TITANIUM DIBORIDE IN-SITU COMPOSITES

Get Complete Project Material File(s) Now! »

Results and Discussion

This chapter presents the results of the predictions and experiments and explains the observed results.

Thermodynamic-based Model / Predictions

Theoretical calculations were performed to predict the behavior of the Ti-xAl + yV% TiB₂ composite system with varying aluminide and reinforcement content. Each of the predictions in this section was calculated assuming an ignition temperature of 1000K.
Figure 4 shows the enthalpy diagram for Ti-34Al + 40V% TiB₂. Shown on the diagram are T_Ad and ∆H_f for this composition.
In Figure 5 the results of the calculations for how T_Ad (assuming T_ig = 1000K) varies across aluminide stoichiometry are shown for composites with 0, 10, 20, 30, 40 and 100V% TiB₂. T_Ad increases with increasing amounts of TiB₂. In the case where there is 0V% TiB₂, the adiabatic temperature is that of just the Ti-xAl. The reason for the plateaus in that curve is that the melting temperature of TiAl is greater that those of Ti₃Al and TiAl₃.
Figure 6 shows the calculated results for ∆H_f across the aluminum content range for 0, 20, 40, 60, 80, and 100 V% TiB₂ in Ti-xAl. With the increasing amounts of TiB₂ ∆H_f becomes more exothermic, approaching the behavior of pure TiB₂ at high reinforcement percentages.
Figure 7 presents the predictions of T_Ad across the range of reinforcement percentage for matrix compositions of Ti-25Al (Ti₃Al), Ti-51Al (TiAl), and Ti-75Al (TiAl₃). For all three matrix compositions increasing amounts of TiB₂ increases T_Ad, the plateaus in the curves are associated with the thermal arrest that occurs during the melting of a constituent. Though the same thermodynamic data was used to calculate the Ti-25Al and Ti-75Al curves, they are distinctly different because the atomic weights of Ti₃Al and TiAl₃ are different and the adiabatic temperature calculation is inherently dependent on atomic weight.

 Formulations

The reaction behavior of varying formulations was measured to understand the effect of the changing formulation and its relation to the theoretical. The average particle sizes of the elemental powder constituents are presented in Table 7. For the discussion of the formulations, the average particle size will be used to identify the powder used in the blend.

Monolithic Formulations

In this section the results from the DSC experiments on monolithic formulations are presented and discussed. Section 3.2.1.1 presents the results of the elemental constituents and Section 3.2.1.2 presents the results of the TiAl and TiB₂ experiments.
 

Elemental Constituents

To be confident in the analysis of the composite formulations it is important that the behavior of each of the elemental constituents is observed so that when are all combined together for reaction, the behavior can be attributed to the constituents or to the reaction. To achieve this goal, the titanium, aluminum and boron powders were each subjected to the temperature profile used in the reaction experiments in an air environment. Figure 8, Figure 9, and Figure 10 show the DSC and TG curves for titanium, aluminum, and boron respectively. The titanium curve exhibits an exotherm around 760˚C, the aluminum curve exhibits an endotherm around 670˚C and a small exotherm around 1000˚C, and the boron curve exhibits an exotherm around 600˚C. The increasing TG corresponding with the exotherm in each case is likely due to the oxidation of the material.

Composite Component Blends

Similarly to the characterization of the elemental constituents, blends were formulated to see if TiAl and TiB₂ would form independently under the standard heating profile. Figure 11 shows the DSC and TG curves for the TiAl experiment formulated with 128µm Al and 46µm Ti. Figure 12 shows the DSC and TG curves for the TiB₂ experiment formulated with 46µm Ti and 0.35µm B.
In Figure 11 we can distinctly see the aluminum melting endotherm followed by a two-humped exotherm. In the corresponding TG curve it can be seen that the material increased in weight following the melting of the aluminum. The endotherm and the latter exotherm appear to be from the aluminum, while the exotherm following the aluminum melting endotherm appears to be in the same location as the exotherm from the elemental titanium. Therefore, by comparison of this curve to the curves for the elemental aluminum and titanium we see that this curve appears to be the combination of the behaviors of titanium and aluminum as opposed to the formation of titanium aluminide. This was expected as T_Ad for TiAl is 1733K which, by the Merzhanov criterion [7], is not expected to react spontaneously.
As for the TiB₂ experimental results shown in Figure 12, when compared to the elemental titanium and boron curves we can see that the curve appears very similar to that of the elemental amorphous boron curve, while no trace of the titanium behavior is noticed. It is initially unexpected that this reaction did not appear to occur because the TiB₂ reaction is known to be extremely violent [4]; however it is likely that this reaction did not initiate due to its size, as it has been shown that small diameter specimens of TiB₂ precursors do not initiate due to radial heat losses [6].

Composite Blends

Ignition Experiments

Figure 3 is a DSC curve for Ti-50Al + 40V% TiB₂ from which several characteristic values for this reaction system can be extracted. It is important to understand which of these values can and should be compared to the theoretical values of T_ig, ∆H_f, and T_Ad. Due to the nature of the DSC measurements, the maximum temperature of reaction cannot be characterized in this instrument. Therefore the two-color infrared pyrometer was used as described previously. That leaves T_ig and ∆H_f to be characterized using the DSC. There are three distinct features in Figure 13, those being an endotherm and two exotherms. In Figure 14 the curves for elemental titanium, aluminum, and boron are overlaid to aid in the comparison of the composite blend to its constituents. It should be noted that the DSC units are in mW/mg and that each of these specimens were approximately the same size, therefore the magnitude of the elemental constituents is inflated when compared to the composite blend. As such, only the location and relative magnitude of these peaks should be considered in the comparison.
The endotherm is the aluminum melting endotherm, the peak of it being taken as the melting temperature of aluminum (T_{m, Al}). It should be noted that the observed peak in the composite blend is lower than that of the elemental aluminum. This can be accounted for by the fact that the endotherm is within a heat producing exotherm, and due to a lag in response time of the thermocouple within the DSC from heat transfer the temperature at which the aluminum melts appears lower than it actually is.
The first, smaller exotherm could be an initial reaction; however when it is compared to the DSC curves from the three elemental constituents in Figure 14 the location of the exotherm matches that of the amorphous boron curve. Therefore, the specimen still appears to be a mixture of elemental titanium, aluminum and boron through the first exotherm.
The second exotherm appears larger than the first exotherm, and this is expected as the aTi + bAl + cTi + dB → zTiAl + yTiB₂ reaction is known to be exothermic, therefore the area of that exotherm is taken to be the heat of reaction. To calculate the area of the exotherm the complex peak fit on the NETZSCH Proteus Thermal Analysis (Version 4.8.1) software was used. A procedure of maximizing the area term was used to ensure uniformity in the calculations. The resulting area value was taken to be ∆H_rxn. The value was calculated by the software in units of J/g which were converted to units of kJ/mol by multiplying the value by the calculated theoretical atomic weight of the reacted composite and dividing by 1000.
Finally, the issue of ignition temperature remains. In the literature it has been assumed that ignition in the Ti-Al-B system occurs when the aluminum melts [11]. That does not appear to be the case presented in Figure 14. To determine if ignition occurs due to the aluminum melting, an experiment was designed to run under the same conditions as before, but with an isothermal hold at 750˚C for 20 minutes to see if the second reaction exotherm occurred. If the second exotherm appears then aluminum melting may be considered the ignition of the reaction. Figure 15 shows the results of this experiment. The secondary exotherm is not observed in this DSC curve. Another experiment was performed using the same profile as before with the addition of a 40 K/min ramp to 1150˚C following the 20 minute hold at 750˚C. The DSC curve for this experiment is shown in Figure 16. The second exotherm is observed in this case, and as such indicates that the reaction is not initiated by the melting of aluminum, but rather requires more thermal energy to initiate. For the case of this report, the onset temperature calculated in the complex peak fit of the reaction exotherm is considered to be T_ig. It should be noted that similar to the suppression of the aluminum melting temperature this value may also be suppressed by the thermal lag within the instrument and as such it could be a source of error.

Dedication.
Acknowledgements 
0 Objectives
1 Background
1.1 Reaction Synthesis
2 Experimental Procedure
2.1 Selection of System
2.2 Thermodynamic-based Model / Predictions
2.3 Formulations
2.4 Reaction Measurements
2.5 Extrinsic Reactant Variables
3 Results and Discussion
3.1 Thermodynamic-based Model / Predictions
3.2 Formulations
3.3 Extrinsic Reactant Variables
4 Conclusions 
5 Future Work
6 Appendix A: Thermochemical Data
7 Appendix B: Matlab Program Code
8 References
GET THE COMPLETE PROJECT
REACTION SYNTHESIS OF TITANIUM ALUMINIDE / TITANIUM DIBORIDE IN-SITU COMPOSITES