Adhesion and Cohesion Strength of Phenol-Formaldehyde Resin Mixed with Different Types and Levels of Catalyst for Wood Composites

Abstract

Phenol-formaldehyde (PF) resin is one of the most well-known adhesives for exterior use. PF adhesive is one of the commercial thermoset polymers that is used extensively due to its many benefits. This study investigated the influence of different types and levels of catalysts, i.e., NaOH and CaCO₃ at 1% and 5% based on the solids content of PF resin on the adhesive properties, adhesion, and cohesion strength of PF resin. The results show that the catalyst type significantly influenced the PF adhesive viscosity and pH. Furthermore, the catalyst level significantly affected the PF adhesive’s solids content, viscosity, and gelation time. The cohesion strength of PF-CaCO₃-1% was more significant than the PF control at 75 °C. According to the DMA analysis, the mixed PF-NaOH-5% provided the highest storage modulus, followed by the PF-CaCO₃-1%, with values that were not statistically different from one another. The adhesion strength of PF-CaCO₃-1% was the highest, whereas the adhesion strength of the PF control was the lowest, as confirmed by the cohesion strength. According to the findings, adding CaCO₃-1% as a catalyst of PF resin would result in better adhesive adhesion and cohesion properties for wood composites in the future.

1. Introduction

Adhesives are often used to join two objects (adherents) through surface bonding. Many types of adhesives have been developed, but they are generally categorized as elastomeric, thermoplastic, and thermosetting adhesives [1]. Unlike the first two, thermosetting adhesives are cured at an elevated temperature and cannot be softened by reheating. During the curing process, the molecular chains of this polymer are cross-linked by strong intermolecular covalent bonds, making them irreversible in their final state [2]. Once the polymer has cured, heating tends to cause degradation and loss of strength rather than melting. Thermosetting has been used in the wood-based and natural fiber industries since the early 1900s [3].

The oldest synthetic thermosetting adhesives to be developed, phenol-formaldehyde resin, i.e., a polycondensation product between phenol and formaldehyde, is one of the most significant in the wood-based sector [4]. This adhesive has been used as an adhesive in wood composite industries, such as particle board and plywood, laminated veneer lumber (LVL), and oriented strand board (OSB) [5]. Besides its excellent mechanical properties, PF adhesive can be used in moist conditions and is suitable for exterior use. Compared to other formaldehyde-based thermosetting, PF has lower formaldehyde emissions [6]. Other advantages of PF include resistance to high temperatures, bacteria, fungi, termites, and microorganisms, as well as resistance to chemicals, such as oils and wood preservatives. The disadvantages of PF are that it gives a dark color to the product and the glue lines are relatively thick. However, PF resin’s main challenge in wood composite panels is its slow cure rate [7,8].

The curing of the adhesive is very important to ensure that the resulting bond has optimal strength. Imperfect curing causes low bonding quality, but on the other hand, the long hot-pressing process to achieve good adhesive curing involves high costs. Various catalysts, additives, and modifiers can be employed to speed up the curing of PF resins [7,9]. Several studies have shown that the addition of additives such as organic phenolic substances and amino-based resins effectively accelerates the curing of adhesives compared to conventional PF [10,11]. Adjustment of the pH in the curing process using a catalyst and the addition of a carbonate accelerator is also reported as a simple method to accelerate curing effectively. The catalyst releases heat, thereby accelerating the hardening of the adhesive mixture [12]. The common base catalyst used for the synthesis of PF adhesives is shown in

Table 1. Types of catalysts commonly used in the synthesis of PF resin.

Catalyst Type [6]
Sodium hydroxide, potassium hydroxide, lithium hydroxide Magnesium hydroxide, calcium hydroxide, barium hydroxide Sodium carbonate Calcium oxide, magnesium oxide Tertiary amines, triethylamine 2-Dimethylamino-2-methyl-1-propanol, 2-dimethylamino-2-hydroxymethyl-1,3-propanediol Tri(p-chlorophenyl)phosphine, triphenylphosphine Tetraalkylammonium hydroxide Aqueous ammonia Organic amines

. Duval et al. [13] evaluated the influence of several catalysts and additives on the MW distribution of PF resins but did not show how it affected the curing rate. Another study reported that the NaOH catalyst was more effective than Ca(OH)₂ for accelerating the curing rate [14]. However, the use of NaOH catalyst is more often done in phenol and formaldehyde condensation processes. The effectiveness of adding NaOH catalyst to the final PF resin before application to wood has not been investigated.

Park et al. [7] reported that the curing rate of PF was increased by using additives of three types of carbonates, i.e., sodium carbonate, potassium carbonate, and propylene carbonate, with the most effective increase shown by propylene carbonate. The application of this adhesive to medium-density fiberboard showed an optimum increase in the panel properties with the addition of 3% additive [9]. Similar results were also shown for high-ortho phenol-formaldehyde with carbonate additives [15]. Meanwhile, the use of calcium carbonate (CaCO₃) as a catalyst in the curing process of PF adhesives has not been investigated. Calcium carbonate is generally white in color and is often found in limestone, calcite, and marble. It is widely used as an extender in paints with a content of up to 30% of the weight of the paint. This material is also a well-known filler in the manufacture of plastics. In addition, CaCO₃ is also routinely used as a filler in thermosetting resins. This compound can also function as a pH corrector to maintain alkalinity.

This study investigates the effect of NaOH and CaCO₃ as curing process catalysts on the adhesion and cohesion strength of commercially available PF adhesive. The addition of the catalyst was carried out at the level of 1% and 5% of the solids content of the PF adhesive. It is expected that the results of this study can provide information regarding simple, applicable, and inexpensive methods to improve the curing and properties of PF adhesives.

2. Materials and Methods

2.1. Materials

Commercial phenol-formaldehyde (PF) adhesive with a solids content of 47.39% was purchased from PT Pamolite Adhesive Industry, Probolinggo, Indonesia. Distilled water, technical-grade NaOH flakes, and technical-grade CaCO₃ powders were used for catalyst solution preparation. Rubber (Hevea Brasiliensis) wood veneers 100 × 100 × 2 mm in size and with a moisture content of 4.9% were used for plywood manufacturing.

2.2. Adhesive Preparation

About 200 g of NaOH and CaCO₃ each were dissolved in 1000 mL of distilled water to obtain a solution of 20% NaOH and 20% CaCO₃. Five PF adhesive formulations consisting of control (without catalyst), a mixture with 20% NaOH at the 1% level, a mixture with 20% NaOH at the 5% level, a mixture with 20% CaCO₃ at the 1% level, and a mixture with 20% CaCO₃ at the 5% level were prepared. The catalyst level was determined based on the solid content of commercial PF adhesive with the formulation as shown in

Table 2. Formulation of PF adhesive at different types and levels of catalysts.

Treatment Type	PF (g)	NaOH 20% (g)	CaCO3 20% (g)
Control *	10	-	-
NaOH-1%	10	0.24	-
NaOH-5%	10	1.18	-
CaCO3-1%	10	-	0.24
CaCO3-5%	10	-	1.18

* Control is PF adhesive without catalyst with a solids content of 47.39%.

. Figure 1 shows an illustration of the PF adhesive mixed with different types and levels of catalysts.

2.3. Characterization of Adhesives

2.3.1. Solids Content

An adhesive sample of 1 g was put on aluminum foil and then placed in an oven (Memmert Celsius 10.0, Memmert, Germany) at a temperature of 103 ± 3 °C for 3 h. After the sample dried, the aluminum foil was transferred to a desiccator and weighed. The solids content was calculated using the formula:

$$\mathrm{Solids} \mathrm{content} (\%)=\frac{\mathrm{O}\mathrm{v}\mathrm{e}\mathrm{n} \mathrm{d}\mathrm{r}\mathrm{i}\mathrm{e}\mathrm{d} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}{\mathrm{I}\mathrm{n}\mathrm{i}\mathrm{t}\mathrm{i}\mathrm{a}\mathrm{l} \mathrm{w}\mathrm{e}\mathrm{i}\mathrm{g}\mathrm{h}\mathrm{t}}\times 100\%$$

2.3.2. Gelation Time

The PF adhesive with various treatments was put in a test tube. The gel time meter (Techne GT-6, Coleparmer, Vernon, IL, USA) was positioned so the needle was submerged in the sample. Dimethyl sulfoxide (DMSO) was used as the material in the water bath so that the temperature reached 135 °C. After that, the time required for the adhesive to gelate was observed. The adhesive gelation time limit was obtained when the timer stopped automatically and showed the gelation time number marked with “gel” on the screen.

2.3.3. Viscosity

About 20 mL PF adhesive samples were put in a glass gauge (C-CC27, AntonPaar, Graz, Austria) and mounted on a rotational rheometer (RheolabQC, AntonPaar, Graz, Austria). Viscosity was measured with a concentric cylinder (cc)-type spindle no. 27 with a rotation speed of 150/s. Tests were carried out at 25 °C, 50 °C, and 75 °C to determine the effect of increasing temperature on viscosity. Dynamic viscosity was measured for 120 s.

2.3.4. pH Value

The pH value of the adhesive was determined using a pH meter (Laqua pH 1200, Horiba, Kyoto, Japan). The pH value was shown on the screen a few moments after the electrode probe of the pH meter was dipped in the adhesive sample placed in a container.

2.3.5. Functional Group Analysis

Functional group analysis of liquid and cured PF adhesive was performed using Fourier transform infrared (FTIR SpectrumTwo, Perkin Elmer, Waltham, MA, USA) using the universal attenuated total reflectance (UATR) method. The samples were placed in a sample holder, pressed, and scanned from 400 to 4000 cm⁻¹ with an average of 16 scans at a resolution of 4 cm⁻¹.

2.3.6. Dynamic Mechanical Analysis

The PF adhesive samples were spread on filter paper (CAT No. 1005-125, Whatman, Buckinghamshire, UK) and then air dried. DMA measurements were carried out using a DMA instrument (DMA 8000, PerkinElmer, Waltham, MA, USA) in dual cantilever mode with a constant frequency of 1 Hz at 30–300 °C. The viscoelastic response was expressed in dynamic modulus (E) and damping ability (tan δ), with a heating rate of 5 °C/min for each sample.

2.4. Fabrication and Evaluation of Plywood

The adhesive strength test was carried out by applying the adhesive to a rubber wood veneer measuring 100 × 100 × 6 mm, which was arranged into three-ply plywood. The PF adhesive with 180 g/m² glue spread was applied to the plywood veneer using a double spread method. The plywood was then hot pressed at a pressure of 1 MPa at 140 °C for 9 min. Testing of plywood tensile shear strength was carried out using a universal testing machine (UTM AGS-X 50 kN, Shimadzu, Kyoto, Japan) with a sample size of 75 × 25 × 18 mm (Figure 2).

2.5. Data Analysis

The obtained data were analyzed using a completely randomized design with two factors: type of catalyst and level of catalyst added to the commercial PF adhesive. Each treatment was made in three repetitions. Data processing was carried out using Microsoft Excel 2019 software and IBM SPSS Statistics 20. Data analysis was completed using analysis of variance (ANOVA) with a 95% confidence level. To determine the significant difference among treatments, Duncan’s multiple-range test (DMRT) was performed on parameters that showed a significant effect.

3. Results and Discussions

3.1. Characteristics of Adhesives

The characteristics of PF adhesives with various catalyst treatments are shown in

Table 3. Basic characteristics of PF adhesive mixed with different types and levels of catalysts.

Catalyst Types	Characteristics of Adhesive
	Solids Content (%)	Gelation Time		Viscosity (mPa.s, T = 25 °C)	pH
		(min, T = 135 °C)	(min, T = 25 °C)
Control	47.4 ± 2.06	12.7 ± 1.71	60.5 ± 5.40	487.1 ± 28.34	13.4 ± 0.03
NaOH 1%	44.5 ± 0.28	6.1 ± 0.29	80.7 ± 6.70	357.4 ± 32.02	13.4 ± 0.08
NaOH 5%	40.9 ± 2.23	10.6 ± 0.55	130.4 ± 10.50	125.5 ± 14.60	13.7 ± 0.06
CaCO3 1%	44.1 ± 0.25	5.9 ± 0.44	135.6 ± 12.60	284.1 ± 0.26	13.2 ± 0.05
CaCO3 5%	40.6 ± 0.77	11.1 ± 1.18	160.2 ± 17.50	95.5 ± 0.37	13.6 ± 0.02

. The solids content (SC) of the adhesive identifies the number of particles in the adhesive. The more adhesive particles that react with wood in the gluing process, the stronger the bond strength. The average SC of the control PF adhesive was 47.39%, whereas the PF adhesive with NaOH and CaCO₃ catalyst had a lower SC, ranging from 40.58% to 44.54%. The percentage of SC adhesive with catalyst met the solids content SNI 06-4567 standard for PF adhesive, which ranges from 40 to 45% [16]. Analysis of variance (ANOVA) showed that only the catalyst level had a significant effect on the SC of the adhesive (p-value 0.001), whereas the type of catalyst and the interaction between the type and level of catalyst had no significant effect on SC. This shows that the addition of catalyst at the 5% level caused the adhesive to become thinner compared to the addition at the 1% level and the control PF. This condition was caused by the concentration of the catalyst solution used being 20%, so the water content in the adhesive increased and the SC of the adhesive decreased. Adhesive with a high SC and good viscosity can form an optimal bond, resulting in satisfactory adhesion. The solids content indicates the resin content contained in the adhesive; the value tends to be directly proportional to the viscosity, specific gravity, and gelation time. The high resin content causes the viscosity and specific gravity of the adhesive to be high, whereas the gelation time tends to be longer [17]. Adhesives with a high solids content will have high adhesion.

The thickening and ripening process of adhesive is called the curing reaction. This process can occur through gelation time and hardening time. Gelation time is the time required by accelerator for the adhesive to form a gel. Hardening time is the time required by the adhesive to improve the gel time process, a hardening process conducted through hot or cold settings. Hot setting is the improvement of the gelation and hardening process (acceleration and completion of hardening) on adhesives by using a catalyst and heating, whereas cold setting is a repaired gelation and hardening process (acceleration and completion of hardening) adhesive using a catalyst and an accelerator.

The gelation time of PF adhesives with various types and levels of catalysts is shown in Table 3. The control PF had the longest gelation time, which was 12.67 min, compared to the PF mixed with catalyst. The NaOH and CaCO₃ catalysts at the 1% level had a gelation time range of 5.90–6.07 min, whereas at the 5% level the gelation time was longer, ranging from 10.57 to 11.10 min. Analysis of variance showed that only the catalyst-level factor had a significant effect on the gelation time of the adhesive (p-value 0.001), whereas the type of catalyst and the interaction between the type and level of catalyst had no significant effect. The gelation time of the PF adhesive with the treatment was faster than that of the PF control. A decrease in gelation time was also reported for PF adhesives with magnesium hydrate, sodium carbonate, and propylene carbonate (PC) as an accelerator, with the most significant reduction obtained by adding PC [18]. The study suggests that PC may have undergone transesterification, which could have boosted the ortho-hydroxymethyl group’s reactivity.

According to previous research [19,20], esters or their decomposition residue can attack the negatively charged phenolic core in polycondensation reactions, increasing the functionality for activating methylol groups and hence reducing the gel time of PF resins and increasing curing speed. This might also take place when CaCO₃ is introduced to PF adhesive. During the cross-linking of PF adhesive, a minor change in pH value has a significant impact on the amount of methylene ether and the number of methylene linkages. The presence of Na⁺ and Ca²⁺ regenerates reactive methylol groups and carbanions, which are then able to continue reacting to methylene linkages, allowing the PF reaction to proceed faster [19,20,21].

The short gelation time indicates that the adhesive no longer required a long setting time during hot pressing in the manufacture of composite products. However, the short gelation time also means that the adhesive would have coagulated quickly, which would shorten the shelf life of the adhesive. SNI 06-4567 requires a PF gelation time of between 30 and 60 min [16]. The gelation time of various adhesive treatments was also observed as 25 °C, with a controlled time of 60 min. The results show that each treatment remained stable for 60 min of testing and began to form a gel after more than 60 min.

The time for the adhesive to form a gel is also related to its solids content. A higher SC results in a shorter gelation time. Higher resin content causes high viscosity and specific gravity of the adhesive, and the gelation time tends to be longer [16]. However, in this study, it was found that among the four catalyst treatments, samples with lower SC required a longer gelation time. This could have been caused by the level of catalyst added to the adhesive, where more solvent was contained in the catalyst, so the adhesive became thinner and took longer to form a gel [17].

The viscosity value affects the ability of the adhesive to penetrate the pores of the wood and the storage life of the adhesive. Adhesives with high viscosity have a short storage life because they harden faster and the quality of the adhesive is low [18]. Adhesives with high solids content and good viscosity can form optimal bonds, resulting in satisfactory adhesion. Figure 3 shows the viscosity and cohesive strength of the adhesive with various types and levels of catalyst for the temperature increase that occurred. Analysis of variance showed that the type of catalyst (p-value 0.001), the level of catalyst (p-value 0.001), and the interaction between the two (p-value 0.001) had a significant effect on the viscosity of the adhesive. The graph shows that at room temperature (25 °C) the control PF yielded the highest value compared to the catalyst-mixed PF adhesive. Duncan’s multiple-range test (DMRT) was conducted to determine whether the average value of each factor had a significant effect. The DMRT results for the effect of the type of catalyst at the 1% and 5% levels on the adhesive viscosity value are shown in

Table 4. Mean effect of catalyst type at the 1% and 5% levels to viscosity.

Catalyst Type	Level		Average
	1%	5%
NaOH	357.37 a *	125.49 c	241.43
CaCO3	284.13 b	95.51 d	189.82
Average	320.75	110.50	(+)

* The mean value with different letter is significantly different each other

.

The viscosity value of PF adhesive required by SNI 06-4567 is 130–300 cps [16]. Only PF-CaCO₃-1% yielded the standard compliant viscosity value. The PF control had the highest viscosity and cohesion strength. High viscosity can affect the storage period of the adhesive and higher viscosity can shorten the storage period. The type and level of catalyst used affect the viscosity and cohesive strength. The graph shows that NaOH had a higher effect than CaCO₃, whereas the 1% level of each catalyst had a higher effect on both viscosity and cohesive strength compared to the addition of 5%. This condition is also supported by previous reports, which stated that the addition of NaOH as a catalyst reduced the viscosity of PF. In general, a concentration of 20% of a catalyst solution contributes to adding a large amount of water, thereby changing the viscosity of the adhesive. This shows that at a high level (5%), the added catalyst contains more hydrogen groups from the solvent, so the viscosity of the adhesive becomes lower. The appropriate viscosity and cohesive strength of the adhesive provide a good application to wood. Therefore, the determination of the optimal catalyst level must be considered as an important factor.

Determination of pH was carried out to determine the adhesive alkalinity. It is important to ensure that alkaline conditions are maintained so that the PF adhesive avoids deposition and has a high tolerance to water. In addition, PF needs to be stored in alkaline conditions to avoid curing reactions (copolymerization) so that the liquid adhesive can be stable for a relatively long storage life [17]. The pH value of the adhesive with various treatments was obtained, ranging from 13.2 to 13.7 (Table 3). The addition of NaOH as a catalyst yielded a higher pH increase compared to the CaCO₃ catalyst, and the 5% level for each catalyst also yielded a higher pH value than the 1% level. The results of the analysis of variance (ANOVA) showed that only the type of catalyst had a significant effect (p-value 0.001) on the pH of the adhesive.

3.2. Functional Group Analysis

Analysis of PF adhesive functional groups was carried out using FTIR in the liquid and cured conditions. Liquid adhesives provided a different functional group analysis compared to cured adhesives (Figure 4). The peak at wave number 3600–3200 cm⁻¹ indicates the presence of OH groups in PF liquid. The absorption peak of phenolic hydroxyl was located at 3300 cm⁻¹. Another peak at 1000–1033 cm⁻¹ can be assigned to C–O. In addition, the absorption peak at 1447 cm⁻¹ could have been caused by a C=C benzene ring obscured by a -CH₂- methylene bridge. The strong absorption peaks between 1600 and 1700 cm⁻¹ can be attributed to the C=C vibration strains on the aromatic and at 3200–3600 cm⁻¹ to the hydroxyl OH groups [21]. The peak in the range of 1050–1300 cm⁻¹ was also a C-O functional group with a strong intensity, and in the range of 690–900 cm⁻¹ there was an aromatic ring group (C-H) with a strong intensity. Overall, there was no significant difference in intensity between each treatment applied to the PF adhesive.

Compared to the control PF, the intensity at wavenumber 1433 cm⁻¹ increased in PF mixed with NaOH catalyst but decreased in PF mixed with CaCO₃ catalyst (Figure 4). The wave number between 1340 and 1470 cm⁻¹ indicates the presence of alkane (CH) groups with strong intensity. Some intensity changes occurred in solid samples compared to liquid samples. There was a peak in the range of 3600–3200 cm⁻¹, which indicates the presence of OH groups in the PF adhesive. It can be seen that the absorption intensity was lower at 3134–3166 cm⁻¹. This could have been caused by reduced OH groups in the cured (solid) adhesive sample. The strong absorption peaks between 1600 and 1700 cm⁻¹ can be attributed to the C=C vibration strains on the aromatic and at 3200–3600 cm⁻¹ to the hydroxyl OH groups [21]. The peak of 1148 cm⁻¹ indicates the presence of the C-O functional group, which had a strong intensity.

3.3. Dynamic Mechanical Analysis

DMA is a thermal analysis method for evaluating the thermo-mechanical and viscoelastic behavior of polymers. Apart from providing information about the molecular structure and mechanical properties of polymers, DMA is very effective for determining the hardening properties of thermosetting resins, including PF adhesives. DMA usually provides information about resin hardening, gelation time, vitrification, and decomposition, which are very important for resin preparation, modification, and application [22]. The results of DMA analysis of PF adhesive mixed with various types and levels of catalysts are shown in Figure 5.

DMA provided information related to the temperature glass transition (Tg), storage modulus (E′), loss modulus (E″), and delta tangent (tan d = E″/E′), which are related to the properties of thermo-mechanics and viscoelastic adhesive. The tan delta peak represented the midpoint transition or inflection point from the decrease in the curve log of E′, whereas the peak of loss modulus showed an initial drop in E′ in the glassy state moving to the transition step. Peak E″ at Tg, showed, in general, the intersection of the two tangential to the log E′ curve obtained from the glassy area and transition, called the onset temperature [23]. The E′ value decreased with increasing temperature. The value of E′ dropped significantly while warming up, going from a temperature of 30 °C to 120 °C, due to softening adhesive activity in this region [24,25]. Softening temperature was achieved at a temperature of 80–120 °C [26]. After softening, E′ decreased gradually between 120 and 200 °C, and another significant decrease occurred at a temperature of 200–300 °C with the transition phase of the adhesive, whereas at a temperature of 265–300 °C the carbonization process of the adhesive took place [25]. The peak of E′_min is defined as the gelation point, where the crosslinks progressed to form an infinite molecular weight network, and the peak of E′_max was taken to be the vitrification point of the system [27]. The value of E′ indicates the stiffness property of the adhesive. This is shown by the higher E′ values and glassier state, which were obtained at temperatures below 200 °C rather than at higher temperatures. Tan delta data show a balance among elastic and viscous phases from the polymer constituent compound used [28]. The strength and hardness of the thermoset were not significantly affected by the increase in temperature or the deformation rate. The combined method of storage modulus (E′), tan delta, and strain curve could explain the hardening properties from PF resin expanding compared with only using parameter E′ and the tan delta curve. This is because every increase in temperature yielded only a small change in the resulting E′ value, so it was difficult to determine accurate onset temperature of the tested adhesives [22].

3.4. Tensile Shear Strength of Plywood

PF adhesives with various types and levels of catalyst are applied in the manufacture of plywood. The tensile strength of plywood was tested to evaluate the bond strength. This property is the main criterion in analyzing the quality of the adhesive. The adhesive strength indicates the strength that can be achieved or maintained by wood joints. Factors affecting adhesion in the bond strength test include wood properties, adhesive properties, and gluing techniques. The results of the tensile strength test on plywood using adhesives with various types and levels of catalysts are shown in Figure 6.

The value of the tensile strength of plywood increased with the addition of a catalyst to the PF adhesive. The tensile strength value of plywood with control PF adhesive was 0.94 MPa, whereas the highest value was shown for plywood with PF adhesive mixed with CaCO₃ catalyst at the 1% level, with a value of 1.98 MPa. The results of the analysis of variance (ANOVA) show that the interaction between the type and level of catalyst had a significant effect (p-value 0.03) on the tensile strength of plywood. The DMRT results show that plywood with 1% PF-CaCO₃ had a significantly different tensile strength than other adhesives. The addition of NaOH to the adhesive yielded a higher value at the 5% level than at the 1% level, but the value obtained was still lower than with the addition of CaCO₃. The standard value of adhesive firmness required by SNI 01-5008.2 is 0.7 MPa regardless of the percentage of damage that occurs. Thus, the addition of a catalyst can not only accelerate the curing time of the adhesive but also reduce the consumption of adhesive to obtain a certain adhesive strength. This of course increases the cost savings due to reduced adhesive consumption costs and due to reduced energy when hot pressing.

3.5. Wood Failure Analysis

Wood failure is a supporting value in analyzing bond strength. The percentage of wood failure is the ratio of the surface area of the tested bonded board or veneer that contains fiber from the material to which it is bonded [29]. The image of the wood failure shape of plywood can be seen in Figure 7, and the failure percentage values for each sample can be seen in

Table 5. Percentage of wood failure.

Treatment	Wood Failure (%)
PF-Control	6.38
PF-NaOH-1%	26.52
PF-NaOH-5%	36.34
PF- CaCO3-1%	37.00
PF- CaCO3-5%	36.23

.

The form of the sample damage indicates the cause of the failure, such as structural failure, cohesive failure, and adhesive failure. The control PF sample (Figure 7) shows that the damage that occurred was a structural failure, which means that the damage occurred due to the mechanical properties of rubber wood, which are lower than the bond strength of PF. This condition was also demonstrated by the structural failure of the PF-NaOH-1% and PF-CaCO₃-1% samples. Adhesive failure was found in the PF-NaOH-5% and PF-CaCO₃-5% samples, which means that this damage arose due to the weak bond between the adhesive and the plywood that occurred, so the surface area of the plywood that failed was even greater. Table 5 shows the percentage of failure of plywood after the bond strength test was carried out.

Wood samples that were assembled using PF adhesive with the addition of a catalyst had a greater percentage of damage after testing compared to wood that was only glued with control PF adhesive. The addition of a CaCO₃-type catalyst at the 1% level obtained the greatest damage value, namely, 37.00% of the surface area tested, followed by samples with PF-NaOH-5% adhesive, which was equal to 36.34% d; PF-CaCO₃-5%, of 36.23%; NaOH-1%, of 26.52%; and PF-Control, of 6.38%. The percentage of wood failure is related to its bond strength. The higher the bond strength of the sample, the greater the percentage of failure that occurs in the sample. Therefore, PF-CaCO₃-1%, which had the highest strength, had the highest percentage of wood failure.

4. Conclusions

The type of catalyst has a significant effect on the viscosity and pH of the PF adhesive, whereas the catalyst level has a significant effect on the solids content, viscosity, and gelation time of the PF adhesive. DMA analysis shows that PF-NaOH-5% produced the highest storage modulus, followed by PF-CaCO₃-1%, with values that were not significantly different. The highest tensile strength was PF-CaCO₃-1% and the lowest was PF-Control, indicated by the percentage of wood failure, which was directly proportional to the failures. Based on the results of the study, the addition of CaCO₃-1% to PF adhesive was more efficient than that of NaOH-5%.