Flame Retardance and Smoke Suppression of CFA/APP/LDHs/EVA Composite

A new intumescent flame-retardants (IFR) system including the charing-foaming agent (CFA), ammonium polyphosphate (APP) and modified-layered double hydroxides (LDHs) with different transition metals (Ni, Co, Cu) were used in the ethylene vinyl acetate (EVA) matrix. Both the limiting oxygen index and the vertical burning tests indicate that the CFA/APP system and LDHs have significant synergistic flame retardant effects. The morphology of combustion residues indicates that the many pores of residues can prevent the melt dripping. The thermal analysis shows that the flame retardants obviously enhanced the thermal degradation temperature of ethylene-based chains of the composites. The cone calorimeter test reveals that the CO2 and combustion residues have an important influence on the reduction of heat release rate (HRR), the smoke production rate (SPR), the production rate and the mean release yield of CO. The composite containing Cu (ELDH-Cu) delivers an 82% reduction in peak heat release rate, while ELDH-Ni has the best CO suppression among all composites. This work not only confirms the flame retardance and smoke suppression of CFA/APP/LDH/EVA, but also provides an effective method for producing new flame retardants and smoke suppressants.


Introduction
The ethylene vinyl acetate (EVA) polymer is widely used as an insulator for wires and cables. However, the EVA not only is easily flammable, but also releases the toxic CO gas during combustion [1,2]. To reduce flammability and the release of CO for EVA, the addition of flame retardants is effective [3,4]. Brominated flame retardants are the most effective and show a good ratio of property to price, however, their uses have been limited to the consideration of life safety and environmental problems due to a large amount of smoke and hydrogen bromide produced during burning [5,6]. The layered double hydroxides (LDHs) is a large class of anionic clays with a general formula of [M 2+ 1−x M 3+ x (OH) 2 ] x+ [A n− ] x/n ·mH 2 O, where M 2+ and M 3+ are divalent and trivalent metal cations, x = M 3+ /(M 2+ +M 3+ ) molar ratio and A n− is an n − valent interlayer anion [7,8]. During combustion, LDHs lose the adsorbed water and interlayer water, which can dilute and cool the combustible products within the flame zone. Meanwhile, the mixed metal oxide and char residue generated by the polymer/LDHs composite can protect the polymer from the air exposure and in turn suppress smoke production [9,10]. Furthermore, since the composition of the metal layers of LDHs can be varied, there are enormous potential applications as flame retardants [11]. Based on the above advantages of LDHs, we have paid significant attention to the design and preparation of LDHs as a non-halogenated additive of EVA [12][13][14][15].
A triazines derivative (CFA) containing hydroxylethylamino, triazine rings and ethylenediamine into groups was synthesized by our group [16]. It is used both as a charring agent and as a foaming agent in intumescent flame retardants (IFR) due to its excellent flame retardant properties and water resistance. To improve the flame retardance and thermal stability of polymer (polypropylene, EVA, polyethylene), CFA and ammonium polyphosphate (APP), systems with different synergistic agents (such as zinc borate, zeolites, organic montmorillonite) have been investigated in the previous report [17][18][19][20][21][22]. However, the research on the combination of LDHs and CFA/APP as synergistic flame retardants in EVA composites has never been reported. In the present work, a new composite flame retardant CFA/APP/LDHs was used in the ethylene vinyl acetate (EVA) matrix. Furthermore, the schematic of CFA/APP/LDHs and EVA was shown in Figure 1. The LOI, UL-94 and HRR results confirm that the synergistic flame retardant effects between CFA/APP and LDHs. The release yield and production rates of CO2 essentially mirror the HRR data. Moreover, the relationship between the CO2 and CO release is observed by the cone calorimeter test.

Synthesis of the Modified-LDHs with M 2+ and M 3+ in a Molar Rate of 4:1
A mixed aqueous solution containing transition metal nitrate, Mg(NO3)2 and Al(NO3)3 with a molar ratio of transition metal cations (Ni 2+ or Co 2+ or Cu 2+ ): Mg 2+ :Al 3+ of 0.05:3.95:1, was slowly added to 100 mL of deionized water at 70 °C with continuously magnetic stirring. In the synthesis, the pH value was adjusted to the range of 8 to 9 by adding the NaOH-Na2CO3 mixed solution (1.2 mol/L NaOH and 0.9 mol/L Na2CO3). After the titration, a heavy suspension gel was obtained and then crystallized at 70 °C for 30 min in a microwave oven (XH-300A, the maximum power of 1000 W and a frequency of 2.45 GHz, Beijing Xianghu science and Technology Development Co. Beijing, China). The obtained precipitate was washed to pH 7 by deionized water, and then filtered. The sample was then dried in an oven at 70 °C until a constant weight was achieved. The above synthesized precursor 10 g and 0.15 g stearate sodium were dispersed into 200 mL of deionized water. The mixture was crystallized at 70 °C for 30 min in a microwave oven. Then the precipitate was washed continuously

Synthesis of the Modified-LDHs with M 2+ and M 3+ in a Molar Rate of 4:1
A mixed aqueous solution containing transition metal nitrate, Mg(NO 3 ) 2 and Al(NO 3 ) 3 with a molar ratio of transition metal cations (Ni 2+ or Co 2+ or Cu 2+ ): Mg 2+ :Al 3+ of 0.05:3.95:1, was slowly added to 100 mL of deionized water at 70 • C with continuously magnetic stirring. In the synthesis, the pH value was adjusted to the range of 8 to 9 by adding the NaOH-Na 2 CO 3 mixed solution (1.2 mol/L NaOH and 0.9 mol/L Na 2 CO 3 ). After the titration, a heavy suspension gel was obtained and then crystallized at 70 • C for 30 min in a microwave oven (XH-300A, the maximum power of 1000 W and a frequency of 2.45 GHz, Beijing Xianghu science and Technology Development Co. Beijing, China). The obtained precipitate was washed to pH 7 by deionized water, and then filtered. The sample was then dried in an oven at 70 • C until a constant weight was achieved. The above synthesized precursor 10 g and 0.15 g stearate sodium were dispersed into 200 mL of deionized water. The mixture was crystallized at 70 • C for 30 min in a microwave oven. Then the precipitate was washed continuously with water to eliminate the excessive stearate sodium. The final pH was washed to 7 and then filtered. The resulting LDHs were named LDH-Ni, LDH-Co, and LDH-Cu, respectively.

Preparation of Flame-Retarded EVA Composites
A new IFR consists of APP, CFA and LDHs. The mass ratio of CFA to APP was fixed at 1:3. Simultaneously, the loading of IFR is kept at 20 wt % in the composite. All CFA/APP/LDH/EVA composites were melted and mixed in an RM-200A torque rheometer (Harbin University of Science and Technology, Haebin, China) with a rotor speed of 60 rpm at 140 • C for 10 min. The composites are named as ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu, respectively. Table 1 shows the composition, limiting oxygen index (LOI) and vertical burning test (UL-94) data of the pure EVA and composites with 20 wt % of flame retardant.

Measurements
X-ray diffraction (XRD) was characterized using a D/MAX 2200 diffractometer (Rigaku, Tokyo, Japan) with λ = 1.5406 Å. Data were collected at the rate of 4 • /min and step 0.02 • with Cu Kα irradiation operated at 40 kV and 45 mA. Fourier transform infrared spectra (FTIR) were collected using a Nicolet FTIR360 spectrometer (Thermo Nicolet, Madison, WI, USA) (KBr pellet method, 4 cm −1 resolution, number of scans is 32 times). Scanning electron microscope (SEM) (FEI, Eindhoven, Holland) observations were carried out on a FEI-Sirion with a field emission of 20 kV. Thermal analysis was carried out with a thermogravimetric analyzer (TGA pyris 1 from Perkin Elmer, Perkin Elmer Co., Walther M, MA, USA) using a constant heating rate of 10 • C/min under the pure nitrogen atmosphere from 50 to 800 • C. The weight of the samples was kept within 3-4 mg. All cone calorimeter tests were carried out in a cone calorimeter (FTT0007, Hongkong Universal Technology Co., Hongkong, China) at an incident heat flux of 50 kW/m 2 according to the ISO 5660-1 standard. The polymer sample (100 × 100 × 5 mm 3 ) is placed horizontally on a balance holder. Figure 2A shows the XRD patterns of three LDHs. All patterns exhibit (003), (006), (012), (015), (018), (110) and (113) characteristic reflections of the LDHs structure, which can be indexed in a 3R polytype [23,24]. The basal spacing (d basal ) of LDH-Ni, LDH-Co and LDH-Cu is 7.67, 7.56 and 7.67 Å, respectively. The thickness of a brucite-like layer is 4.8 Å [25], the interlayer spacing of stearate and nitrate are approximately 26.72 Å [26] and 4.23 Å [27], respectively. Therefore, the d basal values of LDH-Ni, LDH-Co, and LDH-Cu indicate that the interlayer anions are CO 3 2− , without the coexistence Figure 2B gives the IR spectra of three LDHs. The bands at 3445 cm −1 and 1630 cm −1 are attributed to the O-H stretching vibrations of layer hydroxyl groups and interlayer water molecules, respectively [28]. The band at 451 cm −1 is ascribed to deformation of M-OH in the brucite-like layers. The band at 1376 cm −1 is assigned to the antisymmetric stretching mode ν 3 of the interlayer carbonate. Two bands around 839 cm −1 and 680 cm −1 also are characteristic for the ν 2 (out-of-plane deformation) and the ν 2 (in-plane bending) of carbonate [29]. The IR analysis also confirms that the interlayer anions are only CO 3 2− without the coexistence of NO 3 − (at 1384 cm −1 ).

XRD and IR Analysis of the Prepared LDHs
In addition, two bands at 2927 cm −1 and 2848 cm −1 are associated with the C-H stretching vibrations of stearate anions. The weak bands at 1768 cm −1 are the C=O vibrations of stearate anions [30,31]. These weak bands confirmed that the H-bonding interaction occurred between the carboxyl and the hydroxyl [32]. The combination of IR and XRD characterization indicates that the stearate was grafted on the surface of LDHs.
Appl. Sci. 2016, 6, 255 4 of 11 carbonate. Two bands around 839 cm −1 and 680 cm −1 also are characteristic for the ν2 (out-of-plane deformation) and the ν2 (in-plane bending) of carbonate [29]. The IR analysis also confirms that the interlayer anions are only CO3 2− without the coexistence of NO3 − (at 1384 cm −1 ). In addition, two bands at 2927 cm −1 and 2848 cm −1 are associated with the C-H stretching vibrations of stearate anions. The weak bands at 1768 cm −1 are the C=O vibrations of stearate anions [30,31]. These weak bands confirmed that the H-bonding interaction occurred between the carboxyl and the hydroxyl [32]. The combination of IR and XRD characterization indicates that the stearate was grafted on the surface of LDHs.  Figure 3 illustrates the TGA and DTG curves of three LDHs. The first stage corresponds to the mass losses of physical absorbed water and interlayer water. The mass losses of LDH-Ni, LDH-Co, and LDH-Cu are 14.8%, 15.3% and 15.6% in the TGA curves (30-200 °C), respectively. The second stage is associated with the dehydroxylation of the metal hydroxide layers, the degradation of carbonates and stearate at a higher temperature interval (200-800 °C). Furthermore, the maximum mass loss rate temperatures of hydroxyls and carbonates are at 315 and 349 °C for LDH-Ni and LDH-Co, respectively. In addition, the maximum mass loss rate temperatures of the stearate, residual hydroxyls and carbonates are at 375, 375 and 339 °C for LDH-Ni, LDH-Co and LDH-Cu in the DTG curves, respectively.    The second stage is associated with the dehydroxylation of the metal hydroxide layers, the degradation of carbonates and stearate at a higher temperature interval (200-800 • C). Furthermore, the maximum mass loss rate temperatures of hydroxyls and carbonates are at 315 and 349 • C for LDH-Ni and LDH-Co, respectively. In addition, the maximum mass loss rate temperatures of the stearate, residual hydroxyls and carbonates are at 375, 375 and 339 • C for LDH-Ni, LDH-Co and LDH-Cu in the DTG curves, respectively. carbonate. Two bands around 839 cm −1 and 680 cm −1 also are characteristic for the ν2 (out-of-plane deformation) and the ν2 (in-plane bending) of carbonate [29]. The IR analysis also confirms that the interlayer anions are only CO3 2− without the coexistence of NO3 − (at 1384 cm −1 ). In addition, two bands at 2927 cm −1 and 2848 cm −1 are associated with the C-H stretching vibrations of stearate anions. The weak bands at 1768 cm −1 are the C=O vibrations of stearate anions [30,31]. These weak bands confirmed that the H-bonding interaction occurred between the carboxyl and the hydroxyl [32]. The combination of IR and XRD characterization indicates that the stearate was grafted on the surface of LDHs.      Figure 4a,b illustrate the TGA and DTG curves of EVA and its composites with 20 wt % loading of flame retardants, respectively. The TGA curve of EVA showed two main mass loss steps with two mass loss rate maximums at about 353 • C and 433 • C in the DTG. The first step is attributed to the deacetylation of acetate side groups. The second step is due to the chain scission of ethylene leading to full thermal degradation at high temperatures [33,34]. From the DTG curves, it can be seen that, although the maximum mass loss rate temperature of the acetate side groups for all composites is almost the same as that of the pristine EVA at 353 • C, their maximum mass loss rate temperature of ethylene-based chains is higher than 433 • C of the pristine EVA, for ELDH-Ni and ELDH-Co at 473 • C, ELDH-0 and ELDH-Cu at 470 • C. The above thermal analysis results indicated that the flame retardants obviously enhanced the thermal degradation temperature of ethylene-based chains of the composites. The reason is that the flame retardants' released H 2 O, CO 2 , N 2 and NH 3 effectively delay the scission of ethylene-based chains. Furthermore, it can be seen that the maximal mass loss speed rate of ethylene-based chains for ELDH-Ni and ELDH-Cu is lower than the EVA, ELDH-Co and ELDH-0. to full thermal degradation at high temperatures [33,34]. From the DTG curves, it can be seen that, although the maximum mass loss rate temperature of the acetate side groups for all composites is almost the same as that of the pristine EVA at 353 °C, their maximum mass loss rate temperature of ethylene-based chains is higher than 433 °C of the pristine EVA, for ELDH-Ni and ELDH-Co at 473 °C, ELDH-0 and ELDH-Cu at 470 °C. The above thermal analysis results indicated that the flame retardants obviously enhanced the thermal degradation temperature of ethylene-based chains of the composites. The reason is that the flame retardants' released H2O, CO2, N2 and NH3 effectively delay the scission of ethylene-based chains. Furthermore, it can be seen that the maximal mass loss speed rate of ethylene-based chains for ELDH-Ni and ELDH-Cu is lower than the EVA, ELDH-Co and ELDH-0.

Flame Retardance and Smoke Suppression Mechanism of Composites
From Table 1, it can be seen that the LOI value of ELDH-0, from 18% which is of the pure EVA, increases rapidly to 28.5%. Then LOI values of ELDH-Ni, ELDH-Co and ELDH-Cu was increased to 30.3%, 29.7%, and 30.8% due to the addition of LDHs. The LOI test indicates that the CFA/APP system and LDHs have significant synergistic flame retardant effects. The vertical burning (UL-94) test shows that ELDH-Ni, ELDH-Co, and ELDH-Cu reach UL-94 V-2, V-0 and V-2 ratings, respectively. However, ELDH-0 without a UL 94 rating was achieved due to the presence of melts dripping. The UL-94 result also confirms that the synergistic flame retardant effects between CFA/APP and LDHs.
To further identify the synergetic effects between LDHs and CFA/APP on the flame retardance and the smoke suppression behavior of composites, Figure 5 shows the heat release rate (HRR) curves of the pure EVA and composites by the cone calorimeter test. The HRR is considered to have a significant influence on fire hazards. Simultaneously, the peak value of HRR (pK-HRR) is used to assess the intensity of a fire. It can be seen that the pure EVA burned rapidly after ignition, and its pK-HRR value rises to 1120 kW/m 2 . However, the pk-HRR values of ELDH-0, ELDH-Ni, ELDH-Co, and ELDH-Cu decrease to 258, 214, 231, and 206 kW/m 2 , respectively. The reductions in pk-HRR for composites are 77% greater than EVA. In particular, the reduction in pk-HRR of ELDH-Cu reaches 82%. The HRR results also indicate that the flame retardant synergism was observed when LDHs are added into the composites with CFA/APP.

Flame Retardance and Smoke Suppression Mechanism of Composites
From Table 1, it can be seen that the LOI value of ELDH-0, from 18% which is of the pure EVA, increases rapidly to 28.5%. Then LOI values of ELDH-Ni, ELDH-Co and ELDH-Cu was increased to 30.3%, 29.7%, and 30.8% due to the addition of LDHs. The LOI test indicates that the CFA/APP system and LDHs have significant synergistic flame retardant effects. The vertical burning (UL-94) test shows that ELDH-Ni, ELDH-Co, and ELDH-Cu reach UL-94 V-2, V-0 and V-2 ratings, respectively. However, ELDH-0 without a UL 94 rating was achieved due to the presence of melts dripping. The UL-94 result also confirms that the synergistic flame retardant effects between CFA/APP and LDHs.
To further identify the synergetic effects between LDHs and CFA/APP on the flame retardance and the smoke suppression behavior of composites, Figure 5 shows the heat release rate (HRR) curves of the pure EVA and composites by the cone calorimeter test. The HRR is considered to have a significant influence on fire hazards. Simultaneously, the peak value of HRR (pK-HRR) is used to assess the intensity of a fire. It can be seen that the pure EVA burned rapidly after ignition, and its pK-HRR value rises to 1120 kW/m 2 . However, the pk-HRR values of ELDH-0, ELDH-Ni, ELDH-Co, and ELDH-Cu decrease to 258, 214, 231, and 206 kW/m 2 , respectively. The reductions in pk-HRR for composites are 77% greater than EVA. In particular, the reduction in pk-HRR of ELDH-Cu reaches 82%. The HRR results also indicate that the flame retardant synergism was observed when LDHs are added into the composites with CFA/APP. For a long time, the char layer quality is considered to have an important influence on the flame retardance. Figure 6 shows the digital photographs of the residues after cone calorimeter tests. The residue of ELDH-0 has some holes. The residues of ELDH-Ni, ELDH-Co, and ELDH-Cu are not only thin, but also broken into several big pieces. To exhibit the internal morphology of the char residue, the SEM images, after cone calorimeter tests, are shown in Figure 7. The internal residue morphologies of four composites are obviously different. The residue of ELDH-Ni has a smooth structure with some holes of different sizes. The residues of ELDH-Co have many pores. These pores prevented melt dripping of ELDH-Co during combustion. The residues of ELDH-Cu and ELDH-0 have a compact morphology. All of the residues have no intumescent char layer, but metal oxide, char and APP promote charring via the cross-linking of reactive EVA fragments impeding the burning process by reducing the oxygen supply to the bulk phase [35].
The mechanism of the synergistic effects between APP, CFA and LDHs can be included in the following. The formation of carbonized barrier prevents or slows the transfer of heat, oxygen, and combustible volatiles into the pyrolysis zone, retarding the combustion process [35]. APP decomposes at 200-250 °C, it releases nonflammable gas such as N2 and NH3 to dilute the oxygen [36]. CFA decomposes at 319-450 °C, it releases H2O and NH3 [16]. LDHs decompose to form H2O and CO2, which could dilute the flammable gases and oxygen, and decrease the temperature [35]. Therefore, the pk-HRR values of composites are obviously lower than pure EVA. Furthermore, the LDH-Cu has a lower maximum mass loss rate temperature of hydroxyls, carbonates and stearate than LDH-Ni and LDH-Co. The consequent LDH-Cu generated H2O and CO2 can cool, dilute and isolate oxygen during combustion early and rapidly. Thus, the LDH-Cu shows the better flame retardance among composites. For a long time, the char layer quality is considered to have an important influence on the flame retardance. Figure 6 shows the digital photographs of the residues after cone calorimeter tests. The residue of ELDH-0 has some holes. The residues of ELDH-Ni, ELDH-Co, and ELDH-Cu are not only thin, but also broken into several big pieces. To exhibit the internal morphology of the char residue, the SEM images, after cone calorimeter tests, are shown in Figure 7. The internal residue morphologies of four composites are obviously different. The residue of ELDH-Ni has a smooth structure with some holes of different sizes. The residues of ELDH-Co have many pores. These pores prevented melt dripping of ELDH-Co during combustion. The residues of ELDH-Cu and ELDH-0 have a compact morphology. All of the residues have no intumescent char layer, but metal oxide, char and APP promote charring via the cross-linking of reactive EVA fragments impeding the burning process by reducing the oxygen supply to the bulk phase [35].    To verify the effect of CO2, Figure 8 shows the production rate of CO2 (CO2P) curves of EVA and composites. A crucial relationship between HRR and CO2P was obtained. The shapes of CO2P curves are consistent with the HRR curves. The peak values of CO2P (pk-CO2P) time of the pure EVA, ELDH-0, ELDH-Ni, ELDH-Co, and ELDH-Cu are 235, 67, 125, 145 and 112 s, respectively. The above pk-CO2P times are same as the pk-HRR time. The result indicates that when the production rate of CO2 reaches to the pk-CO2P values, the HRR of samples reaches the pk-HRR values. Additionally, from the cone calorimeter test data, one can know that the MCO2Y is the mean release yield of CO2 of burning materials. The MCO2Y values of EVA, ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu are 2.315, 2.408, 2.857, 2.638 and 2.834 kg/kg, respectively. The MCO2Y values of composites significantly increase because of the addition of LDHs. ELDH-Ni, ELDH-Co and ELDH-Cu release more CO2, which dilutes the flammable gases during combustion. Therefore, the composites with LDHs show reduced pk-HRR values compared to pure EVA and ELDH-0. This finding confirmed that the release yield and production rates of CO2 essentially mirror the HRR data [2]. The mechanism of the synergistic effects between APP, CFA and LDHs can be included in the following. The formation of carbonized barrier prevents or slows the transfer of heat, oxygen, and combustible volatiles into the pyrolysis zone, retarding the combustion process [35]. APP decomposes at 200-250 • C, it releases nonflammable gas such as N 2 and NH 3 to dilute the oxygen [36]. CFA decomposes at 319-450 • C, it releases H 2 O and NH 3 [16]. LDHs decompose to form H 2 O and CO 2 , which could dilute the flammable gases and oxygen, and decrease the temperature [35]. Therefore, the pk-HRR values of composites are obviously lower than pure EVA. Furthermore, the LDH-Cu has a lower maximum mass loss rate temperature of hydroxyls, carbonates and stearate than LDH-Ni and LDH-Co. The consequent LDH-Cu generated H 2 O and CO 2 can cool, dilute and isolate oxygen during combustion early and rapidly. Thus, the LDH-Cu shows the better flame retardance among composites.
To verify the effect of CO 2 , Figure 8 shows the production rate of CO 2 (CO 2 P) curves of EVA and composites. A crucial relationship between HRR and CO 2 P was obtained. The shapes of CO 2 P curves are consistent with the HRR curves. The peak values of CO 2 P (pk-CO 2 P) time of the pure EVA, ELDH-0, ELDH-Ni, ELDH-Co, and ELDH-Cu are 235, 67, 125, 145 and 112 s, respectively. The above pk-CO 2 P times are same as the pk-HRR time. The result indicates that when the production rate of CO 2 reaches to the pk-CO 2 P values, the HRR of samples reaches the pk-HRR values. Additionally, from the cone calorimeter test data, one can know that the MCO 2 Y is the mean release yield of CO 2 of burning materials. The MCO 2 Y values of EVA, ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu are 2.315, 2.408, 2.857, 2.638 and 2.834 kg/kg, respectively. The MCO 2 Y values of composites significantly increase because of the addition of LDHs. ELDH-Ni, ELDH-Co and ELDH-Cu release more CO 2 , which dilutes the flammable gases during combustion. Therefore, the composites with LDHs show reduced pk-HRR values compared to pure EVA and ELDH-0. This finding confirmed that the release yield and production rates of CO 2 essentially mirror the HRR data [2]. It is well known that most injuries and deaths result from the smoke and toxic gases in the case of fire [37]. From Figure 9A, it can be seen that the peak values of smoke production rate (pk-SPR) of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu are 0.053, 0.054, 0.057 and 0.036 m 2 /s, respectively. These values are significantly lower than that of EVA (0.109 m 2 /s). Moreover, the pk-SPR time of ELDH-Ni and ELDH-Co are extended to 994 and 793 s, respectively. They are much longer than the 89 s of ELDH-0 and 251 s of EVA. Moreover, Figure 9B shows the production rate of CO (COP) curves of EVA and composites. The COP curves of all composites are significantly declined compared with EVA. The peak values of COP (pk-COP) for ELDH-Ni, ELDH-Co and ELDH-Cu are 0.0022, 0.0026, and 0.0025 g/s, respectively. They are obviously smaller than 0.0038 g/s of ELDH-0 and 0.0128 g/s of EVA. Simultaneously, MCOY is the mean release yield of CO of burning materials. The MCOY value of EVA is 0.031 kg/kg, while the MCOY values of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu decrease to 0.0234, 0.00765, 0.0194 and 0.0102 kg/kg, respectively. MCOY values of composites show an evident decrease compared with EVA. In particular, ELDH-Ni has the smallest MCOY value among composites. The extension of pk-SPR time, the reduction of pk-SPR, pk-COP and MCOY for composites are important contributions to people's evacuation and the rescue of firefighters. At the same time, the relationship between the CO2 and CO release is observed by the cone calorimeter test. When the MCO2Y is increased, the pk-COP and MCOY values are decreased. That is, the change tendency of the MCO2Y values is precisely opposite to the change tendency of pk-COP and MCOY values for EVA and composites.  It is well known that most injuries and deaths result from the smoke and toxic gases in the case of fire [37]. From Figure 9A, it can be seen that the peak values of smoke production rate (pk-SPR) of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu are 0.053, 0.054, 0.057 and 0.036 m 2 /s, respectively. These values are significantly lower than that of EVA (0.109 m 2 /s). Moreover, the pk-SPR time of ELDH-Ni and ELDH-Co are extended to 994 and 793 s, respectively. They are much longer than the 89 s of ELDH-0 and 251 s of EVA. Moreover, Figure 9B shows the production rate of CO (COP) curves of EVA and composites. The COP curves of all composites are significantly declined compared with EVA. The peak values of COP (pk-COP) for ELDH-Ni, ELDH-Co and ELDH-Cu are 0.0022, 0.0026, and 0.0025 g/s, respectively. They are obviously smaller than 0.0038 g/s of ELDH-0 and 0.0128 g/s of EVA. Simultaneously, MCOY is the mean release yield of CO of burning materials. The MCOY value of EVA is 0.031 kg/kg, while the MCOY values of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu decrease to 0.0234, 0.00765, 0.0194 and 0.0102 kg/kg, respectively. MCOY values of composites show an evident decrease compared with EVA. In particular, ELDH-Ni has the smallest MCOY value among composites. The extension of pk-SPR time, the reduction of pk-SPR, pk-COP and MCOY for composites are important contributions to people's evacuation and the rescue of firefighters. At the same time, the relationship between the CO 2 and CO release is observed by the cone calorimeter test. When the MCO 2 Y is increased, the pk-COP and MCOY values are decreased. That is, the change tendency of the MCO 2 Y values is precisely opposite to the change tendency of pk-COP and MCOY values for EVA and composites. It is well known that most injuries and deaths result from the smoke and toxic gases in the case of fire [37]. From Figure 9A, it can be seen that the peak values of smoke production rate (pk-SPR) of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu are 0.053, 0.054, 0.057 and 0.036 m 2 /s, respectively. These values are significantly lower than that of EVA (0.109 m 2 /s). Moreover, the pk-SPR time of ELDH-Ni and ELDH-Co are extended to 994 and 793 s, respectively. They are much longer than the 89 s of ELDH-0 and 251 s of EVA. Moreover, Figure 9B shows the production rate of CO (COP) curves of EVA and composites. The COP curves of all composites are significantly declined compared with EVA. The peak values of COP (pk-COP) for ELDH-Ni, ELDH-Co and ELDH-Cu are 0.0022, 0.0026, and 0.0025 g/s, respectively. They are obviously smaller than 0.0038 g/s of ELDH-0 and 0.0128 g/s of EVA. Simultaneously, MCOY is the mean release yield of CO of burning materials. The MCOY value of EVA is 0.031 kg/kg, while the MCOY values of ELDH-0, ELDH-Ni, ELDH-Co and ELDH-Cu decrease to 0.0234, 0.00765, 0.0194 and 0.0102 kg/kg, respectively. MCOY values of composites show an evident decrease compared with EVA. In particular, ELDH-Ni has the smallest MCOY value among composites. The extension of pk-SPR time, the reduction of pk-SPR, pk-COP and MCOY for composites are important contributions to people's evacuation and the rescue of firefighters. At the same time, the relationship between the CO2 and CO release is observed by the cone calorimeter test. When the MCO2Y is increased, the pk-COP and MCOY values are decreased. That is, the change tendency of the MCO2Y values is precisely opposite to the change tendency of pk-COP and MCOY values for EVA and composites.

Conclusions
A novel IFR system consisting of CFA, APP, and LDHs was used in the EVA composites. The many pores of ELDH-Co residues prevented melt dripping of ELDH-Co. For LDH-Cu composite, the generated H 2 O and CO 2 can cool, dilute and isolate oxygen early and rapidly during combustion. The LOI, UL-94 and HRR results confirm the synergistic flame retardant effects between CFA/APP and LDHs. Simultaneously, the change tendency of the MCO 2 Y values is precisely opposite to the pk-COP and MCOY values. The composite ELDH-Ni has the smallest pk-COP and MCOY value, while LDH-Cu has the best HRR, LOI and SPR values among all of the composites. This work not only confirms the flame retardance and smoke suppression mechanism of CFA/APP/LDHs/EVA composite, but also contributes to safety, the environment, and public health.