Hydroisomerization of long chain n-paraffins: the role of the acidity of the zeolite

Abstract

The transformation of n-hexadecane (n-C₁₆) was tested by two series of experiments over bifunctional catalysts with noble metal Pt and a one dimensional ZSM-22 zeolite to investigate the role of the acidity of the zeolite during the hydroisomerization of long chain n-paraffins. In series 1, varying the Si/Al ratio of the zeolite but fixing the zeolite content, reveals that the high acid strength of the zeolite can deteriorate the selectivity of the corresponding catalyst even at an initial conversion and that the low concentration of acid sites on the zeolite is beneficial to the improvement of the maximal value of the isomer yield. In series 2, varying the zeolite content but fixing the Si/Al ratio, reveals that the change of the zeolite content could not affect the selectivity of the catalyst. Both series 1 and series 2 revealed that the activity of the catalyst linearly increases with the acid strength, the concentration of acid sites and the content of the employed zeolite.

---

1. Introduction

Usually, petroleum fractions contain significant amounts of n-paraffins with high pour/melting points, which greatly deteriorates their cold flow properties. Thus, hydroisomerization that can transform long chain n-paraffins into branched paraffins is an important industrial process, especially for the production of high quality lube oil or diesel fuel.^1–4 Generally, the hydroisomerization reaction takes place over bifunctional catalysts containing metallic sites for hydrogenation/dehydrogenation and acidic sites for skeletal isomerization via carbenium ions.^5,6 Due to the acidic sites, the hydroisomerization reaction is always accompanied by a hydrocracking reaction that can lower the yield of the isomerized molecules and cause the degradation of the long chain n-paraffins to less valuable and lighter products.^7–10 Therefore, an efficient hydroisomerization catalyst should minimize the side reaction.

In this sense, more effort has been paid and some achievements have been obtained. For example, it has been reported that a high yield of branched isomers can be obtained when catalysts containing zeolites with 1-dimensional pores and 10-membered ring channels such as ZSM-22, ZSM-23, and SAPO-11 were employed.^11–13 To explain the high selectivity of these materials, many attempts have been made. Amongst them, the pore mouth catalysis and the key-lock catalysis have been popular.^14,15 According to these catalyses, the hydroisomerization of n-paraffins can happen on the pore opening due to the small pore size of these materials. Therefore, only mono-branched isomers with a smaller molecule size can be formed and multi-branched isomers that are susceptible to cracking are suppressed.

Although these theories work well for the explanation of the high isomer yield and the predication of the branched distribution in the product, the role of other aspects of a catalyst should not be ignored. Over the past few decades, many researchers have pointed out that the acidity characteristics of the zeolite contained in the hydroisomerization catalyst has a major influence on the activity and selectivity.^16,17 Furthermore, a common idea is that the proper balance of hydrogenation function and acid function is critical to maximizing the isomer yield. However, a further conclusion is under debate due to the different types of the employed zeolite structure and the reaction conditions, especially for the role of the acid site density and strength. For example, the hydrogen spillover mechanism emphasized that the olefinic intermediates formed on the acidic sites could be hydrogenated by a far Pt cluster, so even a physical mixture of Pt/SiO₂ and zeolite can show a good selectivity.¹⁸ However, N. Batalha et al. considered that the selectivity of a catalyst is determined by the intimacy between the metal and protonic sites i.e., the number of acid sites encountered by the olefinic intermediates during their diffusion between two metallic sites.¹⁶

In this paper, the hydroisomerization activities and selectivities of various catalysts containing the shape selective zeolite ZSM-22, one of the most successful n-paraffin isomerization zeolites, were investigated using n-hexadecane as a model molecule. Due to the elimination of the influence from the pore structure of the zeolite, the relationship between the hydroisomerization behaviors of a catalyst and the zeolite’s acidity characteristics could be well recognized. It is revealed that the acidity characteristics like the acid strength, the concentration of acid sites and the amount of acid sites play different roles in the hydroisomerization of n-paraffins.

2. Experimental

2.1 Zeolite synthesis

The synthesis of ZSM-22 was carried out using 1,6-diaminohexane as the template following the procedure reported in reference.¹⁹ Briefly: 10.64 g of Ludox AS40 (40 wt% silica) was diluted with 18.40 g of distilled water. To this sol, a solution containing X g of Al₂(SO₄)₃·18H₂O, 1.16 g of KOH, 2.48 g of 1,6-diaminohexane, and 26.20 g of distilled water was added under continuous vigorous magnetic stirring, where X was determined by the target Si/Al ratio. The gel’s molar composition was 27NH₂(CH₂)6NH₂ : 13K₂O : XAl₂O₃ : 91SiO₂ : 3670H₂O. The gel was aged for 90 min at room temperature and subsequently the synthesis was carried out under static conditions for 4 days. Finally, the template in the as-synthesized zeolites was removed by calcination in static air at 550 °C for 4 h using a heating rate of 3 °C min⁻¹. Subsequently, the as-made solid was ion exchanged with 1 M NH₄NO₃ solution at 60 °C for 24 h, filtered and then dried at 120 °C for 4 h.

2.2 Catalyst preparation

To prepare the hydroisomerization catalysts, a different amount of zeolite in the ammonium form was mixed with a binder (γ-alumina), this was kneaded with the addition of water containing peptization agent nitric acid. After that, the mixture was extruded in the form of cylindrical pellets (5 mm length, 1.5 mm diameter), dried at 120 °C for 6 h, and, in a further step, calcined at 600 °C for 3 h to obtain the catalyst containing the zeolite in its acidic form. Pt loading was done by wet impregnation using Pt(NH₃)₄Cl₂. For all catalysts, the target Pt loading was fixed at 0.5 wt%. The oxidation and reduction of the catalysts to disperse the metal was done at 400 °C for 4 h before starting the reaction.

2.3 Materials characterization

The crystalline phase identification and the phase purity of the ZSM-22 zeolites were carried out using XRD (Philips, Holland) using nickel-filtered Cu K_α radiation (λ = 1.5406 Å). The surface areas and pore volumes were determined from N₂ adsorption isotherms using a Coulter (Omnisorp 100 CX) instrument. ²⁷Al NMR spectra of the samples were recorded on a Bruker MSL 300 NMR spectrometer. Scanning electron microscopy (SEM) was employed to observe the morphologies of the ZSM-22 zeolites. NH₃ temperature programmed desorption (TPD) experiments were carried out to test the total acidity of the zeolites at atmospheric pressure in a tubular flow reactor using a fixed bed. The Brønsted acid sites of the zeolites and the corresponding catalysts were determined by temperature programmed desorption of pyridine. The dispersion of platinum was estimated by CO adsorption followed by infrared spectroscopy.

2.4 Catalytic test

The catalytic experiment was examined using a single-pass microreactor. The catalysts were crushed and sieved to select particles with a size in the 40–60 mesh range in order to eliminate diffusive effects. 1.5 g of catalyst was loaded in the constant temperature zone. Prior to the catalytic tests, the catalysts were reduced again under a hydrogen pressure of 4.0 MPa (300 mL min⁻¹) for 30 min at 350 °C. After that, the feed was injected. The flow rate of the feed was 0.2 mL min⁻¹. The reaction temperatures were determined by the targeted conversions and the hydrogen pressure was kept at 4.0 MPa. All the reaction products were analyzed off-line using a gas chromatograph (CE 2000) equipped with a capillary column (OV-17, length 25 m, internal diameter 0.25 mm). The data reported in this paper were collected when the steady state was reached (ca. 5 h).

3. Results and discussion

It is well known that the Si/Al ratio is a key factor that effects the acidity characteristics of a zeolite and that is strongly related to the catalytic activity and selectivity of the corresponding catalyst. Therefore, the first series of experiments were arranged by varying the Si/Al ratios of the ZSM-22 contained in the catalysts. Since other properties of a zeolite can be changed due to the varied Si/Al ratio, it is necessary to first investigate the consistency of some key parameters for the employed zeolites. Fig. 1 shows the XRD patterns of the employed ZSM-22 zeolites with various Si/Al ratios. It is clearly revealed that all patterns are well defined and there is no apparent difference in the diffraction patterns, crystallinities or crystal sizes. Further, solid state ²⁷Al NMR was employed to characterize the coordination information of the Al in the samples, as presented in Fig. 2. For the ²⁷Al NMR spectra, a strong signal at about 58 ppm was the only observed prominent peak for all the zeolite samples. This peak corresponds to the tetrahedral coordination of the aluminum and confirms that all aluminum atoms are located in the framework of the ZSM-22 zeolites. This indicates that the Si/Al ratios of the samples can accurately reflect the number of acid sites in the zeolites. On the other hand, the SEM images (Fig. 3) reveal that similar morphologies and sizes can be observed for these zeolites. Thus, it is reasonable to eliminate their influence on the catalytic behaviours of the corresponding catalysts.

Fig. 1 XRD patterns of the ZSM-22 zeolite samples. (a) Si/Al = 28.6; (b) Si/Al = 39.2; (c) Si/Al = 55.4.

Fig. 2 ²⁷Al NMR spectra of the ZSM-22 zeolite samples. (a) Si/Al = 55.4; (b) Si/Al = 39.2; (c) Si/Al = 28.6.

Fig. 3 SEM images of the ZSM-22 zeolites. (a) Si/Al = 55.4; (b) Si/Al = 39.2; (c) Si/Al = 28.6.

The differences in the acidic properties of the zeolites with various Si/Al ratios was evaluated by NH₃-TPD as shown in Fig. 4.

Fig. 4 Temperature programmed desorption of ammonia for the zeolite samples with various Si/Al values: (a) 28.6; (b) 39.2; and (c) 55.4.

The profiles can be differentiated by both the peak areas and the shifts in the peak positions. The former corresponds to the number of acidic sites and the latter indicates the strength of those acidic sites. Fig. 4 shows that two NH₃ desorption peaks are observed over all the profiles of the zeolites. One peak is centred at about 180 °C, the other is centred at 380 °C, corresponding to the weak and strong acid sites, respectively. Further, it can be observed that the peak areas, whether at the lower or higher temperature, increased with the decreasing Si/Al ratio in the zeolites, indicating that more acid sites are present in the zeolites with a lower Si/Al ratio. On the other hand, there is also an obvious peak position shift toward the higher temperature on these profiles with the decreasing Si/Al ratio in the zeolites, suggesting that their are fewer strong acid sites in the zeolite with the high Si/Al ratio.

The TPD of pyridine was employed to assist in the identification of the weak and strong Brønsted acid sites in the ZSM-22 zeolites and the results are listed in Table 1. Reasonably, the number of Brønsted acid sites increased with the decreasing Si/Al ratio in the zeolite samples. However, it should be emphasized that Table 1 also revealed a slight increasing trend for the ratio of strong Brønsted acid sites compared to the total Brønsted acid sites (defined as S/T) with the decreasing Si/Al ratio in the zeolite samples, indicating more strong Brønsted acid sites in the zeolites with lower Si/Al ratios.

Table 1 Results from pyridine adsorption–desorption²⁰

Zeolites	Si/Al	Total Brønsted^a acid/(μmol g^−1)	Srong Brønsted^b acid/(μmol g^−1)	S/T^c
a Obtained by the desorption of pyridine at 200 °C.b Obtained by the desorption of pyridine at 350 °C.c Defined as strong Brønsted acid sites compared to total Brønsted acid sites.
Z-1	28.6	298.9	201.7	0.67
Z-2	39.2	247.6	149.1	0.60
Z-3	55.4	189.6	112.7	0.59

---

Based on the similar textural properties and the different acidity characteristics of the employed zeolites, the hydroisomerization catalysts were prepared at a fixed zeolite content (about 60 wt%) and tested for the transformation of n-hexadecane to investigate the role of the Si/Al ratio of the zeolites. Table 2 lists the main physical and chemical properties of the corresponding catalysts. As shown in Table 2, all of the key parameters of these catalysts are similar in spite of the different zeolites employed in them, so it is possible that only acidity characteristics be considered for the interpretation of the different hydroisomerization behaviours of these catalysts.

Table 2 Physical and chemical properties of the prepared catalysts

Catalysts	Zeolites	Pt loading/%	Pt dispersion/%	SBET/(m^2 g^−1)	Vpore/(mL g^−1)
Cata-1	Z-1	0.51	68	223	0.46
Cata-2	Z-2	0.51	67	220	0.47
Cata-3	Z-3	0.51	68	221	0.47

---

Fig. 5 shows the catalytic activities of the catalysts for the conversion of n-hexadecane. Combined with the Si/Al ratios listed in Table 1, it is clearly revealed that the catalytic activities of these catalysts are strongly related to their concentration of acid sites and the acid strength on the contained zeolites. The catalytic activity increases with the decreasing Si/Al ratios whatever the reaction temperature. According to the hydroisomerization/hydrocracking scheme, this is reasonable considering the greater formation of carbenium ions on the greater number of acid sites and strong acid sites. However, previous works have pointed out that this easily causes more hydrocracking of the carbenium ions.^5,7 Thus, it is necessary to investigate the selectivities of these catalysts. Fig. 6 shows the relationship of the isomerized hexadecane yield against the conversion of n-hexadecane over the various catalysts. As shown in Fig. 6, the curves representing the yields of the isomers firstly increase as a function of the conversion and then they rapidly decrease after passing through a maximum over all the catalysts. After a close examination, two interesting points are observed in Fig. 6. Firstly, it can be seen that the curve (a) for Cata-1 is lower than curve (b) for Cata-2 and curve (c) for Cata-3 even at low conversions. This can be explained based on the strong acid strength of the zeolite contained in (a) as listed in Table 1. According to previous works, the isomerization proceeds through several successive steps.^5,7,14 Amongst them, the isomerization of n-alkenes into i-alkenes on the acid site is a key step. High acid strength allows for a relatively longer residence time of the i-alkenes intermediates on the acid sites, thus providing sufficient time for the intermediates to be cracked. Overall, the first conclusion to be drawn is this: the high acid strength of the zeolite deteriorates the isomer selectivity of the corresponding catalyst even at the initial conversions. Secondly, Fig. 6 shows that the maximal values for the isomer yields over Cata-2 and Cata-3 are different. This confirms that a high concentration of acid sites on the zeolite is not beneficial to the improvement of the maximal isomer yield of the corresponding catalyst even with a similar acid strength. A possible reason is that at a high conversion, the i-alkene intermediates, with their large size, are more numerous and diffuse slowly. Once the concentration of the acid sites is higher, the i-alkene intermediates meet more acid sites and the cracking is boosted. Overall, the second conclusion is this: at a similar acid strength, a higher concentration of acid sites on the zeolite deteriorates the maximal isomer yield of the corresponding catalysts. Additionally, this conclusion implies that we can obtain the maximal isomer yield by increasing the Si/Al ratio of the zeolite. However, it should be noted that this will cause a higher reaction temperature and increase energy consumption.

Fig. 5 Activities of the catalysts containing ZSM-22 with different Si/Al ratios at various temperatures. (a) Cata-1; (b) Cata-2; (c) Cata-3.

Fig. 6 Isomer yield vs. n-C₁₆ conversion for the catalysts containing ZSM-22 with different Si/Al ratios. (a) Cata-1; (b) Cata-2; (c) Cata-3.

Beside the acid strength and the concentration of the acid sites, the zeolite content is another key factor affecting the catalytic behaviors. Thus, the second series of experiments were arranged by the variation of the zeolite content contained in the catalysts but keeping a fixed Si/Al ratio = 55.4. Table 3 lists the main physical and chemical properties of the corresponding catalysts. As shown in Table 3, only slight changes for the pore parameters of the catalysts like S_BET and V_pore and the dispersion of loaded Pt were observed. Apparently, a decrease in the amount of acid sites happens with the decreasing of the zeolite content.

Table 3 Physical and chemical properties of the catalysts

Catalysts	Zeolite content	Pore parameters		Acidity
		SBET (m^2 g^−1)	Vpore (mL g^−1)	Total Brønsted^a acid/(μmol g^−1)	Strong Brønsted^b acid/(μmol g^−1)
a Obtained by the desorption of pyridine at 200 °C.b Obtained by the desorption of pyridine at 350 °C.
Cata-40	40	224	0.44	153.6	88.2
Cata-60	60	217	0.43	247.6	149.1
Cata-70	70	208	0.43	287.5	166.2

---

Fig. 7 shows the hydroisomerization activities of the employed catalysts. As shown in Fig. 7, it can be observed that the activities of the catalysts are strongly related to the zeolite content within them. At the same reaction temperature, the more zeolite the catalyst contains , the higher activity it has. An interesting point concerns the isomer yields over the various catalysts. It can be inferred that the distance between the acid sites in the catalysts was larger due to the decrease of zeolite content. On the other hand, the amount and the dispersion of the loaded Pt was kept stable for all catalysts. Thus, more hydrogenation of the olefinic intermediates during their diffusion, giving the final i-paraffinic products, can be expected. However, a similar trend for isomer yields over all catalysts is observed in our case (as shown in Fig. 8), indicating the same lifetime for the olefinic intermediates for all catalysts. This can be explained based on the local environment of the acid sites in the catalysts. Actually, there is no change in the number of metallic sites surrounding each acid site in spite of the varied zeolite content in the catalysts. This means that the degree of intimacy between the acidic sites and Pt is unvaried in our case. Overall, the third conclusion is drawn that: the zeolite content cannot affect the selectivity of the corresponding catalyst due to the unchanged intimacy between the acid sites and the metal sites. Additionally, this means that to compensate for the low activity caused by the high Si/Al ratio of the zeolite contained in the catalyst, we can increase the zeolite content in it until the crushing strength of the support is allowed. This is meaningful to prepare an efficient hydroisomerization.

Fig. 7 Activities of the catalysts containing different zeolite content at various temperatures. (a) Cata-70; (b) Cata-60; (c) Cata-40.

Fig. 8 Isomer yield vs. n-C₁₆ conversion for the catalysts containing different zeolite content at various temperatures. (a) Cata-70; (b) Cata-60; (c) Cata-40.

4. Conclusions

The hydroisomerization activities and selectivities of catalysts containing various ZSM-22 zeolites with different acidity characteristics have been tested. Based on these catalysts, the role of the acidity characteristics of a zeolite contained in the catalyst during the hydroisomerization of long n-paraffins was investigated and several points have been clearly outlined due to the elimination of any textural difference: firstly, a high acid strength of a zeolite can deteriorate the selectivity of the corresponding catalyst even at an initial conversion. Secondly, the low concentration of acid sites in a zeolite is beneficial to obtain a maximal value of isomer yield at a similar acid strength. Thirdly, the zeolite content in a catalyst cannot affect its selectivity. Finally, the activity of the catalysts linearly increased with the increase in acid strength, the concentration of acid sites and the zeolite content. We think these results pave a way to develop an efficient hydroisomerization catalyst.

Acknowledgements

This work was supported by the project of the major research plan of the Sinopec corporation (Grant no. 113017). Authors thank the discussion from Prof. Li Mingfeng and Prof. Yang Qinghe.

References

1. F. Bauer, K. Ficht, M. Bertmer, W. Einicke, T. Kuchling and R. Gläser, Catal. Sci. Technol., 2014, 4, 4045–4054.
2. D. Verboekend, K. Thomas, M. Milina, S. Mitchell, J. Pérez-Ramírez and J. P. Gilson, Catal. Sci. Technol., 2011, 1, 1331–1335.
3. J. A. Martens, D. Verboekend, K. Thomas, G. Vanbutsele, J. P. Ramírez and J. P. Gilson, Catal. Today, 2013, 218–219, 135–142.
4. R. Yadav and A. Sakthivel, Appl. Catal., A, 2014, 481, 143–160.
5. F. Alvarez, F. R. Ribeiro, G. Perot, C. Thomazeau and M. Guisnety, J. Catal., 1996, 162, 179–189.
6. A. Patrigeon, E. Benazzi, C. Travers and J. Y. Bernhard, Catal. Today, 2001, 65, 149–155.
7. M. Hochtl, A. Jentys and H. Vinek, J. Catal., 2000, 190, 419–432.
8. M. Y. Kim, K. Lee and M. Choi, J. Catal., 2014, 319, 232–238.
9. J. A. Martens, P. A. Jacobs and J. Weitkamp, Appl. Catal., 1986, 20, 239–281.
10. G. G. Martens, G. B. Marin, J. A. Martens, P. A. Jacobs and G. V. Baron, J. Catal., 2000, 195, 253–267.
11. C. H. Geng, F. Zhang, Z. X. Gao, L. F. Zhou and J. L. Zhou, Catal. Today, 2004, 93/95, 485–491.
12. K. C. Park and S. K. Ihm, Appl. Catal., A, 2001, 203, 201–209.
13. S. W. Lee and S. K. Ihm, Ind. Eng. Chem. Res., 2013, 52, 15359–15365.
14. J. A. Martens, G. Vanbutsele, P. A. Jacobs, J. Denayer, R. Ocakoglu, G. Baron, J. A. Muñoz Arroyo, J. Thybaut and G. B. Marin, Catal. Today, 2011, 65, 111–116.
15. H. Deldari, Appl. Catal., A, 2005, 293, 1–10.
16. N. Batalha, L. Pinard, C. Bouchy, E. Guillon and M. Guisnet, J. Catal., 2013, 307, 122–131.
17. S. Gopal and P. G. Smirniotis, J. Catal., 2004, 225, 278–287.
18. T. Kusakari, K. Tomishige and K. Fujimoto, Appl. Catal., A, 2002, 224, 219–228.
19. D. Verboekend, A. M. Chabaneix, K. Thomas, J. P. Gilson and J. P. Ramirez, CrystEngComm, 2011, 13, 3408–3416.
20. C. A. Emeis, J. Catal., 1993, 141, 347–354.

---