Trisphosphine-Chelate-Substituted Molybdenum and Tungsten Nitrosyl Hydrides as Highly Active Catalysts for Olefin Hydrogenations

DOI: 10.1002/chem.201402736

Full Paper

&

Homogeneous Catalysis

Trisphosphine-Chelate-Substituted Molybdenum and Tungsten

Nitrosyl Hydrides as Highly Active Catalysts for Olefin

Hydrogenations

Subrata Chakraborty, Olivier Blacque, Thomas Fox, and Heinz Berke*^[a]

Abstract: Reaction of [M(NO)Cl₃(NCMe)₂] (M=Mo, W) with

(iPr₂PCH₂CH₂)₂PPh (etpⁱp) at room temperature afforded the

syn/anti-[M(NO)Cl₃(mer-etpⁱp)] complexes (M=Mo, a; W, b;

3a,b(syn,anti); syn and anti refer to the relative position of

Ph(etpⁱp) and NO). Reduction of 3a,b(syn,anti) produced

mixture. The 7a,b(syn,anti)/Et₃SiH/B(C₆F₅)₃co-catalytic sys-

tems were highly active in various olefin hydrogenations

(60 bar H₂, 1408C), with maximum TOFs of 5250 h^ꢀ1(7a-

(syn,anti)) and 8200 h^ꢀ1(7b(syn,anti)) for 1-hexene hydroge-

nation. The Et₃SiH/(B(C₆F₅)₃co-catalyst is anticipated to gen-

erate a [Et₃Si]⁺cation attaching to the O_NOatom. This facili-

tates NO bending and accelerates catalysis by providing

a vacant site. Inverse DKIE effects were observed for the 7a-

(syn,anti)/Et₃SiH/(B(C₆F₅)₃(k_H/k_D=0.55) and the 7b(syn,anti)/

Et₃SiH/(B(C₆F₅)₃(k_H/k_D=0.65) co-catalytic mixtures (20 bar

H₂/D₂, 1408C).

[M(NO)Cl₂(mer-etpⁱp)]

(4a,b(syn)),

[M(NO)Cl(NCMe)(mer-

etpⁱp)] (5a,b(syn,anti)), and [M(NO)Cl(h²-ethylene)(mer-etpⁱp)]

(6a,b(syn,anti)) complexes. The hydrides [M(NO)H(h²-ethyl-

ene)(mer-etpⁱp)] (7a,b(syn,anti)) were obtained from 6a,b-

(syn,anti) using NaHBEt₃(758C, THF) or LiBH₄(808C, Et₃N), re-

spectively. 7a,b(syn,anti) were probed in olefin hydrogena-

tions in the absence or presence of a hydrosilane/B(C₆F₅)₃

Introduction

combination of bound H atoms constitutes a H₂equivalent,

which is prone for hydrogenations of polar unsaturated func-

tionalities (C=O or C=N) by subsequent H atom transfer, as

demonstrated by the groups of Bullock,^[1c,2a,6]Norton,^[7]Martins,

and our group.^[8]Very scant attention has been paid to devel-

opments of catalytic molybdenum- and tungsten-based hydro-

genations of less-polar substrates, such as alkenes.^[9]Stoichio-

metric hydrogenations of alkenes and alkynes using Mo and W

complexes were accomplished by Bullock and co-workers.^[10]

Furthermore, Baricelli^[11]and co-workers reported recently that

the [Mo(CO)₃(PPh₃)₂(NCMe)] complex is active in the hydroge-

nation of aromatic and aliphatic alkenes, achieving a maximum

TOF of 48 h^ꢀ1in the respective transformation of styrene. Our

group’s long-standing efforts to apply middle transition ele-

ments in various catalyses using nitrosyl complexes turned out

recently to be successful in the case of rhenium com-

plexes,^[12a–c]demonstrating excellent performance for hydroge-

nations and hydrosilylations, including efficient hydrosilylation

reactions catalysed by a large-bite-angle DPEphos-ligand-teth-

ered molybdenum complex.^[12d]The search for effective rheni-

um hydrogenations started from large-bite-angle diphosphine-

substituted rhenium complexes of the type [ReBrH(h²-

C₂H₄)(P\P)] (P\P=bidentate phosphine ligand), which were

found to exhibit surprisingly good catalytic performance in

olefin hydrogenations.^[13]The olefin and the hydride moiety of

the active catalytic species were expected to enable insertion

of the olefin into the L_nMꢀH bond best when disposed cis,

which was anticipated to be supported by a specially designed

residual coordination sphere. Furthermore, low-valent middle

transition-element complexes generally lack the ability to pro-

Economic constraints, limited availability, and often toxicity

issues of precious-metal catalysts initiated in recent years

global research efforts to replace platinum-group catalysts

with cheap, low-toxicity, and earth-abundant metals, particular-

ly in homogeneous catalysis. For this quite important research

task, Bullock coined the slogan “Cheap Metals for Noble

Tasks”.^[1]In this regard, catalysts with the middle transition ele-

ments iron, molybdenum, and tungsten are attractive replace-

ments.^[2]The hydrogenation of olefins using transition-metal

complexes is one of the most important and widely applied

homogeneous catalysis to synthesise saturated compounds.^[3]

The mechanisms of these catalytic reactions are well-under-

stood and were found to follow mainly Wilkinson- or Osborn-

type courses, which proceed with oxidative addition of H₂to

a metal centre, splitting H₂into equal H atom parts (“homolytic

splitting”).^[4]The majority of the catalyses indeed employ pre-

cious metal compounds based on rhodium, ruthenium, and iri-

dium.^[5]

Molybdenum and tungsten hydrogenation catalysts were

developed in the past few years. They follow mainly so-called

ionic hydrogenation pathways^[1c,4e,5f]with heterolytic splitting

of H₂generating a hydride and a protic function. This polar

[a] Dr. S. Chakraborty, Dr. O. Blacque, Dr. T. Fox, Prof. H. Berke

Department of Chemistry, University of Zꢀrich

Winterthurerstrasse 190, 8057 Zꢀrich (Switzerland)

E-mail: hberke@chem.uzh.ch

Supporting information for this article is available on the WWW under

http://dx.doi.org/10.1002/chem.201402736.

Chem. Eur. J. 2014, 20, 12641 – 12654

12641

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

vide catalytically relevant vacant sites by ligand dissociation.^[14a]

Our group^[15]could recently demonstrate that creation of such

sites could be induced by reversible “nitrosyl bending” by co-

ordination of a strong electrophile^[16,17]to the O_NOatom, which

also dramatically enhanced the catalytic activity of nitrosyl rhe-

nium systems. This effect was then denoted as the “catalytic

nitrosyl effect”.^[14a]For instance, Re^Idiiodide/hydrosilane/

B(C₆F₅)₃co-catalytic systems showed extremely high activities

of up to 600000 h^ꢀ1, which were found superior to any previ-

ous homogeneous olefin hydrogenation system. It could be

made plausible that the catalytic enhancement indeed origi-

nated from reversible NO bending by attachment of a [R₃Si]⁺

cation to the nitrosyl group.^[14a]The vacant site created by the

nitrosyl bending together with another vacant site (created

perhaps by aid of other ligand effects, like the trans effect or

the cis labilisation effect)^[14]was presumed to generate a super-

electrophile, which according to Negishi’s notion was expected

to be extraordinarily catalytic.^[18]

Scheme 1. Preparation of the isomers of [M(NO)Cl₃(mer-etpⁱp)] 3a,b(syn,anti).

a ratio of 9:1 (3b(syn)/3b(anti)) as indicated by ³¹P{¹H} NMR.

The presence of two different sets of doublets and triplets in

the ³¹P{¹H} NMR spectra of the reaction mixtures confirmed ini-

tial formation of these isomers in yields (from spectroscopy) of

58% and 42% for 3a(syn) and 3a(anti) and of 56% and 44%

for 3b(syn) and 3b(anti), respectively. The ³¹P{¹H} NMR spectra

of 3a(syn) and 3a(anti) are shown in Figure 1.

Therefore, trying to turn appropriate complexes of the

middle transition metals molybdenum and tungsten into effi-

cient olefin hydrogenation catalysts, we attempted a “tuning”

of their ligand sphere. We set out to construct the coordina-

tion sphere employing a strongly donating chelate trisphos-

phine to facilitate oxidative additions and to additionally func-

tion as an ancillary ligand. A functional NO ligand in conjunc-

tion with the co-catalytic hydrosilane/B(C₆F₅)₃reagent and

a labile ligand were to preposition vacant sites.

Figure 1. ³¹P{¹H} NMR spectra of the isomeric mixture of the molybdenum

system 3a(syn,anti) at room temperature in CD₂Cl₂. d=76 (²J_PP=72 Hz,

3a(syn)) and 77 ppm (²J_PP=76 Hz, 3a(anti)) for the P_terminalatoms and d=89

(²J_PP=76 Hz, 3a(anti)) and 98 ppm (²J_PP=72 Hz, 3a(syn)) for the P_centralatoms.

Results and Discussion

Preparation of [M(NO)Cl₃(mer-etpⁱp)] complexes (M=Mo,

3a(syn,anti); W, 3b(syn,anti))

Treatment of one equivalent of the (iPr₂PCH₂CH₂)₂PPh ligand

(2, etpⁱp; see the Supporting Information) with THF solutions

of [M(NO)Cl₃(NCMe)₂], (M=Mo, W) at room temperature led to

precipitation of isomeric mixtures of the pentagonal bipyrami-

dal [Mo(NO)Cl₃(mer-etpⁱp)] complexes (3a(syn) and 3a(anti))

and [W(NO)Cl₃(mer-etpⁱp)] (3b(syn) and 3b(anti)) in 55% and

52% yields, respectively (Scheme 1). The isomer notations for

3a(syn) and 3a(anti) and 3b(syn) and 3b(anti) correspond to

the relative position of the phenyl ring at the internal phos-

phorus atom syn and anti to the NO ligand. 3a,b(syn,anti) dis-

solved well only in CH₂Cl₂. 3b(syn,anti) were found to be also

sparingly soluble in THF, which allowed their partial separation

by repeated extractions with THF. We also failed to separate

the syn/anti isomers by chromatography. After several extrac-

tions, 3b(syn) prevailed in the combined THF solutions in

The doublet sets in the ³¹P{¹H} NMR of the isomeric mixture

of 3b(syn,anti) appeared at d=55 (²J_PP=46 Hz, 3b(syn)),

56 ppm (²J_PP=49 Hz, 3b(anti)) and the triplet sets at d=80

(²J_PP=46 Hz, 3b(syn)) and 72 ppm (²J_PP=49 Hz, 3b(anti)). Relat-

1

ed to the ³¹P{¹H} NMR spectra, the H NMR spectra of the 3a,b-

(syn,anti) mixture in CD₂Cl₂showed also doubled sets of mul-

tiplets for the methyl, methylene, methyne, and phenyl pro-

tons, again confirming presence of isomers. The 3a,b(syn,anti)

isomers could not be distinguished in the IR spectra, as the

molybdenum and the tungsten conformers gave rise to only

one strong n(NO) band at 1647 and 1618 cm^ꢀ1, respectively.

The isomers of 3a(syn,anti) crystallised as 3a(syn) and

3a(anti) in form of yellow single crystals suitable for X-ray dif-

fraction analysis upon layering pentane onto a concentrated

CH₂Cl₂solution of the isomeric mixture. A perspective view of

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12642

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

tion of the 3b(syn,anti) mixture. Similar to the 3a(syn,anti) iso-

mers, the basic molecular structure of 3b(syn) revealed a pen-

tagonal-bipyramidal geometry with trans NO/Cl ligands. The

W1–N1 bond length of 3b(syn) is 1.781(2) ꢁ. As the metal cen-

tres of the 3a,b(syn,anti) isomers are in the +II oxidation state,

reduction was attempted to allow access to low oxidation

state complexes and to eventually apply these complexes as

catalyst precursors in homogeneous hydrogenations.

Preparation of [M(NO)Cl₂(mer-etpⁱp)] complexes 4a(syn) and

4b(syn)

Figure 2. Molecular structures of [Mo(NO)Cl₃(mer-etpⁱp)] (3a(anti); left) and

of [W(NO)Cl₂(mer-etpⁱp)] (4b(syn); right) with ellipsoids set at 50% probabili-

ty.^[24]The terminal isopropyl substituents of both complexes and the toluene

solvate of 4b(syn) are omitted for clarity. Selected bond distances(ꢁ) and

bond angles(^o) for 3a(anti): Mo1–N1A 1.788(2), Mo1–P1 2.6415(4), Mo1–P2

2.5202(5), N1A–O1A 1.202(3), Mo1–Cl1A 2.4569(5); N1A-Mo1-Cl1A 178.05(5),

P2-Mo1-P3 71.878(14), N1A-Mo1-P2 82.44(6), Mo1-N1A-O1A 176.16(16). Se-

lected bond distances [ꢁ] and angles [8] for 4b(syn): W1–N1 1.858(5), N1–O1

1.042(5), W1–P1 2.5488(11), W1–Cl1 2.5162(13); N1-W1-Cl1 176.37(13), P1-

W1-P3 155.71(4), N1-W1-P2 99.57(13), O1-N1-W1 175.6(5).

Reduction of the isomers of 3a,b(syn,anti) with one equivalent

of 1% Na/Hg in THF at room temperature afforded the green

17 e^ꢀconformationally unique species [M(NO)Cl₂(mer-etpⁱp)]

(M=Mo, 4a(syn); W, 4b(syn)) and in 59% and 54% yields, re-

spectively (Scheme 2). They were found to be quite soluble in

the structural model of 3a(anti) is depicted in Figure 2 (for

3a(syn), see the Supporting Information). The crystal structure

analysis of the crystalline mixture of both isomers 3a(syn) and

3a(anti) revealed pentagonal-bipyramidal molybdenum centres

with the three phosphorus atoms of the mer-etpⁱp ligand and

of two chloride ligands attached to the metal centre occupying

the pentagonal plane. There were two independent molecules

in the lattice. One showed disorder in the trans NO/Cl axis in

a ratio of 0.935(2):0.065(2) displaying the presence of 3a(anti)

and 3a(syn) at this lattice position, while the other lattice posi-

tion was found to be occupied by 3a(syn) only. It is worth

mentioning that apparently due to a slight kinetic difference in

the crystallisation rates of the 3a(syn) and 3a(anti) isomers,

quick crystallisation of the 3a(syn,anti) mixture furnishes in the

solid state only the 3a(syn) form as revealed by a X-ray diffrac-

tion study of 3a(syn), which did not show any disorder. A sub-

sequent slow crystallisation of the remaining mother liquor

showed co-crystallisation and disorder of the 3a(syn) and

3a(anti) isomers with predominancy of the 3a(anti) isomer. The

³¹P{¹H} NMR spectrum of the pure 3a(syn) crystals in CH₂Cl₂at

room temperature obtained by quick crystallisation exhibited

only one doublet and triplet set, which allowed the aforemen-

tioned structural assignment. The O1A–N1A–Mo angle of

3a(anti) is 176.16 (16)8 and the Mo1–N1–O1 angle of 3a(syn) is

173.73(13)8, which are both close to linearity. The MoꢀN bond

lengths of 3a(syn,anti) were found to be 1.7749(15) ꢁ and

1.788(2) ꢁ, respectively, which are in accord with related bond

lengths reported earlier.^[19]

Scheme 2. Reduction of the isomeric mixtures of 3a(syn,anti) and 3b-

(syn,anti) leading to M^Iand M⁰complexes.

THF and CH₂Cl₂, but sparingly soluble in toluene and benzene.

Owing to the paramagnetic character of 4a,b(syn), no signals

were observable in the ³¹P{¹H} and ¹H NMR spectra. Besides

EPR and magnetic measurements, these compounds were

characterised by IR spectroscopy, elemental analysis, and X-ray

diffraction studies. The IR spectra displayed two sharp bands at

1589 and 1564 cm^ꢀ1for 4a(syn) and 4b(syn), respectively, as-

signed to nitrosyl stretching vibrations, which appeared at

lower wavenumbers than those of the 3a,b(syn,anti) mixtures,

indicating a higher degree of back-donation to the nitrosyl

ligand. The molecular structures of 4a,b(syn) revealed neutral

pseudo-octahedral complexes as just the syn conformers

(Figure 2). The three phosphorus atoms of the mer-etpⁱp ligand

and one chloride ligand attached to the metal centre occupy

“equatorial” positions. The nitrosyl and the other chloride

ligand are disposed trans; NO/Cl disorders were not found.

This may indicate a stronger thermodynamic preference for

the syn configuration in the solid state, but what cannot be

ruled out in these cases is the possibility of dynamic behaviour

in solution with a fast interconversion of the syn and anti iso-

It became also apparent from the structure analysis that the

3a(syn,anti) isomers do not differ much in their bond lengths

and angles parameters, except that there is different dihedral

angles for the two conformations. A X-ray diffraction study was

also carried out on a single crystal of 3b(syn), which is de-

scribed in the Supporting Information. The crystal was ob-

tained by layering pentane onto a concentrated CH₂Cl₂solu-

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12643

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

mers. Selected bond angles and bond distances for 4b(syn) are

given in Figure 2 (for 4a(syn), see the Supporting Information).

The MꢀN bond lengths of the M^Ispecies 4a(syn) and 4b(syn)

were found to be 1.818(3) ꢁ and 1.858(5) ꢁ, respectively, which

are counterintuitively slightly longer than those of their M^IIpre-

cursors 3a(syn) and 3b(syn), as the latter species in higher oxi-

dation state are expected to possess less potential for p back-

donation. These longer MꢀN bond lengths of 4a,b(syn) also

contrast the position of the n(NO) bands at wavenumbers

lower than those of 3a,b(syn,anti), which may be explained by

the effect of enhanced p donation of the three chloride li-

gands. 4a,b(syn) are rare paramagnetic Mo and W complexes

with the metal centres in +I oxidation state and of low-spin d⁵

configurations. To examine the paramagnetic behaviour, solu-

tion EPR measurements of 4a,b(syn) were carried out in tolu-

ene at 298 K, which are depicted in Figure 3.

Figure 4. Temperature-dependent magnetic susceptibility data for 4a(syn)

recorded in the range of 5–300 K.

Preparation of [M(NO)Cl(NCMe)(mer-etpⁱp)] complexes

(M=Mo, 5a(syn,anti); W, 5b(syn,anti))

Reduction of the isomeric mixture of 3a(syn,anti) can be ac-

complished by stirring an acetonitrile solution for 3 days at

room temperature with an excess of zinc to form the diamag-

netic isomeric species [Mo(NO)Cl(NCMe)(mer-etpⁱp)], (52%,

5a(syn) and 48%, 5a(anti) based on ³¹P{¹H} NMR)) in 75% yield

of isolated product (Scheme 2). The ³¹P{¹H} NMR spectrum of

the 5a(syn,anti) mixture revealed two singlets at d=102.5 and

67.7 ppm (1:2 ratio) assigned to the major isomer 5a(syn) and

two other singlets at d=95 and 62.8 ppm corresponding to

the minor isomer 5a(anti). It was surprising to see that no cou-

pling was observed in the ³¹P{¹H} NMR spectra even though

quite different phosphorus environments exist in both cases of

isomers. The IR spectra of 5a(syn,anti) in the range of 1500–

2000 cm^ꢀ1showed only one sharp band at 1598 cm^ꢀ1assigned

to the n(NO) stretching vibrations. It is worth mentioning that

reduction of 4a(syn) with Zn (5 equiv) in CH₃CN at room tem-

perature after 48 h led to the isomeric mixture of 5a(syn,anti)

(Scheme 2).

Figure 3. EPR spectra of 4a(syn) (left) and 4b(syn) (right) in toluene at 298 K.

The spectrum of 4a(syn) (Figure 3, left) showed two signals

with hyperfine splitting. The coupling patterns of the two sig-

nals are due to coupling of one unpaired electron in each

compound with the internal phosphorus atom (A_P=27 G, ³¹P,

I=1/2). Hyperfine splitting arises from the coupling with the

hydrogen atoms of the phenyl ring (A_H=3 G, 1H, I=1/2). An-

other six weak signals were observed owing to the very weak

coupling of the unpaired electron with the molybdenum

centre (^95,97Mo I=5/2, combined 26%).

Single crystals of the isomeric mixture of 5a(syn,anti) suitable

for a X-ray diffraction study could be obtained after several

days by layering pentane onto a concentrated toluene solution

at ꢀ308C. The coordination geometries of the 5a(syn,anti) iso-

mers were found to be pseudo octahedral (Figure 5). The three

phosphorus atoms and the coordinated acetonitrile ligand

were lying in a plane. The NO and Cl ligands were found to be

disordered over two sets of positions with a site-occupancy

factor of 0.5, supporting the presence of both of the 5a(syn)

and 5a(anti) isomers. The structures of the isomers of 5a-

There is no coupling with the terminal phosphorus atoms,

which was interpreted in terms of the unpaired electron resid-

ing in a d orbital in the (Cl₂, NO, P_int) plane. The EPR spectrum

of 4b(syn) showed in toluene at room temperature two sharp

signals along with two weak signals (Figure 3, right). This could

again be explained in terms of the coupling of the unpaired

electron with the central phosphorus atom (A_P=27 G, ³¹P, I =

1/2) along with the coupling with the ¹⁸³W nucleus (I=1/2).

However, unlike in the case of molybdenum, no hyperfine

splitting was observed for 4b(syn) at room temperature. Mag-

netic susceptibility data of 4a(syn) were obtained in the tem-

perature range of 5–300 K and examples are displayed in

Figure 4. At low temperatures, an effective magnetic moment

of 1.73 m_Bwas determined, which is in accord with the low-

spin d⁵electronic configuration bearing one unpaired electron.

1

(syn,anti) could additionally be established by H NMR, ¹³C{¹H}

NMR, CꢀH correlation, and ¹³C DEPT experiments, and its com-

position was revealed by elemental analysis.

When 3b(syn,anti) was reacted with excess of zinc in CH₃CN

at room temperature, the ³¹P{¹H} NMR spectra in CD₃CN

showed two intense new signals at d=52.9 and 86.8 ppm

(with tungsten satellites) appearing in a 2:1 ratio assigned to

the major isomer 5b(syn), along with two weak signals at d=

48 and 76 ppm (2:1 ratio) assignable to the minor isomer

5b(anti). However, the purity of the isomeric mixture of 5b-

(syn,anti) was 90% according to the ³¹P{¹H} NMR spectra in ace-

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12644

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

pearing in a 2:1 ratio. The doublet signal of the P_terminalatoms

of the minor isomer 6a(anti) overlaps with the signal of the

major isomer 6a(syn). This was supported by the ratio of the

three signals (doublet:triplet:triplet) of 6:2:1 in the ³¹P{¹H} NMR

spectra revealing the presence of 70% of the major isomer

6a(syn) and 30% of the minor isomer 6a(anti). The ¹H NMR

spectra of 6a(syn,anti) also confirmed the presence of isomers

displaying several doubled signals in the region of the methyl,

methylene, and methyne protons; these overlapped partially

and could not be easily unravelled. The multiplet at d=

1.31 ppm was assigned to the protons of coordinated ethylene

of 6a(syn,anti) as confirmed also from C/H correlation spectra

and 1D NOE experiments. In the ¹³C{¹H} NMR spectra a singlet

resonance at d=32.7 ppm was assigned to the C_ethyleneatoms

of the major isomer 6a(syn), while a low intensity signal at d=

32.6 ppm was attributed to the same resonance of the minor

isomer 6a(anti).

Figure 5. Molecular structures of [Mo(NO)Cl(NCMe)(mer-etpⁱp)] (5a(syn); left)

and [Mo(NO)Cl(h²-ethylene)(mer-etpⁱp)] 6a(syn); right) with ellipsoids set at

30% and 50% probability, respectively.^[24]The terminal isopropyl substitu-

ents of both and the toluene solvate molecule of 5a(syn) are omitted for

clarity. Selected bond distances [ꢁ] and angles [8] for 5a(syn): Mo1–N1A

1.773(7), Mo1–P1 2.4899(13), Mo1–N2 2.195(4), N1A–O1A 1.232(8); N1A-

Mo1-Cl1A 177.40(2), P1-Mo1-P3 158.57(4), N1A-Mo1-P2 99.3(2). Selected

bond distances [ꢁ] and angles [8] for 6a(syn): Mo1–N1A 1.810(4), Mo1–P1

2.5597(7), Mo1–C1 2.256(2), Mo1–C2 2.275(2), C1–C2 1.401(4), N1A–O1A

1.220(4); N1A-Mo1-Cl1A 177.81(13), P1-Mo1-P3 154.99(2), C1-Mo1-C2

72.74(15).

The ³¹P{¹H} NMR spectra of the 6b(syn,anti) mixture showed

2

two distinct sets of doublets at d=39.8 (d, J_PP=7.5 Hz, PiPr; d,

1

2

satellites, J_PW=231 Hz, 6b(syn)), 41 ppm (d, J_PP=8.2 Hz, PiPr;

1

d, satellites, J_PW=227.4 Hz, 6b(anti)) for the terminal phospho-

rus atoms and two sets of triplets at d=74.3 (t, J_PP=7.5 Hz,

2

1

2

PPh, d, satellites, J_PW=207.2 Hz, 6b(syn)) and 57 ppm (t, J_PP=

8.2 Hz, PPh, d, satellites, J_PW=195 Hz, 6b(anti)) for the central

1

tonitrile solution. A weak signal at d=61 ppm was anticipated

to account for the residual 10% of impurity, which could as

yet not be attributed to a specific compound and has there-

fore remained unidentified in structure. The IR spectra of 5b-

(syn,anti) revealed in the range of 1500–2000 cm^ꢀ1one band

assignable to the n(NO) vibration at 1566 cm^ꢀ1. Further the for-

mation of the 5b(syn,anti) was confirmed by ¹³C{¹H} and CH

correlation spectra. When acetonitrile solutions of 5b(syn,anti)

were evaporated and dissolved again in THF, several other sig-

nals appeared in the ³¹P{¹H} NMR spectra. This was interpreted

in terms of the additional presence of acetonitrile-exchanged

compounds, which partly could have been replaced by THF or

N₂or that a dinuclear chloro-bridged species had formed. The

presence of admixtures of several species also in the solid

state of 5b(syn,anti) is assumed to be the cause for a non-satis-

factory elemental analysis.

phosphorus atoms, confirming the presence of the isomers of

1

6b(syn,anti) in a 3:1 ratio. In the H NMR spectra, the superim-

posed multiplet resonance at d=1.45 ppm was assigned to

the ethylene hydrogen atoms of both the isomers 6b(syn,anti)

and in the ¹³C{¹H} NMR spectrum, a singlet resonance at d=

30 ppm was assigned to the C_ethyleneatoms of the 6b(syn)

isomer. Another very weak signal at d=30.1 ppm could be at-

tributed to the C_ethyleneatoms of 6b(anti). The appearance of

only one ¹³C{¹H} ethylene signal for each isomer points to free

rotation of the ethylene ligand around the W–(ethylene) axis.

In the IR spectra, a strong n(NO) band could be observed at

1531 cm^ꢀ1for both the syn and anti isomer. 6a,b(syn,anti) are

considerably soluble in THF, toluene, and benzene. These com-

plexes could also be prepared in quantitative yields by the ad-

dition of 1 bar of ethylene to THF solutions of 5a,b(syn,anti) at

room temperature (Scheme 2).

Single crystals suitable for X-ray diffraction studies of the iso-

meric mixtures of 6a,b(syn,anti) were obtained from a concen-

trated toluene solution at room temperature after several days.

The crystal structures analyses of 6a,b(syn,anti) revealed

pseudo-octahedral geometries showing an h²binding mode

for the ethylene ligand (Figure 5; for 6b(syn,anti), see the Sup-

porting Information). The trans NO and Cl ligands of the 6a,b-

(syn,anti) complexes were found to be disordered in a ratio of

0.65:0.35 and 0.78:0.22, respectively, supporting the presence

of isomeric mixtures. The C1–C2 bond distance of the ethylene

molecule was found to be 1.401(4) ꢁ for 6a(syn) and 1.419(6) ꢁ

for 6b(syn), indicating substantial p back-bonding from the

metal centre.

Preparation of the isomeric complexes [M(NO)Cl(h²-ethyl-

ene)(mer-etpⁱp)] (M=Mo, 6a(syn,anti); W, 6b(syn,anti))

Treatment of the isomeric mixtures of 3a,b(syn,anti) with

5 equiv of 1% Na/Hg at room temperature resulted in the

presence of 1 bar ethylene in THF in the formation of isomeric

mixtures of the ethylene complexes [M(NO)Cl(h²-ethylene)(mer-

etpⁱp)] (M=Mo, 6a(syn,anti); M=W, 6b(syn,anti)) in 87% and

92% yields, respectively. The IR spectra of 6a,b(syn,anti) dis-

played sharp bands at 1547 and 1531 cm^ꢀ1, respectively, attrib-

uted to the n(NO) vibrations. The ³¹P{¹H} NMR spectra of 6a-

2

(syn,anti) revealed one doublet at d=58.4 ppm (d, J_PP=

17.2 Hz, iPr₂P) assigned to the P_terminalatoms of the mer-etpⁱp

ligand of both isomers 6a(syn,anti). However, for the P_internal

2

atom two triplets were observed at d=81.3 (t, J_PP=18 Hz,

2

PPh, 6a(syn)) and 66 ppm (t, J_PP=18.7 Hz, PPh, 6a(anti)) ap-

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12645

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Preparation of the isomeric ethylene hydride complexes

[M(NO)H(h²-ethylene)(mer-etpⁱp)], (M=Mo, 7a(syn,anti); W,

7b(syn,anti))

ene protons could be observed in the region of d=2.3–

2.4 ppm overlapping with several doubled multiplets of

methyl, methylene, and methyne protons. Two distinct quar-

tets were assigned to the H_Wresonances in the upfield chemi-

2

Reaction of the isomeric mixture of 6a(syn,anti) with NaBHEt₃

at 758C for 30 min produced the light yellow isomeric ethylene

hydride complexes [Mo(NO)H(h²-ethylene)(mer-etpⁱp)] 7a-

(syn,anti) in 85% yield. 7a(syn,anti) could also be prepared at

room temperature, however, longer reaction times were re-

quired (at least 24 h). In the IR spectra, n(NO) bands of 7a-

(syn,anti) were observed at 1571 cm^ꢀ1, indicating coinciding

absorption for the isomers. A n(MH) vibration could not be as-

signed, which was presumably due to a too low intensity.^[20]

The ³¹P{¹H} NMR spectra of 7a(syn,anti) revealed two conspicu-

cal shift range at d=ꢀ1.6 (q, J_PH=34.3 Hz, WH, 7b(anti)) and

at ꢀ4.95 ppm (q, ²J_PH=51.9 Hz, WH, 7b(syn)). In the ¹³C{¹H}

NMR spectrum, a singlet resonance at d=30.5 ppm was attri-

buted to the C_ethyleneatoms of 7b(syn), but the corresponding

signals of 7b(anti) could not be found either owing to its very

low concentration in solution or its accidental synchronicity

with the signal of 7b(syn). The IR spectra exhibited a band at

1622 cm^ꢀ1that was assigned to the n(NO) vibrations of

7b(syn) and 7b(anti). The mixture of isomers was also found to

be soluble in THF, toluene, benzene, but insoluble in pentane.

The composition of 7a(syn,anti) was established by elemental

analysis. We also attempted complete conversion of 7b(anti)

into 7b(syn) at 1208C, which could be achieved applying a re-

action time of 2 h, as seen by the ³¹P{¹H} NMR spectra.

2

ous sets of doublets at d=89.8 (d, J_PP=17.6 Hz, iPr₂P) and

2

79.5 ppm (d, J_PP=21.8 Hz, iPr₂P) for the P_terminalatoms and two

2

sets of triplets at d=96.1 (t, J_PP=17.6 Hz, PPh) and 92.5 ppm

2

(t, J_PP=21.6 Hz, PPh) for the P_centralphosphorus atom also

speak for the presence of an isomeric 7a(syn,anti) mixture (9:1

1

ratio). In the H NMR spectra, two quartet signals appeared at

Hydrogenations of olefins using isomeric mixtures of

[M(NO)H(h²-ethylene)(mer-etpⁱp)] complexes as catalysts in

the absence or presence of B(C₆F₅)₃/hydrosilane co-catalytic

mixture (M=Mo, 7a(syn,anti); W; 7b(syn,anti))

2

d=ꢀ3.9 (q, J_PH=46.4 Hz,) and at ꢀ6.0 ppm (q, J_PH=65.2 Hz)

with distinctly different P,H coupling constants to the metal-at-

1

tached hydride ligands of 7a (anti,syn). In the H NMR spectra

the ethylene protons of 7a(syn) could be assigned as multi-

plets in the range of d=2.19–2.12 ppm, which could be fur-

ther substantiated by C/H correlation spectra and 1D-NOE ex-

periments. A sharp singlet resonance at d=32.7 ppm in the

¹³C{¹H} NMR spectra was attributed to the C_ethyleneatoms of the

attached ethylene moiety of 7a(syn). Furthermore, the com-

plex was characterised by P/H correlation spectrum. Owing to

the very low content of the minor isomer 7a(anti) and owing

to the presence of several other resonances of very low inten-

sities, the signals of the ethylene protons and carbon atoms of

7a(anti) could not be assigned. It should be mentioned at this

point that the minor isomer 7a(anti) could be converted into

7a(syn) at a reasonable speed at temperatures above 758C, as

supported by the ³¹P{¹H} NMR spectra. The purity of the 7a-

(syn,anti) mixture was established by a satisfactory elemental

analysis. 7a(syn,anti) was found to be soluble in THF, toluene,

and benzene and sparingly soluble in pentane.

In previous work of our group applying the large-bite-angle

rhenium nitrosyl DPEphos complexes,^[13]we realised that cis

disposition of the ethylene and the hydride ligand would be

a prerequisite for suitable catalysts or catalyst precursors of

Wilkinson- or Osborn-type olefin hydrogenations, as insertion

of the olefin ligand into the metal hydride bond was expected

to be favoured by the close vicinity of these ligands. The iso-

meric mixtures of 7a,b(syn,anti) show this feature of the ligand

sphere and therefore their catalytic performance in olefin hy-

drogenations seemed feasible.

The catalytic hydrogenation studies were tested with the

prototypical 1-hexene substrate at room temperature under

60 bar of H₂and with a loading of 0.23 mol% of 7a(syn,anti).

But hydrogenation catalysis could not be initiated with these

conditions, which was apparently due to a too low tempera-

ture (Table 1, entry 1).

In a similar way as for 6a(syn,anti), when the isomeric mix-

ture of 6b(syn,anti) was treated with NaBHEt₃and heated to

908C, the ³¹P{¹H} NMR spectra exhibited several unidentified

signals along with signals of the isomeric hydride complexes

of [W(NO)H(h²-ethylene)(mer-etpⁱp)] 7b(syn,anti). The mixture

of isomers could not be isolated in pure form owing to the

presence of as-yet unidentified products. However, the 7b-

(syn,anti) mixture could eventually be prepared from 6b-

(syn,anti) using the LiBH₄/Et₃N reagent at 808C furnishing an

excellent yield (74%). In the ³¹P{¹H} NMR spectra two sets of

doublets and triplets were found for the P_internaland P_terminal

atoms of 7b(syn,anti) appearing in a 4:1 ratio. They were as-

7a(syn,anti) was of a too robust nature, preventing ligand

exchange reactivity or ethylene insertion into the metal hy-

dride bond at temperatures below 1008C. Hydrogenation cat-

alysis applying 0.23 mol% 7a(syn,anti) could eventually be

turned on at 1008C and under 60 bar H₂, but with low activi-

ties reaching 39% conversion into hexane and a TOF of 12 h^ꢀ1

after 14 h (Table 1, entry 2). The catalytic performance of 7a-

(syn,anti) however improved considerably when the tempera-

ture was increased to 1408C under 60 bar of H₂, which fur-

nished a TOF of 910 h^ꢀ1and a conversion of 21% after initial

30 min of reaction time using a loading of 0.045 mol% 7a-

(syn,anti) (Table 1, entry 3). Still the catalyst’s performance was

considered unsatisfactory, and based on earlier experience of

our group,^[15]we rationalised that the hydrosilane/B(C₆F₅)₃re-

agent could be employed as a co-catalytic system, expecting

in situ generation of silylium ions^[16,17]with subsequent attach-

ment to the O_NOatom. The resulting silylium-derivatised NOSi

2

signed as follows: d=76.3 (t, J_PP=6.6 Hz, PPh; d, satellites,

2

¹J_PW=188 Hz, 7b(syn)), 73.1 (t, J_PP=10.5 Hz, PPh; d, satellites,

2

¹J_PW=231.8 Hz, 7b(anti)), 63.4 (d, J_PP=6.6 Hz, PiPr; d, satellites,

¹J_PW=187 Hz, 7b(syn)), 52.4 ppm (d, ²J_PP=10.5 Hz, PiPr; d, satel-

1

lites, J_PW=231.8 Hz, 7b(anti)). In the H NMR spectra, the ethyl-

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12646

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Table 1. Optimisation reactions for the catalytic hydrogenation of olefins using isomeric mixtures of the catalysts 7a,b(syn,anti) in the absence or presence

of B(C₆F₅)₃, R_nSiH_4ꢀn, or a B(C₆F₅)₃/R_nSiH_4ꢀnmixture.

Entry^[a]

Cat./[mol%]

Substrates [mL]

Co-catalyst

Temp.

[8C]

H₂

[bar]

Time

[h]

TOF

[h^ꢀ1

Conv.

[%]^[c]

[b]

]

1

2

3

4

5

6

7

8

7a(syn,anti)/0.23

7a(syn,anti)/0.045

7a(syn,anti)/0.023

–

1-hexene (0.2)

1-hexene (1)

1-hexene (2)

1-hexene (1)

1-hexene (2)

1-hexene (1)

1-hexene (2)

1-hexene (1)

1-hexene (2)

1-hexene (1)

1-hexene (0.2)

1-hexene (3)

1-hexene (0.2)

1-hexene (2)

–

25

100

140

25

60

20

60

22

14

–

12

910

3200

–

39

21

36

–

0.5

2

0.5

20

0.5

11

Me₂PhSiH/B(C₆F₅)₃

Et₃SiH/B(C₆F₅)₃

iPr₃SiH/B(C₆F₅)₃

Ph₂MeSiH/B(C₆F₅)₃

Ph₂SiH₂/B(C₆F₅)₃

MePhSiH₂/B(C₆F₅)₃

PhSiH₃/B(C₆F₅)₃

B(C₆F₅)₃

7a(syn,anti)/0.023

7a(syn,anti)/0.045

7a(syn,anti)/0.023

7a(syn,anti)/0.045

7b(syn,anti)/0.020

7b(syn,anti)/0.04

7b(syn,anti)/0.20

7b(syn,anti)/0.013

7b(syn,anti)/0.20

7b(syn,anti)/0.02

7b(syn,anti)/0.020

7b(syn,anti)/0.02

5250

4600

3370

2320

2270

1380

1990

2220

15

2100

910

8200

100

75

51

38

26

25

16

23

50

14

24

21

80

2

9

10

11

12

13

14

15

16

17

18

19

20

21^[d]

22

23

24

25

26

Et₃SiH

Et₃SiH/B(C₆F₅)₃

–

140

120

140

25

–

Et₃SiH/B(C₆F₅)₃

Me₂PhSiH/B(C₆F₅)₃

iPr₃SiH/B(C₆F₅)₃

B(C₆F₅)₃

160

25

0.5

11

5080

35

33

75

39

28

35

21

04

140

100

140

0.5

3900

2800

3500

2100

400

Et₃SiH/B(C₆F₅)₃

Et₃SiH

[a] Unless otherwise stated, 2 mg of the catalyst 7a,b(syn,anti), 5 equiv of B(C₆F₅)₃, and 5 equiv of silane, volume (mL) of olefins according to Table 1, 0.8 mL

toluene were used. [b] TOFs were determined after initial 30 min of reaction by ¹H NMR using THF as internal standard. [c] Conversions were determined

by ¹H NMR using THF as internal standard on the basis of substrate consumption. [d] TOF was determined after 11 h.

species were thought to facilitate bending at the N atom with

reversible opening of a coordination site crucial for the catalyt-

ic reaction course. A drastic boost of the catalytic performance

was seen in the case of related rhenium nitrosyl systems apply-

ing the silane/B(C₆F₅)₃reagent as a co-catalyst,^[15]particularly in

the presence of the tertiary Me₂PhSiH silane. Indeed, when

0.023 mol% of the catalyst 7a(syn,anti) was mixed with 5 equiv

of the Me₂PhSiH/B(C₆F₅)₃reagent to accomplish hydrogenation

of 1-hexene under 60 bar of H₂and 1408C in toluene, the

¹H NMR test exhibited after initial 30 min 36% conversion cor-

responding to a TOF of 3200 h^ꢀ1(Table 1, entry 4). A control

experiment applying Me₂PhSiH/B(C₆F₅)₃under 60 bar of H₂and

1408C in toluene in the hydrogenation of 1-hexene did not

afford any hydrogenated hexane, as supported by the ¹³C{¹H}

NMR spectra (Table 1, entry 5). Thus various other tertiary si-

lanes, such as Et₃SiH, iPr₃SiH, Ph₂MeSiH, secondary Ph₂SiH₂,

PhMeSiH₂, and the primary PhSiH₃silanes were probed in com-

bination with the 7a(syn,anti)/B(C₆F₅)₃mixture. The results are

listed in Table 1, entries 6–11. This silane screening revealed

that the 7a(syn,anti)/B(C₆F₅)₃/Et₃SiH system (0.023 mol%)

showed the best catalytic performance under 60 bar of H₂and

1408C in toluene exhibiting after 30 min 59% conversion cor-

responding to a TOF of 5250 h^ꢀ1(Table 2, entry 1), although

after 2 h only 75% conversion could be observed. Several

other optimisation experiments were also carried out applying

7a(syn,anti)/B(C₆F₅)₃/Et₃SiH co-catalytic system and are listed in

Table 1 (entries 12–16). Under “neat” conditions, when the hy-

drogenation of 1-hexene was carried out without a solvent, an

initial conversion of 43% corresponding to a TOF of 3730 h^ꢀ1

could be achieved based on a loading of 0.023 mol% 7a-

(syn,anti) in the presence of the B(C₆F₅)₃/Et₃SiH reagents at

1408C under 60 bar H₂(Table 2, entry 3). Congener tungsten

complexes are often more stable than the corresponding mo-

lybdenum compounds due to somewhat stronger bonds to

the metal centre. Under comparable conditions, the tungsten

derived 7b(syn,anti)/B(C₆F₅)₃/Et₃SiH system was found to be

catalytically even more active, and showed higher conversions

than the 7a(syn,anti)/B(C₆F₅)₃/Et₃SiH molybdenum system.

When 0.020 mol% of 7b(syn,anti) and the B(C₆F₅)₃/Et₃SiH mix-

ture were loaded to a toluene solution of 1-hexene (2 mL) at

1408C under 60 bar H₂, the catalytic mixture revealed by

¹H NMR after initial 30 min a conversion of 80% which corre-

sponded to a TOF of 8200 h^ꢀ1, and 97% conversion after 1 h

could be achieved (Table 2, entry 4). The hydrogenated prod-

1

uct hexane was identified by H and ¹³C{¹H} NMR spectroscopy.

On the other hand, when the same reaction was carried out in

the absence of the B(C₆F₅)₃/Et₃SiH mixture, only 2% conversion

could be observed at 1408C under 60 bar H₂using

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12647

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Table 2. Catalytic hydrogenation of olefins using an isomeric mixture of 7a,b(syn,anti) as catalyst in the presence a B(C₆F₅)₃/Et₃SiH mixture.

Entry^[a]

Cat./[mol%]

Substrates [mL]

Co-catalyst

Temp.

[8C]

H₂

[bar]

Time

[h]

TOF

[h^ꢀ1

]

Conv.

[%]^[c]

[b]

1

2

3^[d]

4

5

6

7

8

9^[e]

10^[e]

11^[e]

12^[e]

13

14

15

16

17

18

19

20

21

22

7a(syn,anti)/0.023

7a(syn,anti)/0.045

7a(syn,anti)/0.023

7b(syn,anti)/0.020

7a(syn,anti)/0.028

7b(syn,anti)/0.024

7a(syn,anti)/0.041

7b(syn,anti)/0.037

7a(syn,anti)/0.037

7b(syn,anti)/0.031

7a(syn,anti)/0.047

7b(syn,anti)/0.040

7a(syn,anti)/0.047

7a(syn,anti)/0.053

7b(syn,anti)/0.046

7a(syn,anti)/0.09

7b(syn,anti)/0.08

7b(syn,anti)/0.020

7b(syn,anti)/0.013

1-hexene (2)

1-hexene (1)

1-hexene (2)

1-octene (2)

styrene (1)

cyclohexene (1)

a-methyl styrene (1)

cyclooctene (1)

1,7-octadiene (1)

allylcyclohexane (0.6)

1-hexene (2)

Et₃SiH/B(C₆F₅)₃

140

60

50

40

30

20

60

2

0.5

1

2

1

5250

>4250

3730

8200

3500

6900

2080

2060

200

75

>98

43

97

75

93

84

100

96

60

100

55

65

85

80

100

72

62

41

2

<5

<10

<12

16

<8

3

2

8

3

<3

0.5

480

210

720

1360

1060

2220

1290

1270

7180

6150

4100

2870

7900

1-hexene (2)

1-hexene (3)

29

79

[a] Unless otherwise stated, 2 mg of the catalyst 7a,b(syn,anti), 5 equiv of B(C₆F₅)₃, and 5 equiv of Et₃SiH with respect to the catalyst, volume (mL) of olefins

according to Table 2, 0.8 mL toluene were used. [b] TOFs were determined after initial 30 min of reaction by ¹H NMR using THF as internal standard.

[c] Conversions were determined by ¹H NMR on the basis of substrate consumption using THF as internal standard. [d] The reaction was carried out with-

out solvents. [e] TOFs were determined after 1 h.

a 0.040 mol% loading of 7b(syn,anti) (Table 1, entry 18), and

furthermore we could not observe any hydrogenated product

at room temperature under 60 bar H₂in the absence of the

co-catalytic system using an even higher loading of 0.20 mol%

of 7b(syn,anti) and longer reaction times (Table 1, entry 19). An

attempt to increase the catalytic activity of the 7b(syn,anti)/

Et₃SiH/B(C₆F₅)₃system by raising the temperature to 1608C

showed no improvement in the hydrogenation of 1-hexene re-

vealing an initial TOF of 5080 h^ꢀ1with a loading of 0.013 mol%

7b(syn,anti) in the presence of the Et₃SiH/B(C₆F₅)₃mixture

under 60 bar H₂(Table 1, entry 20). We assume that 1408C is

a kind of threshold temperature; exceeding this temperature

results in a too rapid decomposition of the catalyst.

entry 6). The hydrogenation of styrene afforded 84% conver-

sion and an initial TOF of 2080 h^ꢀ1after 2 h of reaction time

(Table 2, entry 7) applying the 0.041 mol% 7a(syn,anti)/Et₃SiH/

B(C₆F₅)₃co-catalytic system. Styrene could be fully converted

into ethylbenzene in less than 5 h using 0.037 mol% 7b-

(syn,anti) and 5 equiv of Et₃SiH/B(C₆F₅)₃combination achieving

an initial TOF of 2060 h^ꢀ1(Table 2, entry 8). In the cases of di-

substituted olefins, such as cyclohexene and a-methylstyrene,

the hydrogenation rates were found to be more sluggish em-

ploying 7a,b(syn,anti)/Et₃SiH/B(C₆F₅)₃co-catalytic system. For

instance, the hydrogenation of cyclohexene catalysed by

0.037 mol% of 7a(syn,anti) in the presence of the B(C₆F₅)₃/

Et₃SiH mixture brought about an initial TOF of 200 h^ꢀ1, and the

reaction went to complete conversion in less than 10 h

(Table 2, entry 9). On the other hand, 7b(syn,anti)/Et₃SiH/

B(C₆F₅)₃showed an initial TOF of 480 h^ꢀ1with a conversion of

96% in less than 12 h (Table 2, entry 10). For a-methylstyrene,

only 60% conversion could be reached after 16 h and an initial

TOF of 210 h^ꢀ1using 7a(syn,anti)/Et₃SiH/B(C₆F₅)₃co-catalytic

system (Table 2, entry 11). Conversely, the 7b(syn,anti)/Et₃SiH/

B(C₆F₅)₃system furnished complete conversion of a-methylstyr-

ene in less than 8 h with an initial TOF of 720 h^ꢀ1(Table 2,

entry 12). For cyclooctene, an initial TOF of 1360 h^ꢀ1was found

with a loading of 0.047 mol% of 7a(syn,anti) and the B(C₆F₅)₃/

Et₃SiH co-catalytic mixture. However, the reaction revealed

only 55% conversion after 3 h of reaction time (Table 2,

entry 13). A loading of 0.053 mol% of the catalyst 7a(syn,anti)

Based on extensive catalytic optimisation experiments using

the 7a,b(syn,anti)/B(C₆F₅)₃/R_nSiH_4ꢀnsystems (Table 1), a series of

alkenes were then tested in hydrogenations applying the opti-

mum conditions from the 1-hexene reactions: 1408C and

60 bar H₂and the 7a,b(syn,anti)/B(C₆F₅)₃/Et₃SiH co-catalytic

system. With terminal alkenes, such as 1-octene, 75% conver-

sion could be achieved after 2 h by addition of only

0.028 mol% of 7a(syn,anti) in the presence of the Et₃SiH/

B(C₆F₅)₃reagent at 1408C and with 60 bar of H₂pressure fur-

nishing an initial TOF of 3500 h^ꢀ1(Table 2, entry 5). A loading

of 0.024 mol% of 7b(syn,anti) and the Et₃SiH/B(C₆F₅)₃co-cata-

lytic reagent showed 93% conversion of 1-octene to octane

within one hour with an initial TOF of 6900 h^ꢀ1under other-

wise similar conditions as in the experiments before (Table 2,

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12648

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

with the B(C₆F₅)₃/Et₃SiH mixture led to 65% conversion after

2 h in the hydrogenation of 1,7-octadiene, though the hydro-

genation proceeded well in the initial half an hour with a TOF

of 1060 h^ꢀ1(Table 2, entry 14). Excellent conversion of 85%

could be achieved in 1,7-octadiene hydrogenation catalysed

by 7b(syn,anti)/B(C₆F₅)₃/Et₃SiH system at 1408C under 60 bar H₂

(Table 2, entry 15) with an initial TOF of 2220 h^ꢀ1. Allylcyclohex-

ane was also found to be completely hydrogenated in less

than 3 h using 0.08 mol% of 7b(syn,anti) as the catalyst to-

gether with the Et₃SiH/B(C₆F₅)₃reagent at 1408C under 60 bar

H₂with a maximum initial TOF of 1270 h^ꢀ1(Table 2, entry 17).

But the 7a(syn,anti)/B(C₆F₅)₃/Et₃SiH co-catalytic system afforded

80% conversion of allylcyclohexane after 3 h under the afore-

mentioned conditions, showing an initial TOF of 1290 h^ꢀ1

(Table 2, entry 16).

The tungsten system was found to be more stable through-

out the catalytic cycle, reaching higher activities and full con-

versions in almost all the olefin hydrogenations, even with

considerable lower catalyst loadings (Table 2, entries 4, 6, 8, 10,

12, 15, and 17). The tungsten catalysts apparently do not dete-

riorate so fast presumably owing to stronger bonds to the

tungsten centre.

Mechanistic Investigations

We also examined the pressure dependency of the hydrogena-

tion rate. For this purpose, we carried out five hydrogenation

reactions at 1408C using 0.02 mol% of the catalyst 7b(syn,anti)

in the presence of the Et₃SiH/B(C₆F₅)₃mixture and determined

1

the initial rates as TOFs (h^ꢀ1) after 30 min by H NMR spectros-

copy (Table 2, entries 4, 18–21). A plot of the initial TOF (h^ꢀ1

)

Generally, we realised that within the initial half an hour of

these molybdenum-complex-catalysed reactions, the hydroge-

nations went well, but they started to slow down as they pro-

gressed for longer period of times (Table 2, en-

tries 1,5,7,11,13,14, and 16). For instance, in the hydrogena-

tion of 1-hexene (2 mL), a 59% conversion was achieved after

30 min with a loading of 0.023 mol% of the catalyst 7a-

(syn,anti) in the presence of the Et₃SiH/B(C₆F₅)₃reagent under

60 bar H₂at 1408C. But the reaction furnished only 75% con-

version after 2 h. This could be due to the fact that the catalyst

started to decompose at a relatively early stage of the reaction

course. Furthermore, the isomerisation pathway converting ter-

minal olefins to the thermodynamically more stable, but steri-

cally hindered internal olefins was anticipated to be a reason

for the decrease of the rates at later stages of the reaction

courses. Internal olefins bind less strongly to metal centres,

causing lower concentrations of the kinetically relevant olefin

complexes. Their formation (along with the catalyst decay)

hampers full conversion of the initial terminal olefins to the sa-

turated compounds within a given reaction time, as their hy-

drogenation kinetics are considerably slower. Indeed, the

¹H NMR spectra of the 1-hexene hydrogenation using the 7a-

(syn,anti)/B(C₆F₅)₃/Et₃SiH system did not reveal the presence of

1-hexene after 2 h (at 75% conversion of 1-hexene to hexane);

instead, multiplets appeared at d=5 ppm, which could be at-

tributed to the olefinic protons of isomerised internal alkenes.

In the hydrogenations of 1-octene, 1,7-octadiene, and allycy-

clohexane isomerisation took also place, forming internal ole-

fins (Table 2, entries 5, 14, and 16). A hydrogenation of 1-

hexene was carried out applying lower substrate loading

(1 mL) at 1408C under 60 bar H₂in the presence of Mo system

7a(syn,anti)/B(C₆F₅)₃/Et₃SiH to attain complete conversions of

the hydrogenated products. In fact, after the initial 30 min the

reaction mixture was found to consist of 98% hydrogenated

versus H₂pressure (bar) revealed

(Figure 6).

a linear dependency

Figure 6. Plot of initial TOF [h^ꢀ1] versus hydrogen pressure [bar]. 1-Hexene

(2 mL), 7b(syn,anti) (2 mg), of B(C₆F₅)₃(5 equiv), Et₃SiH (5 equiv), toluene

(0.8 mL), 1408C were applied as reaction conditions and the TOFs were de-

1

termined after the initial 30 min by H NMR spectroscopy using THF as inter-

nal standard.

Asymptotic bending of the curve became not visible at

higher H₂pressures, which means that H₂saturation was not

reached under the employed conditions. This also means that

the catalytic performance could in principle be still improved

at higher H₂pressures. Unfortunately we could not test this, as

our hydrogenation autoclave apparatus did not allow higher

pressures at that relatively high temperature. Then we also

checked the dependency of the rates versus substrate concen-

trations (Table 2, entries 4 and 22), and we found that the reac-

tion rates were not dependent on the concentration of the

substrate, indicating strong affinities of the olefins to the metal

centre with saturation of the catalyst at low concentrations of

the substrate. Moreover, the residues of the reaction solution

of the 7b(syn,anti)/Et₃SiH/B(C₆F₅)₃system were found to be

“partially living” after completion of a catalytic batch. A second

hydrogenation run with an additional olefin charge showed

still catalytic activity, although some decrease in activity was

noted. For example, the hydrogenation reaction of 1-hexene

using the 7b(syn,anti)/Et₃SiH/B(C₆F₅)₃co-catalytic system

1

hexane product, as revealed by the H NMR spectra (Table 2,

entry 2), leading to difficulties in initial TOF determination. This

prompted us to try higher substrate loadings. Therefore it

seemed plausible to assume that the full conversions of other

olefins also catalysed by Mo system in combination with

B(C₆F₅)₃/Et₃SiH co-catalytic mixture could be achieved only by

increasing the catalyst loading or decreasing the substrate

loading under the employed conditions.

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12649

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

(1408C and 60 bar H₂) was stopped after 1 h (97% conversion

with 3% isomerised 2-hexene) and then 1 mL of 1-hexene was

added (without work-up of the reaction mixture). The catalytic

reaction re-started, showing 50% conversion of the added

1-hexene after 20 min of reaction time.

Deuterium kinetic isotope effect studies (DKIE)

Deuterium isotope kinetics (DKIE) were studied using 20 bar of

H₂or D₂in the hydrogenation of styrene catalysed by the 7a-

(syn,anti)/B(C₆F₅)₃/Et₃SiH system at 1408C. Addition of

0.041 mol% of the catalyst 7a(syn,anti) and the B(C₆F₅)₃/Et₃SiH

co-catalytic mixture (1:5 molar ratio) to a solution of styrene in

toluene revealed a TOF of 820 h^ꢀ1under 20 bar H₂. On the

other hand, under the same conditions when the reaction was

carried out under 20 bar D₂, a TOF of 1500 h^ꢀ1revealed an in-

verse DKIE of k_H/k_D=0.55. We also studied the DKIE of the

tungsten catalyst 7b(syn,anti)/B(C₆F₅)₃/Et₃SiH system. When

0.044 mol% of the catalyst 7b(syn,anti) and B(C₆F₅)₃/Et₃SiH

were heated in an autoclave to 1408C under 20 bar H₂or D₂in

toluene, the initial TOF values were found to be 766 and

1170 h^ꢀ1, respectively, as indicated by GC-MS. An inverse kinet-

ic isotope effect of k_H/k_D=0.65 similar to the Mo case was at-

tained. The DKIE values of less than one may imply that HꢀH

bond splitting is not rate limiting, because this is generally ex-

pected to be accompanied by relatively high DKIE values of

more than one. Therefore we assume either a late transition

state for the reductive elimination step of the alkane^[15a,21]or

the H₂or D₂ligand addition to be rate limiting,^[8b,22]as DKIE

values of less than one or very low DKIE values of more than

one were indeed observed for this latter case.

Scheme 3. General sketch of the proposed Osborn-type catalytic cycle for

a prototypic ethylene hydrogenation with the syn/anti isomeric mixtures of

the molybdenum and tungsten complexes of type 7 and the R₃SiH/B(C₆F₅)₃

co-catalytic mixture to provide a silylium cation for NO attachment.

tial or full oxidatively added H₂ligand. Via the transition state

TS, reductive elimination of the alkane occurs to form inter-

mediate I-5, which is presumed to be in equilibrium with the

I-5’ structure by M-NOSi bending, forming a Negishi superelec-

trophile.^[18]I-5 or I-5’ can take up an olefin and close the cata-

lytic cycle.

Mechanistic proposal

All these observations would be in accord with an Osborn-type

olefin hydrogenation cycle with “olefin before H₂coordination”

driven by the 7a,b(syn,anti) catalysts in the presence of the

Et₃SiH/B(C₆F₅)₃co-catalytic system (Scheme 3).

Conclusion

Herein we presented synthetic access to several molybdenum

and tungsten nitrosyl complexes [M(NO)Cl₃(mer-etpⁱp)] (M=

Mo, 3a(syn,anti); W, 3b(syn,anti)), [M(NO)Cl₂(mer-etpⁱp)] (M=

Mo, 4a(syn); W, 4b(syn)), [M(NO)Cl(NCMe)(mer-etpⁱp)] (M=Mo,

5a(syn,anti); W, 5b(syn,anti)), and [M(NO)Cl(h²-ethylene)(mer-

etpⁱp)] (M=Mo, 6a(syn,anti); W, 6b(syn,anti)) bearing the che-

lating mer-trisphosphine (iPr₂PCH₂CH₂)₂PPh (etpⁱp) ligand. Most

of the complexes appeared as isomeric mixtures differing in

the syn or anti orientation of the NO ligand and the phenyl

ring attached to the internal phosphorus atom. Two ethylene

hydride complexes [M(NO)H(h²-ethylene)(mer-etpⁱp)] (M=Mo,

7a(syn,anti); W, 7b(syn,anti)) could be accessed and tested for

their performance in olefin hydrogenations. Various types of

olefins could be hydrogenated using the 7a,b(syn,anti)/

B(C₆F₅)₃/Et₃SiH co-catalytic systems. A ligand sphere prone for

olefin insertion into a metal hydride bond triggered an

Osborn-type hydrogenation cycle, which was further accelerat-

ed by attachment of a silylium cation to the NO ligand and in-

ducement of facile reversible bending of the NOSi moiety. An

appropriately “tuned” ligand sphere and the enforcement of

nitrosyl bending (“catalytic nitrosyl effect”) helps to create cata-

Exploiting the “catalytic nitrosyl effect”,^[14a]an in situ generat-

ed silylium cation is anticipated to initially attach to the O_NO

atom of the nitrosyl ligand, enforcing reversible bending of the

linear NOSi moiety to form the unsaturated 16 e^ꢀspecies I-1 (a

linear NOSi moiety is a three-electron donor, whereas a bent

one is a one-electron donor). Then H₂uptake by this species is

anticipated to generate the hydrogen hydride intermediate I-2.

I-2 is well-prepared for H transfer to the olefin ligand, because

the high number of ligands is expected to enforce merging of

ligands by returning back to the more stable octahedral con-

figuration. In this way, bending of the NOSi moiety is playing

a beneficial role for the given catalysis, providing a catalytically

relevant reactive site “on demand”.^[15a]The initial H₂uptake to

a not so strongly electrophilic site created by NO bending

would be in perfect agreement with the pronounced H₂pres-

sure dependency of the hydrogenations (see Figure 6). The

subsequent b-hydride transfer to the olefin generates the M-

alkyl species I-3. I-3 is suggested to relax in equilibrium to the

equilibrium of I-4 possessing a linear M-NOSi moiety and a par-

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12650

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

CH₃), 18 (m, PCH₂). IR (cm^ꢀ1, ATR): 1648 (n_NO). Anal. calcd(%) for

C₂₂H₄₁Cl₃MoNOP₃: C 41.89, H 6.55, N 2.22; found: C 41.75, H 6.30,

N 2.01.

lytically relevant vacant sites at the metal centres, which is pro-

posed to be one of the general strategies to make middle tran-

sition-metal complexes efficient catalysts and thus to prepare

“cheap metals for noble tasks.”

3b(syn,anti): Yield: 52%. ¹H NMR (500 MHz, CD₂Cl₂): d 7.9–7.5 (m,

Ph), 3.79–3.76 (m, PCH₂), 3.0–2.9 (m, PCH₂), 2.68–2.60 (m, PCH),

2.54–2.39 (m, PCH₂), 1.48–1.15 (m, CH₃). ¹³C{¹H} NMR (125.8 MHz,

CD₂Cl₂): d 133.6 (m, Ph), 133.3 (m, Ph), 132.8 (m, Ph), 129.9 (m, Ph),

29.4 (m, PCH₂), 29 (m, CH), 28 (m, CH), 22.1 (m, CH₃), 20.6 (s, CH₃),

20.1 (m, CH₃), 19.6 (s, CH₃), 17.5 (m, PCH₂). ³¹P{¹H} NMR (202.5 MHz,

Experimental Section

General: All manipulations were carried out under an atmosphere

of nitrogen using either dry glove box or Schlenk techniques. Re-

agent-grade solvents benzene, tetrahydofuran, pentane, toluene,

and diethyl ether were dried with sodium benzophenone ketyl and

distilled prior to use under a N₂atmosphere. Dichloromethane and

cyclohexane were dried over calcium hydride and distilled. Deuter-

ated solvents were dried with sodium benzophenone ([D₈]THF,

C₆D₆) and calcium hydride (CD₂Cl₂) and deoxygenated by freeze–

pump–thaw cycles before use. 1-Hexene, 1-octene, cyclohexene,

1,7-octadiene, cyclooctene, and allylcyclohexane were purified by

distillation prior to use. The [M(NO)Cl₃(NCMe)₂] complexes (M=Mo

and W)^[23]were prepared according to previously reported proce-

dures. Other reagents were purchased and used without further

purification. NMR spectra were measured with a Varian Mercury

200 spectrometer (at 200.1 MHz for ¹H, at 81.0 MHz for ³¹P), with

Varian Gemini-300 instrument (¹H at 300.1 MHz, ¹³C at 75.4 MHz),

with a Bruker-DRX 500 spectrometer (500.2 MHz for ¹H, 202.5 MHz

for ³¹P, 125.8 MHz for ¹³C) and a Bruker-DRX 400 spectrometer

(400.1 MHz for ¹H, 162.0 MHz for ³¹P, 100.6 MHz for ¹³C). All chemi-

2

CD₂Cl₂): d=55 (d, J_PP=46 Hz (syn)), 80 (t, J_PP=46 Hz, (syn)), 56 (d,

²J_PP=49 Hz, (anti)), 72 (t, ²J_PP=49 Hz, (anti)). IR (cm^ꢀ1, ATR): 1618

(n_NO). Anal. calcd(%) for C₂₂H₄₁Cl₃NOP₃W: C 36.77, H 5.75, N 1.95;

found: C 36.65, H 5.71, N 1.99.

General procedure for the preparation of [M(NO)Cl₂(mer-

etpⁱp)] (M=Mo, 4a(syn); W, 4b(syn))

[M(NO)Cl₃(mer-etpⁱp)] (0.16 mmol; M=Mo, 3a(syn,anti); W, 3b-

(syn,anti)) was added to a stirred suspension of 1% sodium amal-

gam (1 equiv, 0.16 mmol) in THF (10 mL). The reaction mixture was

kept stirring for 2 h at room temperature. The colour of the solu-

tion turned green. The final supernatant green solution was filtered

off from the mercury containing residue and evaporated to dry-

ness. The residue was washed with pentane (5 mL) and dried. Fi-

nally the product was extracted with toluene (15 mL) and concen-

trated. Recrystallisation from cold toluene/pentane solution

(10 mL/10 mL) at ꢀ308C afforded green crystals.

1

cal shifts for H and ¹³C{¹H} NMR are expressed in ppm relative to

4a(syn): Yield: 59%. IR (cm^ꢀ1, ATR): 1589 (n_NO). Anal. calcd (%) for

C₂₂H₄₁Cl₂MoNOP₃: C 44.38, H 6.94, N 2.35; found: C 44.42, H 6.95,

N 2.39.

4b(syn): Yield: 54%. IR (cm^ꢀ1, ATR): 1564 (n_NO). Anal. calcd(%) for

C₂₂H₄₁Cl₂NOP₃W: C 38.67, H 6.05, N 2.05; found: C 38.75, H 6.06,

N 2.03.

tetramethylsilane (TMS) and for ³¹P{¹H} relative to 85% H₃PO₄as an

external standard reference. Signal patterns are as follows: s sing-

let, d doublet, t triplet, q quartet, m multiplet. IR spectra were ob-

tained either ATR or KBr methods using a Bio-rad FTS-45 instru-

ment. Elemental analyses were carried out at the Department of

Chemistry of the University of Zꢂrich. X-band EPR spectra were ob-

tained using Bruker EMX Electron Spin Resonance system. Magneti-

sation measurements were carried out on a Quantum Design

SQUID magnetometer, molar magnetic susceptibility was calculat-

ed according to the equation c=(M·M_w)/(m·H), where M=experi-

mental magnetisation, M_w=molecular weight, m=sample weight,

H=magnetic field. The GC-MS spectra were recorded on a Varian

Saturn 2000 spectrometer equipped with a Varian 450-GC chroma-

tograph (Phenomenex ZB-5 ms (30 meter), Gradient 708–2708).

Preparation of [Mo(NO)(NCMe)Cl(mer-etpⁱp)], 5a(syn,anti)

5 Equiv of zinc granules were added to a solution of [Mo(NO)-

Cl₃(mer-etpⁱp)] 3a(syn,anti) (50 mg, 0.08 mmol) in MeCN (10 mL).

The resulting mixture was kept stirring for three days at room tem-

perature. After completion of the reaction, the red solution was fil-

tered off from excess zinc and evaporated to dryness. The obtained

solid was washed with pentane twice (2ꢃ3 mL) and dried in vacuo

and then extracted with THF (10 mL) to obtain a red powder in

1

75% (38 mg) yield. H NMR (500 MHz, CD₃CN): d=7.8–7.4 (m, Ph),

General procedure for the preparation of [M(NO)Cl₃(mer-

etpⁱp)] (M=Mo, 3a(syn,anti); W, 3b(syn,anti))

2.59–2.49 (m, PCH₂), 2.42–2.24 (m, PCH₂), 1.83–1.80 (m, CH), 1.65–

1.54 (m, CH), 1.36–1.06 (m, CH₃). ³¹P{¹H} NMR (162 MHz, CD₃CN):

102.5 (s, PPh, 5a(syn)), 67.7 (s, iPr₂P, 5a(syn)), 94.9 (s, PPh, 5a(anti)),

62.8 (s, iPr₂P, 5a(anti)). ¹³C{¹H} NMR (100.6 MHz, CD₃CN): d=133.8

[M(NO)Cl₃(NCMe)₂] (1.6 mmol, M=Mo, W) was dissolved in THF

(15 mL). A solution of etpⁱp ligand (2, 1.6 mmol) in THF (5 mL) was

then added into that solution. The resulting mixture turned deep

red immediately and it was kept stirring for 2 h. A solid precipita-

tion was observed along with the solution turned green. The solid

precipitate was separated from the solution and dried in vacuo.

Washing with THF (5 mL) and pentane (10 mL) followed. Finally the

solid was extracted with CH₂Cl₂(30 mL) and dried in vacuo. Recrys-

tallisation of the solid from a cold CH₂Cl₂/pentane mixture (20 mL/

20 mL) afforded the desired product in crystalline form.

1

2

(d, J_CP=11.5 Hz, C-ipso), 133 (d, J_CP=10.4 Hz C-ortho), 131.5 (s, C-

3

para), 129. (d, J_CP=8.8 Hz, C-meta), 27.9 (m, CH), 27.5 (m, CH₂), 27

(m, CH), 26.8 (m, CH₂), 26.7 (m, CH), 20.7 (m, CH₃), 20 (m, CH₃), 19.9

(CH₃), 18.7 (m, CH₃). IR (cm^ꢀ1, ATR): 1598 (n_NO). Anal. calcd(%) for

C₂₄H₄₄ClMoNO₂P₃: C 47.97, H 7.38, N 4.66; found: C 48.15, H 7.41,

N 4.57.

Preparation of [W(NO)(NCMe)Cl(mer-etpⁱp)] 5b(syn,anti)

1

3a(syn,anti): Yield: 55%. H NMR (500 MHz, CD₂Cl₂): d=8.3–7.5 (m,

Ph), 3.79–3.7 (m, PCH₂), 2.9–2.6 (m, PCH), 2.5–2.3 (m, PCH₂), 1.52–

A

solution of [W(NO)Cl₃(mer-etpⁱp)] (3b(syn,anti); 200 mg,

2

1.17 (m, CH₃). ³¹P{¹H} NMR (202.5 MHz, CD₂Cl₂): d=76 (d, J_PP

=

0.28 mmol) in CH₃CN (15 mL) was added to an excess of zinc gran-

ules (10 equiv) and kept stirring at least for three days at room

temperature. After completion of the reaction, solution was filtered

off from excess zinc and evaporated to dryness. The obtained solid

showed mixture of several products. Yield: 90% based on

2

72 Hz (syn)), 98 (t, J_PP=72 Hz (syn)), 77 (d, J_PP=76 Hz (anti)), 89 (t,

²J_PP=76 Hz (anti)). ¹³C{¹H} NMR (125.8 MHz, CD₂Cl₂): d=133.6 (m,

Ph), 132 (m, Ph), 130 (m, Ph), 129.7 (m, Ph), 30 (m, PCH₂), 29 (m,

PCH), 28 (m, PCH), 22 (s, CH₃), 20.3 (m, CH₃), 20 (m, CH₃), 19.7 (m,

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12651

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

1

³¹P{¹H} NMR in CD₃CN. H NMR (400 MHz, CD₃CN₎: d=7.86 (m, Ph),

sticky red oily product was washed with minimum amount of pen-

tane (1 mL) and dried in vacuo, resulting in a solid product. The

product was then extracted with benzene (5 mL) and then with

toluene (5 mL) and dried in vacuo. Finally it was extracted with

benzene (5 mL) again and dried. The light yellow powder of

7a(syn,anti) was obtained in 85% yield (55 mg).

7.3 (m, Ph), 2.66 (m, PCH₂), 2.27(m, PCH), 1.35–1.10 (m, CH₃).

³¹P{¹H} NMR (162 MHz, CD₃CN): 52.9 (s, PPh, 5b(syn)), 86.8 (s, iPr₂P,

5b(anti)), (48 (s, PPh, 5b(anti)). 76 (s, iPr₂P, 5b(anti)), 61 (unidenti-

fied product). IR (cm^ꢀ1

, ATR): 1566 (n_NO). Anal. calcd(%) for

C₂₄H₄₄ClN₂OP₃W: C 41.85, H 6.44, N 4.07; found: C 39.80, H 5.77,

N 4.83.

¹H NMR (500 MHz, [D₈]THF): d=7.99–7.75 (m, Ph), 7.38–7.37 (m,

Ph), 2.71–2.61 (m, PCH₂), 2.61–2.55 (m, PCH), 2.33–2.30 (m, PCH),

2.28–2.24 (m, PCH₂), 2.19–2.12 (m, ethylene CH₂), 1.32–1.23 (m,

General procedure for the preparation of [M(NO)Cl(h²-ethyl-

ene)(mer-etpⁱp)] M=Mo, 6a(syn,anti); W, 6b(syn,anti)

CH₃), 1.06–0.93 (m, CH₃), ꢀ3.9 (q, ²J_PH=46.4 Hz, MoH), ꢀ6.0 (q,

2

²J_PH=65.2 Hz, MoH). ³¹P{¹H} NMR (202 MHz, [D₈]THF): 96.1 (t, J_PP

=

2

The isomeric mixture of [M(NO)Cl₃(mer-etpⁱp)] (M=Mo, 3a-

(syn,anti); W, 3b(syn,anti); 0.185 mmol) was added to a suspension

of 1% sodium amalgam (5 equiv, 0.93 mmol) in THF (10 mL) in

a Schlenk flask. The flask was taken out from the glove box and

the nitrogen atmosphere was removed by a freeze–pump–thaw

cycle. Then the tube was filled with ethylene (1 bar) and sealed.

Stirring was continued, overnight, at room temperature to ensure

the completion of the reaction. The final supernatant solution was

filtered off from the mercury containing residue and evaporated to

dryness. The residue was washed with pentane (3 mL) and extract-

ed with toluene (10 mL). Recrystallisation from cold toluene/pen-

tane solution (10 mL/10 mL) afforded pure crystals of the isomeric

mixtures of 6a(syn,anti) and 6b(syn,anti) in excellent yields.

17.6 Hz, PPh, 7a(syn)). 89.8 (d, J_PP=17.6 Hz, iPr₂P, 7a(syn)). 92.5 (t,

²J_PP=21.6 Hz, PPh, 7a(anti)). 79.5 (d, ²J_PP=21.8 Hz, iPr₂P, 7a(anti)).

¹³C{¹H} NMR (125.8 MHz, [D₈]THF): d=135 (Ph), 131 (Ph), 130.7 (C-

para), 128.6 (Ph), 33.5 (m, PCH₂), 32.7 (s, ethylene CH₂), 30 (m, CH),

24 (m, CH), 20.6 (PCH₂), 19.5 (CH₃), 19.1 (CH₃), 18.9 (CH₃), 16.5 (CH₃).

Assignments confirmed by C/H, P/H, and 1D NOE experiments. IR

(cm^ꢀ1, ATR): 1571 (n_NO). Anal. calcd(%) for C₂₄H₄₆MoNOP₃: C 52.08,

H 8.38, N 2.53; found: C 52.42, H 8.55, N 2.29.

Preparation of [W(NO)H(h²-ethylene)(mer-etpⁱp)] 7b(syn,anti)

A

solution of [W(NO)Cl(h²-ethylene)(mer-etpⁱp)] (6b(syn,anti);

150 mg, 0.22 mmol) in Et₃N (20 mL) was mixed with LiBH₄(24 mg,

1.11 mmol) in a Schlenk flask. The mixture was kept at 808C with

constant stirring for 4 h. The reaction was monitored by

³¹P{¹H} NMR spectroscopy. After completion of the reaction, the

Schlenk flask was cooled and taken into the glove box. Without fil-

tering, the reaction mixture was dried in vacuo overnight to

remove Et₃N solvent completely. The obtained sticky oily product

was washed with minimum amount of pentane (1 mL). Then it was

extracted with benzene at least twice (2ꢃ5 mL) to remove excess

LiBH₄and then twice with toluene (2ꢃ5 mL). After drying the tolu-

ene solution and washing with pentane (1 mL), the isomeric mix-

ture of 7b(syn,anti) was obtained as yellowish red powder in 74%

yield (105 mg).

1

6a(syn,anti): Yield: 87%. H NMR (500 MHz, [D₈]THF): d=8.25–8.22

(m, Ph), 8.02–7.99 (m, Ph), 8.25–8.22 (m, Ph), 7.46–7.38 (m, Ph),

2.91–2.82 (m, PCH₂), 2.61–2.51 (m, PCH), 2.45–2.36 (m, PCH₂), 2.32–

2.27 (m, PCH), 2.02–1.99 (m, PCH₂), 1.59–1.52 (m, PCH₂), 1.37–1.10

(m, CH₃), 1.31 (ethylene CH₂overlapped with CH₃protons).

2

³¹P{¹H} NMR (162 MHz, [D₈]THF): 81.3 (t, J_PP=18 Hz, PPh, 6a(syn)),

66 (t, ²J_PP=18.7 Hz, PPh, 6a(anti)), 58.4 (d, ²J_PP=17.2 Hz, iPr₂P,

6a(syn) and 6a(anti) overlapped). ¹³C{¹H} NMR (125.8 MHz, [D₈]THF):

d=135.6 (Ph), 134.9 (Ph), 131.6 (Ph), 129 (Ph), 30.7 (s, ethylene car-

bons, 6a(syn)), 30.6 (s, ethylene carbons, 6a(anti)), 30.3 (m, CH₂),

28.7 (m, CH), 26.5 (m, CH₂), 23 (m, CH), 20.5 (CH₃), 20.2 (CH₃), 19.5

(CH₃), 18.3 (CH₃). IR (cm^ꢀ1, ATR): 1547 (n_NO). Anal. calcd(%) for

C₂₄H₄₅ClMoNOP₃: C 49.03, H 7.71, N 2.38; found: C 49.34, H 7.72,

N 2.19.

¹H NMR (500 MHz, [D₈]THF): d=8.02 (m, Ph), 7.39–7.32 (m, Ph),

2.97–2.65 (m, PCH₂), 2.4–2.3 (ethylene CH₂overlapped with CH),

2.29–2.09 (m, CH), 1.87 (m, PCH₂), 1.31–1.24 (m, CH₃), 1.14–0.9 (m,

6b(syn,anti): Yield 92%. ¹H NMR (500 MHz, [D₈]THF): d=8.20–8.17

(m, Ph), 8.05–8.01 (m, Ph), 7.46–7.44 (m, Ph), 3.02–2.89 (m, PCH₂),

2.65–2.57 (m, PCH₂), 2.42–2.32 (m, PCH), 2.24–2.13 (m, PCH), 1.58–

1.51 (m, PCH₂), 1.45 (m, ethylene CH₂), 1.38–1.27 (m, CH₃), 1.24–

2

CH₃), ꢀ1.55 (q, J_PH=34.3 Hz, WH, 7b(anti)), ꢀ4.95 (q, J_PH=51.9 Hz,

WH, 7b(syn)). ³¹P{¹H} NMR (202.5 MHz, [D₈]THF): 76.3 (t, J_PP

=

2

6.6 Hz, PPh; d, satellites, ¹J_PW=188 Hz, 7b(syn)), 73.1 (t, J_PP

2

10.5 Hz, PPh; d, satellites, ¹J_PW=231.8 Hz, 7b(anti)), 63.4 (d, J_PP

2

1.12 (m, CH₃). ³¹P{¹H} NMR (162 MHz, [D₈]THF): 74.3 (t, J_PP=7.5 Hz,

6.6 Hz, PiPr; d, satellites, ¹J_PW=187 Hz, 7b(syn)), 52.4 (d, J_PP

2

PPh; d, satellites, ¹J_PW=207.2 Hz, 6b(syn)), 39.8 (d, ²J_PP=7.5 Hz,

10.5 Hz, PiPr; d, satellites, ¹J_PW=231.8 Hz, 7b(anti)). Selected

¹³C{¹H} NMR (125.8 MHz, [D₈]THF): d=30.5 (s, ethylene carbons of

7b(syn)). Assignments further confirmed by C/H, P/H and 1D NOE

1

2

PiPr; d, satellites, J_PW=231 Hz, 6b(syn)), 57 (t, J_PP=8.2 Hz, PPh; d,

1

2

satellites, J_PW=195 Hz, 6b(anti)), 41 (d, J_PP=8.2 Hz, PiPr; d, satel-

lites, ¹J_PW=227.4 Hz, 6b(anti)). Selected ¹³C{¹H} NMR (125.8 MHz,

[D₈]THF): 30.1 (s, ethylene carbons, 6b(anti)), 30 (s, ethylene car-

bons, 6b(syn)). IR (cm^ꢀ1, ATR): 1531 (n_NO). Anal. calcd (%) for

C₂₄H₄₅ClNOP₃W: C 42.65, H 6.71, N 2.07; found: C 42.88, H 6.56,

N 2.04.

experiments. IR (cm^ꢀ1

, ATR): 1622 (n_NO). Anal. calcd(%) for

C₂₄H₄₆NOP₃W: C 44.94, H 7.23, N 2.18; found: C 45.29, H 7.30,

N 2.31.

General procedure for the catalytic alkene hydrogenations

Preparation of [Mo(NO)H(h²-ethylene)(mer-etpⁱp)]

7a(syn,anti)

A stock solution of freshly made catalysts 7a(syn,anti) or 7b-

(syn,anti) (10 mg in 1 mL of toluene) was prepared. An aliquot

(0.2 mL) of that stock solution was added to a mixture of B(C₆F₅)₃/

R₃SiH (5 equiv relative to catalyst). Then a certain volume (mL) of

alkenes according to Table 1 or Table 2 was added into that mix-

ture. Further toluene (0.8 mL) was added and the entire mixture

was transferred to a 30 mL steel autoclave equipped with a stirring

bar. The autoclave was charged with required hydrogen pressure

and kept in preheated oil bath (1408C). After appropriate reaction

time, the autoclave was immediately taken out and cooled to

room temperature. Then THF (200 mL) was added into that solution

NaHBEt₃(143 mL, 1m in THF) was added to a solution of [Mo-

(NO)Cl(h²-ethylene)(mer-etpⁱp)] (6a(syn,anti); 70 mg, 0.12 mmol) in

THF (10 mL) in a Schlenk flask. The resulting solution was taken

out from the glove box and kept for heating at 758C with constant

stirring for 30 min. The reaction was monitored by ³¹P{¹H} NMR

spectroscopy. After completion of the reaction, the Schlenk flask

was cooled and taken into the glove box. The reaction mixture

was filtered and the solvent was removed in vacuo. The obtained

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12652

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

1

[3] a) J. G. de Vries, C. J. Elsevier, The Handbook of Homogeneous Hydroge-

nation, Wiley-VCH, Weinheim, 2007; b) B. Cornilis, W. A. Hermann, Ap-

plied Homogeneous Catalysis with Organometallic Compounds, VCH,

Weinheim, 1996.

and shaken properly. The mixture was finally analysed by H NMR

immediately to identify the products and to determine the conver-

sions. The TOFs and conversions were determined on the basis of

consumption of substrate by H NMR integration using THF as in-

1

[4] a) J. A. Osborn, F. H. Jardine, J. F. Young, G. Wilkinson, J. Chem. Soc. A

1966, 12, 1711; b) R. R. Schrock, J. A. Osborn, J. Chem. Soc. Chem.

Commun. 1970, 567–568; c) R. R. Schrock, J. A. Osborn, J. Am. Chem.

Soc. 1976 , 98 , 213 4; d) C. R. Landis, J. Halpern, J. Am. Chem. Soc. 1987 ,

109 , 1746; e) H. Berke, ChemPhysChem 2010, 11, 1837–1849.

[5] a) R. H. Crabtree, The Organometallic Chemistry of the Transition Metals,

2ednd edWiley, New York, 1994; b) R. R. Schrock, J. A. Osborn, J. Am.

Chem. Soc. 1971, 93, 3089; c) R. H. Crabtree, Acc. Chem. Res. 1979 , 12 ,

331; d) W. S. Knowles, Angew. Chem. 2002 , 114 , 2096; Angew. Chem. Int.

Ed. 2002 , 41 , 1998; e) R. Noyori, Angew. Chem. 2002 , 114 , 210 8; Angew.

Chem. Int. Ed. 2002 , 41 , 2008; f) S. E. Clapham, A. Hadzovic, R. H. Morris,

Coord. Chem. Rev. 2004 , 248 , 2201.

[6] a) J.-S. Song, D. J. Szalda, R. M. Bullock, C. J. C. Lawrie, M. A. Rodkin, J. R.

Norton, Angew. Chem. 1992 , 104 , 1280; Angew. Chem. Int. Ed. Engl.

1992 , 31 , 1233; b) J.-S. Song, D. J. Szalda, R. M. Bullock, Organometallics

2001 , 20 , 3337; c) M. H. Voges, R. M. Bullock, Dalton Trans. 2002 , 759;

d) B. F. M. Kimmich, P. J. Fagan, E. Hauptman, W. J. Marshall, R. M. Bul-

lock, Organometallics 2005 , 24 , 6220.

[7] a) M. P. Magee, J. R. Norton, J. Am. Chem. Soc. 2001 , 123 , 177 8 –1 779;

b) H. Guan, M. Iimura, M. P. Magee, J. R. Norton, G. Zhu, J. Am. Chem.

Soc. 2005 , 127 , 780 5 –7 814.

[8] a) A. Dybov, O. Blacque, H. Berke, Eur. J. Inorg. Chem. 2011 , 652; b) S.

Chakraborty, O. Blacque, T. Fox, H. Berke, Chem. Asian J. 2014 , 9 , 328;

c) S. Namorado, M. A. Antunes, L. F. Veiros, J. R. Ascenso, M. T. Duarte,

A. M. Martins, Organometallics 2008 , 27 , 4589.

[9] a) T. Tatsumi, M. Shibagaki, H. Tominaga, J. Mol. Catal. 1981 , 13 , 331;

b) A. Fusi, R. Ugo, R. Psaro, J. Mol. Catal. 1982 , 16 , 217; c) T. Fuchikami,

Y. Ubukata, Y. Tanaka, Tetrahedron Lett. 1991 , 32 , 1199.

ternal standard.

General procedure for the deuterium kinetic isotope effect

experiments

In a 30 mL steel autoclave equipped with a stirring bar, molybde-

num or tungsten complex (7a(syn,anti) or 7b(syn,anti)), 2 mg from

freshly prepared stock solution (10 mg in 1 mL toluene), B(C₆F₅)₃

(5 equiv with respect to catalyst), and Et₃SiH (5 equiv with respect

to catalyst 7a(syn,anti) or 7b(syn,anti)) were dissolved in 1 mL of

styrene (in the case of 7b(syn,anti), styrene (0.8 mL) was used). Tol-

uene (0.8 mL) was added into that mixture. Then the system was

charged with 20 bar of D₂and kept stirring at 1408C temperature.

After 30 min of reaction time, the reaction mixture was cooled im-

mediately and without further purification; conversion was deter-

mined by GC analysis on the basis of substrate consumption. In

a similar way and under the same conditions, the reactions were

also carried out with 20 bar H₂to determine the conversions (by

GC-MS) and initial TOFs were calculated after 30 min of reaction

time.

General procedure for the conversion of 7a,b(syn,anti)

isomeric mixture into 7a,b(syn)

The molybdenum or tungsten isomeric complexes 7a(syn,anti) or

7b(syn,anti) (10 mg) were dissolved in THF (0.5 mL) in a J. Young

tap NMR tube. Then the NMR tube was taken out from the glove

box and kept in a pre-heated oil bath (808C for molybdenum 7a-

(syn,anti) or 1208C for tungsten 7b(syn,anti)). The reaction was

monitored by ³¹P{¹H} NMR spectroscopy. For the isomeric mixture

of 7a(syn,anti), conversion of the 7a(anti) isomer to the 7a(syn)

isomer could be observed, and after 10 min 7a(syn) was the only

product as revealed by ³¹P{¹H} NMR spectra. On the other hand,

when 7b(syn,anti) isomeric mixture was heated at 1208C, the

7b(anti) isomer converted into 7b(syn) more sluggishly and com-

plete conversion of 7b(anti) into 7b(syn) could be observed only

after 2 h as indicated by the ³¹P{¹H} NMR spectra.

[10] a) R. M. Bullock, B. J. Rappoli, J. Chem. Soc. Chem. Commun. 1989, 1447;

b) R. M. Bullock, J.-S. Song, J. Am. Chem. Soc. 1994 , 116 , 8602; c) L. Luan,

J.-S. Song, R. M. Bullock, J. Org. Chem. 1995 , 60 , 7170.

[11] P. J. Baricelli, L. G. Melean, S. Ricardes, V. Guanipa, M. Rodriguez, C.

Romero, A. J. Pardey, S. Moya, M. Rosales, J. Organomet. Chem. 2009 ,

694 , 3381.

[12] a) Y. Jiang, J. Hess, T. Fox, H. Berke, J. Am. Chem. Soc. 2010 , 132 , 18233;

b) A. Choualeb, E. Maccaroni, O. Blacque, H. W. Schmalle, H. Berke, Orga-

nometallics 2008 , 27 , 347 4; c) H. Dong, H. Berke, Adv. Synth. Catal. 2009 ,

351 , 1783; d) S. Chakraborty, O. Blacque, T. Fox, H. Berke, ACS Catal.

2013 , 3 , 2208.

[13] B. Dudle, K. Rajesh, O. Blacque, H. Berke, J. Am. Chem. Soc. 2011 , 133 ,

8168.

[14] a) Y. Jiang, H. Berke, Nitrosyl Complexes in Homogeneous Catalysis Struct.

Bond. 2014, 153, 167–228; b) B. J. Coe, S. J. Glenwright, Coord. Chem.

Rev. 2000 , 203 , 5; c) J. D. Atwood, T. L. Brown, J. Am. Chem. Soc. 1976 ,

98 , 3160; d) A. Kovacs, G. Frenking, Organometallics 2001 , 20 , 2510.

[15] a) Y. Jiang, B. Schirmer, O. Blacque, T. Fox, S. Grimme, H. Berke, J. Am.

Chem. Soc. 2013 , 135 , 4088; b) Y. Jiang, W. Huang, H. W. Schmalle, O.

Blacque, T. Fox, H. Berke, Organometallics 2013 , 32 , 7043.

[16] a) S. Rendler, M. Oestreich, Angew. Chem. 2008 , 120 , 6086; Angew.

Chem. Int. Ed. 2008 , 47 , 5997; b) D. T. Hog, M. Oestreich, Eur. J. Org.

Chem. 2009 , 5047; c) H. F. T. Klare, M. Oestreich, J.-i. Ito, H. Nishiyama, Y.

Ohki, K. Tatsumi, J. Am. Chem. Soc. 2011 , 133 , 3312; d) M. Mewald, R.

Froehlich, M. Oestreich, Chem. Eur. J. 2011 , 17 , 940 6; e) K. Mꢂther, R.

Froehlich, C. Mueck Lichtenfeld, S. Grimme, M. Oestreich, J. Am. Chem.

Soc. 2011, 133, 12442.

Acknowledgements

Financial support from the Swiss National Science Foundation,

Lanxess AG, and the University of Zꢂrich are gratefully

acknowledged.

Keywords: hydrogenation · isomers · molybdenum · olefins ·

tungsten

[17] a) J. M. Blackwell, E. R. Sonmor, T. Scoccitti, W. E. Piers, Org. Lett. 2000 , 2 ,

3921; b) D. J. Parks, J. M. Blackwell, W. E. Piers, J. Org. Chem. 2000 , 65 ,

3090; c) A. Berkefeld, W. E. Piers, M. Parvez, J. Am. Chem. Soc. 2010 , 132 ,

10660.

[1] a) R. M. Bullock, Handbook of Homogeneous Hydrogenation, Wiley VCH,

Weinheim, 2007; b) R. M. Bullock, Catalysis without precious metals,

Wiley-VCH, Weinheim, 2010; c) R. M. Bullock, Chem. Eur. J. 2004 , 10 ,

2366.

[2] a) R. M. Bullock, M. H. Voges, J. Am. Chem. Soc. 2000 , 122 , 12594;

b) R. M. Bullock, Science 2013, 342, 1054; c) K. Junge, K. Schrçder, M.

Beller, Chem. Commun. 2011, 47, 4849; d) C. P. Casey, H. Guan, J. Am.

Chem. Soc. 2007 , 129 , 581 6; e) C. P. Casey, H. Guan, J. Am. Chem. Soc.

2009 , 131 , 2499; f) R. Langer, G. Leitus, Y. Ben-David, D. Milstein, Angew.

Chem. 2011 , 123 , 216 8; Angew. Chem. Int. Ed. 2011 , 50 , 2120; g) C. Sui-

Seng, F. Freutel, A. J. Lough, R. H. Morris, Angew. Chem. 2008 , 120 , 954;

Angew. Chem. Int. Ed. 2008 , 47 , 940.

[18] E. Negishi, Chem. Eur. J. 1999 , 5 , 411.

[19] E. Carmona, E. Gutierrezpuebla, A. Monge, P. J. Perez, L. J. Sanchez,

Inorg. Chem. 1989, 28, 2120.

[20] a) F. Liang, H. Jacobsen, H. W. Schmalle, T. Fox, H. Berke, Organometallics

2000 , 19 , 1950; b) F. Liang, H. W. Schmalle, T. Fox, H. Berke, Organome-

tallics 2003 , 22 , 3382.

[21] a) G. Parkin, Acc. Chem. Res. 2009 , 42 , 315; b) M. Zimmer-De Iuliis, R. H.

Morris, J. Am. Chem. Soc. 2009 , 131 , 1126 3; c) W. D. Jones, Acc. Chem.

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12653

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper

Res. 2003 , 36 , 14 0; d) G. L. Gould, D. M. Heinekey, J. Am. Chem. Soc.

1989 , 111 , 5502.

[22] B. R. Bender, G. J. Kubas, L. H. Jones, B. I. Swanson, J. Eckert, K. B. Capps,

C. D. Hoff, J. Am. Chem. Soc. 1997 , 119 , 9179.

this paper. These data can be obtained free of charge from The Cam-

bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_re-

quest/cif.

[23] L. Bencze, J. Kohan, Inorg. Chim. Acta 1982 , 65 , L17.

[24] CCDC 956868 (3a(anti)), 956869 (3a(syn), 956870 (3b(syn)), 956871

(4a(syn)), 956872 (4b(syn)) 956873 (5a(syn)) 956874 (6a(syn)), and

956875 (6b(syn)) contain the supplementary crystallographic data for

Received: March 23, 2014

Revised: May 20, 2014

Published online on August 14, 2014

Chem. Eur. J. 2014, 20, 12641 – 12654

www.chemeurj.org

12654

ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Article Doi

Trisphosphine-Chelate-Substituted Molybdenum and Tungsten Nitrosyl Hydrides as Highly Active Catalysts for Olefin Hydrogenations

DOI: 10.1002/chem.201402736

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.201402736

Full text of DOI:10.1002/chem.201402736

Products guided by the article

R&D Labs maybe for 111-65-9

Relevant to this article

Hot Product