A thermodynamic analysis of rhenium(I)-Formyl C-H bond formation via base-assisted heterolytic H2 cleavage in the secondary coordination sphere

Article

pubs.acs.org/Organometallics

A Thermodynamic Analysis of Rhenium(I)−Formyl C−H Bond

Formation via Base-Assisted Heterolytic H₂Cleavage in the

Secondary Coordination Sphere

Thomas S. Teets, Jay A. Labinger,* and John E. Bercaw*

Arnold and Mabel Beckman Laboratories of Chemical Synthesis, California Institute of Technology, Pasadena, California 91125,

United States

S

* Supporting Information

ABSTRACT: Conversion of synthesis gas, a mixture of

carbon monoxide and hydrogen, into value-added C_n≥2

products requires both C−H and C−C bond-forming events.

Our group has developed a series of molecular complexes,

based on group 7 (manganese and rhenium) carbonyl

complexes, to interrogate the elementary steps involved in

the homogeneous hydrogenative reductive coupling of CO.

Here, we explore a new mode of H₂activation, in which strong

bases in the secondary coordination sphere are positioned to

assist in the heterolytic cleavage of H₂to form a formyl C−H

bond at a rhenium-bound carbonyl. A series of cationic

rhenium(I) complexes of the type [Re^I(P∼B:-κ₁-P)(CO)₅]ⁿ(n

= 0, +1), where P∼B: is a phosphine ligand with a tethered

strong base, are prepared and characterized; measurement of their protonation equilibria demonstrates a pronounced attenuation

of the basicity upon coordination. Formyl complexes supported by these ligands can be prepared in good yield by hydride

delivery to the parent pentacarbonyl complexes, and several of the free-base formyl complexes can be protonated, generating

observable [Re^I(P∼BH-κ₁-P)(CHO)(CO)₄]ⁿcomplexes. Intramolecular hydrogen bonding is evident for one of the complexes,

providing additional stabilization to the protonated formyl complex. By measuring both the hydricity of the formyl, ΔG°_H−, and

its pK_a, the overall free energy of H₂cleavage is calculated from an appropriate cycle and found to be thermodynamically uphill in

all cases (in the best case by only about 8 kcal/mol), although signiﬁcantly dependent upon the properties of the supporting

ligand.

structure, bonding, and energetic proﬁles of key reaction

intermediates.⁹There are some examples of homogeneous

catalytic synthesis gas conversion, with product selectivities that

can favor C_n≥2hydrocarbons¹⁰or alcohols,¹¹depending on the

choice of catalyst and operating conditions. However, to date

homogeneous catalysts are not used industrially, and most of

the focus with molecular species has been on modeling and

studying proposed reaction intermediates. Earlier examples

include studies on the formation, properties, and reactivity of

metal formyl and related oxycarbene complexes, many of which

are stable enough to be isolated;¹²in some cases carbon−

carbon bond formation was observed upon reduction of metal-

bound CO with hydride equivalents.¹³C−C bond formation

was also modeled by studying the migratory insertion of CO

into metal−alkyls, which can be promoted by Lewis or

Brønsted acidic additives.¹⁴

INTRODUCTION

■

Synthesis gas, or syngas, a mixture of carbon monoxide and

hydrogen, can be derived from coal, biomass, or methane and is

the key intermediate in the indirect conversion of these

abundant, low-value natural resources into valuable fuels or

commodity chemicals.¹The best-known method for upgrading

syngas is the Fischer−Tropsch process, which converts syngas

into a complex mixture of hydrocarbon products.²The

Fischer−Tropsch process traces its origins to over 80 years

ago,³but to this day continued improvements are sought,

primarily aimed at the development of more active catalysts⁴

with narrower hydrocarbon distributions,⁵as well as modiﬁed

systems that are selective for oxygenated products.⁶Exper-

imental⁷and computational⁸studies on heterogeneous syn-

thesis gas conversion catalyses have proliferated in recent years,

leading to enhanced understanding of catalyst performance and

selectivity.

In these early examples, strong hydride donors were used to

form the C−H bond, and the Lewis acid additives that

promoted C−C bond formation typically also resulted in the

A complementary strategy for converting syngas into more

valuable products is homogeneous catalysis, which oﬀers the

potential of higher selectivity, milder operating conditions, and

improved mechanistic understanding owing to the wealth of

solution-phase techniques available for interrogating the

Received: August 12, 2013

Published: October 3, 2013

5530

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

formation of strong bonds between the Lewis acid and the acyl

oxygen (perhaps better described as a stabilized oxycarbene in

these cases). Practical application would require the use of H₂

(or hydride equivalents derived from H₂) to form the C−H

bond and the use of milder Lewis acids to promote C−C bond

formation. In recent years, advances in both of these areas have

been made, by both our group and others. Noteworthy

achievements include the use of secondary-coordination-sphere

alkylborane Lewis acids to assist in heterolytic H₂cleavage and

C−C bond formation at rhenium carbonyl complexes;¹⁵

employing main group frustrated Lewis pairs (FLPs) for

homogeneous, stoichiometric CO hydrogenation and C−O

bond cleavage;¹⁶reductive coupling of CO at an iron complex

with participation from the supporting ligand;¹⁷and demon-

strations that milder Lewis acids such as zinc(II) and

calcium(II) can promote C−C bond formation by migratory

insertion.¹⁸In spite of the fundamental advances described in

these previous reports, several limitations remain. In the

systems employing borane Lewis acid promoters, strong B−O

bonds preclude product release and catalytic turnover, and the

later examples that use milder Lewis acids also result in the

formation of strong bonds between the rhenium center and

charge-compensating halide ions that accompanied the Lewis

acid, which could not be broken in a catalytically relevant

fashion.

In this work, we examine a new approach for H₂cleavage and

formyl C−H bond formation, inspired in part by the reports of

C−H bond formation mediated by H₂activation at FLPs,

where the resulting borohydride transfers hydride to CO. The

latter demonstrate that prior activation of H₂at a transition

metal center, long thought to be a requisite, is in fact not

necessary; perhaps the intermediacy of the borohydride can be

bypassed as well? We envisioned a system in which a Brønsted

base, positioned in the secondary coordination sphere of a

metal carbonyl complex, assists in heterolytic H₂cleavage,

delivering hydride directly to a carbonyl carbon and

simultaneously generating the conjugate acid of the appended

base, which should be able to hydrogen bond with the resultant

formyl oxygen and thus stabilize the product.

Obviously, the viability of this approach depends upon both

kinetic and thermodynamic constraints. There is no good way

to assess the former other than actually demonstrating a

successful instance; but the thermodynamics, which depend on

both the acid−base properties of the pendant group and the

hydricity of the metal formyl, can be assessed by an appropriate

combination of observables. As shown in Scheme 1, the desired

overall reaction can be decomposed into a thermodynamic

cycle with three elementary steps. The ﬁrst step, (a), is the

hydride-accepting free energy of the formyl, which is the

negative of the formyl’s hydricity, ΔG°_H−. The hydricities of

several formyl complexes have been tabulated, and they range

between 40 and 55 kcal/mol.¹⁹The free energy of reaction (b)

depends on the pK_aof the rhenium formyl, with a multiplier to

convert from unitless pK_ato kcal/mol; implicit in the measured

pK_ais a contribution from hydrogen bonding between the

formyl oxygen and the acidic proton, if present. Finally, to close

the thermodynamic cycle, the free energy for heterolytic H₂

cleavage is included, which has been determined to be 76.0

kcal/mol in acetonitrile,¹⁹the solvent of choice for these

studies. The overall free energy for H₂cleavage, which is the

sum of the energies for reactions (a)−(c) in Scheme 1, is

abbreviated as ΔG°_H2herein. It should also be noted that the

thermodynamic cycle can be written in a diﬀerent way, starting

Scheme 1. Thermodynamic Cycle for Heterolytic H₂

Cleavage

with protonation of the carbonyl complex and then addition of

hydride to generate the formyl. This leads to the same overall

reaction, but in practice the sequence outlined in Scheme 1 is

more practical for measuring the overall reaction thermody-

namics, as discussed later.

Rhenium(I) pentacarbonyl complexes of the type

[Re^I(P∼B:-κ₁-P)(CO)₅]ⁿ(n = 0, +1), where P∼B: is a

phosphine ligand with an appended strong base, were chosen

as the platform for the examination of these thermodynamic

considerations. We report here the synthesis and reactivity of

rhenium pentacarbonyl complexes with six diﬀerent base-

appended supporting ligands. In all cases, we were able to

obtain complexes with exclusive ligation through phosphorus,

leaving the base free, and to measure the pK_aof the protonated

species. In several cases we have been able to generate the

target structures, [Re^I(P∼BH-κ₁-P)(CO)₄(CHO)]ⁿ, by sequen-

tial reaction with strong hydride donor and weak acid, and

determine their free energies of formation from the parent

pentacarbonyl complex and H₂using the thermodynamic cycle

of Scheme 1. Unfortunately, all these values are unfavorable (in

the best case by only about 8 kcal/mol), due largely to a very

substantial attenuation of the basicity upon coordination, the

extent of which depends on the precise structure of the ligand.

Consistent with those ﬁndings, none of the complexes

[Re^I(P∼B:-κ₁-P)(CO)₅]ⁿare reactive toward H₂. Nonetheless,

the thermodynamic parameters and other results obtained in

this work provide a useful framework and design criteria for this

as well as related, potentially useful transformations.

RESULTS

■

Synthesis of Phosphine Ligands. Chart 1 presents the

phosphine ligands that have been studied here, with the

Chart 1

5531

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

abbreviations that will be used throughout. All ligands feature a

diphenylphosphine ligating group with an appended basic unit.

Most of the ligands in Chart 1 are described in the literature.

Proton Sponge-substituted ligand L1,²⁰N,N-dimethylbenzyla-

mino-substituted L2,²¹and morpholine-substituted L3²²were

prepared as previously outlined. The ligand L5, which features

the strong bicyclic base 1,8-diazabicyclo[5.4.0]undec-7-ene

(DBU), has also been prepared,²³but we opted for an alternate

route whereby DBU was selectively lithiated at the 6-position²⁴

with tert-butyllithium, followed by addition of Ph₂PCl to install

the diphenylphosphino substituent. Substituted phenol ligand

L6 (the protonated form of this ligand) was also prepared by a

modiﬁed procedure: we lithiated the 2-bromo-substituted

MOM-protected phenol precursor, whereas previous reports

of this class of ligand begin their syntheses with direct C−H

lithiation of the protected phenol precursor.^25,26

products, as a result of ligand substitution occurring at a similar

rate to chelation. The desired κ₁-P complexes are accessible

using more reactive cationic rhenium starting materials, which

allow for milder conditions. Scheme 3 summarizes the general

Scheme 3. Synthesis of Cationic Rhenium Pentacarbonyl

Complexes, with Isolated Yields in Parentheses

The only new ligand described in this study is L4, which

contains a 2-phenyl-1,1,3,3-tetramethylguanidine base. L4 was

accessed by a standard method for preparing mixed

triarylphosphines (Scheme 2): the brominated precursor 2-

Scheme 2. Synthesis of L4

synthesis of these compounds. The labile rhenium(I) starting

material Re^I(CO)₅(FBF₃)²⁸showed the greatest utility, and in

combination with L1 and L3−L6 in CH₂Cl₂gave the desired

complexes in good yields. Complex 5, [Re^I(L5-κ₁-P)(CO)₅]-

(BF₄), was observed as the major product in crude reaction

mixtures, but could not be isolated in pure form due to

contamination with the corresponding κ₂-P,N complex (vide

infra). For ligand L2, reaction with Re^I(CO)₅(FBF₃) gave a

complex mixture of products, tentatively attributed to

unselective binding of rhenium by phosphorus and nitrogen,

but treatment with Re^I(CO)₅(OTf) gave gradual conversion (1

week) to the desired complex [Re^I(L2-κ₁-P)(CO)₅](OTf) (2)

in good yield.

In solution, the κ₁-P coordination mode is most readily

discerned from the ³¹P{¹H} NMR chemical shift, and from the

ν_CO

̃

region of the IR spectra. The ³¹P{¹H} NMR spectra for

(2′-bromophenyl)-1,1,3,3-tetramethylguanidine²⁷was lithiated

at −78 °C via lithium−halogen exchange with tert-butyllithium,

and addition of Ph₂PCl gave the desired product L4, which was

isolated in 52% yield in gram quantities and found to be air-

stable.

all complexes show singlets, in some case quite broad (fwhm

∼30−120 Hz), near 0 ppm (vs D₃PO₄). Similar features have

been observed for other phosphine-ligated rhenium pentacar-

bonyl cations.^15c,18In addition, the solution IR features of

complexes 1−6, where three IR-active CO stretching modes are

evident, establish a C_4vgeometry and match closely with other

phosphine-ligated rhenium pentacarbonyl complexes.²⁹

Synthesis of Protonated Ligands. Ligands L1−L5 are all

protonated at nitrogen upon addition of acid. Tetraﬂuoroborate

salts of the ligands (abbreviated (LxH)(BF₄), where “x” is the

numerical designation for the ligand) were prepared as stable

solids by treating the free-base ligand with HBF₄. (L1H)(BF₄)

was prepared in EtOH by a previously reported procedure,²⁰

whereas the remaining ligands were protonated by HBF₄·Et₂O

in aprotic solvents. Good yields of the desired salts were

obtained, and all spectroscopic, mass spectrometric, and

combustion analyses indicate monoprotonation, with no

additional protonation at phosphorus or other available

heteroatoms. In addition, the solid-state structure of (L5H)-

(BF₄) was determined by X-ray crystallography and is shown in

Figure S1 of the Supporting Information. The N−H hydrogen

atom was located in the diﬀerence map, and the presence of a

Solid-state structures of complexes 1−4 and 6 have all been

determined by X-ray crystallography. The structure of 4,

representative of the series, is shown in Figure 1, with the

remaining structures shown as Figures S2−S5 in the

Supporting Information. The structures are unremarkable in

most aspects, with bond angles all very close to 90°, as expected

for a C_4v, axially elongated pseudo-octahedral structure. The

Re−P bond distances are just under 2.5 Å, with Re−C distances

all very near 2.0 Å. In addition, the distances between the

tethered basic heteroatom and the rhenium center are all quite

long; even the shortest such distance, 3.4840(7) Å between

Re(1) and the phenolic O(6) in the structure of 6, is much too

long to be considered even a weak interaction.

Complexes 1−3 and 6 are indeﬁnitely stable at room

temperature in either dichloromethane or acetonitrile solution.

However, complexes 4 and 5 do gradually decompose to their

respective κ₂-P,N complexes when held at room temperature in

solution. In the case of guanidine-substituted 4, the

decomposition is quite slow, occurring with a half-life on the

−

hydrogen-bonding interaction with the BF₄counteranion is

evident.

Synthesis and Properties of Cationic [Re^I(P∼B:-κ₁-

P)(CO)₅]⁺Complexes. Reactions between neutral rhenium(I)

starting materials Re^I(CO)₅Br or [Re^I(CO)₄(μ-Br)]₂and

ligands L1−L5 typically result in κ₂-P,N complexes as major

5532

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

Figure 2. X-ray crystal structure of 7, with thermal ellipsoids drawn at

the 50% probability level. Hydrogen atoms and the BF₄⁻counteranion

are omitted.

Figure 1. X-ray crystal structure of 4, with thermal ellipsoids drawn at

the 50% probability level. Hydrogen atoms and the BF₄⁻counteranion

are omitted for clarity.

Scheme 5. Synthesis of Protonated Rhenium Complexes 8−

11

order of several days, and multiple products are present, as

judged by NMR. For 5, with the DBU-substituted phosphine

ligand, chelation is comparatively more rapid, occurring with a

half-life of ca. 12 h at room temperature, such that complete

conversion is noted within 60 h. The decomposition of

complex 5 is outlined in Scheme 4, with the exclusive product

Scheme 4. Decomposition of 5 to 7

isolated, complexes 8−12 were soluble and indeﬁnitely stable in

acetonitrile solution, with no evidence for decomposition over

prolonged time periods.

Scheme 6. Synthesis of Complex 12

identiﬁed as [Re^I(L5-κ₂-P,N)(CO)₄](BF₄) (7). The broad

³¹P{¹H} resonance of 5 occurs at 1.3 ppm in dichloromethane,

shifting downﬁeld to 23.4 ppm and sharpening considerably in

7. The IR spectrum of 7 is indicative of a pseudo-C_s

coordination geometry brought on by chelation. In pentacar-

bonyl complex 5, three CO stretching frequencies are observed,

at 2151, 2091, and 2050 cm⁻¹, whereas in 7 four CO stretching

frequencies are observed, occurring at signiﬁcantly lower energy

(2109, 2020, 2006, and 1968 cm⁻¹), on account of the more

electron-rich rhenium center. And ﬁnally, complex 7 shows

enhanced crystallinity relative to 5, allowing its structure to be

unambiguously obtained by X-ray crystallography. The

structure of the cation is shown in Figure 2, which clearly

shows the connectivity that was anticipated from the above-

mentioned spectroscopic analyses. The P(1)−Re(1)−N(1)

chelate angle is observed to be 75.70(8)°.

Acid/Base Chemistry of Rhenium Complexes. With

complexes 1−6 in hand, we sought to explore their acid/base

chemistry. Complexes 1−5, which feature neutral nitrogenous

tethered bases, can be protonated at the basic nitrogen to

deliver dicationic complexes 8−12. As summarized in Scheme

5, complexes 8−11 were prepared by addition of HBF₄·Et₂O to

dichloromethane solutions of the corresponding [Re^I(Lx)-

(CO)₅]⁺complex. The desired dicationic [Re^I(LxH)(CO)₅]²⁺

precipitated from solution and could be isolated in high yield.

Owing to the inability to cleanly isolate DBU-substituted

complex 5, the complex [Re^I(L5H)(CO)₅](BF₄)₂(12) was

prepared by direct reaction of (L5H)(BF₄) with

Re^I(CO)₅(FBF₃) at 50 °C in CH₂Cl₂(Scheme 6). Once

Several characteristic spectroscopic features distinguish

complexes 8−12. The ³¹P{¹H} NMR spectra are shifted upon

protonation, downﬁeld for 8, 11, and 12 and upﬁeld for 9 and

10 relative to their free-base analogues, but in all cases still in a

spectral region characteristic of rhenium κ₁-P ligation. In the ¹H

NMR spectra, the N−H resonance is unambiguously located in

each case. The IR spectra in the ν_CO

̃

region are quite similar

to those of complexes 1−5, with hypsochromic shifts on the

order of 5 cm⁻¹, reﬂecting a slight decrease in electron density

at the metal center brought on by protonation.

The crystal structures of 8, 9, 11, and 12 were determined,

and that of 11 is shown in Figure 3 as a representative example.

The coordination geometries about the rhenium centers are

minimally perturbed relative to the structures of the

unprotonated complexes, showing nearly identical bond lengths

and angles. In each case the N−H hydrogen atom was located

in the diﬀerence map and is oriented in such a manner to

permit hydrogen bonding to a nearby counteranion. Analysis of

the hydrogen bonding interactions in 8, 11, and 12 shows a

5533

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

Complex 13 decomposes with a half-life of ca. 3 h at room

temperature in acetonitrile, via a two-stage process. A

metastable intermediate species, identiﬁed as Re^I(L6⁻-κ₂-

P,O)(CO)₄(14), gives way over longer time periods to

Re^I(L6⁻-κ₂-P,O)(CO)₃(NCMe) (15), which is conﬁrmed as

the facial isomer from the downﬁeld CO ¹³C{¹H} NMR

features. Complex 14 is identiﬁed by its ³¹P{¹H} NMR

resonance at 25.6 ppm in CD₃CN, suggestive of κ₂-P,O ligation,

and is the major product when 13 is heated in CH₂Cl₂.

Complex 15, which forms as the dominant product within 24 h

in MeCN at 80 °C from a solution of 13, is identiﬁed by a

³¹P{¹H} NMR shift of 33.2 ppm in CD₃CN. An isolated sample

1

of 15 shows a H NMR signal at 0.00 ppm in C₆D₆for the

bound MeCN ligand, with weak (1.4 Hz) coupling to ³¹P;

when dissolved in CD₃CN, the NMR spectrum shows one

equivalent of free CH₃CN.

Figure 3. X-ray crystal structure of 11, drawn at the 50% probability

level. Carbon-bound hydrogen atoms, the CH₂Cl₂solvent molecule,

−

and the outer-sphere BF₄counteranion are omitted.

pK_aValues of Ligands and Rhenium Pentacarbonyl

Complexes. Table 1 summarizes the quantitative acid/base

single-point hydrogen bond between the acidic proton and a

ﬂuorine atom of the BF₄⁻counterion; this is true even for 8, in

which the proton is “chelated” by Proton Sponge via an

intramolecular hydrogen bond. Complex 9 crystallized as a

Table 1. Summary of Acetonitrile pK_aValues for Ligands

and Rhenium Pentacarbonyl Complexes

mixed OTf⁻/BF₄salt, with preferential hydrogen bonding

−

ligand

pK_a

complex

pK_a

between the N−H proton and the triﬂate counterion.

Elemental analysis conﬁrms that 9 is also isolated as a mixed

salt in the bulk sample.

Conditions were also sought to deprotonate complex 6, to

expose a strongly basic phenoxide unit in the vicinity of the

rhenium center. Treatment of 6 with a suﬃciently strong base,

such as DBU, results in quantitative formation of the

zwitterionic complex Re^I(L6⁻-κ₁-P)(CO)₅(13), which was

identiﬁed spectroscopically. Scheme 7 summarizes the prep-

a

L1H⁺

L2H⁺

L3H⁺

L4H⁺

L5H⁺

L6

18.24(2)

18.0(2)

8

11.02(5)

13.00(3)

5.7(2)

9

17.4(2)

10

11

12

6

19.57(1)

22.62(2)

28.17(5)

13.09(4)

13.34(4)

16.5(1)

a

From ref 20.

properties of the free ligands and their rhenium pentacarbonyl

complexes, measured in acetonitrile. The values were

determined by establishing equilibria with acids or bases of

known strength, measuring equilibrium concentrations by

NMR, and combining the measured equilibrium constant

with the known pK_avia Hess’s law to determine the unknown

pK_a. The free ligand pK_avalues are largely in line with those of

the corresponding unsubstituted base, though it is worth noting

that for ligands 4−6 the observed pK_avalues are ca. 1−1.7 pK_a

units smaller than those of the corresponding unsubstituted

bases 2-phenyl-1,1,3,3-tetramethylguanidine (pK_a= 20.6),³⁰

DBU (pK_a= 24.34),³¹and phenol (pK_a= 29.14),³²

respectively.

Scheme 7. Synthesis and Decomposition of 13

For rhenium complexes 1−5 and 13, the basicity is

attenuated substantially compared to the free ligand. In terms

of ΔpK_a, the smallest attenuation is observed for complex 2,

which also has the largest spacer length between the basic atom

and the ligating phosphorus, with four carbon atoms separating

the two. The largest decreases in basicity occur for 3, which has

only a methylene spacer between the phosphine and the

morpholine base, and for substituted-phenolate complex 13,

where the anionic oxygen of the zwitterion suﬀers a nearly 12

order-of-magnitude decrease in basicity relative to the free

phosphinophenolate L6⁻. For complexes 1 and 4, which

contain three-carbon aromatic spacers between the neutral

nitrogenous base and the phosphine, intermediate pK_a

decreases of 7.22 and 6.48 are observed, respectively. Complex

5, which has a three-carbon nonaromatic spacer between

phosphine and basic nitrogen, shows a decrease in pK_aof 9.28

units.

aration and decomposition of 13. The most identifying NMR

feature for 13 is the ³¹P{¹H} signal, which shifts downﬁeld from

that of 6 (−4.6 to 1.8 ppm) but is still diagnostic of a κ₁-P

1

complex. In the H NMR spectrum, the resonance for the aryl

proton ortho to phosphorus shifts upﬁeld dramatically, from

6.82 ppm to 6.06 ppm, and no O−H signal is detected, both of

which are suggestive of deprotonation of L6 to reveal a

phenoxide. And ﬁnally, the IR spectrum shows three CO

stretches (2100, 2074, 1996 cm⁻¹), consistent with a C_4v

geometry and at substantially lower frequencies than those of

6 (2157, 2102, 2048 cm⁻¹).

Synthesis of Rhenium(I) Formyl Complexes. Com-

plexes 1−5 and 13 are unreactive toward H₂, at room

5534

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

temperature and pressures up to 3 atm. To assess the

thermodynamics of H₂cleavage, as outlined in Scheme 1, we

sought conditions to prepare the neutral formyl complexes,

which would allow a comprehensive study of their ΔG°_H−and

pK_ato calculate the overall thermodynamics. We found that the

anionic tungsten(0) hydride complex (PPN)[cis-W⁰(H)(P-

(OMe)₃)(CO)₄] (W−H), ﬁrst described by M. Darensbourg

and co-workers³³and later utilized by DuBois et al.³⁴to prepare

a Cp*-ligated rhenium formyl, proved to be an eﬀective hydride

reagent for rapidly generating high yields of the neutral formyl

complexes 16−20, as summarized in Scheme 8, with

monitoring of reactions shows that they are formed in

moderate to excellent yields and decompose with half-lives of

two hours or more. Table 2 summarizes the key spectral

Table 2. Summary of Spectral Characteristics of Rhenium

a

Formyl Complexes

2

complex δ_CHO(¹H) (ppm) δ_CHO(¹³C) (ppm) J_CP(Hz) ν_CO(cm⁻¹

)

16

17

18

19

20

21

14.91

14.98

15.14

14.86

15.16

15.14

259.4

259.6

259.2

260.0

9.1

1590

1589

1584

9.4

10.4

9.6

b

ND

Scheme 8. Synthesis of Rhenium Formyl Complexes 16−20

274.9

12.0

a

b

Recorded in acetonitrile. Not determined on account of the low

yield and poor stability of complex 20.

characteristics of rhenium formyl complexes 16−21. In general

the spectral characteristics attributable to the formyl group

show little dependence on the identity of the supporting

1

phosphine. In the H NMR spectra, the characteristic formyl

C−H resonance appears near 15 ppm, whereas in the ¹³C{¹H}

NMR the formyl carbon resonates close to 260 ppm,

substantially downﬁeld of the carbonyl resonances. For

complex 21, the ¹³C{¹H} NMR resonance appears further

downﬁeld, at 274.9 ppm, likely due to the negative charge and/

or an interaction with BEt₃or Li⁺leading to more carbene

character.^12bFurther support for an interaction with BEt₃

approximate yields from NMR integration shown. Within

minutes of mixing equimolar amounts of pentacarbonyl

complexes 1−5 and W−H in acetonitrile all spectral data are

consistent with the formation of unstabilized rhenium formyl

complexes cis-Re^I(Lx)(CHO)(CO)₄(16−20). In all instances,

the tungsten-containing product is identiﬁed as cis-W⁰(P-

(OMe)₃)(NCMe)(CO)₄(W⁰), identiﬁed by its NMR features

and with IR spectral features nearly matching a closely related

complex.³⁵The spectral features attributed to W⁰show no

dependence on the identity of the rhenium formyl, conﬁrming

an absence of interaction between the formyl oxygen and the

tungsten center, which has been previously been observed.³⁴

In contrast to the above, hydride transfer from W−H to

phenol-substituted complex 6 was not observed. With a single

equivalent of W−H, only deprotonation was observed, giving

zwitterion 13 as the sole product. With two equivalents, 13 was

again observed, along with an equivalent of unreacted W−H,

with no hydride transfer on the time scale of the decomposition

of 13. However, the anionic formyl complex Li[cis-Re^I(L6⁻)-

(CHO)(CO)₄] could be generated from 6 by treatment with

two equivalents of LiHBEt₃(Scheme 9).

1

comes from the H NMR spectrum, where the resonances

for BEt₃are shifted upﬁeld relative to those of free BEt₃. It is

also possible that the basic phenoxide oxygen interacts with

BEt₃or Li⁺, though NMR evidence for this this possibility is not

conclusive. The ¹³C NMR shift of the phenoxide C1 remains

far downﬁeld (169.3 ppm), and the ¹H NMR resonance for the

aryl proton ortho to phosphorus is very far upﬁeld (5.82 ppm),

both of which are suggestive of considerable phenoxide

character (see Experimental Section). In all complexes,

couplings of the formyl proton and carbon with the ³¹P

nucleus are quite small, ≤1.4 Hz for ³J_HPand 9−10 Hz for ²J_CP,

indicating a cis geometry. The CO formyl stretch in the IR

spectrum is essentially independent of the identity of the

phosphine, appearing near 1590 cm⁻¹for 16−20 and at slightly

lower energy (1584 cm⁻¹) for 21. And ﬁnally, other spectral

features attributed to the CO ligands are indicative of the

altered coordination geometry and increase in electron density

at the rhenium center; the three distinct ¹³C{¹H} CO peaks in

16−19 are ca. 10−15 ppm downﬁeld of the two observed

Rhenium formyl complexes 16−21 were not isolated, owing

to the diﬃculty of separating them from their respective

byproducts and their instability under vacuum, though spectral

resonances in pentacarbonyl complexes 1−5, and the ν_CO

̃

IR

spectral region shows four bands,³⁶indicative of the decrease to

approximate C_ssymmetry and occurring at lower energies than

those of the parent pentacarbonyl complexes.

Scheme 9. Synthesis of Anionic Borane-Stabilized Formyl

Complex 21

Protonation of Formyl Complexes. Treatment of Proton

Sponge-substituted formyl complex 16 with lutidinium

tetraﬂuoroborate resulted in gradual formation, over a period

of 45 min, of pentacarbonyl complex 1 and H₂, with no NMR

evidence for a protonated formyl complex. Substituting the

stronger acid anilinium tetraﬂuoroborate, or subjecting

complexes 18 and 20 to similar treatments, likewise failed to

produce an observable protonated formyl complex in

appreciable yield. In contrast, treatment of complexes 17, 19,

and 21 with weak acids gave protonated formyl complexes 22−

24 as major products, which were spectroscopically observable.

Optimum yields were obtained when the acid was added to

5535

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

thawing solutions of the formyl complex, which inhibited a side

reaction where the acid directly reacted with the formyl directly

to release H₂. Scheme 10 summarizes the protonation of formyl

complexes 17, 19, and 21 to generate 22−24.

formyl resonance shifts from 14.86 ppm to 15.38 ppm, and a

sharp NH resonance appears at 9.23 ppm. In the upﬁeld region

the N(CH₃)₂resonance shifts downﬁeld from 2.20 ppm in 19

to a much broader signal at 2.73 ppm in 23. The downﬁeld

region of the ¹³C{¹H} NMR spectrum is also diagnostic; the

peak attributed to the formyl carbon shifts from 260.0 ppm in

19 to 269.4 ppm in 23, whereas the CO carbons undergo a very

slight, ca. 2 ppm upﬁeld shift. Complex 23 is moderately stable

in solution, decomposing over the course of ca. 30 h via

dehydrogenation, which produces pentacarbonyl complex 4,

and decarbonylation, which produces protonated hydride

complex 25, as depicted in Scheme 11. Complexes 4 and 25

Scheme 10. Protonation of Formyl Complexes 17, 19, and

21 to Generate 22−24, with Estimated NMR Yields in

Parentheses

Scheme 11. Decomposition Pathways for Protonated Formyl

Complex 23

Complex 22 was formed only in moderate yields (∼50%),

and its poor stability and broad NMR features complicated

1

characterization. Nevertheless, H and ³¹P{¹H} NMR spectra

clearly indicated the formation of 22. In the ¹H NMR

spectrum, the formyl CHO resonance shifts downﬁeld, from

14.98 ppm in 17 to 15.15 ppm in 22, with considerable

broadening in the latter. The ³¹P{¹H} NMR signal of 13.0 ppm

for protonated complex 22 is also distinct from that of the free-

base precursor 17 (11.6 ppm). Once formed, complex 22

decomposes by loss of H₂, and within 6 h at room temperature

it has completely given way to pentacarbonyl complex 2 as the

major product. Neutral complex 24, supported by phenol-based

ligand L6, was formed in better yields (∼75%) and had much

more distinct spectroscopic features than those of complex 22,

but also suﬀered from poor stability. Initial NMR spectra after

addition of 2,6-lutidinium tetraﬂuoroborate to a solution of 21

show complex 24 as the major species. Figure S9 shows the

downﬁeld region of the NMR spectra of 21 and 24, showing

are present in a 2.3:1 ratio at the end of the reaction, suggesting

that dehydrogenation occurs twice as fast as decarbonylation.

Complex 25 is most readily identiﬁed by a ¹H NMR doublet at

−5.51 ppm (²J_HP=20.4 Hz), suggestive of a hydride cis to the

phosphine, and a singlet at 8.71 ppm for the NH proton,

distinct from that of complex 23 and indicating that hydride

complex 25 remains protonated upon decarbonylation.

1

the distinct change that occurs upon protonation. In the H

NMR spectrum, the formyl resonance shifts from 15.14 ppm in

21 to 14.82 ppm in 24, and a distinct O−H singlet is present at

8.42 ppm. In the ³¹P{¹H} NMR spectrum, the singlet at 5.6

ppm in 21 shifts to 7.6 ppm in 24. It should be noted that BEt₃

is still present in these samples, likely interacting with the

formyl and inﬂuencing the observed chemical shifts. In the IR

spectrum, an apparent CO formyl stretch for 24 is located at

1595 cm⁻¹, though other resonances in this region complicate

this assignment. Complex 24 displays poor stability, undergoing

nonspeciﬁc decomposition over the course of 1 h to a complex

mixture of products.

pK_a’s of Formyl Complexes. Using a method analogous to

that described above for the pentacarbonyl complexes, the pK_a’s

of complexes 22−24 were determined in acetonitrile. For

complex 22, the pK_awas determined to be 14.4(1), by

equilibration of free-base formyl 17 with 2,6-lutidinium.

Consistent with this observation, addition of 4-dimethylami-

nopyridinium (DMAPH⁺, pK_a= 17.95)³¹to neutral formyl

complex 17 resulted in no evidence for proton transfer. The

pK_aof 23 was determined by equilibration of 19 with

DMAPH⁺, with a value of 16.6(3) obtained. For 24, the pK_a

measurement was complicated by varying interactions of 21

and 24 with triethylborane, which causes changes in the

equilibrium NMR shifts and interferes with the determination

of the equilibrium concentrations. Nevertheless, it was observed

that treatment of 21 with DMAPH⁺led to incomplete

deprotonation of the acid, and by measuring the equilibrium

populations of DMAP and DMAPH⁺an estimated pK_aof 17.7

was obtained for complex 24. Consistent with this estimate,

treatment of 21 with 2-(2′-tolyl)-1,1,3,3-tetramethylguanidi-

nium (pK_a= 20.5)³⁰left the NMR signals of 21 unperturbed,

indicating no protonation.

Complex 23, supported by the phenylguanidine-containing

ligand L4, proved to be the most stable and amenable to

characterization. Whereas the formyl CO stretching region of

the IR spectrum is obscured by ligand-based resonances,

making it diﬃcult to pinpoint the formyl stretch, multinuclear

NMR spectroscopy clearly reveals 23 as the major product,

obtained in 75% yield when a solution of 19 in thawing

acetonitrile is treated with 2,6-lutidinium tetraﬂuoroborate. The

³¹P{¹H} NMR spectrum shows a single resonance at 10.5 ppm,

distinct from that of free-base precursor 19 (9.3 ppm). Even

1

more telling are the changes to the H and ¹³C{¹H} NMR

spectra that occur upon protonation of 19 to give 23. In the

1

downﬁeld region of the H NMR spectrum, Figure S10, the

5536

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

ΔG°_H−of Rhenium Formyl Complexes. The thermody-

namic hydricities of rhenium formyl complexes 17 and 19 were

determined by equilibration with group 10 hydride complexes.

Reactions between the cationic carbonyl complexes 2 and 4 and

the cationic group 10 hydride complexes are quite sluggish

compared to the hydride transfer from the anionic complex

W−H, making it diﬃcult to judge if equilibrium has been

reached before onset of decomposition of formyl complexes 17

and 19. Fortunately, complexes 17 and 19 are the most stable

of the formyl complexes studied here, persisting for days in

solution. Treatment of 2 and 4 with [Pt(H)(dmpe)₂](PF₆)

(ΔG°_H−= 42.5 kcal/mol) leads to nearly quantitative hydride

transfer; over the course of 8 h (2) or 14 h (4) the

pentacarbonyl complexes are completely consumed, suggesting

that ΔG°_H−for the formyl complexes is at least 2.5 kcal/mol

greater. During these time frames decomposition of the formyl

is observed to a minor extent, with the rhenium-hydride

complex present at <20% relative to the formyl product in each

case. Treatment with [Pt(H)(depe)₂](PF₆) (ΔG°_H−= 44.2

kcal/mol) leads to much slower conversion, but low yields of

formyls 17 and 19 are observed, as well as some of the

rhenium-hydride decarbonylation products. In addition,

combination of 2 or 4 and [Ni^II(H)(dmpe)₂](PF₆) (ΔG°_H−

= 50.9 kcal/mol) shows no sign of hydride transfer over the

course of 24 h or longer, suggesting that ΔG°_H−for the formyl

complexes is at least 2.5 kcal/mol smaller than that of the

nickel-hydride reagent. Taken together, the results of the

reactions with platinum- and nickel-hydride reagents allow us

to determine that the ΔG°_H−for 17 and 19 are the same within

experimental error, with values of 45(2) kcal/mol. In the case

of anionic formyl 21, an accurate value for ΔG°_H−could not be

determined, since 21 can be generated only with borohydride

reagents, and interaction of the formyl with the borane

byproduct skews equilibrium measurements (also, ΔG°_H−for

these hydride reagents have not been experimentally

varying basic groups and spacers, constitute a platform for

investigating the key intermediates in the proposed approach to

homogeneous syngas conversion based on intramolecular base-

assisted heterolytic H₂activation. We have established

conditions for avoiding the tendency of these ligands to

coordinate in κ₂fashion (Scheme 3), with the resulting κ₁-P

complexes showing room-temperature stabilities of several

hours or more. The IR spectra of the complexes suggest

strongly electrophilic carbonyl ligands in 1−5, with ν_CO

̃

values

well above 2000 cm⁻¹in all cases, with little dependence on the

identity of the supporting ligand. Furthermore, these stretching

frequencies are nearly identical to analogous complexes

supported by simple tertiary phosphines,²⁹demonstrating that

the electron-rich basic moieties of L1−L5 do not inﬂuence the

metal center substantially when bound in the κ₁-P mode. In

zwitterion 13 the carbonyl stretching frequencies decrease

substantially (ca. 50 cm⁻¹), reﬂecting the decreased electro-

philicity in the neutral complex.

Complexes 1−5 can all be protonated to yield stable

dicationic complexes 8−12. The spectral properties of the

protonated complexes suggest little change in the electronic

properties of the metal center upon protonation of the pendant

base; both the ¹³C{¹H} NMR and IR spectral features

associated with the carbonyl ligands are aﬀected minimally,

showing that the tethered base can engage in acid/base

chemistry with little impact at the rhenium center. However,

the converse is not at all true: measurement of the pK_avalues of

complexes 6 and 8−12 shows that coordination to rhenium

results in a severe attenuation of the ligand’s basicity, exhibiting

decreases of 5−12 orders of magnitude for the cationic

carbonyl complexes compared to the free ligands (Table 1).

With this small set of complexes it is diﬃcult to discern trends

in the behavior; for example, the single methyl spacer in 3 and

the aryl spacer in 13 both result in a ca. 12-pK_a-unit

attenuation, but the same aryl spacer in 4 tempers the basicity

by only ∼7 pK_aunits. It seems likely that both the identity of

the dangling base and the nature of the spacer between the base

and the ligating group inﬂuence the pK_aof the bound ligand.

We envisioned the H₂-activation reaction outlined in Scheme

1 as proceeding through a similar mechanism to other known

heterolytic H₂-cleavage reactions,³⁷where in this case the CO

π* orbital, which is polarized toward carbon, is the acceptor

orbital. In spite of many examples in the literature of H₂

cleavage to make transition metal or main-group hydrides, there

are no examples of a metal-bound CO directly serving as the

hydride acceptor in a H₂-cleavage reaction, but it is not obvious

whether such a transformation is thermodynamically disfavored

or merely kinetically diﬃcult.

To obtain the thermodynamic parameters needed for the

analysis of Scheme 1, it was important that the metal formyl

complexes be free of any interaction with Lewis acidic species

in solution, as these (which have been shown to occur on

numerous occasions^12b,34,38) would perturb the measured

thermodynamics of the various elementary steps. For cationic

complexes 1−5, treatment with anionic complex W−H

furnished neutral formyl complexes 16−20 in moderate to

excellent yield, showing no interaction with the byproducts (the

18-electron complex cis-W⁰(P(OMe)₃)(NCMe)(CO)₄(W⁰)

and the weakly electrophilic cation PPN⁺). The spectral

characteristics attributed to the formyl and carbonyl groups in

16−20 are all quite similar to each other, once again showing

that the supporting phosphine ligand has a minimal impact on

the electronic properties of the rhenium center.

determined). However, the knowledge that W−H (ΔG°_H−

=

37(2) kcal/mol)³⁴is unable to transfer a hydride to complex 13

(vide supra) suggests that the ΔG°_H−for the anionic formyl

complex 21 is ≤34(2) kcal/mol.

Complete Thermodynamics of H₂Cleavage. By

combining the ΔG°_H−values and pK_avalues of the formyl

complexes, the overall free energy of H₂cleavage, ΔG°_H2, can

be determined as outlined in Scheme 1. Table 3 summarizes

a

Table 3. Thermodynamic Parameters for 22−24

complex

ΔG°_H−(kcal/mol)

pK_a

ΔG°_H2(kcal/mol)

22

23

24

45(2)

14.4(1)

16.6(3)

11(2)

8(2)

≤34(2)

∼17.7

≥17(2)

a

Values for complex 24 are approximate, for reasons given in the text.

the relevant thermodynamic parameters for the formation of

protonated formyl complexes 22−24. As seen from the values

in Table 3, H₂cleavage is unfavorable for all three complexes at

standard conditions (1 atm H₂, 25 °C). The small diﬀerence

between complexes 22 and 23 is due to the diﬀerence in pK_a

between the two complexes, whereas the much larger ΔG°_H2

value for complex 24 is a result of its much smaller ΔG°_H−

.

DISCUSSION

■

The cationic rhenium pentacarbonyl complexes 1−5 and 13,

containing six diﬀerent tethered-base phosphine ligands with

5537

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

Attempts to protonate neutral formyl complexes 16−20 with

weak acids gave mixed results. In some cases, addition of acid

resulted either in rapid H₂generation, with no evidence for the

desired product, or in complex, intractable product mixtures. In

three cases, though, we were able to prepare protonated formyl

complexeswhich would be the products of the proposed

heterolytic H₂cleavageby sequential addition of hydride and

proton equivalents to the pentacarbonyl complexes. Amine-

tethered complex 17 and anionic phenoxide-decorated complex

21 could both be protonated to give observable protonated

formyl complexes 22 and 24, respectively, though a

combination of low yield (22) and poor stability (24) hindered

complete spectroscopic analysis of these complexes. However,

protonation of formyl complex 19, supported by the novel

guanidine-based ligand L4, gave protonated formyl complex 23,

which formed in good yields and was suﬃciently stable for

spectroscopic characterization. In particular, the ¹³C{¹H} NMR

spectrum of 23 is suggestive of intramolecular hydrogen

bonding between the guanidinium moiety and the oxygen of

the formyl: the resonance attributed to the formyl carbon shifts

downﬁeld from 260.0 ppm (19) to 269.4 ppm (23), which is

indicative of enhanced carbene character in the formyl

group.^12b,15aFigure 4 illustrates two resonance structures of

of L1−L5, feature neutral nitrogenous bases whose pK_a’s should

have been high enough to promote heterolytic H₂activation,

given known ΔG°_H−values for similar formyl complexes.

However, coordination to Re reduced their basicities by 5−12

orders of magnitude, enough to take them out of the range

where H₂cleavage was predicted to be favorable. This

motivated pursuit of the phenoxide-based zwitterionic complex

13, which was expected toand doesshow enhanced basicity

compared to the cationic complexes, by several pK_aunits

(Table 1). Unfortunately, that change is accompanied by a

substantial decrease in carbonyl electrophilicity, such that the

ΔG°_H−is much smaller in magnitude, resulting in overall

thermodynamics of H₂cleavage by 13 to form complex 24 that

are the most unfavorable of all. That was a somewhat

surprising, and quite disappointing result: ΔG°_H−values for

the two formyl complexes (22 and 23) for which good

numbers could be measured are the same, 45(2) kcal/mol. The

coincidence of these two values is consistent with the

spectroscopic observations that suggested the properties of

the rhenium-bound carbonyls do not depend signiﬁcantly on

the identity of the nitrogen-base tethered ligand. In contrast,

switching to a phenoxide base does have a major unfavorable

eﬀect on the hydride-accepting properties of the carbonyls,

presumably a consequence of the net change in overall charge.

Clearlyand not so surprisinglytrying to improve thermo-

dynamics by redesigning the system can result in trade-oﬀs.

The simplest inference from these ﬁndings is that a ligand

featuring a much strongerbut still neutralappended base

could improve the thermodynamics and possibly lead to a

successful demonstration of the targeted heterolytic H₂

activation. However, the path forward is not so straightforward;

while (unfunctionalized) bases meeting these criteria are

available,³¹they do not appear readily amenable to attachment

to phosphines or other ligands. Also, there will probably be

additional factors governing the stability of the desired product,

particularly the strength of intramolecular hydrogen bonding

between the protonated base and the oxygen of the formyl,

which may well depend on speciﬁc structure. For example, the

pentacarbonyl complexes of ligands L2 and L4 have nearly

identical acid/base properties (Table 1), whereas the

corresponding formyl complexes exhibit signiﬁcantly diﬀerent

pK_avalues, 14.4(1) and 16.6(3), respectively, suggesting that

the guanidinium unit in 23 may be better positioned to

hydrogen bond with the formyl than is the trialkylammonium

group in 22. Another means of improving the overall

thermodynamics of H₂cleavage is to increase the hydricity

(ΔG°_H−) of the formyl, which is 45(2) kcal/mol for 17 and 19.

On the basis of previously tabulated formyl hydricities, the

Figure 4. Contributing resonance structures for protonated formyl

complex 23.

23, and the sizable shift of the formyl ¹³C{¹H} resonance

relative to 19 indicates that hydroxycarbene structure B is an

important contributor. Further support for intramolecular

1

hydrogen bonding comes from the H NMR spectra, where

the sharp signals for the formyl C−H and the guanidinium N−

H protons suggest rigidity, and the substantial broadening of

the N(CH₃)₂resonance in 23 relative to the sharp signal

observed in 19 is reﬂective of a more hindered interchange

between the two dimethylamino groups in the former.

Having established syntheses and spectral properties of the

protonated formyl species, setting up equilibria utilizing weaker

hydride reagents and acids allowed for the determination of the

thermodynamic parameters ΔG°_H−and pK_afor the formyl

complexes. The combination of these two parameters (Table 3)

shows that, in all cases, the cleavage of H₂is thermodynamically

prohibited at the pressures of H₂attainable in a laboratory

setting. Consistent with those ﬁndings, complexes 1−5 and 13

are all unreactive toward H₂, up to 3 atm at room temperature

in acetonitrile, while the putative products, the protonated

base−formyl complexes (22−24), which could be synthesized

by the alternate stepwise route in several instances (Schemes 8

and 10), all decompose (albeit rather slowly) by loss of H₂.

The last observation provides a small piece of encourage-

ment: since loss of H₂is the microscopic reverse of the target

transformation, we know at least that there is no insurmount-

able kinetic barrier. Might it be possible to use the lessons

learned here to design a system in which the thermodynamics

become favorable? Our initial choices for study, the complexes

analogous manganese(I) pentacarbonyl complexes (ΔG°_H−

≈

50 kcal/mol) and complexes of the type [Re^I(Cp)-

(CO)₂(NO)]⁺(Cp = cyclopentadienyl, ΔG°_H−≈ 55 kcal/

mol) are better hydride acceptors than the rhenium complexes

described here.¹⁹Manganese(I) pentacarbonyl complexes with

base-appended phosphines, in particular L4, are currently being

studied, and we are also looking to develop synthetic strategies

for tethering a guanidine or other strong base onto rhenium(I)-

cyclopentadienyl complexes. In spite of the thermodynamic

challenges outlined herein, we continue to explore this

approach for heterolytic H₂cleavage, as well as other

possibilities for using secondary coordination sphere inter-

actions of ligand-appended Brønsted acids and bases to

promote transformations of relevance to catalytic syngas

conversion.

5538

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

mixture allowed to warm to room temperature and stirred overnight,

during which the color darkened to a deep red-brown color. Addition

of 2 mL of methanol caused the color to fade to yellow. The volatiles

were removed in vacuo, producing a yellow-orange residue. The

remaining workup was performed in ambient atmosphere. The crude

product was dissolved in 50 mL of CH₂Cl₂and washed with 50 mL of

water. The aqueous phase was extracted with an additional 50 mL of

CH₂Cl₂, and the combined organic phases were dried over MgSO₄and

concentrated by rotary evaporation to leave an orange oil. The product

was dissolved in 25 mL of ethanol and chilled to −20 °C for 1 h,

separating some colorless crystals. The walls of the ﬂask were scraped

with a spatula, and with further cooling additional product

precipitated. The white solid was collected by ﬁltration, washed with

25 mL of ethanol, and dried in vacuo. Yield: 3.09 g (52.2%). ¹H NMR

(300 MHz, C₆D₆) δ: 7.57 (m, 4H, ArH), 7.02−7.22 (m, 8H, ArH),

6.79 (m, 1H, ArH), 6.59 (ddd, J = 7.9, 4.7, 1.2 Hz, 1H, ArH), 2.29 (s,

12H, CH₃). ¹³C{¹H} NMR (126 MHz, C₆D₆) δ: 157.4 (d, J_CP= 1.3

Hz, CN), 155.2 (d, J_CP= 19.3 Hz, Ar), 139.4 (d, J_CP= 12.8 Hz, Ar),

134.7 (s, J_CP= 20.0 Hz, Ar), 133.5 (d, J_CP= 1.6 Hz, Ar), 129.7 (d, J_CP

= 7.2 Hz, Ar), 129.4 (s, Ar), 128.4 (d, J_CP= 6.8 Hz, Ar), 128.2 (s, Ar),

120.6 (d, J_CP= 2.4 Hz, Ar), 120.5 (d, J_CP= 0.9 Hz, Ar), 39.4 (s,

NCH₃). ³¹P{¹H} NMR (121 MHz, C₆D₆) δ: −11.8. Anal. Calcd for

C₂₃H₂₆N₃P: C, 73.58; H, 6.98; N, 11.19. Found: C, 73.89; H, 7.11; N,

11.12.

P r e p a r a t i o n o f 6 - ( D i p h e n y l p h o s p h i n o ) - 1 , 8 -

diazabicyclo[5.4.0]undec-7-ene (L5). A solution of DBU (3.50 g,

23.0 mmol) in 85 mL of THF was cooled in a dry ice/acetone bath. A

pentane solution of tert-butyllithium (1.63 M, 15 mL, 24 mmol, 1.05

equiv) was added at a dropwise rate via cannula, giving a yellow

solution, from which a yellow solid precipitated during 1 h of stirring.

At this time the cold bath was removed, and by allowing the solution

to warm to room temperature for 1 h the solid dissolved and most of

the color faded. The nearly colorless solution was rechilled in dry ice/

acetone, and a solution of Ph₂PCl (5.07 g, 23.0 mmol, 1.00 equiv)

dissolved in 20 mL of THF was added, giving an orange solution. After

20 min, the cold bath was removed, and the solution stirred at room

temperature overnight. Addition of 2 mL of methanol completely

bleached the yellow color, and concentrating in vacuo left a pale yellow

residue. The residue was extracted into 50 mL of CH₂Cl₂and ﬁltered

through Celite to remove a white solid. The solvent was removed in

vacuo to give a pale sticky solid, which was dissolved in 100 mL of 1:1

CH₂Cl₂/hexane. The solution was concentrated to one-half the

original volume, and a white solid collected by ﬁltration. A second crop

was isolated from the supernatant. The combined crude products were

suspended in 50 mL of boiling EtOH, which was made basic with 6

mL of 25% KOMe/MeOH. The mixture was cooled to −20 °C

overnight, ﬁltered, and concentrated by rotary evaporation. The yellow

oil was triturated with 50 mL of hexane, releasing a white solid, which

was collected by ﬁltration and dried in vacuo. Yield: 3.18 g (41.1%).

The spectral properties match those previously reported for this

compound, prepared by an alternate route.²³

EXPERIMENTAL SECTION

■

Materials. All reactions were executed in a glovebox ﬁlled with

either N₂or argon or on a Schlenk line under argon. Solvents were

dried by passing through an alumina column using the method of

Grubbs,³⁹and deuterated NMR solvents were passed through a short

column of alumina prior to use. CD₃CN was further dried by storing

over molecular sieves. Hydrogen was passed through columns of

molecular sieves and manganese oxide prior to delivery through a

high-vacuum manifold. The ligands 1,8-dimethylamino-2-diphenyl-

phosphinonapthalene (L1),²⁰2-diphenylphosphino-N,N-dimethylben-

zylamine (L2),²¹and N-(diphenylphosphinomethyl)morpholine

(L3)²²were prepared as described in the literature. The tetraﬂuor-

oborate salt of L1 ((L1H)(BF₄)) was also available by a previously

described synthesis.²⁰The sodium salt of L6 was prepared as

previously outlined for related ligands⁴⁰and was used in the pK_a

determination of L6. The ligand precursors 2-(2′-bromophenyl)-

1,1,3,3-tetramethylguanidine²⁷and methoxymethyl ether-protected 2-

bromo-6-tert-butyl-4-methylphenol⁴¹were also obtained from liter-

ature routes. The rhenium(I) starting materials Re^I(CO)₅(FBF₃)²⁸

and Re^I(CO)₅(OTf)⁴²(OTf = triﬂuoromethanesulfonate) were

likewise prepared by published methods. Commercially available

bases 2,6-lutidine, triethylamine, 2-chloroaniline, aniline, pyridine, and

1,8-diazabicyclo[5.4.0]undec-7-ene were puriﬁed by standard meth-

ods,⁴³whereas 4-dimethylaminopyridine (DMAP) was used as

received. The base 2-(2′-tolyl)-1,1,3,3-tetramethylguanidine was

prepared by a modiﬁed literature procedure,²⁷substituting toluene

for benzene as the reaction solvent and heating to reﬂux. The

tetraﬂuoroborate salts of 2,6-lutidine, aniline, and pyridine, and 2-(2′-

tolyl)-1,1,3,3-tetramethylguanidine were obtained by treating a

solution of the base in Et₂O with one equivalent of HBF₄·Et₂O; the

salts precipitated as white solids, which were deemed pure by ¹H

NMR. The tetraﬂuoroborate salt of DMAP was prepared similarly,

using MeCN as the solvent, and precipitating the product with Et₂O.

Lithium triethylborohydride (Super Hydride) was obtained from

Aldrich as a 1.0 M THF solution. The hydride reagents (PPN)[cis-

W⁰(H)(P(OMe)₃)(CO)₄]³³(W−H, PPN = bis(triphenylphosphine)-

iminium), [Pt^II(H)(diphosphine)₂][PF₆]⁴⁴(diphosphine = bis-

(dimethylphosphino)ethane (dmpe), bis(diethylphosphino)ethane

(depe)), and [Ni^II(H)(dmpe)₂][PF₆]⁴⁴were prepared as previously

described.

1

Physical Methods. H and ³¹P{¹H} NMR spectra were recorded

1

on a Varian Mercury 300 spectrometer, operating at 300 MHz for H

acquisition and 121.5 MHz for ³¹P acquisition. ¹³C{¹H} NMR were

recorded on a Varian Inova 500 spectrometer, operating at 125.7

MHz. ¹H and ¹³C{¹H} spectra were referenced to solvent resonances,

whereas ³¹P NMR spectra were referenced to an external standard of

85% D₃PO₄. All NMR spectra were acquired at room temperature. For

¹³C{¹H} NMR spectra of formyl complexes, the default decoupling

frequency resulted in incomplete decoupling for the formyl resonance.

Hence for all formyl complexes additional scans were recorded with

1

the H decoupling frequency centered at 15 ppm; the insets of the

Preparation of 2-tert-Butyl-6-diphenylphosphino-4-methyl-

phenol (L6). A solution of MOM-protected 2-bromo-6-tert-butyl-4-

methylphenol (8.50 g, 29.6 mmol) in 100 mL of THF was chilled in a

dry ice/acetone bath. A pentane solution of tert-butyllithium (40 mL,

65 mmol, 2.2 equiv) was added via cannula, giving a yellow mixture,

which was stirred for 45 min while kept cold. After allowing the

mixture to brieﬂy warm to room temperature, it was rechilled, and a

solution of Ph₂PCl (6.53 g, 29.6 mmol, 1.00 equiv) in 8 mL of THF

was added. The color initially lightened, and while stirring overnight at

room temperature a deep red color developed. Addition of 2 mL of

methanol caused the color to fade to pale orange. The solution was

concentrated in vacuo, and the resulting residue dissolved in 100 mL

of CH₂Cl₂and washed with 100 mL of 1 M Na₂HPO₄. The organic

layer was dried over MgSO₄, ﬁltered, and concentrated to produce a

pale orange oil. Trituration with 25 mL of methanol released a white

solid, and after cooling to −20 °C for 30 min the MOM-protected

product was ﬁltered, washed with methanol, and dried in vacuo. Yield:

depicted ¹³C{¹H} NMR spectra in the Supporting Information display

the CO region of the spectrum recorded as such. IR spectra were

recorded on a Nicolet 6700 spectrometer. Samples were housed in a

solution cell with KBr windows and a 0.1 mm Teﬂon spacer. High-

resolution mass spectrometry was executed on a JEOL JMS-600H

high-resolution mass spectrometer operating in FAB+ mode with 3-

nitrobenzyl alcohol as the matrix. Elemental analyses were performed

by Midwest Microlab LLC.

Preparation of 2-(2′-Diphenylphosphinophenyl)-1,1,3,3-tet-

ramethylguanidine (L4). A Schlenk ﬂask was charged with 2-(2′-

bromophenyl)-1,1,3,3-tetramethylguanidine (4.26 g, 15.8 mmol)

dissolved in 50 mL of THF. The colorless solution was cooled in a

dry ice/acetone bath, after which a pentane solution of tert-

butyllithium (1.63 M, 21 mL, 34 mmol, 2.2 equiv) was added via

cannula, eﬀecting a color change to bright yellow. After stirring for 1 h

while being kept cold, Ph₂PCl (3.48 g, 15.8 mmol, 1.00 equiv), diluted

with 6 mL of THF, was added via cannula. The initially yellow solution

was stirred for 2 h, at which time the cold bath was removed and the

1

8.16 g (70.3%). H NMR (300 MHz, CDCl₃) δ: 7.22−7.36 (m, 10H,

ArH), 7.18 (d, 1H, ArH), 6.47 (ddquart., J = 4.37, 2.10, 0.66 Hz, 1H,

5539

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

ArH), 5.22 (d, 2H, CH₂), 3.51 (s, 3H, OCH₃), 2.15 (s, 3H, ArCH₃),

1.43 (s, 9H, C(CH₃)₃). ³¹P{¹H} NMR (121 MHz, C₆D₆) δ: −12.8. A

sample of the MOM-protected phosphinophenol (5.46 g, 13.9 mmol)

was deprotected as previously described.²⁵The isolated product L6

was spectroscopically identical to a previously characterized sample.²⁶

Preparation of (L2H)(BF₄). A sample of L2 (100 mg, 0.313

mmol) was dissolved in 4 mL of Et₂O. With stirring, HBF₄·Et₂O (46.1

μL, 0.344 mmol, 1.10 equiv) was added via syringe. A sticky white

solid precipitated immediately, and by scraping with a spatula and

stirring a white powder was obtained. The supernatant was decanted,

and the product dried in vacuo. Yield: 113 mg (89.0%). ¹H NMR (300

MHz, CD₃CN) δ: 9.04 (br, s, 1H, NH), 7.10−7.70 (m, 14H, ArH),

4.44 (d, J = 5.7 Hz, 2H, CH₂), 2.86 (d, J = 5.4 Hz, 6H, CH₃). ¹³C{¹H}

NMR⁴⁵(126 MHz, CD₃CN) δ: 136.2 (br, s, Ar), 135.2 (s, Ar), 134.9

(br, d, J_CP= 18.0 Hz, Ar), 133.0 (d, J_CP= 6.2 Hz, Ar), 131.9 (s, Ar),

130.4 (br, s, Ar), 60.2 (br, d, J_CP= 17.9 Hz, CH₂), 44.2 (s, CH₃).

³¹P{¹H} NMR (121 MHz, CD₃CN) δ: −16.4 (br). HRMS (FAB): m/

z calcd for C₂₁H₂₃NP [M]⁺320.1568, found 320.1552.

of L1 (193 mg, 0.484 mmol, 1.00 equiv), dissolved in 2 mL of CH₂Cl₂,

was added at a rate of one drop per second, initially giving an orange

solution. Over the course of 30 min with continuous stirring, the color

faded to yellow. The solvent was removed in vacuo to leave a yellow

residue, which was redissolved in 2 mL of CH₂Cl₂. Addition of 2 mL

of hexane induced precipitation of a yellow solid, and the mixture was

chilled in the freezer (−35 °C) for 3 h. The supernatant was decanted,

and the product dried in vacuo. Yield: 306 mg (77.9%). ¹H NMR (300

MHz, CD₂Cl₂) δ: 7.85 (dd, J = 8.8, 2.4 Hz, 1H, ArH), 7.49−7.77 (m,

13H, ArH), 7.38 (dd, J = 7.5, 1.3 Hz, 1H, ArH), 2.56 (s, 6H, CH₃),

2.20 (s, 6H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 179.4 (br, d,

²J_CP= 6.9 Hz, cis CO), 175.2 (br, d, ²J_CP= 39.7 Hz, trans CO), 154.7

(d, J_CP= 2.5 Hz, Ar), 153.8 (s, Ar), 140.2 (d, J_CP= 2.1 Hz, Ar), 133.9

(d, J_CP= 54.9 Hz, Ar), 132.6 (d, J_CP= 11.2 Hz, Ar), 132.0 (d, J_CP= 2.8

Hz, Ar), 129.8 (d, J_CP= 11.1 Hz, Ar), 129.4 (d, J_CP= 15.2 Hz, Ar),

129.2 (s, Ar), 128.7 (d, J_CP= 7.0 Hz, Ar), 127.6 (d, J_CP= 60.1 Hz, Ar),

127.5 (d, J_CP= 13.5 Hz, Ar), 125.1 (s, Ar), 118.9 (s, Ar), 47.1 (s, CH₃),

44.6 (s, CH₃). ³¹P{¹H} NMR (121 MHz, CD₂Cl₂) δ: 1.4. IR

Preparation of (L3H)(BF₄). Solid L3 (200 mg, 0.701 mmol) was

dissolved in 6 mL of Et₂O. To the stirred solution was added HBF₄·

Et₂O (98 μL, 0.73 mmol, 1.04 equiv), which caused a sticky white

solid to precipitate. After 10 min the supernatant was decanted, and

the product was triturated with toluene/Et₂O, which liberated a white

solid. After pipetting oﬀ the solvents, the product was dried in vacuo.

(CH₂Cl₂): ν_CO

̃

= 2155 (m), 2094 (w), 2047 (s) cm⁻¹. Anal. Calcd

for C₃₁H₂₇BF₄N₂O₅PRe: C, 45.88; H, 3.35; N, 3.45. Found: C, 45.64;

H, 3.49; N, 3.36.

Preparation of [Re^I(L2-κ₁-P)(CO)₅](OTf) (2). Re^I(CO)₅(OTf)

(100 mg, 0.210 mmol) and L2 (67 mg, 0.21 mmol, 1.0 equiv) were

dissolved in 5 mL of CH₂Cl₂in a 20 mL ﬂask sealed with a Teﬂon

plug. The solution was stirred for 1 week at room temperature, at

which time the crude ³¹P{¹H} NMR spectrum indicated ca. 90%

completion. The solution was ﬁltered to remove a small amount of a

brown impurity and then concentrated in vacuo. The residue was

redissolved in 2 mL of THF, and by adding 8 mL of Et₂O a white solid

precipitated, which was decanted and dried in vacuo. Yield: 116 mg of

2·THF (69.5%). ¹H NMR (300 MHz, CD₂Cl₂) δ: 7.50−7.75 (m, 14H,

ArH), 3.68 (m, THF), 2.98 (s, 2H, CH₂), 1.82 (m, THF), 1.68 (s, 6H,

1

Yield: 189 mg (72.1%). H NMR (300 MHz, CD₃CN) δ: 7.05−7.95

(m, 11H, ArH + NH), 4.00 (br, m, 4H, CH₂), 3.74 (m, 2H, CH₂), 3.58

(br, d, J = 12.6 Hz, 2H, CH₂), 3.18 (br, m, 2H, CH₂). ¹³C{¹H} NMR

(126 MHz, CD₃CN) δ: 134.2 (br, d, J_CP= 20.8 Hz, Ar), 131.3 (br, s,

Ar), 130.8 (br, s, Ar), 130.2 (br, d, J_CP= 18.0 Hz, Ar), 65.6 (s, CH₂),

59.4 (br, s, CH₂) 55.0 (br, s, CH₂). ³¹P{¹H} NMR (121 MHz,

CD₃CN) δ: −27.0 (br). HRMS (FAB): m/z calcd for C₁₇H₂₁ONP

[M]⁺286.1361, found 286.1351.

2

Preparation of (L4H)(BF₄). L4 (200 mg, 0.533 mmol) was

dissolved in 1 mL of CH₂Cl₂. Via syringe, HBF₄·Et₂O (75 μL, 0.56

mmol, 1.05 equiv) was added, leaving a colorless solution. The

solution was stirred at room temperature for 20 min, during which

time some white solid formed. Addition of 10 mL of Et₂O further

precipitated the product, which was separated from the supernatant by

decantation and dried in vacuo. Yield: 240 mg (97.2%). ¹H NMR (300

MHz, CD₃CN) δ: 7.59 (br, s, 1H, NH), 7.20−7.55 (m, 12H, ArH),

7.12 (dd, J = 8.0, 4.7 Hz, 1H, ArH), 7.04 (dd, J = 8.3, 4.1 Hz, 1H,

ArH), 2.79 (s, 12H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ:

160.2 (s, CN), 141.9 (d, J_CP= 22.8 Hz, Ar), 136.3 (s, Ar), 135.9 (d,

J_CP= 8.1 Hz, Ar), 134.6 (d, J_CP= 20.2 Hz, Ar), 132.2 (s, Ar), 131.8 (br,

s, Ar), 130.5 (s, Ar), 130.0 (d, J_CP= 7.3 Hz, Ar), 127.9 (s, Ar), 125.1 (s,

Ar), 40.8 (s, CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: −18.9.

Anal. Calcd for C₂₃H₂₇BF₄N₃P: C, 59.63; H, 5.87; N, 9.07. Found: C,

59.40; H, 5.73; N, 9.04.

CH₃). ¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 179.1 (br, d, J_CP= 6.6

2

Hz, cis CO), 175.3 (br, d, J_CP= 42.1 Hz, trans CO), 142.9 (s, Ar),

135.0 (d, J_CP= 14.6 Hz, Ar), 133.7 (d, J_CP= 8.6 Hz, Ar), 133.1 (d, J_CP

= 11.6 Hz, Ar), 132.8 (s, Ar), 132.6 (d, J_CP= 2.8 Hz, Ar), 132.4 (s, Ar),

132.0 (s, Ar), 130.2 (d, J_CP= 11.2 Hz, Ar), 128.8 (d, J_CP= 12. Hz, Ar),

1

121.4 (quart., J_CF= 322.6 Hz, CF₃), 68.1 (s, THF), 63.1 (s, CH₂),

44.8 (s, CH₃), 25.9 (s, THF). ³¹P{¹H} NMR (121 MHz, CD₂Cl₂) δ:

5.3. IR (CH₂Cl₂): ν_CO

̃

= 2156 (m), 2095 (w), 2048 (s) cm⁻¹. Anal.

Calcd for C₃₁H₃₀F₃NO₉PReS (2·THF): C, 42.96; H, 3.49; N, 1.62.

Found: C, 42.66; H, 3.46; N, 1.49.

Preparation of [Re^I(L3-κ₁-P)(CO)₅](BF₄) (3). A scintillation vial

was charged with Re^I(CO)₅(FBF₃) (200 mg, 0.484 mmol) suspended

in 2 mL of CH₂Cl₂. A solution of L3 (138 mg, 0.484 mmol, 1.00

equiv) in 3 mL of CH₂Cl₂was added. A colorless solution resulted,

which was stirred for 30 min at room temperature. Addition of 15 mL

of Et₂O separated a white residue, from which the supernatant was

decanted. The residue was dissolved in 2 mL of THF, and with

addition of 6 mL of Et₂O a sticky solid separated. The solid was

suspended in 4 mL of Et₂O and stirred overnight, producing a freely

ﬂowing white powder, which was dried in vacuo. Yield: 277 mg

(82.0%). ¹H NMR (300 MHz, CD₂Cl₂) δ: 7.30−7.90 (m, 10H, ArH),

Preparation of (L5H)(BF₄). A scintillation vial was charged with

L5 (500 mg, 1.49 mmol) dissolved in 2 mL of CH₂Cl₂. A sample of

HBF₄·Et₂O (210 μL, 1.56 mmol, 1.05 equiv) was added. The colorless

solution was stirred for 10 min, at which point ca. 15 mL of Et₂O was

added to precipitate a ﬂuﬀy, white solid. The supernatant was

decanted, and the product washed with Et₂O and dried in vacuo. Yield:

623 mg (98.9%). ¹H NMR (300 MHz, CD₃CN) δ: 7.65−7.75 (m, 2H,

ArH), 7.50−7.56 (m, 3H, ArH), 7.36−7.46 (m, 5H, ArH), 7.15 (br, s,

2

3.84 (d, J_HP= 3.9 Hz, 4H, CH₂), 3.54 (m, 4H, CH₂), 2.35 (m, 4H,

CH₂). ¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 178.6 (br, s, cis CO),

175.7 (br, s, trans CO), 133.1 (d, J_CP= 2.5 Hz, Ar), 132.4 (d, J_CP

=

2

1H, NH), 4.70 (dddd, J_HH= 15.2, 11.1, 2.2 Hz, J_HP= 8.8 Hz, 1H,

10.1 Hz, Ar), 130.9 (d, J_CP= 50.3 Hz, Ar), 130.4 (d, J_CP= 10.4 Hz,

Ar), 66.5 (s, CH₂), 60.3 (d, J_CP= 47.5 Hz, CH₂), 55.9 (d, J_CP= 7.8 Hz,

CH₂). ³¹P{¹H} NMR (121 MHz, CD₂Cl₂) δ: 3.8 (br). IR (CH₂Cl₂):

CH), 3.63 (ddd, J_HH= 6.2, 4.1 Hz, J_HP= 2.2 Hz, 1H, CH₂), 3.32−3.56

(m, 3H, CH₂), 3.00−3.12 (m, 1H, CH₂), 2.61−2.73 (m, 1H, CH₂),

1.50−2.10 (m, 8H, CH₂). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ:

165.3 (d, ²J_CP= 7.4 Hz, CN), 135.2 (d, J_CP= 15.5 Hz, Ar), 134.6 (d,

J_CP= 19.6 Hz, Ar), 134.4 (d, J_CP= 18.3 Hz, Ar), 132.1 (d, J_CP= 14.5

Hz, Ar), 131.5 (s, Ar), 131.2 (s, Ar), 130.3 (d, J_CP= 8.2 Hz, Ar), 129.8

(d, J_CP= 7.7 Hz, Ar), 54.7 (d, J_CP= 16.0 Hz, CH₂), 50.5 (s, CH₂), 45.2

(d, J_CP= 22.3 Hz, CH), 39.2 (s, CH₂), 27.2 (d, J_CP= 13.2 Hz, CH₂),

26.8 (s, CH₂), 24.4 (d, J_CP= 7.2 Hz, CH₂), 19.4 (s, CH₂). ³¹P{¹H}

NMR (121 MHz, CD₃CN) δ: −17.5. Anal. Calcd for C₂₁H₂₆BF₄N₂P:

C, 59.46; H, 6.18; N, 6.60. Found: C, 59.38; H, 6.21; N, 6.58.

Preparation of [Re^I(L1-κ₁-P)(CO)₅](BF₄) (1). Re^I(CO)₅(FBF₃)

(200 mg, 0.484 mmol) was suspended in 4 mL of CH₂Cl₂. A solution

ν_CO

̃

= 2157 (m), 2099 (w), 2047 (s) cm⁻¹. Anal. Calcd for

C₂₂H₂₀BF₄NO₆PRe: C, 37.84; H, 2.89; N, 2.01. Found: C, 38.06; H,

3.01; N, 1.96.

Preparation of [Re^I(L4-κ₁-P)(CO)₅](BF₄) (4). To a suspension of

Re^I(CO)₅(FBF₃) (200 mg, 0.484 mmol) in 2 mL of CH₂Cl₂was

added a solution of L4 (182 mg, 0.485 mmol, 1.00 equiv) in 3 mL of

CH₂Cl₂. The solid was drawn into solution, with a pale yellow color

resulting. After stirring for 30 min at room temperature, 15 mL of

Et₂O was gradually added, which separated an eggshell-colored solid.

The supernatant was decanted, and the product washed with 4 mL of

Et₂O and dried in vacuo. Yield: 356 mg (93.2%). ¹H NMR (300 MHz,

5540

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

CD₃CN) δ: 7.35−7.65 (m, 12H, ArH), 6.95 (dddd, J = 8.1, 7.1, 2.1,

1.1 Hz, 1H, ArH), 6.41 (ddd, J = 8.2, 5.8, 1.2 Hz, 1H, ArH), 2.37 (s,

12H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 181.2 (br, s, cis

CO), 178.2 (br, d, ²J_CP= 39.1 Hz, trans CO), 163.6 (s, CN), 156.8

(d, J_CP= 8.6 Hz, Ar), 134.4 (s, Ar), 134.3 (d, J_CP= 10.7 Hz, Ar), 133.7

(d, J_CP= 6.0 Hz, Ar), 132.7 (d, J_CP= 2.5 Hz, Ar), 132.6 (d, J_CP= 53.0

Hz, Ar), 130.1 (d, J_CP= 10.7 Hz, Ar), 122.9 (d, J_CP= 6.9 Hz, Ar),

119.8 (d, J_CP= 9.8 Hz, Ar), 115.9 (d, J_CP= 57.8 Hz, Ar), 39.7 (s, CH₃).

CH₂Cl₂. To the yellow solution was added HBF₄·Et₂O (18 μL, 0.13

mmol, 1.1 equiv), which immediately resulted in the formation of a

white solid. The mixture was stirred for 10 min and then allowed to

settle. The supernatant was decanted, and the product washed with 2

mL of CH₂Cl₂and 2 × 2 mL of Et₂O before drying in vacuo. Yield:

101 mg (91.0%). ¹H NMR (300 MHz, CD₃CN) δ: 17.08 (br, s, NH),

8.24 (dd, J = 3.6, 1.3 Hz, 1H, ArH), 8.21 (dd, J = 2.9, 1.3 Hz, 1H,

ArH), 8.08 (dd, J = 7.8, 1.3 Hz, 1H, ArH), 7.87−8.02 (m, 2H, ArH),

7.59−7.74 (m, 10H, ArH), 3.29 (d, J = 4.2 Hz, 6H, CH₃), 2.50 (d, J =

0.6 Hz, 6H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 180.2 (d,

²J_CP= 8.2 Hz, cis CO), 176.2 (d, ²J_CP= 42.8 Hz, trans CO), 149.3 (s,

Ar), 143.1 (s, Ar), 138.4 (s, Ar), 135.8 (d, J_CP= 11.8 Hz, Ar), 134.7 (d,

J_CP= 18.1 Hz, Ar), 134.2 (d, J_CP= 2.6 Hz, Ar), 133.2 (d, J_CP= 50.9 Hz,

Ar) 132.5 (d, J_CP= 37.9 Hz, Ar), 132.0 (s, Ar), 130.9 (d, J_CP= 11.2 Hz,

Ar), 130.5 (s, Ar), 130.0 (d, J_CP= 13.7 Hz, Ar), 124.1 (s, Ar), 123.4 (d,

J_CP= 8.2 Hz, Ar), 48.2 (s, CH₃), 44.8 (s, CH₃). ³¹P{¹H} NMR (121

³¹P{¹H} NMR (121 MHz, CD₃CN) δ: −7.2. IR (CH₂Cl₂): ν_CO

̃

=

2150 (m), 2088 (w), 2047 (s) cm⁻¹

.

Anal. Calcd for

C₂₈H₂₆BF₄N₃O₅PRe: C, 42.65; H, 3.32; N, 5.33. Found: C, 42.62;

H, 3.41; N, 5.20.

In Situ Preparation of [Re^I(L5-κ₁-P)(CO)₅](BF₄) (5). Because of

its instability with regard to forming the κ₂-P,N complex (7) (see

below), complex 5 was not isolated. We identiﬁed and characterized it

in solution either from the crude reaction between Re^I(CO)₅(FBF₃)

and 1 equivalent of L5 (ca. 80% yield at early time points) or by

deprotonation of complex 12 (see below) with triethylamine

MHz, CD₂Cl₂) δ: 14.9. IR (CH₃CN): ν_CO

̃

= 2160 (m), 2054 (s)

cm⁻¹. Anal. Calcd for C₃₁H₂₈B₂F₈N₂O₅PRe: C, 41.40; H, 3.14; N, 3.11.

Found: C, 40.89; H, 3.17; N, 2.95.

(quantitative yield). ¹H NMR (300 MHz, CD₃CN) δ: 7.65−7.80

2

(m, 2H, ArH), 7.35−7.60 (m, 8H, ArH), 4.10 (app. t, J_HP= J_HH

=

Preparation of [Re^I(L2H-κ₁-P)(CO)₅](BF₄)(OTf) (9). Solid 2·THF

(56 mg, 0.065 mmol) was dissolved in 2 mL of CH₂Cl₂. Via syringe,

HBF₄·Et₂O (10 μL, 0.075 mmol, 1.1 equiv) was added with stirring,

causing a white solid to precipitate. After stirring brieﬂy, the mixture

was diluted with 4 mL of Et₂O and decanted, and the product was

dried in vacuo. Yield: 56 mg (98%). ¹H NMR (300 MHz, CD₃CN) δ:

8.24 (m, 1H, ArH), 7.95 (m, 2H, ArH), 7.63−7.76 (m, 11H, ArH),

7.48 (br, s, 1H, NH), 3.68 (d, J = 6.3 Hz, 2H, CH₂), 2.08 (d, J = 5.1

10.0 Hz, 1H, CH) 3.84 (dd, J = 15.3, 10.4 Hz, 1H, CH₂), 2.95−3.32

(m, 5H, CH₂), 1.35−2.07 (m, 8H, CH₂). ³¹P{¹H} NMR (121 MHz,

CD₃CN) δ: 3.6 (br). IR (CH₂Cl₂) ν_CO

̃

= 2151 (m), 2091 (w),

2041(s) cm⁻¹.

Preparation of [Re^I(L6-κ₁-P)(CO)₅](BF₄) (6). Solid

Re^I(CO)₅(FBF₃) (200 mg, 0.484 mmol) was suspended in 4 mL of

CH₂Cl₂. A solution of L6 (169 mg, 0.485 mmol, 1.00 equiv) dissolved

in 4 mL of CH₂Cl₂was added, producing a colorless solution. After

stirring for 90 min, the solvent was removed in vacuo. The colorless

residue was redissolved in 2 mL of CH₂Cl₂, and with stirring 6 mL of

Et₂O was added. The white solid that precipitated was separated by

decanting the mother liquor and then washed with Et₂O and dried in

Hz, 6H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 180.0 (d, ²J_CP

8.3 Hz, cis CO), 176.7 (d, J_CP= 42.2 Hz, trans CO), 140.7 (d, J_CP

21.7 Hz, Ar), 135.8 (d, J_cp= 2.6 Hz, Ar), 134.7 (s, Ar), 133.8 (d, J_CP

=

2

2.0 Hz, Ar), 132.6 (d, J_CP= 11.0 Hz, Ar), 132.1 (d, J_CP= 6.3 Hz, Ar),

131.8 (d, J_CP= 34.8 Hz, Ar), 131.5 (s, Ar), 131.4 (s, Ar), 128.0 (d, J_CP

= 49.3 Hz, Ar), 121.9 (quart., ¹J_CF= 320.1 Hz, CF₃), 59.0 (d, J_CP= 5.0

Hz, CH₂), 43.7 (s, CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 2.6.

1

vacuo. Yield: 347 mg of 6·CH₂Cl₂(84.6%). H NMR (300 MHz,

CD₃CN) δ: 7.51−7.65 (m, 6H, ArH), 7.35−7.50 (m, 5H, ArH), 6.82

(d, J = 12.0 Hz, 1H, ArH), 6.58 (s, 1H, OH), 5.45 (s, CH₂Cl₂), 2.22 (s,

3H, ArCH₃), 1.37 (s, 9H, C(CH₃)₃). ¹³C{¹H} NMR (126 MHz,

CD₃CN) δ: 180.8 (d, ²J_CP= 6.2 Hz, cis CO), 177.6 (br, d, ²J_CP= 42.1

IR (CH₃CN): ν_CO

̃

= 2160 (m), 2053 (s) cm⁻¹. Anal. Calcd for

C₂₇H₂₃BF₇NO₈PReS: C, 36.75; H, 2.63; N, 1.59. Found: C, 36.76; H,

2.74; N, 1.56.

2

Hz, trans CO), 154.0 (d, J_CP= 5.3 Hz, Ar), 139.6 (d, J_CP= 5.3 Hz,

Preparation of [Re^I(L3H-κ₁-P)(CO)₅](BF₄)₂(10). Complex 3

(100 mg, 0.143 mmol) was dissolved in 2 mL of CH₂Cl₂. HBF₄·Et₂O

(20 μL, 0.15 mmol, 1.05 equiv) was added via syringe, which separated

a colorless residue from the solution. After 10 min, the mixture was

diluted with 8 mL of Et₂O, and by scraping with a spatula a white solid

was produced. The supernatant was decanted, and the product dried in

Ar), 133.9 (d, J_CP= 11.3 Hz, Ar), 133.9 (s, Ar), 133.0 (d, J_CP= 2.5 Hz,

Ar), 132.6 (d, J_CP= 10.2 Hz, Ar), 131.9 (d, J_CP= 5.4 Hz, Ar) 131.4 (d,

J_CP= 54.3 Hz, Ar), 130.5 (d, J_CP= 11.0 Hz, Ar), 119.0 (d, J_CP= 56.1

Hz, Ar), 34.9 (s, C(CH₃)₃), 30.4 (s, C(CH₃)₃), 21.0 (s, ArCH₃).

³¹P{¹H} NMR (121 MHz, CD₂Cl₂) δ: −4.6. IR (CH₂Cl₂): ν_CO

̃

=

1

2157 (m), 2102 (w), 2048 (s) cm⁻¹

. Anal. Calcd for

vacuo. Yield: 99 mg (88%). H NMR (300 MHz, CD₃CN) δ: 7.55−

8.00 (m, 10H, ArH), 7.23 (br, s, 1H, NH), 4.63 (d, J = 5.1 Hz, 2H,

CH₂), 3.83 (d, J = 13.5 Hz, 2H, CH₂), 3.58 (t, J = 12.0 Hz, 2H, CH₂),

3.30 (m, 2H, CH₂), 3.09 (d, J = 12.6 Hz, 2H, CH₂). ¹³C{¹H} NMR

(126 MHz, CD₃CN) δ: 178.7 (d, ²J_CP= 6.3 Hz, cis CO), 176.2 (d, ²J_CP

= 42.5 Hz, trans CO), 135.0 (s, Ar), 133.5 (d, J_cp= 11.5 Hz, Ar), 131.7

(d, J_CP= 11.6 Hz, Ar), 125.2 (d, J_CP= 52.5 Hz, Ar), 63.8 (s, CH₂), 56.9

(s, CH₂), 56.8 (s, CH₂, overlap with previous peak). ³¹P{¹H} NMR

C₂₉H₂₇BCl₂F₄O₆PRe (6·CH₂Cl₂): C, 41.15; H, 3.22; N, 0.00.

Found: C, 40.99; H, 3.16; N, 0.

Preparation of [Re^I(L5-κ₂-P,N)(CO)₄](BF₄) (7). A sample of crude

5 (357 mg), prepared by reaction of Re^I(CO)₅(FBF₃) with L5, was

dissolved in 6 mL of CH₂Cl₂and held at room temperature for ca. 60

h. The solution was concentrated in vacuo to produce a colorless

residue, which was redissolved in 2 mL of CH₂Cl₂. Addition of 8 mL

of Et₂O separated a white solid; the supernatant was removed and the

(121 MHz, CD₃CN) δ: −5.4. IR (CH₃CN): ν_CO

̃

= 2163 (m), 2157

(w), 2056 (s) cm⁻¹. Anal. Calcd for C₂₂H₂₁B₂F₈NO₆PRe: C, 33.61; H,

2.69; N, 1.78. Found: C, 33.70; H, 2.79; N, 1.78.

1

product dried in vacuo. Yield: 325 mg (94.5%). H NMR (300 MHz,

CD₂Cl₂) δ: 7.45−7.85 (m, 8H, ArH), 7.22−7.32 (m, 2H, ArH), 4.49

(ddd, ²J_HP= 14.3 Hz, J_HH= 9.8, 2.4 Hz, 1H, CH), 4.24 (dd, J_HH= 15.8,

10.4 Hz, 1H, CH₂), 3.36−3.76 (m, 5H, CH₂), 1.35−2.15 (m, 8H,

CH₂). ¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 186.6 (d, ²J_CP= 6.7 Hz,

cis CO), 185.3 (d, ²J_CP= 46.9 Hz, trans CO), 185.1 (d, ²J_CP= 8.3 Hz,

Preparation of [Re^I(L4H-κ₁-P)(CO)₅](BF₄)₂(11). A scintillation

vial was charged with 4 (100 mg, 0.127 mmol) dissolved in 2 mL of

CH₂Cl₂. To this solution was added HBF₄·Et₂O (18 μL, 0.13 mmol,

1.05 equiv). A white solid immediately formed, and after 10 min 8 mL

of Et₂O was added to the mixture. After allowing the product to settle,

the supernatant was decanted, and the remaining white powder was

washed with Et₂O and dried in vacuo. Yield: 110 mg of 11·CH₂Cl₂

(90.3%). ¹H NMR (300 MHz, CD₃CN) δ: 8.15 (ddd, J = 17.9, 7.8, 1.6

Hz, 1H, ArH), 7.64−7.89 (m, 12H, ArH), 6.94 (dd, J = 8.1, 4.1 Hz,

1H, ArH), 5.73 (s, 1H, NH), 5.45 (s, CH₂Cl₂), 2.40 (br, s, 12H, CH₃).

2

cis CO), 183.6 (d, J_CP= 8.4 Hz, cis CO), 171.2 (d, J_CP= 12.3 Hz,

CN), 135.3 (d, J_CP= 12.7 Hz, Ar), 134.8 (s, Ar), 133.9 (d, J_CP= 2.6

Hz, Ar), 132.1 (d, J_CP= 2.4 Hz, Ar), 130.6 (d, J_CP= 11.6 Hz, Ar),

130.2 (d, J_CP= 10.4 Hz, Ar) 130.0 (d, J_CP= 10.1 Hz, Ar), 125.6 (d, J_CP

= 56.1 Hz, Ar), 58.1 (d, J_CP= 4.7 Hz, CH₂), 56.1 (s, CH₂), 49.2 (d, J_CP

= 34.6 Hz, CH), 49.1 (s, CH₂), 28.0 (d, J_CP= 9.4 Hz, CH₂), 27.1 (s,

CH₂), 26.8 (s, CH₂), 23.1 (s, CH₂). ³¹P{¹H} NMR (121 MHz,

2

¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 179.8 (d, J_CP= 8.3 Hz, cis

CD₂Cl₂) δ: 23.4. IR (CH₂Cl₂): ν_CO

̃

= 2109 (m), 2020 (s), 2006 (s),

CO), 176.6 (d, ²J_CP= 42.5 Hz, trans CO), 157.4 (s, CN), 140.3 (d,

J_cp= 20.4 Hz, Ar), 140.0 (d, J_CP= 2.3 Hz, Ar), 137.1 (d, J_CP= 2.4 Hz,

Ar), 134.1 (d, J_CP= 2.6 Hz, Ar), 132.6 (d, J_CP= 11.2 Hz, Ar), 131.6 (d,

J_CP= 10.9 Hz, Ar), 129.2 (d, J_CP= 50.6 Hz, Ar), 127.7 (d, J_CP= 14.0

Hz, Ar), 123.9 (s, Ar), 117.8 (d, J_CP= 48.6 Hz, Ar), 55.3 (s, CH₂Cl₂),

1968 (s) cm⁻¹. Anal. Calcd for C₂₅H₂₅BF₄N₂O₄PRe: C, 41.62; H, 3.49;

N, 3.88. Found: C, 40.97; H, 3.56; N, 3.66.

Preparation of [Re^I(L1H-κ₁-P)(CO)₅](BF₄)₂(8). A scintillation vial

was charged with 1 (100 mg, 0.123 mmol) dissolved in 4 mL of

5541

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

40.5 (br, s, CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 1.2. IR

NCCH₃). ³¹P{¹H} NMR (121 MHz, C₆D₆) δ: 33.9. IR (CH₂Cl₂):

(CH₃CN): ν_CO

̃

= 2160 (m), 2053 (s) cm⁻¹. Anal. Calcd for

ν

̃

= 2306 (w); ν

= 2024 (s), 1930 (s), 1892 (s) cm⁻¹. When

̃

CN

CO

dissolved in CD₃CN, the ¹H NMR spectrum of 15 shows one

equivalent of free CH₃CN in addition to the peaks attributed to the

phosphine ligand.

C₂₉H₂₉B₂Cl₂F₈N₃O₅PRe (11·CH₂Cl₂): C, 36.24; H, 3.04; N, 4.37.

Found: C, 36.39, H, 3.13, N, 4.25.

Preparation of [Re^I(L5H-κ₁-P)(CO)₅](BF₄)₂(12). A mixture of

Re^I(CO)₅(BF₄) (200 mg, 0.484 mmol) and (L5H)(BF₄) (205 mg,

0.483 mmol, 1.00 equiv) in 25 mL of CH₂Cl₂was sealed in a 75 mL

ﬂask with a Teﬂon plug. The mixture was heated to 50 °C for 4 h,

initially giving a colorless solution from which a white solid deposited.

The volatiles were removed in vacuo, and the solid was dissolved in

MeCN and transferred to a scintillation vial. The solvent was removed

under vacuum, and the residue twice triturated with CH₂Cl₂/Et₂O to

produce a white powder, which was dried in vacuo. Yield: 346 mg

(85.4%). ¹H NMR (300 MHz, CD₃CN) δ: 7.62−7.84 (m, 10H, ArH),

6.71 (br, s, 1H, NH), 4.40 (ddd, ²J_HP= 9.5 Hz, J_HH= 9.4, 6.0 Hz, 1H,

CH), 3.02−3.47 (m, 5H, CH₂), 2.46 (m, 1H, CH₂), 1.37−2.10 (m,

8H, CH₂). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 179.2 (d, ²J_CP= 8.2

Hz, cis CO), 176.0 (d, ²J_CP= 43.2 Hz, trans CO), 151.4 (br, s, CN),

134.8 (br, d, J_cp= 76.9 Hz, Ar), 133.4 (br, s, Ar), 131.8 (d, J_CP= 10.5

Hz, Ar), 131.3 (d, J_CP= 10.8 Hz, Ar), 51.8 (br, s, CH₂), 50.5 (s, CH₂),

45.5 (d, J_CP= 18.7 Hz, CH), 40.5 (s, CH₂), 27.1 (s, CH₂), 24.2 (s,

CH₂, overlapped with another broad signal), 19.0 (s, CH₂). ³¹P{¹H}

Preparation of Re^IFormyl Complexes from W−H. The general

procedure is described here. Attempts to isolate rhenium formyl

complexes were not successful, owing to the diﬃculty of separating the

products from the tungsten-containing product, as well as hastened

decarbonylation when the formyl complexes are subjected to vacuum

to remove solvent. Therefore the formyl complexes were exclusively

studied in solution by spectroscopic means. To prepare the formyl

complexes 16−20, rhenium complexes 1−5 (25 mg) and 1.1

equivalents of W−H were dissolved each in ca. 0.35 mL of CD₃CN.

Both solutions were frozen in the glovebox cold well, and upon

thawing, the solution of W−H was added dropwise. The solution was

transferred to an NMR tube, where it was allowed to warm to room

temperature. Later experiments showed that the rhenium complex and

W−H could be mixed at room temperature with no eﬀect on the

observed yield of the formyl product. Speciﬁc details and spectral

features pertaining to the formyl complexes prepared in this way are

given below. In all cases, the tungsten-containing product was

identiﬁed as W⁰(P(OMe)₃)(NCMe)(CO)₄(W⁰), consistent with

the observed NMR spectra and possessing a nearly identical IR

spectrum to the closely related P(OⁱPr)₃-substituted analogue.³⁵In

addition, NMR features for the bis(triphenylphosphine)iminium

(PPN⁺) cation were present and likewise match previously reported

data.⁴⁶The spectroscopic data for these two byproducts, which are

present in all samples of rhenium formyl complexes, are as follows. W⁰:

NMR (121 MHz, CD₃CN) δ: 9.0 (br). IR (CH₃CN): ν_CO

̃

= 2161

(m), 2054 (s) cm⁻¹. Anal. Calcd for C₂₆H₂₆B₂F₈N₂O₅PRe: C, 37.30;

H, 3.13; N, 3.35. Found: C, 37.63; H, 3.39; N, 3.24.

In Situ Preparation of Re^I(L6⁻-κ₁-P)(CO)₅(13). Complex 6 can

be deprotonated with a variety of suitably strong bases. In a typical

experiment, complex 6·CH₂Cl₂(25 mg, 0.030 mmol) in 0.7 mL of

CD₃CN was treated with 1.1 equivalents of DBU, delivered from a

stock solution. The initially colorless solution turned pale yellow. The

NMR spectra showed, in addition to peaks for protonated DBU,

quantitative formation of 13, which has a half-life of ca. 3 h at room

3

¹H NMR (300 MHz, CD₃CN) δ: 3.64 (d, J_HP= 11.4 Hz, 9H, CH₃).

2

¹³C{¹H} NMR (126 MHz, CD₃CN): 205.9 (d, J_CP= 50.6 Hz, trans

2

1

CO), 205.4 (d, J_CP= 8.2 Hz, cis CO), 201.4 (²J_CP= 11.0 Hz, J_CW

=

1

temperature (see below for details of the decomposition). H NMR

129.0 Hz, cis CO), 52.4 (s, CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN)

(300 MHz, CD₃CN) δ: 7.50−7.66 (m, 7H, ArH), 7.36−7.45 (m, 4H,

ArH), 6.06 (ddd, J = 11.1, 2.1, 0.8 Hz, 1H, ArH), 2.08 (s, 3H, ArCH₃),

1.42 (s, 9H, C(CH₃)₃). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 1.8. IR

δ: 146.8 (¹J_PW= 384 Hz). IR (CH₃CN): ν_CO

̃

= 2026 (w), 1905 (s),

1862 (m) cm⁻¹. PPN⁺: H NMR (300 MHz, CD₃CN) δ: 7.43−7.70

(m, 30H, ArH). ¹³C{¹H} NMR (126 MHz, CD₃CN): 134.6 (s, Ar),

133.2 (m, Ar), 130.4 (m, Ar), 128.2 (dd, J_CP= 108, 2.1 Hz, Ar).

³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 21.9.

1

(CH₃CN): ν_CO

̃

= 2100 (m), 2074 (m), 1996 (s) cm⁻¹.

Conversion of 13 to κ₂-P,O Products. Within 15 h at room

temperature in acetonitrile, complex 13 decomposes in two stages;

initially two products are evident before gradual conversion to a single

species. At early time points, the intermediate present is Re^I(L6⁻-κ₂-

P,O)(CO)₄(14), which converts slowly to fac-Re^I(L6⁻-κ₂-P,O)-

(NCMe)(CO)₃(15). At 80 °C, a good yield of 15, starting from 13,

occurs within 24 h. Complex 14 is prepared as the major product by

heating 13 to 50 °C in CH₂Cl₂solution. Complexes 14 and 15 can be

separated from protonated DBU by extracting into Et₂O and ﬁltering

to remove the (DBUH)(BF₄). The following spectral characteristics

Preparation of cis-Re^I(L1-κ₁-P)(CO)₄(CHO) (16). Following the

procedure described above, complex 16 was prepared by treating 1 (25

mg, 0.031 mmol) with W−H (33 mg, 0.034 mmol, 1.1 equiv). Formyl

1

16 formed in ca. 90% yield, as judged by integration of the H NMR

1

spectrum. H NMR (300 MHz, CD₃CN) δ: 14.91 (s, 1H, CHO),

7.25−7.83 (m, 15H, ArH, overlapped with PPN⁺), 2.57 (s, 6H, CH₃),

2.28 (s, 6H, CH₃). ¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 259.4 (d,

²J_CP= 9.1 Hz, CHO), 191.1 (d, ²J_CP= 7.4 Hz, cis CO), 190.7 (d, ²J_CP

=

2

1

9.3 Hz, cis CO), 188.1 (d, J_CP= 42.6 Hz, trans CO), 154.5 (s, Ar),

153.4 (s, Ar), 139.9 (s, Ar), 137.0 (d, J_CP= 48.8 Hz, Ar), 135.1 (d, J_CP

= 11.9 Hz, Ar), 134.4 (d, J_CP= 11.9 Hz, Ar) 131.2 (d, J_CP= 2.5 Hz,

Ar), 131.1 (d, J_CP= 15.2 Hz, Ar), 129.2 (d, J_CP= 10.5 Hz, Ar), 128.9

(d, J_CP= 10.5 Hz, Ar), 128.8 (s, Ar), 125.4 (d, J_CP= 12.1 Hz, Ar),

124.6 (s, Ar), 117.8 (s, Ar), 47.0 (s, CH₃), 44.4 (s, CH₃). ³¹P{¹H}

are noted for the two products. 14: H NMR (300 MHz, CD₂Cl₂) δ:

7.36−7.64 (m, 10H, ArH), 7.03 (d, J = 2.2 Hz, 1H, ArH), 6.92 (ddd, J

= 8.7, 2.2, 1.0 Hz, 1H, ArH), 2.21 (s, 3H, ArCH₃), 1.35 (s, 9H,

C(CH₃)₃). ¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 190.0 (br, s, cis

CO), 188.0 (d, ²J_CP= 50.1 Hz, trans CO), 186.8 (d, ²J_CP= 9.1 Hz, cis

CO), 177.2 (d, J_CP= 23.7 Hz, ArCO), 142.0 (d, J_cp= 7.6 Hz, Ar),

134.2 (d, J_CP= 53.5 Hz, Ar), 132.6 (d, J_CP= 11.4 Hz, Ar), 131.7 (d, J_CP

NMR (121 MHz, CD₃CN) δ: 15.8. IR (CH₃CN): ν_CO

̃

= 2089 (m),

1994 (sh), 1983 (s), 1959 (s) cm⁻¹; ν_CO

̃

= 1590 (m) cm⁻¹.

= 2.0 Hz, Ar), 131.1 (d, J_CP= 2.6 Hz, Ar), 130.3 (s, Ar), 129.3 (d, J_CP

=

Preparation of cis-Re^I(L2-κ₁-P)(CO)₄(CHO) (17). Following the

general procedure, complex 17 was prepared from 2·THF (25 mg,

0.029 mmol) and W−H (33 mg, 0.034 mmol, 1.2 equiv) in ca. 95%

10.8 Hz, Ar), 124.2 (d, J_CP= 8.3 Hz, Ar), 112.7 (d, J_CP= 57.9 Hz, Ar),

35.4 (s, C(CH₃)₃), 29.4 (s, C(CH₃)₃), 20.6 (s, ArCH₃). ³¹P{¹H} NMR

(121 MHz, CD₂Cl₂) δ: 25.1. IR (CH₂Cl₂): ν_CO

̃

= 2101 (m), 2001

1

(s), 1933 (m), 1874 (w) cm⁻¹. 15: ¹H NMR (300 MHz, C₆D₆) δ: 7.79

(m, 2H, ArH), 7.68 (m, 2H, ArH), 7.27 (d, J = 2.2 Hz, 1H, ArH), 7.09

(ddd, J = 8.3, 2.3, 0.8 Hz, 1H, ArH), 6.87−7.04 (m, 6H, ArH), 2.17 (s,

3H, ArCH₃), 1.73 (s, 9H, C(CH₃)₃), 0.00 (d, ⁵J_HP= 1.4 Hz, NCCH₃).

¹³C{¹H} NMR (126 MHz, CD₂Cl₂) δ: 196.4 (d, J_CP= 6.6 Hz, cis CO),

194.4 (d, ²J_CP= 7.8 Hz, cis CO), 193.2 (d, ²J_CP= 65.4 Hz, trans CO),

176.9 (d, J_CP= 25.0 Hz, ArCO), 141.2 (d, J_cp= 7.7 Hz, Ar), 133.5 (d,

J_CP= 10.8 Hz, Ar), 133.2 (d, J_CP= 11.5 Hz, Ar), 131.4 (d, J_CP= 1.9 Hz,

Ar), 130.6 (d, J_CP= 2.3 Hz, Ar), 130.5 (s, Ar), 130.3 (d, J_CP= 2.4 Hz,

Ar), 129.0 (d, J_CP= 5.7 Hz, Ar), 128.9 (d, J_CP= 6.8 Hz, Ar), 123.4 (d,

J_CP= 7.6 Hz, Ar), 119.9 (s, NCCH₃), 112.1 (d, J_CP= 54.8 Hz, Ar), 35.4

(d, J_CP= 2.1 Hz, C(CH₃)₃), 29.4 (s, C(CH₃)₃), 20.7 (s, ArCH₃), 2.6 (s,

yield as judged by integration of the H NMR spectrum. H NMR

(300 MHz, CD₃CN) δ: 14.98 (d, ³J_HP= 0.8 Hz, 1H, CHO), 7.37−7.80

(m, 14H, ArH, overlapped with PPN⁺), 3.63 (m, THF, overlapped

with W⁰), 3.02 (s, 2H, CH₂), 1.78 (s, 6H, CH₃), 1.78 (m, THF,

overlapped with previous peak). ¹³C{¹H} NMR (126 MHz, CD₃CN):

259.6 (d, ²J_CP= 9.4 Hz, CHO), 190.7 (d, ²J_CP= 7.3 Hz, cis CO), 190.5

(d, ²J_CP= 9.5 Hz, cis CO), 188.1 (d, ²J_CP= 43.7 Hz, trans CO), 143.8

(d, J_CP= 6.0 Hz, Ar), 135.4 (d, J_CP= 13.3 Hz, Ar), 134.9 (s, Ar), 134.1

(d, J_CP= 11.3 Hz, Ar), 132.2 (d, J_CP= 2.3 Hz, Ar), 131.9 (d, J_CP= 2.4

Hz, Ar), 131.2 (d, J_CP= 7.6 Hz, Ar), 131.2 (s, Ar), 129.9 (d, J_CP= 10.4

Hz, Ar), 127.8 (d, J_CP= 13.7 Hz, Ar), 68.3 (s, THF), 63.2 (d, J_CP= 4.7

Hz, CH₂), 45.3 (s, CH₃), 26.2 (s, THF). ³¹P{¹H} NMR (121 MHz,

5542

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

CD₃CN) δ: 11.6. IR (CH₃CN): ν_CO

̃

= 2090 (m), 1994 (sh), 1984

stability of the product precluded ¹³C{¹H} NMR determination, which

showed very broad features in an initial attempt to collect.

(s), 1964 (s) cm⁻¹; ν_CO

̃

= 1590 (m) cm⁻¹.

Preparation of cis-Re^I(L3-κ₁-P)(CO)₄(CHO) (18). Following the

described procedure, complex 18 was prepared from 3 (25 mg, 0.036

mmol) and W−H (38 mg, 0.040 mmol, 1.1 equiv), forming in ca. 85%

Preparation of cis-[Re^I(L4H-κ₁-P)(CO)₄(CHO)](BF₄) (23). Com-

plex 19 was prepared as described above. After 15 min, the solution

was frozen in the glovebox cold well. The frozen solution was

removed, and upon thawing 2,6-lutidinium tetraﬂuoroborate (5.7 mg,

0.029 mmol, 0.90 equiv) was delivered from a stock solution. The

solution was transferred to a J. Young NMR tube, which was sealed.

The initial NMR spectra indicate the desired product present in 70%

yield. ¹H NMR (300 MHz, CD₃CN) δ: 15.38 (br, s, 1H, CHO), 9.23

(s, 1H, NH), 6.98−7.80 (m, 14H, ArH, overlapped with PPN⁺and

2,6-lutidine), 2.73 (br, s, 12H, CH₃). ¹³C{¹H} NMR (126 MHz,

CD₃CN) δ: 269.4 (d, ²J_CP= 9.4 Hz, CHO), 189.5 (d, ²J_CP= 7.7 Hz, cis

CO), 189.1 (d, ²J_CP= 6.8 Hz, cis CO), 187.4 (d, ²J_CP= 45.0 Hz, trans

CO), 159.3 (s, CN), 158.6 (s, Ar), 140.9 (s, Ar), 138.6 (s, Ar), 137.0

(d, J_CP= 11.9 Hz, Ar), 134.9 (s, Ar), 133.1 (d, J_CP= 4.6 Hz, Ar), 127.4

(d, J_CP= 9.7 Hz, Ar), 126.9 (s, Ar), 124.8 (s, Ar), 121.7 (s, Ar), 41.4 (s,

CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 10.5. IR (CH₃CN):

1

yield as judged by integration of the H NMR spectrum, with cis-

Re^I(L3-κ₁-P)(CO)₄(H) as the major side product. ¹H NMR (300

MHz, CD₃CN) δ: 15.14 (s, 1H, CHO), 7.40−7.80 (m, 10H, ArH,

2

overlapped with PPN⁺), 3.60 (d, J_HP= 3.7 Hz, 2H, CH₂), 3.41 (m,

4H, CH₂), 2.15 (app. t, J = 4.6 Hz, 4H, CH₂). ¹³C{¹H} NMR (126

2

MHz, CD₃CN) δ: 259.2 (d, J_CP= 10.4 Hz, CHO), 190.5 (d, J_CP

=

7.8 Hz, cis CO), 190.1 (d, ²J_CP= 9.5 Hz, cis CO), 188.9 (d, ²J_CP= 43.4

Hz, trans CO), 133.8 (s, Ar), 133.6 (d, J_CP= 9.9 Hz, Ar), 132.0 (s, Ar),

129.8 (d, J_CP= 9.6 Hz, Ar), 67.1 (s, CH₂), 59.7 (d, J_CP= 45.4 Hz,

CH₂), 59.2 (d, J_CP= 6.8 Hz, CH₂). ³¹P{¹H} NMR (121 MHz,

CD₃CN) δ: 6.7. IR (CH₃CN): ν_CO

̃

= 2080 (m), 1993 (sh), 1983 (s),

1953 (s) cm⁻¹; ν_CO

̃

= 1590 (m) cm⁻¹.

Preparation of cis-Re^I(L4-κ₁-P)(CO)₄(CHO) (19). Following the

general procedure, complex 19 was prepared in ca. 85% yield from 4

ν_CO

̃

= 2095 (m), 2008 (sh), 1988 (s) cm⁻¹; ν_CO

̃

= 1580⁴⁸(w)

1

cm⁻¹.

(25 mg, 0.032 mmol) and W−H (31 mg, 0.032 mmol, 1.0 equiv). H

3

Preparation of cis-Re^I(L6-κ₁-P)(CO)₄(CHO) (24). Complex 21

was prepared as described above, in 0.4 mL of CD₃CN. The solution

was frozen in the glovebox cold well. The frozen solution was removed

and allowed to thaw, and 2,6-lutidinium tetraﬂuoroborate (7.0 mg,

0.036 mmol, 1.2 equiv) in 0.3 mL of CD₃CN was added while still

cold. The solution was transferred to a J. Young NMR tube, and initial

NMR (300 MHz, CD₃CN) δ: 14.86 (d, J_HP= 1.4 Hz, 1H, CHO),

7.27−7.80 (m, 12H, ArH, overlapped with PPN⁺), 6.91 (ddd, J = 9.0,

5.3, 2.3 Hz, 1H, ArH), 6.33 (m, 1H, ArH), 2.20 (s, 12H, CH₃).

¹³C{¹H} NMR (126 MHz, CD₃CN) δ: 260.0 (d, ²J_CP= 9.6 Hz, CHO),

2

191.8 (d, J_CP= 8.0 Hz, cis CO), 191.2 (d, J_CP= 9.7 Hz, cis CO),

189.2 (d, ²J_CP= 42.0 Hz, trans CO), 159.6 (s, CN), 155.3 (d, J_CP

=

1

2.8 Hz, Ar), 136.8 (d, J_CP= 15.7 Hz, Ar), 134.5 (d, J_CP= 50.8 Hz, Ar),

133.6 (d, J_CP= 11.0 Hz, Ar), 133.6 (s, Ar), 130.7 (d, J_CP= 2.3 Hz, Ar),

128.8 (d, J_CP= 10.2 Hz, Ar), 122.3 (d, J_CP= 5.4 Hz, Ar), 119.4 (d, J_CP

= 12.4 Hz, Ar), 119.0 (s, Ar), 39.4 (s, CH₃). ³¹P{¹H} NMR (121 MHz,

NMR spectra indicated the desired product present in 75% yield. H

NMR (300 MHz, CD₃CN) δ: 14.82 (s, 1H, CHO), 8.42 (s, 1H, OH),

7.36−7.64 (m, 10H, ArH, overlapped with 2,6-lutidine), 7.32 (d, J =

2.2 Hz, 1H, ArH), 6.45 (dd, J = 11.8, 2.1 Hz, 1H, ArH), 2.11 (s, 3H,

ArCH₃), 1.41 (s, 9H, C(CH₃)₃). ³¹P{¹H} NMR (121 MHz, CD₃CN)

CD₃CN) δ: 9.3. IR (CH₃CN): ν_CO

̃

= 2086 (m), 1982 (s), 1958 (s)

cm⁻¹; ν_CO

̃

= 1590 (m) cm⁻¹.

δ: 7.6. IR (CH₃CN): ν_CO

̃

= 2095 (m), 1986 (s), 1938 (m) cm⁻¹;

Preparation of cis-Re^I(L5-κ₁-P)(CO)₄(CHO) (20). In a slight

departure from the typical procedure, protonated complex 12 (25 mg,

0.030 mmol) was combined with W−H (60 mg, 0.062 mmol, 2.1

equiv) to form 20 in ca. 50% yield. Alternatively, treatment of 5

(contaminated with ca. 15% 7) with 1 equivalent of W−H gave

ν_CO

̃

= 1595 (w) cm⁻¹.

General Procedure for pK_aDetermination. The pK_aof L1 has

been previously determined.²⁰All new pK_avalues reported here were

determined by NMR spectroscopy and are the average of two or more

self-consistent trials. The compound of interest was combined with an

acid or base with a known pK_a, and the equilibrium populations were

determined by NMR. In most cases, proton transfer was rapid on the

NMR time scale, in which case the equilibrium concentration was

1

complex 20 in similar yields. H NMR (300 MHz, CD₃CN) δ: 15.16

3

(d, J_HP= 1.1 Hz, 1H, CHO), 7.30−7.80 (m, 10H, ArH, overlapped

with PPN⁺), 2.80−3.95 (m, 7H, CH + CH₂), 1.25−2.15 (m, 8H,

CH₂). ³¹P{¹H} NMR (121 MHz, CD₃CN) δ: 8.9 (br). IR (CH₃CN):

determined from the chemical shift, using the equation χ_A= (δ_eq

−

ν_CO

̃

= 2080 (m), 1985 (s) cm⁻¹; ν_CO

̃

= 1589 (m) cm⁻¹.

δ_B)/(δ_A− δ_B), where χ_Ais the mole fraction of the conjugate acid, and

δ refers to the measured chemical shift of a given peak at equilibrium

(eq) and for pure samples of the conjugate acid (A) and base (B). The

value of χ_Awas determined using all well-resolved ¹H NMR signals, as

well as from the ³¹P{¹H} spectrum. Good agreement between the

independent calculations of χ_Awas obtained, and the value used to

calculate the equilibrium concentration was an average of the

independently determined values. In some cases, particularly for

rhenium pentacarbonyl complexes, the equilibration between the

acidic and basic forms of the complex was slow, in which case distinct

peaks for the two species could be resolved and concentrations were

determined by NMR integration. Once the equilibrium concentrations

were determined, the equilibrium constant for the reaction between

the compound of interest and the known acid/base was determined,

and by using Hess’s law the pK_aof the compound was calculated. For

the pK_ameasurements of the formyl complexes, minor side reactions

consumed some of the added acid. Thus for these experiments the

ratios of the conjugate acid and base for both the rhenium formyl

complex and the known acid were determined independently from the

equilibrium NMR spectrum, allowing K_eqto be more accurately

determined.

Preparation of (Li⁺)[cis-Re^I(L6⁻-κ₁-P)(CO)₄(CHO)]⁻(21). Com-

plex 6·CH₂Cl₂(25 mg, 0.030 mmol) was dissolved in 0.7 mL of

CD₃CN. To the resulting colorless solution was added LiHBEt₃

solution (1.0 M, 75 μL, 0.075 mmol, 2.5 equiv), which yielded a

yellow solution containing BEt₃-stabilized 21 in ca. 75% yield. ¹H

3

NMR (300 MHz, CD₃CN) δ: 15.14 (d, J_HP= 1.7 Hz, 1H, CHO),

7.18−7.60 (m, 10H, ArH), 7.01 (d, J = 2.4 Hz, 1H, ArH), 5.82 (m, 1H,

ArH), 1.93 (s, 3H, ArCH₃), 1.41 (s, 9H, C(CH₃)₃). ¹³C{¹H} NMR

(126 MHz, CD₃CN) δ: 274.9 (d, ²J_CP= 12.0 Hz, CHO), 191.6 (d, ²J_CP

= 7.6 Hz, cis CO), 189.2 (d, ²J_CP= 41.4 Hz, trans CO),⁴⁷169.3, (d, J_CP

= 7.8 Hz, Ar), 139.6 (d, J_CP= 5.4 Hz, Ar), 135.0 (s, Ar), 134.9 (s, Ar),

131.6 (s, Ar), 130.5 (s, Ar), 130.3 (d, J_CP= 7.8 Hz, Ar), 129.0 (d, J_CP

=

9.9 Hz, Ar), 118.7 (d, J_CP= 11.6 Hz, Ar), 115.9 (d, J_CP= 62.1 Hz, Ar),

35.4 (s, C(CH₃)₃), 30.2 (s, C(CH₃)₃), 21.2 (s, ArCH₃). ³¹P{¹H} NMR

(121 MHz, CD₃CN) δ: 5.6. IR (CH₃CN): ν_CO

̃

= 2088 (m), 1998

(s), 1968 (s), 1948 (s) cm⁻¹; ν_CO

̃

= 1584 (m) cm⁻¹.

Preparation of cis-[Re^I(L2H-κ₁-P)(CO)₄(CHO)](BF₄) (22). Com-

plex 17 was prepared as described above, at the same scale and housed

in a J. Young NMR tube. After 15 min, the tube was frozen in the

glovebox cold well. Pyridinium tetraﬂuoroborate (5.2 mg, 0.031 mmol,

1.0 equiv) was added to the frozen solution from a stock solution. The

tube was allowed to thaw while sealed, and the NMR spectra were

recorded immediately, indicating the presence of the desired product

The following acids/bases were used for determination of unknown

pK_avalues: triethylamine (pK_a= 18.82) for L2H and L3H; 2-(2′-

tolyl)-1,1,3,3-tetramethylguanidine (pK_a= 20.5)³⁰for L4H; DBU (pK_a

= 24.34)³¹for L5H; phenol (pK_a= 28.12) for L6; pyridine (pK_a=

12.53)³¹for 8, 9, and 11; 2-chloroaniline (pK_a= 7.86) for 10; 2,6-

lutidine and 2,6-lutidinium (pK_a= 14.13) for 12 and 22, respectively;

DMAP (pK_a= 17.95)³¹for 6; and DMAPH⁺for 23 and 24.

1

in ca. 50% yield. H NMR (300 MHz, CD₃CN) δ: 15.15 (br, s, 1H,

CHO), 7.15−7.80 (m, 14H, ArH, overlapped with PPN⁺), 3.05 (br, s,

2H, CH₂), 2.29 (br, s, 6H, CH₃). ³¹P{¹H} NMR (121 MHz, CD₃CN)

δ: 13.0. IR (CH₃CN): ν_CO

̃

= 2097 (m), 1983 (s), 1964 (sh) cm⁻¹;

ν_CO

̃

= 1590 (w) cm⁻¹. The low yield of the reaction and poor

5543

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

X-ray Crystallographic Procedures. All crystallizations were

carried out at room temperature. Crystals of (L5H)(BF₄) were

obtained from 2-butanone at room temperature, 1 was crystallized

from a mixture of CH₂Cl₂and chlorobenzene by vapor diﬀusion of

THF, 2, 3, and 6−9 were crystallized from MeCN by vapor diﬀusion

of Et₂O, 4 was crystallized from CH₂Cl₂layered with Et₂O, crystals of

11 were grown from CH₂Cl₂/MeCN by diﬀusion of Et₂O, and crystals

of 12 deposited from MeCN by vapor diﬀusion of CH₂Cl₂. Crystals

were mounted on either a Bruker APEXII four-circle diﬀractometer or

Bruker three-circle diﬀractometer with a SMART 1K CCD detector

using Mo radiation from a sealed-tube 3 kW X-ray generator. The data

were collected at 100(2) K and were processed and reﬁned using the

program SAINT supplied by Siemens Industrial Automation.

Structures were solved by Patterson methods or direct methods in

SHELXS and reﬁned by standard diﬀerence Fourier techniques in the

SHELXTL program suite (6.10 v., Sheldrick G. M., and Siemens

Industrial Automation, 2000). Hydrogen atoms bonded to carbon

were placed in calculated positions using the standard riding model

and reﬁned isotropically; all non-hydrogen atoms were reﬁned

anisotropically. In the structures of (L5H)(BF₄), 6−9, 11, and 12,

oxygen and nitrogen-bound hydrogen atoms were located in the

diﬀerence map, restrained to a distance of 0.84 Å (O−H) or 0.88 Å

(N−H), and reﬁned isotropically with the isotropic displacement

parameter constrained to be 1.2 times greater than that of the atom it

(5) (a) den Breejen, J. P.; Radstake, P. B.; Bezemer, G. L.; Bitter, J.

H.; Frøseth, V.; Holmen, A.; de Jong, K. P. J. Am. Chem. Soc. 2009,

131, 7197−7203. (b) Borg, O.; Dietzel, P. D. C.; Spjelkavik, A. I.;

Tveten, E. Z.; Walmsley, J. C.; Diplas, S.; Eri, S.; Holmen, A.; Rytter, E.

J. Catal. 2008, 259, 161−164. (c) Torres Galvis, H. M.; Koeken, A. C.

J.; Bitter, J. H.; Davidian, T.; Ruitenbeek, M.; Dugulan, A. I.; de Jong,

K. P. J. Catal. 2013, 303, 22−30. (d) Dinse, A.; Aigner, M.; Ulbrich,

M.; Johnson, G. R.; Bell, A. T. J. Catal. 2012, 288, 104−114.

(6) (a) Gong, J.; Yue, H.; Zhao, Y.; Zhao, S.; Zhao, L.; Lv, J.; Wang,

S.; Ma, X. J. Am. Chem. Soc. 2012, 134, 13922−13925. (b) Yin, H.;

Ding, Y.; Luo, H.; Yan, L.; Wang, T.; Lin, L. Energy Fuels 2003, 17,

1401−1406.

(7) (a) Loveless, B. T.; Buda, C.; Neurock, M.; Iglesia, E. J. Am.

Chem. Soc. 2013, 135, 6107−6121. (b) Ojeda, M.; Li, A.; Nabar, R.;

Nilekar, A. U.; Mavrikakis, M.; Iglesia, E. J. Phys. Chem. C 2010, 114,

19761−19770.

(8) (a) De Smit, E.; Cinquini, F.; Beale, A. M.; Safonova, O. V.; van

Beek, W.; Sautet, P.; Weckhuysen, B. M. J. Am. Chem. Soc. 2010, 132,

14928−14941. (b) Zhao, Y.-F.; Rousseau, R.; Li, J.; Mei, D. J. Phys.

Chem. C 2012, 116, 15952−15961.

(9) West, N. M.; Miller, A. J. M.; Labinger, J. A.; Bercaw, J. E. Coord.

Chem. Rev. 2011, 255, 881−898.

(10) (a) Demitras, G. C.; Muetterties, E. L. J. Am. Chem. Soc. 1977,

99, 2796−2797. (b) Collman, J. P.; Brauman, J. I.; Tustin, G.; Wann,

G. S., III. J. Am. Chem. Soc. 1983, 105, 3913−3922.

(11) (a) Dombek, B. D. Adv. Catal. 1983, 32, 325−416. (b) Kiso, Y.;

Tanaka, M.; Hayashi, T.; Saeki, K. J. Organomet. Chem. 1987, 322,

C32−C36.

−

is bonded to. In the structures of 1 and 4, BF₄counterions were

found to be disordered about two positions. For the disordered parts,

bond distances and angles were restrained to be similar using the

“SADI” command, and the rigid bond restraints “SIMU” and “DELU”

were also employed. The structure of 4 also contained a solvent

molecule disordered about a cubic special position. The size and

electron density of this void was consistent with the presence of a

single Et₂O solvent molecule, but the electron density could not be

satisfactorily modeled. The SQUEEZE function within PLATON was

employed for the ﬁnal reﬁnement cycles of this structure. A summary

of crystallographic details for all structures is provided in Tables S1−

S4 in the Supporting Information.

(12) (a) Gladysz, J. A. Adv. Organomet. Chem. 1982, 20, 1−38.

(b) Collman, J. P.; Winter, S. R. J. Am. Chem. Soc. 1973, 95, 4089−

4090. (c) Casey, C. P.; Andrews, M. A.; Rinz, J. E. J. Am. Chem. Soc.

1979, 101, 741−743. (d) Tam, W.; Wong, W.-K.; Gladysz, J. A. J. Am.

Chem. Soc. 1979, 101, 1589−1591. (e) Barger, P. T.; Bercaw, J. E.

Organometallics 1984, 3, 278−284.

(13) Wolczanski, P. T.; Bercaw, J. E. Acc. Chem. Res. 1980, 13, 121−

127.

(14) (a) Butts, S. B.; Holt, E. M.; Strauss, S. H.; Alcock, N. W.;

Stimson, R. E.; Shriver, D. F. J. Am. Chem. Soc. 1979, 101, 5864−5866.

(b) Lindner, E.; von Au, G. Angew. Chem., Int. Ed. Engl. 1980, 19,

824−825.

ASSOCIATED CONTENT

* Supporting Information

Crystallographic summary tables, X-ray crystal structure

depictions for (L5H)(BF₄), 1−3, 6, 7−9, and 12, NMR

spectra for all new compounds. This material is available free of

charge via the Internet at http://pubs.acs.org.

■

S

(15) (a) Miller, A. J. M.; Labinger, J. A.; Bercaw, J. E. J. Am. Chem.

Soc. 2008, 130, 11874−11875. (b) Miller, A. J. M.; Labinger, J. A.;

Bercaw, J. E. J. Am. Chem. Soc. 2010, 132, 3301−3303. (c) Miller, A. J.

M.; Labinger, J. A.; Bercaw, J. E. Organometallics 2010, 29, 4499−4516.

(16) Dobrovetsky, R.; Stephan, D. W. J. Am. Chem. Soc. 2013, 135,

4974−4977.

AUTHOR INFORMATION

Corresponding Authors

*E-mail (J. E. Bercaw): bercaw@caltech.edu.

*E-mail (J. A. Labinger): jal@caltech.edu

Notes

■

(17) Sazama, G. T.; Betley, T. A. Organometallics 2011, 30, 4315−

4319.

(18) (a) West, N. M.; Labinger, J. A.; Bercaw, J. E. Organometallics

2011, 30, 2690−2700. (b) Hazari, A.; Labinger, J. A.; Bercaw, J. E.

Angew. Chem., Int. Ed. 2012, 51, 8268−8271.

The authors declare no competing ﬁnancial interest.

(19) Ellis, W. W.; Miedaner, A.; Curtis, C. J.; Gibson, D. H.; DuBois,

D. L. J. Am. Chem. Soc. 2002, 124, 1926−1932.

ACKNOWLEDGMENTS

We thank BP for their support of this work through the XC²

program.

■

(20) Farrer, N. J.; McDonald, R.; McIndoe, J. S. Dalton Trans. 2006,

4570−4579.

(21) (a) Rauchfuss, T. B.; Patino, F. T.; Roundhill, D. M. Inorg.

Chem. 1975, 14, 652−656. (b) Alonso, M. A.; Casares, J. A.; Espinet,

P.; Soulantica, K.; Orpen, A. G.; Phetmung, H. Inorg. Chem. 2003, 42,

3856−3864.

REFERENCES

■

(1) Keim, W. “Synthesis Gas.” In Industrial Chemicals via C1

Processes; Fahey, D. R., Ed.; American Chemical Society: Washington,

DC, 1987; Vol. 328, pp 1−16.

(22) Song, H.-B.; Zhang, Z.-Z.; Mak, T. C. W. New J. Chem. 2002, 26,

113−119.

(23) O’Reilly, M.; Pattacini, R.; Braunstein, P. Dalton Trans. 2009,

6092−6095.

(2) (a) Rofer-DePoorter, C. K. Chem. Rev. 1981, 81, 447−474.

(b) Khodakov, A. Y.; Chu, W.; Fongarland, P. Chem. Rev. 2007, 107,

1692−1744.

(24) Matsumura, N.; Nishiguchi, H.; Okada, M.; Yoneda, S. J.

Heterocycl. Chem. 1986, 23, 885−887.

(3) Fischer, F.; Tropsch, H. Process for the Production of Paraﬃn-

Hydrocarbons with More Than One Carbon Atom. U.S. Patent

1,746,464, February 11, 1930.

(25) Rauchfuss, T. B. Inorg. Chem. 1977, 16, 2966−2968.

(26) Haddad, M.; Laghzaoui, M.; Welter, R.; Dagorne, S.

Organometallics 2009, 28, 4584−4592.

(4) De Smit, E.; Weckhuysen, B. M. Chem. Soc. Rev. 2008, 37, 2758−

2781.

(27) Pruszynski, P. Can. J. Chem. 1987, 65, 626−629.

5544

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Organometallics

Article

(28) Beck, W.; Raab, K.; Shapley, J. R.; Whittlesey, B. R. Inorg. Synth.

1989, 26, 106−113.

(29) Drew, D.; Darensbourg, D. J.; Darensbourg, M. Y. Inorg. Chem.

1975, 14, 1579−1584.

(30) Leffek, K. T.; Pruszynski, P.; Thanapaalasingham, K. Can. J.

Chem. 1989, 67, 590−595.

(31) Kaljurand, I.; Kutt, A.; Soovali, L.; Rodima, T.; Maemets, V.;

̈

Leito, I.; Koppel, I. A. J. Org. Chem. 2005, 70, 1019−1028.

(32) Kutt, A.; Movchun, V.; Rodima, T.; Dansauer, T.; Rusanov, E.

B.; Leito, I.; Kaljurand, I.; Koppel, J.; Pihl, V.; Koppel, I.; Ovsjannikov,

̈

G.; Toom, L.; Mishima, M.; Medebielle, M.; Lork, E.; Roschenthaler,

̈

G.-V.; Koppel, I. A.; Kolomeitsev, A. A. J. Org. Chem. 2008, 73, 2607−

2620.

(33) Slater, S. G.; Lusk, R.; Schumann, B. F.; Darensbourg, M.

Organometallics 1982, 1, 1662−1666.

(34) 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.

(35) Schenk, W. A. J. Organomet. Chem. 1979, 179, 253−261.

(36) One of the bands appears as a shoulder and was not resolved in

the spectra of 19 and 20.

(37) (a) Henry, R. M.; Shoemaker, R. K.; DuBois, D. L.; Rakowski

DuBois, M. J. Am. Chem. Soc. 2006, 128, 3002−3010. (b) Stephan, D.

W.; Erker, G. Angew. Chem., Int. Ed. 2010, 49, 46−76.

(38) Elowe, P. R.; West, N. M.; Labinger, J. A.; Bercaw, J. E.

Organometallics 2009, 28, 6218−6227.

(39) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;

Timmers, F. J. Organometallics 1996, 15, 1518−1520.

(40) Long, R. J.; Gibson, V. C.; White, A. J. P.; Williams, D. J. Inorg.

Chem. 2006, 45, 511−513.

(41) Golisz, S. R.; Bercaw, J. E. Macromolecules 2009, 42, 8751−8762.

(42) Schmidt, S. P.; Nitschke, J.; Trogler, W. C.; Huckett, S. I.;

Angelici, R. J. Inorg. Synth. 1989, 26, 113−117.

(43) Perrin, D. D. Puriﬁcation of Laboratory Chemicals, 3rd ed.;

Pergamon Press: Oxford, 1988.

(44) Miedaner, A.; DuBois, D. L.; Curtis, C. J.; Haltiwanger, R. C.

Organometallics 1993, 12, 299−303.

(45) The ¹³C{¹H} spectrum was broad, and only six of the expected

10 aromatic signals were resolved.

(46) Lacour, M.-A.; Zablocka, M.; Duhayon, C.; Majoral, J.-P.;

Taillefer, M. Adv. Synth. Catal. 2008, 350, 2677−2682.

(47) The third CO resonance in the ¹³C{¹H} NMR was not located,

but all other spectroscopic data are consistent with the stated

formulation.

(48) The presence of ligand stretches in this region made it diﬃcult

to precisely locate the weak CO stretch.

5545

dx.doi.org/10.1021/om400810v | Organometallics 2013, 32, 5530−5545

Article Doi

A thermodynamic analysis of rhenium(I)-Formyl C-H bond formation via base-assisted heterolytic H2 cleavage in the secondary coordination sphere

DOI: 10.1021/om400810v

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om400810v

Full text of DOI:10.1021/om400810v

Products guided by the article

R&D Labs maybe for 1188417-02-8

Relevant to this article

Hot Product