3.1 The accumulation of stabilized, stress induced martensite
In order to determine the amount of induced martensite, mechanical cycling was performed on Fe43,5Mn34Al15Ni7,5 samples and those in which + -1.5% of Al were replaced with Ni, in the hot rolled state. The results are centralized in Fig.3.1.
It can be seen that the Fe43,5Mn34Al15Ni7,5 sample, from Fig.3.1 (b), has undergone a maximum force 75% higher than the other two samples and the loop closed much faster, suffering permanent elongation increases of only 0.13% in the third cycle and 0.11% in the fourth. In all three samples, the permanent elongation increased more and more slowly, as the number of cycles increased, so that a closed loop was obtained after the fifth cycle.
Fig.3.1 Tensile loading-unloading curves recorded during mechanical cycling of hot-rolled samples with different chemical compositions: (a) Fe43,5Mn34Al16,5Ni6 (b) Fe43,5Mn34Al15Ni7,5 și (c) Fe43,5Mn34Al13,5Ni9
Fig.3.4 DMA thermograms presenting the variation of storage modulus, E’ and internal friction, tanδ, depending on the relative bending amplitude, during cycles 2-5, of amplitude sweep, applied to the hot rolled samples, Fe43,5Mn34Al15Ni7,5, at different temperatures: (a) room temperature; (b) 180 C with martensitic antiferromagnetic structure, (c) 260 C with paramagnetic martensitic structure and (d) 300 C with austenitic structure
3.2 Determination of reversible mechanical behavior by amplitude sweep
The results of the amplitude sweep, for five cycles, performed at the frequency of 1 Hz, with a proportionality factor of 1.3 and the amplitude increase with a ratio 1 μm, between 1 - 14 μm, are centralized in Fig. 3.4.
In all cases, the monotonous and repeated increase of the storage modulus, E', can be observed, with the relative amplitude of the bending deformation. Being a dynamic behavior, this increase of E' can be associated with hardening (Pricop, 2018).
Comparing the variations of E' with the relative amplitude, at the four temperatures, it is observed that the sample with paramagnetic structure, due to the disorder of the magnetic fields, has the most unstable behavior, because the variations of the module and of the internal friction, of the four cycles, do not overlap, as in the other cases.
3.4 Reducing the thickness of the samples (t)
To reduce the thickness of the samples, the same hot rolling procedure was applied at a temperature set at 1060 C. It was found that the alloying with Si reduced the plasticity of the samples, as illustrated in Fig. 3.8.
It can be observed that the cracking tendency, observed during electro-erosion cutting of the Fe43,5Mn34Al11Ni7,5Si4 sample containing the highest amount of silicon, was also maintained during hot rolling, as illustrated in Fig. 3.8 (c).
Fig. 3.8 Hot rolled samples: (a) Fe43,5Mn34Al14Ni7,5Si1; (b) Fe43.5Mn34A13Ni7.5Si2; (c) Fe43,5Mn34Al11Ni7,5Si4 and (d) Fe43,5Mn34Al15Ni7,5
Fig.3.10 Thermogram of the cyclic heat treatment, for the abnormal growth of the crystalline grains
3.6 Application of heat treatment for abnormal grains growth
The essential part of the heat treatment for the abnormal growth of crystalline grains is the cyclic thermal treatment (CHT) illustrated schematically in Fig. 3.10. The hurnace, with a programmable controller, easily maintains the speed of 10 C / min during heating but has some cooling problems due to the weak argon flow. The essential condition for maintaining the integrity of the samples, following CHT, is the encapsulation in argon atmosphere under partial vacuum. In this respect, as shown in the last stage, a procedure for emptying and closing the quartz tubes with oxyacetylene flame was implemented within the present project.
3.9 Illustration of traction behavior through loading-unloading cycles
Three alloy samples with the classical chemical composition, Fe43,5Mn34A15Ni7,5, in hot rolled state, were subjected to tensile loading-unloading cycle, up to a maximum elongation of 3%, at three temperatures. Representative curves are shown in Fig. 3.14.
Fig. 3.14 Tensile load-unloading curves of hot rolled samples of Fe43,5Mn34Al18Ni7,5
Fig. 3.15 Results of tensile cycling at room temperature, of Fe43,5Mn34Al18Ni7,5 hot rolled samples
3.10 The accumulation of stabilized stress induced martensite
The first results obtained by mechanical cycling at room temperature are shown in Fig. 3.15. The first three cycles were applied up to 3% and the next 7 to 4.7%. As in Fig.3.1, there is the gradual closure of the stress-elongation curve, which turns in a loop but retains hysteresis.
The tensile loading-unloading curves recorded at high temperatures had a very different configuration.
3.11 Determination of mechanical behavior by amplitude sweep
The DMA thermogram obtained by temperature scanning corresponding to the FMAN-2Si sample is shown in Fig. 3.22. The following particularities can be observed, which can be considered as effects of the 2% substitution of Al with Si:
• the values of the storage module are higher than in FMAN-1Si but lower than in FMAN and the variation of the heating module follows the same steps observed in FMAN and FMAN-1Si but the maximum increase is approx. 14 GPa, reaching a maximum of 135 GPa in the austenitic phase;
• the internal friction has two clear maxima, like the FMAN sample and reaches, by the second maximum, the value of approx. 0.053
Fig.3.22 DMA thermograms of the variation of elastic storage modulus, E ‘ , and internal friction, tanδ, on heating with 3K / min, for the hot rolled sample, of Fe43,5Mn34Al13Ni7,5Si2, required for three point bending, with 1 Hz and amplitude of 100 μm. The arrows mark the temperatures at which the amplitude sweep was performed
Fig.3.27 DMA thermograms of the variation of elasticit storage modulus, E ‘, and of the internal friction, tanδ, on heating with 3K / min, for the sample Fe43,5Mn34Al15Ni7,5 subjected to cyclic thermal treatment (CHT) during bending in three points, with 1 Hz and the amplitude of 100 μm.
3.12 Determination of critical temperatures by DSC or DMA
Because the calorimetric effect is very low in FeMn-based samples, the most accessible method of thermal analysis applicable for determining critical temperatures is the mechanical-dynamic analysis (DMA) applied in the temperature scanning variant.
A particularity of the FMAN sample in the CHT state is the appearance of two intermediate maxima of the storage module, denoted I and II, as well as the widening of the first maximum of internal friction, marked with a wide blue arrow in Fig.3.27.
By the tangent method, the critical transformation temperatures can be determined based on the two variations, of the module and the internal friction.
3.14 Identification of texture variations induced by accumulation of irreversible deformation, through SEM-EBSD
This activity also included the purchase of a electron back-scattering diffraction (EBSD) detector. After determining the crystal structure of our samples, the images of poles for the main phase, γ-cfc, were recorded, according to Fig. 3.33.
Fig.3.33 Pole images of the γ-cfc phase in FeMnNi, space group Fm-3m, according to the families of planes {100}, {111} and {110}, with the representation of the orientation mode and with a standard stereographic projection (001 ) for the cubic system (Cullity, 1956)
Fig.3.34 Tensile curves recorded using the DEBEN MicroTest 2000 microtension device from CEMS – UPB endowment using samples of Fe43,5Mn34Al15Ni7,5, Fe43,5Mn34Al14Ni7,5Si1 and Fe43,5Mn34Al13Ni7,5Si2, in the (a) CHT state and (b) AT state.
3.15 Identification of voltage-induced martensite α'
Special tests samples were prepared from the 6 samples, (FMAN-CHT, FMAN-1Si-CHT, FMAN-2Si-CHT, FMAN-AT, FMAN-1Si-AT and FMAN-2Si-AT) which were extended using the DEBEN MicroTest 2000 microtension device from CEMS - UPB. The tensile curves are shown in Fig. 3.34.
It is noteworthy that in both cases the samples deformed linearly, due to the low deformation value, 0.45%. The linear approximation, of the form σ = bε + a, (where σ - the stress and ε - the specific elongation) gave errors below 5%, in all cases. In addition, the slopes of variation of these straight lines fall between arctg (473,1225) ≈ 89,880 and arctg (683,8252) ≈ 89,910, so the increase of tension with deformation is very sharp, which denotes the high rigidity of the samples.
3.16 Identification of coherent precipitates
A first set of experiments aimed at highlighting the superposition of Al and Ni atoms, considering that with aging they formed AlNi precipitates (Omori, 2011). For this purpose, the EDS detector was used, the results being centralized in Fig. 3.37, for the same FMAN-AT sample.
In this mappings, we can see certain portions where the atoms of Al and Ni overlap, such as the examples marked by circles in Fig. 3.37 (a) - (c), corresponding to the CHT state of the samples. In the AT state, where the number of NiAl precipitates should be higher, delimiting the two atomic species becomes very difficult, even at magnification powers over 23000 times, as in Fig. 3.37 (d).
Fig. 3.37 SEM-EDS mapping of Al and Ni to the samples: (a) FMAN-CHT; (b) FMAN-1Si-CHT; (c) FMAN-2Si-CHT; (d) FMAN-AT; (e) FMAN-1Si-AT and (f) FMAN-2Si-AT