Electrostatic-Interaction-Driven Assembly of Binary Hybrids towards Fire-Safe Epoxy Resin Nanocomposites

Manganese dioxide (MnO2), as a promising green material, has recently attracted considerable attention of researchers from various fields. In this work, a facile method was introduced to prepare binary hybrids by fabricating three-dimensional (3D) zinc hydroxystannate (ZHS) cubes on two-dimensional (2D) MnO2 nanosheets towards excellent flame retardancy and toxic effluent elimination of epoxy (EP) resin. Microstructural analysis confirmed that the morphologies and structures of MnO2@ZHS binary hybrids were well characterized, implying the successful synthesis. Additionally, the morphological characterization indicated that MnO2@ZHS binary hybrids could achieve satisfactory interfacial interaction with the EP matrix and be well dispersed in nanocomposites. Cone calorimeter test suggested that MnO2@ZHS binary hybrids effectively suppressed the peak of heat release rate and total heat release of EP nanocomposites, performing better than MnO2 or ZHS alone. Condensed-phase analysis revealed that MnO2@ZHS binary hybrids could promote the char density and graphitization degree of char residues and thereby successfully retard the permeation of oxygen and flammable gases. Moreover, through the analysis of gas phase, it can be concluded that MnO2@ZHS binary hybrids could efficiently suppress the production of toxic gases during the degradation of EP nanocomposites. This work implies that the construction of 2D/3D binary hybrids with an interfacial interaction is an effective way to fabricate high-performance flame retardants for EP.


Introduction
Epoxy resin (EP), one of the most important thermosetting polymers, has been widely used in the fields of furniture coatings [1], circuit boarding [2], aeronautical material [3], and so forth [4][5][6] in virtue of its excellent mechanical properties, outstanding optical property, glorious resistance to solvent and chemical corrosion, and high climate resistance. However, analogous to most other

Synthesis of MnO 2 Nanosheets and ZHS Cubes
The synthesis process of MnO 2 was according to the reported work [34]. In a typical process, 15 mmol of KMnO 4 was completely dissolved in 750 mL of deionized water. Then, 200 mL of ethyl acetate was added into the above solution by constantly stirring. The solution was maintained at 95 • C and refluxed overnight. The obtained brown product was filtrated and washed with deionized water and ethanol consecutively, then dried at 80 • C for further use. In the case of ZHS cubes [25], 10 mmol of ZnSO 4 ·7H 2 O was dissolved in 300 mL of deionized water at 25 • C. Once completely dissolved, the equal loading of Na 2 SnO 3 ·H 2 O was added into the above solution. The continuous stirring was maintained for 4 h, and the white suspension was centrifuged and washed with deionized water several times and then dried at 60 • C overnight.

Preparation of MnO 2 @ZHS Binary Hybrid
MnO 2 @ZHS binary hybrid was prepared by a simple electrostatic adsorption method. MnO 2 (1 g) was dispersed in 500 mL of deionized water with ultrasound-assisted mechanical agitation for 2 h. On the other hand, 0.2 g of ZHS was dispersed in 300 mL of deionized water under ultrasonic and mechanical stirring for 1 h. The pH value of ZHS suspension was adjusted to 5 by 0.1 M HCl solution and stirred for another 1 h. Subsequently, the ZHS suspension was dropwise added into the MnO 2 suspension within 30 h by constant mechanical stirring. Finally, the mixed suspension was stirred and maintained for 4 h. The resultant was selected by centrifugation, washed by deionized water and dried at 80 • C overnight.

Preparation of Pure EP and EP Nanocomposites
The preparation of EP nanocomposites was in accordance with our previous works [18]. Typically, 1.22 g of MnO 2 @ZHS binary hybrid was added into 50 mL of acetone under ultrasonically assisted stirring to form a homogeneous suspension. Then, 50 g of premelted EP was poured into the above suspension under mechanical stirring and maintained over 6 h. Subsequently, the suspension was maintained at 100 • C for 12 h to remove the solvent. Thereafter, 10 g of premelted 4,4'-diaminodiphenylmethane was injected into the mixture under constant stirring. Finally, EP nanocomposites could be obtained after being cured at 100 • C for 2 h and 150 • C for 2 h, respectively. The sample was marked as EP/MnO 2 @ZHS 2%, and the others were fabricated by using the same strategy. The preparation process is illustrated in Figure 1.

Characterization
X-ray diffraction (XRD) measurements were performed using a Japan Rigaku D = Max-Ra rotating anode X-ray diffractometer (RIGAKU, Tokyo, Japan) equipped with a Cu-Ka tube and Ni filter (k = 0.1542 nm). Transmission electron microscopy (TEM) images were obtained using a JEOL JEM-100SX transmission electron microscope (JEOL, Tokyo, Japan) with an acceleration voltage of 100 kV. Thermogravimetric analysis (TGA) of as-prepared samples was undertaken using TGA-Q5000 apparatus (TA Co., New Castle, DE, USA) from 50 to 700 °C at a heating rate of 20 °C min −1 . The weight of all samples was maintained within 3-5 mg in an open platinum pan. The morphologies of MnO2 materials with three different dimensions, coated with a gold layer in advance, were observed using scanning electron microscopy (SEM; AMRAY1000B, Beijing R&D Center of the Chinese Academy of Sciences, Beijing, China). TG-IR of the samples was performed using a TGA Q5000IR thermal gravimetric analyzer (TA Co., New Castle, DE, USA) that was interfaced to the Nicolet 6700 FTIR spectrophotometer (Thermo Scientific, Waltham, MA, USA). Approximately 5.0 mg of the sample was placed in an alumina crucible and heated from 30 to 800 °C at a heating rate of 20 °C min −1 (helium atmosphere, flow rate of 45 mL min −1 ). A combustion test was performed on the cone calorimeter (Fire Testing Technology, Derby, UK) according to ISO 5660 standard, using specimens with size of 100 × 100 × 3 mm 3 . Each specimen was exposed horizontally to an external heat flux of 35 kW m −2 . X-ray photoelectron spectroscopy (XPS) was performed on a VG Escalab Mark II spectrometer (Thermo-VG Scientific Ltd., West Sussex, UK), using Al Kα excitation radiation (hυ = 1486.6 eV).

Characterization of MnO2@ZHS Binary Hybrid
TEM technique was employed to observe the morphologies of the MnO2 nanosheets and the MnO2@ZHS binary hybrid, and the morphologies of the MnO2 nanosheets and their hybrid are presented in Figure 2. As can be observed in Figure 2a, the MnO2 nanosheets show layered structure with large surface area, and the wrinkle belonging to the layers can be obviously observed. After the assembly procedure forced by electrostatic adsorption, ZHS cubes are successfully fabricated on the surface of MnO2. As is shown in Figures 2b-d, the ZHS cubes are firmly absorbed on the layers, and no ZHS cubes exist in the area beside the MnO2 nanosheets, indicating that there exist very strong electrostatic forces between the ZHS cubes and the MnO2 nanosheets. Moreover, the EDS spectrum

Characterization
X-ray diffraction (XRD) measurements were performed using a Japan Rigaku D = Max-Ra rotating anode X-ray diffractometer (RIGAKU, Tokyo, Japan) equipped with a Cu-Ka tube and Ni filter (k = 0.1542 nm). Transmission electron microscopy (TEM) images were obtained using a JEOL JEM-100SX transmission electron microscope (JEOL, Tokyo, Japan) with an acceleration voltage of 100 kV. Thermogravimetric analysis (TGA) of as-prepared samples was undertaken using TGA-Q5000 apparatus (TA Co., New Castle, DE, USA) from 50 to 700 • C at a heating rate of 20 • C min −1 . The weight of all samples was maintained within 3-5 mg in an open platinum pan. The morphologies of MnO 2 materials with three different dimensions, coated with a gold layer in advance, were observed using scanning electron microscopy (SEM; AMRAY1000B, Beijing R&D Center of the Chinese Academy of Sciences, Beijing, China). TG-IR of the samples was performed using a TGA Q5000IR thermal gravimetric analyzer (TA Co., New Castle, DE, USA) that was interfaced to the Nicolet 6700 FTIR spectrophotometer (Thermo Scientific, Waltham, MA, USA). Approximately 5.0 mg of the sample was placed in an alumina crucible and heated from 30 to 800 • C at a heating rate of 20 • C min −1 (helium atmosphere, flow rate of 45 mL min −1 ). A combustion test was performed on the cone calorimeter (Fire Testing Technology, Derby, UK) according to ISO 5660 standard, using specimens with size of 100 × 100 × 3 mm 3 . Each specimen was exposed horizontally to an external heat flux of 35 kW m −2 . X-ray photoelectron spectroscopy (XPS) was performed on a VG Escalab Mark II spectrometer (Thermo-VG Scientific Ltd., West Sussex, UK), using Al Kα excitation radiation (hυ = 1486.6 eV).

Characterization of MnO 2 @ZHS Binary Hybrid
TEM technique was employed to observe the morphologies of the MnO 2 nanosheets and the MnO 2 @ZHS binary hybrid, and the morphologies of the MnO 2 nanosheets and their hybrid are presented in Figure 2. As can be observed in Figure 2a, the MnO 2 nanosheets show layered structure with large surface area, and the wrinkle belonging to the layers can be obviously observed. After the assembly procedure forced by electrostatic adsorption, ZHS cubes are successfully fabricated on the surface of MnO 2 . As is shown in Figure 2b-d, the ZHS cubes are firmly absorbed on the layers, and no ZHS cubes exist in the area beside the MnO 2 nanosheets, indicating that there exist very strong electrostatic forces between the ZHS cubes and the MnO 2 nanosheets. Moreover, the EDS spectrum of the MnO 2 @ZHS binary hybrid is presented in Figure 3. The EDS profile confirms the presence of corresponding elements such as Zn, Sn, Mn, and O. Figure 4 presents XRD patterns of the MnO 2 nanosheets, ZHS cubes, and the MnO 2 @ZHS binary hybrid. It is evident that the MnO 2 @ZHS binary hybrid contains all the characteristic peaks of MnO 2 and ZHS [35,36]. These results indicate that the crystalline structures of ZHS and MnO 2 have no changes during the absorption procedure. Moreover, TGA was used to measure the thermal stability of the MnO 2 @ZHS binary hybrid and investigate the changes compared with that of the MnO 2 nanosheets. Figure 5 shows the TGA curves of the MnO 2 nanosheets and MnO 2 @ZHS binary hybrid. The thermal stability of the MnO 2 @ZHS binary hybrid is relatively lower than that of MnO 2 nanosheets, due to the existence of predecomposition compared with MnO 2 . It is probably because of the separation of hydrate water from ZHS [37]. The further chemical state and element composition of MnO 2 and MnO 2 @ZHS were explored by XPS test. of the MnO2@ZHS binary hybrid is presented in Figure 3. The EDS profile confirms the presence of corresponding elements such as Zn, Sn, Mn, and O. Figure 4 presents XRD patterns of the MnO2 nanosheets, ZHS cubes, and the MnO2@ZHS binary hybrid. It is evident that the MnO2@ZHS binary hybrid contains all the characteristic peaks of MnO2 and ZHS [35,36]. These results indicate that the crystalline structures of ZHS and MnO2 have no changes during the absorption procedure. Moreover, TGA was used to measure the thermal stability of the MnO2@ZHS binary hybrid and investigate the changes compared with that of the MnO2 nanosheets. Figure 5 shows the TGA curves of the MnO2 nanosheets and MnO2@ZHS binary hybrid. The thermal stability of the MnO2@ZHS binary hybrid is relatively lower than that of MnO2 nanosheets, due to the existence of predecomposition compared with MnO2. It is probably because of the separation of hydrate water from ZHS [37]. The further chemical state and element composition of MnO2 and MnO2@ZHS were explored by XPS test.   of the MnO2@ZHS binary hybrid is presented in Figure 3. The EDS profile confirms the presence of corresponding elements such as Zn, Sn, Mn, and O. Figure 4 presents XRD patterns of the MnO2 nanosheets, ZHS cubes, and the MnO2@ZHS binary hybrid. It is evident that the MnO2@ZHS binary hybrid contains all the characteristic peaks of MnO2 and ZHS [35,36]. These results indicate that the crystalline structures of ZHS and MnO2 have no changes during the absorption procedure. Moreover, TGA was used to measure the thermal stability of the MnO2@ZHS binary hybrid and investigate the changes compared with that of the MnO2 nanosheets. Figure 5 shows the TGA curves of the MnO2 nanosheets and MnO2@ZHS binary hybrid. The thermal stability of the MnO2@ZHS binary hybrid is relatively lower than that of MnO2 nanosheets, due to the existence of predecomposition compared with MnO2. It is probably because of the separation of hydrate water from ZHS [37]. The further chemical state and element composition of MnO2 and MnO2@ZHS were explored by XPS test.     In Figure 6a, the XPS survey spectra of MnO2 and MnO2@ZHS are presented. The XPS survey spectra indicate that the MnO2@ZHS hybrid is mainly comprised of Mn, O, Sn, Zn, K, and C elements. The high-resolution Zn 2p 3 and Sn 3d are also shown in Figures 6b and c. The binding energies of Zn 2p 3 and Sn 3d are centered at 121.5 and 486.7 eV, respectively, which are attributed to Zn-O and Sn-O bonds [38]. Additionally, the high-resolution O1s spectra is exhibited in Figures 6d-f. The location at 529.8 eV is ascribed to the O-Mn-O bond, and the peak centered at 531.2 eV belongs to the O-Mn-H bond [39,40]. Additionally, it can be clearly found that the binding energy of O1s at around 532.3 eV attributed to the weak bond (-OH) of the MnO2@ZHS binary hybrid is higher than that of MnO2. This is reasonable due to the fabrication of ZHS, which is accompanied by abundant hydroxyl groups.    In Figure 6a, the XPS survey spectra of MnO2 and MnO2@ZHS are presented. The XPS survey spectra indicate that the MnO2@ZHS hybrid is mainly comprised of Mn, O, Sn, Zn, K, and C elements. The high-resolution Zn 2p 3 and Sn 3d are also shown in Figures 6b and c. The binding energies of Zn 2p 3 and Sn 3d are centered at 121.5 and 486.7 eV, respectively, which are attributed to Zn-O and Sn-O bonds [38]. Additionally, the high-resolution O1s spectra is exhibited in Figures 6d-f. The location at 529.8 eV is ascribed to the O-Mn-O bond, and the peak centered at 531.2 eV belongs to the O-Mn-H bond [39,40]. Additionally, it can be clearly found that the binding energy of O1s at around 532.3 eV attributed to the weak bond (-OH) of the MnO2@ZHS binary hybrid is higher than that of MnO2. This is reasonable due to the fabrication of ZHS, which is accompanied by abundant hydroxyl groups. In Figure 6a, the XPS survey spectra of MnO 2 and MnO 2 @ZHS are presented. The XPS survey spectra indicate that the MnO 2 @ZHS hybrid is mainly comprised of Mn, O, Sn, Zn, K, and C elements. The high-resolution Zn 2p 3 and Sn 3d are also shown in Figure 6b,c. The binding energies of Zn 2p 3 and Sn 3d are centered at 121.5 and 486.7 eV, respectively, which are attributed to Zn-O and Sn-O bonds [38]. Additionally, the high-resolution O1s spectra is exhibited in Figure 6d-f. The location at 529.8 eV is ascribed to the O-Mn-O bond, and the peak centered at 531.2 eV belongs to the O-Mn-H bond [39,40]. Additionally, it can be clearly found that the binding energy of O1s at around 532.3 eV attributed to the weak bond (-OH) of the MnO 2 @ZHS binary hybrid is higher than that of MnO 2 . This is reasonable due to the fabrication of ZHS, which is accompanied by abundant hydroxyl groups.

Interfacial Adhesion Beween MnO2@ZHS Binary Hybrid and EP Matrix
The morphology of the fracture surface of the samples is observed by SEM test, in order to further investigate the interfacial interaction between the nanofillers and EP matrix. As can be observed in Figure 7a, the fracture surface of pure EP shows smooth and clean. After the addition of MnO2 nanosheets, the surface of EP/MnO2 with a 2 wt % loading appears rough and wrinkled, indicating the improved interfacial interaction between the MnO2 nanosheets and EP matrix ( Figure  7b). In the case of EP/MnO2@ZHS with different loadings, the wrinkles on the surface show deeper than the addition of MnO2, declaring that the interfacial interactions between the MnO2@ZHS binary hybrid and EP matrix are improved (Figures 7c-e). Moreover, as the loading of the MnO2@ZHS binary hybrid increases, the fish-like folds become more delicate. EP/MnO2@ZHS 2% exhibits the most delicate surface among the samples, which is reasonably due to the increase of MnO2@ZHS acting as stress points. Moreover, the EDS information of Figure 7e is listed in Figure 8, and the elements from the MnO2@ZHS binary hybrid can also be found. Therefore, the results obtained from SEM images demonstrate that the incorporation of MnO2@ZHS can improve the interfacial interaction between MnO2@ZHS and the EP matrix.

Interfacial Adhesion Beween MnO 2 @ZHS Binary Hybrid and EP Matrix
The morphology of the fracture surface of the samples is observed by SEM test, in order to further investigate the interfacial interaction between the nanofillers and EP matrix. As can be observed in Figure 7a, the fracture surface of pure EP shows smooth and clean. After the addition of MnO 2 nanosheets, the surface of EP/MnO 2 with a 2 wt % loading appears rough and wrinkled, indicating the improved interfacial interaction between the MnO 2 nanosheets and EP matrix (Figure 7b). In the case of EP/MnO 2 @ZHS with different loadings, the wrinkles on the surface show deeper than the addition of MnO 2 , declaring that the interfacial interactions between the MnO 2 @ZHS binary hybrid and EP matrix are improved (Figure 7c-e). Moreover, as the loading of the MnO 2 @ZHS binary hybrid increases, the fish-like folds become more delicate. EP/MnO 2 @ZHS 2% exhibits the most delicate surface among the samples, which is reasonably due to the increase of MnO 2 @ZHS acting as stress points. Moreover, the EDS information of Figure 7e is listed in Figure 8, and the elements from the MnO 2 @ZHS binary hybrid can also be found. Therefore, the results obtained from SEM images demonstrate that the incorporation of MnO 2 @ZHS can improve the interfacial interaction between MnO 2 @ZHS and the EP matrix.

Interfacial Adhesion Beween MnO2@ZHS Binary Hybrid and EP Matrix
The morphology of the fracture surface of the samples is observed by SEM test, in order to further investigate the interfacial interaction between the nanofillers and EP matrix. As can be observed in Figure 7a, the fracture surface of pure EP shows smooth and clean. After the addition of MnO2 nanosheets, the surface of EP/MnO2 with a 2 wt % loading appears rough and wrinkled, indicating the improved interfacial interaction between the MnO2 nanosheets and EP matrix ( Figure  7b). In the case of EP/MnO2@ZHS with different loadings, the wrinkles on the surface show deeper than the addition of MnO2, declaring that the interfacial interactions between the MnO2@ZHS binary hybrid and EP matrix are improved (Figures 7c-e). Moreover, as the loading of the MnO2@ZHS binary hybrid increases, the fish-like folds become more delicate. EP/MnO2@ZHS 2% exhibits the most delicate surface among the samples, which is reasonably due to the increase of MnO2@ZHS acting as stress points. Moreover, the EDS information of Figure 7e is listed in Figure 8, and the elements from the MnO2@ZHS binary hybrid can also be found. Therefore, the results obtained from SEM images demonstrate that the incorporation of MnO2@ZHS can improve the interfacial interaction between MnO2@ZHS and the EP matrix.

Thermal Stabilities of Pure EP and Its Nanocomposites Studied by TGA Test
The thermal stability of pure EP and EP nanocomposites is studied by TGA test, as shown in Figure 9. The onset decomposition temperature and the temperature at the maximal weight loss rate are denoted as T-10 and Tmax, respectively. Pure EP and EP nanocomposites exhibit the same degradation process, showing a one-stage degradation process, which is ascribed to the decomposition of macromolecular chains. Additionally, it can be seen that after the incorporation of MnO2, the char residue of EP/MnO2 at 800 °C increases compared with that of pure EP, but the T-10% and Tmax of EP/MnO2 decrease, which is possibly attributed to the catalysis effect from MnO2 promoting the predecomposition of the matrix and the degradation of metal oxides. In the case of EP/MnO2@ZHS nanocomposites, the values of T-10% and Tmax further decrease, which can be due to the predegradation of ZHS, thereby resulting in the predecomposition of the matrix. However, the increased residual content at 800 °C of EP/MnO2@ZHS nanocomposites is enhanced with the increase of the loading. Therefore, the addition of the MnO2@ZHS binary hybrid improves the char residue of the EP matrix.

Thermal Stabilities of Pure EP and Its Nanocomposites Studied by TGA Test
The thermal stability of pure EP and EP nanocomposites is studied by TGA test, as shown in Figure 9. The onset decomposition temperature and the temperature at the maximal weight loss rate are denoted as T -10 and T max , respectively. Pure EP and EP nanocomposites exhibit the same degradation process, showing a one-stage degradation process, which is ascribed to the decomposition of macromolecular chains. Additionally, it can be seen that after the incorporation of MnO 2 , the char residue of EP/MnO 2 at 800 • C increases compared with that of pure EP, but the T -10% and T max of EP/MnO 2 decrease, which is possibly attributed to the catalysis effect from MnO 2 promoting the predecomposition of the matrix and the degradation of metal oxides. In the case of EP/MnO 2 @ZHS nanocomposites, the values of T -10% and T max further decrease, which can be due to the predegradation of ZHS, thereby resulting in the predecomposition of the matrix. However, the increased residual content at 800 • C of EP/MnO 2 @ZHS nanocomposites is enhanced with the increase of the loading. Therefore, the addition of the MnO 2 @ZHS binary hybrid improves the char residue of the EP matrix.

Thermal Stabilities of Pure EP and Its Nanocomposites Studied by TGA Test
The thermal stability of pure EP and EP nanocomposites is studied by TGA test, as shown in Figure 9. The onset decomposition temperature and the temperature at the maximal weight loss rate are denoted as T-10 and Tmax, respectively. Pure EP and EP nanocomposites exhibit the same degradation process, showing a one-stage degradation process, which is ascribed to the decomposition of macromolecular chains. Additionally, it can be seen that after the incorporation of MnO2, the char residue of EP/MnO2 at 800 °C increases compared with that of pure EP, but the T-10% and Tmax of EP/MnO2 decrease, which is possibly attributed to the catalysis effect from MnO2 promoting the predecomposition of the matrix and the degradation of metal oxides. In the case of EP/MnO2@ZHS nanocomposites, the values of T-10% and Tmax further decrease, which can be due to the predegradation of ZHS, thereby resulting in the predecomposition of the matrix. However, the increased residual content at 800 °C of EP/MnO2@ZHS nanocomposites is enhanced with the increase of the loading. Therefore, the addition of the MnO2@ZHS binary hybrid improves the char residue of the EP matrix.

Flame Retardancy Evaluated by Cone Calorimeter
The cone calorimeter is generally regarded as the most powerful tool to evaluate the force burning fire performance of materials through simulating a developing fire scenario on a fixed-sized specimen. The parameters, including peak heat release rate (PHRR) and total heat release (THR), are recognized as the crucial data for evaluating the fire safety of matrix. HRR and THR curves of pure EP and EP nanocomposites are presented in Figure 10. According to Figure 10a, incorporating MnO 2 into the EP matrix can reduce the values of PHRR of EP nanocomposites, which is mainly attributed to the physical barrier effect. In the case of EP/MnO 2 @ZHS nanocomposites, all the values of PHRR are decreased as compared to that of pure EP. Moreover, as the amount of percent add-ons increases, the value of PHRR is further reduced. Thus, in terms of the decreased PHRR values, it can be indicated that the MnO 2 @ZHS hybrid shows the superior inhibition of heat release of the EP matrix. In addition, THR values of all EP nanocomposites shown in Figure 10b decrease obviously, in comparison with the pure one, showing that both MnO 2 and the MnO 2 @ZHS binary hybrid can effectively suppress the heat release of EP nanocomposites during combustion. Moreover, the THR value of EP/MnO 2 @ZHS with a 2 wt % loading can achieve near 40% reduction, compared with the control EP, which is rarely attained for the nanocomposites. The results of cone calorimetry illustrate that the function of charring formation of ZHS is successfully grafted on the MnO 2 nanosheets. The catalytic charring function, combined with the physical barrier effect from the MnO 2 nanosheets, can significantly enhance the flame retardancy of EP nanocomposites.

Flame Retardancy Evaluated by Cone Calorimeter
The cone calorimeter is generally regarded as the most powerful tool to evaluate the force burning fire performance of materials through simulating a developing fire scenario on a fixed-sized specimen. The parameters, including peak heat release rate (PHRR) and total heat release (THR), are recognized as the crucial data for evaluating the fire safety of matrix. HRR and THR curves of pure EP and EP nanocomposites are presented in Figure 10. According to Figure 10a, incorporating MnO2 into the EP matrix can reduce the values of PHRR of EP nanocomposites, which is mainly attributed to the physical barrier effect. In the case of EP/MnO2@ZHS nanocomposites, all the values of PHRR are decreased as compared to that of pure EP. Moreover, as the amount of percent add-ons increases, the value of PHRR is further reduced. Thus, in terms of the decreased PHRR values, it can be indicated that the MnO2@ZHS hybrid shows the superior inhibition of heat release of the EP matrix. In addition, THR values of all EP nanocomposites shown in Figure 10b decrease obviously, in comparison with the pure one, showing that both MnO2 and the MnO2@ZHS binary hybrid can effectively suppress the heat release of EP nanocomposites during combustion. Moreover, the THR value of EP/MnO2@ZHS with a 2 wt % loading can achieve near 40% reduction, compared with the control EP, which is rarely attained for the nanocomposites. The results of cone calorimetry illustrate that the function of charring formation of ZHS is successfully grafted on the MnO2 nanosheets. The catalytic charring function, combined with the physical barrier effect from the MnO2 nanosheets, can significantly enhance the flame retardancy of EP nanocomposites.

Char Residues Analysis of EP and EP/MnO2@ZHS Nanocomposites
Digital photos of char residues of pure EP and EP nanocomposites are plotted in Figure 11, in order to directly observe their morphologies. As shown in Figure 11a, pure EP exhibits lower amount of char residues compared with other EP nanocomposites incorporated by MnO2@ZHS binary hybrids. According to Figure 11d, the char residue of EP/MnO2@ZHS 2% shows a larger volume, illustrating a more effective catalytic charring formation by MnO2@ZHS. Additionally, the char residues of EP nanocomposites are covered by yellow-colored materials, which are the pyrolysis products obtained by the degradation of the MnO2@ZHS binary hybrids. Usually, for inorganic nanocomposites, the inorganic fillers would transfer to the surface of the matrix during the decomposition process. These yellow-colored degradation products can act as the insulation barrier, delaying the permeation of oxygen and flammable gases, thereby reducing the values of both PHRR and THR.

Char Residues Analysis of EP and EP/MnO 2 @ZHS Nanocomposites
Digital photos of char residues of pure EP and EP nanocomposites are plotted in Figure 11, in order to directly observe their morphologies. As shown in Figure 11a, pure EP exhibits lower amount of char residues compared with other EP nanocomposites incorporated by MnO 2 @ZHS binary hybrids. According to Figure 11d, the char residue of EP/MnO 2 @ZHS 2% shows a larger volume, illustrating a more effective catalytic charring formation by MnO 2 @ZHS. Additionally, the char residues of EP nanocomposites are covered by yellow-colored materials, which are the pyrolysis products obtained by the degradation of the MnO 2 @ZHS binary hybrids. Usually, for inorganic nanocomposites, the inorganic fillers would transfer to the surface of the matrix during the decomposition process. These yellow-colored degradation products can act as the insulation barrier, delaying the permeation of oxygen and flammable gases, thereby reducing the values of both PHRR and THR. To investigate the surface morphologies of char residues of pure EP and EP/MnO2@ZHS, the char residues studied here were obtained after cone calorimetry and imaged via SEM test. According to the char residue of pure EP portrayed in Figure 12a, deep cracks and collapses are obviously observed and thereby lead the oxygen and flammable gases to transfer easily. After the incorporation of MnO2@ZHS into the EP matrix, the char residue shows as more intact with less cracks (Figures  12b-d). Furthermore, as the loading increases up to 2 wt %, the surface of EP/MnO2@ZHS is very smooth, and there are no collapses and cracks, which can perfectly retard the permeation of oxygen and toxic and flammable products. Overall, SEM images visually confirm that the addition of MnO2@ZHS can significantly improve and perfect the char formation of EP nanocomposites. Besides, the Raman test was used to further study the inner graphitic structure of the char residues. As shown in Figures 12e-h, the Raman spectra of pure EP and EP nanocomposites all present two visible characteristic bands, namely, the D-band (1353 cm −1 ) and G-band (1592 cm −1 ), respectively attributed to the vibration of carbon atoms with dangling bonds in the plane terminations of disordered graphite or glass carbons and the vibration of sp 2 -bond carbon atoms in graphite layers. Generally, the ratio of the intensity of the D-band and G-band (ID/IG) is used to evaluate the graphitization degree of char residues. It is well recognized that sp 2 -bond carbon atoms in graphite layers can behave as insulation barriers retarding the transfer of heat, oxygen, and flammable products. For the value of ID/IG, the lower value means the better graphitic structure and physical insulation [41]. With the incorporation of MnO2@ZHS in the EP matrix, the value of ID/IG gradually becomes lower. As the loading increases, EP/MnO2@ZHS 2% exhibits the lowest value of ID/IG, compared with the others, indicating the most intact char structure. These results are in good agreement with the SEM results. The XPS test was employed to further explore the carbon structure of interior and exterior char residues of pure EP and EP/MnO2@ZHS 2% and their flame retardancy mechanism, as plotted in Figure 13. For all the high-resolution C 1s spectra, it can be deconvoluted to three characteristic peaks centered at 284.6, 285.6, and 288.3 eV, assigned to aliphatic and aromatic species, hydroxyl group To investigate the surface morphologies of char residues of pure EP and EP/MnO 2 @ZHS, the char residues studied here were obtained after cone calorimetry and imaged via SEM test. According to the char residue of pure EP portrayed in Figure 12a, deep cracks and collapses are obviously observed and thereby lead the oxygen and flammable gases to transfer easily. After the incorporation of MnO 2 @ZHS into the EP matrix, the char residue shows as more intact with less cracks (Figure 12b-d). Furthermore, as the loading increases up to 2 wt %, the surface of EP/MnO 2 @ZHS is very smooth, and there are no collapses and cracks, which can perfectly retard the permeation of oxygen and toxic and flammable products. Overall, SEM images visually confirm that the addition of MnO 2 @ZHS can significantly improve and perfect the char formation of EP nanocomposites. Besides, the Raman test was used to further study the inner graphitic structure of the char residues. As shown in Figure 12e-h, the Raman spectra of pure EP and EP nanocomposites all present two visible characteristic bands, namely, the D-band (1353 cm −1 ) and G-band (1592 cm −1 ), respectively attributed to the vibration of carbon atoms with dangling bonds in the plane terminations of disordered graphite or glass carbons and the vibration of sp 2 -bond carbon atoms in graphite layers. Generally, the ratio of the intensity of the D-band and G-band (I D /I G ) is used to evaluate the graphitization degree of char residues. It is well recognized that sp 2 -bond carbon atoms in graphite layers can behave as insulation barriers retarding the transfer of heat, oxygen, and flammable products. For the value of I D /I G , the lower value means the better graphitic structure and physical insulation [41]. With the incorporation of MnO 2 @ZHS in the EP matrix, the value of I D /I G gradually becomes lower. As the loading increases, EP/MnO 2 @ZHS 2% exhibits the lowest value of I D /I G , compared with the others, indicating the most intact char structure. These results are in good agreement with the SEM results. To investigate the surface morphologies of char residues of pure EP and EP/MnO2@ZHS, the char residues studied here were obtained after cone calorimetry and imaged via SEM test. According to the char residue of pure EP portrayed in Figure 12a, deep cracks and collapses are obviously observed and thereby lead the oxygen and flammable gases to transfer easily. After the incorporation of MnO2@ZHS into the EP matrix, the char residue shows as more intact with less cracks (Figures  12b-d). Furthermore, as the loading increases up to 2 wt %, the surface of EP/MnO2@ZHS is very smooth, and there are no collapses and cracks, which can perfectly retard the permeation of oxygen and toxic and flammable products. Overall, SEM images visually confirm that the addition of MnO2@ZHS can significantly improve and perfect the char formation of EP nanocomposites. Besides, the Raman test was used to further study the inner graphitic structure of the char residues. As shown in Figures 12e-h, the Raman spectra of pure EP and EP nanocomposites all present two visible characteristic bands, namely, the D-band (1353 cm −1 ) and G-band (1592 cm −1 ), respectively attributed to the vibration of carbon atoms with dangling bonds in the plane terminations of disordered graphite or glass carbons and the vibration of sp 2 -bond carbon atoms in graphite layers. Generally, the ratio of the intensity of the D-band and G-band (ID/IG) is used to evaluate the graphitization degree of char residues. It is well recognized that sp 2 -bond carbon atoms in graphite layers can behave as insulation barriers retarding the transfer of heat, oxygen, and flammable products. For the value of ID/IG, the lower value means the better graphitic structure and physical insulation [41]. With the incorporation of MnO2@ZHS in the EP matrix, the value of ID/IG gradually becomes lower. As the loading increases, EP/MnO2@ZHS 2% exhibits the lowest value of ID/IG, compared with the others, indicating the most intact char structure. These results are in good agreement with the SEM results. The XPS test was employed to further explore the carbon structure of interior and exterior char residues of pure EP and EP/MnO2@ZHS 2% and their flame retardancy mechanism, as plotted in Figure 13. For all the high-resolution C 1s spectra, it can be deconvoluted to three characteristic peaks centered at 284.6, 285.6, and 288.3 eV, assigned to aliphatic and aromatic species, hydroxyl group The XPS test was employed to further explore the carbon structure of interior and exterior char residues of pure EP and EP/MnO 2 @ZHS 2% and their flame retardancy mechanism, as plotted in Figure 13. For all the high-resolution C 1s spectra, it can be deconvoluted to three characteristic peaks centered at 284.6, 285.6, and 288.3 eV, assigned to aliphatic and aromatic species, hydroxyl group and/or ether, and carbonyl groups, respectively. It is generally accepted that the ratios of C ox /C a (C ox : oxidized carbon; C a : aliphatic and aromatic carbons) are employed to evaluate the thermal oxidative resistance of EP nanocomposites [18]. The lower value of C ox /C a means better thermal oxidative resistance. It is found that the values of C ox /C a in interior and exterior char residues of pure EP are 0.74 and 0.77, respectively, which are much higher than those of EP/MnO 2 @ZHS 2% (0.19 and 0.41). Therefore, based on these results, it can be well confirmed that MnO 2 @ZHS can promote the char density and graphitization degree of char residues, thereby efficiently enhancing the flame retardancy of the EP matrix.
Polymers 2018, 10, x FOR PEER REVIEW 11 of 16 and/or ether, and carbonyl groups, respectively. It is generally accepted that the ratios of Cox /Ca (Cox: oxidized carbon; Ca: aliphatic and aromatic carbons) are employed to evaluate the thermal oxidative resistance of EP nanocomposites [18]. The lower value of Cox /Ca means better thermal oxidative resistance. It is found that the values of Cox /Ca in interior and exterior char residues of pure EP are 0.74 and 0.77, respectively, which are much higher than those of EP/MnO2@ZHS 2% (0.19 and 0.41). Therefore, based on these results, it can be well confirmed that MnO2@ZHS can promote the char density and graphitization degree of char residues, thereby efficiently enhancing the flame retardancy of the EP matrix. Figure 13. The high-resolution C1s spectra of interior and exterior char residues of (a, b) pure EP and (c, d) EP/MnO2@ZHS 2%.

Gas-Phase Analysis of Pure EP and EP Nanocomposites
It is well known that the pyrolysis products of the EP matrix are extremely harmful to the health of humans. Here, TG-IR technique was adopted to study the composition of pyrolysis products during the degradation process of the EP matrix and further investigate the gas-phase mechanism. The FTIR spectra of the pyrolysis products of pristine EP and EP/MnO2@ZHS 2% are presented in Figure 14. It can be concluded that the composition of the pyrolysis products is rarely changed, and the characteristic peaks are strictly ascribed to the functional groups with unambiguous band positions, such as alkane groups mainly for allyl alcohol; hydrocarbons (2820-3120 m −1 ); aromatic compounds (750, 830, 1460, 1510, and 1610 cm −1 ); CO (2180 cm −1 ); CO2 (2356 cm −1 ); and water and/or phenol (ca. 3640-3670 cm −1 ) [26,33]. d) Figure 13. The high-resolution C1s spectra of interior and exterior char residues of (a,b) pure EP and (c,d) EP/MnO 2 @ZHS 2%.

Gas-Phase Analysis of Pure EP and EP Nanocomposites
It is well known that the pyrolysis products of the EP matrix are extremely harmful to the health of humans. Here, TG-IR technique was adopted to study the composition of pyrolysis products during the degradation process of the EP matrix and further investigate the gas-phase mechanism. The FTIR spectra of the pyrolysis products of pristine EP and EP/MnO 2 @ZHS 2% are presented in Figure 14. It can be concluded that the composition of the pyrolysis products is rarely changed, and the characteristic peaks are strictly ascribed to the functional groups with unambiguous band positions, such as alkane groups mainly for allyl alcohol; hydrocarbons (2820-3120 m −1 ); aromatic compounds (750, 830, 1460, 1510, and 1610 cm −1 ); CO (2180 cm −1 ); CO 2 (2356 cm −1 ); and water and/or phenol (ca. 3640-3670 cm −1 ) [26,33]. Figure 15a shows the Gram-Schmidt curves of EP and EP/MnO 2 @ZHS 2%. The incorporation of MnO 2 @ZHS binary hybrids obviously reduces the absorbance of pyrolysis products, rivalled by that of pure EP. Moreover, Figure 15b,c,f suggest that aromatic compounds and alkane groups are all effectively decreased, implying the reduction of organic gases and improved toxic effluent elimination. Additionally, Figure 15d,e respectively present the absorbance of CO and CO 2 , and it can be seen that as the absorbance of CO decreases, the absorbance of CO 2 increases for all the nanocomposites. This could not have an alternative explanation that less pyrolysis gases penetrate the char, making the combustion more ventilated. Therefore, it might very well be a catalysis effect for conversion of CO into CO 2 . In summary, TG-IR results demonstrate that MnO 2 @ZHS binary hybrids can efficiently suppress the production of toxic gases during the degradation of EP nanocomposites, and the addition of binary hybrids can catalyze conversion of CO into CO 2 . The mechanism of catalytic conversion of CO into CO 2 is proposed in Figure 16. Combined with the XPS results obtained from Figure 6, it is inferred that the capacity of Mn to adapt several oxidation states enables it to undergo oxygen reduction reactions, indicating MnO 2 nanosheets can act as a reservoir for oxygen, analogous to the previous work [42]. Certainly, toxic gases in real scenarios only contain CO instead of a wide array of hydrocarbons. Therefore, the catalytic conversion of CO into CO 2 is of great significance. In addition, the increase of CO 2 can dilute the oxygen and flammable gases in the air and thereby retard the combustion process.  Figure 15a shows the Gram-Schmidt curves of EP and EP/MnO2@ZHS 2%. The incorporation of MnO2@ZHS binary hybrids obviously reduces the absorbance of pyrolysis products, rivalled by that of pure EP. Moreover, Figures 15b,c,f suggest that aromatic compounds and alkane groups are all effectively decreased, implying the reduction of organic gases and improved toxic effluent elimination. Additionally, Figures 15d and e respectively present the absorbance of CO and CO2, and it can be seen that as the absorbance of CO decreases, the absorbance of CO2 increases for all the nanocomposites. This could not have an alternative explanation that less pyrolysis gases penetrate the char, making the combustion more ventilated. Therefore, it might very well be a catalysis effect for conversion of CO into CO2. In summary, TG-IR results demonstrate that MnO2@ZHS binary hybrids can efficiently suppress the production of toxic gases during the degradation of EP nanocomposites, and the addition of binary hybrids can catalyze conversion of CO into CO2. The mechanism of catalytic conversion of CO into CO2 is proposed in Figure 16. Combined with the XPS results obtained from Figure 6, it is inferred that the capacity of Mn to adapt several oxidation states enables it to undergo oxygen reduction reactions, indicating MnO2 nanosheets can act as a reservoir for oxygen, analogous to the previous work [42]. Certainly, toxic gases in real scenarios only contain CO instead of a wide array of hydrocarbons. Therefore, the catalytic conversion of CO into CO2 is of great significance. In addition, the increase of CO2 can dilute the oxygen and flammable gases in the air and thereby retard the combustion process.

Conclusions
In this work, a facile method was introduced to synthesize binary hybrids by fabricating ZHS cubes on MnO2 nanosheets with better flame retardancy and toxic effluent elimination of EP. XRD, TEM, and XPS results confirmed the successful synthesis of MnO2@ZHS binary hybrids. Moreover, the SEM and TEM ultrathin images indicated that MnO2@ZHS binary hybrids could achieve strong interfacial interaction with the EP matrix and be well dispersed in nanocomposites. Cone calorimeter test suggested that MnO2@ZHS binary hybrids could effectively suppress the curves of HRR and THR of EP nanocomposites, performing better than MnO2 or ZHS alone. Condensed-phase analysis revealed that MnO2@ZHS binary hybrids could promote the char density and graphitization degree of char residues and thereby successfully retard the permeation of oxygen and flammable gases. Additionally, it was concluded that MnO2@ZHS binary hybrids could efficiently suppress the production of toxic gases during the degradation of EP nanocomposites through the analysis of gas phase.

Conclusions
In this work, a facile method was introduced to synthesize binary hybrids by fabricating ZHS cubes on MnO 2 nanosheets with better flame retardancy and toxic effluent elimination of EP. XRD, TEM, and XPS results confirmed the successful synthesis of MnO 2 @ZHS binary hybrids. Moreover, the SEM and TEM ultrathin images indicated that MnO 2 @ZHS binary hybrids could achieve strong interfacial interaction with the EP matrix and be well dispersed in nanocomposites. Cone calorimeter test suggested that MnO 2 @ZHS binary hybrids could effectively suppress the curves of HRR and THR of EP nanocomposites, performing better than MnO 2 or ZHS alone. Condensed-phase analysis revealed that MnO 2 @ZHS binary hybrids could promote the char density and graphitization degree of char residues and thereby successfully retard the permeation of oxygen and flammable gases. Additionally, it was concluded that MnO 2 @ZHS binary hybrids could efficiently suppress the production of toxic gases during the degradation of EP nanocomposites through the analysis of gas phase.