Bio-Based Wood Adhesive from Camelina Protein (a Biodiesel Residue) and Depolymerized Lignin with Improved Water Resistance.

The aim of this study was to improve water resistance of camelina protein (CP) for wood adhesives with depolymerized lignin. Kraft lignin was depolymerized by H2O2-induced oxidation in the presence of ultrasound (US) irradiation to reduce lignin's particle size and thermal stability and increase the hydroxyl group. Coupling with depolymerized lignin camelina protein exhibited increased hydrophobicity. Fluorescence spectroscopy analysis revealed that the oxidation treatment of lignin further stimulated the hydrophobization effect of the protein-lignin copolymer due to the increased reactivity of depolymerized lignin to camelina protein. Accordingly, the water resistance of CP-lignin adhesives was significantly improved. When copolymerized with US-induced oxidized lignin, the camelina protein had increased wet shear adhesion strength from 0.28 to 1.43 MPa, with wood panels passing the three-cycle water-soaking test. The CP resin, with depolymerized lignin as an economical, green, and bio-based hydrophobic enhancer, provided an alternative to the petroleum-based and other edible protein-based adhesives, such as soy protein.

Introduction

Plant proteins, such as soybean protein, have been modified and commercially available for bio-based adhesives and coatings. However, soybean proteins are also important human diets worldwide; therefore, the competition between food and nonfood uses of edible proteins urged us to find alternative nonedible protein sources. Camelina is an easy-growing oilseed crop on dry land; in the past few decades, it has drawn increasing attention due to its abundance in polyunsaturated fatty acids, which were known for aviation fuels and health benefits. Accordingly, the ever-accumulating interest for camelina oil is generating increasing desired amounts of camelina meal (CM) that contains 40% crude protein. However, some potential antinutritional toxic compounds in camelina restricted the food uses of the protein extracted from CM. Therefore, camelina protein (CP) becomes a suitable nonedible protein source for bio-based products, such as adhesives.

Plant proteins contain similar amino acids in terms of content; however, they have different amino acid sequences that make the proteins different from each other. Therefore, the modification method good for soybean protein is not necessarily suitable for camelina protein. As a relatively new protein source, there are only limited studies on the modification and application of CP. We previously elucidated the unfolding behaviors of CP by disrupting hydrogen bond and disulfide bonds.[1] By utilizing the cysteine-induced denaturation, Zhao et al.[2] developed CP-based film with improved physicochemical properties for cell culture. Reddy et al.[3] grafted camelina with vinyl monomers to modify the thermoplastic properties. However, similar to other plant protein polymers, there are two major concerns regarding CP’s adhesive application: (1) low mechanical strength, which is due to the weak protein–protein interaction and (2) poor water resistance caused by protein’s hydrophilic nature. To resolve these problems, scientists usually applied classic protein modification techniques, such as unfolding, cross-linking, and side-group grafting, to increase the protein’s hydrophobicity, cross-linking degree, and reactivity.[4] Regarding plant protein-based wood adhesives, such as soy, adding proper hydrophobic additives seems to be most effective and convenient to increase the adhesion properties of the cured protein resins, especially under aqueous conditions.[5,6] Those hydrophobic materials, such as polyvinyl acetate,[7] polyamidoamine-epichlorohydrin,[8] and polyurethane,[9,10] would stabilize the protein structure during the curing progress and contribute to the water resistance of adhesives. Unfortunately, high proportions of toxic additives in the protein-based adhesives significantly destroyed the biodegradability and environment-friendly features of biopolymers. Besides the previously mentioned chemicals, to date, there are growing interests in developing bio-based hydrophobic enhancers for polymer fabrication. Lignin, the most abundant and underutilized biopolymer, exhibited its potential in improving the mechanical, thermal, and hydrophobic properties. From the chemists’ point of view, lignin’s abundant aromatic structures, phenolic hydroxyls, and tough structures should be appropriate to act as copolymers to modify protein’s hydrophobicity, reactivity, and mechanical properties. Thus, the development of lignin-reinforced protein adhesive could meet the expectation for both environment friendliness and water resistance.

In spite of the appealing performance, lignin’s inert nature limits its reactivity and solubility and hinders its further applications; thus, functionalization treatment is necessary. Depolymerization techniques, including pyrolysis, oxidation, enzymatic and ionic liquids treatment, are effective to break down lignin’s heterogeneous structures, release functional groups, and increase reactivity.[11−13] Compared to other lignin depolymerization techniques that require harsh temperature condition or complicated ionic solvents, oxidative depolymerization of lignin under mild and solvent-free reaction conditions should be considered because of its eco-friendly nature and practicality. Hydrogen peroxide (H2O2) was widely used to induce such depolymerization. Tortora et al. utilized H2O2 oxidation to prepare lignin-based microcapsules for drug delivery,[14] whereas many others used this approach to prepare soluble lignin fragments with reactive moiety.[11,15,16] For a higher oxidation efficiency of H2O2, ultrasound (US) technique, a green and efficient means, was utilized to increase free radicals that accelerated the oxidation.[17] In addition, the cavitation force generated by US could also fragment lignin substrates into nanoparticles.[18]

In recent years, development of lignin-based biopolymers has become attractive. However, to the best of our knowledge, little is known about the interaction between plant proteins and depolymerized lignin. Although lignin has been widely used as a hydrophobic enhancer to improve water resistance of protein polymers, the partially depolymerized lignin in this research had further benefits when it was applied in protein-based wood adhesives. At high pH (strong base condition), lignin was dissolved and protonated with increased reactivity; however, in our previous research, we found that soy protein under the base condition exhibited significantly reduced adhesion. However, under neutral and acidic conditions, soy protein had strong adhesion but lignin was insoluble and thus the interaction between soy protein and lignin was limited.[19,20] We observed similar phenomena between camelina protein and lignin in our preliminary experiments. The goal of this research was to improve the interaction of lignin with camelina protiens (CP) by partially depolymerized lignin through lignin oxidation. We believed that the depolymerized lignin would increase its solubility in and miscibility with CP at neutral pH, and consequently, the wet adhesion strength of CP–lignin adhesive would improve.

Our aim was to develop an eco-friendly wood adhesive from the low-value nonedible camelina proteins and improve its water resistance by copolymerizing it with functionalized lignin. Lignin was depolymerized by (ultrasound-assisted) H2O2 oxidation. The modified lignin was analyzed by dynamic light scattering (DLS), transmission electron microscopy (TEM), Fourier transform infrared (FTIR) spectroscopy, and thermogravimetric analysis (TGA) to examine the depolymerization effect. After copolymerization, the resultant adhesives were characterized for hydrophobicity, rheological properties, and wood adhesion strength. Once optimized, this approach provides a green and facile way to stimulate lignin and protein interaction and adds marketable values to camelina meals and lignin.

Results and Discussion

Chemical Reaction Pathways

The proposed chemical reactions of lignin and protein are presented in Schemes and . In the presence of H2O2, the oxidation reaction of kraft lignin would increase its surface area and −OH groups and reduce the particle size. Scheme describes reinforced intermolecular interaction between camelina protein and depolymerized lignin. Initially, the pristine lignin exhibited low reactivity and miscibility with camelina protein due to its hydrophobic nature. It is hypothesized that after the oxidation treatment the reduced particle size and increased functional group of depolymerized lignin would stimulate its conjugation with camelina protein in the presence of poly(ethylene glycol) diglycidyl ether (PEGDE); then, the stabilized copolymer would obtain increased cohesion and hydrophobicity, which further enhances its adhesion strength, especially under the wet condition. Scheme presents the interaction between the wood and copolymer adhesives. The newly appeared −OH groups at the surface of oxidized lignin (OL) molecules would react with the functional group (mostly −OH and −COOH) at wood surface, which contributes to the interfacial adhesion strength between the wood and adhesives. In brief, the camelina protein and oxidized lignin copolymer are expected to increase hydrophobicity and enhance cohesion and interfacial adhesion strength; consequently, the water resistance of its adhesive would be improved.

Scheme 1

Reaction between the Protein and (Oxidized) Lignin

Scheme 2

Protein–Lignin Adhesives Bonding between Wood Panels

Particle Size and Morphology

Particle size is an important parameter to reveal the effect of depolymerization on lignin, which further influences lignin’s reactivity and compatibility with proteins. The effect of oxidation treatments on volume-mean diameter (D43) of lignins and their distribution profiles are shown in Figure a. The heat-induced oxidation reaction reduced the particle size (D43) of lignin significantly from 26 to 3.90 μm and further to 2.59 μm with ultrasound treatment. Three major possible reaction pathways (Figure S1)[21−23] have been proposed about the fragmentation effect on lignin structures by H2O2-induced oxidation. By dealkylation, the C–C and C–O bonds of lignin were first cleaved. Then, hydroxylation and oxidation reactions further increased the amphipathic property of lignin by introducing hydroxyl groups. As a consequence, the depolymerized lignin obtained reduced particle size and thus increased reactivity. Ultrasound treatment led to further particle fragmentation through a physical “cavitation” phenomenon. Locally high pressure and temperature generated free radicals with high energy in the solution and thus lignin decomposed. Napoly et al. also reported and explained the cavitation mechanism of ultrasound-driven particle size reduction of kraft lignin.[17]

Figure 1

Characterization of depolymerized lignin. (a) Particle size distribution of native lignin, oxidized lignin (OL), and ultrasound-induced oxidized lignin (UL). (b) TEM images of (b1) native lignin, (b2) oxidized lignin (OL), (b3) ultrasound-induced oxidized lignin (UL), (b4) zoom-in morphology of OL, and (b5) zoom-in morphology of UL. (c) FTIR spectra of native lignin, oxidized lignin (OL), and ultrasound-induced oxidized lignin (UL) between 4000 and 400 cm–1. (d) Derivative thermogravimetry (DTG) curves of lignin, oxidized lignin (OL), and ultrasound-induced oxidized lignin (UL).

To obtain more morphological details of the depolymerized lignin particles, TEM was used to observe the microstructures of OL and UL in comparison to the native lignin. In consensus with the particle size distribution profiles, native lignin turned into the smaller-particle OL and UL after oxidation reaction (Figure b). Native lignin was a big chunklike particle with no porous or fibrous structure, which was due to lignin’s highly cross-linked structure. Unlike the other biopolymers, such as proteins or gums, whose structures are highly dependent on intermolecular noncovalent bonds,[4] lignin’s phenyl rings are covalently bonded by C–C or C–O. Lignin’s condensed structure results in limited surface area and chemically inert nature. After oxidative depolymerization, the lignin particle was fragmented into small pieces like a shattered rock. The average dimension of the newly formed OL was about 2 μm, which was consistent with DLS results. When applying ultrasound to induce the oxidation reaction, in addition to the H2O2-induced dealkylation effect, the cavitation force further reduced the particle size of lignin by disrupting its intermolecular hydrogen bond and thus UL exhibited more fragmented morphology than OL.

FTIR

FTIR analysis was used to qualitatively examine the changes in lignin’s functional groups. In Figure c, a broad peak was observed ranging from 3050 to 3500 cm–1, which was assigned to the stretching vibration of the hydroxyl group (both aromatic and aliphatic hydroxyl groups) for both pristine and modified lignin, and this peak for both OL and UL was stronger than that for the native lignin. This was due to the increased hydroxyl groups generated by dealkylation and hydroxylation. There are also two peaks at 2864 and 2937 cm–1, assigned to the methoxyl group linked with C–H band. Thus, the band intensity slightly decreased for OL and UL at 2864 cm–1, indicating the removal of methoxyl groups from lignin’s aromatic ring. In addition, the band around 1700 cm–1 was assigned to native lignin’s C=O stretching of acetyl group; however, it significantly reduced for both OL and UL. This C=O group consumption may be due to the repolymerization effect of lignin fragments by H2O2-induced oxidation. Such phenomenon was also reported by other studies on lignin’s oxidative depolymerization.[18,24]

More distinct difference in band shapes can be observed in the fingerprint region (Figure S2). The band intensities increased for OL and UL at 1590 and 1510 cm–1, respectively, which denoted the C=C stretching of the aromatic ring. Bands at 1260 and 1330 cm–1 are attributed to the C–O stretch in lignin for guaiacyl groups and syringyl groups, respectively. For OL and UL, the band intensity at 1268 cm–1 decreased significantly, whereas they were relatively stable at 1215 cm–1. This result indicates that H2O2-induced oxidation had more obvious effects on decomposing guiacyl structures than on syringyl moiety.[17] Thus, by examining the IR spectra, we observed lignin’s three oxidation pathways in the presence of H2O2: hydroxylation, dealkylation, and oxidative coupling (Figure S1).

Thermogravimetric Analysis (TGA)

Derivative thermogravimetry (DTG) curves of native pan class="Chemical">lignin, OL, and UL are shown in Figure d. Native lignin exhibited one big thermal derivate weight loss peak ranging from 250 to 500 °C, whose maximum degradation rate was at 385 °C. Interestingly, after the oxidation and ultrasound treatment, besides the peak at 385 °C, there was a new peak for the DTG curves of OL and UL at about 310 °C. These differences can be explained by the varied stability of lignin chemical bonds. There are three major chemical bonds for lignin’s complex structures: lateral chain, C–C bonds, and aromatic rings. Due to their different chemical stabilities, they exhibited different decomposition temperatures: the lateral chain decomposed at around 245 °C, C–C bond decomposed at around 310 °C, and the aromatic ring with the highest stability decomposed from 350 to 410 °C.[25] Thus, the broad peak of native lignin’s DTG curve was due to the decomposition of these three bonds. For OL and UL, the newly appeared peak at 310 °C was assigned to the C–C bond degradation, and we can still find the aromatic ring degradation peak at about 385 °C. As the oxidation and ultrasound broke down and unfolded the rigid structure of OL and UL, their C–C bond became more exposed than the native lignin for heat decomposition.

Fluorescence

To verify the conjugating effect of CP on different lignins, the hydrophobicity of CP–lignin was evaluated by detecting the 1-anilino-8-naphthalenesulfonate (ANS) binding capacity. Lignin, with a large amount of aromatic groups, contributed to CP’s hydrophobicity, which can be determined by the fluorescence intensity (FI) in the presence of ANS.[1] Thus, increased FI indicated enhanced protein–lignin interaction. As expected, P–CP exhibited the lowest FI (Figure ). In addition, the FI of depolymerized lignin-based copolymers (P–CPOL, P–CPUL) was higher than that of P–CPL, suggesting that the depolymerized lignin was more cross-linked with CP than the native lignin. As previously discussed, lignin was functionalized by the oxidation reaction with increased reactivity and decreased particle size, which facilitated the protein–lignin interaction. Compared to heat-induced oxidation, ultrasound treatment stimulated the depolymerization effect by further reducing lignin’s particle size to induce increased reactivity (Figure a,b). Accordingly, the highest FI was obtained by P–CPUL, indicating the strongest CP–lignin interaction, which further improved the water resistance of the resultant adhesive.

Figure 2

Fluorescence intensity of different protein–lignin dispersions of P–CP, P–CPL, P–CPOL, and P–CPUL.

Rheological Properties

Figure shows the storage modulus (G′) of different CP–lignin copolymers as a function of frequency sweep. For all of the CP–lignin dispersions, G′ increased linearly at a frequency region ranging from 0.1 to 3 PaS and after that G′ exhibited a nonlinear sharp increase. This result was consistent with our previous frequency sweep study of CP.[1] As described by Tunick, the frequency at the boundary point between the linear and sharp-increase regions was defined as critical frequency (fc). After fc point, the protein polymer structure was broken down and thus the sharply increased G′ after fc was not used to evaluate the protein polymers’ intermolecular interactions.[26−28] Due to different degrees of protein–lignin polymerization, their G′ exhibited different frequency dependences and different fc values. Before fc point, protein polymers with strong intermolecular interaction exhibited low frequency dependence and high fc frequency, and accordingly, high storage modulus was observed.[27] To better describe the protein polymer’s G′ variation, in Table , we recorded fc and the G′ modulus at fc (Mfc).

Figure 3

Rheological properties of CP–lignin adhesives. Variation of storage modulus of different protein–lignin dispersions with frequency sweep.

Table 1

Rheological Properties of Different CP–Lignin Dispersions for Frequency Sweep

copolymer	critical frequency fc (PaS)a	modulus at fc (Pa)b
CP	1.00	6.5
P–CP	2.03	10.1
P–CPL	1.19	11.6
P–CPOL	2.42	22.1
P–CPUL	3.46	29.0

The frequency at which protein structure broke down and G′ began to increase sharply.

The modulus of the protein at the critical frequency.

Rhepan class="Chemical">ological properties of CP–lignin adhesives. Variation of storage modulus of different protein–lignin dispersions with frequency sweep.

CP exhibited the weakest protein interaction, whose structure was broken at 1.00 PaS with the lowest G′ of 5.52 Pa. After the addition of PEGDE, the structure of CP was stabilized, and the fc value of P–CP increased to 2.03 PaS with increased Mfc at 10.1 Pa. However, the fc value of P–CPL decreased to 1.19 PaS, indicating that the lignin incorporation destabilized P–CP’s network. As a highly cross-linked macromolecule, lignin’s small surface-to-volume ratio leads to its inert nature and low reactivity with other biomacromolecules. Only a small amount of lignin would be copolymerized with CP, and the unreacted lignin particles would weaken the protein polymerization due to the steric hindrance effect. After oxidation, the weakening effect of lignin disappeared for P–CPOL and P–CPUL. The fc value and storage modulus of both P–CPOL and P–CPUL increased significantly. The fc and Mfc of P–CPOL reached 2.42 PaS and 22.1 Pa, and those of P–CPUL reached 3.46 PaS and 29.0 Pa, respectively. Thus, the increased functionality and surface-to-volume ratio of depolymerized lignins (OL and UL) induced their stronger interaction with camelina protein than the native lignin. This effect also contributed to the adhesion properties of protein–lignin dispersions.

Adhesion Properties of Camelina Protein

Similar to other plant proteins, CP molecules were unfolded in the solution, and their free side groups from the unfolded polypeptide, such as amino groups, carboxyl groups, and hydroxyl groups, formed strong chemical bond with the available hydroxyl groups on the wood surface under hot press condition. This process generated interfacial adhesion between the wood surface and protein adhesives; in addition, the protein molecules were also cross-linked and entangled with each other, thus providing cohesion strength to the adhesives.

Figure shows the dry shear strength and wet shear strength (refers to water resistance) of the CP-based adhesives polymerized with different lignins. The dry shear strength of the unmodified camelina protein was 1.68 Pa. In the presence of PEGDE alone (P–CP), the adhesion strength increased to 2.13 MPa, followed by a small decrease to 1.87 MPa for P–CPL. This is consistent with the results of frequency sweep that pointed the weakening effect of pristine lignin particles to the CP molecular interaction. When the lignin was depolymerized, the dry shear strength of P–CPOL and P–CPUL increased to 2.28 and 2.35 MPa, respectively.

Figure 4

Dry shear strength and wet shear strength of different adhesive samples.

Regarding the wet strength, similar to soy protein, CP-based adhesives are not water-resistant as their adhesion strength decreased drastically after water soaking. The unmodified CP had the lowest wet adhesion strength at 0.2 MPa. After the addition of PEGDE cross-linking agent, camelina protein structure was cross-linked, and its wet shear strength increased significantly to 0.84 MPa for P–CP. The addition of lignin induced further increase in the water resistance of P–CPL to 0.95. Although the unmodified lignin interfered with protein’s cross-linking effect and reduced the dry shear strength of adhesives, it increased the wet strength of CP adhesives as a hydrophobic enhancer. This strengthening effect of native lignin was widely utilized to improve the water resistance of hydrophilic biopolymers. Luo et al.[29] developed a soy meal-based wood adhesive with improved water resistance using kraft lignin. Huang et al. investigated the effect of alkaline lignin and lignosulfonate on soy protein-based composite for enhanced mechanical performance and aqueous absorption. They found that, as a hydrophobic filler, the alkaline lignin increased the tensile strength, thermal stability, and water resistance of soy protein,[30] whereas lignosulfonate contributed to soy protein’s microphase separation and the formation of cross-linked structures.[31]

Instead of simply mixing the lignin and protein, functionalized lignin further increased the protein’s water resistance. In this research, the wet shear strength of P–CPOL and P–CPUL reached 1.35 and 1.43 MPa, respectively. Consistent with the results of wet strength, although a proportional CP, P–CP, and P–CPL failed a three-cycle water-soaking test, all of the specimens of P–CPOL and P–CPUL passed (Table ). The pristine CP exhibited major slippage (3.8 in.). After the copolymerization, the delamination degrees of P–CP (1.72 in.), P–CPL (1.86 in.), and P–CPOL (0.78 in.) decreased significantly. For P–CPUL, only minimal delamination was observed (0.18 in.), indicating its strongest water resistance. As we hypothesized in Schemes and , compared to native lignin, depolymerized lignins (PL and UL) contain newly formed hydroxyl groups that were able to form stronger hydrogen bonds and covalent bonds with the polar groups of camelina protein, which increased the cohesion of the adhesives. In addition, these hydroxyl groups increased the adhesion strength by stimulating the interfacial bonding between the wood surface and protein adhesives[32,33] (details in Schemes and ).

Table 2

Three-Cycle Water-Soaking Test Evaluation of Different CP–Lignin Adhesives*

	wood delamination length after soaking (in.)
	% pass the criteria
samples	1st round	2nd round	3rd round
CP	2.04 (±0.31)a	2.81 (±0.64)a	3.80 (±0.49)a
	50%	25%	0%
P–CP	0.67 (±0.15)a	1.15 (±0.26)b	1.72 (±0.12)b
	100%	100%	100%
P–CPL	0.52 (±0.45)b	1.26 (±0.50)b	1.86 (±0.80)bc
	100%	100%	75%
P–CPOL	0.08 (±0.07)b	0.3 (±0.25)c	0.78 (±0.36)cd
	100%	100%	100%
P–CPUL	0 (±0)b	0.03 (±0.03)c	0.18 (±0.12)d
	100%	100%	100%

Letters indicate significant (p < 0.05) difference within the same column.

Letters indicate significant (p < 0.05) pan class="Chemical">difference within the same column.

For comparison purposes, we duplicated the most commonly used adhesives in our lab, including polymeric amine epichlorohydrine (PAE)- or NaHSO3-modified soy protein-based adhesives and urea-formaldehyde (UF)-based adhesives (Table ). The wet strength of plywood bonded with the newly developed CP–lignin adhesive was 3.8 times higher than that bonded with commercial UF-based adhesives, 1.8 times higher than that bonded with NaHSO3-modified soy protein isolate (SPI), and 13.6% higher than that bonded with the polymeric amine epichlorohydrine (PAE)-modified SPI adhesives. The comparison data indicated that the hydrophobic chemical structures from lignin had high efficiency to improve the water resistance of CP-based adhesives.

Table 3

Comparison of the Water Resistance of Different Adhesives

adhesives	wet strength (MPa)	note
urea-formaldehyde	0.40 ± 0.12	toxic
NaHSO3-modified soy proteina	0.75 ± 0.09	bio-based/food sources
PAE-modified soy proteinb	1.25 ± 0.23	bio-based/nonfood sources
CP–lignin copolymer	1.42 ± 0.20	bio-based/nonfood sources

Soy protein solution was prepared at 10%, and the NaHSO3 content of soy protein was 3%.

PAE content of soy protein was 10%.

Soy protein solution was prepared at 10%, and the pan class="Chemical">NaHSO3 content of soy protein was 3%.

Conclusions

Camelina protein copolymerized with oxidized kraft lignin improved water resistance of wood adhesives. To stimulate the protein–lignin interaction, lignin was depolymerized by H2O2-induced oxidation under elevated temperature (50 °C) and ultrasound irradiation separately. The modified lignin exhibited increased hydroxyl groups and reduced particle size. The oxidation induced by ultrasound irradiation treatment was more effective to induce the decomposition of lignin structure and increase its reactivity. Accordingly, the adhesive derived from camelina protein copolymerized with oxidized kraft lignin induced by ultrasound irradiation (P–CPUL) had the strongest intermolecular interaction and highest water resistance. The nonedible camelina protein copolymerized with oxidized kraft lignin has potential for industrial adhesive applications, which would benefit camelina oil-based energy industry as well as paper industry.

Experimental Section

Materials

Defatted camelina meal (DCM) with 32.4% crude protein (db) and 11% moisture content (db) was provided by Field Brothers Inc. (Pendroy, MT). Crude protein intake was separated from DCM using the method described by Li et al., with some modifications.[33] Lignin (alkali), poly(ethylene glycol) diglycidyl ether (PEGDE; Mn 500), hydrochloric acid (HCl), sodium hydroxide (NaOH), and urea-formaldehyde were purchased from Sigma-Aldrich (St. Louis, MO). Hydrogen peroxide and sodium bisulfite were purchased from Fisher Scientific (Fair Lawn, NJ). Polymeric amine epichlorohydrine (PAE) was provided by Wuhu Hangchen Trading Co. Ltd (Wuhu, China). PAE is in aqueous solution at 12.7% solid content and pH 3.9. Soy protein was obtained using the method described by Liu et al.[6] SPI was precipitated at pH 4.5, redissolved at pH 7.6, and then freeze-dried (freeze dryer Model 6211-0459; The VirTis Company, Inc., Gardiner, NY). The dried SPI was milled into powder, with 95% passed through U.S. #100 mesh. The freeze-dried SPI powder sample had a protein content of 85.6% and a moisture content of 8%. Yellow pine wood veneers with dimensions of 305.50 mm × 305.50 mm × 3.26 mm (width × length × thickness) were provided by Veneer One (Oceanside, NY).

Lignin Depolymerization

To prepare the oxidized lignin (OL) solution, 2 g of lignin substrate was dispersed into 100 mL of distilled water, with pH adjusted to 10.5 by adding NaOH solution (3 M) and gently stirring for 30 min. Then, 1 mL of hydrogen peroxide was added. After conditioning for 10 min, the solution was heated at 50 °C for 16 h to induce the oxidation reaction. To prepare the ultrasound-induced oxidized lignin (UL) solution, an ultrasound processor model VCF 1500 (Sonics & Materials Inc., Newtown, CT) was used to sonicate the lignin dispersion instead of heat treatment to induce oxidation. Sample solutions were treated at 20 kHz and 480 W for 60 min (pulse duration of on-time 5 s and off-time 5 s). The resultant OL solution and UL solution had a pH of 7.2, and they were further used for the adhesive preparation and particle size characterization. Their lyophilized samples were used for IR and TGA tests.

Characterization of Depolymerized Lignin

Particle Size

The particle size of lignin was characterized by both dynamic light scattering (DLS) and transmission electron microscopy (TEM). Native lignin, OL, and UL solutions (2% w/v) were prepared to measure their volume distribution and volume-mean diameters by DLS (HORIBA Laser Scattering Particle Size Distribution Analyzer LA-910). A Philips CM 100 (FEI Company, Hillsboro, OR) transmission electron microscope was used to study the microstructure of native lignin, OL, and UL. All samples were diluted to 0.2% with deionized (DI) water and absorbed onto Formvar/carbon-coated 200-mesh copper grids (Electron Microscopy Sciences, Fort Washington, PA). The morphological properties of each sample were recorded with the transmission electron microscope operating at an accelerating voltage of 100 kV.

FTIR and Thermogravimetric Analysis (TGA)

FTIR spectra of native lignin, OL, and UL samples were acquired with a PerkinElmer Spectrum 400 FT-IR/FT-NIR Spectrometer (Waltham, MA) over the 4000–400 cm–1 region at 64 scan and a resolution of 4 cm–1. The TGA (PerkinElmer, Norwalk, CT) was also carried out in a nitrogen-rich environment, which provided an inert atmosphere during pyrolysis. The samples were heated from 50 to 700 °C at a heating rate of 20 °C /min.

Preparation of CP–Lignin-Based Adhesives

CP powder (1:10 w/v) was dissolved in DI water, lignin solution, OL solution, and UL solution separately and gently stirred for 1 h at pH 7.2. In our preliminary experiment, the content of lignin was fixed at 20% to CP (dry base). Then, PEGDE was added at 2% concentration (v/v), and the resultant dispersion was heated at 85 °C for 30 min to stimulate the copolymerization. After cooling to room temperature, the adhesive was used for wood test and also lyophilized for fluorescence test. In summary, five different formulations for CP-based adhesives are shown in Table (percent lignin was based on the dry weight of CP).

Table 4

Different Formulations of CP–Lignin-Based Adhesives

number	formulation	lignin content (%)	notation
1	CP	0	CP
2	CP cross-linked by PEGDE	0	P–CP
3	CP/native lignin cross-linked by PEGDE	20	P–CPL
4	CP/OL cross-linked by PEGDE	20	P–CPOL
5	CP/UL cross-linked by PEGDE	20	P–CPUL

Surface Hydrophobicity

Surface hydrophobicity was measured using 1-anilino-8-naphthalenesulfonate (ANS) as the fluorescence probe as previously reported with modification.[1] P–CP, P–CPL, P–CPOL, and P–CPUL dispersions (see Table for sample description) were prepared in phosphate-buffered saline (0.2 M, pH 7.5) at 2 mg/mL. After centrifuging to remove any insoluble matter, 60 μL of ANS solution (8.0 mM in 0.1 M phosphate buffer, pH 7.4) was added to 3 mL of sample solution. Sample solutions were excited at 365 nm, and the relative fluorescence intensity of emission spectra was recorded from 400 to 600 nm using a Hitachi F-7000 fluorescence spectrophotometer (Hitachi, Ltd., Tokyo, Japan) with a slit width of 10 nm.

A Bohlin C-VOR 150 rheometer (Malvern Instruments, Southborough, MA) was used to characterize the viscoelastic properties of adhesives. A parallel plate head was used with a 20 mm plate diameter and a 500 μm gap. A thin layer of silicon oil was spread over the circumference of the sample to prevent sample dehydration during testing. All experiments were performed in triplicate at 1% strain (within its linear viscoelastic region). Frequency sweep analyses were performed for a frequency range of 0.1–100 rad/s. The testing temperature was 23 °C.

Preparation of Three-Layer Wood

The 300 × 300 × 3.5 mm dimension yellow pine were preconditioned in a 27 °C, 30% relative humidity (RH) chamber at least 1 week before wood adhesion test. The three veneers were layered up in a way that grain line of the middle panels were perpendicular to the grain lines up in a way that grain of the top and bottom panels. Around 20–22 g/ft2 (wet basis) adhesives were brushed on the two faces of the middle veneer panel only. The assembled wood specimen was standing for 15 min before hot press. The hot press conditions were 150 °C and 10 min at 1.03 MPa. The bonded three-layer wood specimens were conditioned in a chamber at 23 °C and 50% RH for 2 days. Then, they were cut into 10 small pieces (82.6 × 25.6 mm2) and 4 large pieces with dimension of 50 × 127 mm2 for three-cycle water-soaking test.

Shear Strength Measurement and Water Resistance

Five small wood specimens were soaked in water for 24 h at 23 °C to test the wet adhesion strength, whereas five specimens were used to test the dry adhesion strength. Both dry strength and wet strength were tested with an Instron Tester (Model 4465, Canton, MA) accorpan class="Chemical">ding to ASTM Standard Method D906-98[34] at a crosshead speed of 1.6 mm/min. Adhesion strength was recorded as stress at the maximum load. Four large wood panels were used for the three-cycle water-soaking test in accordance with American National Standard for Hardwood and Decorative Plywood (ANSI/HPVA HP-1-2004).[35] In each cycle, the wood specimens were soaked at 23 °C for 4 h and then dried at 50 °C with good air circulation for 19 h. After drying, the wood pieces were evaluated for delamination, and the delamination was recorded. Any wood piece with delamination length value higher than 2 in. was considered as failed.