Effect of Different Size-Modified Expandable Graphite and Ammonium Polyphosphate on the Flame Retardancy, Thermal Stability, Physical, and Mechanical Properties of Rigid Polyurethane Foam.
INTRODUCTIONIn order to prevent burning and delay fire spread, flame retardants (FRs) are commonly used in polymers [1-6]. The graphite intercalation compound (GIC), namely expandable graphite (EG), is known as a new generation of intumescent flame retardant, which is halogen-free, nondripping, and low smoke [7, 8]. In the event of a fire or thermal shock, EG will instantly expand and form "worm-like" expanded graphite. The produced loose "graphite worms" cover polymer pyrolysis zone and prevent molten drop of liquid pyrolysis product as well as mass and heat transfer. At the same time, the expansion process absorbs a large amount of heat, thus lowering the matrix temperature. In addition, C02, [H.sub.2]O, and S02 gases released in the redox reaction between graphite and [H.sub.2]S[O.sub.4]/HS[O.sub.4.sup.-] can dilute the concentration of flammable gas in the flame [9]. Because of its excellent performance, EG is widely used as a flame retardant additive for polymer materials such as ethylene vinyl acetate (EVA) [10], polyolefin [11, 12], acrylonitrile-butadiene-styrene (ABS) [13, 14], epoxy resin [15], and silicone rubber [16], However, the used EG is commodity, and it is prepared by using [H.sub.2]S[O.sub.4] as an intercalating agent, which makes the GIC usually contain high levels of sulfate and release more S[O.sub.2] gas during combustion.
Rigid polyurethane foam (RPUF) is widely used in insulation and space filler due to its excellent properties such as low thermal conductivity, high compression strength, good adhesive strength, and sound absorption property [17, 18]. However, the flammability and release of toxic gases restrict its widespread use. These disadvantages can be overcome by the introduction of FRs. In recent years, researchers have paid more attention to the flame retardant effect of EG on RPUF due to the high efficiency of EG and the matching of expansion temperature with the decomposition temperature of RPUF [19-21]. However, high dosages are usually used to achieve satisfactory flame retardancy, which makes the foaming process difficult, reduces the material compression property, and increases RPUF density [22], Furthermore, when EG is prepared with [H.sub.2]S[O.sub.4] as intercalator, the combustion will release more S02 gas [23, 24], In particular, the loose "graphite worm" residues are prone to collapse and even fly away under the influence of flame pressure or heat convection, which seriously reduces the efficiency in slowing down heat and mass transfer [25].
To overcome the aforementioned shortcomings, investigators have done a lot of work. First, the synergy between EG and other FRs can achieve excellent flame retardancy, so that the amount of EG can be reduced. For examples, the combination of EG with dimethyl methylphosphonate or aluminum hydroxide or ammonium polyphosphate (APP) can significantly improve the flame retardancy of RPUF by the combined actions of the villi-like or spherical matter and the "worm-like" char layer [26-28]. Second, modification of EG can reduce its sulfur content. EG can be modified by graphite intercalation reaction [29]. This is because, in addition to [H.sub.2]S[O.sub.4], the intercalation of the assistant intercalator not only reduces the relative content of sulfur, but also improves the flame retardancy. Furthermore, new substances or functional groups can be introduced by physical or chemical means [30]. For example, boric acid can be bonded on EG surface by the silane coupling agent (KH550) [31]. Finally, it has been confirmed that EG expansion volume and its particle size have the effects on the flame retardancy, mechanical properties, and thermal stability of polymer composites [32, 33].
Sodium tripolyphosphate (STPP) is a phosphate binder used in refractory materials [34]. We found that the EG prepared with [H.sub.2]S[O.sub.4] as main intercalator and STPP as assistant intercalator shows improved dilatability and reduced suffer content [35]. In view of the influence of GIC size on flame retardancy and effect of modification on EG expansion performance, we further prepared three different sizes of STPP-modified EG (EGP), which were marked as [EG.sub.P50], [EG.sub.P80], and [EG.sub.P100], respectively. Since the density of EGP is lower than that of ordinary inorganic FR, there is no significant effect on the foam density. Therefore, the size effect of the EGP on RPUF combustibility, thermal stability, physical and mechanical properties, diathermancy and hydrophobic property was characterized by limiting oxygen index (LOI), vertical burning level, cone calorimeter, thermogravimetry/ differential thermogravimetry (TG/DTG), density, compression strength, thermal conductivity, and contact angle tests. Moreover, the synergy between EG and APP can inhibit fly ash through the conglutination of APP decomposition product for the loose "worm-like" expanded graphite, the interaction between [EG.sub.P] and APP II was also investigated. A component formula to get RPUF with good flame retardancy, physical and mechanical properties and heat insulation performance was suggested.
EXPERIMENTAL
Materials and Reagents
Natural flake graphite with carbon content of 96 wt% and different sizes of 50, 80, and 100 mesh (corresponding to 300, 180, and 150 [micro]m, respectively) were provided by Qingdao Xite Carbon CO. LTD, Shandong, China. APP (II, n > 1,000) was purchased from Xusen Non-halogen Smoking Suppressing Fire Retardants Co. LTD, Zhejiang, China. Sodium tripolyphosphate (STPP) and [H.sub.2]S[O.sub.4] (98 wt%) were all analytical reagents and obtained from the local vendors.
The polyether polyol (YD-4110) supplied by Hebei Shijiazhuang Yadong Chemical Company of China holds a hydroxyl value of 460 [+ or -] 15 mg KOH [g.sup.-1] and a viscosity of 4,300 [+ or -] 500 mPa [s.sup.-1] at 25 [degrees]C. Polymethylene polyphenyl isocyanate (PM-200) with a NCO content of 31.3 wt% and a viscosity of 197 mPa [s.sup.-1] at 25 [degrees]C was provided by Yantai Wanhua Chemical Group Co. Ltd. of Shandong, China. The industry grade foam stabilizer silicone oil-178 was available from Jiangsu Mei Maysta Chemical Co. Ltd. of China. The methylene diamine (A-33) solution with a 33 wt% and N,N,N',N', N"-pentamethyldiethylenetriamine (PC-5) were all industry grade and provided by US Air Products Company, Arlington, USA. The industry grade foaming agent dichlorofluoroethane (141B-1) was supplied by Zhejiang Sanmei Chemical Co. Ltd. of China.
Preparation of EGP and the Reference EG ([EG.sub.0])
Natural graphite (C) with different sizes of 50, 80, and 100 mesh was used as material, respectively. [H.sub.2]S[O.sub.4] (98 wt%), STPP, and KMn[O.sub.4] were weighed in turn according to a mass ratio optimized by single factor experiment; [H.sub.2]S[O.sub.4] (98 wt%) was diluted to a mass concentration with deionized water and cooled to room temperature before use. Then, the quantified reactants were mixed in the order of the diluted [H.sub.2]S[O.sub.4], C, STPP, and KMn[O.sub.4] in beaker and then stirred at a certain temperature controlled by a water bath. After reaction, the products were washed with water until the pH value of the washing solution reached to 6.0-7.0. Continue to soak in deionized water for 2 h. After filtration and drying at 50-60 [degrees]C for about 5.0 h, a series of [EG.sub.P] were obtained and marked as [EG.sub.P50], [EG.sub.P80], and [EG.sub.P100], respectively (the subscript number is the particle size of graphite expressed in "mesh").
In the preparation of the reference [EG.sub.0], the [EG.sub.0.50] and [EG.sub.P50], [EG.sub.0.80] and [EG.sub.P80], [EG.sub.0,100] and [EG.sub.P100] are in one-to-one correspondence. The size of natural graphite, amount of KMn[O.sub.4] and [H.sub.2]S[O.sub.4], and the wt% concentration of [H.sub.2]S[O.sub.4] are the same, except that the amount of STPP is zero.
Preparation of Pure RPUF and its Flame-Retarded Composites
Pure RPUF was prepared by an oneshot, free rise procedure according to the reported method [36]. Firstly, the component I was prepared from that 100 g of polyether polyol, 2.0 g of silicon oil, 0.3 g of N,N,N',N',N"-pentamethyldiethylenetriamine, 1.5 g of triethylene diamine, 2.5 g of potassium acetate, 0.5 g of deionized water, 40 g of dichlorofluoroethane and a certain mass of FRs such as [EG.sub.P], the reference [EG.sub.0], APP, mixture of [EG.sub.P] and APP. FRs were added at different wt% of the total mass of component I and component II (polymethylene polyphenyl isocyanate). Component I was thoroughly mixed under vigorous agitation in a plastic breaker at ambient temperature until it became homogeneous, and then 100 g component II was added and thoroughly mixed with component I under a high-speed mechanical stirrer. The reaction maintained for 15-20 s at room temperature, and then the product was rapidly poured into an open mold with a size of 30 x 25 x 15 [cm.sup.3] (length x width x height) and kept for 24 h at room temperature to insure the foam free rise. Composition of the prepared RPUF foams was listed in Table 1. These samples were cut into the desired size, and then used for the evaluation of different properties.
Measurement and Characterization of Different Properties
Determination of expansion volume (EV) of the prepared GICs: 1.0 g sample in a quartz beaker with scale line was heated in muffle furnace at 800 [degrees]C for 30 s (air atmosphere). Measure the volume and calculate the average value of three parallel samples.
Determination of initial expansion temperature of the GICs: 1.0 g sample in a quartz beaker with scale line was heated in oven for 30 min (air atmosphere). When its volume becomes 1.5 times its original value, the used furnace temperature is defined as the initial expansion temperature. Calculate the average of the three measurements.
XRD data of the prepared GICs were collected using Y2000 X-ray diffractometer (model D8 Advance, Bruker-AXS, Karlsruhe, Germany) under the condition of Cu [K.sub.[alpha]] radiation ([lambda] = 0.15418, 40 kV, 30 mA) and 20 ranging from 20[degrees] to 70[degrees].
In FTIR analysis, the testing sample was ground with KBr into fine powders, and then the mixture was pressed into a disk. The FTIR spectra were recorded using a FTS-40 FTIR spectrometer (NICOLET380, Thermo Electron Corporation, 81 Wyman Street, Waltham, MA, 02454, America).
The macrostructure morphology of the expanded products of the prepared GICs and combustion residues of the flame retardant RPUF foams after cone calorimeter test (CCT) was recorded by a digital camera. Scanning electron microscope (SEM) (TM3000, Hitachi Electronics Co. Ltd., Tokyo, Japan) was used to observe the layer structure of graphite and the GICs. SEM observation of section microstructure of RPUF foams and combustion residues after CCT were also performed. All samples were coated with gold and then analyzed at an accelerating voltage of 15 kV.
Foam density was tested as described by ASTM D1622M2014 standard, and the compression strength at 10% strain in the direction parallel to foam growth was measured with UTM4204 instrument following the ASTM D 1621-94 standard. The sample size was 50 x 50 x 50 [mm.sup.3]. The movement rate of crosshead was fixed at 2.0 mm min-1, and the preload was set to 1 N. In the measurement of density and compression strength, five replicates were measured for each sample and the results were averaged.
Combustion characteristics of the prepared RPUF composites were evaluated in LOI, vertical burning level, and CCT analyses. LOI value was measured using an oxygen index instrument (Suzhou of China) following ASTM D 2863-97 standard on sheets with a size of 127 X 10 x 10 [mm.sup.3]. Vertical burning test was performed using a PX-01-005 vertical burning instrument (Suzhou of China) according to ASTM D 3801-96 standard on sheets sized 127 x 13 x 10 [mm.sup.3]. In the measurement of LOI and vertical burning level, the average values of five parallel samples were recorded. CCT test was carried out with a cone calorimeter (iCone Plus, Fire Test Technology Ltd. UK) according to ISO 5660 standard. The sample with a size of 100 x 100 x 30 [mm.sup.3] was wrapped in aluminum foil and horizontally exposed to an external heat flux of 50 kW [m.sup.-2]. The heat release rate (HRR) and total heat rate (THR) results were recorded.
The thermal stability measurement was performed using a STA 449C thermogravimetric analyzer (STA 449C, Netzsch Company, Selb, Germany) under [N.sub.2] atmosphere with a flux of 60 mL [min.sup.-1]. About 5.0 mg of sample was heated from 40 to 800 [degrees]C at a heating rate of 10 [degrees]C [min.sup.-1]. Change of residual yield along with temperature was reported.
At 22 [degrees]C and relative humidity of 27%, the thermal conductivity of the prepared foam was measured by using the transient hot wire method with conductometric instrument (model TC 3000E, Xiaxi Electronic Tech Co. Ltd., Xian, PRC) (GB/T, 2010). For solid samples, length, width, and height cannot be less than 25 x 25 x 0.3 [mm.sup.3]. Three replicates were measured for each sample and the results were averaged. Thermal conductivity was ensured 3% accuracy.
With distilled water used as contact liquid, a video optical contact angle measuring instrument (OCA15EC, Beijing Oriental Defei Instrument Co., Ltd. of China) was used to observe the contact angle of water, air, and RPUF foam. The changes of contact angle at 0, 5, 10, and 15 min were recorded. Three replicates were measured for each sample. The results were averaged and ensured 3% accuracy.
RESULTS AND DISCUSSION
Optimization of the EGP Preparing Conditions
The good flame retardancy of EG depends on its high EV [37], To ensure that each EGP gets a high EV, the mass ratio of graphite C, [H.sub.2]S[O.sub.4] (98 wt%), KMn[O.sub.4], and STPP, as well as reaction temperature, reaction time, and [H.sub.2]S[O.sub.4] concentration used in the reaction, were optimized by single factor experiments. The results were listed in Table 2. At the same time, to show the influence of STTP on GIC dilatability, a series of referenced [EG.sub.0] of [EG.sub.0,50], [EG.sub.0,80], and [EG.sub.0,100] were prepared. The size of natural graphite, amount of KMn[O.sub.4] and [H.sub.2]S[O.sub.4], and the wt% concentration of [H.sub.2]S[O.sub.4] are the same as each [EG.sub.P]. However, the amount of STPP is zero.
The results of reactant mass ratio, wt% concentration of [H.sub.2]S[O.sub.4], reaction temperature, reaction time, and the EV, initial expansion temperature of the prepared GICs are shown in Table 2 and Fig. 1. It can be seen that small size graphite particles need more KMn[O.sub.4] to get a good EV. Although the mass ratio of [H.sub.2]S[O.sub.4] (98 wt%) to graphite keeps a constant 5:1, [H.sub.2]S[O.sub.4] concentration used in the reaction decrease as graphite size decreases. This is because high concentration of [H.sub.2]S[O.sub.4] makes graphite peroxidize and form graphite oxide. The smaller the graphite, the more structural defects it has. Figure 1 indicates that there is a positive linear relationship between the particle size and EV, and a negative linear relationship between the particle size and initial expansion temperature. Each [EG.sub.P] or [EG.sub.0] with big size exhibits high EV but low initial expansion temperature. On the contrary, the [EG.sub.P] or [EG.sub.0] with small size holds low EV due to the released gases in the reaction between [H.sub.2]S[O.sub.4]/HS[O.sub.4.sup.-] and graphite easily escaping from the edge of the flake [38]. More importantly, the assistant intercalation of STPP has changed the GICs dilatability; each [EG.sub.P] possesses bigger EV than the same size of [EG.sub.0]. Among these GICs, [EG.sub.P50] shows the highest EV.
Structure and Morphology Characterization of the Prepared GICs
XRD analysis was carried out to investigate the influence of oxidation and intercalation effect of the prepared GICs. As shown in Fig. 2, the untreated natural graphite gives two characteristic diffraction peaks at 26.6[degrees] and 54.8[degrees] corresponding to 002 and 001 crystal face, respectively. The [EG.sub.P50], [EG.Psub.80], and [EG.sub.P100] show the same characteristic diffraction peaks like natural graphite. However, the diffraction peaks near 26.6[degrees] entirely move toward a smaller angle and correspond to larger interplanar crystal spacing of 3.53 [Angstrom], 3.47 [Angstrom], and 3.42 [Angstrom], respectively. Moreover, the intensity of 002 and 001 becomes weaker and weaker as the [EG.sub.P] size decreases, and they also show significant broadening due to lower integrity and more lattice defects, which means that the damage degree to periodical structure becomes more evident as the [EG.sub.P] size decreases. In oxidation reaction, the intercalation of anions and molecular into graphite layer structure results in an increase of crystal spacing, which widens the diffraction peak (corresponding to 002) and shifts it toward a smaller diffraction angle. And the more complete the oxidation and intercalation, the larger the interplanar spacing. At the same time, the more complete the oxidation of graphite, the more severe the damage of the 001 crystal plane structure.
Figure 3 is the layer structure photographs of natural graphite (a), [EG.sub.P50] (b), [EG.sub.P80] (c), and [EG.sub.P100] (d). Layers of the untreated natural graphite are compact, and the interlayer space is very small, which implies the strong force between layered structures. However, interlayer space of each [EG.sub.P] is significantly enlarged, and the edge structure becomes very loose due to the oxidation and intercalation reaction, especially for [EG.sub.P50]. It suggests that there is a weak interaction force between the layers.
When GICs are heated at 800 [degrees]C for 10~30 s, it will instantly expand and turn into expanded graphite as shown in Fig. 4. The SEM photograph of "graphite worm" of [EG.sub.P50] (Fig. 4a) indicates that it keeps a thick and complete "worm-like" shape. In contrast, the "graphite worms" of [EG.sub.P80] (Fig. 4b) and [EG.sub.P100] (Fig. 4c) are relatively thin and fragmentized. Figure 4d and f reveal the porous structure of these "graphite worms". As described in the literature [39, 40], the pores of expanded graphite can be classified into four levels. The first-level hole is a v-shaped open pore as shown in Fig. 4d, labeled as I. The size is approximately dozens of [micro]m. The second-level hole is a willow leaf-shaped pore as shown in Fig. 4d, labeled as II. The size ranges from a few [micro]m to dozens of [micro]m. The shape of the third-level hole is an irregular polygon with sizes ranging from 0.1 to 1 [micro]m. The fourth-hole is the nanopore with the size less than 0.1 [micro]m. It is these porous and loose "graphite worms" that cover on the surface of the polymer and form an intumescent barrier to prevent molten drop as well as mass and heat transfer (as shown in Fig. 8). However, it is foreseeable that these loose, low-density residues can easily collapse and even fly away under the influence of flame pressure or heat convection, resulting in fly ash and mass loss.
The TG characteristics of [EG.sub.P50], [EG.sub.P80], and [EG.sub.P100] under [N.sub.2] were investigated. As listed in Table 3, both the temperature corresponding to a 5% weight loss ([T.sub.5]), and the temperature corresponding to a 10% weight loss ([T.sub.10]) all decrease as the [EG.sub.P] size decreases. Among these GICs, [EG.sub.P50] shows the highest thermal stability with a residue yield of 83.86% at 800 [degrees]C. Although [EG.sub.P50], [EG.sub.P80], and [EG.sub.P100] are prepared with [H.sub.2]S[O.sub.4] and STPP as intercalators, their EV increases with the EGP size increases. The results indicate that the loading capacity of graphite for intercalators is different. Furthermore, the mass loss of GIC is mainly caused by the redox reaction between graphite and [H.sub.2]S[O.sub.4]/HS[O.sub.4.sup.-] [41], which leads to the release of C[O.sub.2], [H.sub.2]O, and S[O.sub.2] gases. These gases are easily escaping from the edge of small size [EG.sub.P] and then resulting in low thermal stability.
Influence of FRs on Foam Pore Structure
Figure 5 shows the SEM micrograph of pure RPUF and the flame retardant composites. Holes of the pure RPUF are rich with the diameter ranging from 50 to 250 [micro]m (see Fig. 5a). The closed pore percentage is close to 80% [42]. The addition of APP significantly decreases the cell size and the closed pore numbers. This is because APP can act as a nucleating agent in the foaming process (refer to Fig. 5b) [36]. Due to its small particle size and high density, the APP can be embedded in the wall. On the contrary, the addition of EGP, especially, the [EG.sub.P50] can damage the cellular structure in the direction parallel to foam growth, and then cause tearing, collapsing, and colliding of pores as shown in Fig. 5c and d, which directly reduces the closed cell numbers. The closed pore percentage of 80RPUF/[10EG.sub.P50] and 80RPUF/[10EG.sub.P50]/10APP is 40 and 67%, respectively. The particle size of [EG.sub.P80] is close to the pore size and can be distributed inside the cells (refer to Fig. 5e), thus 80RPUF/[10EG.sub.P80]/10APP shows a high closed pore percentage of about 89%. As for 80RPUF/[10EG.sub.P100]/10APP (refer to Fig. 5f), the worst dispersity in matrix results in bad uniformity in cell structure and size. The closed pore percentage is about 80%.
Influence of FRs on RPUF Density and Compression Strength
Density and compression strength of RPUF mainly depends on the type and amount of foaming agent and other additives [43]. In this work, dichlorofluoroethane and deionized water were used as foaming agents with the total dosage remaining unchanged. Meanwhile, EGp, APP, mixtures of [EG.sub.P] and APP were used as additives in order to improve the flame retardancy of RPUF. Results listed in Table 4 indicate that the addition of these fillers can increase the foam density, and it increases as the additive content increases. This is because the density of [EG.sub.P] and APP is higher than that of pure RPUF. When the total mass fraction is maintained at 20% and [EG.sub.p50] is partly replaced by APP, the foam density increases as the APP wt% increases. As for the influence of the [EG.sub.p] size, whether it is a single [EG.sub.P] system or an [EG.sub.P] and APP binary system, the density decreases as the GIC size decreases. It is mainly ascribed to the fact that the smaller particle size of [EG.sub.P] has a "nucleating" effect during the foam formation [32].
Compression strength is an important parameter determining foam applications in load bearing and packaging materials, and it generally increases with material density increase. Although the addition of [EG.sub.P50], [EG.sub.P80], or [EG.sub.P100] can increase the foam density, it shows a negative effect on foam compression strength (Table 4). Addition of [EG.sub.P50] causes tearing, collapsing, and colliding of pores, and then reduces the foam compression strength. The particle size of [EG.sub.P80] is close to the pore size and can be distributed inside the cells (refer to Fig. 5e). Therefore, the influence of [EG.sub.P80] on compression strength of 80RPUF/20[EG.sub.P80] and 80RPUF/10[EG.sub.P80]/10APP is relatively weak. Although the defect caused by the [EG.sub.P50] particles is more obvious than that of [EG.sub.P100], the increase of [EG.sub.P50] size leads to a decrease of the particle numbers. The reduction of defect number makes the strength of 80RPUF/20[EG.sub.P50] and 80RPUF/ 10[EG.sub.P50]/10APP higher than that of 80RPUF/20[EG.sub.P100] and 80RPUF/10[EG.sub.P100]/10APP. Since APP can be embedded in the wall to increase the wall strength. 80RPUF/20APP has the largest density and compression strength of 0.2226 MPa, increasing by about 68.5% than the pure RPUF. At the same time, the combination of APP and each [EG.sub.p] makes 80RPUF/10[EG.sub.P]/10APP show good mechanical property than 80RPUF/20[EG.sub.P]. Consideration of the low density and high compressive strength of foam, 80RPUF/ 10[EG.sub.P80]/10APP shows a good overall performance.
Influence of FRs on RPUF Combustion Behaviors
The influence of FRs on RPUF combustion LOI and vertical burning level were investigated. As shown in Table 5, pure RPUF is very flammable due to the cellular structure holding high level of radiant flux and large surface areas, and the used reagents in the preparation of foam being very flammable. Addition of the tested FRs can improve the LOI value. First, the LOI value of RPUF increases as the [EG.sub.P50] dosage increases, and it is improved by 50% when the amount is increased by 20%. Secondly, keeping the total dosage at 20%, 80RPUF/20[EG.sub.P50] shows higher LOI than the 80RPUF/20[EG.sub.0.50], 80RPUF/20STPP, and 80RPUF/12[EG.sub.p0,50]/ 8STPP (the mass ratio of [EG.sub.0.50] and STPP in 80RPUF/ 12[EG.sub.P0,50]/8STPP is calculated according to the theoretical intercalated dosage of STPP in the [EG.sub.P50]), and it is also higher than the calculated theoretical LOI result ([LOI.sub.the]) of 24.7%. The [LOI.sub.the] is calculated according to Eq. (1) using 80RPUF/20[EG.sub.0,50], 80RPUF/ 20STPP LOI values and [EG.sub.0,50], STPP wt% in 80RPUF/12[EG.sub.p0.50] 8STPP [44], and assume that there is a linear relationship between the number of FRs and the LOI. At high temperature, graphite will react with the intercalated [H.sub.2]S[O.sub.4]/HS[O.sub.4.sup.-] and release C[O.sub.2], S[O.sub.2], and [H.sub.2]O gases. These evolved gases can cause reduction of the effective concentration of the volatiles near the flame zone, which results in the improved LOI. At the same time, the high vertical burning level is attributed to the excellent char-forming property of GICs. Furthermore, the modification of [EG.sub.0,50] by STPP makes [EG.sub.p50] hold higher LOI value than the mechanical mixing of [EG.sub.0.50] and STPP. Moreover, there is synergistic effect between [EG.sub.0,50] and STPP, because 80RPUF/12[EG.sub.0,50]/8STPP gives a higher LOI than the [LOI.sub.the]. As for the influence of EGP size, the LOI values reduce as the particle size decreases. This is because there is a positive linear relationship between the [EG.sub.P] size and EV, and high EV is the key to improving the level of flame retardancy [45].
[mathematical expression not reproducible] (1)
Here, [LOI.sub.A] and [LOI.sub.B] are LOI value of component A and B in single flame retardant RPUF. The [wt.sub.A] and [wt.sub.B] are wt% of component A and B in A and B combined flame retardant RPUF.
Although the addition of 20 wt% APP can improve the LOI value of 80RPUF/20APP to 25.4%, the combustion is still accompanied by ignition. However, addition of [EG.sub.P50] and APP at different mass ratios not only raise the vertical burning level to V-0, but also significantly improve the LOI value. In particular, the results are significantly higher than the [LOI.sub.the] calculated based on Eq. (1), which confirms the synergistic efficiency. Meanwhile, the mass ratio of APP and [EG.sub.P50] has a certain influence on LOI, and the best ratio is 1:1. 80RPUF/10[EG.sub.P50]/10APP, 80RPUF/10[EG.sub.P80]/ 10APP, and 80RPUF/10[EG.sub.P100]/10APP show gradually reduced LOI. Consideration of the low density, high compressive strength, and good flame retardancy, 80RPUF/10[EG.sub.p80]/10APP shows a good overall performance.
The CCT results of pure RPUF and its flame retardant samples containing 20 wt% [EG.sub.P] or mixture of [EG.sub.P] and APP at 1:1 ratio were investigated. The relationship between HRR and time are shown in Fig. 6, and detailed data are summarized in Table 6. After ignition, pure RPUF quickly reaches the maximum combustion rate and gives the highest HRR of 200.9 KW [m.sup.-2], and the combustion finishes before 165 s with a THR of 12.0 MJ [m.sup.-2]. Addition of [EG.sub.p50] can significantly increase the burning time and reduce the HRR and THR. This may be attributed to the formed expanded graphite after fire treatment, which forms a sufficient barrier layer to hinder the heat penetration process. Although the influence of APP on HRR and burning time of 80RPUF/20APP is the same as [EG.sub.P50], its THR increases by 21.3% over pure RPUF. It is worth noting that the combination of [EG.sub.P50] and APP makes 80RPUF/10[EG.sub.P50]/10APP give the least HRR value of 92.2 KW [m.sup.-2] and THR value of 12.0 MW [m.sup.-2], reduced by 54% and 14.2% respectively, which is consistent with the highest LOI result listed in Table 5. As for 80RPUF/10[EG.sub.P50]/ 10APP, 80RPUF/10[EG.sub.P80]/10APP, and 80RPUF/10[EG.sub.P100]/10APP, the HRR and THR decrease with EGP size increases. This is because the EV and high temperature thermal stability of the [EG.sub.P50], [EG.sub.P80], and [EG.sub.P100] increase with the increase of EGP particle size, and more sufficient barrier layer is formed to hinder the heat transfer.
FTIR Analysis of RPUF Combustion Residues
Composition investigation of the residues can help to reveal the flame retardant mechanism. The FTIR spectrum of fire residues of pure RPUF, 80RPUF/20[EG.sub.P50], and 80RPUF/ 10[EG.sub.P50]/10APP were presented in Fig. 7. They all produce the characteristic stretching vibration absorption peaks of N-H near 3,430 [cm.sup.-1]. Peaks at about 2,350 [cm.sup.-1] are absorption of C[O.sub.2]. Peaks around 2,950-2,850 [cm.sup.-1] are typical for the stretching vibration of C-H, and the absorption strength of 80RPUF/ 20[EG.sub.p50], 80RPUF/10[EG.sub.P50]/10APP is significantly weaker than that of RPUF. At the same time, 1,620 [cm.sup.-1] is typical for an aromatic char structure, which confirms the existence of olefin in the residues. In the FTIR of 80RPUF/20[EG.sub.P50] and 80RPUF/ 10[EG.sub.P50]/10APP, the C=N stretching vibration peaks appear at 1500 [cm.sup.-1]. The result is attributed to the addition of [EG.sub.P50] and APP, which helps to terminate the decomposition reaction of RPUF, and results in incomplete combustion. At the same time, P=O absorption at 1250 [cm.sup.-1] is caused by the formation of polyphosphoric acid in the decomposition reaction of phosphate. In particular, characteristic absorption of P--O--C and P--O--P near 1,067 and 890 [cm.sup.-1] are observed in the FTIR of 80RPUF/ 10[EG.sub.P50]/10APP. This indicates that polyphosphoric acid has undergone a cross-linking reaction with the polyol produced in RPUF pyrolysis reaction [46, 47]. The reaction eventually leads to the formation of char rather than ether, and more phosphorus-containing products are formed in the condensed phase. The above results confirm that the addition of [EG.sub.P] and APP help to prevent decomposition reaction of RPUF, and APP can help to form a phosphor-carbonaceous polyaromatic structure in condensed phase.
Structures of Combustion Residues
To show that how the formation of intumescent char affects the combustion of the RPUF foams, the residues left after CCT were characterized with digital camera and SEM, respectively. The macrostructure and microstructure morphologies of fire residues of pure RPUF Fig. 8a and a', 80RPUF/20APP Fig. 8b and b', 80RPUF/20[EG.sub.p50] Fig. 8c and c', 80RPUF/10[EG.sub.p50]/10APP Fig. 8d and d', 80RPUF/10[EG.sub.P80]/10APP Fig. 8e and e', and 80RPUF/10[EG.sub.P100]/10APP Fig. 8f and f were investigated. The char layer of pure RPUF is very thin, indicating the combustion is very acute, and it contains numerous open holes caused by the released CO, C[O.sub.2], HCN, NO, HCOH gases [48]. Residue of 80RPUF/20[EG.sub.p50], by contrast, is thick and fluffy. The matrix surface is completely covered with "worm-like" particles due to the expansion of [EG.sub.P50]. However, it also shows typical "popcorn effect" due to the weak cohesiveness between graphite particles and RPUF matrix. In comparison, char layer of 80RPUF/20APP is thinner but compacter than the 80RPUF/20[EG.sub.P50]. The combination of [EG.sub.p50] and APP makes 80RPUF/10[EG.sub.P50]/10APP produce a relatively compact and continuous residue due to APP gluing expanded graphite together, and thus stabilizing the fire residue [49, 50]. It is the compact and continuous residue that effectively blocks the transfer of heat and volatile small molecules produced in the polymer decomposition reaction. Therefore, the flame retardancy of 80RPUF/10[EG.sub.P50]/10APP is significantly improved. As for the influence of EGP size, 80RPUF/10[EG.sub.P50]/10APP gives more dense char layer than 80RPUF/10[EG.sub.P80]/10APP and 80RPUF/10[EG.sub.p100]/ 10APP, because the EV increases with the increase of EGP size.
Thermal Stability of RPUFs. TG/DTG curves of pure RPUF, 80RPUF/20APP, 80RPUF/20[EG.sub.0,50], 80RPUF/20[EG.sub.P50], 80RPUF/ 10[EG.sub.P50]/10APP, 80RPUF/10[EG.sub.p80]/10APP, and 80RPUF/ 10[EG.sub.P100]/10APP in [N.sub.2] were tested and shown in Fig. 9, and the related data were listed in Table 7. Although the addition of FRs cannot change the two-step thermal degradation tendency of RPUF, there is a significant difference between pure RPUF and the flame retardant samples. Firstly, addition of [EG.sub.0,50], [EG.sub.P50] can improve [T.sub.5], [T.sub.max] (the temperature which corresponds to the maximum mass loss rate) of stage I and stage II by 11,7, and 13 K for 80RPUF/20[EG.sub.0,50] and 27, 9, and 16 K for 80RPUF/20[EG.sub.P50], respectively. At the same time, the residue yields at 800 [degrees]C ([R.sub.exp]) increase severally by 11.63% and 14.32% over RPUF. [EG.sub.0,50] and [EG.sub.P50] can expand at temperature below 250 [degrees]C, and the "graphite worms" produced in the burned layer can block the mass transfer and heat transfer, resulting in the decrease of the maximum mass loss rate ([R.sub.max]) and the increase of [T.sub.max] [51]. STPP auxiliary intercalator has a melting point higher than 600 [degrees]C, so it shows high thermal stability and can be used as a refractory material [34]. Therefore, modification of [EG.sub.P50] by STPP makes 80RPUF/ 20[EG.sub.P50] show higher [T.sub.5], [T.sub.max], and [R.sub.exp] than the 80RPUF/ 20[EG.sub.0,50].
Secondly, the addition of APP has a completely different influence on foam thermal stability. The [T.sub.5] and [T.sub.max] of 80RPUF/ 20APP are not only lower than 80RPUF/20[EG.sub.0,50] and 80RPUF/ 20[EG.sub.p50], but also lower than pure RPUF. The higher [R.sub.max,I] is mainly related with the decomposition reaction of APP, which leads to release of N[H.sub.3], [H.sub.2]O and formation of a highly cross-linked products of [P.sub.2][O.sub.5]. It is the produced polyphosphoric acid that promotes subsequent dehydration and carbonization reaction of matrix [46]. And then, the cohesive and dense residue blocks the transfer of volatile micromolecules, which further leads to a small [R.sub.max,II] and high [R.sub.exp] of 33.04%. The [R.sub.exp] is not only significantly higher than pure RPUF and APP [52], but also higher than that of 80RPUF/20[EG.sub.0,50] and 80RPUF/20[EG.sub.P50]. The results confirm the chemical char formation effect of APP on RPUF is more important than the char formation of the GICs.
Third, the combination of [EG.sub.P50] and APP at 1:1 fraction makes 80RPUF/10[EG.sub.P50] /10APP show smaller [T.sub.max], [R.sub.max,II] than 80RPUF/20APP and 80RPUF/20[EG.sub.P50]. This is because the produced polyphosphoric acid in APP decomposition reaction not only speeds the subsequent dehydration and carbonization reaction of RPUF, it also presents an adhesion effect for the produced "worm-like" expanded graphite, which enhances the compactness of residues. Eventually, [R.sub.exp] of 80RPUF/10[EG.sub.P50]/10APP is higher than the others, reaching 35.69%. It is also greater than the calculated theoretical value ([R.sub.cal]) of 31.49%, which is calculated according to Eq. (2) using the wt% of APP and EGP and the residue yields of 80RPUF/20APP, 80RPUF/20[EG.sub.P] at 800 [degrees]C. Results confirm that [EG.sub.P50] and APP have a good synergistic effect on improving high temperature thermal stability and residue.
[mathematical expression not reproducible] (2)
Here, [R.sub.exp,A] and [R.sub.exp,B] are [R.sub.exp] value of A and B in single component flame retardant sample.
Finally, the [T.sub.5] and [R.sub.exp] of 80RPUF/10[EG.sub.P50]/10APP, 80RPUF/ 10[EG.sub.P80]/10APP, and 80RPUF/10[EG.sub.P100]/10APP decrease as EGP size decreases. Whereas, the [T.sub.max] exhibits an inverse relationship. The results indicate that the larger the EGP size is, the better high-temperature thermal stability 80RPUF/10[EG.sub.p]/10APP obtains. This is because the large size [EG.sub.p] owns more intercalator and better dilatability, and then forms thicker residue [53].
Thermal Conductivity of RPUFs. Thermal conductivity is used to evaluate the thermal insulation property of foam [54]. Materials with a thermal conductivity lower than 0.030 W [m.sup.-1] [K.sup.-1] are called as insulative materials [55]. As presented in Table 8, the incorporation of the mentioned fillers all slightly increases the RPUFs thermal conductivity, especially for the tested GICs. This is because graphite and its GICs are thermal conductive materials [56]. Addition of 20 wt% [EG.sub.P50] increases the thermal conductivity of 80RPUF/20[EG.sub.P50] to 0.0321 W [m.sup.-1] [K.sup.-1], which is 26.9% higher than the pure RPUF. The incorporation of APP only slightly increases the result, because its addition also increases the number of holes. Meanwhile, the combination of APP and [EG.sub.P50] makes 80RPUF/10[EG.sub.p50]/10APP exhibit a thermal conductivity of 0.0292 W [m.sup.-1] [K.sup.-1]. As for the influence of the GIC particle size on thermal conductivity, it decreases as the [EG.sub.P] size decreases. This is mainly due to the following reasons. First, the density of the RPUF foam increases when these FRs are added, which increases the conduction through the solid phase. Second, the density of RPUF foam increases as GIC particle size increases, which reduces the radiation contribution, and then increases the value of the solid contribution [57]. Finally, as the GIC particle size increases, the open cell content in the foam increases, and this helps to increase the thermal conductivity.
Hydrophobic Property of the RPUF Composites. Application of RPUF is usually limited by the environment, especially by humidity. When the humidity is high, it will absorb moisture and then lead to hydrolysis of ester group, which further causes destruction of main chain and decrease of mechanical property and flame retardancy [58]. The value of the contact angle between material and water and its variation over time can reflect the hydrophilic/hydrophobic nature of the interface. Therefore, experiments were carried out to investigate the influence of these fillers on RPUF hydrophobic property. Results shown in Table 9 indicate that the contact angle of the investigated sample gradually decreases with time. As for pure RPUF, the droplet is completely absorbed by matrix within 15 min, so it shows the worst hydrophobic property. Addition of APP gives the biggest contact angle at 0 min, because the high density and small pore size make 80RPUF/20APP exhibit maximum surface tension in a short time. 80RPUF/ 20[EG.sub.p50] shows a bigger contact angle than the 80RPUF/20APP at 15 min. This may be due to the nonpolar nature of [EG.sub.P50], which helps to prevent water droplets from diffusing into the matrix. Especially, the combination of hydrophobicity of EGP and influence of APP on hole integrity results in the hydrophobicity of the RPUF/EGp/APP system superior to the RPUF/EGP or RPUF/APP system. Furthermore, 80RPUF/10[EG.sub.P80]/10APP always maintains the maximum contact angle within 15 min, thus shows better waterproof performance than other foams. At this point, the synergetic effect between the hydrophobic function of [EG.sub.P80] and influence of APP on hole diameter and integrity is optimal. Therefore, consideration of the low density, high compressive strength, good flame retardancy, low thermal conductivity, and high hydrophobic property, 80RPUF/10[EG.sub.p80]/10APP shows a good overall performance.
Possible Flame Retardant Mechanism. Combining the LOI, vertical burning, CCT, TG/DTG characteristics, residues morphology, and FTIR results, the possible flame retardant mechanism is described as Fig. 10.
As for the GIC, when in contact with heat source/flame source, it instantly expands and forms "graphite worm" that covers the surface of RPUF to slow down heat and mass transfer and interrupt matrix degradation [31]. In addition, the expansion of the [EG.sub.p] will consume a large amount of heat to lower the combustion temperature. Furthermore, the C[O.sub.2], [H.sub.2]O, and S[O.sub.2] gases released in the oxidation reaction between graphite and [H.sub.2]S[O.sub.4]/HS[O.sub.4.sup.-] will reduce combustible gas concentration. All of which will contribute to the formation of protective residue. The APP decomposition reaction can release N[H.sub.3], [H.sub.2]O gases, and its addition can contribute to gas phase flame retardation. It is also active in solid phase by promoting coke formation. The produced polymeric forms of phosphoric acid can not only cause adhesion between the loose "graphite worms," but also exhibit a catalytic action for RPUF de-polycondensation reaction [59], which promotes the formation of phosphor-carbonaceous polyaromatic structure. Therefore, it eventually leads to a significant increase in high-temperature residue. APP changes the pyrolysis behavior of RPUF through both physical synergy and chemical reaction, and its chemical char forming is more important than the physical function of the GIC. Moreover, the combination of GIC and APP not only produces a continuous, compact residual layer, inhibits "popcorn effect" of "graphite worms," but also improves the thermal stability at high temperature.
CONCLUSIONS
Three kinds of different size [EG.sub.p] have been prepared, and the result of dilatability confirms the linear relationship between the GICs size and EV. The size effect on thermal stability, flame retardancy, mechanical properties, pore cell structure, thermal conductivity, and hydrophobicity of RPUF foams are also validated. TG/DTG, LOI, vertical burning level, and CCT results indicate that those additives can significantly improve the thermal stability and flame retardancy of RPUF foam, and the effect increases as the [EG.sub.p] size increases. An approximately liner relationship between [EG.sub.p] size and LOI value is observed. The chemical char formation effect of APP on RPUF is more important than the char formation of the GICs. In particular, [EG.sub.P] and APP show synergistic effect in improvement for RPUF flame retardancy and thermal stability, and it also increases as the [EG.sub.P] size increases. Addition of each EGP will deteriorate the mechanical properties of RPUF. However, 80RPUF/20[EG.sub.p80] and 80RPUF/10[EG.sub.p80]/10APP show good compression. Except 80RPUF/20EGP5o, other foams are insulative materials. The combination of hydrophobicity of [EG.sub.P] and influence of APP on hole integrity and number makes 80RPUF/10[EG.sub.P50]/10APP, 80RPUF/10[EG.sub.P80]/10APP, 80RPUF/ 10[EG.sub.P100]/10APP show better hydrophobicity than 80RPUF/ 20[EG.sub.P50], 80RPUF/20[EG.sub.P80], 80RPUF/20[EG.sub.P100], and 80RPUF/ 20APP. Furthermore, 80RPUF/10[EG.sub.P80]/10APP shows a good overall performance of low density, high compressive strength, good flame retardancy, low thermal conductivity, and high hydrophobic property.
Nomenclature list
ABS acrylonitrile-butadiene-styrene
APP ammonium polyphosphate
C natural graphite
CCTs cone calorimeter tests
EG expandable graphite
[EG.sub.0] the referenced EG
[EG.sub.p] tripolyphosphate modified expandable graphite
EV expansion volume
EVA ethylene vinyl acetate
FR flame retardant
FTIR Fourier transform infrared spectroscopy
GIC graphite intercalation compound
HRR heat release rate, KW [m.sup.-2]
LOI limiting oxygen index, %
[LOI.sub.the] the calculated theoretical LOI result, %
N.D. not detected or not calculated
[R.sub.exp] experimental detected value of residue yield at
800 [degrees]C, %
[R.sub.the] theoretical calculated value of residue yield, %
[R.sub.max] the maximum mass loss rate, wt% [degrees][C.sup.-1]
RPUF rigid polyurethane foam
SEM scanning electron microscope
STPP sodium tripolyphosphate
[T.sub.5] temperature corresponding to a 5% weight loss,
[degrees]C
[T.sub.i0] temperature corresponding to a 10% weight loss,
[degrees]C
[T.sub.max] temperature corresponding to [R.sub.max], [degrees]C
TG/DTG thermogravimetric/differential thermal gravimetric
THR total heat release, MJ [m.sup.-2]
XRD X-ray diffraction spectroscopy
REFERENCES
[1.] A. Yehaskel, J. Hazard. Mater., 3, 357 (1980).
[2.] L. Chen and Y.Z. Wang, Polym. Advan. Techn., 21, 1 (2010).
[3.] H.D. Lu, L. Song, and Y. Hu, Polym. Advan. Techn., 22, 379 (2011).
[4.] S.K. Prashant, D.P. Hansora, B.N. Jitendra, and A. Chatteijee, Polym. Degrad. Stab., 130, 194 (2016).
[5.] B. Schartel and G. Schmaucks, Polym. Eng. Sci., 57, 1099 (2017).
[6.] S.U. Mahajan, A. Anton, C. Pryzbyla, R.D. Kulkarni, and V. A. Bhanu, Pain, 67, 89 (2017).
[7.] H.J. Kruger, W.W. Focke, W. Mhike, A. Taute, and A. Roberson, Fire Mater., 41, 573 (2017).
[8.] M.E. Mngomezulu, A.S. Luyt, S.A. Chappie, and M.J. John, Polym. Eng. Sci., 58, 1619 (2018).
[9.] A. Melezhyk and E. Nmemetov, Mater. Sci. Eng., 98, 012041 (2015).
[10.] M.J. Mochane and A.S. Luyt, Polym. Eng. Sci., 55, 1255 (2016).
[11.] R. Xie and B. Qu, J. Appl. Polym. Sci., 80, 1881 (2010).
[12.] R. Xie and B. Qu, J. Appl. Polym. Sci., 80, 1190 (2010).
[13.] S.G. Hong and S.Y. Chang, Fire Mater., 36, 277 (2012).
[14.] X.Y. Pang, X.Z. Shi, X.O. Kang, M.W. Duan, and M.Q. Weng, Polym. Composite., 37, 2673 (2016).
[15.] T. Benelli, L. Mazzocchetti, E. D'Angelo, M. Lanzi, F. Saraga, L. Sambri, M.C. Franchini, and L. Giorgini, Polym. Eng. Sci., 57, 621 (2017).
[16.] XL. Chen, J.L. Zhuo, W.K. Song, C.M. Jiao, Y. Qian, and S. X. Li, Polym. Advan. Technol., 25, 1530 (2014).
[17.] E. Kabakci, G. Sayer, E. Suvaci, O. Uysal, 1. Giiler, and M. Kaya, J. Appl. Polym. Sci., 134, (2017). https://doi.org/10. 1002/app.44870.
[18.] M.S.M. Azahari, A.Z.M. Rus, M.T. Zaliran, and S. Kormin, IOP Conf. Ser.: Mat. Sci. Eng., 226, 012009 (2017).
[19.] M. Modesti, A. Lorenzetti, F. Simioni, and G. Camino, Polym. Degrad. Stab., 77, 195 (2002).
[20.] C.Q. Wang, F.Y. Ge, J. Sun, and Z.S. Cai, J. Appl. Polym. Sci., 130, 916 (2013).
[21.] L. Liu, X. Wang, G. Wang, B. Xu, and W. Xu, J. Appl. Polym. Sci., 134, 45188 (2017).
[22.] H.J. Kim, Y. Kwon, and C.K. Kim, Nanosci. Nanotechno.. 14, 6048 (2014).
[23.] N.E. Sorokina, M.A. Khaskov, V.V. Avdeev, and I. V. Nikol'Skaya, Russ. J. Gen. Chem., 75, 162 (2005).
[24.] D.D.L. Chung, J. Mater. Sci., 51, 554 (2016).
[25.] J.D. Huang, Q.Q. Tang, W.B. Liao, G.C. Wang, W. Wei, and C. Z. Li, Ind. Eng. Chem. Res., 56, 5253 (2017).
[26.] S. Wang, L. Qian, and F. Xin, Polym. Composite., 39, 329 (2018).
[27.] Y. Yuan, H.Y. Yang, B. Yu, Y.Q. Shi, W. Wang, L. Song, Y. Hu, and Y.M. Zhang, Ind. Eng. Chem. Res., 50, 10813 (2016).
[28.] Y.L. Liu, J.Y. He, and R.J. Yang, Ind. Eng. Chem. Res., 54, 5876 (2015).
[29.] X.Y. Pang, Y. Tian, and M.Q. Weng, Polym. Composite., 36, 1407 (2015).
[30.] Y. Zhou, W.C. Sun, Z.Y. Ling, X.M. Fang, and Z.G. Zhang, Ind. Eng. Chem. Res., 56, 14799 (2017).
[31.] L.Z. Liu, W.L. Li, W.W. Cui, X.R. Zhang, and W. Ling, Pigm. Resin Technol., 4(5), 450 (2016).
[32.] Y. Li, J. Zou, S.T. Zhou, Y. Chen, H.W. Zou, M. Liang, and W. Z. Luo, J. Appl. Polym. Sci., 131, 39885 (2014).
[33.] A. Lorenzetti, B. Dittrich, B. Schartel, M. Roso, and M. Modesti, J. Appl. Polym. Sci., 134, 45173 (2017).
[34.] S. Goberis, I. Pundene, and V. Antonovich, Refract. Ind. Ceram., 46, 403 (2005).
[35.] Z.X. Zhai, X.Y. Pang, R.N. Lin, S.Y. Sun, and M.Q. Weng, Asian J. Chem., 27, 2971 (2015).
[36.] X.Y. Pang, W.S. Chang, R. Chang, and M.Q. Weng, Int. Polym. Proc., 33, 117 (2018).
[37.] M.Y. Shen, W.J. Chen, K.C. Tsai, C.F. Kuan, H.C. Kuan, H. W. Chou, and C.L. Chiang, Polym. Composite., 38, 2378 (2017).
[38.] X.C. Wang, T. Geng, J. Han, C.T. Liu, C.Y. Shen, L.S. Turng, and H.E. Yang, Polym, Composite., 39, E1429 (2018).
[39.] H. Zhao, W. Zhou, N.Z. Cao, W.C. Shen, and Y.P. Zheng, Chinese Mater. Sci. Eng.. 20, 153 (2002).
[40.] B. Redondo-Foj, P. Ortiz-Serna, M. Carsi, M.J. Sanchis, M. Culebras, C.M. Gomez, and A. Cantarero, Polym. Int., 64, 284 (2015).
[41.] Y.P. Chen, S.Y. Li, R.Y. Luo, X.M. Lv, and X.J. Wang, New Carbon Maters., 28, 435 (2013).
[42.] X. Wang, W. Wei, Y.S. Wang, and F.C. Wang, China Synthetic Resin Plastics, Sinopec Group Asset Management Co., Ltd., Beijing, 34, 80 (2017).
[43.] M. Thirumal, D. Khastgir, G.B. Nando, Y.P. Naik, and N. K. Singha, Polym. Degrad. Stab., 95, 1138 (2010).
[44.] M. Lewin, J. Fire. Sci., 17, 3 (1999).
[45.] B. Schilling, Kunstst-plast Eur., 87, 1004 (1997).
[46.] S. Zhou, L. Song, Z.Z. Wang, Y. Hu, and W.Y. Xing, Polym. Degrad. Stab., 93, 1799 (2008).
[47.] L.V. Pin, Y. Hu, K.L. Hu, and Z.Z. Wang, J. Anal. Appl. Pyroly., 86, 207 (2009).
[48.] S. Duquesne, M.L. Bras, S. Bourbigot, R. Delobel, G. Camino, B. Eling, C. Lindsay, T. Roels, and H. Vezin, J. Appl. Polym. Sci., 82, 3262 (2001).
[49.] Y. Zhang, X.L. Chen, and Z.P. Fang, J. Appl. Polym. Sci., 128, 2424 (2013).
[50.] Z.H. Zheng, J.T. Yan, H.M. Sun, Z.Q. Cheng, W.J. Li, H. Y. Wang, and X.J. Cui, Polym. Int., 63, 84 (2014).
[51.] X.L. Zhang, H.J. Duan, D.X. Yan, L.Q. Kang, W.Q. Zhang, J. H. Tang, and Z.M. Li, J. Appl. Polym. Sci., 132, 42364 (2015).
[52.] X.Y. Pang, R. Chang, and M.Q. Weng, Polym. Eng. Sci., 58, 2008 (2018).
[53.] L. Shi, Z.M. Li, B.H. Xie, J.H. Wang, C.R. Tian, and M. B. Yang, Polym. Int., 55, 862 (2006).
[54.] GB/T 10297-2015: Test method for thermal conductivity of nonmetal solid materials--hot-wire method.9-42 (2010).
[55.] GB/T: 50538-2010. Ministry of housing and urban-rural development of PRC. (2010).
[56.] S.T. Zhou, Y.Z. Lei, H.W. Zou, and M. Liang, Polym. Composite., 34, 1816 (2013).
[57.] W.Z. Xu. Study of preparation and performance of hydrophobic flame retardant polyurethane elastomer. A Dissertation for Doctor Degree of Engineering, Hefei: University of Science and Technology of China. (2014).
[58.] P. Acuna, M. Santiago-Calvo, F. Villafane, M.A. Rodriguez-Perez, J. Rosas, and D.Y. Wang, Polym. Composite., 40, E1705 (2018). https://doi.org/10.1002/pc.25127.
[59.] M. Modesti and A. Lorenzetti, Polym. Degrad. Stab., 78, 341 (2002).
Xiu-Yan Pang [iD], (1,2) Ya-Ping Xin, (1) Xiu-Zhu Shi, (1) Jian-Zhong Xu (1,2)
(1) College of Chemistry and Environmental Science, Hebei University, Baoding, 071002, China
(2) Flame Retardant Material and Processing Technology Engineering Technology Research Center of Hebei Province, Key Laboratory of Analytical Science and Technology of Hebei Province, Hebei University, Baoding, 071002, China
Correspondence to: X.-Y. Pang; e-mail: [email protected]
Contract grant sponsor: Natural Science Foundation of Hebei Province; contract grant number: B2015201028.
DOI 10.1002/pen.25123
Caption: FIG. 1. The relationship between particle size and EV and initial expansion temperature. [Color figure can be viewed at wileyonlinelibrary.coml
Caption: FIG. 2. XRD analysis of natural graphite, [EG.sub.P50], [EG.sub.P80]. and [EG.sub.P100]. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 3. SEM analyses for natural graphite (a), [EG.sub.P50] (b), [EG.sub.P80] (c), and [EG.sub.P100] (d).
Caption: FIG. 4. The "graphite worms" images of [EG.sub.p50] (a, d), [EG.sub.P80] (b, e), and [EG.sub.P100] (c, f). [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 5. The SEM morphologies of pure RPUF (a), 80RPUF/20APP (b), 80RPUF/20[EG.sub.P50] (c), 80RPUF/10[EG.sub.P5O]/10APP (d), 80RPUF/10[EG.sub.P80]/10APP (e), and 80RPUF/10[EG.sub.P100]/10APP (f).
Caption: FIG. 6. The relationship between HRR and time. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 7. Residue FTIR of RPUF (a), 80RPUF/20EGP5O (b), and 80RPUF/ 10[EG.sub.p50]/10APP (c). [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 8. Residual macrostructure and microstructure morphologies of pure RPUF (a, a'), 80RPUF/20APP (b, b'), 80RPUF/20[EG.sub.p50] (c, c'), 80RPUF/10[EG.sub.P50]/10APP (d, d'), 80RPUF/10[EG.sub.p80]/10APP (e, e'), and 80RPUF/10[EG.sub.P100]/ 10APP (f, f). [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 9. The TG and DTG curves of RPUF, 80RPUF/20APP, 80RPUF/20[EG.sub.0,50], 80RPUF/20[EG.sub.P50], 80RPUF/10[EG.sub.P50]/10APP, 80RPUF/10[EG.sub.P80]/10APP, and 80RPUF/10[EG.sub.P100]/10APP under [N.sub.2] condition. [Color figure can be viewed at wileyonlinelibrary.com]
Caption: FIG. 10. The flame retardant mechanism of EGP and APP for RPUF. [Color figure can be viewed at wileyonlinelibrary.com]
TABLE 1. Composition of the prepared RPUF foams.
Sample FR (wt %)
[EG.sub.P50] [EG.sub.0,50] STPP
RPUF 0 0 0
95RPUF/5[EG.sub.P50] 5 0 0
90RPUF/10[EG.sub.P50] 10 0 0
85RPUF/15[EG.sub.P50] 15 0 0
80RPUF/20[EG.sub.P5O] 20 0 0
80RPUF/20[EG.sub.0,50] 0 20 0
80RPUF/20STPP 0 0 20
80RPUF/12 [EG.sub.050/8STPP] 0 12 8
80 RPUF/20[EG.sub.P80] 0 0 0
80 RPUF/20[EG.sub.P100] 0 0 0
80RPUF/20APP 0 0 0
80RPUF/5E[EG.sub.P50]15APP 5 0 0
80RPUF/10[EG.sub.P50]/10APP 10 0 0
80RPUF/15[EG.sub.P80]5APP 15 0 0
80RPUF/10[EG.sub.P80]/10APP 0 0 0
80 RPUF/10[EG.sub.P100]/10APP 0 0 0
Sample FR (wt %)
[EG.sub.P80] [EG.sub.P100] APP
RPUF 0 0 0
95RPUF/5[EG.sub.P50] 0 0 0
90RPUF/10[EG.sub.P50] 0 0 0
85RPUF/15[EG.sub.P50] 0 0 0
80RPUF/20[EG.sub.P5O] 0 0 0
80RPUF/20[EG.sub.0,50] 0 0 0
80RPUF/20STPP 0 0 0
80RPUF/12 [EG.sub.050/8STPP] 0 0 0
80 RPUF/20[EG.sub.P80] 20 0 0
80 RPUF/20[EG.sub.P100] 0 20 0
80RPUF/20APP 0 0 20
80RPUF/5E[EG.sub.P50]15APP 0 0 15
80RPUF/10[EG.sub.P50]/10APP 0 0 10
80RPUF/15[EG.sub.P80]5APP 0 0 5
80RPUF/10[EG.sub.P80]/10APP 10 0 10
80 RPUF/10[EG.sub.P100]/10APP 0 10 10
TABLE 2. Preparation condition of each GIC (a)
KMn[O.sub.4] [H.sub.2]S[O.sub.4]
GIC (g [g.sup.~1]) (g [g.sup.-1])
[EG.sub.P50] 0.18 5.0
[EG.sub.0.50] 0.18 5.0
[EG.sub.P80] 0.2 5.0
[EG.sub.0.80] 0.2 5.0
[EG.sub.P100] 0.3 5.0
[EG.sub.0.100] 0.3 5.0
STPP Concentration
GIC (g [g.sup.-1]) of [H.sub.2]
S[O.sub.4] (wt%)
[EG.sub.P50] 0.7 80
[EG.sub.0.50] 0 80
[EG.sub.P80] 0.6 75
[EG.sub.0.80] 0 75
[EG.sub.P100] 0.5 75
[EG.sub.0.100] 0 75
Reaction Reaction EV
GIC temperature time (min) (mL
([degrees]C) [g.sup.-1])
[EG.sub.P50] 40 40 630
[EG.sub.0.50] 40 40 450
[EG.sub.P80] 40 40 580
[EG.sub.0.80] 40 40 320
[EG.sub.P100] 40 30 500
[EG.sub.0.100] 40 30 245
(a) As for the mass ratio between [H.sub.2]S[O.sub.4] and graphite
(g [g.sup.-1]), it is corresponding to the 98 wt% [H.sub.2]S[O.sub.4].
TABLE 3. Thermoanalysis data of each [EG.sub.P] in [N.sub.2]
atmosphere (a).
Specimen [T.sub.5] [T.sub.10] Residue yield (%)
([degrees]C) ([degrees]C)
[EG.sub.P50] 288 377 83.86
[EG.sub.P80] 262 346 82.35
[EG.sub.P100] 238 344 81.69
(a) [T.sub.5], [T.sub.10] the temperature corresponding to a 5%,
100% weight loss, respectively, [degrees]C.
TABLE 4. Influence of FRs on RPUF density and compression strength.
Sample Density (kg Compression strength
[m.sup.-3]) (MPa)
RPUF 31.0 0.1321
95RPUF/5[EG.sub.P50] 31.3 0.1197
90RPUF/10[EG.sub.P50] 34.3 0.1146
85RPUF/15[EG.sub.P50] 38.0 0.0991
80RPUF/20[EG.sub.P50] 40.0 0.0982
80RPUF/20[EG.sub.P80] 35.2 0.1076
80RPUF/20[EG.sub.P100] 33.4 0.0865
80RPUF/15[EG.sub.P50]/5APP 40.7 0.1092
80RPUF/10[EG.sub.P50]/10APP 43.1 0.1288
80RPUF/5E[EG.sub.P50]/15APP 47.1 0.1626
80RPUF/10[EG.sub.P80]/10APP 40.4 0.1392
80RPUF/10[EG.sub.P100]/10APP 37.2 0.1098
80RPUF/20APP 48.2 0.2226
TABLE 5. Results of LOI and vertical burning level of RPUFs (a).
LOI (%)
Sample LOI [LOI.sub.the] Vertical
burning
level
RPUF 19.5 N.D. N.D.
95RPUF/5[EG.sub.P50] 22.7 N.D. N.D.
90RPUF/10[EG.sub.P50] 25.5 N.D. V-0
85RPUF/15[EG.sub.P50] 27.7 N.D. V-0
80RPUF/20[EG.sub.P50] 28.3 24.7 V-0
80RPUF/20[EG.sub.0.50] 27.5 N.D. N.D.
80RPUF/20STPP 20.5 N.D. N.D.
80RPUF/12[EG.sub.P50]/8STPP 25.2 24.7 N.D.
80RPUF/20[EG.sub.P80] 27.9 N.D. V-0
80RPUF/20[EG.sub.100] 26.7 N.D. V-0
80RPUF/20APP 25.4 N.D. N.D.
80RPUF/5E[EG.sub.P50]/15APP 28.4 26.1 V-0
80RPUF/10[EG.sub.P50]/10APP 29.7 26.9 V-0
80RPUF/15[EG.sub.P50]/5APP 28.6 27.6 V-0
80RPUF/10[EG.sub.P80]/10APP 28.5 26.9 V-0
80RPUF/10[EG.sub.P100]/10APP 27.5 26.4 V-0
N.D.: not detected or not calculated.
(a) [LOI.sub.the]: the calculated theoretical LOI result.
TABLE 6. CCT results of the HRR, THR, and burning time (a).
Sample HRR (KW THR (MJ Burning
[m.sup.-2]) [m.sup.-2]) time (s)
RPUF 200.9 14.0 165
80RPUF/20[EG.sub.P50] 101.5 13.8 315
80RPUF/20APP 118.7 17.0 330
80RPUF/10[EG.sub.P50]/10APP 92.2 12.0 305
80RPUF/10[EG.sub.P80]/10APP 115.2 13.8 330
80RPUF/10[EG.sub.100]/10APP 126.0 16.6 250
(a) HRR: heat release rate, KW [m.sup.-2] THR: total heat release,
MJ [g.sup.-1].
TABLE 7. TG/DTG data of RPUF samples under [N.sub.2] condition (a).
[T.sub.max]
([degrees]C)
Sample [T.sub.5] I II
([degrees]C)
RPUF 233 333 458
80RPUF/20[EG.sub.0.50] 244 340 471
80RPUF/20[EG.sub.P50] 260 342 474
80RPUF/10[EG.sub.P50]/10APP 247 312 443
80RPUF/10[EG.sub.P80]/10APP 214 323 451
80RPUF/10[EG.sub.P100]/10APP 190 325 454
80RPUF/20APP 230 320 449
[R.sub.max] (wt%
[degrees][C.sup.-1])
Sample I 11
RPUF -7.94 -1.46
80RPUF/20[EG.sub.0.50] -7.33 -1.55
80RPUF/20[EG.sub.P50] -6.86 -1.56
80RPUF/10[EG.sub.P50]/10APP -8.10 -1.11
80RPUF/10[EG.sub.P80]/10APP -8.20 -0.91
80RPUF/10[EG.sub.P100]/10APP -8.20 -0.81
80RPUF/20APP -10.7 -1.22
Residual yield (%)
Sample [R.sub.exp] [R.sub.cal]
RPUF 15.62 N.D.
80RPUF/20[EG.sub.0.50] 27.25 N.D.
80RPUF/20[EG.sub.P50] 29.94 N.D.
80RPUF/10[EG.sub.P50]/10APP 35.69 31.49
80RPUF/10[EG.sub.P80]/10APP 33.82 N.D.
80RPUF/10[EG.sub.P100]/10APP 32.62 N.D.
80RPUF/20APP 33.04 N.D.
[T.sub.max]: temperature corresponding to the maximum mass loss
rate, [degrees]C; [R.sub.max]; the maximum mass loss rate, wt%
[degrees] [C.sup.-1]; [R.sub.max]: the maximum mass loss rate, wt%
[degrees] [C.sup.-1; [R.sub.exp]: the detected residue yields at
800 [degrees]C, %; [R.sub.cal]: the calculated residue yields, %.
(a) [T.sub.5]: temperature corresponding to a 5% weight loss,
[degrees]C.
TABLE 8. Thermal conductivity of RPUF composites.
Sample Thermal conductivity
(W [m.sup.-1] [K.sup.-1])
RPUF 0.0253
80RPUF/20[EG.sub.P50] 0.0321
80RPUF/20APP 0.0287
80RPUF/10[EG.sub.P50]/10APP 0.0292
80RPUF/10[EG.sub.P80]/10APP 0.0268
80RPUF/10[EG.sub.P100]/10APP 0.0264
TABLE 9. Changes of contact angle over time.
Sample Contact angle/0
0 min 5 min 10 min 15 min
RPUF 118.2 104.0 57.7 0
80RPUF/20[EG.sub.p50] 125.1 116.2 95.8 57.5
80RPUF/20APP 128.1 116.4 101.9 46.2
80RPUF/10[EG.sub.P50]/10APP 117.4 105.9 90.7 69.7
80RPUF/10[EG.sub.P80]/10APP 132.1 123.2 108.8 104.5
80RPUF/10[EGP.sub.100]/10APP 112.1 97.2 63.8 46.2
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Pang, Xiu-Yan; Xin, Ya-Ping; Shi, Xiu-Zhu; Xu, Jian-Zhong |
|---|---|
| Publication: | Polymer Engineering and Science |
| Article Type: | Report |
| Geographic Code: | 9CHIN |
| Date: | Jul 1, 2019 |
| Words: | 10785 |
| Previous Article: | Effects of SCF Content, Injection Speed, and CF Content on the Morphology and Tensile Properties of Microcellular Injection-Molded CF/PP Composites. |
| Next Article: | The Influence of the Dispersed Phase on the Morphology, Mechanical and Thermal Properties of PLA/PE-LD and PLA/PE-HD Polymer Blends and their... |
| Topics: | |