Supplementary Section S1 of Supporting Information shows the catalytic reactor experimental setup. The calcined mixed oxide spinel catalyst (MgAl₂O₄) was not pretreated. It was stabilized by feeding ethanol at 300°C, atmospheric pressure (0.39 bar of ethanol partial pressure), and LHSV = 0.235 h⁻¹. The total flow of ethanol fed was 0.0106 ml/min to obtain an LHSV = 0.235 h⁻¹, since the catalyst volume is 2.7 ml. In tests at different LHSV values, the total flow of ethanol fed can be easily calculated, taking into account that the catalyst volume is the one mentioned above.

Once the catalyst was stabilized, all the tests were carried out without changing the catalyst sample. Experimental data were obtained operating continuously by averaging the product reaction values of each experiment with a minimum of 10 chromatographic analyses (each one every 38 min).

The ethanol conversion, the product distribution, and the butanol productivity were respectively calculated as follows: Conversion , XEtOH ( % ) = C   mol   of   ethanol   fed − C   mol   of   ethanol   at   reactor   outlet   C   mol   of   ethanol   fed · 100 Carbon   product   selectivity ( % ) = C   mol   of   product   i C   mol   of   total   products   · 100 Butanol   productivity , PButOH ( g kg cat · h ) = Butanol out ( g h ) catalyst   weight ( k g )

The magnesium–aluminum hydrotalcite precursor was used to obtain a mixed oxide MgAl₂O₄ catalyst by thermal decomposition. The appropriate ratio of magnesium/aluminum was used in the precursor synthesis to obtain a stoichiometric spinel (MgAl₂O₄) based on the phase diagram (72 wt% Al₂O₃ and 28 wt% MgO) (Braulio et al., 2011). The hydrotalcite was prepared by co-precipitation method using an aqueous solution of 2 M NaOH and 0.5 M Na₂CO₃, as a precipitation agent. The synthesis took place in an automatically controlled titration system. The solution was added in a constant drip into an aqueous solution of metal nitrate salts, Mg(NO₃)₂·6H₂O and Al (NO₃)₃·9H₂O (molar ratio Mg²⁺/Al³⁺ = 0.5), and the mixture was maintained under continuous stirring at 70°C until reaching pH 10. The final solution was aged 18 h at the same temperature in stirring. After that, the precipitate solution was filtrated and washed with deionized water. Nitrates and carbonates were removed during the calcination process at high temperatures. The filter cake obtained was dried at 100°C overnight and finally calcined at 900°C (10°C/min) for 24 h, thus obtaining the calcined MgAl₂O₄ mixed oxide catalyst. The Mg/Al ratio is selected to achieve an adequate balance acid-basic sites as showcased in the acidity studies vide infra in the Results section. The use of Mg–Al spinel catalyst brings some novelty since the spinel structure is scarcely studied in this reaction compared to standard mixed oxide systems.

Nitrogen adsorption–desorption isotherm was used to analyze the textural properties of the calcined catalyst. The isotherm was realized at liquid nitrogen temperature in a Micromeritic Tristan II apparatus. First, the sample was degassed at 250°C for 4 h in a vacuum. The surface area was determined using the Brunauer–Emmett–Teller (BET) method, and the pore size was calculated using the Barret–Joyner–Halenda (BJH) method.

XRD was conducted to identify the crystalline structure and phases present in the fresh (hydrotalcite precursor), calcined, and spent catalysts. The measurement was carried out on a Siemens D-500 diffractometer using a Ni-filtered Cu kα radiation (40 mA and 45 kV), from 10º to 90° 2θ (0.05° step size and 300 s per step).

NH₃-TPD and CO₂-TPD were performed in homemade equipment using a U-shape quartz tube with 100 mg of sample. In both cases, the samples were pretreated in an inert gas (50 ml/min He) until 200°C for 1 h to remove the water physisorbed and then were cooled down to 50°C. Afterward, 30 ml/min of CO₂ or NH₃/He (5 vol%) was passed through the sample until saturation (30 min). Then the samples were purged with He (30 ml/min) for 1 h. In addition, the temperature was increased from 50 to 900°C in He (30 ml/min) with 10°C/min heating ramp. The effluent gases from the CO₂ or NH₃ desorption were monitored by mass spectrometry (MS PFEIFFER Vacuum Prisma Plus controlled by QUadera^® software). The mass/charge ratios followed in NH₃-TPD were m/z = 16, 17, and 18. Owing the presence of water signals (m/z = 18, 17) in the gas flow, it is necessary to make a signal correction for an accurate ammonia formation profile to deduct m/z = 17 contribution of water. The difference between the total signal and the water m/z = 18, representing 26% of the signal, was considered as a correction factor. For CO₂-TPD, the CO₂ (m/z = 44) signal was monitored.

Carbon deposition and thermal stability were studied by TGA on a TA Instrument SDT Q600 equipment. The experiment covered a range of temperatures from RT to 900°C (rate 10°C/min) under airflow (100 ml/min).

Textural properties are presented in Supplementary Section S3 of Supporting Information, showcasing the mesoporous nature of the synthesized catalyst. As for the structural analysis, Figure 1 presents the XRD pattern of the fresh catalyst (referred to as hydrotalcite precursor) and the calcined and the spent mixed oxide catalyst (MgAl₂O₄). As shown in Figure 1, the hydrotalcite precursor obtained using co-precipitation methods presents a mixture of different phases. The formation of some side phases apart from a pure hydrotalcite structure was associated with the ratio of M²⁺/M³⁺, as suggested by Cavani et al. (1991) in a very early study. It was identified that the reflection planes (003), (006), and (009) of the layered hydrotalcite structure at 2θ 11.6, 23.6, and 35° corresponding with the 3R rhombohedral plane’s reflection ascribed to the magnesium/aluminum hydrotalcite (MgAl₂CO₃(OH)₁₆·4H₂O) (Pérez-Ramírez et al., 2007). Together with hydrotalcite diffractions, β-Al(OH)₃ (bayerite) and AlO(OH) (boehmite) diffraction peaks are observed, and the presence of Mg(OH)₂ phase cannot be discarded owing to overlapping with aluminum species (Dębek et al., 2016; Lee et al., 2016). These species are considered promoters of the double-layer hydrotalcite compound in which the Al and Mg ions are integrated into the layered structure, giving rise to the hydrotalcite formation (Lee et al., 2016). This fact suggested that the hydrotalcite phase is not formed almost completely or it is defective, which could affect the structure of the mixed oxide. The mixed oxide pattern (MgAl₂O₄), obtained after the hydrotalcite calcination at 900°C, presents crystalline and well-defined peaks of the plane reflections (111), (220), (311), (222), (400), (422), (511), and (440) ascribed to magnesium aluminate spinel. Furthermore, we identify small reflections of the MgO phase at = 42.97 and 62.29°. The presence of a MgO phase out of the spinel structure could be associated to the aluminum species observed in the hydrotalcite compound, generating a non-stoichiometric mixed oxide. Further evaluation of the XRD analysis as suggested by Carvalho et al. (2012) would be beneficial. Nevertheless, this is beyond the objectives of this work whose main aim is to cover broad parametric studies of the reaction conditions for a fine catalyst.

For the sake of crystalline structure stability analysis during the reaction, the spent catalyst was also studied. As depicted in Figure 1, there are no significant changes in the X-ray pattern of the sample whose dominant phases remain as MgAl₂O₄ and MgO in fair agreement with its calcined unreacted counterpart. A slight shift, though, to lower 2θ angles was appreciated in comparison to the calcined sample. This could indicate a possible inversion in the spinel, where the Mg²⁺ and Al³⁺ exchange positions from the tetrahedral and octahedral sites. It is important to highlight that we did not observe diffractions ascribed to carbon phases after the reaction. Hence, no crystalline carbon is formed, indicating that catalyst deactivation by carbon poisoning might not be an issue in this material under the studied reaction conditions.

The surface acid–base properties of the mixed oxide catalyst (MgAl₂O₄) were evaluated by NH₃-TPD and CO₂-TPD. Both profiles are shown in Figure 2. NH₃-TPD was carried out to evaluate the strength of acid sites in the catalyst surface. We differentiate two zones associated with the acid strength, depending on the NH₃ temperatures desorption: moderate (200–400°C) and strong (400–800°C) (Madduluri et al., 2020). The peak at a lower temperature, 256°C, is associated with reversible adsorption H-bonded ascribed to hydroxyl groups presence over the surface, considered as BrØnsted acid sites (Song et al., 2021). Alternatively, the peaks at high temperatures (450–510°C) are related to Lewis acid sites, showcasing the interaction with accessible Al⁺³ cations in Al³⁺–O²⁻–Mg²⁺, because of possible defects present over the spinel surface (He et al., 2015). Furthermore, Coleman et al. (2009) associated the Lewis acid sites with the presence of amorphous AlO _y species that provoke an electron-deficient Al³⁺.

CO₂-TPD was performed to study the surface basicity in the mixed oxide (Figure 2). The catalyst exhibits low–moderate basic sites at lower temperatures mainly, although it is possible to distinguish weak desorption at 618°C, indicating strong basic sites formation. Song et al. (2021) ascribe these peaks to the formation of the different carbonate geometries because of reactive CO₂ absorption. The peak at 180°C could be related to the bicarbonate anions associated with the presence of hydroxyl groups; meanwhile, the peak at 371°C is ascribed to bidentate carbonates on acid–base sites (Al³⁺–O²⁻ or Mg²⁺–O²⁻) (Díez et al., 2003). Furthermore, formation of monodentate carbonates with strong basicity (high temperature) associated with the presence of pure MgO phase outside the spinel structure cannot be disregarded. Overall, our catalyst presents both medium and strong acidic sites, with medium strength sites being the dominant ones.