Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Vitrification of Liquid Metal‐in‐Oil Emulsions Using Nano‐Mineral Oxides

Vitrification of Liquid Metal‐in‐Oil Emulsions Using Nano‐Mineral Oxides IntroductionGa and Ga‐based alloys, such as EGaIn (75.5 wt% Ga and 24.5 wt% In) and Gallinstan (68.5 wt% Ga, 21.5 wt% In, and 10.0 wt% Sn), are referred to as “liquid metals (LMs)”[1] because of their low melting points (near or even below room temperature). These materials have low toxicity, in contrast with well‐known LM Hg, while possessing both high electrical conductivity (as “metals”) and high fluidity (as “liquids”).[2] These characteristics afford high potential in a wide variety of cutting‐edge technologies.[3–10] Although unmodified LMs are promising for various applications,[3,11,12] modification can extend the application scope.[13–27] LMs can be tailored into the form of colloidal microdroplets,[17–27] which enables their dispersion in another carrier liquid or in solidifiable matrices. Despite their high interfacial tension, the dispersed LM microdroplets can remain as discrete particles (without immediate coalescence) owing to the solid oxide layer that spontaneously forms on the outermost particle surface,[2] unless this layer is intentionally destroyed.[24,28–30] This oxide layer, known to be mostly Ga2O3,[2,31] enables shape transformation of the LM microdroplets,[19–21] surface functionalization,[32–35] adhesion to substrates,[2,36,37] and templated synthesis of other particulate matter.[10] Intuitively, an oxide passivation layer with an ultrawide bandgap (≈4.8 eV)[18] may seem undesirable in applications where electrical percolation via the core metal phase of the microdroplets is important; however, the surface oxide layer may be useful in certain classes of applications where fine tuning of the dielectric properties or electrical breakdown strength of the electronic devices embedding LMs is required.[38,39] Furthermore, the electronic properties of the oxide skin, in conjunction with the core metal phase, are prospectively key factors in the photothermal conversion and photocatalytic activity of LM microdroplets,[17,18,40,41] with significant implications in human health and environmental applications such as targeted therapeutics[9,23,42–44] and pollutant removal.[45] Thus, the unique tunability of the electronic properties by employing external oxide materials[17,18,46–48] calls for follow‐up research on various combinations of LMs and other particulate additives.LM microdroplets, when suspended as discrete particles with a solid oxide skin in a liquid medium, increase the viscosity (ɳ) of the resulting colloidal systems in a loading‐dependent manner, and even induce elasticity at high loadings.[24,25,49,50] Such rheological properties of colloidal LM suspensions (or “emulsions,” considering the liquid nature of the dispersed phase) may be a key parameter for successfully harnessing the aforementioned functions with various form factors in emerging manufacturing technologies, such as multidimensional printing.[51,52] Neumann et al.[25] attempted to load LM microdroplets into a liquid silicone medium up to a maximum concentration of 90 wt%, achieving a 3D‐printable colloidal LM emulsion with an estimated yield stress (τy) of ≈64 Pa. More concentrated LM emulsions could possess higher τy values of a few hundred pascals.[24,49] Such emulsions have successfully been patterned in stretchable electronics,[24] or used in the fabrication of 3D heat dissipation composites,[49] via the direct ink writing (DIW) technique. Moreover, adding small amounts of liquid bridges with adequate wettability on the oxide skin of LM microdroplets can induce capillary attraction between the LM microdroplets and further enhance the rheological stiffness of LM emulsions (up to τy values of a few thousand pascals), which may potentially enable novel manufacturing conditions.[50]One promising yet relatively underexplored method of tuning the rheological properties of LM emulsions is by adding external solid particulate matter (metals,[26] ceramics,[49] and so on) to LM emulsions. The rheological effects of such solid additives, which were directly added to or mixed with “bulk” LM (as a continuous phase or a matrix, not a dispersed phase such as LM microdroplets suspended in another liquid continuous phase), have been extensively studied.[13–16,53] The solid additives suspended in the LM matrix readily modify the intrinsic rheological properties of bulk LM (e.g., high surface tension and low viscosity) and often induce a unique fluid‐to‐solid transition, enabling versatile LM processing. The resulting amalgam also demonstrated enhanced “functionality” such as electromagnetic interference shielding[16] and thermal management[14] when appropriate external solid particles were used. Although not extensively studied, external solid particles (which are not reactive with the LM) in a colloidal LM emulsion (comprising LM microdroplets suspended in another liquid medium) are theoretically expected to increase the rheological strength of the overall colloidal system by increasing in the volume fraction (Φ) of the total dispersed phase.[54] Moon et al.[49] recently demonstrated that adding boron nitride (BN) fillers (up to ΦBN = 0.2) to a polymerizable organic matrix with highly loaded LM microdroplets (at ΦLM = 0.7) increased the elastic modulus and τy of the resulting suspensions by an order of magnitude. However, to the best of our knowledge, quantitative investigations of the rheological effects of external solid fillers on colloidal LM emulsions have been conducted only for this single type of filler. More extensive rheological studies on the effects of solid additives using a variety of well‐defined solid particles will be highly useful for understanding the complex behaviors of multicomponent and multiphase colloidal systems containing LM microdroplets with other engineered particles and their practical application in numerous emerging technologies.Herein, we demonstrate that the addition of nanosized mineral oxide (MO) particles readily vitrifies an apolar oil (O)‐based emulsion with suspended LM microdroplets, even at minute dosages (ΦMO < 0.01 and ΦLM+MO = 0.55). At high MO loadings (ΦMO ≥ 0.1), a high‐strength colloidal LM system with an unprecedentedly high τy value of ≈104 Pa is obtained, wherein a partial “sintering” effect derived from localized droplet coalescence is observed, attributed to the high internal stress within the sample. The mechanisms of enhancement of the rheological strength are discussed in terms of the polar interaction between the oxide surfaces submerged in an apolar or dispersive medium. The MO additives not only modify the rheology of LM emulsions, but also improve their thermophysical properties, which are potentially useful in various applications.Results and DiscussionThe base LM‐in‐oil emulsion with ΦLM = 0.55 and ΦO = 0.45 (Figure 1a), consisting of EGaIn (dispersed phase; ɳ ≈ 2 mPa s) and liquid paraffin (continuous phase; ɳ ≈ 20 mPa s), was prepared via probe sonication of the mixture upon incremental addition of LM to the bulk oil phase. The produced LM droplets were observed as polydispersed microspheres (<2 µm, Figure 1e), the shape of which was retained; the droplets did not undergo immediate coalescence, attributed to the solid oxide skin[2] that formed spontaneously on the droplet surface (inset in Figure 1e). Despite the high Φ of the dispersed phase (ΦLM = 0.55), the resulting emulsion exhibited relatively high fluidity (Figure 1a), consistent with prior studies.[25,50] To the as‐prepared LM‐in‐oil emulsion, silica (SiO2, 224.8 ± 105.0 nm; see electron microscopic image in Figure S1 in the Supporting Information) nanospheres were added as rheology modifiers. For the systematic rheology study, Φ for the total dispersed phase was the same as that of the base emulsion, that is, ΦLM+MO = 0.55. Primary LM‐in‐oil emulsions with ΦLM < 0.55 were first prepared, and appropriate amounts of MO particles were added and vortex‐mixed.1FigurePhotographs of a) base LM emulsion and LM–MO emulsions with b) ΦMO = 0.0075, c) ΦMO = 0.0375, and d) ΦMO = 0.15. The total volume fraction of the dispersed phase was fixed at ΦLM+MO = 0.55. SEM image of e) LM microdroplets (inset: TEM image near droplet surface) and f) LM–SiO2 aggregates covered by oils taken from the vitrified emulsion at ΦLM = 0.475 and ΦSiO2 = 0.075. g,h) TEM images of LM–SiO2 aggregates and i) that with elemental mapping by EDS.The apparent stiffness of the LM emulsions increased remarkably with an increase in ΦSiO2 when ΦLM+SiO2 was fixed at 0.55 (Figure 1b–d). (Note that the representative case, in which SiO2 was used as the MO additive, is shown here, but the overall trend was qualitatively similar for other MOs, as discussed hereinafter). There was no visible macroscopic segregation between the LM‐ and MO‐rich phases despite the high physicochemical dissimilarities between the two types of particles. Scanning electron microscopy (SEM) images of the SiO2‐added emulsions (see the representative image of the sample with ΦSiO2 = 0.075 in Figure 1f) revealed raspberry‐like, rough particles, which were likely aggregates of LM and SiO2. The liquid substances covering the aggregates are the oils, which could be removed by deliberate washing with hexane using a squeeze bottle (see the SEM image of the repeatedly washed sample in Figure S2 in the Supporting Information; see also the presence of an oil layer covering the particles in the transmission electron microscopy (TEM) image in Figure S3 of the Supporting Information). TEM imaging (Figure 1g,h) with energy‐dispersive spectroscopy (EDS) elemental mapping (Figure 1i) of the samples after gentle hexane washing revealed the compositional structures of these aggregates. In most images, numerous adhered LM–SiO2 assemblies, that is, LM microdroplets covered with adsorbed SiO2 particles (see also the images of similar aggregate structures with other MOs in Figure S4 of the Supporting Information), some of which resembled Pickering emulsion[55] droplets), were observed, suggesting a high LM–MO affinity. Given the general nonwettability between the “pristine” metal and MOs,[37,56] the observed LM–MO adhesion was plausibly mediated by the oxide skin formed on the outermost surface of the LM droplets.Although adhesion between oxide‐coated LMs and various types of surfaces, including MO in air, has extensively been studied,[36,37,57,58] adhesion in liquid medium has rarely been evaluated. To understand the interaction between the surfaces of the LM and MO submerged in the oil medium, the pull‐off force between the two adhered macroscopic surfaces was measured (see Figure S5a in the Supporting Information). For the measurements, a planar substrate, either a glass slide (as a mimic of the SiO2 surface) or a slide coated with LM, placed on the bottom sample mount was lifted at a constant velocity to touch an LM droplet suspended at the top, either in the air or oil environment. (Note that the LM droplet and the planar LM surface on the glass substrate were pre‐equilibrated for a few minutes under ambient conditions before the measurements, such that the oxide skin could form on the respective surfaces, which is known to occur within a timescale of microseconds).[59] After touching the suspended LM droplet, the substrate was again drawn down at a constant velocity, during which the pulling force between the two adhered surfaces (i.e., LM–SiO2 and LM–LM) was measured using a microbalance.[57,60,61] The negative forces recorded at the initial time periods (Figure S5b–e, Supporting Information) correspond to the loadings exerted by the substrate upon lifting upward, which continually decreased at t > 0 (where t = 0 corresponds to the time at which the two surfaces first touched) as the substrate was drawn downward. As the recorded force moved into the positive region across the zero value (Figure 2a–d), the two adhered surfaces that were forced to separate began to pull each other, and the measured maximum positive force (maximum pull‐off force or Fpull‐off) was used to compare the relative strength of adhesion between different pairs of adhered surfaces. As shown in Figure 2a,b, the Fpull‐off value measured in air for the LM–LM pair (the outermost surfaces of both were likely coated by oxide skins), that is, ≈3300 µN, was significantly higher than that for LM–SiO2, that is, ≈970 µN (the Fpull‐off for LM–SiO2 was close to the literature value obtained using a similar method).[57] Assuming a priori that the surface energy of the oxide skin of LM is of the same order of magnitude as that of SiO2, this higher Fpull‐off value for LM–LM compared to that of LM–SiO2 is plausibly be ascribed to the bulk LM phases underneath the oxide skins of the two adhered surfaces, which, despite the presence of a thin (oxide) passivation layer, could still contribute to the total attractive interaction between the surfaces in close proximity.[62] It is also possible that the LM released through the ruptured sites of the oxide layer made direct contact, increasing the measured Fpull‐off. However, the Fpull‐off values for both LM–LM and LM–SiO2 decreased significantly in the oil medium (Figure 2c,d), possibly because the LM and SiO2 surfaces were primarily covered by oil molecules adhered via dispersive (or London)[62] interactions, preventing direct adhesion between the oil‐submerged surfaces. Notably, the Fpull‐off value for LM–LM in the oil medium (≈22 µN; Figure 2c) was even lower than that for LM–SiO2 (≈54 µN; Figure 2d), which is opposite to the observations for the same pairs in air. The higher adhesion strength between LM and SiO2 in the oil medium, inferred from the higher Fpull‐off, compared to that between LM and LM, suggests that in the oil medium, LM–SiO2 adhesion may be energetically more favorable than LM–LM adhesion. This difference provides a possible explanation for the large number of LM–SiO2 assemblies observed in the microscopic images of the aggregates from the vitrified LM emulsion with added SiO2.2FigureMeasured adhesion forces for a) LM–LM and b) LM–SiO2 surfaces in air, and c) LM–LM and d) LM–SiO2 surfaces in oil.The adhesion strength between a pair of interacting surfaces can be studied theoretically, where the work of adhesion (i.e., the free energy gain upon the adhesion of two interacting surfaces) can be estimated using a surface thermodynamic model.[62] Here, the widely used Owens, Wendt, Rabel, and Kaelble (OWRK) method, based on classical Fowkes theory,[58,63,64] was adopted. In this theory, the surface energy (or surface tension in units of mJ m−2) of a condensed phase of species i, γi, is defined as the sum of the dispersive (γiD)$\left( {\gamma _{\rm{i}}^{\rm{D}}} \right)$ and polar (γiP$\gamma _{\rm{i}}^{\rm{P}}$) components1γi=γiD+γiP\[\begin{array}{*{20}{c}}{{\gamma _{\rm{i}}} = \gamma _{\rm{i}}^{\rm{D}} + \gamma _{\rm{i}}^{\rm{P}}}\end{array}\]The total work of adhesion between two surfaces of species 1 and 2 (W1 − 2; in vacuum or air) is defined as the sum of the dispersive (W1−2D$W_{1 - 2}^{\rm{D}}$) and polar (W1−2P$W_{1 - 2}^{\rm{P}}$) contributions2W1−2=W1−2D+W1−2P\[\begin{array}{*{20}{c}}{{W_{1 - 2}} = W_{1 - 2}^{\rm{D}} + W_{1 - 2}^{\rm{P}}}\end{array}\]where3W1−2D=−2γ1Dγ2D\[\begin{array}{*{20}{c}}{W_{1 - 2}^{\rm{D}} = - 2\sqrt {\gamma _1^{\rm{D}}} \sqrt {\gamma _2^{\rm{D}}} }\end{array}\]and4W1−2P=−2γ1Pγ2P\[\begin{array}{*{20}{c}}{W_{1 - 2}^{\rm{P}} = - 2\sqrt {\gamma _1^{\rm{P}}} \sqrt {\gamma _2^{\rm{P}}} }\end{array}\]By combining Equations (2)–(4) with the well‐known Young's equation and taking species 1 and 2 as the solid (S) and liquid (L) phases, respectively, the following relation is obtained5γL(cosθ+1)2γLD=γSPγLPγLP+γSD\[\begin{array}{*{20}{c}}{\frac{{{\gamma _{\rm{L}}}\left( {\cos \theta + 1} \right)}}{{2\sqrt {\gamma _{\rm{L}}^{\rm{D}}} }} = \frac{{\sqrt {\gamma _{\rm{S}}^{\rm{P}}} \sqrt {\gamma _{\rm{L}}^{\rm{P}}} }}{{\sqrt {\gamma _{\rm{L}}^{\rm{P}}} }} + \sqrt {\gamma _{\rm{S}}^{\rm{D}}} }\end{array}\]where θ is the contact angle of liquid on solid. Thus, in principle, γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for a solid surface of interest can be obtained by solving Equation (5) with the measured θ values of (at least) two probe liquid droplets with known γLD$\gamma _{\rm{L}}^{\rm{D}}$ and γLP$\gamma _{\rm{L}}^{\rm{P}}$.To estimate γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for the glass and LM surfaces that were used in Fpull‐off measurements, the θ values of five probe liquids, i.e., water, ethylene glycol, glycerol, dimethyl sulfoxide, and diiodomethane, with known γLD$\gamma _{\rm{L}}^{\rm{D}}$ and γLP$\gamma _{\rm{L}}^{\rm{P}}$ were measured on each surface (Table S1, Supporting Information). (Note that using this model that does not include the metallic surface energy component, that is, γiM$\gamma _{\rm{i}}^{\rm{M}}$, with a series of nonmetallic probe liquids does not enable the extraction of any information on the metallic properties of the LM surface. Here, we assumed an ideal LM surface that is completely coated by the oxide skin, where the underlying metal phase has no influence on the direct adhesion with the other surface in the oil medium; the latter was in part supported by the result of Fpull‐off measurement in oil, and will be further discussed later). Using Equation (5), the γSD$\gamma _{\rm{S}}^{\rm{D}}$ values were preliminarily calculated using the θ values of the purely dispersive probe liquid (diiodomethane); the γSP$\gamma _{\rm{S}}^{\rm{P}}$ values were inferred via linear regression using the fixed γSD$\gamma _{\rm{S}}^{\rm{D}}$ values and the θ values for the other four probe liquids. As summarized in Table 1, both the glass and LM surfaces possessed large γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ values. Note that γSP$\gamma _{\rm{S}}^{\rm{P}}$ for the LM surface was comparable to that for the glass, attributed to the oxide (most likely Ga2O3) layer that spontaneously formed at the outermost surface of the LM, given that the “pristine” metal would not possess the “polar” component.1TableγSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ values of surfaces investigated using the OWRK method, along with calculated work of adhesion valuesSurfaceSurface energy component[mJ m−2]Work of adhesion[mJ m−2]γSD$\gamma _{\rm{S}}^{\rm{D}}$γSP$\gamma _{\rm{S}}^{\rm{P}}$WS−oilD$W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}}$WS−oilP$W_{{\rm{S}} - {\rm{oil}}}^{\rm{P}}$WS − oilWS−oil-LMD$W_{{\rm{S}} - {\rm{oil - LM}}}^{\rm{D}}$WS−oil−LMP$W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$WS − oil − LMGlass49.416.8−76.30−76.3−5.2−33.2−38.4LM(on glass)49.816.4−76.60−76.6−5.3−32.8−38.1SiO2(pellet)49.116.1−76.10−76.1−5.1−32.5−37.7TiO2(pellet)45.318.4−73.10−73.1−4.2−34.8−39.0Al2O3(pellet)49.616.6−76.50−76.5−5.2−33.0−38.3ZnO(pellet)50.014.8−76.80−76.8−5.3−31.2−36.5Using the known γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for both surfaces, in the same theoretical framework, the work of adhesion of the respective surfaces with any counterpart with known γiD$\gamma _{\rm{i}}^{\rm{D}}$ and γiP$\gamma _{\rm{i}}^{\rm{P}}$ (in vacuum or air) can be calculated using Equations (2)–(4). When initially submerged in the purely dispersive oil medium, with γoil=γoilD+γoilP=29.5+0 = 29.5 mJ m−2${\gamma _{{\rm{oil}}}} = \gamma _{{\rm{oil}}}^{\rm{D}} + \gamma _{{\rm{oil}}}^{\rm{P}} = 29.5 + 0\; = \;29.5\;mJ\;{{\rm{m}}^{ - 2}}$, the adhesion between the respective surfaces and oil would take place first exclusively via the dispersive interaction, with WS−oil=−2γSDγoilD−2γSP0=WS−oilD${W_{{\rm{S}} - {\rm{oil}}}} = - 2\sqrt {\gamma _{\rm{S}}^{\rm{D}}} \sqrt {\gamma _{{\rm{oil}}}^{\rm{D}}} - 2\sqrt {\gamma _{\rm{S}}^{\rm{P}}} \sqrt 0 = W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}}$, the absolute value of which is an energy barrier to be overcome for the two solid surfaces to directly adhere to each other in the oil medium (Figure 3). It is notable that the WLM−oil=WLM−oilD${W_{{\rm{LM}} - {\rm{oil}}}} = W_{{\rm{LM}} - {\rm{oil}}}^{\rm{D}}$ of − 76.6 mJ m−2 for LM–oil was lower than the WSiO2−oil=WSiO2−oilD${W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}}}} = W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}}}^{\rm{D}}$ of − 76.3mJ m−2 for SiO2–oil, suggesting that the initial oil layer adhered more strongly on the LM surface than on the SiO2 surface. Taking the energy level for the adhered S–oil and LM–oil (the counterpart LM when S for the S–oil is also LM) to be zero, the work of adhesion for S–LM in oil can be given by6WS−oil−LM=WS−oil−LMD+WS−oil−LMP\[\begin{array}{*{20}{c}}{{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}} = W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} + W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}}\end{array}\]where7WS−oil−LMD=−WS−oilD−WLM−oilD+WS−LMD+Woil−oilD\[\begin{array}{*{20}{c}}{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} = - W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}} - W_{{\rm{LM}} - {\rm{oil}}}^{\rm{D}} + W_{{\rm{S}} - {\rm{LM}}}^{\rm{D}} + W_{{\rm{oil}} - {\rm{oil}}}^{\rm{D}}}\end{array}\]and8WS−oil−LMP=−WS−oilP−WLM−oilP+WS−LMP+Woil−oilP\[\begin{array}{*{20}{c}}{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}} = - W_{{\rm{S}} - {\rm{oil}}}^{\rm{P}} - W_{{\rm{LM}} - {\rm{oil}}}^{\rm{P}} + W_{{\rm{S}} - {\rm{LM}}}^{\rm{P}} + W_{{\rm{oil}} - {\rm{oil}}}^{\rm{P}}}\end{array}\]3FigureSchematic of energetics of two interacting surfaces submerged in apolar oil medium.As shown in Table 1, the polar work of adhesion (WS−oil−LMP≈−33mJ m−2)$\left( {W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}} \approx - 33mJ\;{{\rm{m}}^{ - 2}}} \right)$ for the two surfaces was dominant over the dispersive work of adhesion (WS−oil−LMD≈−5 mJ m−2)$\left( {W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} \approx - 5\;mJ\;{{\rm{m}}^{ - 2}}} \right)$, which suggests that the polar interaction is the main driving force for the adhesion between the two solid surfaces (with significant γSP$\gamma _{\rm{S}}^{\rm{P}}$) in the dispersive oil medium. Notably, the WSiO2−oil−LM${W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}}$ of −38.4 mJ m−2 for SiO2–LM in oil is lower than the WLM − oil − LM of −38.1 mJ m−2 for LM–LM in oil, which is qualitatively consistent with the results of the Fpull‐off measurements. Note that if the metallic work of adhesion was considered for LM–LM, that is, WLM−oil−LMM=WLM−LMM=−2γiM≪−600 mJ m−2$W_{{\rm{LM}} - {\rm{oil}} - {\rm{LM}}}^{\rm{M}} = W_{{\rm{LM}} - {\rm{LM}}}^{\rm{M}} = - 2\gamma _{\rm{i}}^{\rm{M}} \ll - 600\;mJ\;{{\rm{m}}^{ - 2}}$,[62,65] the LM–LM adhesion in the oil medium would always dominate LM–SiO2, which was obviously not the case, as shown in Figure 2c versus Figure 2d. Thus, it seems reasonable to assume that the metallic contribution to the interparticle interaction was likely suppressed in the oil medium, possibly because of the presence of the oxide skin and the adhered oil layer, and the overall interparticle interaction was dominated by that between the polar surface sites.Provided that surface charging and electrostatic interparticle repulsion are unlikely in an oil medium with a low dielectric constant (ε ≈ 2.2 for liquid paraffin),[31,66] the attraction[67–70] mediated by the polar surface sites would be the main colloidal force affecting the rheological behavior of the emulsion. The addition of nanosized MOs with high thermodynamic affinity for the oxide skin of the LM droplets, as well as a high specific surface area, was expected to increase the overall attractive interaction between the suspended particles, thereby enhancing the rheological strength of the emulsions. Furthermore, although the native oxide skin of the LM droplets was too thin to confer a high degree of surface rigidity, the adsorbed solid MO particles more effectively enhanced the mechanical rigidity of the droplet surface, contributing to the increase in the overall emulsion strength. (Note: Increases in the rheological strength of liquid‐in‐liquid emulsions by interfacially adsorbed solid particle stabilizers, i.e., Pickering emulsifiers, have also been suggested for systems consisting of more “ordinary” liquid dispersed phases, e.g., oil or water,[69,71,72] although the mechanism of particle adsorption at the liquid–liquid interface differs from that of the LM–MO adhesion discussed here).Figure 4a shows the elastic (G′) and viscous (G″) moduli of LM emulsions with various ΦSiO2 values at ΦLM+SiO2 = 0.55, obtained via oscillatory viscoelasticity measurements using a stress ramp at a constant frequency (1 Hz). The measured G′ and G″ values of the base emulsion (ΦLM = 0.55 with no added MO) were ≈103 and ≈102 Pa, respectively, in the linear viscoelastic region. Notably, replacing only ≈1 vol% of LM with SiO2 (i.e., ΦSiO2 ≈ 0.01 at ΦLM+SiO2 = 0.55) increased the G′ and G″ values to >104 and >103 Pa, respectively. Furthermore, by assigning the crossover point of the G′ and G″ curves as the yield point of the sample, it was inferred that the τy value of the base emulsion (≈10 Pa) increased by two orders of magnitude at ΦSiO2 ≈ 0.01. For example, a τy value of 802.1 ± 253.6 Pa was obtained for the sample with ΦSiO2 = 0.0075 (Figure 4c), which is of the same order of magnitude as that of the LM‐in‐oil emulsion with polymeric liquid‐bridged LM microdroplets.[50] The τy value systematically increased with a further increase in (Figure 4c), exceeding 104 Pa at ΦSiO2 ≥ 0.1, which is an unprecedented value for colloidal LM emulsions. The measurements often exceeded the instrumental limit at higher SiO2 loadings (data not shown). The size of the MOs affects the efficiency of emulsion vitrification. When ≈10 µm sized SiO2 particles were used, which were larger than the LM droplets (<2 µm), the emulsion still vitrified but at a much lower rate than the emulsion of nanosized SiO2 (≈200 nm) particles (Table S2, Supporting Information). This is because the nanosized MOs conferred a much larger total surface area, and the interaction with LM droplets was more effectively facilitated.4Figurea) Elastic (G′, solid lines) and viscous (G″, dotted lines) moduli of LM–SiO2 emulsions with various ΦSiO2 (data for the representative MO‐free pristine LM emulsion are illustrated by gray color; the numbers on top of G′ curves of LM–SiO2 emulsions indicate the corresponding ΦSiO2). b) G′ and G″ curves for LM–MO emulsions at a low ΦMO of 0.0037. c) Yield stress (τy) for LM–MO emulsions with various ΦMO (red spheres: SiO2, brown diamonds: TiO2, green stars: Al2O3, blue squares: ZnO, red stars: hydrophobically modified SiO2). In panels (a) and (b), the data were obtained via oscillatory measurements using stress ramp at a constant frequency of 1 Hz. In panel (c), the data were obtained by assigning the cross‐over points of the G′ and G″ curves. All the investigated sample emulsions had a fixed dispersed phase volume fraction of 0.55, i.e., ΦLM+MO = 0.55.Figure 4b,c shows an increasing trend for rheological parameters with other types of MO additives, including titania (TiO2, 120.3 ± 41.3 nm), alumina (Al2O3, 193.8 ± 72.6 nm), and zinc oxide (ZnO, 154.8 ± 64.6 nm) (the SEM images of the MOs and TEM images of the LM–MO aggregate structures are shown in Figures S1 and S4 in the Supporting Information, respectively). Despite differences in the particle shape and size distribution, all the investigated MOs have average sizes in the same order of magnitude, resulting in significant emulsion vitrification, although the rate was dependent on the type of MO used. The γMOD$\gamma _{{\rm{MO}}}^{\rm{D}}$ and γMOP$\gamma _{{\rm{MO}}}^{\rm{P}}$ values of these MOs (Table 1) were estimated by applying the OWRK method, for the macroscopic surfaces of the pelletized MOs prepared by compressing the MO powders at ≈ 382 MPa. (The measured θ values of the five probe liquids on these surfaces are listed in Table S1 in the Supporting Information) All the investigated MOs exhibited significant γMOD$\gamma _{{\rm{MO}}}^{\rm{D}}$ and γMOP$\gamma _{{\rm{MO}}}^{\rm{P}}$ values, comparable in magnitude to those of the glass substrate and the manually spread LM surface discussed above. Thus, the calculated WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ values for these MOs similarly dominated WMO−oil−LMD$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}}$, suggesting the importance of the polar contribution to the LM–MO interaction. The relative magnitudes of WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ were consistent with the order of the increasing rates of τy shown in Figure 4c (i.e., TiO2 > Al2O3 > ZnO), except in the case of SiO2, for which WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ was found to be (comparable to but) slightly higher than that of Al2O3. However, we note that the “Hamaker constant” (the prefactor for the van der Waals interaction) for LM–MO interaction in the oil medium, calculated using the Lifshitz theory and exclusively considering the dispersive contribution,[62] would predict an absurdly lower attraction between LM and SiO2 (or even “repulsion,” i.e., ASiO2−oil−LM≤0${A_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}} \le 0$) than that between LM and the other MOs (ATiO2−oil−LM≈1.0 ×10−19J${A_{{\rm{Ti}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}} \approx 1.0\; \times {10^{ - 19}}{\rm{J}}$, AZnO − oil − LM ≈ 5.2 × 10−20J, and AAl2O3−oil−LM≈3.3 ×10−20J${A_{{\rm{A}}{{\rm{l}}_2}{{\rm{O}}_3} - {\rm{oil}} - {\rm{LM}}}} \approx 3.3\; \times {10^{ - 20}}{\rm{J}}$; see more details in the Supporting Information). This is in contrast with the favorable LM–SiO2 assemblies observed in the electron microscopic images (Figure 1f–i) and the highest τy values of the LM–SiO2 emulsions at various ΦSiO2 values (Figure 4c). Thus, we suggest that the polar interparticle interaction may be a more relevant parameter than the dispersive interaction in the attraction‐driven rheological stiffening of LM–MO emulsions based on the dispersive medium. This type of interaction is also considered the main driving force for various self‐assembly processes in apolar media, such as the adsorption of surfactants in the “head‐down” configuration or the formation of inverse micelles.[66,73,74] As a control, SiO2 particles modified with hexadecyl triethoxysilane (SiO2@HDTS) were added to LM emulsions with various ΦSiO2@HDTS at a fixed ΦLM+MO of 0.55, as described above. Modification with HDTS completely altered the wettability of SiO2 by capping the polar sites of the bare SiO2 surface and covalently implanting apolar hydrocarbon chains.[75] The resulting SiO2@HDTS exhibited excellent dispersibility in the oil medium (Figure S6b, Supporting Information), in stark contrast with unmodified SiO2 (Figure S6a, Supporting Information). For the LM emulsions with added SiO2@HDTS, the increase in τy was much less pronounced (Figure 4c; Figure S6c, Supporting Information), attributed to screening of the polar interaction by the surface‐anchored apolar moieties, which corroborate the importance of polar interactions in emulsion vitrification.High shape retention of vitrified LM–MO emulsions is beneficial for several processing methods, including extrusion printing and injection molding. As a proof of concept, facile lettering via DIW using the LM–MO emulsion with τy ≈ 104 Pa is demonstrated in Figure 5a. Notably, emulsions with even higher particle loadings, for which the τy values are expected to be higher (data not obtained due to instrumental limitations), often exhibited visibly shiny regions when processed under ambient conditions (Figure 5c; see also the SEM image of the microstructure of the processed high‐τy emulsion in Figure S7 (Supporting Information); this was not as clear for the emulsion with a lower particle loading shown in Figure 5b). This is attributed to the high internal stress in the vitrified emulsion, which causes the rupture of the weak oxide skin (note that the known yield stress for the isolated oxide skin of LM is only ≈102 Pa,[76] two orders of magnitude smaller than the τy (>104 Pa) of a highly vitrified emulsion) and the coalescence of LM microdroplets. Nevertheless, this localized sintering effect was induced without any highly sample‐destructive external stimuli; therefore, the overall free‐standing structure of the processed high‐τy emulsion was retained not only in air (Figure 5d), but also in various polymeric matrices subjected to high‐temperature curing (Figure 5e).5FigurePhotographs of a) direct‐printed letters and b) lines using LM–MO emulsion with ΦLM = 0.475 and ΦMO = 0.075 and c) lines of the emulsion with ΦLM = 0.55 and ΦMO = 0.15. Various molded structures of highly vitrified LM–MO emulsions d) in air and e) in polymer matrices (PDMS: polydimethylsiloxanes, TMPTMA: trimethylolpropane trimethacrylate, and PEGDA: polyethylene glycol diacrylate). f) Thermal conductivity (k) of LM–MO emulsions with ΦLM = 0.55 and various ΦMO. g) Infrared images of manually spread (top) and extrusion‐printed (bottom) LM–Al2O3 emulsion (ΦLM = 0.55 and ΦAl2O3 = 0.15) under 1 sun illumination (inset: photographs of actual samples deposited on glass substrates). h) Temperature rise and fall of extrusion‐printed LM–Al2O3 emulsions (ΦLM = 0.55 and various ΦAl2O3) under light‐on (1 sun) and light‐off conditions, respectively.Solid MO particles can dramatically alter the rheological properties of LM emulsions, thereby impacting subsequent processing, and can also improve other unique functions, enabling novel applications. One example is the thermal conductivity (k). Figure 5f illustrates the k values of the vitrified LM–MO emulsions at ΦLM = 0.55, with variation of ΦMO. Despite the high k of bulk LM (kLM = 26.4 W m−1 K−1), the MO‐free base emulsion (ΦLM = 0.55) had a low k of 1.23 W m−1 K−1 due to the low‐k oxide skin and oil medium, both of which prevented direct contact between the dispersed LM droplets, and consequently, effective thermal transport. When ΦMO was increased to 0.15, the overall k for the emulsion increased remarkably. With SiO2, for example, the emulsion k reached the maximum value of 2.23 W m−1 K−1 at ΦSiO2 = 0.15 (note that the decrease in k at a higher ΦSiO2, which was consistently found for other MOs, is possibly due to the increased phonon scattering by excess MOs). Given the low k of SiO2 (kSiO2 ≈ 1 W m−1 K−1), this higher emulsion k is attributed to enhanced thermal percolation[77] via the dispersed LM phase, plausibly originating from the increased association of the LM microdroplets and the local sintering effect[78] in the vitrified emulsion. The highest k value of 3.20 W m−1 K−1 was obtained for the emulsion with Al2O3, which had a higher k value (kAl2O3 ≈ 30 W m−1 K−1) than that of SiO2, at the same MO concentration. Thus, MOs with desired thermophysical properties can enable modification of the rheological properties as well as fine tuning of the overall k value of LM emulsions, which may be useful in various thermal management applications, including thermal interface materials (TIMs). We highlight that the vitrified LM–MO emulsions, locally sintered by internal stress on the order of ≈104 Pa, did not permit electrical percolation, which is undesirable in most practical thermal management applications (note that it has been suggested that a critical stress of ≈ 1 MPa is required for electrical percolation via the high degree of LM sintering.[28,30,78] No such high pressure, which could not only allow for electrical percolation but also destroy the processed sample structure, was used to achieve the high k values shown in Figure 5f).Another functional feature of LM emulsions that can be improved by MO additives is photothermal conversion, which may have significant implications in biomedical therapy[9,23,42–44] and antimicrobial applications.[79] Figure 5g displays infrared images of the processed LM–Al2O3 emulsion that was either manually spread (top) or extrusion‐printed (bottom) on a glass substrate, under 1 sun illumination (1 kW m−2; AM 1.5G filter), where selective heating of the patterned regions was confirmed. The rate of the temperature rise of the processed LM–Al2O3 emulsion (at a fixed ΦLM of 0.55 with the variation of ΦAl2O3) under 1 sun illumination systematically increased as ΦAl2O3 increased to 0.15 (Figure 5h), attributed to the increased light–matter interaction. (The reason for the less pronounced temperature rise with higher Al2O3 concentrations, that is, ΦAl2O3 = 0.2, requires further investigation; this phenomenon is seemingly coincident with the decrease in the emulsion k at the same Al2O3 concentration shown in Figure 5f) Thus, MO additives may boost the photothermal conversion of LM emulsions, rather than simply acting as rheology modifiers, warranting follow‐up research on compounding LM emulsions with more precisely tailored and well‐defined photoresponsive materials.ConclusionSolid MO nanoparticles, as rheology modifiers, can readily vitrify colloidal systems of LM microdroplets suspended in apolar oil media. Despite the high physicochemical dissimilarities between LM and MOs, the MO nanoparticles exhibit high affinity for the surfaces of the suspended LM droplets, attributed to the oxide skin formed at the LM surface. Although direct contact between the particles submerged in the oil was suppressed owing to the oil layer adhering to the particle surfaces via the dispersive interaction, experimental and theoretical evaluations suggest that attractive interparticle interactions could still be formed via the polar interaction between the polar surface sites of the MO and the oxide‐coated LM. This oxide‐mediated polar LM–MO adhesion is considered crucial for the observed increase in the rheological strength of the emulsions. Oscillatory rheological measurements showed that even minute amounts of MO additives (ΦMO < 0.01) remarkably increased the G′ and G″ values of the LM‐in‐oil emulsion at ΦLM+MO = 0.55 by an order of magnitude, and the τy by two orders of magnitude. With high MO loadings (ΦMO ≥ 0.1), an emulsion with unprecedentedly high rheological strength was obtained, characterized by τy ≈104 Pa. The free‐standing structures of such high‐strength emulsions, prepared via extrusion or injection molding under ambient conditions, displayed excellent shape retention both in air and in polymeric matrices subjected to thermal curing, which may be beneficial for processing colloidal LM in various form factors. In highly vitrified emulsions, partial sintering effects are induced by the high internal sample stresses, which could improve the thermophysical properties of the emulsions, enabling several practical applications. We believe that the results and discussion presented in this study will be useful for compounding colloidal LM suspended in another liquid medium with a variety of functional particulate matter for desired applications.Experimental SectionMaterialsEGaIn (75.5 wt% Ga and 24.5 wt% In) was purchased from Changsha Santech Materials (China). Four types of MOs, SiO2 (99.9%, 2.65 g cm−3, 224.8 ± 105.0 nm, stock#: US1161M), TiO2 (anatase, 99.9%, 3.97 g cm−3, 120.3 ± 41.3 nm, stock#: US3411), Al2O3 (99.9%, 3.97 g cm−3, 193.8 ± 72.6 nm, stock#: US3003), and ZnO (99.9%, 5.606 g cm−3, 154.8 ± 64.6 nm, stock#: US3555), were purchased from US Research Nanomaterials, Inc. (USA). Average particle sizes and standard deviations were estimated from the SEM images (Figure S1, Supporting Information). Liquid paraffin and azobisisobutyronitrile (AIBN) were purchased from Samchun Chemicals (Korea). Trimethylolpropane trimethacrylate (TMPTMA) and polyethylene glycol diacrylate (PEGDA, average Mn = 700) were purchased from Sigma–Aldrich (USA). The Sylgard 184 Silicone Elastomer Kit was purchased from Dow Corning (USA).Sample PreparationTo prepare the base LM‐in‐oil emulsion, the desired amounts of EGaIn were added to a 20 mL scintillation vial charged with liquid paraffin, and the mixture was probe‐sonicated in an ice bath at an amplitude of 63 µm and a frequency of 20 kHz for 45 min with continuous magnetic stirring at 700 rpm. To vitrify the LM‐in‐oil emulsion, the desired amount of MO powder was added to the sample emulsion, followed by vortex mixing for several minutes using a Vortex‐Genie 2 shaker (Scientific Industries, USA). The vitrified LM–MO emulsions were manually processed via either syringe extrusion or injection molding under ambient conditions to demonstrate their shape retention properties. To prove the shape retention in the polymeric matrix, free‐standing structures of the processed emulsions were submerged in uncured liquid monomers or prepolymers (i.e., TMPTMA, PEGDA, or a mixture of Sylgard 184 and the curing agent in a 10:1 ratio; both TMPTMA and PEGDA contained dissolved initiator AIBN at 1 wt%), then heated at an elevated temperature of 80 °C for 5 h.Rheological MeasurementsTo study the viscoelastic properties of the vitrified emulsions, oscillatory measurements were performed using a HAAKE MARS‐40 rheometer (Thermo Fischer Scientific, USA) equipped with parallel plates (35 mm diameter and 1 mm gap distance). For the measurements, a shear‐stress ramp was used within the range of 10−1–2.4 × 104 Pa at a constant frequency of 1 Hz to obtain the storage (G′) and loss (G″) moduli of the samples. The yield stress (τy) of the samples was obtained by assigning crossover points of the G′ and G″ curves.Pull‐Off Force MeasurementsThe pull‐off forces between the macroscopic glass or LM surface and the LM droplet were measured using a Sigma 701 instrument (Biolin Scientific, Sweden) equipped with a microbalance and a ring‐shaped Pt–Ir probe. As the glass substrate, a plain microscope glass slide (Cat. No. 1 000 412; Marienfeld, Germany) was used. To prepare the LM surface, a droplet of EGaIn (≈30 µL) was sandwiched between two glass slides, which were slowly moved across each other multiple times to completely spread the LM on the glass surface. The substrates were placed at the bottom of a glass container, which was either empty (exposed to air) or charged with the oil medium, on a sample mount that could be lifted up or drawn down automatically by the instrument. The EGaIn droplet intended for contact with the substrate was suspended from the ring‐shaped Pt–Ir probe, which was hung at the top of the instrument. Both the substrate and suspended droplet were equilibrated in the respective measurement environments (air or oil) for at least a few minutes before starting the measurements. The sample mount with the substrate placed on it was then lifted at a constant velocity of 0.13 mm s−1 until the substrate touched the suspended droplet; the force exerted between the two surfaces was first detected by a microbalance installed in the instrument (corresponding to t = 0). The sample mount was then drawn back down at a constant velocity of 0.03 mm s−1, and the measured forces between the two adhered surfaces were continually recorded and represented as a function of time at t > 0.Contact Angle MeasurementsThe contact angles of sessile droplets of five probe liquids (water, ethylene glycol, glycerol, dimethyl sulfoxide, and diiodomethane) on different types of surfaces were measured using a Phoenix‐MT goniometer (SEO, Korea). To prepare the macroscopic MO surfaces, the MO powders were pelletized by compression at ≈382 MPa using a Pixie Hydraulic Pellet Press (PIKE Technologies, USA).ImagingAn SU‐70 (Hitachi, Japan) scanning electron microscope was used to characterize the morphologies of the LM microdroplets, MO particles, and vitrified LM–MO emulsions. A JEM‐2100F (JEOL, Japan) transmission electron microscope equipped with an energy‐dispersive spectroscope was used to observe the structures of the LM–MO aggregates.Thermal Conductivity MeasurementsThe thermal conductivities of the vitrified LM–MO emulsions were measured using the transient plane heat source method (or “Hot Disk” method, ISO 22007‐2:2022) using a TPS‐2500 instrument (Hot Disk AB, Sweden) equipped with a Kapton‐insulated sensor (Model 5465 F1). The measurements were performed in isotropic mode using a heating power of 240 mW. The measurement time and frequency were 10 s and 60 Hz, respectively.Photothermal EffectsTo demonstrate the photothermal effect of the vitrified LM–MO emulsions, the sample emulsions were deposited on glass substrates, either via manual spreading or via syringe extrusion, then subjected to 1 sun illumination (1 kW m−2; AM 1.5G filter) using a SciSun‐300 solar simulator (Sciencetech, Canada). The temperature increase versus time was monitored using a CX320 infrared camera (COX, Korea).AcknowledgementsThis work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2021R1F1A1048634). This research was supported by Nano∙Material Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (2009‐0082580).Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available from the corresponding author upon reasonable request.K. Kalantar‐Zadeh, J. Tang, T. Daeneke, A. P. O'Mullane, L. A. Stewart, J. Liu, C. Majidi, R. S. Ruoff, P. S. Weiss, M. D. Dickey, ACS Nano 2019, 13, 7388.T. Daeneke, K. Khoshmanesh, N. Mahmood, I. A. de Castro, D. Esrafilzadeh, S. J. Barrow, M. D. Dickey, K. Kalantar‐zadeh, Chem. Soc. Rev. 2018, 47, 4073.M. D. Dickey, Adv. Mater. 2017, 29, 1606425.M. D. Bartlett, N. Kazem, M. J. Powell‐Palm, X. Huang, W. Sun, J. A. Malen, C. Majidi, Proc. Natl. Acad. Sci. USA 2017, 114, 2143.X. Wang, R. Guo, J. Liu, Adv. Mater. Technol. 2019, 4, 1970009.S. Chen, H.‐Z. Wang, R.‐Q. Zhao, W. Rao, J. Liu, Matter 2020, 2, 1446.K. Yamagishi, W. Zhou, T. Ching, S. Y. Huang, M. Hashimoto, Adv. Mater. 2021, 33, 2008062.K. Zuraiqi, A. Zavabeti, J. Clarke‐Hannaford, B. J. Murdoch, K. Shah, M. J. S. Spencer, C. F. McConville, T. Daeneke, K. Chiang, Energy Environ. Sci. 2022, 15, 595.X.‐L. Ding, M.‐D. Liu, Q. Cheng, W.‐H. Guo, M.‐T. Niu, Q.‐X. Huang, X. Zeng, X.‐Z. Zhang, Biomaterials 2022, 281, 121369.R. Aukarasereenont, A. Goff, C. K. Nguyen, C. F. McConville, A. Elbourne, A. Zavabeti, T. Daeneke, Chem. Soc. Rev. 2022, 51, 1253.C. Ladd, J.‐H. So, J. Muth, M. D. Dickey, Adv. Mater. 2013, 25, 5081.Y.‐G. Park, H. S. An, J.‐Y. Kim, J.‐U. Park, Sci. Adv. 2019, 5, eaaw2844.U. Daalkhaijav, O. D. Yirmibesoglu, S. Walker, Y. Manguc, Adv. Mater. Technol. 2018, 3, 1700351.W. Kong, Z. Wang, M. Wang, K. C. Manning, A. Uppal, M. D. Green, R. Y. Wang, K. Rykaczewski, Adv. Mater. 2019, 31, 1904309.H. Chang, Z. R. Guo, Y. Cui, Y. Hou, Z. Sun, W. Rao, ACS Appl. Mater. Interfaces 2020, 12, 14125.C. Wang, Y. Gong, B. V. Cunning, S. Lee, Q. Le, S. R. Joshi, O. Buyukcakir, H. Zhang, W. K. Seong, R. S. Ruoff, Sci. Adv. 2021, 7, abe3767.W. Zhang, J. Z. Ou, S.‐Y. Tang, V. Sivan, D. D. Yao, K. Latham, K. Khoshmanesh, A. Mitchell, A. P. O'Mullane, K. Kalantar‐Zadeh, Adv. Funct. Mater. 2014, 24, 3799.W. Zhang, B. S. Naidu, J. Z. Ou, A. P. O'Mullane, A. F. Chrimes, B. J. Carey, Y. Wang, S.‐Y. Tang, V. Sivan, A. Mitchell, S. K. Bhargava, K. Kalantar‐Zadeh, ACS Appl. Mater. Interfaces 2015, 7, 1943.A. Yamaguchi, Y. Mashima, T. Iyoda, Angew. Chem., Int. Ed. 2015, 54, 12809.Y. Lu, Q. Hu, Y. Lin, D. B. Pacardo, C. Wang, W. Sun, F. S. Ligler, M. D. Dickey, Z. Gu, Nat. Commun. 2015, 6, 10066.Y. Lin, Y. Liu, J. Genzer, M. D. Dickey, Chem. Sci. 2017, 8, 3832.K. Chu, B. G. Song, H.‐I. Yang, D.‐M. Kim, C. S. Lee, M. Park, C.‐M. Chung, Adv. Funct. Mater. 2018, 28, 1800110.J. Yan, X. Zhang, Y. Liu, Y. Ye, J. Yu, Q. Chen, J. Wang, Y. Zhang, Q. Hu, Y. Kang, M. Yang, Z. Gu, Nano Res. 2019, 12, 1313.L. Zhou, J. Fu, Q. Gao, P. Zhao, Y. He, Adv. Funct. Mater. 2020, 30, 1906683.T. V. Neumann, E. G. Facchine, B. Leonardo, S. Khan, M. D. Dickey, Soft Matter 2020, 16, 6608.A. Uppal, W. Kong, A. Rana, R. Y. Wang, K. Rykaczewski, ACS Appl. Mater. Interfaces 2021, 13, 43348.V. Okatenko, L. Castilla‐Amoros, D. C. Ctoian, J. Vavra, A. Loiudice, R. Buonsanti, J. Am. Chem. Soc. 2022, 144, 10053.J. W. Boley, E. L. White, R. K. Kramer, Adv. Mater. 2015, 27, 2355.S. Liu, S. N. Reed, M. J. Higgins, M. S. Titus, R. Kramer‐Bottiglio, Nanoscale 2019, 11, 17615.X. Li, M. Li, J. Xu, J. You, Z. Yang, C. Li, Nat. Commun. 2019, 10, 3514.J. Kim, J. Jeong, Y. Hyun, S. K. Chumg, J. Lee, Small 2021, 17, 2104143.J. Yan, M. H. Malakooti, Z. Lu, Z. Wang, N. Kazem, C. Pan, M. R. Bockstaller, C. Majidi, K. Matyjaszewski, Nat. Nanotechnol. 2019, 14, 684.Z. J. Farrell, C. J. Thrasher, A. E. Flynn, C. E. Tabor, ACS Appl. Nano Mater. 2020, 3, 6297.Q. Wei, M. Sun, Z. Wang, J. Yan, R. Yuan, T. Liu, C. Majidi, K. Matyjaszewski, ACS Nano 2020, 14, 9884.H. Bark, M. W. M. Tan, G. Thangavel, P. S. Lee, Adv. Energy Mater. 2021, 11, 2101387.D. Kim, P. Thissen, G. Viner, D.‐W. Lee, W. Choi, Y. J. Chabal, J.‐B. Lee, ACS. Appl. Mater. Interfaces 2013, 5, 179.I. D. Joshupura, H. R. Ayers, G. A. Castillo, C. Ladd, C. E. Tabor, J. J. Adams, M. D. Dickey, ACS Appl. Mater. Interfaces 2018, 10, 44686.C. Pan, E. J. Markvicka, M. H. Malakooti, J. Yan, L. Hu, K. Matyjaszewski, C. Majidi, Adv. Mater. 2019, 31, 1900663.Y. Liu, W. Zhang, H. Wang, Mater. Horiz. 2021, 8, 56.P. Wang, Q. Hu, B. Lv, J. Zhu, W. Ma, Z. Dong, J. Wei, J. Sun, Nanotechnology 2020, 21, 145703.P. Wang, Q. Hu, B. Lv, Y. Liu, J. Yin, J. Sun, Mater. Adv. 2021, 2, 4414.N. Xia, N. Li, W. Rao, J. Yu, Q. Wu, L. Tan, H. Li, L. Gou, P. Liang, L. Li, X. Meng, Nanoscale 2019, 11, 10183.P. Zhu, S. Gao, H. Lin, X. Lu, B. Yang, L. Zhang, Y. Chen, J. Shi, Nano Lett. 2019, 19, 2128.J.‐J. Hu, M.‐D. Li, Y. Chen, F. Gao, S.‐Y. Peng, B.‐R. Xie, C.‐X. Li, X. Zheng, X.‐Z. Zhang, Biomaterials 2019, 207, 76.H. Li, Z. Bian, J. Zhu, Y. Huo, H. Li, Y. Lu, J. Am. Chem. Soc. 2007, 129, 4538.V. Sivan, S.‐Y. Tang, A. P. O'Mullane, P. Petersen, N. Eshtiaghi, K. Kalantar‐Zadeh, A. Mitchell, Adv. Funct. Mater. 2013, 23, 144.M. B. Ghasemian, M. Mayyas, S. A. Idrus‐Saidi, M. A. Jamal, J. Yang, S. S. Mofarah, E. Ababifiroozjaei, J. Tang, N. Syed, A. P. O'Mullane, T. Daeneke, K. Kalantar‐Zadeh, Adv. Funct. Mater. 2019, 29, 1901649.Y. Zhang, Z. Guo, H. Zhu, W. Xing, P. Tao, W. Shang, B. Fu, C. Song, Y. Hong, M. D. Dickey, T. Deng, J. Am. Chem. Soc. 2022, 144, 6779.S. Moon, H. Kim, K. Lee, J. Park, Y. Kim, S. Q. Choi, iScience 2021, 24, 103183.J. Kim, J. Lee, Small 2022, 18, 2108069.T. V. Neumann, M. D. Dickey, Adv. Mater. Technol. 2020, 5, 2000070.C. P. Ambulo, M. J. Ford, K. Searles, C. Majidi, T. H. Ware, ACS Appl. Mater. Interfaces 2021, 13, 12805.D. Wu, D. Liu, X. Tian, C. Lei, X. Chen, S. Zhang, F. Chen, K. Wu, Q. A Fu, Small Methods 2022, 6, 2200246.R. G. Larson, The Structure and Rheology of Complex Fluids, Oxford University Press, Inc, New York, NY 1999.R. Aveyard, B. P. Binks, J. H. Clint, Adv. Colloid Interface Sci. 2003, 100–102, 503.J. Park, T. Phongpreecha, J. Nicholas, Y. Qi, Acta Mater. 2020, 199, 551.M. Yunusa, G. J. Amador, D.‐M. Drotief, M. Sitti, Nano Lett. 2018, 18, 2498.I. D. Joshipura, A. Persson, V. K. Truong, J.‐H. Oh, M. Kong, M. H. Vong, C. Ni, M. Alsafatwi, D. P. Parekh, H. Zhao, Langmuir 2021, 37, 10914.M. Dickey, Phys. Today 2021, 74, 30.Y. Sun, Y. Li, X. Dong, X. Bu, J. W. Drelich, Colloids Surf. A 2020, 591, 124562.J. Long, Z. He, P. Zhou, X. Xie, C. Zhou, W. Hong, W. Hu, Adv. Mater. Interfaces 2018, 5, 1800353.J. N. Israelachvili Intermolecular and Surface Forces, Academic Press, Oxford 2011.F. M. Fowkes, Ind. Eng. Chem. 1964, 56, 40.D. K. Owens, R. C. Wendt, J. Appl. Polym. Sci. 1969, 13, 1741.S. Handschuh‐Wang, Y. Chen, L. Zhu, X. Zhou, ChemPhysChem 2018, 19, 1584.J. Lee, Z.‐L. Zhou, G. Alas, S. H. Behrens, Langmuir 2015, 31, 11989.S. S. Datta, D. D. Gerrard, T. S. Rhodes, T. G. Mason, D. A. Weitz, Phys. Rev. E 2011, 84, 041404.E. Irani, P. Chaudhuri, C. Heussinger, Phys. Rev. Lett. 2014, 112, 188303.S. F. Velandia, P. Marchal, C. Lemaitre, V. Sadtler, T. Roques‐Carmes, J. Colloid Interface Sci. 2021, 589, 286.P. L. Fuhrmann, S. Breunig, G. Sala, L. Sagis, M. Stieger, E. Scholten, J. Colloid Interface Sci. 2022, 607, 389.M. Destribats, S. Gineste, E. Laurichesse, H. Tanner, F. Leal‐Calderon, V. Heroguez, V. Schmitt, Langmuir 2014, 30, 9313.J. Xiao, X. Wang, A. J. Perez Gonzeles, Q. Huang, Food Hydrocolloids 2016, 54, 30.I. A. Morrison, Colloids Surf. A 1993, 71, 1.M. M. Gacek, J. C. Berg, Adv. Colloid Interface Sci. 2015, 220, 108.S. Lim, H. Park, J. Yang, C. Kwak, J. Lee, Colloids Surf. A 2019, 579, 123648.Q. Xu, N. Oudalov, Q. Guo, H. M. Jaeger, E. Brown, Phys. Fluids 2012, 24, 063101.F. Kargar, Z. Barani, R. Salgado, B. Debnath, J. S. Lewis, E. Aytan, R. K. Lake, A. A. Balandin, ACS Appl. Mater. Interfaces 2018, 10, 37555.A. Uppal, M. Ralphs, W. Kong, M. Hart, K. Rykaczewski, R. Y. Wang, ACS Appl. Mater. Interfaces 2020, 12, 2625.X. Xu, M. Fan, Z. Yu, Y. Zhao, H. Zhang, J. Wang, M. Wu, F. Sun, X. Xu, C. Ding, J. Li, Chem. Eng. J. 2022, 429, 132491. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Advanced Materials Interfaces Wiley

Vitrification of Liquid Metal‐in‐Oil Emulsions Using Nano‐Mineral Oxides

Download PDF
11 pages

Loading next page...
 
/lp/wiley/vitrification-of-liquid-metal-in-oil-emulsions-using-nano-mineral-0VJKrSSYkI

References (121)

Publisher
Wiley
Copyright
© 2023 Wiley‐VCH GmbH
eISSN
2196-7350
DOI
10.1002/admi.202202527
Publisher site
See Article on Publisher Site

Abstract

IntroductionGa and Ga‐based alloys, such as EGaIn (75.5 wt% Ga and 24.5 wt% In) and Gallinstan (68.5 wt% Ga, 21.5 wt% In, and 10.0 wt% Sn), are referred to as “liquid metals (LMs)”[1] because of their low melting points (near or even below room temperature). These materials have low toxicity, in contrast with well‐known LM Hg, while possessing both high electrical conductivity (as “metals”) and high fluidity (as “liquids”).[2] These characteristics afford high potential in a wide variety of cutting‐edge technologies.[3–10] Although unmodified LMs are promising for various applications,[3,11,12] modification can extend the application scope.[13–27] LMs can be tailored into the form of colloidal microdroplets,[17–27] which enables their dispersion in another carrier liquid or in solidifiable matrices. Despite their high interfacial tension, the dispersed LM microdroplets can remain as discrete particles (without immediate coalescence) owing to the solid oxide layer that spontaneously forms on the outermost particle surface,[2] unless this layer is intentionally destroyed.[24,28–30] This oxide layer, known to be mostly Ga2O3,[2,31] enables shape transformation of the LM microdroplets,[19–21] surface functionalization,[32–35] adhesion to substrates,[2,36,37] and templated synthesis of other particulate matter.[10] Intuitively, an oxide passivation layer with an ultrawide bandgap (≈4.8 eV)[18] may seem undesirable in applications where electrical percolation via the core metal phase of the microdroplets is important; however, the surface oxide layer may be useful in certain classes of applications where fine tuning of the dielectric properties or electrical breakdown strength of the electronic devices embedding LMs is required.[38,39] Furthermore, the electronic properties of the oxide skin, in conjunction with the core metal phase, are prospectively key factors in the photothermal conversion and photocatalytic activity of LM microdroplets,[17,18,40,41] with significant implications in human health and environmental applications such as targeted therapeutics[9,23,42–44] and pollutant removal.[45] Thus, the unique tunability of the electronic properties by employing external oxide materials[17,18,46–48] calls for follow‐up research on various combinations of LMs and other particulate additives.LM microdroplets, when suspended as discrete particles with a solid oxide skin in a liquid medium, increase the viscosity (ɳ) of the resulting colloidal systems in a loading‐dependent manner, and even induce elasticity at high loadings.[24,25,49,50] Such rheological properties of colloidal LM suspensions (or “emulsions,” considering the liquid nature of the dispersed phase) may be a key parameter for successfully harnessing the aforementioned functions with various form factors in emerging manufacturing technologies, such as multidimensional printing.[51,52] Neumann et al.[25] attempted to load LM microdroplets into a liquid silicone medium up to a maximum concentration of 90 wt%, achieving a 3D‐printable colloidal LM emulsion with an estimated yield stress (τy) of ≈64 Pa. More concentrated LM emulsions could possess higher τy values of a few hundred pascals.[24,49] Such emulsions have successfully been patterned in stretchable electronics,[24] or used in the fabrication of 3D heat dissipation composites,[49] via the direct ink writing (DIW) technique. Moreover, adding small amounts of liquid bridges with adequate wettability on the oxide skin of LM microdroplets can induce capillary attraction between the LM microdroplets and further enhance the rheological stiffness of LM emulsions (up to τy values of a few thousand pascals), which may potentially enable novel manufacturing conditions.[50]One promising yet relatively underexplored method of tuning the rheological properties of LM emulsions is by adding external solid particulate matter (metals,[26] ceramics,[49] and so on) to LM emulsions. The rheological effects of such solid additives, which were directly added to or mixed with “bulk” LM (as a continuous phase or a matrix, not a dispersed phase such as LM microdroplets suspended in another liquid continuous phase), have been extensively studied.[13–16,53] The solid additives suspended in the LM matrix readily modify the intrinsic rheological properties of bulk LM (e.g., high surface tension and low viscosity) and often induce a unique fluid‐to‐solid transition, enabling versatile LM processing. The resulting amalgam also demonstrated enhanced “functionality” such as electromagnetic interference shielding[16] and thermal management[14] when appropriate external solid particles were used. Although not extensively studied, external solid particles (which are not reactive with the LM) in a colloidal LM emulsion (comprising LM microdroplets suspended in another liquid medium) are theoretically expected to increase the rheological strength of the overall colloidal system by increasing in the volume fraction (Φ) of the total dispersed phase.[54] Moon et al.[49] recently demonstrated that adding boron nitride (BN) fillers (up to ΦBN = 0.2) to a polymerizable organic matrix with highly loaded LM microdroplets (at ΦLM = 0.7) increased the elastic modulus and τy of the resulting suspensions by an order of magnitude. However, to the best of our knowledge, quantitative investigations of the rheological effects of external solid fillers on colloidal LM emulsions have been conducted only for this single type of filler. More extensive rheological studies on the effects of solid additives using a variety of well‐defined solid particles will be highly useful for understanding the complex behaviors of multicomponent and multiphase colloidal systems containing LM microdroplets with other engineered particles and their practical application in numerous emerging technologies.Herein, we demonstrate that the addition of nanosized mineral oxide (MO) particles readily vitrifies an apolar oil (O)‐based emulsion with suspended LM microdroplets, even at minute dosages (ΦMO < 0.01 and ΦLM+MO = 0.55). At high MO loadings (ΦMO ≥ 0.1), a high‐strength colloidal LM system with an unprecedentedly high τy value of ≈104 Pa is obtained, wherein a partial “sintering” effect derived from localized droplet coalescence is observed, attributed to the high internal stress within the sample. The mechanisms of enhancement of the rheological strength are discussed in terms of the polar interaction between the oxide surfaces submerged in an apolar or dispersive medium. The MO additives not only modify the rheology of LM emulsions, but also improve their thermophysical properties, which are potentially useful in various applications.Results and DiscussionThe base LM‐in‐oil emulsion with ΦLM = 0.55 and ΦO = 0.45 (Figure 1a), consisting of EGaIn (dispersed phase; ɳ ≈ 2 mPa s) and liquid paraffin (continuous phase; ɳ ≈ 20 mPa s), was prepared via probe sonication of the mixture upon incremental addition of LM to the bulk oil phase. The produced LM droplets were observed as polydispersed microspheres (<2 µm, Figure 1e), the shape of which was retained; the droplets did not undergo immediate coalescence, attributed to the solid oxide skin[2] that formed spontaneously on the droplet surface (inset in Figure 1e). Despite the high Φ of the dispersed phase (ΦLM = 0.55), the resulting emulsion exhibited relatively high fluidity (Figure 1a), consistent with prior studies.[25,50] To the as‐prepared LM‐in‐oil emulsion, silica (SiO2, 224.8 ± 105.0 nm; see electron microscopic image in Figure S1 in the Supporting Information) nanospheres were added as rheology modifiers. For the systematic rheology study, Φ for the total dispersed phase was the same as that of the base emulsion, that is, ΦLM+MO = 0.55. Primary LM‐in‐oil emulsions with ΦLM < 0.55 were first prepared, and appropriate amounts of MO particles were added and vortex‐mixed.1FigurePhotographs of a) base LM emulsion and LM–MO emulsions with b) ΦMO = 0.0075, c) ΦMO = 0.0375, and d) ΦMO = 0.15. The total volume fraction of the dispersed phase was fixed at ΦLM+MO = 0.55. SEM image of e) LM microdroplets (inset: TEM image near droplet surface) and f) LM–SiO2 aggregates covered by oils taken from the vitrified emulsion at ΦLM = 0.475 and ΦSiO2 = 0.075. g,h) TEM images of LM–SiO2 aggregates and i) that with elemental mapping by EDS.The apparent stiffness of the LM emulsions increased remarkably with an increase in ΦSiO2 when ΦLM+SiO2 was fixed at 0.55 (Figure 1b–d). (Note that the representative case, in which SiO2 was used as the MO additive, is shown here, but the overall trend was qualitatively similar for other MOs, as discussed hereinafter). There was no visible macroscopic segregation between the LM‐ and MO‐rich phases despite the high physicochemical dissimilarities between the two types of particles. Scanning electron microscopy (SEM) images of the SiO2‐added emulsions (see the representative image of the sample with ΦSiO2 = 0.075 in Figure 1f) revealed raspberry‐like, rough particles, which were likely aggregates of LM and SiO2. The liquid substances covering the aggregates are the oils, which could be removed by deliberate washing with hexane using a squeeze bottle (see the SEM image of the repeatedly washed sample in Figure S2 in the Supporting Information; see also the presence of an oil layer covering the particles in the transmission electron microscopy (TEM) image in Figure S3 of the Supporting Information). TEM imaging (Figure 1g,h) with energy‐dispersive spectroscopy (EDS) elemental mapping (Figure 1i) of the samples after gentle hexane washing revealed the compositional structures of these aggregates. In most images, numerous adhered LM–SiO2 assemblies, that is, LM microdroplets covered with adsorbed SiO2 particles (see also the images of similar aggregate structures with other MOs in Figure S4 of the Supporting Information), some of which resembled Pickering emulsion[55] droplets), were observed, suggesting a high LM–MO affinity. Given the general nonwettability between the “pristine” metal and MOs,[37,56] the observed LM–MO adhesion was plausibly mediated by the oxide skin formed on the outermost surface of the LM droplets.Although adhesion between oxide‐coated LMs and various types of surfaces, including MO in air, has extensively been studied,[36,37,57,58] adhesion in liquid medium has rarely been evaluated. To understand the interaction between the surfaces of the LM and MO submerged in the oil medium, the pull‐off force between the two adhered macroscopic surfaces was measured (see Figure S5a in the Supporting Information). For the measurements, a planar substrate, either a glass slide (as a mimic of the SiO2 surface) or a slide coated with LM, placed on the bottom sample mount was lifted at a constant velocity to touch an LM droplet suspended at the top, either in the air or oil environment. (Note that the LM droplet and the planar LM surface on the glass substrate were pre‐equilibrated for a few minutes under ambient conditions before the measurements, such that the oxide skin could form on the respective surfaces, which is known to occur within a timescale of microseconds).[59] After touching the suspended LM droplet, the substrate was again drawn down at a constant velocity, during which the pulling force between the two adhered surfaces (i.e., LM–SiO2 and LM–LM) was measured using a microbalance.[57,60,61] The negative forces recorded at the initial time periods (Figure S5b–e, Supporting Information) correspond to the loadings exerted by the substrate upon lifting upward, which continually decreased at t > 0 (where t = 0 corresponds to the time at which the two surfaces first touched) as the substrate was drawn downward. As the recorded force moved into the positive region across the zero value (Figure 2a–d), the two adhered surfaces that were forced to separate began to pull each other, and the measured maximum positive force (maximum pull‐off force or Fpull‐off) was used to compare the relative strength of adhesion between different pairs of adhered surfaces. As shown in Figure 2a,b, the Fpull‐off value measured in air for the LM–LM pair (the outermost surfaces of both were likely coated by oxide skins), that is, ≈3300 µN, was significantly higher than that for LM–SiO2, that is, ≈970 µN (the Fpull‐off for LM–SiO2 was close to the literature value obtained using a similar method).[57] Assuming a priori that the surface energy of the oxide skin of LM is of the same order of magnitude as that of SiO2, this higher Fpull‐off value for LM–LM compared to that of LM–SiO2 is plausibly be ascribed to the bulk LM phases underneath the oxide skins of the two adhered surfaces, which, despite the presence of a thin (oxide) passivation layer, could still contribute to the total attractive interaction between the surfaces in close proximity.[62] It is also possible that the LM released through the ruptured sites of the oxide layer made direct contact, increasing the measured Fpull‐off. However, the Fpull‐off values for both LM–LM and LM–SiO2 decreased significantly in the oil medium (Figure 2c,d), possibly because the LM and SiO2 surfaces were primarily covered by oil molecules adhered via dispersive (or London)[62] interactions, preventing direct adhesion between the oil‐submerged surfaces. Notably, the Fpull‐off value for LM–LM in the oil medium (≈22 µN; Figure 2c) was even lower than that for LM–SiO2 (≈54 µN; Figure 2d), which is opposite to the observations for the same pairs in air. The higher adhesion strength between LM and SiO2 in the oil medium, inferred from the higher Fpull‐off, compared to that between LM and LM, suggests that in the oil medium, LM–SiO2 adhesion may be energetically more favorable than LM–LM adhesion. This difference provides a possible explanation for the large number of LM–SiO2 assemblies observed in the microscopic images of the aggregates from the vitrified LM emulsion with added SiO2.2FigureMeasured adhesion forces for a) LM–LM and b) LM–SiO2 surfaces in air, and c) LM–LM and d) LM–SiO2 surfaces in oil.The adhesion strength between a pair of interacting surfaces can be studied theoretically, where the work of adhesion (i.e., the free energy gain upon the adhesion of two interacting surfaces) can be estimated using a surface thermodynamic model.[62] Here, the widely used Owens, Wendt, Rabel, and Kaelble (OWRK) method, based on classical Fowkes theory,[58,63,64] was adopted. In this theory, the surface energy (or surface tension in units of mJ m−2) of a condensed phase of species i, γi, is defined as the sum of the dispersive (γiD)$\left( {\gamma _{\rm{i}}^{\rm{D}}} \right)$ and polar (γiP$\gamma _{\rm{i}}^{\rm{P}}$) components1γi=γiD+γiP\[\begin{array}{*{20}{c}}{{\gamma _{\rm{i}}} = \gamma _{\rm{i}}^{\rm{D}} + \gamma _{\rm{i}}^{\rm{P}}}\end{array}\]The total work of adhesion between two surfaces of species 1 and 2 (W1 − 2; in vacuum or air) is defined as the sum of the dispersive (W1−2D$W_{1 - 2}^{\rm{D}}$) and polar (W1−2P$W_{1 - 2}^{\rm{P}}$) contributions2W1−2=W1−2D+W1−2P\[\begin{array}{*{20}{c}}{{W_{1 - 2}} = W_{1 - 2}^{\rm{D}} + W_{1 - 2}^{\rm{P}}}\end{array}\]where3W1−2D=−2γ1Dγ2D\[\begin{array}{*{20}{c}}{W_{1 - 2}^{\rm{D}} = - 2\sqrt {\gamma _1^{\rm{D}}} \sqrt {\gamma _2^{\rm{D}}} }\end{array}\]and4W1−2P=−2γ1Pγ2P\[\begin{array}{*{20}{c}}{W_{1 - 2}^{\rm{P}} = - 2\sqrt {\gamma _1^{\rm{P}}} \sqrt {\gamma _2^{\rm{P}}} }\end{array}\]By combining Equations (2)–(4) with the well‐known Young's equation and taking species 1 and 2 as the solid (S) and liquid (L) phases, respectively, the following relation is obtained5γL(cosθ+1)2γLD=γSPγLPγLP+γSD\[\begin{array}{*{20}{c}}{\frac{{{\gamma _{\rm{L}}}\left( {\cos \theta + 1} \right)}}{{2\sqrt {\gamma _{\rm{L}}^{\rm{D}}} }} = \frac{{\sqrt {\gamma _{\rm{S}}^{\rm{P}}} \sqrt {\gamma _{\rm{L}}^{\rm{P}}} }}{{\sqrt {\gamma _{\rm{L}}^{\rm{P}}} }} + \sqrt {\gamma _{\rm{S}}^{\rm{D}}} }\end{array}\]where θ is the contact angle of liquid on solid. Thus, in principle, γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for a solid surface of interest can be obtained by solving Equation (5) with the measured θ values of (at least) two probe liquid droplets with known γLD$\gamma _{\rm{L}}^{\rm{D}}$ and γLP$\gamma _{\rm{L}}^{\rm{P}}$.To estimate γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for the glass and LM surfaces that were used in Fpull‐off measurements, the θ values of five probe liquids, i.e., water, ethylene glycol, glycerol, dimethyl sulfoxide, and diiodomethane, with known γLD$\gamma _{\rm{L}}^{\rm{D}}$ and γLP$\gamma _{\rm{L}}^{\rm{P}}$ were measured on each surface (Table S1, Supporting Information). (Note that using this model that does not include the metallic surface energy component, that is, γiM$\gamma _{\rm{i}}^{\rm{M}}$, with a series of nonmetallic probe liquids does not enable the extraction of any information on the metallic properties of the LM surface. Here, we assumed an ideal LM surface that is completely coated by the oxide skin, where the underlying metal phase has no influence on the direct adhesion with the other surface in the oil medium; the latter was in part supported by the result of Fpull‐off measurement in oil, and will be further discussed later). Using Equation (5), the γSD$\gamma _{\rm{S}}^{\rm{D}}$ values were preliminarily calculated using the θ values of the purely dispersive probe liquid (diiodomethane); the γSP$\gamma _{\rm{S}}^{\rm{P}}$ values were inferred via linear regression using the fixed γSD$\gamma _{\rm{S}}^{\rm{D}}$ values and the θ values for the other four probe liquids. As summarized in Table 1, both the glass and LM surfaces possessed large γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ values. Note that γSP$\gamma _{\rm{S}}^{\rm{P}}$ for the LM surface was comparable to that for the glass, attributed to the oxide (most likely Ga2O3) layer that spontaneously formed at the outermost surface of the LM, given that the “pristine” metal would not possess the “polar” component.1TableγSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ values of surfaces investigated using the OWRK method, along with calculated work of adhesion valuesSurfaceSurface energy component[mJ m−2]Work of adhesion[mJ m−2]γSD$\gamma _{\rm{S}}^{\rm{D}}$γSP$\gamma _{\rm{S}}^{\rm{P}}$WS−oilD$W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}}$WS−oilP$W_{{\rm{S}} - {\rm{oil}}}^{\rm{P}}$WS − oilWS−oil-LMD$W_{{\rm{S}} - {\rm{oil - LM}}}^{\rm{D}}$WS−oil−LMP$W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$WS − oil − LMGlass49.416.8−76.30−76.3−5.2−33.2−38.4LM(on glass)49.816.4−76.60−76.6−5.3−32.8−38.1SiO2(pellet)49.116.1−76.10−76.1−5.1−32.5−37.7TiO2(pellet)45.318.4−73.10−73.1−4.2−34.8−39.0Al2O3(pellet)49.616.6−76.50−76.5−5.2−33.0−38.3ZnO(pellet)50.014.8−76.80−76.8−5.3−31.2−36.5Using the known γSD$\gamma _{\rm{S}}^{\rm{D}}$ and γSP$\gamma _{\rm{S}}^{\rm{P}}$ for both surfaces, in the same theoretical framework, the work of adhesion of the respective surfaces with any counterpart with known γiD$\gamma _{\rm{i}}^{\rm{D}}$ and γiP$\gamma _{\rm{i}}^{\rm{P}}$ (in vacuum or air) can be calculated using Equations (2)–(4). When initially submerged in the purely dispersive oil medium, with γoil=γoilD+γoilP=29.5+0 = 29.5 mJ m−2${\gamma _{{\rm{oil}}}} = \gamma _{{\rm{oil}}}^{\rm{D}} + \gamma _{{\rm{oil}}}^{\rm{P}} = 29.5 + 0\; = \;29.5\;mJ\;{{\rm{m}}^{ - 2}}$, the adhesion between the respective surfaces and oil would take place first exclusively via the dispersive interaction, with WS−oil=−2γSDγoilD−2γSP0=WS−oilD${W_{{\rm{S}} - {\rm{oil}}}} = - 2\sqrt {\gamma _{\rm{S}}^{\rm{D}}} \sqrt {\gamma _{{\rm{oil}}}^{\rm{D}}} - 2\sqrt {\gamma _{\rm{S}}^{\rm{P}}} \sqrt 0 = W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}}$, the absolute value of which is an energy barrier to be overcome for the two solid surfaces to directly adhere to each other in the oil medium (Figure 3). It is notable that the WLM−oil=WLM−oilD${W_{{\rm{LM}} - {\rm{oil}}}} = W_{{\rm{LM}} - {\rm{oil}}}^{\rm{D}}$ of − 76.6 mJ m−2 for LM–oil was lower than the WSiO2−oil=WSiO2−oilD${W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}}}} = W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}}}^{\rm{D}}$ of − 76.3mJ m−2 for SiO2–oil, suggesting that the initial oil layer adhered more strongly on the LM surface than on the SiO2 surface. Taking the energy level for the adhered S–oil and LM–oil (the counterpart LM when S for the S–oil is also LM) to be zero, the work of adhesion for S–LM in oil can be given by6WS−oil−LM=WS−oil−LMD+WS−oil−LMP\[\begin{array}{*{20}{c}}{{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}} = W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} + W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}}\end{array}\]where7WS−oil−LMD=−WS−oilD−WLM−oilD+WS−LMD+Woil−oilD\[\begin{array}{*{20}{c}}{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} = - W_{{\rm{S}} - {\rm{oil}}}^{\rm{D}} - W_{{\rm{LM}} - {\rm{oil}}}^{\rm{D}} + W_{{\rm{S}} - {\rm{LM}}}^{\rm{D}} + W_{{\rm{oil}} - {\rm{oil}}}^{\rm{D}}}\end{array}\]and8WS−oil−LMP=−WS−oilP−WLM−oilP+WS−LMP+Woil−oilP\[\begin{array}{*{20}{c}}{W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}} = - W_{{\rm{S}} - {\rm{oil}}}^{\rm{P}} - W_{{\rm{LM}} - {\rm{oil}}}^{\rm{P}} + W_{{\rm{S}} - {\rm{LM}}}^{\rm{P}} + W_{{\rm{oil}} - {\rm{oil}}}^{\rm{P}}}\end{array}\]3FigureSchematic of energetics of two interacting surfaces submerged in apolar oil medium.As shown in Table 1, the polar work of adhesion (WS−oil−LMP≈−33mJ m−2)$\left( {W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}} \approx - 33mJ\;{{\rm{m}}^{ - 2}}} \right)$ for the two surfaces was dominant over the dispersive work of adhesion (WS−oil−LMD≈−5 mJ m−2)$\left( {W_{{\rm{S}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}} \approx - 5\;mJ\;{{\rm{m}}^{ - 2}}} \right)$, which suggests that the polar interaction is the main driving force for the adhesion between the two solid surfaces (with significant γSP$\gamma _{\rm{S}}^{\rm{P}}$) in the dispersive oil medium. Notably, the WSiO2−oil−LM${W_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}}$ of −38.4 mJ m−2 for SiO2–LM in oil is lower than the WLM − oil − LM of −38.1 mJ m−2 for LM–LM in oil, which is qualitatively consistent with the results of the Fpull‐off measurements. Note that if the metallic work of adhesion was considered for LM–LM, that is, WLM−oil−LMM=WLM−LMM=−2γiM≪−600 mJ m−2$W_{{\rm{LM}} - {\rm{oil}} - {\rm{LM}}}^{\rm{M}} = W_{{\rm{LM}} - {\rm{LM}}}^{\rm{M}} = - 2\gamma _{\rm{i}}^{\rm{M}} \ll - 600\;mJ\;{{\rm{m}}^{ - 2}}$,[62,65] the LM–LM adhesion in the oil medium would always dominate LM–SiO2, which was obviously not the case, as shown in Figure 2c versus Figure 2d. Thus, it seems reasonable to assume that the metallic contribution to the interparticle interaction was likely suppressed in the oil medium, possibly because of the presence of the oxide skin and the adhered oil layer, and the overall interparticle interaction was dominated by that between the polar surface sites.Provided that surface charging and electrostatic interparticle repulsion are unlikely in an oil medium with a low dielectric constant (ε ≈ 2.2 for liquid paraffin),[31,66] the attraction[67–70] mediated by the polar surface sites would be the main colloidal force affecting the rheological behavior of the emulsion. The addition of nanosized MOs with high thermodynamic affinity for the oxide skin of the LM droplets, as well as a high specific surface area, was expected to increase the overall attractive interaction between the suspended particles, thereby enhancing the rheological strength of the emulsions. Furthermore, although the native oxide skin of the LM droplets was too thin to confer a high degree of surface rigidity, the adsorbed solid MO particles more effectively enhanced the mechanical rigidity of the droplet surface, contributing to the increase in the overall emulsion strength. (Note: Increases in the rheological strength of liquid‐in‐liquid emulsions by interfacially adsorbed solid particle stabilizers, i.e., Pickering emulsifiers, have also been suggested for systems consisting of more “ordinary” liquid dispersed phases, e.g., oil or water,[69,71,72] although the mechanism of particle adsorption at the liquid–liquid interface differs from that of the LM–MO adhesion discussed here).Figure 4a shows the elastic (G′) and viscous (G″) moduli of LM emulsions with various ΦSiO2 values at ΦLM+SiO2 = 0.55, obtained via oscillatory viscoelasticity measurements using a stress ramp at a constant frequency (1 Hz). The measured G′ and G″ values of the base emulsion (ΦLM = 0.55 with no added MO) were ≈103 and ≈102 Pa, respectively, in the linear viscoelastic region. Notably, replacing only ≈1 vol% of LM with SiO2 (i.e., ΦSiO2 ≈ 0.01 at ΦLM+SiO2 = 0.55) increased the G′ and G″ values to >104 and >103 Pa, respectively. Furthermore, by assigning the crossover point of the G′ and G″ curves as the yield point of the sample, it was inferred that the τy value of the base emulsion (≈10 Pa) increased by two orders of magnitude at ΦSiO2 ≈ 0.01. For example, a τy value of 802.1 ± 253.6 Pa was obtained for the sample with ΦSiO2 = 0.0075 (Figure 4c), which is of the same order of magnitude as that of the LM‐in‐oil emulsion with polymeric liquid‐bridged LM microdroplets.[50] The τy value systematically increased with a further increase in (Figure 4c), exceeding 104 Pa at ΦSiO2 ≥ 0.1, which is an unprecedented value for colloidal LM emulsions. The measurements often exceeded the instrumental limit at higher SiO2 loadings (data not shown). The size of the MOs affects the efficiency of emulsion vitrification. When ≈10 µm sized SiO2 particles were used, which were larger than the LM droplets (<2 µm), the emulsion still vitrified but at a much lower rate than the emulsion of nanosized SiO2 (≈200 nm) particles (Table S2, Supporting Information). This is because the nanosized MOs conferred a much larger total surface area, and the interaction with LM droplets was more effectively facilitated.4Figurea) Elastic (G′, solid lines) and viscous (G″, dotted lines) moduli of LM–SiO2 emulsions with various ΦSiO2 (data for the representative MO‐free pristine LM emulsion are illustrated by gray color; the numbers on top of G′ curves of LM–SiO2 emulsions indicate the corresponding ΦSiO2). b) G′ and G″ curves for LM–MO emulsions at a low ΦMO of 0.0037. c) Yield stress (τy) for LM–MO emulsions with various ΦMO (red spheres: SiO2, brown diamonds: TiO2, green stars: Al2O3, blue squares: ZnO, red stars: hydrophobically modified SiO2). In panels (a) and (b), the data were obtained via oscillatory measurements using stress ramp at a constant frequency of 1 Hz. In panel (c), the data were obtained by assigning the cross‐over points of the G′ and G″ curves. All the investigated sample emulsions had a fixed dispersed phase volume fraction of 0.55, i.e., ΦLM+MO = 0.55.Figure 4b,c shows an increasing trend for rheological parameters with other types of MO additives, including titania (TiO2, 120.3 ± 41.3 nm), alumina (Al2O3, 193.8 ± 72.6 nm), and zinc oxide (ZnO, 154.8 ± 64.6 nm) (the SEM images of the MOs and TEM images of the LM–MO aggregate structures are shown in Figures S1 and S4 in the Supporting Information, respectively). Despite differences in the particle shape and size distribution, all the investigated MOs have average sizes in the same order of magnitude, resulting in significant emulsion vitrification, although the rate was dependent on the type of MO used. The γMOD$\gamma _{{\rm{MO}}}^{\rm{D}}$ and γMOP$\gamma _{{\rm{MO}}}^{\rm{P}}$ values of these MOs (Table 1) were estimated by applying the OWRK method, for the macroscopic surfaces of the pelletized MOs prepared by compressing the MO powders at ≈ 382 MPa. (The measured θ values of the five probe liquids on these surfaces are listed in Table S1 in the Supporting Information) All the investigated MOs exhibited significant γMOD$\gamma _{{\rm{MO}}}^{\rm{D}}$ and γMOP$\gamma _{{\rm{MO}}}^{\rm{P}}$ values, comparable in magnitude to those of the glass substrate and the manually spread LM surface discussed above. Thus, the calculated WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ values for these MOs similarly dominated WMO−oil−LMD$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{D}}$, suggesting the importance of the polar contribution to the LM–MO interaction. The relative magnitudes of WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ were consistent with the order of the increasing rates of τy shown in Figure 4c (i.e., TiO2 > Al2O3 > ZnO), except in the case of SiO2, for which WMO−oil−LMP$W_{{\rm{MO}} - {\rm{oil}} - {\rm{LM}}}^{\rm{P}}$ was found to be (comparable to but) slightly higher than that of Al2O3. However, we note that the “Hamaker constant” (the prefactor for the van der Waals interaction) for LM–MO interaction in the oil medium, calculated using the Lifshitz theory and exclusively considering the dispersive contribution,[62] would predict an absurdly lower attraction between LM and SiO2 (or even “repulsion,” i.e., ASiO2−oil−LM≤0${A_{{\rm{Si}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}} \le 0$) than that between LM and the other MOs (ATiO2−oil−LM≈1.0 ×10−19J${A_{{\rm{Ti}}{{\rm{O}}_2} - {\rm{oil}} - {\rm{LM}}}} \approx 1.0\; \times {10^{ - 19}}{\rm{J}}$, AZnO − oil − LM ≈ 5.2 × 10−20J, and AAl2O3−oil−LM≈3.3 ×10−20J${A_{{\rm{A}}{{\rm{l}}_2}{{\rm{O}}_3} - {\rm{oil}} - {\rm{LM}}}} \approx 3.3\; \times {10^{ - 20}}{\rm{J}}$; see more details in the Supporting Information). This is in contrast with the favorable LM–SiO2 assemblies observed in the electron microscopic images (Figure 1f–i) and the highest τy values of the LM–SiO2 emulsions at various ΦSiO2 values (Figure 4c). Thus, we suggest that the polar interparticle interaction may be a more relevant parameter than the dispersive interaction in the attraction‐driven rheological stiffening of LM–MO emulsions based on the dispersive medium. This type of interaction is also considered the main driving force for various self‐assembly processes in apolar media, such as the adsorption of surfactants in the “head‐down” configuration or the formation of inverse micelles.[66,73,74] As a control, SiO2 particles modified with hexadecyl triethoxysilane (SiO2@HDTS) were added to LM emulsions with various ΦSiO2@HDTS at a fixed ΦLM+MO of 0.55, as described above. Modification with HDTS completely altered the wettability of SiO2 by capping the polar sites of the bare SiO2 surface and covalently implanting apolar hydrocarbon chains.[75] The resulting SiO2@HDTS exhibited excellent dispersibility in the oil medium (Figure S6b, Supporting Information), in stark contrast with unmodified SiO2 (Figure S6a, Supporting Information). For the LM emulsions with added SiO2@HDTS, the increase in τy was much less pronounced (Figure 4c; Figure S6c, Supporting Information), attributed to screening of the polar interaction by the surface‐anchored apolar moieties, which corroborate the importance of polar interactions in emulsion vitrification.High shape retention of vitrified LM–MO emulsions is beneficial for several processing methods, including extrusion printing and injection molding. As a proof of concept, facile lettering via DIW using the LM–MO emulsion with τy ≈ 104 Pa is demonstrated in Figure 5a. Notably, emulsions with even higher particle loadings, for which the τy values are expected to be higher (data not obtained due to instrumental limitations), often exhibited visibly shiny regions when processed under ambient conditions (Figure 5c; see also the SEM image of the microstructure of the processed high‐τy emulsion in Figure S7 (Supporting Information); this was not as clear for the emulsion with a lower particle loading shown in Figure 5b). This is attributed to the high internal stress in the vitrified emulsion, which causes the rupture of the weak oxide skin (note that the known yield stress for the isolated oxide skin of LM is only ≈102 Pa,[76] two orders of magnitude smaller than the τy (>104 Pa) of a highly vitrified emulsion) and the coalescence of LM microdroplets. Nevertheless, this localized sintering effect was induced without any highly sample‐destructive external stimuli; therefore, the overall free‐standing structure of the processed high‐τy emulsion was retained not only in air (Figure 5d), but also in various polymeric matrices subjected to high‐temperature curing (Figure 5e).5FigurePhotographs of a) direct‐printed letters and b) lines using LM–MO emulsion with ΦLM = 0.475 and ΦMO = 0.075 and c) lines of the emulsion with ΦLM = 0.55 and ΦMO = 0.15. Various molded structures of highly vitrified LM–MO emulsions d) in air and e) in polymer matrices (PDMS: polydimethylsiloxanes, TMPTMA: trimethylolpropane trimethacrylate, and PEGDA: polyethylene glycol diacrylate). f) Thermal conductivity (k) of LM–MO emulsions with ΦLM = 0.55 and various ΦMO. g) Infrared images of manually spread (top) and extrusion‐printed (bottom) LM–Al2O3 emulsion (ΦLM = 0.55 and ΦAl2O3 = 0.15) under 1 sun illumination (inset: photographs of actual samples deposited on glass substrates). h) Temperature rise and fall of extrusion‐printed LM–Al2O3 emulsions (ΦLM = 0.55 and various ΦAl2O3) under light‐on (1 sun) and light‐off conditions, respectively.Solid MO particles can dramatically alter the rheological properties of LM emulsions, thereby impacting subsequent processing, and can also improve other unique functions, enabling novel applications. One example is the thermal conductivity (k). Figure 5f illustrates the k values of the vitrified LM–MO emulsions at ΦLM = 0.55, with variation of ΦMO. Despite the high k of bulk LM (kLM = 26.4 W m−1 K−1), the MO‐free base emulsion (ΦLM = 0.55) had a low k of 1.23 W m−1 K−1 due to the low‐k oxide skin and oil medium, both of which prevented direct contact between the dispersed LM droplets, and consequently, effective thermal transport. When ΦMO was increased to 0.15, the overall k for the emulsion increased remarkably. With SiO2, for example, the emulsion k reached the maximum value of 2.23 W m−1 K−1 at ΦSiO2 = 0.15 (note that the decrease in k at a higher ΦSiO2, which was consistently found for other MOs, is possibly due to the increased phonon scattering by excess MOs). Given the low k of SiO2 (kSiO2 ≈ 1 W m−1 K−1), this higher emulsion k is attributed to enhanced thermal percolation[77] via the dispersed LM phase, plausibly originating from the increased association of the LM microdroplets and the local sintering effect[78] in the vitrified emulsion. The highest k value of 3.20 W m−1 K−1 was obtained for the emulsion with Al2O3, which had a higher k value (kAl2O3 ≈ 30 W m−1 K−1) than that of SiO2, at the same MO concentration. Thus, MOs with desired thermophysical properties can enable modification of the rheological properties as well as fine tuning of the overall k value of LM emulsions, which may be useful in various thermal management applications, including thermal interface materials (TIMs). We highlight that the vitrified LM–MO emulsions, locally sintered by internal stress on the order of ≈104 Pa, did not permit electrical percolation, which is undesirable in most practical thermal management applications (note that it has been suggested that a critical stress of ≈ 1 MPa is required for electrical percolation via the high degree of LM sintering.[28,30,78] No such high pressure, which could not only allow for electrical percolation but also destroy the processed sample structure, was used to achieve the high k values shown in Figure 5f).Another functional feature of LM emulsions that can be improved by MO additives is photothermal conversion, which may have significant implications in biomedical therapy[9,23,42–44] and antimicrobial applications.[79] Figure 5g displays infrared images of the processed LM–Al2O3 emulsion that was either manually spread (top) or extrusion‐printed (bottom) on a glass substrate, under 1 sun illumination (1 kW m−2; AM 1.5G filter), where selective heating of the patterned regions was confirmed. The rate of the temperature rise of the processed LM–Al2O3 emulsion (at a fixed ΦLM of 0.55 with the variation of ΦAl2O3) under 1 sun illumination systematically increased as ΦAl2O3 increased to 0.15 (Figure 5h), attributed to the increased light–matter interaction. (The reason for the less pronounced temperature rise with higher Al2O3 concentrations, that is, ΦAl2O3 = 0.2, requires further investigation; this phenomenon is seemingly coincident with the decrease in the emulsion k at the same Al2O3 concentration shown in Figure 5f) Thus, MO additives may boost the photothermal conversion of LM emulsions, rather than simply acting as rheology modifiers, warranting follow‐up research on compounding LM emulsions with more precisely tailored and well‐defined photoresponsive materials.ConclusionSolid MO nanoparticles, as rheology modifiers, can readily vitrify colloidal systems of LM microdroplets suspended in apolar oil media. Despite the high physicochemical dissimilarities between LM and MOs, the MO nanoparticles exhibit high affinity for the surfaces of the suspended LM droplets, attributed to the oxide skin formed at the LM surface. Although direct contact between the particles submerged in the oil was suppressed owing to the oil layer adhering to the particle surfaces via the dispersive interaction, experimental and theoretical evaluations suggest that attractive interparticle interactions could still be formed via the polar interaction between the polar surface sites of the MO and the oxide‐coated LM. This oxide‐mediated polar LM–MO adhesion is considered crucial for the observed increase in the rheological strength of the emulsions. Oscillatory rheological measurements showed that even minute amounts of MO additives (ΦMO < 0.01) remarkably increased the G′ and G″ values of the LM‐in‐oil emulsion at ΦLM+MO = 0.55 by an order of magnitude, and the τy by two orders of magnitude. With high MO loadings (ΦMO ≥ 0.1), an emulsion with unprecedentedly high rheological strength was obtained, characterized by τy ≈104 Pa. The free‐standing structures of such high‐strength emulsions, prepared via extrusion or injection molding under ambient conditions, displayed excellent shape retention both in air and in polymeric matrices subjected to thermal curing, which may be beneficial for processing colloidal LM in various form factors. In highly vitrified emulsions, partial sintering effects are induced by the high internal sample stresses, which could improve the thermophysical properties of the emulsions, enabling several practical applications. We believe that the results and discussion presented in this study will be useful for compounding colloidal LM suspended in another liquid medium with a variety of functional particulate matter for desired applications.Experimental SectionMaterialsEGaIn (75.5 wt% Ga and 24.5 wt% In) was purchased from Changsha Santech Materials (China). Four types of MOs, SiO2 (99.9%, 2.65 g cm−3, 224.8 ± 105.0 nm, stock#: US1161M), TiO2 (anatase, 99.9%, 3.97 g cm−3, 120.3 ± 41.3 nm, stock#: US3411), Al2O3 (99.9%, 3.97 g cm−3, 193.8 ± 72.6 nm, stock#: US3003), and ZnO (99.9%, 5.606 g cm−3, 154.8 ± 64.6 nm, stock#: US3555), were purchased from US Research Nanomaterials, Inc. (USA). Average particle sizes and standard deviations were estimated from the SEM images (Figure S1, Supporting Information). Liquid paraffin and azobisisobutyronitrile (AIBN) were purchased from Samchun Chemicals (Korea). Trimethylolpropane trimethacrylate (TMPTMA) and polyethylene glycol diacrylate (PEGDA, average Mn = 700) were purchased from Sigma–Aldrich (USA). The Sylgard 184 Silicone Elastomer Kit was purchased from Dow Corning (USA).Sample PreparationTo prepare the base LM‐in‐oil emulsion, the desired amounts of EGaIn were added to a 20 mL scintillation vial charged with liquid paraffin, and the mixture was probe‐sonicated in an ice bath at an amplitude of 63 µm and a frequency of 20 kHz for 45 min with continuous magnetic stirring at 700 rpm. To vitrify the LM‐in‐oil emulsion, the desired amount of MO powder was added to the sample emulsion, followed by vortex mixing for several minutes using a Vortex‐Genie 2 shaker (Scientific Industries, USA). The vitrified LM–MO emulsions were manually processed via either syringe extrusion or injection molding under ambient conditions to demonstrate their shape retention properties. To prove the shape retention in the polymeric matrix, free‐standing structures of the processed emulsions were submerged in uncured liquid monomers or prepolymers (i.e., TMPTMA, PEGDA, or a mixture of Sylgard 184 and the curing agent in a 10:1 ratio; both TMPTMA and PEGDA contained dissolved initiator AIBN at 1 wt%), then heated at an elevated temperature of 80 °C for 5 h.Rheological MeasurementsTo study the viscoelastic properties of the vitrified emulsions, oscillatory measurements were performed using a HAAKE MARS‐40 rheometer (Thermo Fischer Scientific, USA) equipped with parallel plates (35 mm diameter and 1 mm gap distance). For the measurements, a shear‐stress ramp was used within the range of 10−1–2.4 × 104 Pa at a constant frequency of 1 Hz to obtain the storage (G′) and loss (G″) moduli of the samples. The yield stress (τy) of the samples was obtained by assigning crossover points of the G′ and G″ curves.Pull‐Off Force MeasurementsThe pull‐off forces between the macroscopic glass or LM surface and the LM droplet were measured using a Sigma 701 instrument (Biolin Scientific, Sweden) equipped with a microbalance and a ring‐shaped Pt–Ir probe. As the glass substrate, a plain microscope glass slide (Cat. No. 1 000 412; Marienfeld, Germany) was used. To prepare the LM surface, a droplet of EGaIn (≈30 µL) was sandwiched between two glass slides, which were slowly moved across each other multiple times to completely spread the LM on the glass surface. The substrates were placed at the bottom of a glass container, which was either empty (exposed to air) or charged with the oil medium, on a sample mount that could be lifted up or drawn down automatically by the instrument. The EGaIn droplet intended for contact with the substrate was suspended from the ring‐shaped Pt–Ir probe, which was hung at the top of the instrument. Both the substrate and suspended droplet were equilibrated in the respective measurement environments (air or oil) for at least a few minutes before starting the measurements. The sample mount with the substrate placed on it was then lifted at a constant velocity of 0.13 mm s−1 until the substrate touched the suspended droplet; the force exerted between the two surfaces was first detected by a microbalance installed in the instrument (corresponding to t = 0). The sample mount was then drawn back down at a constant velocity of 0.03 mm s−1, and the measured forces between the two adhered surfaces were continually recorded and represented as a function of time at t > 0.Contact Angle MeasurementsThe contact angles of sessile droplets of five probe liquids (water, ethylene glycol, glycerol, dimethyl sulfoxide, and diiodomethane) on different types of surfaces were measured using a Phoenix‐MT goniometer (SEO, Korea). To prepare the macroscopic MO surfaces, the MO powders were pelletized by compression at ≈382 MPa using a Pixie Hydraulic Pellet Press (PIKE Technologies, USA).ImagingAn SU‐70 (Hitachi, Japan) scanning electron microscope was used to characterize the morphologies of the LM microdroplets, MO particles, and vitrified LM–MO emulsions. A JEM‐2100F (JEOL, Japan) transmission electron microscope equipped with an energy‐dispersive spectroscope was used to observe the structures of the LM–MO aggregates.Thermal Conductivity MeasurementsThe thermal conductivities of the vitrified LM–MO emulsions were measured using the transient plane heat source method (or “Hot Disk” method, ISO 22007‐2:2022) using a TPS‐2500 instrument (Hot Disk AB, Sweden) equipped with a Kapton‐insulated sensor (Model 5465 F1). The measurements were performed in isotropic mode using a heating power of 240 mW. The measurement time and frequency were 10 s and 60 Hz, respectively.Photothermal EffectsTo demonstrate the photothermal effect of the vitrified LM–MO emulsions, the sample emulsions were deposited on glass substrates, either via manual spreading or via syringe extrusion, then subjected to 1 sun illumination (1 kW m−2; AM 1.5G filter) using a SciSun‐300 solar simulator (Sciencetech, Canada). The temperature increase versus time was monitored using a CX320 infrared camera (COX, Korea).AcknowledgementsThis work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (2021R1F1A1048634). This research was supported by Nano∙Material Technology Development Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning (2009‐0082580).Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available from the corresponding author upon reasonable request.K. Kalantar‐Zadeh, J. Tang, T. Daeneke, A. P. O'Mullane, L. A. Stewart, J. Liu, C. Majidi, R. S. Ruoff, P. S. Weiss, M. D. Dickey, ACS Nano 2019, 13, 7388.T. Daeneke, K. Khoshmanesh, N. Mahmood, I. A. de Castro, D. Esrafilzadeh, S. J. Barrow, M. D. Dickey, K. Kalantar‐zadeh, Chem. Soc. Rev. 2018, 47, 4073.M. D. Dickey, Adv. Mater. 2017, 29, 1606425.M. D. Bartlett, N. Kazem, M. J. Powell‐Palm, X. Huang, W. Sun, J. A. Malen, C. Majidi, Proc. Natl. Acad. Sci. USA 2017, 114, 2143.X. Wang, R. Guo, J. Liu, Adv. Mater. Technol. 2019, 4, 1970009.S. Chen, H.‐Z. Wang, R.‐Q. Zhao, W. Rao, J. Liu, Matter 2020, 2, 1446.K. Yamagishi, W. Zhou, T. Ching, S. Y. Huang, M. Hashimoto, Adv. Mater. 2021, 33, 2008062.K. Zuraiqi, A. Zavabeti, J. Clarke‐Hannaford, B. J. Murdoch, K. Shah, M. J. S. Spencer, C. F. McConville, T. Daeneke, K. Chiang, Energy Environ. Sci. 2022, 15, 595.X.‐L. Ding, M.‐D. Liu, Q. Cheng, W.‐H. Guo, M.‐T. Niu, Q.‐X. Huang, X. Zeng, X.‐Z. Zhang, Biomaterials 2022, 281, 121369.R. Aukarasereenont, A. Goff, C. K. Nguyen, C. F. McConville, A. Elbourne, A. Zavabeti, T. Daeneke, Chem. Soc. Rev. 2022, 51, 1253.C. Ladd, J.‐H. So, J. Muth, M. D. Dickey, Adv. Mater. 2013, 25, 5081.Y.‐G. Park, H. S. An, J.‐Y. Kim, J.‐U. Park, Sci. Adv. 2019, 5, eaaw2844.U. Daalkhaijav, O. D. Yirmibesoglu, S. Walker, Y. Manguc, Adv. Mater. Technol. 2018, 3, 1700351.W. Kong, Z. Wang, M. Wang, K. C. Manning, A. Uppal, M. D. Green, R. Y. Wang, K. Rykaczewski, Adv. Mater. 2019, 31, 1904309.H. Chang, Z. R. Guo, Y. Cui, Y. Hou, Z. Sun, W. Rao, ACS Appl. Mater. Interfaces 2020, 12, 14125.C. Wang, Y. Gong, B. V. Cunning, S. Lee, Q. Le, S. R. Joshi, O. Buyukcakir, H. Zhang, W. K. Seong, R. S. Ruoff, Sci. Adv. 2021, 7, abe3767.W. Zhang, J. Z. Ou, S.‐Y. Tang, V. Sivan, D. D. Yao, K. Latham, K. Khoshmanesh, A. Mitchell, A. P. O'Mullane, K. Kalantar‐Zadeh, Adv. Funct. Mater. 2014, 24, 3799.W. Zhang, B. S. Naidu, J. Z. Ou, A. P. O'Mullane, A. F. Chrimes, B. J. Carey, Y. Wang, S.‐Y. Tang, V. Sivan, A. Mitchell, S. K. Bhargava, K. Kalantar‐Zadeh, ACS Appl. Mater. Interfaces 2015, 7, 1943.A. Yamaguchi, Y. Mashima, T. Iyoda, Angew. Chem., Int. Ed. 2015, 54, 12809.Y. Lu, Q. Hu, Y. Lin, D. B. Pacardo, C. Wang, W. Sun, F. S. Ligler, M. D. Dickey, Z. Gu, Nat. Commun. 2015, 6, 10066.Y. Lin, Y. Liu, J. Genzer, M. D. Dickey, Chem. Sci. 2017, 8, 3832.K. Chu, B. G. Song, H.‐I. Yang, D.‐M. Kim, C. S. Lee, M. Park, C.‐M. Chung, Adv. Funct. Mater. 2018, 28, 1800110.J. Yan, X. Zhang, Y. Liu, Y. Ye, J. Yu, Q. Chen, J. Wang, Y. Zhang, Q. Hu, Y. Kang, M. Yang, Z. Gu, Nano Res. 2019, 12, 1313.L. Zhou, J. Fu, Q. Gao, P. Zhao, Y. He, Adv. Funct. Mater. 2020, 30, 1906683.T. V. Neumann, E. G. Facchine, B. Leonardo, S. Khan, M. D. Dickey, Soft Matter 2020, 16, 6608.A. Uppal, W. Kong, A. Rana, R. Y. Wang, K. Rykaczewski, ACS Appl. Mater. Interfaces 2021, 13, 43348.V. Okatenko, L. Castilla‐Amoros, D. C. Ctoian, J. Vavra, A. Loiudice, R. Buonsanti, J. Am. Chem. Soc. 2022, 144, 10053.J. W. Boley, E. L. White, R. K. Kramer, Adv. Mater. 2015, 27, 2355.S. Liu, S. N. Reed, M. J. Higgins, M. S. Titus, R. Kramer‐Bottiglio, Nanoscale 2019, 11, 17615.X. Li, M. Li, J. Xu, J. You, Z. Yang, C. Li, Nat. Commun. 2019, 10, 3514.J. Kim, J. Jeong, Y. Hyun, S. K. Chumg, J. Lee, Small 2021, 17, 2104143.J. Yan, M. H. Malakooti, Z. Lu, Z. Wang, N. Kazem, C. Pan, M. R. Bockstaller, C. Majidi, K. Matyjaszewski, Nat. Nanotechnol. 2019, 14, 684.Z. J. Farrell, C. J. Thrasher, A. E. Flynn, C. E. Tabor, ACS Appl. Nano Mater. 2020, 3, 6297.Q. Wei, M. Sun, Z. Wang, J. Yan, R. Yuan, T. Liu, C. Majidi, K. Matyjaszewski, ACS Nano 2020, 14, 9884.H. Bark, M. W. M. Tan, G. Thangavel, P. S. Lee, Adv. Energy Mater. 2021, 11, 2101387.D. Kim, P. Thissen, G. Viner, D.‐W. Lee, W. Choi, Y. J. Chabal, J.‐B. Lee, ACS. Appl. Mater. Interfaces 2013, 5, 179.I. D. Joshupura, H. R. Ayers, G. A. Castillo, C. Ladd, C. E. Tabor, J. J. Adams, M. D. Dickey, ACS Appl. Mater. Interfaces 2018, 10, 44686.C. Pan, E. J. Markvicka, M. H. Malakooti, J. Yan, L. Hu, K. Matyjaszewski, C. Majidi, Adv. Mater. 2019, 31, 1900663.Y. Liu, W. Zhang, H. Wang, Mater. Horiz. 2021, 8, 56.P. Wang, Q. Hu, B. Lv, J. Zhu, W. Ma, Z. Dong, J. Wei, J. Sun, Nanotechnology 2020, 21, 145703.P. Wang, Q. Hu, B. Lv, Y. Liu, J. Yin, J. Sun, Mater. Adv. 2021, 2, 4414.N. Xia, N. Li, W. Rao, J. Yu, Q. Wu, L. Tan, H. Li, L. Gou, P. Liang, L. Li, X. Meng, Nanoscale 2019, 11, 10183.P. Zhu, S. Gao, H. Lin, X. Lu, B. Yang, L. Zhang, Y. Chen, J. Shi, Nano Lett. 2019, 19, 2128.J.‐J. Hu, M.‐D. Li, Y. Chen, F. Gao, S.‐Y. Peng, B.‐R. Xie, C.‐X. Li, X. Zheng, X.‐Z. Zhang, Biomaterials 2019, 207, 76.H. Li, Z. Bian, J. Zhu, Y. Huo, H. Li, Y. Lu, J. Am. Chem. Soc. 2007, 129, 4538.V. Sivan, S.‐Y. Tang, A. P. O'Mullane, P. Petersen, N. Eshtiaghi, K. Kalantar‐Zadeh, A. Mitchell, Adv. Funct. Mater. 2013, 23, 144.M. B. Ghasemian, M. Mayyas, S. A. Idrus‐Saidi, M. A. Jamal, J. Yang, S. S. Mofarah, E. Ababifiroozjaei, J. Tang, N. Syed, A. P. O'Mullane, T. Daeneke, K. Kalantar‐Zadeh, Adv. Funct. Mater. 2019, 29, 1901649.Y. Zhang, Z. Guo, H. Zhu, W. Xing, P. Tao, W. Shang, B. Fu, C. Song, Y. Hong, M. D. Dickey, T. Deng, J. Am. Chem. Soc. 2022, 144, 6779.S. Moon, H. Kim, K. Lee, J. Park, Y. Kim, S. Q. Choi, iScience 2021, 24, 103183.J. Kim, J. Lee, Small 2022, 18, 2108069.T. V. Neumann, M. D. Dickey, Adv. Mater. Technol. 2020, 5, 2000070.C. P. Ambulo, M. J. Ford, K. Searles, C. Majidi, T. H. Ware, ACS Appl. Mater. Interfaces 2021, 13, 12805.D. Wu, D. Liu, X. Tian, C. Lei, X. Chen, S. Zhang, F. Chen, K. Wu, Q. A Fu, Small Methods 2022, 6, 2200246.R. G. Larson, The Structure and Rheology of Complex Fluids, Oxford University Press, Inc, New York, NY 1999.R. Aveyard, B. P. Binks, J. H. Clint, Adv. Colloid Interface Sci. 2003, 100–102, 503.J. Park, T. Phongpreecha, J. Nicholas, Y. Qi, Acta Mater. 2020, 199, 551.M. Yunusa, G. J. Amador, D.‐M. Drotief, M. Sitti, Nano Lett. 2018, 18, 2498.I. D. Joshipura, A. Persson, V. K. Truong, J.‐H. Oh, M. Kong, M. H. Vong, C. Ni, M. Alsafatwi, D. P. Parekh, H. Zhao, Langmuir 2021, 37, 10914.M. Dickey, Phys. Today 2021, 74, 30.Y. Sun, Y. Li, X. Dong, X. Bu, J. W. Drelich, Colloids Surf. A 2020, 591, 124562.J. Long, Z. He, P. Zhou, X. Xie, C. Zhou, W. Hong, W. Hu, Adv. Mater. Interfaces 2018, 5, 1800353.J. N. Israelachvili Intermolecular and Surface Forces, Academic Press, Oxford 2011.F. M. Fowkes, Ind. Eng. Chem. 1964, 56, 40.D. K. Owens, R. C. Wendt, J. Appl. Polym. Sci. 1969, 13, 1741.S. Handschuh‐Wang, Y. Chen, L. Zhu, X. Zhou, ChemPhysChem 2018, 19, 1584.J. Lee, Z.‐L. Zhou, G. Alas, S. H. Behrens, Langmuir 2015, 31, 11989.S. S. Datta, D. D. Gerrard, T. S. Rhodes, T. G. Mason, D. A. Weitz, Phys. Rev. E 2011, 84, 041404.E. Irani, P. Chaudhuri, C. Heussinger, Phys. Rev. Lett. 2014, 112, 188303.S. F. Velandia, P. Marchal, C. Lemaitre, V. Sadtler, T. Roques‐Carmes, J. Colloid Interface Sci. 2021, 589, 286.P. L. Fuhrmann, S. Breunig, G. Sala, L. Sagis, M. Stieger, E. Scholten, J. Colloid Interface Sci. 2022, 607, 389.M. Destribats, S. Gineste, E. Laurichesse, H. Tanner, F. Leal‐Calderon, V. Heroguez, V. Schmitt, Langmuir 2014, 30, 9313.J. Xiao, X. Wang, A. J. Perez Gonzeles, Q. Huang, Food Hydrocolloids 2016, 54, 30.I. A. Morrison, Colloids Surf. A 1993, 71, 1.M. M. Gacek, J. C. Berg, Adv. Colloid Interface Sci. 2015, 220, 108.S. Lim, H. Park, J. Yang, C. Kwak, J. Lee, Colloids Surf. A 2019, 579, 123648.Q. Xu, N. Oudalov, Q. Guo, H. M. Jaeger, E. Brown, Phys. Fluids 2012, 24, 063101.F. Kargar, Z. Barani, R. Salgado, B. Debnath, J. S. Lewis, E. Aytan, R. K. Lake, A. A. Balandin, ACS Appl. Mater. Interfaces 2018, 10, 37555.A. Uppal, M. Ralphs, W. Kong, M. Hart, K. Rykaczewski, R. Y. Wang, ACS Appl. Mater. Interfaces 2020, 12, 2625.X. Xu, M. Fan, Z. Yu, Y. Zhao, H. Zhang, J. Wang, M. Wu, F. Sun, X. Xu, C. Ding, J. Li, Chem. Eng. J. 2022, 429, 132491.

Journal

Advanced Materials Interfaces – Wiley

Published: Apr 1, 2023

Keywords: adhesion; colloid; EGaIn; emulsion; liquid metal; oxide; rheology

There are no references for this article.