Methanol is a feedstock for producing many demanded petrochemical products: lower olefins, gasoline products, and aromatic compounds. Lower olefins (ethylene, propylene, butylenes) are the main monomers in the modern petrochemical industry. They are mainly used for producing polymers, and also epoxides, alkylaromatic compounds, alcohols, etc. [1]. The traditional procedure for producing C2–C4 olefins is thermal pyrolysis of hydrocarbon feedstock. However, crude oil resources are limited; therefore, search for alternative feedstock is topical. One of the most promising alternatives is the use of methanol as a feedstock; it can be produced from methane, coal, or biomass.

Catalysts based on SAPO-34 microporous silicoaluminophosphate (pore size 4 Å), exhibiting high heat resistance and hydrothermal durability, show high selectivity in methanol conversion to olefins [1]. However, the major drawback of catalysts based on SAPO-34 is their rapid deactivation due to coking [2].

The methanol conversion to olefins on various kinds of zeolite-containing catalysts based on ZSM-11 [3], ZSM-22 [4], and ZSM-5 [5] was studied. In particular, ZSM-5 is a catalyst for producing lower olefins and liquid hydrocarbons owing to its three-dimensional structure and strong acid sites. ZSM-5 has the MFI structure consisting of straight (5.6 × 5.3 Å) and sinusoidal (5.5 × 5.1 Å) channels [6]. An important advantage of ZSM-5 over SAPO-34 is its low deactivation rate due to favorable pore size; therefore, catalysts based on ZSM-5 can preserve the activity for a considerably longer time. However, they are less selective with respect to lower olefins [1].

One of the main parameters influencing the catalyst selectivity is its acidity [7]. Brønsted acid sites ensure the reaction pathway toward formation of lower olefins, whereas Lewis acid sites favor the occurrence of side reactions [7]. To reduce the concentration of Lewis acid sites, catalysts are modified with nonmetals and metals. Various elements such as Fe [8], P [9], Mn [10], and Ga [11] were used for modifying methanol conversion catalysts based on ZSM-5. Their interaction with aluminum as a zeolite constituent decreases the amount of acid sites [1, 12].

Small size of inlet windows of ZSM-5-type zeolite leads to rapid deactivation of the catalyst as a result of coke accumulation in narrow pores. This problem can be solved by using micro-mesoporous catalysts. Halloysite can be used as a mesoporous component. Halloysite is a natural aluminosilicate consisting of kaolin plates rolled into nanotubes 0.5–1.2 μm long with the outside diameter of 40–60 nm and inside diameter of 10–30 nm [1416].

Previously we synthesized an H–ZSM-5–halloysite catalyst [13] that showed high selectivity with respect not only to propylene but also to С5–С8 in dimethyl ether (DME) conversion (11 and 32%, respectively, at DME conversion of 80%), which may be due to the formation of the micro-mesoporous structure on introducing halloysite.

It can be anticipated that introduction of halloysite aluminosilicate nanotubes into ZSM-5 zeolite will enhance its stability and eliminate the diffusion hindrance, and “softer,” compared to the zeolite, acidity of halloysite will reduce the catalyst deactivation caused by coking [14]. In this study, the catalyst based on halloysite mesoporous aluminosilicate nanotubes and H–ZSM-5 microporous zeolite was examined in methanol conversion to hydrocarbons.

EXPERIMENTAL

H–ZSM-5–halloysite catalyst was prepared as described previously [13] from H–ZSM-5 zeolite (SiO2/Al2O3 = 37, Zeolyst, the United Kingdom) and halloysite mineral (Sigma–Aldrich, the United States). The peptizer was a 1 M aqueous nitric acid solution (EKOS-1) containing 2.5 wt % polyethylene glycol (Fluka Analytical, the United States). The calculated halloysite content in terms of dry sample weight was 33 wt %. After mixing the components, the mass was extruded through a die 1 mm in diameter, dried at room temperature for a day and then at 60, 80, 110, and 140°С for 2 h at each temperature, and calcined at 550°С for 3 h.

As a reference catalyst we took pure H–ZSM-5 zeolite, which was formed at a pressure no higher than 2.5 MPa (to prevent the structure degradation) in a mold; the catalyst was finely divided, and the 0.2–0.5 mm fraction was taken.

X-ray diffraction analysis was performed with a Rigaku SmartLab device in the range 2θ = 5°–55° with a step of 0.05°.

Transmission electron microscopic (TEM) images were taken with a JEOL JEM-2100 microscope with the 50–1 500 000 magnification and image resolution of 0.19 nm at 200 kV.

Low-temperature nitrogen adsorption–desorption was studied with a Gemini VII 2390t device (Micromeritics, the United States). The specific surface area was determined using the Brunauer–Emmett–Teller (BET) equation, and the micropore volume and external surface area were calculated by the t-plot method.

The acidity was determined by temperature-programmed ammonia desorption (TPD–NH3). Measurements were performed with an AutoChem 2950 HP chemisorption analyzer (Micromeritics, the United States) with the signal recording using the thermal conductivity detector. The amount of weak and medium-strength acid sites (ASs) was determined from the amount of ammonia desorbed at 100–300 С, and the amount of strong ASs, from the amount of ammonia desorbed at 300–550°С.

Catalytic experiments were performed in a flow-through installation with a fixed catalyst bed (3 mL) in the temperature interval 380–460°С under nitrogen pressure (0.1–0.5 MPa) with the feed space velocity (FSV) of 0.5–1 h–1. As a feed we used methanol (chemically pure grade, Khimmed, Russia), which was added to the nitrogen flow (30 mL min–1). The installation was brought to the steady-state operation mode, and two product samples were taken for analysis at 0.5 h interval. The averaged result of analyzing these two samples was taken as a final result.

The gaseous products were analyzed with a laboratory chromatographic complex for natural gas (Khromos-RGU), equipped with a Valco PLOT VP-Alumina Na2SO4 capillary column (50 m×0.53 mm×10.0 μm) and a flame ionization detector. Liquid products were analyzed with a gas–liquid laboratory chromatographic complex for analysis of petroleum products (Khromos-RGU) equipped with a MEGA-WAX Spirit capillary column (0.32 mm × 60 m × 0.25 μm) and a flame ionization detector.

The methanol conversion was calculated using the formula

$${X_{\rm{M}}} = {{m_{\rm{F}}^{\rm{M}} - m_{\rm{P}}^{\rm{M}}} \over {m_{\rm{F}}^{\rm{M}}}} \times 100\% ,$$

where mFM is the methanol weight in the feed, g, and mPM is the methanol weight in the product, g.

The selectivity with respect to hydrocarbons Х, SX, was determined as

$${S_X} = {\rm{\;}}{{{{\rm{\omega }}_X}} \over {{{\rm{\omega }}_1} + {{\rm{\omega }}_2} + \ldots + {{\rm{\omega }}_n}}} \times 100{\rm{\% }},$$

where ωX is the molar concentration of component X, and ω1, ω2, …, ωn are the molar concentrations of all the components in the product mixture.

RESULTS AND DISCUSSION

The structure of the synthesized material, H–ZSM-5– halloysite, was confirmed by X-ray diffraction (XRD) (Table 1); the interplanar spacings d were determined by the Bragg equation. The reflections at 2θ = 7.9°, 8.8°, and 23.1° correspond to ZSM-5-type zeolite [17, 18], and those at 2θ = 8.9°, 24.7°, and 25.7° are characteristic of halloysite [19].

Table 1. Interplanar spacings in H–ZSM-5–halloysite material

Phases of ZSM-5 zeolite and aluminosilicate nanotubes are clearly seen in the TEM images of the H–ZSM-5–halloysite catalyst (Figs. 1a1d).

Fig. 1.
figure 1

(a–d) TEM images of H–ZSM-5–halloysite catalyst, taken at different magnifications.

The results of studying the samples by low-temperature N2 adsorption–desorption and TPD–NH3 are given in Table 2.

Table 2. Textural characteristics and acid properties of halloysite, H–ZSM-5 zeolite, and H–ZSM-5–halloysite catalyst

The zeolite is characterized by high content of acid sites (1222 μmol g–1), the majority of which are strong acid sites. Strong acid sites are usually correlated with the ammonia desorption from Brønsted acid sites [24]. The total amount of acid sites in halloysite is 4.5 times smaller than in zeolite, and the fraction of strong sites among them is 69%. In the ZSM-5–halloysite catalyst, the total amount of acid sites is also smaller (677 μmol g–1) than in zeolite, but the fraction of strong sites among them is 63%.

The pore volume and specific surface area of H–ZSM-5 are virtually totally ensured by micropores. The textural characteristics of halloysite are essentially different: It contains virtually no micropores, and its adsorption isotherm (Fig. 2a) belongs to type IV characteristic of mesoporous materials [2023]. The H–ZSM-5–halloysite catalyst has micro-mesoporous structure. In its nitrogen adsorption isotherm, there is both a portion of a sharp increase in the adsorption at low relative pressures, characteristic of microporous materials, and a hysteresis loop at relative pressures in the range 0.4–1.0, characteristic of mesopores [20]. The micropores originate from ZSM-5 zeolite, and the mesoporous structure, from halloysite nanotubes. The volume and area of micropores in the ZSM-5–halloysite catalyst obtained are approximately two times smaller than in the zeolite.

Fig. 2.
figure 2

(a) Nitrogen adsorption–desorption isotherms and (b) TPD-NH3 spectra for halloysite, H–ZSM-5 zeolite, and H–ZSM-5–halloysite catalyst.

The catalytic properties of the catalyst obtained and of H–ZSM-5 zeolite were studied at temperatures of 380–460°С, pressures of 0.1–0.5 MPa, and FSV of 0.5– 1 h–1. The results are shown in Fig. 3. As the pressure is increased from 0.1 to 0.5 MPa, the methanol conversion on both catalysts decreases (Fig. 3), and the product distribution is shifted toward an increase in the molecular mass, which agrees with the published data [25]. The methanol conversion on the H–ZSM-5–halloysite catalyst is 10–15% higher than on H–ZSM-5 throughout the ranges of the varied parameters.

Fig. 3.
figure 3

Comparison of the product distribution on (a, c, e) H–ZSM-5–halloysite and (b, d, f) H–ZSM-5 catalysts at 380–460°С, P = 0.1–0.5 MPa, and FSV = 0.5–1 h–1. (a) H–ZSM-5–halloysite (P = 0.5 MPa, FSV = 1 h–1), (b) H–ZSM-5 (P = 0.5 MPa, FSV = 1 h–1), (c) H–ZSM-5–halloysite (P = 0.1 MPa, FSV = 1 h–1), (d) H–ZSM-5 (P = 0.1 MPa, FSV = 1 h–1), (e) H–ZSM-5–halloysite (P = 0.1 MPa, FSV = 0.5 h–1), and (f) H–ZSM-5 (P = 0.1 MPa, FSV = 0.5 h–1).

The methane formation selectivity is somewhat higher on H–ZSM-5–halloysite, whereas the selectivity of the formation of С2–С4 alkanes is higher on H–ZSM-5. At a pressure of 0.5 MPa, the methane amount obtained on the H–ZSM-5–halloysite catalyst increases with temperature, whereas the selectivity of the formation of С2–С4 alkanes remains approximately constant. On the H–ZSM-5 catalyst, the selectivity of the formation of С2–С4 alkanes is virtually constant, whereas the methane formation selectivity is considerably lower than on H–ZSM-5–halloysite.

The yield of С6+ hydrocarbons (mainly aromatic) on the modified catalyst is considerably higher than on H–ZSM-5, probably because of the fact that the formation of large molecules in H–ZSM-5 micropores is impossible. It is believed that liquid aromatic compounds are precursors of coke whose formation leads to the catalyst deactivation [26]; however, the H–ZSM-5–halloysite catalyst did not underwent coking, which may be due to facilitated mass transfer in halloysite voids [13].

Figure 4 shows the selectivity of the formation of lower olefins on the catalysts tested in the examined ranges of parameters. At a pressure of 0.5 MPa and FSV of 1 h–1, propylene is formed on the H–ZSM-5–halloysite catalyst with the highest selectivity among olefins (up to 20%); the ethylene formation selectivity is somewhat lower (up to 15%), and the selectivity of the formation of butylenes does not exceed 10%. On H–ZSM-5, butylenes are formed with the highest selectivity (up to 19%); the propylene formation selectivity is somewhat lower (6–15%), and the ethylene formation selectivity does not exceed 5%.

Fig. 4.
figure 4

Olefin formation selectivity on (a, c, e) H–ZSM-5–halloysite and (b, d, f) H–ZSM-5 catalysts as a function of temperature. T = 380–460°С, P = 0.1–0.5 MPa, and FSV = 0.5–1 h–1. (a) H–ZSM-5–halloysite (P = 0.5 MPa, FSV = 1 h–1), (b) H–ZSM-5 (P = 0.5 MPa, FSV = 1 h–1), (c) H–ZSM-5–halloysite (P = 0.1 MPa, FSV = 1 h–1), (d) H–ZSM-5 (P = 0.1 MPa, FSV = 1 h–1), (e) H–ZSM-5–halloysite (P = 0.1 MPa, FSV = 0.5 h–1), and (f) H–ZSM-5 (P = 0.1 MPa, FSV = 0.5 h–1).

As the pressure is decreased to 0.1 MPa, the ethylene formation selectivity on the H–ZSM-5–halloysite catalyst decreases to 8%, the selectivity of the formation of butylenes increases to 15%, and propylene remains the major product (selectivity up to 20%). On H–ZSM-5, the trends are similar, but the propylene formation selectivity is lower than on H–ZSM-5–halloysite, whereas the selectivity of the formation of ethylene and butylenes is higher. Thus, addition of halloysite imparting mesoporous structure to the catalyst leads at the given process parameters to an increase in the methanol conversion and in the propylene formation selectivity.

CONCLUSIONS

A comparative study of the methanol conversion on the H–ZSM-5–halloysite catalyst and on straight H–ZSM-5 was made. The use of halloysite as a catalyst component leads to an increase in the selectivity of the formation of propylene and С6–С8 aromatic hydrocarbons due to its physicochemical properties, namely, its mesoporous structure and acid properties. The halloysite-based catalyst shows promise for the production of both lower olefins and aromatic hydrocarbons.