*2.3. Reaction Time*

The impact of three reaction times on HC formation was tested (t1 = 1 h, t2 = 3 h and t3 = 6 h), the XRD and FTIR-ATR results of which are shown in Figure 8a and Figure 8b, respectively.

**Figure 8.** (**a**) XRD results of the time series t1, t2 and t3 between 5◦ 2*θ* and 90◦ 2*θ*. Each scan is y-shifted by 2500 counts. The inset depicts the primary LDH peaks and change in position. HC = hydrocalumite, Kat = katoite, P = portlandite and C = calcite. (**b**) FTIR-ATR scans of the time series t1, t2 and t3 between 4000 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup>1. The three insets show the regions: (**1**) 3700 cm<sup>−</sup><sup>1</sup> – 3600 cm<sup>−</sup>1, (**2**) 1500 cm<sup>−</sup><sup>1</sup> – 1250 cm<sup>−</sup>1, (**3**) 1000 cm<sup>−</sup><sup>1</sup> – 550 cm<sup>−</sup><sup>1</sup> in detail. Dashed and solid grey lines indicate the maxima of vibrations. t1 = 1 h, t2 = 3 h and t3 = 6 h.

XRD showed a large similarity between the phases formed; however, the LDH phase formed in t1 was shifted to the right (see inset of Figure 5a, more closely resembling the primary reflection position of the phase formed in T2. The best-fit LDH crystal structure for all three, t1, t2 and t3, was again anorthic [Ca4Al2(OH)12][(CO3) · 5 H2O]. A small additional phase with a similar crystal structure to chloride-intercalated HC was detected in t3 ([Ca4Al2(OH)12][(CO3)0.5Cl4 · 8H2O]). Overall, Rietveld refinement showed that an increase in reaction time led to an increase in LDH phase and decrease in katoite and portlandite present. Calcite content also decreased slightly with an increase in reaction time. Table 2 depicts the Rietveld refinement results obtained for the different reaction times.

**Table 2.** Rietveld refinement of the time series of LDHs t1 = 1 h, t2 = 3 h and t3 = 6 h. HC indicates the phase [Ca4Al2(OH)12][(CO3) · 5 H2O], HC2 indicates the phase [Ca4Al2(OH)12][(CO3)0.5Cl4 · 8H2O], Kat is katoite, P is portlandite and C is calcite. All values are given in percentages of the crystalline phases.


FTIR scans of the three LDHs were similar; most notably showing an increase in vibration in the doublet at 1414 cm<sup>−</sup>1/1361 cm<sup>−</sup>1, corresponding to a decreased amount of calcite and possible increased intercalation of carbonate into the interlayer. Most vibrations increased in strength with an increase in reaction time, except the vibration at 3243 cm<sup>−</sup>1, which showed a stronger vibration in t1. Further, the vibration at 3641 cm<sup>−</sup><sup>1</sup> (linked to unreacted Ca(OH)2) was notably stronger in t2, even though this LDH contained less Ca(OH)2 according to Rietveld refinement.

Morphologically, the three materials were very similar; the only easily noticeable differences being the amount of particulate matter clinging to the other phases, improved crystallisation of the platelets and a weak correlation between the amount of katoite and calcite observed, and Rietveld refinement results (Figure 9).

**Figure 9.** SEM micrographs of the time series of HC LDHs t1 = 1 h, t2 = 3 h and t3 = 6 h at 1 keV and 2k magnification. The scale bar is indicated under the label.

Recording the pH for the duration of the 6 h t3 experiment showed that the pH approximately levels out after 3 h (Figure 10). Small differenced between the pH measurements were observed between t1, t2 and t3 in their respective time-frames.

**Figure 10.** pHs of the time series of LDHs t1 = 1 h, t2 = 3 h and t3 = 6 h. The pH was adjusted to 25 ◦C to facilitate comparison.

#### *2.4. Molar Calcium-to-Aluminium Ratio*

The reaction of Ca(OH)2 and Al(OH)3 to form HC stands in competition with the formation of katoite (Ca3Al2(OH)12) [26,27]. Three molar calcium-to-aluminium ratios were, thus, tested to determine the effect of this ratio on the formation of HC. MR1 (3:2) should favour katoite formation (based on stoichiometry), while MR2 (4:2) should favour HC formation and MR3 (6:2) would provide an excess of Ca(OH)2 to the reaction. Out of all results, the result of this series was most surprising, inverting expectations. XRD (shown in Figure 11) showed that, once again, similar phases formed, just with different ratios and that the [Ca4Al2(OH)12][(CO3) · 5 H2O] phase again provided the best fit for the crystal structure of the LDH formed.

**Figure 11.** (**a**) XRD results of the molar calcium-to-aluminium ratio series MR1, MR2 and MR3 between 5◦ 2*θ* and 90◦ 2*θ*. Each scan is y-shifted by 2500 counts. The inset depicts the primary LDH peaks and change in position. HC = hydrocalumite, Kat = katoite, P = portlandite and C = calcite. (**b**) FTIR-ATR scans of the molar calcium-to-aluminium ratio series MR1, MR2 and MR3 between 4000 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup>1. The three insets show the regions: (**1**) 3700 cm<sup>−</sup><sup>1</sup> – 3600 cm<sup>−</sup>1, (**2**) 1500 cm<sup>−</sup><sup>1</sup> – 1250 cm<sup>−</sup>1, (**3**) 1000 cm<sup>−</sup><sup>1</sup> – 550 cm<sup>−</sup><sup>1</sup> in detail. Dashed and solid grey lines indicate the maxima of vibrations. MR1 = 3:2, MR2 = 4:2 and MR3 = 6:2.

The primary MR3 LDH peak was slightly shifted to the right of MR1 and MR2 (inset of Figure 11a). No significant amounts of other LDH phases were formed in this series. However, Rietveld refinement revealed that MR1 led to the greatest purity HC phase (>80%), while MR3 resulted in a purity of less than 50%. The Rietveld refinement results of this series are shown in Table 3.

**Table 3.** Rietveld refinement of the molar calcium-to-aluminium series of LDHs MR1 = 3:2, MR2 = 4:2 and MR3 = 6:2. HC indicates the phase [Ca4Al2(OH)12][(CO3) · 5 H2O], Kat is katoite, P is portlandite and C is calcite. All values are given in percentages of the crystalline phases.


FTIR-ATR results (Figure 11b showed good correlation between the Rietveld results and the vibrational intensities of each sample. The vibration at 3641 cm<sup>−</sup>1, linked to unreacted Ca(OH)2, showed excellent correlation between the Rietveld and FTIR results, increasing in intensity with an increase in left-over portlandite. The twin-peak at 1414 cm<sup>−</sup>1/1361 cm<sup>−</sup><sup>1</sup> correlated well with the shift in primary LDH reflection position and possible carbonate content. In the region between 1000 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup><sup>1</sup> the vibration strength decreased with an increase in molar ratio.

Morphologically, clear differences were visible between the amount of particulate matter present (Ca(OH)2 or amorphous Al(OH)3) in each sample (Figure 12). MR3 contained the largest amount of particulate matter, correlating well with the Rietveld refinement results. MR3 also contained visibly more katoite and calcite, as expected. The platelets showed some differences too. MR1 produced smaller, rugged platelets with many broken edges and the agglomeration of different platelet sizes. MR2 produced much smoother, better defined and larger platelets than MR1 and MR3. The platelets of MR3 were less rugged than those of MR1, but due to the coverage in particulate matter, proper description of their definition was challenging.

**Figure 12.** SEM micrographs of the series of molar calcium-to-aluminium ratio MR1 = 3:2, MR2 = 4:2 and MR3 = 6:2 at 1 keV and 2k magnification. The scale bar is indicated under the label.

The pHs of MR1 and MR3 were very similar and dropped slightly with time in comparison with MR2. MR3 started with a slightly elevated pH (as expected due to the large amount of Ca(OH)2 present). Conversely, MR1 had a lower starting pH, as expected, due to the lower amount of Ca(OH)2 present and MR2 lay between the two at the start—as portrayed in Figure 13. With progression in time, the pHs of MR1 and MR3 fell slightly below that of MR2.

**Figure 13.** pHs of the molar calcium-to-aluminium series of LDHs MR1 = 3:2, MR2 = 4:2 and MR3 = 6:2. The pH was adjusted to 25 ◦C to facilitate comparison.

#### *2.5. Chemical Morphology/Crystallinity of the Al(OH)3*

Three different Al(OH)3 sources were tested in this work. The first (A1), from Sigma Aldrich (SA) was used as the standard source in all experiments. This Al source proved to be amorphous by XRD and have a surface area of 59.35 m<sup>2</sup> g<sup>−</sup>1. The second Al source (A2) from ACE Chemicals (ACE) proved to be minimally crystalline boehmite with a surface area of 69.12 m<sup>2</sup> g<sup>−</sup><sup>1</sup> (due to the low crystallinity it is possible that some of the phase consisted of other amorphous crystal structures or polymorphs). The third Al source (A3) from Merck (M) proved to be highly crystalline gibbsite with a surface area of 0.75 m<sup>2</sup> g<sup>−</sup>1. Figure 14 shows the XRD results for these three sources and isotherms obtained for the determination of the Brunauer–Emmett–Teller (BET) surface area.

**Figure 14.** (**a**) XRD results of the Al sources used for the synthesis of A1, A2 and A3. SA = amorphous, ACE = boehmite and M = gibbsite—Sigma Aldrich (SA), ACE Chemicals (ACE) and Merck (M). Each scan is y-shifted by 2500 counts. (**b**) Isotherms obtained for each Al source. Sigma Aldrich (SA), ACE Chemicals (ACE) and Merck (M).

The Al sources used thus had very different crystallinities and crystal structures. SEM further showed that the morphologies of the three Al sources were very different (see Figure 15).

**Figure 15.** SEM micrographs of the three different Al(OH)3 sources used from Sigma Aldrich (SA), ACE Chemicals (ACE) and Merck (M) taken at 1 keV and 10k (top) and 500 times (bottom) magnification. The scale bar is indicated on each micrograph.

The SA and ACE reagents had similarly sized particles but different "macro" morphologies. Where SA consisted of a carpet-like covering of small dot-like particles (agglomerated into random shapes), ACE consisted of the same size particles but agglomerated into a myriad of variably sized balls. These balls were damaged in the sample preparation process, breaking up/squashing and revealing some of the internal structure. The M source, on the other hand, consisted of large, thick chunks of gibbsite.

These three materials caused the formation of three very different reaction outcomes. The outcome of A1 (the standard sample S3) has already been discussed. In comparison, A2 consisted of two HC phases, being mainly the typical one with a crystal structure similar to [Ca4Al2(OH)12][(CO3) · 5 H2O] but also a small fraction of the HC phase with a crystal structure similar to [Ca4Al2(OH)12][(CO3)0.5Cl · 4.8H2O], as already present in t3. These two phases are clearly distinguishable on the XRD inset in Figure 16a depicting the primary LDH reflections.

**Figure 16.** (**a**) XRD results of the chemical morphology/crystallinity of Al(OH)3 series A1, A2 and A3 between 5◦ 2*θ* and 90◦ 2*θ*. Each scan is y-shifted by 2500 counts. The inset depicts the primary LDH peaks and change in position. HC = hydrocalumite, HC2 = hydrocalumite phase 2, Kat = katoite, P = portlandite, C = calcite and G = gibbsite. (**b**) FTIR-ATR scans of the chemical morphology/crystallinity of Al(OH)3 A1, A2 and A3 between 4000 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup>1. The three insets show the regions: (**1**) 3700 cm<sup>−</sup><sup>1</sup> – 3600 cm<sup>−</sup>1, (**2**) 1500 cm<sup>−</sup><sup>1</sup> – 1250 cm<sup>−</sup>1, (**3**) 1200 cm<sup>−</sup><sup>1</sup> – 550 cm<sup>−</sup><sup>1</sup> in detail. Dashed and solid grey lines indicate the maxima of vibrations. A1 = Al(OH)3-SA, A2 = Al(OH)3-ACE and A3 = Al(OH)3-SA.

A3 consisted of only one phase of LDH but at a lower overall purity—with large amounts of Ca(OH)2 remaining unreacted as shown through Rietveld refinement in Table 4. A2 produced the most katoite and A3 contained no calcite, but large amounts of unreacted gibbsite. Both A2 and A3 showed slight right-shifts in their primary LDH reflection in comparison to A1.

**Table 4.** Rietveld refinement of the morphology/crystallinity series of LDHs A1 = SA, A2 = ACE and A3 = M. HC indicates the phase [Ca4Al2(OH)12][(CO3) · 5 H2O], HC2 is [Ca4Al2(OH)12][(CO3)0.5Cl4 · 8H2O], Kat is katoite, P is portlandite, C is calcite and G is gibbsite. All values are given in percentages of the crystalline phases.


There were also differences between the FTIR-ATR scans of A1, A2 and A3 depicted in Figure 16b. While the scans for A1 and A2 were quite similar, A3 deviated from almost all major vibrations with either a decreased or increased vibration. The large amount of unreacted portlandite in A3 was clearly visible at 3641 cm<sup>−</sup>1. Most other vibrations, however, were weaker in A3—most notably those indicating the OH str. vibration of water bonded to interlayer carbonate (3010 cm<sup>−</sup>1) and the combination band at 2828 cm<sup>−</sup><sup>1</sup> as well as the twin peak at 1414 cm<sup>−</sup>1/1361 cm<sup>−</sup><sup>1</sup> associated with both, interlayer carbonate and interlayer hydroxide anions. The scan for A3 gained an additional band at 1014 cm<sup>−</sup><sup>1</sup> which correlates to the strongest vibration of the Al(OH)3-M source itself. The remaining vibrations between 1000 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup><sup>1</sup> were less strongly defined than those of A1 and A2. There also existed some stark differences between the FTIR scans of the three Al(OH)3 sources, which are also depicted in Figure 16b. Al(OH)3-M showed defined vibrations in the 3700 cm<sup>−</sup><sup>1</sup> to 3300 cm<sup>−</sup><sup>1</sup> region and between 1200 cm<sup>−</sup><sup>1</sup> and 550 cm<sup>−</sup>1, while Al(OH)3-SA and Al(OH)3-ACE were very similar and had much less strongly defined features. Both consisted of a broad band in the 3700 cm<sup>−</sup><sup>1</sup> to 3300 cm<sup>−</sup><sup>1</sup> region and immolated well the 1000 cm<sup>−</sup><sup>1</sup> to 550 cm<sup>−</sup><sup>1</sup> region. Both materials, however, had a doublet feature blue-shifted from the 1414 cm<sup>−</sup>1/1361 cm<sup>−</sup><sup>1</sup> doublet. This feature has been linked to adsorbed carbonate species [36–38] and will be discussed in-depth later in the text.

Morphological examination of these three materials showed that A1 and A2 formed similar structures but that A3 formed much larger HC platelets (Figure 17). A far greater quantity of particulate matter was also present in A3, as expected from the sample consisting of almost 50% of portlandite and gibbsite. No large chunks of gibbsite (as initially present from the reactant) were visible, though, indicating that these chunks of gibbsite must have broken up during synthesis.

**Figure 17.** SEM micrographs of the chemical morphology/crystallinity of Al(OH)3 series A1 = SA, A2 = ACE and A3 = M taken at 1 keV and 2k magnification. The scale bar is indicated below the label.

Even though the phase compositions were very different for A1, A2 and A3, the synthesis pHs observed were almost identical throughout the reaction, as shown in Figure 18—indicating that Al(OH)3 crystallinity and surface area, and the subsequent dissoluted Al-species, do not contribute to the mixture pH significantly.

**Figure 18.** pHs of the morphology/crystallinity of Al(OH)3 series of LDHs A1 = SA, A2 = ACE and A3 = M. The pH was adjusted to 25 ◦C to facilitate comparison.
