Mechanism of N,N,N-Amide Ruthenium(II) Hydride Mediated Acceptorless Alcohol Dehydrogenation: Inner-Sphere β-H Elimination versus Outer-Sphere Bifunctional Metal-Ligand Cooperativity

Research Article

pubs.acs.org/acscatalysis

Mechanism of N,N,N‑Amide Ruthenium(II) Hydride Mediated

Acceptorless Alcohol Dehydrogenation: Inner-Sphere β‑H

Elimination versus Outer-Sphere Bifunctional Metal−Ligand

Cooperativity

Kuei-Nin T. Tseng, Jeﬀ W. Kampf, and Nathaniel K. Szymczak*

Department of Chemistry, University of Michigan, Ann Arbor, Michigan 48109, United States

*

S Supporting Information

ABSTRACT: The reversible transformations between ke-

tones and alcohols via sequential hydrogenation−dehydrogen-

ation reactions are eﬃciently achieved using a single

precatalyst HRu(bMepi)(PPh ) (bMepi = 1,3-bis(6′-methyl-

3

2

′-pyridylimino)isoindolate). The catalytic mechanism of

HRu(bMepi)(PPh ) mediated acceptorless alcohol dehydro-

3

2

genation (AAD) has been investigated by a series of kinetic

and isotopic labeling studies, isolation of intermediates, and

evaluation of Ru(b4Rpi)(PPh ) Cl (R = H, Me, Cl, OMe,

3

2

OH) complexes. Two limiting dehydrogenation scenarios are

interrogated: inner-sphere β-H elimination and outer-sphere

bifunctional double hydrogen transfer. Isotopic labeling

experiments demonstrated that the proton and hydride

transfer in a stepwise manner. Catalyst modiﬁcations suggest that the imine group on the bMepi pincer scaﬀold is not

necessary for catalytic alcohol dehydrogenation. Evaluation of the kinetic experiments and catalyst modiﬁcations suggests a

pathway whereby HRu(bMepi)(PPh ) operates via the inner-sphere β-H elimination mechanism. Following a single PPh

3

2

3

dissociation, an alcohol substrate can bind and undergo proton transfer followed by a turnover-limiting β-H elimination step.

⧧

Analysis of the Eyring plot established activation parameters for the β-H elimination reaction as ΔH = 15(1) kcal/mol and ΔS

=

−41(3) eu. AAD reactions using a series of Ru(b4Rpi)(PPh ) Cl complexes indicated that the ortho-substituted methyl groups

of bMepi slightly impede catalytic activity, and electronic modiﬁcations of the pincer scaﬀold have a minimal eﬀect on the

3

2

reaction rate.

KEYWORDS: acceptorless alcohol dehydrogenation, ruthenium, ligand eﬀects, inner-sphere mechanism, outer-sphere mechanism,

metal−ligand cooperativity

INTRODUCTION

with the ligand via aromatization−dearomatization of the

central pyridinyl group concomitant with protonation−

■

Transition-metal-catalyzed acceptorless alcohol dehydrogen-

ation (AAD) with the liberation of H is an atom-economical

1a,4

deprotonation of the methylene arm (Scheme 2, left panel).

2

Computational studies revealed that 1 favors an outer-sphere

and selective route to generate a variety of organic carbonyl

1

bifunctional double hydrogen-transfer pathway rather than an

synthons. In the context of the “hydrogen energy economy”,

5

inner-sphere β-H elimination process. More recently, a

AAD also provides a highly desirable strategy for promoting H₂

release from suitable biomass feedstocks for chemical energy

computational study by Yamaguchi, Fujita, and co-workers

demonstrated that the AAD reaction of benzyl alcohol

catalyzed by Cp*Ir(bpyO)(2, bpyO = α,α′-bipyridonate) also

operates via an outer-sphere pathway, where the metal center

and the bipyridonate motif work synergistically to oxidize

2

storage applications.

To achieve high atom economy (no exogenous additives),

promoterless AAD reactions are most often mediated by

bifunctional catalysts that operate via a metal−ligand

cooperative mechanism. This ligand-assisted, transition-metal-

catalyzed process diﬀers from the classical inner-sphere

mechanism by not requiring coordination of the substrate,

thus enabling outer-sphere proton transfer to a ligand-based

benzyl alcohol en route to H elimination with the aid of an

2

6

alcohol bridge (Scheme 2, right panel). Collectively, these and

related studies establish the importance of a cooperative

mechanism to achieve eﬃcient dehydrogenation activity.

basic site with concurrent hydride transfer to the metal center

3

(Scheme 1). For example, Milstein’s group developed a series

Received: May 6, 2015

Revised: August 4, 2015

of pyridyl PNE (E = PR or NR ) Ru pincer complexes (1,

2

HRu(PNE)(CO)) that employ cooperation of the metal center

©

XXXX American Chemical Society

5468

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Scheme 1. Generalized Catalytic Cycles for the Inner-Sphere

and Outer-Sphere Dehydrogenation Pathways

intermediates in the dehydrogenation catalytic cycle? (3) What

impact do the steric (ortho-substituted methyl groups) and

electronic (electron donating and withdrawing groups in the

secondary coordination sphere) proﬁles of the bMepi pincer

ligand have on alcohol dehydrogenation?

RESULTS AND DISCUSSION

■

3

Limiting Mechanistic Scenarios for AAD Catalyzed by

8

. Based on prior PPh studies and the observation of a ﬁrst-

3

order dependence of the rate of 1-phenylethanol (1PhEtOH)

10

dehydrogenation on [3], two limiting monometallic alcohol

dehydrogenation pathways mediated by 3 are proposed in

Scheme 3. For either pathway, phosphine dissociation from 3

Scheme 3. Proposed Inner-Sphere and Outer-Sphere

Dehydrogenation Pathways

However, in systems that contain bifunctional groups, it may be

ambiguous whether a cooperative pathway is actually required

for eﬃcient dehydrogenation.

7

We recently reported an N,N,N-bMepi (bMepi =1,3-bis(6′-

II

methyl-2′-pyridylimino)isoindolate) Ru hydride complex (3,

HRu(bMepi)(PPh ) ) capable of catalyzing promoterless and

3

2

8

,9

chemoselective AAD reactions. Of particular note, precatalyst

promotes acceptorless dehydrogenation of secondary alcohols

3

to ketones, acceptorless dehydrogenative coupling of primary

alcohols to esters, and diols to lactone products with high

conversion eﬃciencies. Importantly, neither of these reactions

requires exogenous base or hydrogen acceptor additives, and

the catalyst system is unusually selective for the dehydrogen-

ation of secondary alcohols in the presence of primary alcohols.

Preliminary analysis of the alcohol dehydrogenation reaction

revealed two key ﬁndings: (1) a homogeneous active catalyst, as

assessed by mercury and substoichiometric ligand poisoning

experiments, and (2) the release of PPh₃under catalytic

conditions. In this article, we use these observations as an entry

point to disclose a detailed mechanistic analysis of a series of

kinetic rate data including isotopic labeling studies, stoichio-

metric reactions to probe catalytic intermediate species, and

new ligand variants to understand the steric and electronic

eﬀects of the bMepi pincer ligand on the activity of the Ru

complex. We aim to answer the following key questions: (1)

Does precatalyst 3 participate in an inner- or outer-sphere

dehydrogenation pathway? (2) What are the details of the

Scheme 2. Proposed AAD Reaction Pathway Mediated by 1 and 2

5

469

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

generates a coordinately unsaturated Ru species that is able to

participate in either an inner-sphere β-H elimination pathway

(

Scheme 3, top panel) or an outer-sphere concerted pathway

Scheme 3, bottom panel). In the inner-sphere cycle, proton

transfer from the alcohol to the Ru hydride aﬀords a Ru-

alkoxide species (likely via a transient Ru-H intermediate),

2

which undergoes β-H elimination to complete the cycle. An

alternative pathway to the inner-sphere β-H elimination

mechanism is the outer-sphere pathway, where both proton

and hydride transfer occur without requiring coordination to

Ru. Complex 3 operating via this pathway may involve proton

transfer to the imine (or isoindolate) group on the bMepi-

ligand backbone with concurrent hydride transfer to the Ru-

metal center. Both of these AAD mechanistic scenarios are

evaluated by a series of kinetic experiments, catalyst

modiﬁcations, as well as isolation of proposed intermediates.

Standard Conditions for Kinetic Studies. In order to

examine the operative pathway for catalysis by 3, 1PhEtOH was

selected as a standard substrate because its low volatility

permits heating in an open system. Additionally, the reverse

reaction, reduction of acetophenone to 1PhEtOH, is generally

accepted as a standard test for (transfer) hydrogenation

Figure 1. Inﬂuence of [PPh ] on the reaction rates for 1PhEtOH

dehydrogenation catalyzed by 3.

3

The zero-order [PPh ] dependence (up to 20 equiv)

3

suggests that PPh dissociation from 3 is not included in the

3

turnover-limiting step and furthermore that phosphine bind-

ing/release is not in an equilibrium with the turnover-limiting

step under these conditions. Therefore, the turnover-limiting

step must be either β-H elimination (inner-sphere) or H₂

formation reaction (outer-sphere). Alternatively, for this to be

true for a proton/hydrogen-transfer turnover-limiting step, the

alcohol binding would have to be irreversible, which is highly

3

,11

catalysis.

The dehydrogenation reaction was performed in

an open system inside an inert-atmosphere glovebox, and the

1

conversion of 1PhEtOH to acetophenone was monitored by H

NMR spectroscopy using phenyltrimethylsilane (PhTMS) as an

internal standard (Scheme 4). The observed reaction rates were

12

Scheme 4. Standard Reaction Conditions of AAD of

improbable.

-Phenylethanol Dependence. In both the inner- and

1

PhEtOH Catalyzed by 3

1

outer-sphere dehydrogenation scenarios (Scheme 3), the next

step following PPh dissociation involves the alcohol substrate.

3

A rate dependence on [1PhEtOH] would be anticipated if the

turnover-limiting step was alcohol binding followed by

deprotonation (inner-sphere) or proton and hydride transfer

from the alcohol (outer-sphere). The inﬂuence of 1PhEtOH

concentration on the catalytic rate was examined by changing

the [1PhEtOH] while holding the initial concentration of 3

constant. Over the range of 6.5−8 M 1PhEtOH, the observed

reaction rate proﬁle displayed a linear dependence on

[1PhEtOH] from 6.5 to 7.5 M 1PhEtOH, then the rate

reached culmination after 7.5 M 1PhEtOH with an averaged

obtained using the method of initial rates. All kinetic

experiments were simultaneously performed in triplicate.

Standard reaction conditions for kinetic studies employed a

vial containing 7.5 M 1PhEtOH and 0.01 mol % 3. After the

reaction mixture was heated to 120 °C for 4 h, acetophenone

was observed in 11.6% conversion, which corresponds to an

catalytic rate of 5.2(2) × 10⁻M·s (Figure 2). The reaction

rate dependence on [1PhEtOH] suggests a pre-equilibrium

model in which at high [1PhEtOH] the equilibrium is driven to

the right, and the dependence on [1PhEtOH] drops from the

rate law. This model could ﬁt either the inner- or outer-sphere

pathways, where a pre-equilibrium alcohol binding and proton/

hydrogen transfer is followed by a slow metal-based reaction

(e.g., β-H elimination or H₂formation). Given that two

diﬀerent mechanistic regimes exist at high and low [1PhEtOH],

it is important to note that the kinetic experiments were

performed at an initial [1PhEtOH] of 7.5 M (Scheme 4, unless

otherwise stated), and under these conditions, the catalyst

operated in the linear regime when the reaction proceeded 10−

15%.

5

−1

13

−5

−1

initial rate of 5.2(2) × 10 M·s , a turnover number (TON)

−1

of 1213, and a turnover frequency (TOF) of 303 h . To

establish conﬁdence in the method of initial rates in this system,

−5

−1

a reaction rate of 5.8 × 10 M·s (within 10% error of the

initial rate) at 4 h was obtained from the ﬁrst derivative of an

exponential ﬁ t of the complete dehydrogenation reaction

proﬁle (see Figure S2).

Triphenylphosphine Dependence. Based on our prior

studies that showed free PPh during catalysis, the dependence

3

on PPh concentration was examined to determine whether

3

PPh dissociation contributed to a turnover-limiting step in

3

either of the proposed alcohol dehydrogenation cycles. The

Temperature Dependence. The activation parameters for

1PhEtOH dehydrogenation mediated by 3 at low and high

[1PhEtOH] were analyzed to interrogate the transition-state

structures. The reaction rates were measured over a 40 °C

temperature range and plotted according to an Eyring analysis

(Figure 3). In the low [1PhEtOH] regime, analysis of the

order in [PPh ] was determined by measuring the observed

3

rates for 1PhEtOH dehydrogenation over several PPh₃

concentrations under the reaction conditions listed in Scheme

4

. No dependence on the rate of 1PhEtOH dehydrogenation

was observed up to 20 equiv of PPh (14.5 mM) relative to 3

3

⧧

(Figure 1).

Eyring plot revealed a free energy activation barrier (ΔG ) of

5

470

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Isotopic Labeling Studies. A series of deuterium isotopic

substitutions of 1PhEtOH were used to interrogate proton and

hydride transfer and inner- versus outer-sphere pathways. The

scenario of 1PhEtOH deprotonation as the turnover-limiting

step was examined by monitoring the dehydrogenation of the

1

PhEtOH isotopologue, 1PhEtOD, catalyzed by 3. A normal

primary kinetic isotope eﬀect (KIE) would be anticipated if the

O−H bond cleavage is involved in the turnover-liming step or

precedes it. Dehydrogenation of 1PhEtOD at 120 °C in the

presence of 0.01 mol % 3 yielded an observed rate of 2.7(1) ×

−5

−1

1

0

M·s (Scheme 6, eq 4). This reduced reaction rate,

compared to the rate for the perprotio isotopologue, aﬀorded a

KIE (r_O_H_C_H/r_O_D_C_H) of 1.9(2), thus supporting either O−H

bond cleavage in the turnover-limiting step (inner- or outer-

sphere) or a β-H elimination turnover-limiting step with a

proton-transfer pre-equilibrium (inner-sphere). Hence, it

should be noted that the observed isotope eﬀect is likely a

Figure 2. Inﬂuence of [1PhEtOH] on the reaction rates for 1PhEtOH

dehydrogenation catalyzed by 3.

18

composite of both equilibrium and kinetic isotope eﬀects.

Another set of isotopic labeling experiments were performed

with a second isotopologue of 1PhEtOH, 1PhCH CDOH. An

3

observed reaction rate of 3.6(1) × 10⁻M·s was obtained for

5

−1

⧧

3

1

1(3) kcal/mol at 120 °C, an activation enthalpy (ΔH ) of

⧧

the dehydrogenation of 1PhCH CDOH at 120 °C resulting in

8(1) kcal/mol, and an activation entropy (ΔS ) of −32(3)

eu. In the high [1PhEtOH] regime, analysis of the Eyring plot

3

1

4

^a^K^I^E⁽^rOHCH^/^rOHCD) of 1.4(1) (Scheme 6, eq 5). This

observed KIE is consistent with the cleavage of the C−H bond

in the turnover-limiting step (β-H elimination or outer-sphere

pathway) and is too large for a secondary isotope eﬀect.

Depending on the nature of the transition state, varying

magnitudes of KIE (>1.3) have been measured for β-H

⧧

revealed a ΔG of 31(3) kcal/mol at 120 °C, an ΔH of 15(1)

⧧

kcal/mol, and an ΔS of −41(3) eu. An activation enthalpy of

this magnitude is consistent with bond-breaking character in

15

the turnover-limiting transition structure. In addition, the

relatively large negative entropy of activation suggests not only

an associative process but also a higher degree of organization

in the transition state than in the ground state. Analysis of the

activation parameters at low and high [1PhEtOH] revealed

equivalent ΔG values. However, between the low and high

1PhEtOH] regimes, the magnitude of ΔH decreases, whereas

that of ΔS increases. This implies that at high [1PhEtOH] the

AAD catalysis is entropically controlled. Unfortunately, these

Eyring data could not be used to unambiguously diﬀerentiate

between the two proposed mechanisms. For example, a highly

ordered transition structure could be expected for an alcohol-

assisted proton/hydrogen-transfer turnover-limiting step in the

18c

elimination from metal-alkoxides. However, a KIE with a

larger magnitude (2.6) was observed for an outer-sphere

19

concerted pathway. In addition, the measured KIE is also

consistent with a proton-transfer turnover-limiting step where a

Ru−H(D) species could participate in deprotonation of the

alcohol after the ﬁrst turnover in an inner-sphere pathway.

The outer-sphere concerted pathway can be evaluated using

⧧

[

⧧

̈

a series of isotopic labeling experiments. For example, Backvall

and Johnson demonstrated that Shvo’s catalyst operated via an

outer-sphere concerted mechanism by analyzing the KIE for the

doubly deuterium labeled isotopologue of 1PhEtOH,

16

19

two pathways as proposed in Scheme 5. Furthermore, a range

of ΔS values (+12 to −30 eu) have been reported for a β-H

elimination turnover-limiting step from metal-alkoxide spe-

1PhCH CDOD. For a concerted pathway, the observed

3

⧧

isotope eﬀect for 1PhCH CDOD should be the product of the

3

two individual isotope eﬀects (r_O_H_C_H/r_O_D_C_H× r_O_H_C_H/r_O_H_C_D).

1

7

cies.

When 1PhCH CDOD was subjected to the standard

3

Figure 3. Eyring plots for 1PhEtOH dehydrogenation catalyzed by 3. Left panel: [1PhEtOH] = 7.5 M. Right panel: [1PhEtOH] = 8.2 M.

0

5

471

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Scheme 5. Proposed 1PhEtOH Deprotonation with 1PhEtOH as a Proton-Transfer Shuttle in the Inner- and Outer-Sphere

Pathways

Me

Scheme 6. Isotopic Labeling Experiments for the

Dehydrogenation of 1PhEtOH

Ru(bMepi )(PPh )OTf (5) was prepared by treating a

3 2

DCM solution of Ru(bMepi)(PPh )Cl (4) with 10 equiv of

3

MeOTf at room temperature for 18 h. Complex 5 was isolated

1

31

in 62% yield and characterized by H and P NMR

spectroscopy, elemental analysis, and X-ray crystallography.

31

1

The P{ H} spectrum displays a singlet at 47.8 ppm, and the

1

H NMR spectrum is consistent with the asymmetry of the

Me

bMepi ligand. In particular, two distinct resonances for the

ortho-CH groups were located at 1.57 and 1.78 ppm and a

3

singlet was detected at 3.71 ppm and assigned as the methyl

group on the ligand backbone. Complex 5 has a slightly

22

distorted square-based pyramid (τ = 0.10) structure, which

contains a triﬂate anion trans to the PPh ligand (Figure 4).

3

−

5

In addition to the covalent methylated imine bond in 5,

noncovalent interactions of the ortho-methyl groups were

found. Agostic M−H−C interactions are characterized by

dehydrogenation conditions, a reaction rate of 2.6(1) × 10

−1

M·s was observed, providing a kinetic isotope eﬀect (r_O_H_C_H

/

r_O_D_C_D) of 2.0(2) (Scheme 6, eq 6). In contrast, the product of

the individual isotope eﬀects is 2.7 (1.9 × 1.4) and thus

inconsistent with the measured combined KIE. Furthermore, to

normalize for any secondary isotope eﬀects on the C−H KIE,

we averaged the r

/r

(1.44) and r

/r

(1.04)

OHCH OHCD

ODCH ODCD

values to provide a C−H KIE of 1.24. The product of the O−H

and averaged C−H KIE is 2.4(1), which also does not match

the observed r

/r

. These analyses provide strong

OHCH ODCD

evidence against pathways in which both proton and hydride

transfer in a concerted manner and are in support of an inner-

sphere, stepwise pathway.

Me

Synthesis and Reactivity of Ru(bMepi )(PPh )OTf :

3

2

Possible Role of Metal−Ligand Cooperativity. To

complement the kinetic isotope studies that discounted an

II

outer-sphere concerted pathway, we targeted Ru compounds

Me

with the ligand bMepi , in which one of the imine groups is

methylated. Prior studies from our laboratory found that a late

stage modiﬁcation can be used to alkylate the imine backbone,

thus providing complementary complexes that feature similar

primary coordination environments yet diﬀer in charge of the

2

0

pincer ligand. Furthermore, the site of protonation or

21

Figure 4. Synthesis and crystal structure (thermal ellipsoids of 5

alkylation in these and related complexes is the imine

nitrogen, rather than the amido nitrogen, which indicates the

former as the favored kinetic site for protonation. Hence,

metal−ligand cooperative pathways involving proton transfer to

the central isoindoline nitrogen are unlikely AAD mechanisms

for the bis(pyridylimino)isoindolate framework.

Me

depicted at 50% probability) of Ru(bMepi )(PPh )OTf (5). The

3

2

outer-sphere triﬂate anion, PPh phenyl groups, and hydrogen atoms

3

are omitted for clarity. Selected bond distances (angstroms): Ru1−P1,

2

.2933(7); Ru1−N1, 2.081(3); Ru1−N3, 1.955(2); Ru1−N5,

2.063(3); Ru1−O3, 2.221(2); N2−C7, 1.488(4); N2−C8, 1.351(4);

N3−C8, 1.340(4); N3−C15, 1.410(4); and N4−C15, 1.289(4).

5

472

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

relatively short M−H distances (1.8 to 2.3 Å), small M−H−C

bond angles (90 to 140°), and upﬁeld chemical shifts of the

Scheme 7. Proposed Proton Transfer via Step-Wise Metal−

Ligand Cooperativity

23

agostic hydrogen atoms. All three criteria for an agostic

interaction are met in complex 5. The crystal structure of 5

24

reveals a short M−H distance of 2.29 Å and a small M−H−C

bond angle of 119° from one hydrogen of the ortho-CH₃

groups. In addition, the two singlet resonances for the methyl

groups are upﬁeld of the free HbMepi ligand (the hydrogen

atoms of the methyl groups on the free HbMepi ligand were

1

observed in the H NMR spectrum as a singlet at 2.50 ppm, Δ

=

0.93, 0.72 ppm). The analogous protonated complex

(

Ru(HbMepi)(PPh )Cl[PF ]) was also synthesized by heating

3 6

a THF solution containing HbMepi, RuCl (PPh ) , and TlPF

2

3 3

6

(Figure 5). In contrast to sharp ligand resonances observed in

species and could not undergo β-H elimination without losing

another ligand (or dissociation of the alkoxide or a pyridine

arm). Dissociation of the PPh ligand would deviate from the

3

observed zero-order [PPh ] dependence.

3

To assess the potential participation of the backbone imine

group on the bMepi ligand in the dehydrogenation of alcohols

via bifunctional metal−ligand catalysis, the dehydrogenation of

1

PhEtOH catalyzed by 5 was evaluated. Of key importance to a

bifunctional metal−ligand pathway is proton transfer to the

backbone imine nitrogen. This protonation event would seem

energetically unfavorable for the alkylated complex (Scheme 7,

eq 8) given that heating 4 in excess MeOTf aﬀorded only the

monomethylated complex, which suggests that the remaining

imine functionality is less basic in the bMepi ligand than in

the parent bMepi ligand.

The proposed hybrid metal−ligand cooperative pathway was

evaluated by comparing the reaction rates of 1PhEtOH

dehydrogenation catalyzed by 4 and 5 (Scheme 8). Heating a

Figure 5. Synthesis and crystal structure (thermal ellipsoids of 5

depicted at 50% probability) of Ru(HbMepi)(PPh )Cl[PF ]. The

3

6

outer-sphere PF anion, PPh phenyl groups, and hydrogen atoms,

6

3

except the NH, are omitted for clarity. Selected bond distances

Me

(

angstroms): Ru1−P1, 2.3356(4); Ru1−N1, 2.048(1); Ru1−N3,

1

.948(1); Ru1−N5, 2.100(3); Ru1−Cl1, 2.4696(3); N2−C7,

.283(2); N3−C7, 1.412(2); N3−C14, 1.329(2); and N4−C14

.339(2).

1

the H NMR spectrum of 5, the analogous protonated complex,

Scheme 8. Reaction Rate Comparison between the bMepi

Ru(HbMepi)(PPh )Cl[PF ], contains broad ligand resonances

3

6

Me

and bMepi Ligated Ruthenium Complexes

at room temperature, which are consistent with a dynamic

protonation equilibrium between the imine nitrogens. The

solid-state structure shows square-based pyramid geometry

about the Ru center with a chloride ligand trans to PPh . For

3

Ru(HbMepi)(PPh )Cl[PF ], two out of the three criteria are

3

6

met for an agostic interaction; the M−H distance is slightly

longer (2.44 Å). However, the M−H−C bond angle of 107°

and the upﬁeld shift of the ortho-CH groups (1.78 and 1.70

3

ppm; Δ = 0.72, 0.80 ppm) are consistent with an agostic M−

H−C interaction. Complex 4 exhibited similar structural and

spectroscopic properties as the protonated complex, such as the

M−H distance (2.41 Å), a M−H−C bond angle of 113°, and

the upﬁeld shift of the ortho-CH groups (1.72 ppm; Δ = 0.78

3

ppm).

Because the isotopic labeling studies were not consistent with

a concerted dehydrogenation pathway (vide supra), a stepwise,

hybrid metal−ligand cooperative pathway was evaluated. This

bifunctional hybrid mechanism is a combination of the inner-

sphere β-H elimination and the outer-sphere bifunctional

pathway, in which proton transfer takes place at the backbone

imine group on the bMepi ligand, aﬀording a Ru-alkoxide

intermediate (Scheme 7, eq 7) that could undergo β-H

7.8 M 1PhEtOH solution containing 0.024 mol % of 4 and

t

0.048 mol % NaO Bu to 120 °C for 4 h resulted in an averaged

TON of 1441, which corresponds to a reaction rate of 7.7(3) ×

−

5

−1

10 M·s . The alkylated Ru complex (5) was also a

competent dehydrogenation precatalyst. For instance, under

the same reaction conditions, complex 5 oxidized 1PhEtOH to

−

5

acetophenone and H with a reaction rate of 8.7(1) × 10 M·

2

−1

25

s . These results demonstrate that a cooperative interaction

involving the imine functionality is not necessary to achieve

elimination. Unless H is eliminated, this pathway is not

probable because the Ru-alkoxide intermediate is an 18 e

2

−

26

eﬃcient rates for catalytic AAD reaction.

5

473

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Hammett Studies. The electronic character of the

turnover-limiting transition state in catalytic AAD promoted

by 3 was investigated by conducting a linear free energy analysis

using initial rates of dehydrogenation of para-substituted

diﬀerence in magnitude of the ρ values may be explained by the

nature of the transition state. The smaller ρ value of −0.43

observed for Ru-hydroxyapatite indicates that electronic

changes have a subtle eﬀect on β-H elimination, which is

consistent with a late transition state with almost complete C−

H bond cleavage and Ru−H bond formation. For our system 3,

electron-donating groups increase the nucleophilicity of the

benzylic hydrogen atom, which acquires hydridic character

during β-H elimination. Stabilization of the positive charge

buildup at the benzylic carbon as the hydride is transferred in

the transition state suggests an early transition-state model.

Isolation of a Ruthenium(II)−Alkoxide Complex.

Following the kinetic experiments, which support an inner-

sphere β-H elimination pathway, stoichiometric reactions were

performed to examine the intermediate species during catalysis.

In particular, a Ru alkoxide species was implicated as the

catalyst resting state that undergoes β-H elimination in an

inner-sphere pathway. To trap such a species prior to β-H

elimination, triﬂuoroethanol, whose conjugate base is resistant

1PhEtOH substrates (Figure 6). The ρ value has previously

30

to β-H elimination, was selected. The addition of 1.1 equiv of

triﬂuoroethanol to a solution of 3 in THF resulted in the clean

conversion to Ru(bMepi)(PPh )(OCH CF ) (6), which was

3

isolated as a dark blue solid in 76% yield after heating at 70 °C

Figure 6. Hammett plot for 1PhEtOH dehydrogenation catalyzed by

1

for 2 days (Figure 7). The H NMR spectrum features a single

3.

been used to diﬀerentiate between limiting mechanistic regimes

of alcohol dehydrogenation (Scheme 9). For instance, distinct

Scheme 9. Comparison of Hammett Parameters

Figure 7. Synthesis and crystal structure (thermal ellipsoids of 6

depicted at 50% probability) of Ru(bMepi)(PPh )(OCH CF ) (6).

3

2

3

PPh₃phenyl groups and hydrogen atoms are omitted for clarity.

Selected bond distances (angstroms): Ru1−P1, 2.3224(5); Ru1−O1,

2

.086(1); Ru1−N1, 2.079(2); Ru1−N3, 1.946(2); Ru1−N5, 2.075(2);

N2−C7, 1.298(3); N3−C7, 1.380(3); N3−C14, 1.374(3); and N4−

C14, 1.304(3).

set of bMepi resonances with the methyl resonances at 1.75

31

1

ppm, and the P{ H} NMR spectrum of 6 exhibits a singlet at

3.8 ppm, which is similar to the ³¹P spectrum observed for

ρ values were reported for Ru-catalyzed alcohol dehydrogen-

ation reactions that operate through turnover-limiting β-H

4

Ru(bMepi)(PPh )Cl (4, 43.5 ppm). Crystals suitable for single-

3

27

elimination (ρ = −0.43), free-radical H atom transfer (ρ =

crystal X-ray diﬀraction were obtained from vapor diﬀusion of

pentane into a PhMe solution of 6. The solid-state structure

2

8

29

−

0.30), or outer-sphere pathway (ρ = −0.89). In contrast

to these values, the Hammett analysis for 3 aﬀorded a ρ value of

1.69(5). The negative ρ value signiﬁes a positive charge

II

shows a square-based pyramid geometry about the Ru center

22

−

(τ = 0.01) with the −OCH CF ligand trans to PPh (Figure

2

3

buildup in the transition state, supporting a β-H elimination

turnover-limiting step in the inner-sphere pathway. Although 3

and the heterogeneous Ru-hydroxyapatite system are proposed

to undergo a β-H elimination turnover-limiting step, the

7). The shortest M−H distance is 2.69 Å with a M−H−C bond

angle of 99°, and the chemical shift of the methyl groups is

upﬁeld of the free HbMpi ligand. The structural and

spectroscopic properties satisfy two out of three criteria for

5

474

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

determining an agostic interaction between the Ru center and

the methyl C−H group.

that at low alcohol concentrations proton transfer is much

slower and becomes competitive with β-H elimination. Thus, as

reactant alcohol is consumed during catalysis, the turnover-

limiting step is proposed to change from β-H elimination to

proton transfer.

Given that complex 6 is similar to the proposed Ru-alkoxide

intermediate in the AAD catalytic cyclc, intermediates species

prior to H liberation were investigated by allowing 6 to react

2

with H . When a J. Young NMR tube containing a toluene-d₈

With the results of catalyst resting state at low and high

2

solution of 6 and PPh was charged with 30 psi of H , the

[1PhEtOH] in hand, an in-depth analysis of [PPh ] depend-

3

2

3

immediate formation of triﬂuoroethanol was detected as a

ence, [1PhEtOH] dependence, and activation parameters was

pursued. The reaction rate dependence on [1PhEtOH] ﬁts a

pre-equilibrium model with an equilibrium proton-transfer step

occurring before turnover-limiting β-H elimination. This is

1

9

triplet at −76.5 ppm in the proton-coupled F spectrum, and 6

1

31

and 3 were the only complexes observed by H and P NMR

spectroscopy. In addition, no reaction was observed when the

same experiment was performed at −75 °C. Upon slowly

warming the J. Young tube in the NMR spectrometer, the

formation of triﬂuoroethanol and 3 resulted from the clean

consistent with the zero-order dependence on [PPh ] and our

3

previous analysis showing that phosphine binding is not in an

equilibrium with the turnover-limiting step. At high [1PhE-

tOH], the forward rate of proton transfer is fast and β-H

elimination is the turnover-limiting step. As [1PhEtOH]

decreases, proton transfer becomes slower and eventually

becomes turnover-limiting. This implies that the Eyring data

collected at low [1PhEtOH] contains contributions from both

proton transfer and β-H elimination with β-H elimination as

conversion of 6 and H . No Ru intermediate species were

2

observed at low temperature, and 3 was the only Ru species

observed when 6 reacted with H . These observations suggest

2

that both alcohol and η -H adducts are short-lived

2

intermediates with respect to the alkoxide and/or 3.

Furthermore, no reaction (β-H elimination or decomposition)

was observed when a solution of 6 with and without 100 equiv

of triﬂuoroethanol in C D was heated to operating temper-

⧧

the major component. Hence, the activation parameters (ΔH

⧧

= 15 kcal/mol and ΔS = −41 eu) determined at high

[1PhEtOH] exclusively describe the transition-state structure

for a β-H elimination turnover-limiting step.

6

atures for catalytic AAD reaction (120 °C for 3 h), which is

consistent with an increase (∼15 kcal/mol) in the activation

30a

barrier eﬀected by the triﬂuoromethyl group.

Although experimental evidence supports a β-H elimination

⧧

Catalyst Resting State and Mechanistic Discussion.

With known spectroscopic features of a Ru-bMepi alkoxide

species in hand, NMR experiments were performed to observe

the catalyst resting state in situ. A solution of 8.3 M 1PhEtOH

containing 0.1 mol % of 3 inside a J. Young tube was monitored

turnover-limiting step, the large negative ΔS diﬀers signiﬁ-

cantly from the previously reported values for β-H elimination

17

from metal-alkoxide species. The classic β-H elimination

process involves cleavage of a β-C−H of the coordinated

alkoxide, with concomitant formation of hydride on an empty

cis coordination site and coordinated ketone (or aldehyde)

ligand. It follows that β-H elimination reactions are typically

unimolecular and largely enthalpically controlled (considerable

bond making and breaking character in the transition state). To

account for the atypical activation parameters for 3, alternative

1

31

by H and P NMR spectroscopy at ambient temperature and

1

00 °C. After 10 min at ambient temperature, 73% of 3 was

31

converted to a new species with a P resonance at 41.7 ppm

with concomitant formation of free PPh . The hydride region

3

1

of the H NMR spectrum showed no new species. This new

32

species at 41.7 ppm is consistent with the chemical shift of the

mechanisms, such as binuclear hydride abstraction and

33

isolated Ru alkoxide 6 and thus is proposed as Ru(bMepi)-

alcohol-assisted alkoxide dissociation, for β-H elimination

were considered. Our kinetic experiments discredit both

processes as possible β-H elimination pathways for our catalyst.

A binuclear mechanism is inconsistent with the observed ﬁrst-

order dependence on [3]. An alcohol-assisted alkoxide

dissociative pathway is also unlikely because of the zero-order

dependence on [1PhEtOH] at high [1PhEtOH]. Alternatively,

we propose that 3 operates under a traditional β-H elimination

process in which the highly negative activation entropy reﬂects

contributions from solvent reorganization likely imparted by

31

(

PPh )(OCHPhMe). Heating the J. Young tube inside the

3

NMR spectrometer for 10 min at 100 °C resulted in the full

31

conversion of 3 to the proposed Ru alkoxide. At 100 °C the P

NMR spectrum exhibited only two resonances corresponding

to the Ru alkoxide (40.6 ppm) and free PPh₃with 1:1

integration values. Observation of the catalyst resting state as

Ru-alkoxide species at high [1PhEtOH] further supports a β-H

elimination turnover-limiting step during catalysis.

The catalyst resting state was also examined at low

34

[

1PhEtOH]. A C D solution containing 0.83 M 1PhEtOH

hydrogen bonding. Such large entropic contributions are

consistent with the reactions performed in neat alcohol solvent

that allows the formation of a network of hydrogen bonds with

a coordinated alkoxide ligand. Comparison of the activation

parameters at low and high [1PhEtOH] revealed that entropy,

not enthalpy, is the main contributor to the β-H elimination

process. Thus, reorganization in the transition state, as reﬂected

6

1

31

and 0.1 mol % of 3 was monitored by H and P NMR

spectroscopy at ambient temperature and 100 °C. After 10 min

at ambient temperature, 13% of a species consistent with

3

1

formulation of the proposed alkoxide ( P = 41.7 ppm) was

31

observed by P NMR spectroscopy. The hydride region of the

1

H NMR spectrum showed only the hydride resonance of 3 as

⧧

a triplet, which was broadened and suggestive of a dynamic

process associated with ligand substitution and/or proton

transfer. The equilibrium constant between 3 and the proposed

Ru-alkoxide species is invariant from 0.83 to 6.0 M 1PhEtOH,

which is consistent with a pre-equilibrium process (Table S1 ).

After the NMR tube was heated for 10 min at 100 °C, a 1:1

ratio of the Ru alkoxide to 3 was observed. Continued

monitoring of the reaction mixture at 100 °C for 1 h

by the negative and large ΔS , must be due to rearrangement of

the hydrogen bonds interacting with the Ru-alkoxide species in

order for the β-C−H to migrate onto the Ru center.

Proposed Mechanism. Based on the series of kinetic and

isotopic labeling experiments, an inner-sphere catalytic cycle for

1PhEtOH dehydrogenation mediated by 3 is implicated. We

propose that a single PPh dissociation from 3 generates a

3

II

coordinatively unsaturated Ru -hydride species that can

(

corresponding to ca. 0.3% acetophenone) resulted in no

reversibly bind 1PhEtOH. This species likely undergoes a fast

change in the ratio of the alkoxide species to 3. This suggests

proton-transfer event in equilibrium with a β-H elimination

5

475

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

turnover-limiting step to yield the ketone product and generate

the coordinatively unsaturated Ru -hydride species to re-enter

in the dehydrogenation cycle (Scheme 10).

result is consistent with our previously reported ﬁndings that

dehydrogenation of secondary alcohols is favored in the

presence of primary alcohols when using temperatures lower

than those required for acceptorless dehydrogenative coupling

II

8

reactivity to aﬀord esters. However, because these conditions

Scheme 10. Proposed Mechanism for AAD Catalyzed by 3

do not aﬀord coupling reactivity of primary alcohols, more

forcing conditions were used to promote dehydrogenation

activity of primary and secondary alcohols. Heating a PhMe

solution containing BnOH and 3 (5 mol %) to 120 °C for 3 h

aﬀorded 65% benzyl benzoate and 5% benzaldehyde (Scheme

1

1, eq 12).

When a competition experiment using BnOH and 1PhEtOH

was performed under identical reaction conditions, 10% benzyl

benzoate and >99% acetophenone were observed (Scheme 11,

eq 11). These results demonstrate that 3 chemoselectively

dehydrogenates secondary alcohols in the presence of primary

alcohols under conditions which promote the dehydrogenation

of both primary and secondary alcohols. The origin of this

preference was considered to arise from diﬀerences in rates of

either (a) proton transfer or (b) β-H elimination. In addition to

implications of β-H elimination, rather than proton transfer, as

the turnover-limiting step, the pK diﬀerence between primary

a

i

(

pK (n-PrOH) = 16.0) and secondary (pK ( PrOH) = 16.5)

a

alcohols also cannot account for the observed chemoselectivity

because primary alcohols are more acidic and should be easily

37

deprotonated. To further support this hypothesis, in situ

3

1

examination of the catalyst resting state by P NMR

spectroscopy revealed the quantitative formation of a primary

31

Ru-alkoxide species ( P = 42.5 ppm) when 0.2 mol % of 3 was

dissolved in a 4.3 M BnOH C D₆solution at room

6

temperature. In a diﬀerent experiment under identical

conditions using 4.3 M 1PhEtOH, only 45% of 3 was

converted to the secondary Ru-alkoxide species as observed

in the ³¹P NMR spectrum. Thus, the origin of the observed

chemoselectivity must occur after the formation of the Ru-

alkoxide intermediate.

Primary versus Secondary Alcohol Dehydrogenation.

An unusual feature of precatalyst 3 is the high selectivity for

secondary alcohol dehydrogenation in the presence of primary

3

5

alcohols. To a ﬁrst approximation, thermodynamic arguments

might be invoked to support the formation of the ketone over

To address chemoselectivity-inducing β-H elimination-

dependent reactions, we monitored the catalyst resting state

and product(s) formation of an AAD reaction of a 1:1 mixture

of BnOH (4.3 M) to 1PhEtOH (4.3 M) catalyzed by 0.2 mol %

36

the aldehyde product. However, this consideration assumes

equilibrium conditions are met, which is not likely under the

catalytic conditions. To gain further insight into the origin of

the chemoselectivity bias, we evaluated competition experi-

ments between benzyl alcohol (BnOH) and 1PhEtOH.

Heating an equimolar (0.5 mmol) mixture of BnOH and

3

1

of 3. Prior to heating the reaction to 120 °C, the P NMR

spectrum exhibited only two resonances: one for free PPh and

another at 42.5 ppm, which was previously identiﬁed as the

primary Ru-alkoxide species (Ru(bMepi)(PPh )(OCH Ph).

3

1PhEtOH containing 5 mol % 3 to 100 °C for 3 h resulted in

3

2

the quantitative conversion of 1PhEtOH to acetophenone,

while BnOH remained unreacted (Scheme 11, eq 10). This

Upon heating the reaction to 120 °C, a 9:1 ratio of the

primary to secondary alkoxide species were observed.

Continuous monitoring of the reaction showed no changes in

the ratio of the primary to secondary alkoxides species in the

Scheme 11. Competition Experiment between BnOH and

a

31

1

PhEtOH

P NMR spectrum and only the production of acetophenone

1

(

34 turnovers in 30 min) was observed in the H NMR

spectrum (Scheme 12).

The competition experiments implicate slower β-H elimi-

nation from the primary alkoxide compared to the secondary

alkoxide, which is consistent with the absence of BnOH

dehydrogenation activity at lower (<120 °C) temperatures, and

38

the stronger BDE of the C−H bond cleaved. In addition, the

activation parameters at high alcohol concentration for β-H

elimination suggests that hydride transfer is dominated by

entropic factors derived from hydrogen-bonding solvation

eﬀects. Hydrogen bonding with the alcohol solvent would be

more favorable for the primary alkoxide species because of the

decreased steric environment surrounding the oxygen atom

a

Note that reaction yields are calculated based on the corresponding

alcohol reactants.

(−OCPhH versus −OCPhMeH). Therefore, a higher degree

2

5

476

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Scheme 12. Catalyst Resting State and Activity for a

Competitive Experiment between BnOH and 1PhEtOH

of reorganization in the transition state for β-H elimination

would be anticipated for the primary alkoxide species, thus

leading to a larger kinetic barrier.

Steric and Electronic Eﬀects of the bMepi Ligand on

Dehydrogenation Activity. The requirement of the ortho-

methyl units around the primary coordination sphere was

interrogated with the ligands 1,3-bis(4′,6′-methyl-2′-

pyridylimino)isoindolate (b4,6-Mepi) and 1,3-bis(2′-

pyridylimino)isoindolate (bpi). Complementary to evaluation

of a steric eﬀect, the eﬀects of electronically rich and deﬁcient

ligands were also examined using para-substituted variants,

b4Rpi (R = H, Cl, Me, OMe, OH). A series of Ru(b4Rpi)-

Figure 8. Synthesis of Ru(b4Rpi)(PPh ) Cl (7-R) and crystal

structures (thermal ellipsoids depicted at 50% probability) of 7-H

3

2

and 7-Me. The PPh phenyl groups and hydrogen atoms are omitted

3

for clarity. Selected bond distances for 7-H (angstroms): Ru1−Cl1,

2.4782(4); Ru1−P1, 2.4275(4); Ru1−P2, 2.3874(4); Ru1−N1,

2.092(1); Ru1−N3, 2.009(1); Ru1−N5, 2.116(1); N2−C6,

1.302(2); N3−C6, 1.367(2); N3−C13, 1.371(2); and N4−C13,

(

PPh ) Cl (7-R, R = H, Cl, Me, OMe) complexes were

3 2

1.299(2). Selected bond distances for 7-Me (angstroms): Ru1−Cl1,

2.4824(5); Ru1−P1, 2.4076(5); Ru1−P2, 2.4254(5); Ru1−N1,

2.111(2); Ru1−N3, 2.004(2); Ru1−N5, 2.101(2); N2−C6,

1.304(3); N3−C6, 1.365(3); N3−C13, 1.379(3); and N4−C13,

synthesized by heating a THF solution containing Hb4Rpi,

RuCl (PPh ) , and TlPF to 60−70 °C for 16−24 h, followed

2

3 3

6

t

by the addition of 1.05 equiv of NaO Bu (Figure 8). After

1

.296(3).

isolation of complexes 7-R, the composition and purity were

1

13

31

conﬁrmed by H, C, and P NMR spectroscopy, infrared

1

spectroscopy, and elemental analysis. The H NMR spectra of

solution dissociation dynamics was suppressed with the

addition of excess PPh . For example, in the presence of 1

equiv of PPh , the singlet at 68.38 ppm was absent in the

P{ H} NMR spectrum. Crystals suitable for single-crystal X-

7

-R feature a single set of ligand-based resonances with the

3

1

absence of the isoindole proton, and the P{ H} NMR spectra

exhibit a singlet at 26.1, 25.1, 26.1, and 26.4 ppm (7-R, R = H,

Cl, Me, OMe, respectively), consistent with trans disposed

phosphorus atoms and meridional binding of the b4Rpi ligand.

3

31

1

ray diﬀraction were obtained from vapor diﬀusion of pentane

into a THF solution of 7-OH at −35 °C. Analogous to the

single-crystal structures for 7-H and 7-Me, the solid-structure

Single crystals of 7-H and 7-CH were subjected to X-ray

3

II

diﬀraction experiments, and the solid-state structures conﬁrm

reveals an octahedral geometry around the Ru center with the

II

octahedral geometry around the Ru center with a chloride

b4OHpi ligand meridionally coordinated with two trans PPh₃

ligands and a chloride, thus conﬁrming the identity of the

product (Figure 9).

In the proposed AAD mechanism shown in Scheme 10, the

steric proﬁle of the methyl groups may play a blocking role to

impede hydride transfer in the turnover-limiting step. This

steric eﬀect imposed by the methyl groups was examined by

comparing the reaction rates of 1PhEtOH dehydrogenation

ligands trans to the isoindolate nitrogen atom (Figure 8).

To enhance the stability of a coordinatively unsaturated Ru

species during AAD catalysis, we targeted an electron-rich

metal environment by synthesizing a Ru compound with the

ligand b4OHpi, in which strongly electron-donating hydroxyl

groups are substituted para to the pyridyl nitrogens.

Deprotonation of the hydroxyl groups to generate aryl-oxide

groups in situ would further enhance the electron-richness of

the metal environment. Ru(b4OHpi)(PPh ) Cl (7-OH) was

catalyzed by 4, 4-Me (Ru(b4,6-Mepi)(PPh )Cl), 7-H, and 7-

3

−

5

−1

Me (Table 1). The reaction rates for 4 (7.7(3) × 10 M·s )

3

2

and 4-Me (7.7(3) × 10⁻M·s ) were identical (Table 1,

entries 1 and 2), which suggests that the weakly electron-

donating methyl groups have no electronic eﬀect on the

dehydrogenation activity. This observation allowed evaluation

of the steric inﬂuence of the ortho-methyl substituents of the

bMepi ligand by comparing the reaction rates of 4 and 7-H.

The reaction rate was increased by 42% when 7-H (10.9(2) ×

5

−1

prepared by allowing a THF solution containing Hb4OHpi,

t

RuCl (PPh ) , and NaO Bu to stir at 70 °C for 16 h. 7-OH was

2

3 3

isolated in 79% yield as a dark blue solid with only slight

solubility in THF. The H NMR spectrum reveals multiple

broad resonances, and the P{ H} NMR spectrum in THF

1

3

1

displays resonances at 68.38, 26.19, and −5.51 ppm (free

PPh ), suggesting a dynamic process likely caused by the

3

−5

−1

−5

−1

dissociation of one of the PPh ligands in solution. This

10 M·s , Table 1, entry 3) or 7-Me (10.9(2) × 10 M·s ,

3

5

477

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

−5

reaction rate was increased when 7-OMe (11.8(2) × 10 M·

−

1

40

s ; σ (OMe) = −0.27; Table 1, entry 6) and 7-OH

para

−5

−1

40

(

12.0(2) × 10 M·s ; σ (OH) = −0.37; Table 1, entry 7)

−5 −1

para

were used instead of 7-Cl (8.0(5) × 10 M·s ; σ_p_a_r_a(Cl) =

40

0

.23; Table 1, entry 5). To further enhance the donor

−

40

strength, the aryl-oxide (7-O ; σ (O ) = −0.81) was

prepared. 7-OH was allowed to react with 5 equiv of NaO Bu,

which resulted in a further 11% increase in rate (13.3(2) × 10

M·s ; Table 1, entry 8) of 1PhEtOH dehydrogenation.

Although, these results indicate that a more electron-rich Ru

environment exhibits higher AAD activity, the changes to the

reaction rate are small; thus, the electronic environment at the

Ru center has a minimal eﬀect on the turnover-limiting step in

the AAD catalytic cycle.

para

t

−5

−1

Base-Promoted AAD Catalysis. Benchtop-stable reagents

provide greater synthetic utility and accessibility and thus are

more commonly and easily handled by most synthetic

laboratories. The air-sensitive precatalyst 3 is capable of

mediating promoterless AAD reactions, likely due in part to

the Ru-hydride, an internal basic site. Alternatively, entry into

the AAD catalytic cycle should also be possible using the air-

stable complex 4 in the presence of an external base or using an

in situ preparation of the Ru-bMepi catalytic species. Indeed,

when a toluene solution containing 0.5 mmol of 1PhEtOH, 1

Figure 9. Synthesis and crystal structure (thermal ellipsoids depicted

at 50% probability) of Ru(b4OHpi)(PPh ) Cl (7-OH). The PPh

3

phenyl groups and hydrogen atoms are omitted for clarity. Selected

3

2

bond distances (angstroms): Ru1−Cl1, 2.4824(5); Ru1−P1,

2

.4076(5); Ru1−P2, 2.4254(5); Ru1−N1, 2.111(2); Ru1−N3,

.004(2); Ru1−N5, 2.101(2); N2−C6, 1.304(3); N3−C6, 1.365(3);

N3−C13, 1.379(3); and N4−C13, 1.296(3).

a

Table 1. Reaction Rates of 1PhEtOH Dehydrogenation

t

mol % of 4, and 2 mol % of NaO Bu was heated to reﬂux for 4

h, acetophenone was observed in quantitative (>99%) yield

5

entry

catalyst

rate (× 10⁻M·s⁻¹

)

(

Scheme 13, eq 13). This reactivity demonstrates the synthetic

1

2

3

4

5

6

7

4

7.7(3)

4-Me

7-H

10.9(2)

8.0(5)

Scheme 13. Base-Promoted AAD Catalysis

7-Me

7-Cl

7-OMe

7-OH

7-H

11.8(2)

12.0(2)

13.3(2)

12.4(2)

b

8

b

9

applicability of 4 as a dehydrogenation catalyst that can be

prepared using air-stable reagents. Furthermore, AAD catalysis

by in situ formation of the catalytically active Ru-bMepi species

was evaluated. In the presence of 1 mol % RuCl (PPh ) , 1 mol

a

Reaction conditions: 1PhEtOH (7.8 M), PhTMS (0.38 M), [Ru]

t

2

3 3

(

1.9 mM), and NaO Bu (3.8 mM) were stirred at 120 °C in an open

vial inside an inert-atmosphere glovebox. The conversion of 1PhEtOH

to acetophenone was monitored by H NMR spectroscopy. Reaction

performed with 9.5 mM NaO Bu.

t

%

HbMepi, and 3 mol % NaO Bu, 1PhEtOH was converted to

1

b

acetophenone in >99% yield after heating for 4 h in reﬂuxing

toluene (Scheme 13, eq 14). Control experiments showed no

reaction in the absence of HbMepi. Therefore, the broad

applicability of our system to promote dehydrogenation by

well-deﬁned precatalysts as well as in situ generation from air-

stable precursors highlights the robustness of the system.

In addition to the alcohol dehydrogenation activity of the in

situ prepared catalyst, stoichiometric reactions were performed

to uncover intermediates en route to the well-deﬁned

t

Table 1, entry 4) was used instead of 4. The rate enhancement

is consistent with a sterically blocking eﬀect of the methyl

groups, which hinders β-H elimination in the turnover-limiting

step (Scheme 10).

While the addition of a modestly donating methyl group in

the para-position had no eﬀect on the overall rate, we evaluated

the eﬀect of adding highly donating substituents to the ﬂanking

pyridine rings. Increased electron donor strength of the ligand

is expected to concomitantly enhance the hydricity of any

precatalyst 4. Allowing HbMepi and RuCl

dichloroethane at 70 °C for 4 h generated Ru(HbMepi)-

(PPh Cl (8) in 70% yield as a green solid (Figure 10). The

P{ H} spectrum displays a singlet at 24.9 ppm, and the H

(PPh ) to react in

2

3 3

)

3

2

31

1

39

metal-based hydrides. However, because the turnover-limiting

step is β-H elimination, the ligand electronic eﬀect on the

turnover-limiting step may be less than dramatic.

NMR spectrum reveals a solution structure consistent with

asymmetric binding of the HbMepi ligand. For example, two

resonances for the ortho-CH substituents were observed at

3

The electronic variations at the Ru center imposed by

electronically rich and deﬁcient bpi ligands were evaluated by

examining the dehydrogenation rates of 1PhEtOH. The

0.99 and 2.31 ppm, and nine distinct resonances were observed

for the HbMepi scaﬀold in the aromatic region of the H NMR

spectrum. Crystals suitable for a single X-ray diﬀraction

1

5

478

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Figure 10. Synthesis of Ru(HbMepi)(PPh ) Cl (8) and thermal

3

2

ellipsoids of 8 depicted at 50% probability. PPh phenyl groups and

3

hydrogen atoms, except for the N−H, are omitted for clarity. Selected

bond distances (angstroms): Ru1−Cl1, 2.410(2); Ru1−Cl2, 2.411(2);

Ru1−P1, 2.384(2); Ru1−P2, 2.371(2); Ru1−N1, 2.088(7); Ru1−N2,

Figure 11. Synthesis of complex 9 and thermal ellipsoids of 9 depicted

at 50% probability. PPh phenyl groups and hydrogen atoms, except

3

for the methanide, are omitted for clarity. Selected bond distances

2

.092(8); N2−C7, 1.29(1); N3−C7, 1.38(1); N3−C14, 1.40(1); and

(

1

angstroms): Ru1−P1, 2.392(1); Ru1−N1, 2.128(4); Ru1−N3,

N4−C14, 1.28(1).

.968(4); Ru1−N5, 2.027(4); Ru1−C21, 2.221(5); N2−C7,

.300(6); N3−C7, 1.371(8); N3−C14, 1.385(5); and N4−C14,

experiment were obtained from slow evaporation of a DCM

1.302(7).

solution of 8 at 5 °C. The solid-state structure exposes an

II

2

octahedral geometry around the Ru center, supported by a κ -

of 1PhEtOH to acetophenone (eq 15). Hence, complex 9 is a

precursor to generate a catalytically active HRu(bMepi)(PPh₃)

species that operates via the AAD catalytic cycle proposed in

Scheme 10.

HbMepi, two PPh and two chloride ligands. Addition of 1.05

equiv of NaO Bu to a THF solution containing 8 cleanly

aﬀorded complex 4, which is a benchtop-stable precatalyst in

the base-promoted AAD reaction.

3

t

Isolation of an Alternative Promoterless AAD Cata-

lyst. In (de)hydrogenation catalysis, active Ru complexes are

4

1

often generated using exogenous base additives. We

previously showed that 4 was activated in the presence of

NaO Bu to provide an active dehydrogenation catalyst;

t

Reversible Catalytic Hydrogenation−Dehydrogena-

tion Reactions. Catalytic hydrogenation and dehydrogen-

ations reactions are attractive candidates to target for reversible

however, the mechanism of activation was unclear. Stoichio-

metric reactions between 4 and base were performed to

examine the reaction pathway. Under basic conditions,

deprotonation of the ortho-methyl group was achieved by

2

,42

energy storage.

Although a myriad of catalysts can mediate

the forward or reverse reaction, very few systems are capable of

catalyzing reversible hydrogenation−dehydrogenation reaction-

−

t

O Bu base, and the deprotonated intermediate was sub-

2

a,7,43

sequently trapped by coordination with another Ru complex to

aﬀord a dimer (9), which was isolated in 78% yield (Figure 11).

Crystals suitable for single-crystal X-ray diﬀraction were

obtained from vapor diﬀusion of pentane into a benzene

solution of 9, and the solid-state structure reveals a square-

based pyramidal geometry about the Ru center (τ = 0.02,

s.

The ability of 3 to eﬀect a catalytic transfer

i

hydrogenation reaction using PrOH as the H

surrogate was

previously demonstrated. Thus, we hypothesized that the

hydrogenation reactions should be possible using H given that

entry to 3 was also gained by treating the Ru-alkoxide (6) with

. Indeed, acetophenone was completely consumed to aﬀord

1PhEtOH within 1 h at 110 °C using 1 mol % 3 and 30 psig H

in PhMe-d inside a J. Young NMR tube. Following the

2

8

2

H

2

0

.03) with the pincer ligand meridionally coordinated and the

2

II

pyridinylmethanide motif coordinated to another Ru center

Figure 11). The asymmetry of the pincer ligand is conﬁrmed

8

(

hydrogenation reaction, the solution was transferred from the J.

Young NMR tube to a Schlenk ﬂask to assess the ability to

promote the dehydrogenation of 1PhEtOH. Reﬂuxing the

toluene solution under an inert atmosphere for 4 h restored

acetophenone quantitatively (Scheme 14). This demonstrates

the complete and reversible transformations between aceto-

phenone and 1PhEtOH via successive hydrogenation−dehy-

drogenation reactions using complex 3 as the single catalyst.

in solution by >10 distinct resonances in the aromatic region of

1

the H NMR spectrum.

To investigate complex 9 as a precursor en route to the

catalytically active HRu(bMepi)(PPh ) species, a reaction with

3

H₂was examined. H₂was heterolytically cleaved by the

pyridinylmethanide group and the Ru center at low pressure

(

30 psig) and reacted with another PPh molecule to aﬀord 3

3

(

Figure 11). This demonstrates that complex 9 is a precursor to

CONCLUSION

the HRu(bMepi)(PPh ) species and that the methanide motif

■

3

is an internal basic site that may promote dehydrogenation

reactions without requiring any additives. To illustrate the latter

point, reﬂuxing a 0.25 M 1PhEtOH toluene solution containing

The mechanism of the AAD reaction catalyzed by complex 3

was studied by a series of kinetic and isotopic labeling

experiments, isolation of intermediates, and catalyst modiﬁca-

tions. Experimental evidence supported an inner-sphere,

0

.5 mol % 9 for 4 h resulted in quantitative conversion (>99%)

5

479

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Scheme 14. Reversible Hydrogenation−Dehydrogenation

Reactions Catalyzed by 3

ligand have a minimal eﬀect on the rate of catalytic

dehydrogenation.

In addition to delineating a detailed mechanistic under-

standing of dehydrogenative catalysis mediated by 3, we also

showed 3 as an eﬃcient hydrogenation precatalyst. By coupling

the hydrogenation and the dehydrogenation abilities of 3, we

have thus demonstrated that completely reversible trans-

formations between ketones and alcohols are achieved and

are dictated by hydrogen input or release. Overall, such

reversible catalytic reactions are of broad interest to the ﬁeld of

hydrogen storage as well as chemical synthesis.

stepwise pathway for proton and hydride transfers with a β-H

elimination turnover-limiting step. Selective isotopic labeling

experiments combined with catalyst modiﬁcation (methylation

of the pincer ligand backbone) demonstrated that a cooperative

metal−ligand pathway involving the imine functionality is not

necessary for eﬃcient dehydrogenation. The activation

parameters suggested an associative pathway involving a highly

ordered transition-state structure. Thus, we propose that a

EXPERIMENTAL SECTION

■

General Considerations. All manipulations were con-

ducted under a nitrogen atmosphere on a Schlenk manifold or

in a glovebox using standard Schlenk techniques, unless

otherwise stated. All reagents were purchased from commercial

t

vendors. Anhydrous dichloroethane (DCE, Acros), NaO Bu

single PPh dissociation event from 3 generates a coordinatively

(Sigma-Aldrich), and MeOTf (Sigma-Aldrich) were used

without further puriﬁcation. 1-Phenylethanol, phenyltrimethyl-

silane, acetophenone, 1-(4-methylphenyl)ethanol, 1-(4-

methoxyphenyl)ethanol, 1-(4-ﬂuorophenyl)ethanol, 1PhEtOD,

3

unsaturated HRu(bMepi)(PPh ) species, which undergoes a

3

proton-transfer equilibrium to generate a transient Ru-(H₂)

alkoxide species. H loss aﬀords a Ru-alkoxide intermediate that

2

can participate in a turnover-limiting β-H elimination reaction

to complete the catalytic cycle (Scheme 15). Moreover,

modiﬁcations to the pincer ligand revealed that the steric

proﬁles of the methyl groups on bMepi slightly impeded

catalytic activity, while electronic modiﬁcations of the pincer

1PhCH CDOH, 1PhCH CDOD, and 2,2,2-triﬂuoroethanol

3 3

were distilled from CaH under a nitrogen atmosphere and

2

then stored over 3 Å molecular sieves for at least 24 h. The

following compounds were synthesized according to literature

8

methods: HRu(bMepi)(PPh₃)₂(3), Ru(bMepi)(PPh )Cl

3

Scheme 15. Proposed Cycle for Catalytic AAD and Isolation of Precursors

5

480

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

8

19

44

(

4), 1PhEtOH isotopologues, and the Hb4Rpi ligands.

solvent was evaporated under vacuum, and the residue was

The 3 Å molecular sieves were dried at 250 °C under dynamic

vacuum for 24 h. Dichloromethane (DCM), diethyl ether

puriﬁed through a plug of silica gel eluting with Et O (5 mL).

2

Evaporation of the Et O solution aﬀorded acetophenone as a

2

(

Et O), pentane, benzene (C H ), dimethoxyethane (DME),

colorless oil. The purity and identity were conﬁrmed by

2

6

and tetrahydrofuran (THF) were puriﬁed using a Glass

Contour solvent puriﬁcation system consisting of a copper

catalyst, neutral alumina, and activated molecular sieves then

passed through an in-line, 2 μm ﬁlter immediately before being

dispensed. Toluene (PhMe) and hexanes (Hex) were sparged

using nitrogen and then stored over 3 Å molecular sieves for at

least 24 h.

comparison to previously reported NMR data.

Ru(HbMepi)(PPh )Cl[PF ]. THF (3 mL) was added to a vial

3

6

charged with HbMepi (51.6 mg, 0.158 mmol), RuCl (PPh )

2

3 3

(137.4 mg, 0.143 mmol), TlPF (50.1 mg, 0.143 mmol), and a

6

stir bar. The reaction solution was allowed to stir at 70 °C for 2

days. After the solution was cooled to ambient temperature,

TlCl was ﬁltered using a ﬁne frit and the THF solvent was

removed under vacuum. The crude product was washed with

NMR spectra were recorded on Varian Inova 500, Varian

MR400, Varian vnmrs 500, and Varian vnmrs 700 spectrom-

C

6

H

6

(4 × 10 mL) and Et

was dissolved in minimum DCM and layered with Et

24 h at ambient temperature, the precipitates were collected

and washed with Et O (4 × 10 mL). Evaporation of the

volatiles under vacuum aﬀorded the product as a dark blue

solid. Crystals were obtained from vapor diﬀusion of Et O into

a DCM solution at ambient temperature. Yield: 94 mg (75%).

2

O (4 × 10 mL). The crude product

1

13

eters at ambient temperature. H and C shifts are reported in

parts per million (ppm) relative to TMS, with the residual

2

O. After

3

1

19

solvent peak used as an internal reference. P and F NMR

2

1

spectra were referenced on a uniﬁed scale to their respective H

3

1

NMR spectra. At elevated temperatures, P spectra were

2

referenced relative to an internal standard of PPh at −5.6 ppm.

3

1

The following abbreviations are reported as follows: singlet (s),

doublet (d), doublet of doublets (dd), triplet (t), quartet (q),

multiplet (m), methyl (Me), methoxy (OMe), and triphenyl-

H NMR (700 MHz, CD

Cl ): δ 10.85 (s, 1H, NH), 8.17 (t,

2 2

J_H_H= 7.0 Hz, 1H), 7.94 (d, J_H_H= 7.7 Hz, 1H), 7.69 (d, J_H_H

=

5.6 Hz, 1H) 7.62 (bs, 1H), 7.51 (bs, 1H), 7.45−7.44 (m, 2H),

1

3

phosphine (PPh₃). C NMR resonances were observed as

singlets unless otherwise stated. For atom numbering of the bpi

ligand in complexes 4−7, see Figure 12.

7.34 (d, J = 7.0 Hz, 1H), 7.27 (bs, 1H), 7.17 (t, J = 7.0 Hz,

HH

3H, PPh ), 6.92 (t, J = 7.0 Hz, 6H, PPh ), 6.52 (t, J = 9.1

3

HH

3

HH

1

3

1

Hz, 6H, PPh ), 1.78 (s, 3H, Me), 1.70 (s, 3H, Me). C{ H}

3

(

176 MHz, CD Cl ): δ 160.99, 160.90, 156.363, 153.37,

2 2

1

47.34, 146.49, 139.11, 138.98, 138.57, 132.71, 132.58, 132.53,

31.93, 131.04, 130.86, 129.65, 129.13, 129.08, 124.43, 123.46,

22.55, 121.30, 115.64, 23.56, 22.76. P{ H} NMR (162 MHz,

31

1

CD Cl ): δ 39.17 (s, PPh ), −144.30 (septet, J = 710 Hz,

2

3

PF

1

9

PF ). F NMR (376 MHz, CD Cl ): δ −72.77 (d, J = 710

6

2

FP

−1

Hz, PF ). IR (ATR, cm ): 3331, 3059, 1631, 1600, 1552, 1532,

6

Figure 12. Atom numbering of the bpi ligand for the NMR

characterization of 4−7.

1463, 1449, 1433, 1372, 1292, 1210, 1163, 1104, 1088, 998,

833, 795, 740, 692. Anal. Calculated (found): C, 52.39 (52.51);

H, 3.70 (3.80); N, 8.04 (7.85).

Solid-state IR spectra were collected using a Nicolet iS10

spectrometer equipped with a diamond attenuated total

reﬂectance (ATR) accessory. Elemental analyses were

performed by Midwest Microlab, LLC.

General Procedure for 1PhEtOH Dehydrogenation

Catalyzed by 3. 1PhEtOH (5 mL, 41.4 mmol) was added to a

Preparation of Ru(bMepi)(PPh )Cl (4) from Ru-

3

(HbMepi)(PPh )Cl[PF ]. THF (5 mL) was added to a 20

3

6

mL vial charged with Ru(HbMepi)(PPh )Cl[PF ] (10 mg,

3

6

t

0.0115 mmol), NaO Bu (1.1 mg, 0.0115 mmol), and a stir bar.

The reaction solution was allowed to stir at ambient

temperature for 30 min. Solvent was removed under vacuum,

and the crude product was extracted with DCM (10 mL). The

DCM solvent was removed under vacuum, and the product was

2

0 mL vial charged with 3 (3.9 mg, 0.00409 mmol), PhTMS

(

0.5 mL, 2.9 mmol), and a stir bar. The vial was capped with a

septum and pierced with a 27-gauge needle. Then the vial was

heated to the desired temperature (90, 100, 110, 120, or 130

washed with Et O (4 × 10 mL). Evaporation of the volatiles

2

under vacuum aﬀorded the product as a dark purple powder.

The purity and identity were conﬁrmed by comparison to

previously reported NMR data. Yield: 7.5 mg (90%).

°

C) using an aluminum heating block inside an inert-

atmosphere glovebox. The formation of acetophenone was

monitored by sampling 0.5 mL aliquots and then analyzed by

Ru(b4,6-Mepi)(PPh )Cl (4-CH ). THF (10 mL) was added to

3

1

H NMR spectroscopy against PhTMS as an internal standard.

a 20 mL vial charged with Kb4,6-Mepi (103.7 mg, 0.264

mmol), RuCl (PPh ) (240.6 mg, 0.251 mmol), and a stir bar.

To conﬁrm reproducibility, all kinetic experiments were

performed in triplicate.

2

3 3

The resulting solution was stirred at room temperature for 17

h. Solvent was removed under vacuum. The crude product was

General Procedure for Base-Promoted 1PhEtOH

Dehydrogenation. PhMe (2 mL) was added to a 10 mL

Schlenk ﬂask charged with [Ru] (0.005 mmol; 4, 3.6 mg; 5, 5.0

washed with Et O (4 × 5 mL) and extracted with DCM (4 × 5

2

mL). The DCM solvent was removed under vacuum, and the

product was washed with pentane (4 × 10 mL), aﬀording the

product as a dark blue powder. The product was recrystallized

from layering pentane on top of a DCM solution at −35 °C.

t

mg; RuCl (PPh ) , 4.8 mg; and HbMepi, 1.6 mg), NaO Bu (2

2

3 3

mol %, 0.01 mmol, 1.0 mg; 3 mol %, 0.015 mmol, 1.4 mg), and

a stir bar. 1PhEtOH (60 μL, 0.5 mmol) was added to the

mixture. A reﬂux condenser was connected to the Schlenk ﬂask,

capped with a septum, and pierced with a 27-gauge needle in

addition to a nitrogen inlet 16-gauge needle. The reaction

solution was heated to 120 °C using an aluminum heating

block. After 4 h, the reaction mixture was cooled to ambient

temperature and exposed to air to quench the reaction. The

1

Yield: 126 mg (68%). H NMR (400 MHz, C D ): δ 7.98 (dd,

6

5

4

J_H_H= 5.6, 3.2 Hz, 2H, H ), 7.51 (s, 2H, H ), 7.05 (dd, J

5.2, 2.8 Hz, 2H, H ), 6.86 (t, J = 8.4 Hz, 6H, PPh ), 6.77 (t,

=

HH

6

HH

3

J_H_H= 7.2 Hz, 3H, PPh ), 6.67 (t, J = 6.8 Hz, 6H, PPh ), 6.24

3

HH

3

2

13

1

(s, 2H, H ), 1.98 (s, 6H, p-Me), 1.78 (s, 6H, o-Me). C{ H}

(176 MHz, CD Cl ): δ 158.69, 154.87, 153.30, 147.36, 141.48,

2

5

481

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

1

34.26, 134.03, 133.01, 129.53, 128.92, 128.20, 125.82, 121.44,

crude product was washed with Et O (4 × 20 mL), aﬀording

2

31

1

19.96. P{ H} NMR (162 MHz, C D ): δ 45.34 (s, PPh ). IR

Ru(Hbpi)(PPh ) Cl[PF ] in 83% yield (932 mg). THF (15

6

3

3 2

6

−1

(

ATR, cm ): 3052, 1622, 1566, 1498, 1447, 1430, 1372, 1327,

mL) was added to a 20 mL vial charged with Ru(Hbpi)-

t

1

9

6

289, 1251, 1229, 1204, 1182, 1109, 1087, 1030, 1001, 969,

00, 843, 773, 745, 726, 694. Anal. Calculated (found): C,

3.78 (63.59); H, 4.68 (4.48); N, 9.30 (9.17).

(PPh ) Cl[PF ] (925.5 mg, 0.837 mmol), NaO Bu (84.5 mg,

3

2

6

0.879 mmol), and a stir bar. The reaction solution was allowed

to stir at ambient temperature for 30 min. The THF solvent

was removed under vacuum, and the crude product was

Me

Ru(bMepi )(PPh )(OTf) (5). MeOTf (150 μL, 1.39 mmol)

3

2

was added to a 20 mL vial containing DCM solution of 4 (101

mg, 0.139 mmol) and a stir bar. The reaction solution was

allowed to stir at ambient temperature for 16 h. The DCM

solvent was removed under vacuum, and the crude product was

washed with Et O (4 × 10 mL). The crude product was

dissolved in minimum DCM and layered with Et O. After 24 h

at ambient temperature, the precipitates were collected and

washed with Et O (4 × 10 mL). Evaporation of the volatiles

under vacuum aﬀorded the product as a dark purple crystalline

solid. Crystals were obtained from vapor diﬀusion of pentane

into a DCM/C H solution at ambient temperature. Yield: 111

extracted with C H (50 mL). The C H solution was

6 6 6 6

lyophilized, and the product was washed with pentane (4 ×

20 mL). Evaporation of the volatiles under vacuum aﬀorded the

product as a green powder. Crystals were obtained from slow

evaporation of a DCM solution at ambient temperature

2

1

(

DCM/Hex). Yield: 562 mg (70%). H NMR (400 MHz,

2

1

C D ): δ 10.60 (d, J = 6.4 Hz, 2H, H ), 7.95 (dd, J = 4.8,

6

HH

5

2

.4 Hz, 2H, H ), 7.37−7.33 (m, 14H), 6.88 (t, J = 8.4 Hz,

HH

2

3

2

H, H ), 6.81−6.71 (m, 18H), 6.00 (t, J = 8.8 Hz, 2H, H ).

HH

13

1

C{ H} (176 MHz, CD Cl ): δ 158.47, 157.43, 152.63, 141.83,

2

134.43, 133.89, 132.69 (t, JCP = 17.2 Hz, ipso-CP), 128.86,

6

3

1

28.56, 127.59, 127.21, 119.93, 116.59. P{ H} NMR (162

mg (79%). H NMR (500 MHz, CD Cl ): δ 8.30−8.27 (m,

2

−

1

MHz, C D ): δ 26.09 (s, PPh ). IR (ATR, cm ): 3053, 1568,

2

7

H), 8.13 (t, J = 7.5 Hz, 1H), 7.95 (d, J = 8.0 Hz, 1H),

.83−7.80 (m, 2H), 7.72−7.65 (m, 2H), 7.39−7.35 (m, 5H),

.14 (t, J = 8.0 Hz, 6H, PPh ), 6.81 (t, J = 10.5 Hz, 6H,

6

3

HH

1

552, 1513, 1454, 1434, 1378, 1305, 1290, 1210, 1186, 1105,

087, 1007, 909, 843, 770, 744, 696. Anal. Calculated (found):

HH

3

HH

C, 67.60 (67.25); H, 4.41 (4.40); N, 7.30 (7.20).

PPh ), 3.73 (s, 3H, N-Me), 1.78 (s, 3H, Me), 1.57 (s, 3H, Me).

3

1

3

1

Ru(b4Mepi)(PPh ) Cl (7-Me). THF (10 mL) was added to a

0 mL vial charged with Hb4Mepi (103.7 mg, 0.317 mmol),

C{ H} (176 MHz, CD Cl ): δ 165.67, 160.83, 160.25, 153.97,

3 2

2

1

52.62, 148.97, 141.13, 140.89, 139.70, 134.86, 132.81, 132.76,

32.39, 131.55, 131.05, 130.77, 129.55, 129.49, 124.94, 124.59,

RuCl (PPh ) (303.7 mg, 0.317 mmol), TlPF (110.7 mg,

2

3 3

6

3

1

0.317 mmol), and a stir bar. The reaction solution was allowed

to stir at 60 °C for 24 h. After the solution cooled to ambient

temperature, TlCl was ﬁltered using a ﬁne frit and the THF

solvent was removed under vacuum. The crude product was

24.44, 123.76, 116.42, 44.46, 22.97, 21.76. P{ H} NMR (202

−1

MHz, CD Cl ): δ 47.84 (s, PPh ). IR (ATR, cm ): 3062,

2

3

1

609, 1565, 1519, 1459, 1435, 1400, 1308, 1266, 1230, 1206,

187, 1156, 1117, 1090, 1015, 909, 811, 797, 779, 742, 696.

washed with Et O (4 × 10 mL), aﬀording Ru(Hb4Mepi)-

2

Anal. Calculated (found): C, 49.10 (48.97); H, 3.42 (3.51); N,

(

PPh ) Cl[PF ] in 87% yield (312 mg). THF (15 mL) was

3 2 6

6.98 (6.88).

added to a 20 mL vial charged with Ru(Hb4Mepi)(PPh ) Cl-

3

2

Ru(bMepi)(PPh )(OCH CF ) (6). CF CH OH (4.3 μL,

3

2

3

2

t

[

PF ] (122 mg, 0.108 mmol), NaO Bu (10.9 mg, 0.113 mmol),

6

0.0563 mmol) was added to a 20 mL vial containing THF

and a stir bar. The reaction solution was allowed to stir at

ambient temperature for 30 min. The THF solvent was

removed under vacuum, and the crude product was extracted

with C H . The C H solution was lyophilized, and the product

solution of 3 (48.8 mg, 0.051 mmol) and a stir bar. The

reaction solution was allowed to stir at 70 °C for 2 days. After

cooling to ambient temperature, the THF solvent was removed

6

under vacuum and the crude product was washed with Et O (4

2

was washed with pentane (4 × 10 mL). Evaporation of the

volatiles under vacuum aﬀorded the product as a green powder.

Crystals were obtained from layering pentane on top of a DCM

×

10 mL) and pentane (4 × 10 mL). The product was

extracted with C H (15 mL). The C H solution was

6

lyophilized, aﬀording the product as a purple powder. Crystals

were obtained from vapor diﬀusion of pentane into a PhMe

solution at 5 °C. Yield: 22 mg (55%). H NMR (700 MHz,

1

solution at −35 °C. Yield: 90 mg (85%). H NMR (400 MHz,

1

C D ): δ 10.44 (d, J = 6.4 Hz, 2H, H ), 7.98 (dd, J = 5.2,

1

6

HH

5

4

2

6

1

.8 Hz, 2H, H ), 7.44−7.39 (m, 12H, PPh ), 7.26 (s, 2H, H ),

5

3

C D ): δ 8.09 (dd, J = 5.6, 3.5 Hz, 2H, H ), 7.69 (d, J

=

2

6

HH

.81−6.73 (m, 18H, PPh ), 5.92 (d, J = 6.8 Hz, 2H, H ),

4

3

HH

7

5

7

.7 Hz, 2H, H ), 7.19 (t, J = 7.7 Hz, 2H, H ), 7.06 (dd, J

.6, 2.8 Hz, 2H, H ), 6.77−6.72 (m, 9H, PPh ), 6.65 (t, J

.0 Hz, 6H), 6.35 (d, J = 7.0 Hz, 2H, H ), 3.28 (q, J = 7.7,

13

1

HH

.76 (s, 6H, Me). C{ H} (176 MHz, CD Cl ): δ 157.44,

6

2

3

HH

⁵⁶^.⁹⁴^,¹⁵²^.⁹⁵^,¹⁴⁵^.⁹³^,¹⁴¹^.⁸⁵^,¹³³^.⁹⁶^,¹³³^.⁰⁸⁽^t^,^JCP = 17.0

Hz, ipso-CP), 128.72, 128.45, 127.49, 119.75, 118.50. P{ H}

NMR (162 MHz, C D ): δ 26.14 (s, PPh ). IR (ATR, cm ):

2

31

1

HH

HF

1

3

1

J_H_H= 2.1 Hz, 2H, OCH CF ), 1.75 (s, 6H, Me). C{ H} (176

MHz, C D ): δ 159.69, 155.41, 152.68, 142.22, 136.03, 135.83,

−1

2

3

6

3

6

3

1

053, 1552, 1501, 1481, 1462, 1431, 1403, 1375, 1293, 1189,

102, 1088, 1007, 941, 840, 817, 746, 693. Anal. Calculated

1

34.59, 132.92, 132.86, 125.59, 120.31, 118.69, 67.59, 23.48.

3

1

P{ H} NMR (283 MHz, PhMe-d ): δ 43.88 (s, PPh ).

8

3

(found): C, 68.11 (68.38); H, 4.70 (4.79); N, 7.09 (6.99).

9

1

F{ H} NMR (376 MHz, PhMe-d ) δ −76.52 (s, OCH CF ).

8

2

3

Ru(b4Clpi)(PPh ) Cl (7-Cl). THF (10 mL) was added to a 20

3

2

−1

IR (ATR, cm ): 3041, 2814, 2714, 1568, 1538, 1512, 1460,

mL vial charged with Hb4Clpi (88 mg, 0.239 mmol),

RuCl (PPh ) (218.2 mg, 0.228 mmol), TlPF (79.5 mg;

1

7

4

433, 1388, 1264, 1184, 1153, 1121, 1107, 1009, 949, 905, 794,

69, 742, 693. Anal. Calculated (found): C, 60.91 (60.88); H,

.22 (4.29); N, 8.88 (8.63).

2

3 3

6

0.228 mmol), and a stir bar. The reaction solution was allowed

to stir at 60 °C for 18 h. After the solution cooled to ambient

temperature, TlCl was ﬁltered using a ﬁne frit and the THF

solvent was removed under vacuum. The crude product was

Ru(bpi)(PPh ) Cl (7-H). THF (15 mL) was added to a 20 mL

3

2

vial charged with Hbpi (319.4 mg, 1.07 mmol), RuCl (PPh )

2

3 3

(

974.4 mg, 1.02 mmol), TlPF (355 mg, 1.02 mmol), and a stir

washed with Et O (4 × 10 mL), aﬀording Ru(Hb4Clpi)-

6

2

bar. The reaction solution was allowed to stir at 70 °C for 21 h.

After cooling to ambient temperature, TlCl was ﬁltered using a

ﬁne frit and the THF solvent was removed under vacuum. The

(PPh ) Cl[PF ] in 91% yield (242 mg). THF (10 mL) was

3

2

6

added to a 20 mL vial charged with Ru(Hb4Clpi)(PPh ) Cl-

3

2

t

[PF ] (242 mg, 0.206 mmol), NaO Bu (20.8 mg, 0.216 mmol),

6

5

482

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

and a stir bar. The reaction solution was allowed to stir at

ambient temperature for 30 min. The THF solvent was

removed under vacuum, and the crude product was extracted

with C H . The C H solution was lyophilized, and the product

(136.3 mg, 0.142 mmol), and a stir bar. The reaction solution

was allowed to stir at ambient temperature for 24 h. The

precipitate was collected on a frit and washed with Et O (4 ×

2

6

10 mL), and the product was extracted with DCM. Evaporation

of the volatiles under vacuum aﬀorded the product as a green

solid. Crystals were obtained from slow evaporation of a DCM

was washed with pentane (4 × 10 mL). Evaporation of the

volatiles under vacuum aﬀorded the product as a green powder.

1

Yield: 131 mg (62%). H NMR (700 MHz, CD Cl ): δ 9.87 (d,

1

2

solution at 5 °C (DCM/Hex). Yield: 113 mg (78%). H NMR

1

5

^JHH = 7.0 Hz, 2H, H ), 7.60 (dd, J = 5.6, 2.8 Hz, 2H, H ),

HH

(400 MHz, CD Cl ): δ 10.78 (s, 1H), 7.79 (d, J = 7.6 Hz,

2

HH

6

4

7

6

.39 (dd, J_H_H= 5.6, 2.8 Hz, 2H, H ), 7.17 (d, J = 2.1 Hz, H ),

.07 (t, J_H_H= 7.0 Hz, 6H, PPh ), 6.94−6.88 (m, 24H, PPh ),

.27 (dd, J_H_H= 7.0, 2.8 Hz, 2H, H ). C{ H} (176 MHz,

HH

1

H), 7.73−7.69 (m, 12H), 7.57 (t, J = 7.6 Hz, 1H), 7.51 (t,

HH

3

^JHH = 7.6 Hz, 2H), 7.30 (t, J ⁼⁷^.⁸^H^z^,¹^H⁾^,⁷^.²²⁽^t^,^JHH = 7.8

2

13

1

HH

Hz, 1H), 7.13−7.04 (m, 19H), 6.89 (d, J = 7.6 Hz, 1H), 6.30

HH

CD Cl ): δ 158.61, 157.72, 153.38, 141.84, 141.55, 133.84,

2

(

d, J = 8.0 Hz, 1H), 6.21 (d, J = 8.0 Hz, 1H), 2.31 (s, 3H,

HH HH

1

2

1

8

6

³²^.⁴⁶⁽^t^,^JCP = 17.5 Hz, ipso-CP), 129.11, 129.02, 127.72,

13

1

Me), 0.99 (s, 3H, Me). C{ H} (176 MHz, CD Cl ): δ 159.83,

3

1

2

26.49, 120.25, 116.92. P{ H} NMR (162 MHz, C D ): δ

6

−1

159.34, 148.76, 138.27, 136.19, 135.56, 135.02, 134.19, 131.71,

5.13 (s, PPh ). IR (ATR, cm ): 3056, 1544, 1514, 1492,

3

1

31.10, 130.14, 129.86, 129.17, 128.76, 127.65, 127.09, 124.64,

446, 1372, 1318, 1299, 1210, 1186, 1121, 1089, 1001, 914,

87, 868, 808, 776, 737, 694. Anal. Calculated (found): C,

3.07 (62.64); H, 3.92 (4.02); N, 6.81 (6.63).

23.27, 120.49, 118.49, 108.24, 24.23, 20.54. ³P{ H} NMR

1

−1

(

162 MHz, CD Cl ): δ 24.94 (s, PPh ). IR (ATR, cm ): 3170,

2 2 3

3

1

056, 1640, 1609, 1581, 1548, 1482, 1467, 1432, 1305, 1270,

214, 1150, 1107, 1089, 1030, 1007, 805, 760, 748, 685, 655.

Ru(b4OMepi)(PPh ) Cl (7-OMe). THF (10 mL) was added

3

2

to a 20 mL vial charged with Hb4OMepi (99 mg, 0.276 mmol),

RuCl (PPh ) (251.6 mg, 0.262 mmol), TlPF (91.7 mg, 0.262

Anal. Calculated (found): C, 65.69 (65.46); H, 4.63 (4.65); N,

2

3 3

6

.84 (6.65).

mmol), and a stir bar. The reaction solution was allowed to stir

at 60 °C for 16 h. After the solution cooled to ambient

temperature, TlCl was ﬁltered using a ﬁne frit and the THF

solvent was removed under vacuum. The crude product was

Dimer 9. THF (5 mL) was added to a 20 mL vial charged

t

with 4 (110 mg, 0.152 mmol), NaO Bu (18.9 mg, 0.197 mmol),

and a stir bar. The reaction solution was allowed to stir at

ambient temperature for 18 h. The THF solvent was removed

under vacuum, and the crude product was washed with pentane

(4 × 20 mL) then extracted with C H . The C H solution was

washed with Et O (4 × 10 mL), aﬀording Ru(Hb4OMepi)-

2

(

PPh ) Cl[PF ] in 87% yield (267 mg). THF (10 mL) was

3 2 6

added to a 20 mL vial charged with Ru(Hb4OMepi)(PPh ) Cl-

3

2

6

t

[

PF ] (264 mg, 0.227 mmol), NaO Bu (22.9 mg, 0.238 mmol),

lyophilized, and the product was washed with Et O (5 mL) and

6

2

and a stir bar. The reaction solution was allowed to stir at

ambient temperature for 30 min. The THF solvent was

removed under vacuum, and the crude product was extracted

with C H . The C H solution was lyophilized and the product

pentane (4 × 10 mL). Evaporation of the volatiles under

vacuum aﬀorded the product as a green powder. Crystals were

obtained from vapor diﬀusion of pentane into a C H solution

6

1

6

at ambient temperature. Yield: 64 mg (62%). H NMR (700

was washed with pentane (4 × 10 mL). Evaporation of the

MHz, C D ): δ 8.46 (t, J = 7.0 Hz), 7.23−7.18 (m), 6.09−

6

HH

volatiles under vacuum aﬀorded the product as a dark blue

6

.86 (m, PPh ), 6.75−6.71 (m, PPh ), 6.62−6.57 (m, PPh ).

1

3

powder. Yield: 192 mg (83%). H NMR (400 MHz, C D ): δ

6

Major species: 7.99 (d, J = 7.7 Hz, 1H), 7.94 (d, J = 7.0

1

HH

1

2

0.32 (d, J = 7.2 Hz, 2H, H ), 7.99 (dd, J = 5.6, 3.2 Hz,

HH HH

^H^z^,¹^H⁾^,⁷^.¹⁴⁽^t^,^JHH ⁼⁷^.⁰^H^z^,¹^H⁾^,⁶^.⁸¹⁽^d^,^JHH = 7.7 Hz, 1H),

^.⁶⁸⁽^t^,^JHH ⁼⁷^.⁷^H^z^,¹^H⁾^,⁶^.¹⁰⁽^d^,^JHH = 7.0 Hz, 1H), 5.41 (d,

^JHH = 7.7 Hz, 1H), 2.37 (t, J = 9.1 Hz, 1H), 1.78 (s, 3H, Me),

2.62 (t, J = 9.1 Hz, 1H). Minor species: 7.98 (d, J_H_H= 7.0 Hz,

H), 7.92 (d, J_H_H= 7.7 Hz, 1H), 7.09−7.06 (m, 2H), 7.00 (d,

J_H_H= 7.7 Hz, 1H), 6.65 (t, J_H_H= 7.7 Hz, 1H), 5.83 (d, J

5

H, H ), 7.50−7.46 (m, 12H, PPh ), 6.99 (d, J = 2.8 Hz, 2H,

3

HH

6

4

H ), 6.79−6.78 (m, 18H, PPh ), 5.86 (dd, J = 7.2, 3.2 Hz,

3

HH

2

13

1

2

1

H, H ), 3.10 (s, 6H). C{ H} (176 MHz, CD Cl ): δ 165.35,

2 2

−

1

58.46, 158.10, 153.57, 141.82, 133.99, 133.23 (t, J_C_P= 16.6

Hz, ipso-CP), 128.76, 128.57, 127.55, 119.76, 109.69, 106.69,

₅_.₇₆_.₃1P{ H} NMR (162 MHz, C D ): δ 26.40 (s, PPh ). IR

1

=

HH

5

(

6 6 3

−1

7.0 Hz, 1H), 5.69 (d, J_H_H= 7.7 Hz, 1H), 3.10 (t, J = 9.1 Hz,

ATR, cm ): 3050, 1619, 1558, 1509, 1467, 1436, 1378, 1328,

178, 1091, 1039, 1002, 845, 744, 694. Anal. Calculated

1

3

1

H), 0.70 (s, 3H, Me), −2.77 (t, J = 9.1 Hz, 1H). C{ H} (176

1

MHz, CD Cl ): δ 174.34, 159.04, 157.10, 151.97, 150.29,

(

found): C, 65.98 (65.45); H, 4.55 (4.44); N, 6.87 (6.86).

2

1

49.06, 142.89, 141.59, 134.77, 134.62, 133.51, 133.45, 133.13,

32.49, 126.08, 120.14, 119.67, 118.20, 116.30, 112.59, 27.60,

Ru(b4OHpi)(PPh ) Cl (7-OH). THF (10 mL) was added to a

3

2

0 mL vial charged with Hb4OHpi (63.6 mg, 0.192 mmol),

t

31

1

RuCl (PPh ) (184.1 mg, 0.192 mmol), NaO Bu (19.4 mg,

24.78, 20.37, 19.95. P{ H} NMR (283 MHz, C D ): δ 33.05

6 6

2

3 3

−1

0.202 mmol), and a stir bar. The reaction solution was allowed

(s, minor), 31.34 (s, major). IR (ATR, cm ): 3046, 2968,

2900, 2613, 1533, 1560, 1504, 1461, 1431, 1387, 1324, 1286,

1239, 1188, 1159, 1112, 1090, 1030, 998, 976, 903, 792, 764,

739, 693. Anal. Calculated (found): C, 66.27 (66.06); H, 4.39

(4.35); N, 10.17 (10.16).

to stir at 70 °C for 16 h. After the solution was cooled to

ambient temperature, the THF solvent was removed under

vacuum. The crude product was washed with DCM (4 × 10

mL), H O (4 × 10 mL), and Et O (4 × 10 mL). Evaporation of

2

the volatiles under vacuum aﬀorded the product as a dark blue

powder. Crystals were obtained from vapor diﬀusion of pentane

ASSOCIATED CONTENT

31

1

■

into a THF solution at −35 °C. Yield: 151 mg (79%). P{ H}

−1

S

*

Supporting Information

NMR (162 MHz, THF): δ 26.40 (s, PPh ). IR (ATR, cm ):

3

1

046, 1555, 1517, 1478, 1432, 1324, 1296, 1186, 1103, 1091,

017, 972, 912, 862, 795, 744, 693. Anal. Calculated (found):

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acscatal.5b00952.

C, 65.42 (65.45); H, 4.27 (4.13); N, 7.06 (7.12).

Kinetic plots and NMR spectra (PDF)

Chemical characterization (CIF )

Ru(HbMepi)(PPh ) Cl (8). DCE (5 mL) was added to a vial

3

2

charged with HbMepi (51.2 mg, 0.156 mmol), RuCl (PPh )

2

3 3

5

483

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

AUTHOR INFORMATION

Research Article

6

4

(

7

891. (c) Gom

857−4963.

19) Johnson, J. B.; Bac

684.

(20) Tseng, K.-N. T.; Kampf, J. W.; Szymczak, N. K. ACS Catal.

2015, 5, 411−415.

21) Csonka, R.; Speier, G.; Kaizer, J. RSC Adv. 2015, 5, 18401−

18419.

́

ez-Gallego, M.; Sierra, M. A. Chem. Rev. 2011, 111,

■

Corresponding Author

E-mail: nszym@umich.edu.

̈

kvall, J.-E. J. Org. Chem. 2003, 68, 7681−

*

Notes

The authors declare no competing ﬁnancial interest.

(

ACKNOWLEDGMENTS

■

(

22) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor,

G. C. J. Chem. Soc., Dalton Trans. 1984, 1349−1356.

23) Brookhart, M.; Green, M. L. H.; Parkin, G. Proc. Natl. Acad. Sci.

We thank the anonymous reviewers for providing valuable

suggestions during the submission process. Acknowledgment is

made to the Donors of the American Chemical Society

Petroleum Research Fund for support of this research (53760-

DNI3), the University of Michigan Department of Chemistry,

and the National Science Foundation (CHE-0840456) for X-

ray instrumentation. N.K.S. is an Alfred P. Sloan Research

Fellow and a Dow Corning Assistant Professor.

(

U. S. A. 2007, 104, 6908−6914.

(

24) The hydrogen atoms were reﬁned assuming an idealized

geometry for the sp carbon; thus, the M−H distances are approximate

3

values.

(25) An identical rate (7.7(3) × 10⁻M·s ) was obtained for 4

when 2 equiv of NaOTf was added, which suggests that the NaOTf

generated when 5 was used had no eﬀect on the catalytic rate of

1PhEtOH dehydrogenation.

26) It should be noted that although the metal−ligand cooperative

pathway was eﬀectively turned oﬀ for 5, this hybrid hydrogen-transfer

mechanism may still be a plausible competing pathway for 3.

27) Kaneda, K.; Mori, K.; Hara, T.; Mizugaki, T.; Ebitani, K. Catal.

Surv. Asia 2004, 8, 231−239.

28) Vijayasri, K.; Rajaram, J.; Kuriacose, J. C. J. Mol. Catal. 1987, 39,

03−217.

29) Thorson, M. K.; Klinkel, K. L.; Wang, J.; Williams, T. J. Eur. J.

Inorg. Chem. 2009, 2009, 295−302.

30) Fluorination has a destabilizing eﬀect (15 kcal/mol) on the

transition state for β-H elimination. See references: (a) Gellman, A. J.;

Dai, Q. J. Am. Chem. Soc. 1993, 115, 714−722. (b) Johnson, T. J.;

Huffman, J. C.; Caulton, K. G. J. Am. Chem. Soc. 1992, 114, 2725−

726.

31) Further conﬁrmation of the ruthenium alkoxide complex was

5

−1

REFERENCES

■

(

1) (a) Gunanathan, C.; Milstein, D. Science 2013, 341, 249−260.

(

b) Dobereiner, G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681−

7

(

03. (c) Friedrich, A.; Schneider, S. ChemCatChem 2009, 1, 72−73.

2) (a) Bonitatibus, P. J.; Chakraborty, S.; Doherty, M. D.; Siclovan,

O.; Jones, W. D.; Soloveichik, G. L. Proc. Natl. Acad. Sci. U. S. A. 2015,

(

2

(

1

2

3

12, 1687−1692. (b) Marr, A. C. Catal. Sci. Technol. 2012, 2, 279−

87. (c) Johnson, T. C.; Morris, D. J.; Wills, M. Chem. Soc. Rev. 2010,

9, 81−88.

(

3) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.

(

2

(

004, 248, 2201−2237.

4) Gunanathan, C.; Milstein, D. Acc. Chem. Res. 2011, 44, 588−602.

5) (a) Li, H.; Wang, X.; Huang, F.; Lu, G.; Jiang, J.; Wang, Z.-X.

Organometallics 2011, 30, 5233−5247. (b) Li, H.; Zheng, B.; Huang,

2

(

K.-W. Coord. Chem. Rev. 2015, 293−294, 116−138.

(

6) Zeng, G.; Sakaki, S.; Fujita, K.-i.; Sano, H.; Yamaguchi, R. ACS

not possible by 2D NMR correlation experiments, such as HMBC. For

a discussion, see: Phosphorous-31 NMR Spectroscopy; Kuhl, O., Ed.;

Springer: New York, 2008; pp 7−23.

Catal. 2014, 4, 1010−1020.

7) Zhang, G.; Vasudevan, K. V.; Scott, B. L.; Hanson, S. K. J. Am.

Chem. Soc. 2013, 135, 8668−8681.

8) Tseng, K.-N. T.; Kampf, J. W.; Szymczak, N. K. Organometallics

013, 32, 2046−2049.

9) Complex 3 and Ru(bMepi)(PPh )Cl (4) are now commercially

(

6

(

4

32) Ritter, J. C. M.; Bergman, R. G. J. Am. Chem. Soc. 1998, 120,

(

2

(

826−6827.

33) Blum, O.; Milstein, D. J. Organomet. Chem. 2000, 593−594,

79−484.

34) For a recent example of large entropic factors in an unimolecular

reaction performed in a protic solvent, see: Byers, J. A.; Jamison, T. F.

Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 16724−16729.

35) For other examples demonstrating similar chemoselectivity see:

3

available.

(

10) See Figure S1 .

(

11) Mikhailine, A. A.; Maishan, M. I.; Lough, A. J.; Morris, R. H. J.

Am. Chem. Soc. 2012, 134, 12266−12280.

(

12) Concentrations higher than 20-fold excess of PPh with respect

3

(

a) Chakraborty, S.; Lagaditis, P. O.; Fo

Hazari, N.; Holthausen, M. C.; Jones, W. D.; Schneider, S. ACS Catal.

014, 4, 3994−4003. (b) Manzini, S.; Urbina-Blanco, C. A.; Nolan, S.

P. Organometallics 2013, 32, 660−664.

36) The standard enthalpy of formation in the gas phase for

̈

rster, M.; Bielinski, E. A.;

to 3 are likely to have an eﬀect on the dehydrogenation rate; however,

those scenarios deviate greatly from the standard conditions (Scheme

2

4

) and thus are not meaningful to obtaining relevant analysis on the

AAD mechanism.

13) The neat concentration of 1PhEtOH is 8.3 M. Low 1PhEtOH

concentrations are operationally challenging because replacing

PhEtOH impacts other factors, such as dielectric constant, solvation,

(

acetophenone and benzaldehyde are −87 and −37 kJ/mol,

respectively. See http://webbook.nist.gov/chemistry/name-ser.html .

(37) Olmstead, W. N.; Margolin, Z.; Bordwell, F. G. J. Org. Chem.

1980, 45, 3295−3299.

1

and polarizability, which inﬂuence the kinetic studies and the AAD

mechanism.

(

14) Morse, P. M.; Spencer, M. D.; Wilson, S. R.; Girolami, G. S.

Organometallics 1994, 13, 1646−1655.

15) Wilkins, R. G. Kinetics and Mechanism of Reactions of Transition

Metal Complexes; Wiley-VCH: Weinheim, Germany, 2003; pp 199−

56.

16) Casey, C. P.; Singer, S. W.; Powell, D. R.; Hayashi, R. K.;

Kavana, M. J. Am. Chem. Soc. 2001, 123, 1090−1100.

17) (a) Mueller, J. A.; Goller, C. P.; Sigman, M. S. J. Am. Chem. Soc.

004, 126, 9724−9734. (b) Blum, O.; Milstein, D. J. Am. Chem. Soc.

(38) Golden, D. M.; Benson, S. W. Chem. Rev. 1969, 69, 125−134.

(39) Ellis, W. W.; Ciancanelli, R.; Miller, S. M.; Raebiger, J. W.;

Rakowski DuBois, M.; DuBois, D. L. J. Am. Chem. Soc. 2003, 125,

12230−12236.

(40) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.

(41) (a) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed.

2012, 51, 2772−2775. (b) He, L.-P.; Chen, T.; Gong, D.; Lai, Z.;

Huang, K.-W. Organometallics 2012, 31, 5208−5211. (c) Wang, Z.;

Belli, J.; Jensen, C. M. Faraday Discuss. 2011, 151, 297−305.

(d) Nielsen, M.; Kammer, A.; Cozzula, D.; Junge, H.; Gladiali, S.;

Beller, M. Angew. Chem., Int. Ed. 2011, 50, 9593−9597.

(

2

(

2

́

1

995, 117, 4582−4594. (c) Martínez-Prieto, L. M.; Avila, E.; Palma,

́

P.; Alvarez, E.; Cam

́

pora, J. Chem. - Eur. J. 2015, 21, 9833−9849.

(

18) For further discussions of equilibrium isotope eﬀects at

(42) (a) Grellier, M.; Sabo-Etienne, S. Dalton Trans. 2014, 43, 6283−

6286. (b) Crabtree, R. H. Energy Environ. Sci. 2008, 1, 134−138.

(43) (a) Chakraborty, S.; Brennessel, W. W.; Jones, W. D. J. Am.

Chem. Soc. 2014, 136, 8564−8567. (b) Fujita, K.-i.; Tanaka, Y.;

transition-metal centers see: (a) Slaughter, L. M.; Wolczanski, P. T.;

Klinckman, T. R.; Cundari, T. R. J. Am. Chem. Soc. 2000, 122, 7953−

7975. (b) Janak, K. E.; Parkin, G. J. Am. Chem. Soc. 2003, 125, 6889−

5

484

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

ACS Catalysis

Research Article

Kobayashi, M.; Yamaguchi, R. J. Am. Chem. Soc. 2014, 136, 4829−

4832. (c) Kawahara, R.; Fujita, K.-i.; Yamaguchi, R. Angew. Chem., Int.

Ed. 2012, 51, 12790−12794. (d) Yamaguchi, R.; Ikeda, C.; Takahashi,

Y.; Fujita, K.-i. J. Am. Chem. Soc. 2009, 131, 8410−8412.

(

44) Siegl, W. O. J. Org. Chem. 1977, 42, 1872−1878.

5

485

DOI: 10.1021/acscatal.5b00952

ACS Catal. 2015, 5, 5468−5485

Article Doi

Mechanism of N,N,N-Amide Ruthenium(II) Hydride Mediated Acceptorless Alcohol Dehydrogenation: Inner-Sphere β-H Elimination versus Outer-Sphere Bifunctional Metal-Ligand Cooperativity

DOI: 10.1021/acscatal.5b00952

Source and publish data:

Authors:

Article abstract of DOI:10.1021/acscatal.5b00952

Full text of DOI:10.1021/acscatal.5b00952

Products guided by the article

R&D Labs maybe for 1428232-72-7

Relevant to this article

Hot Product