# Fileset

[083903_1_5.0304815.pdf](https://mdr.nims.go.jp/filesets/64aaf72d-fdbe-4abd-a1e5-4bf8dcf4dc35/download)

## Creator

[Hirofumi Suto](https://orcid.org/0000-0003-4387-5862), [Keita Katayama](https://orcid.org/0000-0003-4278-8863), Yohei Tanaka, Dolly Taparia, [Nattamon Suwannaharn](https://orcid.org/0000-0003-1285-599X), [Tomoya Nakatani](https://orcid.org/0000-0001-9590-216X), [Taisuke T. Sasaki](https://orcid.org/0000-0002-5952-7638), [Hisato Yabuta](https://orcid.org/0000-0001-6008-7054), [Yuya Sakuraba](https://orcid.org/0000-0003-4618-9550)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Excimer-laser-annealing-induced crystallization and atomic ordering of Co2Mn0.5Fe0.5Ge Heusler alloy thin films for spintronic applications](https://mdr.nims.go.jp/datasets/9e53e8f1-49a1-4cd8-ab53-684c46bd0640)

## Fulltext

Excimer-laser-annealing-induced crystallization and atomic ordering of Co2Mn0.5Fe0.5Ge Heusler alloy thin films for spintronic applicationsViewOnlineExportCitationRESEARCH ARTICLE |  FEBRUARY 26 2026Excimer-laser-annealing-induced crystallization and atomicordering of Co2Mn0.5Fe0.5Ge Heusler alloy thin films forspintronic applicationsHirofumi Suto   ; Keita Katayama  ; Yohei Tanaka; Dolly Taparia; Nattamon Suwannaharn  ;Tomoya Nakatani  ; Taisuke T. Sasaki  ; Hisato Yabuta  ; Yuya Sakuraba J. Appl. Phys. 139, 083903 (2026)https://doi.org/10.1063/5.0304815Articles You May Be Interested InTemperature-dependence of current-perpendicular-to-the-plane giant magnetoresistance spin-valves usingCo2(Mn1−xFex)Ge Heusler alloysJ. Appl. Phys. (April 2016)Large magnetoresistance and high spin-transfer torque efficiency of Co2MnxFe1−xGe (0 ≤ x ≤ 1) Heusleralloy thin films obtained by high-throughput compositional optimization using combinatorially sputteredcomposition-gradient filmAPL Mater. (November 2024)Improved current-perpendicular-to-plane giant magnetoresistance outputs by heterogeneous Ag-In:Mn-Zn-O nanocomposite spacer layer prepared from Ag-In-Zn-O precursorJ. Appl. Phys. (November 2019) 20 April 2026 06:32:19https://pubs.aip.org/aip/jap/article/139/8/083903/3381364/Excimer-laser-annealing-induced-crystallizationhttps://pubs.aip.org/aip/jap/article/139/8/083903/3381364/Excimer-laser-annealing-induced-crystallization?pdfCoverIconEvent=citejavascript:;https://orcid.org/0000-0003-4387-5862javascript:;https://orcid.org/0000-0003-4278-8863javascript:;javascript:;javascript:;https://orcid.org/0000-0003-1285-599Xjavascript:;https://orcid.org/0000-0001-9590-216Xjavascript:;https://orcid.org/0000-0002-5952-7638javascript:;https://orcid.org/0000-0001-6008-7054javascript:;https://orcid.org/0000-0003-4618-9550https://crossmark.crossref.org/dialog/?doi=10.1063/5.0304815&domain=pdf&date_stamp=2026-02-26https://doi.org/10.1063/5.0304815https://pubs.aip.org/aip/jap/article/119/15/153903/141912/Temperature-dependence-of-current-perpendicular-tohttps://pubs.aip.org/aip/apm/article/12/11/111114/3321148/Large-magnetoresistance-and-high-spin-transferhttps://pubs.aip.org/aip/jap/article/126/17/173904/595140/Improved-current-perpendicular-to-plane-gianthttps://servedbyadbutler.com/redirect.spark?MID=188841&plid=3615298&setID=1044475&channelID=0&CID=1697306&banID=524364881&PID=0&textadID=0&tc=1&rnd=4963046619&scheduleID=3830852&placementScheduleId=3830852&adItemScheduleId=0&adSize=1640x440&data_keys=%7B%22%22%3A%22%22%7D&metadata=%5B%5D&mt=1776666739928894&spr=1&referrer=http%3A%2F%2Fpubs.aip.org%2Faip%2Fjap%2Farticle-pdf%2Fdoi%2F10.1063%2F5.0304815%2F20919572%2F083903_1_5.0304815.pdf&request_uuid=f2971c1e-2b90-464c-8987-3ffa22c7f374&hc=81d81509c91adef6f0061d015ba3393778907229&location=Excimer-laser-annealing-induced crystallizationand atomic ordering of Co2Mn0.5Fe0.5Ge Heusleralloy thin films for spintronic applicationsCite as: J. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815View Online Export Citation CrossMarkSubmitted: 30 September 2025 · Accepted: 14 January 2026 ·Published Online: 26 February 2026Hirofumi Suto,1,a) Keita Katayama,2,3 Yohei Tanaka,3 Dolly Taparia,1 Nattamon Suwannaharn,1Tomoya Nakatani,1 Taisuke T. Sasaki,1 Hisato Yabuta,2,3 and Yuya Sakuraba1AFFILIATIONS1Research Center for Magnetic and Spintronic Materials, National Institute for Materials Science (NIMS), Tsukuba, Japan2Graduate School of Information Science and Electrical Engineering, Kyushu University, Fukuoka, Japan3Department of Gigaphoton Next GLP, Kyushu University, Fukuoka, Japana)Author to whom correspondence should be addressed: SUTO.Hirofumi@nims.go.jpABSTRACTMagnetic Heusler alloys are highly attractive for spintronics; however, realizing their full potential requires high-temperature annealing,which is often incompatible with practical device fabrication. Excimer laser annealing (ELA) potentially addresses this temperature con-straint by making use of the short annealing time and temperature gradient along the depth direction. We investigated the effect of ELA onCo2Mn0.5Fe0.5Ge half-metallic Heusler-alloy thin films and demonstrated that ELA successfully induces crystallization and B2 atomic order-ing. Optimized ELA condition using low fluence with a high number of laser irradiations achieved reduced resistivity and negative aniso-tropic magnetoresistance, indicating improved atomic ordering and high spin polarization, while maintaining flatness of the films. Thesefindings establish ELA as a viable annealing method for integrating high-performance Heusler alloys into the device structure with strictthermal budget. Moreover, ELA offers additional advantages such as enhanced throughput and selective area annealing, thereby broadeningthe scope of Heusler-alloy applications in spintronic devices.© 2026 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license(https://creativecommons.org/licenses/by/4.0/). https://doi.org/10.1063/5.0304815I. INTRODUCTIONHeusler alloys exhibit intriguing properties such as high spinpolarization found in Co-based half-metallic Heusler alloys, such asCo2Mn0.5Fe0.5Ge (CMFG),1–6 Co2FeGa0.5Ge0.5,7–10 and others,11,12and a large anomalous Hall effect found in Co2MnAl13,14 andCo2MnGa,15,16 making them highly attractive materials for spin-tronic applications.17,18 One major challenge in implementingHeusler-alloy-based thin film devices is the process temperaturerequired to achieve their full potential for the following reason. Inmany Co-based Heusler alloys, denoted as Co2YZ, the L21-orderedstructure with atomic ordering in all the Co, Y, and X atoms is ener-getically most stable in thermal equilibrium. Deposited thin films,however, usually exhibit atomic disorder, leading to a B2 structurewith disorder between Y and Z atoms, an A2 structure with disorderamong all constituent atoms, or an amorphous structure. Suchdisorders are known to degrade the material properties, and anneal-ing processes such as postdeposition furnace annealing or deposi-tion at an elevated temperature are generally employed to promoteatomic ordering. In the case of current-perpendicular-to-plane giantmagnetoresistance (CPP-GMR) devices, the magnetoresistance(MR) ratio shows an increasing trend with respect to the processtemperature up to 600 °C,6 indicating the advantage of the improvedatomic ordering at high process temperatures. However, practicalapplications impose temperature limitations. For example, inCPP-GMR read heads for hard-disk-drives, one of the most exten-sively studied spintronic applications of Heusler alloys, the processtemperature is limited to 300 °C due to the temperature tolerance ofthe soft magnetic shield layers fabricated near the CPP-GMR devicefor improved spatial resolution.17 Similarly, in the case of magneto-resistive random access memory, the process temperature of MRmemory cells fabricated in the back-end-of-line process is limited toJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-1© Author(s) 2026 20 April 2026 06:32:19https://doi.org/10.1063/5.0304815https://doi.org/10.1063/5.0304815https://pubs.aip.org/action/showCitFormats?type=show&doi=10.1063/5.0304815http://crossmark.crossref.org/dialog/?doi=10.1063/5.0304815&domain=pdf&date_stamp=2026-02-26https://orcid.org/0000-0003-4387-5862https://orcid.org/0000-0003-4278-8863https://orcid.org/0000-0003-1285-599Xhttps://orcid.org/0000-0001-9590-216Xhttps://orcid.org/0000-0002-5952-7638https://orcid.org/0000-0001-6008-7054https://orcid.org/0000-0003-4618-9550mailto:SUTO.Hirofumi@nims.go.jphttps://creativecommons.org/licenses/by/4.0/https://creativecommons.org/licenses/by/4.0/https://doi.org/10.1063/5.0304815https://pubs.aip.org/aip/jap400 °C.19 In the case of spintronic devices on flexible substrates,temperature limitation ranges from 200–450 °C depending on thesubstrate materials. Currently, the implementation of the aboveapplications is sought mainly by using conventional CoFe-basedmaterials.17,19–22 To address the above temperature constraint ofHeusler alloys, research has been conducted to lower the requiredtemperature through optimizing materials and structures. Forexample, the amorphization of CMFG by an amorphous seedlayer2,3 or alloying Co2FeSi with a small amount of Ag23,24 wasreported effective to improve atomic ordering at relatively lowannealing temperatures. In addition, annealing Heusler-based filmsat sufficient temperature on a high-temperature compatible structurefollowed by transferring and bonding to the device structure hasbeen studied.25Excimer laser annealing (ELA) is an alternative annealingmethod that can overcome the temperature constraints. In ELA,pulsed laser is applied to the sample to induce an annealing effect.Because of the short pulse width typically in the nanosecond region,ELA can prevent the sample from sustained exposure to high tem-perature. Furthermore, laser absorption and heat generation occurprimarily near the surface, creating a temperature gradient along thedepth direction. Therefore, by optimizing the laser condition andthermal design of devices, conducting ELA right after the depositionof Heusler-based device films can allow the films to reach a suffi-cient temperature, while keeping the underlying structures coolerthan the temperature limitation. ELA has been extensively studiedfor the fabrication of low-temperature polysilicon thin films on glassor flexible substrates, where amorphous Si layers are rapidly heatedto the melting point for crystallization before heat diffusion into thesubstrate causes damage.26–28 Moreover, the ELA application hasbeen explored for a wide variety of functional materials, with moti-vations such as integrating the materials into devices with lowthermal budget or increasing heating and cooling rates.29–34 Laserannealing also offers the advantage of selected-area annealing bydefining the irradiated area.34 This selected-area annealing enablesthe local control of magnetization direction of the pinned layer inspin-valve MR devices.35–37 However, the use of ELA in orderedalloys such as Heusler alloys remains unexplored.In this study, we investigate the effect of ELA on sputter-deposited CMFG Heusler-alloy thin films for spintronic applications.Among half-metallic Co-based Heusler alloys, CMFG attracts atten-tion owing to high atomic ordering obtained at relatively low anneal-ing temperature. X-ray diffraction (XRD) measurements revealedthat the CMFG film was amorphous in the as-deposited state andchanged into a polycrystalline structure with B2 atomic ordering byELA. The fluence required for crystallization decreases with increas-ing number of laser irradiation because the annealing effect is accu-mulated by multiple laser irradiations. An excess increase in fluenceresulted in increased film roughness, being detrimental to practicalapplications. The effect of ELA was examined also by measuringanisotropic magnetoresistance (AMR) and resistivity. Based on thecomparison of the AMR and resistivity between the samples treatedwith ELA and furnace annealing, ELA was found to induce theannealing effect equivalent to a temperature of 300 °C at whichCMFG should show relatively good device performance.Transmission electron microscopy (TEM) observation shows the filmafter ELA has grain with a size of a few tens of nanometers withoutnoticeable interdiffusion between seed and capping layers. Theseresults demonstrate the potential of ELA as a viable annealing tech-nique, broadening the scope of Heusler alloy applications.II. EXPERIMENTALThe samples were prepared by depositing Ta (2 nm)/Co38Fe38B19Ta5 (1.5 nm)/CMFG (50 nm)/Ru (2 nm) on 2 cm × 2 cmSi substrates with a 500 nm thermally oxidized layer by using mag-netron sputtering. The order of the layers is from bottom to top.The composition of the CMFG layer was Co49.7Mn17.8Fe9.3Ge23.2,as measured by X-ray fluorescence analysis. ELA experiments wereconducted using a KrF excimer laser (Gigaphoton Inc.) with awavelength (λ) of 248 nm and a pulse duration of 83 ns by varyingthe following three ELA parameters: fluence, number, and fre-quency of laser pulses. The pulsed laser was introduced to homoge-nizing optics to form a uniform rectangular fluence profile with thearea of 0.4 × 1.2 mm2 on the samples. This size of the irradiationarea was chosen to allow the testing of various laser conditionswithin a single substrate and can be adjusted depending on theintended use in applications. The samples were placed in achamber with constant Ar flow at the atmospheric pressure to sup-press the oxidation of the samples, and the laser was introducedthrough the window of the chamber. By moving the samplechamber, laser irradiation was conducted on the designated posi-tion on the sample. Based on the complex refractive index of Ruused as the surface layer (n = 1.05 and k = 2.75), the reflectance atthe surface defined by ((n� 1)2 þ k2)/((nþ 1)2 þ k2) is calculatedto be 0.64, and the penetration depth defined by λ/4πk is calculatedto be 7.2 nm. The penetration depth is larger than the Ru thickness,meaning that a substantial amount of laser penetrated into theCMFG layer. Although the detailed optical constant of CMFG isunknown, based on the fact that light penetration depth in metal inthe UV region is typically 10 nm or shorter, the laser was mostlyabsorbed in the CMFG layer with little reaching the substrate. As areference, the samples with a size of 10 × 10 mm were also pro-cessed with the conventional annealing in a vacuum furnace for30 min at various post-annealing temperatures (Tp) from 200 to500 °C. During furnace annealing, the temperature was rampedslowly from RT over the course of an hour to prevent overshoot.Crystallization and atomic ordering were examined by XRDmeasurements (Rigaku SmartLab) with Cu-Kα radiation with awavelength of 1.5406 Å. 2θ−ω scans were conducted using a two-dimensional detector (Rigaku HyPix-3000) with an active area of77.5 × 38.5 mm2 placed at a camera length of 150 mm. The XRDprofiles were obtained by integrating the ring pattern along the βangle using the whole two-dimensional images except for the 2θrange below 30° where the β range of 150°–210° was used. The βangle corresponds to the angle of the Debye ring with 180° beingthe perpendicular direction, and thus, the β angle dependenceserves similar to the rocking curve measurement. The β range of179°–181° was excluded in the profiles to get rid of the strong sub-strate peaks with Si 004 appearing approximately at 69°. The Si 004peaks also appeared at 66° and 62° due to Cu-Kβ and W-Lα ener-gies contained in the source x ray, respectively, and the forbiddenSi 002 peak sometimes appeared at 33° due to multiple reflections.Surface morphology was measured by atomic force microscopyJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-2© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/jap(AFM) with a scan area of 1 × 1 μm2 to estimate the average rough-ness (Ra). AMR and resistivity measurements were conducted byapplying a current of 1 mA to the strips with a width of 40 μm andlength of 170 μm and measuring the resistance by the four-terminalmethod. The strips were patterned by Ar ion milling after ELA orfurnace annealing. AMR ratio is defined as (ρ//–ρ⊥)/ρ⊥, where ρ//and ρ⊥ are the resistivity of films when the directions of the currentand magnetic fields are parallel and orthogonal, respectively.According to theoretical and experimental studies, the cross-sectional scanning transmission electron microscopy (STEM)images, energy dispersive x-ray spectroscopy (EDS) elementalmaps, selected-area electron diffraction (SAED), and nanobeamelectron diffractions (NBED) were obtained for microstructureanalysis utilizing FEI Titan G2 80-200 TEM operating at 200 kV.III. RESULTS AND DISCUSSIONA. Effect of ELA and furnace annealing oncrystallization, atomic ordering, and surface roughnessWe first investigated the effect of conventional furnace anneal-ing on the structure and morphology of the samples. Figure 1(a)shows the XRD profiles for as-deposited and furnace-annealsamples for Tp = 200–500 °C. In the as-deposited state, a very broadpeak at around 2θ = 44° appears originating from short-range localatomic arrangements, which indicates the amorphous structure ofthe film without long-range order. After annealing at 200 °C, funda-mental 004 and 202 peaks and a superlattice 002 peak appear, indi-cating that the CMFG layer changed to a poly-crystalline structurewith B2 atomic ordering. No significant change in the XRD profileswas observed for Tp = 200–500 °C. Figure 1(b) shows the 2D XRDimages of the Tp = 450 °C sample obtained for χ = 0° and 45°. Forχ = 0°, the ring patterns of 202 and 004 are observed, and theformer shows a higher intensity at the upper and lower edges of the2D image, whereas the latter shows a higher intensity near thecenter. On the other hand, only the 202 ring pattern is observed forχ = 45°. This angle dependence of the peak intensities is summa-rized in the inset, where the center of the 004 peak appears in theperpendicular direction and that of the 202 peak is tilted by 45°,with the full width at half maximum of both peaks being approxi-mately 30°. These results indicate a weak (001) texture of the CMFGfilm, with the {110} planes naturally tilted by 45° from theout-of-plane direction. The 002 peak is visible only in the profileand not in the 2D image due to its small intensity. According to theprevious reports, L21 atomic ordering might exist at higher Tp, butthe 111 superlattice peak reflecting the L21 ordering was notobserved due to its small intensity. Note that all the furnace-annealsamples show good surface specularity, indicating their flat surface.We next conducted ELA on the as-deposited sample.Figure 2(a) shows the photo image of the ELA sample for 10 Hz,10 shots, and various fluences from 40 to 300 mJ/cm2. The positionmarks were made by damaging the film with high fluence, denotedby the red box, and the 2 mm wide area next to the position markis the ELA area made by sweeping the laser, denoted by the bluebox. When the fluence is 200 mJ/cm2 or more, the surface specular-ity was degraded, suggesting a change in the surface morphology.Figure 2(b) shows the corresponding XRD profiles. For fluence upto 100 mJ/cm2, the profiles are the same as that of the as-depositedsample, indicating that the amorphous structure was unchanged.Above 120 mJ/cm2, 004, 202, and 002 peaks appear, demonstratingthat the crystallization and atomic ordering of the CMFG layerswere successfully induced by ELA. At 260 mJ/cm2, the peaksdo not appear despite the visible change for unknown reasons.Figures 2(c) and 2(d) show the results from the same measure-ments when the shot number was increased to 1000. The degrada-tion of surface specularity occurs from 160mJ/cm2, and severedamage was observed at 300 mJ/cm2. XRD peaks appear from100 mJ/cm2. The peaks abruptly become sharp and intense from200 mJ/cm2. These strong peaks suggest the growth of larger grainsby agglomeration, which can lead to the degradation of surfacespecularity. The fluence values at which the XRD peaks appearedFIG. 1. (a) Out-of-plane XRD profiles obtained for the furnace-anneal samplesat various post-annealing temperatures. Values shown to the right of the CMFG004 peaks represent their integral width, with errors corresponding to the lastdigit of the main value indicated in parentheses. (b) 2D XRD image obtained atTp = 450 °C for χ = 0° (top) and 45° (bottom). The inset shows the polar angleprofile of the peak intensities for 202 and 004 rings.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-3© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japand the surface specularity degraded decreased by increasing theshot number from 10 to 1000, indicating these effects of ELA accu-mulate with a number of laser irradiations.Figure 3(a) shows the out-of-plane lattice constants (c) as afunction of Tp and fluence, evaluated from the 004 peak positionsin the XRD results. In this analysis, the considered β angle rangewas reduced to 170°–179° and 181°–190° in determining the peakposition to suppress the effect of change in lattice constantsbetween the out-of-plane and in-plane directions, while maintain-ing a sufficient signal-to-noise ratio. The furnace-anneal sampleshows the c values around 0.571 nm, which is slightly smaller thanthe reported bulk value of 0.574 nm. The ELA samples exhibit cvalues below 0.57 nm, which are smaller than the furnace-annealsample, and exhibit a decreasing trend with fluence, suggesting thatlattice strain is induced in the ELA samples, presumably due tomuch faster rate of temperature change. To evaluate the strain, thein-plane lattice constant (a) of the furnace-anneal (Tp = 450 °C)sample was calculated from the c value and the 202 peak positionat χ = 45° with the β angle range of 130°–140°, yieldinga = 0.578 nm. For further confirmation, in-plane grazing-incidenceXRD with an incidence angle of 0.4° was also conducted tomeasure the 400 peak position (data not shown), yieldinga = 0.58 nm. These two a values are consistent and slightly largerthan the c value, and the c/a ratio is estimated to be 0.98–0.99.Similarly, the a value was evaluated to be 0.584 nm for the ELAsample (10 shots and 180 mJ/cm2) using the two XRD scans atχ = 0 and 45°. In-plane grazing-incidence XRD could not be per-formed due to the small ELA area. The c/a ratio is estimated to be0.96, confirming that ELA induces larger strain than furnaceannealing. The fluence dependence of the lattice constant changesthe trend at 200 mJ/cm2 for 1000 shots, which can be attributed tothe relaxation of the lattice strain by agglomeration.The integral width of the 004 peaks was also evaluated, asshown in Figs. 1(a), 2(b), and 2(d). The furnace-anneal samplesexhibit width in the range of 0.7°–0.9°. The ELA samples generallyshow a broader linewidth than the furnace-anneal samples, exceptfor high-fluence region above 200 mJ/cm2 for 1000 shots. Thebroader linewidth indicates that the ELA samples have a smallercrystallite size, larger lattice strain, and/or higher number ofdefects. In the high fluence range for 1000 shots, the linewidthdecreases, suggesting that increased atomic diffusion by high-fluence laser resulted in strain relaxation and formation of largergrains. In addition, for 220 mJ/cm2 for 10 shots, where the a valueshows the minimum among all the samples, the 004 peak exhibitsFIG. 2. (a) Photo image of the ELA sample for 10 Hz, 10 shots, and various fluences. The red square indicates the position mark made by strong laser, and the bluesquare indicates the ELA area for 40 mJ/cm2. (b) Corresponding out-of-plane XRD profiles. Values shown to the right of the CMFG 004 peaks represent their integralwidth, with errors corresponding to the last digit of the main value indicated in parentheses. (c) and (d) Results from the same measurement for 10 Hz and 1000 shots. In(a), due to the experimental mistake, 220 mJ/cm2 was conducted in a different position, and its image is copied as indicated by the black dashed square.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-4© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japan asymmetric shape with a tail to the lower angle, suggestingstrain-related defects.AFM observations were conducted to quantitatively evaluatethe surface roughness. Figure 4 shows the fluence dependence of Rafor 10 and 1000 shots, and the inset shows the Tp dependence ofRa of the furnace-anneal samples. In the as-deposited state, thesample surface is very flat with an Ra of approximately 0.15 nm.For furnace annealing, the roughness increases slightly with Tp toapproximately 0.4 nm at Tp = 500 °C. For ELA, the roughnessremains unchanged up to 140 mJ/cm2 and starts to increase, andthe increase is higher for 1000 shots due to the accumulated effectfrom multiple laser irradiations. The ELA conditions with increasedroughness show overall agreement with the visible change of thesamples in Figs. 2(a) and 2(c), indicating that the degraded surfacespecularity arose from the roughness. Increased roughness can bedetrimental in practical applications, especially for MR devicesFIG. 3. Tp dependence of c for the furnace-anneal samples and fluence depen-dence of c for the ELA samples.FIG. 4. Fluence dependence of Ra for 10 and 1000 shots (10 Hz) for the ELAsamples. The inset shows the Tp dependence of Ra for the furnace-annealsamples.FIG. 5. (a) Out-of-plane XRD profiles obtained for the ELA samples for variousshot number for 100 and 200 mJ/cm2. The irradiation frequency was 10 Hz for10, 100, and 1000 shots, and 1000 Hz for 10000 and 50000 shots. Shotnumber dependence of Ra for 100 and 200 mJ/cm2. (c) 2D XRD image for100 mJ/cm2 and 50 000 shots (1000 Hz).Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-5© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japconsisting of stacked nanometer-scale multilayers, where interfacequality is crucial to the device performance.The effect of shot number is examined by extending its upperrange to 50 000. Figures 5(a) and 5(b) show the dependence of theXRD profiles on the shot number and the corresponding Ra,respectively. Frequency of laser irradiation was set to 10 Hz for 10,100, and 1000 shots and 1000 Hz for 10 000 and 50 000 shots. For200 mJ/cm2, XRD peaks already appear for ten shots and intensifywith the shot number, which accompanies with an increase in theroughness. On the other hand, for 100 mJ/cm2, the peaks appearfirst for 1000 shots and intensify with the shot number, whichcaused no roughness increase even at 50 000 shots. These resultsindicate that, by limiting the fluence below a certain threshold,crystallization and atomic ordering can be promoted by repeatingthe laser irradiation without affecting the roughness. Figure 5(c)shows the 2D XRD image obtained for 100 mJ/cm2 and 50 000shots, which exhibit the ring patterns of 004 and 202 similar tothat of the furnace-anneal sample.B. Effect of ELA and furnace annealing on resistivityand AMRWe next investigate the effect of furnace annealing and ELAon the electrical and magneto-transport properties. Figures 6(a)and 6(b) show the Tp dependence of the resistivity and AMR ratiofor as-deposited and furnace-anneal samples, respectively. Bothresistivity and AMR ratio decrease monotonically with Tp up to450 °C. By further increasing Tp to 500 °C, the resistivity increases,whereas the AMR ratio continues to decrease. Resistivity generallyreflects the film structure, grain size, and atomic ordering.38Considering that the surface flatness was maintained in thefurnace-anneal samples, the decrease in resistivity is likely due tothe promoted grain growth, reduced defects, and improved atomicordering induced by annealing. Consistent with this interpretation,the linewidth of the XRD peaks in Fig. 1(a) shows overall narrow-ing trend with Tp. With regard to the AMR effect, its physicalorigin has been theoretically understood by considering s–d scatter-ing of conduction electrons.39 In particular, the model developedFIG. 6. Tp dependence of (a) resistivity and (b) AMR ratio obtained for the furnace-anneal samples. Fluence dependence of (c) resistivity and (d) AMR ratio obtained forthe ELA samples with different shot numbers. Arrows show the conditions for samples used in TEM observation.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-6© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japby Kokado et al. predicts half-metallic materials to exhibit negativeAMR, originating from the dominant s–d scattering of s↑→ d↑ ors↓→ d↓.40 This model has been experimentally validated in avariety of Co-based Heusler alloys, where the magnitude of negativeAMR correlates with half-metallicity.41–43 Based on these previousfindings, enhanced negative AMR observed at higher Tp can beattributed to improved half-metallicity as a result of improvedatomic ordering. Despite the indication of improved atomic order-ing from the above discussion, the 002 superlattice peak XRDresults in Fig. 1(a) show little Tp dependence because XRD is insen-sitive to change in the atomic ordering among atoms with closeatomic numbers such as Co, Fe, and Mn.Figures 6(c) and 6(d) show the fluence dependence of the resis-tivity and AMR ratio for the ELA samples with different shotnumbers, respectively. The frequency of laser irradiation was set to10Hz for 10, 100, and 1000 shots and 1000Hz for 5000, 10 000,and 50 000 shots. As will be discussed in detail later, changing fre-quency from 10 to 1000Hz affects the results due to increased heataccumulation by fast repetition of laser irradiation. However, theoverall trend remains unchanged, and the fluence shift originatingfrom heat accumulation is typically around 50mJ/cm2, confirmingthat frequency does not affect the interpretation of data in Figs. 6(c)and 6(d). Assuming all measurements were conducted at the samefrequency, the gap of the results for 1000 and 5000 shots wouldshrink. In the case of ten shots, the resistivity is unchanged up to140 mJ/cm2, gradually decreases to the minimum at 220 mJ/cm2,and then slightly increases. The AMR ratio exhibits a similar trend.Notably, the onset of resistivity reduction and negative AMR at160 mJ/cm2 coincides with the emergence of XRD peaks, indicatingthat the decreases in resistivity and the enhancement in negativeAMR originate from the promoted crystallization and atomic order-ing due to ELA. In the fluence range where the resistance decreasesand the negative AMR enhances, the XRD peak width shows anoverall narrowing trend, supporting that these changes partly origi-nate from the grain growth and reduced defects. An increase inresistivity after reaching the minimum can be attributed to increasedsurface roughness, as supported by the AFM results. In addition,interdiffusion among the layers may also contribute to the resistivityincrease. Regarding the horizontal axis, the position of both theonset of the resistivity change and the minimum resistivity shifts tothe lower fluence side with the shot number. Regarding the verticalaxis, the minimum resistivity decreased with the shot number,achieving 92 μΩ cm for 10 000 and 50 000 shots and the fluencerange of 100–140 mJ/cm2. The trend in the AMR ratio follows thatof the resistivity, with the largest negative AMR ratio being−0.095%. Notably, this fluence range does not affect the roughness.Comparison between ELA and furnace annealing resultsreveals that the minimum resistivity achieved by ELA correspondsto the values obtained at furnace annealing temperatures between300 and 350 °C. Similarly, the maximum negative AMR ratio byELA corresponds to values between 250 and 300 °C, indicating thatthe effect of ELA is equivalent to annealing at approximately300 °C. However, the minimum resistivity and largest negativeAMR of the ELA samples under the tested conditions could notreach those of the furnace-anneal samples at higher Tp, indicatingthat the furnace annealing at higher Tp can achieve higher atomicordering and half-metallicity than ELA. Nevertheless, the studieson CPP-GMR devices with CMFG reported that furnace annealingat 300 °C exhibited relatively good MR output, due to the improvedB2 ordering.3,5 Our findings suggest that CMFG films processedwith the optimal ELA condition can potentially exhibit good deviceperformance.The total laser irradiation time for 50 000 shots is 4.15 ms(83 ns × 50 000 shots), and the total process time is 50 s (50 000shots/1000 Hz). These times are remarkably shorter than that ofthe furnace annealing, which takes 30 min for the constant temper-ature along with the additional 2 h for heating and cooling thesystem, suggesting that the ELA can enhance the throughput of theannealing process.We also investigated the effect of the frequency of laser irradi-ation. Figures 7(a) and 7(b) show the fluence dependence of resis-tivity and AMR ratio obtained for 1000 shot by setting thefrequency to 10, 1000, and 2000 Hz and for 50 000 shots by settingthe frequency to 1000 and 2000 Hz. In the case of 1000 shots, theFIG. 7. Fluence dependence of (a) resistivity and (b) AMR ratio obtained forthe ELA samples with different frequencies of laser irradiation.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-7© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japonset of decrease in the resistivity and AMR shifts to the lowerfluence side by increasing the frequency from 10 to 1000 Hz. Thisshift can be attributed to heat accumulation by repeated laser irra-diation. As the interval between laser irradiation shortens with fre-quency, the heat from the prior laser irradiation remains, and theresultant temperature increase enhances the annealing effect. Theresults for 1000 and 2000 Hz also show a shift, although the shift issmall, indicating that the temperature change by the residual heatis also small due to the small interval difference between 1000 and2000 Hz.C. Microstructure analysis by TEMTEM analysis was conducted to investigate the microstruc-ture and interface quality. Figures 8(a)–8(c) show the bright-field(BF) STEM images and their corresponding diffraction patternsfor the as-deposited and two ELA samples with the followingconditions: 10 shots, 10 Hz, 180 mJ/cm2 and 50 000 shots,1000 Hz, 100 mJ/cm2. These two ELA conditions were selected torepresent the lowest and highest shot numbers, while maintainingfluence below the threshold to affect surface roughness. The AMRresults of these three TEM samples are marked by the arrows inFig. 6(d) for ease of reference. The as-deposited sample exhibitsno distinct grain boundaries, and the SAED acquired in the samearea showed a clear halo feature superimposed with diffractionspots, which confirmed that the structure is mostly amorphouswith some crystallization, consistent with the XRD results. In theELA samples, crystalline grains were observed, and the SAED pat-terns exhibit rings of clear diffraction spots without a halo, indi-cating their polycrystalline nature and absence of an amorphousstructure. Furthermore, the 002 superlattice spots confirm B2atomic ordering in SAED or NBED taken from particular grainsalong [110] zone axis, consistent with the XRD results. The weak(001) texture observed in the XRD is not confirmed in the SAEDpatterns due to the low counting statistics of the grains. Thestructure of the Ta seed layer was amorphous regardless of theELA condition (data not shown). Grain size varies depending onthe laser conditions: the sample in (b) exhibits grains rangingfrom 20–40 nm, while the sample in (c) exhibits larger grainsranging from 30 to 60 nm. The sample in (b) was irradiated withFIG. 8. Cross-sectional STEM images and electron diffraction patterns for (a) the as-deposited sample and the two ELA samples (b) 10 shots, 10 Hz, and 180 mJ/cm2and (c) 50 000 shots, 1000 Hz, and 100 mJ/cm2. Dashed and dotted rings in the SAED patterns represent the positions of fundamental and B2 superlattice reflections,respectively. (d) EDS elemental map for the sample in (c) and corresponding compositional profile along the thickness direction.Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-8© Author(s) 2026 20 April 2026 06:32:19https://pubs.aip.org/aip/japthe higher-fluence laser, resulting in a higher peak temperatureand a faster ramping rate. Such annealing conditions are expectedto create a larger number of nucleation sites, leading to a fine-grained structure. This difference in the grain size is consistentwith the resistivity measurements in which the sample in (b)exhibited higher resistivity than the sample in (c). Figure 8(d)shows the EDS elemental maps of the 50 000-shot sample in (c)and the corresponding compositional profile along the thicknessdirection. There is no noticeable interdiffusion among the seed,CMFG, and cap layers. This suggests that ELA is applicable toHeusler-based multilayer devices, where interdiffusion is prob-lematic. STEM analysis revealed no clear non-uniformity, such asgrain size, along the depth direction. Although the actual temper-ature profile was unmeasurable, these results suggest that theeffect of temperature gradient is not significant in these experi-ments, which can be attributed to the thermal properties of oursamples. Because the sample structure consists of metallic seed,CMFG, and cap layers with high thermal conductivity and a SiO2underlayer with low thermal conductivity, the generated heat isconfined in the CMFG layer, contributing to the uniform temper-ature. To actively make use of the steep temperature gradient,designing the thermal and optical properties of samples and opti-mizing ELA condition are required. For example, introducing athermal barrier layer with low thermal resistivity above a lowthermal-budget region and below Heusler-based device film canenhance the temperature difference between them. Additionally,incorporating a heat sink structure or reducing the laser pulsewidth could be explored.IV. CONCLUSIONSWe investigated the effect of ELA as a thermal treatment forHeusler-alloy thin films because of its potential to address the tem-perature constraint in practical applications due to the shortannealing time and temperature gradient. XRD analysis confirmedthat ELA successfully transformed the as-deposited amorphousCMFG film into a polycrystalline structure with B2 atomic order-ing, which is crucial to derive the material’s potential. Althoughexcess fluence resulted in increased surface roughness, we identifieda process window, where atomic ordering is induced withoutincreasing the roughness. Within this window, STEM and EDSanalyses revealed no noticeable interdiffusion between the CMFGlayer and the adjacent seed and capping layers. The high film flat-ness without interdiffusion in the ELA samples indicates the appli-cability of ELA for important applications using nanometer-scalemultilayers, such as MR devices. Analysis by resistivity and AMRmeasurements revealed that combining a modest fluence laser witha large number of laser irradiations realizes higher annealingeffects, and the optimized ELA process induced annealing effectsequivalent to conventional furnace annealing at approximately300 °C. Although depth temperature profile remains unmeasured,the nature of ELA suggests that underlying layers experience lowertemperatures than the CMFG layer, implying that, under thepresent conditions, it is feasible to form a CMFG layer with proper-ties equivalent to 300 °C furnace annealing on structures with lowthermal budget. Together with the established advantages of theELA process such as enhanced throughput and selective areaannealing, these findings demonstrate that ELA is a promisingalternative annealing technique for Heusler alloys in spintronicapplications and can broaden their scope of use. However, furnaceannealing at higher temperatures was found more effective thanELA under the conditions tested in this study, presumably due tothe limited kinetics in the nanosecond-scale ELA process. Futurework should focus on designing sample structures and materialsfrom thermal and optical perspectives and optimizing ELA condi-tions to enhance the annealing effects and actively make use of thethermal gradient.ACKNOWLEDGMENTSThis work was partially supported by Advanced StorageResearch Consortium (ASRC), ARIM of MEXT (JPMXP1225NM5220), MEXT Initiative to Establish Next-generation NovelIntegrated Circuits Centers (X-NICS) (Grant No. JPJ011438), andGigaphoton Inc. The authors thank M. Inoue of NIMS for techni-cal support.AUTHOR DECLARATIONSConflict of InterestThe authors have no conflicts to disclose.Author ContributionsHirofumi Suto: Conceptualization (equal); Investigation (equal);Project administration (equal); Visualization (equal); Writing –original draft (equal). Keita Katayama: Investigation (equal);Methodology (lead); Writing – review & editing (equal). YoheiTanaka: Investigation (equal); Methodology (equal); Writing –review & editing (equal). Dolly Taparia: Investigation (equal).Nattamon Suwannaharn: Investigation (equal); Visualization(equal); Writing – original draft (equal). Tomoya Nakatani:Investigation (equal); Writing – review & editing (equal). TaisukeT. Sasaki: Investigation (equal); Writing – review & editing(equal). Hisato Yabuta: Conceptualization (equal); Supervision(equal); Writing – review & editing (equal). Yuya Sakuraba:Conceptualization (equal); Supervision (equal); Writing – review &editing (equal).DATA AVAILABILITYThe data that support the findings of this study are availablefrom the corresponding author upon reasonable request.REFERENCES1M. R. Page, T. M. Nakatani, D. A. Stewart, B. R. York, J. C. Read, Y. S. Choi,and J. R. Childress, “Temperature-dependence of current-perpendicular-to-the-plane giant magnetoresistance spin-valves using Co2(Mn1−xFex)Ge Heusleralloys,” J. Appl. Phys. 119, 2 (2016).2Y.-s. Choi, T. Nakatani, J. C. Read, M. J. Carey, D. A. Stewart, andJ. R. Childress, “Enhancement of current-perpendicular-to-plane giant magneto-resistance by insertion of amorphous ferromagnetic underlayer in Heusler alloy-based spin-valve structures,” Appl. Phys. Express 10, 013006 (2017).3S. Li, T. Nakatani, K. Masuda, Y. Sakuraba, X. D. Xu, T. T. Sasaki, H. Tajiri,Y. Miura, T. Furubayashi, and K. Hono, “Enhancement ofcurrent-perpendicular-to-plane giant magnetoresistive outputs by improvingJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-9© Author(s) 2026 20 April 2026 06:32:19https://doi.org/10.1063/1.4947119https://doi.org/10.7567/APEX.10.013006https://pubs.aip.org/aip/japB2-order in polycrystalline Co2(Mn0.6Fe0.4)Ge Heusler alloy films with the inser-tion of amorphous CoFeBTa underlayer,” Acta Mater. 142, 49 (2018).4T. Nakatani, S. Li, Y. Sakuraba, T. Furubayashi, and K. Hono, “AdvancedCPP-GMR spin-valve sensors for narrow reader applications,” IEEE Trans.Magn. 54, 3300211 (2018).5T. Nakatani, S. K. Narayananellore, L. S. R. Kumara, H. Tajiri, Y. Sakuraba, andK. Hono, “Thickness dependence of degree of B2 order of polycrystallineCo2(Mn0.6Fe0.4)Ge Heusler alloy films measured by anomalous x-ray diffractionand its impacts on current-perpendicular-to-plane giant magnetoresistance prop-erties,” Scr. Mater. 189, 63 (2020).6V. Barwal, H. Suto, R. Toyama, K. Simalaotao, T. Sasaki, Y. Miura, andY. Sakuraba, “Large magnetoresistance and high spin-transfer torque efficiencyof Co2MnxFe1−xGe (0≤ x≤ 1) Heusler alloy thin films obtained by high-throughput compositional optimization using combinatorially sputteredcomposition-gradient film,” APL Mater. 12, 111114 (2024).7S. Li, Y. K. Takahashi, T. Furubayashi, and K. Hono, “Enhancement of giantmagnetoresistance by L21 ordering in Co2Fe(Ge0.5Ga0.5) Heusler alloycurrent-perpendicular-to-plane pseudo spin valves,” Appl. Phys. Lett. 103,042405 (2013).8Y. Du, T. Furubayashi, T. T. Sasaki, Y. Sakuraba, Y. K. Takahashi, and K. Hono,“Large magnetoresistance in current-perpendicular-to-plane pseudo spin-valvesusing Co2Fe(Ga0.5Ge0.5) Heusler alloy and AgZn spacer,” Appl. Phys. Lett. 107,112405 (2015).9J. W. Jung, Y. Sakuraba, T. T. Sasaki, Y. Miura, and K. Hono, “Enhancement ofmagnetoresistance by inserting thin NiAl layers at the interfaces inCo2FeGa0.5Ge0.5/Ag/Co2FeGa0.5Ge0.5 current-perpendicular-to-plane pseudospin valves,” Appl. Phys. Lett. 108, 102408 (2016).10D. Taparia et al., “Improvement in CPP-GMR read head sensorperformance using [001]-oriented polycrystalline half-metallic Heusler alloyCo2FeGa0.5Ge0.5 and CoFe bilayer electrode,” Sci. Technol. Adv. Mater. 25,2388503 (2024).11Y. Sakuraba, K. Izumi, T. Iwase, S. Bosu, K. Saito, K. Takanashi, Y. Miura,K. Futatsukawa, K. Abe, and M. Shirai, “Mechanism of large magnetoresistancein Co2MnSi/Ag/Co2 MnSi devices with current perpendicular to the plane,”Phys. Rev. B 82, 094444 (2010).12T. Kubota, Y. Ina, Z. Wen, H. Narisawa, and K. Takanashi, “Currentperpendicular-to-plane giant magnetoresistance using an L12 Ag3Mg spacer andCo2Fe0.4Mn0.6Si Heusler alloy electrodes: Spacer thickness and annealing temper-ature dependence,” Phys. Rev. Mater. 1, 044402 (2017).13E. Vilanova Vidal, G. Stryganyuk, H. Schneider, C. Felser, and G. Jakob,“Exploring Co2MnAl Heusler compound for anomalous Hall effect sensors,”Appl. Phys. Lett. 99, 132509 (2011).14P. Li et al., “Giant room temperature anomalous Hall effect and tunable topol-ogy in a ferromagnetic topological semimetal Co2MnAl,” Nat. Commun. 11,3476 (2020).15A. Sakai et al., “Giant anomalous Nernst effect and quantum-critical scaling ina ferromagnetic semimetal,” Nat. Phys. 14, 1119 (2018).16K. Sumida et al., “Spin-polarized Weyl cones and giant anomalous Nernsteffect in ferromagnetic Heusler films,” Commun. Mater. 1, 89 (2020).17G. Albuquerque, S. Hernandez, M. T. Kief, D. Mauri, and L. Wang, “HDDreader technology roadmap to an areal density of 4 Tbpsi and beyond,” IEEETrans. Magn. 58, 1 (2022).18T. Nakatani, P. D. Kulkarni, H. Suto, K. Masuda, H. Iwasaki, and Y. Sakuraba,“Perspective on nanoscale magnetic sensors using giant anomalous Hall effect intopological magnetic materials for read head application in magnetic recording,”Appl. Phys. Lett. 124, 070501 (2024).19D. Edelstein et al., “A 14 nm embedded STT-MRAM CMOS technology,” inTechnical Digest—International Electron Devices Meeting, IEDM, Vols.2020-December (Institute of Electrical and Electronics Engineers Inc., 2020),pp. 11.5.1–11.5.4.20L. M. Loong, W. Lee, X. Qiu, P. Yang, H. Kawai, M. Saeys, J.-H. Ahn, andH. Yang, “Flexible MgO barrier magnetic tunnel junctions,” Adv. Mater. 28,4983 (2016).21J.-Y. Chen, Y.-C. Lau, J. M. D. Coey, M. Li, and J.-P. Wang, “High perfor-mance MgO-barrier magnetic tunnel junctions for flexible and wearable spin-tronic applications,” Sci. Rep. 7, 42001 (2017).22S. Ota, A. Ando, T. Sekitani, T. Koyama, and D. Chiba, “Flexible CoFeB/MgO-based magnetic tunnel junctions annealed at high temperature (≥350 °C),”Appl. Phys. Lett. 115, 202401 (2019).23S. Bosu, Y. Sakuraba, T. T. Sasaki, S. Li, and K. Hono, “Enhancement of L2 1order and spin-polarization of Heusler alloy Co2MnSi thin film by Ag alloying,”Scr. Mater. 110, 70 (2016).24S. Li, Y. Sakuraba, T. Sasaki, J. Chen, S. Bosu, and K. Hono, “Enhancedcurrent-perpendicular-to-plane giant magnetoresistance by improvement ofatomic order of Co2FeSi Heusler alloy film through Ag doping,” AIP Adv. 8,075230 (2018).25J. Chen et al., “Fully epitaxial giant magnetoresistive devices with half-metallicHeusler alloy fabricated on poly-crystalline electrode using three-dimensionalintegration technology,” Acta Mater. 200, 1038 (2020).26T. Sameshima, S. Usui, and M. Sekiya, “XeCl excimer laser annealingused in the fabrication of poly-Si TFTs,” IEEE Electron Device Lett. 7, 276(1986).27H. Watanabe, H. Miki, S. Sugai, K. Kawasaki, and T. Kioka, “Crystallizationprocess of polycrystalline silicon by KrF excimer laser annealing,” Jpn. J. Appl.Phys. 33, 4491 (1994).28M. Miyasaka and J. Stoemenos, “Excimer laser annealing of amorphous andsolid-phase-crystallized silicon films,” J. Appl. Phys. 86, 5556 (1999).29E. E. Marinero, “Material transformations in semiconductor and magneticthin films,” Appl. Surf. Sci. 43, 117 (1989).30S. S. N. Bharadwaja, T. Dechakupt, S. Trolier-Mckinstry, and H. Beratan,“Excimer laser crystallized (Pb,La)(Zr,Ti)O3 thin films,” J. Am. Ceram. Soc. 91,1580 (2008).31L. Grenouillet et al., “Nanosecond laser anneal (NLA) for Si-implanted HfO2ferroelectric memories integrated in back-end of line (BEOL),” in 2020 IEEESymposium on VLSI Technology (IEEE, 2020), pp. 1–2.32T. Okuda, A. Sugimura, O. Eryu, L. K. E. B. Serrona, N. Adachi, I. Sakamoto,and A. Nakanishi, “Nd-Fe-B thin films with perpendicular magnetic anisotropyand high coercivity prepared by pulsed laser annealing,” Jpn. J. Appl. Phys. 42,6859 (2003).33M. Perzanowski, M. Krupinski, A. Zarzycki, Y. Zabila, and M. Marszalek,“Structural ordering of laser-processed FePdCu thin alloy films,” J. AlloysCompd. 646, 773 (2015).34J. Nomoto, T. Koida, I. Yamaguchi, H. Makino, Y. Kitanaka, T. Nakajima, andT. Tsuchiya, “Over 130 cm2/Vs Hall mobility of flexible transparent conductiveIn2O3 films by excimer-laser solid-phase crystallization,” NPG Asia Mater. 14,76 (2022).35S. W. Kim, S. D. Choi, D. H. Jin, K. A. Lee, S. S. Lee, and D. G. Hwang, “Localmagnetization reversal in exchange biased film by laser annealing,” J. Magn.Magn. Mater. 272–276, 376–377 (2004).36I. Berthold, M. Müller, S. Klötzer, R. Ebert, S. Thomas, P. Matthes,M. Albrecht, and H. Exner, “Investigation of selective realignment of the pre-ferred magnetic direction in spin-valve layer stacks using laser radiation,” Appl.Surf. Sci. 302, 159–162 (2014).37A. Sharma et al., “Exchange bias and diffusion processes in laserannealed CoFeB/IrMn thin films,” J. Magn. Magn. Mater. 489, 165390(2019).38H. Kubota, J. Nakata, M. Oogane, Y. Ando, H. Kato, A. Sakuma, andT. Miyazaki, “Fabrication and characterization of Co–Mn–Al Heusler-type thinfilm,” J. Appl. Phys. 97, 10C913 (2005).39I. A. Campbell, A. Fert, and O. Jaoul, “The spontaneous resistivity anisotropyin Ni-based alloys,” J. Phys. C 3, S95 (1970).40S. Kokado, M. Tsunoda, K. Harigaya, and A. Sakuma, “Anisotropic magneto-resistance effects in Fe, Co, Ni, Fe 4N, and half-metallic ferromagnet: A system-atic analysis,” J. Phys. Soc. Jpn. 81, 024705 (2012).41Y. Sakuraba, S. Kokado, Y. Hirayama, T. Furubayashi, H. Sukegawa, S. Li,Y. K. Takahashi, and K. Hono, “Quantitative analysis of anisotropicJournal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-10© Author(s) 2026 20 April 2026 06:32:19https://doi.org/10.1016/j.actamat.2017.09.046https://doi.org/10.1109/TMAG.2017.2753221https://doi.org/10.1109/TMAG.2017.2753221https://doi.org/10.1016/j.scriptamat.2020.08.002https://doi.org/10.1063/5.0226638https://doi.org/10.1063/1.4816382https://doi.org/10.1063/1.4930229https://doi.org/10.1063/1.4943640https://doi.org/10.1080/14686996.2024.2388503https://doi.org/10.1103/PhysRevB.82.094444https://doi.org/10.1103/PhysRevMaterials.1.044402https://doi.org/10.1063/1.3644157https://doi.org/10.1038/s41467-020-17174-9https://doi.org/10.1038/s41567-018-0225-6https://doi.org/10.1038/s43246-020-00088-whttps://doi.org/10.1109/TMAG.2021.3081042https://doi.org/10.1109/TMAG.2021.3081042https://doi.org/10.1063/5.0191974https://doi.org/10.1002/adma.201600062https://doi.org/10.1038/srep42001https://doi.org/10.1063/1.5128952https://doi.org/10.1016/j.scriptamat.2015.08.003https://doi.org/10.1063/1.5045175https://doi.org/10.1016/j.actamat.2020.04.002https://doi.org/10.1109/EDL.1986.26372https://doi.org/10.1143/JJAP.33.4491https://doi.org/10.1143/JJAP.33.4491https://doi.org/10.1063/1.371560https://doi.org/10.1016/0169-4332(89)90200-6https://doi.org/10.1111/j.1551-2916.2008.02313.xhttps://doi.org/10.1143/JJAP.42.6859https://doi.org/10.1016/j.jallcom.2015.05.190https://doi.org/10.1016/j.jallcom.2015.05.190https://doi.org/10.1038/s41427-022-00421-4https://doi.org/10.1016/j.jmmm.2003.12.454https://doi.org/10.1016/j.jmmm.2003.12.454https://doi.org/10.1016/j.apsusc.2014.02.133https://doi.org/10.1016/j.apsusc.2014.02.133https://doi.org/10.1016/j.jmmm.2019.165390https://doi.org/10.1063/1.1852329https://doi.org/10.1088/0022-3719/3/1S/310https://doi.org/10.1143/JPSJ.81.024705https://pubs.aip.org/aip/japmagnetoresistance in Co2MnZ and Co2FeZ epitaxial thin films: A facile way toinvestigate spin-polarization in half-metallic Heusler compounds,” Appl. Phys.Lett. 104, 172407 (2014).42V. K. Kushwaha, S. Kokado, S. Kasai, Y. Miura, T. Nakatani, R. Kumara,H. Tajiri, T. Furubayashi, K. Hono, and Y. Sakuraba, “Prediction of half-metallicgap formation and Fermi level position in Co-based Heusler alloy epitaxial thinfilms through anisotropic magnetoresistance effect,” Phys. Rev. Mater. 6, 064411(2022).43R. Toyama, V. K. Kushwaha, T. T. Sasaki, Y. Iwasaki, T. Nakatani, andY. Sakuraba, “Combinatorial optimization for high spin polarization in Heusleralloy composition-spread thin films by anisotropic magnetoresistance effect,”APL Mater. 11, 101127 (2023).Journal ofApplied PhysicsARTICLE pubs.aip.org/aip/japJ. Appl. Phys. 139, 083903 (2026); doi: 10.1063/5.0304815 139, 083903-11© Author(s) 2026 20 April 2026 06:32:19https://doi.org/10.1063/1.4874851https://doi.org/10.1063/1.4874851https://doi.org/10.1103/PhysRevMaterials.6.064411https://doi.org/10.1063/5.0169124https://pubs.aip.org/aip/jap