On the Origin and Structure Characteristics of Tri-coordinated

Subscriber access provided by UOW Library

Article

On the Origin and Structure Characteristics of Tri-coordinated Extraframework Aluminum Species in Dealuminated Zeolites Xianfeng Yi, Kangyu Liu, Wei Chen, Junjie Li, Shutao Xu, Chengbin Li, Yao Xiao, Haichao Liu, Xinwen Guo, Shang-Bin Liu, and Anmin Zheng J. Am. Chem. Soc., Just Accepted Manuscript • Publication Date (Web): 02 Aug 2018 Downloaded from http://pubs.acs.org on August 2, 2018

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 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 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 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.

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 34 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

Journal of the American Chemical Society

On

the

Origin

and

Structure

Characteristics

of

Tri-coordinated Extraframework Aluminum Species in Dealuminated Zeolites

Xianfeng Yi,a Kangyu Liu,b Wei Chen,a,f Junjie Li,c Shutao Xu,d Chengbin Li,a Yao Xiao,a,f Haichao Liu,b Xinwen Guo,c Shang-Bin Liu,e and Anmin Zhenga,g,*

a

State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, National

Center for Magnetic Resonance in Wuhan, Key Laboratory of Magnetic Resonance in Biological Systems, Wuhan Institute of Physics and Mathematics, Chinese Academy of Sciences, Wuhan 430071, P. R. China; b

Beijing National Laboratory for Molecular Sciences (BNLMS), State Key Laboratory for

Structural Chemistry of Stable and Unstable Species, College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, P. R. China; c

State Key Laboratory of Fine Chemicals, PSU-DUT Joint Center for Energy Research, School of

Chemical Engineering, Dalian University of Technology, Dalian, 116024, P. R. China. d

National Engineering Laboratory for Methanol to Olefins, Dalian National Laboratory for

Clean Energy, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, Dalian 116023, P. R. China e

Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan

f

University of Chinese Academy of Sciences, Beijing 100049, P. R. China;

g

Wuhan-Oxford Joint Catalysis Laboratory, Wuhan 430071, P. R. China.

*Corresponding author: [email protected];

ACS Paragon Plus Environment

Journal of the American Chemical Society 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 34

Abstract Post-synthetic dealumination treatment is one of the common tactics adopted for industrialized zeolitic catalysts to improve their catalytic performances through the enhancements in acidity and stability. However, among the possible extra-framework aluminum (EFAL) species in dealuminated zeolites such as Al3+, Al(OH)2+, Al(OH)2+, AlO+, AlOOH, and Al(OH)3, the presence of tri-coordinated EFAL-Al3+ species, which exhibit large quadrupolar effect due to lack of hydrogen bonding species, were normally undetectable by conventional one- and two-dimensional 1H and/or

27

Al

solid-state nuclear magnetic resonance (SSNMR) techniques. By invoking a combined theoretical density functional theory (DFT) calculations with experimental 31

P SSNMR using trimethylphosphine (TMP) as the probe molecule, we report herein

a comprehensive study to certify the origin, fine structure, and possible location of tri-coordinated EFAL-Al3+ species in dealuminated HY zeolite. It is conclusive that the spatial proximities and synergies between the Brønsted and various Lewis acid sites were clearly identified, and the origin for the observed EFAL-Al3+ species with ultra-strong Lewis acidity was deduced to be attributed to the expense of adjacent Brønsted acid sites. The excellent performances for such tri-coordinated EFAL species were furthermore confirmed by glucose isomerization reaction.

KEYWORDS: Dealuminated zeolite; Tri-coordinated extra-framework aluminum; Acidity; NMR spectroscopy; Trimethylphosphine; DFT calculations.


Page 3 of 34 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


1. Introduction Owing to their unique acidic and structural properties, zeolites have been extensively employed as solid acid catalysts in chemical and petrochemical industries for various reaction processes such as cracking, alkylation, disproportionation, and isomerization etc.[1−6] It is well-known that a post-synthesis hydrothermal treatment of zeolites, which leads to the formations of Lewis acidic extra-framework aluminum (EFAL) species, is beneficial for improving their thermal stabilities and catalytic performances.[7−9] As revealed by acidity characterization using solid-state 13C NMR of the adsorbed 2-13C-labeled acetone as probe,[10] the enhanced interactions between the EFAL species and structural framework of dealuminated zeolites readily provoke enhancement in Brønsted acidity.[11] However, detailed interaction mechanisms invoking synergistic effects of Brønsted and Lewis acid sites, arising from the bridging hydroxyl (Si−OH−Al) and EFAL species, respectively, remain lacking. Thus far, a group of possible EFAL species in dealuminated zeolites have been proposed,[11−16] including cationic AlO+, Al(OH)2+, AlOH2+, and Al3+ moieties as well as some neutral species such as AlOOH and Al(OH)3. For example, extra-framework Al(OH)3 and AlOH2+ species have been identified in dealuminated HY zeolites[11] by means of 1H double quantum magic-angel spinning (DQ MAS) NMR technique[17] through information on H-H proximities of hydroxyl groups. In addition,

27

Al DQ

MAS NMR has also been employed to explore the Al-Al proximities between the framework aluminum (FAL) and EFAL species for clarifying the presences of Al(OH)3, Al(OH)2+ and AlOH2+ moieties in different dealuminated zeolites.[15,16]



Page 4 of 34

Alternatively, the acid properties of various EFAL species may be probed by

31

P

solid-state NMR (SSNMR) of adsorbed trimethylphosphine (TMP) as the probe molecule.[18−27] Such a 31P-TMP SSNMR approach has been shown to be a practical tool for probing Lewis acidic strengths of various acid catalysts, which normally span over a wide 31P chemical shift (δ31P) range of ca. −20 to −60 ppm.[18−27] Moreover, as verified by results obtained from theoretical density functional theory (DFT) calculations,[20] a linear correlation between the δ31P of the adsorbed TMP probe molecule and Lewis acidic strengths may be inferred, making the

31

P-TMP SSNMR

approach a reliable technique for probing Lewis acidity. Whilst in the case of Brønsted acidity, couplings of the TMP probe molecule with Brønsted acidic protons in zeolites normally lead to the formation of TMPH+ ionic complexes, which typically give rise to

31

P resonances within a narrow δ31P range of only ca. −2 to −5 ppm. In

this context, the

31

P SSNMR approaches using TMP as probe molecules are

preferable choices for synergistically probing Brønsted and Lewis acidic properties and structure-activity correlations of solid acid catalysts.[20−25] Aside from various aforementioned EFAL species identifiable by different SSNMR techniques, the possible formation of tri-coordinated EFAL (Al3+) species has also been hypothesized.[13] Nevertheless, owing to the lack of hydrogen bonding, the

27

Al nuclei of the tri-coordinated Al3+ species were found to possess large

quadrupole coupling constants (QCCs),[26] thus, undetectable by conventional one-dimensional (1D) and two-dimensional (2D) 1H and/or 27Al SSNMR methods. By exploiting the

31

P-TMP SSNMR approach and 2D multinuclear SSNMR techniques


Page 5 of 34 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


together with DFT calculations, this comprehensive study aims to explore the origin, structural characteristics, and acidic property of tri-coordinated Al3+ species in dealuminated HY and HUSY zeolites. As will be shown later, this may be accomplished by probing the interactions between various Lewis acid sites arising from EFAL species and Brønsted acid sites originated from the framework bridging hydroxyl groups. Accordingly, a mechanism of formation was also proposed for the highly active, ultra-strong Lewis acidic EFAL-Al3+ species.

2.

Results and discussion

2.1. Structural properties The structural properties of HY and HUSY zeolites before and after the dealumination treatment were examined by X-ray diffraction (XRD) and SSNMR spectroscopy. As revealed by the XRD results in Fig. S1a of the Supplementary Information (hereafter denoted as SI), the structural integrity of the HY and HUSY zeolites remained practically intact after the dealumination treatment. On the other hand, notable differences in the

29

Si MAS NMR spectra of HY and HUSY were

observed compared to their respective dealuminated counterpart (Fig. S1b; SI). This is owing to the dealumination treatment, which resulted in partial collapse of tetrahedral-coordinated FAL species to form octahedral-coordinated EFAL species, as revealed by the

27

Al MAS NMR spectra (Fig. S1c; SI). The parent HY and HUSY

zeolites exhibited mostly FAL species with a

27

Al chemical shift (δ27Al) of 60 ppm,

whereas additional resonances at −1 ppm was observed for dealuminated samples,



indicating the presence of EFAL species. For the dealuminated HUSY-d450, extra broad signal at ca. 40 ppm was also observed, which may be ascribed due to penta-coordinated EFAL or distorted tetrahedral FAL species.[15] Through analyses of peak areas in the deconvoluted

29

Si MAS NMR spectra (Fig. S1b; SI), information

and concentration of various silicon local chemical environments, namely Si(1Si,3Al)), Si(2Si,2Al), Si(3Si,1Al), and Si(4Si) may be obtained, as depicted in Table S1 (SI). A notable increase in Si/Al ratio after the dealumination treatment was observed for both HY and HUSY zeolites, as expected.

2.2. Acid properties It has been demonstrated that 31P NMR of adsorbed TMP as the probe molecule is a reliable technique for characterization of Lewis acidity in solid acid catalysts.[18−27] As shown in Fig. 1a, the 31P MAS NMR spectra of TMP adsorbed on the parent HY, parent HUSY zeolites and their respective dealuminated counterparts revealed overlapping broad resonances, which may be deconvoluted up to five peaks with δ31P at 28, −5, −33, −47, and −57 ppm. To obtained spectra with enhanced sensitivity and resolution, spectra obtained with cross-polarization (CP) were also obtained, as shown in Fig. 1b. The

31

P CP/MAS NMR spectrum of TMP adsorbed on the parent HY

revealed only a single resonance at −5 ppm, which may be unambiguously attributed to the formation of TMPH+ ionic complex associated with the Brønsted acid sites. On the other hand, additional peaks at 28, −33, −47, and −57 ppm were observed for TMP adsorbed on dealuminated HY-d450. The weak peak at ca. 28 ppm has been


Page 6 of 34

Page 7 of 34 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


assigned due to P(CH3)4+ species formed during the disproportionation of TMP probe molecules.[19] Whereas,

31

P resonance peaks at −33, and −47 ppm should be arose

from TMP bound to Lewis acid sites with different acid strengths.[20] The signal at −33 ppm whose δ31P exceeds that of superacidic AlCl3 (−41 ppm),[27] revealing the presence of ultra-strong Lewis acid sites in both dealuminated HY-d450 and HUSY-d450 zeolites, which may be temporarily assigned to Lewis acidic EFAL-Al3+ species (as confirmed by the DFT theoretical calculations; vide infra). It is noteworthy that the assignment of Lewis acidity may further be verified by partial hydration of the respective sample, as illustrated in Fig. S2a (SI). Owing to the highly water-affinic nature, Lewis acid sites will quickly interact with water to transform into Brønsted acidity when exposed to moisture, leading to a notable diminishing of the

31

P

resonances associated with Lewis acid sites. Clearly, dealumination treatment of zeolite indeed create EFAL species with characteristics of Lewis acidity. Interestingly, a notable enhancement in the peak at the intensity of

31

P signal at −33 ppm was

observed in the dealuminated HUSY-d450 sample (Fig. 1b), indicating a significant increase in the amount of this strong Lewis acid sites, in excellent agreement with results obtained from acidity characterization by the 31P-TMP SSNMR approach (vide infra). To afford quantitative information on concentration and distribution of various acid sites in the parent and dealuminated HY and HUSY samples, spectra obtained in the absence of CP (Fig. 1a) were used for deconvolution analysis. Accordingly, the relative peak areas (i.e., concentrations) of different resonance species in various samples were obtained, as depicted in Table S2. It is indicative that the acidic amounts



of EFAL-Al3+ species (δ31P ∼ −33 ppm) readily increased with the dealumination treatment. In addition, supplementary results were also obtained by conventional acidity characterization methods such as temperature programed desorption of ammonia (NH3-TPD; Fig. S2b) and infrared spectroscopy of adsorbed pyridine (Py-IR; Fig. S3), as summarized in Table S3 of the Supplementary Information. In brief, although NH3-TPD and Py-IR provide consistent trends in variations of total acidity and distribution of acid sites for samples before and after the dealumination treatment, they gave rise to incomplete information and inferior results compared to those obtained from the

31

P-TMP SSNMR approach. For the latter, detailed acid

features such as type, distribution, concentration, and strength of acid sites may be simultaneously obtained.[20−23] In addition, the exact amounts of various acid sites may readily be attained through deconvolution of NMR spectra in conjunction with reliable elemental analysis data.


Page 8 of 34

Page 9 of 34 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


Fig. 1. 31P MAS NMR spectra of TMP adsorbed on the parent HY, parent HUSY zeolites and their respective dealuminated counterparts, namely HY-d450 and HUSY-d450, obtained (a) without and (b) with cross-polarization (CP; contact time of 4.0 ms). The dashed lines in (a) represent Gaussian deconvolution results.

2.3. Correlations among various acid sites The assignments of the

31

P signals responsible for TMP adsorbed on Brønsted

and Lewis acid sites were further verified by using the two-dimensional 1H-31P heteronuclear correlation (2D HETCOR) NMR technique.[17] The HETCOR spectrum recorded with a short contact time (τc = 0.1 ms) for the parent and dealuminated HY and HUSY samples are depicted in Fig. S4 (SI), whereas those recorded with prolonged contact time (τc = 4.0 ms) are shown in Fig. 2. At τc = 0.1 ms, the parent HY and dealuminated HY-d450 samples both showed only two correlation peaks at (−5, 7.3)/(−5, 6.0) and (−5, 8.6)/(−5, 7.1) ppm associated with TMP adsorbed on bridge hydroxyl (Si−OH−Al) protons (i.e., TMP interact with Brønsted acid sites to form TMPH+ complexes), which correspond to

31

P resonance (F2 dimension) with

δ31P of −5 ppm and 1H resonance (F1 dimension) with δ1H of 6.0 ∼ 8.6 ppm, respectively. It is anticipated that, upon increasing contact time, progressive increase in correlations peaks were observed, as expected. For the parent HY recorded with τc = 4.0 ms, only an additional correlation peak at (−5, 1.4) ppm was observed, indicating the correlation of TMPH+ with 1H resonance of methyl groups (δ1H = 1.4 ppm). On the other hand, the dealuminated HY-d450 revealed more complex correlation peaks with increasing contact time (Fig. S5; SI). In addition to the correlation peaks at (−5, 6.0), (−5, 7.1), and (−5, 1.7) ppm (Fig. 3a), additional peaks



correspond to TMP adsorbed on various active sites may readily be identified when the contact time was prolonged to 4.0 ms (Fig. 2b). Correlation peaks at (−33, 1.7) and (−47 ∼ −53, 1.4) ppm may be attributed to TMP adsorbed on Lewis acid sites associated with the methyl protons (Fig. 3b). While the correlation peaks at (−56 ∼ −60, 1.4) ppm were ascribed to TMP adsorbed on weakly acidic Si−OH groups, which was further confirmed by the correlation peak at (−56, 6.9) ppm due to the hydrogen-bonding interactions (Fig. 3c). As such, three possible interaction schemes may be envisaged for the adsorption of TMP on dealuminated zeolites, namely chemisorbed on Brønsted acid sites (i.e., formation of TMPH+ complexes), chemisorbed on Lewis acid site, and physisorbed on hydroxyl groups (i.e., hydrogen bonding interactions), as illustrated in Fig. 3. It is noteworthy that owing to their difference in acidic strength, a slight shift in δ1H (ca. 0.3 ppm) was observed between signals arising from methyl protons of TMP adsorbed on Brønsted and Lewis acid sites. Furthermore, comparing the HETCOR spectrum recorded with various contact times (0.1 ∼ 4.0 ms), no spatial correlation between Brønsted acid sites (δ1H = 6.0 ∼ 7.1 ppm) and strong Lewis acid sites (δ31P = −33 ppm) was found in dealuminated HY-d450 sample.


Page 10 of 34

Page 11 of 34 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


Fig. 2. 2D 1H-31P HETCOR NMR spectra of TMP adsorbed on (a) parent HY, (b) dealuminated HY-d450, (c) parent HUSY, and (d) dealuminated HUSY-d450 zeolites recorded with a contact time of 4.0 ms.

Fig. 3. Possible schemes for adsorption of TMP on dealuminated zeolites: (a) chemisorbed on Brønsted acid sites, (b) chemisorbed on Lewis acid centers, and (c) physisorbed on hydroxyls (hydrogen bonding interactions; M = Al or Si). The corresponding 1H and 31P NMR chemical shifts (in ppm) obtained from the NMR experiments are also shown.

Unlike the parent HY (cf. Fig. S4a; SI), the 1H-31P HETCOR NMR spectrum of the parent HUSY exhibited a broad distribution of correlation peaks at (−33, 7.3) ppm



in addition to the correlation at (−5, 7.3) ppm even at short contact time of 0.1 ms (Fig. S4c; SI). The former correlation peak may be attributed to interactions between tri-coordinated EFAL-Al3+ species with neighboring bridging hydroxyl (Si−OH−Al) as well as silanol (Si−OH) groups, as illustrated in Fig. 3. At prolonged contact time (4 ms), additional correlations at (−5, 1.7), (−33, 1.6), and (−47, 1.4) ppm were observed, which may be attributed to TMP adsorbed on Brønsted and Lewis acid sites associated with methyl protons. Moreover, more discrete correlation peaks may be inferred, the peak at (−33, 7.3) ppm remain visible even at prolonged contact time (4 ms). Similar conclusions may also be drawn for the dealuminated HUSY-d450 (Figs. 2d and S4d; SI), except, in this case, much better signals were observed compared to that observed for the parent HUSY. Moreover, compared to that of HY-d450, the correlation peak at (−56, 6.9) ppm (which was ascribed due to interactions between TMP with Si-OH group; see Fig. 3c) was absent in HUSY-d450 whilst a correlation peak at (−33, 7.3) ppm was emerged.

2.4. Proximity of the tri-coordinated EFAL-Al3+ The interactions among various adsorbed TMP over the dealuminated HY-d450 zeolite were further studied by means of 2D

31

P-31P correlation based on

proton-driven spin diffusion (PDSD) MAS NMR experiment.[28] Compared to the 2D 31

P-31P PDSD MAS NMR spectrum observed for TMP adsorbed on the parent HY

zeolite (Fig. 4a), which showed only one autocorrelation (i.e., peaks along the diagonal axis; F2 = F1) peak at (−5, −5) ppm, up to five autocorrelation peaks at (−5,


Page 12 of 34

Page 13 of 34 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


−5), (−33, −33), (−47, −47), (−53, −53), and (−56, −56) ppm may be identified for the dealuminated HY-d450 (Fig. 4b). In addition, three off-diagonal correlation (F2, F1) peak pairs at (−5, −47), (−5, −53), (−5, −56) and (−47, −5), (−53, −5), (−56, −5) ppm were also observed. The presences of these crossed peaks clearly indicate that the 31P resonances associated with Brønsted acid sites (δ31P at −5 ppm) are in close proximity to those associated with Lewis acid sites with different acidic strengths (with δ31P at −47, −53 ppm), and weakly acidic Si−OH groups (δ31P = −56 ppm). In other word, the spatial proximity and the synergy between the Brønsted and various active sites (including Lewis acids with different acidic strengths) in the dealuminated HY zeolite may readily be inferred. Nonetheless, it is noteworthy that the spatial correlation between

31

P resonances of Brønsted (at −5 ppm) and the strong Lewis acid sites (at

−33 ppm) was absent in the 31P-31P PDSD MAS NMR spectra. This indicates that the two types of acid sites are distance away from one another in dealuminated HY-d450 zeolite; in close agreement with the result obtained from the 1H-31P HETCOR NMR spectra (Fig. 2; vide supra). Further measurements by 2D

31

P double-quantum (DQ)

MAS NMR technique[17,29] have also been made. In brief, the results shown in Fig. 5 revealed that only interactions among Brønsted acid sites exist in the parent HY zeolite. On the other hand, Brønsted (δ31P = −5 ppm) and weak Lewis acid sites (δ31P = −47 ppm) are in close proximity and are spatially correlated in dealuminated HY-d450 zeolite. However, no interaction between Brønsted acid sites and tri-coordinated EFAL-Al3+ (i.e., ultra-strong Lewis acid sites at δ31P = −33 ppm) was found. In other word, the EFAL-Al3+ species appear to be isolated in dealuminated



HY-d450 zeolite.

Fig. 4. 2D 31P-31P PDSD MAS NMR spectra of TMP adsorbed on the (a) parent HY, (b) HY-d450, (c) parent HUSY, and (d) HUSY-d450 zeolites recorded with a mixing time of 100 ms.


Page 14 of 34

Page 15 of 34 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


Fig. 5. 31P-31P DQ MAS NMR spectra of TMP adsorbed on (a) the parent HY, and (b) dealuminated HY-d450 zeolite samples.

The proximity of EFAL-Al3+ species in the parent HUSY and dealuminated HUSY-d450 zeolites was also studied by 2D 31P-31P PDSD MAS NMR, as shown in Figs. 4c and 4d, respectively. In addition to the autocorrelation peaks at the diagonal axis, off-diagonal correlation peaks at (−5, −33) and (−47, −33) ppm due to couplings of EFAL-Al3+ with Brønsted and Lewis acid sites were both observed for HUSY-d450 (Fig. 4d). Since only correlation peak at (−5, −30) ppm was found for the parent HUSY (Fig. 4c), it is indicative that the dealumination treatment of HUSY resulted in not only enhanced formation of EFAL-Al3+ species but also their couplings with Lewis acid sites in dealuminated HUSY-d450 zeolite.



Based on the NMR results, it is conclusive that, unlike HUSY, tri-coordinated EFAL-Al3+ species present in dealuminated HY-d450 were found to be mostly isolated (Figs. 2b and 4b). On the other hand, EFAL-Al3+ species was also found in the parent HUSY sample (Fig. 4c), in which aside from isolated EFAL-Al3+, relatively small portion of these tri-coordinated Al3+ were weakly coupled with Brønsted acid sites in close proximity. Upon further dealumination treatment additional spatial correlations between EFAL-Al3+ and Lewis acidic sites with modest strength were also observed for HUSY-d450 sample (Fig. 4d).

2.5. Adsorbed structures of TMP on the EFAl species Theoretical calculation methods, particularly those by density functional theory (DFT), have been shown to be an effective and indispensable tool in verifying and/or predicting the interactions and relevant NMR parameters associated with acid properties of zeolite catalysts.[11,13,14,20−23,30,31] As such, it is anticipated that theoretical DFT calculations should render new insights in the role of EFAL species on catalytic activity and reaction pathway involved during heterogeneous catalysis. As mentioned above, cationic compounds such as Al3+, AlOH2+, AlO+, and Al(OH)2+ as well as some neutral compounds such as Al(OH)3 and AlOOH have been identified as possible EFAL species in dealuminated zeolites.[11−16] Moreover, these EFAL species may be accountable for the Lewis acidity of the dealuminated zeolite catalysts.[11,20−23] In addition, tri-coordinate framework aluminum (TFAL) species, which tend to form through severe sample calcination treatment at elevated temperatures, were also


Page 16 of 34

Page 17 of 34 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


proposed as possible origins of Lewis acidity. By adopting the EFAL and FAL theoretical models proposed by Mota and co-workers[13,14] and Sokol et al.,[32] respectively, the adsorption configurations as well as relevant NMR parameters associated with the adsorption system of TMP on dealuminated zeolite were investigated by means of DFT calculations based on 8T as well as 10T clustered zeolite model, as shown in Figs. S6 and 6, respectively. In view of theoretical calculations, it is anticipated that all EFAL species should be coordinated with the oxygen atoms nearby to the FAL in the dealuminated zeolite. Whereas, in the case of TFAL species, the central tri-coordinated Al atom was assumed to bond with the framework oxygen (O) atoms or hydroxyl (OH) groups. Moreover, to keep the cluster model neutral, adequate framework Al atoms were incorporated during the DFT calculations to compensate the charge of EFAL species. The optimized geometries of TMP adsorbed on possible EFAL and TFAL species of the dealuminated Y-type zeolite models, representing various adsorption complexes associated with different Lewis acid sites with varied P−Al distances, are illustrated in Fig. 6. It is found that, in the case of TMP-EFAL adsorption complexes, the adsorption of TMP probe molecule onto Lewis acid sites of oxoaluminum cationic and neutral EFAL species, resulting in variations of the P−Al distance of the adsorption complexes in the range of 2.370−2.519 Å. It is noteworthy that, in these TMP-EFAL adsorption complexes, the EFALs were coordinated with three O atoms and one P atom (on TMP) to form four-coordinate centers (Figs. 6a−f). By comparison, in the case of TMP-TFAL adsorption complexes (Figs. 6g−i), variations



of the P−Al distances in the range of 2.431−2.456 Å were observed. Among various adsorption complexes, TMP adsorbed on isolated tri-coordinated EFAL (EFAL-Al3+) species resulted in a shortest P−Al distance of 2.370 Å (Fig. 6a), indicating the case of strongest interaction. The above findings are in excellent agreement with the results obtained from LUMO energy calculations (Table 1), in which a much lower energy (−6.65 eV) was derived for the EFAl-Al3+ Lewis acid site compared to the other TFAL and EFAL clusters, which spanned over the range of −0.68 to −3.40 eV. Thus, it is indicative that the tri-coordinated EFAL-Al3+ species presented in dealuminated HY zeolite indeed showed the strongest Lewis acidity surpassing the other TFAL and EFAL species. A similar trend may be inferred in terms of the fluoride ion affinity (FIA),[33] the strength of the Lewis acid sites associated with the tri-coordinated EFAL-Al3+ species was found to have the highest FIA value (192.8 kcal/mol) compared to its TFAL and EFAL counterparts (77.1−155.6 kcal/mol; see Table 1).


Page 18 of 34

Page 19 of 34 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


Fig. 6. Optimized geometries of TMP adsorbed on various 10T Lewis acid sites of dealuminated Faujasite zeolite, including extra-framework aluminum (EFAL) species such as (a) Al3+, (b) AlO+, (c) AlOH2+, (d) Al(OH)2+, (e) Al(OH)3, and (f) AlOOH, and those associated with tri-coordinated framework aluminum (TFAL) species such as (g) AlSi3, (h) AlSi2, and (i) AlSi1, The calculated P−Al interatomic distances (in Å) are indicated.

Table 1. List of LUMO and fluoride ion affinity (FIA) energies of various 10T Lewis acid sites and corresponding parameters of the TMP adsorption complexes. The 1H and 31P chemical shifts of TMP adsorbed on Brønsted acid sites and weakly acidic Si−OH groups are also given. Energy Type Al3+ AlO+ AlOH2+ Al(OH)2+ Al(OH)3 AlOOH

Parameters of adsorption complex δ31P (ppm)a

δ1H (ppm)d

LUMO (eV)

FIA (kcal/mol)

rP−Al (Å)

JP−Al (Hz)

HFb

HFc

CH3

H+

−6.65 −1.84 −3.40 −0.68 −1.06 −1.52

192.8 111.3 155.6 91.9 80.1 77.1

2.370 2.426 2.402 2.519 2.471 2.498

696.4 267.7 496.2 288.9 274.4 155.3

−35.3 −52.8 −50.3 −45.2 −49.2 −54.2

−34.3 −52.6 −49.6 −45.2 −49.2 −54.2

2.7 1.8 2.0 1.6 1.6 1.7

-------------

AlSi3 AlSi2 AlSi1

120.5 2.431 357.6 −50.7 −50.8 1.7 −1.74 101.1 2.446 342.8 −49.4 −49.5 1.7 −1.10 95.1 2.456 318.1 −49.5 −49.5 1.6 −1.00 --------2.2 Si−OH−Al −5 --------1.4 Si−OH −60 a Calculations performed at HF/6-311++G(2d,2p)//B3LYP/DZVP2 level of theory. b Based on deprotonated 10T clustered HY zeolite model (See Fig 6). c Based on deprotonated 8T clustered HY zeolite model (See Fig S6). d Calculated 1H chemical shifts of the methyl (CH3) group and Brønsted acidic proton (H+).

------8.3 7.4

2.6. Confirmation of tri-coordinated EFAL species On the basis of results obtained from various optimized TMP-Lewis acid site adsorption structures (Fig. 6), relevant NMR parameters, namely the

31

P-27Al

J-coupling constant (JP−Al) and δ31P were also obtained, as depicted in Table 1. Comparing to TMP adsorbed on tri-coordinated EFAL-Al3+ Lewis acid sites, which gave rise to the highest predicted JP−Al and δ31P values of 696.4 Hz and −35.3 ppm, respectively, the corresponding values observed for TMP adsorbed on TFAL and other



EFAL species showed substantially lower JP−Al (155.3−496.2 Hz) and δ31P (−54.2 to −45.2 ppm) values (Table 1). Moreover, the range of calculated δ31P obtained from various TMP-Lewis acid site adsorption complexes in dealuminated HY zeolite were also in good agreement with the experimental values (−33, −47, −53, and −57 ppm) observed for TMP adsorbed on HY-d450 sample. As discussed above, being has the lowest LUMO energy, highest FIA value, shortest P−Al bond length, and largest JP−Al values in the TMP adsorption structure (Table 1), the tri-coordinated EFAl-Al3+ species indeed revealed Lewis acid sties with the strongest acidic strength compared to other TFAL and EFAL species. As a result, Lewis acid sites associated with tri-coordinated EFAl-Al3+ species also resulted in the highest predicted δ31P value (−35.3 ppm), which is in excellent agreement with the experimental value of −33 ppm. This observation is also in line with a previous study of dealuminated zeolite by Van Bokhoven et al. using the in situ X-ray absorption near edge spectroscopy (XANES).[34] Recently, such tri-coordinated aluminum defect sites were also observed for dehydroxylated alumina based on experimental infrared (IR) spectroscopy using nitrogen (N2) as the probe.[35] Moreover, it is noteworthy that the EFAL-Al3+ species also exhibits excellent stability, as confirmed by washing treatments with water (Fig. S7; SI) and HNO3 solutions (Fig. S8; SI). It is indicative that washing of dealuminated zeolites by HNO3 may potentially lead to influence the properties of the catalysts, especially when its acid concentration exceeded 0.02 M (cf. Fig. S8). Nonetheless, the signal intensity of the resonance peak responsible for the EFAL-Al3+ species (δ31P ∼ −33 ppm) in dealuminated HUSY-d450 sample remained


Page 20 of 34

Page 21 of 34 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


practically unchanged when subjected to washing with diluted HNO3 solution (≤ 0.02 M). By comparison, the structural characteristic of the HUSY-d450 remained unchanged even when subjected to washing treatment by boiling water (373 K) for extended period of time (12 h), as shown in Fig. S7. This together with the strong Lewis acidic strength observed making tri-coordinate EFAL-Al3+ species present in dealuminated zeolites the highly active sites for low-temperature activation of alkanes.[36,37] To further verify the interaction between the TMP probe molecule with Lewis acid sites with varied acidic strengths in dealuminated zeolite, we also employed the 2D electron localization function (ELF) analysis [38,39] and the local electron energy density functions H(r)[40]. The ELF value, which is within the range of 0 to 1, represents the localization degree of the electron. A null value of ELF would represent a complete delocalization of electron, while ELF = 1 should be the case of a perfectly localized electron. As such, a larger observed ELF value normally corresponds to more covalent characteristics of a chemical bond. The 2D ELF images observed for molecular fragments of TMP adsorbed at tri-coordinated EFAL-Al3+, EFAL-AlOH2+, and TFAL-AlSi3 Lewis acid sites are shown in Fig. 7. It is clear that a higher degree of electron localization in the corresponding P−Al bond of the TMP/EFAL-Al3+ adsorption

complex

compared

to

that

of

its

TMP/EFAL-AlOH2+,

and

TMP/TFAL-AlSi3 counterparts may be inferred. In other word, upon adsorption of TMP, the electrons associated with the TMP probe molecule tend to be much more closer to the Al atom of the EFAL-Al3+ adsorbent compared to that of EFAL-AlOH2+



and TFAL-AlSi3, revealing a much stronger interactions of the former guest-host system. Again, these results are in excellent agreement with the aforementioned discussions derived from the

31

P NMR results. Similar conclusion may be drawn as

well by examining their local electron energy density functions, as confirmed in Fig. 7 that an evident covalent bond property is present in the TMP/EFAL-Al3+ adsorption complex compared to that of other counterparts. Moreover, based on the experimental and theoretical results, a larger spatial distance between Brønsted acid site and tri-coordinated EFAL-Al3+ Lewis acid site was observed compared to interactions with other Lewis acidic EFAL and TFAL species in dealuminated HY-d450 zeolites. In view of the abundant positive charge possessed by the tri-coordinated EFAL-Al3+ active sites, the notion proposed above is reasonable since the tri-coordinated EFAL-Al3+ Lewis acid sites were formed at the expenses of adjacent Brønsted acid sites, as illustrated in the proposed formation mechanism in Scheme 1. Clearly, the presence of more abundant acidic protons in the zeolite frameworks is favorable for the formation of EFAL-Al3+ species during dealumination treatments as indicated that such EFAL-Al3+ species cannot be formed in the H-mordenite and H-ZSM-5 zeolites with relative high Si/Al rations (Fig. S9; SI).


Page 22 of 34

Page 23 of 34 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


Fig. 7. 2D electron localization function (ELF) contours (a, b, and c) and local electron energy density functions (d, e, and f) of TMP adsorbed on (a, d) EFAL-Al3+, (b, e) EFAL-AlOH2+, and (c, f) TFAL-AlSi3 Lewis acid sites.

Scheme 1. Illustration of the proposed location of tri-coordinated EFAL-Al3+ Lewis acid site in dealuminated zeolite.

2.7. Catalytic performance of tri-coordinated EFAL species



Page 24 of 34

As a typical biomass-derived platform molecule, 5-hydroxymethylfurfural (HMF) has attracted much attention due to its wide use for producing valuable chemicals and polymer precursors. Typically, producing HMF from glucose generally experiences tandem steps, while the first one is the isomerization of glucose to fructose over Lewis acidic catalysts.[41-44] Hence, the catalytic performance of tri-coordinated EFAL species with Lewis acidity inside HY and HUSY zeolites before and after the dealumination treatment were examined by this important reaction at 363 K and 1 MPa N2. As shown in Table 2, two major products, mannose and fructose, are explicitly detected in methanol respectively from the epimerization and isomerization of glucose, indicating the presence of Lewis acid sites on these zeolite catalysts.[41,44] Therein, parent HY exhibits negligible glucose conversion (0.5%) during the reaction condition, which should be due to the absence of Lewis centers on this catalyst. After the dealumination treatment, a slight improvement of the glucose conversion can be observed for HY-d450, which is in good agreement with the appearance of Lewis acidic signal as illustrated in Fig. 1b (−33 ~ −57 ppm), especially for the weak resonance at −33 ppm. Interestingly, it is noteworthy that the glucose conversion remarkably increases to 7.6%, and even 20.9% for parent HUSY, and dealuminated HUSY-d450 samples, respectively, being consistent with the enhancement of

31

P

signal intensity at −33 ppm in Fig. 1b. On the basis of the previous experimental results that the isomerization of glucose to fructose is a typical Lewis acid-catalyzed reaction during which Brønsted acidity was found to show negligible effect during the catalytic process.[41-44] Therefore, the excellent catalytic performance of dealuminated


Page 25 of 34 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


HUSY-d450 was attributed to the tri-coordinated EFAL species with strong Lewis acidity.

Table 2. Conversion and product selectivity during isomerization of glucose over HY and HUSY zeolites before and after dealumination treatment (at 450 oC).* Zeolite

Conversion (%)

Selectivity (%) Mannose

Fructose

Parent HY

0.5

57.6

40.3

HY-d450

0.9

58.6

41.3

Parent HUSY

7.6

36.3

58.0

HUSY-d450

20.9

49.7

43.0

*

Reaction conditions: catalyst 50 mg, glucose 0.2 g, methanol 20 mL, temperature 363 K, pressure 1 MPa N2, reaction time 2h.

3. Conclusions In summary, the origin, structure, and possible locations of tri-coordinated EFAL-Al3+ Lewis acid site in dealuminated HY zeolite have been comprehensively investigated by means of advanced solid-state multinuclear NMR techniques in conjunction with theoretical DFT calculations. Acidity characterization using the 31

P-TMP NMR probe molecule approach revealed that the EFAL-Al3+ species

possesses Lewis acidity with ultra-strong acidic strength surpassing that of other EFAL and TFAL species presented in dealuminated zeolites. In contrast to dealuminated HY, the tri-coordinated EFAL-Al3+ species observed in both parent and dealuminated HUSY zeolites are relatively less isolated; Furthermore, it is experimentally confirmed that the tri-coordinated EFAL species exhibits excellent catalytic performance in the isomerization reaction of glucose to fructose. Further



work has been undertaken to explore detailed reaction mechanism for activation of alkane over the EFAL-Al3+ acid sites in dealuminated zeolites.

4.

Experimental and Theory Section

4.1. Preparation and Characterization of Dealuminated HY NaY (Si/Al = 2.8) and HUSY (Si/Al ≈ 3.5) zeolites were obtained from Nankai University, China. The NaY zeolite was exchanged in 1 mol/L aqueous solution of NH4NO3 at 353 K for 10 h, and then washed with distilled water. This procedure was repeated four time to obtain the NH4Y zeolite after drying in air at 353 K overnight. Subsequently, the obtained NH4Y sample was carefully deaminated and dehydrated on a vacuum manifold (pressure < 10-3 Pa). To avoid dealumination of the zeolite framework, the H-form Y zeolite was obtained from the parent NH4Y by gradual heating at a temperature ramp rate of 1 K/min till reaching the target temperature (673 K), and maintained at the final temperature for ca. 10 h before it was cooled to room temperature. For the preparation of dealuminated HY and HUSY zeolites, the as-prepared NH4Y (or HUSY) zeolite sample was placed in a quartz crucible in a tube furnace, followed by a calcination treatment in air by gradual heating from room temperature to 723 K (450 oC), then, maintained at the final temperature for ca. 3.5 h. The dealuminated samples so obtained are hereafter denoted as HY-d450 and HUSY-d450. The structural properties of the parent HY and HUSY, and their dealuminated counterparts were characterized by using X-ray diffraction (XRD; Bruker D8 Advance, Cu Kα radiation) and solid-state 27Al and 29Si nuclear magnetic resonance (NMR) spectroscopy. Temperature programmed desorption of ammonia (NH3-TPD) was performed on a CHEMBET 3000 chemical absorber (Quantachrome, USA). About 0.1 g of the catalyst sample was pretreated in helium at 773 K for 1 h, cooled to 393 K, then exposed to an ammonia–helium mixture (8% NH3-92% He) for 30 min. The physically adsorbed NH3 was removed by helium flow at 393 K for 1 h. The TPD plot was obtained at a heating rate of 10 K/min from 393 K to 923 K. The desorbed ammonia was detected by gas chromatography with a thermal conductivity detector. Infrared spectroscopy with pyridine (Py) adsorption was carried out using an EQUINOX55 Fourier transform infrared spectrometer (FT-IR) (Bruker Corp.) with a resolution of 4 cm-1. The samples were ground to fine powders and pressed into a thin self-supporting wafer with a weight of 10 mg and a diameter of 10 mm. Before the base adsorption, the samples were evacuated at 673 K for 1 h. Subsequently, the samples were exposed to pyridine for 10 min for adsorption, then the samples were desorbed under vacuum at 573 K, and the corresponding spectra were recorded. The P elemental analysis of the zeolites with TMP adsorption was conducted by using ICP-AES (Leeman Prodigy 7), each sample was determined for 3 times to obtain an average value. Furthermore, to intentionally remove the possible debris retained in the zeolite pores, dealuminated HUSY-d450 samples were washed with either water or nitric acid (HNO3) solutions. For the latter, the concentration of the acid was varied from 0.01 M to 0.25 M. Typically, the dealuminated zeolite sample was first washed with HNO3, then reflux at 333 K


Page 26 of 34

Page 27 of 34 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


for 2 h, followed by washing with distilled water before it was dried in air at 353 K overnight. In the case of samples washed with water, the stability tests performed under different conditions were accomplished, e.g., at 333 K for 4 h and 12h, and at 373 K for 12 h. To assess the influence of zeolite framework structure on formation of EFAL species during a dealumination treatment, HZSM-5 (Si/Al = 11.5 and 40) and HMOR (Si/Al = 10) zeolites were also examined under experimental conditions similar to that of HY and HUSY samples. 4.2. NMR Experiments All the NMR experiments were performed on a Bruker AVANCE-III 500 MHz spectrometer operating at a Larmor frequency of 500.57, 202.63, 130.44, and 99.44 MHz for the 1H, 31P, 27Al, and 29Si nucleus, respectively. Solid-state 1H, 31P, and 27Al NMR spectra were recorded using a 4 mm magic-angle-spinning (MAS) probe operating at a spinning rate of 12 kHz. A single-pulse sequence with a π/2 pulse length of 4.0 µs and a recycle delay of 5 s was used for the 1H NMR experiments. All one-dimensional (1D) 1H-31P cross-polarization (CP)/MAS measurements were performed with a contact time of 4 ms and a recycle delay of 3 s. Two-dimensional (2D) 1H-31P heteronuclear correlation (HETCOR)[17] experiments were carried out with a varying CP contact times (0.1 ~ 4 ms), whereas 2D 31P-31P proton driven spin diffusion (PDSD)[28] correlation experiments were performed with a mixing time of 100 ms. For the 31P double quantum (DQ) MAS NMR[17] experiments, DQ coherences were excited and reconverted with a POST-C7[45] pulse sequence following the common scheme for 2D multiple-quantum (MQ) NMR spectroscopy. The pulse sequences for all the 2D NMR experiments were illustrated in Fig. S10 (SI). All 1D 27Al MAS NMR spectra were recorded using a single-pulse sequence with a pulse length of 0.26 µs (π/12) and a recycle delay of 1 s, whereas 1D 29Si MAS NMR experiments were conducted with high power proton (1H) decoupling using a π/2 pulse of 3.9 us and a recycle delay of 80 s on a 7 mm triple-resonance MAS probe with a spinning rate of 4 kHz. The chemical shift of the 1H, 31P, 27Al, and 29Si nucleus was externally referenced to adamantane, (NH4)2HPO4 (1 ppm), 1 M aqueous Al(NO3)3, and kaolinite (−91.5 ppm), respectively. For acidity characterization by means of the 31P-trimethylphosphine (TMP) NMR approach, sample was placed in a glass tube and connected to a vacuum manifold for a dehydration treatment prior to the adsorption of TMP probe molecule. Typically, the dehydration treatment was carried by gradual heating at a ramp rate of 1 K/min till reaching the final temperature (673 K), then maintained at the temperature for at least 10 h under vacuum, then cooled to ambient temperature. Subsequently, a known amount of volatile TMP molecule was transfer onto the sample frozen over a liquid N2 bath. Then, the sample was evacuated at room temperature for 1 h to remove physisorbed TMP molecules. To verify the presence of Lewis acidity, the TMP adsorbed dealuminated zeolite samples were subjected to partial hydration treatments, the sample tube was then flame-sealed. Prior to the 31P NMR experiments, the sample was transferred into a ZrO2 rotor, then sealed with a Kel-F MAS cap in a nitrogen glovebox. 4.3 Catalytic Experiments Glucose (Alfa Aesar, anhydrous, 99%) isomerization reactions were carried out in a Teflon-lined stainless steel autoclave (50 mL) under agitation at a speed of 600 rpm. Typically, 50 mg of zeolite catalysts and 0.20 g of glucose were introduced to 20 mL absolute



methanol (Tongguang, GR) in the autoclave. Afterwards, the autoclave was fully purged with N2 (Beijing Haipu, 99.999%), pressurized with N2 to 1 MPa, and then heated to 363 K. The reactants and liquid products were analyzed by High Performance Liquid Chromatography (Shimadzu LC20-AD) using Ca2+ exchanged LC Column (Phenomenex Rezex RCM-Monosaccharide Ca2+) with RID detector. Selectivity of products was calculated on the carbon basis. 4.4. Theoretical calculations Cationic extra-framework aluminum (EFAL) species such as Al3+, AlOH2+, AlO+, and Al(OH)2+ as well as neutral species such as Al(OH)3 and AlOOH were selected as possible Lewis sites. Their coordination with the oxygen atoms nearby to the framework aluminum (FAL) species in the deprotonated 10T clustered zeolite model were assigned as EFAL-Al3+, EFAL-AlOH2+, EFAL-AlO+, EFAL-Al(OH)2+, EFAL-Al(OH)3, and EFAL-AlOOH, respectively. In the calculations, both the 8T and 10T cluster models have been used. It is observed that the structures and δ31P are in good agreement with each other, which confirmed the reliability of the theoretical models in this work (see Figs. 6 and S6; SI). In addition, three types of tri-coordinated framework aluminum (TFAL) species in which the central Al atom was connected with framework oxygen (O) atoms or hydroxyl (OH) groups were also included as possible Lewis acid sites and are designated as TFAL-AlSi3, TFAL-AlSi2, and TFAL-AlSi1, respectively. It is noteworthy that similar Lewis acid models were adopted in previous studies.[11,13,14,32,46] To keep the cluster model electrically neutral, as many FAL atoms as necessary were adopted to compensate the positive charges of various EFAL species. On the other hand, for the Lewis acid model associated with neutral EFAL species, namely EFAL-Al(OH)3 and EFAL-AlOOH sites, an additional proton was included and linked to the oxygen atom near the FAL. During the calculation, all terminal Si atoms were saturated with a H atoms at a Si−H bond length of 1.47 Å. Nevertheless, for all tetrahedral Al atoms of the framework, we used hydroxyl groups to complete the valence for a better description of the electron density near the aluminum atom. Moreover, only the isolated acid sites rather than its synergetic interactions with the neighboring Brønsted acid sites were taken into account in the calculations.[11,46] Accordingly, the correlation between the 31P chemical shift (δ31P) of the adsorbed TMP with Lewis acidic strength (i.e., LUMO energy and/or fluoride ion affinity) may be established. During the structural optimization of bare acid sites and TMP adsorption complexes, all atoms except for terminal −SiH3 and −AlOH groups were allowed to relax. All geometry optimizations were employed by means of the B3LYP hybrid density functional[47,48] combined with DZVP2[49] basis sets, which has been demonstrated to be a reliable method for predicting the structures of probe molecules adsorbed on zeolites.[30,50] On the basis of the optimized structures, the 31P NMR chemical shifts (δ31P) and 31P-27Al J-coupling constant (JAl-P) were calculated using the gauge independent atomic orbital (GIAO)[51] method at the HF/6-311++G(2d,2p) level.[20,52] Here, the experimental δ31P value of the physisorbed TMP was adopted as an internal standard for converting the calculated isotropic absolute shielding constant to the calculated chemical shift.[20] All geometry optimizations, NMR parameters, and fluoride ion affinity calculations were performed with the Gaussian 09 software package.[53] The 2D electron localization function (ELF) contours and local electron energy


Page 28 of 34

Page 29 of 34 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


density functions were calculated with the Multiwfn software.[54]

Additional information The Supporting Information is available free of charge on the ACS Publications website at DOI: xxx/xxx.xxx. Additional data on XRD patterns, and 27Al/29Si MAS NMR spectra of HY and HUSY zeolites before and after the dealumination treatment; 31 P CP/MAS NMR spectra of TMP adsorbed on dealuminated zeolites with varied H2O loadings, and proposed mechanism for the formation of various EFAL species inside HY zeolite. Acknowledgements

This work was supported by the National Natural Science Foundation of China (No. 21522310, 21473244 and 91645112), and Natural Science Foundation of Hubei Province of China (2018CFA009), Key Research Program of Frontier Sciences, CAS (No. QYZDB-SSW-SLH026), Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund (the second phase) under Grant No.U1501501, and Sinopec Corp. (417012-4).

References [1] Corma, A. Inorganic solid acids and their use in acid-catalyzed hydrocarbon reactions. Chem. Rev. 1995, 95, 559−614. [2] Svelle, S.; Kolboe, S.; Swang, O. Theoretical investigation of the dimerization of linear alkenes catalyzed by acidic zeolites. J. Phys. Chem. B 2004, 108, 2953−2962. [3] Macht, J.; Carr, R. T.; Iglesia, E. Consequences of acid strength for isomerization and elimination catalysis on solid acids. J. Am. Chem. Soc. 2009, 131, 6554−6565. [4] Arata, K.; Matsuhashi, H.; Hino, M.; Nakamura, H. Synthesis of solid superacids and their activities for reactions of alkanes. Catal. Today 2003, 81, 17−30. [5] Marcus, D. M.; Hayman, M. J.; Blau, Y. M.; Guenther, D. R.; Ehresmann, J. O.; Kletnieks, P. W.; Haw, J. F. Mechanistically significant details of the H/D exchange reactions of propene over acidic zeolite catalysts. Angew. Chem., Int. Ed. 2006, 45, 1933−1935. [6] Clark, L. A.; Sierka, M.; Sauer, J. Computational elucidation of the transition state shape



Page 30 of 34

selectivity phenomenon. J. Am. Chem. Soc. 2004, 126, 936−947. [7] DeCanio, S. J.; Sohn, J. R.; Fritz, P. O.; Lunsford, J. H. Acid catalysis by dealuminated zeolite-Y: I. Methanol dehydration and cumene dealkylation. J. Catal. 1986, 101, 132−141. [8] Sohn, J. R.; DeCanio, S. J.; Fritz, P. O.; Lunsford, J. H. Acid catalysis by dealuminated zeolite Y. 2. The roles of aluminum. J. Phys. Chem.1986, 90, 4847−4851. [9] Beyerlein, R. A.; McVicker, G. B.; Yacullo, L. N.; Ziemiak, J. Influence of framework and nonframework aluminum on the acidity of high-silica, proton-exchanged FAU-framework zeolites J. Phys. Chem. 1988, 92, 1967−1970. [10] Haw, J. F.; Xu, T.; Nicholas, J. B.; Gorgune, P. W. Solvent-assisted proton transfer in catalysis by zeolite solid acids. Nature 1997, 389, 832-835. [11] Li, S. H.; Zheng, A. M.; Su, Y. C.;. Zhang, H. L.; Chen, L.; Yang, J.; Ye, C. H.; Deng, F. Brønsted/Lewis acid synergy in dealuminated HY zeolite: A combined solid-state NMR and theoretical calculation study. J. Am. Chem. Soc. 2007, 129, 11161−11171. [12] Shannon, R. D.; Gardner, K. H.; Staley, R. H.; Bergeret, G.; Gallezot, P.; Auroux, A. The nature of the nonframework aluminum species formed during the dehydroxylation of H-Y. J. Phys. Chem. 1985, 89, 4778−4788. [13] Bhering, D. L.; Ramirez-Solis, A.; Mota, C. J. A. A density functional theory based approach to extraframework aluminum species in zeolites. J. Phys. Chem. B. 2003, 107, 4342−4347. [14] Mota, C. J. A.; Bhering, D. L.; Rosenbach, N., Jr. A DFT study of the acidity of ultrastable Y zeolite: Where is the Brønsted/Lewis acid synergism? Angew. Chem., Int. Ed. 2004, 43, 3050−3053. [15] Yu, Z.; Zheng, A.; Wang, Q.; Chen, L.; Xu, J.; Amoureux, J. P.; Deng, F. Insights into the dealumination of zeolite HY revealed by sensitivity-enhanced

27

Al DQ-MAS NMR

spectroscopy at high field. Angew. Chem., Int. Ed. 2010, 49, 8657−8661. [16] Yu, Z.; Li, S.; Wang, Q.; Zheng, A.; Xu, J.; Chen, L.; Deng, F. Brønsted/Lewis acid synergy in H-ZSM-5 and H-MOR zeolites studied by 1H and

27

Al DQ-MAS solid-state NMR

spectroscopy. J. Phys. Chem. C 2011, 115, 22320−22327. [17] Brown, S. P.; Spiess, H. W. Advanced solid-state NMR methods for the elucidation of structure and dynamics of molecular, macromolecular, and supramolecular systems. Chem. Rev. 2001, 101, 4125−4156. [18] Lunsford, J. L.; Rothwell, W. P.; Shen, W. Acid sites in zeolite Y: A solid-state NMR and infrared study using trimethylphosphine as a probe molecule. J. Am. Chem. Soc. 1985, 107, 1540−1547. [19] Haw, J. F.; Zhang, J.; Shimizu, K.; Venkatraman, T. N.; Luigi, D.-P.; Song, W.; Barich, D.


Page 31 of 34 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


H.; Nicholas, J. B. NMR and theoretical study of acidity probes on sulfated zirconia catalysts. J. Am. Chem. Soc. 2000, 122, 12561−12570. [20] Chu, Y.; Yu, Z.; Zheng, A.; Fang, H.; Zhang, H.; Huang, S. J.; Liu, S. B.; Deng, F. Acidic strengths of Brønsted and Lewis acid sites in solid acids scaled by 31P NMR chemical shifts of adsorbed trimethylphosphine. J. Phys. Chem. C 2011, 115, 7660−7667. [21] Zheng, A.; Huang, S. J.; Liu, S. B.; Deng, F. Acid properties of solid acid catalysts characterized by solid-state

31

P NMR of adsorbed phosphorous probe. Phys. Chem. Chem.

Phys. 2011, 13, 14889−14901. [22] Zheng, A.; Li, S.; Liu, S. B.; Deng, F. Acidic properties and structure-activity correlations of solid acid catalysts revealed by solid-state NMR spectroscopy. Acc. Chem. Res. 2016, 49, 655−663. [23] Zheng, A.; Liu, S. B.; Deng, F. 31P NMR chemical shifts of phosphorous probes as reliable and practical acidity scales for solid and liquid catalysts. Chem. Rev. 2017, 117, 12475−12531. [24] Peng, Y. K.; Ye, L.; Qu, J.; Zhang, L.; Fu, Y. Y.; Teixeira, I. F.; McPherson, I. J.; He, H. Y.; Tsang, S. C. E. Trimethylphosphine-assisted surface fingerprinting of metal oxide nanoparticle by 31P solid-state NMR: A zinc oxide case study. J. Am. Chem. Soc. 2016, 138, 2225–2234. [25] Kreissl, H. T.; Li, M. M. J.; Peng, Y. K.; Nakagawa, K.; Hooper, T. J. N.; Hanna, J. V.; Shepherd, A.; Wu, T. S.; Soo, Y. L.; Tsang, S. C. E. Structural studies of bulk to nanosize niobium oxides with correlation to their acidity. J. Am. Chem. Soc. 2017, 139, 12670−12680. [26] Kao, H. M.; Grey, C. P. Determination of the

31

P-27Al J-coupling constant for

trimethylphosphine bound to the Lewis acid site of zeolite HY. J. Am. Chem. Soc. 1997, 119, 627−628. [27] Sang, H.; Chu, H. Y.; Lunsford, J. H. An NMR study of acid sites on chlorided alumina catalysts using trimethylphosphine as a probe. Catal. Lett. 1994, 26, 235–246. [28] Manolikas, T.; Herrmann, T.; Meier, B. H. Protein structure determination from

13

C

spin-diffusion solid-state NMR spectroscopy. J. Am. Chem. Soc. 2008, 130, 3959−3966. [29] Peng, L. M.; Chupas, P. J.; Grey, C. P. Measuring Brønsted acid densites in zeolite HY with diphosphine molecules and solid state NMR spectroscopy. J. Am. Chem. Soc. 2004, 126,

12254−12255. [30] Barich, D. H.; Nicholas, J. B.; Xu, T.; Haw, J. F. Theoretical and experimental study of the 13

C chemical shift tensors of acetone complexed with Brønsted and Lewis acids. J. Am. Chem.

Soc. 1998, 120, 12342−12350.



[31] Haw, J. F.; Nicholas, J. B.; Ferguson, D. B. Physical organic chemistry of solid acids: Lessons from in situ NMR and theoretical chemistry. Acc. Chem. Res. 1996, 29, 259−267. [32] Sokol, A. A.; Catlow, C. R. A.; Garcés, J. M.; Kuperman, A. Computational investigation into the origins of Lewis acidity in zeolites. Adv. Mater. 2000, 12, 1801−1805. [33] Christe, K. O.; Dixon, D. A.; McLemore, D.; Wilson, W. W.; Sheehy, J. A.; Boatz, J. A. On a quantitative scale for Lewis acidity and recent progress in polynitrogen chemistry. J. Fluo. Chem. 2000, 101, 151−153. [34] Van Bokhoven, J. A.; Vander Eerden, A. M. J.; Koningsberger, D. C. Three-coordinate aluminum in zeolites observed with in situ X-ray absorption near-edge spectroscopy at the Al K-edge: Flexibility of aluminum coordinations in zeolites. J. Am. Chem. Soc. 2003, 125, 7435−7442. [35] Wischert, R.; Coperet, C.; Delbecq, F.; Sautet, P. Dinitrogen: A selective probe for tri-coordinate Al "defect" sites on alumina. Chem. Commun. 2011, 47, 4890−4892. [36] Wischert, R.; Coperet, C.; Delbecq, F.; Sautet, P. Optimal water coverage on alumina: A key to generate Lewis acid-base pairs that are reactive towards the C-H bond activation of methane. Angew. Chem., Int. Ed. 2011, 50, 3202−3205. [37] Valyon, J.; Engelhardt, J.; Kallo, D.; Hegedus, M. The activation of CD4 for H/D exchange over H-zeolites. Catal. Lett. 2002, 82, 29−35. [38] Silvi, B.; Savin, A. Classification of chemical-bonds based on topological analysis of electron localization functions. Nature 1994, 371, 683−686. [39] Petit, L.; Joubert, L.; Maldivi, P.; Adamo, C. A comprehensive theoretical view of the bonding in actinide molecular complexes. J. Am. Chem. Soc. 2006, 128, 2190−2191. [40] Borocci, S.; Giordani, M.; Grandinetti, F. Bonding motifs of noble-gas compounds as described by the local electron energy density. J. Phys. Chem. A 2015, 119, 6528−6541. [41] Román-Leshkov, Y.; Moliner, M.; Labinger, J. A.; Davis, M. E. Mechanism of glucose isomerization using a solid Lewis acid catalyst in water. Angew. Chem., Int. Ed. 2010, 49, 8954−8957. [42] Wang, L.; Wang, H.; Liu, F. J.; Zheng, A. M.; Zhang, J.; Sun, Q.; Lewis, J. P.; Zhu, L. F.; Meng, X. J.; Xiao, F. S. Selective catalytic production of 5-hydroxymethylfurfural from glucose by adjusting catalyst wettability. ChemSusChem 2014, 7, 402–406. [43] Pagán-Torres, Y. J.; Wang, T. F.; Gallo, J. M. R.; Shanks, B. H.; Dumesic, J. A. Production of 5-hydroxymethylfurfural from glucose using a combination of Lewis and Brønsted acid catalysts in water in a biphasic reactor with an alkylphenol solvent. ACS Catal. 2012, 2, 930−934.


Page 32 of 34

Page 33 of 34 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


[44] Christianson, J. R.; Caratzoulas, S.; Vlachos, D. G. Computational insight into the effect of Sn-beta Na exchange and solvent on glucose isomerization and epimerization. ACS Catal. 2015, 5, 5256−5263. [45] Hohwy, M.; Jakobsen, H. J.; Eden, M.; Levitt, M. H.; Nielsen, N. C. Broadband dipolar recoupling in the nuclear magnetic resonance of rotating solids: A compensated C7 pulse sequence. J. Chem. Phys. 1998, 108, 2686−2694. [46] Fang, H.; Zheng, A.; Chu, Y.; Deng, F. 13C chemical shift of adsorbed acetone for measuring the acid strength of solid acids: A theoretical calculation study. J. Phys. Chem. C 2010, 114, 12711−12718. [47] Becke, A. D. Becke’s three parameter hybrid method using the LYP correlation functional. J. Chem. Phys. 1993, 98, 5648−5652. [48] Lee, C.; Yang, W.; Parr, R. G. Development of the colle-salvetti correlation-energy formula into a functional of the electron density. Phys. Rev. B 1988, 37, 785−789. [49] Godbout, N.; Salahub, D. R.; Andzelm, J.; Wimmer, E. Optimization of Gaussian-type basis sets for local spin density functional calculations. Part I. Boron through neon, optimization technique and validation. Can. J. Chem. 1992, 70, 560−571. [50] Zheng, A.; Zhang, H.; Chen, L.; Yue, Y.; Ye, C.; Deng, F. Relationship between 1H chemical shifts of deuterated pyridinium ions and Brønsted acid strength of solid acids. J. Phys. Chem. B 2007, 111, 3085−3089. [51] Wolinski, K.; Hinton, J. F.; Pulay, P. Efficient implementation of the gauge-independent atomic orbital method for NMR chemical shift calculations. J. Am. Chem. Soc. 1990, 112, 8251−8260. [52] Zhang, Y.; Oldfield, E.

31

P NMR chemical shifts in hypervalent oxyphosphoranes and

polymeric orthophosphates. J. Phys. Chem. B 2006, 110, 579−586. [53] Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg,



J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian09, Revision B.01; Gaussian, Inc.: Wallingford, CT, 2010. [54] Lu, T.; Chen, F. Multiwfn: a multifunctional wavefunction analyzer. J. Comput. Chem. 2012, 33, 580−592.

TOC Figure


Page 34 of 34

On the Origin and Structure Characteristics of Tri-coordinated

Recommend Documents