Improving the ablation resistance of epoxy modified organosilicon resin synergistic modified with B2O3 and ZrSi2

The boron oxide and zirconium silicide synergistic modified epoxy modified organosilicon resin composites were prepared and the ablation resistance properties of the composites were analyzed. With the synergistic effect of boron oxide and zirconium silicide, the properties, such as: the residue yield, hardness, tensile strength and ablation resistance (oxyacetylene ablation and static muffle furnace ablation) of the samples are markedly enhanced. The oxyacetylene linear ablation rate of the modified composite was observed to the minimum (0.04 mm s−1), when the contents of resin, boron oxide and zirconium silicide are 100 g, 50 g, and 150 g, respectively. The skeleton-like compact structure formed in the surface of composites after oxyacetylene ablation, which results in the best ablation resistance.


Introduction
Resin based composites with high specific strength, excellent efficiency of thermal insulation, and good designability have been widely used as thermal protection materials in aerospace thermal protection field [1][2][3].However, the operating environment such as heat flux becomes much severer ascribed to the higher speed of new generation aircraft, and consequently increases the ablation rate of resin based composite materials [4][5][6].Under this condition, some efforts need to be made to improve the corresponding ablation resistance to meet the needs of application demands.
In order to improve the ablation resistance of resin based composite materials, numerous researches were carried out on matrix resin molecular structure modification [7,8], addition of ceramic particles [9][10][11] and so on.Adding ceramic particles to resin for blending modification has become one of the current research hotspots whose advantages were simple process, low cost and obvious modification effect.In recent years, many kinds of ceramic particles such as silica [12,13], carbon nanotube [14][15][16], graphene oxide [17,18], silicon carbide [19], zirconium carbide [20], boron nitride [21,22], molybdenum disilicide [23], zirconium diboride [24,25], and zirconium silicide [26,27] etc, have been employed to improve anti-ablation performance of resin-based thermal protection materials.Some of these ceramic particles, for example zirconium diboride, zirconium silicide, boron nitride and molybdenum disilicide etc. can react to form new phases in the oxyacetylene flame environment, which is better in improving the ablation performance.
The effects of zirconium diboride (ZrB 2 ) particle added into the carbon-phenolic composite, when exposed to the ablation test, were explored by Chen et al [24].ZrB 2 modified carbon-phenolic composite exhibited higher ablation resistance with the formation of zirconia and boron oxide during the test of oxyacetylene ablation.Ding J et al studied the effect of zirconium silicide (ZrSi 2 ) particles on the ablation properties of carbonphenolic composite [26].ZrSi 2 reacts with the oxygen-containing molecules to form SiO 2 -ZrO 2 layer and molten SiO 2 covering on the ablated surface, thus the ablation resistance is significantly enhanced.The average linear and mass ablation rates of carbon-phenolic composites modified by ZrSi 2 reduce by about 63% and 42% respectively, compared to those of unmodified carbon-phenolic composites.The boron nitride was added into phenolic-carbon fiber composites by Daniel et al [21], and composites with 5 wt% BN loadings showed maximum improvement of ablation performance.Performance enhancement resulted from the oxidation of boron nitride and phase transformation.Amirsardari et al investigated the effects of GO and ZrB 2 on thermal stabilities of C/Ph composites [28].The presence of ZrB 2 with formation of ZrO 2 in C/Ph composites could markedly decrease the ablation rate, and protect the underlying unoxidized material from the structural damage caused by ablation.It can be concluded that the addition of ceramic particles can significantly enhance the ablation resistance of resin-based thermal protection composites, especially reactive ceramic particles.
At present, a type of ceramic particle is mainly used in the modification of composite materials.It will be an effective and promising way to improve ablation resistance by the synergistic modification with two different ceramic particles.Boron oxide (B 2 O 3 ) with low melting point (about 450 °C) and good wettability can improve the interfacial compatibility between the char and ceramic phases, and form molten liquid protective layer for the char during high-temperature ablation [29,30].ZrSi 2 can react with the oxygen-containing molecules to form SiO 2 and ZrO 2 .SiO 2 and ZrO 2 with high melting point (about 1723 °C and 2700 °C, respectively) resist the erosion of combustion-gas flow.All these results indicate that B 2 O 3 and ZrSi 2 are ideal materials to be used for composite synergistic modification, and the anti-ablation property of epoxy modified organosilicon resin is likely to be remarkably improved by the combined addition of B 2 O 3 and ZrSi 2 .
To the best of our knowledge, few works have been conducted on the synergistic modification.Up until this point, the effects of B 2 O 3 and ZrSi 2 synergistic modification on the ablative property of epoxy modified organosilicon resin composites are unclear.On this basis, different amounts of B 2 O 3 and ZrSi 2 were introduced to synergistic modify the epoxy modified organosilicon resin.The mechanical, thermal and ablative properties of composites were analyzed in detailed.

Fabrication
The formulation of composites is shown in table 1 [27].According to the weight ratio of ZrSi 2 and B 2 O 3 , the ceramic powder was weighed, then mixed with the epoxy modified organosilicon resin and polyamide by mechanical stirring for 15 min under vacuum.Then, the evenly mixed slurry was poured into the PTFE mold, and cured at room temperature for 7 days.Finally, the ablative samples (Ø30 × 10 mm) and the tensile standard samples were obtained.

Characterizations
The tensile properties was tested according to ISO 527, and the loading speed was 2 mm/min.Thermogravimetric tests were performed by a TG analyzer (Q600, TA Instrument) in nitrogen at heating rates (10 °C/min).Each analysis was carried out from room temperature to 800 °C.The micromorphology of composites after ablation were observed by scanning electron microscopy (SEM, HITACHI S4800, Japan) equipped with an energy-dispersive spectroscope (EDS).The phase analysis of samples was carried out by x-ray Diffractometer (XRD, D8 ADVANCE, Bruker, Germany).
The ablation resistance of the composites was carried out using oxyacetylene torch.The test samples were subjected to the torch for a duration of 10 s.The gas pressures and flow rates of acetylene and oxygen were 0.095 MPa and 0.4 MPa, 1116 l h −1 and 1512 l h −1 , respectively.The oxyacetylene torch nozzle was 2 mm in diameter.The distance from torch to the surface of sample was 10 mm.The heat flux was 4.2 MW m −2 .The linear ablation rate was calculated by the following formula: where Δl is the change in the thickness of the sample before and after ablation; t was the time of ablation.The static muffle furnace ablation test was performed.The samples were ablated in air condition at 800 °C for 5 min.The static muffle furnace mass ablation rate Ka was calculated by the following formula: where Δm is the change in the mass of the sample before and after static muffle furnace ablation; m 0 is the mass of samples before static ablation.The ablated micro-structure is closely related to the anti-ablation property of the composite.Therefore, analyzing the effect of oxyacetylene flame on the micro-structure of the composite is of great significance.Figure 8 shows the micro-morphologies of samples after ablation.As shown in figure 8, there are the different ablative surface in Samples 1, 2, 3, and 4 as a result of adding to the B 2 O 3 and ZrSi 2 .Some small holes were formed in the ablated surface of Sample 2. The porous structure appears in the surface of Sample 3 after ablation.Compared with Sample 3, the surface structure of Sample 4 is denser than that of Sample 3.

Results and discussion
In order to deeply understand the influence of B 2 O 3 and ZrSi 2 powder on the ablation resistance of composites, the detailed structures of the ablated composites were evaluated.Figure 9 shows the surface microstructure of the composites after ablation.As shown in figures 9(a) and (b), the loose and feathery structure was formed on the surface of Sample 2 during ablation, which is easy to be washed away by the high speed oxyacetylene flame.Therefore, the ablation resistance of the Sample 2 is poor and the linear ablation rate is high as shown in  In order to understand the phase component of ablated surfaces of samples, the XRD pattern was carried out.Figure 10 illustrates the XRD patterns of Samples 2, 3 and 4 after ablation.As shown in figure 10, it is observed that only ZrSi 2 peaks appear for Sample 2. Besides, except for ZrSi 2 peaks, ZrO 2 , SiO 2 and ZrB 2 peaks also appear in the surface of Sample 3 and 4.This is because the ZrSi 2 and B 2 O 3 could take part in thermochemical reactions as shown in the following equations (3) and (4).
After a thermodynamics calculation, the Gibbs free energy changes for reactions (3) and (4) at 2000 K are −1872.7 kJ mol −1 and −463.92kJ mol −1 , respectively.It can be informed that ZrSi 2 could react with oxygen and B 2 O 3 to form ZrB 2 , ZrO 2 , and SiO 2 during oxy-acetylene ablation, which is consistent with [26].The chemical stability of ZrB 2 is better than that of ZrSi 2 .During the ablation process, the formation of ZrB 2 is helpful in improving the ablation resistance of the epoxy modified organosilicon resin.It can be calculated from the formulations of samples in table 1 that the volume fraction of ZrSi 2 for Samples 3 and 4 is 21.40 vol% and 13.33 vol%, respectively, and the volume fraction of B 2 O 3 is 14.15 vol% and 26.44 vol%, respectively.The ratio of B 2 O 3 volume fraction of Samples 3 and 4 to the total volume fraction of B 2 O 3 and ZrSi 2 is 0.398 and 0.665, respectively.As can be seen from figure 3, the residual weight of Sample 1 in aerobic environment is 22.2 wt%.Therefore, in the process of oxyacetylene ablation, the resin pyrolysis residue on the sample surface is less.If the resin pyrolysis residue is not taken into account, it can be seen from the comparison of volume fractions that the surface of Sample 3 is dominated by ZrSi 2 .ZrSi 2 can react to form ZrO 2 and SiO 2 during oxy-acetylene ablation.ZrSi 2 reacts completely with oxygen to form the corresponding oxide, which becomes about 2.5 times its original volume.Therefore, the volume fraction of B 2 O 3 on the surface of Sample 3 is lower in the ablation process, and the surface of Sample 3 is dominated by ZrO 2 , SiO 2 and ZrB 2 .ZrO 2 , SiO 2 and ZrB 2 have good chemical stability and high melting point, which help to improve the ablative resistance of the composites.The small amount of B 2 O 3 with low melting point and good wettability forming molten liquid phase, plays a certain role in structural filling and binding action.Therefore, the skeleton-like compact structure formed in the surface of Sample 3, which improves the structural strength of the ablative surface.Therefore, the ablative resistance of Sample 3 is the best by the synergistic modification of appropriate amount of B 2 O 3 and ZrSi 2 .For Sample 2, there is no B 2 O 3 , which cannot perform the roles of structural filling and binding action.Moreover, the volume fraction of resin is high and the pyrolysis shrinkage is large, so its ablative resistance is low.The volume fraction of B 2 O 3 in Sample 4 is nearly 2 times that of ZrSi 2 .Because of excess B 2 O 3 , too much liquid phase formed in the surface of Sample 4, as shown in figure 9(e).Therefore, its ablative resistance is worse than that of Sample 3.

Conclusions
Epoxy modified organosilicon resin composites synergistic modified with boron oxide and zirconium silicide were successfully prepared.The effects of B 2 O 3 and ZrSi 2 on the mechanical properties, thermal stability and ablation properties of composites were investigated in this work.The addition of B 2 O 3 and ZrSi 2 leads to the significant improvement of hardness, strength, thermal residue yield and ablation performance of composites.During oxy-acetylene ablation, ZrSi 2 could react with oxygen and B 2 O 3 to form solid phases with high melting point, and B 2 O 3 forming molten liquid phase plays a certain role in structural filling and binding action.When the contents of resin, boron oxide and zirconium silicide are 100 g, 50 g, and 150 g, respectively, the skeleton-like compact structure formed in the surface of composites by the synergistic modification of appropriate amount of B 2 O 3 and ZrSi 2 after oxyacetylene ablation, which improves the structural strength of the ablative surface and results in the significant improvement of ablation resistance.

3. 1 .
Thermal and mechanical performanceThe density, shore hardness and tensile strength of the samples are presented in table 2. With the addition of B 2 O 3 and ZrSi 2 , the density of the composites increases obviously.Due to the low density of B 2 O 3 , the density of the Samples 3 and 4 decreases with increasing content of B 2 O 3 .The shore hardness and tensile strength of the samples also improve markedly because of the particle reinforcement of B 2 O 3 and ZrSi 2 .Figure1shows the tensile curves of the samples, and the complete tensile curve of Sample 1 is displayed in top right of figure 1.The tensile elongation of Sample 1 is too large and exceeds 50%.It could be observed that the tensile elongation of the samples modified by B 2 O 3 and ZrSi 2 reduces sharply.Sample 3 achieves the maximum tensile strength of 4.37 MPa and the elongation of Sample 3 is larger than that of Sample 4.According to the formulation of samples, the resin volume fraction of Sample 2 is 69.23 vol%.As the density of B 2 O 3 is significantly lower than that of ZrSi 2 , the resin volume fraction of Sample 3 and Sample 4 gradually decreases with the increase of B 2 O 3 .The resin volume fraction of Sample 3 and Sample 4 is 64.45 vol% and 60.23 vol%, respectively.As the volume fraction of the resin reduced, the integrity and continuity of matrix resin were obviously reduced.As a result, the elongation of the Samples 3 and 4 decreased significantly.Due to particle strengthening by the addition of ZrSi 2 and B 2 O 3 , the Samples 3 and 4 have the excellent tensile strength.The thermal degradation behaviors of the samples in a nitrogen atmosphere were investigated by TG analysis and the results are shown in figure2.As shown in figure2, the residue yields at 800 °C are 27.11wt%, 83.09 wt%, 82.63 wt% and 83.11 wt% for Samples 1, 2, 3 and 4, respectively.There is little difference in mass loss rate of

Figure 1 .
Figure 1.Tensile curves of the samples.

Figure 2 .
Figure 2. TG curves of the samples in N 2 atmosphere.

Figure 3 .
Figure 3. Static muffle furnace mass ablation rate of the samples.

Figure 6 .
Figure 6.Linear ablation rate of the samples.

Table 1 .
Formulation of the samples.

Table 2 .
[26]ity, Shore hardness and tensile properties of the samples.According to residue yield of sample 1 and the mass loss of B 2 O 3 and ZrSi 2 was not considered, the residue yields of Samples 2, 3 and 4 is 74.13 wt%.The residue yield of composites in the nitrogen atmosphere has dramatically improved with the addition of B 2 O 3 and ZrSi 2 .3.2.Ablation propertiesThe static muffle furnace mass ablation rate of the samples are displayed in figure3.The static mass ablation rate of composites is sharply reduced by the addition of B 2 O 3 and ZrSi 2 .The mass ablation rate of Sample 3 is only 13.64 wt%, which is lower than that of Samples 1, 2, and 4. The residuals of Sample 1 is very little after static muffle furnace ablation.The micro-structure of Samples 2, 3 and 4 is displayed in figure4after static muffle furnace ablation.Due to the thermal decomposition of the matrix resin, porous structures were formed.As illustrated in figure4(a), the residuals of Sample 2 is porous structure and the aperture wall is comparatively thin.It is observed that large area of big size holes is in figure 4(c) for Sample 4. The volume ratio of the hole is high, therefore the static mass ablation rate of Samples 2 and 4 is larger than that of Sample 3. The micro-structure of Sample 3 is denser than that of others.Consequently, Sample 3 has the lowest static muffle furnace mass ablation rate.The XRD patterns of composites after static muffle furnace mass ablation test are shown in figure5.There is no evident peak for Sample 1.It is observed that only ZrSi 2 peaks appear for Samples 2, 3 and 4.There is no peaks of B 2 O 3 , because an amorphous state B 2 O 3 is very difficult to form crystals.It can be inferred from the XRD patterns that the peaks of ZrSi 2 after the static muffle furnace ablation are no obvious difference.The oxyancetylene linear ablation rate of the samples are shown in figure6.Compared to Sample 1, the linear ablation rate of Samples 2, 3 and 4 obviously decreased after the addition of B 2 O 3 and ZrSi 2 .The linear ablation rate of Sample 3 was lowest and reduced by 91.67% compared with that of Sample 1.The linear ablation rate of Sample 3 is close to that of the classical carbon fiber reinforced phenolic composites (0.046 mm s −1[26]