Hydronium-Ion-Catalyzed Elimination Pathways of Substituted


Hydronium-Ion-Catalyzed Elimination Pathways of Substituted...

0 downloads 69 Views 740KB Size

Subscriber access provided by LAURENTIAN UNIV

Article

Hydronium ion catalyzed elimination pathways of substituted cyclohexanols in zeolite H-ZSM5 Peter Heinrich Hintermeier, Sebastian Eckstein, Mariefel V. Olarte, Donald M. Camaioni, Eszter Barath, and Johannes A. Lercher ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b01582 • Publication Date (Web): 04 Oct 2017 Downloaded from http://pubs.acs.org on October 4, 2017

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

ACS Catalysis 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 24

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

ACS Catalysis

Hydronium ion catalyzed elimination pathways of substituted cyclohexanols in zeolite H-ZSM5 Peter H. Hintermeier,† Sebastian Eckstein,† Donghai Mei,‡ Mariefel V. Olarte,‡ Donald M. Camaioni,‡ Eszter Baráth,†,* and Johannes A. Lercher†,‡* †

Technische Universität München, Department of Chemistry and Catalysis Research Center,

Lichtenbergstraβe 4, Garching, D-85747, Germany ‡

Institute for Integrated Catalysis, Pacific Northwest National Laboratory, 902 Battelle

Boulevard, Richland, WA 99352, USA

ABSTRACT Hydronium ions in the pores of zeolite H-ZSM5 show high catalytic activity in the elimination of water from cyclohexanol in aqueous phase. Substitution induces subtle changes in rates and reaction pathways, which are concluded to be related to steric effects. Exploring the reaction pathways of 2-, 3-, 4-methylcyclohexanol (2-McyOH, 3-McyOH, 4-McyOH), 2-, 4ethylcyclohexanol (2-EcyOH, 4-EcyOH), 2-n-propylcyclohexanol (2-PcyOH) and cyclohexanol (CyOH) it is shown that the E2 character increases with closer positioning of the alkyl- and the hydroxyl-group. Thus, 4-McyOH dehydration proceeds via an E1 type elimination, while cis-2McyOH preferentially reacts via an E2 pathway. The entropy of activation decreased with increasing alkyl chain length (ca. 20 J mol-1 K-1 per CH2-unit) for 2-substituted alcohols, which

ACS Paragon Plus Environment

1

ACS Catalysis

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 24

is concluded to result from constraints influencing the configurational entropy of the transition states.

KEYWORDS Dehydration, Elimination reactions, Zeolite, Heterogeneous catalysis, Homogeneous catalysis 1. INTRODUCTION Elimination reactions in the gas phase such as the dehydration of (cyclic) alcohols on acidic catalysts have been extensively studied, not only because of significant synthetic interest, but also because it helped to elucidate fundamental mechanistic principles.1-3 A seminal study of Macht et al. showed that 1- and 2-butanol on Keggin-type polyoxometalate (POM) catalysts involve late carbenium-ion-type transition states in the kinetically relevant elimination step.4 Similar catalysis was also observed for zeolites.5 The presence of H2O led, however, to the formation of stable and less reactive coordination complexes of water and alcohol.5 This suggests that the local environment and the interactions between reactant influence the reaction pathways. The deprotonation energy of the acid catalyst, the protonation enthalpies of the alcohols and the stability of these complexes determine the enthalpies as well as entropies of activation along the elimination reaction path.5 Therefore, the structure and the degree of substitution of the alcohol markedly influence the elimination pathway, as both affect the stabilization of ground and transition states.4 Dehydration in aqueous phase is catalyzed by hydrated hydronium ions, independent whether they are provided by molecular or solid acids. For solids immersed in water, the hydronium ions remain associated with the surface or pores of the solid acid, as the energy required for charge separation prevents complete delocalization of the hydrated hydronium ions (H3O+).5 The

ACS Paragon Plus Environment

2

Page 3 of 24

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

ACS Catalysis

presence of zeolite pores, induces constraints that enhance the dehydration rates of these hydronium ions compared to reactions in aqueous phase. This is exemplified by the significantly higher dehydration rates of cyclohexanol (CyOH) by hydronium ions in zeolite BEA compared to phosphoric acid, while the principal E1 mechanism remains unchanged (Figure 1).6

Figure 1. Schematic energy diagram of possible dehydration pathways (E1 and E2) on solid acid catalysts (adapted from ref. 6). The association of the alcohol with the hydrated hydronium ion is nearly thermoneutral and is hypothesized to resemble the substitution of one water molecule in the hydration shell.6 The stabilization of alcohols in the zeolite pores leads to a marked enrichment compared to the aqueous solution, leading to a zero order for each of the alcohols studied. As a note in passing, we would like to emphasize that the exact coordination of the alcohol has not been experimentally shown so far. The subsequent protonation (B) is correlated with an increase in ∆G°‡ passing the first transition state (TS1). Then, the E1-type elimination proceeds via a stepwise cleavage of the Cα–O (TS2) and Cβ–H bonds (TS3) with a stable intermediate (carbenium ion; C). A simultaneous cleavage of the Cα–O and Cβ–H bonds (TS4) forming the C=C double bond and water (D) would follow in an E2-type elimination pathway.6, 7 The alkyl

ACS Paragon Plus Environment

3

ACS Catalysis

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 4 of 24

substituents of cyclohexanol influence the hydronium ion catalyzed dehydration, leading to different reactivity of the isomers.8 In order to tailor catalysts for more complex, functionalized alcohols, we explore here the impact of the substituents of cyclohexanol on the reaction pathways and rates of hydronium catalyzed dehydration within the confines of the MFI structure, as these confines led to the highest rates for simple alcohols9 and were expected to exert marked influence on catalytic activity and selectivity. Varying the size (chain length) and the position of the alkyl substituent relative to the OH-group was chosen as approach to qualitatively and quantitatively analyze dehydration pathways. 2. RESULTS and DISCUSSION Characterization and stability of H-ZSM5 zeolite The most important physicochemical properties of the MFI zeolite, i.e., the micropore volume and the concentration of acid sites are compiled in Table 1. The concentration of Brønsted acid sites (BAS, 360 µmol g-1) and Lewis acid sites (LAS, 45 µmol g-1) were determined by IR spectra of adsorbed pyridine. Both, the position and the relative intensity of XRD peaks assured that, the microcrystalline structure was maintained during the catalytic experiments (SI, Figure S1-S3). The acid site concentration also did not decrease markedly. Thus, we conclude that the zeolite was stable under the conditions used for the presented experiments.

ACS Paragon Plus Environment

4

Page 5 of 24

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

ACS Catalysis

Table 1. Characterization of the investigated MFI-45 zeolite (Si/Al = 45) Characterization type

MFI-45 (Si/Al = 45)

Particle size (nm)

100 - 150 2

-1

Surface area – total (m g )

389 2

-1

Surface area – micropores (m g )

302

Surface area – mesopores (m2 g-1)

87

a

-1

c (acid sites) (µmol g ) b

-1

b

-1

c (BAS) (µmol g )

400 360

c (LAS) (µmol g ) 45 a Determined by TPD of NH3 (see SI). bDetermined by IR spectroscopy with adsorbed pyridine (see SI).

Dehydration of alkylcyclohexanols in aqueous phase The dehydration of alkyl cyclohexanols substituted at the 2-, 3-, or 4-positions led to various alkenes (Scheme 1). The product selectivities are compiled in Table 2. Main products were isomers of alkylcyclohexenes and, at higher temperatures, additional small amounts of ringcontracted alkylcyclopentenes (selectivity < 2%). Primary data and experimental details (determination of reaction rates and data analysis, Equation S1-S11, Figure S1-S27, Table S1S27) can be found in the Supporting Information. Scheme 1. Dehydration products of 2-, 3-, 4-McyOH and CyOH

The product with the more highly substituted double-bond, e.g., 1-methylcyclohexene, (1MCH, Saytzeff-product10) dominated (ca. 80% for 2-McyOH). The product with the less substituted double bond, e.g., 3-methylcyclohexene (3-MCH, Hofmann-product) was always

ACS Paragon Plus Environment

5

ACS Catalysis

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

Page 6 of 24

found in lower concentrations (ca. 13% for 2-McyOH).11 We want to emphasize that dehydration in aqueous phase is a reversible reaction.6 The addition of water to 1-MCH forms 1-McyOH (selectivity < 6%), however this reaction is not significant at low conversions.12 Table 2. Selectivitya of the dehydration products of 2-, 3-, 4-McyOH and CyOH Reactant Cyclohexene (CH)

Selectivity (%) 4-, 3-Methyl-cyclohexene (4-MCH, 3-MCH)

1-Methyl1-Methylcyclohexene cyclohexanol (1-MCH) (1-McyOH) CyOH 100 2-McyOH (3-MCH) 14 80 6 3-McyOH (4-MCH and 3-MCH)b 70 28 2 4-McyOH (4-MCH) 72 26 2 a Selectivity was determined at 10% conversion, 170 °C, 0.5 M alcohol, 100 mL H2O, 50 bar, MFI zeolite. b4-MCH and 3-MCH cannot be separated via GC-MS analysis.

Dehydration of 2-, 3-, 4-McyOH and CyOH In a first step dehydration of 2-, 3-, 4-McyOH and CyOH was kinetically investigated using the four as benchmark substrates in aqueous phase. A reaction order of zero in alcohol was determined for 2-, 4-McyOH as well as CyOH (SI, Figure S6 – S8 and Table S1 – S3), which implies that the reaction enthalpies and entropies represent intrinsic values.13 Figure 2 illustrates the temperature dependence of the dehydration rates catalyzed by hydronium ions in zeolite MFI.

ACS Paragon Plus Environment

6

Page 7 of 24

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

ACS Catalysis

Figure 2. Arrhenius plots for the dehydration of 2-, 3-, 4-McyOH (cis/trans mixtures) and CyOH (MFI). The corresponding activation energies (Ea), the turnover frequencies (TOFs), enthalpies (∆H°‡) as well as entropies of activation (∆S°‡) of all methyl-substituted reactants and CyOH are compiled in Table 3. The differences in Ea, ∆H°‡ and ∆S°‡ among the substrates with varying substituent position are significant. Table 3. Comparison of Eaintr, ∆H°‡, ∆S°‡, and TOF for the dehydration of CyOH and McyOHs (MFI)a

a

CyOH

Eaintr (kJ mol-1) 152 (±2)

∆H°‡ (kJ mol-1) 148 (±2)

∆S°‡ (J mol-1 K-1) +74 (± 6)

TOF (170 °C) (s-1) 0.24 (±0.02)

2

4-McyOH

146 (±3)

142 (±3)

+53 (± 6)

0.10 (±0.02)

3

3-McyOH

133 (±3)

129 (±3)

+28 (± 7)

0.15 (±0.01)

4

2-McyOH

116 (±3)

112 (±3)

-8 (± 7)

0.25 (±0.01)

Entry

Substrate

1

Reaction conditions: 120 – 160 °C, 0.5 M alcohol (cis/trans mixtures), 100 mL H2O, 50 bar.

Compared to CyOH, the introduction of a methyl-group in position 4 affects ∆S°‡ more than ∆H°‡ (Table 3, Entries 1 and 2). The slightly higher enthalpy of activation (148 vs. 142 kJ mol-1) for CyOH is overcompensated by a higher entropy of activation (+74 vs. +53 J mol-1 K-1) such that the rates and TOFs at every reaction temperature are higher than those of 4-McyOH. We would like to emphasize that the activation parameters for 3-McyOH (Table 3, Entry 3) differed significantly from both CyOH and 4-McyOH. The activation enthalpy (129 kJ mol-1) and the entropy (+28 J mol-1 K-1) were both significantly lower compared to the two previously discussed substrates. Although the entropy was lower (+28 vs. +53 J mol-1 K-1), the reduced activation enthalpy (129 vs. 142 kJ mol-1) resulted in higher TOFs compared to 4-McyOH (Table 3, Entries 2 and 3). The magnitude of the enthalpic and entropic barriers of 3-McyOH dehydration were between those of 2- and 4-methylcyclohexanol (Table 3, Entry 2, 3 and 4).

ACS Paragon Plus Environment

7

ACS Catalysis

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

Page 8 of 24

Dehydration of 2-McyOH (Table 3, Entry 4) led to a low enthalpy (116 kJ mol-1) and a negative entropy of activation ∆S°‡(-8 J mol-1 K-1). The differences of 30 kJ mol-1 in ∆H°‡ and ca. 60 J mol-1 K-1 in ∆S°‡ between 2- and 4-McyOH are remarkable (Table 3, Entry 2 and 4). The higher reactivity of the cis isomer in a cis/trans mixture (1:1) of 2-McyOH point to an important steric effect of the substituent.8c The experiments previously introduced (Table 3) were performed with cis/trans mixtures. However, fundamental understanding of dehydration reactivity requires the analysis of the pure isomers. Based on the largest difference in activation parameters such as Ea, ∆H°‡, ∆S°‡, TOF, (Table 3) among 2- and 4-McyOH the isomers of these two substrates were investigated in detail. Table 4 compiles the kinetic data of the hydronium ion catalyzed dehydration of cis and trans isomers of 2-McyOH and 4-McyOH. Cis and trans isomers of 4-McyOH showed the same reactivity in dehydration (Table 4, Entry 1 and 2). Not only the rates per active site (TOFs: 5.4×10-3 s-1 and 4.4×10-3 s-1), but also ∆H°‡ (144 and 145 kJ mol-1) and ∆S°‡ (+58 and +59 J mol1

K-1) were nearly identical (Table 4, Entry 2).

Table 4. Comparison of Eaintr, ∆H°‡, ∆S°‡, and TOF for the dehydration of 2- and 4-McyOH isomers (MFI)a

a

Entry

Substrate

Eaintr (kJ mol-1)

1

trans-4-McyOH

2

∆S°‡ (J mol-1 K-1) +59 (± 4)

TOF (140 °C) (s-1)

148 (± 2)

∆H°‡ (kJ mol-1) 144 (± 2)

cis-4-McyOH

148 (± 2)

145 (± 2)

+58 (± 2)

4.4 (±1) ×10-3

3

trans-2-McyOH

144 (± 2)

141 (± 2)

+43 (± 2)

2.1 (±1) ×10-3

4

cis-2-McyOH

114 (± 3)

111 (± 3)

-14 (± 7)

1.2 (±1) ×10-2

5.4 (±2) ×10-3

Reaction conditions: 120 – 160 °C, 0.05 M alcohol, 100 mL H2O, 50 bar.

Significant differences were observed, however, between the two isomers of 2-McyOH (Table 4, Entry 3). Dehydration of trans-2-McyOH surprisingly showed the same order of magnitude in

ACS Paragon Plus Environment

8

Page 9 of 24

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

ACS Catalysis

TOF (2.1×10-3 s-1) and similar enthalpy of activation (141 kJ mol-1) as the isomers of 4-McyOH. The reactivity of cis-2-McyOH (Table 4, Entry 4) was about one order of magnitude higher (1.2×10-2 s-1), most markedly induced by a lower activation enthalpy. While this shows that the cis isomer dominates initially in mixtures of the 2-McyOH cis/trans mixture, let us discuss why it is easier to convert in detail (Scheme 2). Scheme 2. E2 dehydration pathways for cis- and trans-2-McyOH

Elimination of water via an E1 type mechanism should not lead to a preference for cis and trans isomers of the alcohol.14 Therefore, we hypothesize that the elimination from 2-McyOH occurs in a concerted reaction step (E2 mechanism, Scheme 2). Dehydration via this route requires an anti-periplanar configuration of the protonated hydroxyl-group and the adjacent β-H (Scheme 2).14

ACS Paragon Plus Environment

9

ACS Catalysis

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 10 of 24

In case of cis-2-McyOH both reaction products can be formed, the Saytzeff-product with the more highly substituted double bond (energetically favored) and the Hofmann-product (energetically less favored) (Table 5). By contrast, for trans-2-McyOH the formation of the Saytzeff-product is not possible via an E2 mechanism. Thus, the pathway leading to the Hofmann-product is the only accessible. The higher activation barrier of trans-2-McyOH is attributed to the formation of the more energetically demanding Hofmann-product. Table 5. Saytzeff/Hofmann-product ratios for the dehydration of cis- and trans-2-McyOH (MFI) Substrate

120 °C

140 °C

160 °C

cis-2-McyOH (Saytzeff/Hofmann)

14.4

13.6

11.2

trans-2-McyOH (Saytzeff/Hofmann)

1.7

1.5

1.6

Scheme 3. E1 dehydration pathways for cis- and trans-2-McyOH

ACS Paragon Plus Environment

10

Page 11 of 24

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

ACS Catalysis

The substantial excess of the Saytzeff-product (11- to 14-fold) with cis-2-McyOH allows to conclude, therefore, that the elimination occurs concertedly, i.e., without the possibility of isomerization of an intermediately formed carbenium ion by hydride shift. On first sight, one could argue that the lower activation energy is caused by the fact that charged species are avoided in the reaction path on the expense of a lower entropy of activation. However, we would tend to rule out this argument, as it would also suggest that a concerted pathway should also have been dominating for the non-substituted alcohol.6 Thus, the question arises, why the substitution at the 2 position leads to an overall lower ∆G°‡. The anti-periplanar orientation is of critical importance for the concerted elimination step. An E2 pathway can only occur when the adjacent C-H bond and the leaving group (OH2+ in the present case) have ~180° dihedral angle (such as in case of cis-2-McyOH). In the concerted elimination mechanism the anti-periplanar geometry is favored, because the σC-H and σ*C-O orbitals are parallel to each other. Thus, the parallel σC-H bonding and the σ*C-O anti-bonding orbitals can overlap, and allow hyperconjugation with the C-H bonds of the methyl group. Note that based on the calculated charge distribution within the substituted cyclic carbocations, we conclude that the cyclohexyl carbenium ions are not significantly better stabilized when methyl groups are present in the 2-, 3- or 4-positions (see SI, Figure S27). Additional factors contribute to make the E2 pathway favorable, i.e., the cis-2-McyOH being 4 kJ mol-1 less stable than trans2-McyOH.

15

This, combined with the greater stability of 1-methylcyclohexene compared to its

other isomers (6 and 8 kJ mol-1 more stable than 3-MCH and 4-MCH, respectively, 7 kJ mol-1 more stable than methylenecyclohexane),16 make the dehydration of cis-2-McyOH the least endothermic. Therefore, the TS may occur earlier in the E2 path than for the other alcohols, leading to the most negative ∆S°‡.

ACS Paragon Plus Environment

11

ACS Catalysis

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

Page 12 of 24

Scheme 4 and Scheme 5 illustrate the pathways leading to the observed product distributions. Dehydration of trans-2-McyOH results in a slight excess of the Saytzeff-olefin. The statistic distribution of products is in line with a pathway via a carbocation. In contrast, the high excess of the Saytzeff-product for cis-2-McyOH is characteristic for a concerted reaction pathway. Following the product yields as a function of conversion, further supports the mechanistic proposal. For trans-2-McyOH (SI, Figure S4) dehydration conversion was positively correlated with the formation of the cis isomer. The formation of ethylcyclopentene was already detected at low conversions (X < 3%), as it is characteristic for the presence of a carbenium ion. For cis-2McyOH (SI, Figure S5), however, neither the fraction of trans-2-McyOH changed with conversion nor was ethylcyclopentene observed. In all cases, the amount of 1-McyOH increased with rising conversion. As outlined above, the anti-periplanar orientation of both leaving groups is not required in an E1 elimination (Scheme 3). The resulting higher number of abstractable β-protons favors the formation of the Hofmann-product statistically (2:1) for both isomers of 2-McyOH (Table 4).

ACS Paragon Plus Environment

12

Page 13 of 24

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

ACS Catalysis

Scheme 4. Reaction pathways of trans-2-McyOH via a carbocation

1-McyOH can be either formed by the addition of water to the olefinic product via the most stable carbenium ion according to Markovnikov’s rule or the secondary carbocation can be transformed into a tertiary carbocation by a hydride-shift and a subsequent attack of water (Scheme 4 and Scheme 5). A separate set of experiments using 1H-NMR to analyze the products support these conclusions (SI, Table S22). Conversion of cis-2-McyOH did not lead to the trans isomer (Scheme 5), whereas the experiments with pure trans-2-McyOH formed the cis isomer with increasing reaction time (Scheme 4). The cis and trans isomers of 4-McyOH are also interconverted as a function of reaction time, by rehydration of 4-McyOH cation as well as the product 4-methylcyclohexene.

ACS Paragon Plus Environment

13

ACS Catalysis

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

Page 14 of 24

Scheme 5. E2 dehydration pathway of cis-2-McyOH

Together with the information of the product distribution of the pure isomers, the activation parameters like Ea, ∆H°‡ and ∆S°‡ (Table 3 and 4) can be evaluated in more detail. The variation in activation energy of about 30 kJ mol-1 among both 2-McyOH isomers seems to be the consequence of two different dehydration routes. The negative entropy of activation (∆S°‡ = -8 J mol-1 K-1) in case of cis-2-McyOH is hypothesized to represent an early and substrate-like transition state. The low entropy of activation is associated with the constrained and highly ordered E2 arrangement consisting of the substrate, the proton (converting R-OH into the reactive leaving group R-OH2+) and the water cluster (β-H abstracting base). Contrary to this, the positive entropy of activation in case of trans-2-McyOH (∆S°‡ = +43 J mol-1 K-1) is interpreted as a late and product-like transition state with a stepwise C-O cleavage and a consecutive proton abstraction.

ACS Paragon Plus Environment

14

Page 15 of 24

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

ACS Catalysis

The concerted E2 pathway avoids the formation of a carbenium ion and stabilizes the early transition state complex, which leads to a low enthalpy of activation (Figure 1). In case of trans2-McyOH the concerted pathway leading to the thermodynamically favored Saytzeff-product is forbidden by the absence of the anti-periplanar orientation of C-OH2+ and C-H. As a result, the dehydration of trans-2-McyOH follows the pathway with a higher enthalpy of activation, the stepwise reaction sequence via the carbenium ion (E1). Interestingly, the isomers of 4-McyOH are not restricted in possible dehydration pathways, but nevertheless react via E1 (∆S°‡ = +53 J mol-1 K-1 and ∆H°‡ = 142 kJ mol-1), comparable to the data of trans-2-McyOH. Same is observed for CyOH (∆S°‡ = +74 J mol-1 K-1 and ∆H°‡ = 148 kJ mol-1) with slight deviation occurring from the absence of the methyl substituent. In contrast to the 2-McyOH cis/trans mixture, which is kinetically determined by the cis isomer (E2), the activation parameters of 3-McyOH (∆S°‡ = +29 J mol-1 K-1 and ∆H°‡ = 128 kJ mol-1) led to values that are intermediate between both extreme cases (E1 and E2) (Table 3). Experiments with heavy water and deuterated cyclohexanol previously showed that the C-H bond cleavage from the carbenium ion is the rate determining step within the E1 route.6 This implies that independently of the elimination mechanism, the step of proton abstraction is kinetically relevant. To explore whether the differences in activation enthalpy and entropy results primarily from the molecular structure or additionally from the zeolite confinement, dehydration kinetics of 2and 4- McyOH were analyzed in the presence of hydronium ions generated by a molecular acid (H3PO4) under identical reaction conditions. Contrary to the MFI catalyzed reactions, the

ACS Paragon Plus Environment

15

ACS Catalysis

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

Page 16 of 24

reaction orders were first order in substrate (SI, Figure S13, S16 and Table S8, S11). Table 6 compiles TOF, Eaapp, ∆S°‡ and ∆H°‡ for the dehydration of trans-2-McyOH and cis-2-McyOH. The comparison of the reaction rates normalized to the concentrations of hydronium ions17 shows that the difference in reactivity between the cis (k = 1.8×10-4 L mol-1 s-1) and the trans (k = 1.9×10-5 L mol-1 s-1) isomer of 2-McyOH is one order of magnitude, i.e., equivalent to the differences in the zeolite catalyzed dehydration studies (Table 6). In line with this, a significantly higher activation barrier (Eaapp = 167 kJ mol-1) for trans-2-McyOH was determined than for the cis isomer (Eaapp = 133 kJ mol-1). Similar to the zeolite results discussed above, the entropy of activation was higher for the trans isomer (cis: ∆S°‡ = +18 J mol-1 K-1; trans: ∆S°‡ = +69 J mol-1 K-1). The calculations to estimate the intrinsic activation parameters for reactions in aqueous phase from the apparent values were done in analogy to ref.6 While the association equilibrium between the alcohol and the hydronium ion is enthalpically neutral, differences in the entropy of ca. 24 J mol-1 K-1 have to be considered. The normalized intrinsic values for transand cis-2-McyOH of +18 and +69 J mol-1 K-1 are similar to the numbers obtained by MFI zeolite catalysis. Thus, the confinement in the zeolite changes the ∆H°‡ more than ∆S°‡. Table 6. Homogeneously catalyzed dehydration (H3PO4): comparison of TOF, Eaapp, ∆H°‡ and ∆S°‡ Entry

Substrate

1a

cis-2-McyOH

b

Eaapp (kJ mol-1) 133 (±3)

∆H°‡ (kJ mol-1) 129 (±3)

∆S°‡ (J mol-1 K-1) +18 (±7)

k (140 °C)c (L mol-1 s-1) 1.8 (±0.1) ×10-4

1.9 (±0.2) ×10-5 Reaction conditions: 150 – 190 °C, 0.2 M alcohol, 50 bar, 3 mM H3PO4. Reaction conditions: 170 – 210 °C, 44 mM alcohol, 50 bar, 50 mM/17 mM H3PO4. cReaction conditions: 44 mM alcohol, 50 bar, 73 mM H3PO4. 2

a

trans-2-McyOH

167 (±3)

163 (±3)

+69 (±6)

b

1

H NMR analysis (SI, Table S22) after catalytic conversion of pure isomers with hydronium

ions generated by H3PO4 showed that the dehydration of cis-2-McyOH did not form the trans-2McyOH (E2), whereas trans-2-McyOH produced cis-2-McyOH (E1 via carbenium ion).

ACS Paragon Plus Environment

16

Page 17 of 24

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

ACS Catalysis

Therefore, it is concluded that the dehydration mechanism does not depend on the steric constraints of the hydronium ion. The selectivity observed with phosphoric acid is comparable to zeolite catalysis, with Saytzeff/Hofmann-product ratios between 1.3 and 1.5 for trans-2-McyOH and between 14 and 19 for cis-2-McyOH. These results were to be expected, because the differences in reactivity between both isomers of 2-McyOH in TOF (MFI)/ k (H3PO4) (about one order of magnitude) as well as in enthalpic (ca. 30 kJ mol-1) and entropic barriers (ca. 50 J mol-1 K-1) were constant and independent of the confinement (Table 4, Entry 3 and 4; Table 6, Entry 1 and 2). Thus, we conclude that the specific activity of the isomers does not depend on the environment of the hydrated hydronium ion. It solely depends on the steric factors outlined above.

Impact of larger alkyl-substituents As we showed that the environment has a positive influence and may constrain the ground and transition states of reacting substrates in their interactions with hydrated hydronium ions, we explored the impact of increasing steric limitations by replacing the methyl substituent with an ethyl- and propyl-group for 2- and 4-substituted CyOHs. Turnover frequencies, entropy and enthalpy of activation for the ethyl- and propyl-substituted cyclohexanols are compiled in Table 7. The dehydration reactions catalyzed by MFI were zero order with respect to the alcohol. Similar to methyl-substituted cyclohexanols, all substrates formed alkyl cyclohexenes and only trace quantities of five-membered ring products (mainly at higher temperatures). The Saytzeffisomer was the major olefinic product for all 2-substituted substrates (SI, 2-McyOH: Table S17, Table S25, Figure S22; 2-EcyOH: Table S18, Table S26, Figure S23; 2-PcyOH: Table S19, Table S27, Figure S24). For 2-EcyOH the excess was 9 – 14 fold, similar to the selectivity of 2-

ACS Paragon Plus Environment

17

ACS Catalysis

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

Page 18 of 24

McyOH. Interestingly, the Saytzeff/Hofmann isomer ratios for 2-PcyOH were lower, i.e., between 4 and 7. In case of 4-substituted cyclohexanols, the primary products were 4alkylcyclohexenes, although 3- and 1-alkylcyclohexenes were observed in small concentrations. Table 7. Comparison of Eaintr, ∆H°‡, ∆S°‡ and TOF for the dehydration of larger alkylsubstituents (MFI)a

a

4-McyOH

Eaintr (kJ mol-1) 146 (±3)

∆H°‡ (kJ mol-1) 142 (±3)

∆S°‡ (J mol-1 K-1) +53 (±6)

TOF (170 °C) (s-1) 0.10 (±0.02)

2

4-EcyOH

143 (±3)

139 (±3)

+50 (±7)

0.15 (±0.01)

3

2-McyOH

116 (±3)

112 (±3)

-8 (±7)

0.25 (±0.01)

4

2-EcyOH

107 (±2)

104 (±2)

-25 (±4)

0.28 (±0.01)

5

2-PcyOH

104 (±2)

101 (±2)

-43 (±7)

0.07 (±0.01)

Entry

Substrate

1

Reaction conditions: 140 – 190 °C, 0.5 M alcohol (cis/trans mixture), 50 bar.

Increasing the size of the substituent did not strongly influence the rates or enthalpies and entropies of activation in position 4. We conclude, therefore, that the size of an n-alkyl group hardly influences the ground and transition state along the reaction pathway. The n-propyl-substituent in position 2, had a greater impact on the TOF (0.07 s-1; Table 7, Entry 5) lowering it approximately four fold. The decrease in the reaction entropy with chain length is attributed to a rather limited configurational entropy in the transition state. Correlating all dehydration enthalpies and entropies of the investigated alcohols, two groups of substrates proceed via two different reaction pathways (Figure 3A). The alcohols CyOH, cis/trans-4-McyOH and trans-2-McyOH follow the pathway via a carbenium ion (E1) with high enthalpic barriers to overcome. These elimination reactions are supported by high entropies (late transition states). On the other hand there are concerted elimination reactions for the cis-2substituted compounds. Based on the absence of a carbenium ion, the enthalpic barriers are significantly lower (by 30 kJ mol-1). The drawback of a concerted pathway is a highly ordered

ACS Paragon Plus Environment

18

Page 19 of 24

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

ACS Catalysis

arrangement in the transition state consisting of the substrate, the proton and the base (water molecule(s)). The three-dimensional-correlation of TOF, ∆H°‡ and ∆S°‡ (Figure 3B) additionally highlights the correlation between TOFs and the activation barriers. With decreasing enthalpic barriers, higher dehydration rates were observed in a linear correlation. However, three exceptions are noted, i.e., the dehydration of CyOH and 2-PcyOH (formation of Saytzeff-product) as well as all dehydration pathways of the 2-subsituted alcohols leading to the Hofmann-products (Figure 3B). In case of cyclohexanol, the absence of the alkyl substituent leads to significantly higher TOFs compared to the methyl-substituted counterparts, as methylcyclohexanols gain intrinsically less entropy in their transition states than cyclohexanol does.

Figure 3. A) Correlation of ∆H°‡ and ∆S°‡ of MFI catalyzed dehydration. B) Correlation of ∆H°‡, ∆S°‡ and TOF (170 °C) values of MFI catalyzed dehydration. The propyl substituent (2-PcyOH) is sterically demanding and seems to impact the dehydration process; the observed smaller TOFs are attributed to a shielding effect of the extended alkyl

ACS Paragon Plus Environment

19

ACS Catalysis

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

Page 20 of 24

chain affecting the protonation/deprotonation by the (hydronium ion) water cluster. The dehydration rates to the Hofmann-products are significantly smaller based on the thermodynamic preference of forming higher substituted olefins (Saytzeff-products). Furthermore, a combination of several factors determines the catalytic conversion, i.e., (i) the position of the alkyl-substituents, (ii) their steric configuration to the OH group, and (iii) their size.

3. Conclusions Elimination of water from secondary methyl substituted cyclohexanols catalyzed by hydronium ions in the medium pore zeolite H-ZSM5 depends on the position and the size of the alkyl substitution. Dehydration of pure isomers show that trans-2-McyOH and cis-/trans-4McyOH react via a carbocation intermediate (E1 pathway). This conclusion is based on the formation of the cis-isomer during conversion and the nearly balanced Saytzeff/Hofmann ratios (ca. 2:1) via a late transition state with a high positive entropy of activation (+43 J mol-1 K-1). Dehydration of cis-2-McyOH proceeds via a concerted dehydration (E2) avoiding the interconversion of isomers and leading to a high excess of Saytzeff-product (ca. 14:1) via an early transition state with a negative entropy of activation (-8 J mol-1 K-1). Regardless of mechanism, a compensating effect exists in which higher activation barriers for E1 pathways are systematically offset by larger entropies of activation, (i.e., greater frequencies of reactive encounters). Independent of the dehydration mechanism, catalysis of hydronium ions generated by H3PO4 in water led to TOFs that were two orders of magnitude lower and had higher activation energies compared to hydronium ions in the confines of MFI pores.

ACS Paragon Plus Environment

20

Page 21 of 24

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

ACS Catalysis

The results show, how the chemical nature of the reacting substrate can experience local differences despite the identical nature of the catalytically active site. The steric constraints and the chemical activity of the active site determine the rate constants. While we begin to understand these localized activities of sites in a system resembling the confines of enzyme pockets, the presented study also points to the complexity and challenges to quantitatively describe and understand the conversions on a molecular scale. AUTHOR INFORMATION Corresponding Author *[email protected] *[email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. The authors declare no competing financial interest. ASSOCIATED CONTENT Supporting Information. Detailed experimental procedures and characterization data, kinetical data collection for all the substrates including Figure S1-S27, Table S1-S27 and Equation S1S11. The Supporting Information is available free of charge on the ACS Publications website (PDF). ACKNOWLEDGMENT This work was supported by the U.S. Department of Energy (DOE), Energy Efficiency and Renewable Energy Office, Bioenergy Technologies Office in the framework of the PNNLTUM/CN 212303 project for catalytic upgrading of biomass using lower temperature and

ACS Paragon Plus Environment

21

ACS Catalysis

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 22 of 24

pressure. J.A.L. and D.M.C. acknowledge support by the U.S. Department of Energy (DOE), Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences and Biosciences for exploring alternative oxidic supports for deoxygenation reactions. Special gratitude is expressed to Prof. G. L. Haller and Prof. A. Jentys for fruitful discussions, to F. Kirchberger and H. Renges for their help in performing experiments. Pacific Northwest National Laboratory is operated by Battelle for the U.S. DOE. REFERENCES (1)

(a) Noller, H.; Thomke, K. J. Mol. Catal. 1979, 6, 375‒392. (b) Noller, H.; Kladnig, W.

Catal. Rev. Sci. Eng. 1976, 13, 149‒207. (c) Salvador, P.; Kladnig, W. J. Chem. Soc., Faraday Trans. 1: Phys. Chem. in Cond. Phases 1977, 73, 1153‒1168. (d) Thomke, K.; Noller, H. Z. Naturforsch. B. 1972, 27, 1462‒1464. (2)

(a) He, J.; Zhao, C.; Lercher, J. A. J. Catal. 2014, 309, 362‒375. (b) Zhao, C.; Song, W.;

Lercher, J. A.; ACS Catal. 2012, 2, 2714‒2723. (c) Zhao, C.; Lercher, J. A. Angew. Chem., Int. Ed. 2012, 24, 6037‒6042. (d) Zhao, C.; Kou, Y.; Lemonidou, A. A.; Li, X.; Lercher, J. A. Angew. Chem., Int. Ed. 2009, 22, 4047‒4050. (e) Zhao, C.; Kasakov, S.; He, J.; Lercher, J. A. J. Catal. 2012, 296, 12‒23. (f) Zhao, C.; He, J.; Lemonidou, A. A.; Li, X.; Lercher, J. A. J. Catal. 2011, 280, 8‒16. (g) Zhao, C.; Camaioni, D. M.; Lercher, J. A. J. Catal. 2012, 288, 92‒103. (h) Vjunov, A.; Hu, M. Y.; Feng, J.; Camaioni, D. M.; Mei, D.; Hu, J. C.; Zhao, C.; Lercher, J. A. Angew. Chem., Int. Ed. 2014, 53, 479‒482. (3)

(a) Bautista, F.; Campelo, J.; Garcia, A.; Luna, D.; Marinas, J.; Quiros, R.; Romero, A.;

Appl. Catal. A: Gen. 2003, 243, 93‒107. (b) Clearfield, A.; Thakur, D. S. J. Catal. 1980, 65, 185‒194. (c) Johnstone, R. A.; Liu, J.-Y.; Whittaker, D. J. Chem. Soc., Perkin Trans. 2 1998, 1287‒1288. (d) Thakur, D.; Clearfield, A. J. Catal. 1981, 69, 230‒233.

ACS Paragon Plus Environment

22

Page 23 of 24

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

ACS Catalysis

(4)

(a) Macht, J.; Janik, M. J.; Neurock, M.; Iglesia, E. J. Am. Chem. Soc. 2008, 130, 10369‒

10379. (b) Janik, M. J.; Macht, J.; Iglesia, E.; Neurock, M. J. Phys. Chem. C 2009, 113, 1872‒ 1885. (5)

(a) Zhi, Y.; Shi, H.; Mu, L.; Liu, Y.; Mei, D.; Camaioni, D. M.; Lercher, J. A. J. Am.

Chem. Soc. 2015, 137, 15781‒15794. (b) Mei, D.; Lercher, J. A. Am. Inst. Chem. Eng. 2017, 63, 172‒184. (6)

Liu, Y.; Vjunov, A.; Shi, H.; Eckstein, S.; Camaioni, D.; Mei, D.; Baráth, E.; Lercher, J.

A. Nat. Commun. 2017, 8, 14113. (7)

Thibblin, A. Chem. Soc. Rev. 1993, 22, 427‒433.

(8)

(a) Clennan, M. M.; Clennan, E. L. J. Chem. Educ. 2011, 88, 646‒648. (b) Friesen, J. B.;

Schretzman, R. J. Chem. Educ. 2011, 88, 1141‒1147. (c) Todd, D. J. Chem. Educ. 1994, 71, 440‒440. (d) Kochloefl, K.; Kraus, M.; Chin-Shen, C.; Beranek, L.; Bažant, V. Collect. Czech. Chem. Commun. 1962, 27, 1199‒1209. (9)

Shi, H.; Eckstein, S.; Vjunov, A.; Camaioni, D.; Lercher, J. A. Nat. Commun. 2017, 8,

15442. (10)

Saytzeff, A. Liebigs Ann. Chem. 1875, 179, 296‒301.

(11)

Hofmann, A. Liebigs Ann. Chem. 1851, 79, 11‒39.

(12)

Markownikoff, W. Liebigs Ann. Chem. 1870, 153, 228‒259.

(13)

Laidler, K. J.; Chemical Kinetics: Theories of Reaction Rates, 3rd ed.; Harper & Row:

New York, 1987; pp 89‒115. (14)

Smith, M. B.; March, J.; March’s Advanced Organic Chemistry: Reactions, Mechanisms

and Structure, 3rd ed.; Wiley & Sons Inc.: New Jersey, 2007; pp 1477‒1558. (15)

Kabo, G. Ya.; Frenkel, M. L. J. Chem. Thermodynamics 1983, 15, 377‒381.

ACS Paragon Plus Environment

23

ACS Catalysis

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

(16)

Yursha, I. A.; Kabo, G. Ya.; Andreevskii, D. N. Neftekhimiya 1974, 14, 688‒693.

(17)

Rudolph, W. W. J. Solution Chem. 2012, 41, 630‒645.

Page 24 of 24

For Table of Contents use Only (TOC)

ACS Paragon Plus Environment

24