Dramatic Effect of Carboxylic Acid on the Electronic ... - ACS Publications

Jul 11, 2016 - (1) Recently, White and Chen have reported that the catalyst system [((S,S)-PDP)Fe(CH3CN)2](SbF6)2 (1SbF6)/H2O2/CH3COOH (Chart 1) can ...
0 downloads 3 Views 980KB Size
Subscriber access provided by The University of British Columbia Library

Letter

Dramatic Effect of Carboxylic Acid on the Electronic Structure of the Active Species in Fe(PDP)-Catalyzed Asymmetric Ep-oxidation Alexandra M. Zima, Oleg Y. Lyakin, Roman V. Ottenbacher, Konstantin P. Bryliakov, and Evgenii P. Talsi ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b01473 • Publication Date (Web): 11 Jul 2016 Downloaded from http://pubs.acs.org on July 12, 2016

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 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

ACS Catalysis

Dramatic Effect of Carboxylic Acid on the Electronic Structure of the Active Species in Fe(PDP)-Catalyzed Asymmetric Epoxidation Alexandra M. Zima, Oleg Y. Lyakin, Roman V. Ottenbacher, Konstantin P. Bryliakov, Evgenii P. Talsi* Novosibirsk State University, Pirogova 2, Novosibirsk 630090, Russian Federation Boreskov Institute of Catalysis, Pr. Lavrentieva 5, Novosibirsk 630090, Russian Federation ABSTRACT: The electronic structure of the iron-oxygen intermediates responsible for catalytic transformations in the biomimetic catalyst systems [((S,S)-PDP)FeII(OTf)2]/H2O2/RCOOH has been found to strongly depend on the structure of the carboxylic acid RCOOH. For carboxylic acids with primary and secondary α-carbon atom (acetic acid, butyric acid, caproic acid), the active species exhibit EPR spectra with large g-factor anisotropy (g1 = 2.7, g2 = 2.4, g3 = 1.7), whereas for those with tertiary α-carbon atom (2-ethylhexanoic acid, valproic acid, 2-ethylbutyric acid), the active species display EPR spectra with small g-factor anisotropy (g1 = 2.07, g2 = 2.01, g3 = 1.96). The EPR spectra of the latter intermediates are very similar to those of the intermediates previously assigned to oxoiron(V) species. The systems featuring intermediates of the second type ensure higher enantioselection in the epoxidation of electron-deficient olefins. KEYWORDS: asymmetric epoxidation, bioinspired catalysis, enantioselectivity, EPR, iron, mechanism

The search for robust and selective biomimetic catalysts, capable of oxidizing organic substrates in the same fashion as metalloenzymes do has been a challenging task for years.1 Recently, White and Chen have reported that the catalyst system [((S,S)-PDP)Fe(CH3CN)2](SbF6)2 (1SbF6)/H2O2/CH3COOH (Chart 1) can selectively oxidize inactivated sp3 and methylene C–H bonds of complex organic molecules and natural products with considerable level of predictability and in preparatively useful yields.2a–e Bryliakov, Talsi and coworkers showed that the catalyst system [((S,S)-PDP)Fe(CF3SO3)2] (1)/H2O2/RCOOH (Chart 1) is capable of epoxidizing prochiral electron-deficient olefins with high enantioselectivity level,3a where RCOOH is carboxylic acid, typically introduced in the systems as a co-catalytic additive.2f–k The epoxidation enantioselectivity substantially increased when branched carboxylic acids (such as 2-ethylhexanoic acid) were used instead of CH3COOH. Costas and co-workers discovered that the yields and enantioselectivities of epoxidations catalyzed by aminopyridine-iron complexes of the PDP family can be improved by introducing electron-donating substituents at the pyridine rings of the PDP ligand.3b–c Complexes 2 and 3 (Chart 1) were identified as the best catalysts in the series, the enantioselectivities exhibited by catalyst 3 approaching 99% ee for some substrates. It would be tempting to unravel the effect of electron-donors on the nature and properties of the catalytically active sites of the Fe(PDP)-based catalyst systems. Previously, we demonstrated that EPR spectroscopy may be efficiently used to detect the elusive iron-oxygen species in several catalyst systems based on non-heme iron complexes.3a,4

Chart 1. Structures of Aminopyridine Iron Complexes In the course of our studies, the extremely unstable intermediates 1a and 1aAA were observed in the catalyst systems 1/H2O2 and 1/H2O2/CH3COOH, respectively, exhibiting rather similar EPR spectra: g1 = 2.65–2.66, g2 = 2.42, g3 = 1.71–1.73 (Table 1, the superscript AA denotes that acetic acid was used as the additive). The self-decay of those intermediates, which took several minutes, was monitored by EPR at low temperature (−85 °C).

ACS Paragon Plus Environment

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 8

Table 1. EPR Spectroscopic Data for S = 1/2 Iron Species Formed in the Catalyst Systems Based on Iron Complexes with N4donor Tetradentate Ligands no.

g1 +a

g2

g3

ref

1

[(TMC)Fe =O(NC(O)CH3)]

2.053

2.010

1.971

6a

2

[(TAML)FeV=O]− b

1.99

1.97

1.74

6b

3

1a

2.65

2.42

1.73

3a

4

2.66

2.42

1.71

3a

5

1aAA 1aAA type c

2.66–2.72

2.42

1.66–1.71

this work

6

[((S,S)-PDP)FeV=O(OC(O)R1)]2+ (1aEHA type) d

2.069

2.007

1.961–1.963

this work

7

[((S,S)-PDP)FeIII(OC(O)R2)(R2COOH)]2+ (1b type) e

2.75–2.78

2.41–2.42

1.62–1.64

this work

8

4a

2.67

2.42

1.72

4a, 4b

9

4aAA

2.69

2.42

1.70

4a, 4b 8a

V

2+

AA

10

[((S,S)-PDP*)Fe =O(OC(O)CH3)] (5a )

2.071

2.008

1.960

11

[((S,S)-PDP*)FeV=O(OC(O)R3)]2+ (5a type) f

2.069–2.070

2.007– 2.008

1.957–1.958

2.07

2.01

1.95

12 a

compound V

V

2+

[(PyNMe3)Fe =O(OC(O)CH3)] (6a)

g

b

c

this work 9

d

TMC – tetramethylcyclam. TAML – macrocyclic tetraamide ligand. RCOOH = AA, BA, CA, IBA, CHA. R1COOH = EHA, EBA, PVA. e Complexes of the type 1b were observed for all carboxylic acids. f R3COOH = BA, CA, IBA, CHA, EHA, EBA, PVA. g PyNMe3 – neutral aminopyridine ligand.

In the catalyst systems 4/H2O2 and 4/H2O2/CH3COOH, based on the structurally closely related complex [(BPMEN)Fe(CH3CN)2](ClO4)2 (4, Chart 1), the EPR active intermediates 4aAA and 4a with very similar parameters (g1 = 2.67–2.69, g2 = 2.42, g3 = 1.70–1.72, Table 1) were observed by EPR.4a–b 4a was somewhat more stable than 1a, allowing monitoring its self-decay at temperatures as high as −70 °C. Moreover, it was demonstrated that 4a reacts with cyclohexene at this temperature, to afford cyclohexene epoxide.4a The formation of the active species 1a, 1aAA, 4a and 4aAA from 1 and 4 was shown to be a stepwise process.3a,4b At the initial stage of reaction between 1 and H2O2, the hydroperoxo complex [((S,S)-PDP)FeIII(OOH)]2+ is formed. The presence of acetic acid in the sample noticeably facilitated the conversion of [((S,S)-PDP)FeIII(OOH)]2+ to the intermediate 1aAA, enhancing its steady-state concentration. In effect, the yield of cyclohexene epoxide formed in the catalyst system 1/H2O2/CH3COOH was much higher than for the system 1/H2O2.3a In the course of the above studies, the highly reactive intermediates 1a, 1aAA, 4a and 4aAA were tentatively assigned to oxoiron(V) complexes.3a,4,5 Intriguingly, their EPR parameters (Table 1, entries 3, 4, 8, 9) were sharply different from those of the previously reported oxoferryl, formally FeV=O species (the latter exhibited much lower g-factor anisotropies: Table 1, entries 1, 2).6,7 More recently, it has been found that the introduction of electron-donating substituents into the pyridine rings of the PDP-ligand drastically changed the EPR spectrum of the active iron-oxygen species originating from the corresponding substituted non-heme complex. In particular, complex 5 (Chart 1) in the presence of acetic acid and H2O2 afforded the highly reactive intermediate 5aAA,8 exhibiting rather small g-factor anisotropy, which was drastically different from that of 1aAA (Table 1, cf. entries 4 vs. 10) but very similar to that of [(TMC)FeV=O(NC(O)CH3)]+ (Table 1, entry 1).

Very recently, the detection of intermediate 6a with EPR spectrum (g1 = 2.07, g2 = 2.01, g3 = 1.95) very close to that of 5aAA (g1 = 2.071, g2 = 2.008, g3 = 1.960) was reported for the catalyst system [(PyNMe3)Fe(CF3SO3)2] (6)/CH3CO3H (Chart 1).9 Intermediate 6a, capable of hydroxylating strong alkyl C– H bonds with high stereoretention, was assigned to the ironoxo species with iron in the formal +5 oxidation state. It is logical to assign 5aAA, exhibiting almost identical EPR spectrum, to an analogous iron-oxo species.

Chart 2. Structures and Abbreviations of Carboxylic Acids Studied Herein So far, only the effect of the donor properties of the aminopyridine ligand on the anisotropy of the EPR spectra of the active iron-oxygen species has been considered.8a At the same time, the other factor, strongly affecting the enantioselectivity of those species – the structure of the added carboxylic acid RCOOH – has not been examined with respect to its influence on the anisotropy of the EPR signal of the active oxygentransferring species. Herewith, we are presenting such study.

ACS Paragon Plus Environment

Page 3 of 8

ACS Catalysis 2.75

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

2.41

2.069 2.007

EHA

1b

EHA

1a 1bEHA

1.963

1.64

A)

−75 °C: 0 min 2 min 3 min 4 min 5 min

B)

R.T.

1bEHA

2500

3000

3500

4000

4500

H/G

Figure 1. (A) EPR spectra (−196 °C) of the sample 1/H2O2/EHA ([1]:[H2O2]:[EHA] = 1:3:10, [1] = 0.04 M) frozen 2.5 min after mixing the reagents at −65 °C in a 1.8:1 CH2Cl2/CH3CN mixture and storing the sample at −75 °C for various times. (B) EPR spectrum (−196 °C) of the sample in “A” after storing at room temperature for 2 min.

Given the reported exceptional ability of 2-ethylhexanoic acid (EHA, Chart 2) to enhance the enantioselectivity of olefin epoxidations with H2O2, catalyzed by non-heme iron complexes,3 we took EHA as the first candidate for our EPR studies. In Figure 1A, the EPR spectrum (−196 °C) of the system 1/H2O2/EHA = 1:3:10, frozen just after mixing the reagents for 2.5 min at −65 °C is presented, displaying resonances of two low-spin (S = 1/2) iron complexes 1aEHA and 1bEHA. 1bEHA is stable even at room temperature (Figure 1B), and displays rhombically anisotropic EPR spectrum (g1 = 2.75, g2 = 2.41, g3 = 1.64). The stability and EPR parameters of 1bEHA suggest its assignment to low-spin iron(III) complex with the proposed structure [((S,S)-PDP)FeIII(OC(O)R)(RCOOH)]2+ (for details of the assignment of 1bEHA see Figure S1 and the corresponding discussion in the Supporting Information). 1aEHA is highly unstable and decays with half-life time (τ1/2) of 4 min at −75 °C (Figure 1A). In the course of the experiment, the concentration of 1aEHA did not exceed 2% of total iron concentration. The addition of 0.5 equiv (0.02 M) of cyclohexene substantially accelerated the decay of 1aEHA (τ1/2 < 0.5 min at −75 °C). The EPR spectrum of 1aEHA is very similar to those for 5aAA (Table 1), exhibiting low g-factor anisotropy. It is reasonable to assume that the electronic structure of the active species of the catalyst system 1/H2O2/EHA, 1aEHA, should be similar to those of 5aAA. This picture is different from that documented for the system 1/H2O2/CH3COOH = 1:3:10, in which the low-spin intermediate 1aAA with large g-factor anisotropy was detected by EPR (Figure 2A).3a 1aAA decayed with τ1/2 = 5 min at −85 °C (Figure S2A, Supporting Information). After the decay of 1aAA, the buildup of the EPR signal of stable complex 1bAA was detected (Figure S2B). The EPR parameters of 1bAA (g1 = 2.78, g2 ≈ 2.41, g3 = 1.62) are very close to those of 1bEHA (g1 = 2.75, g2 = 2.41, g3 = 1.64), which in combination with its stability at room temperature suggests its assignment to the low-spin iron(III) species (Table 1, entry 7). Intrigued by the observation of species 1aEHA with the lowanisotropic g-factor, we examined the effect of a broad set of

carboxylic acids (Chart 2) on the EPR parameters of the active species formed in the catalyst systems 1/H2O2/RCOOH. In the presence of EHA, EBA and PVA, the EPR study revealed a close similarity of the EPR parameters of the extremely unstable (τ1/2 = 4–5 min at −75 °C) active species of the type 1aEHA (g1 = 2.069, g2 = 2.007, g3 = 1.961–1.963; Table S1, Supporting Information). All these systems also exhibited EPR spectra of stable complexes of the type 1b (g1 = 2.75, g2 = 2.41, g3 = 1.64) (Figure 2B). The situation was drastically different for catalyst systems 1/H2O2/RCOOH, where RCOOH were unbranched carboxylic acids, i.e. AA, BA, CA (Chart 2): in these cases, only the active species of the type 1aAA (with large g-factor anisotropy) were observed (Figures S2, S3, Table S1, Supporting Information). The observed maximum concentration of the intermediates of the type 1aAA was not higher than 3% of total iron concentration. The distinctive feature of EHA, EBA, and PVA, on the one hand, and AA, BA, and CA, on the other hand, is the presence of either tertiary α-carbon atom (EHA, EBA, and PVA), or secondary (BA, CA), or primary (AA) α-carbon atom. It would be logical to conclude that the presence of tertiary αcarbon atom in the acid is crucial for the formation of the active species with small g-factor anisotropy. 2.42 2.66

1aAA

1aAA

*

* 2.75

2.41

A)

1.71

2.069

1aAA

1aPVA 1aEBA 1aEHA 2.007 1.961–1.963

1.64

1bEBA 1bPVA 1bEHA

2500

B)

1bEBA 1bPVA 1bEHA

3000

3500

4000

4500

H/G

Figure 2. (A) EPR spectrum (−196 °C) of the sample 1/H2O2/CH3COOH ([1]:[H2O2]:[CH3COOH] = 1:3:10, [1] = 0.04 M) frozen 1 min after mixing the reagents at −75 °C in a 1.8:1 CH2Cl2/CH3CN mixture and storing the sample at −85 °C for 1 min. Signals denoted by asterisks belong to previously described ferric hydroxo complex [((S,S)-PDP)FeIII–OH(CH3CN)]2+ (g1 = 2.44, g2 = 2.21, g3 = 1.89, ref. 3a). (B) EPR spectra (−196 °C) of the samples 1/H2O2/RCOOH ([1]:[H2O2]:[RCOOH] = 1:3:10, [1] = 0.04 M) frozen 2.5 min after mixing the reagents at −65 °C in a 1.8:1 CH2Cl2/CH3CN mixture (RCOOH = PVA, red; EBA, green; EHA, blue).

ACS Paragon Plus Environment

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

At the same time, the presence of tertiary α-carbon is a necessary but not sufficient condition: with isobutyric acid (IBA) and cyclohexanecarboxylic acid (CHA) (Chart 2), only the intermediates similar to 1aAA, with large g-factor anisotropy (g1 = 2.69–2.72, g2 ≈ 2.42, g3 = 1.66–1.67) were observed (Figure S4, Table S1, Supporting Information). These intermediates of the type 1aAA disappeared with τ1/2 ≈ 5 min at −80 °C, which decay was accompanied by formation of stable complexes with EPR signatures similar to those of 1b-type species (g1 = 2.75, g2 ≈ 2.42, g3 = 1.64). One can conclude that the electronic/spin structure of the active species operating in the catalyst systems 1/H2O2/RCOOH crucially depends on the structure of R. While the use of carboxylic acids with primary and secondary α-carbons leads to the intermediates with large g-factor anisotropy, the use of carboxylic acids with tertiary α-carbon (e.g. EHA, EBA, PVA) can favor the formation of intermediates with small g-factor anisotropy. At the same time, in spite of branching at the α-carbon, the acids with short alkyl groups (IBA), or of cyclic structure (CHA), were unable to generate the intermediates with low g-factor anisotropy. Catalyst systems 1/H2O2/RCOOH were tested in the epoxidation of chalcone (Table 2). The systems exhibiting the active species with small g-factor anisotropy (i.e. with RCOOH = EHA, EBA, PVA) showed enantioselectivities noticeably higher than those featuring active species with large g-factor anisotropy (i.e. with RCOOH = AA, BA, CA, IBA, CHA), with the ratios of the rates of formation of the major and minor enantiomers ranging from 4.26 (with AA) to 11.50 (with PVA). The epoxidation of benzylideneacetone in the presence of the same set of carboxylic acids demonstrated a congenial enantioselectivity pattern (Table 3).10

Page 4 of 8

Table 3. Effect of Carboxylic Acid Structure on Asymmetric Epoxidation of Benzylideneacetone with H2O2 Catalyzed by Complex 1a

entry

carboxylic acid b

conversion (%) / epoxide yield (%)

e.r. c

1

AA

50 / 39

61 : 39

2

BA

69 / 53

63 : 37

3

CA

67 / 50

64 : 36

4

IBA

72 / 59

68 : 32

5

CHA

83 / 71

69 : 31

6

EBA

79 / 68

76 : 24

7

EHA

81 / 70

77 : 23

8

PVA

86 / 75

79 : 21

a

At 0 °C, [substrate]:[H2O2]:[carboxylic acid] = 100 µmol : 200 µmol : 55 µmol, catalyst load 2 mol.%, oxidant was added by a syringe pump over 30 min, and the mixture was stirred for an additional 2.5 h followed by LC analysis. b For the structures and abbreviations of carboxylic acids see Chart 2. c Enantiomeric ratio. Absolute configuration of benzylideneacetone epoxide was not assigned. active species of the type 5a

Table 2. Effect of Carboxylic Acid Structure on Asymmetric Epoxidation of Chalcone with H2O2 Catalyzed by Complex 1a

b

conversion (%) / epoxide yield (%)

1

AA

77 / 75

81 : 19

2

BA

65 / 63

82 : 18

entry

a

carboxylic acid

2.069–2.070

2.007–2.008

1.957–1.958

e.r. c

3

CA

68 / 65

82 : 18

4

IBA

85 / 85

86 : 14

5

CHA

91 / 90

86 : 14

6

EBA

89 / 89

91 : 9

7

EHA

90 / 89

91 : 9

8

PVA

92 / 92

92 : 8

At 0 °C, [substrate]:[H2O2]:[carboxylic acid] = 100 µmol : 200 µmol : 55 µmol, catalyst load 1 mol.%, oxidant was added by a syringe pump over 30 min, and the mixture was stirred for an additional 2.5 h followed by LC analysis. b For the structures and abbreviations of carboxylic acids see Chart 2. c Enantiomeric ratio. Absolute configuration of chalcone epoxide was (2R,3S).

EHA PVA CHA IBA EBA BA CA AA

3100

3200

3300

3400

3500

3600

3700 H / G

Figure 3. EPR spectra (−196 °C) of the samples 5/H2O2/RCOOH ([Fe]:[H2O2]:[RCOOH] = 1:3:10, [5] = 0.04 M) frozen 1.5 min after mixing the reagents at −75 °C in a 1.8:1 CH2Cl2/CH3CN mixture and storing the samples at −85 °C for a few minutes to reach maximum concentration of the active species.

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

ACS Catalysis

Table 4. Asymmetric Epoxidation of Chalcone with H2O2 Catalyzed by Complex 5: Enantioselectivity vs. Concentration of the Active Species a

entry

carboxylic conversion (%) / acid b epoxide yield (%)

e.r. c

concentration of the active species d

1

AA

74 / 73

84 : 16

1.0

2

CA

48 / 46

85 : 15

1.0

3

BA

61 / 59

85 : 15

1.4

4

IBA

92 / 92

89 : 11

2.2

5

CHA

72 / 71

89 : 11

2.2

6

EBA

54 / 54

91 : 9

2.0

7

PVA

69 / 69

91 : 9

2.6

8

EHA

73 / 73

92 : 8

3.0

a

At 0 °C, [substrate]:[H2O2]:[carboxylic acid] = 100 µmol : 200 µmol : 55 µmol, catalyst load 0.5 mol.% (1 mol.% Fe), oxidant was added by a syringe pump over 30 min, and the mixture was stirred for an additional 2.5 h followed by LC analysis. b For the structures and abbreviations of carboxylic acids see Chart 2. c Enantiomeric ratio. Absolute configuration of chalcone epoxide was (2R,3S). d Maximum concentration of the active species calculated from EPR spectra, in % of total iron concentration in the sample.

In contrast to catalyst systems 1/H2O2/RCOOH, the systems 5/H2O2/RCOOH exhibited only the intermediates with small g-factor anisotropy (5aAA, 5aBA, 5aCA, 5aIBA, 5aCHA, 5aEBA, 5aPVA, 5aEHA), irrespective of the carboxylic acid used (Figure 3 and Table S1, Supporting Information). The EPR parameters of these intermediates were virtually identical; however, the maximum concentrations of the observed intermediates generated under similar conditions were different for different acids (Figure 3). The enantioselectivities demonstrated by the systems 5/H2O2/RCOOH well correlated with the observed maximum concentrations of the intermediates of the type 5a in these systems (Table 4). This is a clear illustration for the widely-used Hammond-Leffler principle, which for the present catalyst system may be expressed as follows: the less reactive oxygen-transferring species (i.e. those approaching higher steady-state concentration) leads to more product-like transition states, resulting in higher epoxidation enantioselectivity. The noticeably higher epoxidation enantioselectivities (Tables 2 and 3) of catalysts systems 1/H2O2/(RCOOH = EHA, EBA, PVA), exhibiting intermediates of the type 1aEHA, as compared to the systems 1/H2O2/(AA, BA, CA), displaying intermediates of the type 1aAA, can be due to the additional stabilization of the active species via unpaired electron delocalization over the carboxylic moiety, which leads to more product-like transition states (vide supra) and eventually to higher enantioselectivities. We speculate that this stabilization effect may be more significant than the steric effect caused by the carboxylic moieties. Indeed, the use of EBA and CHA (with close steric properties, Chart 2) leads to noticeably different epoxidation enantioselectivities in the catalyst systems

1/H2O2/RCOOH (e.r.(EBA)/e.r.(CHA) = 1.7 and 1.4, cf. entries 5 vs. 6 of Table 2 and 5 vs. 6 of Table 3), which is ascribed to the different nature of the observed active species, 1aEBA (small g-factor anisotropy) and 1aCHA (large g-factor anisotropy). At the same time, for the catalyst systems 5/H2O2/EBA and 5/H2O2/CHA, this stabilization effect could not be significant since the PDP* ligand itself promotes the formation of the intermediate with small g-factor anisotropy. In the issue, the observed enantioselectivity difference was much smaller (e.r.(EBA)/e.r.(CHA) = 1.2, entries 5 and 6 of Table 4). Let us now discuss the nature of the proposed active species of epoxidation. The intermediate 5aAA with small g-factor anisotropy was previously assigned to species [((S,S)PDP*)•FeIV=O(OC(O)CH3)]2+ exhibiting antiferromagnetic coupling between the FeIV=O center (S = 1) and the ligand cation radical (S = 1/2) (Scheme 1).11 Recently, DFT calculations of Shaik and co-workers showed the possibility of interconversion of the S = 1/2 intermediate (L)FeIV=O(•OC(O)CH3) (L = TPA*) into its (L)FeV=O(OC(O)CH3) electronic isomer.7b Considering our data, the observed strong dependence of the electronic/spin structure of the active species in catalyst systems 1/H2O2/RCOOH on the structure of the carboxylic acid RCOOH suggest that intermediate of the type 1aEHA with low g-factor anisotropy may be better represented as [((S,S)PDP)FeIV=O(•OC(O)R)]2+, the latter featuring preferential localization of the unpaired electron at the Fe-coordinated branched carboxylic moiety with tertiary α-carbon atom (Scheme 1). The ease of conversion between intermediates with high and low g-factor anisotropy (just the replacement of acetic acid by e.g. 2-ethylhexanoic acid as additive) gives evidence in favor of close structures of the intermediates with high and low g-factor anisotropy.

Scheme 1. Plausible Mechanistic Scenarios for the Formation of the Active Species of the Type 1aEHA and 5a (with Small g-Factor Anisotropy)11

In summary, the data reported here show that, depending on the nature of added carboxylic acid, two types of active species can be detected in bioinspired catalyst systems Fe(PDP)/H2O2/RCOOH. The use of unbranched acids (acetic, butyric, caproic) results in the formation of intermediates with large g-factor anisotropy (g1 ≈ 2.7, g2 ≈ 2.4, g3 ≈ 1.7), while the use of branched carboxylic acids featuring tertiary αcarbon (2-ethylhexanoic, 2-ethylbutyric, valproic) triggers the formation of the intermediates with small g-factor anisotropy.

ACS Paragon Plus Environment

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

The latter intermediates are more enantioselective, and can be assigned to oxo-iron species with iron in formally +5 oxidation state, [((S,S)-PDP)FeV=O(OC(O)R)]2+ or [((S,S)PDP)FeIV=O(•OC(O)R)]2+. Their higher enantioselectivity can be caused by the additional stabilization via delocalization of the unpaired electron over the carboxylic moiety. This work represents the first case where the transient, formally “FeV=O” species with low g-factor anisotropy, responsible for the enantioselective epoxidations have been directly observed in the family of biomimetic catalyst systems Fe(PDP)/H2O2/RCOOH. The unprecedented dependence of the electronic structure of the reported intermediates on the nature of carboxylic acid, which correlates with the epoxidation enantioselectivity, delineates the directions for rational tuning the stereoselectivity of bioinspired iron and structurally related manganese based catalysts.3a,3d,8a

ASSOCIATED CONTENT Experimental details, additional data on the catalytic epoxidation of chalcone, extended EPR data, EPR spectra. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author * E-mail: [email protected].

Notes The authors declare no competing financial interest.

ACKNOWLEDGMENT Complexes 1 and 5 were synthesized, carboxylic acids (Chart 2) purchased, and catalytic studies were conducted with the aid of the Russian Science Foundation (#14-13-00158). EPR studies were performed using the equipment of the Russian Academy of Sciences and Federal Agency of Scientific Organizations (project V.44.2.4).

REFERENCES (1) (a) Costas, M.; Mehn, M. P.; Jensen, M. P.; Que, Jr., L. Chem. Rev. 2004, 104, 939–986. (b) Ortiz de Montellano, P. R. Chem. Rev. 2010, 110, 932–948. (c) Costas, M. Coord. Chem. Rev. 2011, 255, 2912–2932. (d) Company, A.; Gómez, L.; Costas M. In Iron containing enzymes: Versatile catalysts of hydroxylation reactions in Nature; de Visser, S. P.; Kumar D., Eds.; RCS publishing: Cambridge, 2011, pp 148–208. (e) Talsi, E. P.; Bryliakov, K. P. Coord. Chem. Rev. 2012, 256, 1418–1434. (f) Groves, J. T. Nat. Chem. 2014, 6, 89–91. (g) Sorokin, A. B. Chem. Rev. 2013, 113, 8152–8191. (h) Bryliakov, K. P. Environmentally Sustainable Catalytic Asymmetric Oxidations, CRC Press: Boca Raton, 2014, pp 109–142. (i) Cussó, O.; Ribas, X.; Costas, M. Chem. Commun. 2015, 51, 14285–14298. (2) (a) Chen, M. S.; White, M. C. Science 2007, 318, 783–787. (b) Vermeulen, N. A.; Chen, M. S.; White, M. C. Tetrahedron 2009, 65, 3078–3084. (c) Chen, M. S.; White, M. C. Science 2010, 327, 566–

Page 6 of 8

571. (d) Bigi, M. A.; Reed, S. A.; White, M. C. Nat. Chem. 2011, 3, 216–222. (e) White, M. C. Science 2012, 335, 807–809. (f) Lindsay Smith, J. R.; Shul’pin, G. B. Tetrahedron Lett. 1998, 39, 4909–4912. (g) Romakh, V. B.; Therrien, B.; Süss-Fink, G.; Shul’pin, G. B. Inorg. Chem. 2007, 46, 1315–1331. (h) Nizova, G. V.; Shul’pin, G. B. Tetrahedron 2007, 63, 7997–8001. (i) Shul’pin, G. B.; Matthes, M. G.; Romakh, V. B.; Barbosa M. I. F.; Aoyagi, J. L. T.; Mandelli, D. Tetrahedron 2008, 64, 2143–2152. (j) Shul’pin, G. B.; Kozlov, Y. N.; Shul’pina, L. S.; Strelkova, T. V.; Mandelli, D. Catal. Lett. 2010, 138, 193–204. (k) Mandelli, D.; Kozlov, Y. N.; Carvalho, W. A.; Shul’pin, G. B. Catal. Commun. 2012, 26, 93–97. (3) (a) Lyakin, O. Y.; Ottenbacher, R. V.; Bryliakov, K. P.; Talsi, E. P. ACS Catal. 2012, 2, 1196–1202. (b) Cussó, O.; Garcia-Bosch, I.; Ribas, X.; Lloret-Fillol, J.; Costas, M. J. Am. Chem. Soc. 2013, 135, 14871–14878. (c) Cussó, O.; Ribas, X.; Lloret-Fillol, J.; Costas, M. Angew. Chem. Int. Ed. 2015, 54, 2729–2733. (d) Ottenbacher, R. V.; Samsonenko, D. G.; Talsi, E. P.; Bryliakov, K. P. ACS Catal. 2016, 6, 979–988. (4) (a) Lyakin, O. Y.; Bryliakov, K. P.; Britovsek, G. J. P.; Talsi, E. P. J. Am. Chem. Soc. 2009, 131, 10798–10799. (b) Lyakin, O. Y.; Bryliakov, K. P.; Talsi, E. P. Inorg. Chem. 2011, 50, 5526–5538. (5) We note that the other example of catalytically active oxoferryl(V) species with high g-factor anisotropy, [(Me2PyTACN)FeV=O(OH)]2+, was reported in (a) Prat, I.; Mathieson, J. S.; Güell, M.; Ribas, X.; Luis, J. M.; Cronin, L.; Costas M. Nat. Chem. 2011, 3, 788–793. (b) Lyakin, O. Y.; Prat, I.; Bryliakov, K. P.; Costas, M.; Talsi, E. P. Catal. Commun. 2012, 29, 105–108. (6) (a) Van Heuvelen, K. M.; Fiedler, A. T.; Shan, X.; De Hont, R. F.; Meier K. K.; Bominaar, E. L.; Münck, E.; Que, L., Jr. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 11933–11938. (b) De Oliveira, F. T.; Chanda, A.; Banerjee, D.; Shan, X.; Mondal, S.; Que, L., Jr.; Bominaar, E. L.; Münck, E.; Collins, T. J. Science 2007, 315, 835–838. (7) (a) Wang, Y.; Janardanan, D.; Usharani, D.; Han, K.; Que, Jr., L.; Shaik, S. ACS Catal. 2013, 3, 1334–1341. (b) Oloo, W. N.; Meier, K. K.; Wang, Y.; Shaik, S.; Münck, E.; Que, Jr., L. Nat. Commun. 2014, 5, 3046–3055. (8) (a) Lyakin, O. Y.; Zima, A. M.; Samsonenko, D. G.; Bryliakov, K. P.; Talsi, E. P. ACS Catal. 2015, 5, 2702–2707. (b) 5aAA was active toward electron-rich alkenes even at −85 °C. (9) Serrano-Plana, J.; Oloo, W. N.; Rueda-Acosta, L.; Meier, K. K.; Verdejo, B.; García-España, E.; Basallote, M. G.; Münck, E.; Que, Jr., L.; Company, A.; Costas, M. J. Am. Chem. Soc. 2015, 137, 15833– 15842. (10) To probe chiral “matching-mismatching” effect for the asymmetric epoxidation of chalcone, several experiments were performed, using enantiomeric (R,R)- and (S,S)-forms of catalysts 1 and 5, and enantiomerically pure catalytic additives. The influence of switching the chiralities of the catalyst and/or the additives on the epoxide yield and enantioselectivity for the above catalysts was insignificant. See Supporting Information (Table S2 and the corresponding discussion) for details. (11) The mechanism of the formation of active species of the type 5a (with small g-factor anisotropy) was previously discussed in ref. 8a.

ACS Paragon Plus Environment

Page 7 of 8

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 ACS Paragon Plus Environment

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

TOC graphic in TIFF format 591x255mm (96 x 96 DPI)

ACS Paragon Plus Environment

Page 8 of 8