Nanocavity effects of various zeolite frameworks on n-pentane cracking to light olefins: combination studies of DFT calculations and experiments

Anawat Thivasasitha, Thana Maihomb, Sitthiphong Pengpanichc and Chularat Wattanakit*a
aSchool of Energy Science and Engineering, Nanocatalysts and Nanomaterials for Sustainable Energy and Environment Research Network of NANOTEC, Vidyasirimedhi Institute of Science and Technology, Rayong 21210, Thailand. E-mail: chularat.w@vistec.ac.th
bDepartment of Chemistry, Faculty of Liberal Arts and Science, Kasetsart University, Kamphaeng Saen Campus, Nakhon Pathom 73140, Thailand
cPTT Global Chemical Public Company Limited (PTTGC), Bangkok 10900, Thailand

Received 10th July 2019 , Accepted 19th August 2019

First published on 20th August 2019


Better control of the product selectivity of light olefins (e.g., ethylene and propylene) obtained from the n-pentane catalytic cracking process has attracted considerable attention from both scientific and petrochemical industrial points of view. In this context, we report insights into the effects of the nanocavities of various zeolite frameworks, including H-FER, H-ZSM-5, and H-FAU, representing small, medium, and large cavities, on the reaction mechanism of n-pentane cracking to light olefins by using M06-2X/6-31G(d,p) density functional calculations, eventually leading to fine-tuning the product distribution of light olefins. The reaction mechanism consists of the following two main steps: (i) the protolytic cracking of n-pentane to form a pentonium intermediate; and (ii) the subsequent dissociation of the intermediate to either ethane–propylene or ethylene–propane. The key reaction pathways controlling the product distribution of light olefins relate to the dissociation of the pentonium intermediate, which can produce selectively either propylene (P) or ethylene (E), resulting in a controllable P/E ratio. The differences in the activation energies for ethylene production compared with those of propylene production over H-FER, H-ZSM-5, and H-FAU are 6.7, 5.0, and 0.5 kcal mol−1, respectively. Compared with H-ZSM-5 and H-FAU, the higher difference in the activation energy of these two pathways over H-FER implies that the preferable production of ethane–propylene compared with ethylene–propane is more pronounced. It is therefore reasonable to conclude that a smaller pore zeolite such as H-FER eventually leads to a high ratio of production of propylene to ethylene, in accordance with experimental observations.


1 Introduction

The development of light naphtha catalytic cracking to light olefins such as ethylene, propylene and butenes has attracted considerable attention from both scientific and petrochemical industrial points of view. Currently, the steam-cracking process is mainly used in the conversion of naphtha to light olefins. However, this process requires high energy consumption due to high operating temperatures (>850 °C) and it is difficult to control the product distribution of light olefins.1 To circumvent this problem, the catalytic cracking process has been intensively developed to reduce the energy consumption, to control the product distribution and to improve the yields of light olefins. Typically, solid acid catalysts such as zeolites, metals supported on zeolites and metal-zeolites mixed with Al2O3 are widely used in the catalytic cracking process.1,2 In particular, Brønsted acid zeolites (e.g., ZSM-5, ZSM-22, ZSM-48, MOR, and SAPO-34), which exhibit the advantage of molecular shape selectivity, can be used to produce high yields of light olefins such as ethylene and propylene at lower temperatures (550–650 °C).1

Zeolite catalysts composed of 8-, 10- and 12-membered ring (MR) channels with pore diameters in the range of 3.8 to 7.5 Å such as CHA, AEI, MFI, FER, FAU, MOR, and BETA are widely used in the catalytic cracking process because of their unique pore channels, eventually resulting in different product distributions. Small-pore zeolites such as CHA and AEI composed of 8 MR straight channels (a diameter of 3.8 × 3.8 Å)3,4 are suitable for methanol to light olefins (MTO).4 A medium-pore zeolite such as MFI with two intersections of zigzag and straight 10 MR channels (a diameter of 5.1 to 5.6 Å3,4) is suitable for catalytic cracking and conversion of hydrocarbons to aromatics/olefins, and methanol conversion to light olefins (MTO).4 Small- and medium-pore zeolites such as FER composed of two perpendicular intersections of 8 MR (a diameter of 3.4 × 5.4 Å) and 10 MR (a diameter of 4.3 × 5.5 Å) straight channels3,4 are suitable for nitrous oxide decomposition and reduction, methanol to olefin, n-paraffin cracking, alkane hydroisomerization, and skeletal isomerization of n-alkene to isoalkene.4–9 However large-pore zeolites, such as FAU and BETA with 12 MR with a pore diameter of 6.5 to 7.4 Å,3,4 are suitable for the alkylation of aromatics, isomerization, and catalytic cracking of hydrocarbons.4 This information makes it clear that the different architectures of zeolite structures match different catalytic reactions, involving different molecular structures of reaction schemes.

There are many reports demonstrating the study of insights into the nanocavity effect for the catalytic cracking of hydrocarbons on various types of zeolites. For example, Swisher and co-workers10 studied the kinetics of alkane cracking (C3–C6) over ZSM-5 and FAU zeolites. It was found that the apparent rate coefficient for alkane cracking over ZSM-5 is higher than that of the FAU zeolite. Maihom and co-workers11 studied the mechanism of n-hexane cracking to propane and propylene on ZSM-5 and FAU zeolites. The adsorption energy of n-hexane in medium pores of ZSM-5 is higher than that of large-pore FAU due to the van der Waals interaction of the adsorbate and zeolitic wall, while the catalytic activity over both catalysts is slightly different.11 Hou and co-workers12 studied the reaction pathways for n-pentane cracking on different zeolites such as ZSM-35, BETA, and ZSM-5 to produce light olefins. It was found that the ZSM-5 zeolite with intersecting medium pores exhibited high selectivity for light olefins because of the advantages of molecular shape selectivity of ZSM-5. Bortnovsky and co-workers13 studied the catalytic performance of n-pentene cracking to C2–C4 light olefins over different types of zeolites such as FER, MOR, ZSM-12, BETA, ZSM-5 and ZSM-11. They reported that only ZSM-5 and ZSM-11 composed of intersecting medium 10 MR channels and acid concentrations in the range 0.03–0.1 mmol g−1 showed a high yield of C2–C4 olefins. Apart from the catalytic cracking of hydrocarbons, the confinement effects of various zeolite frameworks for other chemical reactions have been also studied. Indeed, the confinement effects of zeolites with different pore diameters, structures, and pore-connection can control the diffusion of reactants, intermediates, and products as well as chemical reaction processes, eventually affecting the product distribution.14–16

The catalytic cracking of hydrocarbons to light olefins on Brønsted acid zeolites was typically proposed to occur via bimolecular or monomolecular mechanisms.11,12,17–19 The bimolecular mechanism involves hydride transfer between alkanes and adsorbed trivalent carbenium ions on the zeolite surface followed by isomerization and β-scission.17–19 This reaction mechanism typically occurs under high partial pressure of hydrocarbons at low temperature (<400 °C).20 In contrast, the monomolecular mechanism is of crucial importance to produce light olefins and is typically obtained at low partial pressure, high temperature (>400 °C) and low alkane conversion.20 This reaction mechanism is proposed via the pentavalent carbonium ion intermediate generated by the protonation of alkanes over Brønsted acid sites. Then, C–C bond cleavage is observed to form alkane and alkene products.17,19 Because high selectivity of light olefins can be achieved only in the case of the monomolecular pathway, we focus on the study of the catalytic cracking of hydrocarbons to light olefins via the monomolecular mechanism under conditions of low partial pressure, high temperature, and low alkane conversion.

In the present study, we report on the nanocavity effect of various zeolite frameworks ranging from small and medium to large pore zeolites such as H-ZSM-5, H-FER, and H-FAU on the catalytic activity and product selectivity of n-pentane cracking to ethylene and propylene using M06-2X/6-31G(d,p) density functional calculations including dispersion energy. The catalytic cracking process occurs through the monomolecular mechanism via the direct protonation of the C–C bond of n-pentane over Brønsted acid sites to produce a pentonium intermediate. Subsequently, the C–C bond cleavage of the pentonium is observed to form different product distributions of alkanes and alkenes including ethane/propylene and ethylene/propane. This theoretical study is also discussed along with experimental data to clearly understand the effect of different ranges of zeolite pore size on the product distribution of light olefins (the propylene/ethylene ratio).

2 Computational methods and experimental studies

2.1 Models and computational details

The catalyst models of H-ZSM-5, H-FER, and H-FAU were generated from crystallographic data21–23 as shown in Fig. 1. To represent the MFI structure, H-ZSM-5 with a 34T quantum cluster model was covered at the intersection cavities between the 10-membered ring (10 MR) window of 5.1 × 5.5 Å in diameter of straight channels and the 10-membered ring (10 MR) window of 5.3 × 5.6 Å in diameter of zigzag channels and one aluminum atom was substituted at the T12 position in order to generate the Brønsted acid site (Fig. 1a).5,11,24 H-FER with a 37T quantum cluster model was covered at the perpendicular intersection of straight channels between the 10-membered ring (10 MR) window of 4.2 × 5.4 Å and 8-membered ring (8 MR) window of 3.5 × 4.8 Å in diameter and one aluminum atom was substituted at the T2 position (Fig. 1b).5,11,24 H-FAU with a 36T quantum cluster model composes of the 12-membered ring (12 MR) window of 7.4 Å in diameter connecting between two supercages of FAU and one aluminum atom was substituted at T2 position (Fig. 1c).5,11,24
image file: c9cp03871j-f1.tif
Fig. 1 Quantum cluster models representing various zeolite structures of: (a) H-ZSM-5 (34T quantum cluster), (b) H-FER (37T quantum cluster), and (c) H-FAU (36T quantum cluster).

All calculations were carried out using the M06-2X density functional,11,25–27 which also describes the van der Waals interactions between adsorbate molecules and the zeolitic wall and is suitable for studying a chemical reaction system.26,27 The 6-31G(d,p) basis set was applied to all atoms of adsorbed species and zeolite frameworks. During the geometry optimizations, the 5T clusters of the active site region of all zeolite models and adsorbate molecules were able to be relaxed, while the rest of the atoms were fixed at the crystallographic coordinates to retain the structure of their corresponding frameworks. Frequency calculations were performed at the same level of theory for the identification of the local minima (INT) and the transition states (TS). All calculations were performed with the GAUSSIAN 09 code.28

2.2 Catalyst preparation and catalytic tests

ZSM-5 (a Si/Al ratio of 70) was synthesized by using a previously reported method.29,30 FER (a Si/Al ratio of 18) and FAU (a Si/Al ratio of 5) were taken from the Zeolyst International Company with a code of CP914C and CBV720, respectively. All zeolite catalysts were transformed to the H+ form by ion-exchange with 0.1 M NH4NO3 solution at 80 °C for 2 h with repetition of steps three times. Subsequently, the zeolites were dried and calcined at 550 °C for 4 h to remove impurities. The zeolite catalysts were defined as H-ZSM-5, H-FER, and H-FAU.

The catalytic cracking of n-pentane to light olefins was carried out using a continuous down-flow tubular fixed-bed reactor. A 0.2 g of the zeolite catalyst was placed in the middle part of the reactor and then it was pretreated with 3% (v/v) H2 in He for 3 h. The n-pentane feed was introduced using a total flow rate of 0.025 mL min−1. All reactions were carried out under atmospheric pressure at a temperature of 550 °C and a weight hourly space velocity (WHSV) of 5 h−1.31 The reaction products were analyzed by an on-line gas chromatograph (GC) (Agilent 7820A) equipped with an FID detector and a GS-GASPRO capillary column (60 m × 0.320 mm).

3 Results and discussion

3.1 Catalytic conversion of n-pentane to light olefins over various zeolite frameworks

To investigate the effect of various zeolite frameworks on the product distribution of light olefins obtained from the catalytic cracking of n-pentane, proton-exchanged zeolites with various frameworks (H-ZSM-5, H-FER, and H-FAU) were prepared by an ion-exchange method as mentioned above. As can be seen in Fig. 2, the product distribution of equal conversion of n-pentane to light olefins over several zeolites (19.0, 20.0 and 18.5% for H-ZSM-5, H-FER, and H-FAU, respectively) is reported. Obviously, the different zeolite frameworks can control the product formation, eventually leading to different product distributions of light olefins. For example, in the case of H-ZSM-5 composed of intersecting 10 MR cavities of zigzag and straight channels, the ethylene and propylene products are 17.8 and 30.1%, respectively, whereas for H-FER composed of perpendicular intersecting straight 8 MR and 10 MR channels, the amount of ethylene product (13.2%) is lower than that of H-ZSM-5 and the amount of propylene (40.7%) is significantly higher. In strong contrast to this, in the case of H-FAU containing 12 MR pore structures with uniform supercages, the ethylene and propylene products are much lower than those of H-ZSM-5 and H-FER and their selectivities are only 10.9 and 13.3%, respectively. In all cases, the cracked alkane products such as methane, ethane and propane appear to be almost equally described as 40.1, 39.8, and 39.7% for H-ZSM-5, H-FER, and H-FAU respectively. In addition, the other products such as butane, butene, iso-pentane, hexane, hexene, and BTX obtained when using H-FAU can be produced significantly (36.1%), whereas only 12.0 and 6.4% are obtained over H-ZSM-5 and H-FER, respectively. The selectivity of all alkanes and alkenes is summarized as shown in Table S1 in the ESI.
image file: c9cp03871j-f2.tif
Fig. 2 n-Pentane conversion (in parentheses) and product selectivity of light olefins over various types of zeolites: obtained under experimental conditions of T = 550 °C, WHSV = 5 h−1, and n-pentane feed = 0.025 mL min−1.

From these observations, it is clearly seen that the size of the zeolite frameworks is directly related to the light olefin distribution. Large pore cavities are not suitable to control the light olefin yield. In addition, the propylene to ethylene (P/E) ratio can be tuned easily depending on the framework structures. For example, the P/E ratios obtained when using H-FER, H-ZSM-5, and H-FAU are 3.1, 1.7, and 1.2, respectively. The small pore zeolite can greatly enhance the P/E ratio. To understand the product distribution as a function of the zeolite framework, the details of the catalytic reaction mechanism on various types of zeolites are explained in the Computational section.

3.2 Reaction mechanism of n-pentane cracking to light olefins on various zeolite frameworks

It is well-known that there are several possible pathways of n-pentane cracking to small alkene and alkane products as shown in Scheme 1. However, under appropriate conditions for light olefin production, pathway 1 can be excluded, and therefore pathways 2 and 3 are the key point to control the distribution of light olefins. The reaction mechanisms of n-pentane catalytic cracking on Brønsted acid sites of zeolites are shown in Scheme 2. The C–C bond of n-pentane is broken to produce smaller alkane and alkene products. In this work, it is reasonable to consider two different pathways to produce light olefins from the n-pentane catalytic cracking process via: (i) ethane–propylene (PROD) production through transition state 2 (TS2); and (ii) ethylene–propane (PROD′) production through transition state 2′ (TS2′), because both pathways exhibit a main contribution as can be seen from obtained experimental data demonstrating ethane, ethylene, propane, and propylene as main products.
image file: c9cp03871j-s1.tif
Scheme 1 Illustration of possible pathways of n-pentane cracking to alkane and alkene products.

image file: c9cp03871j-s2.tif
Scheme 2 The reaction mechanism of n-pentane catalytic cracking to ethane–propylene (PROD) and ethylene–propane products (PROD′).

The reaction mechanism of n-pentane catalytic cracking to alkanes and alkenes proceeding via the monomolecular mechanism consists of the following two main steps: (i) the protolytic cracking of n-pentane to form a pentonium intermediate; and (ii) the dissociation of the pentonium intermediate to produce ethane–propylene and ethylene–propane products. For the former step, n-pentane is initially adsorbed on the Brønsted acid site of the zeolites (Scheme 2, ADS). Subsequently, a proton of the zeolite surfaces transfers to the C2–C3 bond of the pentane molecule (Scheme 2, TS1), and a pentonium intermediate can be then produced (Scheme 2, INT1). After that, the dissociation step is associated with the decomposition of the pentonium intermediate to either ethane–propylene or ethylene–propane products. For the production of the ethane–propylene product, ethane is formed by accepting a proton from the Brønsted acid site, and propylene is formed by giving a proton back to the zeolite framework (Scheme 2, TS2). In the ethylene–propane production, ethylene is formed by transferring a proton back to the zeolite framework and propane is formed by accepting a proton from the zeolite surface (Scheme 2, TS2′). Finally, ethane–propylene and ethylene–propane products are produced on the zeolite surfaces (Scheme 2, PROD and PROD′ for ethane–propylene and ethylene–propane products, respectively).

3.2.1 Effect of the zeolite frameworks on the adsorption of n-pentane. The optimized structures of n-pentane adsorption on H-ZSM-5, H-FER, and H-FAU are shown in Fig. 3a–c. Selected geometrical structural parameters for the adsorption complexes are shown in Tables S2–S4 in the ESI. The reaction starts from n-pentane adsorption on the Brønsted acid site (Hz) of the zeolite at the C2⋯C3 position of n-pentane.
image file: c9cp03871j-f3.tif
Fig. 3 Optimized structures of n-pentane adsorption on different zeolite frameworks: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 36T.

For the H-ZSM-5 model, n-pentane adsorbs on the Brønsted acid site (Hz) of ZSM-5 at the C2–C3 position with the bond distance of C2⋯Hz and C3⋯Hz around 2.88 Å and 2.62 Å, respectively (Fig. 3a). The adsorption energy of n-pentane on the H-ZSM-5 framework is −15.6 kcal mol−1. The obtained value agrees well with previous calculated and experimental data, which are in the range of −14 to −17 kcal mol−1.32–34 In the case of H-FER, n-pentane adsorbs on the Brønsted site (Hz) of H-FER at the same position with the bond distance of C2⋯Hz and C3⋯Hz approximately 3.40 Å and 2.62 Å, respectively (Fig. 3b). However, the adsorption energy of n-pentane on H-FER is significantly stronger than that of H-ZSM-5 because of the stronger effect of the van der Waals interaction of narrow pore-cavities. The adsorption energy of n-pentane on the H-FER framework is −17.7 kcal mol−1, which is in agreement with the reported data in the range of −16 to −17 kcal mol−1.35 For H-FAU having the largest pore size cavity, n-pentane adsorbs on the Brønsted acid site (Hz) of H-FAU with the bond distance of C2⋯Hz and C3⋯Hz around 2.40 Å and 3.11 Å, respectively (Fig. 3c). The adsorption energy of n-pentane is −12.7 kcal mol−1, which is lower than those of H-ZSM-5 and H-FER. The results are again in the acceptable range of previous reports (−9 to −14 kcal mol−1).10,33,34

3.2.2 Protolytic cracking of n-pentane on various zeolite frameworks. The optimized structures of related compounds in the protolytic cracking of n-pentane on H-ZSM-5, H-FER, and H-FAU are shown in Fig. 4a–c. After the adsorption state, the proton (Hz) of the zeolite moves to the C2–C3 bond of n-pentane to form a pentomium intermediate via transition state 1 (TS1). For the H-ZSM-5 model, the bond distance between Hz and O1 of the zeolite is elongated from 0.97 Å to 1.70 Å. Meanwhile, the bond distances of C2⋯Hz and C3⋯Hz are 1.38 and 1.29 Å, respectively, as shown in Fig. 4a. In the cases of H-FER and H-FAU, there are similar reaction behaviors with respect to H-ZSM-5. For example, the bond distance between Hz and O1 of the zeolite is elongated from 0.97 Å to 1.68 Å. Meanwhile, the bond distances of C2⋯Hz and C3⋯Hz are 1.34 and 1.39 Å, respectively, as shown in Fig. 4b. For the H-FAU model, the bond distance between Hz and O1 of the zeolite is elongated from 0.97 Å to 2.34 Å and the bond distances of C2⋯Hz and C3⋯Hz are 1.24 and 1.25 Å, respectively, as shown in Fig. 4c. Subsequently, the movement of Hz of the zeolites to the C2–C3 bond of n-pentane to form a pentonium intermediate is confirmed by a single imaginary frequency of −633.2, −749.1 and −544.5i cm−1 for H-ZSM-5, H-FER, and H-FAU, respectively. The activation energies of the protonation step on H-ZSM-5, H-FER, and H-FAU are 42.1, 39.6, and 42.9 kcal mol−1, respectively.
image file: c9cp03871j-f4.tif
Fig. 4 Optimized structures of transition state (TS1) according to the protolytic cracking step of n-pentane on various zeolite frameworks: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 36T.

Subsequently, the pentonium intermediate (INT1) is formed on the zeolite surfaces by interacting with the negative charge of framework oxygen (O1) as shown in Fig. 5a–c. The bond distances of C2⋯Hz and C3⋯Hz at an intermediate state are almost equal around 1.25 Å and the bond distance between C2 and C3 is elongated to 1.98 Å with a C2–Hz–C3 bridging angle of 104.8° for H-ZSM-5 as shown in Fig. 5a. The bond distance of C2⋯C3 over the H-FER model is 1.93 Å with a C2–Hz–C3 bridging angle of 100.7° (Fig. 5b), while the bond distance of C2⋯C3 is 2.00 Å with a C2–Hz–C3 bridging angle of 105.2° for the H-FAU model (Fig. 5c). These observations demonstrate that the elongation of the C2⋯C3 bond of the pentonium intermediate, which relates to the protolytic process, depends on the porous structures of the zeolites. The complexation energy of INT1 is highly endothermic with values of 25.8, 20.0, and 30.0 kcal mol−1 for H-ZSM-5, H-FER, and H-FAU, respectively. This makes it clear that the pentonium intermediates over all frameworks are very reactive species and their relative energies are slightly lower than those of their corresponding transition states (TS1) by 0.7, 1.9, and 0.3 kcal mol−1 for H-ZSM-5, H-FER, and H-FAU, respectively. Therefore, the pentonium intermediates can be decomposed easily to either ethylene–propane or ethane–propylene products.


image file: c9cp03871j-f5.tif
Fig. 5 Optimized structures of pentonium intermediates adsorbed on various zeolites: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 36T.
3.2.3 Dissociation of a pentonium intermediate on various zeolite frameworks. As the above-mentioned pentonium intermediates are very reactive, the ethane–propylene and ethylene–propane products can be produced through transition state 2 (TS2) and transition state 2′ (TS2′), respectively (Fig. 6 and 7).
image file: c9cp03871j-f6.tif
Fig. 6 Optimized structures of transition state 2 (TS2) according to the dissociation of a pentonium intermediate to ethane–propylene over various zeolite frameworks: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 36T.

image file: c9cp03871j-f7.tif
Fig. 7 Optimized structures of transition state 2′ (TS2′) according to the dissociation of a pentonium intermediate to ethylene–propane over various zeolite frameworks: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 38T.

In the ethane–propylene production pathway via TS2, the pentonium intermediate is decomposed to ethane–propylene products by transferring the proton (Hz) towards C2 of the ethyl group to form ethane and the proton (Hp) at the C4 of the propyl group to the framework oxygen of the zeolite to form propylene (Fig. 6a–c). For the H-ZSM-5 model, the bond distance of C4⋯Hp is elongated from 1.09 to 1.30 Å and Hp⋯O2 is shortened from 2.89 to 1.28 Å as shown in Fig. 6a, while the bond distance of C4⋯Hp is elongated from 1.09 to 1.47 Å and 1.09 to 1.13 Å and Hp⋯O2 is shortened from 2.67 to 1.17 Å and 2.49 to 1.79 Å for H-FER and H-FAU, respectively (Fig. 6b and c). There are single imaginary frequencies of −445.8, −157.3, and −278.6i cm−1 for H-ZSM-5, H-FER, and H-FAU, respectively, and the frequency is related to the movement of the proton (Hp) of the propyl group to the zeolite framework. The activation energies of this step are 0.6, 0.3, and 5.1 kcal mol−1 for H-ZSM-5, H-FER, and H-FAU, respectively (Fig. 8a and b, TS2). Subsequently, the ethane–propylene products are formed on the zeolite surfaces as shown in Fig. S1a–c (ESI). The desorption energies of the ethane–propylene products are 23.4, 26.6, and 23.2 kcal mol−1 for H-ZSM-5, H-FER, and H-FAU, respectively.


image file: c9cp03871j-f8.tif
Fig. 8 Energy profiles of the overall steps of n-pentane catalytic cracking via ethane–propylene and ethylene–propane production over various zeolite frameworks: (a) H-ZSM-5 34T, (b) H-FER 37T, and (c) H-FAU 36T (the black line and blue dash line show the mechanisms through the ethane–propylene pathway and ethylene–propane pathway, respectively).

In the ethylene–propane production pathway, the pentonium intermediate is converted to ethylene–propane products via TS2′. The Hz atom moves to C3 of the propyl group to produce a propane molecule (Fig. 7). Meanwhile, the proton (He) from C1 of the ethyl group transfers to the framework oxygen of the zeolite to form an ethylene molecule. The bond distances of C1⋯He are elongated from 1.09 to 1.24 Å, 1.09 to 1.25 Å, and 1.09 to 1.13 Å for H-ZSM-5, H-FER, and H-FAU, respectively. The bond distances of He⋯O2 are shortened from 2.87 to 1.50 Å, 3.52 to 1.52 Å, and 2.93 to 1.53 Å for H-ZSM-5, H-FER, and H-FAU, respectively, as shown in Fig. 7a–c. The activation energies of this step over H-ZSM-5, H-FER, and H-FAU are 5.6, 7.0, and 5.6 kcal mol−1, respectively (Fig. 8a–c, TS2′). The transition state structures are confirmed by a single imaginary frequency of −469.2, −432.1, and −333.7i cm−1 for H-ZSM-5, H-FER, and H-FAU, respectively. Finally, the ethylene–propane products are formed on the zeolite surfaces (Fig. S2a–c, ESI) with desorption energies of 24.6, 24.5, and 18.5 kcal mol−1 for H-ZSM-5, H-FER, and H-FAU, respectively.

3.2.4 Effect of the zeolite frameworks on the product distribution of light olefins. As mentioned above the different possible ways to produce light olefins from n-pentane catalytic cracking are either via ethane–propylene or ethylene–propane production. To study the effect of the zeolite frameworks on the light olefin distribution, the energy profiles of n-pentane cracking to ethane–propylene and ethylene–propane over various zeolites are compared as shown in Fig. 8a–c. The ethane–propylene products are formed through transition state 2 (TS2) while the ethylene–propane products are produced via transition state 2′ (TS2′). Compared with the activation barrier of the ethylene–propane pathway, the production of ethane–propylene is much easier with a significantly lower activation energy in the range of 5 to 7 kcal mol−1 for some cases. For example, the difference in activation energy between the production pathways of ethane–propylene (via TS2) and ethylene–propane (TS2′) over H-ZSM-5 is 5.0 kcal mol−1 as shown in Fig. 8a and this value is also in good agreement with the available data.12 Compared to the experimental results, the proportion of ethane–propylene products is higher than that of ethylene–propane products (see Fig. 2 and Table S1 in the ESI). A similar tendency of the difference in activation energy between the production pathways of ethane–propylene (via TS2) and ethylene–propane (TS2′) over H-FER can be observed. Importantly, this value is even much larger (6.7 kcal mol−1) than that of H-ZSM-5 (Fig. 8b). A larger difference in the energy barriers of the ethane–propylene and ethylene–propane production pathways over H-FER with respect to H-ZSM-5 correlates with the higher ratio of ethane–propylene to ethylene–propane over H-FER compared with H-ZSM-5. These observations also confirm the experimental results in which the obtained ratios of ethane–propylene to ethylene–propane over H-FER and H-ZSM-5 are 3.1 and 1.3, respectively, as shown in Table S1 in the ESI.

However, for the H-FAU model having the largest pore cavity (12 MR), the difference in the activation energy of the ethane–propylene and ethylene–propane pathways is almost negligible. The small difference in the activation energies through TS2 and TS2′ is 0.5 kcal mol−1 over H-FAU (see Fig. 8c), implying that an equivalent proportion of ethane–propylene and ethylene–propane can be produced. However, in such cases not only the protolytic pathway but also other side reactions such as oligomerization, and oligomerization-cracking pathways can be observed as can be seen from a high amount of side-products over the large pore zeolite (H-FAU) (Table S1, ESI).

Moreover, the overall reaction steps of n-pentane cracking to light olefins over H-FER, H-ZSM-5 and H-FAU are calculated in term of free energy (ΔG) at 298.15 K as shown in Fig. S3a–c in the ESI. The energy value of each reaction step of the free energy profiles (Fig. S3a–c, ESI) slightly decreases by approximately 1–3 kcal mol−1 compared with the electronic energies (Fig. 8a–c). However, the free energy profiles show the same trend as the electronic energy profiles. The differences in the activation free energies for ethylene production compared with those of propylene production over H-FER, H-ZSM-5, and H-FAU are 5.1, 3.9, and 0.5 kcal mol−1, respectively. These values show the same trend as the difference in the activation electronic energies of 6.7, 5.0 and 0.1 kcal mol−1, for H-FER, H-ZSM-5, and H-FAU, respectively. Although the overall reaction mechanisms of n-pentane cracking to light olefins over H-FER, H-ZSM-5 and H-FAU are considered at 298 K, basically the activation energies do not change at the reaction temperature of 823 K compared with the experimental condition. However, the reaction temperature affects the rate of reactions, which is increased with enhancing the reaction temperature (k298 = 4.33 × 10−2 s−1 and k823 = 4.33 × 10−2 s−1) as can be seen in eqn (1) in the ESI.

In summary, the different pore structure and size of various types of zeolites strongly affect the product distribution of light olefins obtained from the n-pentane catalytic cracking process. From the computational results, it was found that the dissociation of a pentonium intermediate on the zeolite surfaces is the key state to control the light olefin distribution (Fig. 8a–c). To control the ratio of propylene and ethylene (P/E), two reaction pathways have been proposed: (i) ethane–propylene production through transition state 2 (TS2); and (ii) ethylene–propane production through transition state 2′ (TS2′). The difference in activation energy between the two pathways corresponds to the product distribution of propylene and ethylene. In the catalytic cracking process on the zeolite surfaces, the catalytic pathway related to ethane–propylene production is more pronounced with respect to the formation of ethylene–propane. Furthermore, a significant difference in activation energy between the two pathways of ethane–propylene and ethylene–propane production, in particular in the case of H-FER, demonstrates the higher distribution of propylene compared with ethylene. As can be seen in the experimental results, the ratio of ethane–propylene to ethylene–propane and P/E over H-FER is remarkably high and the further side products due to oligomerization and aromatization are almost negligible, and therefore the monomolecular mechanism would be a predominant pathway. When comparing such experimental results together with the computational data, the ratio of P/E over H-FER is related to a significantly higher difference in activation barriers between the ethane–propylene production through transition state 2 (TS2) and the ethylene–propane production through transition state 2′ (TS2′) where the ethylene–propane pathway requires a significantly higher energy. Therefore, this study opens up perspectives to understand the effect of zeolite frameworks on the product distribution of light olefins obtained from n-pentane catalytic cracking.

4 Conclusions

The n-pentane catalytic cracking mechanism controlling the selectivity of light olefins over various zeolite frameworks such as H-FER, H-ZSM-5, and H-FAU representing small, medium and large pore zeolites was investigated by using the M06-2X/6-31G(d,p) density functional and compared with the experimental results. The reaction mechanism of n-pentane cracking to light olefins proceeds via the monomolecular mechanism, which requires two consecutive steps: (i) the protolytic cracking of n-pentane to form a pentonium intermediate; and (ii) the dissociation of the pentonium intermediate to produce either ethane–propylene or ethylene–propane products. The key step to control the product distribution of light olefins is the dissociation of the pentonium intermediate. The light olefin production from the pentonium intermediate can be divided into two routes: (i) via transition state 2 (TS2) producing ethane–propylene products; and (ii) via transition state 2′ (TS2′) producing ethylene–propane products. The ethane–propylene products are much easier to form with respect to the ethylene–propane products. The differences in the activation energies of the ethane–propylene and ethylene–propane products on H-FER, H-ZSM-5, and H-FAU are 6.7, 5.0, and 0.5 kcal mol−1, respectively. The larger difference value over H-FER implies that the ratio of ethane–propylene to ethylene–propane is significantly higher with respect to the cases of H-ZSM-5 and H-FAU, eventually resulting in the preferable production of a high P/E ratio. These observations correspond to the high proportion of the ethane–propylene compared with the ethylene–propane products, in particular over H-FER, obtained from the experimental measurements. Therefore, this example demonstrates insights into the effect of zeolite frameworks on the reaction mechanism to understand the product distribution of light olefins obtained from n-pentane catalytic cracking.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by grants from Vidyasirimedhi Institute of Science and Technology (VISTEC), PTT Global Chemical Public Company Limited, Thailand Research Fund (TRF) (MRG6180099), and the Office of Higher Education Commission (OHEC). In addition, this work has been partially supported by the National Nanotechnology Center (NANOTEC), NSTDA, Ministry of Science and Technology, Thailand, through its program of Research Network NANOTEC (RNN).

References

  1. Y. Yoshimura, N. Kijima, T. Hayakawa, K. Murata, K. Suzuki, F. Mizukami, K. Matano, T. Konishi, T. Oikawa, M. Saito, T. Shiojima, K. Shiozawa, K. Wakui, G. Sawada, K. Sato, S. Matsuo and N. Yamaoka, Catal. Surv. Jpn., 2001, 4, 157–167 CrossRef.
  2. N. Rahimi and R. Karimzadeh, Appl. Catal., A, 2011, 398, 1–17 CrossRef CAS.
  3. C. Baerlocher, L. B. McCusker, D. H. Olson, C. Baerlocher, A. W. Burton, W. H. Baur, R. W. Broach, D. L. Dorset, R. X. Fischer, H. Gies, R. M. Kirchner, R. F. Lobo, L. B. McCusker, W. M. Meier, R. E. Morris, W. J. Mortier, M. O'Keeffe, D. H. Olson, M. M. J. Treacy, H. van Koningsveld and O. Terasaki, Atlas of Zeolite Framework Types, Elsevier Science B.V., Amsterdam, 2007 Search PubMed.
  4. J. Shi, Y. Wang, W. Yang, Y. Tang and Z. Xie, Chem. Soc. Rev., 2015, 44, 8877–8903 RSC.
  5. P. Pantu, B. Boekfa, B. Sunpetch and J. Limtrakul, Chem. Eng. Commun., 2008, 195, 1477–1485 CrossRef CAS.
  6. M. A. Aramendía, V. Borau, C. Jimenez, J. M. Marinas, R. Roldán, F. Romero and F. Urbano, Chem. Lett., 2002, 672–673 CrossRef.
  7. B. G. Anderson, R. R. Schumacher, R. van Duren, A. P. Singh and R. A. van Santen, J. Mol. Catal. A: Chem., 2002, 181, 291–301 CrossRef CAS.
  8. C. Jiménez, F. J. Romero, R. Roldán, J. M. Marinas and J. P. Gómez, Appl. Catal., A, 2003, 249, 175–185 CrossRef.
  9. K. Föttinger, G. Kinger and H. Vinek, Appl. Catal., A, 2003, 249, 205–212 CrossRef.
  10. J. A. Swisher, N. Hansen, T. Maesen, F. J. Keil, B. Smit and A. T. Bell, J. Phys. Chem. C, 2010, 114, 10229–10239 CrossRef CAS.
  11. T. Maihom, P. Pantu, C. Tachakritikul, M. Probst and J. Limtrakul, J. Phys. Chem. C, 2010, 114, 7850–7856 CrossRef CAS.
  12. X. Hou, Y. Qiu, X. Zhang and G. Liu, Chem. Eng. J., 2017, 307, 372–381 CrossRef CAS.
  13. O. Bortnovsky, P. Sazama and B. Wichterlova, Appl. Catal., A, 2005, 287, 203–213 CrossRef CAS.
  14. Z. Liu, Y. Chu, X. Tang, L. Huang, G. Li, X. Yi and A. Zheng, J. Phys. Chem. C, 2017, 121, 22872–22882 CrossRef CAS.
  15. Z. Liu, X. Yi, G. Wang, X. Tang, G. Li, L. Huang and A. Zheng, J. Catal., 2019, 369, 335–344 CrossRef CAS.
  16. B. Song, Y. Chu, G. Li, J. Wang, A.-Y. Lo, A. Zheng and F. Deng, J. Phys. Chem. C, 2016, 120, 27349–27363 CrossRef CAS.
  17. S. Kotrel, H. Knözinger and B. C. Gates, Microporous Mesoporous Mater., 2000, 35–36, 11–20 CrossRef CAS.
  18. A. Corma and A. V. Orchillés, Microporous Mesoporous Mater., 2000, 35–36, 21–30 CrossRef CAS.
  19. W. O. Haag and R. M. Dessau, Proceedings, 8th International Congress on Catalysis Berlin, 1984, 2, 305–316.
  20. D. B. Lukyanov, V. I. Shtral and S. N. Khadzhiev, J. Catal., 1994, 146, 87–92 CrossRef CAS.
  21. H. Van Koningsveld, H. Van Bekkum and J. C. Jansen, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 1987, 43, 127–132 CrossRef.
  22. R. E. Morris, S. J. Weigel, N. J. Henson, L. M. Bull, M. T. Janicke, B. F. Chmelka and A. K. Cheetham, J. Am. Chem. Soc., 1994, 116, 11849–11855 CrossRef CAS.
  23. D. H. Olson and E. Dempsey, J. Catal., 1969, 13, 221–231 CrossRef CAS.
  24. W. Thongnuam, T. Maihom, S. Choomwattana, Y. Injongkol, B. Boekfa, P. Treesukol and J. Limtrakul, Phys. Chem. Chem. Phys., 2018, 20, 25179–25185 RSC.
  25. B. Boekfa, S. Choomwattana, P. Khongpracha and J. Limtrakul, Langmuir, 2009, 25, 12990–12999 CrossRef CAS.
  26. Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157–167 CrossRef CAS.
  27. Y. Zhao and D. G. Truhlar, J. Phys. Chem. C, 2008, 112, 6860–6868 CrossRef CAS.
  28. G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, Gaussian, Inc., Wallingford, CT, USA, 2009.
  29. S. Shetsiri, A. Thivasasith, K. Saenluang, W. Wannapakdee, S. Salakhum, P. Wetchasat, S. Nokbin, J. Limtrakul and C. Wattanakit, Sustainable Energy Fuels, 2019, 3, 115–126 RSC.
  30. W. Wannapakdee, D. Suttipat, P. Dugkhuntod, T. Yutthalekha, A. Thivasasith, P. Kidkhunthod, S. Nokbin, S. Pengpanich, J. Limtrakul and C. Wattanakit, Fuel, 2019, 236, 1243–1253 CrossRef CAS.
  31. A. Thivasasith, T. Maihom, S. Pengpanich, J. Limtrakul and C. Wattanakit, Phys. Chem. Chem. Phys., 2019, 21, 5359–5367 RSC.
  32. J. F. Denayer, W. Souverijns, P. A. Jacobs, J. A. Martens and G. V. Baron, J. Phys. Chem. B, 1998, 102, 4588–4597 CrossRef CAS.
  33. B. A. De Moor, M.-F. Reyniers, O. C. Gobin, J. A. Lercher and G. B. Marin, J. Phys. Chem. C, 2011, 115, 1204–1219 CrossRef CAS.
  34. F. Eder, M. Stockenhuber and J. A. Lercher, J. Phys. Chem. B, 1997, 101, 5414–5419 CrossRef CAS.
  35. W. J. M. van Well, X. Cottin, B. Smit, J. H. C. van Hooff and R. A. van Santen, Catal. Lett., 1998, 102, 3952–3958 CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cp03871j

This journal is © the Owner Societies 2019