Acidity Effects in Positron Annihilation Lifetime Spectroscopy of Zeolites

Subscriber access provided by READING UNIV

Article

Acidity Effects in Positron Annihilation Lifetime Spectroscopy of Zeolites Robbie Warringham, Lars Gerchow, David Cooke, Paolo Crivelli, Richard S. Vallery, Sharon Mitchell, and Javier Perez-Ramirez J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b11336 • Publication Date (Web): 19 Jan 2018 Downloaded from http://pubs.acs.org on January 20, 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 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 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.

The Journal of Physical Chemistry C 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 40 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

The Journal of Physical Chemistry

Acidity Effects in Positron Annihilation Lifetime Spectroscopy of Zeolites Robbie Warringham,† Lars Gerchow,‡ David Cooke,‡ Paolo Crivelli,*,‡ Richard S. Vallery,§ Sharon Mitchell,*,† and Javier Pérez-Ramírez*,† †

Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences,

ETH Zurich, Vladimir-Prelog-Weg 1, 8093 Zurich, Switzerland. ‡

Institute for Particle and Astrophysics, Department of Physics, ETH Zurich, Otto-Stern-Weg 5,

8093 Zurich, Switzerland. §

Department of Physics, Grand Valley State University, 118 Padnos Hall, Allendale, Michigan

49401, USA.

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 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 40

ABSTRACT Positron annihilation lifetime spectroscopy (PALS) is a complementary tool to study pore networks in zeolites as the diffusion of metastable ortho-positronium (o-Ps, an electron-positron bound pair) provides insights into connectivity unobtainable by other techniques. The accurate assessment of porosity requires knowledge of the interaction of o-Ps with acid centers commonly present in these materials. Although previous studies have highlighted a potential effect, the specific impact of the nature and concentration of acid sites remains unclear. By preparing a series of well-crystallized aluminum- and tin-containing MFI type zeolites, and the study of commercial materials, we map the effects of Brønsted and Lewis acidity on the PALS response. The results reveal that both types of acid sites decrease the amount of o-Ps detected, but through different mechanisms. Brønsted acid sites strongly affect the amount of o-Ps annihilating in the micropore network, but only nominally influences the amount out-diffusing from the crystal, which is attributed to an energy threshold for the interaction. Lewis acid sites originating from the incorporation of framework tin have a more substantial, but uniform impact ascribed to the suppressed formation of o-Ps. A similar effect is observed due to the Lewis acid sites in sodiumexchanged aluminum containing MFI, but the introduction of larger extraframework cations also hinders the diffusion of o-Ps. A preliminary model is proposed to describe the changes in the PALS response in the presence of acid centers.


2

Page 3 of 40 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


INTRODUCTION The development of improved tools for porosity assessment is of continued interest to guide the design of materials with enhanced properties. Positron annihilation lifetime spectroscopy (PALS) has attracted interest due to its ability to dynamically probe open volumes,1-3 providing complementary insights to more widespread equilibrium techniques as gas sorption. This has been demonstrated for silica powders4-6 and films,4,7,8 aluminosilicate gels,9 metal-organic frameworks,10,11 zeolites,12-19 and layered materials.20,21 The high sensitivity of PALS stems from the possibility of positrons to bind with electrons in the bulk forming metastable orthopositronium (o-Ps) atoms, which are sufficiently long-lived to diffuse through open pore networks and annihilate with a lifetime characteristic of the pore size.3 This property has recently been exploited in the characterization of zeolitic materials to discriminate the effectiveness of hierarchical pore networks in enhancing transport properties,14-16 to study coking and detemplation mechanisms,15,17 to assess the micropore topology,18,19 and monitor pore evolution during structural transformations.20 An interesting, but less explored facet of PALS is the sensitivity of o-Ps to acid sites present in zeolites, which could have significant implications for porosity assessment using this technique.2,22-25 Early works by Goldanskii et al., reported a chemical quenching effect, correlating the concentration of exposed Brønsted acid sites in partially hydrated amorphous silica-alumina catalysts with the lifetime of o-Ps.26,27 Specifically, it was proposed that o-Ps could be oxidized by the following reaction; H+ + o-Ps → H + e+

(1)

A later study of FAU-type zeolites by Nakanishi et al., using a combination of PALS and Doppler broadening spectroscopy (DBS), related variations in the lifetime and intensity of o-Ps


3


Page 4 of 40

contributions upon thermal treatment to the exposure of Brønsted acid sites upon dehydration.28 The presence of the latter was postulated to have two possible effects, either to inhibit the formation of o-Ps, or if formed to catalyze their oxidation. Similar findings were also reported by Gao et al. for a series of FAU-type zeolites.29 Huang et al. published a series of studies on various zeolite frameworks utilizing different positron spectroscopy methods, including PALS, DBS, and angular correlation of annihilation radiation (ACAR).30-34 The authors concluded that the strength of Brønsted acidity determined the rate of o-Ps oxidation and proposed a mechanistic cycle involving the decomposition and subsequent regeneration of the acid site.34 The role of Lewis acidity has rarely been addressed. Huang et al. reported that Lewis acid sites formed upon thermal treatment of zeolites played no role in the oxidation of o-Ps.34 In contrast, later calculations concluded that a higher rate of o-Ps oxidation would be expected over Lewis than Brønsted acid centers.35 The limited understanding of the effects of acidity on the PALS response is exacerbated by the fact that many previous studies focused on a narrow range of the lifetime spectra. Furthermore, the studies often failed to provide sufficient characterization data (e.g. type and concentration of acid sites present, crystal size, morphology etc.) to enable complete rationalization. In this study, well-crystallized MFI-type zeolites with tailored Brønsted (varying Si/Al ratio) and Lewis (lattice substitution of tin or sodium exchange) acidity were prepared to decouple the effects of the pore and acid-site structure on the PALS measurement. Zeolites with MFI-type framework were chosen due to the possibility of varying the acidic properties over a wide range while preserving the crystal size and morphology. A variable energy slow positron beam was utilized to differentiate annihilation in the micropore. A mechanistic discussion is developed based on the results.


4

Page 5 of 40 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


METHODS Zeolite Synthesis. A series of ZSM-5 zeolites comprised of coffin-shaped single crystals of ca. 20×5×0.5 µm with varying silicon-to-aluminum (purely siliceous, Si:Al = 80 and 50, coded Sil, Z80, and Z50, respectively) and silicon-to-tin (Si:Sn = 96, Sn-MFI) ratios were synthesized in fluoride media. For the synthesis of aluminum-containing samples, sodium aluminate (NaAlO2, Sigma Aldrich) and tetrapropylammonium bromide (TPABr, abcr, 98%) were dissolved in distilled water until a clear solution was attained. Ammonium fluoride (NH4F, Acros Organics, 98%) was added and fumed silica (Aerosil 130, Evonik) was slowly introduced in the mixture under vigorous stirring. The pH was adjusted to 7 through the addition of hydrofluoric acid (HF, Sigma Aldrich, 40wt.%). The resulting mixture was aged for 2 h at room temperature, leading to a final gel composition of 100 SiO2: 1 NaAlO2: 56 NH4F: 4.3 TPABr: 7990 H2O for Z80 and 100 SiO2: 3 NaAlO2: 112 NH4F: 3.5 TPABr: 7990 H2O for Z50. For the Sil sample, the same procedure was repeated in the absence of NaAlO2 leading to a final gel composition of 100 SiO2: 100 NH4F: 8 TPABr: 2000 H2O. Hydrothermal syntheses were conducted in Teflonlined stainless steel autoclaves at 448 K for 48 h, after which the solid products were recovered by filtration, thoroughly washed with distilled water, and dried at 338 K overnight. Subsequently the template was removed by calcining the samples at 823 K (5 K min−1) for 5 h. The assynthesized zeolites were converted to the ammonium form by three consecutive treatments in aqueous solutions of NH4NO3 (0.1 M, 100 cm3 g−1zeolite). Sn-MFI was prepared following a previously reported recipe.36 Briefly an aqueous solution of tin tetrachloride (SnCl4, Sigma Aldrich, 98%) was added dropwise under rapid stirring to an aqueous solution of NH4F. Thereafter, TPABr dissolved in distilled water was slowly added to the mixture and stirred for 30 min.

Fumed

silica

was

then

added,

leading

to

a

final

gel

composition

of


5


Page 6 of 40

200 SiO2 : 1 SnO2: 52 TPABr: 200 NH4F : 7000 H2O. In this case, hydrothermal synthesis was conducted at 473 K for 144 h. Collection of the zeolite and template removal was performed as described above. Commercial ZSM-5 zeolites of varying Si:Al ratio were obtained for comparative analysis; ZC1000 (HSZ-890HOA, H+-form, Si:Al = 1060, TOSOH), ZC40 (CBV8014, NH4-form, Si:Al = 40, Zeolyst International), and ZC15 (CBV302-E, NH4-form, Si:Al = 15, Zeolyst International). These samples comprise of crystal aggregates with a particle size range of 0.5-2.0 µm in diameter. A tin-containing sample of similar morphology was also obtained via the alkaline-assisted stannation of ZC1000 (ZC-Sn) following a previously described protocol.36 This was achieved by adding the parent material to an aqueous solution of sodium hydroxide (NaOH, Sigma Aldrich, 97%) and tin sulfate (SnSO4, abcr, 95%). The mixture was stirred at 338 K for 0.5 h in an Easymax 102 reactor system (Mettler Toledo). Thereafter, the treatment was quenched in distilled water and the suspended material filtered and washed until neutral pH of the filtrate was achieved. Calcination and ion exchange were performed on these solids following the same protocols outlined for Z80 and Z50. Partially (ZC40-Na-0.5) and fully (ZC40-Na-1) sodium-exchanged samples were obtained by three consecutive treatments of the as-received ZC40 in aqueous solutions of NaNO3 (0.1 mM and 0.1 M for ZC40-Na-0.5 and ZC40-Na-1 respectively, 100 cm3 g−1zeolite). The resulting slurries were stirred for 12 h at room temperature before the zeolite was collected by filtration, washed with deionized water, and dried overnight at 338 K. All samples were calcined in static air at 823 K (5 K min−1) for 5 h before further characterization. Characterization Methods. Inductively coupled plasma optical emission spectroscopy (ICP-OES) was conducted using a Horiba Ultra 2 instrument. Samples were dissolved in HF (Sigma Aldrich, >99.99%) and neutralized in boric acid solution (Sigma Aldrich, >99.8%)


6

Page 7 of 40 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


before analysis. Nitrogen sorption at 77 K was performed using a Micromeritics 3Flex instrument after sample evacuation at 473 K for 3 h. X-ray diffraction (XRD) was performed using a PANalytical X’Pert PRO-MPD diffractometer operated in Bragg-Brentano geometry using Ni-filtered Cu Kα (λ = 0.1541 nm) radiation. Data were recorded in the range of 5-60° 2θ with an angular step size of 0.05° and a counting time of 2 s per step. Scanning electron microscopy (SEM) was undertaken by using a Zeiss Leo 1530 microscope operated at 5 keV. Infrared spectroscopy of adsorbed pyridine was undertaken using a Bruker IFS 66 spectrometer. Self-supporting zeolite wafers (1 cm2) were degassed at 10−3 mbar and 693 K for 4 h prior to analysis. Following adsorption at room temperature, weakly bound molecules were evacuated at 473 K for 0.5 h. The concentrations of Brønsted and Lewis acid sites were determined using extinction coefficients of εBrønsted = 1.67 cm µmol−1 and εLewis = 2.94 cm µmol−1 for aluminum containing samples,37 and εLewis = 1.42 cm µmol−1 for tin-containing samples.38 Diffuse reflectance infrared Fourier Transform spectroscopy (DRIFTS) was conducted under constant air flow using a Bruker Vertex 70 spectrometer equipped with a Harrick Praying Mantis high-temperature diffuse reflectance accessory. Spectra were acquired in the range of 4000-650 cm−1 with a nominal resolution of 4 cm−1 for 32 scans. Temperature programmed surface reaction (TPSR) of n-propylamine and the temperature programmed desorption

of

ammonia (NH3-TPD) were undertaken using a Micromeritics AutoChem 2920 system connected to a MKS Cirrus 2 residual gas analyzer. For TPSR analysis samples were pretreated in situ in flowing He (50 cm3 min−1) at 773 K for 2 h, followed by saturation with n-propylamine at 473 K. Physisorbed amine was removed by purging with He. Decomposition of n-propylamine was monitored in the range of 473-773 K using a heating rate of 10 K min−1. For NH3-TPD measurements, samples were pretreated in situ in flowing He (50 cm3 min−1) at 773 K for 2 h


7


Page 8 of 40

before exposure to three consecutive saturation cycles of ammonia (10 vol% NH3 in 20 cm3 min−1 He flow) at 473 K for 30 min, followed by purging in He at the same temperature for 1 h. Desorption of NH3 was monitored in the range of 473-873 K, ramping at 10 K min−1. X-ray photoelectron spectroscopy (XPS) on the sample powder was performed using a Physical Electronics Instruments Quantum 2000 spectrometer using mono-chromatic Al Kα radiation generated from an electron beam operated at 15 kV and 32.3 W. The spectra were collected under ultrahigh vacuum conditions at a pass energy of 46.95 eV. Positron Annihilation Lifetime Spectroscopy. PALS measurements were performed using the ETHZ slow positron beam.15 Powdered samples (ca. 100 mg) were suspended in acetone (50 cm3 g−1) and deposited on the sample stage resulting in a uniform layer of ca. 1 mm thick and degassed in situ under vacuum (Lx or y>Ly or z>Lz, the distribution is folded back on the surface from which it exits. The generated three-dimensional profile is then convoluted along the crystal boundary to account for every possible incoming angle at every point of the crystal surface. The resulting profile describes the positron and thus the approximate initial positronium distribution averaged over all crystals. The simulations indicate that for the large coffin shaped crystals of 20×5×0.5 µm, the only appreciable differences in implantation occur along the shorter axis. Thus


9


Page 10 of 40

we reduce the three-dimensional distribution to two-dimensions by averaging along the long axis to get P(y,z).

RESULTS Properties of the Zeolite Samples. To quantify the impact of the acidic properties on the PALS response a series of well-crystallized coffin-shaped ZSM-5 zeolites of varying Si:Al ratios (Sil, Z80, and Z50) and a tin-containing MFI sample of similar morphology (Sn-MFI) were synthesized (Figure 1, Table 1, Figure S1). Substitution of Si by Al or tin could be expected to impact the number of electrons available to form positronium, therefore distorting the PALS comparison between Sil and the aluminum- /tin-containing samples. However, the variation in electron numbers is relatively small with the percentage increase from Sil to Z50 being 0.07%, and for Sn-MFI is 1.25%. In terms of the expected effect on positronium formation, the variation in electron numbers would be negligible. The study of large crystals is important to minimize kinetic effects due to the out-diffusion of o-Ps from the zeolite, which can shorten the lifetime and reduce the amount of o-Ps sampling the micropore network.19 It also ensures greater control over the positron stopping profile, since the mean implantation depth is smaller than the crystal size. This is illustrated in Figure 2 and Figure S2 which considers the modelled threedimensional positron implantation profiles for crystals of different sizes at 3, 5, 7.5, and 10 keV incident positron energy. In contrast to typical one-dimensional models where positron implantation occurs perpendicular to an infinitely large plane, in the presented model positrons are implanted across all surface at each point and every angle, which is more representative of the randomly-orientated distribution of crystals that will be measured experimentally. At 5 keV a homogeneous positron distribution is observed for small crystals (0.5×0.5 µm in diameter). This


10

Page 11 of 40 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


results in a reduced intensity of o-Ps annihilation in the zeolite micropore and an increased intensity of out-diffused o-Ps (>10 ns). Comparatively, with dimensions similar to the large crystal samples studied here (20×5×0.5 µm), a higher density of positrons is observed close to the surface at low implantation energies, but the distribution becomes more uniform upon deeper positron implantation at 10 keV (Figure S2). The isomorphous substitution of aluminum into the MFI-type framework leads to the formation of Brønsted acid sites when the zeolite is in protonic form, while the introduction of tin gives rise to Lewis acid sites. ICP-OES analysis confirmed the Si:Al and Si:Sn ratios of the synthesized materials were close to the targeted values (Table 1). Assessment by N2 sorption evidenced the similar porous properties of the samples, which all exhibited type-I isotherms characteristic of purely microporous materials (Table 1, Figure S1), with a micropore volume in the range of 0.15-0.16 cm3 g ̶ 1. The high crystallinity of the samples was measured by XRD, exhibiting patterns consistent with MFI-type zeolites (Figure S1). The coffin-shaped morphology was visible by SEM, with the aluminum-containing samples all displaying average dimensions of approximately 20×5×0.5 µm (Figure 1a). The Sn-MFI is slightly smaller (10×2.5×0.5 µm) but the dimensions are still large with respect to the range of positron implantation depths (0.2-1.6 µm) studied. Multi-technique assessment confirmed the model acidic properties of the samples (Figure 1b, Figure 1c, Figure 1d, Figure 1e). In particular, an increased concentration of Brønsted acid sites from 0 in Sil to 144 µmol g−1 in Z50 was observed by the IR spectroscopy study of adsorbed pyridine (Table 1). These trends were reflected in the DRIFTS spectra in the OH stretching region (Figure 1b), where only small intensities relating to internal and external silanol groups were measured for Sil, Z80, and Z50, with negligible intensity associated with silanol nests at 3500 cm−1 in line with the low amount of defects present


11


Page 12 of 40

in the crystals.41 This can be clearly seen upon comparison with the characterization data of the commercially available smaller crystal samples (Figure S3). Aside from the concentration, both the strength and distribution of acid sites could impact the interaction with o-Ps. Analysis by XPS revealed a close agreement between the Si/Al ratio determined at the near surface and that identified in the bulk by ICP-OES for Z50 (Table 1), pointing towards a homogeneous distribution of aluminum in the sample. Equivalent measurements of Z80 were not fruitful as the low concentration of aluminum in these samples is close to the detection limit for the technique. The comparable strength of Brønsted acid sites in the model samples was confirmed by TPSR of n-propylamine (Figure 1d). The temperature of propene evolution due to the decomposition of the stoichiometrically adsorbed reactive probe molecule, via a Hofmann-type elimination, decreases over stronger acid sites.42 Comparison of the mass spectrometry profiles for Z50 and Z80 (Figure 1d) shows that the peak maximum occurs at similar temperatures. The larger peak observed for Z50 is consistent with the higher concentration of Brønsted acid sites in the sample. Further analysis by NH3-TPD (Figure 1e) is consistent with the previously discussed results, both Z50 and Z80 exhibiting a similar desorption peak shape and temperature maximum. Impact of Brønsted Acid Centers. PALS measurements of Sil, Z80, and Z50 were performed to assess the impact of varying Brønsted acidity on the annihilation behavior of o-Ps. Additional information can be obtained upon varying the incident positron energy, thereby controlling the depth of positron implantation to assess different volumes within the zeolite crystal. The contour plot in Figure 3 show the correlation between Io-Ps (i.e. the total measured amount of o-Ps), Brønsted acid site concentration (cBrønsted) and implantation depth. No significant variation in Io-Ps was observed for Sil, consistent with the high structural uniformity


12

Page 13 of 40 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


and absence of acid centers in these samples. However, increases in cBrønsted clearly lead to a significant drop in Io-Ps, which is more pronounced at higher positron implantation depths. This is consistent with previous reports of the oxidative effect of Brønsted acidity on positronium.22-34 Interestingly, comparison of the absolute intensities of the different components reveals a stronger effect of the presence of Brønsted acid sites on Imicro (the intensity of o-Ps annihilation in the zeolite micropores) compared to Iout (the combined intensity of annihilation components >10 ns) (Figure 4a), revealing a non-uniform impact on o-Ps species. In particular, these observations indicate that the presence of Brønsted acid centers does not affect the diffusion of o-Ps out of the crystal. This is consistent with the comparable τmicro values within each of the samples, which would be expected to vary if Brønsted acidity disrupted the diffusion of o-Ps in the micropore (Figure 4b, Table S1-S3). This would imply that a fraction of the out-diffused oPs is energy independent and could therefore originate from a delocalized Bloch state.11,18 To generalize the observations for the Sil, Z80, and Z50, a series of commercially available samples with differing acidity were studied (ZC1000, ZC40, and ZC15, Table 1, Figure S3, Figure S4). The morphology of these samples consists of smaller crystals (ca. 2 µm for ZC1000 and 0.5 µm for ZC40 and ZC15) arranged in larger crystal aggregates of >2 µm average diameter. The effect of these types of morphologies and the incident positron energy on the PALS response have been previously discussed (Figure 2, Figure S2), and larger kinetic effects are expected due to the increased out-diffusion of o-Ps.19 Comparing the fits (Table S4), an additional component between 30-60 ns appears, consistent with the annihilation of o-Ps in the intercrystalline voids present in these samples. Additionally the intensity of the longest component, Ivac, is much higher due to the reduced diffusion path out of the crystals. Nonetheless, as for the large crystal samples, a reduction in the measured amount of o-Ps (from


13


Page 14 of 40

40.8% to 22.5%) was observed with increasing cBrønsted (Figure 5, Table S4). Another comparable trend is the substantial drop in Imicro, from 9.5% in ZC1000 to 1.6% in ZC15, indicating that the general response of o-Ps with cBrønsted occurs irrespective of crystal size. More notable discrepancies between the large crystal and commercial samples occur in the lifetime values for o-Ps. However, the fitting procedure of these components can be influenced by the corresponding low intensity and higher level of out-diffusion.7,19 Impact of Lewis Acid Centers. Lewis acidity can be introduced in zeolites in different ways. A major factor limiting the understanding of the effect of Lewis acidity from previous studies is that the majority of samples studied contained both Brønsted and Lewis acid sites, and the latter were primarily related to extraframework aluminum species the structure of which was poorly characterized. A more elegant approach to quantify the potential influence of Lewis acid sites is to prepare a purely Lewis acidic material, which in MFI can be readily achieved through the introduction of tin into the framework while preserving a similar crystal size and coffin-shaped morphology (Sn-MFI, Table 1, Figure 1, Figure S1). Comparison with Sil as a non-acidic reference, a substantial drop in intensity was observed in the PALS spectra acquired for Sn-MFI (Figure S5). This observation is reflected in the fitted Io-Ps values with ca. 42% loss compared to Sil (Table S5). Remarkably, inspection of the lifetime and intensity values of the individual components show negligible variation across the energies studied, suggesting that the presence of tin in the framework has a uniform effect on o-Ps annihilation. This indicates a difference in the mechanism of interaction with Brønsted and Lewis acid centers. Possible hypotheses for this behavior are discussed later in this paper. The influence of Lewis acid centers was extended to the commercial materials by introducing tin centers into ZC1000 via alkaline-assisted stannation (ZC-Sn), leading to an observable


14

Page 15 of 40 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


increase in cLewis (Table 1).36 Unlike Sn-MFI, ZC-Sn also exhibits a low concentration of Brønsted acid sites due to the presence of a small amount of aluminum. From the PALS analysis (Table S6), the introduction of Lewis acidity considerably reduces the amount of o-Ps detected compared to ZC1000 (ca. 49% at 5 keV). The observed drop is comparable to that evidenced between Sil and Sn-MFI at 5 keV (ca. 42%), further confirming the strong effect of Lewis acid centers on the PALS response. Impact of Sodium Exchange. A common strategy for the elimination of Brønsted acid centers in zeolites is the exchange of acidic protons with cationic species such as Na+. Goldanskii et al. proposed that, in terms of positron measurements, the reaction; Na+ + Ps → Na + e+

(4)

in silica and alumina gels is endothermic and therefore plays no role in the oxidation of o-Ps.22 Similar findings have also been reported from the study of different forms of a FAU-type zeolite Y (Na-, NH4-, and H-), where it was proposed that Brønsted acid centers favor the formation and annihilation of o-Ps at the pore surface of the zeolite, leading to shortening of the lifetime.29 A considerable increase in the intensity of o-Ps annihilation has also been reported upon sodium exchange of a zeolite Y sample.43 The preceding studies primarily focused on the chemical reaction between o-Ps and an acid center and therefore the discussion often did not consider the effects on o-Ps out-diffusion and the implications for porosity assessment by PALS. To address this, the Z50 sample was fully exchanged with sodium (Z50-Na, Table 1). The absence of Brønsted acidity was confirmed by IR spectroscopy of adsorbed pyridine (Table 1). An additional band ascribed to pyridine adsorbed on a sodium cation appears at 1445 cm−1.44-46 To the best of our knowledge no molar extinction coefficient has been determined to quantify the amount of Na+ centers, but this can be estimated from the concentration of sodium in the


15


Page 16 of 40

samples, which as expected was in good agreement with the aluminum content (cNa+ = 160 µmol g−1, Na:Al = 1.1 in Z50-Na). PALS analysis of the exchanged sample indicates a similar distribution of components as in the Brønsted acidic Z50 (Table S7). As expected, the removal of Brønsted acid centers led to an increase in Io-Ps across all the studied energies. Nonetheless, the values remain substantially lower than Sil suggesting that the removal of Brønsted acid centers by ion exchange does not fully recover the acid-site free values. Interestingly, the measured Io-Ps are comparable to those observed for Sn-MFI pointing towards a similar influence of the distinct Lewis acid center (Na+ vs. framework Sn). Considering the individual components, Z50-Na displays a marked increase in Imicro, which constitutes ca. 83% of the total o-Ps measured compared to ca. 57% for Z50. Comparatively, the increase in Imicro for Z50-Na (19.1% at 3 keV to 29.7% at 10 keV) is more significant than that observed for Sn-MFI, which is steady at ca. 27% from 5 keV upwards. This could suggest the presence of sodium in the micropore volume impedes the diffusion of positronium, increasing the extent of o-Ps annihilation in the micropore. A series of sodium-exchanged ZC40 samples were also prepared where the Brønsted acid centers were partially (ZC40-Na-0.5) or fully (ZC40-Na-1) exchanged (Table 1, Figure S6). Consistent with the progressive incorporation of the larger sodium cations, a linear correlation was observed between Vads and cBrønsted (Figure 5). As with the Z50-Na sample, the partial or full exchange of Brønsted acid centers led to an overall increase in Io-Ps, but the values remained lower than ZC1000 (Figure 5, Table S4). However, τmicro dropped from ca. 5 ns in ZC40 to 3 ns for ZC40-Na-1, which is closer to the micropore value observed in the larger crystal, further suggesting that the presence of sodium may impede out-diffusion of o-Ps and slightly mitigate the larger kinetic effects observed for the small crystals.


16

Page 17 of 40 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


DISCUSSION Distinct Effects of Brønsted and Lewis Acidity. An overview of the relative effects of Brønsted and Lewis acidity can be obtained by comparing the trends in o-Ps intensity of the large crystal zeolites with implantation depth after normalization to the values measured at 5 keV (Io-Ps,N = Io-Ps/Io-Ps,5keV, Imicro,N = Imicro/Imicro,5keV, Iout,N = Iout/Iout,5keV). Note that the values measured at 3 keV are not used due to the non-negligible contribution of backscattered positrons and positronium at this energy. Consistent with its non-acidic character, minor variation in Io-Ps,N is observed for Sil (Figure 6a). Comparatively Imicro,N also stays fairly constant (Figure 6b), while Iout,N exhibits a slight drop (Figure 6c). The variation in Iout,N confirms that some o-Ps are confined within and diffusing through the micropores, and thus the probability of escape from the crystal decreases with increasing implantation depth. An associated increase in Imicro is not obvious from the normalized trends due to the significantly higher absolute values with respect to Iout. Considering Z80 and Z50 i.e. the samples with high Brønsted acidity, a marked drop in Imicro,N is observed with respect to Sil (Figure 6b). This trend is less exaggerated for Z50 due to the already low absolute value of Imicro at 5 keV. Interestingly, the trend of Iout,N is comparable to Sil (Figure 6c). In the case of the Lewis acidic samples, Sn-MFI and Z50-Na are found to display very similar behavioral trends, despite the significant distinctions in absolute intensity values. A slight increase in Imicro with incident energy for Z50-Na likely relates to the impeded diffusion of o-Ps in the micropore due to the presence of sodium cations, but would require further investigation to confirm. To explain the observed trends, it is necessary to consider the mechanisms of o-Ps formation, thermalization, and diffusion. As illustrated in Figure 7, the formation of positronium in materials is described by the Ore1 and Spur, or Bubble model,47-49 which proposes that high


17


Page 18 of 40

energy positrons implanted into the material collide with and ionize atoms. Positronium can form by picking up an electron after ionization, with the energy required for formation dependent on the ionization potential of the atom and the binding energy of positronium. Thus, the direct annihilation of positrons is favored in comparison to positronium formation, resulting in a large intensity of lifetime components 0

(5)

where ∆Hdis is the enthalpy of positronium dissociation. For this study p-Ps is considered to be 0 then positronium will require a certain kinetic energy (EPs) to overcome this threshold of interaction. Additionally if this process can happen after many interactions, at a time consistent with the expected lifetime


18

Page 19 of 40 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


within the MFI framework (ca. 2.5-3 ns), then the observed τmicro would be expected to decrease with increasing acid concentration. However, no such dependence is observed (Table S1-S3). A proposed explanation is that localized o-Ps enter the pore with a few eV of energy

50

and

quenching occurs before it makes many interactions. The following relation can be drawn; Eout ≤ ∆HB.ox ≤ Emicro

(7)

where Eout and Emicro are the kinetic energies of out-diffused positronium and positronium annihilating in the micropores respectively. It is also possible that high-energy localized positronium is able to sufficiently thermalize below the energy threshold, introducing additional paths for annihilation within the micropore and out diffusion (Figure 7). However, as the trends in Iout are similar for all samples (Figure 6c), these fractions are considered to be small. An equivalent expression to Equation 8 which describes the effects of Lewis acidity on the PALS response is more difficult to propose, as although the variation between Imicro and Iout with energy is identical between Sil and Sn-MFI (Figure 6), the absolute values exhibit a consistent decrease with respect to Sil (Table S5) i.e. both Emicro and Eout decrease equally. As Brønsted acidity only impacts Imicro significantly, it is proposed that Lewis acidity has a more uniform effect on the PALS response and could directly impact the positronium formation process either by scavenging spur electrons that would normally be available for the positron to interact or by scavenging positrons themselves (Figure 7). Decoupling Porosity and Acidity Effects. The previous application of PALS to probe pore connectivity in zeolites was primarily based on the assessment of the normalized fraction of out-diffused o-Ps (Cpore = Iout/Io-Ps).14 Considering the identified impacts, correction for the effects of Brønsted and Lewis acidity will be essential for the precise quantification of pore architecture. In the presence of Brønsted acid sites, the fact that the oxidation only impacts


19


Page 20 of 40

localized high-energy o-Ps is advantageous as it means that the diffusion through and sampling of the micropore network is unaltered. However, the differing effects on Imicro and Iout cause variations in Cpore unrelated to pore structure. Fortuitously, in the case of purely Lewis acidic materials, assuming that the acid centers are homogeneously distributed and therefore suppress o-Ps formation uniformly throughout the sample, they should not affect the determination of Cpore. However, as outlined below, accounting for the reduced formation of o-Ps may be necessary if mixed with Brønsted acid sites. In this regard, cation exchange with small alkali metals such as sodium can be a good strategy to avoid the need for correcting for Brønsted acidity, but the species introduced should not significantly impede o-Ps diffusion and therefore the effectiveness will be greater for large and medium-pore zeolites. Looking towards correcting for the effects of Brønsted and Lewis acidity, models have been proposed for the effect of electrons scavengers in other materials on o-Ps formation.51,52 As Lewis acidity was seen to effect both Imicro and Iout equally (Figure 6), the following expressions are proposed; ILout =I0out /(1+(σL×cLewis)α)

(8)

ILmicro =I0micro /(1+(σL×cLewis) α)

(9)

where ILout and ILmicro are the intensities of the out-diffused and micropore o-Ps respectively in the presence of Lewis acidity, I0out and I0micro are the intensities without acid sites, σL is the electron scavenging co-efficient, and α is a fitting parameter. However, as Brønsted acidity only affects high-energy positronium in the micropore (vide supra, Figure 6, Figure 7), an additional coefficient (σB) for Imicro must be considered in materials containing both types of site; α 0 σB ×cBrønsted IB,L +IB micro =Imicro /(1+(σL×cLewis) )×e

(10)


20

Page 21 of 40 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


B,L where Imicro is the intensity of o-Ps after interacting with the acid sites and IB is the asymptotic

intensity observed at high cBrønsted. The inverse exponential function attached to σB is derived from the experimental observation with Imicro in Figure 4. Given that the majority of out-diffused positronium is thought to originate from the delocalized Bloch state, which has too low an energy to interact with the Brønsted acidity, the impact on ILout is negligible and can be ignored. Thus, the measured out-diffused intensity Iout = ILout (Equation 8) and the micropore intensity Imicro = IB,L micro (Equation 9). As a preliminary assessment, the data for the large-crystal Sil, Z80, Z50, Sn-MFI, and Z50-Na obtained at 5 keV and 10 keV have been fitted simultaneously using a least-squares fitting routine obtaining correlations between Iout and cLewis (Figure 8a) and Imicro with cBronsted and cLewis (Figure 8b and Figure S6). Considering the different trends observed for Iout and Imicro (Figure 6), σB has been fitted independently at different incident positron energies, whereas σL has been kept constant. Additionally, fitting of the α parameter obtained a value of 0.5, which is consistent with previous literature.51,52 Due to the limited data set the fits have large 90% confidence bands and so interpretation remains qualitative. In general the experimental and the estimated trends show good agreement and similar profiles are observed at both energies. The proposed model stresses the possible implications of acidity in interpreting PALS results in zeolites. A larger sample series is required to further parameterize the multiple system variables that can influence the PALS response with sufficient confidence. For example, it was not possible to unite the trends for the large-crystal and commercial samples. An improved description of kinetic effects requires a better understanding of the impacts of crystal size and morphology.19 Although attaining reasonable quantitative trends, the broad definition of Brønsted and Lewis acidity in the current study does not consider variations in strength relating to the specific origin or possible heterogeneity in their distribution. It was also not yet possible to


21


Page 22 of 40

discriminate the relative fractions of localized and delocalized o-Ps. Finally, it is noted that this study was possible due to the flexibility of synthesizing MFI-type zeolites with model acidity and equivalent porosity. While similar behavior is expected for zeolites of different framework type, knowledge of the influence of the former factors (i.e. kinetic effects, acid strength, distribution etc.) and an improved understanding of topology effects on the delocalized state of o-Ps will be critical to fully interpret PALS measurements.

CONCLUSIONS This study has quantified the relative effects of Brønsted and Lewis acidity on the analysis of zeolites by PALS. A series of MFI-type samples with controlled Brønsted and Lewis acidity were prepared with large crystal sizes to minimize the kinetic effects associated with o-Ps outdiffusion. The total amount of measured o-Ps was found to decrease with increasing acidity, irrespective of the nature of the acid sites. Deconvolution of the distinct annihilation events revealed that Brønsted acidity diminishes the amount of o-Ps annihilating in the micropore more than the number of longer-lived out-diffused components. This observation was attributed to the selective oxidation of high-energy positronium species in localized states within the micropore. The study of tin-containing samples demonstrated that Lewis acid centers exhibit a stronger effect on the PALS response than Brønsted acid centers. However, the equivalent impact on all o-Ps components revealed that this resulted from suppressed positronium formation, likely due to either scavenging spur electrons from the positron thermalization process or scavenging positrons. The results were generalized by studying a series of commercial MFI-type zeolites, where compatible trends were observed. Sodium exchange of the zeolites led to an increase in the intensity of o-Ps, but did not recover the values of the non-acidic sample due to presence of


22

Page 23 of 40 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


Lewis acidity in this sample. The observed effects have direct implications for porosity analysis and require the development of a correction model. A first empirical attempt to account for the distinct impacts of Brønsted and Lewis acid centers considering the relative concentrations and interaction cross-sections with o-Ps obtained a reasonable description for the large-crystal samples, but requires further parameterization to become generally applicable. In particular, the impacts of kinetic effects, acid strength, and zeolite framework type still need to be addressed. Further, improved models of the effect of introducing heteroatoms in the zeolite framework on the PALS response are required to properly understand the observed behavior. The ability to evaluate both the porosity and acidity of zeolites using positrons will widen the scope of the technique for the analysis of functional materials.

ASSOCIATED CONTENT Supporting Information. Full results from the spectral fits of the PALS data and additional characterization data. The following files are available free of charge. Supporting Information (PDF)

AUTHOR INFORMATION Corresponding Author * E-mail: [email protected]. Tel.: +41 4463 33511. * E-mail: [email protected]. Tel.: +41 4463 23123. * E-mail: [email protected]. Tel.: +41 4463 37120. Author Contributions


23


Page 24 of 40

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGEMENTS The authors acknowledge ETH Zurich (grant ETH-33 15-1) for funding. ScopeM for access to its facilities. Dr. G. M. Lari is thanked for synthesis and characterization of Sn-MFI, Dr. A. J. Martín is thanked for SEM measurements, and Dr. R. Hauert is thanked for XPS analysis.

REFERENCES 1. Mogensen, O. E. Positron Annihilation in Chemistry; Springer-Verlag: Berlin, Germany, 1995. 2. Schmidt, W. In Handbook of Porous Solids; Schüth, F., Sing, K. S. W., Weitkamp, J., Eds.; Wiley-VCH: Weinheim, Germany, 2002; Vol. 1, pp 508−510. 3. Gidley, D. W.; Peng, H.-G.; Vallery, R. S. Positron Annihilation as a Method to Characterize Porous Materials. Annu. Rev. Mater. Res. 2006, 36, 49−79. 4. Gidley, D. W.; Frieze, W. E.; Dull, T. L.; Yee, A. F.; Ryan, E. T.; Ho, H.-M. Positronium Annihilation in Mesoporous Thin Films. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 60, R5157−R5160. 5. Babu, G. P.; Manohar, V.; Sudarshan, K.; Pujari, P. K.; Manohar, S. B.; Goswami, A. Positron Annihilation Studies in Silica Supported Nickel Carbonate Systems. J. Phys. Chem. B. 2002, 106, 6902−6906.


24

Page 25 of 40 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


6. Zaleski, R.; Wawryszczuk, J.; Goworek, J.; Borówka, A.; Goworek, T. Vacuum Removal of the Template in MCM-41 Silica Studied by the Positron Annihilation Method. J. Colloid Interface Sci. 2003, 262, 466−473. 7. Liszkay, L.; Corbel, C.; Raboin, L.; Boilot, J. P.; Perez, P.; Brunet-Bruneau, A.; Crivelli, P.; Gendotti, U.; Rubbia, A.; Ohdaira, T.; et al. Mesoporous Silica Films With Varying Porous Volume Fraction: Direct Correlation Between Ortho-Positronium Annihilation Decay and Escape Yield Into Vacuum. Appl. Phys. Lett. 2009, 95, 124103. 8. Liszkay, L.; Guillemot, F.; Corbel, C.; Boilot, J.-P.; Gacoin, T.; Barthel, E.; Pérez, P.; Barthe, M.-F.; Desgardin, P.; Crivelli, P.; et al. Positron Annihilation in Latex-Templated Macroporous Silica Films: Pore Size and Ortho-Positronium Escape. New J. Phys. 2012, 14, 065009. 9. Kajcsos, Z.; Kosanović, C.; Bosnar, S.; Subotić, B.; Major, P.; Liszkay, L.; Bosnar, D.; Lázár, K.; Havancsák, H.; Luu, A. T.; et al. Monitoring the Crystallization Stages of Silicalite by Positron Lifetime Spectroscopy. Mater. Sci. Forum. 2009, 607, 173−17631. 10. Dutta, D.; Feldblyum, J. I.; Gidley, D. W.; Imirzian, J.; Liu, M.; Matzger, A. J.; Vallery, R. S.; Wong-Foy, A. G. Evidence of Positronium Bloch States in Porous Crystals of Zn4OCoordination Polymers. Phys. Rev. Lett. 2013, 110, 197403. 11. Crivelli, P.; Cooke, D.; Barbiellini, B.; Brown, B. L.; Feldblyum, J. I.; Guo, P.; Gidley, D. W.; Gerchow, L.; Matzger, A. J. Positronium Emission Spectra from Self-Assembled MetalOrganic Frameworks. Phys. Rev. B. 2014, 89, 241103. 12. Ito, Y.; Takano, T.; Hasegawa, M. Positron Annihilation in Synthetic Zeolites. Appl. Phys. A: Solids Surf. 1988, 45, 193−201.


25


Page 26 of 40

13. Nakanishi, H.; Ujihira, Y. Application of Positron Annihilation to the Characterization of Zeolites. J. Phys. Chem. 1982, 86, 4446−4450. 14. Milina, M.; Mitchell, S.; Cooke, D.; Crivelli, P.; Pérez-Ramírez, J. Impact of Pore Connectivity on the Design of Long-Lived Zeolite Catalysts. Angew. Chem., Int. Ed. 2015, 54, 1591−1594. 15. Milina, M.; Mitchell, S.; Crivelli, P.; Cooke, D.; Pérez-Ramírez, J. Mesopore Quality Determines the Lifetime of Hierarchically Structured Zeolite Catalysts. Nat. Commun. 2014, 5, 3922−3931. 16. Kenvin, J.; Mitchell, S.; Sterling, M.; Warringham, R.; Keller, T. C.; Crivelli, P.; Jagiello, J.; Pérez-Ramírez, J. Quantifying the Complex Pore Architecture of Hierarchical Faujasite Zeolites and the Impact on Diffusion. Adv. Funct. Mater. 2016, 26, 5621−5630. 17. Warringham, R.; Gerchow, L.; Zubiaga, A.; Cooke, D.; Crivelli, P.; Mitchell, S.; PérezRamírez, J. Insights Into the Mechanism of Zeolite Detemplation by Positron Annihilation Lifetime Spectroscopy. J. Phys. Chem. C. 2016, 120, 25451−25461. 18. Zubiaga, A.; Warringham, R.; Mitchell, S.; Gerchow, L.; Cooke, D.; Crivelli, P.; PérezRamírez, J. Pore Topology Effects in Positron Annihilation Spectroscopy of Zeolites. ChemPhysChem, 2017, 18, 470−479. 19. Zubiaga, A.; Warringham, R.; Boltz, M.; Cooke, D.; Crivelli, P.; Gidley, D.; Pérez-Ramírez, J.; Mitchell, S. The Assessment of Pore Connectivity in Hierarchical Zeolites using Positron Annihilation Lifetime Spectroscopy: Instrumental and Morphological Aspects. Phys. Chem. Chem. Phys. 2016, 18, 9211−9219. 20. Jagiello, J.; Sterling, M.; Eliášová, P.; Opanasenko, M.; Zukal, A.; Morris, R. E.; Navaro, M.; Alvaro Mayoral.; Crivelli, P.; Warringham, R.; et al. Structural Analysis of IPC Zeolites


26

Page 27 of 40 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


and Related Materials Using Positron Annihilation Spectroscopy and High-Resolution Argon Adsorption. Phys. Chem. Chem. Phys. 2016, 18, 15269−15277. 21. Warringham, R.; Mitchell, S.; Murty, R.; Schaeublin, R.; Crivelli, P.; Kenvin, J.; PérezRamírez, J. Mapping the Birth and Evolution of Pores Upon Thermal Activation of Layered Hydroxides. Chem. Mater. 2017, 29, 4052−4062. 22. Goldanskii, V. I.; Firsov, V. G. Chemistry of New Atoms. Annu. Rev. Phys. Chem. 1971, 22, 209−258. 23. Kobayashi, Y.; Ito, K.; Oka, T.; Hirata, K. Positronium Chemistry in Porous Materials. Radiat. Phys. Chem. 2007, 76, 224−230. 24. Ache, H. J. Chemistry of the Positron and of Positronium. Angew. Chem., Int. Ed. 1972, 11, 179−199. 25. Levin, B. M.; Shantarovich, V. P.; Agievskii, D. A.; Landau, M. V.; Chukin, G. D. Investigation of Electron-Acceptor Properties of Proton Centers in Type-Y Zeolites by Time Method of Annihilation of Positrons. Kinet. Catal. 1977, 18, 1260−1265. 26. Goldanskii, V. I.; Levin, B. M.; Mokrushin, A. D.; Kaliko, M. A.; Pervushina, M. N. Effect of the Chemical State of the Surface on the Annihilation Characteristics of Positronium in Porous Systems. Dokl. Akad. Nauk. SSSR 1970, 191, 855−858. 27. Goldanskii, V. I.; Levin, B. M.; Pazdzerskii, V. A.; Shantarovich, V. P. Positron Annihilation Processes in Systems Containing Electron Donors and Acceptors. Dokl. Akad. Nauk. SSSR 1976, 228, 1372−1375. 28. Nakanishi, H.; Ujihira, Y. Application of Positron Annihilation to the Characterization of Zeolites. J. Phys. Chem. 1982, 86, 4446−4450.


27


Page 28 of 40

29. Gao, Z.; Yang, X.; Cui, J.; Wang, Y. Application of Positron Annihilation Spectroscopy to the Study of Zeolites. Zeolites 1991, 11, 607−611. 30. Huang, W. F.; Huang, D. C.; Tseng, P. K. Surface Acidity by Positron Annihilation Lineshapes. Catal. Lett. 1994, 26, 269−275. 31. Huang, W. F.; Huang, D. C.; Hung, K. J.; Tseng, P. K. Study of Brønsted Acidity of NaHY Zeolite by Positron Spectroscopy. Catal. Lett. 1996, 40, 31−34. 32. Huang, W. F.; Hung, K. J.; Huang, D. C.; Huang, C. C.; Tseng, P. K. Brønsted Acidic Strength in Y-zeolite by Positron Spectroscopy. Catal. Lett. 1999, 59, 213−216. 33. Huang, W. F.; Hung, K. J.; Huang, D. C. Application of ACAR Spectroscopy to Acidic Catalysts. Mater. Sci. Forum 2001, 363, 263−265. 34. Huang, W. F.; Ochoa, R.; Miranda, R. Positronium-Brønsted Site Interaction in SilicaAlumina Catalysis. Phys. Lett. A 1991, 158, 417−424. 35. Paiziev, A. A. Oxidation of Positronium Atoms on a Surface of Oxidic Catalyst Carrier Containing Acid Centres. Radiat. Phys. Chem. 2000, 58, 787−790. 36. Lari, G. M.; Dapsens, P. Y.; Scholz, D.; Mitchell, S.; Mondelli, C.; Pérez-Ramírez, J. Deactivation Mechanisms of Tin-Zeolites in Biomass Conversions. Green Chem. 2016, 18, 1249−1260. 37. Emeis, C. A. Determination of Integrated Molar Extinction Coefficients for Infrared Absorption Bands of Pyridine Adsorbed on Solid Acid Catalysts. J. Catal. 1993, 141, 347−354. 38. Harris, J. W.; Cordon, M. J.; Di Iorio, J. R.; Vega-Vila, J. C.; Ribeiro, F. H.; Gounder, R. Titration and Quantification of Open and Closed Lewis Acid Sites in Sn-Beta Zeolites that Catalyze Glucose Isomerization. J. Catal. 2016, 335, 141−154.


28

Page 29 of 40 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


39. Pascual-Izarra, C.; Dong, A. W.; Pas, S. J.; Hill, A. J.; Boyd, B. J.; Drummond, C. J. Advanced Fitting Algorithms for Analysing Positron Annihilation Lifetime Spectra. Nucl. Instrum. Methods Phys. Res., Sect. A 2009, 603, 456−466. 40. Kajcsos, Z.; Duplâtre, G.; Liszkay, L.; Billard, I.; Bonnenfant, A.; Azenha, E.; Lázár, K.; Pál-Borbély, G.; Caullet, P.; Patarin, J.; et al. On the Peculiarities of Positron Annihilation Features in Silicalite-1 and Y-zeolites. Radiat. Phys. Chem. 2000, 58, 709−714. 41. Holm, M. S.; Svelle, S.; Joensen, F.; Beato, P.; Christensen, C. H.; Bordiga, S.; Bjørgen, M. Assessing the Acid Properties of Desilicated ZSM-5 by FTIR using CO and 2, 4, 6trimethylpyridine (collidine) as Molecular Probes. Appl. Catal. A: Gen. 2009, 356, 23−30. 42. Milina, M.; Mitchell, S.; Michels, N. L.; Kenvin, J.; Pérez-Ramírez, J. Interdependence Between Porosity, Acidity, and Catalytic Performance in Hierarchical ZSM-5 Zeolites Prepared by Post-Synthetic Modification. J. Catal. 2013, 308, 398−407. 43. Zhu, J.; Ma, L.; Wang, S.-J.; Lou, X.-H. Reaction of Positronium with Proton in Ultra-stable Y Zeolite Studied by Positrons. Chin. Phys. Lett. 2000, 17, 134−136. 44. Ward, J. W. A Spectroscopic Study of the Surface of Zeolite Y: the Adsorption of Pyridine. J. Colloid Interface Sci. 1968, 28, 269−278. 45. Lefrancois, M.; Malbois, G. The Nature of the Acidic Sites on Mordenite: Characterization of Adsorbed Pyridine and Water by Infrared Study. J. Catal. 1971, 20, 350−358. 46. Take, J.; Ueda, T.; Yoneda, Y. An Infrared Study of Acid Sites on Silica-Alumina Catalysts Poisoned with Sodium Hydroxide. Bull. Chem. Soc. Jpn. 1978, 51, 1581−1584. 47. Mogensen, O. E. Spur Reaction Model of Positronium Formation. J. Chem. Phys. 1974, 60, 998−1004.


29


Page 30 of 40

48. Stepanov, S. V.; Byakov, V. M.; Zvezhinskiy, D. S.; Duplâtre, G.; Nurmukhametov, R. R.; Stepanov, P. S. Positronium in a Liquid Phase: Formation, Bubble State and Chemical Reactions. Adv. Phys. Chem. 2012, 2012, 1-17. 49. Tao, S. J. The Formation of Positronium in Molecular Substances. Appl. Phys. 1976, 10, 67−79. 50. Nagashima, I.; Morinaka, Y.; Kurihara, T.; Nagai, Y.; Hyodo, T.; Shidara, T.; Nakahara, K. Origins of Positronium Emitted from SiO2. Phys. Rev. B. 1998, 58, 12676−12679. 51. Levay, B.; Mogensen, O. E. Correlation Between the Inhibition of Positronium Formation by Scavenger Molecules, and Chemical Reaction Rate of Electrons with these Molecules in Nonpolar Liquids. J. Phys. Chem. 1977, 81, 373−377. 52. Wang, C. L.; Kobayashi, Y.; Zheng, W.; Zhang, C.; Nagai, Y.; Hasegawa, M. Positronium Formation in a Polymer Blend of Polyethylene and Chlorinated Polyethylene. Phys. Rev. B. 2001, 63, 064204.


30

Page 31 of 40 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


Table 1. Composition, porosity, and acidity of the zeolites studied. cBrønstedf Vmesoe Smesod 3 −1 2 −1 (cm g ) (m g ) (µmol g−1)

cLewisf (µmol g−1)

Si:Xa (mol mol-1)

Vadsd (cm3 g−1)

Vmicrod (cm3 g−1)

Sil

-

0.18

0.16

0.02

24

0

0

Z80

81

0.18

0.16

0.02

25

74

6

0.17

0.15

0.02

29

144

14

0.18

0.16

0.02

17

0

46

Sample

Z50

b

53 [46] c

Sn-MFI

95

Z50-Na

-

0.17

0.14

0.03

24

0

160g

ZC1000

961

0.19

0.14

0.05

59

19

10

ZC40

39

0.25

0.17

0.08

68

168

18

ZC15

15

0.29

0.14

0.15

76

305

43

ZC-Sn

94c

0.29

0.08

0.21

98

15

58

-

0.24

0.16

0.08

67

102

66g

ZC40-Na-0.5

ZC40-Na-1 0.23 0.14 0.09 70 0 168g a Determined from ICP-OES, where X = Al unless indicated. bDetermined by XPS, cX = Sn. dt-plot method applied to the N2 isotherm. eVmeso = Vpore − Vmicro. fConcentrations of Brønsted and Lewis acid sites determined from IR spectroscopy of adsorbed pyridine. gConcentration of sodium determined from ICP-OES. Average uncertainties are 2a, 0.005 for Vadsd, Vmicrod, Vmesoe, 5 for Smesod, and 10f.


31


Page 32 of 40

Figure 1. (a) Scanning electron micrographs of selected zeolites and acidity characterization including (b) DRIFTS in the OH stretching region, (c) IR spectroscopy of adsorbed pyridine, (d) the propene signal from TPSR of n-propylamine and (e) the ammonia signal from NH3-TPD. The scale bar in all micrographs represents 5 µm.


32

Page 33 of 40 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


Figure 2. Three-dimensional positron implantation profiles estimated for the (a) small and (b) large crystal samples at 3 and 5 keV. Note the profiles have been averaged along one dimension (Lx) to produce two-dimensional representations. P(y,z) has been normalized in each case to the equivalent 3 keV profile in the series.


33


Page 34 of 40

Figure 3. Contour representation of the variation of the total measured intensity of o-Ps with the concentration of Brønsted acid sites and distance of positron implantation.


34

Page 35 of 40 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


Figure 4. The variation in o-Ps (a) intensity and (b) lifetime values with the concentration of Brønsted acid sites for Sil, Z80 and Z50, obtained at 10 keV. Iout represents the sum total of Idef and Ivac.


35


Page 36 of 40

Figure 5. Total measured intensity of o-Ps as a function of the concentration of Brønsted acid centers for the commercial samples. The inset shows the correlation of Vads with cBrønsted for ZC40, ZC40-Na-0.5, and ZC40-Na-1.


36

Page 37 of 40 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


Figure 6. Variation in (a) Io-Ps, (b) Imicro, and (c) Iout with the energy of the incident positrons for Sil, Z80, Z50, Sn-MFI, and Z50-Na. Each intensity set is normalized to the intensity value obtained at 5 keV.


37


Page 38 of 40

Figure 7. Schematic representation of positronium formation and the various annihilation paths in zeolites. After thermalization most positrons directly annihilate with the solid, but a small fraction form positronium in para- or ortho- spin states. The presence of Brønsted and Lewis acid sites have distinct impacts. The former reduce the amount of o-Ps by oxidation of high energy localized positronium (Ilocal) in the micropores, while the latter suppress the formation of positronium. The color scale indicates the estimated kinetic energy of positronium.


38

Page 39 of 40 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


Figure 8. Application of the proposed numerical model to estimate the effect of (a) Lewis acidity on Iout at 5 keV (blank symbols), 10 keV (solid symbols), and (b) Brønsted and Lewis acidity on Imicro at 5 keV. The red lines in (a) indicate the average value from the model (solid = 5 keV, dashed = 10 keV) with the colored shaded regions representing the calculated error (blue = 5 keV, yellow = 10 keV) whilst the black frame in (b) indicate the estimated error.


39


Page 40 of 40

TOC Graphic


40

Acidity Effects in Positron Annihilation Lifetime Spectroscopy of Zeolites

Recommend Documents