Arylketone π-Conjugation Controls Enantioselectivity in Asymmetric

19 mins ago - The origin of enantioselectivity in the asymmetric alkynylation of trihalomethyl ketones catalyzed by octahedral stereogenic-at-rutheniu...
0 downloads 9 Views 9MB Size
Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Arylketone #-Conjugation Controls Enantioselectivity in Asymmetric Alkynylations Catalyzed by Centrochiral Ruthenium Complexes Shuming Chen, Yu Zheng, Tianjiao Cui, Eric Meggers, and Kendall N. Houk J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.8b00485 • Publication Date (Web): 26 Mar 2018 Downloaded from http://pubs.acs.org on March 26, 2018

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

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

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

Journal of the American Chemical Society

Arylketone π-Conjugation Controls Enantioselectivity in Asymmetric Alkynylations Catalyzed by Centrochiral Ruthenium Complexes Shuming Chen,† Yu Zheng,‡ Tianjiao Cui,‡ Eric Meggers,‡ and K. N. Houk*,† † ‡

Department of Chemistry and Biochemistry, University of California, Los Angeles, California 90095-1569, United States Fachbereich Chemie, Philipps-Universität Marburg, Hans-Meerwein-Strasse 4, 35043 Marburg, Germany

ABSTRACT: The origin of enantioselectivity in the asymmetric alkynylation of trihalomethyl ketones catalyzed by octahedral stereogenic-at-ruthenium complexes has been investigated through density functional theory calculations. Computational results support a mechanism involving formation of a ruthenium acetylide, followed by pre-coordination of the trihalomethyl ketone through the carbonyl oxygen and intramolecular attack of the acetylide via a compact four-membered transition state. Differences in computed free energies of activation for the formation of the major and minor propargyl alcohol enantiomers are in good agreement with the experimentally observed levels of asymmetric induction. Analysis of fragment distortion energies show that disfavored transition states are destabilized due to the more severe distortion and loss of π-conjugation in the coordinated arylketone fragments. Examination of the different substitution patterns in the ketone substrate and the catalyst revealed the key steric factors that control the enantioselectivity. Finally, calculations indicate promising directions for the simplification of the catalyst scaffold while preserving the high levels of enantioselectivity of these alkynylation reactions.

INTRODUCTION Asymmetric synthesis provides entry into highly desirable enantiopure compounds important to the production of bioactive agents and pharmaceuticals.1 Transition metal complexes that catalyze enantioselective transformations are typically employed in conjunction with carefully designed chiral ligands for asymmetric induction.1 However, an alternative design can be envisioned in which enantioselectivity is achieved via the creation of a chiral environment through an asymmetric arrangement of achiral ligands around a transition metal center.2 Chiral Lewis acids featuring octahedral, biscyclometalated iridium3 or rhodium4 metal centers have been shown to be highly competent catalysts for a variety of synthetically useful reactions.5 Broader applicability of this design principle was recently demonstrated in the development of a new class of centrochiral complexes Δ- or Λ-Ru1 and Ru2 (Figure 1a), which have been shown to be effective catalysts for asymmetric alkynylations of trihalomethyl ketones 1 (Figure 1b).6–7 In these Ru complexes, two achiral bidentate N-(2pyridyl)-substituted N-heterocyclic carbene (PyNHC) ligands coordinate to the Ru center in a propeller-like arrangement which generates helical chirality.8 The strongly σ-donating PyNHC ligands render the acetonitrile ligands trans to them highly labile (trans effect). The acetonitrile ligands thus dissociate readily during reactions to provide coordination sites for substrates. Coordinated substrates are closely subjected to the chiral environment created by the configurationally inert PyNHC units, which presumably leads to high levels of asymmetric induction. Figure 1b shows the scope of the Ru-catalyzed asymmetric alkynylation of trihalomethyl ketones 1 to yield enantiopure propargyl alcohols 3.6 The Ru complex Λ-Ru1, which is un-

substituted at the C3 position, provided good asymmetric induction (97% ee) in the alkynylation of trifluoroacetophenone to give 3a. This high enantioselectivity was a surprising result given the absence of sterically demanding substituents on the PyNHC ligand framework. Further improved stereodistinction (99% ee) was achieved with Λ-Ru2, which possesses bulky 3,5-dimethylphenyl groups at the C3 positions of the pyridine rings. Excellent yields and enantioselectivities were observed in the alkynylations of trifluoroacetophenone with both alkyl and aryl acetylenes (3a–m). Aryl-substituted trifluoroacetophenones gave excellent enantioselectivities and high yields (3n, 3o, 3p–u), with the exception of 2'-methyl-2,2,2trifluoroacetophenone, which gave the alcohol product 3p in much lower yields of 55% with Λ-Ru1 and 27% with Λ-Ru2. These diminished yields indicate the catalysts to be sensitive to steric changes in the substrate. The more electron-rich acetophenone was found to be unreactive (3z), and the use of alkyl or ester-substituted trifluoromethyl ketones resulted in significantly diminished ee values (3v–w). Finally, 2-chloro2,2-difluoroacetophenone was found to be a competent substrate for the reaction (3x), whereas 2,2,2trichloroacetophenone was not reactive toward the alkynylation (3y).

ACS Paragon Plus Environment

Journal of the American Chemical Society

(a) configurationally inert ligands

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

(PF6)2

R1 N

(PF6)2

R1

N

NN

N

N N

Ru

N

labile ligands

N

N

NN

Ru

N N

N N R1

R1

Δ-Ru1: R1 = H Δ-Ru2: R1 = 3,5-Me2Ph

Λ-Ru1: R1 = H Λ-Ru2: R1 = 3,5-Me2Ph

(b)

R2 O

R2

R1

CF3

Λ-Ru2 (cat.)

H

1

HO

Et3N (0.2 equiv) THF, 60 °C

(3.0 equiv)

R1

2

CF3 3

For R1 = Ph: 3a: R = H (Λ-Ru1: 97% ee) Λ-Ru2: 99% ee 3b: R = 4’-Me: 99% ee 3c: R = 3’-Me: 99% ee 3d: R = 2’-Me: 99% ee 3e: R = 4’-OMe: 99% ee 3f: R = 4’-Br: 97% ee 3g: R = 2’-F: 99% ee

R

3a–g 75–99% yield

S

3h 88% yield 96% ee

3i 98% yield 99% ee

SiMe3 3j 96% yield 99% ee

3k 97% yield 97% ee

3l 98% yield 99% ee

3m 77% yield >99% ee

For R2 = Ph: 3n: R = 4’-Me: 93% yield, 98% ee 3o: R = 3’-Me: 96% yield, 99% ee 3p: R = 2’-Me: 55% yield, 94% ee (Λ-Ru1)a 27% yield, 91% ee (Λ-Ru2) 3q: R = 4’-OMe: 90% yield, 99% ee 3r: R = 4’-Br: 99% yield, 99% ee 3s: R = 3’-Br: 99% yield, 99% ee 3t: R = 3’-CF3: 99% yield, 99% ee

R

3n–t

CO2Et 3u 96% yield 99% ee

3v Λ-Ru1: 48% yield, 30% eea Λ-Ru2: 44% yield, 62% ee Ph

HO Ph

Ph HO

CF2Cl

3x 99% yield, 99% ee

3w 70% yield 7% ee

Ph

Ph HO

CCl3

3y 0% yielda

Ph

CH3

3z 0% yielda

Figure 1. (a) Structure and design principle of octahedral Ru catalysts with metal-centered chirality. (b) Enantioselective alkynylations of trifluoromethyl ketones catalyzed by centrochiral Ru complexes bearing achiral PyNHC ligands. a Experimental results from this work.

Density functional theory calculations provide in-depth understanding of the stereocontrolling elements in transition states, which is essential for the rational improvement of catalyst designs. Recent computational studies have shed light on the mechanism and source of stereodistinction in asymmetric transformations catalyzed by centrochiral, C2-symmetric rhodium complexes relying on similar design principles as the ruthenium catalysts shown in Figure 1a.9 However, all previous computational studies have focused on substrates that coordinate in a bidentate manner to the metal catalyst. As such, these computational studies do not provide any models of asymmetric induction that are transferrable to the alkynylation reaction shown in Figure 1b. In this paper, we report the first computational study elucidating the mechanism and the origin of the observed enantioselectivity in the alkynylation of trihalomethyl ketones catalyzed by stereogenic-at-ruthenium complexes shown in Figure 1a. COMPUTATIONAL METHODS All computations were performed with the Gaussian 0910 software package. Ground state and transition state geometries were optimized in the gas phase using the B3LYP11 functional augmented with the D3 version of Grimme’s empirical dispersion correction12 using the LANL2DZ13 effective core potential for Ru, and 6-31G(d) for all other atoms. Frequency calculations were carried out at the same level of theory as that used for geometry optimization to characterize the stationary points as either minima (no imaginary frequencies) or saddle points (one imaginary frequency) on the potential energy surface, and to obtain thermal corrections to the Gibbs free energies. Intrinsic Reaction Coordinate (IRC) calculations were performed to ensure that the saddle points found were true transition states connecting the reactants and the products. Single point energies were calculated with the M0614 functional augmented with the D3 version of Grimme’s empirical dispersion correction using the SDD15 basis set for Rh and 6-311G++(d,p) for all other atoms. Solvation effects were modeled using the SMD16 solvation model with tetrahydrofuran as the solvent. For fragment distortion analysis, single-point energies of the molecular fragments were computed at the M06-D3/6311G++(d,p)-SDD level using the B3LYP-D3/6-31G(d)LANL2DZ geometries obtained in the gas phase. Molecular structures were visualized using CYLview.17 Steric Isotope Effect (SIE) calculations were carried out using the default Gaussian 09 thermochemistry algorithm accompanying frequency calculations, and adjusted for temperature and isotopic content. Monte Carlo conformational searches were performed with the Merck molecular force field (MMFF) implemented in Spartan ’16 to ensure that the lowest energy conformations of intermediates and transition states are presented in the manuscript. A more detailed discussion on the choice of computational methods and alternative conformers explored in this study can be found in the Supporting Information. RESULTS AND DISCUSSION A plausible mechanism for the alkynylation of trihalomethyl ketones by the centrochiral ruthenium complexes Λ-Ru or Δ-Ru is outlined in Scheme 1. Dissociation of an acetonitrile ligand from the Ru center enables the coordination of acetylene 2 and enhances the acidity of the sp C–H proton (intermediate 4). Deprotonation by triethylamine leads to the for-

ACS Paragon Plus Environment

Page 2 of 8

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

Journal of the American Chemical Society

mation of ruthenium acetylide 5, which undergoes ligand exchange to yield complex 6, with the trihalomethyl ketone coordinated to the Ru center through the carbonyl oxygen. Intramolecular delivery of the ruthenium acetylide in a compact four-membered transition state TS-1, followed by protonation and ligand exchange, furnishes the propargyl alcohol product 3. The close coordination between the Ru center and the ketone substrate throughout the C–C bond formation process is proposed to account for the high levels of asymmetric induction observed experimentally. To test the validity of this mechanistic model, we first computed the energetics of the formation of ruthenium acetylide complex 6a from Δ-Ru1, propyne, and trifluoroacetophenone. The calculated free energy of formation for complex 6a was 16.0 kcal/mol. This energy will not necessarily reflect well on the energy in solution since the anionic counterions are not included in the calculation and are expected to interact more strongly with ammonium than the ruthenium complex. Transition states for the re-face alkynylation (TS-1aR) and the si-face alkynylation (TS-1aS) were computed to have free energies of activation of 25.4 and 28.0 kcal/mol, respectively. These relatively high barriers are in agreement with the experimental observation that elevated temperatures (60 °C) and long reaction times (16 h) are required for good conversion to the products. 2+

R1 N NN

R2

N

Ru

2

N

N N

N NN

N Ru

N N

– MeCN

Et3N

Δ-Ru1

CF3

4a

5a

8.1

8.3

MeCN

TS-1aS

28.0

TS-1aR

25.4 6a

MeCN

+

L

16.0

O

L

Ru

O CF3

Et3NH+

Ph

0.0

Figure 2. Free energy diagram of the formation of ruthenium acetylide complex 6a from Δ-Ru1, and alkynylation transition states TS-1aR and TS-1aS. Figure 3 shows the optimized geometries of the transition states for the re-face alkynylation (TS-1aR) and the si-face alkynylation (TS-1aS) of trifluoroacetophenone. The forming C–C bonds are 1.96 Å in TS-1aR and 2.01 Å in TS-1aS, respectively. The computed 2.6 kcal/mol difference in free energies of activation between TS-1aR and TS-1aS corresponds to a computed ee of 99%, which is in excellent agreement with the experimental ee value of 97%. TS-1aR

N Ru

O

TS-1aS

+

CF3 Ph

N N

N NN

N Ru

+

Ph O CF3

N N

4

R1

Δ-Ru

O

Ph

NN

R2

R1

Ru

Ph CF3

N

N H

+

L L

2+

R1 +

N

ΔG (kcal/mol)

B B H +

R1 N

N

NN

O

Ru

+

N Ru

CF3

– MeCN

N NN

N

O

N N

R2 R1

+

R2 R1

TS-1

5

ΔG‡ = +9.4



R2

CF3 R3

N N

N

Ru

6

R1

NN

1

R2 R1

N

R3

R3

N N

+

R1 O

CF3

HO R3

CF3 3

Scheme 1. Proposed Mechanism of the Alkynylation of Trihalomethyl Ketones Catalyzed by Centrochiral Ruthenium Complexes.

ΔG‡ = +12.0

Figure 3. Optimized transition state geometries for the alkynylation of trifluoroacetophenone catalyzed by Δ-Ru1. Hydrogen atoms are omitted for clarity. Atomic distances are denoted in Ångströms; free energies are with respect to the coordinated complex 6a, and denoted in kcal/mol. To elucidate the origin of the 2.6 kcal/mol difference in the free energies of activation between TS-1aR and TS-1aS, we calculated distortion energies of the molecular fragments participating in the two transition states. Fragment distortion energies are computed by comparing gas-phase electronic energies of the transition state geometries and that of their corresponding ground state geometry in the ruthenium acetylide

ACS Paragon Plus Environment

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

complex 6a. As highlighted in Figure 4, the trifluoroacetophenone molecule in TS-1aS is more distorted than in TS-1aR by 2.2 kcal/mol, despite TS-1aS being the earlier transition state where the C–C bond formation is less advanced. Two geometric parameters account for the higher distortion energy. Firstly, the O1–C3–C4–C5 dihedral angle is 13.1° in TS-1aR and 35.6° in TS-1aS, which shows that the π-conjugation in trifluoroacetophenone is significantly more weakened in TS-1aS. Secondly, the C2–C3–C4 angle is 114.2° in TS-1aR and 111.9° in TS-1aS, indicating more advanced pyramidalization of C3 in TS-1aS, which also reflects the loss of sp2 planarity from decreased π-conjugation. The 2.2 kcal/mol difference in the ketone distortion energy is comparable to the 2.6 kcal/mol overall difference in the free energies of activation between TS-1aR and TS-1aS, suggesting that the main source of stereodistinction is the degree of distortion in the ketone geometry, primarily the loss of π-conjugation. In contrast, the distortion energies of the acetylide segment are essentially the same in both transition states, and the distortion energies of the Ru catalyst framework with its twin propeller structure are within 0.2 kcal/mol of each other.

An examination of fragment distortion energies reveals that the difference in ΔEdist (ketone), the distortion energy of the ketone fragment, increases to a striking 5.4 kcal/mol between TS-1bR and TS-1bS. This value is again comparable to the overall difference in free energies of activation (5.1 kcal/mol) between TS-1bR and TS-1bS. The O1–C3–C4–C5 dihedral angle is 8.5° in TS-1bR and 67.8° in TS-1bS, a much more pronounced difference compared to that between TS-1aR and TS-1aS. The C2–C3–C4 angle is similarly smaller in the disfavored transition state (115.3° in TS-1bR and 112.4° in TS1bS), indicating a more pyramidalized C3 due to loss of πconjugation. These results indicate that severely weakened πconjugation in the trifluoroacetophenone segment in the disfavored transition state is the main source of enantioselectivity. Compared to Δ-Ru1, Δ-Ru2 improves upon the stereodiscrimination by creating an even larger discrepancy between the degrees of π-conjugation in the ketone fragments in the two transition states. TS-1bR

TS-1bS

+ R

N

TS-1aR +

L

Page 4 of 8

NN

N Ru

N

CF3

O

NN

Ph

N N

L

Ru

N

Ph O

Ru

CF3

N N

O R

CF3 Ph

R

R = 3,5-Me2Ph

R = 3,5-Me2Ph

1

2

3 5

4

ΔEdist (Ru) = -0.8 ΔEdist (acetylide) = 0.2

+ R

ΔEdist (ketone) = 21.4

TS-1aS +

L L

Ru

O Ph CF3

2

1 3 4

5

ΔG‡ = +9.1 ΔEdist (Ru) = -0.6 ΔEdist (acetylide) = 0.2

ΔG‡ = +14.2

ΔEdist (ketone) = 23.6

Figure 4. Fragment distortion energies in TS-1aR and TS-1aS in the alkynylation of trifluoroacetophenone. Atomic distances are denoted in Ångströms; energies are shown in kcal/mol. We next computed the transition states for the alkynylation of trifluoroacetophenone catalyzed by Δ-Ru2, with 3,5-Me2Ph groups at the C3 position of the pyridine rings (Figure 5). Starting from the ruthenium acetylide complex 6b, the re-face alkynylation (TS-1bR) and the si-face alkynylation (TS-1bS) were calculated to have free energies of activation of 9.1 and 14.2 kcal/mol, respectively. The forming C–C bonds are 1.96 Å in TS-1bR and 2.02 Å in TS-1bS, which are comparable to the corresponding C–C bond lengths in TS-1aR and TS-1aS. The calculated difference in free energies of activation is 5.1 kcal/mol, which predicts >99% ee, a result that is in good agreement with the experimental value (99% ee).

1 2

1 5

3 4

ΔEdist (ketone) = 21.4 ΔEdist (Ru) = -1.8 ΔEdist (acetylide) = 0.2

2

5

3 4

ΔEdist (ketone) = 26.8 ΔEdist (Ru) = -1.6 ΔEdist (acetylide) = 0.2

Figure 5. Optimized transition state geometries and fragment distortion energies for the alkynylation of trifluoroacetophenone catalyzed by Δ-Ru2. Hydrogen atoms are omitted for clarity. Atomic distances are denoted in Ångströms; free energies are computed with respect to the coordinated complex 6b, and denoted in kcal/mol.

ACS Paragon Plus Environment

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

Journal of the American Chemical Society

Figure 6 shows the major van der Waals contacts in the transition states TS-1aR, TS-1aS, TS-1bR and TS-1bS. In TS1aR, the pyridine C2 hydrogen on the PyNHC ligand has a steric interaction with the ketone CF3 group at 2.48 Å. The ketone Ph group has a weakly attractive π-π stacking interaction at 3.71 Å with the other pyridine moiety on the ligand framework (Figure 3), and is largely free from unfavorable steric interactions. In TS-1aS, the pyridine C2 hydrogen has two unfavorable C–H steric interactions with the ketone Ph group (2.50 Å, 2.66 Å). These interactions are responsible for the distortion of the the ketone geometry and the breaking of its π-conjugation. It is worth noting that due to the highly compact nature of the transition states, the ketone CF3 group experiences significant steric clashes with the catalyst ligand framework, the acetylide fragment, and the Ph group on the ketone itself. This steric sensitivity accounts for the inactivity of trichloroacetophenone, which has a larger CCl3 group, under the same experimental conditions. In the case of the CF2Cl variant, the lone chlorine can occupy the least sterically demanding positions in both transition states, which explains why the CF2Cl ketone is a competent substrate for this reaction. To further quantify the steric repulsion experienced by individual hydrogens in these transition states, we computed their Steric Isotope Effect (SIE) values (Figure 6). Due to the less sterically demanding nature of C–D bonds compared to C–H bonds, chemical transformations proceeding through sterically strained transition states are known to exhibit inverse kH/kD SIE values.18 Unsurprisingly, the most inverse SIE values were exhibited by the pyridine C2 hydrogen atom on the pyridine ring, with a kH/kD of 0.951 in TS-1aR and 0.949 in TS-1aS. In TS-1bR and TS-1bS, the most inverse SIE values were exhibited by the C2 hydrogen atom on the top 3,5-Me2Ph substituent, at 0.963 and 0.967 respectively. The C2 hydrogens on the bottom 3,5-Me2Ph substituent have the second most inverse SIE values, indicating that the 3,5-Me2Ph groups create a tighter steric pocket around the ketone substrate and improve stereodiscrimination.19,20

TS-1aR

TS-1aS (0.949)

(0.951)

(0.967) (0.982)

(0.963)

(0.986)

(0.975) (0.979)

TS-1bR

TS-1bS

Figure 6. Interatomic distances (in Å) of van der Waals contacts in alkynylation transition states. Red numbers in parentheses indicate computed Steric Isotope Effect (SIE) values (kH/kD) of select hydrogen atoms. To probe the effects of the electronics and sterics of the ketone substrate on the reaction, we computed the transition states TS-1R and TS-1S for the alkynylations of a range of ketones catalyzed by Δ-Ru1 or Δ-Ru2. Overall, the calculated ΔΔG‡ values are in good qualitative agreement with experimental ΔΔG‡ values, and the errors are within 1.0 kcal/mol. Given that the change in π-conjugation is the major source of stereodistinction in the alkynylation of arylketones, the low observed ee in alkynylation of alkylketones can be rationalized due to their lack of π-conjugation. Indeed, computed transition states TS-1cR and TS-1cS for the alkynylation of 1,1,1trifluoroacetone showed low levels of stereodistinction. The higher calculated barriers for acetophenone (16.3 and 19.8 kcal/mol for TS-1dR and TS-1dS, respectively) accounts for its lack of reactivity towards alkynylation under the experimental conditions. Finally, calculation trends show that ortho substitution on the arylketone does not have a significant negative impact on the stereodistinction (TS-1eR and TS-1eS, TS-1fR and TS-1fS). Table 1. Calculated free energies of activation (in kcal/mol) for the alkynylation of ketones. Free energies are computed with respect to the corresponding ruthenium acetylide complex 6.

ACS Paragon Plus Environment

Journal of the American Chemical Society

re face alkynylation

si face alkynylation +

R1 N

N

NN

Ru

N

R3

O

+

R1

NN

R2

N N

N Ru

O

R3 R2

N N R1

TS-1R

R1

Calculated ΔG‡

TS

R1

R2

R3

TS-1aR

H

Ph

CF3

9.4

TS-1aS

H

Ph

CF3

12.0

TS-1bR

3,5Me2Ph

Ph

CF3

9.1

TS-1bS

3,5Me2Ph

Ph

CF3

14.2

TS-1cR

H

Me

CF3

10.9

TS-1cS

H

Me

CF3

10.3

TS-1dR

H

Me

Ph

16.3

TS-1dS

H

Me

Ph

19.8

TS-1eR

H

2-ClPh

CF3

9.0

TS-1eS

H

2-ClPh

CF3

10.5

TS-1fR

H

2-MePh

CF3

8.2

TS-1fS

H

2-MePh

CF3

11.1

Calculated ΔΔG‡

TS-1S

Exper- Experimental imental ee (%) ΔΔG‡

2.6

97

2.4

5.1

>99

>2.5

-0.6

30a

0.4

3.5

-

-

1.5

95b

2.1

ΔEdist (Ru) values are calculated with respect to the geometry of the Ru–PyNHC framework in Δ-Ru1. The rigidity of the Ru–PyNHC catalyst frameworks in ΔRu1 and Δ-Ru2 could be attributed both to a tightly interlocked propeller structure and to attractive π-π stacking interactions between the pyridyl and the mesityl moieties. We were interested in seeing how far the catalyst scaffold could be simplified without diminishing the levels of asymmetric induction. Calculated ΔG‡ and ΔΔG‡ values for the alkynylation of trifluoroacetophenone catalyzed by Δ-Ru complexes bearing simpler imidazole N substituents are shown in Table 2. The ΔΔG‡ values predict high levels of stereodistinction for both N-Ph and N-Me variants of the catalyst. Because the ligand framework is not stabilized by π-π stacking in the N-Me catalyst, this result shows that π-π stacking within the catalyst scaffold is not critical for enantiodiscrimination. Although the synthetic methodology used for the preparation of Δ-Ru1 and Δ-Ru2 cannot be applied to the N-Ph and N-Me variants, the remarkable insensitivity of the stereoinduction to the imidazole N-substitution nonetheless paint a promising picture for simpler and more economical catalyst designs. Table 2. Calculated free energies of activation (in kcal/mol) for the alkynylation of trifluoroacetophenone with simplified Ru catalyst designs. Free energies are computed with respect to the corresponding ruthenium acetylide complex 6. re face alkynylation

2.9

91

1.9

Experimental result obtained with 1,1,1-trifluoro-4phenylbutan-2-one. b Experimental result obtained with 2',5'dichloro-2,2,2-trifluoroacetophenone. Figure 7 shows an overlay of the geometries adopted by the Ru–PyNHC metal-ligand frameworks in Δ-Ru1, complex 6a and transition state TS-1aR. The ΔEdist (Ru) values show that the most distorted Ru–PyNHC framework occurs upon ketone coordination in the ruthenium acetylide complex 6a. The compactness of the four-membered transition state relieves some of the steric strain on the Ru-PyNHC framework, leading to a smaller ΔEdist (Ru) value in TS-1aR (1.8 kcal/mol compared to 2.6 kcal/mol in 6a). Overall, the geometrical changes in the Ru–PyNHC framework are small in magnitude, indicating the exceeding rigidity of these catalyst scaffolds.

N NN

N Ru

N N

si face alkynylation +

N R NN

a

configurationally inert ligands

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 8

L1 L2

Red: Δ-Ru1, ΔEdist (Ru) = 0.0 Green: 6a, ΔEdist (Ru) = 2.6 Blue: TS-1aR, ΔEdist (Ru) = 1.8

Figure 7. Overlay of the Ru–PyNHC metal-ligand frameworks in Δ-Ru1, complex 6a and transition state TS-1aR.

R

N Ru

O

CF3 Ph

N N

+ N N R O NN Ru R N N

TS-1R

Ph CF3

TS-1S

TS

R

Calculated ΔG‡

TS-1aR

Mesityl

9.4

TS-1aS

Mesityl

12.0

TS-1gR

Ph

9.7

TS-1gS

Ph

12.0

TS-1hR

Me

9.6

TS-1hS

Me

12.1

Calculated ΔΔG‡ 2.6

2.3

2.5

CONCLUSIONS We present the first computational study elucidating the origin of enantioselectivity in the asymmetric alkynylation of trihalomethyl ketones catalyzed by octahedral stereogenic-atruthenium complexes, where the asymmetric induction arises exclusively from the stereogenic metal center and the resulting helical chirality of the catalyst structure. Our results support a mechanism involving the intramolecular delivery of a ruthenium acetylide to a coordinated ketone through a highly compact four-membered transition state. In this transition state, both substrates are coordinated to the stereogenic ruthenium center, which is crucial to obtaining high asymmetric induction. Computed free energies of activation are generally in good agreement with the experimentally determined ee values. We show that transition states leading to the minor enantiomers are destabilized by disruption of p-conjugation in the arylketone. Key steric interactions between the catalyst

ACS Paragon Plus Environment

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

Journal of the American Chemical Society

framework and the substrate in the transition states have been identified. Finally, calculations predict high levels of asymmetric induction for N-Ph and N-Me variants of the catalyst, a result that offers promising directions of simplifying the catalyst scaffold while preserving enantioselectivity.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Computational Details, Energies and Cartesian Coordinates of computed structures (PDF)

11.

AUTHOR INFORMATION Corresponding Author 12.

*[email protected]

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT K.N.H is grateful to the National Science Foundation (Grant CHE-1059084) for financial support. E.M. thanks the Deutsche Forschungsgemeinschaft for financial support (ME1805/15-1). Calculations were performed on the Hoffman2 cluster at the University of California, Los Angeles, and the Extreme Science and Engineering Discovery Environment (XSEDE), which is supported by the National Science Foundation (Grant OCI-1053575).

REFERENCES 1.

Walsh, P. J.; Kozlowski, M. C. Fundamentals of Asymmetric Catalysis; University Science Books: Sausalito, CA, 2009. 2. For a recent review on asymmetric catalysts which derive their chirality from a stereogenic metal center, see: Zhang, L. Meggers, E. Chem. Asian J. 2017, 12, 2335. 3. Huo, H.; Fu, C.; Harms, K.; Meggers, E. J. Am. Chem. Soc. 2014, 136, 2990. 4. Wang, C.; Chen, L.-A.; Huo, H.; Shen, X.; Harms, K.; Gong, L.; Meggers, E. Chem. Sci. 2015, 6, 1094. 5. Zhang, L.; Meggers, E. Acc. Chem. Res. 2017, 50, 320. 6. Zheng, Y.; Tan, Y.; Harms, K.; Marsch, M.; Riedel, R.; Zhang, L.; Meggers, E. J. Am. Chem. Soc. 2017, 139, 4322. 7. For reviews on enantioselective alkynylation reactions, see: (a) Peshkov, V. A.; Pereshivko, O. P.; van der Eycken, E. Chem. Soc. Rev. 2012, 41, 3790; (b) Yoo, W.-J.; Zhao, L.; Li, C.-J. Aldrichimica Acta, 2011, 44, 43; (c) Trost, B. M.; Weiss, A. H. Adv. Synth. Catal. 2009, 351, 963; (d) Cozzi, P. G.; Hilgraf, R.; Zimmerman, N. Eur. J. Org. Chem. 2004, 20, 4095; (e) Pu, L. Tetrahedron 2003, 59, 9873. 8. The Ru–PyNHC structural scaffolds in Λ-Ru1 and Λ-Ru2 are based on a scaffold reported in: Kaufhold, O.; Hahn, F. E.; Pape, T.; Hepp, A. J. Organomet. Chem. 2008, 693, 3435. 9. For computational studies on the origin of enantioselectivity in reactions catalyzed by complexes with exclusive metal-centered chirality, see: (a) Tutkowski, B.; Meggers, E.; Wiest, O. J. Am. Chem. Soc. 2017, 139, 8062; (b) Fernandez-Alvarez, V. M.; Maseras, F. Org. Biomol. Chem. 2017, 15, 8641. (c) Chen, S.; Huang, X.; Meggers, E.; Houk, K. N. J. Am. Chem. Soc. 2017, 139, 17902.. 10. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.;

13. 14. 15.

16. 17. 18.

19. 20.

Nakai, H.; Vreven, T.; Montgomery, J. A.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09; Gaussian Inc.: Wallingford, CT, 2009. (a) Head-Gordon, M.; Pople, J. A.; Frisch, M. J. Chem. Phys. Lett. 1988, 153, 503; (b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648; (c) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785; (d) Vosko, S. H.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200; (e) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys. Chem. 1994, 98, 11623. Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys. 2010, 132, 154104. Hay, P. J.; Wadt, W, R. J. Chem. Phys. 1985, 82, 299. Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215. (a) Häussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.; Schwerdtfeger, P.; Pitzer, R. M. Mol. Phys. 1993, 78, 1211. (b) Küchle, W.; Dolg, M.; Stoll, H.; Preuss, H. J. Chem. Phys. 1994, 100, 7535. Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2009, 113, 6378. Legault, C. Y. CYLview, 1.0b; Université de Sherbrooke, 2009 (http://www.cylview.org). (a) Kuchitsu, K.; Bartell, L. S. J. Chem. Phys. 1962, 36, 2470. (b) Allinger, N. L.; Flanagan, H. L. J. Comput. Chem. 1983, 4, 399. (c) Dunitz, J. D.; Ibberson, R. M. Angew. Chem., Int. Ed. 2008, 47, 4208. The analogue of Ru1 in which both pyridine C2 positions are deuterated resulted in identical enantioselectivities as the undeuterated Ru1. See the Supporting Information for details.. The smaller kH/kD values observed for TS-1aS than TS-1bS are likely due to a combination of the van der Waals contact distances being smaller, and the clashes being more “side-on” than “head-on” in TS-1aS (kH/kD values are more sensitive to out-ofplane bending of C-H bonds than stretching). See: (a) Streitwieser, A. Jr. Solvolytic Displacement Reactions; McGraw-Hill Book Co: New York, NY, 1956. (b) Streitwieser, A. Jr.; Jagow, R. H.; Fahey, C. R.; Suzuki, S. J. Am. Chem. Soc. 1958, 80, 2326.

ACS Paragon Plus Environment

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

Page 8 of 8

1.

Table of Contents (TOC) Graphic R1

+

L

O

R2 CF3

H L

(3.0 equiv)

Ru

O CF3

Δ-Ru (cat.) Et3N (0.2 equiv) THF, 60 °C

Ph

2

R2

4

HO R1

1 3

CF3 favored TS

ACS Paragon Plus Environment

5