3.1.2. Powder X-Ray Diffraction
The PXRD patterns of the synthesised samples are included in the Figures S1–S4
. All of the studied samples showed diffraction patterns typical for layered double hydroxides with hydrotalcite- and hydrocalumite-like structures for our MA–X and CA–X samples.
Mg,Al-LDHs showed a first peak at ca. 10°2θ (d-spacing = 8.7 Å for MA–MnO4, 8.8 Å for MA–S2O8), which was due to the reflection of the (003) planes in hydrotalcite-like structures with a rhombohedral 3R stacking of the main layers [21
]. For these samples, the lattice parameter c
corresponded to three times the spacing value of this peak (i.e., c
= 26.1 Å for MA–MnO4 and c
= 26.4 Å for MA–S2O8). Lattice parameter a
gives the average distance between two adjacent cations in the main layers. It can be calculated as twice the d-spacing of the reflexion of the (110) planes, which is responsible for the first peak of a doublet at approximately 60°2θ; a
= 3.03 Å for MA–MnO4 and a
= 3.02 Å for MA–S2O8, which were in relatively good agreement with the reported data for similar LDHs [18
]. The typical doublet at approx. 60°2θ could not be easily detected in the PXRD diagrams of these samples, and only a single peak was recorded. Hence, the here stated values must be considered with caution and might deviate due to the quality of the studied samples. Possible explanations are the low crystallinity of the Mg,Al-samples due to the selected aging time and stacking faults in the crystal lattice, most likely as a turbostratic disorder of the cations in the layers in in c-direction,, which generates the so called “shark’s fin” peak shape of the (h
) reflections in the range from 35–50°2θ.
Ca,Al-LDHs likewise showed a first peak at approximately 10°2θ, which corresponded to the reflection of the (003) planes of CA–MnO4 (d-spacing = 8.6 Å) and (006) planes for the sample CA–S2O8 (d-spacing = 8.7 Å). The variation in the planes responsible for the diffractions originates because Ca,Al-LDHs can exhibit two different stacking sequences, namely, rhombohedral 3R polytype with X = MnO4−
and rhombohedral 6R polytype with X = S2
]. Consequently, the lattice parameter c
corresponds to three times the spacing of this first peak for the permanganate intercalated sample (c
= 25.8 Å) and six times the spacing for the peroxydisulphate intercalated sample (c
= 52.0 Å). In these samples, the lattice parameter a
differs largely from those stated for the Mg,Al-LDHs as the presence of the larger calcium cations, if compared to the magnesium counterparts, in the main layer of the LDHs result in increased distances between neighbouring atoms in the brucite-like layers [20
]. For the Ca,Al-LDHs samples, a
values of 5.73 Å (CA-MnO4) and 5.75 Å (CA-S2O8) were observed, which were in very good agreement with the values reported in the literature [20
The longer aging time of the Ca,Al-LDHs samples led to the development of intense, sharp peaks, which indicated a well crystallized material had formed for these samples. In contrast, Mg,Al-LDHs samples rather showed broader, weaker peaks, indicating a less crystalline material. The crystallite size was calculated for the first peak, corresponding to the spacing of the LDH in the c
-direction, using the Scherrer equation [33
]. The resulting crystallite sizes are given in Table 2
. It should be noticed that despite the different nature of the intercalated anion, the crystallite size in the direction perpendicular to the layers was roughly the same for both Mg,Al-samples on one side, and for both Ca,Al-samples on the other. It should also be noted that the values for the Ca,Al-samples were significantly larger than for the Mg,Al-LDHs, probably as a result of the longer reaction time, allowing a slow increase in the crystallite size from the mother liquor during synthesis.
3.1.3. Thermal Analysis and Evolved Gas Analysis
The thermal decomposition of the synthesised materials was studied using TG-DTA coupled to a MS to determine the gases evolved upon increasing the temperature. The curves are included in the Figures S5–S8
and that for sample MA–MnO4 is included in Figure 2
. All curves of the studied material revealed the expected mass losses that are typical for Mg,Al- and Ca,Al-LDHs [34
], respectively (i.e., three mass loss steps for the Mg,Al-LDHs, while four for the calcium counterparts, respectively), a difference clearly related to the nature of the layers. Significant differences can be also observed on changing the nature of the intercalated anion. The thermal behaviour of these solids was also different from that reported for nitrates or carbonates containing LDHs, where the interlayer anions are removed as a gaseous species upon heating at or above ca. 500–600 °C. However, in the current study, the decomposition of permanganate or peroxydisulphate led to permanent anions, which were not expected to be removed upon heating in the experimental conditions and temperature range here applied.
The values of the mass losses recorded in each step are summarized in Table 3
In all cases, the first mass loss can be ascribed to the removal of externally adsorbed water (if any) and that of the interlayer water, as confirmed by mass spectrometry of the evolved gases for sample CA–MnO4, which showed a signal at m/z = 18 (H2O+).
The amount of interlayer water (see Table 4
) was calculated from the mass content of the layer metal cations and interlayer anions (Table 1
) and this first mass loss. To perform this study, it was assumed that MnO4−
) were the only species existing in the interlayer, together with water molecules. The error was not too large, making this approach or considering the presence of nitrate and/or carbonate, to fully balance the positive charge of the layers, especially in the case of the S2O8 samples. The DTA curves of the studied samples showed medium intensity endothermic peaks for the Mg,Al-LDHs and high intensity endothermic peaks for the Ca,Al-LDHs in the same temperature range. In all cases, a small mass loss was recorded above ca. 550 °C, which might correspond to the release of occluded water molecules. However, the main mass loss, at intermediate temperatures, occurred in a single step for the Mg,Al-samples, but in two steps for the Ca,Al-LDHs.
The main loss took place between ca. 150 and 600 °C, but the behaviour observed for the permanganate and peroxydisulfate intercalated samples was quite different and are discussed separately.
For the Mg,Al-LDHs, a second mass loss was recorded up to 530 °C, which was due to the decomposition of the interlayer phases and evolution of water by condensation of the hydroxyl groups of the main layer. Only NO2, as co-intercalated nitrate to balance the excess of the net positive charge of the main layers and being present due to the use of Mg2+ and Al3+ nitrate salts during the synthesis of the reactants, was detected in this second mass loss step, together with water vapour, evolved for sample MA–MnO4. While for sample MA–S2O8, NO2, together with a minor amount of CO2, was evolved, together with SO3 coming from the decomposition of peroxydisulphate. In this last case, it seems that the interlayer anions were S2O82− and NO3−, which came from the initially used magnesium and aluminium nitrates used for the synthesis of the samples as well as CO32−, most likely intercalated from atmospheric CO2. This amount was qualitatively rather low as the amount of evolved gases was not quantified. The presence of nitrate was confirmed by FTIR spectroscopy (see below).
The question remains about the thermal decomposition of interlayer permanganate anion. It is well known [36
] that thermal decomposition of KMnO4
takes place in the same temperature range, leading to a mixture of KMnO4
, and O2
. Formation of birnessite should be discarded, as an oxidising atmosphere was used in the current case. In our case, evolution of O2
was not monitored, as oxygen being the carrier gas, small changes in its partial pressure from the thermal decomposition of MnO4−
would be negligible. However, it should be noted that this is the pattern for the decomposition of bulk KMnO4
, not of MnO4−
existing in the interlayer, which can undergo decomposition in a different fashion. The recorded mass loss was caused by the transformation of the intercalated permanganate to some sort of manganese oxide, most likely Mn3
, haussmanite, or MnO2
, pyrolusite, which is confirmed by the evolution of O2
, together with the release of NO2
, from co-intercalated nitrate, and water from main layer hydroxyl groups. This mass loss was accompanied in the DTA curve by a strong endothermic effect. The shape of the DTA curves for samples MA–MnO4 and MA–S2O8 were rather similar as well as the corresponding TG curves. However, the gases detected by MS in the second mass loss effect for sample MA–S2O8 corresponded to CO2
, and SO3
, together with H2
O. This means that, in this case, in addition to S2
, the interlayer also contained small portions of nitrate and carbonate, the tentative origin of which has been described above. In addition, the partial decomposition of the S2
anion was also observed, with the evolution of SO3
The Ca,Al-LDHs samples showed a different behaviour from that shown by the permanganate-containing sample. The first and last steps extended in similar temperature ranges for both sets of samples, but the main difference is that the intermediate mass loss splits into two well defined processes. From the MS of the evolved gases, it is concluded that the second mass loss (13.8% of the initial sample mass for the sample CA–MnO4, and 9.4% for sample CA–S2O8) corresponded to the removal of water mainly through the condensation of a layer of hydroxyl groups. For the third mass loss (15.4% and 23% for samples CA–MnO4and CA–S2O8, respectively), the MS showed the release of NO2 for sample CA–MnO4and of NO2 and SO3 for sample CA–S2O8. For the reasons given above, the evolution of MnO4was not monitored.
All synthesised samples showed similar patterns in their thermal decomposition. The Mg,Al-LDH samples showed a three step thermal decomposition of the solids. The first one corresponded to the removal of the interlayer and eventually externally adsorbed water molecules. The second one corresponds to the condensation of the layer of hydroxyl groups with the evolution of water, together with NO2 from the interlayer nitrate, the anions in the original reagents, which complete the electrical balance in the solid, and SO2 for sample MA–S2O8. The third step probably corresponds to the evolution of gases formed during the early stage of the decomposition, which resulted being occluded and are released only at high temperatures.
On the contrary, the CA-LDHs showed a decomposition in four steps. The first and the fourth step probably had the same origin as the first and the third one in the decomposition of the Mg,Al-LDH samples. The second one seems to correspond to the evolution of water vapour from the layer of hydroxyl groups, and the third one to the decomposition of the interlayer counter anions (i.e., peroxydisulphate and nitrate). Processes involving the interlayer species occur simultaneously with the condensation of layer hydroxyls for the MA–LDH samples, but not for the CA-LDH ones.
The calculated formulas of the studied compounds, which were based on the results of the TG-DTA-MS analyses, are given in Table 4
. The content of magnesium, calcium, aluminium, and the interlayer anions MnO4−
were calculated based on the results of the ICP-OES analyses (see the raw data in Table 1
). The water content in the interlayer was calculated from the results of the thermal analyses of the solids. The nitrate content included was calculated based on the electrical balance between the layers and the interlayer, considering that NO2
evolution was observed in all cases.
3.1.4. FTIR Spectroscopy
All samples studied showed absorption bands typical for layered materials related to the hydrotalcite- and hydrocalumite-subgroups. These bands can be assigned to vibrations of the hydroxyl groups of the main layers, water molecules of the interlayer, the intercalated anions, and of the lattice of the main layers [37
]. The precise positions of the bands in all recorded spectra are given in Tables S2 and S3
shows the FTIR spectra of the studied samples. For the Mg,Al-LDHs samples, a broad absorption band was recorded between 3575–3345 cm−1
, which corresponded to the stretching vibration of hydroxyl groups in the interlayer and the main layer of the LDH. For Ca,Al-LDHs, this broad absorption band was recorded as a double band with vibrations of the hydroxyl groups in the main layer corresponding to the first band and those of the two interlayers in the second band. The broadening of this absorption band is caused by hydrogen bonds. The absorption band at 1632 cm−1
corresponds to the bending vibration of the interlayer water molecules. All samples showed an intense absorption band at 1385 cm−1
, which is most likely caused by the stretching vibrations of nitrate species [29
] due to insufficient washing of the samples. From the chemical elemental analysis, it can be assumed that some of these nitrate anions are in the interlayer to completely balance the positive charge of the brucite-like layers. The absorption band at 785 cm−1
corresponded to bending vibrations of the metal hydroxide groups of the main layer [29
]. These could only be found in the spectra of the Ca,Al-samples. Absorption bands recorded below 700 cm−1
were caused by the metal-hydroxyl translation modes in the lattice of the LDHs.
In addition to the absorption bands typical for every LDH, these spectra showed further bands caused by their respective interlayer anion. Hence, permanganate intercalated samples showed absorption bands caused by the stretching vibrations of the tetrahedral-coordinated permanganate-anion at 904 cm−1
) and 826 cm−1
]. Sample CA–MnO4 additionally showed an absorption band at 786 cm−1
, corresponding to the symmetric stretching vibration of the tetrahedrally-coordinated permanganate-ion in the interlayer of the LDH [42
The peroxydisulphate intercalated samples showed additional adsorption bands caused by different vibrations in the peroxydisulphate-ion. Bands at 1312 cm−1
, 1271 cm−1
, and 1111 cm−1
corresponded to the stretching vibrations of the SO42−
]; sample CA–S2O8 only showed the last mentioned band. Bending vibrations of the sulphate groups led to bands at 685 and 667 cm−1
], which were only recorded for sample MA–S2O8. The stretching vibrations of the S–O–O–S bridge caused absorption bands at 1060 cm−1
) and 836 cm−1
]. The absorption band at 423 cm−1
can also be assigned to bending vibrations of the S–O–O–S bridge in the peroxydisulphate-ion [41
The evaluation of the FTIR spectra of the synthesised samples confirmed the successful intercalation of the desired interlayer anions permanganate and peroxydisulphate, together with the probable presence of nitrate anions.
3.1.5. Particle Size Distribution
The particle size distribution was measured for all synthesised samples on the dried sample material. All samples were treated under ultrasound for 5 up to 15 min to minimize the aggregation of the primary particles. All diagrams are included in the Figures S9–S12
The Mg,Al-LDH samples revealed a maximum at around 300 µm with a sharp decrease in the curve for values above this maximum. Below the maximum, the recorded curve decreased more steadily, revealing shoulders at approximately 30 and 2 µm. Application of 15 min ultrasound resulted in the evolution of a second maximum at 30 µm for sample MA–S2O8, whereas no significant differences could be seen in the respective curve of sample MA–MnO4 after 15 min of ultrasound treatment. The evolution of the second maxima in sample MA–S2O8 might be due to a re-agglomeration of small particles, which were separated during the ultrasound treatment [44
Ca,Al-LDHs samples showed a maximum at 100–100 µm in the curve of the untreated sample. Towards a larger or smaller particle size, the curve decreased slowly. The curve for sample CA–MnO4 showed one shoulder at ca. 5 µm, which was absent in the curve of sample CA–S2O8. However, this curve showed shoulders at 0.5 and 125 µm. Upon 15 min ultrasound treatment, the mentioned shoulders evolved into new maxima at 0.5 and 3.5 µm in both CA–LDHs. Small shoulders could be seen at 20 µm and 80 µm.
Sample MA–MnO4 was the only sample to show a monomodal particle size distribution at 110 µm, the other synthesised samples revealed bimodal particle size distributions.
(0.5) values defined as the median particle size in µm, which divides the population exactly into two equal halves for all samples, are given in Table 5
As shown in Table 5
, the Mg,Al-LDHs samples generally had larger particle sizes than the Ca,Al-LDHs ones. The peroxydisulphate intercalated samples showed smaller average particle sizes compared to the permanganate intercalated LDHs. Upon ultrasound treatment, the average particle size of all samples decreased significantly, the largest decrease being detected for sample CA–MnO4 (reduced by a factor of 34). Such a behaviour should be closely related to the specific nature of the layers and to a lesser extent, to the interlayer anion. It is probable that small differences in the surface properties could account for these findings. It seems that the development of particles and their size distribution is strongly dependent on the precise stirring conditions during synthesis and aging as well as on the precise drying conditions.