Exploring the Wet Mechanochemical Synthesis of Mg-Al, Ca-Al, Zn-Al and Cu-Al Layered Double Hydroxides from Oxides, Hydroxides and Basic Carbonates

The synthesis of Mg-Al, Ca-Al, Zn-Al and Cu-Al layered double hydroxides (LDHs) was investigated with a one-step wet mechanochemical route. The research aims to expand on the mechanochemical synthesis of LDH using a mill designed for wet grinding application. A 10% slurry of solids was added to a Netzsch LME 1 horizontal bead mill and milled for 1 h at 2000 rpm. Milling conditions were selected according to machine limitations and as an initial exploratory starting point. Precursor materials selected consisted of a mixture of oxides, hydroxides and basic carbonates. Samples obtained were divided such that half was filtered and dried at 60 °C for 12 h. The remaining half of the samples were further subjected to ageing at 80 °C for 24 h as a possible second step to the synthesis procedure. Synthesis conditions, such as selected precursor materials and the MII:MIII ratio, were adapted from existing mechanochemical methods. LDH synthesis prior to ageing was successful with precursor materials observably present within each sample. No Cu-Al LDH was clearly identifiable. Ageing of samples resulted in an increase in the conversion of raw materials to LDH product. The research offers a promising ‘green’ method for LDH synthesis without the production of environmentally harmful salt effluent. The synthesis technique warrants further exploration with potential for future commercial up-scaling.


Introduction
Layered double hydroxides (LDHs) are clay-like minerals commonly referred to as anionic clays with a wide range of physical and chemical properties. They are represented by the general formula [M II 1−x M III x (OH) 2 ][X q− x/q ·H 2 O] in which M II and M III represent the selected divalent and trivalent metal elements and [X q− x/q ·H 2 O] denotes the interlayer composition. LDHs often find application in pharmaceuticals, as polymer additives, as additives in cosmetics, and in catalysis. This is due to having variable layer charge density, reactive interlayer space, ion exchange capabilities, a wide range of chemical compositions and rheological properties [1]. LDH materials can be synthesised using various different techniques of which the most common are co-precipitation, reconstruction, hydrothermal methods and urea decomposition-homogenous precipitation. The primary principle associated with these methods include the precipitation of various types of metal ions which makes large scale production difficult. Challenges associated with these methods include differing precipitation rates of metal ions, need for inert environments, production of environmentally harmful waste and high production costs [2]. Novel, 'green' synthesis techniques are therefore often sought 525 L·h −1 . A water slurry consisting of 10% solids (reactants) was added to the milling chamber and milled for 1 h at 2000 rpm. Samples obtained were divided such that half was filtered and dried at 60 • C for 12 h. The remaining half of the sample was subjected to ageing at 80 • C for 24 h. The principle of the mill is similar to that of agitator bead mills in which the grinding media is accelerated with the use of an agitator shaft. The energy supplied to the media is then transferred to the solids via collisions and de-acceleration. The vessel is placed in a horizontal position to allow for even grinding activity and activation. Figure 1 depicts a technical schematic of the Netzsch LME 1 horizontal bead mill. The product inlet and outlet to the grinding chamber were sealed to allow for batch milling. Raw materials and grinding media were added to the vessel through the 'bead filling connection'. At the end of every experimental run, the 'tank floor' or front cap of the milling chamber was removed and the sample and beads collected. The grinding media and mill were washed in preparation for the next experimental run. The pump set-up provided by Netzsch was not used.

Ageing Process
The ageing step was conducted by making use of a bench-top Lasec digital hotplate stirrer. Samples obtained from the milling chamber were divided such that half was immediately filtered and dried and the other half subjected to an ageing step. Samples were placed in a glass beaker and agitated, at 400 rpm, for 24 h. Sample temperature was elevated and kept constant at 80 • C. A thin plastic film was placed over the beaker to prevent excessive moisture loss. Experiments were performed without the use of an inert gas under air atmosphere. All samples were filtered and dried at 60 • C for 12 h.

Ca-Al LDH
Commercial grade Ca(OH) 2 (LimeCo. Minerals, JHB, GP, ZA) was first calcined at a temperature of 900 • C for 1 h to remove any hydroxide and carbonate impurities, to form CaO. This was then reacted with 100 mL water for 15 min to form Ca(OH) 2 . This step eliminated the possibility of vapor formation within the milling chamber due to the extremely exothermic CaO hydration reaction. The Ca(OH) 2 and Al(OH) 3

Scanning Electron Microscopy (SEM)
SEM imaging was used to observe the morphology of the prepared samples. A Zeiss Gemini 1 cross beam 540 FEG SEM (Oberkochen, Germany). Powdered samples were placed secured onto an aluminium sample holder and graphite coated 5 times with a Polaron Equipment E5400 SEM auto-coating sputter system (Quorum, East Sussex, UK).

X-ray Diffraction Analysis (XRD)
Reaction products of powdered samples were identified using a PANalytical X'Pert Pro powder diffractometer in θ-θ configuration fitted with an X'Celerator detector and variable divergence-and fixed receiving slits (Malvern Panalytical, Malvern, UK ). The system made use of Fe filtered Co-Kα (λ = 1.789Å) source. Samples were prepared using the standardised PANalytical backloading system, providing a random distribution of particles. Samples were scanned from 5 • to 90 • with a step size of 0.008 • . Sample mineralogy was determined using the ICSD database in correlation with X'Pert Highscore plus software.

Fourier Transform Infrared Spectroscopy (FT-IR)
FT-IR spectra for the samples were obtained using a PerkinElmer 100 Spectrophotometer (Massachusetts, USA) over a range of 550-4000 cm −1 and represent an average of 32 scans, at a resolution of 2 cm −1 . 2.7.5. X-ray Fluorescence (XRF) XRF was used for elemental analysis of the samples. Samples were dried at 100 • C and roasted at 1000 • C to determine mass loss on ignition. In addition, 1 g of the sample was mixed with 6 g Lithumtetraborate flux and fused at 1050 • C to form a stable fused glass bead. Analysis was conducted using a Thermo Fisher ARL Perform 'X Sequential instrument (Massachusetts, USA). Samples were characterised using UNIQUANT software.

Particle Size Analysis
The particle size of the raw material mixtures as well as that of the sample obtained is depicted in Tables 1 and 2, respectively. It was noted that overall particle size reduction occurred for most samples, with the exception of S3, with an increase in the grinding time as expected. This could possibly be attributed to the formation and agglomeration of Ca-Al LDH present within the sample. Raw material mixtures exhibited large D 90 measurements that could be attributed to immediate reaction with water, as well as agglomeration.

X-ray Fluorescence
The elemental composition and metal ratios of the samples were obtained via XRF analysis and listed in Table 3. All samples were found to have a small amount of zirconium, yttrium, and iron contamination from the milling media and the milling chamber. Samples S1 and S2 contained SiO 2 introduced by the selected commercial grade MgO reagent. XRF analysis was conducted to ensure that the correct metal ratios were applied to the raw materials added to the system and are therefore not an indication of the composition of the LDH phases present within each sample. They are an indication of the metal ratios within the overall sample obtained. Calculated metal ratios were observed to correlate with those adapted from literature.

X-ray Diffraction Analysis
Samples analysed were observed to have minor unindexed peaks present. This could be attributed to unidentified phases present or impurities due to mill degradation. Future studies should be conducted in attempt to adequately investigate all phases present within samples obtained.
Mg-Al LDH. Figures 2 and 3 show the XRD spectra obtained for samples S1 and S2. The reaction had not yet reached completion at the selected synthesis conditions prior to ageing. The sample synthesised with a 2:1 metal ratio (S1) exhibited no clear LDH peaks with no crystalline LDH phase detected within the sample. Ageing of the sample resulted in a clear LDH pattern with primary and secondary peaks located at 2θ values of 13.49 • and 27.22 • , respectively. This correlated with a basal spacing of 0.759 nm. Comparatively, S2 exhibited primary and secondary LDH peaks prior to ageing at 2θ values of 13.37 • and 26.99 • , respectively. This correlated with a basal spacing of 0.767 nm. Ageing of the sample resulted in more complete conversion of raw materials to LDH product, with clear peaks observed at 2θ values of 13.48 • and 27.10 • . The basal spacing was calculated to be 0.760 nm. Basal spacing has been found to be influenced by numerous factors including milling time, the amount of water present and possible carbonate contamination [15]. Lattice imperfections as a result of mechanically induced amorphitisation could further contribute to larger basal spacing values [5]. The addition of water has been known to decrease the degree of supersaturation, which could negatively impact morphology and crystallinity of the synthesised LDHs. It has further been observed that the crystallinity of LDHs can pass through a maximum, with lattice imperfections increasing with an increase in milling time [26]. The calculated basal spacing values obtained for both S1 and S2, after ageing, were found to be similar with those reported in literature [15]. Similarly, spectra for the aged samples obtained further correlated with existing literature [9,15,27].  Ca-Al LDH. XRD reflection patterns for samples S3 and S4 are as depicted in Figures 4 and 5, respectively. The presence of LDH was observed prior to ageing for S3 with a primary peak at a 2θ of 13.50 • . This correlated to a basal spacing of 0.759 nm. Ageing of the sample improved conversion of raw materials to LDH product, with a decrease in Al(OH) 3 and Ca(OH) 2 peak intensity observed. Twinning primary peaks were observed at 2θ values of 13.54 • and 13.20 • corresponding to basal spacing's of 0.757 nm and 0.776 nm, respectively. Comparatively, the XRD spectra for S4 depicted the presence of LDH within the sample despite the lack of a direct carbonate source. The primary peak was observed to occur at 2θ of 13.47 • , with a basal spacing of 0.761 nm. The presence of CaCO 3 was further noted and likely due to atmospheric carbonate contamination. Ageing of the sample resulted in the formation of katoite (Ca 3 Al 2 (OH) 12 ) as well as twinning primary LDH peaks. These were observed to occur at 2θ of 13.27 • and 13.58 • , corresponding to basal spacing of 0.772 nm and 0.755 nm. Twinning peaks could be attributed to the presence of different LDH phases within the sample and differ through carbonate content [22,27,28]. Basal spacing values obtained for both samples, before and after ageing, suggest the presence of either or a mixture of calcium monocarboaluminate and a dehydrated polymorph of calcium hemicarboaluminate that forms upon ageing. Basal spacing reported for each of these were 0.750 nm [22,27] and 0.760-0.770 nm [22,28], respectively. Ca-Al LDH synthesised in the presence of a carbonate source formed more readily when compared to that synthesised with no CaCO 3 . Previous studies have shown that Ca-Al LDH and katoite (tricalcium aluminate) can result when reacting Ca(OH) 2 and Al(OH) 3 in the absence of an additional phase or carbonate source, with little to no LDH formation occurring at times [11,21,29]. Studies have also shown that, upon the addition of a third phase such as CaCO 3 or CaCl 2 ·2H 2 O LDH, formation occurs more readily with little to no katoite formation. This suggests a complex relationship between the formation of Ca-Al LDH and katoite. It has been suggested that the presence of pillared anions such as Cl − and CO 3 2− assist in the formation of the layered structure of Ca-Al LDH. The third phase therefore stabilises the Ca-Al LDH structure allowing for formation to occur more readily [21].  Zn-Al LDH. The XRD patterns of samples S5 are depicted in Figure 6. The primary LDH peak was observed to occur at a 2θ value of 13.86 • , corresponding to a basal-spacing of 0.749 nm. Conversion prior to ageing was observed to be incomplete with Zn 5 (CO 3 ) 2 (OH) 6 and Al(OH) 3 peaks observed at 2θ values of 15.08 • and 21.31 • , respectively. Ageing of the sample resulted in the increase in LDH peak intensity with a primary peak at 2θ of 13.70 which corresponds to a basal spacing of 0.748 nm. A decrease in raw material peaks were observed with ageing; however, conversion remained incomplete for the selected synthesis conditions. Metal ratios (M II :M III ) typically employed for the synthesis of Zn-Al LDH are between 2:1 and 4:1 for conventional methods such as co-precipitation [18]. It has recently been suggested that molar ratios suitable for mechanochemical synthesis range between 1:1 and 2:1, with nearly pure phase Zn-Al LDH as the result [18]. Slight differences were observed for basal spacing values obtained. These were observed to differ from those reported in literature (0.758 nm, Zn 4 CO 3 (OH) 6 ·H 2 O as starting material) with the Zn content influencing the basal spacing obtained [18]. Cu-Al LDH. The XRD spectra obtained for samples S6 and Cu2(OH)2CO3 can be seen in Figure  7. The results for S6 prior to ageing were considered to be inconclusive as no obvious LDH peaks were identified. A decrease in Cu2(OH)2CO3 peak intensities were, however, observed to occur after 1 h of milling activity. Ageing of the sample resulted in the formation of a peak at 13.74° with a second peak present at approximately 27.76°. Identification of the LDH peaks were difficult due to prominent and overlapping Cu2(OH)2CO3 peaks. Basal spacing associated with the observed primary peak was determined to be 0.746 nm. This was observed to be smaller than that reported in literature (0.754 nm) [17]. The formation of Cu-Al LDH was reported to be dependent on the selected rotational speed and therefore the degree of amorphitization [17].  Cu-Al LDH. The XRD spectra obtained for samples S6 and Cu 2 (OH) 2 CO 3 can be seen in Figure 7. The results for S6 prior to ageing were considered to be inconclusive as no obvious LDH peaks were identified. A decrease in Cu 2 (OH) 2 CO 3 peak intensities were, however, observed to occur after 1 h of milling activity. Ageing of the sample resulted in the formation of a peak at 13.74 • with a second peak present at approximately 27.76 • . Identification of the LDH peaks were difficult due to prominent and overlapping Cu 2 (OH) 2 CO 3 peaks. Basal spacing associated with the observed primary peak was determined to be 0.746 nm. This was observed to be smaller than that reported in literature (0.754 nm) [17]. The formation of Cu-Al LDH was reported to be dependent on the selected rotational speed and therefore the degree of amorphitization [17].

Fourier Transform Infrared Spectroscopy
The main purpose of the FT-IR data was to support the notion that LDH was present within each sample. This was due to the fact that not all LDH peaks were easily identifiable when conducting XRD analysis.
Mg-Al LDH. The FT-IR spectra for S1 and S2, before and after ageing, are depicted in Figures 8 and 9. Prior to ageing peaks were observed to occur between 3500 cm −1 and 3700 cm −1 for both samples and could be attributed to the stretching vibrations of free -OH groups [15]. Similarly, peaks located between 3250 cm −1 and 3600 cm −1 are likely due to bonded -OH within both samples [15]. Peaks located at 1367 cm −1 (S1) and 1365 cm −1 (S2) could be attributed to carbonate interactions (CO 3 2− v3 vibrations) [9,15,20]Ageing of the samples at 80 • C for 24 h, resulted in the intensification of these peaks. A broad peak, from 3250-3700 cm −1 , specifically 3425 cm −1 (S1) and 3460 cm −1 (S2), was observed to form upon ageing of both samples. This could be assigned to the -OH stretching vibrations that occur within the layered brucite like structure of the LDH as well as interlayer water molecules [9,15,20]. Peaks observed between 500 and 900 cm −1 could be attributed to M-O and MO-H (M = Mg, Al) vibrations [30]. Peaks located from 1100-900 cm −1 are typical of Si-O interactions from SiO 2 impurities in the MgO raw material [30]. The FT-IR spectra for both S1 and S2 after ageing coincide with spectra observed in literature [9,15,20,27].  Ca-Al LDH. FT-IR spectra for S3 and S4, before and after, ageing are depicted in Figures 10 and 11. Peaks observed between 3700-3300 cm −1 could be due to MO-H vibrations within each sample [5,21] as well as the vibration of -OH (v2) within the inorganic main layers of the LDH structure [5]. Prior to ageing, S3 depicted peaks at 1418 cm −1 and 876 cm −1 , which could be due to carbonate vibrations on the surface of the LDH structure present [5,21]. Similarly peaks at 1370 cm −1 could be attributed to CO 3 2− v3 vibrations within the interlayer of the LDH structure [21]. Ageing of the sample resulted in similar spectra to that obtained prior to ageing. The twinning carbonate interactions observed near 1366 cm −1 have been suggested further to be the result of two different environments for carbonate present, likely due to different Ca-Al LDH phases present [27]. The spectra for S4 prior to ageing were observed to be similar to that of S3. Peaks observed at 1414 cm −1 could be the result of carbonate within the system [5,21]. Synthesis, drying, and ageing were conducted, without the use of an inert gas, under air atmosphere. Carbonate contamination was therefore possible. Ageing of the sample resulted in the formation of twinning peaks at 1366 cm −1 and 1415 cm −1 . These could once again be attributed to interlayer and surface carbonate interactions of LDH formed within the system [5,21].  Zn-Al LDH. The FT-IR spectra for S5, as well as that of Zn 5 (CO 3 ) 2 (OH) 6 , is depicted in Figure 12. The peaks at and before 3468 cm −1 could be attributed to the stretching vibrations of -OH groups within the sample [18]. Spectra of the sample was observed to resemble that of the Zn 5 (CO 3 ) 2 (OH) 6 raw material before implantation of the ageing step. Ageing of the sample resulted in more complete conversion of raw materials to LDH product. Broadening of peaks between 3000 cm −1 and 3700 cm −1 were noted and are due to O-H stretching of hydroxyl groups [31]. Peak formation at approximately 1356 cm 1 and 1620 cm −1 was observed and is likely due to the asymmetric stretching vibrations of CO 3 2− within the interlayer of the LDH [18,30,31]. Peaks below 1000 cm −1 could further be attributed to M-O vibrations (M = Zn, Al) [18,31]. Cu-Al LDH. The FT-IR spectra for S6 and Cu 2 (OH) 2 CO 3 , before and after ageing, are depicted in Figure 13. Samples obtained depicted similar spectra to that of the Cu 2 (OH) 2 CO 3 . Identification of bond interactions associated specifically with LDH was therefore difficult and inconclusive. It was noted, however, that additional and broadening of peaks occurred between 3300 cm −1 and 3700 cm −1 and was consistent with Cu-Al LDH spectra reported in literature [17,32]. Additional peaks could further be attributed to bonded and free -OH groups within the sample [30]. Ageing resulted in the formation of a minor peak at 1632 cm −1 which could be due to the vibrations of water molecules [17].

Scanning Electron Microscopy
SEM imaging was conducted to provide insight on the morphology of the samples obtained before and after ageing and is depicted in Figure 14. Prior to ageing, both S1 and S2 depicted similar morphology with no obvious differences observed. Ageing of the samples resulted in the formation of thin platelet like structures. These were observed to correlate with SEM images associated with Mg-Al LDH reported in literature [9]. No obvious morphological differences were observed between S3 and S4 prior to ageing. Ageing of S3 resulted in the formation of thin crystalline platelets which correlate to those reported in literature for Ca-Al LDH 1 . Comparatively, S4 depicted structures associated with katoite, calcite, and Ca-Al LDH, and was observed to coincide with the XRD data obtained. No clear morphological differences were observed for S5 and S6 before and after ageing. Platelet like structures could be construed as being present.

Conclusions
A facile mechanochemical method for the synthesis of Mg-Al, Ca-Al, Zn-Al, and Cu-Al LDH was successfully developed. Implementation of ageing resulted in more complete conversion of raw materials to LDH product. Starting materials selected were a mixture of oxides, hydroxides, and basic carbonates, promoting 'green' synthesis of LDH materials. The synthesis method eliminates the use of pH control and avoids the production of salt rich effluent. The selected mill offers potential for commercial up-scaling with a variety of iso mills available, capable of processing large volumes of liquid reagent. Mill operation is versatile offering the potential for batch, semi-batch, or continuous synthesis of LDH materials. Traditional methods have predominantly made use of ball mills, mixer mills, or a mortar and pestle [2]. Wet milling allows for ease of transfer of reagents to downstream processes with little to no process interruption. Inclusion of the ageing step has allowed for an increase in conversion of raw materials to LDH product and improved sample morphology. This offers potential for converting the one-step wet mechanochemical route to that including a second step. Previous studies conducted have indicated that a grinding step is necessary for LDH formation to occur more readily [9,17,18]. The grinding activity allows for mechanochemical activation of raw materials, creating active sites. The selected horizontal bead mill has successfully resulted in the formation of multiple types of LDH product; however, raw material peaks were still observably present under the selected synthesis conditions. Samples S1, S2, S3, and S5 resulted in successful LDH formation with high intensity peaks observed after ageing. Samples S4 resulted in LDH formation prior to ageing; however, katoite formation occurred during the ageing step. The presence of a carbonate source was therefore necessary for LDH formation to occur and continue successfully. Cu-Al LDH did not form readily, with no clear indication of its presence prior to ageing. LDH peaks were observed to have formed upon ageing with precursor materials still notably present. The mill therefore offers potential for a successful one-step wet mechanochemical technique. Further investigation of a change in milling parameters such as retention time, jacket water temperature, bead size, solids loading and bead loading is warranted.