Formulation of Shaped ZSM–5/Al₂O₃ Composites: Comparison of Pseudoboehmite and Nitrate-Derived Aluminum Binders

Abstract

This study focuses on the shaping of ZSM–5-based granules using conventional pseudoboehmite as well as aluminum hydroxynitrate precursors obtained by partial thermal treatment of Al(NO₃)₃·9H₂O. The starting materials and nitrate-derived precursors were characterized by X-ray diffraction, thermal analysis, infrared spectroscopy, scanning electron microscopy, and low-temperature nitrogen adsorption. Model pseudoboehmite systems were used to evaluate the effects of moisture, peptization equivalent, and HNO₃ concentration in the initial liquid portion on pore volume, apparent density, shrinkage, and axial and radial crushing strength. Acid peptization showed a strongly non-linear effect: small HNO₃ additions increased strength and densification, whereas excess acid promoted structural heterogeneity and granule disintegration during thermal treatment. Even at constant moisture and total acid dosage, strength depended markedly on the initial peptizer concentration. In unpeptized ZSM–5/pseudoboehmite composites, increasing zeolite content reduced pore volume but increased density and strength, indicating formation of a more compact framework. Aluminum hydroxynitrate precursors enabled effective shaping of ZSM–5-containing pastes, demonstrating their potential as alternative binder precursors, while the optimized pseudoboehmite systems provided the most favorable overall balance between porosity and mechanical stability.

1. Introduction

Synthetic ZSM–5 zeolites are widely used in acid-base catalytic processes in petroleum refining and petrochemistry due to their high thermal stability, well-developed microporous structure, and tunable acidic properties. In most practical applications, zeolites are used not as powders but as shaped bodies, including granules, extrudates, pellets, and monolithic structures suitable for operation in fixed or moving beds [1,2,3]. Converting a zeolite powder into a shaped body is a distinct materials-engineering challenge since the resulting material must simultaneously possess sufficient mechanical strength, preserve the accessibility of the zeolitic micropores, and contain a well-developed network of transport meso- and macropores [3,4,5].

The shaping of zeolitic materials cannot be reduced to the simple mechanical consolidation of a powder. The addition of a binder inevitably modifies the structure of the shaped body. In addition to the intrinsic micropores of the zeolite phase, secondary porosity develops as a result of the packing of zeolite and binder particles, the distribution of liquid within the shaping paste, shrinkage during drying, and transformations of the binder during thermal treatment. This secondary pore structure largely controls reactant transport and the accessibility of the zeolite crystal surfaces. Therefore, the binder should provide sufficient mechanical strength while minimizing micropore blockage and excessive densification of the interparticle space [3,4].

The most common approach to shaping zeolitic materials involves the addition of an inorganic binder, water, a peptizing agent, and, where required, additional plasticizing or pore-forming additives. In this context, the binder cannot be regarded merely as an inert mechanical component. It determines the plasticity of the shaping paste, the strength of the dried and calcined granules, the total pore volume, the pore size distribution, and the accessibility of the active phase [3,6]. Moreover, interactions between the binder and the zeolite may alter the acidic properties, promote aluminum redistribution, and modify the surface of the shaped material, which is particularly important for synthetic zeolites with a pronounced acid function [3,4,5].

Alumina-based binders occupy a particularly important position among the materials used for shaping zeolitic systems. On the one hand, they provide high mechanical strength of extrudates; on the other hand, γ-Al₂O₃ and its precursors may introduce their own Lewis acid sites and modify the surface properties of the zeolite-containing material [7,8,9]. For ZSM–5-based systems, the nature of the binder, the shaping procedure, and the heat treatment conditions have been shown to affect not only the textural characteristics but also the distribution of acid sites and the stability of the material during subsequent operation [9,10,11]. Consequently, the selection of an alumina-based binder and the conditions governing its transformation during drying and calcination are key factors in the development of shaped zeolite-containing materials.

Pseudoboehmite, AlOOH·nH₂O, is one of the most widely used precursors of alumina binders. Its technological advantages arise from its ability to undergo peptization, form a plastic shaping paste, and transform during thermal treatment to transition aluminas, primarily γ-Al₂O₃, while retaining a high specific surface area. In zeolite-based systems, the conditions of pseudoboehmite peptization, the water content, and the amount of acid strongly affect paste extrudability, mechanical strength, and the pore structure of the final material [10,12]. The origin and phase composition of the alumina precursor are also important: industrial aluminum hydroxides containing different proportions of amorphous phase, boehmite, pseudoboehmite, and other hydroxide phases may exhibit different paste-forming abilities and produce supports with different strength and porosity values [13]. Formulation design is therefore not merely an empirical processing step but an optimization problem requiring a compromise between the structural and mechanical properties of the shaped body and the preservation of a well-developed pore structure [3,10].

The peptizing agent plays a particularly important role in the shaping of these systems. Water alone provides wetting of the powder mixture but does not produce a sufficiently plastic and homogeneous paste. A peptizing agent promotes the dispersion of aggregated hydrous alumina particles, modifies the surface charge, and facilitates the formation of a colloidal structure suitable for shaping. Various mineral and organic acids may be used to peptize pseudoboehmite, as may salts whose hydrolysis produces acidic aqua and hydroxo complexes [14,15]. Nitric acid is nevertheless the most widely used and technologically convenient option since it effectively regulates the plasticity of alumina-based pastes, promotes granule strengthening, and is removed during thermal treatment without introducing residual chloride, phosphate, or sulfate ions that could modify the acidic and textural properties of the material [14,15]. The nature and concentration of the peptizing agent affect the paste viscosity and shear stress. An insufficient amount results in incomplete dispersion, whereas an excess may promote coagulation-induced structuring and gel formation [15]. Acid peptization generally increases granule strength but may also be accompanied by a pore volume decrease [14].

Increasing the zeolite fraction in the shaped material is an important practical objective because the zeolite phase provides the primary functionality of the catalytic system. However, increasing the zeolite content reduces the relative amount of binder and may impair both paste formability and granule strength. Recent studies of shaped γ-Al₂O₃/ZSM–5 composites have shown that the dependence of granule properties on zeolite content is non-linear and that the shaping process itself may affect the properties of the zeolite phase [16]. It is therefore necessary to identify a processing window that provides an acceptable balance between high zeolite content, mechanical strength, and preservation of the pore structure.

Along with pseudoboehmite, aluminum hydroxynitrate compounds with the general composition Al₂O₃·xN₂O₅·yH₂O are considered promising precursors for alumina binders [17,18]. These compounds occupy an intermediate position between soluble aluminum salts and hydrated aluminum oxyhydroxides such as pseudoboehmite. Unlike a mechanical mixture of pseudoboehmite and nitric acid, aluminum hydroxynitrate precursors contain nitrate and hydroxyl groups within a single polymerized system. During subsequent thermal treatment, they undergo successive dehydration, partial denitration, and transformation into aluminum oxyhydroxide or transition alumina phases [17,18]. Hydroxynitrate precursors of different compositions contain different amounts of nitrate and hydroxyl groups, while the composition of the resulting product depends not only on temperature but also on treatment duration, conditions of gas removal, and ambient humidity.

The use of such precursors as binders is of interest for several reasons. First, aluminum hydroxynitrates may provide a more uniform distribution of the alumina component within the shaping system because in this case the alumina binder phase is formed in situ during drying and calcination. Second, decomposition of the nitrate groups and removal of bound water may additionally affect the development of the pore structure [17,18]. Third, variation in the hydroxynitrate precursor preparation conditions may enable control over its composition and degree of dehydration without the addition of other agents. In this respect, comparison of pseudoboehmite and aluminum hydroxynitrates as binders may contribute to the development of more flexible strategies for shaping zeolite-containing catalysts.

Despite the large number of studies devoted to the catalytic properties of powdered ZSM–5 and modified zeolites, the shaping stage itself is often described only briefly. Typically, the binder type, component mass ratio, presence of a peptizing agent, and calcination conditions are reported, whereas the formulation space of the shaping paste is not systematically examined [1,2,3,10]. In studies where shaping parameters are varied, their effects are generally evaluated directly in zeolite-containing compositions, where changes in zeolite fraction, binder content, water demand, and peptization may overlap [10,16]. This complicates the interpretation of the individual factors governing the balance between pore structure and mechanical strength.

Accordingly, the present work applies a sequential formulation strategy. A model pseudoboehmite system was first examined in order to establish the effects of paste moisture, peptization equivalent, and the concentration of HNO₃ during its initial contact with the binder. The identified relationships were transferred to ZSM–5/pseudoboehmite compositions, while nitrate-derived aluminum hydroxynitrates were additionally evaluated as alternative binder precursors. This approach may potentially allow distinguishing binder-related effects from those associated with the introduction of the zeolite phase.

Therefore, this study aimed to develop and compare approaches to the shaping of ZSM–5-containing materials using pseudoboehmite and aluminum hydroxynitrate precursors. The present study intentionally isolates the shaping stage prior to the introduction of metal promoters, allowing structural and mechanical effects to be evaluated without the superposition of promoter-dependent catalytic effects.

2. Results and Discussion

2.1. Characterization and Study of the Starting Materials

2.1.1. Characterization of the ZSM–5 Zeolite and the Pseudoboehmite Binder

The starting components used to prepare the shaped materials, namely pseudoboehmite and ZSM–5 zeolite, were characterized by X-ray diffraction (XRD), and the resulting patterns are shown in Figure 1.

Figure 1. XRD patterns of the starting materials: *—MFI-type ZSM–5 zeolite; ○—pseudoboehmite, AlOOH·nH₂O.

X-ray diffraction analysis revealed substantial differences in the crystalline structures of the starting components (Figure 1). The XRD pattern of ZSM–5 consisted of a series of sharp and intense reflections characteristic of the MFI framework (PDF 85–1208). The most prominent reflections occurred at 2θ values of ~ 8–11° and 26–29°, confirming the crystalline structure of ZSM–5. No crystalline impurity phases were detected by XRD. In contrast, the pseudoboehmite pattern exhibited broad, low-intensity reflections characteristic of poorly crystalline AlOOH·nH₂O (PDF 21–1307). The broadening of the reflections indicates the defective structure of the commercial pseudoboehmite and a relatively small coherent scattering domain size (c.s.d), which, according to calculations, was equal to ~ 6 nm. This poorly ordered hydrated structure is important for shaping because it facilitates acid peptization and the formation of a plastic paste [15].

The results of thermal analysis of the initial pseudoboehmite are presented in Figure 2.

Figure 2. Thermal analysis curves of the initial pseudoboehmite. The red and blue arrows indicate the corresponding ordinate axes for the TGA and DTA curves, respectively.

Thermal analysis confirmed the hydrated state of pseudoboehmite (Figure 2). Mass loss occurred in several stages. In the low-temperature region, a pronounced endothermic effect with a maximum at ~85 °C was accompanied by a rapid decrease in mass. This effect can be attributed to the removal of physically adsorbed and weakly bound water from the interparticle space and surface hydration layers.

Further mass loss over the approximate temperature range of 150–450 °C was more gradual and was associated with the removal of more strongly bound water and the progressive dehydroxylation of pseudoboehmite. This process corresponded to a broad endothermic effect with a maximum at ~450 °C in the DTA curve, reflecting the transformation of AlOOH·nH₂O into transition alumina phases through the removal of structural hydroxyl groups [19,20]. The total mass loss by 750 °C was 24.37 wt.%, in good agreement with the loss on ignition of 24.8 wt.% determined by calcining the initial pseudoboehmite in a muffle furnace for 4 h.

The presence of a substantial amount of bound water is particularly important for shaping because pseudoboehmite acts not only as a solid alumina-containing component but also as a hydrated aluminum oxyhydroxide precursor capable of acid peptization and subsequent consolidation during drying and calcination [13,14,15].

The low-temperature nitrogen adsorption–desorption isotherms of the starting components are shown in Figure 3.

Figure 3. Nitrogen adsorption–desorption isotherms of the starting materials: the left axis corresponds to ZSM–5 zeolite, and the right axis corresponds to pseudoboehmite. The arrows indicate the corresponding ordinate axes.

Low-temperature nitrogen adsorption revealed substantial differences between the pore structures of the starting components (Figure 3). The ZSM–5 isotherm exhibited a sharp increase in nitrogen uptake at low relative pressures, P/P_S, corresponding to the filling of the zeolitic micropores. After this initial region, adsorption increased more gradually, while the hysteresis loop was negligible, indicating predominantly microporous character with only a limited contribution from mesopores and intercrystalline voids.

Pseudoboehmite exhibited a different isotherm shape (Figure 3). Nitrogen adsorption increased moderately at low relative pressures, whereas a sharp increase in adsorbed volume and a pronounced hysteresis loop were observed at high P/P_S. This behavior indicates a developed meso- and macropore structure associated with the packing of pseudoboehmite particle aggregates.

Unlike zeolite, the porosity of pseudoboehmite is not determined by a microporous crystalline framework but by voids between particles and aggregates of hydrated aluminum oxyhydroxide [15,20]. Therefore, the final texture of shaped ZSM–5-containing granules is expected to be governed by the combined contributions of zeolite microporosity, the interparticle porosity of the binder, and the degree of consolidation of the alumina framework during drying and calcination.

The SEM images of the starting powders shown in Figure 4 also reveal pronounced morphological differences between the components.

Figure 4. SEM micrographs of ZSM–5 zeolite (a,b) and pseudoboehmite (c,d), magnifications 10,000× (a,c) and 3000× (b,d).

Pseudoboehmite consisted of aggregates of fine particles that tended to form loose secondary agglomerates (Figure 4). This morphology is favorable for peptization because acid addition partially disperses the aggregates, thereby imparting plasticity to the shaping paste. In contrast, ZSM–5 consisted of more faceted crystalline particles and did not independently form a plastic binding phase under the investigated conditions.

2.1.2. Selection of Conditions for Preparing the Nitrate-Derived Aluminum Precursors

Along with pseudoboehmite, the use of a nitrate-derived aluminum precursor obtained by partial thermal conversion of aluminum nitrate nonahydrate was investigated. This approach was selected because moderate heating of Al(NO₃)₃·9H₂O results not only in the removal of crystallization water but also in the gradual hydrolysis of nitrate complexes and the formation of basic aluminum salts [17,18]. Upon subsequent calcination, these products can transform into amorphous Al₂O₃ and, at higher temperatures, into finely dispersed transition aluminas, including γ-Al₂O₃ and θ-Al₂O₃ [19,20,21], thereby forming an alumina binder phase directly within the interparticle space of the granule.

Published data on the chemical composition of the products obtained by thermal treatment of Al(NO₃)₃·9H₂O were analyzed to select an appropriate temperature range. Drobot et al. reported the compositions of products formed after treating this salt at different temperatures, expressed as Al₂O₃·xN₂O₅·yH₂O [21]. For the present study, these tabulated data were recalculated as the N₂O₅/Al₂O₃ and H₂O/Al₂O₃ molar ratios and plotted in the Al₂O₃–N₂O₅–H₂O ternary system (Figure 5). This graphical reconstruction of the literature data was conducted to identify a rational temperature range for the subsequent experiments.

Figure 5. Changes in the composition of Al(NO₃)₃·9H₂O thermal-treatment products after heating to constant mass, reconstructed from published tabulated data [21]: (a) N₂O₅/Al₂O₃ and H₂O/Al₂O₃ molar ratios; (b) composition trajectory in the Al₂O₃–N₂O₅–H₂O system during thermal treatment.

The plotted data clearly show that the most substantial transformation of aluminum nitrate occurs within the temperature range of 100–200 °C. Both ratios decrease sharply in this region, indicating the intensive decomposition of hydrated nitrate species and the formation of basic aluminum salts with variable compositions. This behavior is consistent with a stepwise transformation mechanism in which dehydration is followed by hydrolytic decomposition of the nitrate salt, accompanied by the formation of basic salts and amorphous alumina [17,21].

Based on this analysis, the temperature range of 100–160 °C was selected for further investigation since the most pronounced compositional changes associated with dehydration and denitration of the initial aluminum nitrate occurred within this interval (Figure 5). To compare the rates of thermal conversion of Al(NO₃)₃·9H₂O, sample mass changes were monitored during treatment in air at 100, 125, 150, and 160 °C. The treatment time was varied depending on the temperature. The results of the gravimetric measurements are presented in Figure 6.

Figure 6. Thermal decomposition kinetics of Al(NO₃)₃·9H₂O: (a) relative mass-loss curves at 100, 125, 150, and 160 °C; (b) Arrhenius plot for the observed pseudo-first-order rate constant.

Over the entire temperature range studied, cumulative mass loss increased with thermal treatment time, whereas both the rate and extent of mass loss depended strongly on temperature (Figure 6a).

The key differences between the curves were observed during the initial stage of treatment. At 100 and 125 °C, the relative mass loss increased almost uniformly, whereas at 150 and 160 °C, a substantial fraction of the volatile components was removed within the first 1–2 h. The process subsequently slowed, which may be attributed both to the decreasing contents of water and nitrate groups and to the formation of a denser intermediate product that hindered the transport of gaseous products from the bulk material.

The recorded mass loss reflects the combined contribution of several processes. At relatively low temperatures, the principal contribution was most likely associated with the removal of crystallization and coordinatively bound water. As the temperature increased, decomposition of the nitrate complexes accompanied by partial denitration became increasingly significant [17,19].

For a quantitative comparison of the temperature dependence, the experimental curves were described using a pseudo-first-order kinetic model. The observed rate constant, k_app (min^–1), was calculated using Equation (1):

$$k_{\mathrm{a}\mathrm{p}\mathrm{p}}={\displaystyle \frac{1}{60\mathsf{\tau }}}\ln \left({\displaystyle \frac{100}{100-X}}\right),$$

(1)

where τ is the thermal-treatment time (h), and X is the relative mass loss of the material (wt.%).

The calculated k_app values increased with temperature and were satisfactorily described by the Arrhenius equation (Figure 6b). The apparent activation energy of the overall decomposition process was calculated to be 62.4 kJ/mol. This value should be regarded as an apparent parameter of the coupled dehydration and denitration processes. The satisfactory description of the curves by the same model and the linear Arrhenius dependence suggest that no pronounced change in the dominant kinetic regime occurred within the investigated range.

The kinetic data show that increasing the temperature from 100 to 160 °C substantially accelerated the formation of the aluminum hydroxynitrate precursors. The treatment time used for preparing the samples for further investigation was therefore fixed at 4 h. This made it possible to compare products obtained at different temperatures under the same treatment duration and to avoid combining the effects of temperature and processing time. The phase composition and structure of aluminum nitrate and its thermal conversion products were examined by XRD. The corresponding diffraction patterns are shown in Figure 7.

Figure 7. XRD patterns of aluminum nitrate decomposition products obtained by thermal treatment at 100, 125, 150, 160, and 200 °C for 4 h.

The XRD patterns of thermally treated Al(NO₃)₃·9H₂O confirmed substantial changes in the phase state of the salt within the temperature range of 100–160 °C (Figure 7). After treatment at 100 °C, the diffraction pattern retained the reflections of the initial aluminum nitrate nonahydrate phase (PDF 12–472). This indicates that the mass loss observed at this temperature was primarily associated with partial water removal and was not accompanied by complete destruction of the crystalline structure of the initial salt.

When the treatment temperature was increased to 125 °C, the reflections of the initial aluminum nitrate almost completely disappeared, and the diffraction pattern became predominantly diffuse, with weak reflections attributable to a newly formed phase. The same reflections were observed for the samples treated at 150 and 160 °C, where they became better resolved, indicating progressive crystallization of this phase with increasing treatment temperature. The positions of these reflections agree with those reported by Guilliatt and Brett for a high-nitrate dispersible alumina gel (PDF 31–28) [22]. Those authors were unable to assign the reflections to any known aluminum hydroxide or oxide but suggested that they could be associated with basic hydroxy salts and a layered type of structural ordering. In a later study, Shi et al. observed the same set of reflections after the thermal treatment of aluminum nitrate at 160 °C [18]. Based on XRD, FTIR, and gravimetric data, they assigned the product to a basic aluminum nitrate salt with the composition Al(NO₃)₃·3Al(OH)₃·2.5H₂O.

After treatment at 200 °C, the reflections attributed to this hydroxynitrate phase disappeared, with the diffraction pattern dominated by a broad amorphous halo. This indicates that the decomposition of the intermediate hydroxynitrate crystalline phase, suggesting that this is restricted to a relatively narrow temperature interval. Further heating leads to the formation of amorphous and transition alumina products [20].

The XRD results indicate that increasing the temperature from 125 to 160 °C did not produce additional crystalline phases but promoted greater structural ordering and an increase in the relative contribution of the crystalline Al_x(OH)_3x−1NO₃ phase. The higher reflection intensities observed for AHN(150) and AHN(160) are consistent with the extensive dehydration and structural rearrangement indicated by the gravimetric measurements (Figure 6). No reflections characteristic of pseudoboehmite, AlOOH·nH₂O, or transition alumina phases were detected in the investigated samples. Therefore, after 4 h of treatment at temperatures up to 150–160 °C, the transformation was limited to dehydration and the formation of aluminum hydroxynitrate products, without reaching the extensive denitration required for the formation of crystalline aluminum hydroxide or oxide phases detectable by XRD.

These results demonstrate that the precursor treatment temperature determines not only the amount of retained water and nitrate groups but also the degree of structural ordering of the binder precursor. To further clarify the structural changes and confirm the simultaneous presence of nitrate, hydroxyl, and Al–O groups, the hydroxynitrate-containing products were additionally analyzed by FTIR spectroscopy. The resulting spectra are presented in Figure 8.

Figure 8. FTIR spectra of aluminum nitrate decomposition products obtained at 125, 150, and 160 °C for 4 h.

The FTIR spectra shown in Figure 8 demonstrate that the products obtained within the investigated temperature range retained both hydroxyl and nitrate components. All spectra contained a broad absorption band in the range of 3600–3000 cm^–1, assigned to ν(OH) vibrations, and a band at 1600–1640 cm^–1 associated with δ(H₂O) vibrations. The intense band at 1380–1400 cm^–1 was assigned to the asymmetric stretching vibrations of nitrate groups, ν(NO₃). Bands and shoulders near 1040, 830, and 600 cm^–1 may be associated with vibrations of Al–OH groups and the Al–O framework containing aluminum atoms in different coordination environments [18].

Increasing the thermal-treatment temperature led to a progressive decrease in the area and intensity of the broad ν(OH) band, together with attenuation of the δ(H₂O) band. The shape of the broad O–H band also changed, indicating a decrease in the contents of water and hydroxyl groups and a redistribution of hydroxyl species involved in hydrogen bonds of different strengths.

AHN(125) exhibited the most intense absorption in the ν(OH) region, in agreement with its poorly ordered state indicated by XRD. The nitrate band of this sample was comparatively broad, probably because of the heterogeneous local environments of the NO₃⁻ ions. When the treatment temperature was increased to 150 °C, the ν(NO₃) band became narrower and more intense, while the bands in the low-wavenumber region became better resolved, consistent with structural ordering of the product.

Treatment at 160 °C resulted in a decrease in the intensity of both the O–H-related bands and the principal nitrate band, most likely because of further dehydration and partial denitration of the basic aluminum nitrate. Nevertheless, the persistence of the ν(NO₃) band confirms that AHN(160) retained a hydroxynitrate composition and did not completely transform into AlOOH or Al₂O₃.

Consistent with these structural changes, the approximate compositions estimated from combined EDS and gravimetric data changed from Al₂O₃·2.10N₂O₅·11.13H₂O for AHN(125) to Al₂O₃·0.24N₂O₅·1.49H₂O for AHN(160), confirming pronounced dehydration and denitration with increasing treatment temperature.

Thus, the XRD and FTIR results allow several distinct states of the nitrate-derived aluminum precursor to be identified:

• At 100 °C, partially dehydrated initial aluminum nitrate;
• At 125 °C, a poorly ordered aluminum hydroxynitrate product;
• At 150–160 °C, a more crystalline aluminum hydroxynitrate phase;
• At 200 °C, an amorphous product formed after loss of the ordered hydroxynitrate structure.

These differences may affect precursor dispersion within the paste, the nature of interparticle bonding, shrinkage, and the development of the pore structure during subsequent drying and calcination.

2.2. Effect of Shaping Conditions on the Properties of Extrudates

2.2.1. Effect of Shaping Conditions on the Properties of Pseudoboehmite Extrudates

To establish the role of the alumina binder, the shaping process was first investigated using a model system containing no zeolite. This approach enabled the independent evaluation of the effects of water and nitric acid on the properties of pseudoboehmite shaping pastes while excluding the contribution of the non-plastic crystalline ZSM–5 phase. A total of 25 formulations were prepared, each containing 20 g of the initial pseudoboehmite and differing in the amounts of added water and nitric acid and in the procedure used to introduce the acid solution during the initial mixing stage.

For selected samples, the acid solution was diluted with water to a predetermined initial volume and gradually added to the dry pseudoboehmite, while the remaining water was introduced after the acid had been distributed throughout the powder. In particular, samples PB-(6–9) contained the same total amount of HNO₃ but differed in the acid concentration in the initial liquid portion. This series was therefore used to evaluate the effect of the local peptizer concentration during its initial contact with pseudoboehmite. The detailed starting formulations are provided in Table S1.

The initial formulations were recalculated in terms of three basic processing factors:

• Paste moisture content, φ (wt.%);
• Peptization equivalent, PE (mol HNO₃/mol Al₂O₃);
• HNO₃ concentration in the initial liquid portion, $C_{\mathrm{H}\mathrm{N}{\mathrm{O}}_3}$ (wt.%).

The corresponding calculation procedures are given in Appendix A. The initial acid concentration was considered only for samples PB-(6–9). The calculated processing factors and the resulting granule properties are summarized in Table 1.

Table 1. Processing factors of the shaping pastes and properties of the pseudoboehmite extrudates.

Sample	φ (wt.%)	PE (mol/mol)	${\mathit{C}}_{\mathbf{H}\mathbf{N}{\mathbf{O}}_3}$ (wt.%)	δ (g/cm3)	${\mathit{V}}_{\mathsf{\Sigma }}$ (cm3/g)	${\mathit{P}}_{\square }$ (MPa)	${\mathit{P}}_{○}$ (MPa)	L (%)
PB–1	42.5	0	–	0.71	0.78	17.3	5.1	12.1
PB–2	43.5	0	–	0.71	0.79	20.1	4.0	12.1
PB–3	44.8	0	–	0.72	0.82	16.7	4.5	12.3
PB–4	45.9	0	–	0.69	0.86	15.1	3.7	13.2
PB–5	46.8	0	–	0.74	0.78	15.8	4.6	12.4
PB–6	39.4	0.19	23.2	1.17	0.39	81.8	19.1	29.6
PB–7	39.4	0.19	14.1	1.07	0.34	49.0	5.6	26.3
PB–8	39.4	0.19	37.8	1.21	0.36	16.1	9.8	25.9
PB–9	39.4	0.19	65.0	1.13	0.41	8.3	3.2	24.2
PB–10	40.9	0.19	–	1.18	0.42	63.6	22.6	25.6
PB–11	38.2	0.19	–	1.08	0.40	29.3	21.9	24.6
PB–12	41.8	0.19	–	1.17	0.45	47.5	16.7	27.1
PB–13	40.9	0.19	–	1.32	0.43	58.3	7.6	29.6
PB–14	38.2	0.19	–	1.11	0.42	20.9	5.6	26.4
PB–15	36.9	0.19	–	1.10	0.43	12.1	4.4	25.5
PB–16	38.0	0.10	–	0.99	0.43	43.1	17.1	23.7
PB–17	40.1	0.10	–	1.02	0.44	86.3	16.0	23.4
PB–18	42.6	0.10	–	0.97	0.46	55.6	23.4	23.0
PB–19	34.5	0.29	–	1.02	0.41	–	–	28.9
PB–20	36.5	0.29	–	1.09	0.40	–	–	30.7
PB–21	38.1	0.29	–	1.06	0.39	–	–	29.5
PB–22	37.6	0.05	–	1.03	0.52	53.0	23.5	22.8
PB–23	39.3	0.05	–	0.93	0.61	45.5	16.9	19.7
PB–24	41.1	0.05	–	1.07	0.53	21.5	12.5	20.7
PB–25	41.1	0.05	–	0.99	0.56	29.2	12.7	19.6

Note: δ, apparent density; $V_{\mathsf{\Sigma }}$, total pore volume; $P_{\square }$, axial crushing strength; $P_{○}$, radial crushing strength; L, relative shrinkage. A dash in the $C_{\mathrm{H}\mathrm{N}{\mathrm{O}}_3}$ column indicates that this parameter was not considered for the corresponding sample. Dashes in the $P_{\square }$ and $P_{○}$ columns indicate disintegration of the corresponding samples during drying or thermal treatment.

Pairwise correlation analysis was performed as a preliminary assessment of the relationships between the processing factors and the granule properties. The Pearson correlation matrix is provided in Table S2.

The correlation analysis indicated that increasing the peptization equivalent was associated with a decrease in total pore volume and an increase in apparent density and shrinkage, whereas increasing paste moisture generally produced the opposite effect. The relationship between peptization equivalent and mechanical strength was less straightforward, indicating the combined effects of moisture content, degree of peptization, and acid addition procedure, as well as the possible non-linearity of these relationships.

The relationships between the peptization equivalent and the pore structure characteristics, including pore volume, apparent density, and shrinkage, are shown in Figure 9.

Figure 9. Effect of peptization equivalent on the properties of pseudoboehmite extrudates: (a) total pore volume; and (b) apparent density and shrinkage. The arrows indicate the corresponding ordinate axes.

The effect of peptization on the structural characteristics of the granules was strongly non-linear (Figure 9). The most pronounced changes were already observed upon transition from the unpeptized paste to the lowest investigated value of PE = 0.05 mol/mol. The mean pore volume decreased from ~0.80 to 0.56 cm³/g, whereas the apparent density and shrinkage increased markedly. When PE was further increased to 0.10–0.19 mol/mol, the pore volume decreased to ~ 0.40–0.45 cm³/g and subsequently approached a plateau. Apparent density and shrinkage changed in the opposite direction, confirming consolidation of pseudoboehmite particles due to acid peptization.

These changes may be attributed to acid-induced dispersion of pseudoboehmite aggregates and the formation of a more homogeneous plastic paste [15]. During drying and calcination, the system undergoes more extensive particle rearrangement and consolidation, resulting in a decrease in interaggregate pore space and an increase in shrinkage. Under excessive peptization, no substantial further decrease in pore volume was observed, whereas shrinkage continued to increase. This indicates that excess acid intensified granule deformation and densification without producing a change in pore volume.

The relationship between moisture content and axial crushing strength is shown in Figure 10.

Figure 10. Effect of paste moisture content at different PE values on the axial crushing strength of pseudoboehmite extrudates.

Granule crushing strength as a function of paste moisture varied considerably with the peptization equivalent (Figure 10). For the unpeptized PB–(1–5) series, axial crushing strength remained comparatively low, at ~15–20 MPa, irrespective of moisture content. Thus, although water provided sufficient pseudoboehmite formability, it did not promote the development of strong interparticle bonding in the calcined material in the absence of acid peptization.

The addition of nitric acid substantially increased granule strength, although the dependence on moisture content became non-linear. At PE = 0.05 mol/mol, the highest axial crushing strength, 53 MPa, was obtained at the lowest investigated moisture content of 37.6 wt.%, corresponding to the lower boundary of the studied moisture range. Increasing the moisture content to 39.3 wt.% decreased the strength to 45 MPa, while at 41.1 wt.% it declined to 21–29 MPa. Therefore, in weakly peptized pastes, excess water reduced mechanical strength, probably because it decreased the solids concentration and weakened the contacts between particles after removal of the liquid phase.

At PE = 0.10 mol/mol, a pronounced maximum was observed: axial crushing strength increased from 43 to 86 MPa with increasing moisture content and subsequently decreased to 56 MPa. A similar trend was observed at PE = 0.19 mol/mol, for which the strength increased from 12 to 82 MPa and then declined to 48 MPa. These results indicate that moderately peptized systems require an optimum moisture content sufficient to distribute the peptized phase uniformly and form a dense binder framework without excessively diluting the paste. For samples prepared at PE = 0.10 mol/mol, the rational moisture range was ~38–40 wt.%, whereas at PE = 0.19 mol/mol it was ~40–41 wt.%.

Notably, a further increase in PE to 0.29 mol/mol did not improve the mechanical strength. Samples PB–(19–21) retained their integrity immediately after shaping but disintegrated during subsequent drying and thermal treatment. This behavior was most likely caused by the formation of a heterogeneous gel-like structure under excessive peptization, which experienced substantial shrinkage stresses during water removal [15]. Decomposition of residual nitrate groups accompanied by the release of gaseous products during drying and calcination may have additionally promoted crack development in defective regions of the granules.

To separately evaluate the effect of the peptizer addition procedure, the mechanical and textural properties of samples PB–(6–9) were analyzed. These samples had identical peptization equivalents, PE = 0.194 mol/mol, and paste moisture contents, φ = 39.4 wt.%, but differed in the degree of dilution of the acid in the initial liquid portion. Therefore, the key variable within this series was the HNO₃ concentration during the initial contact of the solution with pseudoboehmite. The resulting relationships are presented in Figure 11.

Figure 11. Effect of HNO₃ concentration in the initial liquid portion at constant φ = 39.4 wt.% and PE = 0.19 mol/mol on the properties of pseudoboehmite extrudates: (a) pore volume and apparent density; and (b) crushing strength.

The effect of nitric acid concentration in the initial liquid portion exhibited a pronounced maximum (Figure 11). Increasing the concentration from 14 to 23 wt.% sharply increased both strength characteristics: the axial crushing strength reached ~82 MPa, while the radial crushing strength reached ~ 19 MPa. Further increasing the HNO₃ concentration to 38–65 wt.% caused both strength values to decrease sharply.

The consistent changes in $P_{\square }$ and $P_{○}$ indicate that the observed maximum was not merely associated with the selected loading geometry or granule dimensions but reflected an actual change in the structure of the shaped material. At the same time, the total pore volume and apparent density changed to a considerably smaller extent. Thus, the pronounced changes in mechanical strength cannot be explained solely by overall granule densification but were probably governed by the uniformity of the peptized phase distribution, the quality of interparticle contacts, and the concentration of local defects, which are not reflected by integral pore volume and density values.

Consequently, even at constant total acid content and paste moisture, the mechanical properties of the granules depended strongly on the peptizer concentration during its initial interaction with the dry material.

When the initial liquid portion was excessively diluted, the local acidity was probably insufficient for rapid and uniform dispersion of pseudoboehmite aggregates. Conversely, the introduction of an overly concentrated solution caused local over-peptization, the formation of heterogeneous gel-like regions, and non-uniform distribution of the binder phase. During subsequent drying and calcination, this heterogeneity could generate internal stresses and weaken individual interparticle contacts.

The highest mechanical strength was obtained at an intermediate HNO₃ concentration in the initial liquid portion of ~20–25 wt.%. This range apparently provided a balance between the acidity required for pseudoboehmite dispersion and the uniform distribution of the peptizing agent throughout the shaping paste. However, owing to the limited number of concentrations investigated, this maximum should be regarded as a rational operating range rather than a precisely established optimum.

Overall, the investigation of model pseudoboehmite shaping pastes demonstrated that their structural and mechanical properties were governed by the combined effects of water content, degree of peptization, and nitric acid concentration during the initial contact stage. The results indicate that pseudoboehmite-based formulations should be defined by a rational processing window rather than by a single optimum composition. Within the investigated range, this window corresponded to PE = 0.05–0.10 mol/mol, depending on the required pore volume, a paste moisture content of 38–41 wt.%, and controlled addition of HNO₃ at a moderate concentration.

2.2.2. Effect of Shaping Conditions on the Properties of ZSM–5/Pseudoboehmite Extrudates

In order to evaluate the effect of introducing the zeolite component on granule properties and shaping behavior, ZSM–5/pseudoboehmite formulations were prepared following the approach used for the pseudoboehmite-based samples. Each formulation contained a total of 20 g of dry solids and differed in the zeolite-to-binder ratio and in the amounts of added water and nitric acid. The detailed starting formulations are provided in Table S3.

The formulations were recalculated in terms of the overall Si/Al atomic ratio, Si/Al (mol/mol), paste moisture content, φ (wt.%), and peptization equivalent, PE (mol/mol). The corresponding calculation procedures are described in Appendix A. The calculated processing factors and the properties of the resulting ZSM–5/pseudoboehmite granules are summarized in Table 2.

Table 2. Processing factors of the shaping pastes and properties of the ZSM–5/pseudoboehmite extrudates.

Sample	Si/Al (mol/mol)	φ (wt.%)	PE (mol/mol)	δ (g/cm3)	${\mathit{V}}_{\mathsf{\Sigma }}$ (cm3/g)	${\mathit{P}}_{\square }$ (MPa)	${\mathit{P}}_{○}$ (MPa)	L (%)
PB–26	0.13	39.9	0	0.90	0.68	26.1	5.0	13.7
PB–27	0.13	42.9	0	0.79	0.72	25.4	5.6	13.9
PB–28	0.13	41.9	0	0.81	0.68	28.1	2.8	14.6
PB–29	0.26	38.5	0	0.94	0.59	33.5	3.9	16.0
PB–30	0.26	28.4	0.13	1.30	0.38	55.1	21.0	20.8
PB–31	0.51	31.4	0.09	1.07	0.43	40.6	11.1	16.9
PB–32	0.39	32.7	0.08	1.01	0.42	36.8	5.3	13.7
PB–33	0.51	33.4	0.19	1.09	0.42	39.8	14.9	14.1
PB–34	0.60	34.4	0.12	1.04	0.48	36.7	14.3	11.4
PB–35	0.69	34.6	0.16	0.96	0.52	34.7	4.7	11.6
PB–36	0.69	35.1	0	1.06	0.46	42.0	7.3	21.4

Note: δ, apparent density; $V_{\mathsf{\Sigma }}$, total pore volume; $P_{\square }$, axial crushing strength; $P_{○}$, radial crushing strength; L, relative shrinkage.

As in the pseudoboehmite model system, pairwise correlation analysis was performed to assess the relationships between the processing factors and granule properties. The Pearson correlation matrix is provided in Table S4.

To separately evaluate the contribution of the zeolite phase, ZSM–5/pseudoboehmite composites prepared without nitric acid were initially considered. This enabled samples PB–(26–29) and PB-36 to be compared with the unpeptized pseudoboehmite granules PB–(1–5). The effects of the overall Si/Al on the textural and mechanical properties are presented in Figure 12.

Figure 12. Effect of the overall Si/Al on the properties of ZSM–5/pseudoboehmite extrudates shaped without a peptizing agent (PE = 0): (a) pore volume and apparent density; and (b) crushing strength.

An increase in Si/Al, corresponding to a higher ZSM–5 fraction in the shaping paste, was accompanied by a decrease in total pore volume from ~0.78–0.86 to 0.46 cm³/g and an increase in density from ~0.69–0.74 to 1.06 g/cm³ (Figure 12a). This change in the textural properties may be associated with the mixed packing of the two types of particles. Unpeptized pseudoboehmite forms a comparatively loose system of aggregates with a developed interaggregate pore space. The larger and more rigid crystalline ZSM–5 particles may occupy some of these voids and form a more compact framework. The decrease in V_Σ and increase in apparent density supports this interpretation.

Both mechanical strength characteristics increased simultaneously with structural densification (Figure 12b). The axial crushing strength increased from ~15–20 to 42 MPa, while the radial crushing strength increased from ~4–5 to 7 MPa. This increase in strength despite the decreasing proportion of the pseudoboehmite binder was unexpected. It can be assumed that increasing the zeolite fraction and reducing the pore space increased the number of effective interparticle contacts and improved the organization of the resulting composite framework, with the rigid ZSM–5 particles acting as load-bearing inclusions and the remaining pseudoboehmite occupying interparticle contact regions and, after calcination, forming alumina bridges between adjacent particles. A larger number of effective solid–solid contacts together with a lower fraction of large interaggregate voids can therefore improve stress transfer through the composite even as the nominal binder fraction decreases.

Thus, although increasing the ZSM–5 fraction generally impaired paste formability and required greater force during extrusion, it did not reduce the mechanical strength of the unpeptized composites. Instead, it promoted the formation of a denser structure and improved their structural and mechanical characteristics. Nevertheless, granules prepared from the unpeptized pastes exhibited limited mechanical strength, which prompted further investigation of the effects of peptization and paste moisture on the properties of ZSM–5/pseudoboehmite composites.

Increasing the paste moisture content was accompanied by an increase in pore volume and a decrease in apparent density (Figure 13a).

Figure 13. Effect of paste moisture content on the properties of ZSM–5/pseudoboehmite extrudates: (a) total pore volume and apparent density; and (b) crushing strength. Circles and diamonds correspond to the quantities indicated in the respective panel legends. Filled symbols correspond to compositions prepared without peptization, while open symbols correspond to peptized compositions.

The observed relationships were consistent with the correlation-analysis results and showed that the amount of liquid phase was one of the principal factors determining the packing of solid particles in the ZSM–5/pseudoboehmite composites. Removal of a larger amount of water during drying produced a more developed interparticle pore space, while decreasing the density of the final material.

The general dependence of the mechanical characteristics on moisture content showed the opposite trend: granule strength predominantly decreased with increasing φ (Figure 13b). This tendency was most pronounced for the axial crushing strength, $P_{\square }$. Increasing the water content reduced the solids concentration and probably decreased the number of effective contacts formed between the zeolite particles and alumina binder during drying and calcination. Thus, the increase in pore volume was accompanied by weakening of the composite framework.

A considerably greater scatter around the general trend was observed for the radial crushing strength. The points corresponding to PB–33 and PB–34 in Figure 13b, whose deviations from the trend are indicated by black arrows, show that increasing the amount of nitric acid at similar moisture contents substantially increased $P_{○}$, thereby reducing the strength anisotropy of the resulting granules. Acid peptization therefore partially compensated for the adverse effect of moisture by promoting the formation of stronger contacts involving the pseudoboehmite binder.

Radial crushing of cylindrical extrudates is known to be particularly sensitive to macroscopic defects and structural heterogeneity because this loading geometry may produce brittle fracture modes initiated by stress concentrations around pre-existing defects [23]. The increase in $P_{○}$ resulting from peptization may therefore indicate the formation of a more homogeneous and better-connected interparticle framework. However, the low $P_{○}$ value of PB–35, which was also prepared with a relatively high nitric acid content, demonstrates that the effect of the peptizing agent was not independently monotonic but was governed by the combined effects of moisture content, formulation composition, and binder phase distribution.

Acid peptization may increase mechanical strength through two complementary effects: increasing the number and strength of effective interparticle contacts and homogenizing the distribution of the binder within the extrudate. The latter effect is particularly important under radial loading. Reduction in a relatively small population of critical defects can thus produce a disproportionately large increase in radial crushing strength without a comparable change in total pore volume or apparent density.

Overall, the paste moisture content primarily controlled particle packing density and pore volume, whereas mechanical strength depended on the combined influence of moisture and the other formulation variables. Increasing the water content loosened the structure and generally decreased the mechanical strength. Rational peptization could nevertheless improve granule strength and, in particular, reduce strength anisotropy. These findings demonstrate the need to simultaneously control paste moisture and HNO₃ content when shaping zeolite-containing composites.

2.2.3. Effect of Shaping Conditions on the Properties of ZSM–5/Hydroxynitrate Extrudates

As alternatives to pseudoboehmite, the nitrate-derived aluminum precursors AHN(125) and AHN(160), obtained by thermal treatment of Al(NO₃)₃·9H₂O at 125 and 160 °C, respectively, for 4 h, were investigated as binders. As discussed above, these temperatures resulted in different degrees of conversion of the initial salt. AHN(125) was predominantly a poorly ordered and highly hydrated product, whereas AHN(160) contained a more crystalline aluminum hydroxynitrate phase.

To evaluate the applicability of these precursors as aluminum-containing binders, shaping pastes with different ZSM–5/AHN ratios and water contents were prepared. No additional nitric acid was introduced in order to evaluate whether the AHN precursors could provide sufficient shaping behavior under the selected conditions without an externally added peptizer.

The investigation of the ZSM–5/AHN compositions was conducted as a screening study aimed at assessing the general feasibility of nitrate-derived aluminum precursors as binders and identifying the effect of their initial state on the properties of the shaped granules. The detailed paste formulations and AHN treatment temperatures are provided in Table S5.

The formulations were recalculated in terms of the overall Si/Al atomic ratio, Si/Al (mol/mol), and paste moisture content, φ (wt.%). The calculations accounted for the different mass losses and estimated water contents of AHN(125) and AHN(160). The corresponding calculation procedures are described in Appendix A. The processing factors and the properties of the resulting ZSM–5/AHN extrudates are summarized in Table 3.

Table 3. Processing factors of the shaping pastes and properties of the ZSM–5/AHN extrudates.

Sample	Si/Al (mol/mol)	φ (wt.%)	${\mathit{T}}_{\mathbf{A}\mathbf{H}\mathbf{N}}$ (°C)	δ (g/cm3)	${\mathit{V}}_{\mathsf{\Sigma }}$ (cm3/g)	${\mathit{P}}_{\square }$ (MPa)	${\mathit{P}}_{○}$ (MPa)	L (%)
AHN–1	0.54	33.6	160	0.99	0.40	24.6	1.7	17.7
AHN–2	0.54	35.9	160	0.95	0.43	22.9	4.7	21.1
AHN–3	0.76	34.9	160	0.91	0.44	15.1	4.0	21.6
AHN–4	0.65	35.5	160	0.94	0.43	35.6	5.3	20.6
AHN–5	0.65	35.5	160	0.95	0.43	25.0	6.4	24.2
AHN–6	0.76	37.3	160	0.84	0.48	7.6	3.3	21.3
AHN–7	0.69	33.8	160	0.81	0.45	15.4	6.0	23.7
AHN–8	0.69	36.2	160	0.89	0.42	26.4	3.7	14.3
AHN–9	0.42	34.4	160	0.92	0.46	19.6	3.3	17.1
AHN–10	0.42	37.0	160	1.04	0.45	19.1	4.3	21.9
AHN–11	0.54	43.7	125	1.06	0.38	19.0	2.7	21.2
AHN–12	0.54	44.3	125	1.02	0.37	21.0	3.8	19.4
AHN–13	0.76	40.4	125	0.93	0.52	5.4	1.1	15.2
AHN–14	0.76	39.7	125	0.96	0.55	6.8	1.4	17.1
AHN–15	0.65	41.4	125	1.02	0.45	22.3	4.0	18.7

Note: δ, apparent density; $V_{\mathsf{\Sigma }}$, total pore volume; $P_{\square }$, axial crushing strength; $P_{○}$, radial crushing strength; L, relative shrinkage.

The Pearson correlation matrix obtained for this series is provided in Table S6. It should be noted that the formulation variables were not varied fully independently. In particular, the use of AHN(125) resulted in pastes with higher calculated moisture contents than those prepared with AHN(160). Therefore, the observed relationships were interpreted comparatively rather than as independent factor effects. Nevertheless, the results revealed several consistent trends and identified promising directions for further optimization. The textural and mechanical properties of the granules as functions of the overall Si/Al ratio are presented in Figure 14.

Figure 14. Effect of the overall Si/Al ratio on the properties of ZSM–5/AHN extrudates: (a) total pore volume and apparent density; and (b) crushing strength. Circles and diamonds correspond to the quantities indicated in the respective panel legends. Filled symbols correspond to compositions prepared with AHN(160), and open symbols correspond to compositions prepared with AHN(125).

No unambiguous monotonic dependence of the properties on Si/Al ratio was observed for the AHN–(1–10) series since the moisture content and component ratio varied simultaneously. Nevertheless, the total pore volumes of most samples remained within a comparatively narrow range, whereas their mechanical strengths varied much more substantially. This result demonstrates that similar integral pore structure characteristics do not necessarily ensure comparable stability of the interparticle framework.

Among the compositions prepared with AHN(160), the most balanced properties were obtained at Si/Al ≈ 0.65 for samples AHN–4 and AHN–5. These samples combined a total pore volume of ~0.43 cm³/g with the highest measured strength values. At higher Si/Al values, the mechanical properties generally decreased, which may be attributed to the lower effective amount of the aluminum-containing binder phase and the increased sensitivity of the structure to paste preparation conditions. Taken together with the ZSM–5/pseudoboehmite data, these results suggest that an overall Si/Al ratio of ~0.6–0.7 may provide favorable mechanical properties, although the available data are insufficient to establish a definitive optimum.

Comparison of the two precursors at similar compositions revealed the most pronounced differences at high Si/Al values. At Si/Al ≈ 0.76, the use of AHN(125) resulted in more porous but considerably less mechanically robust granules, particularly under radial loading. At lower zeolite contents, the differences between AHN(125) and AHN(160) were less pronounced.

These findings indicate that the poorly ordered and more highly hydrated AHN(125) precursor favored the retention of a larger pore volume but did not provide a sufficiently strong binder framework at high ZSM–5 contents. In contrast, the more extensively thermally treated AHN(160) generally produced more mechanically stable granules under the investigated conditions. However, this difference cannot be attributed exclusively to the precursor phase state since the pastes prepared with AHN(125) also had higher moisture contents.

Overall, the screening experiments confirmed the feasibility of using nitrate-derived aluminum precursors to shape ZSM–5-containing granules. Although the AHN-based systems were generally inferior to the optimized pseudoboehmite compositions in terms of mechanical strength and reproducibility, they demonstrated the potential of aluminum hydroxynitrates as alternative alumina binder precursors. Further optimization may include variation in the precursor treatment temperature and duration, partial addition of a peptizing agent, the use of mixed PB/AHN binder systems, or the introduction of organic and inorganic binders to control formability, porosity, and mechanical strength. From a practical perspective, the AHN route also requires consideration of the nitrogen-containing volatile products released during precursor preparation and calcination, as well as possible effects of nitrate-derived species on the acidic properties of the zeolite phase.

3. Materials and Methods

3.1. Preparation of Shaped Samples

3.1.1. Starting Materials

Synthetic NH₄-ZSM–5 zeolite (Meryer, Shanghai, China) with a SiO₂/Al₂O₃ molar ratio of 38–40 was used as the zeolitic component.

Commercial pseudoboehmite, AlOOH·nH₂O, supplied by SKTB Katalizator (Novosibirsk, Russia), was used as the conventional alumina-based binder. The loss on ignition (LOI) of the pseudoboehmite at 600 °C was 24.8 wt.%, and its specific surface area was 175 m²/g.

Commercial nitric acid, HNO₃, with a concentration of 65 wt.% and a density of 1.39 g/cm³ was used as the peptizing agent. Distilled water was used to prepare the shaping pastes.

Aluminum hydroxynitrate obtained from aluminum nitrate nonahydrate, Al(NO₃)₃·9H₂O, by thermal treatment at either 125 or 160 °C for 4 h was used as an alternative alumina binder precursor. The resulting products are hereafter denoted as AHN(T), where AHN refers to aluminum hydroxynitrate, and T is the heat treatment temperature. After thermal treatment, the products were ground, sieved, and stored in a desiccator.

3.1.2. Shaping of Samples

In the first stage, model samples based on pseudoboehmite without zeolite were prepared to determine the effects of water and peptizer contents on paste formability and on the strength, shrinkage, density, and pore structure of the resulting granules.

A 20 g portion of pseudoboehmite was placed in a porcelain mortar, after which nitric acid and water were added. When nitric acid was included in the formulation, it was diluted with a specified amount of water. The peptizing solution and the remaining water were then introduced into the paste in small portions (2–3 cm³), with thorough mixing after each addition. After the liquid phase had been completely added, the paste was mixed and kneaded for 10 min until a homogeneous plastic mass suitable for extrusion was obtained. Across the formulation series, the water content was varied within the range that yielded a cohesive and extrudable paste. Thus, the resulting moisture content should be regarded not only as a formulation parameter but also as an indicator of the water demand of a given system under the selected shaping conditions.

Shaping was performed using a laboratory ram extruder equipped with a die 2.7 mm in diameter. The resulting extrudates were cut into cylindrical granules approximately 3–5 mm in length. The granules were aged overnight at room temperature, dried at 150 °C for 4 h using a heating rate of 2 °C/min, and subsequently calcined in air at 500 °C for 2 h using a heating rate of 10 °C/min.

Shaped ZSM–5/pseudoboehmite composites were prepared using a similar method. ZSM–5 and pseudoboehmite were mixed at predetermined mass ratios, with the total mass of the dry mixture fixed at 20 g. The amount of water was adjusted to obtain a plastic paste suitable for extrusion. Nitric acid was introduced as a peptizing agent in selected formulations. The mixing, extrusion, overnight aging, drying, and calcination procedures were identical to those described above.

Shaped composites were prepared using aluminum hydroxynitrate as an alternative binder precursor. ZSM–5 and aluminum hydroxynitrate were mixed at different mass ratios, and distilled water was subsequently added until a formable paste was obtained. No additional nitric acid was introduced. The pastes were extruded, aged overnight at room temperature, dried, and calcined under the same conditions as those described above.

3.2. Characterization of Shaped Samples

The phase composition and structural characteristics of the samples were studied by X-ray diffraction (XRD). Diffraction patterns were recorded on a DRON-4-07 powder diffractometer (Burevestnik, Saint Petersburg, Russia) using Co Kα radiation over a 2θ range of 10–90° with a step size of 0.02°. Phase identification was performed using the PDF database and Crystallographica Search-Match v. 2.0.3.1 software (Oxford Cryosystems, Oxford, UK).

The mean coherent scattering domain (c.s.d.) sizes were determined from the broadening of the diffraction maxima using the Scherrer equation [14]:

$$d={\displaystyle \frac{0.9\mathsf{\lambda }}{B\cos \mathsf{\theta }}},$$

(2)

where d is the coherent scattering domain size (nm), λ = 0.1789 nm is the wavelength, B is the full width at half maximum of the peak (rad), and θ is the diffraction angle (°).

Differential thermal analysis of the initial pseudoboehmite was performed to evaluate water removal and the thermal transformation of AlOOH·nH₂O. The analysis was carried out using a DTG–60H thermal analyzer (Shimadzu, Akishima, Tokyo, Japan) over the temperature range of 25–800 °C at a heating rate of 10 °C/min.

The surface morphology of the samples was studied by scanning electron microscopy (SEM). The measurements were performed on a JSM–6610LV scanning electron microscope (JEOL, Akishima, Tokyo, Japan). The elemental composition of the aluminum hydroxynitrate precursors was estimated by energy-dispersive X-ray spectroscopy (EDS) using the analytical system coupled to the scanning electron microscope. The N/Al atomic ratios obtained by SEM–EDS were combined with the experimentally determined mass losses during precursor preparation to estimate the bulk empirical compositions of aluminum hydroxynitrates.

The specific surface area and pore characteristics of the samples were studied by low-temperature nitrogen adsorption–desorption at 77 K using a Thermo Scientific Surfer gas adsorption analyzer (Thermo Fisher Scientific, Milan, Italy). Before analysis, the samples were degassed at 200 °C. The specific surface area was calculated using the Brunauer–Emmett–Teller (BET) method. Pore-size distributions were calculated using the Barrett–Joyner–Halenda (BJH) method applied to the desorption branch of the isotherm.

FTIR spectra were recorded on a Nicolet iS5 FTIR spectrometer (Thermo Fisher Scientific, Milan, Italy) at room temperature in the range of 4000–400 cm^–1 in transmission mode. The samples for analysis were prepared by pressing with KBr.

The total pore volume of the shaped granules, $V_{\mathsf{\Sigma }}$ (cm³/g), was determined by water titration. A weighed portion of the calcined sample was placed in a glass flask, and distilled water was added dropwise until the pore space was completely filled. Adhesion of the granules to the flask wall was used as the endpoint criterion. Each sample was measured in duplicate. The total pore volume was calculated as follows:

$$V_{\mathsf{\Sigma }}={\displaystyle \frac{V_{{\mathrm{H}}_2\mathrm{O}}}{m}},$$

(3)

where $V_{{\mathrm{H}}_2\mathrm{O}}$ is the volume of water consumed during titration (cm³) and m is the mass of the dry calcined sample (g).

The mechanical strength of the shaped granules was determined using a LinteL PK–21 automated crush-strength tester (BSKB Neftekhimavtomatika, Ufa, Russia). Before testing, the diameter and length of each granule were measured using a vernier caliper. The tests were performed in two loading geometries: axial compression between the end faces of the cylindrical granule and radial compression along the generatrix.

The axial crushing strength, $P_{\square }$ (MPa), was calculated as follows:

$$P_{\square }={\displaystyle \frac{4F_{\square }}{\pi D^2}},$$

(4)

where $F_{\square }$ is the measured failure load under axial compression (N), and D is the granule diameter (mm).

The radial crushing strength, $P_{○}$ (MPa), was calculated using the longitudinal projected area of the cylindrical granule:

$$P_{○}={\displaystyle \frac{F_{○}}{DH}},$$

(5)

where $F_{○}$ is the measured failure load under radial compression (N), D is the granule diameter (mm), and H is the granule height (mm).

Granule shrinkage, L (%), was evaluated from the change in extrudate diameter after drying and calcination relative to the initial diameter measured immediately after extrusion. The diameter of the fresh extrudates, D₀ (mm), was measured directly after shaping, whereas the diameter of the calcined granules, D (mm), was measured after thermal treatment. Shrinkage was calculated using Equation (6):

$$L={\displaystyle \frac{D_0-D}{D_0}}\cdot 100\mathrm{\%}.$$

(6)

The apparent density, δ (g/cm³), of the cylindrical granules was calculated from the mass of each granule, m (g), and its geometric volume determined from its diameter, D (mm), and height, H (mm):

$$\mathsf{\delta }={\displaystyle \frac{4m}{\pi D^{2}H}}.$$

(7)

Ten granules were used to determine axial and radial crushing strength, shrinkage, and apparent density. The final value was calculated as the mean of individual measurements.

3.3. Data Processing and Comparison Criteria

The main formulation variables were the ZSM–5 content, the type and amount of the alumina-based binder, and the water and nitric acid contents. The response variables included total pore volume, apparent density, shrinkage, axial crushing strength, and radial crushing strength.

Axial crushing strength was used as the primary structural and mechanical criterion for the cylindrical extrudates, whereas total pore volume was selected as the key pore structure parameter. Apparent density and shrinkage were additionally considered as indicators of consolidation of the shaped bodies during their drying and calcination.

Pearson linear correlation coefficients were calculated to evaluate the relationships between the processing factors and the resulting extrudate properties. Data correlation was performed using Wolfram Mathematica, v. 13.3.1.0 (Wolfram Research, Champaign, IL, USA).

4. Conclusions

A systematic study was conducted on the shaping of ZSM–5-based materials using two types of aluminum-containing binders: pseudoboehmite and nitrate-derived aluminum hydroxynitrate precursors. Examination of a model system containing only pseudoboehmite allowed to separately assess the roles of water content, degree of peptization, and nitric acid concentration during the initial mixing stage. The resulting relationships were subsequently applied to the analysis of more complex zeolite-containing compositions.

Acid peptization was shown to have a strongly non-linear effect. The addition of a small amount of HNO₃ induced restructuring of the pseudoboehmite aggregates, accompanied by a decrease in pore volume and increase in apparent density, shrinkage, and mechanical strength. Further increasing the acid content produced minor changes in the integral textural characteristics while making the outcome more sensitive to paste moisture and the peptizing agent concentration. Under excessive peptization, the system became unstable, presumably due to heterogeneous structure formation and the development of internal stresses during drying and calcination. Accordingly, a suitable paste formulation should be defined by a rational processing window that accounts for the peptizer amount, the moisture content, and the acid concentration during its initial contact with the dry mass.

Total pore volume and apparent density were found to primarily characterize the overall packing and consolidation of particles but did not fully determine the mechanical stability of the granules. Samples with similar integral textural characteristics exhibited substantially different crushing strengths, particularly under radial loading. This behavior demonstrates the importance of binder-phase distribution, connectivity of the interparticle framework, structural orientation induced by extrusion, and local defects. Radial crushing strength thus represents a more sensitive indicator of the quality of shaped catalyst bodies than axial crushing strength alone.

Shaping ZSM–5 with unpeptized pseudoboehmite decreased the pore volume and increased the apparent density but, contrary to the expected weakening associated with a lower binder fraction, also increased granule strength. This finding shows that the properties of the shaped material depend not only on the binder content but also on the organization of the zeolite and alumina particles within the composite framework. Provided that a sufficient amount of pseudoboehmite is retained, the zeolite phase may contribute to the formation of a dense structure. In peptized ZSM–5/pseudoboehmite pastes, moisture primarily controlled interparticle porosity and density, whereas mechanical strength was governed by the combined effects of water content, formulation composition, and the state of the binder phase.

Aluminum hydroxynitrate (AHN) precursors obtained by thermal treatment of Al(NO₃)₃·9H₂O in the range of 125–160 °C were also evaluated for shaping ZSM–5. The precursor obtained at 125 °C remained predominantly poorly ordered and more highly hydrated, whereas a more crystalline phase formed at 150–160 °C. Both precursor types enabled the preparation of ZSM–5-based granules, while AHN(160) provided a more favorable balance between porosity and mechanical strength. Although the investigated AHN systems were generally inferior to the optimized pseudoboehmite compositions in terms of strength and reproducibility, they demonstrated the potential of nitrate-derived precursors as tunable alternative binders whose performance may be improved through further optimization of the shaping conditions.

Overall, catalyst shaping should be regarded as an independent stage of directed structure formation rather than merely as a means of converting a powder into granules. Water content, precursor type, and peptization procedure jointly determine the pore structure and mechanical stability of the final zeolite-containing catalyst. The proposed approach provides a practical basis for selecting formulations with a controlled balance among active phase content, porosity, and mechanical strength. It also establishes a framework for subsequently correlating structural and mechanical properties with preservation of zeolite microporosity, acid-site distribution, and the catalytic performance of the shaped materials.