Effect of Phenolic Resin Oligomer Motion Ability on Energy Dissipation of Poly (Butyl Methacrylate)/Phenolic Resins Composites

Poly (butyl methacrylate) (PBMA) was blended with a series of phenolic resins (PR) to study the effect of PR molecular weight on dynamic mechanical properties of PBMA/PR composites. Differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA) found a similar variation of glass transition temperature (Tg). The maximum loss peak (tanδmax) improved in all PBMA/PR blends compared with the pure PBMA. However, tanδmax reduced as the molecular weight increased. This is because PR with higher molecular weight is more rigid in the glass transition zone of blends. The hydrogen bonding between PBMA and PR was characterized by Fourier transform infrared spectroscopy (FTIR). Lower molecular weight PR formed more hydrogen bonds with the matrix and it had weaker temperature dependence. Combined with the results from DMA, we studied how molecular weight affected hydrogen bonding and thus further affected tanδmax.


Introduction
In order to reduce the harm caused by vibration in people's daily life and industrial production, the development of damping materials is becoming more and more important [1]. Polymer damping materials, as a kind of passive damping material, have attracted wide attention because it is simpler to implement and more cost-effective than semi-active and active techniques [2]. When vibration and noise are generated, it can convert mechanical energy into heat energy through internal friction between molecule around its T g . However, in practical application, polymer damping materials are usually required to play a high damping capacity in a specific temperature range. In order to achieve this goal, some methods like blends, interpenetrating networks, and copolymerization were adopted [3][4][5][6][7][8]. Among them, blending is an effective and simple method to regulate the damping properties of materials.
Some scholars have tried to combine polymers with organic or inorganic fillers in order to obtain high performance polymer damping materials. Yin et al. blended petroleum resin with chlorinated butyl rubber (CIIR), thus improving its damping temperature range. The contact angle experiments revealed that van der Waals interactions play an important role in improving the damping properties [9]. Wu et al. [10] found that OMMT form a filler network in CIIR/MMT composites which can suppresses the maximum of loss peak and have little influence on the glass transition temperature. Jiang [11] got different results by blending GO with CIIR. The existence of GO reduced the damping temperature

Preparation of PBMA via Emulsion Polymerization
BMA was passed through neutral alumina columns before use to remove the polymerization inhibitor. In the first step, 160 mL deionized water, 100 mL BMA monomer, 4 g emulsifier were added into a 500-mL three-necked flask. Nitrogen was stirred in at room temperature for 20 min and then heated to 72 °C . At the same time, 0.3 g initiator was dissolved in 40 mL deionized water and dripped into the emulsion through a constant pressure drip funnel. When the liquid surfance appears light blue, the reaction was continued for 4 h. The product was precipitated with anhydrous ethanol. Then filtered with a vacuum pump until the liquid in the lower layer is clarified. The product is then placed on a evaporation dish and dried in a vacuum oven.

Preparation of PBMA/PR Composites via Solvent Blending
Ethyl acetate was selected as the co-solvent for PBMA and PR. First, PBMA (3 g) and PR (3 g) were dispersed in 50 mL of ethyl acetate at room temperature for 8 h to form an even solution. Then 250 mL of deionized water was added to precipitate the sediment. The sediment was collected by watch glass and then dried in a vacuum oven at 80 °C for 24 h.

Characterization
FTIR was carried out on Nicolet 6700 (Thermo Fisher Scientifc, Waltham, MA, USA). Total of 0.2 g blends was dissolved in 2 ml tetrahydrofuran (THF) and smeared on the CaF2 windows. After the solvent is completely volatilized, the sample was fixed and tested. For the temperature scans, the samples were placed in a heating cell connected to a temperature controller. The temperature range was from 30 °C to 170 °C. The carbonyl stretching region was fitted by two Gaussian bands, corresponding to free and hydrogen bonded carbonyl band near 1728 cm −1 and 1704 cm −1 , respectively.
DSC test was performed on DSC-822 e (Mettle Toledo, Zurich, Switzerland). The sample was first heated from −20 °C to 120 °C at a rate of 15 °C/min and then cooled toward −20 °C at the same rate. Finally, a heating process at a heating rate of 3 °C/min to 120 °C was performed to measure the

Preparation of PBMA via Emulsion Polymerization
BMA was passed through neutral alumina columns before use to remove the polymerization inhibitor. In the first step, 160 mL deionized water, 100 mL BMA monomer, 4 g emulsifier were added into a 500-mL three-necked flask. Nitrogen was stirred in at room temperature for 20 min and then heated to 72 • C. At the same time, 0.3 g initiator was dissolved in 40 mL deionized water and dripped into the emulsion through a constant pressure drip funnel. When the liquid surfance appears light blue, the reaction was continued for 4 h. The product was precipitated with anhydrous ethanol. Then filtered with a vacuum pump until the liquid in the lower layer is clarified. The product is then placed on a evaporation dish and dried in a vacuum oven.

Preparation of PBMA/PR Composites via Solvent Blending
Ethyl acetate was selected as the co-solvent for PBMA and PR. First, PBMA (3 g) and PR (3 g) were dispersed in 50 mL of ethyl acetate at room temperature for 8 h to form an even solution. Then 250 mL of deionized water was added to precipitate the sediment. The sediment was collected by watch glass and then dried in a vacuum oven at 80 • C for 24 h.

Characterization
FTIR was carried out on Nicolet 6700 (Thermo Fisher Scientifc, Waltham, MA, USA). Total of 0.2 g blends was dissolved in 2 mL tetrahydrofuran (THF) and smeared on the CaF 2 windows. After the solvent is completely volatilized, the sample was fixed and tested. For the temperature scans, the samples were placed in a heating cell connected to a temperature controller. The temperature range was from 30 • C to 170 • C. The carbonyl stretching region was fitted by two Gaussian bands, corresponding to free and hydrogen bonded carbonyl band near 1728 cm −1 and 1704 cm −1 , respectively.
DSC test was performed on DSC-822 e (Mettle Toledo, Zurich, Switzerland). The sample was first heated from −20 • C to 120 • C at a rate of 15 • C/min and then cooled toward −20 • C at the same rate. Finally, a heating process at a heating rate of 3 • C/min to 120 • C was performed to measure the T g . For PR, we are going to approximate its softening point as T g because its molecular weight is so small that there is no rubber platform. DMA was carried out on Q800 (TA Instruments, New Castle, DE, USA) by using a dual cantilever clamp and a testing method of temperature step-frequency sweep under a heating rate of 5 • C/min within a temperature range of −20 • C to 200 • C and a frequency of 1 Hz.

Structure Characterization of Blends and PBMA
The FTIR spectra were used to characterize the structures of PBMA and B01. As shown in Figure 2, the characteristic absorption peak at 1729 cm −1 can be attributed to the stretching vibrations of carbonyl. B01 shows the characteristic absorption peaks at 1500 cm −1 and 1600 cm −1 which can assign to the stretching vibrations of conjugate double bonds in aromatic ring framework. The characteristic absorption peak of hydroxyl appeared at 3485 cm −1 and the peak of carbonyl splits into two peaks located at 1704 cm −1 and 1728 cm −1 respectively for B01. These phenomena indicate the formation of hydrogen bonding between PR and PBMA. Tg. For PR, we are going to approximate its softening point as Tg because its molecular weight is so small that there is no rubber platform. DMA was carried out on Q800 (TA Instruments, New Castle, DE, USA) by using a dual cantilever clamp and a testing method of temperature step-frequency sweep under a heating rate of 5 °C /min within a temperature range of −20 °C to 200 °C and a frequency of 1 Hz.

Structure Characterization of Blends and PBMA
The FTIR spectra were used to characterize the structures of PBMA and B01. As shown in Figure 2, the characteristic absorption peak at 1729 cm −1 can be attributed to the stretching vibrations of carbonyl. B01 shows the characteristic absorption peaks at 1500 cm −1 and 1600 cm −1 which can assign to the stretching vibrations of conjugate double bonds in aromatic ring framework. The characteristic absorption peak of hydroxyl appeared at 3485 cm −1 and the peak of carbonyl splits into two peaks located at 1704 cm −1 and 1728 cm −1 respectively for B01. These phenomena indicate the formation of hydrogen bonding between PR and PBMA.

Glass Transition of PBMA/PR Composites
Glass transition is the cooperative motion of local segments, it is sensitive to the environment. Except for A01, the softening point of all phenolic resins is higher than Tg of PBMA in Figure 3a. With increasing molecular weight, it gradually moves to higher temperature. As for composites, all samples show only one glass transition in Figure 3b, so the composites might not have apparent heterogeneity. B01 has the lowest Tg. With the increase of molecular weight, it increases rapidly at first and then rises slightly. In order to further understand this phenomenon. The Tg measured by DSC is compared with the Tg calculated by Fox equation. As shown in Figure 4, it is obvious that blends display higher Tg than that predicted by the Fox equation. This is mostly caused by the strong intermolecular interaction between fillers and matrix.

Glass Transition of PBMA/PR Composites
Glass transition is the cooperative motion of local segments, it is sensitive to the environment. Except for A01, the softening point of all phenolic resins is higher than T g of PBMA in Figure 3a. With increasing molecular weight, it gradually moves to higher temperature. As for composites, all samples show only one glass transition in Figure 3b, so the composites might not have apparent heterogeneity. B01 has the lowest T g . With the increase of molecular weight, it increases rapidly at first and then rises slightly. In order to further understand this phenomenon. The T g measured by DSC is compared with the T g calculated by Fox equation. As shown in Figure 4, it is obvious that blends display higher T g than that predicted by the Fox equation. This is mostly caused by the strong intermolecular interaction between fillers and matrix.

Molecular Dynamics of PBMA/PR Composites
Dynamic mechanical analysis (DMA) is widely used to detect dynamic mechanical properties and segmental dynamics of polymer composites. Tanδ is an important parameter to evaluate damping properties. Some studies on molecular dynamics have found that tanδ consists of three molecular motions (LSM, Sub-rouse, Rouse model) and the area of tanδ is influenced by them together [24][25][26]. Generally, the area where tanδ > 0.3 is called the effective damping region [27,28]. In this paper, tanδmax and tanδ peak area (TA) were used to evaluate the damping properties and other damping parameters of blends are listed in Table 2. It can be found in Table 2 that the addition of PR can effectively improve the damping properties of PBMA. Tanδmax improves from 1.481 to 2.352, TA increases from 40.4 to 71.64. On the other hand, the molecular weight has a great influence on the damping property. Higher molecular weight leads to smaller improvement of damping properties. By adjusting the molecular weight properly, PR can improve the damping capacity of the polymer more effectively.

Molecular Dynamics of PBMA/PR Composites
Dynamic mechanical analysis (DMA) is widely used to detect dynamic mechanical properties and segmental dynamics of polymer composites. Tanδ is an important parameter to evaluate damping properties. Some studies on molecular dynamics have found that tanδ consists of three molecular motions (LSM, Sub-rouse, Rouse model) and the area of tanδ is influenced by them together [24][25][26]. Generally, the area where tanδ > 0.3 is called the effective damping region [27,28]. In this paper, tanδmax and tanδ peak area (TA) were used to evaluate the damping properties and other damping parameters of blends are listed in Table 2. It can be found in Table 2 that the addition of PR can effectively improve the damping properties of PBMA. Tanδmax improves from 1.481 to 2.352, TA increases from 40.4 to 71.64. On the other hand, the molecular weight has a great influence on the damping property. Higher molecular weight leads to smaller improvement of damping properties. By adjusting the molecular weight properly, PR can improve the damping capacity of the polymer more effectively.

Molecular Dynamics of PBMA/PR Composites
Dynamic mechanical analysis (DMA) is widely used to detect dynamic mechanical properties and segmental dynamics of polymer composites. Tanδ is an important parameter to evaluate damping properties. Some studies on molecular dynamics have found that tanδ consists of three molecular motions (LSM, Sub-rouse, Rouse model) and the area of tanδ is influenced by them together [24][25][26]. Generally, the area where tanδ >0.3 is called the effective damping region [27,28]. In this paper, tanδ max and tanδ peak area (TA) were used to evaluate the damping properties and other damping parameters of blends are listed in Table 2. It can be found in Table 2 that the addition of PR can effectively improve the damping properties of PBMA. Tanδ max improves from 1.481 to 2.352, TA increases from 40.4 to 71.64. On the other hand, the molecular weight has a great influence on the damping property. Higher molecular weight leads to smaller improvement of damping properties. By adjusting the molecular weight properly, PR can improve the damping capacity of the polymer more effectively.
The temperature dependence of tanδ of PBMA/A04 composites is shown in Figure 5a. With the increase of A04 content, tanδ peak position moves to higher temperature and tanδ max increases. Hydrogen bonding plays a role in limiting molecular motion and increasing energy consumption at the same time. Figure 5b shows the variation of tanδ with PR molecular weight. A01 has a plasticizing effect on PBMA which is different from other PR, this can be attributed to the weaker interaction between A01 and PBMA. All composites show a higher tanδ max than pure PBMA and the variation of tanδ Polymers 2020, 12, 490 6 of 11 peak position is consistent with those measured by DSC. It is worth noting that tanδ max can be easily suppressed by a larger molecular weight. We think this is because that the PR with higher molecular weight has weaker motion ability that can reduce the energy consumption from hydrogen bonds.  Figure 5a. With the increase of A04 content, tanδ peak position moves to higher temperature and tanδmax increases. Hydrogen bonding plays a role in limiting molecular motion and increasing energy consumption at the same time. Figure 5b shows the variation of tanδ with PR molecular weight. A01 has a plasticizing effect on PBMA which is different from other PR, this can be attributed to the weaker interaction between A01 and PBMA. All composites show a higher tanδmax than pure PBMA and the variation of tanδ peak position is consistent with those measured by DSC. It is worth noting that tanδmax can be easily suppressed by a larger molecular weight. We think this is because that the PR with higher molecular weight has weaker motion ability that can reduce the energy consumption from hydrogen bonds. In order to test our ideas, ∆T' (Tg-tanδ-Tg-dsc) is used to evaluate the motion ability of PR in the glass transition region of PBMA/PR composites (Tg-dsc refers to the Tg of PR measured by DSC, Tg-tanδ refers to the Tg of PBMA/PR blends measured by DMA). As shown in Table 3, ∆T' reduces with the increase of molecular weight. So the PR with higher molecular weight appears to be more rigid in the glass transition region. According to Radmard's reports [29], flexible polymer chains are more conducive to the formation of hydrogen bonding. Therefore, lower molecular weight PR is more likely to form hydrogen bonds with PBMA and generate more energy loss through repeated formation and destruction of hydrogen bonding. As shown in Figure 6, a schematic diagram is used to represent this phenomenon. When the original hydrogen bonds are destroyed, the low molecular weight PR is more likely to form new ones with the matrix because of its flexibility. In order to test our ideas, ∆T (Tg -tanδ -Tg -dsc ) is used to evaluate the motion ability of PR in the glass transition region of PBMA/PR composites (T g-dsc refers to the T g of PR measured by DSC, T g-tanδ refers to the T g of PBMA/PR blends measured by DMA). As shown in Table 3, ∆T reduces with the increase of molecular weight. So the PR with higher molecular weight appears to be more rigid in the glass transition region. According to Radmard's reports [29], flexible polymer chains are more conducive to the formation of hydrogen bonding. Therefore, lower molecular weight PR is more likely to form hydrogen bonds with PBMA and generate more energy loss through repeated formation and destruction of hydrogen bonding. As shown in Figure 6, a schematic diagram is used to represent this phenomenon. When the original hydrogen bonds are destroyed, the low molecular weight PR is more likely to form new ones with the matrix because of its flexibility.    Figure 7a reveals the temperature dependence of the storage modulus (E') for composites. In the glassy state, the blends have a higher modulus than pure PBMA. Hydrogen bonding plays a reinforcing role. As the temperature increases, the intermolecular force weakens and the fillers begin to soften. At this time, PR can reduce the entanglements in PBMA and the storage modulus of blends become lower than pure PBMA in the rubbery state. Inspection of Figure 7a reveals the temperature dependence of the storage modulus (E ) for composites. In the glassy state, the blends have a higher modulus than pure PBMA. Hydrogen bonding plays a reinforcing role. As the temperature increases, the intermolecular force weakens and the fillers begin to soften. At this time, PR can reduce the entanglements in PBMA and the storage modulus of blends become lower than pure PBMA in the rubbery state.   Figure 7a reveals the temperature dependence of the storage modulus (E') for composites. In the glassy state, the blends have a higher modulus than pure PBMA. Hydrogen bonding plays a reinforcing role. As the temperature increases, the intermolecular force weakens and the fillers begin to soften. At this time, PR can reduce the entanglements in PBMA and the storage modulus of blends become lower than pure PBMA in the rubbery state. On the other hand, the storage modulus of the composites improves with the increase of PR molecular weight in the rubbery state in Figure 7b. Storage modulus (G • N ) of rubber platform is directly related to entanglements, as follow: where M e represents entanglement molecular weight, ρ is the polymer density, R is the gas constant, and T is the temperature. That means the PBMA blended with larger molecular weight PR produce more entanglements and have less free volume. However, different scales of molecular motion have different responses to the space limited. The reduction in free volume can mainly confine the molecule motion with long scale and has little effect on small motion units such as segments corresponding to T g [30,31]. This matches the phenomenon in Figure 5b. The weakening of large-scale molecule motion reduces tanδ max . Figure 8 shows the infrared spectra of the carbonyl stretching region heating from 30 • C to 170 • C for B01 and B04. All spectra exhibit two bands at the carbonyl stretching region. One band centered at around 1728 cm −1 comes from the free carbonyl groups, another centered at around 1704 cm −1 corresponds to the hydrogen bonded carbonyl [32]. With increasing temperature, the intensity of hydrogen bonded carbonyl decreases, while the intensity of free carbonyl improves. To further study the temperature dependence of hydrogen bonds, the carbonyl stretching region was fitted with two Gaussian bands. On the other hand, the storage modulus of the composites improves with the increase of PR molecular weight in the rubbery state in Figure 7b. Storage modulus (G o N) of rubber platform is directly related to entanglements, as follow:

Intermolecular Interaction between PBMA and PR
where Me represents entanglement molecular weight, ρ is the polymer density, R is the gas constant, and T is the temperature. That means the PBMA blended with larger molecular weight PR produce more entanglements and have less free volume. However, different scales of molecular motion have different responses to the space limited. The reduction in free volume can mainly confine the molecule motion with long scale and has little effect on small motion units such as segments corresponding to Tg [30,31]. This matches the phenomenon in Figure 5b. The weakening of large-scale molecule motion reduces tanδmax. Figure 8 shows the infrared spectra of the carbonyl stretching region heating from 30 °C to 170 °C for B01 and B04. All spectra exhibit two bands at the carbonyl stretching region. One band centered at around 1728 cm −1 comes from the free carbonyl groups, another centered at around 1704 cm −1 corresponds to the hydrogen bonded carbonyl [32]. With increasing temperature, the intensity of hydrogen bonded carbonyl decreases, while the intensity of free carbonyl improves. To further study the temperature dependence of hydrogen bonds, the carbonyl stretching region was fitted with two Gaussian bands. As shown in Figure 9, the carbonyl stretching region of B04 recorded at 170 °C was divided into two parts. The total area (AT) of the carbonyl region can be expressed by the following formula [33][34][35]:

Intermolecular Interaction between PBMA and PR
AT=AB+AF (2) where AB and AF are the area of bonded and free carbonyl groups, respectively. The fraction of hydrogen-bonded carbonyl is given by: As shown in Figure 9, the carbonyl stretching region of B04 recorded at 170 • C was divided into two parts. The total area (A T ) of the carbonyl region can be expressed by the following formula [33][34][35]: where A B and A F are the area of bonded and free carbonyl groups, respectively. The fraction of hydrogen-bonded carbonyl is given by: Figure 10 shows the variation of X B with temperature for B01 and B04. B01 forms more hydrogen bonds and has a weak dependence on temperature than B04, especially above 130 • C. This is consistent with our prediction. Combined with previous tests, we can draw a preliminary conclusion that lower molecular weight phenolic resin is more likely to form hydrogen bonds with PBMA because of its greater motion ability, which can effectively improve tanδ max .  Figure 10 shows the variation of XB with temperature for B01 and B04. B01 forms more hydrogen bonds and has a weak dependence on temperature than B04, especially above 130 °C. This is consistent with our prediction. Combined with previous tests, we can draw a preliminary conclusion that lower molecular weight phenolic resin is more likely to form hydrogen bonds with PBMA because of its greater motion ability, which can effectively improve tanδmax.

Conclusions
The influence of PR molecular weight on the dynamic mechanical properties of PBMA/PR composites was investigated in this work. We used poly (butyl methacrylate) (PBMA) as matrix and a series of PR with different molecular weights as fillers for investigation. Solvent blending was used as a blending method. DSC and DMA found similar variation in Tg. PR with higher molecular weight had weaker motion ability in the glass transition zone for PBMA/PR composites, which can weaken the energy dissipation from hydrogen bonding and reduce tanδmax. The variable temperature IR spectra found that PR with smaller molecular weight can form more hydrogen bonds with matrix and shows weaker temperature dependence. By analyzing the results, we found that increasing tanδmax by hydrogen bonding energy requires a premise, the fillers need to have sufficient motion ability in the glass transition zone. The conclusion can be used as a guide for the preparation of high performance polymer damping materials.  Figure 10 shows the variation of XB with temperature for B01 and B04. B01 forms more hydrogen bonds and has a weak dependence on temperature than B04, especially above 130 °C. This is consistent with our prediction. Combined with previous tests, we can draw a preliminary conclusion that lower molecular weight phenolic resin is more likely to form hydrogen bonds with PBMA because of its greater motion ability, which can effectively improve tanδmax.

Conclusions
The influence of PR molecular weight on the dynamic mechanical properties of PBMA/PR composites was investigated in this work. We used poly (butyl methacrylate) (PBMA) as matrix and a series of PR with different molecular weights as fillers for investigation. Solvent blending was used as a blending method. DSC and DMA found similar variation in Tg. PR with higher molecular weight had weaker motion ability in the glass transition zone for PBMA/PR composites, which can weaken the energy dissipation from hydrogen bonding and reduce tanδmax. The variable temperature IR spectra found that PR with smaller molecular weight can form more hydrogen bonds with matrix and shows weaker temperature dependence. By analyzing the results, we found that increasing tanδmax by hydrogen bonding energy requires a premise, the fillers need to have sufficient motion ability in the glass transition zone. The conclusion can be used as a guide for the preparation of high performance polymer damping materials.

Conclusions
The influence of PR molecular weight on the dynamic mechanical properties of PBMA/PR composites was investigated in this work. We used poly (butyl methacrylate) (PBMA) as matrix and a series of PR with different molecular weights as fillers for investigation. Solvent blending was used as a blending method. DSC and DMA found similar variation in T g . PR with higher molecular weight had weaker motion ability in the glass transition zone for PBMA/PR composites, which can weaken the energy dissipation from hydrogen bonding and reduce tanδ max . The variable temperature IR spectra found that PR with smaller molecular weight can form more hydrogen bonds with matrix and shows weaker temperature dependence. By analyzing the results, we found that increasing tanδ max by hydrogen bonding energy requires a premise, the fillers need to have sufficient motion ability in the glass transition zone. The conclusion can be used as a guide for the preparation of high performance polymer damping materials.