Preparation and Characterization of Phenolic Foam Modified with Bio-Oil

Bio-oil was added as a substitute for phenol for the preparation of a foaming phenolic resin (PR), which aimed to reduce the brittleness and pulverization of phenolic foam (PF). The components of bio-oil, the chemical structure of bio-oil phenolic resin (BPR), and the mechanical performances, and the morphological and thermal properties of bio-oil phenolic foam (BPF) were investigated. The bio-oil contained a number of phenols and abundant substances with long-chain alkanes. The peaks of OH groups, CH2 groups, C=O groups, and aromatic skeletal vibration on the Fourier transform infrared (FT-IR) spectrum became wider and sharper after adding bio-oil. These suggested that the bio-oil could partially replace phenol to prepare resin and had great potential for toughening resin. When the substitute rate of bio-oil to phenol (B/P substitute rate) was between 10% and 20%, the cell sizes of BPFs were smaller and more uniform than those of PF. The compressive strength and flexural strength of BPFs with a 10–20% B/P substitute rate increased by 10.5–47.4% and 25.0–50.5% respectively, and their pulverization ratios decreased by 14.5–38.6% in comparison to PF. All BPFs maintained good flame-retardant properties, thermal stability, and thermal isolation, although the limited oxygen index (LOI) and residual masses by thermogravimetric (TG) analysis of BPFs were lower and the thermal conducticity was slightly greater than those of PF. This indicated that the bio-oil could be used as a renewable toughening agent for PF.


Introduction
Phenolic foam (PF) is increasingly used in building structural materials due to its good thermal isolation, high dimensional stability, and particularly outstanding flame-retardant properties (no dripping combustion, low flammability, low smoke density and smoke toxicity) [1][2][3][4][5][6]. However, the application of PF is severely restricted by its high brittleness and pulverization, which is related to the lack of flexible functional groups in its chemical structures [2,4,5]. Thereby, a great number of research efforts have focused on increasing the toughness of PF to overcome its brittleness and pulverization.
Toughening of PF can be summarized into two methods: Physical modification and chemical modification. The physical modification focuses on introducing external toughening agents, such as inert fillers [7][8][9] and chopped fibers [3,10,11], into PF by physical blending. Chemical modification is a technique that concentrates on introducing flexible long chains into the molecular chain of PF by a chemical reaction, which has attracted extensive attention due to its notable toughening effect [2,[12][13][14]. Chemical toughening agents, such as polyurethane [2,14], polyethylene glycol [4,12], and polyether [13], have been widely used to toughen PFs. However, considering the high price of the modifiers discussed above, modifiers from renewable natural compounds, such as lignin [6,15,16], tannin [17,18], and cardanol [1], have been a focus of research in recent years. and the time until the stirring bar could not move was recorded. Each test above was repeated at least three times. The characterization of BFRs are summarized in Table 1. Table 1. Basic characteristics of phenolic resin (PR) and bio-oil phenolic resin (BPR).

Resins
Viscosity ( 3 ). Then, the sample was pushed back and forth on a 300 mesh abrasive paper 30 times at a constant force and the distance of each single-pass friction was 250 mm. The pulverization ratio was measured by the weight loss of a sample after friction [7]. At least five replicates were used for these tests.

Analysis
A gas chromatographic-mass spectrometric (GC-MS) analysis of bio-oil was recorded on a Shimadzu GC/MS-QP system (Kyoto, Japan). FT-IR analysis of the bio-oil and cured resin were obtained by a Nicolet iS5 FT-IR (Nicolet, Wisconsin, USA). The microstructure of the foam was observed using a WV-CP230/G polarizing microscope (Panasonic, Suzhou, China). The cell size distributions were calculated on ImageJ 1.47. Thermogravimetric (TG) analysis of the foam was examined at a heating rate of 10 • C/min using a Q5000IR analyzer (TA Instruments, the USA) under a nitrogen atmosphere.

Components of the Whole Bio-Oil
The moisture of bio-oil is 29.82%, and the main organic compounds of bio-oil characterized by GC-MS are displayed in Table 2. As seen, the phenols represent the major peak area (33.08%) and numerous substances in bio-oil had good reactivity with formaldehyde, such as phenol, cresols, guaiacol, and resorcinol. Besides, many phenols with long unsaturated alkane chains, such as guaiacol, could bring toughening groups into the molecular structure of resin when reacting with formaldehyde or reactive phenol hydroxyl and methylol groups of PR. Moreover, some ketones, aldehydes, esters, alcohols, and acids with long-chain alkanes in bio-oil also had toughening effects [25]. Furthermore, the aldehydes, like formaldehyde, acetaldehyde, and furaldehyde of bio-oil, could react with the unreacted phenol to reduce the free phenol and improve the polymerization of resin. Nevertheless, the low boiling point substance and esters of bio-oil could be used as blowing agent and surfactants respectively, which further reduced the cost of PR. These demonstrated that the bio-oil had great potential for partially replacing phenol to prepare resin while toughening resin. It is also important to note that the bio-oil contains a large number of acids (9.24%), resulting in a low pH value. Therefore, bio-oil should be added at the later stage of the synthesis process of PR to reduce the influence on the addition reaction [22].

FT-IR Analysis of BPRs
The FT-IR spectra of bio-oil and cured PR and BPRs are depicted in Figure 1, and the functional groups that correspond to the major peaks have been identified and listed in Table 3 [26,27]. As seen in Figure 1, the three large peaks of bio-oil at 3435 cm −1 , 1704 cm −1 , and 1612 cm −1 were assigned to the vibration of OH groups, C=O groups, and aromatic skeleton respectively. These peaks were ascribed to the phenol, ketone, aldehyde, and ester components in bio-oil, which was consistent with the GC-MS result of bio-oil in Part 3.1. Additionally, the peak of CH 2 at 2925 cm −1 could prove the existence of long-chain alkanes in bio-oil.

Microstructure of BPFs
The PF and BPFs were examined by an optical microscope and their cell size distributions were calculated in order to study the effect of the B/P substitute rate on the microstructure. As shown in Figure 2, the BPFs and PF were made up of a great number of closed cells. The mean cell sizes of 10%BPF and 20%BPF were 0.192 mm and 0.166 mm, respectively, which are 19-30% smaller than that of PF (0.238 mm). The smaller cell sizes might be due to the much larger molecular compounds in bio-oil and their dragging effect on long side chains [28,29], increasing the viscosity of resin (Table 1) and limiting the growing and merging of cells. Meanwhile, the cell sizes of 10%BPF and 20%BPF, particularly 20%BPF (0.10-0.25 mm), are more uniform than that of PF (0.10-0.45 mm). This was caused by the abundant low volatile compounds in bio-oil, which widened the boiling point range of the foaming agent in the forming progress. However, in the case of 30%BPF, the cell sizes were larger and less uniform. Furthermore, some fragments from the bubble collapses appear. These might be due to: (i) The longer curing time of 30%BPR (Table 1) leading to the failure of achieving the appropriate viscosity of resin in time to stop cells growing and merging [9], and (ii) the lower solids  When using the whole bio-oil to partly replace phenol, all of the prepared BPRs presented similar curves to that of the PR, indicating that the BPRs and PR had similar chemical structures. However, some differences between BPR and PR are also found in Figure 1, e.g., the new peak of C=O stretching vibration at the region of 1704 cm −1 appeared and became wider with the increase of the B/P substitute rate. Moreover, the peak at 1643 cm −1 assigned to C=O also became wider. These indicated more different compounds with C=O groups in resins after adding bio-oil. In other words, numbers of C=O groups were introduced into the chemical structure or formed during the sysnthesis of PR owing to the bio-oil. In addition, the stonger peaks of BPRs than PR at the peaks at 1612 cm −1 and 1494 cm −1 were assigned to the aromatic skeletal vibration, which indicated that the phenolic compounds in bio-oil were involved in the synthesis of resin. These meant that abundant flexible functional groups could be introduced into the chemical structure of PR with the reactions between bio-oil and polyformaldehyde or resin intermediate, such as the dimethylphenol and trimethylphenol. The introduction of flexible long chains could also be demonstrated by the increase of the CH 2 peak after adding bio-oil. The peak of CH 2 at 2925 cm −1 of BPRs, especially 10%BPR, was stronger than that of PR. Another possible reason for the larger CH 2 peaks of BPRs was the facilitate impact of bio-oil on the balance of synthetic reaction. For example, the long-chain alkanes in the opposite site of the phenol hydroxyl in guaiacol made the electron cloud density on the benzene ring migrate to the side chain, thus improving the activation of C-O-C bonds to turn into the more stable CH 2 bond [26]. These results indicated that adding bio-oil could not only toughen resin, but also improve the polymerization of resin. However, compared to 10%BPR and 20%BPR, the CH 2 peak at 2925 cm −1 of 30%BPR decreased, suggesting that there was an optimum B/P substitute rate that best improved the polymerization of PRs.

Microstructure of BPFs
The PF and BPFs were examined by an optical microscope and their cell size distributions were calculated in order to study the effect of the B/P substitute rate on the microstructure. As shown in Figure 2, the BPFs and PF were made up of a great number of closed cells. The mean cell sizes of 10%BPF and 20%BPF were 0.192 mm and 0.166 mm, respectively, which are 19-30% smaller than that of PF (0.238 mm). The smaller cell sizes might be due to the much larger molecular compounds in bio-oil and their dragging effect on long side chains [28,29], increasing the viscosity of resin (Table 1) and limiting the growing and merging of cells. Meanwhile, the cell sizes of 10%BPF and 20%BPF, particularly 20%BPF (0.10-0.25 mm), are more uniform than that of PF (0.10-0.45 mm). This was caused by the abundant low volatile compounds in bio-oil, which widened the boiling point range of the foaming agent in the forming progress. However, in the case of 30%BPF, the cell sizes were larger and less uniform. Furthermore, some fragments from the bubble collapses appear. These might be due to: (i) The longer curing time of 30%BPR (Table 1) leading to the failure of achieving the appropriate viscosity of resin in time to stop cells growing and merging [9], and (ii) the lower solids content (Table 1) and the decreased polymerization of 30%BPR (Figure 1).

Basic Characteristics of BPFs
The apparent density, pulverization ratio, compressive strength, and flexural strength of PF and BPFs are summarized in Table 4. The apparent densities of 10%BPF and 20%BPF were higher than that of PF, which was due to the smaller and more uniform cell sizes. However, further increasing the B/P substitute rate decreased the apparent density. Compared with PF, the pulverization ratio of BPFs decreased first and increased with the increase of the B/P substitute rate. The pulverization ratio of 20%BPF dropped to its lowest point (8.9%) and decreased by 38.6% in comparison with PF. This was due to the improved toughness of BPFs because of the long side chains in bio-oil. However, with the B/P substitute rate increasing to 30 wt.%, the fragments from the foam collapses led to the higher pulverization ratio. With the increase of the B/P substitute rate, the compressive strength and flexural strength of BPFs first increased, maximizing at 20% B/P substitute rate, and then decreased. Compared with PF, the compressive strength and flexural strength of 20%BPF increased by 47.4% and 50.0%, respectively. These were mainly related to the more uniform cell sizes of 20%BPF than other foams. In addition, the improvement of toughness of 20%BPF by bio-oil also led to the higher compressive strength and flexural strength. However, further increasing the B/P substitute rate weakens the compressive strength and flexural strength owing to the cell collapses.

Basic Characteristics of BPFs
The apparent density, pulverization ratio, compressive strength, and flexural strength of PF and BPFs are summarized in Table 4. The apparent densities of 10%BPF and 20%BPF were higher than that of PF, which was due to the smaller and more uniform cell sizes. However, further increasing the B/P substitute rate decreased the apparent density. Compared with PF, the pulverization ratio of BPFs decreased first and increased with the increase of the B/P substitute rate. The pulverization ratio of 20%BPF dropped to its lowest point (8.9%) and decreased by 38.6% in comparison with PF. This was due to the improved toughness of BPFs because of the long side chains in bio-oil. However, with the B/P substitute rate increasing to 30 wt.%, the fragments from the foam collapses led to the higher pulverization ratio. With the increase of the B/P substitute rate, the compressive strength and flexural strength of BPFs first increased, maximizing at 20% B/P substitute rate, and then decreased. Compared with PF, the compressive strength and flexural strength of 20%BPF increased by 47.4% and 50.0%, respectively. These were mainly related to the more uniform cell sizes of 20%BPF than other foams. In addition, the improvement of toughness of 20%BPF by bio-oil also led to the higher compressive strength and flexural strength. However, further increasing the B/P substitute rate weakens the compressive strength and flexural strength owing to the cell collapses.

Thermal Analysis
The PF is widely used because of its outstanding flame-retardant properties. However, a considerable amount of research reported that most of the toughing agents would deteriorate the flame resistance of PF [2,5]. Therefore, the LOI, thermal conductivity, and thermal stability of BPFs were investigated. As shown in Figure 3, the LOI value of PF foam is 43.2%, whereas the values of BPFs decrease to 40.5%, 38.7%, and 35.4%, respectively. It is well known that the benzene rings are difficult to burn owing to the easy charring when exposed to the flame [13]. Therefore, the weakened flame resistance of BPFs was due to the decrease of the benzene rings on the backbone chains by using bio-oil as a substitute for phenol. Fortunately, all the LOI values of BPFs were larger than the B1 standard value (≥30%) according to the China National Standards (GB 8624-2012), which meant that the BPFs still maintained good flame-retardant property. The thermal conductivity of foams was little affected by the B/P substitute rate and the difference in value between BPFs and PF only ranges from 0.002 W/(m·K) to 0.008 W/(m·K). The increased thermal conductivity of BPFs was due to the substances with low thermal conductivity in bio-oil. Additionally, the cell collapses could be another reason for the increased thermal conductivity of 30%BPF. little affected by the B/P substitute rate and the difference in value between BPFs and PF only ranges from 0.002 W/(m·K) to 0.008 W/(m·K). The increased thermal conductivity of BPFs was due to the substances with low thermal conductivity in bio-oil. Additionally, the cell collapses could be another reason for the increased thermal conductivity of 30%BPF. The TG and derivative thermogravimetric (DTG) curves of PF and BPFs under a nitrogen atmosphere are illustrated in Figure 4, and the relevant degradation data of PF and BPFs are addressed in Table 5. Compared with PF, the initial degradation temperatures (T−5%, the temperatures at 5% weight loss) [2,4] of 10%BPF and 20%BPF were slightly higher, which indicated that the incorporation of 10-20% bio-oil improved the thermal stability of foams at lower temperatures. The reason was related to the thick cell walls of 10%BPF and 20%BPF (Figure 3), which made it more difficult for the water and volatiles to evaporate. However, the T−5% of 30%BPF decreased. The maximum weight loss temperature (Tmax) of the BPFs slightly shifted to a lower temperature with an The TG and derivative thermogravimetric (DTG) curves of PF and BPFs under a nitrogen atmosphere are illustrated in Figure 4, and the relevant degradation data of PF and BPFs are addressed in Table 5. Compared with PF, the initial degradation temperatures (T −5% , the temperatures at 5% weight loss) [2,4] of 10%BPF and 20%BPF were slightly higher, which indicated that the incorporation of 10-20% bio-oil improved the thermal stability of foams at lower temperatures. The reason was related to the thick cell walls of 10%BPF and 20%BPF (Figure 3), which made it more difficult for the water and volatiles to evaporate. However, the T −5% of 30%BPF decreased. The maximum weight loss temperature (T max ) of the BPFs slightly shifted to a lower temperature with an increasing B/P substitute rate, and the residual masses at 600 • C of BPFs were lower than that of PF. These were due to the decrease of the benzene rings in modified foams because of the replacement of bio-oil to phenol. However, all the residual masses at 600 • C of BPFs were nearly above 60%. The decline proportion of BPFs was only 0.2-14.9% in comparison with PF, which meant that the BPFs still maintained good thermal stability. The TG and derivative thermogravimetric (DTG) curves of PF and BPFs under a nitrogen atmosphere are illustrated in Figure 4, and the relevant degradation data of PF and BPFs are addressed in Table 5. Compared with PF, the initial degradation temperatures (T−5%, the temperatures at 5% weight loss) [2,4] of 10%BPF and 20%BPF were slightly higher, which indicated that the incorporation of 10-20% bio-oil improved the thermal stability of foams at lower temperatures. The reason was related to the thick cell walls of 10%BPF and 20%BPF (Figure 3), which made it more difficult for the water and volatiles to evaporate. However, the T−5% of 30%BPF decreased. The maximum weight loss temperature (Tmax) of the BPFs slightly shifted to a lower temperature with an increasing B/P substitute rate, and the residual masses at 600 °C of BPFs were lower than that of PF. These were due to the decrease of the benzene rings in modified foams because of the replacement of bio-oil to phenol. However, all the residual masses at 600 °C of BPFs were nearly above 60%. The decline proportion of BPFs was only 0.2-14.9% in comparison with PF, which meant that the BPFs still maintained good thermal stability.

Conclusions
Whole bio-oil had great potential for replacing phenol and toughening resin because of its numbers of phenols and abundant substances with long-chain alkanes. Using bio-oil to partly replace phenol introduced abundant flexible functional groups into the chemical structure of PR. This was proved by the smaller pulverization ratio, larger compressive strength, and flexural strength of 10%BPF and 20%BPF in comparison with PF. Adding bio-oil also made the cell sizes of foams smaller and more uniform. These indicated that bio-oil had a positive impact on the toughness of foams. However, the decrease of LOI and residual masses at 600 • C, as well as the light increase of thermal conducticity of BPFs, suggesting that the bio-oil, like most toughing agents, deteriorated the flame resistance of PF. Fortunately, the negative effect of bio-oil was slight and the BPFs still maintained good flame-retardant property, thermal isolation, and thermal stability. Therefore, the bio-oil could be used as a renewable toughening agent for PF.