Metal-redox for MnAl-Based ternary magnetic nanocrystals

Abstract

We report the unique metal-redox nanosynthesis of ligand-free MnAl nanocrystals for the first time, where the doped cobalt element is shown to stabilize metastable tetragonal MnAl. The magnetic characteristics of MnAl nanocrystals can be tuned by the atomic percentage of cobalt dopant. The ternary CoMnAl nanoalloys have a potential application in the critical-energy related magnet application.

---

Rare-earth-free MnAl magnets have attracted wide attention, due to their earth abundancy, high magnetocrystalline anisotropy and moderate energy product. In this study, we report unique metal-redox nanosynthesis of ligand-free MnAl nanocrystals for the first time, where the doped cobalt element is shown to stabilize metastable tetragonal MnAl. The magnetic characteristics of MnAl nanocrystals can be tuned by the atomic percentage of cobalt dopant. The ternary CoMnAl nanoalloys have a potential application in the critical-energy related magnet application.

Manganese based alloys (MnAl or MnBi, etc.) have attracted extensive interests as the alternative for high energy product rare-earth based permanent supermagnets.¹ Particularly, the metastable L1₀ ordered MnAl alloys in the vicinity of equalatom (Mn atomic percentage of 51–58%, ferromagnetic τ phase) exhibit high magnetocrystalline anisotropy and a moderate energy product (BHmax: 12 MGOe).² The chemical ordering of τ-MnAl phase is achieved from phase transformation of the intermediate B19-structured ε′-phase, which originates from hexagonal ε-MnAl phase via rapidly cooling from high temperature.³ At present, high energy non-equilibrium processing techniques,⁴ such as ball-mill and melt spinning, are widely used for the growth of ferromagnetic metastable τ-MnAl alloys through two-step thermal annealing treatments. In addition, the stabilization of τ-MnAl phase has been realized by using the chemical doping, such as carbon, boron, zinc or iron dopant.⁵ However, the doped-MnAl alloys always exhibit a decrease of the magnetocrystalline anisotropy energy and Curie temperature.^5a,6

In the past, solution nanosynthesis has emerged as one scalable manufacturing approach for the metal alloy nanocrystals, which are generally synthesized by reducing the metal salts through the harsh reducing agents and organic surfactants.⁷ However, during the solution synthesis, organic residue or reducing agents could eventually get incorporated into the final nanocrystals, which would be detrimental for their magnetic performance. Recently, in order to eliminate such effects, our group has developed the metal-redox strategy to grow metallic nanocrystals, which utilize the inherent reducing power of zerovalent molecular-scale metal precursor to reduce the metallic ions for the alloy formation.^2,8 In this study, we apply metal-redox strategy to synthesize τ-MnAl phase for the first time in the solution. By incorporating the cobalt dopant, the metastable L1₀ τ-MnAl phase could be stabilized by the formation of MnAlCo nanoalloys.

All the reactions were carried on standard Schlenk-line. The typical synthesis route of MnAl as follows: 0.05 mmol AlCl₃ was added into a vial with 2 ml diphenyl ether. After refill and vaccum for 10 minutes, the vial was kept stirring with 60 °C on the hot plate. 0.05 mmol LiAlH₄ was dissolved into a vial with 2 ml THF, then refill and vacuum for another 10 minutes. 0.1 mmol Mn₂(CO)₁₀ was added into a flask in the glove-box. After sealing and transforming it, 4 ml degas diphenyl ether was injected into the flask. At the same time, the reaction flask with 10 ml diphenyl ether was also prepared. Then, the flasks were connected with the standard Schlenk – line, refilled and vacuumed for 10 minutes. When the temperature of reaction flask increasing up to 240 °C, the AlCl₃, Mn₂(CO)₁₀ and LiAlH₄ was injected in sequence. When the temperature restore to 240 °C, the reaction was kept for 1 hour. After that, the reaction flask cooled down to the room temperature. For chemical additives doping, the corresponding chemicals were injected into the reaction flask after LiAlH₄ injection. All the products were centrifuged with the rate of 4000 RPM for 10 minutes, washed by hexane for 3 times and dried by Argon. To improve crystallinity and reduce oxide, the products were annealed at 773 K for 1 h in the forming gas with a mixture of 5%H₂ and 95%N₂. The sample morphology was carried on transmission electron microscope (TEM) and high resolution TEM (HRTEM) by field emission FEI Tecnai F20 XT with an accelerating voltage of 200 kV. The crystalline information was obtained from room temperature X-ray diffraction (XRD) characterization were obtained using monochromated Cu-K_α radiation (λ = 1.54178 Å). A MircoSense EV7 high sensitivity vibrating sample magnetometer (VSM) was used to test magnetic properties of all samples.

In this study, the zerovalent metal precursor (Mn₂(CO)₁₀, E^o = −1.19 V) is selected to reduce aluminum chloride (AlCl₃, E^o = −1.66 V). To enhance the aluminum salt reduction, lithium aluminium hydride (LiAlH₄) is used to facilitate the reduction towards to the zerovalent Al formation (the synthetic schematic route is shown in Fig. 1a). In addition, it is critical to achieve the matched decomposition rate between Mn₂(CO)₁₀ and the aluminum salt, and therefore we have optimized the precursor ratios, the injection and reaction temperatures to obtain the stoichiometry-controlled MnAl nanocrystals. The corresponding magnetic performance of MnAl nanocrystals is provided in the ESI (Fig. S1†). The saturation magnetization (Ms) and coercivity of synthesized MnAl nanocrystals are 1.67 emu g⁻¹ and 213 Oe, respectively (Fig. 1b). X-ray diffraction (XRD) is used to confirm the phase composition (Fig. 1c), where three major phases are identified for Al, Mn and MnAl nanostructures. The two-theta diffraction peaks in 41.2° and 43.8° represent (103) and (232) planes of MnAl phase, respectively, indicating the formation of τ phase.^4b However, the weight percentage of MnAl phase is around 5.6%, which is much lower than that of Al (68.5%) and Mn (25.9%), leading to the overall weak magnetic performance. Transmission electron microscopy (TEM) and energy-dispersive X-ray spectroscopy (EDX) elemental mapping show that the MnAl nanocrystals are embedded into the Mn matrix, which limits the magnetic performance of MnAl nanoalloys.

Fig. 1 (a) Schematic route of MnAl nanoalloy synthesis, (b) magnetic hysteresis (M–H) loops of MnAl nanoalloy, the inset shows the coercivity of MnAl nanoalloy, (c) XRD patterns of MnAl nanoalloy, (d) TEM image of MnAl nanoalloy, (e) STEM elemental mapping images of MnAl nanoalloy, (f) combined Al/Mn elemental mapping image, (g) Al elemental mapping images in red, (h) Mn elemental mapping images in yellow.

The chemical doping of MnAl alloys have been predicted to stabilize the τ-MnAl phase for high magnetic performance. In this study, we select the Fe, Co and Ni elements to study their doping effect on the magnetic characteristics of MnAl nanoalloys. As shown in Fig. 2a, the are −1.19, −0.44, −0.28 and −0.25, respectively, indicating that Fe, Co and Ni ions can be reduced by zerovalent manganese precursor (Mn₂(CO)₁₀). To synthesize ternary MnAlM (M: Fe, Co and Ni) nanoalloys, we utilize the inherent reducing power of zerovalent manganese by incorporating the third precursor during the metal-redox process. Fig. 2b shows that the magnetic hysteresis (M–H) loops of ternary MnAlM nanostructures, where the Co doping induces a drastic increase of magnetic performance of MnAl nanoalloys (Fig. 1b). To exclude the potential contribution of CoMn and CoAl phases, the magnetic characteristics of CoMn and CoAl control samples are shown in Fig. S2,† confirming that the measured magnetic performance results from the role of Co doping into the MnAl phase. To explain the magnetic performance difference after Fe, Ni and Co doping, we utilize the EDX mapping to illustrate the elemental distribution. As shown in Fig. 2c–e, the Fe and Co elements are incorporated into the MnAl phase. However, the Ni element is coated on the surface of MnAl phase. Though the Fe element in indeed incorporated into the MnAl phase (Fig. 2c), antiferromagnetic ordering takes place in the Mn-based magnets due to the Fe incorporation.⁹ Therefore, the Ms of MnAl alloy is deteriorated after Fe doping. The TEM and high resolution TEM (HRTEM, Fig. 2f and g) are used to confirm the Co distribution into the MnAl phase. As shown in Fig. 2f, the MnAlCo nanocrystals display an average size of 20.1 nm (The size distribution is shown in Fig. 2f). As shown in Fig. 2g, the interplanar distance matches with (200)-MnAlCo₂ (0.284 nm), indicating the formation of ternary MnAlCo nanoalloys.¹⁰ The Co doping through metal-redox process provides a suitable method to improve magnetic performance of MnAl nanocrystals. Therefore, in the following discussion, we will focus on magnetic performance and structural analysis of Co-doped MnAl phases.

Fig. 2 (a) Standard reduction potentials of M²⁺/M (M: Mn, Fe, Co and Ni), (b) magnetic hysteresis (M−H) loops of MnAlM ((M: Fe, Co and Ni)) alloy, (c–e) TEM image and combined Al/Mn elemental mapping images of MnAlM ((M: Fe, Co and Ni)) alloy, (f) TEM image of as-synthesized MnAlCo nanoparticles and the size distribution (inset), (g) HRTEM image of as-synthesized MnAlCo nanoalloy.

It is known that the reaction and annealing conditions play an important role in optimizing the magnetic performance of metal nanoalloys, due to the control of their chemical composition and nanostructures. As shown in Fig. 3a and b, the reaction temperature of 513 K and reaction time of 60 minutes offer the optimum magnetic performance of MnAlCo nanoalloys. A high reaction temperature is needed to make the reaction towards nanoalloying direction. Therefore, the Ms and coercivity of MnAlCo nanostructured alloys increase as the reaction temperature increases from 493 K to 513 K. However, further increasing the reaction temperature, the LiAlH₄ phase exhibits dual functions as both reducing power and self-decomposition (the decomposition temperature of 473 K). Therefore, at a relatively high temperature, the Al concentration in the MnAlCo nanoalloys is significantly improved. Due to nonmagnetic characteristic of the Al element, the Ms and coercivity in the ternary MnAlCo alloys are reduced as increasing the reaction temperature or the reaction time above the decomposition point of LiAlH₄. As shown in Fig. 3b, at the reaction temperature of 513 K which is higher than the decomposition temperature of LiAlH₄, a long reaction time (120 and 180 minutes) can produce more Al formation, ultimately resulting in a low Ms and coercivity. The ternary MnAlCo nanoalloys also achieve the optimum magnetic performance after annealing at 823 K, as shown in Fig. S3 (ESI†). In addition, the injection ratio of Mn and Al, as well as the concentration of cobalt doping, play an important role in the magnetic performance of MnAlCo nanocrystals (Fig. S4 of ESI†). To summarize the stoichiometry effect on magnetic performance of MnAlCo nanoalloys, Fig. 3c shows the relationship between the Ms, coercivity and stoichiometry of Mn, Al and Co in the ternary MnAlCo nanoalloys. The stoichiometry of Mn, Al and Co is calculated from the EDX spectrum, which is shown in the Fig. S5 of ESI.† The optimum magnetic performance of ternary CoMnAl nanoalloys is achieved at Mn_90.8Al_0.1Co_9.1, which shows the Ms (15.1 emu g⁻¹) and coercivity (850 Oe), respectively. It should be noted that the Ms and coercivity of Mn_68.2Al_17.9Co_13.9 is less than that of Mn_90.8Al_0.1Co_9.1, where the high Al concentration is believed to cause the low performance in the Mn_68.2Al_17.9Co_13.9 phase.

Fig. 3 (a) Magnetic-hysteresis loops of ternary MnAlCo nanoalloy at different reaction temperature with 1 hour, (b) magnetic–hysteresis loops of MnAlCo nanoalloy at different reaction time under the reaction temperature of 240 °C, (c) the magnetization saturation and coercivity depends on the Mn, Al and Co stoichiometry.

The stoichiometry between aluminum, cobalt and manganese dictates the magnetic performance of MnAlCo nanoalloys. Fig. 4a shows X-ray diffraction (XRD) spectra of ternary MnAlCo nanoalloys at different stoichiometries. For the low concentration of Co dopant, two major phases exist as MnAlCo₂ and τ-MnAl, which is consistent with the EDX mapping for the phase separation between Co and MnAl (Fig. 4b). As increasing the Co concentration, the major phases include MnAlCo₂, τ-MnAl and Mn. The cubic MnAlCo₂ alloy typically shows a soft magnetic behavior, resulting in a low coercivity due to its isotropic-structure and a high magnetic moment (3.97–4.01 μB per formula unit).¹¹ The τ-MnAl phase exhibits a hard magnetic characteristic, where the coercivity could be tuned from 1500 to 4000 Oe.¹² In this study, the optimum Ms and coercivity of MnAlCo alloys (15 emu g⁻¹ and 850 Oe, respectively) suggests the coexistence of MnAlCo₂ and τ-MnAl phases after the cobalt doping. Meanwhile, the EDX mapping shows that Co, Mn and Al elements are superimposed to each other as increasing the concentration of cobalt dopant, indicating ternary MnAlCo nanoalloy formation (Fig. 4b).

Fig. 4 (a) XRD patterns of Mn_90.8Al_0.1Co_9.1, Mn_90.1Al_0.9Co_9.0, Mn_68.2Al_17.9Co_13.9, Mn_84.2Al_8.1Co_7.6, Mn_40.5Al_55.4Co_4.1 (from top to bottom), (b) TEM image, STEM image and the combined Co/Mn/Al elemental mapping image (from left to right) of Mn_90.8Al_0.1Co_9.1, Mn_90.1Al_0.9Co_9.0, Mn_68.2Al_17.9Co_13.9, Mn_84.2Al_8.1Co_7.6, Mn_40.5Al_55.4Co_4.1 (from top to bottom).

Conclusions

In summary, the MnAl nanoalloys are synthesized through the metal-redox strategy. To improve magnetic performance and stabilize MnAl τ phase, the cobalt dopant is incorporated by using three precursor based metal-redox process. When the cobalt concentration is 9.1%, the Ms and coercivity of ternary MnAlCo nanoalloys is improved up to 15.1 emu g⁻¹ and 850 Oe, respectively. The synthesized rare-earth-free MnAlCo magnetic composites offer the potential applications in the magnet-related green energy technologies.

Acknowledgements

S.R. thanks the financial support from the U.S. National Science Foundation (NSF) under the carreer award No: NSF-DMR-1551948.

Notes and references

1. (a) L. Lewis and F. Jiménez-Villacorta, Metall. Mater. Trans. A, 2013, 44, 2; (b) J. B. Yang, Y. B. Yang, X. G. Chen, X. B. Ma, J. Z. Han, Y. C. Yang, S. Guo, A. R. Yan, Q. Z. Huang, M. M. Wu and D. F. Chen, Appl. Phys. Lett., 2011, 99, 082505; (c) J. M. D. Coey, IEEE Trans. Magn., 2011, 47, 4671; (d) A. Kirkeminde, J. Shen, M. Gong, J. Cui and S. Ren, Chem. Mater., 2015, 27, 4677.
2. A. Kirkeminde, S. Spurlin, L. Draxler-Sixta, J. Cooper and S. Ren, Angew. Chem., 2015, 127, 4277.
3. Y. J. Kim and J. H. Perepezko, J. Appl. Phys., 1992, 71, 676.
4. (a) F. Jiménez-Villacorta, J. L. Marion, T. Sepehrifar, M. Daniil, M. A. Willard and L. H. Lewis, Appl. Phys. Lett., 2012, 100, 112408; (b) F. Jiménez-Villacorta, J. Marion, J. Oldham, M. Daniil, M. Willard and L. Lewis, Metals, 2014, 4, 8.
5. (a) Z. W. Liu, C. Chen, Z. G. Zheng, B. H. Tan and R. V. Ramanujan, J. Mater. Sci., 2012, 47, 2333; (b) M. Matsumoto, A. Morisako and N. Kohshiro, IEEE Translat. J. Magn. Jpn., 1991, 6, 134; (c) M. Hosoda, M. Oogane, M. Kubota, T. Kubota, H. Saruyama, S. Iihama, H. Naganuma and Y. Ando, J. Appl. Phys., 2012, 111, 07A324; (d) W. Van Roy, H. Bender, C. Bruynseraede, J. De Boeck and G. Borghs, J. Magn. Magn. Mater., 1995, 148, 97.
6. H. X. Wang, Open J. Microphys., 2011, 01, 19.
7. (a) Y. Yu, W. Yang, X. Sun, W. Zhu, X. Z. Li, D. J. Sellmyer and S. Sun, Nano Lett., 2014, 14, 2778; (b) S. F. Ho, Nanoscale, 2014, 6, 6970.
8. A. Kirkeminde and S. Ren, Chem. Commun., 2015, 51, 8603.
9. P. Manchanda, A. Kashyap, J. E. Shield, L. H. Lewis and R. Skomski, J. Magn. Magn. Mater., 2014, 365, 88.
10. T. Ryba, Z. Vargova, J. Kovac, P. Diko, V. Kavecansky, S. Piovarci, C. Garcia and R. Varga, IEEE Trans. Magn., 2015, 51, 1.
11. (a) K. Özdoğan, E. Şaşıoğlu, B. Aktaş and I. Galanakis, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 172412; (b) A. Hirohata, M. Kikuchi, N. Tezuka, K. Inomata, J. S. Claydon, Y. B. Xu and G. van der Laan, Curr. Opin. Solid State Mater. Sci., 2006, 10, 93–107.
12. P. C. Kuo, Y. D. Yao, J. H. Huang and C. H. Chen, J. Magn. Magn. Mater., 1992, 115, 183–186.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra05853a

---