Values related to thin film oxidation for Fe, Cr, and Ni (Table S1) and for the bulk diffusion couples shown in Figure 6 (Table S2-Table S6) were taken from the literature. Note that in compiling values from multiple references for the diffusion data we have been mindful of not reporting the same data twice, as sometimes values for the prefactor (D 0 ) and the activation energy (E a ) were within very close agreement but with slight differences in temperature or one of the variables.The validity of the application of CM to the oxidation of metallic NPs can be further explored by considering the limiting oxide thickness experimentally observed in previous studies. Reported limiting oxide layer thicknesses for Fe, Ni, and Cr nano-interfaces have been reported to be 2-3 nm, 1-3 1-3 nm, 4-6 and ~ 3nm, 7 respectively. Both Cr and Ni approach this limiting rate at 3 nm, but Fe does approach a limiting thickness implying that the values of W, N, and V M need re-assessment. Interestingly, all of these W's are significantly less than those calculated from diffusion studies of bulk materials; for example, an average W for Fe selfdiffusion from a compilation of bulk experiments was found to be 2.54 eV, while Fehlner's fitting of thin film oxidation found a value of 0.8 eV. 8 Encouragingly, other researchers looking at the oxidation of Fe 3 O 4 NPs to the corresponding Fe 2 O 3 products obtained W in the range of 0.85-0.91 eV, 9,10 implying the use of values lower than bulk measurements is appropriate. Using 3 nm as the limiting thickness when the oxidation rate reaches 10 -5 Å/s, one can simulate the range of reasonable values for W with varying V M (Fig S6 ): taking Fe as our example, if we use V M = 1.6 V, the corresponding range of activation energies would be 1.2 -1.6 eV. However, if we fix W = 0.8 eV, as Fehlner did, the corresponding V M would be < 0.1 V, which is exceedingly small and difficult to reconcile with the high potentials in the Cr and Ni system. It is likely, then, that in our approximation the activation energy for Fe 2+ transport is larger than the 0.8 eV value.

In this work we describe a novel method to prepare multi-domain metal-metal oxide heterostructured nanoparticles (NPs). We investigated the ability of Ni(acac) 2 to undergo galvanic exchange with presynthesized metallic a-Fe NPs. Findings indicate that an asymmetric heterostructure emerges from the exchange, which is followed by rapid oxidation to form a NiFe-M 3 O 4 (M = Fe, Ni) alloy-oxide microstructure. Nucleation and growth was monitored using UV-vis and TEM, and crystal evolution, composition change, and oxidation states were measured by XRD and XPS respectively. Galvanic exchange and growth was studied by varying Ni : Fe molar feed ratios during synthesis. The findings indicate that at low Ni : Fe ratios, the NP forms multiple domains of oxides, whereas at higher ratios form regions with novel Ni-NiFe-M 3 O 4 interfaces. These new heterostructures were highly magnetic, and the extent of magnetization was proportional to composition and morphology, where NPs prepared at high Ni : Fe feed ratios resulted in decreased saturation magnetization and increased magnetic hysteresis. The nickel deposition and NP growth mechanism was considered as a combination of both galvanic exchange and reduction, and the observed rapid oxidation of the remaining a-Fe core was considered in light of electron density change at the heterostructures interfaces. Nanomaterials like these may find use as components in magnetic storage and spintronic devices, probes in biomedicine, additives in corrosion resistant coatings, and even as 3D printing inks.

This work explores how to form and tailor the alloy composition of Fe/FexNi1-x core/alloy nanoparticles by annealing a pre-formed particle at elevated temperatures between 180 – 325 oC. This annealing allowed for a systematic FeNi alloying at a nanoparticle whose compositions and structure began as a alpha-Fe rich core, and a thin gamma-Ni rich shell, into mixed phases resembling gamma-FeNi3 and gamma-Fe3Ni2. This was possible in part by controlling surface diffusion via annealing temperature, and the enhanced diffusion at the many grain boundaries of the nanoparticle. Lattice expansion and phase change was characterized by powder X-ray diffraction (XRD), and composition was monitored by X-ray photoelectron spectroscopy (XPS). Of interest is that no phase precipitation was observed (i.e., heterostructure formation) in this system and the XRD results suggest that alloying composition or alloy gradient is uniform. This uniform alloying was considered using calculations of bulk diffusion and grain boundary diffusion for Fe and Ni self-diffusion, as well as Fe-Ni impurity diffusion is provided. In addition, alloying was further considered by calculations for Fe-Ni mixing enthalpy (Hmix) and phase segregation enthalpy (HSeg) using the Miedema model, which allowed for the consideration of alloying favorability or core-shell segregation in the alloying, respectively. Of particular interest is the formation of stable metal carbides compositions, which suggest that the typically inert organic self-assembled monolayer encapsulation can also be internalized.