betaFeedback

WorksPhysics and Astronomy

Magnetic correlations and Griffith-like phase in Co2_2TiSi0.5_{0.5}Al0.5_{0.5} Heusler alloy

Abstract

We present a comprehensive study aimed at elucidating the complex magnetic correlations in Co2_2TiSi0.5_{0.5}Al0.5_{0.5} Heusler alloy having the partial B2-type structure amid L21_1 cubic main phase. The thermo-magnetization measurements at 100 Oe reveal the presence of two magnetic transitions at TC1=278 K\rm_{C1}=278~K and TC2=270 K\rm_{C2}=270~K, respectively, with saturation magnetization of around 1.2 μBμ\rm_ B/f.u. at 5~K. Our magnetic field dependent studies reveal the dominance of TC1\rm_{C1} transition at lower fields (μ0H⩽0.03 Teslaμ_0\rm H \leqslant 0.03~Tesla); however, at higher fields the TC2\rm_{C2} transition becomes more pronounced. The observation of remnant magnetization above Curie temperature suggests the development of Griffiths-like phase, which is extensively analyzed through ac and dc- magnetic susceptibility (χχ) data. The evaluation of magnetocaloric potential indicates second order phase transition with notable ΔSM=ΔS_{\rm M}= 2.22 Jkg−1^{-1}K−1^{-1} at 7 Tesla. The low-field (μ0H⩽0.1 Teslaμ_0\rm H \leqslant 0.1~Tesla) magnetic entropy (ΔSMΔS_{\rm M}) curves exhibit two non-identical positive peaks. The nature and range of spin interactions near TC\rm_C were scrutinized through rigorous critical phenomenon analysis, and the values of exponents α,β,γα, β, γ and δδ to be 0.063,0.361,1.1080.063, 0.361, 1.108 and 3.9433.943, respectively, which are found to be slightly deviating from mean-field theory towards 3D Heisenberg model. Additionally, the exchange magnetic interactions are found to decay as J(r)∼r−4.6J(r) \sim r^{-4.6}. Furthermore, the density functional theory results reveal the half-metallic nature exhibiting 100% spin polarization (SP). However, the electronic and magnetic properties are greatly affected by the incorporation of structural disorder, which results in drastic reduction of SP to mere 8.3%.

[27]. The exploration of HAs as magnetocaloric materials opens new avenues for the development of eco-friendly and energy-efficient refrigeration systems, cooling units and heat pumps [28].

In this context, the Co2YZ\mathrm{Co}_{2}\mathrm{YZ} HAs are of particular interest due to their high Curie temperature TCT_{\mathrm{C}} and the variation in magnetic moment at Co site, ranging from 0.30.3 to 1 μB1\ \mu_{\mathrm{B}}, which is strongly influenced by the local environment [28, 29]. The Co2TiZ\mathrm{Co}_{2}\mathrm{TiZ} series, in particular, has garnered significant scientific interest for its remarkable properties, including half-metallicity, high spin polarization, and large anomalous Hall effect, highlighting its promising spintronic functionality [30, 31, 32, 33]. The variation in magnetic moment at the Co site is also significantly influenced by the localization of 3d3d states in constituent transition metals, which is considered as the primary reason for different ⟨LZ⟩/⟨SZ⟩\langle L_{Z}\rangle/\langle S_{Z}\rangle contributions at Co site in Co2ZrSn\mathrm{Co}_{2}\mathrm{ZrSn} and Co2TiSn\mathrm{Co}_{2}\mathrm{TiSn}, with the Co-3d3d states in Co2ZrSn\mathrm{Co}_{2}\mathrm{ZrSn} being relatively more localized [34]. Moreover, the theoretically predicted half-metallic characteristics of Co2MSn\mathrm{Co}_{2}\mathrm{MSn} (M=Ti, Zr, Hf\mathrm{M}=\mathrm{Ti},\ \mathrm{Zr},\ \mathrm{Hf}) in [35, 36] are found to be controversial, as few other reports have argued that the presence of Co-dd states near Fermi level of minority spin channel destroys the half-metallicity in some of these materials [37, 38]. The spin-resolved electronic structure analysis reveals that the majority spin channel exhibits metallic character due to Co–Ti interactions. However, direct hybridization between Co–Co second next neighbors results in a finite band gap at Fermi level for the minority spin channel [34, 39]. Furthermore, the anomalous Hall effect (AHE) in Co2TiAl\mathrm{Co}_{2}\mathrm{TiAl} revealed that an extrinsic mechanism is responsible for the profound AHE signal here, even-though the (e-m) scattering introduces a side-jump contribution, but still skew scattering is observed as the dominant mechanism. Additionally, the observation of large, non-saturating negative MR is also attributed to the significant (e-m) scattering at low fields [40]. However, to the best of our knowledge, the magnetocaloric properties of Co2TiSi(Al)\mathrm{Co}_{2}\mathrm{TiSi(Al)} system have not been extensively explored, except in [41].

Therefore, in this paper we present a detailed investigation of the structural and magnetic properties as well as electronic structure of Co2TiSi0.5Al0.5\mathrm{Co}_{2}\mathrm{TiSi}_{0.5}\mathrm{Al}_{0.5} (CTSA) Heusler alloy, utilizing the x-ray diffraction, temperature and field-dependent magnetization, and ac susceptibility measurements. Our primary objective is to unravel the origin of unique Griffiths-like phase and the underlying magnetic correlations. It is crucial to thoroughly investigate the intriguing complexities arising from substitution and structural disorder along with their profound influence on the magnetic characteristics. The synergy between our experimental investigations and density functional theory (DFT) simulations aims to bridge the gap between macroscopic observations and understanding microscopic properties of Heusler Alloys.

Experimental and Computational Details

The polycrystalline Co2TiSi0.5Al0.5\mathrm{Co}_{2}\mathrm{TiSi}_{0.5}\mathrm{Al}_{0.5} sample was prepared by arc melting technique (CENTORR vacuum, USA) using water cooled Cu hearth. The constituent elements (Co, Ti, Si, Al) from Alfa Aesar and Sigma Aldrich with purity 99.99+% taken in the stoichiometric amount were melted in flowing argon environment. To ensure homogeneity of sample, the ingot was turned over and re-melted several times. The ingot wrapped in molybdenum foil was then annealed under vacuum in quartz tube at 1123 K for 7 days in Nabotherm furnace, and we recorded the total weight loss of <1%<1\%.

The x-ray diffraction (XRD) pattern was recorded using Malvern PANalytical conventional diffractometer with Cu Kα_{\alpha} radiation (λ=1.5406\lambda=1.5406Å) to study the crystallographic structure. The morphology and elemental composition were characterized using field emission scanning electron microscope (FE-SEM) and energy dispersive x-ray (EDX) measurements (see Figure 1 in [42]). The magnetic measurements were recorded using MPMS3 SQUID magnetometer from Quantum Design, USA.

Image not extracted

Figure 1. (a) The XRD pattern for CTSA sample recorded at room temperature, with the observed data points, simulated curve, difference in observed and simulated data and the Bragg positions, denoted by open red circles, solid black line, solid blue line, and vertical green markers, respectively. The inset highlights the zoomed view of (111) and (200) peaks. (b) The crystal structure of CTSA, crystallized in Fm3ˉ\bar{3}m space group, with Co (pink), Ti (dark green), Si (orange), Al (light green) atoms occupying the (1/4, 1/4, 1/4), (1/2, 1/2, 1/2), (0,0,0) and (0,0,0) sites, respectively.

The first-principles calculations were performed using the DFT as implemented in the Vienna ab initio simulation package (VASP) [43] employing a projector augmented wave (PAW) plane wave basis [44]. For the exchange-correlation, we used Perdew, Burke and Ernzerhof (PBE) functional within the generalized-gradient-approximation framework. Further, to obtain the equilibrium structure for ground state, the unit cell was optimized using full relaxation calculation with energy convergence and atomic force tolerance used as 10−810^{-8} eV and 10−710^{-7} eV/Å, respectively. The self-consistent-field calculations were executed using a k-mesh of 12×12×1212\times12\times12 along-with integration method having 0.050.05 eV smearing and 500500 eV plane wave energy cut off.

Results and discussion

Crystal Structure and Magnetic Properties

The XRD pattern recorded at room temperature is shown in Figure 1(a) where the inset highlights an additional narrow scan of (111) and (200) peaks at 26.7∘26.7^{\circ} and 30.9∘30.9^{\circ}, respectively. The Rietveld refinement using Full-prof package is performed considering Fm3ˉm\mathrm{Fm}\bar{3}\mathrm{m} space group (no. 225) and the extracted lattice parameter is a=5.8038a=5.8038 Å, with corresponding fitting parameters as RP=2.2R_{\mathrm{P}}=2.2, Rexp=2.48R_{\mathrm{exp}}=2.48 and χ2=1.28\chi^{2}=1.28. The Bragg reflections corresponding to h+k+l=4nh+k+l=4n, are independent of the state-of-order and are called primary reflections. The supper-lattice reflections, which are order-dependent, can be classified into two sets, h,k,lh,k,l are odd or h+k+l=4n+2h+k+l=4n+2. The observation of (111) and (200) peaks [see the inset of Figure 1(a) for clarity] indicates the presence of ordered L212_{1} structure; however, the small difference between the atomic scattering factors of Co and Ti, and absorption corrections, render only a qualitative estimate of the chemical order from x-ray diffraction pattern [45, 46]. In Co2_2TiZ samples the commonly observed preferential disorder is B2-type between the Y(Ti) and Z atoms, hence in the present case, the relatively similar electro-negativities and atomic radii of Ti (1.54, 147 pm) and Al (1.61, 143 pm) atoms in contrast with Si (1.9, 111 pm), leads to favorable cross-site occupation of Ti and Al atomic positions [47]. However, the quantification of the anti-site disorder by refining the respective atomic occupancies at Wyckoff positions could not be concluded as there was hardly any improvement in the refined profiles. Therefore, we try to estimate the degree of disorder from order parameters S and α\alpha, using the relations S2=(I200/I220)exp/(I200/I220)thS^2 = (I_{200}/I_{220})_{\mathrm{exp}}/(I_{200}/I_{220})_{\mathrm{th}}, and (1−2α)2=(I111/I220)exp/(I111/I220)th(1 - 2\alpha)^2 = (I_{111}/I_{220})_{\mathrm{exp}}/(I_{111}/I_{220})_{\mathrm{th}}, respectively [45, 48]. The extracted value of α\alpha (fraction of Ti atoms occupying Al sites) from the observed XRD pattern is 0.05 indicating partial B2-type order amid L21_1 main cubic phase [45]. Note that, if the order parameter had been α=0.5\alpha=0.5, it would have symbolized complete order between Ti and Z atoms [45].

In order to investigate the magnetic phase transition and intrinsic magnetic interactions of CTSA, we measure the temperature-dependent magnetization data in zero-field cooled (ZFC) and field-cooled (FC) modes, during warming with an applied field of 100 Oe after cooling the sample in zero field and 100 Oe, respectively. Interestingly, we observe a sharp increase in the magnetic susceptibility (χ\chi) curves below 280 K and as evident in Figure 2(a), the branching of ZFC and FC data points suggests the presence of competing interactions and/or inhomogeneous ferromagnet [49]. The bifurcation however is observed to diminish with increasing magnetic field (see Figure 2 in [42]). Furthermore, the first-order temperature derivative of χ\chi, clearly shows the presence of two magnetic transitions around TC1_{\mathrm{C1}} = 278 K and TC2_{\mathrm{C2}} = 270 K, indicated by dashed lines in the inset of Figure 2(a). The isothermal magnetization curve recorded at 5 K in Figure 2(b) illustrates low coercivity (around 22 Oe) and the saturation magnetization value is found to be 1.2 μB\mu_{\mathrm{B}}/f.u., which is to some extent below the expected Slater-Pauling value of 1.5 μB\mu_{\mathrm{B}}/f.u. [2, 50]. This slight difference in the magnetic moment values could be ascribed to the presence of atomic disorders in the system [51, 52]. The significant influence of structural disorder on magnetic properties has been extensively investigated and discussed in the subsequent DFT section. The temperature-dependent susceptibility (χ\chi – T) curves extracted from the magnetization isotherms are presented

Image not extracted

Figure 2. (a) The thermo-magnetization curve recorded across the temperature range 2–350 K at 100 Oe magnetic field for ZFC (open red squares) and FC (open blue circles) protocols. The inset illustrates the temperature derivative of susceptibility versus temperature. (b) The isothermal magnetization measured at 5 K, with inset showing the zoomed view emphasizing the soft ferromagnetic character and the red dashed line symbolizes the Slater-Pauling value.

in Figure 3(a, Figure 3c) and the corresponding dχ/dTd\chi/dT curves are presented in Figure 3(b, Figure 3d), respectively, to illustrate the evolution of the transitions at TC1T_{\mathrm{C1}} and TC2T_{\mathrm{C2}} with applied magnetic field. The low-field (χ−T)(\chi- T) in Figure 3(a) (0.01–0.1 Tesla) undergo a notable change in the slope of the curves around 270 K, followed by another transition at about 280 K, which is more evidently observed in the respective dχ/dTd\chi/dT plot, Figure 3(b) where the transition at TC1T_{\mathrm{C1}} appears to be dominant upto 0.03 Tesla; however, further increase in the magnetic field leads to broadening of this feature accompanied with systematic shift of transition temperature towards higher values [see inset Figure 3(a)], hence evidencing the ferromagnetic transition. On the other hand, the TC2T_{\mathrm{C2}} transition strengthens beyond 0.04 Tesla with a slight dip in the transition temperature at 0.04 Tesla field followed by systematic increase with the field [see inset Figure 3(a)], thereby manifesting as another competing ferromagnetic interaction. The possible anti-site disorder between Ti and Al atoms, thereby allocates different exchange-coupling constants between Co–(Ti/Al)–Co sites, which might be associated with the emergence of these magnetic transitions [53]. Similar observations are reported in DyMn2_2Si2_2 where the presence of small defects in crystal structure and/or in-homogeneity resulted in the occurrence of multiple ferromagnetic/anti-ferromagnetic transitions [53].

Image not extracted

Figure 3. The temperature-dependent magnetic susceptibility curves constructed from magnetic isotherms for temperature range 250–290K corresponding to (a) H≤0.1H \le0.1 Tesla and (c) 0.1≤H≤70.1 \le H \le7 Tesla. (b, d) The first-order temperature derivative of susceptibility plotted as a function of temperature at various values of applied field, H≤0.1H \le0.1 Tesla and 0.1≤H≤70.1 \le H \le7 Tesla, respectively. The solid purple line represents the scaled ZFC recorded data corresponding to 0.01 Tesla applied field, for substantiating the analysis.

To elucidate the fundamental magnetic interactions, the inverse susceptibility (χ−1)(\chi^{-1}) is plotted as a function of temperature in Figure 4(a). The high-temperature behavior exhibits a slight curvature, often associated with the presence of short-range magnetic correlations and/or temperature-independent susceptibility (χ0)(\chi_{0}) [54]. We fit the susceptibility data in 340–350 K range with Curie Weiss (CW) law, as below:

Image not extracted

Figure 4. (a, c–e) The thermo-magnetization curves recorded across temperature range 2–350 K at 100, 500, 1000, 20000 Oe magnetic fields, χ\chi (open red squares) and χ−1\chi^{-1} (open blue circles). The solid black line represents CW fit in high temperature regime. The arrows denote the respective Griffith temperatures for respective applied field. (b) The log⁡(χ−1)\log(\chi^{-1}) versus log⁡(T−TCR)\log(T-T_{\mathrm{C}}^{R}) curve representing the Griffith phase singularity at 100 Oe along with linear fit (solid red line).

χ=C(T−θCW)(1)\chi= \frac{C}{(T-\theta_{\mathrm{CW}})} \tag*{(1)}

where CC and θCW\theta_{\mathrm{CW}} are the Curie constant, and Curie temperature, respectively. The obtained value of the effective magnetic moment (μeff=2.13 μB)(\mu_{\mathrm{eff}} = 2.13\,\mu_{\mathrm{B}}), and the positive value of θCW=304.3\theta_{\mathrm{CW}} = 304.3 K indicates the ferromagnetic-type interactions in the sample. Interestingly, the χ−1\chi^{-1} shows a downward deviation from CW fit and takes a sharp downturn before reaching the Curie temperature. This feature implicitly suggests a Griffiths-like phase (GP) transition, as observed in Fe2−x_{2-x}Mnx_xCrAl [23], where the anti-site disorder between Fe–Al atoms resulted in random allocation of different exchange coupling constants at different lattice sites. This led to the coexistence of competing magnetic correlations hence suppressing the PM ordering by forming small ferromagnetic clusters well above TCT_{\mathrm{C}} [22, 55, 56]. The deviation temperature of χ−1\chi^{-1} from CW fit, referred to as the Griffith temperature TGT_{\mathrm{G}}, marks the onset of short-range magnetic ordering in the system. Therefore, we examine the field dependence of χ−T\chi- T, which was found to be similar across different fields, as evident from Figure 4(a, Figure 4c, Figure 4d, Figure 4e). However, as the field increases above 100 Oe, the prominent downturn softens and completely vanishes at 2 Tesla, exhibiting CW behavior at high fields. Likewise, Barnabha et al. observed an identical suppression in the downturn with increasing applied field (up to 2 kOe) in CoFeVAl, due to the dominant paramagnetic contribution at higher fields [55].

Intriguingly, in the present case, the TGT_{\mathrm{G}} is found to shift towards lower temperatures from 338 K to 328 K, respectively, with increasing the field from 100 Oe to 1000 Oe, indicated with arrows in the Figure 4(a, Figure 4c, Figure 4d). These features hence align well with the characteristic signatures of Griffiths-like phase, which is observed to emerge due to the nucleation or local alignment of small ferromagnetic clusters within globally paramagnetic regime [22, 24, 57, 58]. The magnetic response of a system exhibiting GP phase is governed by the largest cluster or correlated volume [18, 59]. Therefore, examining the logarithmic dependence of inverse susceptibility on temperature scale is crucial, where the GP singularity (χ−1)(\chi^{-1}) can be characterized by the power law [59]

χ−1∝(T−TCR)1−λ,(2)\chi^{-1} \propto(T-T_{\mathrm{C}}^{\mathrm{R}})^{1-\lambda}, \tag*{(2)}

here, the TCRT_{\mathrm{C}}^{\mathrm{R}} and λ\lambda are the characteristic temperature of the random ferromagnet and exponent of the power law, respectively. The exponent λ\lambda symbolizes the deviation from CW nature and lies between 0<λ<10<\lambda<1 for GP. Meanwhile, the characteristic temperature TCRT_{\mathrm{C}}^{\mathrm{R}} plays a significant role in the aforementioned analysis and lies between the actual and the highest ordering temperature, TC<TCR<TGT_{\mathrm{C}} < T_{\mathrm{C}}^{\mathrm{R}} < T_{\mathrm{G}}, permissible by exchange bond distribution. In the present analysis, we use TCR=θCWT_{\mathrm{C}}^{\mathrm{R}} = \theta_{\mathrm{CW}}, as it ensures that equation (2) reduces to CW nature in PM region for λ∼0\lambda\sim0. The log⁡(χ−1)\log(\chi^{-1}) versus log⁡(T−TCR)\log(T-T_{\mathrm{C}}^{R}) data presented in Figure 4(b) for 100 Oe demonstrates a small linear region at high temperatures followed by a downturn and another linear region (fitted solid red line) at lower temperatures. The linear fitting of the low-temperature segment of the curve produces λ=0.867±0.027\lambda= 0.867 \pm0.027, embodying the GP regime. The large observed value of λ\lambda corresponds to a greater deviation from the CW behavior, indicating the robustness of GP in the present sample. Recently, Nag et. al. have reported a similar robust GP state with λ=0.88\lambda= 0.88, corresponding to the presence of quenched disorder and coexisting AFM/FM interactions in CrFeVGa Heusler alloy [22]. So we can conclude that the sharp downturn observed above TCT_{\mathrm{C}} in Figure 4 could be attributed to the dominant contribution from small ferromagnetically coupled clusters over PM ordering. The crystallographic disorder, here in the form of randomly distributed local B2-type phase, suppresses the long range PM ordering temperature, because of the random allocation of exchange-coupling constants

at different lattice sites. This develops an intermediate temperature region where, the FM clusters are dispersed within the paramagnetic matrix, thereby resulting in the downward deviation of χ−1\chi^{-1} from the CW fit [60].

Moreover, the acac magnetic susceptibility (χac=χac′+ιχac′′\chi_{ac} = \chi'_{ac} + \iota\chi''_{ac}) provides a powerful tool to probe the magnetization and dynamics of a system as a function of temperature, frequency, amplitude and direction of acac and bias dcdc magnetic fields. Figure 5(a) presents the data for χac′\chi'_{ac} and χac′′\chi''_{ac} recorded at 2.5 Oe field and an excitation frequency of 498 Hz. Here, the χac′\chi'_{ac} shows a frequency-independent peak at about 280 K corresponding to the magnetic phase transition. However, χac′′\chi''_{ac} exhibits a low-intensity broad hump at lower temperatures (30–90 K) followed by an asymmetric feature near 280 K, see Figure 5(a). The subsequent frequency-dependent study performed in the vicinity of 280 K reveals an enhancement in the amplitude of this feature along with clear frequency dispersion in the χac′′\chi''_{ac} component, as shown in Figure 5(b). This suggests the presence of a distinct magnetic state developed due to blocking of spins below Griffiths phase. Similarly, in Fe2−x_{2-x}Mnx_xCrAl, χac′′\chi''_{\mathrm{ac}} exhibits frequency dependence near TCT_{\mathrm{C}}, but without shift in the peak position, which was then ascribed to the irreversible domain wall movement and/or pinning effect [23]. Similar frequency dispersion with negative χac′′\chi''_{\mathrm{ac}} anomaly corresponding to the Griffiths phase has been also observed in Tb2_2CoMnO6_6 system [61]. In several other systems, anomalous relaxation processes have often been attributed to mediate such frequency dependence of the χac′′\chi''_{\mathrm{ac}} data [62, 63].

Image not extracted

Figure 5. (a) The real χac′\chi'_{\mathrm{ac}} (open red squares) and imaginary χac′′\chi''_{\mathrm{ac}} component (open blue circles) of ac susceptibility measured at 2.5 Oe ac field and 498 Hz exciting frequency. (b) The frequency dependent χac′′\chi''_{\mathrm{ac}} measured at 2.5 Oe ac field and 9, 99, 198, 498, 599, 699 Hz frequencies in the vicinity of 278 K.

Magnetocaloric Properties

We perform the magnetocaloric analysis to investigate the nexus of these complex magnetic interactions with temperature and magnetic field. The isothermal magnetization measurements were performed near TCT_{\mathrm{C}} (278 K) with a temperature interval of ΔT=1\Delta T = 1 K and the magnetic fields upto 7 Tesla, see Figure 6(a). The isothermal entropy change (ΔSM\Delta S_{\mathrm{M}}) can be determined using the following Maxwell’s equation,

Image not extracted

Figure 6. (a) The isothermal magnetization curves for temperature range 250–290 K, upto 7 Tesla applied magnetic field. (b, c) The magnetic entropy change ΔSM\Delta S_{\mathrm{M}} versus temperature plotted as a function of magnetic field (0.1–7 Tesla) and (0.01–0.1 Tesla), respectively. (d) Arrott plot M1/βM^{1/\beta} versus (H/M)1/γ(H/M)^{1/\gamma} embodying the mean-field model, is illustrated corresponding to the critical exponents β,γ\beta,\gamma as 0.5 and 1, respectively. (e) The variation of exponent “nn”, representing the scaling nature of magnetocaloric properties with applied magnetic fields as a function of temperature. (f) The normalized entropy change plotted against scaled temperature to deduce the order of magnetic phase transition.

ΔSM=μ0∫0Hmax⁡(∂M(T,H)∂T)HdH,(3)\Delta S_{\mathrm{M}} = \mu_0 \int_0^{H_{\max}} \left(\frac{\partial M(T,H)}{\partial T}\right)_H dH, \tag*{(3)}

where, the μ0\mu_0 represents permeability. Figure 6(b) depicts the obtained entropy change for magnetic fields ranging from 0.1 to 7 Tesla across the TCT_{\mathrm{C}}. All the entropy curves exhibit a distinctive peak at temperature TpT_{\mathrm{p}} and its position shifts slightly towards higher temperatures as well as the maximum entropy change increases with increasing magnetic field. For the present sample, the maximum entropy change is found to be 2.22 J.kg−1^{-1}K−1^{-1} at μ0H=7\mu_0H = 7 Tesla, which is comparable to the related Co-based systems, like 2 J.kg−1^{-1}K−1^{-1} @ 9 Tesla for the Co2_2Cr0.75_{0.75}Ti0.25_{0.25}Al [64] and 2.55 J.kg−1^{-1}K−1^{-1} @ 5 Tesla for the Co50_{50}Cr25_{25}Al25_{25} [65]. Interestingly, in the present sample for lower fields (μ0H≤0.1\mu_0H \le0.1 Tesla) the entropy curve exhibits two non-identical positive peaks associated with TC1T_{\mathrm{C1}} and TC2T_{\mathrm{C2}} magnetic transitions, as shown in Figure 6(c), which is consistent with the observations of Figure 3. The relative magnitude of maximum entropy change ΔSM\Delta S_{\mathrm{M}} in these peaks changes with applied magnetic field, and it instantiates a crossover at about 0.06 Tesla, see the inset of Figure 6(c). Therefore, the development of randomly allocated, temperature and magnetic field dependent exchange interactions in the sample, possibly act as the driving force for the occurrence of the observed magnetic transitions [53, 62].

Note that the magnetocaloric properties are strongly influenced by the nature of the magnetic transition where the materials having second-order phase transition (SOPT) are desirable for these applications due to the absence of thermal hysteresis, but they render considerably lower ΔSM\Delta S_{\mathrm{M}} and ΔTad\Delta T_{\mathrm{ad}} values. The order of magnetic phase transition can be determined through magnetic analysis, like the Banerjee criterion, which suggests that for SOPT materials the slope of the Arrott curve for isotherms below and above TCT_{\mathrm{C}} is positive [66]. Here, the Arrott plot (M2M^2 versus H/MH/M) illustrated in Figure 6(d) indicate the positive slope of the isotherms, affirming the second-order nature of the phase transition. However, this criterion is not always reliable because it assumes that the system adheres to the mean-field model, which may not be the case always. To further quantify the nature of the phase transition, we examine the scaling nature of the magnetocaloric properties by probing the magnetic field dependence of the entropy change, defined as ΔSM∝Hn\Delta S_{\mathrm{M}} \propto H^n [67, 68]. This general method is not constrained to any particular equation of state and has been verified and validated by Law et al. for a various materials, including Heusler alloys such as Ni45.7_{45.7}Mn36.6_{36.6}In13.5_{13.5}Co4.2_{4.2} [68]. Therefore, we plot the variation of exponent “nn” as a function of temperature using the following relation [69]

n(T,H)=∂ln⁡(∣ΔSM∣)∂ln⁡(H).(4)n(T,H) = \frac{\partial\ln(|\Delta S_{\mathrm{M}}|)}{\partial\ln(H)}. \tag*{(4)}

For the SOPT, the value of nn should approach 1 below TCT_{\mathrm{C}} and to paramagnetic value 2 for temperatures above TCT_{\mathrm{C}}, whereas, any value of nn greater than 2 is associated with the presence of first-order phase transition. However, before reaching the paramagnetic value, it is expected to exhibit a minimum at T=TCT = T_{\mathrm{C}} [67]. Our computed data for nn as a function of temperature are presented in Figure 6(e), which clearly shows that the nn values are approximately close to 1 for temperatures below TCT_{\mathrm{C}}, and then starts approaching a minimum at around 278 K before increasing again towards the value of 2. This minimum corresponds to the Curie transition from ferromagnetic to the paramagnetic regime, which is found to be consistent with the transition observed in the thermo-magnetization measurement [as shown in Figure 2(a)].

Another qualitative analysis for the SOPT is that after the normalization of entropy change and scaling of temperature, all the field-dependent entropy curves should collapse into a single curve. This analysis is based on the assumption that equivalent points corresponding to different experimental curves should converge to the same point on the universal curve [70]. The equivalent points, denoted as Tr1T_{\mathrm{r1}} and Tr2T_{\mathrm{r2}}, are recognized as points corresponding to the same level relative to the peak on the entropy curve. In the present work, for the analysis of ΔSM(Tr)=xΔSMmax⁡\Delta S_{\mathrm{M}}(T_{\mathrm{r}}) = x\Delta S_{\mathrm{M}}^{\max} we have taken the arbitrary factor xx as 0.8. It is to mention that, the choice of xx does not affect the analysis. The temperature is scaled into θ\theta using the following equation [71]

θ={−(T−Tpk)/(Tr1−Tpk),T≤TC(T−Tpk)/(Tr2−Tpk),T>TC,(5)\theta= \begin{cases} -(T-T_{\mathrm{pk}})/(T_{\mathrm{r1}}-T_{\mathrm{pk}}), & T \leq T_{\mathrm{C}} \\ (T-T_{\mathrm{pk}})/(T_{\mathrm{r2}}-T_{\mathrm{pk}}), & T > T_{\mathrm{C}}, \end{cases} \tag*{(5)}

where, Tr1T_{\mathrm{r1}} and Tr2T_{\mathrm{r2}} represent the temperatures corresponding to ΔSM(Tr1,r2)=0.8∗ΔSM(Tpk)\Delta S_{\mathrm{M}}(T_{\mathrm{r1,r2}})=0.8*\Delta S_{\mathrm{M}}(T_{\mathrm{pk}}) and TpkT_{\mathrm{pk}} is the temperature corresponding to ΔSMmax⁡\Delta S_{\mathrm{M}}^{\max}. Figure 6(e) shows the normalized entropy curves as a function of scaled temperatures, which was found to collapse into a single curve, further confirming the SOPT in the present sample. Note that the relative cooling power (RCP) is considered as the figure of merit for magneto-caloric materials, which is defined as the product of maximum entropy change and temperature at full width half maximum (FWHM) δTFWHM\delta T_{\mathrm{FWHM}} [72], as

RCP=∣−ΔSM∣×δTFWHM.(6)\mathrm{RCP} = |-\Delta S_{\mathrm{M}}| \times\delta T_{\mathrm{FWHM}}. \tag*{(6)}

The value of δTFWHM\delta T_{\mathrm{FWHM}} was extracted by fitting the entropy curves in Figure 6(b), using the Gaussian function. The highest value of RCP is obtained as 64.4 J.kg−1^{-1} (for δTFWHM∼29\delta T_{\mathrm{FWHM}} \sim29 K) at μ0H=7\mu_{0}H = 7 Tesla, which is found to be better than the reported value for Co2TiSi0.75Al0.25\mathrm{Co}_{2}\mathrm{TiSi}_{0.75}\mathrm{Al}_{0.25}, 55.25 J.kg−1^{-1} at μ0H=3\mu_{0}H = 3 Tesla [41].

Critical Analysis

To elucidate the microscopic correlations responsible for the intricate magnetic interactions in the present sample Co2TiSi0.5Al0.5\mathrm{Co}_{2}\mathrm{TiSi}_{0.5}\mathrm{Al}_{0.5}, we now extensively examine the static critical behavior across the FM/PM transition around TC=278T_{\mathrm{C}} = 278 K. According to the Landau mean-field theory, the M2M^{2} vs H/MH/M curves across the TCT_{\mathrm{C}} should constitute of parallel straight lines in high-field region, with T=TCT = T_{\mathrm{C}} isotherm passing through the origin. However, as shown in Figure 6(d), the Arrott curves are not stringently linear in high-field region, indicating that the system cannot be characterized by mean-field model. Therefore, we use the modified Arrott plot (MAP) approach to characterize the phase transition, i.e., the magnetization isotherms are analyzed using the equation of state proposed by Arrott-Noakes [73]:

(H/M)1/γ=aϵ+bM1/β,(7)(H/M)^{1/\gamma} = a\epsilon+ bM^{1/\beta}, \tag*{(7)}

where, aa and bb are temperature-dependent parameters. Near the TCT_{\mathrm{C}}, the spin-spin correlation length (ξ\xi), diverges as ξ=ξ0∣(T−TC)/TC∣−ν\xi= \xi_{0}|(T - T_{\mathrm{C}})/T_{\mathrm{C}}|^{-\nu}, leading to formulation of universal scaling laws governed by asymptotic critical exponents and amplitudes, as [74]

MS(0,T)=M0(−ϵ)βfor ϵ<0, T<TC(8)M_{\mathrm{S}}(0,T) = M_{0}(-\epsilon)^{\beta} \quad\text{for } \epsilon< 0,\ T < T_{\mathrm{C}} \tag*{(8)}
χ0−1(0,T)=h0M0(ϵ)γfor ϵ>0, T>TC(9)\chi_{0}^{-1}(0,T) = \frac{h_{0}}{M_{0}}(\epsilon)^{\gamma} \quad\text{for } \epsilon> 0,\ T > T_{\mathrm{C}} \tag*{(9)}
M(H,TC)=DH1/δfor ϵ=0, T=TC(10)M(H,T_{\mathrm{C}}) = DH^{1/\delta} \quad\text{for } \epsilon= 0,\ T = T_{\mathrm{C}} \tag*{(10)}

here, MSM_{\mathrm{S}}, χ0−1\chi_{0}^{-1} and ϵ=(T−TC)/TC\epsilon= (T - T_{\mathrm{C}})/T_{\mathrm{C}} represent the spontaneous magnetization below TCT_{\mathrm{C}}, initial inverse susceptibility above TCT_{\mathrm{C}}, and reduced temperature, respectively, while the M0M_{0}, h0/M0h_{0}/M_{0}, and DD denote the critical amplitudes. The MAPs [M1/βM^{1/\beta} vs. (H/M)1/γ(H/M)^{1/\gamma}] corresponding to critical exponents predicted for the tricritical (β=0.25\beta= 0.25 and γ=1\gamma= 1), 3D Heisenberg (β=0.365\beta= 0.365 and γ=1.386\gamma= 1.386), mean-field (β=0.5\beta= 0.5 and γ=1\gamma= 1), and 3D Ising (β=0.325\beta= 0.325 and γ=1.24\gamma= 1.24) models [75], are constructed and illustrated in Figure 3(a–Figure 3d) of [42], respectively. The high magnetic field region (H≥2H \geq2 Tesla) was then linearly fitted (indicated by the red lines) and we observed that the universality classes exhibit quasi-straight lines over the wide range of fields. To determine which model produces the best fit, we plot the normalized slope (NS)=S(T)/S(TC)S(T)/S(T_{\mathrm{C}}), where the slope at any given temperature has been normalized to the slope at TC=278T_{\mathrm{C}} = 278 K. Figure 7(a) displays the NS extracted from the corresponding MAPs of Figure 3(a–Figure 3d) of [42]. The sample shows least deviation from unity for the mean-field and 3D Heisenberg models as we move away from the critical point (TCT_{\mathrm{C}}), indicating that either of these models could successfully characterize the magnetic interactions.

Image not extracted

Figure 7. (a) The normalized slopes (NS) for different universality classes in the critical region for Co2_2TiSi0.5_{0.5}Al0.5_{0.5}. (b) The modified Arrott plot using the correct critical exponents for the sample, where blue circles, red lines, and black line represent the experimental data points, linear fit for H≥2H \ge2 Tesla, and the extrapolation of the linear fit at TCT_C down to H=0H = 0, respectively. (c) The temperature-dependent spontaneous magnetization MSM_S (left axis) and inverse initial susceptibility χ0−1\chi_0^{-1} (right axis), where the black solid curves represent the best fit using equations 8 and 9, respectively. (d) Temperature-dependent MS(T)/[dMS(T)/dT]M_S(T)/[dM_S(T)/dT] (left axis) and χ0−1/dχ0−1/dT\chi_0^{-1}/d\chi_0^{-1}/dT (right axis) plots, where the black solid lines represent the linear fit using equations 11 and 12, respectively. (e) The MM-HH curves near TCT_C [278.5(1) K] at 278 and 279 K, inset shows the log-log plot, where solid black lines represent the linear fit for H≥2H \ge2 Tesla. (f) The reduced magnetization mm versus reduced field hh plots, insets show the log-log plot of the same, and (g) the m2m^2 versus h/mh/m plot.

Therefore, we employed an iterative approach to estimate the precise values of the critical exponents β\beta and γ\gamma [76, 77]. The linear extrapolation of high-field region gives MS(T)M_{\mathrm{S}}(T) and χ0−1(T)\chi_{0}^{-1}(T) as intercepts on M1/βM^{1/\beta} and (H/M)1/γ(H/M)^{1/\gamma} axes, respectively. By substituting these resultants in equations 8 and 9, we obtain new set of β\beta and γ\gamma values, which are then used to construct new MAPs. This process is repeated until we are able to find the stable values of β\beta and γ\gamma within the error bars. It is imperative to note that these values of the critical exponents are independent of their initial trial values, thereby confirming their reliability. After rigorous iterations, a set of parallel straight lines were generated corresponding to β\beta, γ\gamma, and TCT_{\mathrm{C}} as 0.369(1), 1.095(8), and 278.5(1) K, respectively, and presented in Figure 7(b). The isotherm at 278 K passes through the origin [see the black line in Figure 7(b)], thus representing the TCT_{\mathrm{C}}. However, a slight curvature in the low magnetic field region (not shown here) could be attributed to the complex multi-domain spin structure in this regime [64]. The temperature dependence of final MS(T)M_{\mathrm{S}}(T) and χ0−1(T)\chi_{0}^{-1}(T) values acquired from the MAP is plotted in Figure 7(c). To further substantiate the critical exponents and TCT_{\mathrm{C}}, we investigate the MS(T)M_{\mathrm{S}}(T) and χ0−1(T)\chi_{0}^{-1}(T) data using Kouvel-Fisher (KF) plot [74],

MS(T)dMS(T)/dT=(T−TC)/β(11)\frac{M_{\mathrm{S}}(T)}{dM_{\mathrm{S}}(T)/dT} = (T - T_{\mathrm{C}})/\beta \tag*{(11)}
χ0−1dχ0−1/dT=(T−TC)/γ(12)\frac{\chi_{0}^{-1}}{d\chi_{0}^{-1}/dT} = (T - T_{\mathrm{C}})/\gamma \tag*{(12)}

According to the KF method, the MS(T)/[dMS(T)/dT]M_{\mathrm{S}}(T)/[dM_{\mathrm{S}}(T)/dT] and χ0−1/[dχ0−1/dT]\chi_{0}^{-1}/[d\chi_{0}^{-1}/dT] yield linear temperature dependence with slopes of 1/β1/\beta and 1/γ1/\gamma, respectively, as presented in Figure 7(d). The obtained values of β\beta, γ\gamma, and TCT_{\mathrm{C}}, from the slopes and y-intercepts of linear fitted curves, are 0.361(5), 1.108(6), and 278.5(1) K, respectively, which are consistent with those generated from the MAPs, hence validating their accuracy. Furthermore, the critical isotherm at TCT_{\mathrm{C}} is characterized by the exponent δ\delta. It exhibits a linear behavior on a log-log scale according to equation 10 with slope 1/δ1/\delta. To verify the value of TCT_{\mathrm{C}} [278.5(1) K] extracted from the MAP and KF methods, we performed the critical isotherm analysis at the magnetic ordering temperature in Figure 7(e). The isotherms in asymptotic region were recorded in 1 K steps, therefore, the M-H curves corresponding to 278 K and 279 K are presented in Figure 7(e). The linear fitting in the inset of Figure 7(e) gives an average δ\delta=3.943(4). Similar analysis to extract the critical parameters has been reported for few Co-based Heusler alloys, as listed in Table I. The critical exponent δ\delta can also be calculated using the Widom scaling relation [78] as:

δ=1+γβ(13)\delta= 1 + \frac{\gamma}{\beta} \tag*{(13)}

here, considering the β\beta and γ\gamma values determined from the Figure 7(b), the calculated value of δ\delta using equation 13 is 3.967(8), which shows good agreement with the critical isotherm analysis. Therefore, it can be concluded that while the extracted critical exponents are consistent across different analytical approaches, they do not adhere to any conventional universality class. Therefore, it is crucial to scrutinize whether the obtained critical exponents can generate the scaling magnetic equation of state [79]:

M(H,ϵ)=∣ϵ∣βf±(H∣ϵ∣β+γ)(14)M(H,\epsilon)=|\epsilon|^\beta f_{\pm}\left(\frac{H}{|\epsilon|^{\beta+\gamma}}\right) \tag*{(14)}

where f+f_+ and f−f_- denote the regular functions above and below TCT_C, respectively. For the correct values of critical exponents, the reduced magnetization m=M∣ϵ∣−βm=M|\epsilon|^{-\beta} versus reduced field h=H∣ϵ∣−(β+γ)h=H|\epsilon|^{-(\beta+\gamma)} curves should collapse into two branches, corresponding to temperatures above and below the TCT_C. Figure 7(f) illustrates the mm versus hh plot using the exponents obtained from MAP analysis, showing convergence into two universal curves. This branching is more profoundly observed in the log-log plot [inset of Figure 7(f)]. Furthermore, the m2m^2 versus h/mh/m plot in Figure 7(g), also exhibits similar branching in the data points, hence indicating that the magnetic interactions are properly re-normalized in the critical regime following a scaling equation of state [80].

Furthermore, the nature and range of complex spin interactions play a pivotal role in determining the magnetic properties of a system, which necessitates their investigation. According to the renormalization group theory of exchange-interaction systems, for a dd-dimensional spin system with an isotropic nn-component order parameter, the isotropic interaction decays as J(r)∼r−(d+σ)J(r)\sim r^{-(d+\sigma)}, where rr, dd, and σ\sigma represent the spatial distance, lattice dimensionality, and range of magnetic interactions, respectively. This model classifies the magnetic interaction as long-ranged for σ≤1.5\sigma\le1.5, and short-ranged for σ≥2\sigma\ge2 [81]. The value of σ\sigma can hence be calculated by substituting γ\gamma in the following relation [82, 83, 84]

Sample/modelReferenceApproachα\alphaβ\betaγ\gammaδ\delta
Co2_2TiSi0.5_{0.5}Al0.5_{0.5}This workModified Arrott plot0.369(1)1.095(8)3.967
Kouvel-Fisher0.361(5)1.108(6)4.069
Critical Isotherm3.943(4)
Co2_2TiSi0.75_{0.75}Al0.25_{0.25}[41]Modified Arrott plot0.429(5)1.211(5)3.82(6)
Kouvel-Fisher0.469(1)1.173(1)3.501(4)
Critical Isotherm3.100(2)
Co50_{50}Cr25_{25}Al25_{25}[65]Modified Arrott plot0.488(3)1.144(4)3.336(5)
Kouvel-Fisher0.482(13)1.148(16)3.382(20)
Critical Isotherm3.401(4)
Co2_2Cr0.75_{0.75}Ti0.25_{0.25}Al[64]Modified Arrott plot0.496(4)1.348(12)3.82(6)
Kouvel-Fisher0.491(9)1.327(13)
Critical Isotherm3.581(5)
Co2_2TiGe[85]Modified Arrott plot0.4951.3253.677
Kouvel-Fisher0.495(2)1.324(4)3.675
Critical Isotherm3.671(1)
Co2_2TiSn[86]Modified Arrott plot0.527(3)1.229(2)3.33(2)
Kouvel-Fisher0.537(2)1.255(3)3.33(7)
Critical Isotherm3.261(2)
Co2_2HfSn[87]Modified Arrott plot0.472(3)1.021(2)3.16(2)
Kouvel-Fisher0.477(4)1.029(5)3.15(7)
Critical Isotherm3.273(5)
Mean-field[75]Theoretical00.51.03
3D Heisenberg[75]Theoretical-0.1150.3651.3864.80
3D Ising[75]Theoretical0.110.3251.2414.82
Tricritical mean-field[88]Theoretical0.251.05

Table 1. Critical exponents (α\alpha, β\beta, γ\gamma and δ\delta) determined using modified Arrott plot, Kouvel-Fisher approach and critical isotherm analysis for Co2_2TiSi0.5_{0.5}Al0.5_{0.5} and other systems.

γ=1+4d(n+2n+8)Δσ+8(n+2)(n−4)d2(n+8)2{1+2G(d2)(7n+20)(n−4)(n+8)}Δσ2(15)\gamma= 1 + \frac{4}{d}\left(\frac{n+2}{n+8}\right)\Delta\sigma+ \frac{8(n+2)(n-4)}{d^2(n+8)^2}\left\{1 + \frac{2G\left(\frac{d}{2}\right)(7n+20)}{(n-4)(n+8)}\right\}\Delta\sigma^2 \tag*{(15)}

where Δσ=(σ−d2)\Delta\sigma= \left(\sigma- \frac{d}{2}\right), G(d2)=3−14(d2)2G\left(\frac{d}{2}\right) = 3 - \frac{1}{4}\left(\frac{d}{2}\right)^2, and nn represents the spin dimensionality. By considering the three-dimensional lattice and spin interactions (n=d=3n = d = 3) and using the values of γ\gamma extracted from the MAP, we find σ=1.696\sigma= 1.696, which falls between the mean-field theory and 3D Heisenberg model range, indicating extended long-range magnetic interactions [81, 86]. The extracted values of σ\sigma can further be used to calculate the critical exponent α\alpha using the relation α=2−νd\alpha= 2 - \nu d, where ν\nu is the exponent of correlation length, given as ν=γ/σ\nu= \gamma/\sigma. On the other hand, the calculated value of α\alpha (0.063) suggests the predominance of long range interactions (mean-field theory, i.e., α→0\alpha\to0) in the present sample. The magnetic interactions in this sample decay approximately as J(r)∼r−4.6J(r) \sim r^{-4.6}, suggesting that they do not completely adhere to any theoretical model, but lies within long range (mean-field model) and short range (3D Heisenberg model) FM interactions [28, 86].

berg model) FM interactions [28, 86].

First-principles simulations

Finally, to support our experimental results we perform DFT based first-principles calculations to investigate the electronic structure and magnetic properties. The properties in a solid are strongly governed by its crystal structure and electronic bonding [89, 90]. Therefore, to verify the accuracy of our DFT parameters, we first performed electronic structure calculations for pristine Co2_2TiSi (CTS) system, presented in Figure 3 of [42]. The calculations predict half-metallic character, with 100% spin polarization (SP) and a band gap of 0.60 eV associated with minority spin channel, concurring with reported literature [30, 31]. The majority spin channel exhibits metallic character due to the Co–Ti interactions, while the Co–Co second next neighbor direct hybridization is responsible for semiconducting nature of minority spin channel [34, 39]. The computed lattice parameter is 5.7475 Å, which agrees well with the calculated value of 5.760 Å in [30] and experimentally observed value of 5.743 Å in [91]. Additionally, as evident from the Table Table II, the computed magnetic moment, 1.99 μB\mu_{\mathrm{B}}/f.u., is in very good agreement with refs. [31, 35], which substantiates the exchange-correlation functional and other parameters used in the calculations.

SystemmCom^{\mathrm{Co}} (μB\mu_{\mathrm{B}})mTim^{\mathrm{Ti}} (μB\mu_{\mathrm{B}})mSim^{\mathrm{Si}} (μB\mu_{\mathrm{B}})mAlm^{\mathrm{Al}} (μB\mu_{\mathrm{B}})mtotalm^{\mathrm{total}} (μB\mu_{\mathrm{B}})
Co2_2TiSi0.99-0.040.03-1.99
Co2_2TiSi [30]1.03-0.02--2.00
Co2_2TiSi [31]0.990.030.01-1.99
Co2_2TiSi0.5_{0.5}Al0.5_{0.5}0.822-0.0970.0110.0011.559
Co2_2TiSi0.5_{0.5}Al0.5_{0.5}(D)0.580-0.0120.008-0.0031.160
Co4_4Ti2_2SiAl [33]0.844-0.1110.015-0.013

Table II. Total magnetic moment and atom resolved magnetic moment per formula unit.

Further, using the optimized unit cell of pristine CTS sample, the Co2_2TiSi0.5_{0.5}Al0.5_{0.5} system was constructed by creating a 2×2×22 \times2 \times2 supercell followed by substituting 50% Si by the Al atoms, see Figure 1(b). The substituted structure is further optimized to obtain the global minimum state, where the optimized lattice parameters are obtained as a=b=5.7802a=b=5.7802 Å and c=5.8003c=5.8003 Å, which are consistent with the value of 5.8047 Å in ref. [33]; however, with a slight distortion along the ‘cc’ axis. The calculated magnetic moment 1.56 μB\mu_{\mathrm{B}}/f.u. is found to be close to the Slater-Pauling value, 1.5 μB\mu_{\mathrm{B}}. Also, the calculated total and atom resolved magnetic moments of present sample are in very good agreement with reported values in [33], as summarized in Table II. However, there is a significant difference from the experimentally observed magnetic moment value, 1.2 μB\mu_{\mathrm{B}}/f.u. in Figure 2(b), which could possibly be due to the presence of chemical disorder between the constituents Y and Z elements [32]. Therefore, we examine the impact of anti-site disorder through swapping of sites between Ti and Al atoms by constructing all the possible structures corresponding to a particular disorder level followed by selection of the most energetically favorable structure for subsequent calculations. Interesting, we find a systematic decrease in the total magnetic moment with increasing level of disorder, as summarized in Table I of [42], owing to the corresponding change in electronic structure. The system corresponding to 37.5% disorder-level Co2_2TiSi0.5_{0.5}Al0.5_{0.5}(D), renders the calculated total magnetic moment of 1.16 μB\mu_{\mathrm{B}}/f.u., which is in excellent agreement with the experimental value (1.2 μB\mu_{\mathrm{B}}/f.u.). Furthermore, using the theoretical value of total magnetic moment in equation TC=23+181∗mtotalT_{\mathrm{C}}=23+181*m_{\mathrm{total}} [92], the extracted TCT_{\mathrm{C}} value is 233 K, which underestimates the experimental value of 278 K. This can be attributed to the empirical nature of this relation, as the structural disorder significantly alters the electronic structure and hence the magnetic correlations in the system. As evident in Table II, the atom-resolved calculations reveal that Co atoms have the dominant magnetic moment of 0.58 μB\mu_{\mathrm{B}}/f.u., followed by a considerable antiparallel contribution of -0.012 μB\mu_{\mathrm{B}}/f.u. from Ti, and small contributions of ≈0.008\approx0.008 and -0.003 μB\mu_{\mathrm{B}}/f.u. from Si and Al atoms, respectively. The hybridization of Co-dd states with Ti-dd states, hence induces a small magnetic moment at Ti site, even though in the opposite direction.

The electronic structure of Co2_2TiSi0.5_{0.5}Al0.5_{0.5} sample is presented in Figure 8 showing both atom-projected electronic band structure and density of states (DOS) for CTSA (top panel) and with disorder CTSA(D) (bottom panel). The majority spin channel shows ‘metallic’ characteristics, whereas there is a finite band gap of 0.40 eV in the minority spin states confirming the persistence of half-metallic character, as discernible from the panels (a) and (b), on Al-substitution. From the PDOS for majority spin channel, panel (c), the valence and conduction bands are densely populated with Co-dd and Ti-dd electrons, with nonzero contributions from spin-less Si and Al. However, for the minority channel, there are no states in the vicinity of the Fermi level in the valence band, whereas dominant contributions from Co-dd and Ti-dd electrons are observed for the conduction band. This is in contrast with a previous study, where around 50% Fe doping in Co2_2TiSi was reported to exhibit loss of half-metallicity owing to the emergence of minority states at the Fermi level [92]. The spin polarization calculated from the total density of states for the Co2_2TiSi0.5_{0.5}Al0.5_{0.5} sample, panel (c), renders the value 100%. This implies that upon Al substitution, even-though the band splitting in minority spin channel reduces from 0.60 eV to 0.40 eV; the half-metallic nature is still preserved. On the other hand, when the Ti-Al disorder is introduced, we observe a significant decrease in the SP value from 100% for CTSA to 8.3% for CTSA(D), symbolizing the loss of half-metallicity from the calculated bands and DOS, as shown in panels (d, e, f). The plausible cause of this drastic reduction of SP and loss of half-metallic nature could be associated to the emergence of Co-dd states in the vicinity of Fermi level for minority channel, in the disordered system. Here, we also find that, similar to the undistorted structure, the valence and conduction bands of majority states have dominant contributions from Co-dd and Ti-dd electrons. However, the minority channel are now composed of dominating Co-dd electrons as well as some nonzero Ti-dd contribution at the Fermi level. This study therefore highlights the profound influence of structural disorder on the electronic structure of the system, which could hence be used to tune the magnetic and transport properties of prospective materials for future spintronic applications [93].

Image not extracted

Figure 8. (a, b) The atom-resolved electronic band structure of Co2_2TiSi0.5_{0.5}Al0.5_{0.5} (CTSA), near Fermi level for minority and majority spin states, respectively. (c) The total and partial density of states of majority and minority spin channels are shown corresponding to near Fermi level region for the CTSA system. (d, e) The atom-resolved electronic band structure of with disorder, i.e., CTSA(D) system, near Fermi level for minority and majority spin states, respectively. (f) The total and partial density of states of majority and minority spin channels in CTSA(D) system.

Conclusions

In conclusion, our comprehensive study of Co2_2TiSi0.5_{0.5}Al0.5_{0.5} system reveals a complex interplay between structural disorder and observed magnetic properties. Through a combination of experimental measurements and theoretical simulations, we have demonstrated that the atomic disorder significantly impacts the underlying magnetic interactions, subsequently leading to the emergence of a magnetically in-homogeneous Griffiths-like phase. The Ti-Al anti-site disorder results in random allocation of different exchange-coupling constants between Co-(Ti/Al)-Co atoms, hence manifesting multiple magnetic transitions in the system. The profound field and temperature dependence of these magnetic correlations was observed in the lower magnetic fields (H≤0.1H \le0.1 Tesla) entropy curves, exhibiting multiple peaks corresponding to these transitions. Furthermore, the calculated values of critical exponents α\alpha, β\beta, γ\gamma and δ\delta are 0.063, 0.361, 1.108 and 3.943, respectively, which are slightly away from mean-field theory and deviating towards Heisenberg model. However, the corresponding magnetic interactions decay approximately as J(r)∼r−4.6J(r) \sim r^{-4.6}, reinforcing the notion of extended long-range magnetic ordering in the sample. The persistence of half-metallicity with Al substitution, albeit disrupted by structural disorder, underscores the role of electronic structure in determining the magnetic behavior of the

structure in determining the magnetic behavior of the system. The disorder significantly alters the electronic structure of the system, evident from the reduced magnetic moment 1.16 μB\mu_{\mathrm{B}}/f.u and spin polarization 8.3%, due to emergence of finite Co-d states at Fermi level in the minority spin channel. Our findings therefore contribute to a deeper understanding of the magnetic interactions in Co-based Heusler alloys, paving the way for future explorations of their practical applications in spintronics and magnetic refrigerators.

Acknowledgments

PY thanks MHRD for the fellowship and expresses sincere gratitude towards Dr. Ajay Kumar, Dr. Zeeshan and Madhav for useful discussions. We acknowledge the Department of Physics, IIT Delhi for providing XRD and MPMS facilities and central research facility (CRF) for FE-SEM measurements. The theoretical results presented in this work have been computed using the High Performance Computing cluster Padum, at IIT Delhi. RSD acknowledges the BRNS for financial support through DAE Young Scientist Research Award with Project Sanction No. 34/20/12/2015/BRNS..

References

References

  1. [2]I. Galanakis, P. H. Dederichs, and N. Papanikolaou, Slater-Pauling behavior and origin of the half-metallicity of the full-Heusler alloys, Phys. Rev. B 66, 174429 (2002).
  2. [18]R.B. Griffiths, Nonanalytic behavior above the critical point in a random Ising ferromagnet, Phys. Rev. Lett. 23, 17–19 (1969).
  3. [22]J. Nag, P. C. Sreeparvathy, R. Venkatesh, P. D. Babu, K. G. Suresh, and A. Alam, Nontrivial Topological Features and Griffith’s-Phase Behavior in CrFeVGa Probed by Experiment and Theory, Phys. Rev. Appl. 19, 044071 (2023).
  4. [23]K. Yadav, M. K. Sharma, S. Singh, and K. Mukherjee, Exotic Magnetic Behaviour and Evidence of cluster glass and Griffiths like phase in Heusler alloys Fe₂−ₓMnₓCrAl (0 ≤ x ≤ 1), Sci. Rep. 9, 15888 (2019).
  5. [24]B. Bandyopadhyay, J. Nag, S. Paul, M. K. Chattopadhyay, A. Lakhani, A. Alam, and K. G. Suresh, Griffiths phase-like behavior in nearly half-metallic CoFeVAl: theory and experiment, Phys. Rev. B 109, 094433 (2024).
  6. [27]J. Liu, T. Gottschall, K. P. Skokov, J. D. Moore, and O. Gutfleisch, Giant magnetocaloric effect driven by structural transitions, Nat. Mater. 11, 620 (2012).
  7. [28]P. Nehla, V. K. Anand, B. Klemke, B. Lake, and R. S. Dhaka, Magnetocaloric properties and critical behavior of Co₂Cr₁−ₓMnₓAl Heusler alloys, J. Appl. Phys. 126, 203903 (2019).DOI
  8. [29]G. D. Gupt, and R. S. Dhaka, Structural, magnetic and transport properties of Co₂CrAl epitaxial thin films, J. Vac. Sci. Technol. A, 41, 023402 (2023).arxiv.org/abs/2208.04687
  9. [30]H. C. Kandpal, G. H. Fecher, and C. Felser, Calculated electronic and magnetic properties of the half-metallic, transition metal based Heusler compounds, J. Phys. D: Appl. Phys. 40, 1507 (2007).DOI
  10. [31]V. Sharma, A. K. Solanki, and A. Kashyap, Electronic, magnetic and transport properties of Co₂TiZ (Z=Si, Ge and Sn): a first-principle study, J. Magn. Magn. Mater. 322, 2922 (2010).DOI
  11. [32]A. Jezierski, Magnetic properties of partially ordered Co₂−ₓPtₓTiSn alloys, J. of Magn. Magn. Mater. 164, 381 (1996).DOI
  12. [33]S. M. Zareii, H. Arabi, and R. Sarhaddi, Effect of Si substitution on electronic structure and magnetic properties of Heusler compounds Co₂TiAl₁−ₓSiₓ, Physica B: Condensed Matter 407, 3339 (2012).DOI
  13. [34]M. Okutani and T. Jo, Orbital magnetic moment in superlattices of transition metals, J. Phys. Soc. Jpn. 69, 598 (2000).DOI
  14. [35]H. C. Kandpal, V. Ksenofontov, M. Wojcik, R. Seshadri, and C. Felser, Electronic structure, magnetism and disorder in the Heusler compound Co₂TiSn, J. Phys. D Appl. Phys. 40, 1587 (2007).DOI
  15. [36]H. C. Kandpal, G. H. Fecher, and C. Felser, Calculated electronic and magnetic properties of the half-metallic, transition metal based Heusler compounds, J. Phys. D Appl. Phys. 40, 1507 (2007).DOI
  16. [37]A. Ślebarski, A. Jezierski, S. Lütkehoff, and M. Neumann, Electronic structure of X₂ZrSn- and XZrSn- type Heusler alloys with X = Co or Ni, Phys. Rev. B 57, 6408 (1998).DOI
  17. [38]A. Yamasaki, S. Imada, R. Arai, H. Utsunomiya, S. Suga, T. Muro, Y. Saitoh, T. Kanomata, and S. Ishida, Orbital angular momentum and interpretation of core-absorption magnetic circular dichroism on the band picture in Co-based Heusler alloys Co₂YSn (Y = Ti, Zr, and Nb), Phys. Rev. B 65, 104410 (2002).DOI
  18. [39]A. Aguayo and G. Murrieta, Density functional study of the half-metallic ferromagnetism in Co-based Heusler alloys Co₂MnSn (M = Ti, Zr, Hf) using LSDA and GGA, J. Magn. Magn. Mater. 323, 3013 (2011).DOI
  19. [40]R. P. Jena, D. Kumar, and A. Lakhani, Scaling analysis of anomalous Hall resistivity in the Co₂TiAl Heusler alloy, J. Phys. Condens. Matter 32, 365703 (2020).DOI
  20. [41]S. Datta, S. Guha, S. K. Panda, and M. Kar, Correlation between critical behavior and magnetocaloric effect near paramagnetic to ferromagnetic phase transition of Co₂TiAl₀.₇₅Si₀.₂₅ Heusler alloy, Phys. Status Solidi (b) 257, 2000123 (2020).DOI
  21. [42]see supplementary information for more details of analysis and calculations.
  22. [43]G. Kresse and J. Furthmüller, efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B 54, 11169 (1996).pubmed.ncbi.nlm.nih.gov/9984901
  23. [44]G. Kresse and D. Joubert, From Ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B 59, 1758 (1999).DOI
  24. [45]P. J. Webster, K. R. A. Ziebeck, Magnetic and chemical order in Heusler alloys containing Cobalt and Titanium, J. Phys. Chem. Solids, 34, 1647-1654 (1973).DOI
  25. [46]G. D. Gupt, Y. Kareri, J. Hester, C. Ulrich, and R. S. Dhaka, Unveiling the magnetic structure and phase transition of Cr₂CoAl using neutron diffraction, J. Appl. Phys., 134, 113901 (2023).arxiv.org/abs/2303.11869
  26. [47]C. Felser, and A. Hirohata, Heusler Alloys: Properties, Growth, Applications, Springer Series of Materials Science Vol. 222, Springer International Publishing, Switzerland, (2016).
  27. [48]G. B. Johnston and E. O. Hall, Studies on the Heusler alloys—I. Cu₂MnAl and associated structures, J. Phys. Chem. Solids 29, 193 (1968).
  28. [49]A. Ślebarski and J. Goraus, Electronic Structure and Thermodynamic Properties of Fe₂VGa, Phys. Rev. B 80, 235121 (2009).DOI
  29. [50]G. H. Fecher, H. C. Kandpal, S. Wurmehl, C. Felser, and G. Schönhense, Slater-Pauling rule and Curie temperature of Co₂-Based Heusler compounds, J. Appl. Phys. 99, 08J106 (2006).DOI
  30. [51]V. V. Sokolovskiy, V. D. Buchelnikov, M. A. Zagrebin, P. Entel, S. Sahoo, and M. Ogura, First-principles investigation of chemical and structural disorder in magnetic Ni₂Mn₁+ₓSn₁−ₓ Heusler alloys, Phys. Rev. B 86, 134418 (2012).DOI
  31. [52]M. Baral, M. K. Chattopadhyay, R. Jangir, A. Chakrabarti, and T. Ganguli, Presence of atomic disorder and its effect on the magnetic and electronic properties of NiCrGa half Heusler alloy, J. Magn. Magn. Mater. 475, 675 (2019).DOI
  32. [53]D. C. Dos Reis, E. L. T. França, V. G. De Paula, A. O. Dos Santos, A. A. Coelho, L. P. Cardoso, and L. M. Da Silva, Magnetic and Magnetocaloric Properties of DyMn₂Si₂ Compound with Multiple Magnetic Phase Transition, J. Magn. Magn. Mat. 424, 84 (2017).
  33. [54]S. Mugraneza and A. M. Hallas, Tutorial: a beginner’s guide to interpreting magnetic susceptibility data with the Curie-Weiss law, Commun. Phys. 5, 95 (2022).DOI
  34. [55]B. Bandyopadhyay, J. Nag, S. Paul, M. K. Chattopadhyay, A. Lakhani, A. Alam, and K. G. Suresh, Griffiths phase–like behavior in nearly half-metallic CoFeVAl: theory and experiment, Phys. Rev. B 109, 094433 (2024).
  35. [56]S. Chatterjee, S. Giri, S. Majumdar, P. Dutta, P. Singha, and A. Banerjee, Observation of Griffiths-like phase in the quaternary Heusler compound NiFeTiSn, J. Phys. Condens. Matter 34, 295803 (2023).DOI
  36. [57]A. Slebarski, M. Fijalkowski, J. Deniszczyk, M. M. Maska, and D. Kaczorowski, Off-stoichiometric effect on magnetic and electron transport properties of Fe₂₄VAl₁.₃₅ and Ni₂VAl: a comparative study, Phys. Rev. B 109, 165105 (2024).
  37. [58]M. Ulrich, J. García-Otero, J. Rivas, and A. Bunde, Slow relaxation in ferromagnetic nanoparticles: indication of spin-glass behavior, Phys. Rev. B 67, 024416 (2003).DOI
  38. [59]W. Jiang, X. Zhou, G. Williams, Y. Mukovskii, and R. Privzentsev, The evolution of Griffiths-phase-like features and colossal magnetoresistance in La₁−ₓCaₓMnO₃ across the compositional metal–insulator boundary, J. Phys. Condens. Matter 21, 415603 (2009).
  39. [60]E. V. Sampathkumaran, N. Mohapatra, S. Rayaprol, and K. K. Iyer, Magnetic Anomalies in the Spin-Chain Compound Sr₃CuRhO₆: Griffiths-Phase-like Behavior of Magnetic Susceptibility, Phys. Rev. B 75, 052412 (2007).
  40. [61]K. Anand, M. Alam, A. Pal, P. Singh, S. Kumari, A. G. Joshi, A. Das, A. Mohan, and S. Chatterjee, Existence of Griffiths Phase and unusual spin dynamics in double perovskite Tb₂CoMnO₆, J. Magn. Magn. Mater. 528, 167697 (2021).DOI
  41. [62]E. M. Levin, V. K. Pecharsky, and K. A. Gschneidner, Unusual magnetic behavior in Gd₅(Si₁.₅Ge₂.₅) and Gd₅(Si₂Ge₂), Phys. Rev. B 62, R14625 (2000).DOI
  42. [63]M. Földeáki, G. Higelin, L. Köszegi, B. Schwendenmann, and F. Walz, The behaviour of the imaginary component of the ac magnetic susceptibility in disaccommodation experiments, J. Magn. Magn. Mater. 46, 350 (1985).DOI
  43. [64]P. Nehla, Y. Kareri, G. D. Gupt, J. Hester, P. D. Babu, C. Ulrich, and R. S. Dhaka, Neutron diffraction and magnetic properties of Co₂Cr₁₋ₓTiₓAl Heusler alloys, Phys. Rev. B 100, 144444 (2019).DOI
  44. [65]J. Panda, S. N. Saha, and T. K. Nath, Critical behavior and magnetocaloric effect in Co₅₀₋ₓNiₓCr₂₅Al₂₅ (x=0 and 5) full Heusler alloy system, J. Alloys Compd. 644, 930 (2015).DOI
  45. [66]B. K. Banerjee, On a Generalised Approach to First and Second Order Magnetic Transitions, Phys. Lett. 12, 16 (1964).DOI
  46. [67]J. Y. Law, V. Franco, L. M. Moreno-Ramírez, A. Conde, D. Y. Karpenkov, I. Radulov, K. P. Skokov, and O. Gutfleisch, A quantitative criterion for determining the order of magnetic phase transitions using the magnetocaloric effect, Nat. Commun. 9, 2680 (2018).DOI
  47. [68]J. Y. Law, Á. Díaz-García, L. M. Moreno-Ramírez, and V. Franco, Increased magnetocaloric response of FeMn-NiGeSi high-entropy alloys, Acta Mater. 212, 116931 (2021).
  48. [69]V. Franco, R. Caballero-Flores, A. Conde, Q. Y. Dong, and H. W. Zhang, The influence of a minority magnetic phase on the field dependence of the magnetocaloric effect, J. Magn. Magn. Mater. 321, 1115 (2009).DOI
  49. [70]V. Franco, A. Conde, J. M. Romero-Enrique, and J. S. Blázquez, A universal curve for the magnetocaloric effect: an analysis based on scaling relations, J. Phys. Condens. Matter 20, 285207 (2008).DOI
  50. [71]Q. Y. Dong, H. W. Zhang, J. R. Sun, B. G. Shen, and V. Franco, A phenomenological fitting curve for the magnetocaloric effect of materials with a second-order phase transition, J. Appl. Phys. 103, 116101 (2008).DOI
  51. [72]V. K. Pecharsky and K. A. Gschneidner, Magnetocaloric effect from indirect measurements: magnetization and heat capacity, J. Appl. Phys. 86, 565 (1999).DOI
  52. [73]A. Arrott and J. Noakes, Approximate equation of state for Nickel near its critical temperature, Phys. Rev. Lett. 19, 786 (1967).DOI
  53. [74]J. S. Kouvel and J. B. Comly, Magnetic equation of state for nickel near its curie point, Phys. Rev. Lett. 20, 1237 (1968).DOI
  54. [75]S. N. Kaul, Static critical phenomena in ferromagnets with quenched disorder, J. Magn. Magn. Mater. 53, 5 (1985).DOI
  55. [76]Yu Liu and C. Petrovic, Three-dimensional magnetic critical behavior in Cr₃, Phys. Rev. B 97, 014420 (2018).DOI
  56. [77]L. Zhang, B. S. Wang, Y. P. Sun, P. Tong, J. Y. Fan, C. J. Zhang, L. Pi, and Y. H. Zhang, Critical behavior in the antiperovskite ferromagnet AlCMn₃, Phys. Rev. B 85, 104419 (2012).DOI
  57. [78]B. Widom, Equation of state in the neighborhood of the critical point, J. Chem. Phys. 43, 3898 (1965).DOI
  58. [79]A. K. Pramanik and A. Banerjee, Critical behavior at paramagnetic to ferromagnetic phase transition in Pr₀.₅Sr₀.₅MnO₃: a bulk magnetization study, Phys. Rev. B 79, 214426 (2009).DOI
  59. [80]G. D. Gupt, P. D. Babu, and R. S. Dhaka, Complex magnetic interactions and critical behavior analysis in quaternary CoFeVO₈Mn₀.₂Si Heusler alloy, J. Appl. Phys., 135, 083905 (2024).arxiv.org/abs/2310.01790
  60. [81]S. Tarhouni, R. M’nassri, A. Mleiki, W. Cheikhrouhou-Koubaa, A. Cheikhrouhoua and E. K. Hıl, Analysis based on scaling relations of critical behaviour at PM-FM phase transition and universal curve of magnetocaloric effect in selected Ag doped manganites, RSC Adv. 8, 18294 (2018).DOI
  61. [82]M. E. Fisher, S.-K. Ma, and B. G. Nickel, Critical exponents for long-range interactions, Phys. Rev. Lett. 29, 917 (1972).
  62. [83]M. E. Fisher, The renormalization group in the theory of critical behavior, Rev. Mod. Phys. 46, 597 (1974).DOI
  63. [84]S. F. Fischer, S. N. Kaul, and H. Kronmüller, Critical magnetic properties of disordered polycrystalline Cr₇₅Fe₂₅ and Cr₇₅Fe₃₀, Phys. Rev. B 65, 064443 (2002).
  64. [85]S. Roy, N. Khan, R. Singha, A. Pariari, and P. Mandal, Complex exchange mechanism driven ferromagnetism in half-metallic Heusler Co₂TiGe:evidence from critical behavior, Phys. Rev. B 99, 214414 (2019).DOI
  65. [86]A. Rahman, M. Rehman, Zhang, D. Zhang, M. Zhang, X. Wang, R. Dai, Z. Wang, X. Tao, L. Zhang, and Z. Zhang, Critical behavior in the half-metallic Heusler alloy Co₂TiSn, Phys. Rev. B 100, 214419 (2019).DOI
  66. [87]A. Rahman, M. Rehman, M. Zhao, X. Liu, J. Wang, Y. Lu, K. Ruan, R. Dai, Z. Wang, X. Tao, L. Zhang, and Z. Zhang, Itinerant magnetism in the half-metallic Heusler compound Co₂HfSn: Evidence from critical behavior combined with first-principles calculations, Phys. Rev. B 103, 094425 (2021).DOI
  67. [88]D. Kim, B. Revaz, B. L. Zink, F. Hellman, J. J. Rhyne, and J. F. Mitchell, Tricritical point and the doping dependence of the order of the ferromagnetic phase transition of La₁₋ₓCaₓMnO₃, Phys. Rev. Lett. 89, 227202 (2002).
  68. [89]P. Klaer, M. Kallmayer, C. G. F. Blum, T. Graf, J. Barth, B. Balke, G. H. Fecher, C. Felser, and H. J. Elmers, Tailoring the electronic structure of half-metallic Heusler alloys, Phys. Rev. B 80, 144405 (2009).DOI
  69. [90]M. Baral, V. Srihari, A. Bhakar, M. K. Chattopadhyay, P. Tiwari, A. Chakrabarti, and T. Ganguli, Revealing superstructure ordering in Co₁₊ₓMnSb Heusler alloys and its effect on structural, magnetic, and electronic properties, Phys. Rev. B 105, 184106 (2022).DOI
  70. [91]A. W. Carbonari, W. P. R. de L. N. Attili, and R. N. Saxena, Magnetic hyperfine fields in the Heusler alloys Co₂YZ (Y=Sc, Ti, Hf, V, Nb; Z=Al, Ga, Si, Ge, Sn), Hyperfine Interact. 80, 971 (1993).
  71. [92]Y. Jin, J. Waybright, P. Kharel, I. Tutic, J. Herran, P. Lukashev, S. Valloppilly, and D. J. Sellmyer, Effect of Fe substitution on the structural, magnetic and electron-transport properties of half-metallic Co₂TiSi, AIP Adv. 7, 055812 (2017).
  72. [93]K. Elphick, W. Frost, M. Samiepour, T. Kubota, K. Takanashi, H. Sukegawa, S. Mitani, and A. Hirohata, Heusler Alloys for Spintronic Devices: Review on Recent Development and Future Perspectives, Sci. Technol. Adv. Mater. 22, 235 (2021).DOI

Paper details

vlnc:W_aw9mygtv
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents