METALLIC FOAMS AND METHODS FOR PRODUCING THEM

Abstract

An aqueous suspension for producing porous metallic structures comprises 2-49 vol % of a mixture of at least one chemical compound comprising a metal atom, wherein said at least one compound is solid at room temperature, has the form of a powder, and is suspended in water, and 10-9) to 0.1 mol of a surfactant per mol of said chemical compound comprising a metal atom. The aqueous suspension is part of a foam and/or is part of an oil-in-water emulsion, comprising 30-90 vol % of a lipophilic phase, said lipophilic phase not comprising a polymerizable compound.

Claims

1. An aqueous suspension for producing porous metallic structures, the aqueous suspension comprising 2-49 vol % of a mixture of at least one chemical compound comprising a metal atom, wherein said at least one compound is solid at room temperature, has the form of a powder, and is suspended in water, and 10.sup.9 to 0.1 mol of a surfactant per mol of said chemical compound comprising a metal atom; wherein the aqueous suspension is part of a foam; and/or wherein the aqueous suspension is part of an oil-in-water emulsion, comprising 30-90 vol % of a lipophilic phase, said lipophilic phase not comprising a polymerizable compound.

2. The aqueous suspension according to claim 1, wherein one or more of said chemical compounds comprising a metal atom is a compound that is reducible to metal in a gaseous atmosphere.

3. The aqueous suspension according to claim 1, wherein the mixture of at least one chemical compound comprising a metal atom is a metal oxide powder.

4. The aqueous suspension according to claim 3, wherein the metal oxide powder is Fe.sub.3O.sub.4, or NiO, or a mixture thereof.

5. The aqueous suspension according to claim 4, wherein the particles of the metal oxide powder have a prolate shape.

6. The aqueous suspension according to claim 1, wherein one or more of said chemical compounds comprising a metal atom is a metal hydride.

7. The aqueous suspension according to claim 6, wherein said metal hydride is TiH.sub.x, PdH.sub.x, ZrH.sub.x, or MgH.sub.2.

8. The aqueous suspension according to claim 1, wherein one or more of said chemical compounds comprising a metal atom is a metal carbonyl compound.

9. The aqueous suspension according to claim 8, wherein said metal carbonyl compound is Rh.sub.2(CO).sub.8, or Ru(CO).sub.5.

10. The aqueous suspension according to claim 1, wherein the surfactant is a cationic surfactant, or a non-ionic surfactant.

11. The aqueous suspension according to claim 1, wherein the aqueous suspension comprises 1-7 g/l of a binder.

12. The aqueous suspension according to claim 11, wherein said binder is methylcellulose, PVA (poly vinyl alcohol), or PVP (poly vinyl pyrrolidone), or a mixture thereof.

13. A method for producing a porous metallic material, comprising the steps: providing an aqueous suspension according to claim 1; comprising a mixture of at least one chemical compound comprising a metal atom; foaming said aqueous suspension to a foam, and/or emulsifying said aqueous suspension with a lipophilic compound to an oil-in-water emulsion, comprising 30-90 vol % of a lipophilic phase, said lipophilic phase not comprising a polymerizable compound; forming with said foam or emulsion a three-dimensional structure; drying said three-dimensional structure, resulting in a dry structure; and subjecting the dry structure to process conditions that result in the conversion of the chemical compounds comprising a metal atom to metal, resulting in a metallic structure.

14. The method according to claim 13, wherein the metallic structure is sintered.

15. The method according to claim 13, wherein one or more of said chemical compounds comprising a metal atom is a compound that is reducible to metal in a gaseous atmosphere; and wherein the dry structure is reduced by subjecting it to a reducing atmosphere, resulting in a metallic structure.

16. The method according to claim 15, wherein the reduction is carried out in an atmosphere of 0.5-100 vol % H2 in an inert gas.

17. The method according to claim 13, wherein one or more of said chemical compounds comprising a metal atom is an interstitial metal hydride; and wherein the dry structure is subjected to a temperature under which the hydrogen is released as hydrogen gas, resulting in a metallic structure.

18. The method according to claim 13, wherein the processing step resulting in the metallic structure and the sintering step are carried out in one process step.

19. The method according to claim 13, wherein the three-dimensional structure is formed by additive manufacturing, for example by three-dimensional printing.

20. The method according to claim 13, wherein the surface of the metallic structure is hydrophobized or lipophilized.

21. The method according to claim 13, wherein after the reduction step, the metallic structure is at least partially oxidized.

22. A material produced according to the method of claim 13.

23. Use of a material according to claim 22 as one of: a resistive heating element; a cooling element; a material for absorbing lipophilic substances floating on water; a heterogeneous catalyst; a carrier structure for a heterogenous catalyst; an electrode; a component in alkaline electrolysis; a storage for gases, in particular molecular hydrogen; a filtration element.

24.-30. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0105] In order to facilitate a fuller understanding of the present invention, reference is now made to the appended drawings and figures. These references should not be construed as limiting the present invention and are intended to be exemplary only.

[0106] Components that are identical, or that are identical at least in terms of their function, are designated below by identical or similar reference numbers.

[0107] FIG. 1 schematically shows the manufacturing process of iron-based hierarchical porous structures according to the invention.

[0108] FIG. 2 shows strain-controlled oscillatory rheology measurements on a foamed aqueous suspensions according to the invention.

[0109] FIG. 3 shows the 3D-printing process forming the wet foam body.

[0110] FIG. 4 shows (a) the heating protocol of the reducing step applied to the green body of FIG. 5, and (b) the XRD spectrum of the resulting fully reduced sample.

[0111] FIG. 5 shows the hierarchical pore structure of a green body after 3D printing the ink of FIG. 2.

[0112] FIG. 6 shows the hierarchical pore structure of the metal body after full reduction of the green body of FIG. 5.

[0113] FIG. 7 shows microstructures iron-based porous structures according to the invention, subjected to different processing parameters.

[0114] FIG. 8 shows measurements of different mechanical properties of the iron-based porous structures of FIG. 7.

[0115] FIG. 9 shows also mechanical properties of the iron foams of FIG. 7, (a) stress-strain curves for an extended range of applied strain, and (b) averaged mechanical properties of the samples obtained from the stress-strain data.

[0116] FIG. 10 shows an XRD spectrum of a nickel-iron alloy based metal foam according to the invention (7 wt % Ni, 850 C., 30 h) with purely austenitic atomic structure.

[0117] FIG. 11 shows a SEM image of the nickel-iron alloy metal foam of FIG. 10.

[0118] FIG. 12 shows (a) the averaged relative density, (b) the density, (c) the compressive modulus, and (d) the strength of nickel-iron alloy metal foams with different Ni content, for two different sintering temperatures.

[0119] FIG. 13 shows the Weibull distribution for the strength of the nickel-iron alloy metal foam in FIG. 12 containing 28 wt % Ni and sintered at 1000 C. for 30 h.

[0120] FIG. 14 shows different examples of iron-based foams according to the invention, from left to right: unreduced magnetite foam, maghemite foam obtained upon oxidation of magnetite at 550 C. for 2 h, 37 wt % maghemite and 63 wt % hematite foam obtained upon oxidation of magnetite at 650 C. for 2 h, sintered hematite foam obtained by sintering at 1000 C. for 2 h.

[0121] FIG. 15 shows the composition of crystalline phases present in an iron porous structure (850 C. for 30 h) after corrosion in phosphate buffer (37 C.) for 1-90 days, determined by Rietveld fitted XRD spectra.

[0122] FIG. 16 shows SEM images of an iron based porous structure sample sintered at 850 C. for 30 h, and subsequently corroded in phosphate buffer solution (37 C.) for 3 days.

[0123] FIG. 17 shows (a) the compressive modulus, and (b) the strength of an iron-nickel based metallic foam according to the invention.

[0124] FIG. 18 shows the surface area of printed structures prepared from foams containing different fractions of iron oxide particles and heat-treated at distinct reduction/sintering conditions.

[0125] FIG. 19 shows SEM images of printed structures prepared from metal foams containing different fractions of iron oxide particles and heat-treated at distinct reduction/sintering conditions.

[0126] FIG. 20 shows resistive heating of iron-based foams for vaporization of liquids. (a) shows time-lapsed infrared images of a 35% Fe foam (5.6 at room temperature) subjected to an electric voltage at 3 W power. (b) displays the average temperature profile of the area marked in (a), with the time points of the snapshots in (a) being indicated with dashed lines in (f). (c) shows an optical image and a thermal images (small figure lower right corner) of on-demand liquid evaporation via resistive heating using a foam infused with etheric oil (limonene, 0.4 mL/g).

[0127] FIG. 21 shows iron-based foam elements with low density (<1 g/cm.sup.3) and with their surface modified with hydrophobic fumed silica, absorbing oil while floating on water.

[0128] FIG. 22 shows a metal foam produced by 3D printing of an emulsion based ink, (a) after printing, and (b) SEM image before sintering.

[0129] FIG. 23 shows a plot of a stress-strain measurement of an iron-nickel based metallic foam.

[0130] FIG. 24 shows the relative compressive modulus vs. the relative density of different metal foams after a first and a second measurement.

[0131] FIG. 25 shows different aspects of measurements regarding a novel stiffening effect of metal foams according to the invention.

[0132] FIG. 26 shows changes of the compressive modulus of a metal foam body according to the invention after a series of uniaxial compressions along different axes.

[0133] FIG. 27 shows the engineering strain in low-carbon metal foams extracted from digital image correlation (a,c,f) with the corresponding Froce-Stroke (b) and Stress-Strain curves (d).

[0134] FIG. 28 illustrates an assumed mechanism of strain stiffening, with (a), (b), (c) a schematic stress-strain curve; (d) a schematic depiction of the porosity of an metal foam according to the invention on three different length scales; and (e) and SEM images (second and third image) on a real metal foam sample after a first compression cycle.

[0135] FIG. 29 shows (a) a stress-strain measurement on a metal foam sample, over the complete range of elasticity, yielding and densification, with a repeated release of the sample; (b) the detail view of the measured stress-strain curve of (a) in the elastic range; (c) the elastic modulus E obtained from the measurement in (a), plotted against the strain; and (d) the obtained elastic modulus plotted against the solid content ratio of the sample.

[0136] The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the present invention, in addition to those described herein, will be apparent to those skilled in the art from the foregoing description and accompanying drawings. Thus, such modifications are intended to fall within the scope of the appended claims. Additionally, various references are cited throughout the specification, the disclosures of which are each incorporated herein by reference in their entirety.

DETAILED DESCRIPTION OF THE INVENTION

[0137] An aqueous suspension of metal atom containing precursor particles according to the invention allows to realize an extrusion-based 3D printing process for manufacturing metal foams, for example steel or iron-based porous structures with ultra-high porosity distributed over three hierarchical levels. The precursor particles also stabilize the foamed ink.

[0138] Using particle-stabilized foams of aqueous suspensions according to the invention as 3D printing inks, pores can by generated at multiple length scales. On a first length scale, porosity can be directly controlled by the printing pattern of the foam filament. On a second, smaller length scale, gas bubbles of the foamed inks serve as templates for the formation of a large fraction of mesopores within the printed filaments. The structure can be controlled via the concentration and distribution of the bubbles in the ink. On a third, even smaller length scale, porosity results from the thermal process leading to conversion of the suspended precursor particles (for example reduction with hydrogen gas in the case of iron oxide particles) and an optional sintering process.

[0139] This enables for example the manufacturing of iron and iron-based alloys with an elastic modulus above 300 MPa and a density below 1 g/cm.sup.3, which is in keeping with the mechanical efficiency expected for porous structures.

[0140] The advantageous manufacturing process for such iron-based metal foams is schematically shown in FIG. 1. The manufacturing of iron-based hierarchical porous structures is a two-step process that in a first step involves the preparation of wet foam inks stabilized by modified particles and in a second step the 3D printing of the wet foam inks, to a three-dimensional structure, followed by reduction and sintering of the dried printed structures.

[0141] The stabilization of wet foams with particles can be achieved by decreasing the wettability of the particles in a liquid and thus promoting their adsorption to the air/water interface. Particle adsorption reduces the total interfacial area and decreases the free energy of the system. This energy gain can easily be orders of magnitude higher than thermal energy, implying that the adsorption is an irreversible process [33]. The irreversible particle adsorption renders the foams nearly immune to common foam destabilization processes, such as liquid drainage, collapse by lamellar breakage, and Ostwald ripening [34].

[0142] Oblong magnetite micro particles 11 are functionalized with a short-chain amphiphilic molecule 12, FIG. 1(a), to produce an aqueous 13 suspension, FIG. 1(b). After frothing the suspension, the result is a foamed ink 1 with gas bubbles 14 stabilized by the oblong magnetite particles 11, FIG. 1(c).

[0143] It was experimentally found that the functionalization of magnetite particles with hexylamine (3 l/g particles) provides the ideal wettability to promote adsorption at the air/water interface and thus generate strong, stable foams upon air incorporation. To prepare such stable foams, magnetite particles are first suspended and de-agglomerated in water containing the short-chain amphiphile hexylamine. The positively charged polar head group of the molecule absorbs to the negatively charged particle surface due to electrostatic interactions. This makes the hydrophobic tail of the amphiphile protrude outwards and render the entire particle slightly hydrophobic. Mechanical frothing of the suspension of modified particles eventually leads to the desired ultra-stable foams. A particle volume fraction of 50 wt % and an optimized foaming procedure (10 min at 800 rpm) provide the additional rheological requirements for printing.

[0144] This stabilized viscoelastic ink 1 can then be used in a Direct Ink Writing (DIW) process for additively forming a three-dimensional structure 3a, FIG. 1(d). The particle-stabilized foamed ink 1 is printed layer-by-layer, by extruding the foamed ink 1 through a printing nozzle 21 at room temperature and depositing it as a continuous foam filament 31.

[0145] The DIW approach is especially suitable to fabricate open cell structures, such as grid-like and open cellular designs.

[0146] There are three rheological requirements for most direct ink writing inks: First, the ink must be shear thinning, which allows the material to flow through the nozzle at low pressures. Second, it must have a high yield stress, below which the ink does not flow, to ensure that the material remains in place after printing. Third, the ink must also show a high storage modulus to prevent the sagging of spanning filaments in grid-like structures. For compressible foams, the printing conditions also need to be adjusted to ensure continuous deposition rather than stop-and-go extrusion [35].

[0147] After drying of the printed wet foam body 3a, the resulting green body 3b (FIG. 1(e)) retains the cellular structure of the foam. The green body 3b is reduced and sintered in a tubular 22 oven under constant H.sub.2 gas flow, FIG. 1(e), to a highly porous, light-weight iron foam element 3b, FIG. 1(f).

[0148] To quantify the rheological behavior and the viscoelastic properties of the particle-stabilized wet foams, strain-controlled oscillatory rheology measurements were performed (cf. FIG. 2). The investigated wet foam comprised 50 wt % iron oxide particles (50 wt %) 3 l hexylamine per g iron oxide. The foam shows a clear viscoelastic response with a storage modulus plateau at low strains and a well-defined yield stress (cf. FIG. 2). The measured yield stress of 70 Pa is not far from the range recommended for DIW inks (100-5000 Pa) [36]. With a high plateau storage modulus of 4.86 kPa, the foams are also stiff enough to enable direct ink writing of grid-like architectures.

[0149] DIW printing of the viscoelastic wet foam using a desktop extrusion-based printer equipped with a syringe and nozzle size of 0.84 mm leads to a three-dimensional object 3a (cf. FIG. 3). To create a hierarchical architecture, a grid-like design was chosen, where the print path controls the size of the open cells, which will hereafter be referred to as macropores 32. Such macropores constitute the first level of the hierarchical porosity of the metallic material according to the invention.

[0150] The ability to print distortion-free grids confirms the suitable rheological properties of the ink and indicates that the wet foams are resistant to the high shear forces applied during extrusion-based printing, which could otherwise cause phase separation and bubble collapse. Moreover, the experiments revealed that the high stability and viscoelastic nature of the particle-stabilized wet foams mitigate any compressibility issues under typical printing conditions.

[0151] After printing, the wet foam object 3a is dried at 60 C. for 1 h, leaving in the in the grid-like structure of the resulting green body 3b (cf. FIG. 5(a)) only the solid precursor particles and remaining organic compounds. Drying leads to the formation of the second hierarchical level of the structure, which consists of spherical pores 33 originating from the former foam cells of the wet foam, as can be seen in the SEM images in FIGS. 5(b) and (c). The dried structure of the green body 3b shows a closed-cell porosity with an average pore size of 45.6 m. Because they fall in an intermediate size range, these spherical pores will hereinafter referred to as mesopores 33.

[0152] The size of these mesopores can be adjusted by the volume fraction of particles, amphiphile concentration, and foaming speed. Generally, increasing the particle and the amphiphile content will increase the suspension viscosity, and thereby reduce the size of the bubbles due to the higher shear stresses generated during mechanical frothing. The increased shear resulting from higher foaming speeds can also reduce the bubble size.

[0153] The green body 3b can be converted into a hierarchically porous metal body 3c by thermal treatment under a reducing atmosphere at temperatures ranging from 690 to 1000 C. The green body 3b of FIG. 5 was heat-treated under a forming gas stream of 5% H2 and 95% N2 for 30 hours at 850 C. The corresponding temperature profile is shown in FIG. 4(a). The XRD spectrum of the fully reduced metal body 3c shows clear correlation with the diffraction peaks expected for a ferritic (-Fe) atomic structure, thereby proofing a complete reduction of the initial magnetite to ferritic iron (cf. FIG. 4(b)). Under the given parameters, sintering took take parallel to reduction of the magnetite particles to iron particles. Chemical analysis of the reduced samples revealed that the porous iron structure contains 0.2040.077 wt % carbon, with organic compounds in the wet foam being the carbon source.

[0154] The combined reduction and sintering process leads to a linear shrinkage of approx. 50%, as can be seen when comparing the metal foam body 3c shown in FIG. 6(a) to the green body 3b shown in FIG. 5(a). Reduction and sintering also lead to an opening of the closed-celled mesopores 32 created during foaming, as can be seen in the SEM images of FIGS. 6(b) and (c), resulting in the formation of micropores 34 in the walls of the bubble-templated mesopores 33. Combined with the macropores 32 created by the print path, this leads to a unique open-celled porous structure with three levels of hierarchical porosity.

[0155] To demonstrate the possibility of adjusting the bubble size, and thus of the mesopores of the material according to the invention, the particle concentration of the foam was reduced from 50 wt %, as in the example discussed above, to 40 wt %, while keeping the amphiphile concentration and the foaming procedure constant. Foams with such lower particle concentration were found to be still sufficiently stiff enough for 3D-printing grid-like structures. Foams with 40 wt % particles lead to larger mesopores with an average diameter of 125.1 m.

[0156] The reduction process is affected by several factors, including sample volume, gas flow, H.sub.2 and H.sub.2O content of the reducing gas, oxide particle size, and temperature. Depending on the temperature and particle size, the reduction is also accompanied by sintering processes of the metal particles and metal oxide particles. To explore part of this ample processing parameter space, the effect of different reduction protocols on the microstructure, relative density, and mechanical properties of the resulting hierarchical porous materials was investigated. To this end, specimens produced with the 50 wt % magnetite ink discussed above were systematically heat-treated for 30 hours at 750 C., 850 C., and 1000 C., and for optimized conditions (40 hours at 690 C.). SEM images of the green body prior to reduction and the metal foam after reduction are shown in FIG. 7. Several mechanical properties of the samples have been measured, with the results shown in FIGS. 8(a), (b) and (c). The symbols used in the plots correspond to the square, circle, diamond and hexagon symbols of the respective samples in FIG. 7.

[0157] FIG. 8(a) shows a plot of the stress-strain curves for a sample reduced and sintered under optimized conditions (690 C., 40h, sintered) in comparison to samples reduced at 750 C., 850 C., and 1000 C., for 30 h. FIG. 9(a) shows the stress-strain curves of the samples for an extended range of strain, while FIG. 9(b) shows the elasticity modulus and the yield strength obtained from the stress-strain data.

[0158] FIG. 8(b) shows the absolute and relative compressive modulus (compared to the solid material) plotted vs. the relative density compared to the solid metal. FIG. 8(c) shows the yield strength as plotted vs. the relative density compared to the solid metal. Literature data for iron-based printed structures without templated macroporosity are displayed for comparison. (grey triangles, Taylor et al. [26]; grey star, Sharma et al. [7]). Theoretical scaling dependences expected from an open-cell foam model are also displayed (cf. L. J. Gibson, M. F. Ashby, Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences 1982, 382, 43).

[0159] A marked change can be observed in the microstructure of samples heat-treated between 750 C. and 1000 C. (FIG. 7). Heat treatment at 750 C. was found to be sufficient for the formation of micro-pores on the walls of meso-pores, while preserving the original foam-like architecture of the green body. At 850 C., the micropores increase significantly in size, but the foam-templated mesopores remain visible. Increasing the temperature to 1000 C. results in a lower surface area and in the fusion of the micro-pores with the meso-pores. This leads to an open-celled network with only two levels of hierarchical porosity.

[0160] Along with the microstructural changes, the determined carbon content decreases from 0.200.08% at 850 C. to only 0.050.01% C at 1000 C. Such a reduction is likely caused by the long exposure of the high-surface area structure to high temperatures, which allows for carbon diffusion out of the material. From this microstructural analysis, it is concluded that high temperatures should be avoided in order to retain a high carbon content and a three-level hierarchical porosity.

[0161] Particularly advantageous hierarchical steel structures with 3-level porosity were eventually obtained by employing a two-step procedure, in which samples are first fully reduced at 690 C. for 40 h and afterward sintered for 2 h at 850 C. This provides the microstructure of samples reduced at low temperature, but with the thicker struts resulting from sintering.

[0162] The mechanical characterization reveals that the elastic modulus and the compressive yield strength of the porous metal structures increase with the heat-treatment temperature (FIGS. 8(a), 9(a), (b)). Since the relative density of all samples remained within the comparable range between 0.053 and 0.268, this stiffening effect can be attributed to the stronger solid network formed for samples sintered at high temperatures. The optimum heat treatment protocol provides by far the stiffest and strongest structures, while retaining an very low average density of 0.860.03 g/cm.sup.3.

[0163] To compare these hierarchical porous structures with other iron-based porous materials, the mechanical data are shown in an Ashby-type plot depicting the effect of the relative density on the elastic modulus and compressive strength of previously reported structures (FIGS. 8(b), (c)). Compared to state-of-the-art materials [7, 26], it is found that the presence of three hierarchical levels in the porous structures according to this disclosure leads to significantly lower relative densities, which lie within a previously inaccessible parameter range. Importantly, the elastic modulus and compressive strength of the materials according to this disclosure follow a similar dependence on the relative density as that observed for the reference structures. This dependence is in good agreement with theoretical predictions from the Ashby-Gibson model for open-cell foams, thus indicating that the shown novel materials combine very low weight with the maximum achievable properties. By reducing the relative density without sacrificing the predicted properties, the hierarchical porous structures reach state-of-the-art mechanical efficiency using 30% less solid phase.

[0164] The proposed hierarchical pore design can be extended to a range of other iron oxides and iron-based alloys. To illustrate this, hierarchical porous NiFe alloys using NiO and Fe.sub.3O.sub.4 particles as inorganic precursors in the initial printable foam. Nickel is a common alloying material for steel, as it ensures that the steel retains its tough high-temperature crystal structure, austenite.

[0165] The nickel content in the alloy can be controlled precisely by tuning the relative fraction of the oxide powders used in the foam formulation. This was validated by elemental analysis (Proton Induced X-Ray Emission) of an alloy designed to contain 28 wt % (calculated) nickel. After heat treatment of the dried precursor foam, the alloy was found to contain 27.3 wt % Ni atoms. X-ray diffraction analysis confirmed the formation of austenite in the heat-treated NiFe alloy (FIG. 10). Interestingly, the microstructure of the NiFe porous structure was observed to be independent of the sintering parameters and Ni-alloying content (FIG. 11).

[0166] FIGS. 8(d), (e) and (f) show the results of measurements equivalent to FIGS. 8(a), (b), (c) for metal foam samples produced with different process temperatures and with varying nickel content in the final metal foam, namely 0, 7, 14, 21, 28 and 35 wt % Ni, with the remaining part being iron. Note that in FIG. 8(d), the curve for the 21 wt % Ni sample stops before the sample has yielded, due to technical limits. Again all data are compared to the literature data obtained for iron-based printed structures without templated macroporosity (grey triangles, Taylor et al. [26]; grey star, Sharma et al. [7]). Theoretical scaling dependences expected from an open-cell foam model are also displayed (Gibson, Ashby, 1993). FIG. 12 shows (a) the relative density, (b) the density, (c) the elasticity modulus, and (d) the yield strength for the same nickel-iron alloy metal foams, for two different sintering temperatures. FIG. 13 shows the Weibull distribution for the strength of the nickel-iron alloy metal foam in FIG. 12 containing 28 wt % Ni and sintered at 1000 C. for 30 h.

[0167] In terms of mechanical properties, the measurements show that the addition of nickel improves both the elastic modulus and the compressive strength of the porous Fe-based alloys considerably. Similar to the pure iron compositions, the presence of three hierarchical porosity levels results in much lower relative densities compared to previously reported iron-based porous materials. Because the measured mechanical properties still follow the trend expected from theoretical predictions, this reduction in relative density is achieved without compromising the mechanical efficiency of the porous structure. Compared to the pure iron counterparts with similar densities and processing conditions, the presence of nickel in the iron-based alloy was found to increase the elastic modulus of the porous structures by a factor of 7.5, and the compressive strength by a factor of 5.

[0168] Iron-based porous structures can be used as a pure metal foams, as metal oxide foams or as foams of mixtures of metal and metal oxide, in a wide range of applications. Iron oxide foams with tailored chemical compositions can be obtained by heat-treating the 3D-printed green body objects at intermediate temperatures in the range of 550-650 C. A two-hour heat treatment at 550 C. air atmosphere leads to the partial oxidation of magnetite to brown-colored maghemite. Red-colored hematite structures can be obtained by heat-treating a 3D-printed magnetite foam at 650 C. for 2 hours in air. FIG. 14 shows such variants of 3D-printed lightweight, permeable, and high-surface-area structures according to the invention (from left to right): unreduced magnetite foam, maghemite foam obtained upon oxidation of magnetite at 550 C. for 2 h (brown), 37 wt % maghemite/63 wt % hematite foam obtained upon oxidation of magnetite at 650 C. for 2 h (red), sintered hematite foam obtained by sintering at 1000 C. for 2 h (silver). While these oxide foams are rather soft, their mechanical properties can be increased by the addition of binders or by sintering. For example, sintering the hematite specimen at 1000 C. results in a mechanically stronger porous structure with a smooth and shiny surface whose silver color is characteristic of sintered hematite. Such oxide foams can potentially be utilized as substrates for gas catalysis, such as methanol oxidation [37], toluene cracking [38], or the decomposition of formic acid [39].

[0169] The porous metal structures themselves show the intrinsic magnetic properties and electrical resistivity of iron (0.00125 m).

[0170] Mixed iron/iron oxide porous structures can be obtained through partial oxidation or iron oxide green bodies. The composition of crystalline phases present in an iron porous structure (850 C. for 30 h) after oxidative corrosion in phosphate buffer (37 C.) for 1-90 days, determined by Rietveld fitted XRD spectra, is shown in FIG. 15 and Table 1. SEM images of an iron based porous structure sample sintered at 850 C. for 30 h, and subsequently corroded in phosphate buffer solution (37 C.) for 3 days, is shown in FIG. 16.

TABLE-US-00001 TABLE 1 Composition after corrosion in phosphate buffer (37 C.) for 1-90 days determined by Rietveld fitted XRD spectra Composition [%] Time Wstite Iron oxide Magnetite Iron oxide phosphate [d] Iron FeO FeO Fe.sub.3O.sub.4 Fe.sub.9O.sub.8(PO.sub.4) 1 53.9 10.5 5.7 3.2 26.7 3 52.1 12 5.6 2.9 27.3 7 25.1 19.5 20.7 11.7 22.8 30 35.2 15.5 19 9.2 21.1 90 23.5 20 21.6 12.4 22.5

[0171] The (a) relative and absolute compressive modulus and the (b) relative and absolute yield strength, and (b) the relative strength of an iron-nickel based metallic foam according to the invention are shown in FIG. 17, together with literature values (grey triangles, Taylor et al. [26]) and the theoretical values predicted by Gibson-Ashby for an open-cell structure.

[0172] The determined surface area of printed structures prepared from mixed foams containing different fractions of iron oxide particles and heat-treated at distinct reduction/sintering conditions is shown FIG. 18.

[0173] SEM images of metal foams containing different fractions of iron oxide particles (40, 45, and 50 wt % iron oxide compared to iron) and heat-treated at distinct reduction/sintering conditions are depicted in FIG. 19. (a) 690 C., 40 h, with sintering; (b) 1000 C., 30 h; (c) 850 C., 30 h; (d) 815 C., 30 h; 730 C., 30 h; (f) 730 C., 15 h; (g) 650 C., 70 h; (f) before reduction.

[0174] Because of the higher electrical resistivity of mixed iron/iron oxide foams compared to pure iron counterparts (0.7 m for 35 wt % iron at room temperature), these porous materials can be used for resistive heating, for example as low-power resistive evaporators. It was experimentally observed that Joule heating with only 3 W power is enough to heat up an 8 mm wide cubic porous structure to a steady temperature of 180 C. Steady state is reached due to a balance between the resistive heating and the convective cooling enabled by the high surface area of the structure (1 m.sup.2/g).

[0175] FIG. 20 shows the application of metal/metal oxide foams according to the invention as resistive heating elements, for example for the on-demand vaporization of liquids.

[0176] FIG. 20(a) shows an experimental setup for liquid evaporation via resistive heating. A metal foam body 3c with 35 wt % Fe infused with etheric oil (limonene, 0.4 ml/g) is clamped between two electrodes 23, 23. The metal foam body is subjected to an electric voltage resulting in a 3 W power output. The inserted image in the lower right corner shows an infrared image of the area marked by dotted lines. FIG. 20(b) shows time-lapsed infrared images. The average temperature over time in the sampling area marked with dotted lines is displayed in FIG. 20(c). The dashed lines indicate the time points of the snapshots in FIG. 20(b).

[0177] Moreover, the very low density (<1 g/cm.sup.3) of the metal foam structures according to the invention makes them sufficiently light to swim on water, which is remarkable in view of the density of pure iron (approx. 8 g/cm.sup.3). The hierarchical open-celled structure induced by the reduction procedure acts like a sponge to liquids that can wet the metal surface. Under favorable wetting conditions, capillary forces cause spontaneous absorption of at least 0.4 ml/g of liquid into the porous structure. The capillary forces were observed to be strong enough to suck liquid not only into the micro-pores but also into the 3D printed channels.

[0178] Because such metal foam bodies can float on water, they can be used as environmentally friendly oil absorbers. For this application, the surface of the metallic structure is hydrophobized with fumed silica via dip-coating. This hydrophobization procedure makes the surface wettable by oils and other apolar solvents, ensuring the spontaneous absorption of these liquids into the porous structure. As can be seen in FIG. 21, such hydrophobized iron foam bodies can absorb hexane (dyed for better visibility) floating on a water surface very effectively. An oil droplet 41 on a water surface is absorbed by a metal foam body 3c within 20 s. The oil wets the hydrophobized surface of the metal foam, and the oil is contained within the micro-pores and macro-pores of the structure.

[0179] A particular advantage of such a use of porous structures as oil-absorbers comes from their ability to swim on water, which enables absorption of the oil right where it resides. As the oil replaces the air in the pores and channels, the density of the structures increases until they start to sink. This allows for crude oil-spill clean-up where the oil is contained in the metal bodies and then deposited down to the ocean floor.

[0180] Another method for producing metal foams according to the invention is using an emulsion-based ink system instead of a wet-foam based ink system. Compared to a wet-foam composition, the surfactant concentration is increased by a factor of 5-50 (examples: 1 mmol/g of hexylamine, 0.1 mmol/g of octyl gallate). 30-80 vol % of an hydrophobic organic solvent (such as n-Octane) are added dropwise while mixing. The pH can be adjusted to the pKa of the surfactant (with NaOH, or HCl, respectively) for improving emulsification. Binders like methylcellulose, PVA (poly vinyl alcohol), and PVP (poly vinyl pyrrolidone) can be used to stabilize the resulting emulsion (1-7 g/L). The emulsion based ink can be processed (printed, dried, sintered, reduced) in the same way as the foams. The macropores in this case result from the organic solvent droplets in the emulsion, similar to the air bubbles in the foam. A metal foam produced by 3D printing of such an emulsion based ink system is shown in FIG. 22.

[0181] Furthermore, it was surprisingly found that the porous metal structures according to the invention show a special strain stiffening behavior.

[0182] Iron foams and Nickel-iron alloy foams were subjected to strain-stress measurements, and the elasticity modulus E.sub.1 was determined in the proportional range. The measurement was then repeated. Unexpectedly, the elasticity modulus E.sub.2 found in this second run was considerably higher. For a nickel-iron alloy foam with 7 wt % Ni, reduced at 850 C. for 30 h, the elasticity modulus measured in a first run was E.sub.1=565 MPa, and the second measured elasticity modulus was E.sub.2=2724 MPa (cf. FIG. 23). The same effect was found for all investigated metal foams according to the invention. FIG. 24 shows a plot of the relative elasticity modulus vs. the relative density (in relation to the solid metal) for two iron foams (process parameter sets R2 and R7 in Table 2) and for nickel-iron alloy foams with 7, 14, 21, 28 wt % Ni (process parameter set R6 in Table 2). All samples show a similar increase of the elasticity modulus in combination with an increase of the density.

[0183] The compressive modulus will improve after compression along the compression direction by a factor of 2-20 with a Poisson's ratio <0.01. This uniaxial stiffening will not be accompanied by a change in the density exceeding 2.5% and is permanent, as long as only uniaxial stress is applied. (FIGS. 24, 25) Changing the direction of applied stress will reset the material and the effect reoccurs (FIG. 26)

[0184] The effect of the increased elasticity modulus is permanent, as can be seen in FIG. 23, where the data of a stress-strain measurement on a metal foam are shown. The sample was repeatedly released, and the measurement was subsequently resumed. In contrast to the initial proportional increase, resulting in elasticity modulus E.sub.1, the elasticity modulus E.sub.2 found after the first and all subsequent release and resumption cycles was considerably (almost 500%) higher. The sample as such remains undeformed (no increase in cross-section area).

[0185] FIGS. 25(a) and (b) shows the averaged elasticity modulus and density of three metal foams (four samples each) with different solid loadings, determined in a first, a second and a third measurement run. The solid loading always refers to the weight percentage of the magnetite powder, unless specified otherwise Both values increase with the number of runs, for all metal foam types. However, the increase in density does not exceed 2.5%. FIG. 25(c) shows the corresponding results in a relative modulus vs. relative density plot. The ratio of the elasticity modulus after the second run and after the first run is shown in FIG. 25(d), while FIG. 25(e) shows the corresponding ratio of the density. FIG. 25(f) shows these values in a modulus ratio vs. density ratio plot. The results do not correspond to the predictions according to Ashby, E.sub.spec=c.sub.1 .sub.spec.sup.2, with c.sub.1=1, plotted as a line in FIGS. 25(c), (f).

[0186] The strain stiffening effect is uniaxial. FIG. 26 shows the elasticity modulus determined in a series of uniaxial stress-strain measurements (marked with encircled numbers 1-8) along different axes, namely the normal to the XY plane (Z axis), the YZ plane (X axis), the XZ plane (Y axis), and again the XY plane (Z axis) of the metal foam body sample. The strain stiffening effect takes place independently along all three axes X, Y, Z. Interestingly, the strain stiffening effect, while being permanent when compression axes remain unchanged (cf. FIG. 23), seems to be reset after the compression along all three axes, and the effect takes place a second time, as shown by the measurement pair 7, 8.

[0187] FIG. 27 shows the results of measurements on metal foam samples for investigating local stiffening along the applied direction of stress. A first metal foam sample (low-carbon steel processed at 850 C. with a relative density of 0.11) is subjected to two compression cycles (1) and (2). The measured force is plotted against the stroke length (dashed lines in FIG. 27(b). For both compression cycles, digital image correlation (DIC) shows the engineering strain along the compression direction at a force value of 70 N in a color map (cf. FIG. 27(a)), with the compression axis along the perpendicular in the figures. A second metal foam sample (low-carbon steel processed at 1000 C. with a relative density of 0.22) is subjected to two compression cycles (1) and (2), and the measured force is plotted against the stroke width (full lines in FIG. 27(b). Again, for both compression cycles, DIC strain images are obtained, at a force value of 40 N (FIG. 27(c)).

[0188] FIG. 27(d) shows a stress-strain curve measured on a third metal foam sample (low-carbon steel processed at 850 C. with a relative density of 0.21 and a 3.5 mm diameter hole drilled through the middle), with two subsequent compression cycles (1) and (2). The resulting compression moduli are shown in FIG. 27(e). The corresponding DIC images of the non-dimensional engineering strain along compression direction are shown in FIG. 27(f). The hole serves to show a localized strain stiffening in response to a defect.

[0189] Without wishing to be bound to a certain theory, it is assumed that the strain stiffening effect occurs due to the superposition of stress responses at two different length scales, as is illustrated in FIG. 28 (cf. FIG. 28(a,b,c)). During compression, the local stress forces are large enough to trigger densification of the micropores in the walls of the macropores (cf. FIG. 28(d,e)). This change is plastic, and it increases the connectivity of the macropore walls, which affects the global compression response. Since the densification of the microstructure occurs locally it is anisotropic and only affects the modulus in the compression direction.

[0190] FIG. 29 (a, b) shows a stress-strain measurement curve on a metal foam sample (low-carbon steel processed at 850 C. with a relative density of 0.22), over the complete range of elasticity, yielding and densification, with repeated release of the sample. FIG. 29(c) shows the elastic modulus E values obtained from the measurements as a function of the applied strain; and FIG. 29(d) shows the obtained elastic modulus values in the elastic range, as a function of the solid ratio of the sample. As one can see, E reaches a maximum E.sub.true when E.sub.apparent reaches the change from the elastic to the yielding regime. During the first compression, the stress-strain curve seemingly comprises the standard elastic phase, yielding and then densification found in foamed materials. However, the elastic phase is only pseudo elastic. Cyclic compression shows that the elastic modulus increases logarithmically with compression force until it reaches the true modulus E.sub.true seen in the second compression. This effect occurs without significant global densification and is rather an effect of microstructural improvement.

[0191] It has been known for nanoporous gold structures obtained by dealloying Ag97Au3 to show on the nanoscale and large porosities a similar effect [40, 41]. However, the effect has so far neither been known to exist for microstructures in the length scale range of micrometers and with lower porosities, as they can be obtained with metal foams according to the invention, nor at such low strains (<5%).

Materials and Methods

[0192] Ink preparation: An aqueous suspension of 40-50 wt % metal oxide powder (Magnetite, Fe.sub.3O.sub.4, E8840, DOWA; Nickel (II) oxide green, 99%, abcr GmbH) and 3 l/g of Hexylamine (99%, ACROS Organics) was homogenized by ball-milling with alumina balls (1-5 mm diameter) in a THINKY mixer (ARE-250) at 2000 rpm for 6 min (in two 3 min increments to avoid overheating). The standard composition contained 50 wt % magnetite powder. The slurry was then foamed at 800 rpm with an electronic stirrer (Heidolph, RZR 2102) until the desired stiffness was achieved (5-15 min). The resulting foam was left to rest overnight in a sealed container before printing volumetrically with a modified fused-filament-fabrication (FFF) printer (Ultimaker 2+) using 20 mL syringes and conical nozzles (0.84 mm diameter). Print speed was set at 10 mm/s and the extrusion rate was adjusted manually (100-150%) to achieve maximum print quality. PMMA plates, coated with commercial skin cream (NIVEA) to prevent sample adhesion, were used as substrates. The final samples were either dried at room temperature overnight or in the oven at 60 C. for 1 h.

[0193] Reduction of printed structures: The samples were reduced in a quartz tube oven (Gero, SR-A 100-500/12) with 12 l/min forming gas flow (95% N.sub.2 and 5% H.sub.2, Pangas). Various heating protocols (Table 2) were used to investigate the effect of reduction and sintering conditions on the microstructure and properties of the resulting porous metals (cf. FIGS. 7, 8).

TABLE-US-00002 TABLE 2 Reduction and sintering parameters used to convert printed foam structures into porous metals Heating Rate Temp. Dwell Sintering Sintering [ C./min] [ C.] Time [h] Temp. [ C.] Time [h] R1 1 650 70 R2 3 690 40 850 2 R3 3 750 15 R4 3 750 30 R5 3 850 15 R6 3 850 30 R7 3 1000 30

[0194] Oxidation of printed structures: As-printed magnetite foams were oxidized into different oxide phases (cf. FIG. 14) by heating the specimens for 2 h at 650, 750 or 1000 C. (Naberthem, LE 6/11/P300). In addition to thermal oxidation, porous iron structures were also partially oxidized in a wet state by placing them in Dulbecco's Phosphate Buffered Saline (Sigma) at 37 C. for 1-90 days (cf. FIG. 16, Table 1).

[0195] Surface hydrophobization: Hierarchical porous structures were hydrophobized (cf. FIG. 21) by dip-coating the specimen in a suspension of 1 wt % hydrophobic fumed silica (Wacker, HDK H18) in hexane (>95%, Sigma) and then drying the coated sample at room temperature in vacuum.

[0196] Microstructural characterization: High-resolution SEM images of the foams before and after reduction were taken with a scanning electron microscope (LEO 1530 instrument, Zeiss GmbH, Germany).

[0197] Compositional characterization: X-ray diffraction spectra were collected with a powder diffractometer (PANalytical Empyrean) equipped with a Cu K X-ray tube (15 kV, 40 mA) and a monochromator. Samples were prepared by either crushing them into powder, if brittle, or compressing them into flat sheets, if tough. Particle induced X-Ray emission (PIXE) measurements were performed by Max Dobeli at the Laboratory of Ion Beam Physics, ETH Zrich. The carbon content of the iron samples was measured) with a carbon/sulfur analyzer (LECO CS230) using four 1 g samples per composition.

[0198] Mechanical characterization: Compressive strength and modulus were measured with a universal mechanical testing machine (AGS-X, Shimadzu) equipped with a 1000 N load cell. To ensure the opposing faces were parallel, the samples were ground into shape (Struers, LaboPol-25, 1000 grit SiC paper) using ethanol as a lubricant to prevent sample oxidation. Tests were performed by applying a constant displacement rate of 12 mm/min. The raw data were corrected to take into account the mechanical compliance of the testing machine. The machine compliance was measured with a 998 mm aluminum cube as a placeholder. The data was evaluated using Matlab. The reported compressive yield strength was determined with the 0.2% offset method.

[0199] Surface area characterization: N.sub.2 gas sorption measurements were performed with a pore size analyzer (Quantachrome Autosorb iQ) at 77 K. Prior to the measurement the samples were outgassed in vacuum at 80 C. for 24 h. The surface area was determined by the Brunauer-Emmett-Teller method.

[0200] Absorption experiments: The oil absorption capability of the hydrophobized porous iron structure (cf. FIG. 21) was evaluated by placing a water-floating sample in contact with hexane as an absorbent (>95%, Sigma). Hexane was died purple with 0.01 g/L of both Nile red (Merck) and methylene blue (Merck).

[0201] Conductivity and resistive heating experiments: The resistive heating capabilities of partially oxidized porous iron structures (cf. FIG. 21) were evaluated by pressing the samples between copper tape and applying electrical current with a Laboratory Power Supply (EA-PS-3032-05-B). The temperature of the sample was measured with infrared imaging (IRCAM, Millenium 327k S/M). R-(+)-Limonene (Merck) was used as vaporizing agent.

[0202] The present invention is not to be limited in scope by the specific embodiments described herein. Indeed, various modifications of the present invention, in addition to those described herein, will be apparent to those skilled in the art from the foregoing description and accompanying drawings. Thus, such modifications are intended to fall within the scope of the appended claims. Additionally, various references are cited throughout the specification, the disclosures of which are each incorporated herein by reference in their entirety.

LIST OF REFERENCE NUMERALS

[0203] 1 foamed aqueous suspension, wet foam ink [0204] 11 precursor particle [0205] 12 surfactant molecule [0206] 13 water [0207] 14 gas bubble, foam cell [0208] 21 printing nozzle [0209] 22 tubular oven [0210] 23, 23 electrodes [0211] 3a three-dimensional structure [0212] 3b green body [0213] 3c metal foam element [0214] 31 foam filament [0215] 32 first hierarchy pore, macropore [0216] 33 second hierarchy pore, mesopore [0217] 34 third hierarchy pore, micropore [0218] 41 oil droplet

REFERENCES

[0219] [1] L.-P. Lefebvre, J. Banhart, D. C. Dunand, Adv. Eng. Mater. 2008, 10, 775. [0220] [2] J. Erlebacher, M. J. Aziz, A. Karma, N. Dimitrov, K. Sieradzki, Nature 2001, 410, 450. [0221] [3] M. E. Cox, D. C. Dunand, Mat Sci Eng a-Struct 2011, 528, 2401. [0222] [4] E. Linul, L. Marsavina, P. A. Linul, J. Kovacik, Compos Struct 2019, 209, 490. [0223] [5] M. H. Sun, S. Z. Huang, L. H. Chen, Y. Li, X. Y. Yang, Z. Y. Yuan, B. L. Su, Chem Soc Rev 2016, 45, 3479. [0224] [6] N. Kleger, M. Cihova, K. Masania, A. R. Studart, J. F. Loffler, Adv Mater 2019, 31. [0225] [7] P. Sharma, P. M. Pandey, Mater Design 2018, 160, 442. [0226] [8] R. Orinakova, R. Gorejova, Z. O. Kralova, L. Haverova, A. Orinak, I. Maskal'ova, M. Kupkova, M. Dzupon, M. Balaz, M. Hrubovcakova, T. Sopcak, A. Zubrik, M. Orinak, Appl Surf Sci 2020, 505. [0227] [9] S. Ray, U. Thormann, M. Eichelroth, M. Budak, C. Biehl, M. Rupp, U. Sommer, T. El Khassawna, F. I. Alagboso, M. Kampschulte, M. Rohnke, A. Henss, K. Peppler, V. Linke, P. Quadbeck, A. Voigt, F. Stenger, D. Karl, R. Schnettler, C. Heiss, K. S. Lips, V. Alt, Biomaterials 2018, 157, 1. [0228] [10] Y. Conde, J. F. Despois, R. Goodall, A. Marmottant, L. Salvo, C. San Marchi, A. Mortensen, Adv Eng Mater 2006, 8, 795. [0229] [11] A. R. Studart, A. Nelson, B. Iwanovsky, M. Kotyrba, A. A. Kundig, F. H. Dalla Torre, U. T. Gonzenbach, L. J. Gauckler, J. F. Loffler, Journal of Materials Chemistry 2012, 22, 820. [0230] [12] Y. J. Quan, F. M. Zhang, H. Rebl, B. Nebe, O. Kessler, E. Burkel, Mat Sci Eng a-Struct 2013, 565, 118. [0231] [13] A. E. Jakus, N. R. Geisendorfer, P. L. Lewis, R. N. Shah, Acta Biomater 2018, 72, 94. [0232] [14] A. R. Studart, Chemical Society Reviews 2016, 45, 359. [0233] [15] A. R. Studart, R. M. Erb, R. Libanori, in Hybrid and Hierarchical Composite Materials, DOI: 10.1007/978-3-319-12868-9_8 (Eds: C.-S. Kim, C. Randow, T. Sano), Springer International Publishing 2015, Ch. 8, p. 287. [0234] [16] B. Gorny, T. Niendorf, J. Lackmann, M. Thoene, T. Troester, H. J. Maier, Materials Science and Engineering: A 2011, 528, 7962. [0235] [17] L. S. Bertol, W. K. Jnior, F. P. d. Silva, C. Aumund-Kopp, Materials & Design 2010, 31, 3982. [0236] [18] L. Yang, O. Harrysson, D. Cormier, H. West, H. Gong, B. Stucker, JOM 2015, 67, 608. [0237] [19] L. E. Murr, S. M. Gaytan, F. Medina, H. Lopez, E. Martinez, B. I. Machado, D. H. Hernandez, L. Martinez, M. I. Lopez, R. B. Wicker, J. Bracke, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 2010, 368, 1999. [0238] [20] P. Heinl, L. Mller, C. Krner, R. F. Singer, F. A. Mller, Acta Biomater. 2008, 4, 1536. [0239] [21] C. Xu, Q. H. Wu, G. L'Esperance, L. L. Lebel, D. Therriault, Mater Design 2018, 160, 262. [0240] [22] M. Calvo, A. E. Jakus, R. N. Shah, R. Spolenak, D. C. Dunand, Adv Eng Mater 2018, 20. [0241] [23] A. E. Jakus, S. L. Taylor, N. R. Geisendorfer, D. C. Dunand, R. N. Shah, Adv Funct Mater 2015, 25, 6985. [0242] [24] B. Y. Ahn, D. Shoji, C. J. Hansen, E. Hong, D. C. Dunand, J. A. Lewis, Adv. Mater. 2010, 22, 2251. [0243] [25] S. L. Taylor, A. E. Jakus, R. N. Shah, D. C. Dunand, Adv. Eng. Mater. 2017, 19, 1600365. [0244] [26] S. L. Taylor, A. E. Jakus, R. N. Shah, D. C. Dunand, Adv Eng Mater 2017, 19. [0245] [27] C. Minas, D. Carnelli, E. Tervoort, A. R. Studart, Adv Mater 2016, 28, 9993. [0246] [28] C. Minas, J. Carpenter, J. Freitag, G. Landrou, E. Tervoort, G. Habert, A. R. Studart, Acs Sustain Chem Eng 2019, 7, 15597. [0247] [29] D. G. Moore, L. Barbera, K. Masania, A. R. Studart, Nature Materials 2020, 19, 212. [0248] [30] L. Alison, S. Menasce, F. Bouville, E. Tervoort, I. Mattich, A. Ofner, A. R. Studart, Sci Rep-Uk 2019, 9. [0249] [31] M. R. Sommer, M. Schaffner, D. Carnelli, A. R. Studart, Acs Applied Materials & Interfaces 2016, 8, 34677. [0250] [32] C. Kenel, N. R. Geisendorfer, R. N. Shah, D. C. Dunand, Addit Manuf 2021, 37. [0251] [33] R. Murakami, S. Kobayashi, M. Okazaki, A. Bismarck, M. Yamamoto, Front Chem 2018, 6. [0252] [34] U. T. Gonzenbach, A. R. Studart, E. Tervoort, L. J. Gauckler, Angewandte Chemie-International Edition 2006, 45, 3526. [0253] [35] H. Ovaici, M. R. Mackley, G. H. Mckinley, S. J. Crook, J Rheol 1998, 42, 125. [0254] [36] D. Kokkinis, M. Schaffner, A. R. Studart, Nat Commun 2015, 6. [0255] [37] M. Bowker, E. K. Gibson, I. P. Silverwood, C. Brookes, Faraday Discuss 2016, 188, 387. [0256] [38] X. H. Zou, T. H. Chen, H. B. Liu, P. Zhang, D. Chen, C. Z. Zhu, Fuel 2016, 177, 180. [0257] [39] S. A. Halawy, S. S. Al-Shihry, M. A. Mohamed, Catal Lett 1997, 48, 247. [0258] [40] S. Shi, Y. Li, B. N. Ngo-Dinh, J. Markmann, J. Weissmuller, Science 371, 1026-1033 (2021). [0259] [41] B. Zandersons, L. Lhrs, Y. Li, J. Weissmller, Acta Mater 2021, 215, 116979.