H for OS/CA, respectively. All functional groups in the spectra after heating at 100 °C for 30 min were attributed to the first exothermic peak temperature (Tp1) of OS adhesives shown by the DSC (Fig. 6).
Fig. 3. Typical FTIR spectra of OS-1.0 with 5 wt% cross-linker after heating at (a) 100 °C for 30 min and (b) 160 °C for 30 min
After reacting the OS with the cross-linkers at 160 °C for 30 min (Fig. 3b), B-pMDI/OS produced peaks for C=O of amide linkages at 1590 cm-1 and C-N of amide at 1350 cm-1, while CA/OS showed a peak for C=O of ester linkages at 1720 cm-1. Notably, the peaks of C=O of ester and aldehyde groups overlapped at 1720 cm-1. However, a new peak of C-O at 1220 cm-1, which was only detected in the CA/OS, indicated the formation of ester linkages in the CA/OS. This result could be due to the –NCO of B-pMDI reacting with the –COOH of OS to form amide (R-NHCO-O-CO-R) linkages (Blagbrough et al. 1986), and the –OH of –COOH of poly-carboxylic acid could have reacted with the –OH or –COOH of OS to form ester (R-COO-R) linkages (Ghosh Dastidar and Netravali 2012). The functional groups in the spectra after heating at 160 °C for 30 min were attributed to the second exothermic peak temperature (Tp2) of the OS adhesives shown by the DSC (Fig. 6). The possible cross-linking reactions of the OS adhesives for the two types of cross-linkers are presented in Fig. 4.
Fig. 4. Possible cross-linking reactions of (a) B-pMDI/OS and (b) CA/OS
Oxidation also affected the solids content, viscosity, and gelation time of the OS adhesives (Table 1). The solids content and viscosity of the OS decreased as the H2O2/starch molar ratio increased, while gelation time increased. This result meant that a greater H2O2/starch molar ratio decreased the reactivity of the OS adhesive. The excess of H2O2 at a greater molar ratio probably remained in the OS, which decreased the solids content and viscosity of the OS and eventually increased the gelation time. Lower viscosity of the adhesive generally increased the gelation time due to the adhesive system needing more time to evaporate the water and solvent. This phenomenon happened to the OS adhesives with greater H2O2/starch molar ratios, corresponding to greater amounts of H2O2 in the system.
Table 1. Solids Content, Viscosity, Gelation Time, and Molecular Weights of OS for Different H2O2/Starch Molar Ratios
Mw: weight average molecular weight, Mn: number average molecular weight, PDI: polydispersity index
The molecular weight distributions of the OS for different H2O2/starch molar ratios are presented in Table 1. The Mw, Mn, and PDI of the OS decreased as the H2O2/starch molar ratio increased. The Mw of the native starch has been reported to be approximately 800,000 g/mol (Zhang et al. 2015a), and the Mw decreased by a factor of 100 after oxidation. It was evident that the oxidation de-polymerized the starch macromolecules into smaller molecules and eventually decreased the molecular weight.
Thermal Properties of the OS
The OS adhesives mixed with two different cross-linkers were characterized using DSC; their DSC thermograms are displayed in Fig. 5. As depicted, the exothermic peak temperature (Tp) values of the OS adhesives increased as the heating rate increased. Moreover, two Tp values were detected in both the B-pMDI/OS and CA/OS adhesives. The first Tp (asterisk, Tp1) at the lower temperature could be attributed to the diffusion of molecules toward each other and the phase change of the adhesive from liquid to gel (Upadhyay 2006; Zhang et al. 2010). The second Tp (arrow, Tp2) at greater temperatures could be due to cross-linking between the –OH of OS and –NCO groups of B-pMDI, or –OH of OS with –COOH of CA. At 2.5°/min, OS/B-pMDI and OS/CA had around 79.8 °C of Tp1, and the temperature increased to 95.7 °C at 10°/min for OS/B-pMDI and 104.5 °C for OS/CA. Similar trend was observed for Tp2 of OS/B-pMDI and OS/CA.
As presented in Fig. 5, the B-pMDI/OS generally led to lower Tp1 values than did the CA/OS, along with lower Tp2 values than those of the CA/OS, indicating greater reactivity of B-pMDI than that of CA. The Tp values of the OS adhesives decreased as the content of cross-linker increased, indicating that a greater amount of cross-linker accelerated the cross-linking reaction. Greater H2O2/starch molar ratios produced lower Tp values than those obtained with lower molar ratios. This result could be due to greater H2O2/starch molar ratios providing more carboxyl groups for cross-linking than lower molar ratios. Regardless of the type of cross-linker, the Tp values of the OS decreased up to a 1.5 H2O2/starch molar ratio and then increased at an H2O2/starch molar ratio of 2.0. This behavior was likely due to the excess of H2O2 in OS-2.0 decreasing the OS viscosity markedly by approximately 60% compared to that of OS-0.5, resulting in the OS-2.0 requiring greater temperatures for cross-linking.