- Email: [email protected]
One-pot synthesis of high silica MCM-22 zeolites and their performances in catalytic cracking of 1-butene to propene
Microporous and Mesoporous Materials 118 (2009) 44–51
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
One-pot synthesis of high silica MCM-22 zeolites and their performances in catalytic cracking of 1-butene to propene Guoliang Xu a,b, Xiangxue Zhu a, Xionglei Niu a,b, Shenglin Liu a, Sujuan Xie a, Xiujie Li a, Longya Xu a,* a b
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Liaoning, Dalian 116023, PR China Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e
i n f o
Article history: Received 31 January 2008 Received in revised form 5 August 2008 Accepted 6 August 2008 Available online 17 August 2008 Keywords: High silica MCM-22 Synthesis 1-butene Catalytic cracking Propene
a b s t r a c t MCM-22 zeolites with high SiO2/Al2O3 (Si/Al2) ratios were synthesized by a one-pot procedure with the assisting of boron. Based on this boron-containing method, the Si/Al2 ratios of the MCM-22 zeolites could be greatly extended from 30 to 600 in control by tuning the Si/Al2 ratios of the starting gels. ICP-AES and 11 B MAS NMR results demonstrated that a small part of the boron still existed in the framework of the MCM-22 after calcination and ammonium exchange. However, the residual boron species had little influence on the physicochemical properties of the MCM-22 samples evidenced by XRD, BET, TG–DTA, NH3– TPD, and Py-IR determinations. The reaction of catalytic cracking of 1-butene to propene was carried out to study the performances of the high silica MCM-22 catalysts. Increasing of the Si/Al2 ratios in the MCM22 zeolites suppressed the formation of byproducts such as propane and aromatics effectively, and the optimal selectivity towards propene was obtained on the sample with a Si/Al2 ratio of 158. Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction MCM-22, a kind of layered zeolite invented by Mobil in 1990 [1], belongs to MWW type and possesses two independent pore systems. One consists of two-dimensional sinusoidal channels composed of slightly elliptical 10-membered rings (MR), and the other is characteristic of 12-MR supercages accessible by 10-MR windows [2]. Due to its unique structure and physicochemical properties, MCM-22 has been widely used in many hydrocarbon catalytic transforming processes such as isomerization [3], alkylation [4], aromatization [5] and cracking [6,7]. Recently, MCM-22 has been found to be a rather active and selective catalyst in the process of catalytic cracking of C4+ alkenes to propene, with which the C4+ streams from petrochemicals can be upgraded and the production of propene can be enhanced [7–9]. Conventionally, MCM-22 zeolites were synthesized under hydrothermal conditions using hexamethyleneimine (HMI) as an organic structure directing agent (OSDA) [10–12]. Most of the research works have revealed that suitable SiO2/Al2O3 (Si/Al2) ratios for the formation of MCM-22 zeolites are ranging from 15 to 80, preferably 20 to 50 [10–14]. Thus, MCM-22 catalysts with such Si/Al2 ratios might be unsuitable in some applications, especially in reactions which need only mild acidity, for instance, the cata-
* Corresponding author. Tel./fax: +86 411 84693292. E-mail addresses: [email protected], [email protected], [email protected] (L. Xu). 1387-1811/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2008.08.019
lytic cracking of the C4+ alkenes to propene [15,16] and the isomerization of the n-alkenes to their branched isomers [3]. During the past several years, much effort had been dedicated to enlarge the Si/Al2 ratios of the MCM-22 zeolites, of which post-treatments to low silica MCM-22 were widely applied including hydrothermal treating [17], organic or inorganic acid leaching [18,19] and silanizing with SiCl4 [20]. Previously, the Si/Al2 ratios of MCM-22 were effectively enhanced to 133 by post-treatment using ammonium hexafluorosilicate in our earlier work [15]. However, this method encountered a dilemma that the zeolite structure may be partially destroyed upon higher Si/Al2 ratios of the products. Different from the procedures mentioned above, Bao et al. succeeded in synthesizing MCM-22 zeolites with high Si/Al2 ratios (high silica MCM-22, Si/Al2 = 24–1000) via a complicated threestep strategy: first, preparation of ERB-1 (MWW borosilicate); then, removal of the boron species by calcination and HNO3 reflux; and finally, alumination of the deboronated ERB-1 with further hydrothermal treatment [21]. On their studying of the intercalating properties of ERB-1, Millini et al. have suggested a boron-containing approach to obtain high silica MWW zeolites using piperidine as an OSDA [22]. Inspired by their work, a one-pot procedure for the synthesis of high silica MCM-22 has been developed by our group, which utilized the assisting action of boron and HMI. In this paper, this one-pot synthesis method is described in detail. The existing states of boron in the zeolites as well as the deboronation process, and their influence on the physicochemical properties of the MCM-22 samples,
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
were studied. Also, the performances of the high silica MCM-22 catalysts for catalytic cracking of 1-butene were investigated.
2. Experimental 2.1. Synthesis of the MCM-22 zeolites
45
Thermo-gravimetric and differential thermal analyses (TG– DTA) of the as-synthesized MCM-22 samples were performed on a Perkin-Elmer Diamond TGS-2 and DTA1700 apparatus. The experiments were carried out in the temperature range of 293– 1323 K, with a heating rate of 10 K/min in flowing air (100 ml/ min). Acidity of the H-form MCM-22 samples was determined by the temperature programmed desorption of ammonia (NH3–TPD) and pyridine adsorption infrared spectra (Py-IR) techniques. The measurements were carried out using the same procedures as in reference [7].
The one-pot procedure (hereafter referred to as the boron-containing method to distinguish it from the conventional boron free method) for the synthesis of high silica MCM-22 zeolites is as follows: first, silica sol (25.6 wt% SiO2, 0.06 wt% Al2O3, Qingdao Haiyang Chemical Co. Ltd.), Al(NO3)3 9H2O (99.0 %, Shanghai Zhenxin Reagent Factory), sodium hydroxide (96.0%, Shenyang Chemical Reagent Factory), and H3BO3 (99.5%, Shenyang Chemical Reagent Factory), together with HMI (95.0%, Sheyang Chemical Factory) were well mixed and stirred vigorously for half an hour at room temperature to give a gel with molar composition of 1.0SiO2:(1/ r)Al2O3:0.1Na2O:0.5B2O3:1.0HMI:40.0H2O (r represents the Si/Al2 ratio). Then, the gel was transferred into a stainless-steel autoclave, heated to 448 K and held at this temperature for 168 h while being rotated at 60 rpm. After completion of the crystallization, the autoclave was cooled by tap water and the solid product was obtained by filtrating, washing and drying. The samples synthesized were designated as BMCM-22-r, e.g., BMCM-22-150. For comparison, the synthesis of boron free samples was conducted according to the conventional method using the same raw materials of silicon, aluminum, alkaline and HMI mentioned above. The final gel had a molar ratio of 0.1Na2O: 1.0SiO2: (1/r)Al2O3: 0.35HMI: 40H2O, and was crystallized at 443 K for 48 h. Similar to the boron-containing samples, the samples of this series were designated as MCM-22-r.
Catalytic cracking of 1-butene over the catalysts were carried out in a continuous-flow stainless-steel fixed-bed reactor with an inner diameter of 7 mm. Before a catalytic evaluation, the catalyst was pretreated at 773 K for 1 h at a constant flow of N2 (20 ml/ min). Then, 1-butene (>99.5 wt%, Qilu Petrochemical Co. Ltd.) was introduced at the desired temperature. The products were analyzed on-line by a Varian 3800 GC equipped with a FID and a 100 m PONA column. All isomerized butenes in the products were considered as the unconverted feed stocks. The conversion of the butenes (X) and the selectivity (S) of a particular product CxHy were calculated as follows:
2.2. Preparation of the catalysts
3. Results and discussion
The as-synthesized samples were calcined at 823 K for 3 h in air for burning off the OSDA. Then the as-calcined zeolites were ion exchanged triply with 0.8 mol/L of an NH4NO3 solution, dried and calcined at 823 K for 2 h to give the H-form zeolites. Finally, the Hform zeolites were pelletized, crushed and sieved to 20–40 mesh particles before employing in the catalytic cracking of 1-butene.
3.1. Synthesis of high silica MCM-22 zeolites
2.3. Characterization of the zeolites X-ray diffraction (XRD) patterns of the samples were recorded on an X Pert Pro X-ray diffractometer using Cu Ka radiation and operating at 40 kV and 40 mA. Scanning electron microscopy (SEM) images were recorded on a Quanta 200F scanning electron microscopy (FEI Company) at 30 kV. Chemical compositions of the zeolites were analyzed on an inductively coupled plasma atomic emission spectrometer (ICPAES, TJA, IRIS Advantage). Prior to the measurements, the samples were dissolved in a mixed solution of HF (1.0 mol/L) and HCl (0.5 mol/L) at room temperature in sealed plastic bottles to avoid volatilization of the SiF4 and BF3. Nitrogen adsorption measurements were carried out at 77 K with a Micrometrics ASAP 2010 equipment. The samples were degassed at 573 K and 103 Pa for 4 h prior to the adsorption measurements. The specific surface areas and micropore volumes were obtained by the BET and the t-plot methods, respectively. 11 B magic-angle spinning nuclear magnetic resonance (MAS NMR) spectra were collected on an Infinityplus-400 spectrometer at 128.3 MHz using a 4 mm ZrO2 rotor with a spinning rate of 10 kHz. The spectra were recorded for 200 scans each sample with a p/4 flip angle and 2 s pulse delay, and the chemical shifts being referenced to H3BO3.
2.4. Catalytic evaluation
X = (C4H8% in the feed C4H8% in the products)/(C4H8% in the feed), P S (CxHy) = (CxHy% in the products)/( CiHj% in the products), where all the percentages were in weight percent.
A series of high silica MCM-22 samples with various Si/Al2 ratios (30–600) were synthesized by the boron-containing method. XRD patterns of the as-synthesized and the as-calcined BMCM22-r zeolites were exhibited in Fig. 1. As shown in Fig. 1A, all the as-synthesized samples showed both sharp and broad peaks which matched well with those of the layered precursors of MCM-22 (also designated as MCM-22(P)) [23–25]. Also, XRD patterns of the corresponding as-calcined samples (Fig. 1B) exhibited exact diffraction peaks of the MCM-22 (PDF Card number: 490627). After calcination, one noticeable phenomenon was that the 0 0 l peaks shifted to higher 2h degrees with reduced intensities, which was regarded as indications for the process of 2D layered MCM-22 precursors transforming to 3D microporous structures [22,24]. In addition, the XRD peaks of the as-synthesized and the corresponding as-calcined samples became sharper and the intensities increased gradually with increasing of the Si/ Al2 ratios, especially for the 0 0 l peaks (e.g. 001, 002 and 004). As the XRD peaks are correlated with the ordering of the structure, this phenomenon undoubtedly suggests a more ordered crystal structure for the samples with higher Si/Al2 ratios, especially along the c-dimension. The SEM images of several as-synthesized samples were given in Fig. 2. It can be seen that the crystals of the MCM-22 zeolites mainly existed in typical disk-like shape. From literature [26], we know that the c-axis of MCM-22 is perpendicular to the plate surface. From Fig. 2, one noticeable phenomenon was observed that the thickness of the plates, viz. the c-dimension of the MCM-22 zeolite, increases obviously with the increasing of the Si/Al2 ratios, which is quite consistent with the XRD results discussed above.
46
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
zeolites by the conventional method. Thus, it is tentatively suggested that boron could play a similar role as aluminum in the synthesis of the BMCM-22 samples. Besides the Si/(B2 + Al2) ratios, HMI also plays an important role in the formation of the MCM-22 precursors, which can be divided into two aspects [10,21]. One part of the HMI serves as a framework charge compensating agent in the protonated form (HMI+) and the other as a framework supporting agent in the neutral molecular form. In the conventional synthesizing system, the negative charges of the MCM-22 precursors are originated from either the Si–O–Al sites (formed by aluminum atoms inserted into the framework) or the Si–O defect sites (formed by the reaction between the OH and silanol groups). If the aluminum content is relatively high (Si/Al2 < 15), the amount of the HMI+ required for the formation of stable MCM-22 precursors is not sufficient to match the amount of the Si–O–Al sites, leading to the formation of other phases such as MOR. When the aluminum content is relatively low (Si/Al2 > 50), the Si–O–Al sites would be totally compensated by HMI+. At the same time, a number of Si–O defect sites are produced to compensate the surplus HMI+, which makes the MCM22 precursor unstable and results in the formation of other phases such as kenyaite and ZSM-5. As for the boron-containing system with low aluminum contents, the boron can also enter the framework of MCM-22 (proved by the 11B MAS NMR results below) and form the Si–O–B sites similar to the behaviors of aluminum. As the Si/(B2 + Al2) ratios are in the range of 15–50, the amount of the framework charges (Si–O–B together with the Si–O–Al sites) matches well with the amount of the HMI+ required for the formation of stable MCM-22 precursors, leading to the successful synthesis of MCM-22 zeolites with high Si/Al2 ratios (up to 600). 3.3. Physicochemical properties of high silica MCM-22 zeolites
Fig. 1. XRD patterns of (A) as-synthesized and (B) as-calcined B-MCM-22-r samples, where r is (a) 30, (b) 40, (c) 50, (d) 100, (e) 150, (f) 200, (g) 300 and (h) 600, respectively.
3.2. Difference in optimal Si/Al2 ratio range for the synthesis of MCM22 between the boron-containing and the conventional methods For comparison, several samples (Si/Al2 ranging from 15 to 90) were synthesized by the conventional methods, and the XRD patterns of the as-calcined samples were presented in Fig. 3. When the Si/Al2 ratio in the gel was as low as 15 or higher than 60, MOR or MCM-22 with impurities (i.e., ZSM-5 and kenyaite) was synthesized, respectively, which was consistent with the results obtained in the literatures [12,27]. Only two pure MCM-22 samples were obtained with the Si/Al2 ratios of 30 and 50, which were among the favorable Si/Al2 range (20–50) for the synthesis of MCM-22 zeolites [11,12,27]. By comparing the results of the boron-containing and the conventional methods, it can be deduced that the existence of boron in the synthesizing system favors the formation of the MCM-22 framework within a wider Si/Al2 range. To investigate the nature of the difference in optimal Si/Al2 ranges between the two methods, the Si/(B2 + Al2) ratios in the as-synthesized BMCM-22 samples were first considered. Table 1 listed the chemical compositions of several BMCM-22 samples. The results revealed that the Si/(B2 + Al2) ratios of the samples were in the range of 15–50, which was just the suitable Si/Al2 range for the synthesis of MCM-22
3.3.1. Texture and chemical compositions of high silica MCM-22 The textural properties of several as-calcined high silica MCM22 samples were given in Table 2. As can be seen, each sample showed a total specific surface area more than 460 m2/g, a microporous surface area larger than 339 m2/g and a micropore volume beyond 0.16 cm3/g. Moreover, no obvious decrease in the surface area and the micropore volume was observed with the increasing of the Si/Al2 ratios from 50 to 600, which was one merit of the boron-containing method as compared with those post-treatment methods [15]. In the synthesis of MCM-22 zeolites by the conventional method, the Si/Al2 ratios of the products were usually lower than those of the starting gels [9,12], which was possibly due to inefficient utilization of the silicon atoms. This problem could well be resolved by the boron-containing method. As shown in Fig. 4, the data points were very close to the 1:1 line, indicating that the Si/Al2 ratios of the products were close to those of the starting gels. As has been discussed above, in the boron-containing system, the formation of the Si–O–B sites inhibits the formation of Si–O defect sites, thus the utilization efficiency of silicon is improved, making it possible to control the Si/Al2 ratios of the MCM-22 samples precisely. 3.3.2. Existing states of boron in high silica MCM-22 It has been reported that deboronation could occur easily during the post-treatments of the boron-containing zeolites [28,29]. In this section, the existing states of boron in the high silica MCM-22 zeolites were investigated by the 11B MAS NMR technique. The sample BMCM-22-150 with a final Si/Al2 ratio of 158 was chosen as the specimen, and the 11B MAS NMR spectra of its as-synthesized, as-calcined and H-form samples were presented in Fig. 5. From Fig. 5a, there was only one main resonance peak at 3 ppm in the as-sythesized sample which was assigned to the tet-
47
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 2. SEM images of the as-synthesized samples of (a) BMCM-22-50, (b) BMCM-22-150 and (c) BMCM-22-600.
Table 1 Chemical compositions of selected as-synthesized BMCM-22 samples Sample
Si/Al2
Si/B2
Si/(B2 + Al2)
BMCM-22-30 BMCM-22-50 BMCM-22-150 BMCM-22-600
30 52 158 573
37 27 25 44
17 18 22 41
Table 2 Textural properties of the as-calcined high silica MCM-22 samples Sample
SBET (m2/g)
Vmicropore (cm3/g)
Smicro
Stotal
BMCM-22-50 BMCM-22-150 BMCM-22-600
354 339 378
499 477 464
0.168 0.162 0.183
Note: SBET = surface area, Smicro = microporous surface area, Stotal = total surface area, Vmicropore = micropore volume.
Fig. 3. XRD patterns of the as-calcined MCM-22-r samples, where r is (a) 15, (b) 30, (c) 50, (d) 60, (e) 70 and (f) 90, respectively. Note: (a) is an XRD pattern of MOR. Peaks belong to impurities in (d), (e), and (f): (;) Kenyaite and (*) ZSM-5.
ra-coordinated, i.e., framework boron [29]. After calcination, the peak intensity of framework boron decreased slightly and three
new signals centered at 0, 4 and 15 ppm appeared as shown in Fig. 5b. The first two peaks were attributed to the tri-coordinated framework boron in zeolite and the broad resonance at 15 ppm was associated with the extra-framework tri-coordinated boron [29]. It can be deduced that part of the tetra-coordinated boron was extracted from the framework during the OSDA removal pro-
48
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 4. Evolution of Si/Al2 ratios in the resulting BMCM-22 samples and in the starting gels. Note: the straight line is a 1:1 line.
Fig. 6. XRD patterns of (a) as-calcined and (b) H-form BMCM-22-150.
the calcination and ammonium exchange processes, but it had little influence on the structure of the high silica MCM-22.
Fig. 5. 11B MAS NMR profiles of (a) as-synthesized, (b) as-calcined and (c) H-form BMCM-22-150.
cess. In Fig. 5c, peaks at 0, 4 and 15 ppm of the H-form MCM-22 became very weak (as shown in the inset of Fig. 5 in which the signal was amplified by 4 times), indicating that there was little amount of tri-coordinated boron in the H-form sample. At the same time the peak intensity of tetra-coordinated boron decreased a lot as compared with that in Fig. 5b, suggesting that serious deboronation occured in the ammonium exchange processes. This was further proved by the ICP-AES results which revealed that the Si/B2 ratio increased from 25 in the as-synthesized form to 180 in the H-form. The XRD patterns of the as-calcined and the H-form BMCM-22150 samples were presented in Fig. 6 to examine the effects of the deboronation on the crystalline phases. As shown in Fig. 6, the intensities of the peaks at 2h = 22.9 ± 0.1° and 26.2 ± 0.1° of the H-form sample were a little weaker than those of the as-calcined form, indicating that during the ammonium exchange process the relative crystallinity of the sample decreased slightly, which was possibly due to the deboronation. Even so, both the XRD curves of the as-calcined and the H-form BMCM-22-150 matched exactly with the XRD patterns of the MCM-22 (PDF Card number: 49-0627), indicating that they possessed good 3D structures of the MCM-22. From the 11B MAS NMR, ICP-AES and XRD results discussed above, it could be concluded that deboronation occurred during
3.3.3. Thermal stability and acidity of high silica MCM-22 zeolites For zeolite catalysts, thermal stability and acidity are the most important properties for their catalytic applications. In this section, we compare the thermal stability and the acidity between the BMCM-22-40 and MCM-22-50, both of which possess close final Si/Al2 ratios of 45 and 40, respectively. Two methods were employed to study the thermal stability of the samples, i.e., a calcination method and a TG–DTA method. For the calcination method, the samples were calcined in air at different temperatures for 1 h, followed by XRD analyses (Fig. 7). As can be seen, both the BMCM-22-40 and the MCM-22-50 samples calcined below or equal to 1173 K maintained a MCM-22 crystalline phase, and the relative crystallinity decreased slightly with the increasing of the calcination temperature. However, both of the samples became an amorphous phase when the calcination temperature reached 1273 K. TG–DTA experiments were performed to further investigate the influence of the existing boron on the thermal stabilities of the MCM-22 zeolites. As shown in Fig. 8, the as-synthesized BMCM22-40 and the MCM-22-50 had roughly similar TG and DTA profiles. The obvious weight loss with broad endothermic DTA peaks before 430 K was attributed to desorption of physically adsorbed water. The weight loss between 430 K and 580 K might be caused by several processes, such as desorption of strongly adsorbed water, desorption of HMI [24], and condensation of the interlayer silanol groups along the c-axis [22]. Thus, the appearing (as shown in Fig. 8a at 526 K) or not appearing (Fig. 8b) of the weak exothermic DTA peaks at this temperature range depends on the comprehensive effects of the reactions mentioned above. Further investigation is underway to make this problem more clearly. The three obvious weight loss stages accompanied with obvious exothermic DTA peaks ranging from 580 K to 970 K were attributed to the oxidation of the occluded OSDA in the MCM-22 samples [21]. In agreement with the results of the calcination experiments, weak exothermic DTA peaks without weight losses in the TG curves at about 1240 K, ascribed to the framework destruction of the two samples, were observed. The NH3–TPD technique was employed to determine the acidity of the two H-form samples of BMCM-22-40 and MCM-22-50, and the profiles were shown in Fig. 9. The relative intensity and the distribution of the NH3 desorption peaks of the two samples were found to be very similar. According to Katada’s approach, the TPD profile of each sample was fitted into two peaks [30], and the
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
49
Fig. 7. XRD patterns of (a) BMCM-22-40 and (b) MCM-22-50 calcined at different temperatures for 1 h.
dashed curves in Fig. 9 was an example of the deconvoluted profiles of BMCM-22-40. The peak at low temperatures (the L-peak) was attributed to the weakly held ammonia species such as physically adsorbed NH3 and NHþ 4 nNH3 (n P 1) associations [31]. The peak at high temperature (the H-peak) was ascribed to the ammonia species desorbed from the acid sites, i.e., the Brönsted and Lewis acid sites (proved by the Py-IR results below) relating to aluminum species. The ratio of the H-peak area of the MCM-2250 to that of the BMCM-22-40 was 1.16, which was rather consis-
tent with the ratio of the aluminum content between them (1.12), indicating that the existing boron had little influence on the acidity of the MCM-22 zeolites. That could possibly be due to the much weaker acidity of the Si–OH–B sites than that of the Si–OH–Al sites, as reported in the literatures [32–34]. Combining with the results of XRD, TG–DTA and NH3–TPD, it could be concluded that the residual boron had little influence on the thermal stability and the acidity of the MCM-22 samples. In other words, it is not necessary to remove the boron completely from the BMCM-22 zeolites before their application as solid acid catalysts. 3.3.4. Acidity of the MCM-22 zeolites with various Si/Al2 ratios NH3–TPD profiles of the H-form BMCM-22 samples with various Si/Al2 ratios were shown in Fig. 10. The results showed that both the desorption peak areas and the maximal desorption peak
Fig. 8. TG (- - -) and DTA (—) profiles of the as-synthesized samples of (a) BMCM-2240 and (b) MCM-22-50.
Fig. 9. NH3–TPD profiles of the H-form samples of BMCM-22-40 and MCM-22-50. Note: The fitted L-peak for the sample BMCM-22-40 is superimposed as (L), and the H-peak is shown as (H).
50
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 10. NH3–TPD profiles of the H-form BMCM-22-r samples, where r is (a) 50, (b) 100, (c) 150, (d) 200, (e) 300 and (f) 600, respectively.
temperatures decreased with the increasing of the Si/Al2 ratios, indicating that the acid amount and the acid strength decreased simultaneously with the decreasing amount of aluminum in the MCM-22 zeolites. The Py-IR spectra of several high silica MCM-22 samples degassed at 623 K were shown in Fig. 11. Three main absorption bands could be observed at about 1545, 1490 and 1456 cm1 on the samples of BMCM-22-50 and BMCM-22-150. The 1490 cm1 absorption consisted of bands assigned to the Brönsted and Lewis sites altogether, while those at 1545 and 1456 cm1 were ascribed to the Brönsted and Lewis acid sites, respectively [35]. The absorption intensities of the BMCM-22-150 were obviously weaker than the corresponding bands of the BMCM-22-50. And no obvious characteristic acid absorption bands appeared on the sample of the BMCM-22-600 after pyridine desorption at 623 K. The decreasing tendency of the acid amount with the increasing of the Si/Al2 ratio was also in accordance with the NH3–TPD results discussed above. 3.4. Performances of high silica MCM-22 catalysts in the catalytic cracking of 1-butene to propene The catalytic cracking of 1-butene to propene was chosen as the model reaction to study the influence of the Si/Al2 ratios on the cat-
Fig. 11. Py-IR of the H-form samples of (a) BMCM-22-50, (b) BMCM-22-150 and (c) BMCM-22-600 at a pyridine desorption temperature of 623 K.
Fig. 12. Influence of Si/Al2 ratios on the performances of the MCM-22 catalysts for catalytic cracking of 1-butene to propene. Reaction conditions: T = 853 K, P = 0.1 MPa, WHSV = 5 h1, TOS = 2 h.
alytic performances of the high silica MCM-22 zeolites. The reaction conditions were as follows: reaction temperature T = 853 K, 1-butene pressure P = 0.1 MPa, the weight hourly space velocity of 1-butene WHSV = 5 h1, and time on stream TOS = 2 h. As shown in Fig. 12, the conversion of butenes deceased almost linearly from 83 % to 25 % upon the increasing of the Si/Al2 ratios from 52 to 573 and the selectivity for ethene showed the same trend but in a more moderate way. However, the propene selectivity at first increased upon increasing of the Si/Al2 ratios, passed through a maximum (41.6%) at the ratio of 158, and then decreased smoothly with higher Si/Al2 ratios. In the process of C4 alkenes cracking on zeolites, the objective product propene could further undergo secondary reactions such as hydrogen transfer and aromatization. With the increasing of the Si/Al2 ratios (up to 158) of the MCM-22 catalysts, the hydrogen transfer and aromatization reactions of propene were suppressed due to the decreasing of the acidity, resulting in an obvious decrement in selectivity of the propane and aromatics, and a remarkable increment in propene selectivity. However, with further increasing of the Si/Al2 ratios (>158), the acidity of the MCM-22 was not strong enough to crack the oligomeric intermediates, resulting in a rapid increment in the selectivity of C5+ species (non-aromatic hydrocarbons with carbon numbers P5) and a gradual decrement of the propene selectivity. Thus the most suitable Si/Al2 ratio of the MCM-22 catalyst was around 158 for the catalytic cracking of 1butene to propene under our experimental conditions.
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
4. Conclusions High silica MCM-22 zeolites with high crystallinity were synthesized with a one-pot boron-containing method. Compared with conventional methods, the Si/Al2 ratios of the MCM-22 zeolites could be broadly extended from 30 to 600 in control by tuning those of the starting gels. All the boron atoms existed as tetra-coordinated framework species in the as-synthesized MCM-22 zeolites and part of them became tri-coordinated framework or extraframework boron species after calcination evidenced by 11B MAS NMR results. Furthermore, most of the tri-coordinated and a large part of the tetra-coordinated boron atoms could be removed during the ammonium exchange process. Residual boron in the samples had little influence on the physicochemical properties of the MCM-22 zeolites. When high silica MCM-22 zeolites were applied in the catalytic cracking of 1-butene to propene, the formation of propane and aromatics could be notably suppressed by increasing of the Si/Al2 ratios. The optimal propene selectivity was obtained on the MCM-22 catalyst at a Si/Al2 ratio of 158 under the selected reaction conditions.
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]
Acknowledgments The authors would like to thank Dr. Zhang Weiping and Liu Yong from the Nano and Interfacial Catalysis Group of Dalian Institute of Chemical Physics for their assistance on NMR characterizations. We also thank the National 973 Project of China (No. 2003CB615802, 2009CB623501) and the National Natural Science Foundation of China (No. 20773120) for financial support. References [1] [2] [3] [4]
M.K. Rubin, P. Chu, US 4954325, 1990. M.E. Leonowicz, J.A. Lawton, S.L. Lawton, M.K. Rubin, Science 264 (1994) 1910. M.A. Asensi, A. Corma, A. Martinez, J. Catal. 158 (1996) 561. J. Rigoreau, S. Laforge, N.S. Gnep, M. Guisnet, J. Catal. 236 (2005) 45.
[25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]
51
Y.Y. Shu, R. Ohnishi, M. Ichikawa, Catal. Lett. 81 (2002) 9. A. Corma, J. Martinez-Triguero, J. Catal. 165 (1997) 102. X.X. Zhu, S.L. Liu, Y.Q. Song, S.J. Xie, L.Y. Xu, Appl. Catal. A: Gen. 290 (2005) 191. J.S. Plotkin, Catal. Today 106 (2005) 10. G.L. Zhao, J.W. Teng, N. Xu, Y. Tang, Z.K. Xie, W.M. Yang, Q.L. Chen, Chem. J. Chin. U 26 (2005) 1140. A. Corma, C. Corell, J. Perezpariente, Zeolites 15 (1995) 2. K. Okumura, M. Hashimoto, T. Mimura, M. Niwa, J. Catal. 206 (2002) 23. Z.Y. Liu, Z.M. Liu, Y. Qi, L. Xu, Y.L. He, Y. Yang, Y.Y. Zhang, Chin. J. Catal. 25 (2004) 542. I. Guray, J. Warzywoda, N. Bac, A. Sacco, Micropor. Mesopor. Mater. 31 (1999) 241. M.J. Cheng, D.L. Tan, X.M. Liu, X.W. Han, X.H. Bao, L.W. Lin, Micropor. Mesopor. Mater. 42 (2001) 307. X.X. Zhu, S.L. Liu, Y.Q. Song, L.Y. Xu, Catal. Commun. 6 (2005) 742. J.S. Buchanan, J.G. Santiesteban, W.O. Haag, J. Catal. 158 (1996) 279. P. Meriaudeau, V.A. Tuan, V.T. Nghiem, F. Lefevbre, V.T. Ha, J. Catal. 185 (1999) 378. J.C. Xia, D.S. Mao, W.C. Tao, Q.L. Chen, Y.H. Zhang, Y. Tang, Micropor. Mesopor. Mater. 91 (2006) 33. P. Wu, T. Komatsu, T. Yashima, Micropor. Mesopor. Mater. 22 (1998) 343. S. Unverricht, M. Hunger, S. Ernst, Stud. Surf. Sci. Catal. 84 (1994) 37. L. Liu, M.J. Cheng, D. Ma, G. Hu, X.L. Pan, X.H. Bao, Micropor. Mesopor. Mater. 94 (2006) 304. R. Millini, G. Perego, J.W.O. Parker, G. Bellussi, L. Carluccio, Micropor. Mater. 4 (1995) 221. A. Corma, V. Fornes, S.B. Pergher, T.L.M. Maesen, J.G. Buglass, Nature 396 (1998) 353. S.L. Lawton, A.S. Fung, G.J. Kennedy, L.B. Alemany, C.D. Chang, G.H. Hatzikos, D.N. Lissy, M.K. Rubin, H.K.C. Timken, S. Steuernagel, D.E. Woessner, J. Phys. Chem. 100 (1996) 3788. L.L. Wang, Y. Wang, Y.M. Liu, L. Chen, S.F. Cheng, G.H. Gao, M.Y. He, P. Wu, Micropor. Mesopor. Mater. 113 (2008) 435. S. Lawton, M.E. Leonowicz, R. Partridge, P. Chu, M.K. Rubin, Micropor. Mesopor. Mater. 23 (1998) 109. Z.C. Liu, S.D. Shen, B.Z. Tian, J.Y. Sun, B. Tu, D.Y. Zhao, Chin. Sci. Bull. 49 (2004) 325. A. Cichocki, W. Lasocha, M. Michalik, Z. Sawlowicz, M. Bus, Zeolites 10 (1990) 583. R. de Ruiter, A.P.M. Kentgens, J. Grootendorst, J.C. Jansen, H. van Bekkum, Zeolites 13 (1993) 128. N. Katada, H. Igi, J.-H. Kim, M. Nawa, J. Phys. Chem. B 101 (1997) 5969. F. Lonyi, J. Valyon, Micropor. Mesopor. Mater. 47 (2001) 293. R. Millini, G. Perego, G. Bellussi, Top. Catal. 9 (1999) 13. C.T.W. Chu, C.D. Chang, J. Phys. Chem. 89 (1985) 1569. X.L. Niu, S.J. Xie, L.Y. Xu, Chin. J. Catal. 27 (2006) 601. C.A. Emeis, J. Catal. 141 (1993) 347.
Contents lists available at ScienceDirect
Microporous and Mesoporous Materials journal homepage: www.elsevier.com/locate/micromeso
One-pot synthesis of high silica MCM-22 zeolites and their performances in catalytic cracking of 1-butene to propene Guoliang Xu a,b, Xiangxue Zhu a, Xionglei Niu a,b, Shenglin Liu a, Sujuan Xie a, Xiujie Li a, Longya Xu a,* a b
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Liaoning, Dalian 116023, PR China Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e
i n f o
Article history: Received 31 January 2008 Received in revised form 5 August 2008 Accepted 6 August 2008 Available online 17 August 2008 Keywords: High silica MCM-22 Synthesis 1-butene Catalytic cracking Propene
a b s t r a c t MCM-22 zeolites with high SiO2/Al2O3 (Si/Al2) ratios were synthesized by a one-pot procedure with the assisting of boron. Based on this boron-containing method, the Si/Al2 ratios of the MCM-22 zeolites could be greatly extended from 30 to 600 in control by tuning the Si/Al2 ratios of the starting gels. ICP-AES and 11 B MAS NMR results demonstrated that a small part of the boron still existed in the framework of the MCM-22 after calcination and ammonium exchange. However, the residual boron species had little influence on the physicochemical properties of the MCM-22 samples evidenced by XRD, BET, TG–DTA, NH3– TPD, and Py-IR determinations. The reaction of catalytic cracking of 1-butene to propene was carried out to study the performances of the high silica MCM-22 catalysts. Increasing of the Si/Al2 ratios in the MCM22 zeolites suppressed the formation of byproducts such as propane and aromatics effectively, and the optimal selectivity towards propene was obtained on the sample with a Si/Al2 ratio of 158. Ó 2008 Elsevier Inc. All rights reserved.
1. Introduction MCM-22, a kind of layered zeolite invented by Mobil in 1990 [1], belongs to MWW type and possesses two independent pore systems. One consists of two-dimensional sinusoidal channels composed of slightly elliptical 10-membered rings (MR), and the other is characteristic of 12-MR supercages accessible by 10-MR windows [2]. Due to its unique structure and physicochemical properties, MCM-22 has been widely used in many hydrocarbon catalytic transforming processes such as isomerization [3], alkylation [4], aromatization [5] and cracking [6,7]. Recently, MCM-22 has been found to be a rather active and selective catalyst in the process of catalytic cracking of C4+ alkenes to propene, with which the C4+ streams from petrochemicals can be upgraded and the production of propene can be enhanced [7–9]. Conventionally, MCM-22 zeolites were synthesized under hydrothermal conditions using hexamethyleneimine (HMI) as an organic structure directing agent (OSDA) [10–12]. Most of the research works have revealed that suitable SiO2/Al2O3 (Si/Al2) ratios for the formation of MCM-22 zeolites are ranging from 15 to 80, preferably 20 to 50 [10–14]. Thus, MCM-22 catalysts with such Si/Al2 ratios might be unsuitable in some applications, especially in reactions which need only mild acidity, for instance, the cata-
* Corresponding author. Tel./fax: +86 411 84693292. E-mail addresses: [email protected], [email protected], [email protected] (L. Xu). 1387-1811/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2008.08.019
lytic cracking of the C4+ alkenes to propene [15,16] and the isomerization of the n-alkenes to their branched isomers [3]. During the past several years, much effort had been dedicated to enlarge the Si/Al2 ratios of the MCM-22 zeolites, of which post-treatments to low silica MCM-22 were widely applied including hydrothermal treating [17], organic or inorganic acid leaching [18,19] and silanizing with SiCl4 [20]. Previously, the Si/Al2 ratios of MCM-22 were effectively enhanced to 133 by post-treatment using ammonium hexafluorosilicate in our earlier work [15]. However, this method encountered a dilemma that the zeolite structure may be partially destroyed upon higher Si/Al2 ratios of the products. Different from the procedures mentioned above, Bao et al. succeeded in synthesizing MCM-22 zeolites with high Si/Al2 ratios (high silica MCM-22, Si/Al2 = 24–1000) via a complicated threestep strategy: first, preparation of ERB-1 (MWW borosilicate); then, removal of the boron species by calcination and HNO3 reflux; and finally, alumination of the deboronated ERB-1 with further hydrothermal treatment [21]. On their studying of the intercalating properties of ERB-1, Millini et al. have suggested a boron-containing approach to obtain high silica MWW zeolites using piperidine as an OSDA [22]. Inspired by their work, a one-pot procedure for the synthesis of high silica MCM-22 has been developed by our group, which utilized the assisting action of boron and HMI. In this paper, this one-pot synthesis method is described in detail. The existing states of boron in the zeolites as well as the deboronation process, and their influence on the physicochemical properties of the MCM-22 samples,
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
were studied. Also, the performances of the high silica MCM-22 catalysts for catalytic cracking of 1-butene were investigated.
2. Experimental 2.1. Synthesis of the MCM-22 zeolites
45
Thermo-gravimetric and differential thermal analyses (TG– DTA) of the as-synthesized MCM-22 samples were performed on a Perkin-Elmer Diamond TGS-2 and DTA1700 apparatus. The experiments were carried out in the temperature range of 293– 1323 K, with a heating rate of 10 K/min in flowing air (100 ml/ min). Acidity of the H-form MCM-22 samples was determined by the temperature programmed desorption of ammonia (NH3–TPD) and pyridine adsorption infrared spectra (Py-IR) techniques. The measurements were carried out using the same procedures as in reference [7].
The one-pot procedure (hereafter referred to as the boron-containing method to distinguish it from the conventional boron free method) for the synthesis of high silica MCM-22 zeolites is as follows: first, silica sol (25.6 wt% SiO2, 0.06 wt% Al2O3, Qingdao Haiyang Chemical Co. Ltd.), Al(NO3)3 9H2O (99.0 %, Shanghai Zhenxin Reagent Factory), sodium hydroxide (96.0%, Shenyang Chemical Reagent Factory), and H3BO3 (99.5%, Shenyang Chemical Reagent Factory), together with HMI (95.0%, Sheyang Chemical Factory) were well mixed and stirred vigorously for half an hour at room temperature to give a gel with molar composition of 1.0SiO2:(1/ r)Al2O3:0.1Na2O:0.5B2O3:1.0HMI:40.0H2O (r represents the Si/Al2 ratio). Then, the gel was transferred into a stainless-steel autoclave, heated to 448 K and held at this temperature for 168 h while being rotated at 60 rpm. After completion of the crystallization, the autoclave was cooled by tap water and the solid product was obtained by filtrating, washing and drying. The samples synthesized were designated as BMCM-22-r, e.g., BMCM-22-150. For comparison, the synthesis of boron free samples was conducted according to the conventional method using the same raw materials of silicon, aluminum, alkaline and HMI mentioned above. The final gel had a molar ratio of 0.1Na2O: 1.0SiO2: (1/r)Al2O3: 0.35HMI: 40H2O, and was crystallized at 443 K for 48 h. Similar to the boron-containing samples, the samples of this series were designated as MCM-22-r.
Catalytic cracking of 1-butene over the catalysts were carried out in a continuous-flow stainless-steel fixed-bed reactor with an inner diameter of 7 mm. Before a catalytic evaluation, the catalyst was pretreated at 773 K for 1 h at a constant flow of N2 (20 ml/ min). Then, 1-butene (>99.5 wt%, Qilu Petrochemical Co. Ltd.) was introduced at the desired temperature. The products were analyzed on-line by a Varian 3800 GC equipped with a FID and a 100 m PONA column. All isomerized butenes in the products were considered as the unconverted feed stocks. The conversion of the butenes (X) and the selectivity (S) of a particular product CxHy were calculated as follows:
2.2. Preparation of the catalysts
3. Results and discussion
The as-synthesized samples were calcined at 823 K for 3 h in air for burning off the OSDA. Then the as-calcined zeolites were ion exchanged triply with 0.8 mol/L of an NH4NO3 solution, dried and calcined at 823 K for 2 h to give the H-form zeolites. Finally, the Hform zeolites were pelletized, crushed and sieved to 20–40 mesh particles before employing in the catalytic cracking of 1-butene.
3.1. Synthesis of high silica MCM-22 zeolites
2.3. Characterization of the zeolites X-ray diffraction (XRD) patterns of the samples were recorded on an X Pert Pro X-ray diffractometer using Cu Ka radiation and operating at 40 kV and 40 mA. Scanning electron microscopy (SEM) images were recorded on a Quanta 200F scanning electron microscopy (FEI Company) at 30 kV. Chemical compositions of the zeolites were analyzed on an inductively coupled plasma atomic emission spectrometer (ICPAES, TJA, IRIS Advantage). Prior to the measurements, the samples were dissolved in a mixed solution of HF (1.0 mol/L) and HCl (0.5 mol/L) at room temperature in sealed plastic bottles to avoid volatilization of the SiF4 and BF3. Nitrogen adsorption measurements were carried out at 77 K with a Micrometrics ASAP 2010 equipment. The samples were degassed at 573 K and 103 Pa for 4 h prior to the adsorption measurements. The specific surface areas and micropore volumes were obtained by the BET and the t-plot methods, respectively. 11 B magic-angle spinning nuclear magnetic resonance (MAS NMR) spectra were collected on an Infinityplus-400 spectrometer at 128.3 MHz using a 4 mm ZrO2 rotor with a spinning rate of 10 kHz. The spectra were recorded for 200 scans each sample with a p/4 flip angle and 2 s pulse delay, and the chemical shifts being referenced to H3BO3.
2.4. Catalytic evaluation
X = (C4H8% in the feed C4H8% in the products)/(C4H8% in the feed), P S (CxHy) = (CxHy% in the products)/( CiHj% in the products), where all the percentages were in weight percent.
A series of high silica MCM-22 samples with various Si/Al2 ratios (30–600) were synthesized by the boron-containing method. XRD patterns of the as-synthesized and the as-calcined BMCM22-r zeolites were exhibited in Fig. 1. As shown in Fig. 1A, all the as-synthesized samples showed both sharp and broad peaks which matched well with those of the layered precursors of MCM-22 (also designated as MCM-22(P)) [23–25]. Also, XRD patterns of the corresponding as-calcined samples (Fig. 1B) exhibited exact diffraction peaks of the MCM-22 (PDF Card number: 490627). After calcination, one noticeable phenomenon was that the 0 0 l peaks shifted to higher 2h degrees with reduced intensities, which was regarded as indications for the process of 2D layered MCM-22 precursors transforming to 3D microporous structures [22,24]. In addition, the XRD peaks of the as-synthesized and the corresponding as-calcined samples became sharper and the intensities increased gradually with increasing of the Si/ Al2 ratios, especially for the 0 0 l peaks (e.g. 001, 002 and 004). As the XRD peaks are correlated with the ordering of the structure, this phenomenon undoubtedly suggests a more ordered crystal structure for the samples with higher Si/Al2 ratios, especially along the c-dimension. The SEM images of several as-synthesized samples were given in Fig. 2. It can be seen that the crystals of the MCM-22 zeolites mainly existed in typical disk-like shape. From literature [26], we know that the c-axis of MCM-22 is perpendicular to the plate surface. From Fig. 2, one noticeable phenomenon was observed that the thickness of the plates, viz. the c-dimension of the MCM-22 zeolite, increases obviously with the increasing of the Si/Al2 ratios, which is quite consistent with the XRD results discussed above.
46
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
zeolites by the conventional method. Thus, it is tentatively suggested that boron could play a similar role as aluminum in the synthesis of the BMCM-22 samples. Besides the Si/(B2 + Al2) ratios, HMI also plays an important role in the formation of the MCM-22 precursors, which can be divided into two aspects [10,21]. One part of the HMI serves as a framework charge compensating agent in the protonated form (HMI+) and the other as a framework supporting agent in the neutral molecular form. In the conventional synthesizing system, the negative charges of the MCM-22 precursors are originated from either the Si–O–Al sites (formed by aluminum atoms inserted into the framework) or the Si–O defect sites (formed by the reaction between the OH and silanol groups). If the aluminum content is relatively high (Si/Al2 < 15), the amount of the HMI+ required for the formation of stable MCM-22 precursors is not sufficient to match the amount of the Si–O–Al sites, leading to the formation of other phases such as MOR. When the aluminum content is relatively low (Si/Al2 > 50), the Si–O–Al sites would be totally compensated by HMI+. At the same time, a number of Si–O defect sites are produced to compensate the surplus HMI+, which makes the MCM22 precursor unstable and results in the formation of other phases such as kenyaite and ZSM-5. As for the boron-containing system with low aluminum contents, the boron can also enter the framework of MCM-22 (proved by the 11B MAS NMR results below) and form the Si–O–B sites similar to the behaviors of aluminum. As the Si/(B2 + Al2) ratios are in the range of 15–50, the amount of the framework charges (Si–O–B together with the Si–O–Al sites) matches well with the amount of the HMI+ required for the formation of stable MCM-22 precursors, leading to the successful synthesis of MCM-22 zeolites with high Si/Al2 ratios (up to 600). 3.3. Physicochemical properties of high silica MCM-22 zeolites
Fig. 1. XRD patterns of (A) as-synthesized and (B) as-calcined B-MCM-22-r samples, where r is (a) 30, (b) 40, (c) 50, (d) 100, (e) 150, (f) 200, (g) 300 and (h) 600, respectively.
3.2. Difference in optimal Si/Al2 ratio range for the synthesis of MCM22 between the boron-containing and the conventional methods For comparison, several samples (Si/Al2 ranging from 15 to 90) were synthesized by the conventional methods, and the XRD patterns of the as-calcined samples were presented in Fig. 3. When the Si/Al2 ratio in the gel was as low as 15 or higher than 60, MOR or MCM-22 with impurities (i.e., ZSM-5 and kenyaite) was synthesized, respectively, which was consistent with the results obtained in the literatures [12,27]. Only two pure MCM-22 samples were obtained with the Si/Al2 ratios of 30 and 50, which were among the favorable Si/Al2 range (20–50) for the synthesis of MCM-22 zeolites [11,12,27]. By comparing the results of the boron-containing and the conventional methods, it can be deduced that the existence of boron in the synthesizing system favors the formation of the MCM-22 framework within a wider Si/Al2 range. To investigate the nature of the difference in optimal Si/Al2 ranges between the two methods, the Si/(B2 + Al2) ratios in the as-synthesized BMCM-22 samples were first considered. Table 1 listed the chemical compositions of several BMCM-22 samples. The results revealed that the Si/(B2 + Al2) ratios of the samples were in the range of 15–50, which was just the suitable Si/Al2 range for the synthesis of MCM-22
3.3.1. Texture and chemical compositions of high silica MCM-22 The textural properties of several as-calcined high silica MCM22 samples were given in Table 2. As can be seen, each sample showed a total specific surface area more than 460 m2/g, a microporous surface area larger than 339 m2/g and a micropore volume beyond 0.16 cm3/g. Moreover, no obvious decrease in the surface area and the micropore volume was observed with the increasing of the Si/Al2 ratios from 50 to 600, which was one merit of the boron-containing method as compared with those post-treatment methods [15]. In the synthesis of MCM-22 zeolites by the conventional method, the Si/Al2 ratios of the products were usually lower than those of the starting gels [9,12], which was possibly due to inefficient utilization of the silicon atoms. This problem could well be resolved by the boron-containing method. As shown in Fig. 4, the data points were very close to the 1:1 line, indicating that the Si/Al2 ratios of the products were close to those of the starting gels. As has been discussed above, in the boron-containing system, the formation of the Si–O–B sites inhibits the formation of Si–O defect sites, thus the utilization efficiency of silicon is improved, making it possible to control the Si/Al2 ratios of the MCM-22 samples precisely. 3.3.2. Existing states of boron in high silica MCM-22 It has been reported that deboronation could occur easily during the post-treatments of the boron-containing zeolites [28,29]. In this section, the existing states of boron in the high silica MCM-22 zeolites were investigated by the 11B MAS NMR technique. The sample BMCM-22-150 with a final Si/Al2 ratio of 158 was chosen as the specimen, and the 11B MAS NMR spectra of its as-synthesized, as-calcined and H-form samples were presented in Fig. 5. From Fig. 5a, there was only one main resonance peak at 3 ppm in the as-sythesized sample which was assigned to the tet-
47
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 2. SEM images of the as-synthesized samples of (a) BMCM-22-50, (b) BMCM-22-150 and (c) BMCM-22-600.
Table 1 Chemical compositions of selected as-synthesized BMCM-22 samples Sample
Si/Al2
Si/B2
Si/(B2 + Al2)
BMCM-22-30 BMCM-22-50 BMCM-22-150 BMCM-22-600
30 52 158 573
37 27 25 44
17 18 22 41
Table 2 Textural properties of the as-calcined high silica MCM-22 samples Sample
SBET (m2/g)
Vmicropore (cm3/g)
Smicro
Stotal
BMCM-22-50 BMCM-22-150 BMCM-22-600
354 339 378
499 477 464
0.168 0.162 0.183
Note: SBET = surface area, Smicro = microporous surface area, Stotal = total surface area, Vmicropore = micropore volume.
Fig. 3. XRD patterns of the as-calcined MCM-22-r samples, where r is (a) 15, (b) 30, (c) 50, (d) 60, (e) 70 and (f) 90, respectively. Note: (a) is an XRD pattern of MOR. Peaks belong to impurities in (d), (e), and (f): (;) Kenyaite and (*) ZSM-5.
ra-coordinated, i.e., framework boron [29]. After calcination, the peak intensity of framework boron decreased slightly and three
new signals centered at 0, 4 and 15 ppm appeared as shown in Fig. 5b. The first two peaks were attributed to the tri-coordinated framework boron in zeolite and the broad resonance at 15 ppm was associated with the extra-framework tri-coordinated boron [29]. It can be deduced that part of the tetra-coordinated boron was extracted from the framework during the OSDA removal pro-
48
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 4. Evolution of Si/Al2 ratios in the resulting BMCM-22 samples and in the starting gels. Note: the straight line is a 1:1 line.
Fig. 6. XRD patterns of (a) as-calcined and (b) H-form BMCM-22-150.
the calcination and ammonium exchange processes, but it had little influence on the structure of the high silica MCM-22.
Fig. 5. 11B MAS NMR profiles of (a) as-synthesized, (b) as-calcined and (c) H-form BMCM-22-150.
cess. In Fig. 5c, peaks at 0, 4 and 15 ppm of the H-form MCM-22 became very weak (as shown in the inset of Fig. 5 in which the signal was amplified by 4 times), indicating that there was little amount of tri-coordinated boron in the H-form sample. At the same time the peak intensity of tetra-coordinated boron decreased a lot as compared with that in Fig. 5b, suggesting that serious deboronation occured in the ammonium exchange processes. This was further proved by the ICP-AES results which revealed that the Si/B2 ratio increased from 25 in the as-synthesized form to 180 in the H-form. The XRD patterns of the as-calcined and the H-form BMCM-22150 samples were presented in Fig. 6 to examine the effects of the deboronation on the crystalline phases. As shown in Fig. 6, the intensities of the peaks at 2h = 22.9 ± 0.1° and 26.2 ± 0.1° of the H-form sample were a little weaker than those of the as-calcined form, indicating that during the ammonium exchange process the relative crystallinity of the sample decreased slightly, which was possibly due to the deboronation. Even so, both the XRD curves of the as-calcined and the H-form BMCM-22-150 matched exactly with the XRD patterns of the MCM-22 (PDF Card number: 49-0627), indicating that they possessed good 3D structures of the MCM-22. From the 11B MAS NMR, ICP-AES and XRD results discussed above, it could be concluded that deboronation occurred during
3.3.3. Thermal stability and acidity of high silica MCM-22 zeolites For zeolite catalysts, thermal stability and acidity are the most important properties for their catalytic applications. In this section, we compare the thermal stability and the acidity between the BMCM-22-40 and MCM-22-50, both of which possess close final Si/Al2 ratios of 45 and 40, respectively. Two methods were employed to study the thermal stability of the samples, i.e., a calcination method and a TG–DTA method. For the calcination method, the samples were calcined in air at different temperatures for 1 h, followed by XRD analyses (Fig. 7). As can be seen, both the BMCM-22-40 and the MCM-22-50 samples calcined below or equal to 1173 K maintained a MCM-22 crystalline phase, and the relative crystallinity decreased slightly with the increasing of the calcination temperature. However, both of the samples became an amorphous phase when the calcination temperature reached 1273 K. TG–DTA experiments were performed to further investigate the influence of the existing boron on the thermal stabilities of the MCM-22 zeolites. As shown in Fig. 8, the as-synthesized BMCM22-40 and the MCM-22-50 had roughly similar TG and DTA profiles. The obvious weight loss with broad endothermic DTA peaks before 430 K was attributed to desorption of physically adsorbed water. The weight loss between 430 K and 580 K might be caused by several processes, such as desorption of strongly adsorbed water, desorption of HMI [24], and condensation of the interlayer silanol groups along the c-axis [22]. Thus, the appearing (as shown in Fig. 8a at 526 K) or not appearing (Fig. 8b) of the weak exothermic DTA peaks at this temperature range depends on the comprehensive effects of the reactions mentioned above. Further investigation is underway to make this problem more clearly. The three obvious weight loss stages accompanied with obvious exothermic DTA peaks ranging from 580 K to 970 K were attributed to the oxidation of the occluded OSDA in the MCM-22 samples [21]. In agreement with the results of the calcination experiments, weak exothermic DTA peaks without weight losses in the TG curves at about 1240 K, ascribed to the framework destruction of the two samples, were observed. The NH3–TPD technique was employed to determine the acidity of the two H-form samples of BMCM-22-40 and MCM-22-50, and the profiles were shown in Fig. 9. The relative intensity and the distribution of the NH3 desorption peaks of the two samples were found to be very similar. According to Katada’s approach, the TPD profile of each sample was fitted into two peaks [30], and the
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
49
Fig. 7. XRD patterns of (a) BMCM-22-40 and (b) MCM-22-50 calcined at different temperatures for 1 h.
dashed curves in Fig. 9 was an example of the deconvoluted profiles of BMCM-22-40. The peak at low temperatures (the L-peak) was attributed to the weakly held ammonia species such as physically adsorbed NH3 and NHþ 4 nNH3 (n P 1) associations [31]. The peak at high temperature (the H-peak) was ascribed to the ammonia species desorbed from the acid sites, i.e., the Brönsted and Lewis acid sites (proved by the Py-IR results below) relating to aluminum species. The ratio of the H-peak area of the MCM-2250 to that of the BMCM-22-40 was 1.16, which was rather consis-
tent with the ratio of the aluminum content between them (1.12), indicating that the existing boron had little influence on the acidity of the MCM-22 zeolites. That could possibly be due to the much weaker acidity of the Si–OH–B sites than that of the Si–OH–Al sites, as reported in the literatures [32–34]. Combining with the results of XRD, TG–DTA and NH3–TPD, it could be concluded that the residual boron had little influence on the thermal stability and the acidity of the MCM-22 samples. In other words, it is not necessary to remove the boron completely from the BMCM-22 zeolites before their application as solid acid catalysts. 3.3.4. Acidity of the MCM-22 zeolites with various Si/Al2 ratios NH3–TPD profiles of the H-form BMCM-22 samples with various Si/Al2 ratios were shown in Fig. 10. The results showed that both the desorption peak areas and the maximal desorption peak
Fig. 8. TG (- - -) and DTA (—) profiles of the as-synthesized samples of (a) BMCM-2240 and (b) MCM-22-50.
Fig. 9. NH3–TPD profiles of the H-form samples of BMCM-22-40 and MCM-22-50. Note: The fitted L-peak for the sample BMCM-22-40 is superimposed as (L), and the H-peak is shown as (H).
50
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
Fig. 10. NH3–TPD profiles of the H-form BMCM-22-r samples, where r is (a) 50, (b) 100, (c) 150, (d) 200, (e) 300 and (f) 600, respectively.
temperatures decreased with the increasing of the Si/Al2 ratios, indicating that the acid amount and the acid strength decreased simultaneously with the decreasing amount of aluminum in the MCM-22 zeolites. The Py-IR spectra of several high silica MCM-22 samples degassed at 623 K were shown in Fig. 11. Three main absorption bands could be observed at about 1545, 1490 and 1456 cm1 on the samples of BMCM-22-50 and BMCM-22-150. The 1490 cm1 absorption consisted of bands assigned to the Brönsted and Lewis sites altogether, while those at 1545 and 1456 cm1 were ascribed to the Brönsted and Lewis acid sites, respectively [35]. The absorption intensities of the BMCM-22-150 were obviously weaker than the corresponding bands of the BMCM-22-50. And no obvious characteristic acid absorption bands appeared on the sample of the BMCM-22-600 after pyridine desorption at 623 K. The decreasing tendency of the acid amount with the increasing of the Si/Al2 ratio was also in accordance with the NH3–TPD results discussed above. 3.4. Performances of high silica MCM-22 catalysts in the catalytic cracking of 1-butene to propene The catalytic cracking of 1-butene to propene was chosen as the model reaction to study the influence of the Si/Al2 ratios on the cat-
Fig. 11. Py-IR of the H-form samples of (a) BMCM-22-50, (b) BMCM-22-150 and (c) BMCM-22-600 at a pyridine desorption temperature of 623 K.
Fig. 12. Influence of Si/Al2 ratios on the performances of the MCM-22 catalysts for catalytic cracking of 1-butene to propene. Reaction conditions: T = 853 K, P = 0.1 MPa, WHSV = 5 h1, TOS = 2 h.
alytic performances of the high silica MCM-22 zeolites. The reaction conditions were as follows: reaction temperature T = 853 K, 1-butene pressure P = 0.1 MPa, the weight hourly space velocity of 1-butene WHSV = 5 h1, and time on stream TOS = 2 h. As shown in Fig. 12, the conversion of butenes deceased almost linearly from 83 % to 25 % upon the increasing of the Si/Al2 ratios from 52 to 573 and the selectivity for ethene showed the same trend but in a more moderate way. However, the propene selectivity at first increased upon increasing of the Si/Al2 ratios, passed through a maximum (41.6%) at the ratio of 158, and then decreased smoothly with higher Si/Al2 ratios. In the process of C4 alkenes cracking on zeolites, the objective product propene could further undergo secondary reactions such as hydrogen transfer and aromatization. With the increasing of the Si/Al2 ratios (up to 158) of the MCM-22 catalysts, the hydrogen transfer and aromatization reactions of propene were suppressed due to the decreasing of the acidity, resulting in an obvious decrement in selectivity of the propane and aromatics, and a remarkable increment in propene selectivity. However, with further increasing of the Si/Al2 ratios (>158), the acidity of the MCM-22 was not strong enough to crack the oligomeric intermediates, resulting in a rapid increment in the selectivity of C5+ species (non-aromatic hydrocarbons with carbon numbers P5) and a gradual decrement of the propene selectivity. Thus the most suitable Si/Al2 ratio of the MCM-22 catalyst was around 158 for the catalytic cracking of 1butene to propene under our experimental conditions.
G. Xu et al. / Microporous and Mesoporous Materials 118 (2009) 44–51
4. Conclusions High silica MCM-22 zeolites with high crystallinity were synthesized with a one-pot boron-containing method. Compared with conventional methods, the Si/Al2 ratios of the MCM-22 zeolites could be broadly extended from 30 to 600 in control by tuning those of the starting gels. All the boron atoms existed as tetra-coordinated framework species in the as-synthesized MCM-22 zeolites and part of them became tri-coordinated framework or extraframework boron species after calcination evidenced by 11B MAS NMR results. Furthermore, most of the tri-coordinated and a large part of the tetra-coordinated boron atoms could be removed during the ammonium exchange process. Residual boron in the samples had little influence on the physicochemical properties of the MCM-22 zeolites. When high silica MCM-22 zeolites were applied in the catalytic cracking of 1-butene to propene, the formation of propane and aromatics could be notably suppressed by increasing of the Si/Al2 ratios. The optimal propene selectivity was obtained on the MCM-22 catalyst at a Si/Al2 ratio of 158 under the selected reaction conditions.
[5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24]
Acknowledgments The authors would like to thank Dr. Zhang Weiping and Liu Yong from the Nano and Interfacial Catalysis Group of Dalian Institute of Chemical Physics for their assistance on NMR characterizations. We also thank the National 973 Project of China (No. 2003CB615802, 2009CB623501) and the National Natural Science Foundation of China (No. 20773120) for financial support. References [1] [2] [3] [4]
M.K. Rubin, P. Chu, US 4954325, 1990. M.E. Leonowicz, J.A. Lawton, S.L. Lawton, M.K. Rubin, Science 264 (1994) 1910. M.A. Asensi, A. Corma, A. Martinez, J. Catal. 158 (1996) 561. J. Rigoreau, S. Laforge, N.S. Gnep, M. Guisnet, J. Catal. 236 (2005) 45.
[25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35]
51
Y.Y. Shu, R. Ohnishi, M. Ichikawa, Catal. Lett. 81 (2002) 9. A. Corma, J. Martinez-Triguero, J. Catal. 165 (1997) 102. X.X. Zhu, S.L. Liu, Y.Q. Song, S.J. Xie, L.Y. Xu, Appl. Catal. A: Gen. 290 (2005) 191. J.S. Plotkin, Catal. Today 106 (2005) 10. G.L. Zhao, J.W. Teng, N. Xu, Y. Tang, Z.K. Xie, W.M. Yang, Q.L. Chen, Chem. J. Chin. U 26 (2005) 1140. A. Corma, C. Corell, J. Perezpariente, Zeolites 15 (1995) 2. K. Okumura, M. Hashimoto, T. Mimura, M. Niwa, J. Catal. 206 (2002) 23. Z.Y. Liu, Z.M. Liu, Y. Qi, L. Xu, Y.L. He, Y. Yang, Y.Y. Zhang, Chin. J. Catal. 25 (2004) 542. I. Guray, J. Warzywoda, N. Bac, A. Sacco, Micropor. Mesopor. Mater. 31 (1999) 241. M.J. Cheng, D.L. Tan, X.M. Liu, X.W. Han, X.H. Bao, L.W. Lin, Micropor. Mesopor. Mater. 42 (2001) 307. X.X. Zhu, S.L. Liu, Y.Q. Song, L.Y. Xu, Catal. Commun. 6 (2005) 742. J.S. Buchanan, J.G. Santiesteban, W.O. Haag, J. Catal. 158 (1996) 279. P. Meriaudeau, V.A. Tuan, V.T. Nghiem, F. Lefevbre, V.T. Ha, J. Catal. 185 (1999) 378. J.C. Xia, D.S. Mao, W.C. Tao, Q.L. Chen, Y.H. Zhang, Y. Tang, Micropor. Mesopor. Mater. 91 (2006) 33. P. Wu, T. Komatsu, T. Yashima, Micropor. Mesopor. Mater. 22 (1998) 343. S. Unverricht, M. Hunger, S. Ernst, Stud. Surf. Sci. Catal. 84 (1994) 37. L. Liu, M.J. Cheng, D. Ma, G. Hu, X.L. Pan, X.H. Bao, Micropor. Mesopor. Mater. 94 (2006) 304. R. Millini, G. Perego, J.W.O. Parker, G. Bellussi, L. Carluccio, Micropor. Mater. 4 (1995) 221. A. Corma, V. Fornes, S.B. Pergher, T.L.M. Maesen, J.G. Buglass, Nature 396 (1998) 353. S.L. Lawton, A.S. Fung, G.J. Kennedy, L.B. Alemany, C.D. Chang, G.H. Hatzikos, D.N. Lissy, M.K. Rubin, H.K.C. Timken, S. Steuernagel, D.E. Woessner, J. Phys. Chem. 100 (1996) 3788. L.L. Wang, Y. Wang, Y.M. Liu, L. Chen, S.F. Cheng, G.H. Gao, M.Y. He, P. Wu, Micropor. Mesopor. Mater. 113 (2008) 435. S. Lawton, M.E. Leonowicz, R. Partridge, P. Chu, M.K. Rubin, Micropor. Mesopor. Mater. 23 (1998) 109. Z.C. Liu, S.D. Shen, B.Z. Tian, J.Y. Sun, B. Tu, D.Y. Zhao, Chin. Sci. Bull. 49 (2004) 325. A. Cichocki, W. Lasocha, M. Michalik, Z. Sawlowicz, M. Bus, Zeolites 10 (1990) 583. R. de Ruiter, A.P.M. Kentgens, J. Grootendorst, J.C. Jansen, H. van Bekkum, Zeolites 13 (1993) 128. N. Katada, H. Igi, J.-H. Kim, M. Nawa, J. Phys. Chem. B 101 (1997) 5969. F. Lonyi, J. Valyon, Micropor. Mesopor. Mater. 47 (2001) 293. R. Millini, G. Perego, G. Bellussi, Top. Catal. 9 (1999) 13. C.T.W. Chu, C.D. Chang, J. Phys. Chem. 89 (1985) 1569. X.L. Niu, S.J. Xie, L.Y. Xu, Chin. J. Catal. 27 (2006) 601. C.A. Emeis, J. Catal. 141 (1993) 347.