Introduction

Layered double hydroxides (LDHs), also known as anionic clays or hydrotalcite-like compounds, are natural or synthetic solid compounds composed by layers of metallic hydroxides and interlayer spaces filled with anions and neutral molecules. Its structure is based on brucite’s (Mg(OH)2), with partial substitution of bivalent cations by trivalent ones that introduces positive charges in the layers, and for the structure to be held together there is a need to be balanced with interlayer anions [1,2,3,4,5].

The LDHs can be described by the formula: \(M_{1 - x}^{\text{II}} M_{x}^{\text{III}} \left( {\text{OH}} \right)_{2} A_{x/n}^{n - } .\delta {\text{H}}_{2} {\text{O}}\), where MII and MIII are, respectively, the bi- and trivalent cations and An− is the n-charged interlayer anion. They can be represented by alternative nomenclature, such as Mills’ (International Mineralogical Association): LDH xMII yMIII.A[B]-Pt, where x and y are the metals proportion in the formula, B is a complex anion and Pt is the crystal polytype of the compound [3].

Usually the ratio MII/MIII is 2:1 or 3:1, always with the majority of MII. The anion influences the stability of the material, being the \({\text{CO}}_{3}^{2 - }\) the most stabilizing and common anion found in these compounds [4, 5].

These materials have received much attention in the last decades, because of its technological applications such as ion exchangers, flame retardancy agents, catalysts and support for metallic nanoparticles [1, 6,7,8,9], whereas they have been used with palladium as catalysts for cross-coupling reactions, as the Suzuki reaction [10,11,12].

C–C and C–N cross-coupling reactions are imperative in the synthesis of many organic structures with biological and electro-optics applications [13–16]. Considering the formation of C–N bonds [17] a composite material composed of Cu/Al LDH and Pd could be active for the N-arylation reaction since both catalytic approaches would be considered, the Ullmann–Goldberg [15] and the Buchwald–Hartwig [16] type reactions. However, one must be aware about the integrity of the layered catalyst at usual temperatures of 100 °C or beyond that are applied in catalytic processes having high-boiling point solvents. This is the temperature range to which the Cu/Al LDH has never had its stability tested. The proposal of the present work is to screen by thermal analysis the LDH stability and, concerning the composites, to infer the mode the LDH supports palladium and how to tune this.

Experimental

Synthesis of the LDH

The LDH was synthesized by coprecipitation at constant pH = 8 simultaneously mixing an aqueous solution 0.225 M Cu(NO3)2.3H2O and 0.075 M Al(NO3)3.9H2O—ratio Cu/Al = 3:1 with an alkali solution 0.50 M NaOH and 0.15 M Na2CO3 into a solution 10−6 M NaOH (pH = 8). The coprecipitation was carried out under magnetic stirring at ambient temperature and the pH remained 8 during the process. A blue gel was precipitated, and it was isolated by centrifugation. The calcinations at different temperatures (120, 200, 400, 600 and 800 °C) were done in furnace for 1 h using alumina crucible.

Synthesis of the composites LDH/Pd

The composites of LDH with Pd were obtained by mixing the LDH during 24 h at 80 °C under magnetic stirring with 0.025 M Na2PdCl4 or 0.012 M Pd2(dba)3 solutions in N,N-dimethylformamide (DMF), affording LDH/Pd1 and LDH/Pd2 materials, and with 0.025 M Na2PdCl4 solution in water, obtaining the LDH/Pd3 material. After each process, the solid was isolated by vacuum filtration and dried at room temperature.

Methods

The powder X-ray diffraction (XRD) patterns were obtained with a Rigaku diffractometer Ultima IV, 2 kW, using CuKα with kβ Ni filter and step of 0.05° in the range of 5°–80°. The FT-infrared spectra (FTIR) were obtained with a Nicolet Magna-IR spectrometer with 4 cm−1 resolution in the range of 400–4000 cm−1, and the samples were mixed with KBr in order to have 1% pellets. The element quantification by energy-dispersive X-ray spectroscopy (EDS) was performed in a JEOL JSM 6460-LV microscope. The CHN elemental analysis was carried out using a ThermoFinnigan elemental analyzer FlashEA 1112. The thermogravimetric (TG) curves were acquired in SHIMADZU DTG-60 analyzer, applying 5 °C min−1 in argon or synthetic air, under a 50 mL min−1 flow rate.

Results and discussion

The XRD pattern of the Cu/Al LDH is shown in Fig. 1a, and it evidences the characteristic hydrotalcite-like structure peaks in 2θ = 11.6° (003), 23.5° (006), 35.5° (012), 40.2° (015), 47.8° (018) and 63.0° (110), indexed in accordance with the literature [1]. It was identified a second phase of Cu2(OH)2CO3, malachite, identified by Crystmet #561579 card.

Fig. 1
figure 1

a Powder XRD pattern and b FTIR spectrum of LDH 2CuAl.CO3. Legend: (filled triangle) Cu2(OH)2CO3

Full size image

Using the Bragg Law it was possible to calculate the interlayer distance of the LDH: d(003) = 0.759 nm. The lattice parameters c and a were also determined: c = 3 × d(003) = 2.275 nm and a = 2 × d(110) = 0.295 nm; these values had good agreement with previously data in the literature [2].

The crystallite size was calculated using the Scherrer formula for the peaks (003), (006) and (012):

$$T = \frac{0.9\lambda }{{\beta \;{ \cos }\theta }}$$
(1)

in which T is the mean crystallite size, λ is the wavelength, θ is the scattering angle and β is the maximum width at half-height. Its calculated value was 25 nm.

In the FTIR spectrum, as shown in Fig. 1b, the following vibration modes can be assigned: νO-H at 3410 cm−1, related to the LDH’s hydroxyls and superficial water; δH-O-H at 1631 cm−1, from interlayered and hydrated water; asymmetric νC-O at 1355 and 1504 cm−1, from interlayer carbonate; asymmetric νN-O at 1384 cm−1 from interlayer nitrate; and νAl-O at 451 cm−1 related to the Al-O bond in octahedral geometry [6].

The elemental analysis by CHN and EDS (Tables 3 and 4, supplementary material) permitted the obtainment of \({\text{CO}}_{3}^{2 - }\) (0.21 mol/100 g), \({\text{NO}}_{3}^{ - }\) (0.01 mol/100 g, neglected due to the low quantity), Cu (0.72 mol/100 g) and Al (0.28 mol/100 g) quantities in the material. With these values it was possible to determine the LDH formula as Cu4Al2(OH)12CO3, and the proportion between LDH and malachite in the sample was 94:6%. The actual Cu/Al ratio in the LDH was 2:1, different from the nominal ratio of 3:1. According to Mill’s nomenclature [3] the LDH was denominated LDH 2CuAl.CO3.

With the TG analysis under synthetic air (Fig. 2) it was identified four temperature segments for the thermal decomposition of the layered compound, shown in Table 1 with its respective mass loss values. The LDH was calcined in each specific temperature. The XRD patterns and FTIR spectra are shown in Fig. 3.

Fig. 2
figure 2

TG curves data from the LDH 2CuAl.CO3

Full size image
Table 1 Mass loss values of the LDH 2CuAl.CO3 (synthetic air)
Full size table
Fig. 3
figure 3

a Powder XRD pattern and b FTIR spectra of LDH 2CuAl.CO3 calcined at different temperatures. Legend: (open circle) LDH 2CuAl.CO3, (filled triangle) Cu2(OH)2CO3, (filled circle) CuO, (open square) CuAl2O4

Full size image

At 120 °C (Fig. 3a) it was observed the beginning of the thermal decomposition process of the LDH, with the displacement of the peak (003) from 11.6° (d(003) = 0.759 nm) to 13.1° (d(003) = 0.676 nm) (stage I). This could be assigned to the interlayer water loss as shown by the weakening of the δHOH mode in the FTIR spectrum (Fig. 3b, r.t. and 120 °C spectra). At 200 °C this XRD peak was displaced to 13.9° (0.637 nm) and the baseline became curved, evidencing the presence of an amorphous phase, probably the double oxi-hydroxide derived from the LDH decomposition. The process related to this stage (II) can be represented by the following chemical reaction:

$${\text{Cu}}_{4} {\text{Al}}_{2} \left( {\text{OH}} \right)_{12} {\text{CO}}_{3(s)} \to {\text{ Cu}}_{4} {\text{Al}}_{2} {\text{O}}_{y + z} \left( {\text{OH}} \right)_{12 - 2y} \left( {{\text{CO}}_{3} } \right)_{1 - z(s)} + \, y{\text{H}}_{2} {\text{O}}_{(g)} + \, z{\text{CO}}_{2(g)} .$$
(2)

Between 200 and 400 °C (stage III) the malachite was decomposed and CuO was formed:

$${\text{Cu}}_{2} \left( {\text{OH}} \right)_{2} {\text{CO}}_{3(s)} \to {\text{ CuO}}_{(s)} + {\text{ CO}}_{2(g)} + {\text{ H}}_{2} {\text{O}}_{(g)}$$
(3)

In addition the decomposition of the LDH ends at 600 °C when no significant carbonate bands are seen in the IR spectrum (stage IV):

$${\text{Cu}}_{4} {\text{Al}}_{2} {\text{O}}_{y + z} \left( {\text{OH}} \right)_{12 - 2y} \left( {{\text{CO}}_{3} } \right)_{1 - z(s)} \to {\text{ Cu}}_{4} {\text{Al}}_{2} {\text{O}}_{7(s)} + \, 6{\text{H}}_{2} {\text{O}}_{(g)} + {\text{ CO}}_{2(g)}$$
(4)

And at 800 °C it was identified the growth of the final phases of CuO and CuAl2O4 phase (stage V):

$${\text{Cu}}_{4} {\text{Al}}_{2} {\text{O}}_{7(s)} \to \, 3{\text{CuO}}_{(s)} + {\text{ CuAl}}_{2} {\text{O}}_{4(s)}$$
(5)

Concerning the water content, x, the process Cu4Al2(OH)12CO3.xH2O → Cu4Al2O7 allowed the calculation of x = 4. Most of the hydrated water loss occurs in the stages I and II.

With the proposed reactions for the stages II, III and IV of the LDH thermal decomposition, it was observed an excellent agreement between the theoretical and the experimental mass losses (25.0 and 25.6%, respectively; Table 1).

Regarding the LDH/Pd composites, the diffraction patterns of LDH/Pd1 and LDH/Pd2 (Fig. 5) show that the LDH matrix was maintained and the (003) peak was kept in the same position, meaning no intercalation process; for the latter sample this was expected since it corresponded to the use of a Pd0 containing precursor that either way would not be intercalated by ion exchange. For both it was observed the presence of Pd0 (111) peak at 39.4° meaning that a Pd reduction step took place for LDH/Pd1. The corresponding FTIR spectra exhibit the same band profile of the LDH matrix, including the presence of carbonate and Al–O bands. This corroborates the results from XRD that there was only a surface impregnation.

Their thermal decomposition profile shows differences in relation to the LDH profile mainly in the stages IV (both) and V (LDH/Pd1): stage IV is related to the final process of dehydration and decarbonation, and in stage V there is an extra mass loss of 3.5% in the 600–800 °C interval related to the formation of molecular chlorine from chloride ions associated with Pd.

$${\text{PdCl}}_{2} \to {\text{ Pd}}_{(\text{s})}^{0} + {\text{ Cl}}_{2(\text{g})}$$
(6)

Therefore, it leads to the conclusion that still some Pd remained unreduced in the LDH/Pd1 sample which makes it differ from the LDH/Pd2 sample. The Pd reduction process during ion exchange in DMF has already been reported by our group in a previous work for Mg/Al LDH [9].

The diffraction pattern of LDH/Pd3 (Fig. 4a) shows a new phase in the material, which could be interpreted as the exchanged form of LDH with formula Cu4Al2(OH)12[PdCl4], with interlayer space of d(003) = 0.546 nm; no Pd0 peak was observed. A decrease in the interlayer distance was also reported in the intercalation process of [PdCl4]2− in Mg/Al LDH by Zhang et al. [12]; however, the reported shift of the (003) peak was not of such high magnitude (from d(003) = 0.759 nm in the LDH 2CuAl.CO3 to 0.546 nm). In Fig. 4a it was noticed that the high intensity (003) peak dislocated from its original position at 2θ = 11.70° for LDH to 11.40° for LDH/Pd3. This could really be showing the intercalation of \(PdCl_{4}^{2 - }\): in this case the peak with d = 0.546 nm would be relative to an unknown new phase, most probably containing Pd. We are researching the identity of this phase to verify this hypothesis. The FTIR spectrum shows that the bands related to carbonate vibration modes have lost much of their intensity, reaffirming the chemical modification of the interlayer space. Its thermogravimetric curve (Fig. 5, green curve) shows significant difference from those of the others LDH signaling that the processes related to the mass losses tend to occur at higher temperatures. This is shown in Table 2: while for the others LDHs significant mass losses occur from 120 °C ahead, for LDH/Pd3 this only happens above 200 °C. The fact that this sample is maintained stable for higher temperatures could probably be related to the intercalation of a non-volatile ion [PdCl4]2−.

Fig. 4
figure 4

a Powder XRD pattern and b FTIR spectra of LDH 2CuAl.CO3 and its composites with Pd. Legend: (open circle) LDH 2CuAl.CO3, (filled diamond) Pd0, (filled circle) Unknown new phase

Full size image
Fig. 5
figure 5

TG curves data from the LDH 2CuAl.CO3 and its composites with Pd

Full size image
Table 2 Mass loss % for the LDH 2CuAl.CO3 and its composites with Pd (argon atmosphere)
Full size table

These results are consistent with the ion exchange process in the interlayer space between \({\text{CO}}_{3}^{2 - }\) and [PdCl4]2−.

Conclusions

The LDH Cu/Al with the formula Cu4Al2(OH)12CO3.4H2O had revealed that its thermal decomposition occurs from 120 to 600 °C, forming after the last step CuO and CuAl2O4, with intermediate phase of mixed oxide. The LDH/Pd1 and LDH/Pd2 presents Pd0 impregnated on the surface, and the LDH/Pd3 presents [PdCl4]2− intercalated in the interlayer space. The thermogravimetric analysis was able to distinguish the mode of supporting palladium between the composites being a reliable characterization for such task.