FIB-SEM on MgO(100) single crystals promoted with NaNO3

https://doi.org/10.1039/D2TA02897B

“To obtain a full 3D representation of the formed MgCO₃ crystal, we performed additional focused ion beam (FIB) SEM tomography measurement of MgCO₃ grown on MgO(100) single crystals promoted with NaNO₃ after 5 hours of CO₂ exposure (Fig. 6). The shape of the crystals/particles formed on the laser ablated surface, Fig. 5, point to a nucleation and crystal growth type process from a supersaturated solution of [Mg²⁺…CO₃²⁻] in agreement with literature.^10,14–16 While MgCO₃ forms through a nucleation and growth type process in both the laser ablated MgO sample, Fig. 5, and MgO(100), Fig. 6, there are still some variations in the growth habit between these two systems. For example, each MgCO₃ particle grown on the ablated MgO surface is composed of a comparatively large number of small crystallites and there is a more extensive out-of-plane 3D growth (crystal thickness of up to 3 μm, see below in Fig. 7) of the MgCO₃ crystals on the ablated MgO surface as compared to the MgCO₃ crystals grown on the surface of MgO(100) single crystals (crystal thickness in the range 0.2–2 μm), in agreement with our previous work.¹⁵ The mean particle size (in-plane) after 5 hours of exposure to CO₂ is 9 ± 1 μm on the ablated surface in comparison to a size of 10–50 μm on the MgO(100) single crystal surface. The morphology of the MgCO₃ formed on the ablated MgO surface can be described as “radiating dendrites” in contrast to the “sector plate” morphology on MgO(100). To rationalize the differences in morphology of the MgCO₃ formed on these two model systems it is worth looking at the growth habits of ice crystals and ice analog crystals. Their growth habits have been mapped with respect to the level of supersaturation. For example, the crystal growth of Na₂SiF₆ from an aqueous solution e.g. can be induced by evaporation of the solution, whereby the supersaturation level is controlled by the rate of evaporation.²⁵ Under such conditions, sodium fluorosilicate crystals exhibit a “sector plate” morphology for lower levels of supersaturation while when increasing the degree of supersaturation “radiating dendrite” morphologies are observed. Assuming that there is a similar link between the degree of supersaturation and the growth habit for MgCO₃, we can infer that the laser ablated MgO surface provides an environment with a higher supersaturation level (radiating dendrite morphology with a larger number of small crystals) compared to a flat MgO(100) surface (sector plate morphology with a few large crystallites). The higher level of supersaturation of [Mg²⁺…CO₃²⁻] ionic pairs is likely a result of the more defective nature of the laser ablated surface promoting in turn the dissolution of MgO, yielding [Mg²⁺…O²⁻] ionic pairs which are the precursors to carbonate ionic pairs, as dissolution preferentially occurs on surface defects.³²”

“Fig. 6 FIB-SEM tomography of a MgCO₃ crystal grown on the surface of MgO(100) promoted with NaNO₃ after 5 hours of exposure to CO₂i.e. NaNO₃–MgO(100)_5hours-CO2. Prior to the tomographic measurement the sample was rinsed in water. (a) SEM Image of the MgCO₃ crystal grown on a MgO(100) single crystal (the grey box shows the area that was used for the tomographic measurement; resolution in x, y, z is, respectively, 10 nm, 25 nm, and =10 nm). (b and c) Surface contours of MgCO₃ and MgO are determined by segmenting MgCO₃ and MgO through a combination of thresholding by grey-level (using the backscattered electron, BSE, signal) and a directional search for grey-level edges. (d) Segmentation of MgCO₃ and MgO in a cross-section (window size [x,z] = 22.7 × 2.7 μm) along the y-direction. Each cross section is taken at a y-spacing of 25 nm and is shown at selected positions, i.e. y = 0, 1.25, 2.5, 3.75, 5, and 6.25 μm.”

“Fig. 7 Quantification of the volume of the MgCO₃ crystals by 3D profilometry measurements after removal of NaNO₃ by rinsing with H₂O. The volume is calculated by the sum of “voxels” of MgCO₃, i.e. the voxels that are at a height z > 0 μm with respect to the flat MgO surface at the bottom of the groove (z = 0 μm, roughness subtracted) multiplied by the dimension of the pixel in the xy direction (0.69 × 0.69 μm²). As the area at the bottom of a groove is smaller in deeper grooves, the MgCO₃ volume is normalized by the area of the bottom part of the groove. Note that the optical profile measurements rely on the reflection of light at the surface. Hence, only MgCO₃ grown at the bottom of the groove can be quantified, as the steep angle of the groove wall with respect to the incoming light minimizes reflection, which hinders quantification of the volume of MgCO₃ grown on the walls of the groove. (a and b) MgCO₃ formation after 5 hours of exposure to CO₂ in grooves with a depth of 11 μm and 44 μm, respectively. (c) Total volume of MgCO₃ crystals formed at the bottom of the grooves normalized by the area of the bottom of the groove as a function of the depth of the groove (d_groove = 6, 11, 11, 22, 44, 84, 122, 155, 180 μm) and carbonation time (0.5, 1, 2, 5 hours).”