Letter pubs.acs.org/NanoLett
Phase Transformation-Induced Tetragonal FeCo Nanostructures Maogang Gong,† Alec Kirkeminde,† Manfred Wuttig,‡ and Shenqiang Ren*,† †
Department of Chemistry, University of Kansas, Lawrence, Kansas 66045, United States Department of Materials Science and Engineering, University of Maryland, College Park, Maryland 20742, United States
‡
S Supporting Information *
ABSTRACT: Tetragonal FeCo nanostructures are becoming particularly attractive because of their high magnetocrystalline anisotropy and magnetization achievable without rare-earth elements, . Yet, controlling their metastable structure, size and stoichiometry is a challenging task. In this study, we demonstrate AuCu templated FeCo shell growth followed by thermally induced phase transformation of AuCu core from face-centered cubic to L10 structure, which triggers the FeCo shell to transform from the body-centered cubic structure to a body-centered tetragonal phase. High coercivity, 846 Oe, and saturation magnetization, 221 emu/g, are achieved in this tetragonal FeCo structure. Beyond a critical FeCo shell thickness, confirmed experimentally and by lattice mismatch calculations, the FeCo shell relaxes. The shell thickness and stoichiometry dictate the magnetic characteristics of the tetragonal FeCo shell. This study provides a general route to utilize phase transformation to fabricate high performance metastable nanomagnets, which could open up their green energy applications. KEYWORDS: Iron−cobalt alloy, nanomagnetism, magnetocrystalline anisotropy, core/shell
I
reduction of the magnet weight while keeping the performance of generators or motors.17 Recently, from the first-principle calculations, Burkert et al. have predicted that the Ku of FeCo alloys could be tuned by the composition and the c/a ratio of its tetragonal distortion.14 At an optimized parameter of c/a = 1.20−1.25, FeCo can achieve a high value of Ku ≈ 1 × 107 J· m−3, three orders of magnitude higher than that of bcc-Fe, surpassing L10 FePt. More importantly, the predicted easy magnetization axis is along the c axis, which indicates Ku, is along (001) orientation and implies an approach to realize its high Ku by introducing simple epitaxial strain. Indeed, recently, tetragonally distorted FexCo1−x thin films with larger Ku have been shown using the epitaxial growth on Pt (001),18 Pd (001),19 Rh (001),20 Ir (001),21,22 and AuCu3 (001)12 substrates. In this letter, we report the templated synthesis of AuCu/ FeCo (core/shell) nanocrystals, which exhibit a thermally induced phase transformation to L10-AuCu thereby creating a tetragonal FeCo nanostructure. This approach shows a unique feature that the tetragonal distortion of the epitaxially grown FeCo shell evolves in response to thermally induced L10 ordering of the AuCu core (at 380 °C annealing). The order−disorder transition of AuCu is well-known and occurs in the temperature ranging from 322 to 437 °C dependent on the composition.23 In this temperature range the AuCu disordered face-centered cubic (fcc) structure transforms to the L10
ron-based precious metal alloys with a large uniaxial magnetocrystalline anisotropy (Ku) have been widely studied as permanent magnets, which have great potential for green energy applications, such as traction-based electric motors,1 wind generators,2 and efficient power electronics.3 The energy density of permanent nanomagnets is dictated by their saturation magnetization and magnetocrystalline anisotropy energy (MAE).4−6 The tetragonal L10-ordered FePd and FePt alloys possess a large value of Ku due to their structural asymmetry.7,8 Despite the unique magnetic properties (large Ku) of L10 FePd or FePt, Pd and Pt compounds as rare and precious materials have limited their broad utilization. Therefore, alternative magnetic alloys with large Ku and free of rare metals are attracting much attention in the magnetic community. FeCo alloys, an important traditional soft magnetic material, are very attractive due to their abundance on the earth, high permeability, and saturation magnetization, Ms,9−11 as well as the alloy structure-dependent Ku.12 However, these alloys are magnetically “soft” meaning that their coercive field is small and its MAE energy is low, typically Ku ≈ 0.1 J·m−3.8 To increase MAE and therefore the energy product, the coercive field of FeCo alloys must be increased.13 It has been predicted that creating a tetragonal or L10-ordered structure of FeCo will increase its coercivity much like its alloy counterparts L10 FePt and FePd.14−16 Therefore, a tetragonal or L10 FeCo phase with high MAE, combined with its highest Ms, could contribute significantly to the development of rare-earth-free high energy product nanomagnets. L10 FeCo alloys can have an energy product three times higher than current state-of-the-art Nd− Fe−B magnets. Consequently, their use could lead to a 3-fold © XXXX American Chemical Society
Received: August 7, 2014 Revised: September 19, 2014
A
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
S1). The L10-ordered AuCu reverts to the disordered fcc structure if the annealing temperature is further increased. Therefore, to confirm the phase transformation effect on the tetragonal distortion of FeCo, we anneal the AuCu/FeCo core/ shell structure at 500 °C, into the stability range of the cubic AuCu phase and obtained a low coercivity confirming the importance of the L10-AuCu structure distortion. Though significant achievements have been previously shown in the growth of the bcc and fcc FeCo epitaxial thin films using the lattice-matched metal substrates (Figure 1a),18,22 the solution-based bulk synthesis of FeCo nanocrystals with the large MAE is still a challenging task (Figure 1b). The tetragonal distortion realized within the FeCo epilayer could enable a large MAE, due to the in-plane expansion from the lattice-mismatched interface. A list of lattice-matched metal substrates with FeCo is shown in Figure 1c, where the Bain strain lattice correspondence with the FeCo phase makes AuCu a prime candidate for an epitaxial growth template for the FeCo shell. Our hypothesis for this study is to test that the lattice expansion due to L10 ordering transformation of AuCu core could enable the tetragonal distortion of FeCo shell, leading to a large MAE. Figure 2a shows the synthetic scheme of AuCu/ FeCo (core/shell) nanocrystals, where the AuCu core guides the templated growth of the FeCo shell. Figure S2, Supporting Information, shows A transmission electron microscopy (TEM) image of AuCu core nanoparticles with an average diameter of 10.2 nm, where the pristine AuCu nanoparticles have typically a face-centered cubic (fcc) structure with (200)
Figure 1. Schematic design route of FeCo thin film growth on fcc metal substrate (a) and (b), Schematic design route of FeCo thin film growth on fcc metal substrate, and bct FeCo shell growth on the core metals, respectively. (c) Lattice mismatch table between metal templates and FeCo.
ordered “AuCu I” phase. In this study, L10-ordered AuCu core structure is attained under low temperature 380 °C annealing. The evolution of the tetragonal distortion strain of the L10AuCu phase transformation triggers the body-centered cubic (bcc) FeCo shell to transform into the body-centered tetragonal (bct) phase. The magnetic properties can be tuned by the FeCo shell thickness and the stoichiometry of the Fe− Co phase, i.e., by altering the growth time and chemical precursor loading ratios. After the tetragonal transformation, the coercivity of optimized L10-AuCu/FeCo core/shell nanocrystals can reach a value as high as 846 Oe, about three times higher than the previously reported value,24 and a nearly six times increase compared to that of the FeCo control sample in this study (details are shown in Supporting Information Figure
Figure 2. (a) Illustration of AuCu/FeCo (core/shell) nanocrystal synthesis. (b−d) TEM and high-resolution TEM (HRTEM) images of AuCu/ FeCo (core/shell). (e) STEM elemental mapping images of AuCu/FeCo core/shell nanocrystals. (f) AuCu core elemental mapping images in yellow. (g) FeCo shell elemental mapping images in blue. (h) Combined AuCu/FeCo core/shell elemental mapping image. B
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 3. (a) Magnetic−hysteresis loops of L10-AuCu/FeCo core/shell nanocrystals with an overall Fe/Co ratio at 47:53 under different annealing conditions. The inset shows the annealing temperature-dependent coercivity of L10-AuCu/FeCo. (b) XRD patterns of AuCu/FeCo core/shell nanocrystals: pristine and after 380 °C annealing. The black and red labeled peaks represent AuCu and FeCo diffraction peaks. (c,d) TEM and HRTEM images of L10-AuCu/tetragonal FeCo nanocrystals annealed at 380 °C.
shows the magnetic hysteresis (M−H) loops of the AuCu/ FeCo samples (Fe/Co ratio of 47:53, the mass contribution of AuCu core has been accounted for). The inset of Figure 3a exhibits the annealing temperature-dependent coercivity (Hc) of L10-AuCu/FeCo core/shell nanocrystals with an overall Fe/ Co ratio at 47:53. The epitaxial growth of AuCu and FeCo could influence the order/disorder temperature of nanoparticles.25 Therefore, the AuCu/FeCo nanocrystals annealed at 300 °C could be ordered due to a decrease of the order/ disorder temperature (Supporting Information, Figure S5), which combined with the thermally induced crystallinity will enhance the coercivity. The coercivity of AuCu/FeCo without annealing is 54.3 Oe, while the 380 °C annealed sample shows Hc = 626.6 Oe. This increase can be attributed to the L10 ordering of the AuCu core, which induces the tetragonal distortion in the FeCo shell structure. The X-ray diffraction (XRD) studies (Figure 3b) confirm the phase transformation upon annealing at 380 °C of the AuCu/FeCo core/shell nanocrystals. The (200) diffraction peak in the pristine XRD pattern represents the AuCu (200) peak of the fcc structure, as the (002) and (201) peaks after annealing signal the L10-AuCu structure.23,26 The XRD patterns confirm a fcc-L10 phase transition of the AuCu core at 380 °C annealing for 20 h. The HRTEM images of pristine (Figure 1d) and annealed AuCu/ FeCo nanocrystals (Figures 3d and S6, Supporting Information) further confirm the fcc-L10 and bcc-bct phase transformations of AuCu core and FeCo shell, respectively. Figure 3c,d shows the HRTEM image of L10-AuCu/FeCo (L10-core/ tetragonal-shell) nanocrystals annealed at 380 °C for 20 h. It should be noted that there are two nanoscale domains with the lattice constant of 0.184 and 0.187 nm, which are representative of the (200) plane in L10-AuCu and (110) plane of tetragonal FeCo.12,27,28 However, with a further increase of the annealing
growth planes. Figure 2b,c indicates the TEM images of the AuCu/FeCo core/shell nanocrystals, where the FeCo shell thickness can be tuned from 2.9 to 5.4 nm by controlling the shell growth time as discussed in more detail in Supporting Information Figure S3. The HRTEM image (Figure 2d) of one single AuCu/FeCo nanocrystal reveals that each particle is composed of the AuCu core and FeCo shell. The interplanar distance of the AuCu core is 0.194 nm, corresponding to (200) lattice constant of fcc-AuCu. The lattice distance of the shell FeCo is 0.202 nm, consistent with the (110) surface plane of bcc-FeCo shell. In addition to the growth-time-dependent FeCo shell thickness, the Fe/Co stoichiometry is controlled by the amount of Co2(Co)8 and Fe(CO)5 precursors used in the synthesis. The elemental Fe/Co ratio in the shell can be tuned by changing the amount of the iron precursor (37:63 and 47:53 atomic ratios for Fe/Co by increasing iron precursor from 1 to 2 mmol (details are shown in Supporting Information Table 1). Figure 2e−h shows the scanning TEM (STEM) images of AuCu/FeCo core/shell nanocrystals, which further confirm the core/shell structure by elemental mapping analysis. The Au and Cu elements are centered in the core region of each nanocrystal (Figure 2f), and Fe and Co elements are located in the shell region (Figure 2g). The elemental mapping image of Au, Cu, Fe, and Co in the AuCu/FeCo (core/shell) nanostructures is shown in Figure S4 of Supporting Information. A composite image (Figure 2h) shows a uniform coating of FeCo shell on the AuCu core confirming the formation of the AuCu/FeCo core/shell structure. It is important to note that the phase transformation is dependent on the annealing temperature, which can further affect the magnetic properties of AuCu/FeCo nanocrystals.23 Room temperature magnetic properties of thermally annealed AuCu/FeCo core/shell nanocrystals were studied. Figure 3a C
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
Figure 4. (a) Heteroepitaxial interfacial structure of AuCu/FeCo core/shell systems: pristine (left) and after 380 °C annealing (right). (b) The coercivity depends on the FeCo shell thickness and stoichiometry. (c) M−H loops of AuCu/FeCo with an overall Fe/Co ratio at 53:47. The inset shows the annealing temperature-dependent coercivity of L10-AuCu/FeCo. (d) SEM image of AuCu/FeCo core/shell particles at 53:47 annealed at 380 °C.
⎡ 4h ⎤ (1 − v)x + ln c ⎥ /8π (1 + v)m hc /x = ⎢1 + ln ⎣ 2πd x ⎦
temperature from 380 to 500 °C, the Hc of AuCu/FeCo nanocrystals drops the coercivity to 202.2 Oe due to the disordering transformation to the fcc-AuCu structure in the core (Figure 3a), which retracts the tetragonal transformation of the FeCo shell. This can be attributed to the fcc structure formation of the AuCu core under a high temperature annealing (500 °C), upon which a small tetragonal distortion of FeCo shell yields a low Hc of the core/shell nanocrystals (the HRTEM and FFT images of the AuCu/FeCo nanocrystals annealed at 500 °C are shown in Figure S7 of Supporting Information). The magnetic transition obtained from all samples indicates that the coercivity of FeCo shell originates from the structure evolution from cubic to tetragonal during thermal annealing, which is due to the disorder/order transition of the AuCu core from fcc to L10 structure. The (200)−(110) structure relationship between the AuCu core and FeCo shell suggests a Bain relationship between the two phases. The Bain strain mismatch m = (b − a)/a has been utilized to quantify the distortion between the two phases; a and b are the lattice parameters of (200) AuCu core and (110) FeCo shell, respectively. The resulting mismatches between the L10-AuCu/tetragonal FeCo and L10-AuCu/bcc FeCo are 2.1% and 10.0% (a detailed calculation is shown in the Supporting Information), respectively, meaning that the L10-AuCu/bctFeCo interface is more stable than that of the L10-AuCu/bcc FeCo interface. Figure 4a shows the schematic interfacial structure of heteroepitaxially grown AuCu/FeCo nanocrystals. As the volume energy of the mismatch-strained bct-FeCo increases, i.e., as its thickness grows, it will relax into the unstrained A2 structure at a critical thickness. The critical thickness hc is determined by the misfit m and elastic constant29 (details are shown in the Supporting Information)
where intermediate lattice constant x = 2ab/(a + b), v as Poisson’s ratio of FeCo, and d as the spacing between atomic planes on each side of the interface. In our L10-AuCu/ tetragonal FeCo systems, the calculated critical thickness hc is about 3.17 nm. In tetragonal FeCo alloys, a strongly enhanced Ku can be expected when the crystal constant ratio c/a ≈ 1.22 and the alloy stoichiometry is chosen such that the Fermi energy is located in the band crossing.4,6,16 In that case, the maximum of Ku appears at the alloy concentration of 0.5.6,30 By varying the FeCo shell thickness and Fe−Co stoichiometry, L10-AuCu/bctFeCo core/shell nanocrystals are thus synthesized to match the results of the theoretical calculations. The final shell thickness and stoichiometry are determined by HRTEM and EDX analysis. The details for the synthetic control are discussed in the Supporting Information. The results of the thickness and concentration tuning summary of annealed samples are shown in Figure 4b. The FeCo shell thickness is adjusted from 2.9 to 5.3 nm at constant stoichiometry of Fe47Co53. The Hc of the annealed samples decreases as the FeCo shell thickness increases, consistent with the critical film thickness of 3.17 nm at which the tetragonal distortion relaxes. When the shell thickness surpasses the relaxed distance, the outer atomic layers of FeCo prefer the equilibrium A2 structure weaken tetragonal effects, leading to a reduced Hc. By varying the concentration of Co from 100% to 0% at a constant shell thickness (2.9 ± 0.5 nm), Hc first increases and reaches its highest value at cCo = 0.53, consistent with the predicted value 0.5 for the highest Hc in the tetragonal structure.14 D
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
■
ACKNOWLEDGMENTS S.R. is thankful for the financial support from the National Science Foundation (NSF) under Award no. NSF-CMMI1332658, and M.W. is thankful for a grant from NSF-DMR0705368.
To further improve the coercivity of FeCo shell through lowtemperature induced phase transformation, we adopt a novel ligand-free synthesis of AuCu/FeCo core/shell nanocrystals. It is known that organic surfactants as reducing agent or stabilizer usually play an important role in the solution growth. However, a low-temperature annealing as used in this study will not fully decompose the organic residue, which can increase the volume of the magnetic shell thereby reducing the saturation magnetization. In addition, the organic surfactant barrier at the interface could also interrupt the order/disorder transformation-induced tetragonal formation and decrease the MAE. Therefore, an optimized synthesis without organic ligands could enable a low temperature annealing (380 °C in this study), which will be needed to enable L10 ordering of AuCu core to reach high performance AuCu/FeCo nanomagnets (detailed synthesis is described in Supporting Information). Figure 4c shows the M−H loops of AuCu/FeCo core/shell with an overall Fe/Co ratio at 53:47 synthesized ligand free. Both Hc and Ms of the 380 °C annealed samples for 20 h show a significant further increase of Hc as the tetragonal FeCo phase forms (Hc = 846 Oe and Ms = 221 emu/g). Figure 4d shows the optimal SEM image of AuCu/FeCo core/shell structure annealed at 380 °C. Note that the ligandless synthesis generates clean nanoparticle surfaces. Hence, the AuCu/FeCo (core/ shell) nanocrystals tend to aggregate. This mode of local fusion does not form local ordered structures. Therefore, the magnetic properties reflect those of randomly packed core/shells. In conclusion, a versatile templating and phase transformation approach is reported to synthesize tetragonal FeCo nanomagnets, exhibiting high Ms = 221 emu/g and Hc = 846 Oe. The approach shows a novelty that a tetragonal FeCo shell can be induced by the L10 phase transformation of the AuCu templating core under mild thermal annealing. The structure evolution of the AuCu core dictates the tetragonal distortion of the FeCo shell, leading to its tunable magnetic characteristics. The close lattice match between the L10-ordered AuCu core and tetragonal (110) FeCo shell forms a stable interface. The Fe/Co stoichiometry and FeCo shell thickness play an important role in determining the coercivity of the FeCo shell layer. Above a critical shell thickness (3.17 nm), the strain relaxation and the FeCo shell returns to the bcc structure. This work provides a general route to synthesize heteroepitaxial L10AuCu/tetragonal FeCo nanomagnets. This synthetic effort could be expanded to other magnetic composite systems, which will allow the fabrication of high energy density magnets for magnetic storage and green energy applications.
■
■
REFERENCES
(1) Scott, J. H. J.; Chowdary, K.; Turgut, Z.; Majetich, S. A.; Mchenry, M. E. J. Appl. Phys. 1999, 85, 4409−4401. (2) Cui, W. B.; Takahashi, Y. K.; Hono, K. Adv. Mater. 2012, 24, 6530−6535. (3) Johnson, F.; Hughes, P.; Gallagher, R.; Laughlin, D. E.; McHenry, M. E.; Willard, M. A.; Harris, V. G. IEEE Trans. Magn. 2001, 37, 2261−2263. (4) Antoniak, C.; Gruner, M. E.; Spasova, M.; Trunova, A. V.; Romer, F. M.; Warland, A.; Krumme, B.; Fauth, K.; Sun, S.; Entel, P.; Farle, M.; Wende, H. Nat. Commun. 2011, 2, 528. (5) Wang, D.; Wu, R.; Freeman, A. Phys. Rev. B 1993, 47, 14932. (6) Manchanda, P.; Kashyap, A.; Shield, J. E.; Lewis, L. H.; Skomski, R. J. Magn. Magn. Mater. 2014, 365, 88−92. (7) Inoue, K.; Shima, H.; Fujita, A.; Ishida, K.; Oikawa, K.; Fukamichi, K. Appl. Phys. Lett. 2006, 88, 102503. (8) Shima, H.; Oikawa, K.; Fujita, A.; Fukamichi, K.; Ishida, K.; Sakuma, A. Phys. Rev. B 2004, 70, 224408. (9) Yu, Y.; Sun, K.; Tian, Y.; Li, X. Z.; Kramer, M. J.; Sellmyer, D. J.; Shield, J. E.; Sun, S. Nano Lett. 2013, 13, 4975−4979. (10) Bozorth, R. M. Ferromagnetism; van Nostrand: Princeton, NJ, 1951. (11) Skomski, R.; Coey, J. M. D. Permanent Magnetism; Institute of Physics: Bristol, U.K., 1999. (12) Ohtsuki, T.; Kojima, T.; Kotsugi, M.; Ohkochi, T.; Mizuguchi, M.; Takanashi, K. J. Appl. Phys. 2014, 115, 043908. (13) Sort, J.; Suriñach, S.; Muñoz, J. S.; Baró, M. D.; Nogués, J.; Chouteau, G.; Skumryev, V.; Hadjipanayis, G. C. Phys. Rev. B 2002, 65, 174420. (14) Burkert, T.; Nordström, L.; Eriksson, O.; Heinonen, O. Phys. Rev. Lett. 2004, 93, 027203. (15) Laughlin, D. E.; Srinivasan, K.; Tanase, M.; Wang, L. Scr. Mater. 2005, 53, 383−388. (16) Andersson, G.; Burkert, T.; Warnicke, P.; Björck, M.; Sanyal, B.; Chacon, C.; Zlotea, C.; Nordström, L.; Nordblad, P.; Eriksson, O. Phys. Rev. Lett. 2006, 96, 037205. (17) Gutfleisch, O.; Willard, M. A.; Brück, E.; Chen, C. H.; Sankar, S. G.; Liu, J. P. Adv. Mater. 2011, 23, 821−842. (18) Warnicke, P.; Andersson, G.; Björck, M.; Ferré, J.; Nordblad, P. J. Phys.: Condens. Matter 2007, 19, 226218. (19) Winkelmann, A.; Przybylski, M.; Luo, F.; Shi, Y.; Barthel, J. Phys. Rev. Lett. 2006, 96, 037205. (20) Blizak, S.; Bihlmayer, G.; Blügel, S. Phys. Rev. B 2012, 86, 094436. (21) Yildiz, F.; Luo, F.; Tieg, C.; Abrudan, R.; Fu, X.; Winkelmann, A.; Przybylski, M.; Kirschner, J. Phys. Rev. Lett. 2008, 100, 037205. (22) Yildiz, F.; Przybylski, M.; Ma, X. D.; Kirschner, J. Phys. Rev. B 2009, 80, 064415. (23) Anraku, T.; Sakaihara, I.; Hoshikawa, T.; Taniwaki, M. Mater. Trans. 2009, 50, 683−688. (24) Chaubey, G. S.; Barcena, C.; Poudyal, N.; Rong, C. B.; Gao, J. M.; Sun, S. H.; Liu, J. P. J. Am. Chem. Soc. 2007, 129, 7214−7215. (25) Zhu, Y.; Cai, J. W. Appl. Phys. Lett. 2005, 87, 032504. (26) Sra, A. K.; Schaak, R. E. J. Am. Chem. Soc. 2004, 126, 6667− 6672. (27) Pramanik, S.; Mishra, M. K.; De, G. CrystEngComm 2014, 16, 56−63. (28) Farnan, G.; Fu, C.; Gai, Z.; Krcmar, M.; Baddorf, A.; Zhang, Z.; Shen, J. Phys. Rev. Lett. 2003, 91, 226106. (29) King, D. A.; Woodruff, D. P. The Chemical Physics of Solid Surfaces; Elsevier Science B.V.: The Netherlands, 1997.
ASSOCIATED CONTENT
S Supporting Information *
Materials and Methods, Table S1, Figure S1−S7, and lattice mismatch calculations. This material is available free of charge via the Internet at http://pubs.acs.org.
■
Letter
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Author Contributions
The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes
The authors declare no competing financial interest. E
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
Nano Letters
Letter
(30) Turek, I.; Kudrnovský, J.; Carva, K. Phys. Rev. B 2012, 86, 174430.
F
dx.doi.org/10.1021/nl5030485 | Nano Lett. XXXX, XXX, XXX−XXX
