Materials & Design 213 (2022) 110369 Contents lists available at ScienceDirect Materials & Design journal homepage: www.elsevier.com/locate/matdes Lightweight macroporous Co-Pt electrodeposited films with semi-hard-magnetic properties Cristina Navarro-Senent a, Konrad Eiler a, Salvador Pané b, Jordi Sort a,c,⇑, Eva Pellicer a,⇑ a Departament de Física, Facultat de Ciències, Universitat Autònoma de Barcelona, E-08193 Bellaterra (Cerdanyola del Vallès), Barcelona, Spain b Institute of Robotics and Intelligent Systems, Swiss Federal Institute of Technology ETH Zurich, Tannenstraße 3, 8092 Zurich, Switzerland c Institució Catalana de Recerca i Estudis Avançats (ICREA), Pg. Lluís Companys 23, E-08010 Barcelona, Spain h i g h l i g h t s g r a p h i c a l a b s t r a c t  Equiatomic macroporous Co-Pt films obtained by electrodeposition on titanium surfaces patterned with latex spheres of 215 nm in diameter.  Annealing at 600 °C for 80 min results in coercivity enhancement from 148 to 1328 Oe while avoiding pore collapse.  Increase in coercivity due to partial transformation of the initial disordered face-centered cubic phase to ordered face-centered tetragonal phase.  Dense Co-Pt counterparts exhibit higher coercivity of 6.2 kOe upon annealing due to complete phase conversion. a r t i c l e i n f o a b s t r a c t Article history: Macroporous, partially L10-ordered Co-Pt films with nearly equiatomic composition were successfully Received 20 September 2021 synthesized by electrodeposition from an aqueous sulfate–chloride electrolyte on colloidal crystal- Revised 27 December 2021 templated substrates, followed by annealing in vacuum. The colloids deposited on the substrate consisted Accepted 27 December 2021 of amidine-functionalized polystyrene spheres of 215 ± 13 nm in diameter, which were self-assembled by Available online 29 December 2021 electrophoresis. As-deposited Co-Pt films obtained after the removal of the spheres showed a highly- packed arrangement of macropores. Structurally, the films showed the A1-disordered face-centered cubic Keywords: (fcc) Co-Pt solid solution, accompanied by small amounts of fcc/hexagonal close-packed (hcp)–Co. Upon Porous films Co-Pt alloys annealing at 600 °C, the A1-disordered phase partly transformed into the L10-ordered (face-centered Electrodeposition tetragonal, fct) phase. As a result, the coercivity significantly increased from 148 Oe to 1328 Oe. Magnetic properties Importantly, the porosity of the films was preserved after annealing. Optimum annealing temperature and time were selected on the basis of a prior parametric study with electroplated dense counterparts. This work demonstrates that the combination of colloidal crystal templating and electrodeposition is a very convenient pathway towards lightweight semi-hard magnets with potential technological applica- tions in automotive and aerospace industries, portable sensors or spectrometers, magnetic levitation sys- tems, or magnetoelectric devices, among others. Ó 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). ⇑ Corresponding authors at: Departament de Física, Facultat de Ciències, Universitat Autònoma de Barcelona, E-08193 Bellaterra (Cerdanyola del Vallès), Barcelona, Spain (J. Sort). E-mail addresses: Jordi.Sort@uab.cat (J. Sort), Eva.Pellicer@uab.cat (E. Pellicer). https://doi.org/10.1016/j.matdes.2021.110369 0264-1275/Ó 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 1. Introduction The vast majority of studies have focused on the electrodeposition of thick and thin films from electrolytes containing either the dini- Hard-magnetic materials are of fundamental interest for their tritodiammineplatinum (Pt(NH3)2(NO2)2) [24,30–34] or the hex- significant performance or key role in biomedical devices, motors, achloroplatinate salts [35,36], the Co precursor and a few generators, recording media and microelectromechanical systems additives. Although electrodeposition has proven effective for the (MEMS) [1]. Currently, there is an increasing demand for small size, fabrication of low-dimensional materials [37], the preparation of lightweight and low-cost hard magnets due to their potential miniaturized L10-ordered Co-Pt alloy in the form of nanostructured applications in spacecraft, aircraft and portable/wireless devices, films [38] and submicron/nanostructures such as nanoparticles among others [2,3]. For this reason, there has been a flurry of and pillars [39,40] has been scarcely reported. Notably, the influ- research on the development of hard-magnetic micro/nanomateri- ence of electrode surface (Ru, Cu, brass) has been investigated in als [4–8]. Rare-earth containing magnets (e.g. Nd-Fe-B, Sm-Co) are a number of studies, aimed at achieving large coercivities already known as the strongest hard-magnetic materials due to their high in the as deposited state [41]. remanent magnetization and maximum energy product (BH)max Several approaches have been proposed for the preparation of [1,9]. Unfortunately, rare-earth magnets are expensive and non- porous metallic films via electrodeposition, using (i) soft templates sustainable, suffer from high corrosion rates and their fabrication such as hydrogen bubbles [42,43] or block copolymers [44,45], and mainly relies on vacuum-based fabrication methods such as sput- (ii) hard templates like anodic alumina membranes [46,47]. Among tering or other physical vapor deposition techniques, which are these, colloidal crystal templating (also known as nanoscale lithog- known for their high complexity and cost. In addition, they are dif- raphy) is a cost-effective, straightforward and facile approach for ficult to be nanostructured or patterned into well-controlled the fabrication of 3D porous structures with controllable pore size. shapes, which limits their integration in real devices [1]. Electrode- This approach relies on the self-assembly of colloid particles of position of rare-earth alloy magnets is a good alternative to phys- varying size (typically from hundreds to tens of nanometers) onto ical deposition methods. However, their electrosynthesis from a substrate to create a pattern of colloids, which serve as a soft or aqueous baths is not optimal because of the associated low current hard template [48]. Several deposition techniques can be utilized efficiency and incorporation of oxygen and other impurities into to self-assemble the colloids on a surface, such as dip-coating the films, which heavily deteriorate the hard-magnetic properties [49], spin coating [50] or electrophoretic deposition [51]. Essen- [10–12]. Alternatively, the electrodeposition of rare-earth-based tially, the combination of two electrically driven processes, herein magnets (in the form of films and nanowires) from deep eutectic electrophoretic colloidal crystal templating followed by electrode- solvents has also been demonstrated [13]. Yet, the resulting coer- position, represents a versatile and powerful bottom-up method to civity values, mainly reported for the samarium-cobalt alloy, are fabricate pseudo-ordered macro- or mesoporous alloy films [52– much lower than those achieved with sputtered films [14,15]. 54]. From an industrial point of view, the two techniques are suf- As an alternative to rare-earth magnets, L10-ordered alloys have ficiently mature, which makes this approach compatible with a emerged as candidate permanent magnets owing to their high wealth of manufacturing approaches [55,56]. Colloidal crystal tem- magnetocrystalline anisotropy and large coercivity. L10-ordered plating could be also combined with physical vapor deposition alloys consist of equiatomic compositions of iron-series transitions techniques, but shadowing effects that create discontinuities in metals (e.g. Fe or Co) and heavy transition metals (e.g. Pt or Pd), the film would then be encountered. Moreover, since electrodepo- which form an asymmetric face-centered tetragonal (fct) structure sition works at ambient pressure, it is generally more cost (also known as L10 phase) [16,17]. Generally, the formation of L10- effective. ordered alloys (e.g. Co-Pt, Co-Pd, Fe-Pt and Fe-Pd) requires either a The introduction of porosity into hard magnets, such as L10- high-temperature deposition or a post-deposition annealing of the ordered alloys, opens a new route to develop advanced magnetic as-deposited A1-disordered cubic phase to induce structural trans- materials which could find applications in several technological formation into the L10-ordered phase. As opposed to rare-earth fields. Given their large (BH)max and porous structure, these mate- magnets, L10-ordered alloys can be prepared more easily by elec- rials have a great potential for applications which require low- trochemical deposition, both in the form of continuous thin/thick density magnets such as spacecraft, automobile or microbial fuel films and micro-/nanostructures, thus facilitating their implemen- cell technologies [57,58]. Also, the possibility to develop magnetic tation in micro-/nanosystems [18,19]. Although one could argue composites using porous hard magnets as a scaffold to be filled that platinum group metals (PGM) are as expensive as rare-earth with other materials having different physical properties is of tech- metals and therefore do not represent a cheaper alternative, the nological relevance. For instance, the pores could incorporate a supply risk of the former is lower compared to light rare-earth ele- soft-magnetic phase (yielding an exchange spring magnet mate- ments such us neodymium and samarium, according to the ‘‘Study rial) or could be filled with an antiferromagnetic material, giving on the EU’s list of Critical Raw Materials” final report launched by rise to a magnetic composite exhibiting exchange bias [59], a phe- the European Commission in 2020 [20]. Therefore, L10-ordered nomenon commonly exploited in spintronic systems, magnetic alloys certainly represent a valid alternative. random-access memories and magnetic read heads [60]. Co-Pt is among the most investigated L10-ordered systems. It Herein, we present the fabrication of macroporous Co-Pt thin exhibits a high uniaxial magnetocrystalline anisotropy (4.9 erg films via colloidal crystal template-assisted electrodeposition. A cm 3) and (BH)max of 200 kJ m 3 [21]. L10-ordered Co-Pt alloys close-packed colloid assembly of charged amidine polystyrene are usually fabricated by physical vapor deposition [22,23] or elec- (PS) spheres with a nominal diameter of 200 nm was obtained trochemical deposition techniques [24–26], mostly in the form of on a titanium surface by electrophoretic deposition. Subsequently, dense thick and thin films owing to the high temperature treat- equiatomic Co-Pt alloy was grown by potentiostatic electrodeposi- ment required for the L10-ordered phase transformation. High- tion from a simple, additive-free aqueous electrolyte. Finally, selec- temperature annealing is known to cause interdiffusion or even tive etching of the spheres yielded a macroporous, nearly the collapse, for example, of arrays of substrate-supported 1D equiatomic Co-Pt alloy film. For the sake of comparison, fully dense nanostructures like nanowires and nanorods [27,28]. films were deposited on unpatterned titanium surfaces. Both fully From a manufacturing perspective, electrodeposition of L10- dense and macroporous Co-Pt films were subjected to a parametric ordered Co-Pt alloys has garnered significant attention because it study to determine the optimal conditions (temperature and time) offers good shape fidelity and easy scalability at low cost [29]. to maximize the extent of conversion of the as-deposited fcc phase 2 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 into the L10-ordered (fct) phase, which was followed by both graz- the beads was 13.2 mC cm 2. Remarkably, positively charged PS ing incidence X-ray diffractometric (GIXRD) and vibrating sample spheres were selected in order to be able to negatively polarize magnetometric (VSM) studies. the Ti conductive layer and avoid its passivation during elec- trophoretic deposition, which otherwise would happen under pos- itive polarization. An aqueous solution of 0.05% w/v of PS spheres 2. Materials and methods suspension was added to the custom cell. The positively charged amidine PS spheres were deposited by applying an electric field 2.1. Cell and electrodes preparation strength of + 40 V cm 1 during 5 min. Subsequently, the samples were dried on a hot plate at 80 °C for 10 min to evaporate the Electrodeposition was carried out in a one-compartment, ther- water. mostated three-electrode cell connected to a PGSTAT204 Autolab potentiostat/galvanostat (Metrohm-Autolab). A double-junction Ag|AgCl (E= +0.210 V/SHE) reference electrode (Metrohm AG) with 2.2. Electrodeposition of Co-Pt 3 M KCl inner solution and 1 M Na2SO4 outer solution was used. A platinum spiral served as the counter electrode. All potentials The electrolyte for Co-Pt deposition consisted of 0.002 M Na2- quoted throughout the text are referred to Ag|AgCl (3 M KCl). Cyc- PtCl66H2O (Merck, 98% purity), 0.2 M CoSO47H2O (Merck,  95% lic voltammetry (CV) curves were recorded on a glassy carbon elec- purity) and 0.2 M Na2SO4 (Merck,  99% purity). The salts were dis- trode of 0.0314 cm2 working area. For the preparation of dense Co- solved in Milli-Q (MQ) water (18.2 MX cm). The pH was adjusted Pt thin films, flat Si/Ti (150 nm) substrates were used as working to 2.88 by adding a few drops of 1 M H2SO4 solution to the elec- electrodes. Titanium was chosen as the metallization layer in order trolyte. Deposition was conducted potentiostatically at 0.55 V to provide electrical conductivity while minimizing or avoiding during 420 s on both unpatterned and patterned Si/Ti substrates, atomic interdiffusion in the Co-Pt thin films during post- under mild agitation (x = 100 rpm) while simultaneously bubbling annealing treatment. It was observed that if gold was used as the N2 into the solution to de-aerate the electrolyte (Fig. 1b(i)). The metallization layer, it diffused towards the Co-Pt alloy upon heat temperature was maintained at 25 °C by circulating water treatment. The working area for electrodeposition was measured throughout the external jacket of the electrochemical cell by to be 0.25 ± 0.01 cm2. For the growth of macroporous Co-Pt thin means of an F12 Julabo thermostat. After deposition, the resulting films, Si/Ti (150 nm) substrates coated with amidine- dense thin films were rinsed with MQ water and dried in air. In the functionalized PS spheres were used. case of deposition onto the electrophoretically patterned sub- Electrophoretic deposition of the PS spheres was performed strates, the PS spheres were selectively etched away by immersing using an Agilent B2902A power supply as voltage source and a the samples in tetrahydrofuran overnight under agitation home-made 3D-printed cell consisting of poly(methyl methacry- (x = 400 rpm) (Fig. 1b(ii)). The resulting macroporous films where late) (PMMA) chamber attached to a counter electrode made of thoroughly cleaned with acetone and isopropanol, rinsed with MQ platinized titanium sheet (anode). Si/Ti (150 nm) substrates (cath- water and finally dried with N2. ode) were fixed at 0.5 cm from the counter electrode (Fig. 1a). Pos- itively charged amidine-terminated PS sub-micrometer spheres of 2.3. Post-deposition heat treatment 200 nm nominal diameter were purchased from ThermoFisher. According to the manufacturer specifications, the particles of the Heat treatments of the as-deposited films were carried out in a acquired batch had a diameter of 220 ± 15 nm, as estimated by home-made set-up consisting of a high-vacuum system connected transmission electron microscopy. The surface charge density of to a quartz tube located inside a 653.03 MTS high-temperature fur- Fig. 1. (a) Schematic drawing depicting the electrophoretic deposition of PS spheres. By applying an electric field strength of 40 V cm 1, positively charged amidine- terminated PS spheres move towards the cathode, resulting in the assembly of the spheres onto the Si/Ti substrate. (b) Schematic representation of colloidal crystal template- assisted electrodeposition: (i) The colloidal crystal layer is used as template for the electrodeposition of Co-Pt alloy, (ii) the PS spheres are removed, leaving the macroporous Co-Pt film behind. 3 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 nace. Annealing at different temperatures (550, 600 and 650 °C) and variable times (20, 60, 80 and 100 min) was performed under vacuum (pressure < 10 6 mbar). First, the furnace was heated up to the desired temperature using a ramp rate of 5 °C min-1. Then, the quartz tube containing the sample under vacuum was inserted into the furnace and kept there for certain periods (20, 60, 80 and 100 min). Finally, the furnace was left to cool down to room temperature. 2.4. Characterization A diluted suspension of PS spheres was deposited dropwise on a silicon substrate and finely metallized for imaging on a Zeiss Mer- lin field-emission scanning electron microscope (FE-SEM). For the sake of comparison with manufacturer specifications, 60 particles were counted to determine the mean size and standard deviation. The morphology and composition of the dense and macroporous Co-Pt films were characterized on the same microscope, which was equipped with an energy dispersive X-ray (EDX) detector. EDX analyses were carried out at 15 keV. GIXRD measurements were performed on a Bruker-AXS, model A25 D8 Discover equipped with a LinxEye XE-T detector using Cu Ka radiation and a grazing incidence angle of 1°. XRD patterns were Rietveld-fitted using the MAUD software [61] to extract the volume percentage of phases, crystallite size and microstrains. X-ray photoelectron spectroscopy (XPS) analyses were carried out on a PHI 5500 Mul- titechnique System spectrometer from Physical Electronics with a monochromatic X-ray source Al Ka line of 1486.6 eV energy and 350 W power, placed perpendicular to the analyzer axis and cali- brated by the 3d5/2 line of Ag with a full width at half maximum (FWHM) of 0.8 eV. The analyzed area was a circle of about 0.8 mm in diameter. Magnetic measurements were carried out using a VSM from Micro sense (LOT–Quantum Design). Hysteresis loops were recorded at room temperature along the in-plane direc- tion, with a maximum applied magnetic field of 20 kOe. Magneti- zation versus temperature curves were recorded under an applied Fig. 2. (a) CV curves recorded under stationary conditions at 50 mV s-1 on vitreous magnetic field of 10 kOe from 300 K to 800 K under argon atmo- carbon electrode from the optimized sulfate–chloride electrolyte (with 0.002 M Na2PtCl66H2O, 0.2 M CoSO4 7H2O and 0.2 M NaSO4, pH = 2.88; blue continuous sphere to prevent from sample oxidation. Samples were attached line) and the electrolyte with halved concentrations of the salts (0.001 M Na2- to the quartz sample holder using an alumina-based bonding cera- PtCl66H2O, 0.1 M CoSO47H2O and 0.1 M NaSO4, pH = 2.88; red dashed line). (b) j-t mic cement (Resbond 901 adhesive from Cotronics Corp.) that curves recorded from the optimized electrolyte on flat Si/Ti substrates (black, withstands 1650 °C. The corrosion behavior of the films was stud- magenta and green curves) under stirring conditions, at the indicated applied potentials. The corresponding Pt content (determined by EDX) in the resulting films ied by potentiodynamic polarization (PDP) at room temperature in is also given. The blue curve corresponds to the deposition on the patterned Si/Ti de-aerated 3.5 wt% NaCl solution. The films were first immersed in substrate at 0.55 V, for which the current was normalized to the geometric area of the NaCl solution for 30 min to determine the open circuit poten- the substrate. (For interpretation of the references to colour in this figure legend, tial (OCP). Immediately afterwards, the potential was swept from – the reader is referred to the web version of this article.) 300 mV below the OCP towards 1.5 V beyond the OCP at 0.5 mV s 1 under stagnant conditions. these salts (blue continuous line). Both CVs were recorded on a glassy carbon electrode under stagnant conditions and showed a 3. Results and discussion similar profile. In the negative sweep, the different reduction peaks can be attributed to the discharge of Pt(IV) ions, followed by Co(II) The bath utilized for the deposition of Co-Pt alloy was adapted co-deposition and hydrogen co-evolution [63]. In the anodic scan, a from a formulation previously devised for the deposition of Fe-Pt main oxidation peak centered at –0.015 V and a much weaker peak magnets [18]. It is an additive-free electrolyte, which simply con- at more positive potentials (for the blue curve) were recorded. tains sodium hexachloroplatinate and cobalt sulfate salts as elec- According to the CV fingerprint, cathodic potentials close to the troactive species and sodium sulfate to increase bath onset of Co and Pt co-deposition (roughly around –0.5 V) should conductivity. This bath formulation had not been used before for be screened for subsequent potentiostatic deposition of nearly the synthesis of L10-ordered Co-Pt alloys. The use of complexing equiatomic alloy compositions on Si/Ti substrates (i.e., to avoid agents and other additives was avoided on purpose to avoid the the growth of Co-rich deposits). Deposits grew on the flat Si/Ti sub- inclusion of impurities, which could deteriorate the magnetic strates at potentials of –0.5 V and more negative when deposition properties. Impurities in electrodeposited magnets form non- from the diluted electrolyte was attempted, but the deposition rate magnetic species with iron group metals, hence diluting mag- was very low. For this reason, the bath from doubled concentration netism within the films [62]. Fig. 2a shows the CV curves recorded of the salts was selected for subsequent experiments, since it from an electrolyte with the same concentrations of salts as those yielded deposits at a higher deposition rate. Following a parametric reported in [18] with the replacement of FeSO4 by CoSO4 (red study of the Co-Pt composition dependence on the deposition dashed curve), and an electrolyte with doubled concentrations of potential (Fig. 2b), a constant potential value of 0.55 V was cho- 4 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 Fig. 3. Representative low and high magnification FE-SEM images of equiatomic, as-deposited (a) dense and (b) macroporous Co-Pt films deposited at 0.55 V. sen for preparing Co-Pt deposits with equiatomic composition on oxygen content is lower than that found in nearly equiatomic, unpatterned Si/Ti substrates. Deposition was conducted potentio- dense, Fe-Pt layers electrodeposited from acidic aqueous solutions, statically since (i) the real working area of the colloidal crystal which is around 25–30 at.% [64]. There is probably not a sole rea- templated substrates used for the deposition of the macroporous son to explain why the here-deposited Co-Pt films show limited Co-Pt counterparts could not be precisely determined, (ii) the oxygen incorporation. First of all, iron electrochemistry is particu- active surface area changes as the process proceeds, and (iii) such larly tricky and the inclusion of moderately large amounts of oxy- processes are usually driven at a relatively low rate as compared to gen in iron and iron alloys electrodeposited from acidic or neutral the deposition on flat substrates because of the sluggish transport electrolytes is quite common [65,66]. Secondly, unless Fe(II) is conditions. The blue curve in Fig. 2b corresponds to the j-t curve complexed via the addition of, e.g. ascorbic acid, Fe(II) oxidation recorded during Co-Pt deposition on the patterned Si/Ti substrates. to Fe(III) takes place in the electrolyte, which eventually leads to Note that the current was normalized by the geometric area of the the precipitation of ferric hydroxides [67]. substrate. If the real working area could be determined, more neg- The increase in both the Co and the oxygen contents in the ative current densities would have been registered. macroporous films can be attributed to regions of increased cur- Fig. 3 shows representative FE-SEM images of the as-deposited, rent density at the edges of the template. In addition, the macrop- equiatomic, macroporous Co-Pt films and their dense counterparts. orous films exhibit a larger surface area. It is highly likely that the Dense films exhibit fine submicrometer grains with round-shaped morphology and small cracks over the entire surface (Fig. 3a). Cracks would appear due to sample drying and incipient hydrogen co-evolution. Fig. 3b shows the FE-SEM images of the macroporous Co-Pt films after PS spheres removal. The films displayed an architecture consisting of nearly a monolayer of hemispherical randomly arranged pores. Interestingly, these were virtually crack-free, as opposed to the dense counterparts. The pore size (200 nm) matched the diameter of the PS spheres. For comparison purposes, Fig. 4 shows a FE-SEM image of PS spheres deposited on a silicon substrate, whose diameter was estimated as 215 ± 13 nm, hence close to the manufacturer value (220 ± 15 nm). Cross-sectional measurements of the Co-Pt coatings revealed a thickness of 201 ± 7 nm for dense films and 116 ± 4 nm for macroporous thin films. Elemental compositional analysis disregarding oxygen revealed a composition of 49 ± 4 at.% Co and 51 ± 4 at.% Pt for the fully dense films, whereas macroporous films showed a com- position of 52 ± 5 at.% Co and 48 ± 5 at.% Pt. Moreover, a relatively Fig. 4. FE-SEM image of the PS spheres utilized to produce macroporous Co-Pt low concentration of oxygen was found in both cases, namely 9 at. coatings. A diluted aqueous suspension of the spheres was prepared and deposited % for the dense films and 15 at.% for the macroporous films. The dropwise on a silicon substrate. 5 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 small Co grains in these films tend to passivate and form an oxide was used instead of standard Bragg-Brentano geometry because layer when they are exposed to air, similar to what was observed in of low deposit thickness (200 nm), i.e., to avoid or minimize a previous study on the magnetic properties of macroporous Co the otherwise intense signal coming from the substrate. For the films [68]. sake of clarity, only the GIXRD patterns corresponding to samples GIXRD analyses were performed to determine the crystallo- annealed under optimal conditions, i.e. 600 °C for 100 min for the graphic phases of dense and macroporous Co-Pt films both before dense films, and 600 °C for 80 min for the macroporous Co-Pt coun- and after annealing (Fig. 5a and b respectively). Note that GIXRD terparts, are shown. These are optimal annealing conditions as it Fig. 5. XRD patterns of (a) the equiatomic dense Co-Pt films in the as-prepared state (black curve) and after being annealed at 600 °C for 100 min (red curve), and (b) the macroporous Co-Pt films in the as-prepared state (black curve) and after being annealed at 600 °C for 80 min (red curve). The peaks denoted with (+) and (*) belong to fcc-CoO and cobalt oxyhydroxides/oxides, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.) Fig. 6. (a) Dependence of HC on temperature for a fixed annealing time of 60 min for the equiatomic dense Co-Pt films. (b) Dependence of HC on annealing time for a fixed temperature of 600 °C. (c) In-plane hysteresis loops of the as-prepared dense Co-Pt films (black curve) and annealed at 600 °C for 20, 60, 80, 100 and 120 min. 6 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 Table 1 Summary of morphological, structural (main phases) and magnetic properties of as prepared and annealed dense and macroporous Co–Pt films. < D > stands for crystallite size, and vol.% for volume percentage of phase. Sample morphology A1 Co–Pt L10 Co–Pt fcc Co MR/MS HC (Oe) (nm) vol.% microstrains (nm) vol.% microstrains (nm) vol.% microstrains As-prepared dense dense, nodular 5 100 8.510–3 – – – – – – 0.47 85 Annealed dense grains, cracks – – – 16 100 2.110–3 – – – 0.84 6212 (600 °C, 100 min) As-prepared porous rounded pores, 8 92 5.610–3 – – – 16 8 2.510–4 0.47 148 Annealed porous crack-free 10 40 4.510–4 9 60 1.610–4 – – – 0.62 1328 (600 °C, 80 min) will be demonstrated later in the magnetic characterization part which matches the position of fcc-CoO phase. Compared to the (Figs. 6 and 9, respectively). fully dense counterpart, Rietveld fitting of the diffractogram (figure In the GIXRD patterns of the as-prepared dense Co-Pt films, two S1) revealed that a fraction of untransformed A1 phase (40 vol.%) wide main peaks, located at 41.0° and 47.4°, that matched the remained in the coating (Table 1). In addition, the cubic intermetal- (1 1 1) and (2 0 0) peaks of the fcc phase of equiatomic Co-Pt (i.e., lic CoPt3 phase (PDF# 029–0499) could also be present, having its the so-called A1-disordered phase), were observed [38]. Further- main peaks at 40.5° and 47.1° angular positions, as indicated in more, the crystallite size obtained from Rietveld fitting (see Fig- Fig. 5(b). This phase is typically formed in non-equiatomic Co-Pt ure S1 from S.I.) was about 5 nm (Table 1). After annealing at alloys with deviations of the order of 5 at.% [70]. The annealed 600 °C for 100 min, the corresponding GIXRD patterns showed nar- macroporous samples remained nanocrystalline, with crystallite rower peaks at 24.2°, 33.4°, 41.6° and 48.2°, which are assigned to sizes of 10 nm and 9 nm for the fcc and fct phases, respectively the fct L10-ordered phase. EDX analyses did not show signs of Ti (Table 1). seed-layer interdiffusion, in contrast to other studies in which 14 The experimental I(1 1 1)/I(2 0 0) ratio for the as-prepared dense at.% of Au was detected in Co-Pt films on annealing Co-Pt/Au/Si film (2.7) is close to that of the theoretical fcc-Co50Pt50 phase with samples annealed at 500 °C [69]. The crystallite size of the fct randomly oriented grains (2.5). The same is true for the as- phase was 16 nm (Table 1). These results prove that the chosen prepared macroporous film (I(1 1 1)/I(2 0 0) = 2.6). Therefore, the annealing conditions efficiently promote the rearrangement of Co fcc phase is not or little textured. Meanwhile, although the (1 1 1) and Pt atoms from the A1-disordered phase into the chemically reflection is also theoretically the most intense reflection of the ordered L10 structure. Hosoiri et al. achieved conversion of the fct phase (PDF no. 00–043-1358), the annealed macroporous film chemically disordered fcc phase into the fct L10-ordered in is slightly textured along the (2 0 0) plane. The theoretical I(1 1 1)/ 35 lm-thick electroplated Co-Pt films with 51.1 at.% Pt at 450 °C I(2 0 0) ratio of the isotropic L10 phase is 2.7 and the corresponding for 60 min [35]. In turn, Oniku et al. applied higher annealing tem- ratios for the dense and macroporous films are 2.7 and 2.0, peratures (675–700 °C) for 30 min in H2/N2 atmosphere [24,30] respectively. and these were optimum conditions for equiatomic Co-Pt films As-deposited, dense Co-Pt thin films were magnetically soft up to 6 lm in thickness to fully convert to L10 phase. In our work, with a coercivity (HC) of 85 ± 5 Oe. A first parametric study to maximum conversion of thinner (0.2 lm) dense films was determine the optimal annealing conditions was performed on achieved at an intermediate temperature (600 °C) held during a the fully dense films (Fig. 6). First, the temperature was optimized longer time period (100 min). The diffractograms of the as- for a fixed annealing time of 60 min, as seen in Fig. 6a. As the deposited macroporous Co-Pt films show two main peaks located annealing temperature was made higher, HC values increased, sug- at 41.6° and 48.0°, which are compatible with the existence of an gesting that the disordered phase gradually converted into the L10- A1-disordered solid solution slightly richer in Co than the equia- ordered phase. The HC reached the highest value of 4.8 kOe when tomic composition [35]. Additionally, the pattern shows a small the dense films were annealed at 600 °C but it dropped to 2.0 peak at 44.4° matching the position of fcc/hcp-Co and a peak at kOe when the annealing temperature was set at 650 °C. From the 37.5° (marked with an asterisk) that can be attributed to Co oxyhy- observed results, it was concluded that 600 °C was the optimum droxides/oxides. Considering that the chemical composition of the temperature for the structural conversion from the as deposited macroporous films was 52 ± 5 at.% Co and 48 ± 5 at.% Pt, a solid fcc to the fct phase. Later, annealing time was optimized while solution richer in Pt should also be present. According to PDF# keeping the optimum temperature of 600 °C fixed (Fig. 6b and c). 004–0802, pure fcc-Pt has (1 1 1) and (2 0 0) reflections at 39.8° On increasing the annealing time from 20 to 100 min, HC increased and 46.2°, respectively. With a small amount of Co dissolved in monotonically from 1.06 kOe after 20 min to 6.2 kOe after 100 min, the fcc-Pt lattice, the corresponding reflections would shift and then dropped to 2.8 kOe when the film was annealed for towards higher 2h angles, hence partly overlapping with the reflec- 120 min, as shown in Fig. 6b. Similar observations were made by tions of the A1-disordered solid solution above mentioned. The other authors which attributed the decline in coercivity to grain films were also nanocrystalline, with a crystallite size of 8 nm for growth [70,71]. Our maximum coercivity values are in agreement the main fcc-Co-Pt phase and 16 nm for the minor pure Co phase, with other studies devoted to the electrodeposition of Co-Pt and as obtained from Rietveld fitting of the XRD pattern (Table 1 and Fe-Pt films followed by annealing in vacuum to induce the fcc to figure S1). The occurrence of various phases in the films can be fct transformation. For example, Yanai et al. achieved HC values explained by the absence of a complexing agent in the electrolyte of 8.5 kOe on average with annealing equiatomic Fe-Pt thick films and the more sluggish mass-transfer conditions compared to the [72]. Yet, higher coercivities have been reported for nanocrystalline unpatterned substrates. electrodeposited Fe-Pt films upon annealing under a reducing Upon annealing, the corresponding XRD patterns showed peaks atmosphere, on the order of 11–14 kOe [73,74]. located at 23.7°, 33.0°, 41.1° and 47.6° which matched the fct L10- In contrast to as-deposited dense films, annealed counterparts ordered phase. These peaks, though, are shifted towards smaller showed squarer hysteresis loops with a squareness ratio closer to angles (by  0.5°) indicating enrichment in Pt. A small peak is 1, which is a characteristic of hard-magnetic materials (Fig. 6c). observed at 42.6° (denoted with a ‘‘+” symbol in the diffractogram), Moreover, all the annealed films showed a nearly single-phase 7 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 magnetic character. From these results, it was concluded that time increased and, most importantly, porosity was preserved in 600 °C and 100 min were the optimal annealing conditions to all the samples after the heat treatment. This is opposite to what achieve maximum conversion of the as-obtained phase into the it had been observed in mesoporous Co-Pt films (20-nm pore size) L10-ordered one for the dense Co-Pt thin films. Interestingly, no prepared by micelle-assisted electrodeposition [76], where anneal- L10 ordering was observed in electrodeposited Co-Pt films ing at 600 °C for 60 min resulted in a collapse of the mesopores, annealed up to 500 °C by Rozman et al. [75], who observed that giving rise to non-porous deposits. Thus, the selection of the proper the formation of the fct phase started upon annealing at 600 °C pore size is important in this kind of material to avoid pores col- for 60 min and maximum conversion was achieved at 700 °C lapsing during heat treatments, which are aimed for the semi (1.17 T), hence under conditions similar to our work. hard-magnetic phase evolution. Fig. 7 shows the FE-SEM images of the dense Co-Pt films before Fig. 9a shows the evolution of HC as a function of the annealing and after annealing at 600 °C for 100 min. Interestingly, no pro- time (keeping Tannealing = 600 °C) for the macroporous Co-Pt thin nounced changes in morphology were observed after heat treat- films. As the annealing time increased, HC augmented progressively ment. Yet, grains became a bit less rounded and the width of the and reached a maximum of 1328 Oe after 80 min. Afterwards, HC cracks increased (Fig. 7b), which could be attributed to recrystal- dropped to 770 Oe when the annealing time was extended to lization induced by the annealing process. 100 min. The increase in HC with the annealing time was primarily Concerning macroporous Co-Pt thin films, they are magnetically explained by the extent of the conversion from fcc into fct phase. soft in the as-prepared state, with a slightly higher HC (148 ± 9 Oe) Rietveld fitting of the XRD pattern of the film heat-treated for compared to the fully dense films, probably due to the hindrance 20 min showed the presence of 43 vol.% of L10 phase (see Figure S2 imposed by the pore walls on the propagation of magnetic domain in the S.I. file), whereas this percentage increased to 60 vol.% for walls. Since the annealing temperature had been previously opti- the macroporous coating treated for 80 min (Table 1). In any case, mized for the dense films, we decided to subject the macroporous the increase in HC was much less pronounced than for the fully thin films to the same temperature for varying annealing times. dense counterparts. First, the as-deposited macroporous films Fig. 8 shows the corresponding in-plane hysteresis loops and were structurally different from the dense counterparts and con- the FE-SEM images of the resulting annealed films. As the anneal- sisted of a mixture of phases, as evidenced by GIXRD. Secondly, ing time was increased, wider loops were obtained, analogous to the transformation resulted in a mixture of nearly equiatomic fcc what was observed for the dense films. In addition, the macrop- Co-Pt phase, L10-ordered Pt-rich Co-Pt alloy and CoO phases. orous Co-Pt film annealed at 600 °C shows a two-phase magnetic Meanwhile, the percentage of L10-ordered phase for the macrop- behavior (Fig. 8c), which is particularly obvious for an annealing orous coating was 60 vol.%, whereas for the dense film full conver- time of 80 min. This could be anticipated from the GIXRD analysis, sion was achieved (Table 1). As can be seen in Fig. 9b, the which revealed the co-existence of several phases (Pt-rich Co-Pt macroporosity was preserved across the film surface after the heat L10-ordered and soft magnetic CoPt3 besides the nearly equiatomic treatment. Roughly half empty spheres are seen from the cross- fcc phase). Ying et al. also observed a shoulder in the hysteresis section (top panel in Fig. 9c), which corresponds to half the diam- loops upon annealing electrodeposited Fe-Pt films at 800 °C, which eter of the spheres (215 nm) and approximately matches the mea- the authors attributed to the coexistence of the soft-magnetic fcc sured film thickness (116 nm). Yet, less rounded macropores were phase and the hard-magnetic L10 phase [74]. Regarding the mor- found after annealing which could be attributed to the deformation phology, no significant variations were observed as the annealing of the pore wall (cf. Fig. 3a and 8d). The same can be said in terms Fig. 7. Low and high magnification FE-SEM images of the equiatomic dense Co-Pt thin films (a) before and (b) after annealing at 600 °C for 100 min. 8 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 Fig. 8. In-plane hysteresis loops of the as-prepared macroporous Co-Pt films (black curve) and after annealing at 600 °C for (a) 20 (b) 60 (c) 80 and (d) 100 min and their corresponding FE-SEM images. 9 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 of film thickness for the annealed dense counterpart (bottom panel in Fig. 9c), which is added here for the sake of comparison. Finally, no pronounced differences were observed between hysteresis loops measured along different angles (from in-plane to perpendicular-to-plane directions) (Fig. 10). Only a slight reduc- tion in the squareness ratio (MR/MS) and a slightly more tilted loop shape were observed for the dense films (from MR/MS = 0.84 along the film plane to MR/MS = 0.73 along the perpendicular-to-plane direction) (Fig. 10a). The porous films exhibit a very isotropic mag- netic response (Fig. 10b). This can be due to the polycrystalline character of the films and the eventual role of porosity in disrupt- ing the in-plane geometry of the films. The magnetization versus temperature curves of the annealed films show a progressive decrease of magnetization till 800 K, at which the magnetization Fig. 10. Angular dependence of the normalized hysteresis loops for annealed (a) dense and (b) macroporous Co-Pt films. is almost zero (Figure S3 in the S.I. file). This agrees with the reported Curie temperature (TC) for L10 and fcc Co-Pt alloys (TC in the range 750 – 900 K) [77,78]. Fig. 11 shows SEM-EDX mappings of the annealed dense and porous samples. Specimens were tilted inside the SEM chamber to capture signal from both on top and cross-sections. The distribu- tion of Co and Pt elements was homogeneous in both samples and the O signal was a bit higher for the macroporous film, in agree- ment with the findings from the as-deposited state. Note that oxy- gen signal has two sources (the Co-Pt layer itself and the substrate beneath, as revealed after mapping a naked substrate). The Ti and Si signals are particularly elevated in the region covered by the fer- romagnetic macroporous film, which is due to the fact that map- ping was done on one edge of the specimen, and the porous character of the Co-Pt film. Considering the presence of oxygen in the coatings, XPS analy- ses were carried out to probe the oxidation state of Co in the films. The Co 2p core-level binding energies of 778 and 780 eV corre- spond to metallic and oxidized Co, respectively (Fig. 12). Note that Fig. 9. (a) Dependence of HC on annealing time for a fixed temperature of 600 °C for all common cobalt oxides and hydroxides exhibit very similar the macroporous Co-Pt thin films, (b) Low magnification top-view, and (c) cross- binding energies around 780 eV, so that an exact determination section FE-SEM images of (top panel) annealed macroporous nearly equiatomic Co- of the oxidized species is not possible and may even include Co Pt film at 600 °C for 80 min and (bottom panel) annealed dense counterpart at (III) (such as in CoOOH) [79]. Notably, metallic Co is the main sig- 600 °C for 100 min. The red semicircles in panel (c) show the profile of half empty spheres, which is the cross-sectional pattern left in the material following removal nal and the presence of cobalt oxides can be largely ascribed to sur- of the 215 nm-sized spheres. (For interpretation of the references to colour in this face passivation. For completeness, the O1s signal is given in figure legend, the reader is referred to the web version of this article.) Figure S4 of the S.I. file. After sputtering the films for 1 min with 10 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 Fig. 11. SEM image and corresponding EDX mappings (Co, Pt, O, Si, and Ti) taken on the annealed (a) dense and (b) macroporous films. Ar+ to remove the outmost passivation layer, the oxygen signal declined . For the sake of clarity, the morphological, structural and mag- netic properties of as deposited and annealed dense and macrop- orous films at the optimal temperatures and times are gathered in Table 1. Although annealing does not endow the macroporous Co-Pt films with hard-magnetic properties, the films are crack- free. The HC values seem to largely correlate with the extent of fcc to fct conversion, this being complete for dense films and par- tial for macroporous counterparts. As a result, HC increases from 148 to 1328 Oe for the macroporous films, and from 85 to 6212 Oe for the dense films. Therefore, the annealed macroporous films can be regarded as semi-hard ferromagnetic. The microstrains were of the order of 10-3-10-4, which are typical values for elec- trodeposited films [80]. Fig. 12. Co 2p core-level XPS spectra for as prepared and annealed macroporous and dense films. Annealing was conducted at the optimal conditions (600 °C, Fig. 13. PDP curves recorded in de-aerated 3.5 wt% NaCl solution for the as 100 min for dense film and 600 °C, 80 min for macroporous film). deposited and annealed Co-Pt films. 11 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 Table 2 Acknowledgements Ecorr and jcorr values extracted from the PDP curves of Fig. 13. Ecorr (mV vs. Ag/AgCl) jcorr (lA cm 2 ) This work was funded by the Spanish Government (Project As-deposited dense film 47 16 MAT2017-86357-C3-1-R, PID2020-116844RB-C21 and associated Annealed dense film 92 3.8 FEDER), the Generalitat de Catalunya (2017-SGR-292), and the As-deposited macroporous film 111 5.9 European Research Council (SPIN-PORICS 2014-Consolidator Annealed macroporous film 109 2.2 Grant, Agreement n. 648454). Finally, electrochemical corrosion tests were carried out to dis- Data availability close, if so, differences in the corrosion behavior of the coatings. Fig. 13 shows the PDP curves of the as-deposited and annealed The raw/processed data required to reproduce these findings Co-Pt films in 3.5 wt.% NaCl solution. The samples showed similar cannot be shared at this time as the data also forms part of an Ecorr values in the range between 50 mV and 111 mV. Interestingly, ongoing study. The data that support the findings of this study the macroporous samples showed slightly more positive Ecorr val- are available from the corresponding authors on request. ues than their dense counterparts. The naked Si/Ti substrate gave a significantly more negative Ecorr value of –131 mV, hence the PDP curves are quite representative of the behavior of the Co–Pt Appendix A. Supplementary data films. The jcorr values were in the range of a few lA cm 2 (see Table 2). Importantly, the corrosion resistance of the macroporous Supplementary data to this article can be found online at films did not deteriorate in spite of the remarkable increase in the https://doi.org/10.1016/j.matdes.2021.110369. surface-to-volume (S/V) ratio. Yet, the real surface area of the macroporous films is unknown and, hence, the current density val- References ues are not fully meaningful. Vallés and co-workers measured Ecorr values between 130 mV and + 30 mV vs. Ag/AgCl in 5.0 wt.% NaCl [1] T.-S. Chin, Permanent magnet films for applications in microelectromechanical for Co-Pt deposits depending on their crystalline structure (fcc or systems, J. Magn. Magn. Mater. 209 (1-3) (2000) 75–79, https://doi.org/ hcp) and Pt content (39 wt.% 77 wt.%) [81]. Hence, our values 10.1016/S0304-8853(99)00649-6. [2] A. Philip, J.-P. Niemelä, G.C. Tewari, B. Putz, T.E.J. Edwards, M. Itoh, I. Utke, M. are in agreement with the literature. Karppinen, Flexible e-Fe2O3-terephthalate thin-film magnets through ALD/ MLD, ACS Appl. Mater. Interfaces 12 (19) (2020) 21912–21921, https://doi.org/ 4. Conclusions 10.1021/acsami.0c0466510.1021/acsami.0c04665.s001. [3] R.A. Murphy, J.R. Long, T.D. Harris, A hard permanent magnet through molecular design, Commun. Chem. 4 (2021) 70, https://doi.org/10.1038/ Macroporous, semi-hard-magnetic Co-Pt thin films were pre- s42004-021-00509-y. pared by colloidal crystal template-assisted electrodeposition from [4] E.A. Kervendal, D.R. Kirk, R.B. Meinke, Spacecraft radiation shielding using ultralightweight superconducting magnets, J. Spacecr. Rockets 46 (5) (2009) an aqueous sulfate–chloride electrolyte followed by annealing. A 982–988, https://doi.org/10.2514/1.37490. parametric study of the optimal annealing conditions taking elec- [5] M. Gich, L.l. Casas, A. Roig, E. Molins, J. Sort, S. Suriñach, M.D. Baró, J.S. Muñoz, trodeposited dense counterparts as a reference revealed that a L. Morellon, M.R. Ibarra, J. Nogués, High-coercivity ultralight transparent magnets, Appl. Phys. Lett. 82 (24) (2003) 4307–4309, https://doi.org/10.1063/ maximum HC value of 1328 Oe was achieved after annealing at 1.1578538. 600 °C for 80 min. Importantly, the selected patterning method [6] M. Janulevicius, V. Klimkevičius, L. Mikoliunaite, B. Vengalis, R. Vargalis, S. (colloidal crystal templating) and bead size (215 nm) secured the Sakirzanovas, V. Plausinaitiene, A. Zilinskas, A. Katelnikovas, Ultralight magnetic nanofibrous GdPO4 aerogel, ACS Omega 5 (23) (2020) 14180– preservation of porosity in the Co-Pt films upon annealing. Other 14185, https://doi.org/10.1021/acsomega.0c0198010.1021/ patterning approaches and/or smaller pore sizes could result in acsomega.0c01980.s001. pore collapse. GIXRD analyses evidenced the polycrystalline struc- [7] N. Chen, Q. Pan, Versatile fabrication of ultralight magnetic foams and application for oil–water separation, ACS Nano 7 (8) (2013) 6875–6883, ture of the as-deposited macroporous films, which showed a nearly https://doi.org/10.1021/nn4020533. equiatomic A1-disordered Co-Pt solid solution with some addi- [8] Y. Li, Q. Liu, A.J. Hess, S. Mi, X. Liu, Z. Chen, Y. Xie, I.I. Smalyukh, Programmable tional minor phases. In turn, the existence of different magnetic ultralight magnets via orientational arrangement of ferromagnetic and crystallographic phases (hard-magnetic Pt-rich L10-ordered nanoparticles within aerogel hosts, ACS Nano 13 (12) (2019) 13875–13883, https://doi.org/10.1021/acsnano.9b04818. and soft-magnetic A1-disordered and CoPt3 phases) was seen in [9] D.P. Arnold, N. Wang, Permanent magnets for MEMS, J. the annealed macroporous films. Potentiodynamic polarization Microelectromechanical Syst. 18 (2009) 1255–1266, https://doi.org/10.1109/ curves were indicative of the noble character of the films. The JMEMS.2009.2034389. [10] J. Chen, L. Rissing, Electroplating hard magnetic SmCo for magnetic increase of the S/V ratio brought by the introduction of porosity microactuator applications, J. Appl. Phys. 109 (7) (2011) 07A766, https://doi. in the Co-Pt films was not deleterious for the corrosion resistance. org/10.1063/1.3565414. This study demonstrates that the combination of electrodeposition [11] M. Xie, Electrodeposition of Sm-Co alloy films with nanocrystalline/ amorphous structures from a sulphamate aqueous solution, Int. J. and colloidal templating technique is a suitable strategy for fabri- Electrochem. Sci. 12 (2017) 11330–11342, https://doi.org/10.20964/ cating semi-hard-magnetic porous (and hence, more lightweight) 2017.12.07. films. These materials are appealing as low-density magnets or [12] L. Kang, C. Cui, W. Yang, Y. Zhang, M. Guo, The properties and microstructure of Nd-Fe-B nanowires fabricated by electrochemical deposition using porous magnetic composites with a great potential for implementation alumina templates, Mater. Chem. Phys. 242 (2020) 122470, https://doi.org/ in spacecraft, automobile or magnetoelectric devices. This syn- 10.1016/j.matchemphys.2019.122470. thetic strategy could be well extended to other L10-ordered alloys [13] X. Xu, S. Sturm, J. Zavasnik, K.Z. Rozman, Electrodeposition of a rare-earth iron alloy from an ionic-liquid electrolyte, ChemElectroChem 6 (11) (2019) 2860– (i.e., Co-Pd, Fe-Pt and Fe-Pd). Also, because of the presence of noble 2869, https://doi.org/10.1002/celc.201900286. metals and large S/V ratio, the macroporous films could be used for [14] P. Cojocaru, L. Magagnin, E. Gomez, E. Vallés, Using deep eutectic solvents to electrocatalytic purposes like the hydrogen evolution reaction. electrodeposit CoSm films and nanowires, Mater. Lett. 65 (23-24) (2011) 3597–3600, https://doi.org/10.1016/j.matlet.2011.08.003. [15] G. Panzeri, M. Tresoldi, C. Rinaldi, L. Magagnin, Electrodeposition of magnetic Declaration of Competing Interest SmCo films from deep eutectic solvents and choline chloride-ethylene glycol mixtures, J. Electrochem. Soc. 164 (13) (2017) D930–D933, https://doi.org/ The authors declare that they have no known competing finan- 10.1149/2.0111714jes. [16] A. Kashyap, R. Skomski, A.K. Solanki, Y.F. Xu, D.J. Sellmyer, Magnetism of L10 cial interests or personal relationships that could have appeared compounds with the composition MT (M=Rh, Pd, Pt, Ir and T=Mn, Fe Co, Ni), J. to influence the work reported in this paper. Appl. Phys. 95 (11) (2004) 7480–7482, https://doi.org/10.1063/1.1687631. 12 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 [17] R. Skomski, Phase formation in L10 magnets, J. Appl. Phys. 101 (9) (2007) highly active electrocatalysts, Chem. Mater. 23 (9) (2011) 2457–2465, https:// 09N517, https://doi.org/10.1063/1.2711801. doi.org/10.1021/cm200382s. [18] F.M.F. Rhen, J.M.D. Coey, Electrodeposition of coercive L10 FePt magnets, J. [45] B.O. Jiang, C. Li, V. Malgras, M. Imura, S. Tominaka, Y. Yamauchi, Mesoporous Pt Magn. Magn. Mater. 322 (9-12) (2010) 1572–1575, https://doi.org/10.1016/j. nanospheres with designed pore surface as highly active electrocatalyst, jmmm.2009.09.027. Chem. Sci. 7 (2) (2016) 1575–1581, https://doi.org/10.1039/C5SC03779D. [19] Y.H. Huang, H. Okumura, G.C. Hadjipanayis, D. Weller, CoPt and FePt [46] H. Wang, H.Y. Jeong, M. Imura, L. Wang, L. Radhakrishnan, N. Fujita, T. Castle, nanowires by electrodeposition, J. Appl. Phys. 91 (10) (2002) 6869, https:// O. Terasaki, Y. Yamauchi, Shape- and size-controlled synthesis in hard doi.org/10.1063/1.1447524. templates: sophisticated chemical reduction for mesoporous [20] European Commission (2020) 3. accessed 10 December 2021 https://op. monocrystalline platinum nanoparticles, J. Am. Chem. Soc. 133 (37) (2011) europa.eu/en/publication-detail/-/publication/c0d5292a-ee54-11ea-991b- 14526–14529, https://doi.org/10.1021/ja2058617. 01aa75ed71a1/language-en. [47] M. Pérez-Page, E. Yu, J. Li, M. Rahman, D.M. Dryden, R. Vidu, P. Stroeve, [21] N.M. Dempsey, Hard magnetic materials for MEMS applications, in: J.P. Liu, E. Template-based syntheses for shape controlled nanostructures, Adv. Colloid Fullerton, O. Gutfleisch, D.J. Sellmyer (Eds.), Nanoscale Magnetic Materials and Interface Sci. 234 (2016) 51–79, https://doi.org/10.1016/j.cis.2016.04.001. Applications, Springer, US, 2009, pp. 661–683. DOI:10.1007/978-0-387- [48] S. Rasouli, N. Rezaei, H. Hamedi, S. Zendehboudi, X. Duan, Superhydrophobic 85600-1_22 and superoleophilic membranes for oil-water separation application: A [22] M.R. Visokay, R. Sinclair, Direct formation of ordered CoPt and FePt compound comprehensive review, Mater. Des. 204 (2021) 109599, https://doi.org/ thin films by sputtering, Appl. Phys. Lett. 66 (1995) 1692–1694. DOI: 10.1063/ 10.1016/j.matdes.2021.109599. 1.113895 [49] Z.-Z. Gu, A. Fujishima, O. Sato, Fabrication of high-quality opal films with [23] R.A. Ristau, K. Barmak, L.H. Lewis, K.R. Coffey, J.K. Howard, On the relationship controllable thickness, Chem. Mater. 14 (2) (2002) 760–765, https://doi.org/ of high coercivity and L10 ordered phase in CoPt and FePt thin films, J. Appl. 10.1021/cm0108435. Phys. 86 (1999) 4527–4533. DOI: 10.1063/1.371397 [50] P. Jiang, T. Prasad, M.J. McFarland, V.L. Colvin, Two-dimensional nonclose- [24] O.D. Oniku, B. Qi, D.P. Arnold, Electroplated thick-film cobalt platinum packed colloidal crystals formed by spincoating, Appl. Phys. Lett. 89 (1) (2006) permanent magnets, J. Magn. Magn. Mater. 416 (2016) 417–428, https://doi. 011908, https://doi.org/10.1063/1.2218832. org/10.1016/j.jmmm.2016.05.044. [51] A.L. Rogach, N.A. Kotov, D.S. Koktysh, J.W. Ostrander, G.A. Ragoisha, [25] A. Garraud, O.D. Oniku, D.P. Arnold, Influence of temperature on the magnetic Electrophoretic deposition of latex-based 3D colloidal photonic crystals: a properties of electroplated L1°CoPt thick films, J. Appl. Phys. 117 (2015) technique for rapid production of high-quality opals, Chem. Mater. 12 (2000) 17C718, https://doi.org/10.1063/1.4913890. 2721–2726, https://doi.org/10.1021/cm000274l. [26] J. Ewing, Y. Wang, D.P. Arnold, High-current-density electrodeposition using [52] P.V. Braun, P. Wiltzius, Electrochemically grown photonic crystals, Nature 402 pulsed and constant currents to produce thick CoPt magnetic films on silicon (6762) (1999) 603–604, https://doi.org/10.1038/45137. substrates, AIP Adv. 8 (5) (2018) 056711, https://doi.org/10.1063/1.5007272. [53] D.W. Bruce, D. O’Hare, R.I. Walton, Porous Materials, John Wiley & Sons, UK, [27] S.-J. Wang, C.-H. Chen, S.-C. Chang, K.-M. Uang, C.-P. Juan, H.-C. Cheng, Growth Inorganic Materials Series, 2010. and characterization of tungsten carbide nanowires by thermal annealing of [54] A. Nicolenco, Y.u. Chen, N. Tsyntsaru, H. Cesiulis, E. Pellicer, J. Sort, Mechanical, sputter-deposited WCx films, Appl. Phys. Lett. 85 (12) (2004) 2358–2360, magnetic and magnetostrictive properties of porous Fe-Ga films prepared by https://doi.org/10.1063/1.1791322. electrodeposition, Mater. & Design 208 (2021) 109915, https://doi.org/ [28] G. Shukla, S. Angappane, Dimensional constraints favour high temperature 10.1016/j.matdes.2021.109915. anatase phase stability in TiO2 nanorods, Appl. Surf. Sci. 577 (2022) 151874, [55] A. Baux, G. Couégnat, G.L. Vignoles, D. Lasseux, A. Kuhn, C. Carucci, N. Mano, T. https://doi.org/10.1016/j.apsusc.2021.151874. D. Le, Digitization and image-based structure-properties relationship [29] H.H. Lou, Y. Huang, Electroplating, in: S. Lee (Ed.), Encyclopedia of Chemical evaluation of a porous gold micro-electrode, Mater. & Des. 193 (2020) Processing, Taylor & Francis, New York, 2006, pp. 839–848. 108812, https://doi.org/10.1016/j.matdes.2020.108812. [30] O.D. Oniku, B. Qi, D.P. Arnold, Electroplated L1°CoPt thick-film permanent [56] R. Ramachandramoorthy, M. Mieszala, C.V. Manzano, X. Maeder, J. Michler, L. magnets, J. Appl. Phys. 115 (2014) 17E521, https://doi.org/10.1063/1.4867119. Philippe, Dual-templated electrodeposition and characterization of regular [31] F.M.F. Rhen, E. Backen, J.M.D. Coey, Thick-film permanent magnets by metallic foam based microarchitectures, Appl. Mater. Today 20 (2020) 100667, membrane electrodeposition, J. Appl. Phys. 97 (11) (2005) 113908, https:// https://doi.org/10.1016/j.apmt.2020.100667. doi.org/10.1063/1.1923587. [57] F. Li, D. Wang, Q. Liu, B. Wang, W. Zhong, M. Li, K. Liu, Z. Lu, H. Jiang, Q. Zhao, C. [32] O. Berkh, Y.u. Rosenberg, Y. Shacham-Diamand, E. Gileadi, Electrodeposited Xiong, The construction of rod-like polypyrrole network on hard magnetic near-equiatomic CoPt thick films, Electrochem. Solid-State Lett. 11 (4) (2008) porous textile anodes for microbial fuel cells with ultra-high output power D38, https://doi.org/10.1149/1.2838047. density, J. Power Sources 412 (2019) 514–519, https://doi.org/10.1016/j. [33] Y. Wang, J. Ewing, D.P. Arnold, Ultra-thick electroplated CoPt magnets for jpowsour.2018.11.090. MEMS, J. Micro. Syst. 28 (2) (2019) 311–320, https://doi.org/10.1109/ [58] S.M. Jones, Aerogel: Space exploration applications, J. Sol-Gel Sci. Technol. 40 JMEMS.2018.2888846. (2-3) (2006) 351–357, https://doi.org/10.1007/s10971-006-7762-7. [34] T. Yanai, R. Hamamura, J. Honda, A. Tomita, H. Yamada, N. Fujita, K. Takashima, [59] J. Sort, J. Nogués, X. Amils, S. Suriñach, J.S. Muñoz, M.D. Baró, Room M. Nakano, H. Fukunaga, Electroplated Co–Pt thick-film magnets prepared in temperature coercivity enhancement in mechanically alloyed citric-acid-based plating baths, IEEE Trans. Magn. 53 (11) (2017) 1–3, https:// antiferromagnetic-ferromagnetic powders, Appl. Phys. Lett. 75 (20) (1999) doi.org/10.1109/TMAG.2017.2710351. 3177–3179, https://doi.org/10.1063/1.125269. [35] K. Hosoiri, F. Wang, S. Doi, T. Watanabe, Preparation and characterization of [60] J. Nogués, I.K. Schuller, Exchange bias, J. Magn. Magn. Mater. 192 (2) (1999) electrodeposited Co-Pt binary alloy film, Mater. Trans. 44 (4) (2003) 653–656, 203–232, https://doi.org/10.1016/S0304-8853(98)00266-2. https://doi.org/10.2320/matertrans.44.653. [61] L. Luterotti, P. Scardi, Simultaneous structure and size-strain refinement by the [36] V. Georgescu, M. Daub, Magnetism and magnetoresistance in electrodeposited Rietveld method, J. Appl. Crystallogr. 23 (1990) 246–252, https://doi.org/ (L10) CoPt superlattices, J. Optoelectron. Adv. Mater. 7 (2005) 853–858. 10.1107/S0021889890002382. [37] Y. Wang, R. Bowrothu, Y.-K. Yoon, D.P. Arnold, Patterning of thick electroplated [62] N.V. Myung, D.-Y. Park, B.-Y. Yoo, P.T.A. Sumodjo, Development of CoPt magnets using SU-8 micromoulds, Micro & Nano Lett. 14 (2019) 1393– electroplated magnetic materials for MEMS, J. Magn. Magn. Mater. 265 (2) 1396, https://doi.org/10.1049/mnl.2019.0287. (2003) 189–198, https://doi.org/10.1016/S0304-8853(03)00264-6. [38] F. Wang, K. Hosoiri, S. Doi, N. Okamoto, T. Kuzushima, T. Totsuka, T. Watanabe, [63] K. Eiler, S. Suriñach, J. Sort, E. Pellicer, Mesoporous Ni-rich Ni–Pt thin films: Nanostructured L1°Co–Pt thin films by an electrodeposition process, Electrodeposition, characterization and performance toward hydrogen Electrochem. Commun. 6 (2004) 1149–1152, https://doi.org/10.1016/ evolution reaction in acidic media, Appl. Catal. B Environ. 265 (2020) j.elecom.2004.09.009. 118597, https://doi.org/10.1016/j.apcatb.2020.118597. [39] N. Yasui, A. Imada, T. Den, Electrodeposition of (001) oriented CoPt L10 [64] K. Leistner, S. Oswald, J. Thomas, S. Fähler, H. Schlörb, L. Schultz, Potential columns into anodic alumina films, Appl. Phys. Lett. 83 (2003) 3347–3349, dependence of composition and structure of electrodeposited Fe–Pt films, https://doi.org/10.1063/1.1622787. Electrochim. Acta 52 (1) (2006) 194–199, https://doi.org/10.1016/ [40] A. Hannour, L. Bardotti, B. Prével, F. Tournus, D. Mailly, J.-P. Bucher, A. Nafidi, j.electacta.2006.04.058. Nanostructured L10-CoPt dot arrays with perpendicular magnetic anisotropy, [65] A. Nicolenco, N. Tsyntsaru, H. Cesiulis, Fe (III)-based ammonia-free bath for Mater. Lett. 193 (2017) 108–111, https://doi.org/10.1016/j.matlet.2017.01.114. electrodeposition of Fe-W alloys, J. Electrochem. Soc. 164 (9) (2017) D590– [41] D.-Y. Park, N.V. Myung, Magnetic properties of electrodeposited cobalt- D596, https://doi.org/10.1149/2.1001709jes. platinum (CoPt) and cobalt-platinum-phosphide (CoPtP) thin films, Front. [66] I. Golvano-Escobal, R. de Paz-Castany, C.C.J. Alcantara, S. Pané, E. Garcia-Lecina, Chem. 9 (2021), https://doi.org/10.3389/fchem.2021.733383 733383. J. Sort, E. Pellicer, Functional macroporous iron-phosphorous films by [42] Y. Li, W.-Z. Jia, Y.-Y. Song, X.-H. Xia, Superhydrophobicity of 3D porous copper electrodeposition on colloidal crystal templates, Electrochim. Acta 313 films prepared using the hydrogen bubble dynamic template, Chem. Mater. 19 (2019) 211–222, https://doi.org/10.1016/j.electacta.2019.05.009. (23) (2007) 5758–5764, https://doi.org/10.1021/cm071738j. [67] T. Müller, J. Grimwood, A. Bachmaier, R. Pippan, Electrodeposition of Fe-C [43] J. Zhang, M.D. Baró, E. Pellicer, J. Sort, Electrodeposition of magnetic, alloys from citrate baths: structure, mechanical properties, and thermal superhydrophobic, non-stick, two-phase Cu–Ni foam films and their stability, Metals 8 (2018) 363, https://doi.org/10.3390/met8050363. enhanced performance for hydrogen evolution reaction in alkaline water [68] I.N. Krivorotov, H. Yan, E.D. Dahlberg, A. Stein, Exchange bias in macroporous media, Nanoscale 6 (21) (2014) 12490–12499, https://doi.org/10.1039/ Co/CoO, J. Magn. Magn. Mater. 226–230 (2001) 1800–1802, https://doi.org/ C4NR03200D. 10.1016/S0304-8853(00)00934-3. [44] L. Wang, Y. Yamauchi, Strategic synthesis of trimetallic Au@Pd@Pt core shell [69] K.Ž. Rožman, J. Kovač, P.J. McGuiness, Z. Samardžija, B. Markoli, S. Kobe, nanoparticles from poly(vinylpyrrolidone)-based aqueous solution toward Microstructural, compositional and magnetic characterization of 13 C. Navarro-Senent, K. Eiler, S. Pané et al. Materials & Design 213 (2022) 110369 electrodeposited and annealed Co–Pt-based thin films, Thin Solid Films 518 (6) [76] C. Navarro-Senent, J. Fornell, E. Isarain-Chávez, A. Quintana, E. Menéndez, M. (2010) 1751–1755, https://doi.org/10.1016/j.tsf.2009.11.025. Foerster, L. Aballe, E. Weschke, J. Nogués, E. Pellicer, J. Sort, Large [70] X. Sun, Z.Y. Jia, Y.H. Huang, J.W. Harrell, D.E. Nikles, K. Sun, L.M. Wang, magnetoelectric effects in electrodeposited nanoporous microdisks driven by Synthesis and magnetic properties of CoPt nanoparticles, J. Appl. Phys. 95 (11) effective surface charging and magneto-ionics, ACS Appl. Mater. Interfaces 10 (2004) 6747–6749, https://doi.org/10.1063/1.1667441. (51) (2018) 44897–44905, https://doi.org/10.1021/acsami.8b1744210.1021/ [71] S. Jeong, Y.-N. Hsu, D.E. Laughlin, M.E. McHenry, Atomic ordering and acsami.8b17442.s001. coercivity mechanism in FePt and CoPt polycrystalline thin films, IEEE Trans. [77] DongEung Kim, J.E. Saal, L. Zhou, ShunLi Shang, Y. Du, Z.-K. Liu, Magn. 37 (2001) 1299, https://doi.org/10.1109/20.950823. Thermodynamic modeling of fcc order/disorder transformations in the CoPt [72] T. Yanai, J. Honda, R. Hamamura, Y. Omagari, H. Yamada, N. Fujita, K. system, CALPHAD 35 (3) (2011) 323–330, https://doi.org/10.1016/ Takashima, M. Nakano, H. Fukunaga, Improvement in surface conditions of j.calphad.2011.04.005. electroplated Fe-Pt thick-film magnets, AIP Adv. 8 (5) (2018) 056437, https:// [78] X.B. Liu, Z. Altounian, Exchange interaction in L10-ordered FePt and CoPt from doi.org/10.1063/1.5007196. first-principles, J. Appl. Phys. 109 (7) (2011) 07B762, https://doi.org/10.1063/ [73] K. Leistner, J. Thomas, H. Schlörb, M. Weisheit, L. Schultz, S. Fähler, Highly 1.3564953. coercive electrodeposited FePt films by postannealing in hydrogen, Appl. Phys. [79] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, Handbook of X-ray Lett. 85 (16) (2004) 3498–3500, https://doi.org/10.1063/1.1807958. photoelectron spectroscopy, Physical Electronics Inc, MI, 1995. [74] Y. Ying, H. Wang, J. Zheng, J. Yu, W. Li, L. Qiao, W. Cai, S. Che, Preparation, [80] E. Pellicer, A. Varea, S. Pané, K.M. Sivaraman, B.J. Nelson, S. Suriñach, M.D. Baró, microstructure, and magnetic properties of electrodeposited nanocrystalline J. Sort, Sort, A comparison between fine-grained and nanocrystalline L10 FePt films, J. of Supercon. Nov. Magn. 33 (11) (2020) 3563–3570, https:// electrodeposited Cu–Ni films. Insights on mechanical and corrosion doi.org/10.1007/s10948-020-05624-w. performance, Surf. Coat Technol. 205 (23-24) (2011) 5285–5293, https://doi. [75] K. Žužek Rožman, A. Krause, K. Leistner, S. Fähler, L. Schultz, H. Schlörb, org/10.1016/j.surfcoat.2011.05.047. Electrodeposition and hard magnetic properties of Co–Pt films in comparison [81] M. Cortés, E. Gómez, E. Vallés, Magnetic properties of nanocrystalline CoPt to Fe–Pt films, J. Magn. Magn. Mater. 314 (2) (2007) 116–121, https://doi.org/ electrodeposited films. Influence of P incorporation, J. Sol. State, Electrochem. 10.1016/j.jmmm.2007.02.146. 14 (12) (2010) 2225–2233, https://doi.org/10.1007/s10008-010-1055-3. 14