Selective Isomerization of n-Butane over Mordenite Nanoparticles

Oct 15, 2017 - For the first time, a sequential fabrication step involving mechanochemistry, recrystallization, and dealumination was developed to con...
0 downloads 0 Views 2MB Size
Subscriber access provided by La Trobe University Library

Article

Selective isomerization of n-butane over mordenite nanoparticles fabricated by sequential ball milling-recrystallization-dealumination route Teguh Kurniawan, Oki Muraza, Abbas S. Hakeem, Idris A Bakare, Toshiki Nishitoba, Toshiyuki Yokoi, Zain H. Yamani, and Adnan M. Al-Amer Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b02555 • Publication Date (Web): 15 Oct 2017 Downloaded from http://pubs.acs.org on October 16, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Subscriber access provided by La Trobe University Library

and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35

Parent H-MOR

Nano-sized dealuminated H-MOR

18

15

80

15

12 9 6 3 0

10

50 C3

100 i-C4

150 C4=

Selectivity [%]

100

Conversion [%]

18

9 40

i-C5

6

20

3

0

0 10

200

Time on stream [min]

12

60

Conversion

C1-2

50 C3

100 i-C4

C4=

150 i-C5

Conversion [%]

500 nm

500 nm

Selectivity [%]

1 2 3 4 5 6 7 8 9 10 11 12 13 100 14 15 16 80 17 18 60 19 20 21 40 22 23 24 20 25 26 0 27 28 C1-2 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

200 Conversion

Time on stream [min]

n-butane isomerization

Natural mordenite was milled and recrystallized followed by dealumination. The nanosized ACS Paragon Plus Environment dealuminated mordenite exhibited a high catalytic activity and selectivity to isobutane.

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 35

Selective isomerization of n-butane over mordenite nanoparticles fabricated by sequential ball milling-recrystallization-dealumination route Teguh Kurniawan1,3, Oki Muraza1*, Abbas S. Hakeem1, Idris A. Bakare1, Toshiki Nishitoba2, Toshiyuki Yokoi2, Zain H. Yamani1, Adnan M. Al Amer1 1

Center of Research Excellence in Nanotechnology and Chemical Engineering Department King Fahd University of Petroleum & Minerals, Dhahran 31261, Saudi Arabia 2 Chemical Resources Laboratory, Tokyo Institute of Technology, 4259 Nagatsuta, Midori-ku,Yokohama 226-8503, Japan 3 Chemical Engineering Department, University of Sultan Ageng Tirtayasa, Cilegon 42435, Indonesia *Corresponding author: [email protected]

Abstract For the first time, a sequential fabrication step involving mechanochemistry, recrystallization, and dealumination was developed to convert low-cost natural zeolites to mordenite nanoparticles. Natural zeolites are mostly found having poor textural properties and high aluminum content, which are not suitable for most of industrial catalytic reactions. The parent natural zeolites within the size of 1-10 µm was treated by ball milling to obtain nanosized particles with size in the range of 20-160 nm. The nitrogen physisorption study revealed that the external surface area and intercrystalline mesopore volume of the milled nanoparticles increased by 4-fold and 7-fold, respectively. Recrystallization by hydrothermal treatment in basic silicate solution was applied to recover the mordenite crystallinity at 170 oC for 6 h. The recrystallized MOR samples were further subjected to acid dealumination treatment over different periods. The H-MOR samples were evaluated in a fixed bed reactor for n-butane isomerization. The isobutane selectivity increased from 11% to 28%, when the parent microparticle was substituted by the recrystallized nanoparticles. Moreover, the catalyst stability improved over the recrystallized nanoparticles. The dealuminated-recrystallized nanoparticle exhibited the highest selectivity of ca. 58% to isobutane and less deactivation rate due to low acid sites density and small nanoparticles size. Keywords: Natural mordenite; Ball milling; Nanoparticles; Recrystallization; Dealumination; Isomerization.

1 ACS Paragon Plus Environment

Page 3 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

1. Introduction Natural zeolites are abundantly available throughout the world particularly as sedimentary deposit around volcanic regions. Open pit mining operation, an easy mining technique, is being used to exploiting the natural zeolites. Due to these reasons, cost of the natural zeolites is far cheaper than the synthetic zeolites. However, the natural zeolites have several drawbacks, such as high impurities, poor textural properties, lower adsorption capacity, and inconsistency of composition.1-3 Consequently, natural zeolites are mostly commercialized for applications with non-strictly specification and required in bulky scale. Natural zeolites are applied as an adsorbents in water purification and wastewaters treatment, a fertilizer in agriculture, for ammonium removal in aquaculture, and raw materials for construction.2-4

Mordenite (MOR) is one of important frameworks that frequently occurring in nature as sedimentary rocks. In fact, mordenite is commercially used as industrial catalysts. Mordenite comprises of two pore types, which are 12-membered ring (MR) with size 6.5 x 7 Å and 8-MR with compressed size 2.6 x 5.7 Å along with the c-axis. The 12-MR and the 8-MR are bridged by another channel, which is 8-MR side pockets with size 3.4 x 4.8 Å along with the b-axis.5, 6 Due to the small pores size, those 8-MRs are not accessible for most of the molecules, hence the mordenite is often considered as a uni-dimensional pore. There are two types of mordenite based on adsorption capacity, i.e., large-port and small-port. The large-port is capable of adsorbing more molecules than the small-port, which excludes the molecules more than 4.5 Å.7 The small-port mordenite blocks large molecules due to the structural fault.8 Recently, Simoncic and Amrbruster6 reported that both the small and large-port MORs have similar defects, hence, the behavior of natural mordenite acting as small-port is still enigmatic. The synthetic MOR is known as the large-port mordenite, which is synthesized at temperature less than 260 oC. On the other hand, the small-port mordenite is produced at higher temperature within of 275 to 300 oC.

2 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Mordenite is commercially used as industrial catalysts for toluene disproportionation, amination, C2-C3 transalkylation, naphtha isomerization, ethylbenzene isomerization, and n-butane isomerization.9 Isomerization of n-butane to isobutane over Pt/chlorinated alumina is important reaction in industry of which the isobutane is converted to alkylates.10 The chemical is capable to improve octane number of gasoline. Unfortunately, the Pt/chlorinated alumina is a non-regenerable catalyst and sensitive to contaminants such as sulfur and nitrogen.9 One of potential candidate to replace the Pt/chlorinated alumina catalyst is Pt/H-mordenite. The mordenite catalyst offers some advantageous, i.e., tolerate to water and sulfur and easy regeneration.11

Over the past decade, nanosized zeolites have been attracting researchers due to its superiority over the conventional microsized crystals.12-14 Nanoparticles mordenite has large specific surface area owing to its small crystal size. This gives several benefits, such as less mass transfer limitation, prolong the catalyst lifetime, and high catalyst activity.15 The nanoparticles also created intercrystalline mesopores, which are formed by the stacking of nanoparticles. Recently, Chu et al.16 reported that the intercrystalline mesopores of ferrierite zeolite increased the stability and product selectivity in 1-butene isomerization. The nanosized zeolite could be prepared through bottom-up strategies by hydrothermal synthesis or top-down approach by milling treatment.12, 17 The top-down approach is a potential method to obtaining nanoparticles from natural zeolites. However, the milling technique reduces the zeolites crystalline phase remarkably. The combination of milling with hydrothermal recrystallization has been reported successful in producing high crystalline nanoparticles mordenite. 18, 19

High silicon to aluminum ratio (Si/Al) of mordenite favored the isobutane selectivity and catalyst stability in the butane isomerization. On the other hand, aluminum-rich mordenite favored selectivity to propane and less stability as cokes developed faster during the reaction.20-22 Postmodification treatment that commonly practiced to increase the Si/Al is acid dealumination. The 3 ACS Paragon Plus Environment

Page 4 of 35

Page 5 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

natural mordenite is commonly found as a small-port type, which can be easily converted to the large-port mordenite by acid dealumination technique.23,

24

The acid dealumination will also

decrease the number of acid sites, particularly the Brønsted acid sites, create micropores, and mesopores in the MOR framework.25-27 The mesopores are important to shortening of the molecules pathlength in the zeolite pores.

In this paper, we investigated the effect of mordenite nanoparticles in n-butane isomerization. The nanosized mordenite samples were derived from low-cost natural zeolites through ball millingrecrystallization and ball milling-recrystallization-dealumination technique. To the best of our knowledge, the systematic study of mordenite nanoparticles for n-butane isomerization is rarely reported.

2. Experimental The natural zeolites samples were modified through various treatments, i.e., high-energy ball milling, hydrothermal recrystallization, and acid dealumination. The flow chart of the experiment is presented in Figure 1. The detailed explanation of experiment is provided as below.

2.1 Nanosized MOR preparation High-energy ball milling attritor (Union Process HDDM-01) was used to reduce the particle size by using water as a dispersion medium. The natural zeolites were obtained from Klaten, Central Java, Indonesia. The particles size of the as-received natural zeolites was 4 to 5 mm. Prior to milling, the sample was crushed and sieved into a smaller size, less than 0.1 mm. 50 g of natural zeolite was milled by controlling the agitator speed at 2000 rpm for 8 h. The ball material was made of zirconia (ZrO2) with fine ball size 650 micron in diameter. The heat generated from the milling process was removed by water maintained at 13 oC circulated through the jacket of the milling tank. The slurry 4 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 35

was dried overnight at 80 oC. The milled nanoparticles sample was called as M. The dried milled powder was ion exchanged twice with 2 M NH4NO3 with ratio 1:20 by weight at temperature 85 oC under reflux and vigorous stirring action for 2 h. Afterward, the powder was dried at 110 oC overnight and calcined in static air at 550 oC for 8 h (H-M).

2.2 Recrystallization The milled sample (M) was treated hydrothermally by using alkaline silicate solution (18 SiO2:12 NaOH:780 H2O).18 Fumed silica (Sigma-Aldrich) was dissolved in sodium hydroxide solution. Afterward, 1 g of the milled powder was mixed with the alkaline silicate solution. After mixing for 15 minutes, the solution was heated at 170 oC for 6 h in an autoclave (Parr). The slurry was centrifuged and washed with deionized (DI) water several times until the neutral pH was reached. Solid particles were separated and dried overnight at 110 oC. After ammonium nitrate ion exchange and calcination as described in section 2.1, the sample was labeled as H-R.

2.3 Acid dealumination The recrystallized sample (H-R) was dealuminated using 1 M hydrochloric acid to partially remove the aluminum and other metals. The dealumination was performed under reflux with constant stirring at 350 rpm for 8 h and 24 h and maintained temperature at 85 oC. The samples washed several times with deionized water. Next, the samples dried overnight at 110 oC and calcined for 8 h at 550 oC under static air labeled as H-R-8 and H-R-24 for dealumination time of 8 h and 24 h, respectively.

2.4 Material characterizations Zeolite phase has been studied by using powder X-ray diffraction (XRD, Miniflex-Rigaku) with angle 2θ from 5o to 50o. Scanning rate was 3o/min with a step size 0.03o. The morphology and particle sizes of samples were examined by field-emission scanning electron microscopy (FE-SEM, LYRA 3 Dual Beam, Tescan). Aluminum structure was studied by the

27

Al Magic-Angle Spin 5

ACS Paragon Plus Environment

Page 7 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Nuclear Magnetic Resonance (27Al MAS-NMR). Elemental composition was measured by X-ray fluorescence (XRF) (Bruker).

Physisorption adsorption-desorption was carried out in ASAP 2020 (Micromeritics). The samples were heated up to 350 oC and maintained for 6 h before nitrogen adsorption. The liquid nitrogen was used to maintain the temperature of sample at -196 oC during the nitrogen gas adsorptiondesorption. Pore size distributions were estimated by DFT-Tarazona method.

Acid strength was performed by ammonia-temperature program desorption (NH3-TPD, Micromeritics ChemiSoft TPx V1.02). The sample was heated up to 600 oC and dwelled for degassing 30 min under flowing of 25 ml/min helium. Thereafter, temperature was decreased to 120 oC. Ammonia introduced to the sample for 30 min, then flush by helium for 1 h. Subsequently, the temperature was raised by rate 10 oC/min from 120 oC to 700 oC for ammonia desorption.

The Nicolet 6700 spectrometer was used for pyridine-FTIR study. The samples for the pyridineFTIR study were prepared by pelletizing 20 mg of sample. The pellet was dried at 500 oC for 30 min. After cooling down to 150 oC, pyridine was introduced into the sample chamber for 15 min followed by evacuating the pyridine that physically adsorbed for 30 min. Spectra were taken at 150 o

C, 250 oC, and 400 oC. The parameters of resolution 8 cm-1 in the range wavenumbers of 400–

4000 cm-1 and 100 scans were applied during the analysis. Calculation of pyridine FTIR adsorbed followed equation published elsewhere.28

2.5 Catalyst evaluation The catalyst powder (0.5 g) was tested in a fixed bed reactor made of Inconel ID 5.16 mm equipped with a gas chromatography (GC) on line analysis. The temperature was increased from ambient 6 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 35

temperature to 500 oC for 30 min and dwell for 30 min under a flow rate of 20 mL/min nitrogen. The temperature was reduced to 350 oC for 30 min followed by feeding 2 mL/min of n-butane (99.5%) and 20 mL/min nitrogen and dwelled 200 min for the n-butane isomerization under atmospheric pressure. The products was observed by using GC (Agilent 7890a) with two columns (Agilent HP-INNOWax PEG and Agilent J&W GC-GasPro) and flame ionization detectors.

3. Results and discussion 3.1 Characterization of the samples The XRD pattern of the parent sample (H-P) shows that mordenite is the major phase with several accessories, i.e., clinoptilolite (HEU) and quartz (Figure 2). The crystallinity of all phases was reduced after the high-energy ball milling treatment as the intensity peaks were decreased as shown by the XRD pattern of the H-M sample at 2θ: 9.6o, 22.5o, 25.8o, 27o and 28o. The hydrothermal recrystallization step has successfully recovered back the MOR crystalline phase. This is indicated by the higher intensity of XRD peaks of mordenite on the H-R as compared with the H-M and H-P. Interestingly, some peaks, which belong to HEU in the parent, disappeared after milling and did not recover after recrystallization such as the peaks on 2θ: 11, 13, 17.3, 26.2, and 30.5 degrees. The solution composition, temperature, and time applied on the hydrothermal treatment was selective to the growth of mordenite crystals. Acid dealumination over the recrystallized nanoparticles reduced the crystallinity as showed in the XRD patterns of the H-R-8 and H-R-24 due to the partial aluminum removal of the MOR framework.

The

27

Al NMR spectra of H-MOR samples are presented in Figure 3. The parent natural zeolites

sample showed a typical zeolite tetrahedral (AlO4) peak at 55 ppm with a small peak at 0 ppm attributed to the extra framework aluminum (EFAL) of the octahedral (AlO6). The ball milling led to the partial distortion of AlO4 structure as shown by the higher intensity of shoulder on the main 7 ACS Paragon Plus Environment

Page 9 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

peak at 55 ppm of the H-M curve. This finding is in agreement with the literature.29 A slight increase in the AlO6 peak at 0 ppm was also observed on the milled sample, which indicated that the ball milling positively affected the EFAL formation. Recrystallization step removed the AlO6 peak and reduced the shoulder peak at 55 ppm which suggested that the tetrahedral AlO4 has been recovered. The peak at 0 ppm in the dealuminated mordenite is attributed to the EFAL formation of AlO6. The peak became higher as the time of acid dealumination was applied longer as shown in the H-R-24. The intensity of AlO4 also decreased after dealumination as shown on the 55 ppm.

Recrystallization of the milled nanoparticles using basic silicate solution has increased the silicon to aluminum ratio from 6 to 9 based on the XRF analysis. Silica was introduced in the recrystallization solution to recover the crystallinity of the milled sample hence it was not surprising that the silica content increased in the H-R. Despite the high alkalinity, the high silica content in the recrystallization solution avoided desilication of the milled powder. The dealumination with 1 M HCl over the recrystallized nanoparticles partially removed the aluminum. The Si to Al ratio increased from 9 to 31, when acid dealumination applied for 8 h. The longer acid treatment time (24 h) led to the less aluminum content in the sample and the silicon to aluminum ratio of H-R-24 increased to 39.

Textural properties of the H-MOR samples were analyzed by nitrogen physisorption (Table 1). The isotherms of all samples were presented in Figure 4. The external surface area increased more than 4-fold after milling step from 23 to 100 m2/g, indicating the particles size reduced significantly. Unfortunately, the micropore was partially demolished after the milling step, as shown by the decreasing of micropore volume. The micropores recovered and increased after recrystallization as compared with the milled and parent samples (Table 1). The dealumination increased both the micropore and mesopores volume. It can be seen after 8 h dealumination, the micropore increased from 0.090 cm3/g to 0.105 cm3/g and the mesopore increased from 0.213 cm3/g to 0.235 cm3/g. The 8 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

dealumination for 24 h increased the micropore from 0.105 cm3/g to 0.115 cm3/g, while the mesopore remains unchanged. The increasing in micropore over mild acid dealumination was also reported elsewhere due to the opening pore of the side pocket 8-MR and creation the secondary micropore of the MOR channel after aluminum removal from its framework.30

Pore size distribution by using the Density Functional Theory (DFT) method is presented in Figure 5. The micropore size was within the size range 5.9 Å to 6.7 Å, which is typical of the mordenite pore. The mesopores in the milled and recrystallized samples most probably were voids formed as the nanosized particles stacking, which called the intercrystalline mesopore. It was the milling step that created hierarchical pore system of zeolite as the nanosized was formed during the ball milling. The aluminum was partially removed after dealumination of the recrystallized nanoparticles, creating new micropores and intracrystalline mesopores channels. This is in agreement with our previous work on dealumination of the natural mordenite which has large size crystals.25, 31 The acid dealumination increased the micropore volume within the size range from 6.3 to 6.7 Å. The mesopores started from size 3.5 nm and increased gradually to 45 nm. The hierarchical pore system consisted of micropore, intercrystalline mesopore, and intracrystalline mesopore was observed in the dealuminated nanosized samples (H-R-8 and HR-24).

Ammonia-TPD was conducted to study the acidity of samples (Figure 6). The weak acid sites (WAS) are shown from 140 oC to 340 oC and the strong acid sites (SAS) are shown in the curve from 340 oC to 700 oC. The milling step reduced the total number of acid sites from 326 to 125 µmol NH3/g (Table 2) due to the collapse of the micropore channels as shown by the nitrogen physisorption study. After recrystallization, the total number of acid sites increased to 362 µmol NH3/g, which is higher than the milled and parent sample. This fact is in agreement with the higher micropore volume of the recrystallized sample over the milled and parent sample (Table 1). The dealumination step reduced the total acid sites both the weak and strong acid sites (Figure 6). In 9 ACS Paragon Plus Environment

Page 10 of 35

Page 11 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

fact, the decreasing of the weak acid sites was observed in significant amount as compared with the SAS. Before dealumination, the ratio of WAS to SAS of the samples (H-P, H-M, and H-R) were more than 1. After dealumination, the ratio of WAS to SAS of the dealuminated samples were less than 1. Niwa et.al32 reported that the BAS in the 12-MR of mordenite were reported weaker than the one in the 8-MR. The data suggested that the partial removal of aluminum affected much the weak acid sites in the large pore channels 12-MR. The dealumination probably open the access to the other strong acid sites in the small 8-MR, which was not accessible before the aluminum removal. As the result, the decreasing of strong acid sites was not observed in a large number as happened in the weak acid sites. To further investigate the dealumination effect on the acidity, we performed FTIR with pyridine as a probe, which is larger than the ammonia.

Pyridine-FTIR was conducted to identify the Brønsted and Lewis acid sites as presented in Table 3. The Brønsted acid sites (BAS) type was found to be dominant in all samples except the H-M, which showed a higher amount of Lewis acid sites (LAS). The Brønsted acid sites were not accessible as the micropore channels demolished after the ball milling treatment. The extraframework of aluminum formed after the sequential process of ball milling-ion exchangecalcination. The EFAL in the milled sample was confirmed by the 27Al NMR study (Figure 3). The LAS also increased significantly after HCl dealumination for 8 h. It was created as the aluminum partially removed from the framework and formed the EFAL as indicated by the

27

Al NMR

analysis. The LAS over the dealuminated sample for 24 h were higher than the 8 h acid treatment as the removal of the aluminum from the framework became intense. As we expected, the BAS over the dealuminated H-R-8 and H-R-24 were higher than the non-dealuminated MOR (H-R). The dealumination of recrystallized MOR has opened the access to the acid sites in the side pockets 8MR channel for large probe molecules such as pyridine ca. 5.7 Å due to the partial removal of aluminum of the side pockets.33 In contrast, the small probe ammonia (ca. 3 Å) was able to penetrate into the small 8-MR channel of the non-dealuminated MOR.34 As the result, the total acid 10 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 35

sites of the recrystallized MOR (H-R) was higher as compared with the dealuminated MOR in the ammonia-TPD analysis (Table 2). We can conclude that the acid dealumination on MOR has opened the access to the acid sites on the side pockets 8-MR channel of MOR, which contains strong acid sites.

The SEM images were studied to measure morphology and particle size of the samples (Figure 7 & 8). The parent crystal shape (H-P) was in irregular form, however, a few of them represented the needle shape (indicated by arrows in Figure 7a). After the milling step, the particle shape became more regular with the spherical dominant shape (Figure 7b). Particle size distribution based on the SEM images is presented in histograms (Figure 8). The parent H-P consisted of large particle size was in within the range of 1 to 10 µm. High-energy ball milling was effectively reduced the particle size into nanosized with the size of 20 to 160 nm with 80% of the particle size less than 100 nm (Figure 8b). The recrystallization process was slightly increased the particle sizes due to the growth of crystalline MOR in the hydrothermal step as it confirmed by XRF with the increasing of silica content on the sample of H-R (Figure 8c). The particle size distribution of dealuminatedrecrystallized 8 h (H-R-8) and 24 h (H-R-24) showed similar size with the recrystallized MOR (Figure 8d and 8e). The average particle size of the H-R, H-R-8, and H-R-24 were about 90 nm.

3.2 The particle size effect on n-butane isomerization Initially, the n-butane conversion over the nanosized recrystallized (H-R) was slightly higher ca. 14% as compared with the microsized parent (H-P) ca.13% at a time on stream (TOS) 10 min (Figure 9). It was due to the H-R sample contained a slightly higher total acid sites indicated by the pyridine-FTIR as compared with the H-P. After 50 min reaction, the n-butane conversion over the H-P was significantly decreased to only 0.1%. In contrast, the catalyst deactivation over the H-R was slower with the n-butane conversion remained high, ca. 8% at TOS 50 min. The unidimensional pore of mordenite is prone to the coke formation, which blockage the pore hence the 11 ACS Paragon Plus Environment

Page 13 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

molecules unable to reach the acid sites. We observed a more darkening color of the spent parent catalyst as compared to the spent recrystallized sample and the spent milled sample, indicating a severe coke deposition in the parent catalyst (Figure S1). Wulfers and Jentoft35 studied the coke deposition on mordenite in the n-butane isomerization. They found that the coke mainly consisted of methyl-substituted naphthalenes, anthracenes, tetracene and polycyclic aromatic species. The HR nanoparticles has a large external surface area and shorter diffusional pathway, as a consequence, the molecules easily to react on the surface and diffuse out hence retarded the coke deposition.

The particle size of H-MOR has also affected the product distribution of n-butane conversion (Figure 9). The n-butane transformation over the nanosized H-R exhibited a higher isobutane product with 28% as compared with the microsized H-P with only 11% at TOS 200 min. It was the particle size that governed the higher isobutane selectivity of H-R nanoparticles. The bimolecular mechanism was most likely the main route as it required a high external surface area, which is provided in the nanosized H-R. This finding is supported by the literature discussed elsewhere.26 The side products of the bimolecular mechanism are propane and pentane via disproportionation. In fact, pentane and butane will undergo secondary reaction to produce more propane e.g., ‫ܥ‬ସ + ‫ܥ‬ହ ↔ 3‫ܥ‬ଷ .20 The nanoparticle size inhibited the secondary reaction as the pore length shorter than the microsized mordenite. It can be seen from Figure 9 that the propane selectivity at TOS 10 min was 60% over the H-R lower than the H-P with 75% at relatively similar conversion. These facts suggested that the nanoparticles size increased the isobutane formation and decreased the propane selectivity.

The propane selectivity was the most decreased products selectivity over the TOS in the recrystallized sample (H-R). It decreased from 58% (TOS=10 min) to 47% (TOS=200 min) (Figure 9c and Table 4). In contrast, the isobutane selectivity was slightly increase over the TOS from 25% to 27%. It can be explained that the bimolecular pathway occurred both on the mesopores/external 12 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 35

surfaces and micropores in the fresh catalyst. The reaction in micropores with high acid density led to high propane and low isobutane selectivity. Over the time, the microporous channel became no longer accessible due to the cokes blockage on the pore, which indicated by the sharp decrease of activity at TOS = 50 min. Practically, the butane transformation mainly occurred on the mesopores and external surfaces area, which reduced the propane formation of the secondary reaction.

The H-M exhibited the lowest activity since the beginning of reaction, although the particle size within the nanosized range (Figure 9b). It was due to the total acid sites in the H-M sample was the lowest among others as indicated by the NH3-TPD. It was also confirmed by pyridine-FTIR that the Brønsted of H-M was the lowest among other samples. The isomerization reaction of n-butane to isobutane was greatly affected by the BAS as reported elsewhere.36 At 10 min reaction, the selectivity to propane reached the highest with 87%, while the selectivity to isobutane was only 6%. This probably due to the acid sites on the external surface area were mostly damaged in the milling step, hence the isomerization through a bimolecular reaction which mainly favored on the external surface area reduced significantly. Only the small part of acid sites retained in the one-dimensional (1-D) micropores channel was active in the transformation of n-butane. The catalytic activity in the 1-D mordenite led to a high propane selectivity through the bimolecular pathway followed by the secondary reaction in the micropore.20 Although the external surfaces area contributed to the reaction products, however, the numbers were not significant as compared with the catalytic activity on the acid sites of micropore. This was confirmed through the product distribution after blockage of the micropore by coke led the catalyst deactivation which occurred at 50 min (Figure 9b). The product distribution shifted with high selectivity to isobutane ca. 36% and low to the propane ca. 32%. This was occurred most probably due to the decreasing of the propane formation in the micropore much higher instead of the decreasing of the isobutane product.

13 ACS Paragon Plus Environment

Page 15 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

The n-butane transformation over H-MOR was proposed through bimolecular mechanism by Asuquo et.al37 as follows: (i) Formation of a butyl carbenium ion (via protonation, hydride abstraction and thermal cracking) (ii) Formation of C8 carbenium ion via dimerization (iii) Formation of n-butane and isobutane via isomerization and (iv) Formation of propane and pentane via disproportionation. The intermediate large molecule C8 carbenium ion has to be formed in the bimolecular mechanism hence the plausible way of the bimolecular mechanism taking place is on the external surface area and mesopore.38 Product distribution is greatly affected by the mesopore presence on the dealuminated mordenite which favored the selectivity to isobutane as it prevents the secondary transformations of the reaction products during their diffusion throughout the onedimensional pore of the MOR.20, 39 We found that the high-density acid sites (Si/Al = 9) with large external surfaces area and high intercrystalline mesopore volume on the recrystallized mordenite (H-R) increased the selectivity to isobutane as compared with the parent.

A report suggested another explanation that the product distribution was also driven by butene concentration on the catalysts bed.40 The authors proposed that the low concentration of butene (120 ppm) favors the bimolecular mechanism. Figure 10 shows that butene concentration was remarkably high, which led to the conclusion that the bimolecular pathway occurred significantly with products of isobutane, propane, and pentane. The investigation of transformation propane over mordenite with Si/Al ratio of 10 showed that the propane conversion was very low41, indicating the high stability of propane. For low Si/Al ratio mordenite, the bimolecular pathway most probably occurred with secondary reactions of the isobutane with pentane in the micropore channel lead to high propane. These explain why the propane selectivity was very high over the microsized particles H-P. In fact, the propane selectivity over H-R was almost double of the isobutane.

14 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 35

3.3 Dealuminated of nanoparticles recrystallized mordenite The as-received natural zeolites were rich in aluminum (Si/Al = 6). The high aluminum content, particularly when the Si/Al ratio less than 6.1, disfavored the isobutane selectivity.20, 39 For low Si/Al ratio, it is necessary to have high mesopore and high external surface area to obtain a high selectivity to isobutane. Dealumination with mild acid concentration increased the activity of mordenite to convert n-butane and favored selectivity to isobutane. The silicon to aluminum ratio was increase to 31 after 8 h acid dealumination. The smaller particle size gives benefit in the dealumination process as the removal of the aluminum from the pore become faster.42 The ammonia-TPD study suggested that the acid sites reduced after the acid dealumination (Figure 6). The monomolecular reaction which is more selective to isobutane occurred on the high Si/Al ratio along with the bimolecular pathway. The monomolecular pathway demands the high strength acid sites, which are provided as the aluminum become more isolated (less acid site density) based on the NNN (next nearest neighbor) theory. The pyridine-FTIR study suggested that the accessibility of the probe molecules to the BAS in the side pocket 8-MR were increase after the dealumination. The BAS in the 8-MR were reported stronger than the one in the 12-MR.32 The ammonia-TPD study also revealed that the ratio of strong acid sites to the weak acid sites increased significantly after dealumination. In addition, the micropore and mesopore volume of the dealuminated samples were higher than the recrystallized sample. The increasing in the accessibility to the high strength acid sites in the dealuminated MOR favored the isomerization reaction through the monomolecular pathway. Furthermore, the bimolecular pathway most likely prefers the isomerization reaction instead of the disproportionation at low acid sites concentration as reported elsewhere.37 As the result, the selectivity to isobutane significantly increased to 58% and propane dropped to 27% at TOS 200 min in both of the dealuminated nanoparticles the H-R-8 and H-R-24 (Table 4).

15 ACS Paragon Plus Environment

Page 17 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

We found that the time of acid dealumination affected on the activity and stability of the catalyst. The more aluminum was removed from the framework, as dealumination applied in a longer time 24 h that led to higher silicon to aluminum ratio 39. The nitrogen physisorption study showed that the micropore volume increased along with the time of dealumination. The activity of H-R-24 was higher as compared with the H-R-8 (Figure 10b and 10c). It is probably due to the increasing access to the strong Brønsted acid sites in the side pocket 8-MR of the micropore as suggested by the nitrogen physisorption, ammonia-TPD and pyridine-FTIR study (Table 1, Table 2 and Table 3). In addition, the H-R-24 stability was higher as compared with the H-R-8 as the less acid concentration reduced the coke formation rate. It is worth mentioning that there was no increase in mesopore volume from 8 to 24 h acid dealumination treatment. Hence, we can conclude that the high stability of H-R-24 was most likely not only due to the nanosized but also caused by the less of acid sites density, which inhibited the coke formation.

The extraframework aluminum (EFAL) in the milled sample (H-M) and the dealuminated samples were high as indicated by the

27

Al NMR study. The EFAL acts as Lewis acid sites (LAS) which

might play a significant role in a catalytic reaction. However, the LAS in the milled sample had no positive effect on the n-butane isomerization as the activity was very low over the H-M sample. This is in agreement with a study by Babůrek and Nováková 36, which revealed that the presence of Lewis acid sites in zeolites decreased the n-butane conversion and selectivity to isobutane. The results of n-butane isomerization over H-M supports the conclusion that the high conversion of nbutane and selectivity to isobutane over the dealuminated samples was due to the high access to strong Brønsted acid sites in the side pocket.

We also investigated the performance of dealuminated parent (H-P-8) and dealuminated nanoparticles (H-R-8) under temperature 420 oC to further confirm that the particle size was important in the n-butane isomerization. The conversion of dealuminated nanoparticles was higher 16 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 35

as compared with the dealuminated parent as shown in Figure S2. It is suggested that the high external surface area of the H-R-8 led the molecules to easily reach the acid sites. The selectivity to isobutane over the H-R-8 was also higher than the H-P-8. This confirmed that the particle size of the sample before dealumination was important as the performance of the n-butane isomerization reaction was better on the dealuminated nanosized mordenite.

The high temperature has increased the conversion of n-butane over the dealuminated nanoparticle (H-R-8). However, selectivity to isobutane was reduced at high temperature (Figure S3). The yield of isobutane at 420 oC was slightly higher as compared with the temperature of 350 oC. At high temperature, the reaction rate increased, hence the conversion of n-butane was also higher. However, from the thermodynamic point of view, the equilibrium conversion of n-butane isomerization to the product of isobutane is decreased along with the increasing of temperature, which indicated that the n-butane isomerization is an exothermic reaction (Figure S4). Another thing is that the high temperature promoted disproportionation pathway which produced propane and pentane.37

4. Conclusions The mordenite nanoparticles have been fabricated through ball milling from natural mordenite followed by hydrothermal recrystallization to recover the crystallinity. The mordenite nanoparticles exhibited higher selectivity and stability in n-butane isomerization as compared with the microsized mordenite. The large surface area and high intercrystalline mesopore volume of the mordenite nanoparticles favored the isomerization through

the

bimolecular reaction

mechanism.

Dealumination over the mordenite nanoparticles further increased the selectivity to isobutane. The monomolecular pathway favored the selectivity to isobutane over the dealuminated samples due to the higher access to the strong acid sites in the 8-MR. The short diffusion pathway of the nanoparticles and the low acid sites density remarkably diminished the rate of coke deposition. 17 ACS Paragon Plus Environment

Page 19 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Acknowledgement The authors would like to thank the funding provided by Saudi Aramco for supporting this work through project contract number 6600011900 as part of the Oil Upgrading theme at King Fahd University of Petroleum and Minerals. References 1. Colella, C., Natural zeolites. In Studies in Surface Science and Catalysis, Čejka, J.; Bekkum, H. v., Eds. Elsevier: 2005; Vol. Volume 157, pp 13-40. 2. Mumpton, F. A., La roca magica: Uses of natural zeolites in agriculture and industry. Proceedings of the National Academy of Sciences of the United States of America 1999, 96, (7), 3463-3470. 3. Colella, C., Applications of Natural Zeolites. In Handbook of Porous Solids, Wiley-VCH Verlag GmbH: 2008; pp 1156-1189. 4. Ghasemi, Z.; Sourinejad, I.; Kazemian, H.; Rohani, S., Application of zeolites in aquaculture industry: a review. Reviews in Aquaculture 2016, n/a-n/a. 5. Baerlocher, C.; McCusker, L. B.; Olson, D. H., MOR - Cmcm. In Atlas of Zeolite Framework Types (Sixth Edition), Baerlocher, C.; Olson, L. B. M. H., Eds. Elsevier Science B.V.: Amsterdam, 2007; pp 218-219. 6. Simoncic, P.; Armbruster, T., Peculiarity and defect structure of the natural and synthetic zeolite mordenite: A single-crystal X-ray study. American Mineralogist 2004, 89, 10. 7. Sand, L. B., Synthesis of large-port and small-port mordenites. In Molecular sieves, Society of Chemical Industry, London, 1968; pp 71-77. 8. Raatz, F.; Marcilly, C.; Freund, E., Comparison between small port and large port mordenites. Zeolites 1985, 5, (5), 329-333. 9. Vermeiren, W.; Gilson, J. P., Impact of Zeolites on the Petroleum and Petrochemical Industry. Topics in Catalysis 2009, 52, (9), 1131-1161. 10. Adeeva, V.; Sachtler, W. M. H., Mechanism of butane isomerization over industrial isomerization catalysts. Applied Catalysis A: General 1997, 163, (1–2), 237-243. 11. Liu, H.; Lei, G. D.; Sachtler, W. M. H., Alkane isomerization over solid acid catalysts Effects of one-dimensional micropores. Applied Catalysis A: General 1996, 137, (1), 167-177. 12. Hincapie, B. O.; Garces, L. J.; Zhang, Q.; Sacco, A.; Suib, S. L., Synthesis of mordenite nanocrystals. Microporous and Mesoporous Materials 2004, 67, (1), 19-26. 13. Konno, H.; Okamura, T.; Kawahara, T.; Nakasaka, Y.; Tago, T.; Masuda, T., Kinetics of nhexane cracking over ZSM-5 zeolites – Effect of crystal size on effectiveness factor and catalyst lifetime. Chemical Engineering Journal 2012, 207, 490-496. 14. Jamil, A. K.; Muraza, O.; Yoshioka, M.; Al-Amer, A. M.; Yamani, Z. H.; Yokoi, T., Selective Production of Propylene from Methanol Conversion over Nanosized ZSM-22 Zeolites. Industrial & Engineering Chemistry Research 2014, 53, (50), 19498-19505. 15. Yuan, Y.; Wang, L.; Liu, H.; Tian, P.; Yang, M.; Xu, S.; Liu, Z., Facile preparation of nanocrystal-assembled hierarchical mordenite zeolites with remarkable catalytic performance. Chinese Journal of Catalysis 2015, 36, (11), 1910-1919. 16. Chu, W.; Li, X.; Zhu, X.; Xie, S.; Guo, C.; Liu, S.; Chen, F.; Xu, L., Size-controlled synthesis of hierarchical ferrierite zeolite and its catalytic application in 1-butene skeletal isomerization. Microporous and Mesoporous Materials 2017, 240, 189-196. 17. Charkhi, A.; Kazemian, H.; Kazemeini, M., Optimized experimental design for natural clinoptilolite zeolite ball milling to produce nano powders. Powder Technology 2010, 203, (2), 389396. 18 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 35

18. Kurniawan, T.; Muraza, O.; Hakeem, A. S.; Al-Amer, A. M., Mechanochemical Route and Recrystallization Strategy To Fabricate Mordenite Nanoparticles from Natural Zeolites. Crystal Growth & Design 2017, 17, (6), 3313-3320. 19. Kurniawan, T.; Muraza, O.; Miyake, K.; Hakeem, A. S.; Hirota, Y.; Al-Amer, A. M.; Nishiyama, N., Conversion of Dimethyl Ether to Olefins over Nanosized Mordenite Fabricated by a Combined High-Energy Ball Milling with Recrystallization. Industrial & Engineering Chemistry Research 2017, 56, (15), 4258-4266. 20. Tran, M. T.; Gnep, N. S.; Szabo, G.; Guisnet, M., Isomerization ofn-Butane over HMordenites under Nitrogen and Hydrogen: Influence of the Acid Site Density. Journal of Catalysis 1998, 174, (2), 185-190. 21. Villegas, J. I.; Kumar, N.; Heikkilä, T.; Smiešková, A.; Hudec, P.; Salmi, T.; Murzin, D. Y., A highly stable and selective Pt-modified mordenite catalyst for the skeletal isomerization of nbutane. Applied Catalysis A: General 2005, 284, (1–2), 223-230. 22. Yori, J. C.; D'Amato, M. A.; Costa, G.; Parera, J. M., Influence of platinum and hydrogen on n-butane isomerization on H-Mordenite. Reaction Kinetics and Catalysis Letters 1995, 56, (1), 129135. 23. Van Geem, P. C.; Scholle, K. F. M. G. J.; Van der Velden, G. P. M.; Veeman, W. S., Study of the transformation of small-port into large-port mordenite by magic-angle spinning NMR and infrared spectroscopy. J. Phys. Chem. 1988, 92, (6), 1585-1589. 24. Raatz, F.; Freund, E.; Marcilly, C., Study of small-port and large-port mordenite modifications. Part 1.-Preparation of the HM forms. Journal of the Chemical Society, Faraday Transactions 1: Physical Chemistry in Condensed Phases 1983, 79, (10), 2299-2309. 25. Nasser, G.; Kurniawan, T.; Miyake, K.; Galadima, A.; Hirota, Y.; Nishiyama, N.; Muraza, O., Dimethyl ether to olefins over dealuminated mordenite (MOR) zeolites derived from natural minerals. J. Nat. Gas Sci. Eng. 2016, 28, 566-571. 26. Cañizares, P.; de Lucas, A.; Dorado, F., n-Butane isomerization over H-mordenite: role of the monomolecular mechanism. Applied Catalysis A: General 2000, 196, (2), 225-231. 27. Viswanadham, N.; Kumar, M., Effect of dealumination severity on the pore size distribution of mordenite. Microporous and Mesoporous Materials 2006, 92, (1–3), 31-37. 28. Emeis, C. A., Determination of Integrated Molar Extinction Coefficients for Infrared Absorption Bands of Pyridine Adsorbed on Solid Acid Catalysts. Journal of Catalysis 1993, 141, (2), 347-354. 29. Sato, K.; Wakihara, T.; Kohara, S.; Ohara, K.; Tatami, J.; Endo, A.; Inagaki, S.; Kawamura, I.; Naito, A.; Kubota, Y., Characterization of Amorphized Zeolite A by Combining High-Energy Xray Diffraction and High-Resolution Transmission Electron Microscopy. The Journal of Physical Chemistry C 2012, 116, (48), 25293-25299. 30. Viswanadham, N.; Dixit, L.; Gupta, J. K.; Garg, M. O., Effect of acidity and porosity changes of dealuminated mordenites on n-hexane isomerization. Journal of Molecular Catalysis A: Chemical 2006, 258, (1–2), 15-21. 31. Nasser, G. A.; Kurniawan, T.; Tago, T.; Bakare, I. A.; Taniguchi, T.; Nakasaka, Y.; Masuda, T.; Muraza, O., Cracking of n-hexane over hierarchical MOR zeolites derived from natural minerals. Journal of the Taiwan Institute of Chemical Engineers 2015. 32. Niwa, M.; Suzuki, K.; Katada, N.; Kanougi, T.; Atoguchi, T., Ammonia IRMS-TPD Study on the Distribution of Acid Sites in Mordenite. The Journal of Physical Chemistry B 2005, 109, (40), 18749-18757. 33. Nesterenko, N. S.; Thibault-Starzyk, F.; Montouillout, V.; Yuschenko, V. V.; Fernandez, C.; Gilson, J. P.; Fajula, F.; Ivanova, I. I., Accessibility of the acid sites in dealuminated small-port mordenites studied by FTIR of co-adsorbed alkylpyridines and CO. Microporous and Mesoporous Materials 2004, 71, (1–3), 157-166.

19 ACS Paragon Plus Environment

Page 21 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

34. Bradley, S. A.; Broach, R. W.; Mezza, T. M.; Prabhakar, S.; Sinkler, W., Zeolite Characterization. In Zeolites in Industrial Separation and Catalysis, Wiley-VCH Verlag GmbH & Co. KGaA: 2010; pp 85-171. 35. Wulfers, M. J.; Jentoft, F. C., Identification of carbonaceous deposits formed on Hmordenite during alkane isomerization. Journal of Catalysis 2013, 307, 204-213. 36. Babůrek, E.; Nováková, J., Isomerization of n-butane over acid zeolites: Role of Broensted and Lewis acid sites. Applied Catalysis A: General 1999, 185, (1), 123-130. 37. Asuquo, R. A.; Edermirth, G.; Lercher, J. A., n-Butane Isomerization over Acidic Mordenite. Journal of Catalysis 1995, 155, (2), 376-382. 38. Cañizares, P.; de Lucas, A.; Dorado, F.; Durán, A.; Asencio, I., Characterization of Ni and Pd supported on H-mordenite catalysts: Influence of the metal loading method. Applied Catalysis A: General 1998, 169, (1), 137-150. 39. Tran, M. T.; Gnep, N. S.; Szabo, G.; Guisnet, M., Comparative study of the transformation of n-butane, n-hexane and n-heptane over H-MOR zeolites with various Si/Al ratios. Applied Catalysis A: General 1998, 170, (1), 49-58. 40. Wulfers, M. J.; Jentoft, F. C., Mechanism of n-butane skeletal isomerization on H-mordenite and Pt/H-mordenite. Journal of Catalysis 2015, 330, 507-519. 41. Guisnet, M.; Bichon, P.; Gnep, N. S.; Essayem, N., Transformation of propane, n-butane and n-hexane over H3PW12O40 and cesium salts. Comparison to sulfated zirconia and mordenite catalysts. Topics in Catalysis 2000, 11, (1), 247-254. 42. Kuznicki, S. M.; Whyte, J. R., Ion-exchange agent and use thereof in extracting heavy metals from aqueous solutions. In Google Patents: 1993.

20 ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Crushing & Sieving natural zeolite (100300 µm)

P

Ball milling

Page 22 of 35

M

M

P

Ion exchange NH4NO3 & Calcination H-R

Acid (1 M HCl) dealumination 8 h & 24 h - Calcination

H-R-8 H-R-24

H-P H-M H-R

Catalyst Testing

Figure 1. Flow chartPlus ofEnvironment the experiment. ACS Paragon

Recrystallization R

Page 23 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

H-R-24 H-R-8 H-R m c

m c

c

m

5

m m

10

m = Mordenite c = Clinoptilolite q = Quartz

15

c

m

mm q

H-M m

q

20

H-P

cm c

25

30

35

40

45

2θ [o]

Figure 2. XRD patterns of the parent and treated samples. ACS Paragon Plus Environment

50

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Page 24 of 35

H-R

Shoulder

H-M

H-R-24

H-P

H-R-8

100

50 0 Chemical Shift [ppm]

-50

100

50 0 Chemical Shift [ppm]

Figure 3. 27Al NMR spectra of the parent and treated samples. ACS Paragon Plus Environment

-50

Page 25 of 35

400 H-R-24

350

H-R-8 300

N2 Adsorbed [mL/g]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

H-R

250

H-P

200

H-M

150 100 50 0 0

0.2

0.4 0.6 Relative Pressure [-]

0.8

Figure 4. Nitrogen isotherms of the parent and treated samples. ACS Paragon Plus Environment

1

Page 26 of 35

0.5

H-R-24

0.4

H-R-8

0.3

H-R

0.2

H-P H-M

0.1 0 0

Cumulative pore volume [mL/g]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Cumulative pore volume [mL/g]

Energy & Fuels

75

150 225 300 Pore size [Å]

375

450

0.25 0.2

H-R-24

0.15

H-R-8

0.1

H-R

0.05

H-P H-M

0 5

6

7 8 Pore size [Å]

9

10

Figure 5. Cumulative pore volume versus pore size of the parent and treated samples by DFT ACS Paragon Plus Environment method.

Page 27 of 35

Weak acid sites

0.012

Strong acid sites

0.01

TCD signal [a.u.]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

0.008 0.006

0.004 0.002 0 100 H-R

200 H-P

300 400 500 Temperature [oC] H-M H-R-8

600

700 H-R-24

Figure 6. Acidity of the parent and treated samples by ammonia-TPD. ACS Paragon Plus Environment

Energy & Fuels

(a) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

(b)

500 nm

(c)

500 nm

(d)

500 nm

Page 28 of 35

(e)

500 nm

500 nm

Figure 7. SEM images of the parent ACS andParagon treated samples: (a) H-P, (b)H-M, (c) H-R, (d) H-R-8, Plus Environment and (e) H-R-24.

Page 29 of 35

50

(c) H-R

40 30 20 10 0 1-2

2-3

3-4

4-5

5-6

6-10

Particle size [µm] Particle size [nm]

(d) H-R-8 50

Frequency [%]

Frequeny [%]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

(b) H-M

(a) H-P

Particle size [nm]

(e) H-R-24

40

30 20 10 0

Particle size [nm]

Particle size [nm]

Figure 8. Particle size distribution of the parent and treated samples. ACS Paragon Plus Environment

Energy & Fuels

(a) H-P

(b) H-M

80

80

60

60

40

40

20

20

0

0 10

50 100 150 200

16 14 12 10 8 6 4 2 0

100

50

100

150

200

80 60

40 20

0 10

10 50 100 150 200 Time on stream [min] C1-2 C3 i-C4 C4= i-C5 16 14 12 10 8 6 4 2 0

0

(c) H-R Selectivity [%]

100

Selectivity [%]

100

Yield/Conversion [%]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Page 30 of 35

Conversion C3 i-C4 C4= i-C5 C1-2 0

50

100

150

200

50 100 150 200

16 14 12 10 8 6 4 2 0 0

50

100

150

Time on stream [min]

ACS Paragon Figure 9. Isomerization of n-butane overPlus (a)Environment H-P, (b) H-M, and (c) H-R at T = 350 oC, C4H10= 2 ml/min, N2 = 20 ml/min.

200

Page 31 of 35

(b) H-R-8

(a) H-R

(c) H-R-24

100

100

100

80

80

80

60

60

60

40

40

40

20

20

20

0

0

Yield/Conversion [%]

Selectivity [%]

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

0 10

50 100 150 200

18

10 50 100 150 200 Time on stream [min] C1-2 C3 i-C4 C4= i-C5

10

18

18

15

15

12

12

9

9

6

6

6

3

3

3

0

0

0

Conversion C3 i-C4 C4= i-C5 C1-2

15 12

9

0

50

100

150

200

0

50 100 150 200 Time on stream [min]

0

50 100 150 200

50

100

150

ACS over Paragon(a) PlusH-R, Environment Figure 10. Isomerization of n-butane (b) H-R-8, and (c) H-R-24 at T = 350 oC, C4H10= 2 ml/min, N2 = 20 ml/min.

200

Energy & Fuels

Page 32 of 35

Table 1. Textural properties and silicon to aluminum ratio of the parent and treated samples.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

SBET (m2g-1)

St (m2g-1)

Sext (m2g-1)

Vtotal (cm3 g-1)

Vmicro (cm3 g-1)

Vmeso (cm3 g-1)

Si/Al* (-)

H-P

183

160

23

0.100

0.068

0.032

6

H-M

131

31

100

0.238

0.013

0.225

6

H-R

286

215

71

0.303

0.090

0.213

9

H-R-8

333

254

79

0.340

0.105

0.235

31

H-R-24

354

272

83

0.350

0.115

0.235

39

Sample

*XRF

ACS Paragon Plus Environment

Page 33 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

Table 2. Number of acid sites of the parent and the treated samples. Sample

Weak acid sites (WAS) [µmol/g]

Strong acid sites (SAS) [µmol/g]

WAS/SAS [-]

Total acid sites by NH3-TPD [µmol/g]

H-P

192

134

1.4

326

H-M

80

45

1.8

125

H-R

210

152

1.4

362

H-R-8

74

110

0.7

184

H-R-24

63

107

0.6

170

ACS Paragon Plus Environment

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Page 34 of 35

Table 3. Brønsted and Lewis acidity of the parent and treated samples.

Sample

Brønsted acid sites (µmol/g)

Lewis acid sites (µmol/g)

150 oC

250 oC

400 oC

150 oC

250 oC

400 oC

H-P

67

48

27

23

17

17

H-M

34

23

2

38

31

26

H-R

107

89

50

25

17

17

H-R-8

125

105

58

51

46

44

H-R-24

111

97

57

57

51

47

ACS Paragon Plus Environment

Page 35 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Energy & Fuels

Table 4. Conversion of n-butane transformation and products distribution over the catalysts at TOS = 10 min and 200 min (in the bracket).

Sample

Conversion (%)

Selectivity (%) C1

C2

C2=

C3

C3=

i-C4

C4=

i-C5

H-P

13 (1)

1 (4)

1 (3)

0 (9)

72 (42)

3 (4)

18 (11)

4 (22)

1 (5)

H-M

2 (0)

0 (0)

0 (0)

2 (0)

85 (0)

2 (0)

6 (0)

5 (0)

0 (0)

H-R

14 (4)

0 (1)

0 (1)

2 (4)

58 (47)

2 (2)

25 (27)

7 (11)

5 (7)

H-R-8

15 (6)

0 (0)

1 (0)

2 (2)

37 (27)

1 (1)

49 (58)

7 (9)

3 (4)

H-R-24

17 (11)

0 (0)

1 (0)

2 (1)

31 (27)

1 (1)

54 (58)

7 (9)

3 (4)

ACS Paragon Plus Environment