Room temperature control of grain orientation via directionally modulated current pulses

Traditional approaches to control the microstructure of materials, such as annealing, require high temperature treatment for long periods of time. In this study, we present a room temperature microstructure manipulation method by using the mechanical momentum of electrical current pulses. In particular, a short burst of high-density current pulses with low duty cycle is applied to an annealed FeCrAl alloy, and the corresponding response of microstructure is captured by using Electron Backscattered Diffraction (EBSD) analysis. We show evidence of controllable changes in grain orientation at specimen temperature around 28 °C. To demonstrate such microstructural control, we apply the current pulses in two perpendicular directions and observe the corresponding grain rotation. Up to 18° of grain rotation was observed, which could be reversed by varying the electropulsing direction. Detailed analysis at the grain level reveals that electropulsing in a specific direction induces clockwise rotation from their pristine state, while subsequent cross-perpendicular electropulsing results in an anticlockwise rotation. In addition, our proposed room temperature processing yields notable grain refinement, while the average misorientation and density of low-angle grain boundaries (LAGBs) remain unaltered. The findings of this study highlight the potentials of ‘convective diffusion’ in electrical current based materials processing science towards microstructural control at room temperature.


Introduction
Controlling the microstructure to tailor the properties of materials has always been a highly desired capability in materials science and engineering.The customization of materials properties caters to specific performance needs and has led to innovative advancements in various industries [1,2].Typically, microstructure control means manipulating grain size, crystal orientation, low-angle grain boundaries (LAGBs) density, etc, which can enhance mechanical and electrical properties such as strength, toughness, and electrical conductivity [3,4].Traditional microstructure modification methods, like annealing, involve the materials to be subjected to controlled heating and cooling cycles in specialized equipment, leading to grain redistribution, increased toughness, and residual stress relief [5,6].However, the process requires long periods of time and often results in limited control due to non-uniform changes in microstructure across the material [7,8].The other heat treatment methods, such as quenching, tempering, and precipitation hardening, lead to the formation of specific microstructures and enhance hardness and strength [9,10].These thermal treatments often induce additional thermal stress into the samples and could also lead to severe oxidation under nonvacuum conditions [11,12].Other than thermal processing, severe plastic deformation (SPD) technique, spark plasma sintering (SPS), Laser Engraving, and surface mechanical treatment are some microstructure modification methods, each tailored to specific needs and applications [13][14][15][16].For example, Li et al [13] employed the SPD technique to obtain refined grains for NiTi-based alloys.
Thermal processes involve random diffusion of defects in materials-a reason why energy minimization process takes a long time.In this study, we ask a fundamental question: what happens if the diffusion process is directional, analogous to convection?There is little or no understanding to this question in the literature, probably because conventional materials processing has always involved high temperature.In this study, we seek the answer to this question by studying materials response to directionally applied electrical current pulses.This is known as electropulsing, a technique shown to change microstructure of materials depending on pulse parameters, such as current density, frequency, and pulse width [17,18].We investigate the 'room temperature' electropulsing condition and its effects on metallic alloys by tuning the parameters.It is important to note the difference in electropulsing conditions typically used in literature, where appreciable temperature rise is part of the process [19][20][21][22].Researchers have shown that properly tuning electropulsing parameters can increase dislocation mobility, accelerate vacancy movements, and reform the microstructure of materials by refining the grain size and structure [23][24][25].Wu et al [23] have shown that the electric pulse can improve the plasticity in cold-rolled pure copper sheets by inducing rapid recrystallization.Song et al [25] explored how the application of pulse current during rolling can enhance the copper sheet's resistance to wear and oxidation.However, these processing conditions raise the temperature up to 250 °C −700 °C, making it difficult to separate the electrical and thermal effects.Therefore, the fundamental cause-effect analysis on electropulsing induced microstructural processing remains to be topic of debate.
We developed a room-temperature processing technique by decoupling the thermal effect from the traditional electropulsing process.During electropulsing, the metallic crystals become exposed to two forces, namely electrostatic and electron wind forces [26].The former can be neglected in most cases as it nullifies itself due to the opposite polarity of positive ions surrounded by negative electrons.The electron wind force, generated by the momentum transfer between the electron and lattice, acts in the direction of current flow [27].In electromigration literature, such directionality creates a convective diffusion mode of atomic mass transfer.The high current density also generates Joule heating, which facilitates void growth and creates a positive feedback loop to accelerate the thermal force [26][27][28].Xu et al [21] reported that the high-density current could increase the material temperature to 600 °C, and the associated thermal force can essentially overcome the activation energy for the grain-boundary diffusion, resulting in grain growth.In most electromigration and electropulsing studies, where the current density is in the order of ( ) -10 10 , this thermal force is assumed to be the main driving force [29,30].We decouple this thermal force by potentially applying very short-pulse width and low-frequency current with a density of ( ) 10 , and use the electron wind force as the primary driving stimulus.Since the temperature rise in sample during electropulsing is proportional to the square of current density [19], we limit the temperature rising by incorporating 10-100 times lower current density than usual electropulsing treatment.In this study, we monitor the sample temperature in real time to ensure that the specimen temperature remains within 30 °C.Our previous studies show promising results in this respect [31][32][33].Using the similar premise, Qin et al [34] have observed crystal rotation towards a certain direction in a duplex steel at room temperature.In this study, we investigate a fundamental question: can we reverse this rotation towards its initial condition to gain control over grain rotation?As a showcase application, we focus on grain orientation control in FeCrAl alloy, a next generation cladding material for nuclear fuels operating at high temperatures [35].The objective of this investigation was to find the effects of electropulsing direction on the crystallographic orientation of the material.By carefully monitoring the temperature and passing electrical current in pre-determined directions, we explore the effects of room-temperature electropulsing on grain orientation dynamics.The proposed room temperature electropulsing offers a stress-free grain modulation technique.The average misorientation and low angle grain boundaries (LAGBs) density is observed to remain unchanged after processing.Notably, we investigate the influence of pulse direction on the resulting grain orientation changes.By selecting perpendicular pulse directions, we demonstrate our ability to control the grain rotation in either direction-clockwise or anticlockwise.Due to the microstructural change induced by the grain rotation, the mechanical properties of material will be affected which we plan to explore in future.The grain rotation was captured from the Euler angles and pole figures obtained from the Electron Backscattered Diffraction (EBSD) analysis [36].While EBSD provides surface microstructures, core morphology analysis requires cross-sectional study, a scope beyond this work and a focus of future publications.However, we believe that the applied uniform pulsed current ensures consistent effectiveness from core to surface.

Experimental details
FeCrAl alloy with a well-confirmed composition (Fe-73%, Cr-21.3%,Al-5.7%) via Energy Dispersive x-ray analysis (figure 1(a)) was employed in this study.The sample was made into a rectangular geometry ( ´2.5 2.3 0.5 mm 3 ), and distinct corners were marked as 1, 2, 3, and 4 for precise identification as shown in figure 1(b).The initial step involved sample preparation to make the surface suitable for Electron Backscatter Diffraction (EBSD) measurements.The sample was prepared in two steps: polishing with a rotary tool and multiple diamond polishing compounds lapping paste (3,000 to 120,000 grit) and subsequent ion milling for 90 min with 4.5kV and 1.5A beam current.At first, a pristine EBSD map of Region of Interest (ROI), shown in figure 1(b), was acquired at a magnification of 2500X.Next, a higher magnification of 10000X was employed to meticulously examine a specific Area of Interest (AOI) containing only a limited number of grains.The EBSD analysis was performed at 3.2nA beam current, 20 kV accelerating voltage, and 0.3 μm step size using a VERIOS Scanning Electron Microscope equipped with the Aztec software suite version 6.2.This comprehensive approach facilitated the detailed characterization of the sample's microstructural features and crystal orientations.Electropulsing was carried out using 40 μs current pulses at a frequency of 2 Hz.At first, current pulses with a current density of 3.36 × 10 3 A cm −2 were applied along the 1-3 axis for 2 min.The associated electropulsing parameters are shown in table 1.During this process, real-time temperature measurements were acquired using an Optris PI-640 thermal microscope.The temperature measurements verified that the sample remained well within the ambient temperature range throughout the pulsing procedure (see figure 1(c)).After electropulsing in 1-3 axis direction, EBSD mapping was repeated for the previously defined ROI and AOI at 2500X and 10,000X magnifications, respectively.Spatial accuracy was ensured by employing a strategic approach where fixed markers were introduced into the sample using a Focused Ion Beam (FIB).This methodology enabled the precise relocation of the analysis area, thereby ensuring consistent comparison across different stages of the experiment.Subsequently, we applied electropulsing along the perpendicular 2-4 axis with a current density of 6.08 × 10 3 A cm −2 .It should be noted that the current densities applied along the 1-3 and 2-4 directions are 100-1000 times lower than the electromigration failure limit of 10 5 -10 6 A cm −2 [37,38].Following this, EBSD mapping was conducted once again to capture the post-electropulsing microstructural features in the specified regions.The postprocessing and comprehensive analysis of the acquired data were undertaken using the specialized Aztec Crystal software suite.This software facilitated the interpretation of crystallographic orientations, grain boundary characteristics, and potential changes in the microstructure induced by the electropulsing process [39].

Results and discussion
Since the main effect of the electron wind force is to mobilize defects, we start the discussion with the types of defects that could dominate the experimental results of this study.Vacancies or dislocations would be difficult to mobilize because the small current density in this study is not able to produce very large wind force.Also, SEM cannot visualize vacancies or vacancy clusters.Grain boundaries are surface defects with largest diffusion coefficients, and therefore expected to experience the highest mobility, manifesting in grain rotation and grain size or shape change.Of these, grain rotation would require minimal external stimulus because grain size change would require more energy to induce atomic mobility and rearrangements in the crystal lattice.We therefore expect grain rotation to be the dominant mechanism behind the lowering of the system energy (or the electrical resistivity [40]).In our study, we observed 16.3% and 34.88% decrease in resistance for 1-3 and 2-4 direction pulsing, respectively.Even though grain size change is energetically unfavorable compared to grain rotation, we observed a considerable amount of grain size decrease due to electropulsing.As we know the grain rotation rate is inversely proportional to the grain size [41], the observed smaller grain sizes further facilitate the grain rotation mechanism [42].Also, the amount of rotation would depend on the crystallographic orientation before the application of the electrical pulses [43].

Microstructural changes in the ROI and AOI
As discussed in the introduction section, electromigration occurs in the direction of current flow [27].Yet, to the best of our knowledge, no prior research has explored the material's behavior when subjected to pulsed current flow in cross directions.Previous studies reveal that grain rotation occurs during electric current treatment [44][45][46].In the present study, we investigate the grain rotation response of a polycrystalline material subjected to electropulsing in cross-perpendicular directions.In the initial phase of our investigation, we present a comprehensive visualization, as depicted in figures  significant change in the grain boundary structure.For further analysis later in this section, we have selected four grains within our AOI, designated as A, B, C, and D. Figure 3 showcases zIPF maps for the AOI, along with {100} pole figures.The IPF maps provide insights into the distribution of crystallographic orientations across the sample [36].In a polycrystalline material, each grain possesses a distinct crystallographic orientation [47], and the IPF reflects the collective orientations by assigning distinct coloring.In this case, the maps are shaded with lattice poles aligned parallel (//) to the z-axis, according to the color-coded key (inserted triangle) presented as part of figure 3. Thus, red colored grains have 〈100〉//z orientation, green grains have 〈110〉//z orientation, blue grains have 〈111〉//z orientation.A careful observation of figures 3(a)-(c) indicates that the IPF coloring of pulsed samples shifts from its pristine map.Each color in the IPF map corresponds to a specific crystallographic orientation, and the transitions between colors signify the shifts in the grain orientation.

Crystallographic orientation and rotation of individual grains
To capture the orientation shift, we have investigated the corresponding pole figures of {100} planes (see figures 3(d)-(f)).The pole figures display the clockwise rotation of poles after 1-3 axial pulsed and their subsequent return to an opposite direction following 2-4 axial pulsed.However, the superposition of orientations from different grains results in a complex pole figure pattern with overlapping features.This makes it challenging to identify specific trends or rotations within the pole figures, especially when trying to isolate the response of individual grains to specific stimuli or treatments.Hence, a meticulous grain-by-grain analysis for four distinct grains (A, B, C, D, see figure 2) is shown in figure 4 to provide a clearer perspective.
Upon careful examination of the {100} pole figures of four individual grains (see figure 4), it is observed that, in response to electropulsing in the 1-3 axial direction, all four individual grains exhibit a remarkable clockwise rotation of their respective poles.Intriguingly, after the application of pulsed current in the cross-perpendicular 2-4 direction, the pole orientations collectively revert to an anticlockwise rotation.This dynamic behavior presents a compelling narrative of crystallographic orientation adjustments driven by the specific pulsing directions.The EBSD technique facilitates us with the crystallographic orientation of individual grains in terms of the Euler Angles ( ) j j j , , 1 2 (see figure 4).Using these Euler Angles, we have computed the rotation angle and rotation axis.
The rotation matrix, based on ZXZ rotation [48] i.e. rotating the Z axis by j , 1 and then the X axis by j and again the Z axis by j , 2 is defined as sin cos sin sin cos cos cos sin sin sin cos sin sin cos cos sin sin cos cos cos cos sin sin sin cos sin cos For each set of Euler Angle, the absolute rotation angle (q), ignoring sample and crystal symmetries, was calculated as and P a b be the rotation matrix and q q q , , p a b be the absolute rotation for the crystallographic orientation of a particular grain for the Pristine, 1-3 direction, and 2-4 direction pulsed conditions, respectively.
We then compute the relative rotation q pa (the rotation of 1-3 direction pulsed grain with respect to pristine orientation) from the rotation matrix Pa p T a and q ab (the rotation of 2-4 direction pulsed grain with respect to the 1-3 direction pulsed grain) from The axis of rotation for each case is computed by the following relation: The obtained results for four individual grains are tabulated in table 2. For example, in the case of Grain A, it rotates from its pristine condition by  18.13 about [−0.58, −0.64, −0.50] axis due to the pulsing in the 1-3 axial direction.When a grain rotates by a specific angle around a particular axis, it essentially adjusts its orientation in space.The rotation axis dictates the direction in which the grain's orientation is altered.The observed clockwise rotation of the crystal following 1-3 axial pulsed treatment can be attributed to the influence of the electron wind force.As the pulsed current flows along the 1-3 axis, the movement of electrons generates a directional pressure on the crystal lattice, inducing a collective rotation of the crystallographic orientations.Upon pulsing in a cross perpendicular 2-4 direction, we observe that the grain rotates by  18.28 from its last position about [0.38, 0.54, 0.75] axis.The subsequent reversal to an anticlockwise rotation post 2-4 axial pulsed treatment can be attributed to the interplay between the lattice structure and the electron wind force.When the current flows along the 2-4 directions, the lattice structure responds differently to the directional pressure exerted by the electron winds.This variation in response leads to an opposite pole rotation compared to the 1-3 axial pulsing.It is to be noted that the opposite sign of two rotation axes denotes distinct directions in three-dimensional space, signifying opposite rotational orientations.While the rotation angle is nearly equal for both cases, the opposite rotationaround different axes does not inherently result in the grain returning to its pristine condition due to the constraints from the neighboring grains.

Other parameters affected by electropulsing
In addition to grain rotation, we observed significant grain refinement due to the proposed room-temperature annealing method.The grain size decreases due to electropulsing irrespective of its direction (see figure 2).The application of short pulse width current, even at low frequencies, imparts energy to the lattice by transferring their momentum, enabling atomic rearrangements and, subsequently, grain refinement [31].The degree of refinement depends on the density of the applied pulsed current [49].Initially, the mean grain size of the entire ROI decreases by 11.2% due to a moderate current density flow ( ) 3360 A cm 2 in 1-3 axial direction.Following that, a higher current density ( ) 6080 A cm 2 was applied in 2-4 directions, resulting in a further 5.7% reduction in grain size.The individual grain size of four designated grains, along with the mean grain size of ROI and AOI, are shown in table 3.
Although there was a significant change in the grain size and orientation due to the room temperature electropulsing, we observed a minor change in misorientation and low angle grain boundaries (LAGBs) density.Figure 5 shows the kernel average misorientation (KAM) distributions of our Area of Interest (AOI) for distinct pulsing conditions.The concentration of the LAGBs (   q   2 1 0) for pristine condition was 5.46%.The LAGBs are displayed by the red lines in figures 2(d)-(f).Followed by pulsing in 1-3 direction, the LAGBs concentrations decreased slightly to 3.81%, and after pulsing in 2-4 direction, the concentration was 5.17%.In addition, the KAM maps demonstrate minimal alteration in their distribution following the processing.

Conclusions
A non-thermal room temperature annealing technique is employed to control the grain rotation of polycrystalline material using short pulse width and low-frequency current passing through pre-defined directions.We investigated the effects of electropulsing direction using the Electron Backscattered Diffraction (EBSD) to demonstrate that the material's response depends on the direction of the pulsed current.Individual pole figure analysis of four designated grains indicates that the grain rotation induced by electropulsing in a given direction can be effectively reverted to the opposite direction by applying electropulsing in a cross-perpendicular direction.The grain orientations and the associated rotations due to electropulsing is achieved through Euler angles, which are obtained from the EBSD analysis.The opposite sign of the rotation axes vectors implies the clockwise and counterclockwise rotations; however, the grains don't necessarily coincide with their pristine condition as the axes of rotations are different.In addition to the observed grain rotation and counter-rotation, a notable (up to 16.2%) grain refinement has also been detected following the application of electropulsing.It is to be noted that the grain refinements are independent of electropulsing direction, rather depend on the current density.However, the low angle grain boundaries concentration and average misorientation remain mostly unchanged.This study shows a potential technique for controlling and/or tailoring the microstructure through electropulsing.

Figure 1 .
Figure 1.(a) EDX spectrum confirming the elements of alloy (Fe-73%, Cr-21.3%,Al-5.7%) in the sample, (b) SEM image of the sample showing four terminals for directional electropulsing.(c) Infrared thermograph showing the temperature distribution across the sample during the electropulsing in 1-3 direction, (arrows indicating directions of the applied pulsed current).
2(a)-(c), wherein Electron Backscatter Diffraction (EBSD) maps, presented as Inverse Pole figure maps in the z-direction (zIPF), and Grain Boundaries (GB) overlays, displaying the microstructural landscape within the encompassing boundaries of the designated Region of Interest (ROI).Additionally, the Grain Boundaries (GB) within the Area of Interest (AOI) are shown in figures 2(d)-(f) for the various conditions.Although the overall micro texture of the grains in the ROI before (figure 2(a)) and after 1-3 (figures 2(b)) and 2-4 (figure 2(c)) electropulsing was fairly the same and represented by the 〈111〉//z crystallographic orientation, some of the individual grains were clearly observed to undergo a change/shift in orientation.Likewise, corresponding grain boundaries (figures 2(d), (e), and (f)) did not show a

Figure 4 .
Figure 4. Individual analyses of Grains A, B, C, and D for various pulsing conditions.

Table 2 .
Rotation angle and rotation axis for various grains.

Table 3 .
Grain size for various pulsing conditions.