Influence of Confinement Effect on Electron Transfers Induced by t

ARTICLE pubs.acs.org/JPCC

Influence of Confinement Effect on Electron Transfers Induced by t-Stilbene Sorption in Medium Pore Acidic Zeolites Matthieu Hureau, Alain Moissette,* Herve Vezin, Claude Bremard, and Maylis Orio Laboratoire de Spectrochimie Infrarouge et Raman UMR-CNRS 8516, Bât. C5, Universite de Lille 1, Sciences et Technologies, 59655 Villeneuve d’Ascq cedex, France

bS Supporting Information ABSTRACT: The mere exposure of trans-stilbene (t-St) to three types of dehydrated medium pore acid zeolites that differ by their pore diameter induces t-St spontaneous ionization in high yield. In situ diffuse reflectance UVvisible, EPR, and Raman spectra recorded over several months highlight the exceptional stability of the charge separated states formed in ferrierite (H-FER), H-MFI, and mordenite (H-MOR). The increase in the pore diameter from H-FER to H-MOR induces different behaviors after radical cation formation. t-St•+ is stabilized for months in the narrow pores of H-FER, whereas in the larger pore H-MFI, relatively fast electron abstraction (hole transfer) takes place from the zeolite framework to create charge transfer complexes. Pulsed EPR experiments were performed using t-St and marked [D12]t-St and [13C2]t-St molecules to reveal the structural environment of the unpaired electrons through the assignment of the couplings with 1H, 2H, 13C, 27Al, and 29Si nuclei.

’ INTRODUCTION Zeolites have been the subject of intense research work because of their particular structures characterized by a controllable internal void space and by their high reactivity.1 The interest for these microporous materials is especially related to their capacity to incorporate molecules by allowing shape selectivity and to the very high internal surface, which is accessible by diffusion to these guest molecules. In particular, zeolites find industrial applications as acid catalysts in oil chemistry, especially for hydrocarbon conversion and gas separation.1,2 Most zeolite properties are known to depend on the material structure, and recently the combined effects of spatial constraints as well as acid site location were studied in medium pore zeolites to clarify their role for alkane conversion.3 Zeolites are microporous crystalline aluminosilicates constituted by cages and channels with molecular dimensions in the range 0.4 to 1.4 nm. The substitution of Al(III) for Si(IV) in the Mn(SiO2)xn(AlO2)n framework requires the presence of charge compensating cations M to ensure the charge neutrality. The charge balancing cations are located in close vicinity of Al framework atoms or are partially bonded to the framework in the case of protons for acid zeolites. These Brønsted acid sites (BAS) correspond to Si-OH-Al bridging hydroxyl groups and are involved in many reactions. Among the parameters that can influence the reactivity, the confinement is of high importance because the guest molecule can exhibit quite different chemical behavior with respect to the interactions in the inner space. Therefore, zeolites can be considered as solid solvents and the judicious choice of these porous materials might allow controlling chemical processes. It should be noted that the confinement effect previously described by Derouane as long-range and attractive interactions cannot be separated from the molecular shape selectivity of zeolites resulting from the short-range interactions due to the interactions of r 2011 American Chemical Society

the occluded molecules with the pore walls.4 Therefore, the understanding and the characterization of hostguest interactions in zeolites are essential to explain the observed reactivity and have already received wide attention.59 In that context, the effects of both high intrachannel electrostatic field due to aluminum and charge balancing cation and confinement on guest molecule give rise to intermolecular noncovalent interactions that can induce many interesting reactions. In particular, these remarkable intrinsic properties can include spontaneous ionization, stabilization of otherwise unstable radicals, polymerization, and hole-catalyzed reactions.1012 Nevertheless, such reactions require specific environment to take place and depend on the appropriateness of the guest molecule with the pore dimension. Therefore, it was demonstrated that spontaneous ionization is a property of the inner surface of zeolites.13 The basis of the spontaneous ionization process of electron donor molecules occluded in zeolites was previously reviewed,10,14 and some of the parameters that influence the reaction are described. The spontaneous ionization depends on the ionization potential of the guest as well as on the ionizing capacity of the host, which is directly related to the Si/Al ratio15 and to the nature of the charge balancing cation.16 The stabilization of charge-separated states within the zeolite internal void space is now well-known and was often observed after sorption of polyaromatics in acidic channeltype zeolites or after photolysis of polyaromatics occluded as intact molecules inside the porous void of nonacidic zeolites. The stabilization of radicals is explained by the compartmentalization of ejected electron trapped away from the initial site of ionization.17 Nevertheless, charge shifting reactions (hole transfer) Received: July 11, 2011 Revised: December 15, 2011 Published: December 17, 2011 1812

dx.doi.org/10.1021/jp2065503 | J. Phys. Chem. C 2012, 116, 1812–1825

The Journal of Physical Chemistry C were reported after the initial ionization and before the final charge recombination15 to create long-lived charge transfer complex.18,19 The stabilization of radical cations highlights the electron acceptor property of zeolite, but the zeolite can also act as an electron donor through AlO4 moieties.20,21 Ways of behaving were shown to depend on the zeolite structure and on the nature of the charge compensating cation, but some data about confinement are still unclear and require additional information. According to these studies, the electron transfer processes observed in acid zeolites suggest that radical cation intermediates could play a crucial role in zeolite hydrocarbon catalytic chemistry.22 Unlike these data, most of the interpretations of the internal reactivity are presently based on proton transfers and do not consider electron transfers. In that context, the present work explores the influence of the confinement effect on the stabilization of charge-separated states and on electron-transfer kinetics by using a probe molecule adsorbed in three different medium-pore-size acid zeolites with various pore diameters but analogous Si/Al ratios (∼10): Ferrierite (H-FER), H-MFI, and Mordenite (H-MOR). The probe molecule is trans-stilbene (E-1,2-diphenylethene, t-St), which is known to induce readily spontaneous ionization in acidic zeolites.19,2326 t-St presents a relatively low ionization potential value (I.P. = 7.65 eV in gas phase) and has the appropriate dimensions to enter the medium pore size zeolites. The mixing of solid t-St crystals and of dehydrated zeolite powder was carried out under argon in the complete absence of solvent. The combined effects of confinement and local electrostatic field on the sorption and charge separation kinetics were monitored as a function of time using in situ continuous and pulsed wave electronic paramagnetic resonance (EPR) spectroscopy, diffuse reflectance UVvisible absorption (DRUVv), and Raman spectroscopy. The location of the guest molecule was also determined by molecular modeling. The consequences of spatial constraints and Brønsted acid site location within zeolite channels on the spontaneous ionization and on the stabilization of charge separated states are thoroughly discussed.

’ EXPERIMENTAL SECTION Materials. As-synthesized MFI samples (Si/Al = 13.5, average particle size ∼1 μm) were obtained according to the template procedure in alkaline medium from VAW aluminum (Schwandorf, Germany). Ferrierite (FER; Si/Al = 10) and mordenite (MOR; Si/Al = 10) were obtained from Zeolyst International (Conhohocken, PA, USA). The NH4+ counterbalancing cations of the aluminated zeolites were completely converted to H+ cations by calcination procedure in flowing air by increasing the temperature up to 723 K and holding for 6 h. trans-Stilbene (t-St, C14H12, Merck-Schuchardt) was purified by sublimation. Pure and dry Ar gas was used. The unit cell (UC) compositions of dehydrated H-MFI, H-FER, and H-MOR zeolites were found to be H6.6(AlO2)6.6 (SiO2)89.4, H3.3(AlO2)3.3(SiO2)32.7, and H4.3(AlO2)4.3(SiO2)43.7, respectively. Powder XRD patterns, 29Si, 27Al MAS NMR, IR, Raman, DRUVv, and EPR spectra of bare zeolites were found to be characteristic of well-crystallized diamagnetic porous compounds with the above formula. Nevertheless, the 27Al MAS NMR experiments carried out on hydrated H-MFI, H-FER, and H-MOR show the presence of small quantities of extraframework hexacoordinated Al species. (See the supporting information of ref 27.) EPR spectroscopic investigations indicate iron impurities

ARTICLE

at trace levels in the commercial zeolite samples. These impurities were not detected by conventional elemental analyses. t-St Sorption Procedure. Weighed amounts of dehydrated zeolite powder were introduced to an evacuable silica cell under dry Ar and heated to 723 K by successive controlled temperature increases. Then, the samples were cooled to room temperature under dry Ar, and weighed amounts of t-St powder were introduced without any solvent to the cell under an inert atmosphere. The powder mixtures were transferred under dry argon into a quartz glass Suprasil cell for optical spectroscopy and into a sealed cylindrical quartz tube for EPR experiments. The samples were kept at 313 K in the dark for long time. The final t-St loadings at equilibrium correspond to 0.4 t-St per UC for FER, 0.5 t-St per UC for MOR, and 1 t-St per UC for MFI. Instrumentation. Sorption Modeling. The molecular modeling of the t-St preferred sorption sites at low coverage in medium pore zeolites H-FER, H-MFI, and H-MOR was performed using Material Studio Modeling package (version 5.0) from Accelrys International. The atomic positions for the zeolite frameworks were obtained from previous X-ray and neutron diffraction determinations of the structures.2831 The structural parameters of t-St were derived from previous structural and theoretical works.3234 In the Monte Carlo (MC) calculations, the Si, Al, O, and H positions were fixed, and the t-St structure was chosen to be rigid. The MC simulations at fixed loading were carried out at 300 K using the conventional Metropolis algorithm and taking into account the nonbonding interactions (EZS). The nonbonding interactions were modeled using the Lennard-Jones (L-J) and Coulombic potentials. The nonbonding L-J force field values and partial charges of Si, Al, O (qi) and C, H (qj) atoms were obtained from previous works.28,35,36 The long-range electrostatic interactions were calculated using the Ewald summation technique. One typical MC run took 1 500 000 steps. From each sorption trajectory, a histogram of the energy distribution for t-St was generated. In the subsequent molecular mechanic (MM) simulations, the zeolite framework was chosen to be rigid and t-St was set up as mobile and flexible. The MM calculations take into account the nonbonding interactions (EZS) and the bond interaction (ES) between t-St atoms. We employ the COMPASS bonding interaction force field with harmonic terms for bond stretching, bond angle bonding, and torsional rotation. The energy minimization of sorbatezeolite energy (ES + EZS) was performed using the conjugate gradient minimization procedure. DRIFTS. The FTIR spectrometer was a Thermo-Nicolet Magna 860 Instrument equipped with liquid-nitrogen-cooled MCT detector. The DRIFT spectra were recorded with 2 cm1 resolution. The key part of the in situ DRIFTS apparatus is a Harrick Scientific Diffuse reflectance attachment “Praying Mantis” combined with cell equipped with CaF2 windows and operating to 700 K under a controlled atmosphere. EPR. The CW X-band EPR spectra of the powders were obtained as a function of time by using a Bruker ELEXYS 580FT spectrometer. The EPR spectra were double-integrated using Bruker software, and the spin concentration was determined relative to a reference standard. This standard sample was a calcined Na6.6(AlO2)6.6(SiO2)89.4 zeolite loaded with various nitroxide amounts (3-carbamoyl-2,2,5,5-tetramethyl-3-pyrrolin-1-yloxy, free radical, 99%, Aldrich). The hyperfine sublevel correlation spectroscopy (2D-HYSCORE) measurements were carried out at 4.2 K and room temperature with the four pulse sequence π/2-τπ/2-t1-π-t2-π/2-τ echo and a four-step phase cycle where the echo is measured as a function of 1813

dx.doi.org/10.1021/jp2065503 |J. Phys. Chem. C 2012, 116, 1812–1825

The Journal of Physical Chemistry C

ARTICLE

t1 and t2 with t1 and t2 being incremented by steps of 16 ns from their initial value. The pulse lengths of the π/2 and π pulses in these experiments were 12 and 32 ns, respectively. The 2DHYSCORE experiments were recorded with different delay τ values (88156 ns) to ensure that no blind spot effects can be observed. Prior to Fourier transformation of the HYSCORE data, the background decay was removed using a polynomial fit and apodized with a Hamming function. DFT Calculations. Quantum chemical calculations were based on density functional theory (DFT) and have been performed with the ORCA program package.37 Geometry optimizations were carried out using the GGA functional BP863840 in combination with the TZV/P41 basis set for all atoms and by taking advantage of the resolution of the identity (RI) approximation in the Split-RI-J variant42 with the appropriate Coulomb fitting sets. Increased integration grids (Grid4 in ORCA convention) and tight SCF convergence criteria were used. For all calculations, solvent effects were accounted for according to the experimental conditions. For that purpose, we used an infinite solvent within the framework of the conductor-like screening (COSMO) dielectric continuum approach.43 EPR parameters were obtained from additional single-point calculations using the hybrid functional B3LYP44,45 and the EPR-II46 basis set. Hyperfine coupling constants were calculated directly from Fermi contact terms and dipolar contributions as the expectation values of the appropriate operator over the spin density. The spinorbit contribution (SOC) of the hyperfine interaction was also calculated, and its isotropic part was added to the Fermi contact term, whereas its anisotropic part was added to the dipolar contribution.47 The MO pictures were generated using natural orbitals, which were localized according to the PipekMezey criterion,48 as implemented in ORCA. DRUVv. After mixing, the system evolution is followed by in situ UVvisible absorption spectrometry between 200 and 900 nm using a Cary 6000i spectrometer. The instrument was equipped with an external integrating sphere (DRA-1800) to study the powdered zeolite samples through diffuse reflectance, and the corresponding bare zeolite was used as reference. The DRUVv spectra were plotted as the KubelkaMunk function FðRÞ ¼ ð1 RÞ2 =2R ¼ K=S

ð1Þ

where R represents the ratio of the diffuse reflectance of the loaded zeolite to that of the dehydrated neat zeolite, K designates an absorption coefficient proportional to the concentration C of chromophore, and S is the scattering coefficient of the powder. F(λ, t) was registered as function of λ (wavelength) at several t (time). Raman Spectrometry. A Bruker RFS 100/S instrument was used as a near-IR FT-Raman spectrometer with a CW Nd:YAG laser at 1064 nm as excitation source. A laser power of 10100 mW was used. The spectra (4000150 cm1) were recorded with a resolution of 2 cm1 using 600 scans. Resonance Raman scattering spectra were collected on a LabRAM spectrometer (Jobin Yvon Horiba Gr.) equipped with a liquid-nitrogencooled charge-coupled device detector. The excitation wavelength of 488 nm was supplied by an argon ion laser. Multivariate Curve Resolution Method. The multivariate curve resolution (MCR) data processing of DRUVv spectral set D(λ,t) was carried out by using the SIMPLISMA (SIMPle-to-use Interactive Self-modeling Mixture Analysis) approach. Primarily, it is necessary to estimate the global rank of D(λ,t) to estimate

the number of pure absorbing species present in the whole data set of the complex mixture. The MCR performs a bilinear decomposition under constraints of multivariate data matrix D(t,λ) according to eq 2 D ¼ C 3 ST þ E

ð2Þ

where C(t) is the data matrix containing time-dependent spectral concentration profiles, S(λ)T is the transpose of the S(λ) matrix formed by the associative pure spectra, and E(t,λ) is the errorrelated matrix containing residual signal after the decomposition. The process aims at minimizing the residues between the original matrix D and the reconstructed matrix D* obtained by multiplication of the spectral concentration profiles by the pure spectra. This algorithm enables us to provide a description of the kinetic process without any a priori chemical or physical knowledge of the data matrix. The estimation of S was obtained by using the SIMPLISMA approach.49 This algorithm allows for the selection of so-called pure variables from the data matrix D. A pure variable is a variable to which one compound of the mixture contributes. All calculations are developed with Matlab 7.7 from the Mathworks.

’ RESULTS Molecular Modeling. MC simulations and energy minimization procedures of the nonbonding interactions between rigid molecules and fixed zeolite framework provide a reasonable structural picture of t-St occluded in acidic H-FER assuming no electron transfer. The framework structure of H-FER contains two perpendicularly intersecting channels. One consists of 10membered rings (10-MR) with dimensions of 0.42 0.54 nm2, and the other is constituted by 8-MR with dimensions of 0.35 0.48 nm2. The MC simulations and subsequent energy minimization carried out at fixed loading (1 t-St/12 UC) indicate that the guest molecule resides in the 10-membered channel within the vicinity of the proton and aluminum atoms of the framework, and the central CdC double bond is oriented along the b direction. The shortest distances between C and H atoms of t-St and Oz and Hz atoms of H-FER pore wall are 0.23 (OzHz 3 3 3 H), 0.27 (OzHz 3 3 3 C), 0.24 (Oz 3 3 3 H), and 0.33 nm (Oz 3 3 3 C), respectively (Table S1 of the Supporting Information). The energy distribution exhibits two sorption sites corresponding either to t-St facially coordinated to the charge compensating proton through a phenyl group or to t-St in interaction with this proton through the central double bond (Figure S1 of the Supporting Information). The main structural role of the zeolite framework appears to constrain the molecular orientation of t-St because of the tight fit of the guest molecule in the pores of the host. The framework structure of MFI contains two types of intersecting channels. Both are formed by rings of 10 oxygen atoms. One channel type is straight and has a nearly circular opening (0.53 0.56 nm2), whereas the other one is sinusoidal and has an elliptical opening (0.51 0.55 nm2). The openings of straight and zigzag channels are sufficiently wide to allow t-St to pass through and diffuse into the void space. The MC simulations of the t-St sorption (1 t-St/UC) as intact molecule in the pores of MFI zeolites provide the expected final structural situation. The shortest distances between C and H atoms of t-St and Oz and Hz atoms of pore wall of H-MFI are 0.30 (OzHz 3 3 3 H), 0.28 (OzHz 3 3 3 C), 0.24 (Oz 3 3 3 H), and 0.32 nm (Oz 3 3 3 C), 1814



ARTICLE

Table 1. Optical DRUVv Data for t-St Occluded in Medium Pore Zeolites (CTC1: Charge Transfer Complex of Type 1; CTC2: Charge Transfer Complex of Type 2) λmax (t-St) λmax (t-St+ 3 ) λmax (CTC1) λmax (CTC2) zeolite

Figure 1. Diffuse reflectance UVvisible absorption (DRUVv) spectra recorded as a function of time after the mixing of solid t-St and H-FER dehydrated at 723 K under argon: (a) 24 h, (b) 3 months, (c) 6 months, and (d) 8 months. The spectra are vertically shifted for clarity.

respectively (Table S1 of the Supporting Information). The preferred sorption site of t-St as intact molecule in H-MFI is the straight channel in close proximity of the charge-balancing H+ with the aromatic ring facially coordinated to the proton. MOR contains two types of channels: a main channel consisting of 12-MR measuring 0.67 0.79 nm2 and a smaller channel is circumscribed by 8-MR measuring 0.26 0.57 nm2. The MC simulations and subsequent energy minimization carried out at fixed loading (1 t-St/12 UC) provide the structure of t-St occluded in the 12-MR channel of MOR in close vicinity of charge compensating proton. The shortest distances between C and H atoms of t-St and Oz and Hz atoms of pore wall of H-MOR are 0.33 (OzHz 3 3 3 H), 0.29 (OzHz 3 3 3 C), 0.26 (Oz 3 3 3 H), and 0.32 nm (Oz 3 3 3 C), respectively (Table S1 of the Supporting Information). DRUVv Absorption Spectroscopy. Several minutes after the mixing of t-St with dehydrated H-FER (Si/Al = 10), the white powder turned light yellowish blue, and the color intensified gradually with time. Figure 1 shows some of the numerous DRUVv spectra recorded as a function of time for 8 months after the mixing of solid t-St with dehydrated Brønsted acidic ferrierite. Prominent bands centered at 310, 481, and 773 nm increased gradually from several minutes to 3 months (Table 1). After 3 months, a new band was clearly observed at 267 nm, and beyond 6 months, two new very weak absorption bands appeared at 567 and 625 nm. To resolve specific absorption of pure species unequivocally, DRUVv spectra set recorded over an extended period of time (8 months) after powder mixing were used for data processing using MCR methods. (See the Experimental Section.) As expected from the matrix rank analysis, three pure spectra were resolved by the MCR treatment with residuals between experimental and calculated values 2νi 13C is estimated at 7.6 MHz from the antidiagonal peak. Indeed, the cross peak splits at 2νi 13C and are centered at A/2 (3.8 MHz). To confirm the interpretation of this later experimental value and its assignment to a 13C-enriched atom, DFT calculations using the ORCA program package were carried out to compute EPR parameters. The calculated hyperfine coupling constants for both carbons and protons are reported in Table 2. According to the data obtained for nonmarked t-St, we assume that the spontaneous ionization of [13C2]t-St in H-FER leads only to the radical cation formation and to the electron trapping in the zeolite. Consequently, the main contribution observed in the HYSCORE spectrum for proton and carbon couplings is due to the Fermi contact term and the anisotropic dipolar term of the hyperfine tensor. From DFT calculations, an Aiso value of 7.3 MHz was obtained for the carbons of the CdC double bond. Moreover, the DFT calculations predict very large 13C coupling and strong anisotropy that cannot be observed in the 2D-HYSCORE feature. The observed signal may also arise from cancellation condition (i.e., Larmor frequency ∼A/2) for which the nuclear modulation is strongest. The compared results between experimental and theoretical values show that they are in the same range and close to the 13C ethylenic carbons of t-St 3 +. A good match is also found between the computed hyperfine couplings for the protons in the surrounding of the CdC double bond and the experimental Aiso values of 13.5 and 7.5 MHz that were previously determined. Additionally, our calculations provided an average Aiso value of 3 MHz for the distal protons that agrees pretty well with the experimentally observed proton pattern (Figure 4). To complete the present analysis, we used DFT calculations to perform the Mulliken spin population analysis of t-St 3 +. The corresponding results are summarized in Table 3 and graphically presented in Figure 5. From Table 3, one can observe that the magnitude of the computed 1H hyperfine couplings are directly related to the individual contribution of the carbon centers to the

with the DRUVv and Raman (see below) spectra recorded as a function of time. To clarify the structural surroundings of the unpaired electrons, we performed pulsed EPR experiments using 2D hyperfine-sublevel correlation experiment (2D-HYSCORE), which enable the proper assignment of various couplings with a large number of nuclei (1H, 13C, 29Si, or 27Al).52 The 2D HYSCORE spectrum recorded at 4.2 K 2 months after the mixing of t-St and H-FER shows a symmetric ridge characterized by five peaks and centered at ν = 14.5 MHz corresponding to the 1H Larmor nuclear frequency (Figure 4a). This symmetric ridge appears in the (+,+), quadrant which indicates that these 1H present a weak hyperfine isotropic coupling Aiso < 2ν (1H). In this spectrum, the 1H cross-peak features are principally dominated by the proton Aiso, and the ridge extent arises from the anisotropic component so that we assume that the t-St 3 +-electron pair is observed. The 1H cross peak ridge coordinates are observed at [8.5, 22]; [11.5, 18]; [14.5, 14.5]; [11.5, 18]; and [22, 8.5] MHz and are principally dominated by the isotropic coupling of 1H of t-St 3 +. The corresponding values of these couplings are estimated at 13.5 and 6.5 MHz, respectively. In Figure 4a, another three-peak ridge centered at ν = 3.9 MHz is observed and is assigned to the 27Al Larmor nuclear frequency. Because of its quadrupolar moment and t-St 3 + free rotation in the channel the 27Al coupling can be observed only at 4.2 K. The appearance of such interaction between unpaired electron and the Al nuclei is the consequence of the charge carrier proximity with the Al nuclei in the framework structure. Nevertheless, this ridge appears only in the (+,+) quadrant, and no contribution is observed in the (,+) quadrant, indicating weak coupling. To discriminate between the contributions of the protons of the Table 2. Calculated 13C and 1H Hyperfine Parameters (MHz) for [13C2]t-St 3 + Individual Contributions of the Isotropic (Aiso) and Anisotropic Dipolar (Ai0 ) Terms to the Calculated HFCsa center

Aiso

Ax0

Ay0

A z0 30.5

C1

7.3

14.4

16.1

C2

7.3

14.4

16.1

30.5

H15

13.5

8.6

0.8

7.8

H14

2.8

1.5

0.1

1.6

H13 H11

3.1 8.3

1.6 5.4

0.1 1.6

1.7 3.7

H9

8.5

5.0

2.2

2.9

H3

13.2

9.7

3.2

6.5

H4

13.3

9.7

3.2

6.5

H20

8.5

5.0

2.2

2.9

H22

8.3

5.4

1.6

3.7

H24

3.1

1.6

0.1

1,7

H25 H26

2.8 13.5

1.5 8.6

0.1 0.8

1,6 7.8

a

Corresponding atom labelling is presented in Figure S5 of the Supporting Information with C1 and C2 being the two 13C ethylenic carbons.

Table 3. Mulliken Spin Population Analysis for [13C2]t-St 3 +a

a

center

C12

spin pop.

0.12

C10

C8

C7

C6

C5

C1

C2

C16

C17

C18

0.1

0.1

0.04

0.14

0.14

0.04

0.1

0.1

C19

C21

C23 0.12

Atom labelling of the corresponding structure is reported in Figure S5 of the Supporting Information. 1818



Figure 5. Spin-density plot (a) and localized single occupied molecular orbital (b) of [13C2]t-St 3 +. Color scheme: C, green; H, white.

total spin density. Such a correlation was expected due to the π nature of the radical cation t-St 3 +53 and further demonstrates the good agreement between theory and experiment. Immediately after exposure of powdered t-St to H-MFI (Si/Al = 13.5), sharp lines superimposed on isotropic broad signal are observed in the g = 2 range of CW-EPR spectra (Figure S6b of the Supporting Information, inset). After 1 day, the sharp lines are probably overlapped by the featureless broad signal. The signal increased with time and reached a maximum after 2 months corresponding to an ionization yield of ∼90%. The CW-EPR spectral set is analogous to that reported previously after exposure of t-St to H-MFI (Si/Al = 27).19 The spectra with sharp lines recorded immediately after the mixings are analogous to the spectra recorded during 2 months after the mixing of t-St with H-FER (Figure S6b of the Supporting Information). The HYSCORE patterns obtained after mixing of t-St or [D12]t-St with H-MFI (Si/Al = 13.5) are analogous to those previously reported for H-MFI (Si/Al = 27)19 and have resemblance to that obtained with H-MFI (Si/Ga = 42).51 All HYSCORE spectra recorded 2 months after mixing exhibit clear change with respect to the HYSCORE patterns recorded several hours after the mixing. As previously reported, the changes of CW-EPR and HYSCORE spectra were attributed to hole transfer from t-St 3 + to the nearest AlO4 moiety of the framework. All spectra recorded after 2 months are relevant of AlO4 3 +-t-St and AlO4 3 entities.19,51 However, it was not possible to determine unequivocally the specific contributions of AlO4 3 +-t-St and AlO4 3 . Immediately after exposure of powdered t-St to dehydrated H-MOR (Si/Al = 10), isotropic broad signal was detected in the g = 2 range of CW-EPR spectra (Figure S6c of the Supporting Information). Contrary to H-MFI and H-FER, no superimposed sharp feature assigned to t-St 3 + was detected. It should be noted that the cooling of the sample to 4.2 K did not reveal any structured pattern. The double integration of the signal indicates that the ionization yield is ∼90% 4 months after the mixing. 2DHYSCORE recorded at 4.2 K is mainly characterized by two major peaks centered at 14.5 MHz corresponding to the 1H Larmor nuclear frequency. The 1H cross peak ridge coordinates are observed at [12, 18]; [18, 12] MHz. Two other peaks centered at ν = 3.9 MHz (27Al) and ν = 2.9 MHz (29Si) are detected when the experiment is carried out at 4.2 K. This symmetric ridge appears in the (+,+) quadrant, indicating weak hyperfine isotropic coupling Aiso < 2ν (1H). The presence of AlO4 3 +-t-St indicates that 1H cross-peak features are principally dominated by the proton Aiso and that the ridge extent arises from the anisotropic component. Moreover, to differentiate between the contributions of 1H from the zeolite and from the t-St, the same experiment was carried out using [D12]t-St. Figure 6b shows the 2D-HYSCORE spectrum obtained after 2 months. This spectrum is very similar to the spectrum recorded after several

ARTICLE

months for [D12]t-St occluded in H-MFI. This spectrum is characterized by two ridges. The first weak ridge is composed by two peaks at coordinates [12, 18]; [18, 12] MHz and centered at ν = 14.5 MHz corresponding to the 1H Larmor nuclear frequency. This weak coupling is attributed to the interactions between AlO4 3 +-t-St and AlO4 3 with 1H of OH groups. The second ridge observed at ν = 2.2 MHz corresponds to the 2 H Larmor nuclear frequency and results from the interaction of AlO4 3 + with the 2H of the [D12]t-St. These values are close to those measured by Goldfarb et al. for 1 H and for 27Al for a copper zeolite system.54 The 1H value is also supported by the coupling observed with deuterated sample where a 6 MHz proton coupling is measured. If we assume that this coupling arises from dipolar coupling then we can approximate an electron nucleus distance of 2.23 Å using single-point dipole approximation. However, in our case, contrary to the copper zeolite system, we have two spin carriers, the radical cation and the ejected electron in the zeolite framework. Then, even we can be sure that the 6 to 6.5 1H MHz coupling arises from zeolite 1 H; it is difficult to determine which is the spin carrier responsible for these coupling because the electron nucleus distance is compatible with both hypothesis. Raman and Resonance Raman Scattering Spectroscopies. The sorption and ionization processes of t-St in the three zeolite topologies were also monitored for several months using FTRaman (λ = 1064 nm) spectroscopy. The spectra recorded as a function of time after mixing t-St and H-FER are reported in Figure 7 and the spectrum of solid t-St is given as reference (Figure 7, spectrum a). The prominent bands observed at 1639, 1591, 1197, and 999 cm1 for solid t-St are assigned to central CdC stretching, aromatic CC stretching, aromatic CCC bending, and aromatic CC stretching, respectively.55,56 The Raman spectra clearly exhibit the slow sorption process and ionization through the total disappearance of the 1639 cm1 line of neutral t-St after 9 months. The MCR procedure applied on the Raman spectra allowed us to resolve two spectra in agreement with the matrix rank analysis. The first spectrum is identical to that of solid t-St and is assigned to the occluded t-St molecule (Figure 8, spectrum a), whereas the second one is assigned to t-St 3 + (Figure 8, spectrum b). Resonance Raman investigations were also performed using the 488.0 nm exciting line within the contour of the visible absorption bands of t-St 3 +. The resulting spectrum exhibits specific resonance Raman bands of t-St 3 + (Figure 8, spectrum g). In terms of wavenumbers, these bands are in good agreement with the extracted t-St 3 + FT Raman spectrum but dramatically differ in terms of relative intensities, especially for the intense Raman lines at 1284 and 1560 cm1 that are only observed by resonance (Table S2 of the Supporting Information). No vibrational modes of the zeolite framework were found to be resonance enhanced. The t-St sorption into H-MFI was also monitored as a function of time using FT-Raman spectrometry. The spectral features of bulk t-St are found to decrease gradually with time concomitantly with the increase in new bands. After 4 months, the t-St characteristic line expected at 1639 cm1 is no longer observed (Figure 9, spectrum f). The rank analysis and MCR treatment provide three pure spectra. The first spectrum is attributed to the neutral t-St. The second spectrum displays bands at 1602, 1585, 1336, and 1190 cm1 and is attributed to the charge-transfer complex of AlO4 3 +-t-St by comparison with the spectrum already observed after photoionization of t-St occluded in Na6.6MFI26 (Figure 8, spectrum c). The third spectrum exhibits bands at 1819



ARTICLE

Figure 6. 2D HYSCORE patterns of (a) t-St@H-MOR 3 + 3 recorded 2 months after the mixing of t-St and dehydrated H-MOR and (b) [D12]t-St@ H-MOR 3 + 3 recorded 2 months after the mixing of [D12]t-St and dehydrated H-MOR. Spectra were recorded at 4.2 K with a τ value of 156 ns.

1622, 1599, 1546, 1513, 1340, 1187, and 996 cm1 and does not correspond to any known species (Figure 8, spectrum d). We tentatively correlate this spectrum to the unknown species previously identified by DRUVv and assigned to a second charge transfer complex of AlO4 3 +-t-St type. No evidence of radical cation t-St 3 + has been observed using off-resonance FT Raman spectrometry (1064 nm). Nevertheless, to demonstrate the initial formation of small amounts of t-St 3 +, resonance Raman investigations were performed using the 488.0 nm laser exciting lines within the contour of the visible absorption bands of t-St 3 +. The resulting spectrum is analogous to the spectrum presented in Figure 8 (spectrum g) and exhibits specific resonance Raman bands of t-St 3 +. This result highlights the presence of small amounts of t-St 3 + (Table S2 of the Supporting Information). The characterization of the system evolution was also carried out using FT-Raman spectrometry after the mixing of t-St with

H-MOR (Figure 10). The spectra show the fast intensity decrease in the t-St band at 1639 cm1, which was not no longer observed after 2 months whereas new bands emerge in the 7001700 cm1 region (Figure 10, spectrum d). The MCR treatment allowed us to resolve three pure spectra corresponding to 3 species. The first spectrum is attributed to neutral t-St. The second one displays bands at 1604, 1587, 1326, and 1186 cm1 and is assigned to AlO4 3 +-t-St moiety by analogy to the spectrum already observed with H-MFI (Figure 8, spectrum e). The third spectrum shows bands at 1621, 1592, 1542, 1511, 1336, 1187, and 994 cm1. It closely resembles the third spectrum resolved for H-MFI and is assigned to a new unknown charge transfer complex of AlO4 3 +-t-St entity (Figure 8, spectrum f). Moreover, it should be noted that the radical cation t-St 3 + was never observed using FT Raman (Table S2 of the Supporting Information). 1820



Figure 7. FT Raman (λ = 1064 nm) spectra recorded as a function of time after the mixing of solid t-St and H-FER dehydrated at 723 K under argon: (a) solid t-St; (b) 48 h; (c) 1 week; (d) 1 month; (e) 2 months; (f) 4 months; (g) 5 months; and (h) 9 months.

Figure 8. Pure Raman spectra resolved using MCR method applied on the FT Raman data set recorded after the mixing of solid t-St and H-FER, H-MFI, and H-MOR. (a) t-St@H-FER; (b) t-St 3 +@H-FER 3 (λex = 1064.0 nm); (c) t-St@H-MFI 3 + 3 of type 1; (d) t-St@H-MFI 3 + 3 of type 2; (e) t-St@H-MOR 3 + 3 of type 1; (f) t-St@H-MOR 3 + 3 of type 2; and (g) resonance Raman spectrum of t-St 3 +@H-FER 3 (λex = 488 nm).

It is worth noting that the total disappearance of the 1639 cm1 line assigned to central CdC stretching of t-St after mixing for the three structures shows evidence of very high ionization and confirms the data obtained using DRUVv and EPR techniques. In addition, the Raman spectra recorded during the system evolution demonstrate the much faster kinetic observed using the DRUVv and EPR techniques in mordenite compared with H-MFI and especially ferrierite.

ARTICLE

Figure 9. FT Raman (λ = 1064 nm) spectra recorded as a function of time after the mixing of solid t-St and H-MFI dehydrated at 723 K under argon: (a) 10 min; (b) 24 h; (c) 1 week; (d) 1 month; (e) 2 months; (f) 4 months; (g) 5 months; and (h) 11 months.

Figure 10. FT Raman (λ = 1064 nm) spectra recorded as a function of time after the mixing of solid t-St and H-MOR dehydrated at 723 K under argon. (a) 10 min; (b) 1 week; (c) 1 month; (d) 2 months; (e) 4 months; (f) 6 months; and (g) 9 months.

modeling through the calculation of preferred sorption sites for the occluded t-St. By contrast, the spectral data demonstrate that the sorption occurs in parallel with spontaneous ionization process in acidic H-FER, H-MFI, and H-MOR after mixing t-St and zeolite under inert atmosphere and without any solvent. Nevertheless, even though the reaction kinetic as well as the system evolution exhibit significant differences according to the zeolite morphology, the sorption and related reactions can be explained by the general following mechanism: After mixing solid t-St and zeolite powder (H-Zeo), sorption takes place essentially in the gas phase by sublimation. t-StðgasÞ þ H-Zeo f t-St@H-Zeo

’ DISCUSSION DRUVv, EPR, and Raman experiment data show that incorporation of t-St takes place in the three medium pore-channeltype zeolites. This feature was expected by the molecular

ð3Þ

Rapidly, ionization of the guest molecules occurs to generate t-St 3 + radical cation (eq 4). t-St@H-Zeo f t-St•þ @H-Zeo• 1821

ð4Þ



ARTICLE

Then, the charge-transfer complex is progressively created through hole transfer (eq 5): t-St•þ @H-Zeo• f t-St@H-Zeo1 • •þ

ð5Þ

The charge-transfer complex of type 1 (t-St@H-Zeo1 3 3 +) directly formed after hole transfer appears to evolve partially by reorganization to a charge transfer complex of type 2 (t-St@H-Zeo2 3 3 +) according to eq 6 t-St@H-Zeo1 • •þ f t-St@H-Zeo2 • •þ 3+

3

ð6Þ

Note that the t-St @H-Zeo moiety is very stable in H-FER, whereas it is only a fast transient species in H-MOR. By contrast, the charge-transfer complexes of type 1 and 2 are observed in H-MFI and H-MOR, whereas type 1 is hardly detected within H-FER. These different behaviors are discussed below. In particular, the first part is focused on the sorption and spontaneous ionization processes, whereas the second part describes the effects of the confinement on the kinetics of the charge transfers and on the fate of the charge separated states. In addition, it should be noted that t-St protonation was not observed after sorption in the H-FER, H-MFI, and H-MOR acid zeolites. Sorption of t-St into Void Space of Channel Zeolites. The sorption process in zeolites has been fully investigated by experimental and theoretical methods and recently described as resulting from a sequence of successive steps.57,58 First, collisions occur between molecules in the gas phase and the external surface of the host material. Then, molecules are trapped via a weak coupling on the external surface and thus can readily move to the pore entry. Finally, the guest molecule diffusion takes place into the pores and the molecules can reach active sorption sites where ionization occurs. The diffusion of aromatics as intact molecules in zeolite porous void was extensively described.59,60 Therefore, the close match between the guest molecule and the walls of the channel might dramatically slow down the diffusion process. The t-St diffusion rates as intact molecule in the channel are expected to increase in the following order FER< MFI< MOR due to the diameter of 10-MR and 12 MR channels. However, in the present study, the experimental data show that sorption occurs in parallel with ionization process, and consequently the diffusion into the pores must account for the migration of charged species t-St 3 + and ejected electron. The diffusion of such radicals is drastically hindered by electrostatic interactions. Indeed, the charged centers induce deep potential wells that are able to trap mobile ions and drastically slow their motion for some time.61 Therefore, the migration of the charged moieties is considerably slower than the diffusion of uncharged species. Spontaneous Ionization. The EPR, DRUVv, and Raman spectra provide convergent data that show evidence of t-St spontaneous ionization upon incorporation in Brønsted acidic medium pore zeolites. The t-St ionization induces radical cation formation in the three zeolites, but the fate of t-St 3 + depends dramatically on the zeolite morphology. The ability of molecules to ionize spontaneously upon mixing with zeolites is now well known and is a property of the inner surface of channel. The spontaneous ionization depends both on the ionization potential values of the guest molecule and on the ionizing power of the host. The experimental data show that the relatively low ionization potential in the gas phase of t-St (I.P. = 7.65 eV) undoubtedly induces molecule ionization in high yield within the three zeolite structures that are simply dehydrated up to 723 K. Lobo

and coauthors have shown that the thermal treatment carried out under an inert atmosphere up to 700 K does not notably generate additional Lewis acid site (LAS) but only releases water.62 The ionizing ability of the host is associated with the acidity strength of the zeolite and with the polarization energy at the sorption AlO(H+)-Si site. In the acid zeolites used here, the high electrostatic field depends not only on the charge compensating proton but also on the Si/Al ratio and on the confinement effect. The small H+ ions are known to induce a strong field in their proximity,21 and the three zeolites present similar and relatively high aluminum content even though the distribution changes from one zeolite to the other. A strong synergy between the polarization energy and the confinement was already evoked; therefore, the accessibility of the acid sites had to be considered. Indeed, even if the total amount of acid and polarizing sites is similar in each zeolite (Si/Al ≈ 10), the accessibility to these acid sites probably has an effect on the ionization. In H-MFI, all acid sites belong to the 10-membered rings (10-MR) straight and sinusoidal channels, and it is assumed that ∼60% of the sites are located either in the straight channel or at the intersection with sinusoidal channel and therefore are easily accessible for t-St.63,64 In H-FER constituted by a framework of 10-MR and 8-MR channels, t-St is only able to penetrate into the 10-MR channels containing ∼40% of the acid sites.65 For H-MOR, ∼40% of the OH groups are present within the 12-MR channels accessible to t-St and 60% in the nonaccessible 8-MR channels.63 These features show that it might be difficult to try to compare directly the high ionization observed within the three zeolites because of various acid site accessibility, although the entire number of acid sites is nearly identical (Si/Al ≈ 10) in each zeolite. The possible role of a low amount of extra-framework aluminum in the spontaneous ionization cannot also be discarded but is assumed to be weak according to previous studies.14,66 In these acid zeolites, the ejected electron is probably trapped on the oxygen atoms of the AlO4H 3 group, and the compartmentalization of electrons trapped away from the initial site of t-St ionization is put forward to explain the slow recombination. In particular, pulsed EPR analyses clearly show the coupling between t-St 3 + and electron pair and 27 Al and 1 H atoms of Al-(OH)-Si bridges of H-FER framework. This interaction indicates that t-St 3 + and electron are located in close proximity of the bridges. Note that H 3 formation was never observed despite the short distance between electron and proton. The lack of convincing evidence of the EPR observation of H-Zeo 3 moiety is a recurrent problem in the study of spontaneous ionization and photoionization of electron donor molecule upon incorporation in zeolites. In the present work, there is no direct specific EPR evidence of the ejected electron, which is indeed the counterpart of t-St 3 +. As reported by Garcia and Roth,10 direct EPR evidence of the fate of the electron transferred to the zeolite in the oxidation step is available only in rare and special cases. Typically, the captured electrons are EPR silent, suggesting that they may be highly delocalized. However, note that the problem of H-Zeo 3 EPR signal was properly resolved after the exposure of dry anthracene to acidic H-MFI under argon and at room temperature.67 In this previous work, pulse EPR techniques were performed after complete anthracene spontaneous ionization upon sorption. The excitation of the CW-EPR complex signal using a two-pulse echo sequence allowed the identification of two moieties resulting from the ionization. The presence in the sample of two nonequivalent chemical magnetic 1822


The Journal of Physical Chemistry C species was demonstrated by performing spinlattice T1 measurements using free induction decay (FID) inversion recovery and echo inversion recovery. The FID signal was typical of the anthracene radical cation in accurate agreement with DRUVv absorption and Raman spectroscopy, whereas the echo signal was assigned to the electron trapped in the zeolite framework. In addition, the differences in T1 values allowed us to characterize separately radical cation and trapped electron by spinecho correlation spectroscopy. Upon heating of the radical cation electron pair, the hyperfine structure of the CW spectra disappeared, and only a featureless signal was obtained after 1 h at 473 K. The hyperfine splitting reappeared upon cooling to room temperature. An identical phenomenon was observed by DRUVv spectrometry. The broad EPR signal was assigned to occluded electronhole pair with respect to the UVvisible absorption spectrum. The CW-EPR spectrum of the electronhole pair was found to be strictly identical to that of the ejected electron isolated by modifying the relaxation times. Therefore, it is important to note that we were not able to differentiate between the EPR signal of the ejected electron and that of the electronic hole. Nevertheless, the radical cation signature is quite different from that of the trapped electrons. The two-pulse echo experiment recorded for the electronhole pair showed the complete disappearance of the FID process in accordance with the disappearance of radical cation. Moreover, T1 echo measurements yield the same value as that measured for the trapped electron of the radical cationelectron, pair demonstrating identical relaxation times for the trapped electron of the radical cationelectron pair and for unpaired electrons involved in the electronhole pair. Such sophisticated pulsed EPR experiments were also carried out in the present work but, unfortunately, did not come off. Charge Transfer within Void Space of Acidic Zeolites. The experimental data obtained throughout several months highlight the exceptional stability of the charge-separated states generated into the three porous frameworks. However, the increase in the pore diameter from H-FER to H-MOR induces significant changes in the system evolution, and three different behaviors are observed after radical cation formation. The high confinement imposed by the narrow channels of H-FER makes the diffusion of the ionized species extremely slow; therefore, the t-St 3 + 3 3 3 AlO4H 3 moiety is stabilized in high yield within H-FER for several months. Note that a very weak hole transfer takes place after 6 months, as shown by the appearance of the characteristic features of the charge transfer band at 567 and 625 nm. By contrast, in H-MFI, the t-St 3 + concentration decreased rapidly by hole transfer from the electron deficient t-St 3 + to another AlO4H group in close proximity and led to an electron hole generation like t-St 3 3 3 AlO4H 3 +. The subsequent hole formation occurring after t-St 3 + depends on the oxidizing power of the radical cation (1.75 V versus SCE in solution) with respect to the AlO4H electron donor site.68 The Eox potential can explain the electron abstraction from AlO4H in close proximity of t-St 3 + to generate the charge-transfer complex; however, it should be noted that the electron donor ability of zeolite upon interaction with alkyl bromide molecules was also shown to occur from the oxygen atoms between two Si atoms.69 In H-MOR, the same reaction sequence is observed, but t-St 3 + is hardly detected because of very fast hole transfer and charge transfer complex formation. Therefore, the charge-transfer complex is observed in H-MFI and H-MOR as well as in H-FER but in very low amount. Indeed, the very slow electron transfer imposed by the slow diffusion of the ionized species and by the high

ARTICLE

confinement effect in H-FER dramatically slows down the electronhole pair formation in this zeolite. The charge-transfer complex is characterized by an intense color and displays characteristic spectral features in the visible spectral range. These bands are tentatively assigned to an electronic transition involving the hole AlO4H 3 + and the t-St molecule in close proximity. The involvement of t-St in the charge transfer is demonstrated by the observation of a vibronic structure with maxima at ca. 565 and 622 nm (H-MFI) separated by ∼1600 cm1 and characteristic of the CdC stretching of the occluded t-St. Note that analogous features were already reported after ionization and subsequent formation of charge transfer complex for diphenylacetlylene, biphenyl, tetracene, and naphthalene incorporated in H-MFI.18,7072 The supplementary charge-transfer complex is probably related to an internal reorganization of the system. In that context, it might correspond to two types of interfacial π interactions associated with the t-St 3 3 3 AlO4H+ 3 moiety: either between H+ and phenyl ring or between H+ and the central double bond of the molecule. The system reorganization required 6 months to take place in H-MFI, whereas it was observed after 2 to 3 months in H-MOR. Effects of Spatial Constraints on Intrazeolite Reactivity. In the present case, the role of the pore diameters on the kinetic of the sorption, ionization, and charge transfer is clearly identified. In particular, the high confinement within H-FER stabilizes the radical cation in high yield, whereas the kinetic of hole transfer to generate charge transfer complexes increases dramatically from H-FER to H-MOR. The higher mobility in the mordenite 10-MR channels and in the H-MFI channels is assumed to enhance the probability of electron transfer. As discussed above, a high synergy exists between the polarization energy and the confinement, and thus the close match between the occluded molecule and the accessible sorption acid sites has to be considered. The polarization energy is closely linked to the local electrostatic field strength at the t-St sorption site. The local electrostatic field depends on the distance between the charge balancing cation and the incorporated t-St via noncovalent interactions. These interactions are in the short-range (van der Waals and electrostatic) and in the long-range (electrostatic). This distance is assumed to decrease significantly according to the spatial constraints from H-MOR to H-FER, inducing lower local electrostatic field in H-MOR than in H-FER (Table S1 of the Supporting Information). These experimental data related to the electron transfer process might also be tentatively interpreted using the nonadiabatic electronic theory developed by Marcus.73 Unfortunately, no information allowing the determination of parameters of the famous equation was obtained here, and no valuable conclusion can be obtained.

’ CONCLUSIONS Diffuse reflectance UVvisible, EPR, and Raman experiments show evidence of the t-St sorption and spontaneous ionization in the straight channels of H-FER, H-MFI, and H-MOR simply dehydrated up to 723 K. The experimental data obtained by this multiple approach demonstrate the key role of the combined effects of confinement and local electrostatic field for generation and stabilization of very long-lived charge separated states. The stability of these species is explained by the restricted mobility of the charged species in the narrow channels as well as the compartmentalization of the trapped electron away from the initial site of ionization. Nevertheless, various electron-transfer 1823


The Journal of Physical Chemistry C behaviors are reported as a function of the zeolite structure. t-St radical cation is stabilized for months in the narrow pores of H-FER, whereas t-St•+ evolves to a charge-transfer complex through hole transfer in H-MFI and in H-MOR. The extremely slow system evolution observed in H-FER is explained by the high resulting electric field and the tighter fit of t-St•+ in the 10-MR pore.

’ ASSOCIATED CONTENT

bS

Supporting Information. MC simulations of t-St occluded in the straight channel of H-FER, DRUVv absorption spectra of pure species resolved by MCR chemometric procedure from the spectral set recorded as a function of time after mixing solid t-St and dehydrated H-FER, DRUVv absorption spectra recorded after the mixing of solid t-St and K-FER dehydrated at 723 K under argon, atom labelling of the optimized structure of [13C2]t-St 3 + occluded in H-FER, EPR cw signals recorded at various times after mixing t-St and H-zeolite, calculated geometric parameters of t-stilbene (t-St) occluded in channel of medium pore zeolites, and characteristic lines observed on the Raman spectra for t-St, t-St+ 3 , charge transfer complex of type 1 (CTC1), and charge transfer complex of type 2 (CTC2) after mixing t-St and medium pore zeolites. This material is available free of charge via the Internet at http://pubs.acs.org.

’ REFERENCES (1) Corma, A. Chem. Rev. 1995, 95, 559–614. (2) Bevilacqua, M.; Montanari, T.; Finocchio, E.; Busca, G. Catal. Today 2006, 116, 132–142. (3) Gounder, R.; Iglesia, E. J. Am. Chem. Soc. 2009, 131, 1958–1971. (4) Derouane, E. G. J. Mol. Catal. A: Chem. 1998, 134, 29–45. (5) Rammamurthy, V. J. Photochem. Photobiol., C 2000, 1, 145–166. (6) Zhang, G.; Thomas, J. K. J. Phys. Chem. B 2003, 107, 7254–7260. (7) Toulhoat, H.; Raybaud, P.; Benazzi, E. J. Catal. 2004, 221, 500–509. (8) Boronat, M.; Concepcion, P.; Corma, A.; Navarro, M. T.; Renz, M.; Valencia, S. Phys. Chem. Chem. Phys. 2009, 11, 2876–2884. (9) Sastre, G.; Corma, A. J. Mol. Catal. A 2009, 305, 3–7. (10) Garcia, H.; Roth, H. D. Chem. Rev. 2002, 102, 3947–4007. (11) Cardin, D. J. Adv. Mater. 2002, 14, 553–563. (12) Bhan, A.; Iglesia, E. Acc. Chem. Res. 2008, 41, 559–567. (13) Moissette, A.; Marquis, S.; Gener, I.; Bremard, C. Phys. Chem. Chem. Phys. 2002, 4, 5690–5696. (14) Marquis, S.; Moissette, A.; Vezin, H.; Bremard, C. C.R. Chimie 2005, 8, 419–440. (15) Hureau, M.; Moissette, A.; Marquis, S. Phys. Chem. Chem. Phys. 2009, 11, 6299–6307. (16) Marquis, S.; Moissette, A.; Vezin, H.; Bremard, C. J. Phys. Chem. B 2005, 109, 3723–3726. (17) O’Neil, M.; Cozens, F. L.; Schepp, N. P. J. Phys. Chem. B 2001, 105, 12746–12758. (18) Marquis, S.; Moissette, A.; Hureau, M.; Vezin, H.; Bremard, C. J. Phys. Chem. C 2007, 111, 17346–17356. (19) Vezin, H.; Moissette, A.; Hureau, M.; Bremard, C. ChemPhysChem 2006, 7, 2474–2477. (20) Yoon, K. B. Chem. Rev. 1993, 93, 321–339. (21) Hashimoto, S. J. Photochem. Photobiol., C 2003, 4, 19–49. (22) McVicker, G. B.; Kramer, G. M.; Ziemiak, J. J. J. Catal. 1983, 83, 286–300. (23) Ramamurthy, V.; Caspar, J. V.; Corbin, D. R. J. Am. Chem. Soc. 1991, 113, 594–600. (24) Kojima, M.; Takeya, H.; Kuriyama, Y.; Oishi, S. Chem. Lett. 1997, 997–998.

ARTICLE

(25) Bonora, M.; Brustolon, M.; Casagrande, M.; Biglino, D.; Itagaki, Y.; Segre, U.; Lenarda, M. Phys. Chem. Chem. Phys. 2000, 2, 4823–4828. (26) Moissette, A.; Bremard, C.; Hureau, M.; Vezin, H. J. Phys. Chem. C 2007, 111, 2310–2317. (27) Belhadj, F.; Moissette, A.; Bremard, C.; Hureau, M.; Derriche, Z. ChemPhysChem 2011, 12, 1378–1388. (28) Beerdsen, E.; Dubbeldam, D.; Smit, B.; Vlugt, T. J.; Calero, S. J. Phys. Chem. B 2003, 107, 12088–12093. (29) Bull, I.; Lightfoot, P.; Villaescusa, L. A.; Bull, L. M.; Gover, R. K. B.; Evans, J. S. O.; Morris, R. E. J. Am. Chem. Soc. 2003, 125, 4342–4349. (30) Martucci, A.; Alberti, A.; Cruciani, G.; Radaelli, P.; Ciambelli, P.; Rapacciulo, M. Microporous Mesoporous Mater. 1999, 30, 95–101. (31) Martucci, A.; Cruciani, G.; Alberti, A.; Ritter, C.; Ciambelli, P.; Rapacciuolo, M. Microporous Mesoporous Mater. 2000, 3536, 405–412. (32) Baehtz, C.; Fuess, H. Phys. Chem. Chem. Phys. 2002, 4, 4543– 4548. (33) Morin, C.; Simon, D.; Sautet, P. J. Phys. Chem. B 2004, 108, 12084–12091. (34) Wiberg, K. B. J. Org. Chem. 1997, 62, 5720–5727. (35) Gener, I.; Buntinx, G.; Bremard, C. Microporous Mesoporous Mater. 2000, 41, 253–268. (36) Olson, D. H.; Khosrovani, N.; Peters, A. W.; Toby, B. H. J. Phys. Chem. B 2000, 104, 4844–4848. (37) Neese, F. ORCA An Ab Initio, Density Functional and Semiempirical Program Package, version 2.8; Universit€at Bonn: Bonn, Germany, 2007. (38) Perdew, J. P. Phys. Rev. B 1986, 33, 8822–8824. (39) Perdew, J. P. Phys. Rev. B 1986, 34, 7406. (40) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (41) Sch€afer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100, 5829–5835. (42) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057–1065. (43) Klamt, A.; Sch€u€urmann, G. J. Chem. Soc., Perkin Trans. 2 1993, 799–805. (44) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (45) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789. (46) Barone, V. Recent Advances in Density Functional Methods, Part I; Chong, D. P., Ed.; World Scientific: Singapore, 1996. (47) Neese, F. J. Chem. Phys. 2003, 118, 3939–3948. (48) Pipek, J.; Mezey, P. G. J. Chem. Phys. 1989, 90, 4916–4926. (49) Windig, W.; Antalek, B.; Lippert, J. L.; Batonneau, Y.; Bremard, C. Anal. Chem. 2002, 74, 1371–1379. (50) Gessner, F.; Scaiano, J. C. J. Photochem. Photobiol., A 1992, 67, 91–100. (51) Moissette, A.; Lobo, R. F.; Vezin, H.; Al-Majnouni, K.; Bremard, C. J. Phys. Chem. C 2010, 114, 10280–10290. (52) H€ofer, P.; Grupp, A.; Nebenf€urh, H.; Mehring, M. Chem. Phys. Lett. 1986, 132, 279–282. (53) Improta, R.; Barone, V. Chem. Rev. 2004, 104, 1231–1253. (54) Carl, P. J.; Vaughan, D. E. W.; Goldfarb, D. J. Phys. Chem. B 2002, 106, 5428–5437. (55) Schneider, S.; Scharnagi, C.; Bug, R.; Baranovic, G.; Meic, Z. J. Phys. Chem. 1992, 96, 9748–9759. (56) Hub, W.; Kl€uter, U.; Schneider, S.; D€orr, F.; Oxman, J. D.; Lewis, F. D. J. Phys. Chem. 1984, 88, 2308–2315. (57) Reitmeier, S. J. Angew. Chem., Int. Ed. 2009, 48, 533–538. (58) Reitmeier, S. J. J. Phys. Chem. C 2008, 112, 2538–2544. (59) Hashimoto, S.; Kiuchi, J. J. Phys. Chem. B 2003, 107, 9763– 9773. (60) Hashimoto, S. J. Phys. Chem. Lett. 2011, 2, 509–519. (61) Krager, J. Diffusion in Zeolites and Other Microporous Solids; Wiley: New York, 1992. (62) Nash, M. J.; Shough, A. M.; Fickel, D. W.; Dore, D. J.; Lobo, R. F. J. Am. Chem. Soc. 2008, 130, 2460–2462. (63) Luchez, F.; Carre, S.; Moissette, A.; Poizat, O. RSC Adv. 2011, 1, 341–350. 1824



ARTICLE

(64) Bhan, A.; Allian, A. D.; Sunley, G. J.; Law, D. J.; Iglesia, E. J. Am. Chem. Soc. 2007, 129, 4919–4924. (65) Domokos, L.; Lefferts, L.; Seshan, K.; Lercher, J. A. J. Mol. Catal. A: Chem. 2000, 162, 147–157. (66) Leu, T. M.; Roduner, E. J. Catal. 2004, 228, 397–404. (67) Vezin, H.; Moissette, A.; Bremard, C. Angew. Chem., Int. Ed. 2003, 42, 5587–5591. (68) Samori, S.; Hara, M.; Tojo, S.; Fujitsuka, M.; Majima, T. J. Photochem. Photobiol., A 2006, 179, 115–124. (69) Li, L.; Zhou, X. S.; Li, G. D.; Pan, X. L.; Chen, J. S. Angew. Chem., Int. Ed. 2009, 48, 6678–6682. (70) Belhadj, F.; Moissette, A.; Bremard, C. J. Photochem. Photobiol., A 2009, 208, 203–212. (71) Gener, I.; Moissette, A.; Bremard, C. Chem. Commun. 2000, 1563–1564. (72) Moissette, A.; Vezin, H.; Gener, I.; Bremard, C. J. Phys. Chem. B 2003, 107, 8935–8945. (73) Marcus, R. A.; Sutin, N. Biochim. Biophys. Acta 1985, 811, 265–322.

1825


Influence of Confinement Effect on Electron Transfers Induced by t

Recommend Documents