Designing a Coupling Agent with Aliphatic Polyether Chain and Exploring Its Effect on Silica/Natural Rubber Nanocomposites under the Action of Non-Rubber Contents

In order to prepare engineering tires with lower rolling resistance and better wet slip resistance in a more environmentally friendly way. In this study, a series of low volatile organic compound (VOC) Mx–Si69 coupling agents (x = 1, 2, 3, 4, 5, 6, which means the number of ethoxy group in bis-(γ-triethoxysilylpropyl)-tetrasulfide (Si69) substituted by the aliphatic polyether chain) were applied to silica/NR nanocomposites to prepare tire tread with excellent performance. Firstly, M1–Si69 was substantiated as the best choice of Mx–Si69 and Si69 to achieve comprehensive optima in the mechanical properties of silica/NR nanocomposites characterized by dynamic and static mechanical properties. Afterwards, the modification of silica with M1–Si69 induced by Non-Rubber Contents (NRCs) in silica/NR nanocomposites was revealed by comparing the filler network, micromorphology, and mechanical properties of isoprene rubber (IR) and NR nanocomposites. Furthermore, compared with Si69, the M1–Si69 coupling agent was found to conspicuously reduce the energy loss and improve the safety performance of engineering tires according to evaluations of the rolling resistance and dynamic thermomechanical properties of the silica/NR nanocomposites. Finally, the critical function of M1–Si69 in reducing ethanol (a kind of volatile organic compound (VOC)) emissions from the reaction of coupling agent and silica was disclosed by gas chromatography–mass spectrometry.


Introduction
In recent years, with the increasing shortage of petrochemical resources, reducing the energy loss of vehicles has received more and more attention [1,2]. The rolling resistance of tires, which is affected by the hysteresis effect of rubber nanocomposites, is an important component of the energy loss of a vehicle [3,4]. Therefore, reducing the hysteresis effect of rubber nanocomposites and improving the dynamic performance of tires have attracted the attention of researchers. In previous studies, "green tires" prepared by the nanocomposites of silica-reinforced solutions of polymerized styrene-butadiene rubber (SSBR) and butadiene rubber (BR) were proven to have extremely low rolling resistance values [5,6]. Road measurement results showed that a 22~35% rolling resistance reduction and 3~8% of fuel saving could be achieved with "green tires" compared with conventional tires [7,8]. In addition, according to Mikhaylov's research, the development of "green tire" technology is also in line with the trend of social and economic development in the future [9,10].
Due to its insufficient tear and tensile strength, silica/BR/SSBR nanocomposites can only be used as the tread of a "green tire" for passenger vehicles [11,12]. However, in only be used as the tread of a "green tire" for passenger vehicles [11,12]. However, in field of engineering vehicles and trucks, natural rubber (NR) nanocomposites are stil replaceable, and the demand to reduce rolling resistance to save the fuel consumptio also urgent the for tires used in these fields. Moreover, NR is an important strategic source, derived from the latex of Hevea brasiliensis, that accounts for more than 30% the world's consumption of rubber hydrocarbons [13,14]. Because of its excellent mech ical properties, about 70% of produced NR is used in the tire industry, especially in en neering tires [15].
In preparing tires for engineering vehicles and trucks, a certain amount of filler m also be added to the NR matrix to improve mechanical strength properties and red costs [16,17]. Silica is also an ideal filler for NR, as it can reduce the hysteresis effect of rubber nanocomposite to meet the need of engineering tires with low rolling resista [18,19]. In our previous research, it was found that silica in rubber must be combined w coupling agents to exert its excellent reinforcing properties, because good silica dispers and proper interfacial interaction in silica/rubber nanocomposites can be achieved [20, In addition, the emission of ethanol (a kind of volatile organic compound (VOC)), gen ated by the reaction of silica and coupling agents and volatilized at high temperatu was also considered by us. Therefore, sulfurized cardanol polyoxyethylene ether, epo terminated polybutadiene, and bis-epoxypropyl polysulfide were designed to avoid emission of ethanol in preparing silica/rubber nanocomposites [22][23][24]. In recent resea we designed a series of Mx-Si69 coupling agents that had different numbers of cha consisting of aliphatic polyethers in place of ethoxy groups in bis-(γ-triethoxysilylprop tetrasulfide (Si69), thereby reducing the volatilized ethanol produced by the hydrolysi ethoxy groups; it was also found that M2-Si69 was the best choice in Mx-Si69 and Si69 decreasing the hysteresis effect of silica/BR/SSBR nanocomposites [25]. In this resea we intended to select the best of a series of Mx-Si69 materials using in silica/NR nanoco posites to improve their dynamic properties and decrease VOC emissions during pre ration. However, the molecular chain of NR contains not only 1000~3000 linear cis-1,4 prene polymer units but also 3~5wt% of phospholipids, proteins, fatty acids and ot non-rubber contents (NRCs; see Figure 1). S. Salina Sarkawi and Tiwen Xu et al. fou that there are hydrogen and covalent bonds between NRCs and silica, which sugg potential effects of NRCs on the mode of silica modified by Mx-Si69 [26,27]. Because M Si69 not only chemically grafts on the silica surface via a condensation reaction betw hydroxyl groups but also physically adsorbs with silica through a strong hydrogen bo with the hydroxyl groups of the silica. Therefore, the effects of Mx-Si69 under the act of NRCs on the properties of silica/NR nanocomposites needed to be clearly revealed. In this work, a series of Mx-Si69 coupling agents were first applied to silica/NR na composites, and their cross-linking density, filler network, dynamic properties, and st In this work, a series of M x -Si69 coupling agents were first applied to silica/NR nanocomposites, and their cross-linking density, filler network, dynamic properties, and static properties were studied. Then, several silica/NR nanocomposites and silica/isoprene rubber (IR) nanocomposites were compared to study the differences between M x -Si69 and Si69 on silica modification under the action of NRCs. The mechanism of M x -Si69 coupling agent-modified silica in NR matrixes was proposed by comparing the differences between Si69 and M x -Si69 applied to silica/NR nanocomposites. Overall, in this research, a "green tire" with extremely low energy loss and ultra-high wet skid resistance was prepared, and the modification mechanism of M x -Si69 in NR matrixes was also revealed.

Preparation of M x -Si69 Coupling Agent
The low VOC coupling agent M x -Si69 was obtained from a reaction between the terminal hydroxyl of fatty alcohol polyoxyethylene ether-9 (AEO-9) and the ethoxy group of coupling agent Si69. The reaction was carried out under the protection of nitrogen. AEO-9 and Si69 were first heated to 130 • C until no ethanol was produced in the reaction, and then the unreacted monomers were removed by vacuuming at 80 • C. By adjusting the molar ratio of AEO-9 to Si69 (1:1~6:1) and the dosage of tetrabutyl titanate of the catalyst (3 wt% of the total reactants), a series of low VOC coupling agents MxSi69 coupling agents containing different numbers of long alkyl chains of polyethers could be obtained.

Characteristics of Coupling Agents
The structures and 1 H NMR spectra of Si69 and M x -Si69 (take M 1 -Si69 as an example; x = 1, 2, 3, 4, 5, 6, which means the number of ethoxy group substituted by chains consisting of aliphatic polyethers) are shown in Figure 2. As previously reported [25], with increases in the aliphatic polyether ratio reacting with Si69, the area of F peak gradually increases and the area of B peak gradually decreases from the original 3.79-3.86 ppm, indicating that the ethoxy group has been replaced by a chain consisting of aliphatic polyethers. The reacting ratio is calculated according to 1 H NMR data. The graft rate is defined as the ratio of AEO-9 participating in the reaction to the ethoxy group in Si69, and the conversion rate is defined as the ratio of AEO-9 participating in the reaction calculated from the 1 H NMR spectrum to the amount of AEO-9 added during the reaction. Taking M 1 -Si69 as an example, if the conversion rate of AEO-9 was 100%, the graft rate of Si69 was 16.7%, as shown in Figure 2, and the conversion of aliphatic polyether in M 1 -Si69~M 6 -Si69 was about 90%, which demonstrated a high reaction efficiency.

Preparation of Rubber Nanocomposites
The formulation of the rubber nanocomposites is shown in Table 1. Silica/rubber nanocompounds were obtained through conventional in situ blending. First, plasticized rubber (NR or IR), silica, and coupling agent (Si69 or M x -Si69) were mixed together in an internal mixer, which has a chamber temperature of 55 • C and a rotational speed of 50 rpm, for 5 min. Then, an activator and an antioxidant were put into the chamber. After 2 min, a preliminary silica/rubber mixture was obtained. Second, the chamber temperature was adjusted to 150 • C and the rotor speed was adjusted to 30 rpm. All of the preliminary silica/rubber mixtures were kneaded for another 5 min to facilitate the reaction between the coupling agent and silica in the rubber matrix. After this process, all of the silica/rubber mixtures were cooled to room temperature. Third, the accelerator and a curing agent were uniformly blended in sequence with the cooled compound in a 6-inch mill at room temperature. The total mixing time was no more than 15 min.

Preparation of Rubber Nanocomposites
The formulation of the rubber nanocomposites is shown in Table 1. Silica/rubber nanocompounds were obtained through conventional in situ blending. First, plasticized rubber (NR or IR), silica, and coupling agent (Si69 or Mx-Si69) were mixed together in an internal mixer, which has a chamber temperature of 55 °C and a rotational speed of 50 rpm, for 5 min. Then, an activator and an antioxidant were put into the chamber. After 2 min, a preliminary silica/rubber mixture was obtained. Second, the chamber temperature was adjusted to 150 °C and the rotor speed was adjusted to 30 rpm. All of the preliminary silica/rubber mixtures were kneaded for another 5 min to facilitate the reaction between the coupling agent and silica in the rubber matrix. After this process, all of the silica/rubber mixtures were cooled to room temperature. Third, the accelerator and a curing agent were uniformly blended in sequence with the cooled compound in a 6-inch mill at room temperature. The total mixing time was no more than 15 min.    The vulcanization of rubber nanocomposites was carried out on a flat vulcanizing machine. The vulcanization pressure was 15 MPa, and the vulcanization time was set according to the t 90 obtained by a disc Vulkameter. The vulcanization time of the 2 mm sheet used to test tensile strength was (t 90 + 3) min, the vulcanization time of the cylinder used to test heat build-up was 2 times (t 90 + 3) min, and the vulcanization time of the solid tire used to test rolling resistance was 3 times (t 90 + 3) min. These silica/rubber nanocomposites were denoted according to their components as NR-PS (pure silica), IR-PS, NR-8% Si69 (4.0 Si69; the dosage of the coupling agent was 8 wt% of the silica), NR-10% Si69 (5.0 Si69), IR-Si69, NR-M x -Si69, and IR-M x -Si69.

Characterization
Bound rubber content is often used to characterize interactions between a filler and rubber. Firstly, about 0.5 g (W 1 ) of unvulcanized rubber composites was cut into 1 mm 3 particles and wrapped into a copper mesh (the copper mesh mass is denoted as W 2 ). Then, the sample was immersed in toluene for 72 h and soaked in a new toluene solution for another 24 h. Finally, the sample was placed in a 70 • C vacuum oven to a constant weight (W 3 ). The bound rubber content of the sample was calculated with the following formula: Bound rubber content = W 3 −W 2 −W 1 ×ε 1 W 1 ×ε 2 where is ε 1 was the mass fraction of filler and ε 2 is the mass fraction of rubber.
The cross−linking density of the rubber nanocomposites was characterized by equilibrium swelling and calculated with the Flory-Rehner equation. A vulcanized sample (mass m 0 ) was soaked in a toluene solution for 72 h, and then the toluene solution adsorbed onto the sample surface was removed with filter paper. Then, the sample was weighed to determine its weight (m 1 ) as soon as possible. Finally, the sample was placed in a vacuum drying oven at 70 • C to a constant weight (m 2 ), and the crosslinking density of the sample was calculated with the following formula: where M c is the molecular weight between molecular crosslinking points, m f is the mass of filler in the test sample, ρ p is the density of the polymer, V s is the molar volume of toluene (106.4 cm 3 /mol), V r is the volume fraction of rubber after swelling, ρ s is the density of toluene (0.866 g/cm 3 ), and V c is the relative crosslinking density. The interaction parameters (χ) of rubber network and toluene was calculated as 0.393. The Mooney viscosity of rubber was tested with a rotorless Mooney viscometer (Gotech Co., Ltd. Guangzhou, Guangdong, China) according to international standard ISO 289. The rubber was preheated at 100 • C for one minute and then tested for four minutes to obtain the corresponding Mooney viscosity value.
Hydrogen nuclear magnetic resonance ( 1 H NMR) tests of the Si69, AEO-9 and M x -Si69 coupling agents were conducted on an AVANCE III 400 MHz NMR spectrometer (Bruker, Daltonik GmbH Co., Ltd. New York, NY, USA), for which, chloroform-d (CDCl 3 ) was used as the solvent for testing.
The optimum vulcanization time of the rubber nanocomposites was determined with an M-3000A rotorless vulcanizer (Gotech Co., Ltd. Guangzhou, China) with a shear angle of 5 • and a frequency of 200 cpm.
The filler network and hysteresis effect of nanocomposites were measured with an American Alpha Company RPA2000. The filler network of the unvulcanized rubber nanocomposites was tested at 60 • C, 1 Hz, and 0-200% strain. The vulcanizate was evaluated under conditions of 60 • C, 10 Hz, and 0~42% strain.
The filler dispersion and microstructure of the rubber nanocomposites were characterized and analyzed with an S-4800 scanning electron microscope from the Hitachi Company of Japan and an H-9500 transmission electron microscope from the JEOL Company of Japan.
Stress-strain curves were created with an AI-7000S1 electronic tensile machine (Gotech Co., Ltd. Guangzhou, Guangdong) according to international standard ISO 37. Samples were cut into dumbbell-shaped test strips with a length of 25 mm and tested at 500 mm/min.
Heat build-up characterization was conducted according to ISO 4666-3:2010 with an RH-2000N Heat Build-Up tester (Gotech Co., Ltd. Guangzhou, China). A cylindrical sample with a height of 25 mm was subjected to a compression test at 55 • C and a displacement of 2.225 for 30 min. Finally, the temperature difference at the bottom of the sample was recorded.
The energy loss of the rubber nanocomposites during rotation was determined with an RRS-II tire rolling resistance meter (Beijing Wanhui Technology Co., Ltd. Beijing, China). The rubber nanocomposites were set with a load of 30 kg and a rotating speed of 600 rpm.
The variation curve of the loss factor with temperature was characterized with a DMAVA-3000 dynamic thermomechanical analyzer of the French 01-dB company. The samples were tested with a deformation of 0.3% and a frequency of 10 Hz.
An Akron abrasion test was conducted with an MZ-4061 Akron abrasion meter (Jiangsu Mingzhu Testing Machinery Co., Ltd. Yangzhou, China). The samples were preground for 800 revolutions and then ground for 3394 revolutions (equivalent to 1.61 km). The anti-wear properties of the rubber nanocomposites were evaluated via the volume difference between the samples before and after abrasion.

Results and Discussion
3.1. Properties of Silica/NR Nanocomposites Containing M x -Si69 3.1.1. Crosslink Density and Filler Network of Silica/NR Nanocomposites As shown in Figure 3a, NR-10% Si69 had the highest crosslink density of all silica/NR nanocomposites, and that of NR-8% Si69 was only lower than NR-10% Si69. With increases in the number of aliphatic polyether chain in M x -Si69, the cross-linking density of NR-M x -Si69 gradually decreased. As sulfur-containing coupling agents, Si69 and M x -Si69 chemically interacted with the silica surface through dehydration condensation during the process of mixing, and they reacted with rubber macromolecules via their active sulfur groups during the process of vulcanizing, resulting in the formation of a cross-linked structure in which multiple rubber molecules were bound to the same silica particle. Therefore, the cross-linking density of silica/NR nanocomposites is related to the actual number of coupling agent molecules used. The molecular weight of the aliphatic polyether chain in M x -Si69 was almost equal to that of Si69, which means that the actual number of M x -Si69 molecules significantly decreased with the increase in "x" at the same usage in these silica/NR nanocomposites, leading to a low crosslink density for NR-M x -Si69. Furthermore, the cross-linking density of NR-M 1 -Si69 was close to that of NR-8% Si69, although M 1 -Si69 had nearly twice the molecular weight of Si69, which means that M 1 -Si69 could more efficiently construct cross-linking networks in the rubber nanocomposites with fewer moles. (Bruker, Daltonik GmbH Co., Ltd. New York), for which, chloroform-d (CDCl3) was used as the solvent for testing. The optimum vulcanization time of the rubber nanocomposites was determined with an M-3000A rotorless vulcanizer (Gotech Co., Ltd. Guangzhou, Guangdong) with a shear angle of 5° and a frequency of 200 cpm.
The filler network and hysteresis effect of nanocomposites were measured with an American Alpha Company RPA2000. The filler network of the unvulcanized rubber nanocomposites was tested at 60 °C, 1 Hz, and 0-200% strain. The vulcanizate was evaluated under conditions of 60 °C , 10 Hz, and 0~42% strain.
The filler dispersion and microstructure of the rubber nanocomposites were characterized and analyzed with an S-4800 scanning electron microscope from the Hitachi Company of Japan and an H-9500 transmission electron microscope from the JEOL Company of Japan.
Stress-strain curves were created with an AI-7000S1 electronic tensile machine (Gotech Co., Ltd. Guangzhou, Guangdong) according to international standard ISO 37. Samples were cut into dumbbell-shaped test strips with a length of 25 mm and tested at 500 mm/min.
Heat build-up characterization was conducted according to ISO 4666-3:2010 with an RH-2000N Heat Build-Up tester (Gotech Co., Ltd. Guangzhou, Guangdong). A cylindrical sample with a height of 25 mm was subjected to a compression test at 55 °C and a displacement of 2.225 for 30 min. Finally, the temperature difference at the bottom of the sample was recorded.
The energy loss of the rubber nanocomposites during rotation was determined with an RRS-II tire rolling resistance meter (Beijing Wanhui Technology Co., Ltd. Beijing). The rubber nanocomposites were set with a load of 30 kg and a rotating speed of 600 rpm.
The variation curve of the loss factor with temperature was characterized with a DMAVA-3000 dynamic thermomechanical analyzer of the French 01-dB company. The samples were tested with a deformation of 0.3% and a frequency of 10 Hz.
An Akron abrasion test was conducted with an MZ-4061 Akron abrasion meter (Jiangsu Mingzhu Testing Machinery Co., Ltd. Yangzhou, Jiangsu). The samples were preground for 800 revolutions and then ground for 3394 revolutions (equivalent to 1.61 km). The anti-wear properties of the rubber nanocomposites were evaluated via the volume difference between the samples before and after abrasion.

Properties of Silica/NR Nanocomposites Containing Mx-Si69
3.1.1. Crosslink Density and Filler Network of Silica/NR Nanocomposites As shown in Figure 3a, NR-10% Si69 had the highest crosslink density of all silica/NR nanocomposites, and that of NR-8% Si69 was only lower than NR-10% Si69. With The bound rubber content and cross-linking density were used to characterize the interactions between the filler and rubber and the network structure of the silica/rubber nanocomposite. As shown in Figure 3b, with the increase in "x" in the series of Mx-Si69, the bound rubber content of the silica/NR composites showed a gradual downward trend. This can be explained by the gradual increase in physical interactions such as hydrogen bonds and van der Waals forces formed by the polyether long alkyl chain of M x -Si69 between the filler and rubber matrix and the decrease in the chemical bonding of the ethoxy and polysulfide groups.
In the rubber nanocomposites, a rapidly decreased storage modulus with increasing strain is known as the Payne effect, which is closely related to the filler-filler network [29,30]. The Payne effect values of the studied unvulcanized rubber nanocomposites are exhibited in Figure 4a; the Payne effect values of NR-8% Si69 and NR-10% Si69 were almost the same and higher than that of NR-M x -Si69, indicating a weak filler-filler network in NR-M x -Si69. A curve of vulcanizate storage modulus versus strain is shown in Figure 4b, and the difference in the corresponding vulcanizate storage moduli is depicted in Figure 4c. NR-M 1 -Si69 showed the lowest Payne effect of the studied vulcanized rubber nanocomposites, which indicated that M 1 -Si69 could effectively weaken the filler network and reduce the agglomeration of nanoparticles (NPs) in the rubber matrix. As a simple chemical coupling agent, Si69 only reacted with one or two hydroxyl groups on the surface of silica. In contrast, the long polyether alkyl chain and ethoxy group were both present in the M 1 -Si69 coupling agent. During vulcanization, M 1 -Si69 could not only form chemical combinations with silanol groups through ethoxy groups but also cover a large number of unreacted silanol groups through its polyether structure. Therefore, a stable dispersion of silica with a weak filler-filler network in the rubber matrix was achieved by using M 1 -Si69.  The stress-strain curves of the silica/NR nanocomposites are shown in Figure 5a, and their corresponding moduli and reinforcement index (modulus at 300% divided by mod-

Static and Dynamic Properties of Silica/NR Nanocomposites
The stress-strain curves of the silica/NR nanocomposites are shown in Figure 5a, and their corresponding moduli and reinforcement index (modulus at 300% divided by modulus at 100%) are depicted in Figure 5b. NR-8% Si69 and NR-10% Si69 showed higher moduli than NR-M x -Si69; meanwhile, the moduli of all NR-M x -Si69 materials decreased with increases in the number of aliphatic polyether chain in M x -Si69. The "coupling bridge" structure between silica and rubber, which was formed by Si69 or M x -Si69, played a significant role in improving the reinforcing efficiency of silica on rubber. NR-M 1 -Si69 had the highest reinforcement index of all silica/NR nanocomposites. The reinforcing efficiency was affected by both silica dispersion and the coupling structure existing between the silica and rubber. In this research, M 1 -Si69 was shown to be a proper combination for promoting silica dispersion and forming a coupling structure between silica and rubber.  Figure 5c shows the loss factor (Tan delta value at 60 °C) of the silica/NR nanocomposites, which is mainly used to reflect the energy loss caused by filler-filler and fillerrubber friction under a load with dynamic characteristics. In the tire industry, the loss factor at 7% strain is commonly used to characterize the rolling resistance of tires [31]. In theory, silica NPs that are adequately combined with rubber molecules under the premise of sufficient dispersion can weaken the friction in filler-filler and filler-rubber networks under dynamic conditions [32]. As shown in Figure 5c, NR-M1-Si69 had the lowest loss factor at 7% strain of all silica/NR nanocomposites, confirming that NR-M1-Si69 had the lowest rolling resistance. This can be explained by the facts that M1-Si69 effectively weakened the silica network in the rubber matrix and that the hysteresis loss caused by the destruction and reconstruction of the filler network was also reduced. In addition, NR-M1-Si69 had a higher bound rubber content and crosslinking density, which further reduced the energy loss caused by the friction between silica NPs and rubber molecular chains. Figure 5d shows the heat build-up of the silica/NR nanocomposites, which reflects the amount of energy loss due to the internal friction in the rubber nanocomposites under dynamic shock at constant pressure. NR-M1-Si69 also had the lowest heat build-up of all silica/NR nanocomposites, which aligned well with the loss factor measured with the RPA.  Figure 5c shows the loss factor (Tan delta value at 60 • C) of the silica/NR nanocomposites, which is mainly used to reflect the energy loss caused by filler-filler and filler-rubber friction under a load with dynamic characteristics. In the tire industry, the loss factor at 7% strain is commonly used to characterize the rolling resistance of tires [31]. In theory, silica NPs that are adequately combined with rubber molecules under the premise of sufficient dispersion can weaken the friction in filler-filler and filler-rubber networks under dynamic conditions [32]. As shown in Figure 5c, NR-M 1 -Si69 had the lowest loss factor at 7% strain of all silica/NR nanocomposites, confirming that NR-M 1 -Si69 had the lowest rolling resistance. This can be explained by the facts that M 1 -Si69 effectively weakened the silica network in the rubber matrix and that the hysteresis loss caused by the destruction and reconstruction of the filler network was also reduced. In addition, NR-M 1 -Si69 had a higher bound rubber content and crosslinking density, which further reduced the energy loss caused by the friction between silica NPs and rubber molecular chains. Figure 5d shows the heat build-up of the silica/NR nanocomposites, which reflects the amount of energy loss due to the internal friction in the rubber nanocomposites under dynamic shock at constant pressure. NR-M 1 -Si69 also had the lowest heat build-up of all silica/NR nanocomposites, which aligned well with the loss factor measured with the RPA.
Ultimately, M 1 -Si69 is the best choice of the Si69 and series of M x -Si69 materials for achieving comprehensive optima in mechanical and dynamic properties. However, this is a different conclusion to that regarding the use of M x -Si69-series coupling agents in silica/SSBR nanocomposites. It can be deduced that the effect of the M x -Si69 coupling agent in silica/NR nanocomposites may be influenced by NRCs.

Mechanism of M x -Si69 on Silica Modification in NR Nanocomposites
In order to further study the effect of NRCs on the silica modification and use of M x -Si69 in silica/rubber nanocomposites, IR, which contains the same constituent units as NR but does not contain NRCs, was used for research. NR and IR additionally have similar molecular weights (as shown in Table 2) and raw rubber Mooney viscosity values (NR: 83; IR: 87).

Cure Characteristics of Silica/NR Nanocomposites and Silica/IR Nanocomposites
Vulcanization curves of the silica/NR nanocomposites and silica/IR nanocomposites are shown in Figure 6. As shown in Figure 6, the torque of NR-PS and IR-PS dramatically increased at the beginning of vulcanization due to the strong interaction between silica NPs that insufficiently dispersed in the rubber nanocomposites. The torque increase of NR-PS in the initial vulcanization stage was smaller than that of IR-PS, indicating that the interaction between silica NPs in NR-PS was weakened. It can be speculated that the NRCs, which were the only components in NR but not IR, play a role in weakening the interaction between silica NPs.
Polymers 2022, 14, x FOR PEER REVIEW 10 of 1 but does not contain NRCs, was used for research. NR and IR additionally have simila molecular weights (as shown in Table 2) and raw rubber Mooney viscosity values (NR 83; IR: 87).

Cure Characteristics of Silica/NR Nanocomposites and Silica/IR Nanocomposites
Vulcanization curves of the silica/NR nanocomposites and silica/IR nanocomposites are shown in Figure 6. As shown in Figure 6, the torque of NR-PS and IR-PS dramatically increased at the beginning of vulcanization due to the strong interaction between silica NPs that insufficiently dispersed in the rubber nanocomposites. The torque increase o NR-PS in the initial vulcanization stage was smaller than that of IR-PS, indicating tha the interaction between silica NPs in NR-PS was weakened. It can be speculated that the NRCs, which were the only components in NR but not IR, play a role in weakening the interaction between silica NPs.

Filler Network and Micromorphology of Silica/NR Nanocomposites and Silica/IR Nanocomposites
The storage modulus versus strain curve of the rubber nanocomposites and the dif ference of the storage moduli are shown in Figure 7a and Figure 7b. The difference of the storage modulus of IR-PS was 51.3% higher than that of NR-PS, indicating that the fille

Filler Network and Micromorphology of Silica/NR Nanocomposites and Silica/IR Nanocomposites
The storage modulus versus strain curve of the rubber nanocomposites and the difference of the storage moduli are shown in Figure 7a,b. The difference of the storage modulus of IR-PS was 51.3% higher than that of NR-PS, indicating that the filler network of IR-PS was more developed than that of NR-PS. These results are consistent with the speculation described in Section 3.2.1. As mentioned by a previous researcher, proteins, which are a main component of NRCs, adsorb onto the silica surface by hydrogen bonds and van der Waals forces, thus hydrophobizing the silica surface and decreasing silica-silica interactions [26,27]. As shown in Figure 7c, the aggregate between silica NPs was more serious in vulcanized IR-PS than that in vulcanized NR-PS, which further proves the role of NRCs in aiding silica dispersion. The Payne effect of IR-M1-Si69 was more obvious than that of IR-Si69; however, the Payne effect of NR-M1-Si69 was less obvious than that of NR-10% Si69. In our previous study, Mx-Si69-series coupling agents also had no more significant effect on reducing the Payne effect than Si69 in silica/SSBR nanocomposites [25]. As coupling agents with ethoxy groups or aliphatic polyether chains at both ends, both Si69 and Mx-Si69 have a chance to react with two different silica NPs. Therefore, silica NPs are isolated from each other by Si69 or Mx-Si69, which is extremely effective in constraining the distance between silica NPs. The molecular number of Si69 is about twice that of M1-Si69 under the same mass usage, leading to more silica NPs being isolated in rubber nanocomposites that use Si69. The aliphatic polyether chain in Mx-Si69 adsorbs onto the surface of silica via hydrogen bonds. Therefore, silica NPs are covered, which is also an effective in reducing the aggregation of silica. In the studied silica/IR nanocomposite, the isolation structure was found The Payne effect of IR-M 1 -Si69 was more obvious than that of IR-Si69; however, the Payne effect of NR-M 1 -Si69 was less obvious than that of NR-10% Si69. In our previous study, M x -Si69-series coupling agents also had no more significant effect on reducing the Payne effect than Si69 in silica/SSBR nanocomposites [25]. As coupling agents with ethoxy groups or aliphatic polyether chains at both ends, both Si69 and M x -Si69 have a chance to react with two different silica NPs. Therefore, silica NPs are isolated from each other by Si69 or M x -Si69, which is extremely effective in constraining the distance between silica NPs. The molecular number of Si69 is about twice that of M 1 -Si69 under the same mass usage, leading to more silica NPs being isolated in rubber nanocomposites that use Si69. The aliphatic polyether chain in M x -Si69 adsorbs onto the surface of silica via hydrogen bonds. Therefore, silica NPs are covered, which is also an effective in reducing the aggregation of silica. In the studied silica/IR nanocomposite, the isolation structure was found to be more effective than the coverage structure in weakening the interaction between silica NPs. In the studied silica/NR nanocomposite, NRCs, which are a component with a low molecular weight, could fully combine with silica in the form of hydrogen bonds under the premise of avoiding the influence of steric hindrance as much as possible. The hydrogen bond between NRCs and polyether had a chance to induce the aliphatic polyether chain in M x -Si69 to more fully cover silica NPs, thus strengthening the coverage structure in NR-M 1 -Si69.
In addition, the Payne effects of these six NR-M x -Si69 materials were almost the same, as shown in Figure 4. M 1 -Si69, with just one aliphatic polyether chain, could fully play the role of covering silica NPs when a large number of hydroxyl groups on the silica surface were occupied by NRCs in the form of hydrogen bonds, which was induced by NRCs to combine with silica. This is the reason why M 1 -Si69 was the best of the M x -Si69-series coupling agents at improving the silica/NR nanocomposites. A schematic diagram of the effect of NRCs on silica modified by M x -Si69 is shown in Figure 8. series coupling agents at improving the silica/NR nanocomposites. A schematic diagram of the effect of NRCs on silica modified by Mx-Si69 is shown in Figure 8.  Figure 9 shows TEM images of vulcanized samples. NR-M1-Si69 had a more even silica dispersion than NR-10% Si69 because NR-10% Si69 contained large amounts of silica aggregates, as represented by dark area in Figure 9. During vulcanization, the polysulfide bond of Si69 or M1-Si69 broke and combined with the double bond of the rubber molecules. In this process, a "coupling bridge" was formed between silica NPs and rubber molecules, and the isolation structure between silica NPs was also destroyed. In contrast, the coverage structure formed by the aliphatic polyether chain in Mx-Si69 for silica remained and played a role in preventing silica from re-aggregating during vulcanization.   Figure 9 shows TEM images of vulcanized samples. NR-M 1 -Si69 had a more even silica dispersion than NR-10% Si69 because NR-10% Si69 contained large amounts of silica aggregates, as represented by dark area in Figure 9. During vulcanization, the polysulfide bond of Si69 or M 1 -Si69 broke and combined with the double bond of the rubber molecules. In this process, a "coupling bridge" was formed between silica NPs and rubber molecules, and the isolation structure between silica NPs was also destroyed. In contrast, the coverage structure formed by the aliphatic polyether chain in M x -Si69 for silica remained and played a role in preventing silica from re-aggregating during vulcanization. series coupling agents at improving the silica/NR nanocomposites. A schematic diagram of the effect of NRCs on silica modified by Mx-Si69 is shown in Figure 8.  Figure 9 shows TEM images of vulcanized samples. NR-M1-Si69 had a more even silica dispersion than NR-10% Si69 because NR-10% Si69 contained large amounts of silica aggregates, as represented by dark area in Figure 9. During vulcanization, the polysulfide bond of Si69 or M1-Si69 broke and combined with the double bond of the rubber molecules. In this process, a "coupling bridge" was formed between silica NPs and rubber molecules, and the isolation structure between silica NPs was also destroyed. In contrast, the coverage structure formed by the aliphatic polyether chain in Mx-Si69 for silica remained and played a role in preventing silica from re-aggregating during vulcanization. The static mechanical properties of the silica/NR nanocomposites and silica/IR nano-

Static and Dynamic Performance of Silica/NR Nanocomposites and Silica/IR Nanocomposites
The static mechanical properties of the silica/NR nanocomposites and silica/IR nanocomposites are shown in Figure 10. NR-PS had a higher modulus than IR-PS, which was also due to the relatively good silica dispersion in NR-PS aided by NRCs. In addition, compared with IR-PS, NR-PS displayed a higher tensile strength, which was due to the NRCs adsorbing onto the surface of silica via hydrogen bonds. During the tensile test, rubber molecules slid across the filler surface with the help of the adsorbed NRCs, which endowed the silica/NR nanocomposites with a higher tensile strength [33]. The dynamic performance of the silica/rubber nanocomposites is displayed in Figure  11. IR-PS exhibited extremely high heat build-up values and severe deformation, which was due to the serious aggregation of silica. NR-PS had a lower heat build-up value than IR-PS because NRCs improved its silica dispersion. NR-M1-Si69 had the lowest heat build-up values of all six silica/rubber nanocomposites, which proves that the coverage structure for silica was formed by the aliphatic polyether chain in Mx-Si69 induced by NRCs and that this can improve the dynamic properties of silica/rubber nanocomposites. Figure 11. Heat build-up of the silica/rubber nanocomposites.
In this research, the effect of NRCs on the use of Mx-Si69 was analyzed by comparing The dynamic performance of the silica/rubber nanocomposites is displayed in Figure 11. IR-PS exhibited extremely high heat build-up values and severe deformation, which was due to the serious aggregation of silica. NR-PS had a lower heat build-up value than IR-PS because NRCs improved its silica dispersion. NR-M 1 -Si69 had the lowest heat build-up values of all six silica/rubber nanocomposites, which proves that the coverage structure for silica was formed by the aliphatic polyether chain in M x -Si69 induced by NRCs and that can improve the dynamic properties of silica/rubber nanocomposites. The dynamic performance of the silica/rubber nanocomposites is displayed in Figure  11. IR-PS exhibited extremely high heat build-up values and severe deformation, which was due to the serious aggregation of silica. NR-PS had a lower heat build-up value than IR-PS because NRCs improved its silica dispersion. NR-M1-Si69 had the lowest heat build-up values of all six silica/rubber nanocomposites, which proves that the coverage structure for silica was formed by the aliphatic polyether chain in Mx-Si69 induced by NRCs and that this can improve the dynamic properties of silica/rubber nanocomposites. In this research, the effect of NRCs on the use of Mx-Si69 was analyzed by comparing the properties of silica/NR nanocomposites and silica/IR nanocomposites. NRCs were found to not only adsorb on the silica surface but also improve the efficiency of aliphatic In this research, the effect of NRCs on the use of M x -Si69 was analyzed by comparing the properties of silica/NR nanocomposites and silica/IR nanocomposites. NRCs were found to not only adsorb on the silica surface but also improve the efficiency of aliphatic polyether chains in improving the dispersion of silica. Therefore, in the silica/NR nanocomposites, M 1 -Si69, which only contained one aliphatic polyether chain, was sufficient to improve the dispersion of silica by covering hydroxyl groups on the silica surface. The sulfur components decreased with increases in the number of aliphatic polyether chains in M x -Si69 under the same dosage, which was detrimental to the cross-linking density of the silica/rubber nanocomposites. Of the M x -Si69-series coupling agents, M 1 -Si69 could balance the silica dispersion and cross-linking density of the nanocomposites better than others. Therefore, NR-M 1 -Si69 exhibited ideal performance.

Rolling Resistance of the Silica/NR Nanocomposites
To further evaluate the energy loss of M 1 -Si69 applied to rubber nanocomposites for tires, five rubber nanocomposites containing different coupling agents were prepared as tire models for testing. As shown in Figure 12, the tires prepared with NR-8% Si69 and NR-10% Si69 had almost the same energy losses, indicating that the 8% Si69 used in the silica/NR nanocomposites was enough to reduce dynamic energy loss because the number of reactive hydroxyl groups on the silica surface was limited. Furthermore, the tire prepared with NR-M 1 -Si69 had a 39% lower energy loss than the tire prepared with NR-8% Si69. According to previous reports, 3% of fuel consumption can be saved when rolling resistance is dropped by 10%. If 100 million tons of fuel is consumed per year, a "green tire" made of NR-M 1 -Si69 instead of NR-8% Si69 could save about 2 million tons of fuel and 5 million tons of carbon dioxide [34]. The energy loss of tires mainly comes from the mutual friction between silica NPs under cyclic reversed loading [35]. M 1 -Si69 was shown to combine good silica dispersion and a sufficient "coupling bridge" between silica and rubber, which effectively weakens the mutual friction of silica NPs. To further evaluate the energy loss of M1-Si69 applied to rubber nanocomposites for tires, five rubber nanocomposites containing different coupling agents were prepared as tire models for testing. As shown in Figure 12, the tires prepared with NR-8% Si69 and NR-10% Si69 had almost the same energy losses, indicating that the 8% Si69 used in the silica/NR nanocomposites was enough to reduce dynamic energy loss because the number of reactive hydroxyl groups on the silica surface was limited. Furthermore, the tire prepared with NR-M1-Si69 had a 39% lower energy loss than the tire prepared with NR-8% Si69. According to previous reports, 3% of fuel consumption can be saved when rolling resistance is dropped by 10%. If 100 million tons of fuel is consumed per year, a "green tire" made of NR-M1-Si69 instead of NR-8% Si69 could save about 2 million tons of fuel and 5 million tons of carbon dioxide [34]. The energy loss of tires mainly comes from the mutual friction between silica NPs under cyclic reversed loading [35]. M1-Si69 was shown to combine good silica dispersion and a sufficient "coupling bridge" between silica and rubber, which effectively weakens the mutual friction of silica NPs.

Wet-Skid Resistance of the Silica/NR Nanocomposites
In the tire industry, a 0 °C loss factor is commonly used to evaluate the wet skid resistance of tires [36,37]. To study the safety performance of tires on wet-skid roads, the DMTA testing of five nanocomposites was conducted. As shown in Figure 13, NR-PS had the lowest 0 °C loss factor, and NR-M1-Si69 and NR-M2-Si69 had higher 0 °C loss factors than NR-10% Si69 and NR-8% Si69. This was because the silica NPs showed good dispersion in the vulcanized rubber nanocomposites due to the help of the coverage structure formed by the aliphatic polyether chain in Mx-Si69, leading more rubber molecular chains in NR-M1-Si69 and NR-M2-Si69 participating in the relaxation motion at 0 °C . The wet skid resistance of NR-M1-Si69 was improved by more than 25% compared with NR-8%

Wet-Skid Resistance of the Silica/NR Nanocomposites
In the tire industry, a 0 • C loss factor is commonly used to evaluate the wet skid resistance of tires [36,37]. To study the safety performance of tires on wet-skid roads, the DMTA testing of five nanocomposites was conducted. As shown in Figure 13, NR-PS had the lowest 0 • C loss factor, and NR-M 1 -Si69 and NR-M 2 -Si69 had higher 0 • C loss factors than NR-10% Si69 and NR-8% Si69. This was because the silica NPs showed good disper-sion in the vulcanized rubber nanocomposites due to the help of the coverage structure formed by the aliphatic polyether chain in M x -Si69, leading more rubber molecular chains in NR-M 1 -Si69 and NR-M 2 -Si69 participating in the relaxation motion at 0 • C. The wet skid resistance of NR-M 1 -Si69 was improved by more than 25% compared with NR-8% Si69, which significantly improved the safety performance of the "green tires".

Akron Abrasion Loss of the Silica/NR Nanocomposites
The wear resistance of tires is affected by multiple factors such as filler dispersion interfacial bonding, and cross-linking density [38][39][40]. Figure 14 shows a model of the Akron abrasion measurement and the Akron wear volume. NR-10% Si69 had the least wear volume because of its high cross-link density. NR-M1-Si69 and NR-8% Si69 had similar relative cross-linking densities, but NR-M1-Si69 had less wear volume and narrower wear stripes ( Figure 15) than NR-8% Si69, meaning that the wear resistance of the silica/NR nanocomposite was improved by M1-Si69 under the same cross-linking density.

Akron Abrasion Loss of the Silica/NR Nanocomposites
The wear resistance of tires is affected by multiple factors such as filler dispersion, interfacial bonding, and cross-linking density [38][39][40]. Figure 14 shows a model of the Akron abrasion measurement and the Akron wear volume. NR-10% Si69 had the least wear volume because of its high cross-link density. NR-M 1 -Si69 and NR-8% Si69 had similar relative cross-linking densities, but NR-M 1 -Si69 had less wear volume and narrower wear stripes ( Figure 15) than NR-8% Si69, meaning that the wear resistance of the silica/NR nanocomposite was improved by M 1 -Si69 under the same cross-linking density.

Akron Abrasion Loss of the Silica/NR Nanocomposites
The wear resistance of tires is affected by multiple factors such as filler dispersion interfacial bonding, and cross-linking density [38][39][40]. Figure 14 shows a model of the Ak ron abrasion measurement and the Akron wear volume. NR-10% Si69 had the least wear volume because of its high cross-link density. NR-M1-Si69 and NR-8% Si69 had similar relative cross-linking densities, but NR-M1-Si69 had less wear volume and narrower wear stripes ( Figure 15) than NR-8% Si69, meaning that the wear resistance of the silica/NR nanocomposite was improved by M1-Si69 under the same cross-linking density.

VOC Contents of Different Coupling Agents Reacted with Silica
The VOC content (ethanol) generated by the reaction of the coupling agent with silica was characterized with gas chromatography-mass spectrometry. As shown in Figure 16, the silica and coupling agent were evenly mixed and placed in a 20 mL sample bottle. These bottles were heated at 150 °C for 30 min, and then the VOC content generated by the reaction was tested. When the ratio of Si69 mixed with silica increased from 8% to 10%, the VOC content produced by the reaction showed no significant change, indicating that the reaction between Si69 and reactive hydroxyl groups on the silica surface was limited. However, when the silica was modified by M1-Si69 and M2-Si69, VOC emissions were reduced by 57% and 80%, respectively, which demonstrates the importance of replacing Si69 with M1-Si69 in reducing VOC emissions during the preparation of rubber nanocomposites.

VOC Contents of Different Coupling Agents Reacted with Silica
The VOC content (ethanol) generated by the reaction of the coupling agent with silica was characterized with gas chromatography-mass spectrometry. As shown in Figure 16, the silica and coupling agent were evenly mixed and placed in a 20 mL sample bottle. These bottles were heated at 150 • C for 30 min, and then the VOC content generated by the reaction was tested. When the ratio of Si69 mixed with silica increased from 8% to 10%, the VOC content produced by the reaction showed no significant change, indicating that the reaction between Si69 and reactive hydroxyl groups on the silica surface was limited. However, when the silica was modified by M 1 -Si69 and M 2 -Si69, VOC emissions were reduced by 57% and 80%, respectively, which demonstrates the importance of replacing Si69 with M 1 -Si69 in reducing VOC emissions during the preparation of rubber nanocomposites.

VOC Contents of Different Coupling Agents Reacted with Silica
The VOC content (ethanol) generated by the reaction of the coupling agent with sil was characterized with gas chromatography-mass spectrometry. As shown in Figure the silica and coupling agent were evenly mixed and placed in a 20 mL sample bot These bottles were heated at 150 °C for 30 min, and then the VOC content generated the reaction was tested. When the ratio of Si69 mixed with silica increased from 8% to 10 the VOC content produced by the reaction showed no significant change, indicating t the reaction between Si69 and reactive hydroxyl groups on the silica surface was limit However, when the silica was modified by M1-Si69 and M2-Si69, VOC emissions w reduced by 57% and 80%, respectively, which demonstrates the importance of replac Si69 with M1-Si69 in reducing VOC emissions during the preparation of rubber nanoco posites. Figure 16. Characterization of the VOC content of coupling agent-modified silica.

Conclusions
In conclusion, NRCs can beneficially weaken silica networks and improve the d Figure 16. Characterization of the VOC content of coupling agent-modified silica.

Conclusions
In conclusion, NRCs can beneficially weaken silica networks and improve the dispersion of silica in rubber matrixes. In addition, the combination of silica and polyether segments in M x -Si69 that is induced by NRCs was found to improve the dispersion of silica in NR matrixes due to a more complete coverage structure for silica. Therefore, M 1 -Si69, which only has one aliphatic polyether chain, can provide enough coverage structure for silica in silica/NR nanocomposites. Moreover, M 1 -Si69 was found to be more efficient than Si69 in improving the dynamic properties of silica/NR nanocomposites. In this research, compared with NR-8% Si69, NR-M 1 -Si69 saved about 39% of energy loss and improved tire wet skid resistance by more than 25%. In addition, compared with Si69, the M 1 -Si69 coupling agent reduced VOC (ethanol) emissions by more than 50%. Therefore, the application of the newly designed M 1 -Si69 coupling agent in silica/NR nanocomposites can not only endow a tire with extremely low energy loss and ultra-high wet skid resistance but also significantly reduce VOC gas emissions, thus providing a new strategy for the manufacturing of high-performance truck tires from the perspective of reducing VOC gas emissions.
Author Contributions: L.Z., X.Y. and X.Z. conceived and designed the experiments; X.Z. and X.C. performed the experiments; F.Z., D.H., J.Z. and X.Y. analyzed the data; F.Z. and X.L. checked and edited the paper; X.Z. and J.Z. wrote the paper. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.