Layered double hydroxides (LDHs) consist of alternate positively charged mixed metal hydroxide layers and negative charged interlayer anions. The stoichiometry of these materials can be formulated as [Mz+1−xM3+x(OH)2]p+[(An−)p/n·mH2O] with z = 2, M = bi- and trivalent metallic elements, A = organic or inorganic anions and m = amount of interlayer H2O depending on the temperature, relative humidity and hydration level  . A special case is Mz+ = Li+(z = 1) and M3+ = Al3+. The ratio between Li and Al is always 1:2  while the ratio between Mz+ and M3+ (z = 2) can vary strongly  depending on which M2+ ion or synthesis parameters are used. These layered materials are able to intercalate negatively charged and neutral molecules or exchange the interlayer anion with organic      or inorganic   anions of different sizes or charges. The [Mz+1−xM3+x(OH)2]p+ main layer remains stable and is not capable of ion exchange once it is formed.
Two well-known and described LDHs are [LiAl2(OH)6] [Cl∙mH2O]    and [Mg2Al(OH)6] [Cl∙mH2O] . Both compounds are generally synthesized by a direct reaction of a LiXor MgX2(X = Cl−, OH−, NO3−, etc) with Al(OH)3[2/4]or by a hydrothermal reaction with higher temperatures and pressures  .
The structure of Al(OH)3is built up of double layered sheets of hexagonally packed O atoms. Two thirds of the octahedral holes are occupied by Al atoms. Using LiX as the reaction partner leads to the formation of [LiAl2(OH)6] [X·mH2O] with Li+ cations entering the vacancies in the aluminum hydroxide layers and A entering the interlayer space   . The structure of the resulting Li-LDH depends directly on the structure of the used aluminium hydroxide. Syntheses using gibbsite as starting material leads to Li-LDHs with hexagonal symmetry, while reactions with bayerite or nordstrandite produce LDHs with rhombohedra symmetry    . In the brucite-like structure of [Mg2Al(OH)6][X·mH2O], Mg2+ is octahedrally coordinated to six OH− anions. These octahedrons share edges and form thereby a layer. Substituting Mg2+ with a trivalent ion like Al3+ leads to a positive charge which can be compensate by interlayer anions  . [Mg2Al(OH)6][X·mH2O] can also be rhombohedra or hexagonal   . The pure [Mg2Al(OH)6] [X·mH2O] phase produced within this work was hexagonal (P6/m).
Almost all publications concerning the interlayer anion exchange or the synthesis and physicochemical properties use a combination of Mz+ (z = 1 or 2) + M3+ in the main layer with a variation of two different elements    -  . The aim of this research is to invent a novel solid solution by adding a Me2+ cation (Mg2+) into the structure of a Li-LDH. The distance between Li+ and O2− ions in a [LiAl2(OH)6] [Cl·mH2O] LDH is 2.129Å and between Al3+ and O2− 1.926Å  . In a [Mg2Al(OH)6][Cl·mH2O] LDH, Mg2+ and Al3+ ions occupy the same positions with the same distance of 2.013Åbetween the cations and O2−  . Comparing both structures and the bonding distances, it should be possible for Mg2+ ions to occupy the Al3+ and the Li+ position in the solid solution.
The starting materials for this work were LiCl (ROTH, purity ≥ 99%), MgCl2・6H2O (AppliChem ≥ 99%), AlCl3・6H2O (Serva ≥ 98%) and NaOH (Fluka ≥ 97%). XRD investigations and loss of ignition (LOI) were done with all chemicals to exclude contaminations and determine the amount of crystal water.
APAN anlytical X’PERT³ Powder diffractometer with Pixcel detector and a Cu radiation (45 kV/40 mA) was used for the X-ray powder diffraction (XRD). The samples were prepared with back loading procedure and recorded from 5˚ - 70˚2θ with a step width of 0.017˚2θ and a irradiation time per step of 19.685 s. Thermogravimetric analysis and differential thermal analysis (TGA/DTA) for the dried samples (relative humidity (RH) 35%) were done simultaneously by the 320U from Seiko Instruments under nitrogen flow and a 2.5 K/min heating rate between 25˚C - 1000˚C. Fourier transform infrared spectra (FTIR) were recorded by an IFS 55 Equinox FTIR spectrometer from Bruker (400 - 4000 cm−1). The scanning electron microscope (SEM) pictures were taken by a JOEL 640 SEM and the chemical compositions of the samples were proven by a Horiba Ultima2 inductively coupled plasma optical emission spectroscopy (ICP-OES).
All mixtures of the initial components were prepared in a glove box with nitrogen atmosphere to avoid carbonatization. The synthesis were carried out in 35 ml PTFE-lined stainless-stealautoclaves  by adding solutions of LiCl, MgCl2∙6H2O and AlCl3·6H2Owith a W/S of 15: 1(a total of 1 g salts with 15 ml deionized/ decarbonized water) and 5M NaOH until an alkaline pH (8.5) was reached and heating it up 10 h and 48 h. A series of experiments with different temperatures, synthesis times and pH were carried out to achieve the best result for a pure solid solution. The synthesis temperature was varied between 100˚C, 120˚C, 140˚C and 160˚C and two different synthesis times (10 h and 48 h) and pH (8.5/9.5) were tested. To synthesize pure [Mg2Al(OH)6][Cl·mH2O] an exact ratio of 2 mol Mg2+: 1 mol Al3+was chosen andthe pure [LiAl2(OH)6][Cl·mH2O] was prepared by adding the Li+ and Al3+ salts in an exact 1 mol : 2 mol ratio. While the Mg containing LDH was prepared without problems, the Li LDH showed a high proportion of an amorphous phase. ICP-OES investigations stated that only 20% of Li+ was incorporated in the LDH structure leaving 80% of Li+ in the solution and the so remaining excess of Al3+ as an amorphous phase. A five times higher concentration of Li+  , as required by stoichiometry for the preparation of the pure [LiAl2(OH)6][Cl·mH2O] LDH, was necessary to compensate the 80 % lack of Li+ in the solid state.
After the synthesis of pure [Mg2Al(OH)6][Cl·mH2O], the amount of Li+ was increased and the amount of Mg2+ was reduced in 10 mol% (X = 0.1) steps until 100 mol% Li ([LiAl2(OH)6][Cl·mH2O]) was reached. The products were filtered, washed with 30 ml deionized water and dried (RH 35%) until a constant mass was reached.
The mineralogical phases were determined by X-ray powder diffraction, the chemical compositions of the products by ICP-OES using the filtrate and the synthesis products dissolved in HNO3   .
3. Results and Discussion
3.1. Stoichiometric Composition
First experiments were carried out at 100˚C, pH 8.5, 10 h synthesis time and a W/S ratio of 15:1. Using an exact stoichiometric ratio of Li+/Mg2+/Al3+ resulted in the synthesis of a high proportion of an amorphous phase with a small amount of a crystalline solid solution. After drying at 80˚C, XRD analysis showed a recrystallized Al(OH)3 phase next to the LDH main phase.
Investigations of the filtered solutions and the dissolved products with ICP-OES stated that, independent from the Mg reactant amount, 99% - 100% of Mg2+ but only 20% of Li+ were build-in into a LDH phase. The other 80% of Li+ remained in the solution. Due to the stoichiometric reactant ratio the leftover Li+ ions in the solution are leading to leftover Al3+. These Al3+ ions formed Al(OH)3 in the basic environment. Using higher temperatures (up to 160˚C) or synthesis times (48 h) showed no positive effect for the crystallization of a pure LDH phase. Increasing pH from 8.5 to 9.5 resulted in a slightly higher amount of a crystalline phase.
3.2. Composition with Increased Li+ Content
After a five times increasement of the stoichiometric amount of Li+ (equal to the pure Li-LDH synthesis), with a resulting ratio of Li: Mg: Al = 5: 1: 1, a pure crystalline LDH phase could be achieved. ICP-OES studies stated that >99% of Mg2+ and Al3+ and the needed 20 % of the five times higher Li+ concentration were build-in into the crystalline phase.
3.2.1. PXRD Analysis
By increasing X from 0 to 1 in 0.1 mol steps in [Li0+xMg2−2xAl1+x(OH)6][Cl∙mH2O], the amount of Mg2+is decreased and replaced by Al3+ and Li+. This leads to a change of the lattice parameter a and the cell volume. Comparing the ion radii of Mg2+ (0.65 Å) with Li+ (0.60 Å) and Al3+ (0.50 Å) it is to be expected that the lattice parameter a starts to decrease with higher Li+/Al3+content  . A dependent change in the lattice parameter c or the basal reflections (00l) is not visible. By means of the (110)/(112) and (300)/(302) peaks, it is easily possible to distinguish the two different phases [Mg2Al(OH)6][Cl·mH2O] (P6/m) and [LiAl2(OH)6][Cl·mH2O] (P63/m). Starting with X = 0 (pure [Mg2Al(OH)6][Cl·mH2O] phase) a separation in two phases is visible between X = 0.1 and X = 0.8 (Figure 1 and Figure 2). The ˚2Θ positions of the 110/112peaks at X = 0.1 - 0.8 show a peak shift to higher ˚2Θ angles in relation to the pure [Mg2Al(OH)6][Cl·mH2O]
Figure 1. XRD pattern of the test series with X for[Li0+xMg2−2xAl1+x(OH)6][Cl·mH2O] (120˚C/10 h) showing no visible phase separation at the (002)/(004) main peaks but two coexisting phases at higher ˚2Θ angle (60 - 65).
Figure 2. XRD pattern in the range of 60˚ to 65 ˚2Θ of the test series with X for [Li0+xMg2−2xAl1+x(OH)6][Cl·mH2O]. The pattern for x = 0.1 until x = 0.8 show two different phases.
phase (Figure 2 and Figure 3) which is also visible in the lattice parameter (Table 1). This shift increases with higher Li+ reactant amounts, which indicates Mg dominated solid solutions with different Li+/Mg2+ ratios.
While the (110)/(112) peaks are completely erased for X = 0.9, the (300)/(302) peaks shifted and the solid solution hasa different lattice parameter compared to [LiAl2(OH)6][Cl·mH2O] at X = 1  (Figure 2 and Figure 3, Table 1). The lattice parameter a is closeto the calculated ideal position of a solid solution. Between X = 0.1 - 0.8 the (300)/(302) peaks have nearly the same position which is shifted to lower ˚2Θ angles and the lattice parameter are also nearly constant. This indicates a stable Li dominated solid solution with a defined amount of
Figure 3. Lattice parameter a of two different phases with X for [Li0+xMg2−2xAl1+x(OH)6] [Cl·mH2O]. The black dashed line shows the theoretical lattice parameter of the solid solutions.
Table 1. Pawley fitted lattice parameter a/c for [Mg2Al(OH)6][Cl·mH2O] (X = 0), [LiAl2(OH)6][Cl·mH2O] (X = 1) and the split Li and Mg dominated solid solutions.
Mg2+ independent from the Mg2+ reactant amount. The miscibility gap for X = 0.1 - 0.8 was observed at all tested synthesis temperatures (100˚C - 160˚C) and times (10 h/48 h).
To synthesize pure solid solution phases, test series between X = 0.9 and X = 1 (in 0.02 mol steps) were conducted. XRD results show a single mineral phase with h0l peak shifts (Figure 4 and Figure 5). This peak shifts follow nearly the
Figure 4. XRD pattern of the test series with X = 0.9 - 1 for[Li0+xMg2−2xAl1+x(OH)6] [Cl·mH2O]. X was increased in 0.02 mol steps (120˚C/10 h). Due to a preferred orientation of 00l, the (100) and (105) peak is no longer visible for X = 0.96; 0.98; 1.
Figure 5. XRD pattern in the range of 60˚ to 65 ˚2Θ of the test series with series with X = 0.9 - 1 for [Li0+xMg2−2xAl1+x(OH)6][Cl·mH2O] with marked peaks. A shift for the (300) and (302) peaks is visible.
calculated shifts for the solid solutions (Figure 6). These experiments were also done at four different temperatures (100˚C, 120˚C, 140˚C, 160˚C). Although there is a shift difference depending on the temperature (Figure 6), no phase separation was observed for all investigated solid solutions (Figure 5).
The optimal results for a pure solid solution phase were achieved at 120˚C/10 h synthesis time/pH 9.5 and W/S ratio 15:1 (Figure 6/Table 2). The measured lattice parameters a differ only slightly from the calculated and the lattice parameters c are nearly constant (Table 2).
The products were fitted by Pawley fit and the space group was determined as P63/m for all pure solid solutions up to X = 0.9. Investigations of the lattice parameter a show a straight increase from ~5.08Å (X = 0)  to ~5.10Å (X = 0.1) as calculated (Figure 6/Table 2).
(a) (b)(c) (d)
Figure 6. Theoretical and measured lattice parameter a for the solid solutions with X = 0.1, 0.8, 0.9 - 0.98 for [Li0+xMg2−2xAl1+x(OH)6] [Cl·mH2O]and the pure Mg/Li LDH at (a) 100˚C; (b) 120˚C; (c) 140˚C; (d) 160˚C. For X = 0.1/0.8 two separated LDH phases are visible.
Table 2. Theoretical and measured/fitted lattice parameter (a) and (c) for the solid solutions with X = 0.9 - 0.98 and [LiAl2(OH)6][Cl·0.51H2O] at X = 1 (120˚C/10 h/pH 9.5/W/S 15: 1).
3.2.2. ICP-OES Analysis
To determine the chemical formula, all products were completely dissolved insuprapur 65% nitric acid and investigated with ICP-OES   . The results were used to calculatethe LDH formulas (Table 4). These calculations also stated a maximum content of an amorphous phase of <1%. Recrystallization tests showed no Al containing phases. Synthesis temperatures higher than 140˚C led to a destabilization of the LDH phase and the formation of AlO(OH) (Figure 7). The test series with 160˚C were repeated several times producing always AlO(OH) next to the LDH. Calculations showed an Al containing amorphous phase and crystalline AlO(OH) proportion of 10 % to 90 % (Table 3). The resulting lack of Al3+ in the solid solution leads to LDH phases with a higher Mg/Al ratio than 2:1 and therefore to the formation of a LDH with higher Mg2+ amounts next to the AlO(OH) phase (Figure 6(d)).
Table 3. Proportion of the amorphous phase/AlO(OH) depending on the synthesis temperature.
Figure 7. Pawley fit of a solid solution [Li0.90Mg0.25Al1.87(OH)6] ][Cl·mH2O] synthesized at 160˚C. A phase of AlO(OH) (*) is visible next to the solid solution (#). The broadening at 40˚ and 47˚ ˚2θ(small picture) is interpreted as stacking faults  .
3.2.3. Thermal Analysis
The amount of interlayer water was determined by TG/DTA for [LiAl2(OH)6] [Cl·0.50H2O], [Mg2Al(OH)6][Cl·0.55H2O]and all pure solid solutions (Table 4). An example for the Li-LDH, Mg-LDH and the solid solution with the highest Mg2+ amount [Li0.9Mg0.2Al1.90(OH)6][Cl·0.51H2O] is shown in Figure 8. Comparing the solid solution with the pure Li- and Mg-LDH, there is a high similarity in mass loss and exothermal reaction. The mass loss at 75˚C - 100˚C is caused by the removal of intercalated interlayer water   . With 4.5% for the pure Li-LDH, 4.2% for the solid solution and 4.7% for the pure Mg-LDH it corresponds with the loss of 0.50 to 0.55 water per formula unit of the LDHs. While the differential thermal analysis of the pure Li- and Mg-LDH show a single endothermic reaction at 275˚C - 325˚C, the solid solution shows two (260˚C and 320˚C). At this temperature, the LDH starts to dehydroxylate which results in the destruction of the metal hydroxide main layer     . Combining Li+ and Mg2+ with Al3+ in the main layer leads to a two-step dehydroxylation.
Figure 8. Thermogravimetric and differential thermal analysis of (a) [LiAl2(OH)6] [Cl·0.51H2O]; (b) [Li0.9Mg0.2Al1.90(OH)6] [Cl·0.50H2O]; (c) [Mg2Al(OH)6][Cl·0.55H2O] (120˚C/10 h/pH 9.5/W/S 15: 1) show the loss of interlayer water at 75˚C - 100˚C). Temperatures above 275˚C destroy the structure of the main layer. Heating rate: 2.5 K/min.
Table 4. Calculated chemical formulas based on ICP-OES results and interlayer water of the solid solutions X for [Li0+xMg2−2xAl1+x(OH)6][Cl∙mH2O] (120°C/10h/pH 9.5/W/S 15: 1).
3.2.4. FTIR Spectroscopy
To prove purity of the products, all samples were investigated by FTIR spectroscopy (Figure 9). Although there are 10 mol% Mg2+ in the solid solution, there is only a slight difference to a pure [LiAl2(OH)6][Cl·0.51H2O] FTIR spectrum visible. All three spectra show the typical H2O/OH− absorption at ~3500 cm−1 and 1630 cm−1    and only the spectra of [Li0.9Mg0.2Al1.90(OH)6] [Cl・0.50H2O] and [Mg2Al(OH)6][Cl・0.55H2O] show an insignificant amount of carbonatization with the absorption at 1380 cm−1    . The absorption of Al (980/720/520 cm−1) related groups is very good visible for the pure Li-LDH but not as distinct for the solid solution   . Mg related absorptions at 415 cm−1 are only visible in the pure Mg-LDH (Table 5). The amount of Mg2+ is high enough to influence the absorption spectra but not to show a clear Mg related absorption.
3.2.5. SEM Analysis
SEM pictures (Figure 10) show flat, (pseudo-) hexagonal particles with different sizes, starting at 2 - 3 µm until nearly nanosize. These particles form cluster in the size of 200 - 600 µm.
3.2.6. Structure of the Solid Solution
Based on the assumption that Mg2+ ions can occupy the positions of Li+ and Al3+ because of the fitting bonding length   , the ion radii  and the determined hexagonal P63/m space group, the structure of the pure phased solid solution should be identical with the Li-LDH (Figure 11). This is also indicated by the chemical composition with the formula [Li0.9Mg0.2Al1.90(OH)6][Cl・0.50H2O]. If Mg2+ ions could not enter one of the two octahedral positions, there would be two possibilities: they would exchange with Li+ ions only, which would reduce the amount of Li+ in the solid solution while the amount of Al3+ would not change, or they would exchange only with Al3+ ions with the opposite result. The results of this work show, that in fact Mg2+ has to be statistically distributed with 5 mol% on the Li+ and 5 mol% on the Al3+ position to provide the measured chemical formula.
Figure 9. FTIR spectrum of (a) [[LiAl2(OH)6][Cl・0.51H2O]; (b) [Li0.9Mg0.2Al1.90(OH)6] [Cl・0.50H2O]; (c) [Mg2Al(OH)6][Cl・0.55H2O] with the typical absorbed water (~3500 cm−1 and 1620 cm−1) and the metal-O and metal-OH vibrations (>1000 cm−1). Absorption at 2400 cm−1 is device related.
Table 5. Observed wavenumbers and the assignment bending.
Figure 10. SEM pictures of [Li0.9Mg0.2Al1.90(OH)6][Cl・0.50H2O] flat hexagonal particles with average crystal size of >3 µm.
Figure 11. View of the unit cell of [Li0.9Mg0.2Al1.90(OH)6]Cl・0.5H2O(based on Li-LDH structure  ―interlayer water excluded) with the octahedral positions of Li+ and Al3+. Both positions are occupied with 5mol% by Mg2+.
It is possible to synthesise a pure [Li0+xMg2−2xAl1+x(OH)6][Cl·mH2O] solid solution using autoclaves with temperatures of 100˚C, 120˚C and 140˚C with a maxi- mum amount of 10 mol% Mg2+ (X = 0.9). Using more Mg2+ in the reactant leads to a parallel formation of an Mg2+ dominated and a Li+ dominated solid solution. Optimal results for a pure solid solution can be achieved at 120˚C, pH 9.5, W/S15: 1, 10 h synthesis time. Changing the temperature to 160˚C provides the formation of an AlO(OH) phase. The pure solid solution with the highest Mg content is [Li0.9Mg0.2Al1.9(OH)6][Cl·0.50H2O].