Ni-based Semiconductors From Ldh-precursors Synthetized Using Amines

1. Introduction

Nanoparticles have attracted extensive research interest due to their great potential for many applications such as photoelectrochemical cells, heterogeneous catalyst, optical switching and single electron transistors [1,2,3,4]. The morphology of the particles has an effect on the properties and also influences the self-assembly process [5,6,7].

Nickel and its oxide nanoparticles display superior magnetic, electrical, thermal, optical, catalytic and mechanical performance [8,9,10,11]. Nanostructured NiO particles are of great interest, since in addition to having important applications as thermistors, sensors and additives, for gas and ceramic [12,13,14,15], they are also of great interest as precursors of Ni-based catalysts [16,17,18,19,20,21]. NiO nanoparticles have also a wide range of applications as a p-type semiconductor due to its stable wide band gap (3.6–4.0 eV) [22,23], although bulk NiO is an antiferromagnetic insulator [24,25].

NiO crystalline nanoparticles have been synthesized by several physical, chemical and biological methods [14]. Some of the synthesis approaches are chemical reactive processes, electrodeposition, solution growth, pulsed-laser deposition, high-temperature nickel oxidation, spray pyrolysis, sputtering, sol–gel technique, the reverse-micellar route, pulsed laser ablation and by microemulsion [10,11,26,27,28,29]. Metal oxides particles in a size between 1 and 100 nm with high surface area are desirable adsorbents, carriers and catalysts. NiO nanoparticles have catalytic activity in pyrolyzing biomass components, a property which is attributed to the different effects in terms of volume, quantum size, surface and macroscopic quantum tunnel. NiO nanoparticles have many important properties over those of bulk and micro-NiO particles [30]. El Kemary et al. [29] synthesized NiO nanoparticles using the reaction of nickel chloride with hydrazine at room temperature and thermal decomposition of the Ni(OH)₂ thus formed. It was reported that upon calcination of Ni(OH)₂ at 400 °C nanoparticles with size of 45 nm and the energy band gap of 3.54 eV are obtained [29]. Salavati-Niasari et al. [3] studied the effect of calcination temperature and the effect of the metal-to-ligand ratio on the particle size of NiO nanoparticles obtained from nickel phthalate complexes by a solid-state thermal decomposition route. Thus, by increasing the ligand-to-metal ratio and decreasing the calcination temperature, the particle size is reduced without agglomeration, obtaining nanoparticles in a range of 15 to 36 nm. Also, Hosny [22] studied the effect of changing the metal anion and metal-to-ligand ratio on the crystal lattice and the particle size of NiO nanoparticles obtained by solid-state decomposition of nickel anthranilic acid semi-solid complexes, obtaining NiO nanoparticles with a size of 8 nm.

The combination of different compounds which have excellent electronic properties can improve the electronic properties of the resulting composite material [31,32]. In recent years, a great number of examples with heterojunctions to modify the photocatalyst and improve its photocatalytic activity has been reported. This type of constructions has been considered as one of the solutions to the serious recombination of photo-generated holes and electrons. Yu’s team proposed the concept of S-scheme heterojunction by an electrostatic interaction, the built-in electric field and the band bending to ensure strong redox photocatalyst [33,34].

Among the rare earth oxide series, lanthanum oxide (La₂O₃) is a p-type semiconductor that has been extensively researched due to its unique chemical and physical properties, which make it suitable for certain electronic applications. It is the only lanthanide oxide with an empty Ln-4f shell [35]. La₂O₃ has a wide band gap of 5.5 eV, is thermally stable and non-toxic, and has a significant relative dielectric constant (K > 20) [36,37]. Because of these properties, La₂O₃ is used in potential applications like biosensors, catalysts, dielectric layers, fuel cells, gas sensors, rechargeable batteries, photoelectric conversion, and optical devices for measuring various body temperatures, and biomedical [37]. Quan et al. [38] designed and constructed a novel S-scheme La₂O₃/AgCl heterojunction catalyst with La-Cl bond by ball milling method. The catalytic activity and stability of La₂O₃/AgCl photocatalyst for BPA degradation were significantly improved; where the formation of heterojunction structure and interfacial La-Cl bond effectively promoted the transfer and quenching of relatively useless holes and electrons in La₂O₃ and AgCl.

Cerium oxide (CeO₂) is another lanthanide oxide with great interest. This oxide is a n-type semiconductor with a wide band gap which has high oxygen storage capacity, strong redox capability and it has been considered as a promising material for photocatalytic applications [39,40,41]. By the combination of CeO₂ with other semiconductors it is possible to improve the photocatalytic efficiency [42,43,44,45]. This improvement is possible by the transference of photoexcited electrons from the conduction band of CeO₂ to the conduction band of the other semiconductor rather than recombining with the holes in the valence band. Meanwhile, the photoexcited holes will flow from the valence band of the semiconductor to the valence band of CeO₂. In this way, Sherly et al. [46] have studied the effect of CeO₂ on the structural, optical and photocatalytic properties of ZnO, observing an improvement on the photocatalytic efficiency of ZnO by the inter particle electron and hole transfer between both semiconductors.

Furthermore, the synthesis of NiO nanoparticles using layered double hydroxides (LDH) as precursors has been described in the literature. Zheng et al. [47] generated in situ NiO and Co_1.29Ni_1.71O₄ by oxidizing and calcining Ni-Co layered double hydroxides (LDHs) to construct an S-scheme heterojunction heterogeneous catalyst in order to improve the electron transfer efficiency and promote the catalyst performance. The resulting catalyst has higher light absorption intensity and photocurrent response than Ni-Co LDH, and also a smaller electronic impedance and good separation efficiency of electrons and holes.

Layered Double Hydroxides, also known as hydrotalcite-like compounds, are a family of solids with at least two different metal cations in their layers and anions occupying the interlayer space [48,49,50]. The layered structure of these materials could be explained from the brucite structure, where, during the coprecipitation process, an isomorphic substitution of divalent cations by trivalent cations takes place in the octahedral environment formed by OH^- ions [M(OH)₆]. The isomorphic substitution gives rise to positively charged layers. The electroneutrality of the compound is achieved by the incorporation of anions in the interlayered space, so that LDH are also known as anionic clays. LDHs could incorporate a great diversity of anions in their interlayer, both organic and inorganic ones, with a variety of size and charge [51]. In addition, water molecules may occupy the remaining free space in the interlayer space. The wide range of anions and divalent and trivalent cations that can be used to prepare LDH, provide them a diversity of composition, based on the general formula ${\left[M_{1-x}^{II}{M}_x^{III}{\left(OH\right)}_2\right]}^{x+}{\left[A^{m-}\right]}_{{\displaystyle \raisebox{1ex}{$x$}\!\left/ \!\raisebox{-1ex}{$m$}\right.}}·n{H}_{2}O$, $M^{II}$ and $M^{III}$ being the divalent and trivalent cations in the octahedral positions, and $A^{m-}$ the interlayer anion, with $x$ defined as the $M^{III}/(M^{II}+M^{III})$ molar ratio [48,52].

LDHs are easily prepared, are cheap and their composition can be easily tuned. Therefore, LDH have emerged as promising materials due to their properties and applications in numerous fields, such as water decontamination [53,54], catalysis [55,56,57,58], drug delivery [59,60], electroactivity [61], biomedicine [62], and others. Moreover, LDHs prove to be excellent precursors of high specific surface area mixed metal oxides (MMOs) with applications in various industrial sectors such as semiconductors, photocatalyst, catalysts, electrodes, adsorbents, etc. [63,64,65,66].

Among the methods reported in the literature to prepare LDH, coprecipitation is the most commonly used. This method is based on the slow addition of a mixed solution of the desired $M^{II}$ and $M^{III}$ cations in a fixed ratio over an alkaline solution, generally an aqueous NaOH solution. The synthesis is carried out at constant pH by the addition of a second alkaline solution, allowing the coprecipitation of both metallic salts [52]. The nature and concentration of the cations and the anions, the precipitation medium, the pH and temperature are important parameters to control the process. Kloprogge et al. [67] reported the influence of the pH in the synthesis of Zn/Al LDH, the range 11-12 being that in which the samples exhibit the best crystallinity.

The alkaline compound used to provide the precipitation medium could influence in the precipitation rate and agglomerate formation of the final material. Studies on the effect of the precipitating agent used during the synthesis process of the LDHs precursors of the Ni-Fe-Al₂O₃ catalysts obtained by calcination and applied in the methanation reaction can be found in the literature. Hwang et al. [68,69] used NaOH, NH₃aq, Na₂CO₃ and (NH₄)₂CO₃ as precipitating agents, showing that the precipitation rate is different depending on the compound used, with Na⁺-based precipitation media leading to a higher precipitation rate than NH₄⁺-based ones. The crystal size varies in the sense NiFeAl-NaOH > NiFeAl-NH₃aq> NiFeAl-Na₂CO₃ > NiFeAl-(NH₄)₂CO₃. As for the catalytic performance, it increases in the sense: NiFeAl-NaOH < NiFeAl-NH₃aq < NiFeAl-Na₂CO₃ < NiFeAl-(NH₄)₂CO₃.

In a previous work, the synthesis of ZnAl LDHs using different amines [methylamine (MMA), dimethylamine (DMA) and trimethylamine (TMA)] in the coprecipitation medium as modifying agents of the morphology and crystallinity of LDH have been reported [70]. The use of organic compounds of a basic nature, such as amines, allows to obtain well crystallized compounds, overall, when DMA or TMA are used.

In this work the results obtained using dimethylamine (DMA) as precipitating agent to prepare NiO-Al₂O₃, NiO-La₂O₃-Al₂O₃ and NiO-Ce₂O₃-Al₂O₃ semiconductors are discussed. The aim is to study the effect of the precipitating agent on the properties of the nanostructured NiO particles using LDH as precursors. It is aimed to achieve a higher dispersion of Al, La and Ce oxides and the formation of highly dispersed NiO nanoparticles.

2. Materials and Methods

2.1. Materials

Ni(NO₃)₂·6H₂O (98-102%), Al(NO₃)₃·9H₂O (98-102%), La(NO₃)₃·6H₂O (98-102%) and NaOH (98%) were purchased from Panreac. Ce(NO₃)₃·6H₂O (98-102%) was purchased from Scharlau. Dimethylamine (99.5%) was purchased from Panreac. All reagents were used as received.

2.2. Synthesis

The samples were synthetized by the co-precipitation method at constant pH 10 [19]. The cation solutions in Ni/Al molar ratio 70/30 and in Ni/X/Al molar ratio 70/5/25, (X = La, Ce), were prepared. The precipitation of the LDH precursors was carried out in a 4.5 M aqueous solution of dimethylamine and 2 M anhydrous sodium carbonate. The synthesis started with the dropwise addition of the cations solution by means of an addition pump that regulates the dripping at a rate of 2 mL/min, on the solution used as the precipitation medium, which had been previously heated to 60 °C. The pH of the medium was kept constant at a value close to 10 by adding the necessary volume of a 3 M NaOH solution using a CRISON pH-Burette 240. The slurry obtained was submitted to an aging process at 60 °C during 1 h under intense agitation. Finally, the solid was separated from the aqueous phase by centrifugation. It was washed repeatedly with distilled water until pH close to 7, in order to remove the counterions of the starting salts. The solids were dried in an air oven at 40 °C and manually grounded using an agate mortar. The solids were then calcined at 600 °C for 6 h in air with a heating ramp of 10 °C/min.

2.3. Characterization

The element chemical analysis for metal ions was carried out at NUCLEUS (University of Salamanca, Spain) by an atomic emission spectrometer with inductively coupled plasma source (ICP-OES) Ultima II of YOBIN IVON.

Powder X-ray diffraction (PXRD) patterns were recorded in a PANalytical X′Pert diffractometer equipped with a Cu-Kα radiation (λ_mean = 0.15418 nm) and a fast X′Celerator detector. The diffractogram was collected over 2θ range from 5° to 80° with a step size of 0.05° and scan time 15.25 s per step. The Scherrer equation, D = kλ/βcosθ [71,72], where k is the shape factor, a constant taken in this case as 0.9; λ is the wavelength of the radiation used; β the full width at half maximum (FWHM) and θ the diffraction angle. Warren correction for instrument line broadening was considered, using diffraction data from standard silicon (Si) powder.

A Perkin-Elmer Spectrum One instrument was used to record the transmission FT-IR spectra from 4000 cm^-1 to 450 cm^-1 with a nominal resolution of 2 cm^-1 using compressed KBr pellets.

Thermogravimetric (TG) and differential thermal analysis (DTA) were performed using an SDT Q600 instrument from TA Instruments. Thermal analysis was performed by heating from room temperature to 900 °C at a rate of 10 °C/min in a continuous flow of oxygen (L'Air Liquide, 99.995%) (50 ml/min).

The specific surface area (SSA) of the samples were estimated from the N₂ adsorption-desorption isotherm (L'Air Liquide, 99.999%) at -196 °C recorded on a Micromeritics ASAP 2020 instrument. Before measurement, a portion of ca. 0.15 g of the powdered calcined sample was degassed in vacuum (< 30 µm Hg) at 250 °C and held for 30 min before measurement. The specific surface area was calculated by the multipoint Brunauer-Emmett-Teller (BET) method [73,74] in the of relative pressure p/p₀ range from 0.05 to 0.3.

A Shimadzu UV1201 UV–vis spectrophotometer was used to record the ultraviolet-visible diffuse reflectance spectra (UV-vis DRS) of the samples at room temperature in the 200-800 nm range and using BaSO₄ powder as a background. The band gap values were calculated by the Tauc method [75,76].

3. Results

3.1. Element Chemical Analysis

The results of the elemental chemical analysis for the LDH precursors are included in Table 1. The M²⁺/M³⁺ molar ratios are also included in this Table; for samples incorporating rare earth elements, the molar percentage of M³⁺ is the sum of the molar percentage of Al³⁺ and the molar percentage of Ce³⁺ or La³⁺. From the element chemical analysis data for the cations composing the sample, the chemical formula of each LDH has been calculated assuming the general chemical formula [Ni_1-y-zX_yAl_z (OH)₂](CO₃)_z/2 · n(H₂O), where X is Ce or La and that carbonate is the only interlayer anion. The water molecules per chemical formula have been calculated from the TG curves of LDHs precursors (discussed below).

The M²⁺/M³⁺ molar ratio obtained is somewhat lower than the expected one from the amounts of reagents in the reaction mixture. Only the sample containing Ce shows a value close to the expected one, with only a very small deviation (<2%). The lowest ratio is found for the NiAl-DMA sample. This deviation from the theoretical molar ratio is probably due to an incomplete precipitation of Ni, which a small part could remain in the precipitation medium solution and be removed in the washing process.

The carbonate anion content has been calculated from the molar ratio M³⁺/(M³⁺+M²⁺) and taking into account that it is a divalent anion. It is also assumed that carbonate is the only interlayered anion balancing the positive charge of the layers, as it is shown by the FT-IR spectroscopy results (discussed below).

3.2. Powder X-ray Diffraction (PXRD)

The PXRD patterns of the LDH precursors are included in Figure 1A. The diffractograms reveal the layered structure of the solids, corresponding to a 3R-polytype of hydrotalcite-like structure [77]. In all cases, the most intense diffraction peak is recorded close to 11.6° (2θ), ascribed to the diffraction by planes (003), with a spacing of 7.65 Å. This spacing is in agreement with the values reported by Miyata et al. [78] for hydrotalcite-like structure compounds with carbonate anion occupying the interlayer space. The diffraction by (006) planes is recorded close to 23.2° (2θ), with a spacing of 3.82 Å. The diffraction by (009) planes would be expected to be between 35° and 40° (2θ), not being possible to identify it due to overlapping with other diffraction peaks. Diffraction peaks by (110) and (113) planes are recorded at 61.2° (2θ) and 62.5° (2θ), with spacing of 1.51 Å and 1.48 Å, respectively.

As observed by Ho et al. [79], the crystallinity of the hydrotalcite-like phase decreases with the incorporation of rare earth elements. Thus, the diffraction peaks (003) and (006) of the NiCeAl-DMA and NiLaAl-DMA samples are less intense and broader than the peaks recorded for the NiAl-DMA sample. Moreover, the diffraction peaks (110) and (113) show a lower resolution in the samples containing cations of these rare earth elements, probably because of a broadening of the peaks due to a lower crystallinity of these samples. This can also be explained by a lower ordering both in the direction of layer stacking and a lower ordering within the layers. Due to the large size of the La³⁺ cation (ionic radius in octahedral coordination: 117.2 pm [80]) compared to Ni²⁺ (83 pm) and Al³⁺ (67.5 pm), it is very likely that its inclusion in the brucite-like layers takes place up to a certain, small, value, as further inclusion would destabilise the structure. So that, the La³⁺ not included in the layers is segregated as La₂(CO₃)(OH)₂ in the NiLaAl-DMA sample [81]. A similar fact takes place in the NiCeAl-DMA sample, where due to the large ion size of Ce³⁺ (115 pm), the cubic CeO₂ phase is formed after oxidising to Ce⁴⁺ (101 pm) [82]. This phenomenon has been observed previously in similar compounds with Y³⁺ (104 pm) [83]. The larger size of these ions compared to the others undoubtedly increases the distortion of the structure, resulting in the broadening of some of the diffraction peaks.

From the powder X-ray diffraction diagrams, the lattice parameters c and a describing the unit cell of the hydrotalcite-like structure compounds can be determined [84,85,86], the values of which are given in Table 2. These parameters have been calculated from the maxima positions of diffraction peaks due to planes (003), (006) and (110) [84]; so that, c = 3 · [1/2 · (d(003) + 2·d(006))] ≈ 22.7 – 23.1 Å, and a = 2 · d(110) ≈ 3.019 − 3.037 Å. Also, the average crystallite size (D) in the direction of the layer stacking (Table 2) has been calculated using the Full Width at Half Maximum (FWHM) value of diffraction peak (003) and the Scherrer equation, D = kλ/βcosθ [71,72]. The differences found in the lattice parameters of the samples containing Ce³⁺ and La³⁺ with respect to the NiAl-DMA sample are less than 1% in all cases, so they can be accepted to be coincident within the experimental error. As for the average crystallite size, it can be observed that for the samples containing Ce³⁺ and La³⁺ it decreases, similarly to their crystallinity.

The PXRD diagrams of the samples calcined at 600 °C are included in Figure 1B. The diagram for the NiAl-DMA-C600 sample shows the formation of the bunsenite (periclase-type) phase of NiO, determined by comparison of the positions and relative intensities of the diffraction peaks with the JCPDS database [87]. The absence of diffraction peaks associated to the Al₂O₃ phase is probably due to the formation of an amorphous Al₂O₃ phase, as observed by other authors for these calcination temperatures [88,89]; however, as described by these authors, when calcination is carried out at higher temperatures it is possible to observe the formation of the corresponding mixed oxide, suggesting that the Al³⁺ is dissolved in the NiO phase [88]. In samples containing Ce³⁺ or La³⁺, the formation of the oxides of these elements, CeO₂ and La₂O₃, respectively, can be observed next to the NiO phase. For all samples, the most intense diffraction peak of the NiO phase, which is attributed to the (002) plane of the crystal structure, is recorded at a position close to 43.5° (2θ), with a spacing of 2.09 Å. As with the LDHs precursor, the crystallinity of the NiO phase is reduced when rare earth element cations are incorporated.

Table 3 includes the lattice parameter a values describing the unit cell of the NiO phase, with cubic NaCl-type structure, for each of the LDH calcined samples. A slightly lower lattice parameter a value than that reported in the literature (4.177 Å) [90] is calculated for the NiAl-DMA-C600 sample. This decrease could be due to the insertion of Al³⁺ cations into the NiO crystal lattice, which, due to the smaller ionic radius of the Al³⁺ cation with respect to the Ni²⁺ cation, would produce a decrease in the unit cell dimensions of the nickel oxide [90,91]. A similar explanation could be applied to the value obtained for the lattice parameter a of the NiLaAl-DMA-C600 sample, in which the insertion of La³⁺ cations into the NiO crystal lattice could take place producing an increase in the unit cell dimensions due to the larger size of the La³⁺ cation (117.2 pm). For the NiCeAl-DMA-C600 sample, a lattice parameter a value similar to that found in the literature for NiO (4.177 Å) is found [90], so a priori it would not be possible to think about the insertion of Ce³⁺ cations (115 pm) in the NiO crystal lattice. However, it could also happen that the two effects mentioned above are taking place, i.e. that on the one hand the insertion of Al³⁺ cations in the lattice leads to a decrease of the lattice parameter a, while on the other hand the insertion of a larger size cation, like Ce³⁺ cation, induces an increase of the lattice parameter, which results in an average value of the parameter a similar to that reported in the literature for the NiO structure.

From the FWHM value of the diffraction peak corresponding to the (002) plane, the average crystal size (D) has been calculated by the Scherrer equation [71,72]; the calculated values are included in Table 3. It is observed that for the three samples studied, the NiO phase shows a crystallite size close to 3.7 nm. The use of DMA as a precipitation agent leads to a smaller NiO crystal size than that found for the LDH precursors prepared with NaOH as precipitation agent [79], and it is not significantly modified by the incorporation of Ce and La, obtaining quite similar crystal sizes in the three samples studied.

3.3. FT-IR Spectroscopy

The FT-IR spectra of the LDH precursors show the characteristic bands of hydrotalcite-like compounds with carbonate as the interlayered anion (Figure 2A). A broad band can be observed around 3440 cm^-1 due to the stretching vibrational modes of the hydroxyl groups in the brucite-like layers and the water molecules in the interlayer space. The width of this band is caused by the different strengths of the O-H bonds due to the formation of multiple hydrogen bonds between the OH groups with each other and with the interlayer H₂O molecules. In addition, the band at 1630 cm^-1 is due to the bending mode of the water molecules in the interlayer space [92].

A shoulder can be observed around 2980 cm^-1 which is due to the hydrogen bonding interaction between the hydroxyl groups and the carbonate anions in the interlayer space [88,93,94,95]. The bands at 1363 cm^-1, 800 cm^-1 and 603 cm^-1 are attributed to the antisymmetric stretching ν₃(E′), out-of-plane bending ν₂(A₂″) and angular bending ν₄(E′) vibrational modes, respectively [92]. The spectra of the LDH precursor promoted with Ce and La (Figure 2A) show a band at 1496 cm^-1 due to the splitting of the band associated to the ν₃(E′) as a consequence of a decrease in the symmetry of the carbonate anion in its interaction with the water molecules that also occupy the interlayer space. This symmetry decrease also leads to the activation of the ν₁(A₁′) mode originally forbidden for a D_3h symmetry, but allowed for both C_2v and C_s [92,96,97]. This vibrational mode is associated to the very weak shoulder identified around 1050 cm^-1.

On the other hand, in the low wavenumber region (below 1000 cm^-1) the bands corresponding to the Metal-O stretching and Metal-OH bending vibrational modes are recorded. The FT-IR spectra show the bands at 564 cm^-1 and 430 cm^-1 due to the Ni-OH and Al-OH vibrational modes, respectively [98]. For LDHs promoted with rare earth elements, a band at 852 cm^-1 due to the stretching vibrational modes of Ce-O and La-O bonds is also observed [99,100].

None of the bands characteristic of DMA are observed in the FT-IR spectra of the synthesised samples, which would indicate a correct washing of the obtained solids.

The FT-IR spectra of the LDH calcined at 600 °C show bands characteristic of metal oxides derived from the LDH precursor composition (Figure 2B). All spectra show a broad band around 3440 cm^-1 and the bands at 1634 cm^-1due to the stretching and bending vibrations of OH groups of water molecules adsorbed on the solids. Furthermore, the spectra show two bands centered at 1510 cm^-1 and 1406 cm^-1 corresponding to the above mentioned vibrational modes for the carbonate anion, which could come from a slight recarbonation process from the adsorption of atmospheric CO₂ in contact with the samples after calcination [101]. The band registered at 857 cm^-1 in the FT-IR spectra of NiLaAl-DMA-C600 sample is ascribed to the out-of-plane bending mode of vibration, which confirms the presence of carbonate [102]. Below 1100 cm^-1 a broad band is found due to the overlapping of different bands due to the Metal-O and Metal-O-Metal vibrational modes. The band recorded at 1087 cm^-1 is typical for γ-alumina due to the Al–O vibration mode [103,104]. The band at 820 cm^-1 is attributed to the vibrational mode of surface Al–O bond and the band at 598 cm^-1 is assigned to the stretching vibrational mode of octahedrally coordinated Al–O [104]. The Ni–O stretching vibration modes are found in the range 500–400 cm^-1 [104,105]. The bands due to the stretching and bending modes of the La-O and Ce-O bonds, which should appear in the 700 to 500 cm^-1 range [106,107], are not observed as they are overlapped with the Ni-O and Al-O modes mentioned above.

3.4. Thermal Analysis

Figure 3 shows the thermogravimetric curves of the LDHs precursors, together with the DTA curves. All samples show a similar behaviour. Below 230 °C, the first two decomposition stages (removal of adsorbed and interlayered water) take place successively and overlapping, which makes difficult their resolution. The endothermic effects of that stages are clearly identified in the DTA curves, in which the first effect appears, in turn, unfolded. Between 230 and 400 °C, the last two decomposition processes (dehydroxylation of the layers and decarbonation) of the layered compound take place. As in the initial stage, these processes occur practically simultaneously. As it can be seen in the DTA curves, these processes are recorded as a single endothermic peak. Only in the sample containing La, a third decomposition process can be observed between 350 and 500 °C. This process could be due to the decomposition of the La₂(CO₃)₂(OH)₂ phase, where the carbonate anion would be forming stronger bonds, which would hinder the decarbonation process, requiring a higher temperature for its decomposition. Like in the previous decomposition processes, this third stage corresponds to an endothermic process.

From the TG curves, the total mass loss percentages were calculated for each sample, finding a mass loss of 38.5%, 36.6% and 38.4% for the samples NiAl-DMA, NiCeAl-DMA and NiLaAl-DMA, respectively. Consequently, the residues after the decomposition process were between 62-64% of the initial sample mass. The mass loss difference between the sample containing Ce and the NiAl-DMA sample could be attributable to the presence of CeO₂, which has been identified by PXRD, segregated during the co-precipitation process. In the sample with La, the presence of a La₂(CO₃)₂(OH)₂ phase is also observed by PXRD, Figure 1A; however, no minor mass loss is observed due to the decomposition process of this second phase and formation of La₂O₃. Furthermore, from the first stage decomposition of the TG curves were calculated the number of water molecules per chemical formula of each sample (Table 1), being practically the same in all three samples.

3.5. Specific Surface Area and Porosity

The textural properties of the LDHs calcined at 600 °C have been studied from the N₂ adsorption-desorption isotherms at -196 °C. The corresponding isotherms are included in Figure 4, where it can be observed how the isotherm recorded for the NiAl-DMA-C600 sample is clearly different from the isotherms of the Ce- and La-containing samples. The isotherm of the NiAl-DMA-C600 sample corresponds to type IV according to the IUPAC classification [108,109]. Type IV isotherms are typical of mesoporous materials, where the most characteristic feature is the hysteresis loop, which is associated to condensation in the pores. At high p/p₀ values an adsorption limit is reached, leading to a plateau, indicating the complete filling of the pores. The initial part of the type IV isotherm can be attributed to monolayer-multilayer adsorption [74]. The hysteresis loop observed in the desorption process corresponds to type H1 according to the IUPAC classification [74,108], generally associated to porous materials with well-defined cylindrical-type pores or compact agglomerates of uniform spheres. In the case of the samples containing Ce and La, the isotherms correspond to type II according to the IUPAC classification [108,109], characteristic of adsorption on non-porous or mesoporous materials, where unrestricted monolayer-multilayer adsorption can occur. In addition, the isotherms of both samples show a hysteresis cycle, corresponding to type H3 according to the IUPAC classification [74,108], whose presence indicates that adsorption takes place in slit-shaped pores formed by layered particles.

The SSA values, calculated by the BET (S_BET) method [73,74], together with the pore volume (V_P) and the average pore diameter calculated by the BJH method [74,110] are included in Table 4. It can be observed how the samples incorporating Ce and La oxides show lower BET specific surface area values compared to the NiAl sample. In contrast to what was reported by Ho et al. [79] for samples prepared using NaOH as the precipitating agent, when DMA is used as precipitating agent, the surface area decreases when rare earth elements are incorporated. This difference can be explained by the distortions in the layers induced by the incorporation of Ce or La, which affect the degree of stacking and agglomeration, leading to changes in the N₂ adsorption-desorption curves and surface area values of the samples. Furthermore, these changes could be also due to the formation of pores with different sizes caused by the segregation of CeO₂ and La₂O₃. Thus, as can be seen in Table 4, the pore volume and the average pore diameter increases when Ce is incorporated in the sample, and even more when La is incorporated. In all cases, from BJH average desorption pore diameter values, the presence of small pore sizes, less than 15 nm, can be confirmed.

3.6. UV-Visible Diffuse Reflectance Spectroscopy

The optical Diffuse Reflectance Spectra (DRS) of the calcined samples are included in Figure S1. The optical band gap was estimated using Tauc’s equation [75,76]: $\alpha h\nu =A{(hy-E_g)}^n$, where $\alpha$ is the absorption coefficient, A is an energy independent constant and $n$ is equal to 1/2 and 2 for direct and indirect transitions, respectively. According to the literature, NiO nanoparticles have a direct band gap [22], so Figure S2 shows the plots of ${\left(\alpha h\nu \right)}^2$ vs. $h\nu$ and the optical band gaps of the samples were calculated by the Tauc method [75] (Figure 7, Table 5). The curves indicate that the values of the direct band gap (Eg) are equal to 3.43, 3.53 and 3.59 eV for NiO nanoparticles derived from LDH precursors of NiAl-DMA, NiCeAl-DMA and NiLaAl-DMA, respectively. The band gap values suggest that NiO nanoparticles are semiconductors.

The lowest band gap value is determined for the NiAl-DMA-C600 sample, indicating an improvement in the conductivity of the material. Although all the samples show similar crystal sizes (≈3.7 nm), the crystallinity of the NiO phase decreases with the incorporation of Ce or La. The improvement in the conductivity of the material could be due to the higher crystallinity of the Ni and Al sample. In the case of the samples incorporating Ce or La, the band gap values do not differ significantly from those obtained for the Ni,Al sample, presenting, in both cases, band gap values lower than those reported in the literature [22,23]. Thus, the synthesis method guarantees the production of new materials with better electrical properties than those of NiO obtained by other means, with the advantage of improving their behaviour in photocalytic processes.

4. Conclusions

Hydrotalcite-like solids as semiconductors precursors were prepared with Ni and Al as major cations, and with the addition of La and Ce, in a 70/30 divalent/trivalent molar ratio using DMA as precipitating agent. This synthesis methodology provides LDH precursors with good crystallinity, being the solid with only Ni and Al in the structure the one that showed better crystallinity against the LDH precursors incorporating Ce or La. This improvement in the crystallinity of the LDH precursor can also be observed in the NiO nanoparticles obtained in each case, which has an impact on the electrical properties of the materials obtained. According to this methodology, p-type semiconductors with band gap values between 3.4 and 3.6 eV are obtained, which better than those reported in the literature. In addition, the incorporation of Ce or La cations in the samples leads to a reduction of the specific surface area of the semiconductors obtained, which together with the good band gap values, make these materials suitable for their application as photocatalysts.