Breaking Linear Scaling Relationships with Secondary Interactions in Confined Space: A Case
Study of Methane Oxidation by Fe/ZSM-5 Zeolite
Item Type Article
Authors Szécsényi, Ágnes;Khramenkova, Elena;Chernyshov, Ivan Yu;Li, Guanna;Gascon, Jorge;Pidko, Evgeny A.
Citation Szécsényi, Á., Khramenkova, E., Chernyshov, I. Y., Li, G., Gascon, J., & Pidko, E. A. (2019). Breaking Linear Scaling Relationships with Secondary Interactions in Confined Space: A Case Study of Methane Oxidation by Fe/ZSM-5 Zeolite. ACS Catalysis, 9(10), 9276–9284. doi:10.1021/acscatal.9b01914
Eprint version Publisher's Version/PDF
DOI 10.1021/acscatal.9b01914
Publisher American Chemical Society (ACS)
Journal ACS Catalysis
Rights This is an open access article published under a Creative
Commons Non-Commercial No Derivative Works (CC-BY-NC-ND) Attribution License, which permits copying and redistribution of the article, and creation of adaptations, all for non-commercial purposes.
Download date 2024-01-12 21:09:35
Item License https://pubs.acs.org/page/policy/
authorchoice_ccbyncnd_termsofuse.html Link to Item http://hdl.handle.net/10754/659069
S
UPPORTINGI
NFORMATIONBreaking Linear Scaling Relationships with Secondary Interactions in Confined Space: a Case Study of Methane Oxidation by Fe/ZSM-5 Zeolite
Ágnes Szécsényi1,2, Elena Khramenkova3,‡, Ivan Yu. Chernyshov3, Guanna Li1, Jorge Gascon1,2, Evgeny A. Pidko*1,3
1 Department of Chemical Engineering, Delft University of Technology, Van der Maasweg 9, 2629 HZ Delft, The Netherlands
2 King Abdullah University of Science and Technology, KAUST Catalysis Center, Advanced Catalytic Materials, Thuwal 23955, Saudi Arabia
3 TheoMAT group, ChemBio Cluster, ITMO University, Lomonosova str. 9, Saint Petersburg, 191002, Russian Federation
‡ Current address: Inorganic Systems Engineering group, Department of Chemical Engineering, Delft University of Technology, Van der Maasweg 9, 2629 HZ Delft, The Netherlands
E-mail: [email protected]
Figure S1: The original [(H2O)2-Fe(III)-(μO)2-Fe(III)-(H2O)2]2+ cluster deposited in the γ position of the zeolite.
Figure S2: Zeolite positions “Site” considered in the paper. The colour code corresponds to the code used in Figure 1 C. α, δ and ε sites are positioned in the main channel, while β and γ sites are located in the sinusoidal channel. In each ring two lattice Si atoms were exchanged with Al to compensate for the charge of the extraframework di-Fe complex.
Figure S3: Frontier Kohn-Sham orbitals of [FeO(H2O)5]2+ cluster in A) S=5/2 and B) S=3/2 spin states. Unpaired electrons are shown in red, left are the α, and right the β orbitals.
The orbitals mentioned in the main text are: Fe(dz2)+O(pz) is shown here as 3σ*α, and Fe(dxz/yz)-O(px/y) is shown here as π*x/yβ orbital. C) The energy diagram of the C-H bond dissociation step (in kcal/mol) and the transition state structures of the σ and π channel. In this case the π channel is the result of the different electronic structure induced by the change in spin state. However as we can see on part A), the π*x/yβ orbital of this structure is not much higher in energy than the 3σ*α. This indicates that in case of steric constraint the reaction might proceed via the π channel. The figure is reproduced from Belanzoni et al.[S1]
Figure S4: Confined TS over γ1*.
Figure S5: Confined site-H + *CH3 for γ1* and γ5.
Figure S6: TSs for γ1, γ6*, and δ1. The secondary C…H interaction with the closest H atom and the respective interatomic distances are depicted in the figure.
BEP relationship
The similar procedure can be performed on the BEP relationship as on the two other linear scaling relationships presented in the main text. The activation energy and reaction energy are calculated as follows:
Reaction energy:
ΔER=E(site-H+•CH3)-E(site+CH4) (S1) Activation energy:
ΔEa=E(TS)-E(site+CH4) (S2)
Thus, the reference point changes from site to site+CH4. This means, that all the phenomena presented in the main text have the effect on the reaction and activation energy. Except for the H-bond forming after *CH3 desorption, as E(site) is not participating in the calculation of the energies. However, there is one more point that needs to be addressed, and that is the confinement of the reference state: site+CH4. According to our simplified qualitative model, presented in Figure 1, this does not result in the deviation from the linear, because if the initial state (IS) is destabilised, both the reaction and activation energy decreases. For this reason, it is marked with yellow in Table S1. However, in our case the steepness of the slope is ~0.5, which means that the same decrease in the reaction energy as in the activation energy will put the point in the blue zone.
Figure S7: The BEP linear scaling relationship.
Table S1:The effects stabilising and destabilising the intermediates and the deviation from the BEP linear scaling relationship. The table is similar to Table 1 in the main text. The numbers in brackets mark the adsorption energy in kJ/mol. R2 and adjusted R2 values of the multiple regression are 0.65 and 0.60, respectively.
Location
Confinement Flexibility and Multifunctionality Energy deviation (kJ/mol) from
nA,TS dA,FS
HB break. @
TS
HB form
@ ‘site- H+*CH3’
OH- coopera-
tivity
BEP
α1 2 2.24 - - - -1
α2 7 2.30 - - - 12
α3 3 2.06 - - - 0
β5 5 2.15 - - - -1
β6 5 2.08 - - + -15
γ1 4 2.08 - - + -14
γ1* 13 1.64 - - - -33
γ1** 5 1.96 - + + 2
γ4 12 2.48 - - - 5
γ4* 8 1.77 - - - -1
γ4** 6 1.97 - + + 18
γ5 11 1.81 - - - -4
γ6 12 1.89 - - + -26
γ6* 6 2.10 - - + -5
γ7 12 2.27 - - - 17
γ7* 9 2.52 - - - 27
δ1 3 2.04 - - + -7
δ2 1 1.86 - - - 1
δ3 2 2.07 - - - 3
ε1 5 2.39 - - - 2
ε2 1 2.17 - - - -5
ε3 5 2.71 + - - 21
o2 4 1.85 - - + -14
o3 3 2.09 + - - 18
BEP regression
coefficients — 34.4(***) — 24.9(**) -12.0(**)
Free energy calculation
Entropy plays an important role in the thermodynamics of the adsorption and desorption steps. However, it was found that the relative entropy contribution of the reaction steps in the adsorbed states is relatively minor, also in zeolites[S2]. Recent works successfully employ different approximation schemes allowing extrapolating the free energy contributions among varied classes of materials (see e.g. ref.[S3,S4]). For example, by examining the work of Bogaard and Olsbye[S2] on ethylene dimerization over Ni-modified zeolite catalysts, one can conclude that the relative Gibbs free energy (∆G) can be reasonably well approximated by correcting the relative electronic energy ∆E(el) with a value of ∆G-∆E(el) of the adsorption step, as represented in Figure S8. ∆E(el) and ∆G were obtained from Table S6 of the Supporting Information of the previously mentioned article, while ∆E(el,corr) was calculated based on the previously mentioned method by approximating the ∆G-∆E(el) adsorption/desorption step to be 70 kJ/mol for ethylene and 83 kJ/mol for butylene for all three sites.
Figure S8: Reaction profile of ethylene dimerization. The corrected electronic energy (∆E(el,corr)) is calculated from the electronic energy (∆E(el)) by adding or subtracting 70 and 83 kJ/mol corresponding to ethylene and butylene adsorption/desorption steps.
To demonstrate that this approximation is also valid to our system we also performed frequency analysis on one C-H bond reaction step to calculate the Gibbs free energies within the harmonic approximation. Frequency analysis of the stationary points was performed by means of the finite difference method as implemented in VASP. Small displacements (0.02 Å) were used to estimate the numerical Hessian matrix. The analysis only includes the atoms of the active site and adsorbate(s). The temperature was 50°C consistent with the experiments. Figure S9 shows the reaction profile of γ1**. ∆E(el,corr) was calculated by adding the adsorption energy, 36 kJ/mol, of methane to the TS and FS.
Figure S9: Reaction profile of γ1**.
In our previous work[S5] this analysis has been extended to methane activation over the same site following heterogeneous and Fenton-type methane activation mechanism.
There the average of ∆G-∆E(el) for the methane adsorption step in for reaction channelsincluding γ1** was calculated to be 43 kJ/mol. The corrected electronic energy was computed in the following way: ∆E(el,corr)=∆E(el)+x∗43kJ/mol, where x is the number of the previously adsorbed methane molecules (0 for the first structure and 1 for the rest).
The presented results demonstrated, that the ∆E(el,corr) approximates ∆G reasonably well, which means that the entropic effect is uniform for all reactions. Consequently, the reaction profile changes the same way, which means that the relative electronic energies can be successfully employed to support the discussion of reaction mechanisms and relative reactivity of different sites.
Supplementary references:
[S1] Belanzoni,P.; Bernasconi, L.; Baerends, E. J., O2 Activation in a Dinuclear Fe(II)/EDTA Complex: Spin Surface Crossing As a Route to Highly Reactive Fe(IV)oxo Species. J. Phys. Chem. A 2009, 113, 11926-11937.
[S2] Brogaard, R. Y.; Olsbye, U., Ethene Oligomerization in Ni-Containing Zeolites:
Theoretical Discrimination of Reaction Mechanisms. ACS Catal. 2016, 6, 1205–1214.
[S3] Latimer, A. A.; Kulkarni, A. R.; Aljama, H.; Montoya, J. H.; Yoo, J. S.; Tsai, C.; Abild- Pedersen, F.; Studt, F.; Nørskov, J. K., Understanding trends in C–H bond activation in heterogeneous catalysis. Nat. Mater. 2016, 16, 225-229.
[S4] Rosen, A.S.; Notestein, J. M.; Snurr, R. Q., Structure−Activity Relationships that Identify Metal−Organic Framework Catalysts for Methane Activation. ACS Catal. 2019, 9, 3576-3587.
[S5] Szécsényi, Á.; Li G.; Gascon, J.; Pidko, E. A., Mechanistic complexity of methane oxidation with H2O2 by single-site Fe/ZSM-5 catalyst. ACS Catal. 2018, 8, 7961-7972.
