Olefin complexes of low-valent rhenium

FULL PAPER

DOI: 10.1002/ejic.200700573

Olefin Complexes of Low-Valent Rhenium

Aldjia Choualeb,^[a]Olivier Blacque,^[a]Helmut W. Schmalle,^[a]Thomas Fox,^[a]

Thomas Hiltebrand,^[a]and Heinz Berke*^[a]

Keywords: Rhenium / Olefin complexes / Nitrosyl complexes / DFT calculation

Replacement of the MeCN group in the acetonitrile-cis-

dibromo(nitrosyl)-trans-bis(phosphane)rhenium compounds

1a,b (R = iPr a, R = Cy b) with ethylene afforded the olefin

derivatives [Re(η²-H₂C=CH₂)(NO)(PR₃)₂Br₂] (2a,b) (R = iPr a,

R = Cy b). Compound 1a could be converted into the di-

methyl species [Re(MeCN)(NO)(PiPr₃)₂Me₂] (3a) applying

MeLi in toluene; the related methylation of 1b, however,

failed. Abstraction of a Br^–ion from 1a,b with [Na][BArЈ₄] in

acetontrile yields the air-stable salts [Re(MeCN)₂(NO)(PR₃)₂-

Br]⁺[BArЈ₄]^–(4a,b) (R = iPr a, R = Cy b) and under 1 bar of H₂

complexes 1a,b were converted into the known dihydrogen

species [Re(η²-H₂)(NO)(PR₃)₂(Br)₂] (5a,b) (R = iPr a, R = Cy

b). Reduction of [Re(MeCN)(NO)(PR₃)₂Br₂] (1a,b) with Na/

Hg under 1 bar of C₂H₄afforded the butadiene complex

[Re(η⁴-C₄H₆)(η²-H₂C=CH₂)(NO)(PR₃)₂] (6a,b) (R = iPr a, R =

Cy b) via oxidative coupling of two coordinated ethylene

groups followed by double β-H shift and subsequent re-

ductive H₂elimination from the formed dihydride complex.

Reduction of the complexes [Re(CO)(NO)(PR₃)₂Cl₂] (7a,b) (R

= iPr a, R = Cy b) with Na/Hg yields the pentacoordinate

species [Re(CO)(NO)(PR₃)₂(η²-H₂C=CH₂)] (8a,b) (R = iPr a, R

= Cy b) under the same conditions as for 6a,b. Reaction of 8a

with 1 equiv. of B(C₆F₅)₃leads to the [Re(CO)₂{NOB(C₆F₅)₃}-

(PiPr₃)₂] compound (9a) and to the carbonyl nitrosyl com-

plexes [Re(CO)₂(NO)(PiPr₃)₂] (10a) with evolution of ethyl-

ene. The same reaction of 8a and 8b, but applying 1 bar of

CO, leads to exclusive formation of 9a,b. Complexes 4a, 4b,

6b, 8a, 9a, and 9b were characterized by X-ray diffraction

studies.

Germany, 2007)

pseudo-octahedral or pentacoordinate olefin complexes

with rhenium in low oxidation states and a partial π ac-

ceptor coordination sphere were studied for their conforma-

tional preferences in olefin rotation,^[4b]photochemical

isomerization processes,^[4c]and in stoichiometric C–H acti-

vations,^[4d]which demonstrated in various ways the basic

capability of such rhenium centers to strongly interact with

olefins. Our group has recently focused on the development

of rhenium bisphosphane nitrosyl systems containing

Introduction

Over a large range of oxidation states rhenium shows

generally high bond strengths to carbon donors. However,

its affinity to π acceptor ligands can vary with the oxidation

state of the metal center, since π back-bonding is expected

to depend on the metal-d-electron count. Complexes with

rhenium in high oxidation states thus show less propensity

for olefin binding, a circumstance, which allows reversible

olefin binding in catalytic intermediates, such as those in

epoxidations,^[1a]olefin metathesis reactions^[1b]and in pro-

cesses, which crucially involve C–H activations.^[2]Com-

pounds with rhenium in lower oxidation states are expected

to more strongly bind olefins, but such rhenium centers

often do not perform well in catalysis, since they frequently

show reluctance to form coordinatively unsaturated species.

Special ancillary ligand sets may, however, assist dissoci-

ations of other ligands. Thus, π acceptors can labilize other

π acceptors sharing the metal π electron density and

furthermore π donor ligands may exert a kinetic cis labiliz-

ing effect on cis ligands.^[3]In this study we therefore wanted

to evaluate ethylene binding to rhenium in lower oxidation

states^[4a]and in various ligand environments. Earlier several

[Re(NO)(PR₃)₂X₂]^[5](X

= halogene) and [Re(NO)₂-

(PR₃)₂]^[6]fragments. The latter fragments were seen to react

with olefins in metathesis catalyses^[6]and thus demonstrated

affinity to olefins. For the earlier types of fragments a high

propensity for olefin uptake was extrapolated based on

[5]

their capability to bind H₂and the close relationship in

binding properties of both ligands.^[7]Our explorations on

ethylene binding were therefore started using [Re(NO)-

(PR₃)₂X₂] fragments possessing a “mixed type” ligand

sphere of σ and π donors and π acceptors.

Results and Discussion

I. Reactions of [Re(CH₃CN)(NO)(PR₃)₂Br₂]

[a] Anorganisch-chemisches Institut Universität Zürich,

Winterthurerstrasse 190, 8057 Zürich, Switzerland

Fax: +41-44-6356802

The reactions of the recently reported [Re(MeCN)-

(NO)(PR₃)₂Br₂] complexes (R = iPr 1a, R = Cy 1b)^[5]with

ethylene produced indeed the ethylene complexes [Re(η²-

E-mail: hberke@aci.uzh.ch

5246

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

Scheme 1.

H₂C=CH₂)(NO)(PR₃)₂Br₂] (R = iPr 2a, R = Cy 2b) in good not afford the monomethyl complex [Re(NO)(MeCN)-

yields (Scheme 1). When ethylene was bubbled through the (PiPr₃)₂MeBr], rather only half an equivalent of 3a. This

solutions of 1a,b at 1 bar the color changed from light yel- could be interpreted in terms of a faster second methyl-

low to purple after 1 h for 2a and 1.5 h for 2b, but compari- ation, faster than the first one. Surprisingly, all attempts to

son of the ³¹P NMR spectra indicated only slight changes.^[5]access the dimethyl rhenium species 2b did not meet with

Complexes 2a,b were characterized further by spectro- success. A variety of inseparable products were observed

1

scopic means (IR, H, and ¹³C NMR) and the pure prod- indicating a complicated course of the reaction with forma-

ucts showed correct elemental analyses. Studies in solution tion of radicals.

were some times more difficult for 2b, since it exhibits low

1

Complex 3a was characterized by IR, H, ³¹P{¹H}, and

solubilities in many organic solvents. Besides a strong ¹³C{¹H} NMR spectroscopy. Because of the liquid nature

ν(NO) band, the solid state IR spectra of 2a,b displayed of the compound, satisfactory elemental analysis could not

ν(CH) bands in the region of 2800 to 3000 cm^–1indicating be obtained. The IR spectrum recorded in CH₂Cl₂revealed

the presence of the Re-coordinated ethylene. ν(C=C) ab- a very strong band at 1588 cm^–1for the NO group, as well

sorptions could not be detected. The ¹H NMR spectrum of as a weak band at 2360 cm^–1for the coordinated MeCN

2a revealed two broad resonances at δ = 2.51 and 2.47 ppm ligand. Compared to 2a the ν(NO) absorption of 3a is

for two chemically different types of protons of the C₂H₄shifted to lower wavenumbers by about 100 cm^–1indicating

1

ligand. Apparently there is hindered rotation around the a relatively strong π-back-bonding.^[9]The H NMR spec-

rhenium–ethylene bond with a preferred orientation of the trum in C₆D₆showed besides the signals of the iPr and

C=C bond parallel to the P–Re–P axis, which leads to dis- MeCN groups, two triplets at δ = 2.36 ppm (J(PH) =

tinction of pairs of Z-protons. Even at temperatures as high 2.2 Hz) and 2.29 ppm (J(PH) = 2.0 Hz) for the chemically

as 80 °C these signals did not coalesce. In the case of com- inequivalent methyl groups. In the ¹³C{¹H} NMR spectrum

1

plex 2b the H NMR signals of the olefin overlapped with they give rise to broad signals at δ = 24.6 and 20.6 ppm.

those of the PCy₃groups so that the hindered rotation of Complex 3a turned out to be very air-sensitive in solution.

1

the olefin ligand could not be traced by H NMR spec-

Abstraction of a bromide atom from 1a,b using 1 equiv.

troscopy. However, DFT calculations (vide infra) suggest a of [Na][BArЈ₄] (ArЈ = m-(CF₃)₂C₆H₃) in acetonitrile oc-

substantial barrier of rotation for such types of bis(phos- curred within 14 h and proceeded with regiospecific replace-

phane) complexes in general. In the ¹³C NMR spectra the ment of the bromide cis to the NO group affording the

two carbon atoms of the C₂H₄moieties are chemically [Re(MeCN)₂(NO)(PR₃)₂Br][BArЈ₄] salts 4a,b. The trans-to-

equivalent and appear for 2a and 2b as broad triplets at δ NO bromide is apparently bound more strongly experienc-

= 39.5 ppm confirming the preferred orientation of the ole- ing a push-pull stabilization of the π donor with the trans

fin in the P–Re–P axis. In the ³¹P NMR spectra both phos- π acceptor. Complexes 4a,b were isolated as pure yellow

phorus atoms are chemically equivalent and exhibit a solids and are quite air-stable. The use of an excess of [Na]-

singlet at δ = –13.3 ppm for 2a and –20.3 ppm for 2b [BArЈ₄] did not lead to abstraction of the second bromide

(40 °C).

ion. Because of the cationic nature of the complexes, the

Complex 1a reacts with 2 equiv. of MeLi in toluene to solid-state IR spectra of 4a,b showed ν(NO) bands shifted

produce the corresponding dimethyl derivative [Re(NO)- to higher wavenumbers in comparison with the starting ma-

(MeCN)(PiPr₃)₂Me₂] (3a) related to the [Re(NO)(CO)- terials. Bands for the two coordinated acetonitriles, which

(PR₃)₂Me₂] complexes (R = OMe, iOPr, Me, Et) reported should appear in the range of 2350 to 2200 cm^–1, could not

earlier.^[8]Reaction of 2a with only 1 equiv. of MeLi does be detected. The NMR spectroscopic data of 4a,b con-

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5247

H. Berke et al.

FULL PAPER

firmed the presence of {Re(MeCN)₂(NO)(PR₃)₂} fragments the two MeCN and the phosphane groups are disposed

with two nitrile ligands. The ¹H NMR spectra displayed trans. Selected bond lengths and angles of 4a,b are summa-

singlets at room temperature at δ = 2.89 ppm for 4a and at rized in Table 1. The nitrosyl groups of both compounds

δ = 2.93 ppm for 4b attributed to the two chemically equiva- were found disordered with the Br ligands.

lent trans acetonitrile ligands. Singlets also appeared in the

³¹P{¹H} NMR spectra at δ = 0.4 for 4a and –10.2 ppm for nated MeCN group in 1a,b producing the corresponding

4b.

dihydrogen derivatives [Re(η²-H₂)(NO)(PR₃)₂Br₂] (5a,b) in

As depicted in Scheme 1, H₂rapidly displaces the coordi-

The molecular structures of 4a,b were also determined quantitative spectroscopic yields. Complexes 5a,b were

1

by X-ray diffraction studies and the cations are shown in identified by H NMR spectroscopy based on the data of

Figure 1. Complex 4b crystallized as a CH₂Cl₂solvate. The their earlier preparations from the reaction of [NEt₄]-

rhenium coordination was found to be pseudo octahedral [Re(NO)Br₅] with the corresponding phosphanes in etha-

with the nitrosyl ligand located trans to the Br atom, while nol.^[5]

Figure 1. ORTEP drawings of the molecular structures of the cations of 4a (left) and 4b (right) (thermal ellipsoids are drawn at the 20

and 30% probability levels, respectively). The rhenium atom of 4a lies on a center of inversion, only half of the molecule has been refined

and the equivalent atoms (*) have been generated by the symmetry transformation 1/2 – x, 1/2 – y, 1/2 – z). The resulting disorder

between the NO and Br ligands, the counterions [BArЈ₄]^–, the CH₂Cl₂solvate molecule of 4b, as well as all hydrogen atoms have been

omitted for clarity.

Table 1. Selected bond lengths [Å] and angles [°] for 4a and 4b.

4a

4b

Re(1)–P(1)

2.5077(12)

1.830(10)

2.056(3)

2.4733(16)

1.208(10)

1.129(5)

180

173.7(3)

180

173.2(12)

178.1(4)

Re(1)–P(1)

2.5319(19)

2.5288(19)

1.716(7)/1.662(17)

2.071(7)

Re(1)–P(1a)^[a]

Re(1)–N(1)

Re(1)–P(2)

Re(1)–N(1)^[b]

Re(1)–N(2)

Re(1)–N(2a)

Re(1)–Br(1)

N(1)–O(1)

Re(1)–N(3)

2.052(6)

Re(1)–Br(1)^[b]

N(1)–O(1)^[b]

2.5582(11)/2.486(7)

1.222(9)/1.178(15)

1.133(10)

1.148(10)

177.25(6)

176.7(2)

173.5(3)

173.2(6)/164(4)

173.1(6)

177.7(7)

N(2)–C(22)

N(2)–C(1)

N(3)–C(3)

N(2a)–C(22a)

P(1)–Re(1)–P(1a)

N(1)–Re(1)–Br(1)

N(2)–Re(1)–N(2a)

Re(1)–N(1)–O(1)

Re(1)–N(2)–C(22)

Re(1)–N(2a)–C(22a)

P(1)–Re(1)–P(2)

N(1)–Re(1)–Br(1)

N(2)–Re(1)–N(3)

Re(1)–N(1)–O(1)

Re(1)–N(2)–C(1)

Re(1)–N(3)–C(3)

[a] Symmetry operation: a = 1/2 – x, 1/2 – y, 1/2 – z. [b] Split atoms due to disorder are labeled N(11), N(12), O(11), O(12), Br(1), Br(2)

in the cif file.

5248

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

Scheme 2.

We then attempted access to olefin complexes with the able differences in the trans influences of ethylene and the

rhenium center in a still lower oxidation state than +I. Re- NO group with the NO ligand having the stronger effect.

duction of 1a,b with sodium amalgam in THF leads at The hydrogen atoms of the coordinated olefins could not

room temperature under 1 bar of ethylene to the butadiene be located in the Fourier difference maps and they have

ethylene species [Re(η⁴-C₄H₆)(η²-H₂C=CH₂)(NO)(PR₃)₂] been calculated after each refinement cycle (riding model).

(6a,b) (Scheme 2). These Re^–Icomplexes were isolated as

pale yellow powders in yields of 31% for 6a and of 63% for

6b.

In Scheme 2 a mechanism for the formation of 6a,b is

proposed. First generation of 2a,b and then reduction to a

transient Re(ethylene)₂complex is suggested, which eventu-

ally converts into a metallacyclopentane structure with oxi-

dative coupling of the two ethylene ligands.^[10–12]Double

β-hydride elimination and H₂reductive elimination^[2,13]is

followed by the ethylene replacement of one phosphane li-

gand.

Single crystals of 6b were obtained from a mixture of

pentane/dichloromethane, 10:1 (Figure 2, Table 2). The X-

ray diffraction study established the structure of 6b in de-

tail. The coordination sphere of 6b can be viewed as a

pseudo tetrahedron counting the butadiene as one ligand

or as a pseudo square pyramid with the butadiene moiety

occupying two coordination sites. According to the latter

view the essentially planar η⁴-butadiene, η²-ethylene, and

the NO ligand would form the base and the phosphane the

Figure 2. ORTEP drawing of the molecular structure of 6b (ther-

apical position of the pyramid. Conformationally, the ethyl-

ene group is in parallel orientation with respect to the Re–

P axis and the cis-butadiene ligand points with its open side

toward the PCy₃ligand. The ethylene and the butadiene

group are disordered. The disorder could properly be re-

solved revealing reasonable bonding parameters. The

average distance between the internal butadiene carbon

atoms of 1.34(2) Å is significantly shorter by 0.09 Å than

the average distances of the terminal carbon atoms of

1.43(2) Å speaking for a strong Re-butadiene back bonding

with superimposition of the π₃orbital of the butadiene moi-

ety. For comparison, the free cis-butadiene bond possesses

an internal C–C distance of 1.463(3) and terminal olefinic

distances of 1.342(2) Å.^[14,15]The Re–C_intand Re–C_termdis-

tances are similar and fall into the range of 2.16(3) to

2.27(2) Å emphasizing a genuine η⁴coordination mode.

The Re–C_termbond length approximately trans to the ethyl-

ene group is shorter by 0.09(2) Å than the Re–C_termdis-

tance trans to the NO group. This can be explained via size-

mal ellipsoids are drawn at the 30% probability level). The disorder

observed between the ethylene and butadiene ligands as well as all

hydrogen atoms have been omitted for clarity.

Table 2. Selected bond lengths [Å] and bond angles [°] for 6b.

Re(1)–N(1) 1.770(3)

Re(1)–C(1A) 2.285(15)

Re(1)–C(2A) 2.25(2)

Re(1)–C(3A) 2.235(13)

Re(1)–C(4A) 2.258(9)

Re(1)–C(5A) 2.18(2)

Re(1)–C(6A) 2.170(18)

Re(1)–C(1B) 2.30(3)

Re(1)–C(2B) 2.28(3)

Re(1)–C(3B) 2.272(18)

Re(1)–C(4B) 2.224(11)

Re(1)–C(5B) 2.21(3)

Re(1)–C(6B) 2.16(3)

N(1)–O(1)

1.197(4)

1.434(17)

1.417(17)

1.344(14)

1.42(2)

1.43(3)

1.47(2)

1.34(2)

1.42(2)

175.8(3)

C(1A)–C(2A)

C(3A)–C(4A)

C(4A)–C(5A)

C(5A)–C(6A)

C(1B)–C(2B)

C(3B)–C(4B)

C(4B)–C(5B)

C(5B)–C(6B)

Re(1)–N(1)–O(1)

C(3A)–C(4A)–C(5A) 118.2(11)

C(4A)–C(5A)–C(6A) 115.4(14)

C(3B)–C(4B)–C(5B) 123.4(14)

C(4B)–C(5B)–C(6B) 111.3(17)

Re(1)–P(1)

2.4585(10)

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5249

H. Berke et al.

FULL PAPER

Principally, butadiene can coordinate to a metal as a cis- lowest energy. B and C of this derivative lie +3.9 and

or trans-butadiene. The trans-η⁴-diene complexes are nor- +8.6 kcal/mol higher in energy. In analogy 6a is assumed to

mally thermodynamically less stable and they may convert exist in the equilibrium conformations A and B. Another

to the corresponding cis-η⁴-diene complexes.^[16,17]The cis- point of discussion concerns the kinetics. The barrier for

η⁴-diene complexes are known mainly with early transition the interconversion of rotamer A and B apparently is low

metal centers, such as in Cp₂M(diene) (M = Ti, Zr, Hf),^[16]enough that room temperature thermal energy (Ͻ15 kcal/

as well as with Ta.^[18]

mol) can equilibrate the two isomers of 6a, however, at a

Two extreme forms of butadiene cis coordination were rate which is slow on the NMR time scale. By the same

found: the cis-η⁴-1,3-diene structure and the metallacyclo- token the energy difference of isomers A and B of 6b and

pent-3-ene structure. NMR spectroscopy is able to distin- their barrier of rotation are expected to be substantially

guish the various coordination modes of the diene ligand higher than those of 6a, since the PCy₃ligand possesses a

on the basis of their coupling constants. In the cis coordina- still larger cone angle than PiPr₃. For the latter complex the

tion the majority of diene complexes reported so far exhibit butadiene rotation is therefore considered to be “frozen

the η⁴-1,3 diene structure possessing bond lengths of C_int

–

out” in the energetically preferred rotamer A.

C_intՆ C_term–C_termand M–C_termՆ M–C_term, the dihedral

angles between the planes defined by C_term, M, C_termand

C

_term, C_int, C_int, C_termatoms are generally Ͻ90°. When the

dihedral angle is greater than 90° and the C_int–C_intbond is

significantly shorter than C_term–C_int, the bent metallacy-

clopent-3-ene structure is adopted.^[12]

1

The H NMR and the ¹³C NMR spectroscopic data of

6a,b indeed confirmed their cis butadiene structure. Also on

the basis of the ³¹P NMR spectra, 6a, however, exists in

solution as a mixture of two stereoisomers appearing in a

3:1 ratio. Suggested by the DFT calculations modeled with

a PMe₃ligand (vide infra) the isomerism is anticipated to

be caused by butadiene rotation. Principally the three rota-

mers A, B, and C are conceivable (Figure 3). For electronic

reasons any of these rotamers are expected to have one leg

of the three-legged pianostool in the diene cleft.^[19]Accord-

Figure 3. Projections of the rotational isomers of 6a,b and 6-Me

ing to the calculations on a PMe₃model isomer A is of and proton assignment for the butadiene ligand.

Figure 4. TOCSY for the butadiene ligand of 6b recorded in C₆D₆(r.t.) with irradiation of one internal CH group.

5250 Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

Olefin Complexes of Low-Valent Rhenium

The ¹H NMR spectrum of 6b indeed points to a solution

structure according to rotamer A, which indeed is the struc-

ture in the solid state (Figure 2). Beside the signals of the

PCy₃ligand, overlapping with those of the C₂H₄group, the

resonances of the butadiene group are attributed to two

anti, two syn, and two internal protons. Any of the isomeric

structures A, B, and C of 6a,b possesses C₁symmetry and

“left/right” asymmetry of the butadiene so that the various

types of butadiene protons are doubled and each butadiene

Scheme 3.

The change from yellow to orange became visible after

proton appears at a distinct chemical shift. For 6b this filtration from the amalgam. The formation of 8a,b was ac-

could nicely be demonstrated by a TOCSY experiment with companied by the formation of a very small amount (about

irradiation of one internal CH group (Figure 4 and Table 3). 1%) of the known dihydride complex [Re(NO)(CO)(H₂)-

For 6a, however, the chemical shift differences between the (PR₃)₂],^[5]which was removed by washing of the solids with

signals of the butadiene and the ethylene groups of the rota- pentane. The dihydride complexes were formed in larger

mers A and B are very small and lead to overlap. The differ- amounts (about 34%) when the reaction was carried out

ences in chemical shifts between the syn and the anti pro- under a pressure of 10 bar of ethylene. It was concluded

tons of 6a,b were found to be in the range of 0.25–1.42 ppm that the formation of the dihydride complex results from

comparable with the ∆δ values of Fe(CO)₃(η⁴-C₄H₆)^[20]and the presence of a H₂impurity in the ethylene gas.

of RhCp(η⁴-C₄H₆) (1.3–2.3 ppm) (Table 3). A coupling be-

tween the syn protons and the phosphorus nuclei was not tions, which are shifted both by more than 100 cm^–1to

The IR spectra of 8a,b reveal ν(CO) and ν(NO) absorp-

observed for both complexes.

lower wavenumbers compared to 7a,b^[8]indicating increased

π-back bonding. ν(C=C) bonds are not observed, but the

coordination of the ethylene group is indicated by the pres-

ence of characteristic ν(C–H) vibrations. The ¹H NMR

spectrum of 8b is less informative with respect to a struc-

tural assignment, since the PCy₃resonances cover other sig-

nals over a large area of chemical shifts. The ¹H NMR reso-

nances of the ethylene group of 8a could not be detected in

the room temperature spectrum, but were observed at

100 °C as a broad triplet at δ = 1.86 ppm for all chemically

and magnetically inequivalent C₂H₄protons.

1

Table 3. H NMR chemical shift [ppm] for the butadiene ligand of

the rotamers A and B of 6a and A of 6b.

H_syn1

H_anti1

H_int1

H_int2

H_anti2

H_syn2

6a(A)

6a(B)

6b(A)

–0.21

–0.17

–0.52

–0.72

–0.42

5.72

5.76

5.25

5.29

0.56

0.43

0.70

1.98

2.06

2.01

The structures of 8a,b were derived from their analytical

and spectroscopic data and for 8a it was confirmed by an

exemplary X-ray diffraction study. The structure of 8a is

shown in Figure 5 and selected bond lengths and angles are

given in Table 4.

The coupling constants J(H_syn1H_anti1), J(H_syn1H_int1),

J(H_anti1H_int1), J(H_int1H_int2), J(H_anti2H_int2), J(H_syn2H_int2) of

isomer A of 6a,b and isomer B of 6a are 7 and 6 Hz and

J(H_anti2H_syn2) 4 and 3 Hz, respectively. On the basis of the

couplings of the anti protons with the phosphorus nuclei

(ca. 7 Hz) it was concluded that the butadiene units in 6a,b

have structures close to a cis-η²-1,3-diene ligand with a

minor admixture of a metallacyclo-3-pentene structure. The

coordination of the ethylene group in 6a,b was confirmed

by ¹³C NMR and by NOE measurements. All ethylene pro-

The coordination geometry around the rhenium center

corresponds to a distorted trigonal bipyramid with the CO

and NO groups and both carbon atoms of the ethylene

group located in the trigonal plane of the molecule. The

two phosphane ligands occupy apical positions with a P(1)–

Re(1)–P(2) angle of 178.16(3)° (Table 4). Contrary to com-

plexes 2a,b the orientation of the C=C bond is perpendicu-

lar to the P–Re–P vector, which is indicated by the dihedral

angles P(1)–Re(1)–C(2) 90.37(10)°, P(1)–Re(1)–C(3)

90.73(9)°, P(2)–Re(1)–C(2) 90.75(10)°, and P(2)–Re(1)–C(3)

91.04(9)°. It is the electronically preferred orientation in d⁸

trigonal bipyramidal structures. The C=C bond length of

1.420(8) Å is comparable to those found in other ethylene

1

tons of 6a appear in the H NMR spectrum as chemically

distinguished signals indicating hindered rotation.

II. Reactions of [Re(NO)(CO)(PR₃)₂Cl₂]

In order to probe the accessibility of other Re^–I–olefin rhenium complexes.^[21]The Re–N bond length of

complexes, we attempted the reduction of the complexes 1.799(3) Å is in accord with the values found in the litera-

[Re(NO)(CO)(PR₃)₂Cl₂] (7a,b) (R = iPr a, Cy b), related to ture.^[22]The N1–O2 bond separation of 1.208(4) Å is in the

1a,b by replacement of the olefin ligand with CO. In the range of 1.10 to 1.38 Å expected for a linear NO⁺type co-

presence of 1.4 bar of ethylene and otherwise under the ordination^[21–23][Re–N–O angle of 172.7(3)°]. The positions

same conditions as for the reductions of 1a,b, the ethylene of the hydrogen atoms of the ethylene ligand were located

derivatives [Re(NO)(CO)(η²-H₂C=CH₂)(PR₃)₂] (8a,b) were in a difference electron-density map and their coordinates

obtained in good yields (Scheme 3) after reaction times of were fixed, but their isotropic displacement parameters

5 d.

were freely refined.

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5251

H. Berke et al.

FULL PAPER

Figure 5. ORTEP drawing of the molecular structure of 8a (thermal ellipsoids are drawn at the 50% probability level). All hydrogen

atoms have been omitted for clarity.

Table 4. Selected bond lengths [Å] and bond angles [°] for 8a.

catalysts, thus increasing activity. One important class of

these zwitterionic complexes is obtained when the counter-

anion is located on the alkyl group occupying the reactive

wedge of the complex. These types of complexes are called

Girdle-type zwitterionic compounds and have been ac-

cessed mainly through electrophilic attack of B(C₆F₅)₃on

suitable hydrocarbon ligands of neutral group 4 bent

metallocenes.^[31,32]

With the objective to generate active rhenium-based cata-

lysts, we exemplarily attempted activation of complex 8a

with a Lewis acid. Treatment of 8a with one equivalent of

B(C₆F₅)₃in toluene or benzene at room temperature did,

however, not lead to the mentioned zwitterionic species

[Re⁺(NO)(CO)(PR₃)₂CH₂CH₂B^–(C₆F₅)₃]. The structures of

Re(1)–P(1)

Re(1)–P(2)

Re(1)–N(1)

Re(1)–C(1)

Re(1)–C(2)

Re(1)–C(3)

C(2)–C(3)

N(1)–O(2)

C(1)–O(1)

P(1)–Re(1)–P(2) 178.16(3)

C(1)–Re(1)–C(2) 120.22(15)

C(1)–Re(1)–P(1) 89.95(11)

C(1)–Re(1)–P(2) 90.75(11)

C(2)–Re(1)–C(3) 36.59(14)

2.4758(9)

2.4744(9)

1.799(3)

1.922(4)

2.239(4)

2.282(3)

1.420(6)

1.208(4)

1.151(5)

C(2)–Re(1)–P(1)

C(2)–Re(1)–P(2)

C(3)–Re(1)–P(1)

C(3)–Re(1)–P(2)

N(1)–Re(1)–C(1) 111.11(15)

N(1)–Re(1)–C(2) 128.67(15)

N(1)–Re(1)–C(3) 165.26(14)

N(1)–Re(1)–P(1)

N(1)–Re(1)–P(2)

Re(1)–C(1)–O(1) 176.4(3)

Re(1)–N(1)–O(1) 172.7(3)

90.37(10)

90.75(10)

90.73(9)

91.04(9)

89.32(9)

88.84(9)

C(2)–C(3)–Re(1)

C(3)–C(2)–Re(1)

70.1(2)

73.3(2)

Olefin complexes can be activated by Lewis acids to be- the product showed that the B(C₆F₅)₃electrophile had at-

come alkene polymerization catalysts.^[24–30]The olefin gets tacked the more Lewis basic NO site to give eventually the

attacked by the Lewis acid forming a zwitterionic species. nitrosyl/Lewis acid adduct {Re(CO)₂[NOB(C₆F₅)₃](PR₃)₂}

The metal cation and the Lewis acid attacked anionic group (9a) and in addition the known complex [Re(CO)₂(NO)-

are covalently linked in some fashion. They have been (PR₃)₂] (10a) with loss of the ethylene ligand in both com-

found to represent a way to modulate ion pairing in these plexes as depicted in Scheme 4. Complex 10a was identified

Scheme 4.

5252

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

by its ³¹P NMR spectrum (δ = 81.6 ppm) and its IR spec- have formed. It apparently originates from the partial de-

trum reported earlier.^{[8] 1}H and ³¹P{¹H} NMR pursuit of composition of the B(C₆F₅)₃adduct of 8a to yield ethylene

the reaction confirmed generation of free ethylene and for- and CO. The presence of only low amounts of CO accounts

mation of 9a and 10a in a 1:1 ratio. Formation of the phos- for the low yield of 9a.

phane adduct iPr₃PB(C₆F₅)₃was not observed.^[33]Complex

9a could be separated from 10a in 38% yield.

With the objective to access 9a,b in a better yield, we

first prepared complex 10a,b by the reaction of 8a,b with

Formation of 9a with replacement of the ethylene group CO. The reaction was very sluggish and took more than

in 8a by CO requires explanation of how free CO could 5 d. But much to our surprise, no stable adduct formation

Figure 6. ORTEP drawings of the molecular structures of 9a and 9b (thermal ellipsoids are drawn at the 30% probability level). All

hydrogen atoms have been omitted for clarity.

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5253

H. Berke et al.

FULL PAPER

could be observed between 10a,b and B(C₆F₅)₃. We have to pound [Re(Cp)(SiMe₂Cl)(NOBCl₃)(PPh₃)].^[36]This com-

assume that the additions of B(C₆F₅)₃to 10a,b are kinet- pound reveals a marked N–O elongation [1.47(1) Å] ac-

ically hindered.

companied by a shortening of the Re–N bond [1.568(9) Å].

When, however, 8a,b, B(C₆F₅)₃, and CO were mixed to- In 9a,b the N–O bond lengths are 1.276(6) and 1.296(8) Å.

gether 9a,b was produced in high yield. This observation Nitrosyl/Lewis acid interactions were shown to play a key

stresses the necessity for the involvement of initial B(C₆F₅)₃role in the catalytic oxidation of alcohols,^[37]as well as in

adducts of 8a,b promoting formation of 9a,b by subsequent catalytic hydrogenation and hydrosilylation of olefins and

ethylene substitution with CO (Scheme 4).

ketones.^[33]

The adducts apparently release ethylene much faster than

8a,b in the presence of CO to trap the unsaturated interme-

diates. Single crystals of 9a,b were obtained from CH₂Cl₂/

pentane solutions at –30 °C.

III. Theoretical Studies on Ethylene Complexes

As a single-faced π acceptor an olefinic ligand quite

often shows barriers to free rotation around the metal–ole-

fin bond. Generally, the C=C bond is preferentially ori-

ented in the trigonal plane of three-coordinate complexes, is

perpendicular to the plane in four-coordinate square-planar

complexes, and is in plane with the trigonal plane of five-

coordinate trigonal-bipyramidal complexes (Scheme 5).^[38]

These prevailing orientations are thought to result mostly

from electronic rather than from steric effects. In d⁶octahe-

dral complexes olefins do only rarely show rotational barri-

ers, if the cis ligands are all the same. Barriers might appear

only for cases where the π donor planes at the metal centers

are electronically strongly distinguished (R₃P–M–PR₃plane

electronically more donating than the L₁–M–L₃plane of

Scheme 5). In d⁸trigonal bipyramidal complexes the olefin

is preferentially oriented in the trigonal plane regardless of

the presence of an axial R₃P–M–PR₃arrangement. For in-

stance in the structure of 8a the olefin orientation is in ac-

cord with this general rule.

The molecular structures of 9a,b were determined and

are shown in Figure 6. The coordination geometries are dis-

torted trigonal bipyramids. The two CO groups and the

NOB(C₆F₅)₃fragment are located in the trigonal plane of

the molecules and the two phosphanes occupy apical posi-

tions. The phosphane ligands are bent towards the two CO

groups [(P(1)–Re–P(2) 161.20(5) and 161.92(6)° for 9a and

9b, respectively] (Table 5). Such substantial bending of the

phosphane molecules may be caused by steric “bumping”

of the phosphanes and the NOB(C₆F₅)₃fragment. Signifi-

cant bending of the phosphanes was also observed in the

hydride Lewis acid adducts [Re(H)(NO)(NOBC₆F₅)₃-

(PiPr₃)₂]^[33]and [Re(H)(NO)(NOBF₃)(PiPr₃)₂]^[5]with P(1)–

Re(1)–P(2) 134.77(3) and 146.60(6)°, respectively. The elec-

tronic and steric factors favor the bending of the axial li-

gands towards the purely σ-donating hydride ligand and

away from the nitrosyl group.^[34,35]Similar to the complexes

{Re(H)(NO)[NOB(C₆F₅)₃](PiPr₃)₂}

and

[Re(H)(NO)

(NOBF₃)(PiPr₃)₂] the Re–N distances of 1.790(4) and

1.785(6) Å are only slightly affected by the coordination of

B(C₆F₅)₃, but compared to other η¹rhenium nitrosyl com-

plexes the N–OBX₃bond is elongated. The nitrosyl ligands

are slightly bent upon B(C₆F₅)₃attachment (Re(1)–N(1)–

O(3) 166.2(4) and 158.6(6)°); however, it was approximately

linear in the cases of {Re(H)(NO)[NOB(C₆F₅)₃](PiPr₃)₂}

[Re–N–O 175.0(3)°] and [Re(H)(NO)(NOBF₃)(PiPr₃)₂] [Re–

N–O 170.0(4)°]. A significant structural effect was recog-

nized upon coordination of BCl₃to a NO group in the com-

Scheme 5.

Five-coordinated d⁶complexes, as for instance (µ²-Cl-

[(AlMe₂)₂Cϵ)]W(CH₃)(PMe₃)₂(C₂H₄),^[39]

(bis-mesityl-

Table 5. Selected bond lengths [Å] and angles [°] for 9a and 9b.

imido)W(PMe₃)₂(C₂H₄),^[40]and bis(2,6-diisopropylphen-

ylimido)Mo(PMe₃)₂(C₂H₄) show reorientation of the ole-

finic group.^[41]In pseudo-octahedral complexes with trans

phosphane ligands cis to the olefin, structures are prevailing

with a parallel orientation of the olefin with respect to the

P–M–P axis.^[42–49]Nevertheless, some examples show the

olefin lying perpendicular to the P–M–P axis.^[50–53]Several

of these latter examples might be governed by special ef-

fects, since they mostly possess two ethylene ligands in

trans^[50,51]or in cis position^[52]to each other with one C=C

bond parallel to the P–M–P vector and the second one per-

pendicular to it. To the best of our knowledge, only in the

ortho-metallated ruthenium complex RuH(C₂H₄)(2-phenyl-

pyridine)(PiPr₃)₂has the coordinated ethylene not aligned

with the P–Ru–P axis.^[53]

9a

9b

Re(1)–P(1)

Re(1)–P(2)

2.4639(17)

2.4535(17)

1.790(4)

1.961(6)

1.912(6)

1.146(8)

1.166(8)

1.276(6)

161.20(5)

101.2(3)

101.06(16)

97.00(16)

166.2(4)

124.1(4)

2.4840(19)

2.480(2)

1.785(6)

1.988(9)

1.916(8)

1.150(10)

1.151(9)

1.296(8)

161.92(6)

104.5(3)

99.5(2)

Re(1)–N(1)

Re(1)–C(a)^[a]

Re(1)–C(b)^[a]

C(a)–O(1)

C(b)–O(2)

N(1)–O(3)

P(1)–Re(1)–P(2)

C(a)–Re(1)–C(b)

N(1)–Re(1)–P(1)

N(1)–Re(1)–P(2)

Re(1)–N(1)–O(3)

N(1)–O(3)–B

98.5(2)

158.6(6)

125.3(6)

[a] C(a) and C(b) are C(39) and C(40) for 9a, and C(1) and C(2)

for 9b.

5254

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

The most widely accepted model for the bonding of an structural observations since the parallel orientation is fa-

olefin to transition metals is that proposed by Dewar^[54]and vored for 2-Me by 23.1 kcal/mol with respect to the corre-

by Chatt and Duncanson.^[55]This model involves formation sponding perpendicular conformation and the perpendicu-

of a σ-bond by donation of π-electrons to the metal and lar orientation is favored for 8-Me (by 10.2 kcal/mol with

formation of a π-bond by back donation from the d orbitals respect to the corresponding parallel conformation). The

of the metal to the olefin π*-orbital. The precise nature of results from the bond analysis of 2-Me show that ∆E_oiis

this bond has been the subject of many theoretical and ex- the main factor determining the strength of the Re-(η²-

perimental studies.^[56]In order to study the structures of 2a, C₂H₄) bond. The steric interaction term ∆E⁰is dominated

2b, and 8a more thoroughly with the olefinic C=C bond by the Pauli repulsion ∆E_Pauli, and thus represents a non-

lying parallel to the P–Re–P axis in the six-coordinate octa- bonding contribution. Despite the fact that the perpendicu-

hedral complexes 2a and 2b and in the trigonal plane of 8a lar orientation of ethylene reduces the steric repulsion be-

DFT calculations^[57]were carried out on the model com- tween fragments from +62 to +37 kcal/mol (2-Me) the at-

plexes Re(PMe₃)₂X₂(NO)(C₂H₄) (X = Br, 2-Me) and tractive orbital interaction ∆E_oigreatly stabilizes the paral-

Re(PMe₃)₂(CO)(NO)(C₂H₄) (8-Me).

lel molecular arrangement from about 120 kcal/mol to only

We made use of a well-established bond partitioning 59 kcal/mol for 2-MeЌ, respectively. In the case of the tri-

scheme,^[58]which divides the overall bond energy BE into gonal-bipyramidal complex 8-Me, the three main contri-

various contributions from different interactions [Equa- butions of 9-Me^//are very similar to those for 2-Me^//(dif-

tion (1)], to look at the bond-forming reaction between the ferences are less than 8 kcal/mol). The major change which

metal fragment and ethylene.

leads to a favored perpendicular arrangement for 8-Me

comes from the electronic contribution ∆E_oi. Calculated

∆E_oiincreases from 56 kcal/mol to 113 kcal/mol on going

from 2-MeЌ to 8-MeЌ, even larger than the value for

8-Me^//(112 kcal/mol). For 8-Me, both steric (+51 vs.

+59 kcal/mol) and electronic terms (–113 vs. –112 kcal/mol)

contribute to a better stabilization of the perpendicular ar-

rangement.

According to the Dewar–Chatt–Duncanson model, two

molecular orbitals from the metallic fragments Re(PMe₃)₂-

(NO)(CO) or Re(PMe₃)₂(NO)Br₂can compete for the for-

mation of the π-bond by back donation to ethylene. The

metal contribution to these two highest occupied molecular

orbitals is mainly d_xzand d_yz(Scheme 6).

∆E_int= –[∆E_elstat+ ∆E_Pauli+ ∆E_oi] = –[∆E⁰+ ∆E_oi]

(1)

The interaction energy between both fragments ∆E_intis

partitioned into three physically meaningful terms, namely

the electrostatic interaction ∆E_elstat, the Pauli repulsion

∆E_Pauli, and the orbital interaction term, ∆E_oi. When two

suitable, bond-forming fragments are brought together to

adapt the geometry of the final molecule, the term ∆E_elstat

describes the classical Coulomb interaction between the un-

modified and interpenetrating charge distributions of the

two fragments. We further have to consider the Pauli repul-

sion ∆E_Pauli, which takes into account destabilizing two-or-

bital four-electron interactions between occupied orbitals

on both fragments. The sum of Pauli repulsion and electro-

static interaction is called the steric interaction ∆E⁰. The last

term in Equation (1), ∆E_oiintroduces the attractive orbital

interaction between occupied and virtual orbitals on the

two fragments, and includes polarization and charge-trans-

fer contributions. Although ∆E_intvalues are not defined in

the same way as bond dissociation enthalpies ∆H, they are

reasonable approximations of bond enthalpy terms, which

in turn provide a good description for the bond strength.^[59]

The bonding analysis for 2-Me and 8-Me is presented in

Table 6.

Table 6. Bond analysis of the [Re]–(η²-C₂H₄) bond for 2-Me and

8-Me.

System

∆E

∆E_Pauli

∆E_elstat∆E⁰

∆E_oi

–120.86 –58.76

–59.23 –22.55

∆E_int

Scheme 6.

//

2-Me

0.0

+225.23 –163.13 +62.10

2-Me Ќ

+23.1 +127.25 –90.57 +36.68

//

8-Me

+10.2 +217.96 –159.16 +58.79

–111.64 –52.84

–113.15 –61.77

Both orbitals can interact with the olefin π*-orbital in a

bonding fashion depending on the orientation of ethylene.

The LUMO is mainly Re-d_z₂and is responsible for the σ-

8-Me Ќ

0.0

+194.45 –143.07 +51.38

For the three studied model complexes, both conforma- bond by interaction with the filled π_C2H4orbital. The en-

tions with the C=C bond of ethylene parallel (^//) or perpen- ergy levels of these crucial orbitals of the metal fragments

dicular (Ќ) to the P–M–P axis have been optimized without of 2-Me and 8-Me in their structures of the optimized mole-

any constraints. The relative total energies confirm the cules are presented in Figure 7 and Table 7. In the pseudo

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5255

H. Berke et al.

FULL PAPER

Table 7. Optimized bond angles [°] and energy levels [eV] of the metallic fragments Re(PMe₃)₂(NO)Br₂of 2-Me and Re(PMe₃)₂(NO)(CO)

of 8-Me.

System Br–Re–Br

Br–Re–N_NO

P–Re–Br

P–Re–P

N_NO–Re–C_CO

LUMO

HOMO

HOMO – 1

2-Me^//91.8

2-MeЌ 88.9

91.4

89.3

81.1

85.9

143.1

168.1

–4.677 (d_z)

–5.685 (d_xz)

–5.805 (d_xz)

–6.648 (d_yz)

–6.590 (d_yz)

2

–4.679 (d_z)

8-Me^//

8-MeЌ

159.1

175.0

136.8

112

–3.462 (d_z)

–5.084 (d_xz)

–4.488 (d_yz)

–5.147 (d_yz)

–5.820 (d_xz)

2

–3.566 (d_z)

octahedral complex 2-Me the three ligands of the equatorial While the P–Re–P angle decreases by 16° from 175.0 (8-

plane, two halides and one nitrosyl keep more or less their MeЌ) to 159.1° (8-Me^//) the N_NO–Re–C_CObond angle can

positions during ethylene rotation. The Br–Re–Br and Br– become narrower by 25° from the parallel to the perpendic-

Re–N_NObond angles remain in the range 88.9–91.8° and ular orientation of ethylene (from 136.8 to 112°). The result

P–Re–Br increases by only 5° when the ethylene lies in the is a greater destabilization of the occupied orbital built with

ReBr₂(NO) plane (Table 7). The orbitals are well separated Re-d_yz, since the overlap between Re-d_yzand p_y/p_zof the

by about 1 eV and the HOMO is mainly built from Re-d_xzcarbonyl and nitrosyl groups is reduced which becomes the

which favors the parallel orientation of ethylene with re- HOMO of the metallic fragment. Back bonding to ethylene

spect to the P–Re–P axis. The second occupied π-type or- becomes stronger in the equatorial plane and favors the per-

bital is low in energy through the bonding interaction with pendicular orientation of ethylene with respect to the P–

the nitrosyl ligand not allowing for efficient further back- Re–P axis.

bonding with the ethylene.

It was also challenging to provide understanding for

the solution structures of the polyolefin complexes

Re(C₂H₄)(C₄H₆)(PR₃)(NO) 6a and 6b. In particular the X-

ray structure of 6b provides a good idea about the ground

states of 6a and 6b which is pseudo tetrahedral with one

ethylene group coordinated to the rhenium center in paral-

lel orientation with respect to the Re–P axis, one cis-butadi-

ene ligand bound in an η⁴-coordination mode with an es-

sentially planar carbon skeleton, one phosphane and one

nitrosyl group. Complex 6a exists in solution as a mixture

of two stereoisomers A and B with either the phosphane or

the olefin in the cleft of the butadiene appearing in a 3:1

ratio on the basis of the ³¹P NMR spectra. Complex 6b is

expected to exclusively exist in the butadiene rotameric

form A with the phosphane ligand oriented in the cleft of

the diolefin. The potential energy surface of the model com-

plex Re(C₂H₄)(C₄H₆)(PMe₃)(NO) 6-Me could not be fully

explored, but the geometries of the global or local minima

corresponding to the three possible rotational isomers A, B,

and C were fully optimized with the help of the density

functional theory (Figure 8).^[60]The computed bonding en-

ergies of 6-Me give isomer A as the most stable calculated

structure, while B and C lie +3.9 and +8.6 kcal/mol higher

in energy and thus confirm the given structural assignments

for rotational isomers of 6a and 6b. Ethylene rotation was

also checked in rotamer A of 6-Me, but no other energetic

minimum than the orientation parallel to the Re–P axis was

found. The observed isomerism of 6a could thus not be

explained on the basis of ethylene rotational isomerism. The

energy differences between structures A and B or C is ex-

pected to have a significant steric component and to depend

on the size of the phosphane. The larger the phosphane the

stronger the preference for structure A, since the cleft of

the butadiene ligand is the only place in the molecule that

provides abundant space. That would explain why 6b with

the bulky PCy₃shows only isomer A and 6a with the

smaller PiPr₃leads to rational isomerism.

2

Figure 7. Energy diagram for the d_xz, d_yz, d orbitals of the frag-

ments Re(PMe₃)₂(NO)(CO) of 2-Me and Re^z(PMe₃)₂(NO)Br₂of 8-

Me in their structures of the optimized molecules.

In the case of the trigonal-bipyramidal complex 8-Me,

the bending distortion of the ligands in both planes and in

the opposite direction of the ethylene is easier than for 2-

Me and polarizes the d_xzor d_yzorbital of the metal for

increased back-donation to the ethylene. Nevertheless, the

degree of bending of the PMe₃is limited by steric repulsions

between PMe₃and CO and NO and between the phospho-

rus ligands themselves whereas the bending of the carbonyl

and nitrosyl group is only limited by electronic and steric

interactions between themselves in the equatorial plane.

5256

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

ters using 5-mm diameter NMR tubes equipped with Teflon valves,

which allow degassing and further introduction of gases into the

probe. Chemical shifts are given in ppm. ¹H and ¹³C{¹H} NMR

spectra were referenced to the residual proton or ¹³C resonances of

the deuterated solvent. ³¹P chemical shifts are relative to 85%

H₃PO₄. Microanalyses were carried out at the Anorganisch-Chem-

isches Institut of the University of Zürich. IR spectra were recorded

with a Bio-Rad FTS-45 spectrometer. The following reagents were

purchased from commercial suppliers and used without further pu-

rification: MeLi (Fluka), PiPr₃(STREM), ethylene (99.999% pu-

rity, MESSER or 99.9995% purity, PanGas), and H₂(99.99995%

purity, PanGas). Complexes 1a,b were obtained according to litera-

ture procedures.^[5]

Synthesis of [Re(NO)(PiPr₃)₂(η²-C₂H₄)Br₂] (2a): Ethylene was

bubbled through a solution of 1a (0.100 g, 0.135 mmol) in toluene

(20 mL) at room temperature for 1 h. After removal of the volatiles

under vacuum, the residue was chromatographed on a SiO₂column

with toluene as a eluent yielding after drying in vacuo 2a as a pale

yellow powder (0.072 g, 0.1 mmol, 74%). IR (KBr): ν = 2963, 2920

˜

1

and 2879 (s, νC–H), 1693 (vs, νNO) cm^–1. H NMR (300.1 MHz,

C₆D₆): δ = 3.12–2.95 (m, 6 H, (CH₃)₂CHP), 2.51 and 2.47 (2 br.

signals, 4 H, H₂C=CH₂), 1.45–1.32 and 1.35–1.19 (2 m, 36 H,

(CH₃)₂CHP) ppm. ³¹P{¹H} NMR (80.9 MHz, C₆D₆): δ = –13.3

(s) ppm. ¹³C{¹H} NMR (75.5 MHz, C₆D₆):

δ

=

39.5 (br.,

H₂C=CH₂), 26.2 (t, J(PC) = 11 Hz, (CH₃)₂CHP), 19.9 and 19.6

(2 s, (CH₃)₂CHP) ppm. C₂₀H₄₆Br₂NOP₂Re (724.55): calcd.

C

33.19, H 6.41, N 1.94; found C 33.39, H 6.52, N 1.93.

Figure 8. Optimized isomeric structures of the model complex

Re(C₂H₄)(C₄H₆)(PMe₃)(NO) 6-Me.

Synthesis of [Re(NO)(PCy₃)₂(η²-C₂H₄)Br₂] (2b): Complex 1b

(0.086 g, 0.088 mmol) in 100 mL of toluene was heated at 60 °C for

one night. After cooling to room temperature, ethylene was

bubbled through the solution for 1.5 h. The precipitate was col-

lected and washed with cold THF (3ϫ5 mL) and MeCN

(3ϫ5 mL) to afford 2b as a pure pale yellow solid (0.066 g,

Conclusions

This work focused on the chemistry of new olefin rhe-

nium complexes. The ethylene complexes 2a,b were isolated

in good yields by replacement of a MeCN group from 1a,b.

The ethylene ligand showed an unexpectedly high prefer-

ence for the orientation parallel to the P–Re–P axis. Re-

duction of 1a,b with sodium amalgam under 1 bar of ethyl-

ene affords the butadiene complexes 6a,b. The salient fea-

tures in bonding and the coordination geometry of 6a,b

have been clarified on the basis of NMR and X-ray analysis

coupled with a theoretical treatment. The pentacoordinate

d⁸-olefin complexes demonstrated the olefin to be oriented

in the trigonal plane of a trigonal bipyramid. This type of

complex revealed a primary attack of the Lewis acid

B(C₆F₅)₃at the nitrosyl group rather than on the olefin,

which eventually furnished Re(CO)₂[NOB(C₆F₅)₃](PR₃)₂

complexes with ethylene replacement in the presence of CO.

0.068 mmol, 78%). IR (KBr): ν = 2926 and 2879 (s, νC–H), 1696

˜

1

(vs, νNO) cm^–1. H NMR (300.1 MHz, CD₂Cl₂, 40 °C): δ = 3.00–

1.32 (H₂C=CH₂and P(C₆H₁₁)₃) ppm. ³¹P{¹H} NMR (80.9 MHz,

CD₂Cl₂, 40 °C): δ = –20.3 (br. s) ppm. ¹³C{¹H} NMR (75.5 MHz,

CD₂Cl₂, 40 °C): δ = 39.5 (br., H₂C=CH₂), 36.5 (t, J(PC) = 9 Hz,

P(C₆H₁₁)₃), 29.4, 28.6 and 27.1 (3 s, P(C₆H₁₁)₃) ppm.

C₃₈H₇₀Br₂NOP₂Re (964.93): calcd. C 47.30, H 7.31, N 1.45; found

C 47.21, H 7.55, N 1.63.

Synthesis of [Re(NO)(PiPr₃)₂(MeCN)Me₂] (3a): To a solution of 1a

(0.081 g, 0.105 mmol) in 10 mL of toluene, cooled to –30 °C was

added rapidly MeLi 1.6  in Et₂O (132 µL, 0.210 mmol). The color

changed immediately from light orange to red. The mixture was

stirred at –30 °C for 30 min then allowed to come to room tempera-

ture and stirred for additional 30 min. The solution was filtered

through celite and the solvent was evaporated under vacuum. The

extraction with pentane affords the red beet viscous complex 3a

(0.078 g, 0.128 mmol, 82%). IR (CH Cl ): ν = 2360 (w, νMeCϵN),

˜

2

1588 (vs, νNO) cm^–1. H NMR (300.1 MHz, C₆D₆): δ = 2.61–2.43

(m, 6 H, (CH₃)₂CHP), 2.36 (t, J(PH) = 2.20 Hz, 3 H, ReMe), 2.29

(t, J(PH) = 2.03 Hz, 3 H, ReMe), 1.97 (s, 3 H, MeCNRe), 1.32–1.22

(m, 36 H, (CH₃)₂CHP) ppm. ³¹P{¹H} NMR (80.9 MHz, C₆D₆): δ

= –15.9 (s) ppm. ¹³C{¹H} NMR (125.8 MHz, C₆D₆): δ = 24.6 (t,

J(PC) = 11 Hz, (CH₃)₂CHP), 20.6 (br. s, ReMe), 20.5 (br. s, ReMe),

20.2 and 20.1 (2s,(CH₃)₂CHP) ppm.

1

Experimental Section

All operations were carried out under a nitrogen atmosphere using

a M. Braun 150 G-B glove box. The solvents were dried with so-

dium/benzophenone (THF, Et₂O, hydrocarbons) or P₂O₅(CH₂Cl₂,

CH₃CN) and distilled under N₂prior to use. The deuterated sol-

vents used in the NMR experiments were dried with sodium/benzo-

phenone (C₆D₆, [D₈]toluene, [D₈]THF) or P₂O₅(CD₂Cl₂, CD₃CN) Synthesis of [Re(NO)(PiPr₃)₂(MeCN)₂Br][BAr^F₄] (4a): A solid Na-

and vacuum transferred for storage in Schlenk flasks fitted with

Teflon valves. NMR experiments were carried out with Varian

Gemini 300, Varian Mercury 200, or Bruker DRX 500 spectrome-

BAr^F(0.131 g, 0.148 mmol) was added to a solution of 1a

4

(0.114 g, 0.148 mmol) dissolved in 10 mL of MeCN. The mixture

was stirred for 14 h at room temperature. The yellow solution was

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5257

H. Berke et al.

FULL PAPER

filtered and the solvent was removed under vacuum. Extraction by

CH₂Cl₂and recrystallization from CH₂Cl₂/pentane at room tem-

perature affords, after one day, yellow crystals of 4a (0.188 g,

Re, isomer A), 23.7 (d, J(CP) = 3 Hz, (HHC=CHH)Re, isomer B),

19.4 (s, (CH₃)₂CHP isomer A), 19.0 (s, (CH₃)₂CHP, isomer B) ppm.

C₁₅H₃₁NOPRe (458.59): calcd. C 39.28, H 6.81, N 3.05; found C

39.46, H 6.93, N 3.18.

1

0.120 mmol, 81%). IR (KBr): ν = 1712 (vs, νNO) cm^–1. H NMR

˜

(300.1 MHz, CD₂Cl₂): δ = 7.73 (m, 8 H, BAr^F₄), 7.56 (m, 4 H,

BAr^F₄), 2.89 (s, 6 H, MeCNRe), 2.83–2.71 (m, 6 H, (CH₃)₂CHP),

1.42–1.33 (m, 36 H, (CH₃)₂CHP) ppm. ³¹P{¹H} NMR

Synthesis of [Re(η⁴-C₄H₆)(η²-H₂C=CH₂)(NO)(PCy₃)₂] (6b): Com-

plex 6b (0.026 g, 0.045 mmol, 63%) was obtained by the same pro-

cedure described for 6a by reduction of 1b (0.070 g, 0.071 mmol)

(121.47 MHz, CD₂Cl₂):

δ =

0.4 (s) ppm. ¹³C{¹H} NMR

under a pressure of 1.4 bar of ethylene. IR (ATR): ν = 2924 and

˜

(125.8 MHz, CD₂Cl₂): δ = 162.0 (q, ¹J(BC) = 50 Hz, ipso-BAr^F),

137.7 (s, H₃CCNRe), 135.1 (s, o-BAr^F), 129.1 (br., m-BAr^F), 126.0

(q, ¹J(CF) = 50 Hz, CF₃), 117.8 (br., p-BAr^F₄), 24.8 (t, J(PC) =

11 Hz, (CH₃)₂CHP), 19.1 (s, (CH₃)₂CHP), 5.2 (s, H₃CCNRe) ppm.

2846 (vs, νC–H), 1643 (vs, νNO) cm^–1. ¹H NMR (500.25 MHz,

C₆D₆): δ = 5.76 (q, J(HH) = 6.4 Hz, 1 H, H_intC=CH_int), 5.29 (q,

³J(HH) = 6.4 Hz, 1 H, H_intC=CH_int), 2.40–1.05 (m, 37 H,

(H₂C=CH₂)Re and P(C₆H₁₁)₃), 2.01 (dd, ²J(H_syn2H_anti2) = 3.3,

J(HH) = 6.4 Hz, 1 H, H_syn2H_anti2CRe), 0.70 (td, J(HH) = 3.3,

J(HH) = J(PH) = 6.4 Hz, 1 H, H_syn2H_anti2CRe), –0.17 (m, 1 H,

¹⁹F

NMR

(125.8 MHz,

CD₂Cl₂):

δ

=

–64.4 ppm.

C₅₄H₆₀BBrF₂₄N₃OP₂Re (1561.91): calcd. C 41.95, H 3.91, N 2.72;

found C 42.03, H 4.01, N 2.67.

Synthesis of [Re(NO)(PCy₃)₂(MeCN)₂Br][BAr^F₄] (4b): 4b (0.078 g,

H

_syn1H_anti1CRe), –0.42 (q, J(HH) = J(PH) = 6.1 Hz, 1 H,

_syn1H_anti1CRe) ppm. ³¹P{¹H} NMR (121.5 MHz, C₆D₆): δ = 10.0

(s) ppm. ¹³C{¹H} NMR (125.8 MHz, C₆D₆):

δ

=

95.7 (s,

_intC=CH_int), 83.6 (s, H_intC=CH_int), 33.6 (d, J(CP) = 5 Hz,

_syn1H_anti1CRe), 32.2 (d, J(CP) = 19 Hz, P(C₆H₁₁)₃), 31.7 (d, J(CP)

0.042 mmol, 84%) was obtained by the same procedure described

for 4a by reaction of 1b (0.050 g, 0.051 mmol) and NaBAr^F

H

4

(0.045 g, 0.051 mmol) in a mixture of THF/MeCN (15:1 mL) at

room temperature after 30 h. IR (ATR): ν = 1703 (vs, νNO) cm^–1.

˜

= 13, P(C₆H₁₁)₃, 30.5 (d, J(CP) = 5, H_anti2H_syn2CRe), 29.8 (br. s,

P(C₆H₁₁)₃, 27.9 (br. s, P(C₆H₁₁)₃, 28.0 (d, J(CP) 9 Hz,

¹H NMR (300.1 MHz, CD₂Cl₂): δ = 2.94 (s, 6 H, CH₃CN-Re),

2.57–1.28 (m, 66 H, P(C₆H₁₁)₃) ppm. ³¹P{¹H} NMR (121.47 MHz,

CD₂Cl₂): = –10.2 (s) ppm. ¹³C{¹H} NMR (75.5 MHz, CD₂Cl₂): δ

= 164.0 (m, ipso-BAr^F), 137.7 (s, H₃CCNRe), 135.2 (s, o-BAr^F),

129.4 (br., m-BAr^F), 129.0 (m, CF₃), 118.0 (br., p-BAr^F₄), 35.2 (t,

J(PC) = 10 Hz, P(C₆H₁₁)₃), 29.6, 28.9 and 26.8 (3 s, P(C₆H₁₁)₃), 5.7

=

P(C₆H₁₁)₃), 26.9 (br. s, P(C₆H₁₁)₃), 25.3 (d, J(CP) = 8 Hz,

(HHC=CHH)Re), 25.1 (d, J(CP) = 3 Hz, (HHC=CHH)Re) ppm.

C₂₄H₄₃NOPRe (578.78): calcd. C 49.80, H 7.48, N 2.42; found C

50.02, H 7.61, N 2.26.

[Re(NO)(CO)(PiPr₃)₂(η²-C₂H₄)] (8a): In a 60 mL Young tap

Schlenk tube 7b (0.100 g, 0.157 mmol) was dissolved in 30 mL of

THF and 10 g of 0.7% sodium amalgam was added. Without stir-

ring, the solution was frozen and kept under pressure of 1.4 bar of

C₂H₄for 5 min. The mixture was warmed to room temperature,

and stirred for 5 d. The mixture was filtered through celite and

the solvent was removed under vacuum. The orange residue was

extracted with toluene and washed with cold pentane to afford pure

8a (0.081 g, 0.136 mmol, 87%). Suitable X-ray crystals were ob-

tained from a solution of 8a in 3 mL of pentane at –30 °C. IR

(s, CH₃CN-Re) ppm. ¹⁹F NMR (125.8 MHz, CD₂Cl₂):

δ =

–63.9 ppm. C₇₂H₈₄BBrF₂₄N₃O₂P₂Re (1818.28): calcd. C 47.56, H

4.65, N 2.31; found C 47.71, H 4.67, N 2.20.

Synthesis of [Re(η⁴-C₄H₆)(η²-H₂C=CH₂)(NO)(PiPr₃)₂] (6a): In a

50 mL Young tap Schlenk tube the complex [Re(NO)(MeCN)-

(PiPr₃)₂Br₂] (1a) (0.100 g, 0.135 mmol) was dissolved in 20 mL of

THF and 10 g of 0.7% sodium amalgam was added. Without stir-

ring, the solution was frozen and kept under pressure of 1.4 bar of

C₂H₄for 5 min. The mixture was warmed to room temperature,

and stirred for 3 d. Filtration over celite and removal of the solvent

left a yellow residue, which was extracted with toluene and then

dried. Final extraction with pentane affords a pure pale yellow

(ATR): ν = 2963, 2927 and 2872 (s, νC–H), 1879 (s, νCO), 1576 (s,

˜

1

νNO) cm^–1. H NMR (300.1 MHz, C₆D₆, room temp.): δ = 2.16–

1.84 (m, 6 H, (CH₃)₂CHP), 1.38–1.19 (m, 36 H, (CH₃)₂CHP) ppm.

1

powder of 6a (0.020 g, 0.042 mmol, 31%). IR (ATR): ν = 2964,

˜

C₂H₄not observed. H NMR (300.1 MHz, [D₈]toluene, 100 °C): δ

2930 and 2872 (s, νC–H), 1646 (vs, νNO) cm^–1. ¹H NMR

= 2.42–1.31 (m, (CH₃)₂CHP partly overlap with signal of toluene

CH₃), 1.86 (br. t, 4 H, H₂C=CH₂), 1.25–1.09 (m, 36 H, (CH₃)₂-

CHP) ppm. ³¹P{¹H} NMR (121.5 MHz, C₆D₆, room temp.): δ =

13.4 (s) ppm. ¹³C{¹H} NMR (75.5 MHz, [D₈]toluene, 90 °C): δ =

218.4 (t, J(PC) = 12 Hz, C=O), 25.1 (t, J(PC) = 23 Hz, (CH₃)₂-

CHP), 23.0 (br. s, H₂C=CH₂), 20.6 and 19.3 (2 s, (CH₃)₂-

CHP) ppm. C₂₁H₄₆NO₂P₂Re (592.75): calcd. C 42.55, H 7.82, N

2.36; found C 42.69, H 8.12, N 2.35.

(500.25 MHz, C₆D₆):

δ = 5.72 (q, J(HH) = 6.7 Hz, 1 H,

H_intC=CH_int), 5.25 (q, J(HH) = 6.7 Hz, 1 H, H_intC=CH_int), 2.20

(m, 1 H, (HHC=CHH)Re), 2.06 (m, 0.34 H, H_syn2H_anti2CRe, iso-

mer B), 1.98 (dd, ³J(HH) = 6.7, ²J(HH) = 3.5 Hz, 0.66 H,

H_syn2H_anti2CRe, isomer A), 1.82 (m,

5 H, (CH₃)₂CHP and

(HHC=CHH)Re), 1.48 (m, 1 H, (HHC=CHH)Re), 1.18–0.98 and

3

0.95–0.87 (2 m, 18 H, (CH₃)₂CHP), 0.56 (td, J(HH) = J(PH) =

2

6.7 Hz, J(HH) = 3.5 Hz, 0.66 H, H_syn2H_anti2CRe, isomer A), 0.43

[Re(NO)(CO)(PCy₃)₂(η²-C₂H₄)] (8b): Complex 8b (0.086 g,

0.103 mmol, 90%) was obtained by the same procedure as de-

scribed for 8a via reduction of [Re(NO)(CO)(PCy₃)₂Cl₂] (0.100 g,

0. 114 mmol) using 10 g of 0.7% sodium amalgam and 1.4 bar

(td, ³J(HH) = ³J(PH) = 6.7 Hz, ²J(HH) = 3.5 Hz, 0.34 H,

H_syn2H_anti2CRe, isomer B), –0.21 (m, 1 H, H_syn1H_anti1CRe), –0.52

(q, J(HH) = J(PH) = 6.7 Hz, 0.66 H, H_syn1H_anti1CRe, isomer A),

–0.72 (q, J(HH) = J(PH) = 6.7 Hz, 0.34 H, H_syn1H_anti1CRe, isomer

B) ppm. ³¹P{¹H} NMR (121.5 MHz, C₆D₆): δ = 21.3 (br. s, isomer

A), 12.9 (br. s, isomer B) ppm. ¹³C{¹H} NMR (125.8 MHz, C₆D₆):

δ = 97.5 (d, J(CP) = 2 Hz, H_intC=CH_int, isomer B), 96.14 (d, J(CP)

= 2 Hz, H_intC=CH_int, isomer A), 83.0 (br. s, H_intC=CH_int, isomer

A), 81.97 (br. s, H_intC=CH_int, isomer B), 34.0 (d, J(CP) = 5 Hz,

C H . IR (ATR): ν = 2925 and 2848 (s, νCO), 1883 (s, νCO), 1585

˜

2

4

(vs, νNO) cm^–1. ¹H NMR (300.1 MHz, C₆D₆, room temp.): δ =

2.17–1.17 (H₂C=CH₂and P(C₆H₁₁)₃) ppm. ³¹P{¹H} NMR

(80.9 MHz, C₆D₆): δ = 5.7 (s) ppm. ¹³C{¹H} NMR (75.5 MHz,

[D₈]toluene, –20 °C): δ = 218.5 (t, J(PC) = 12 Hz, C=O), 34.4 (br.,

P(C₆H₁₁)₃), 30.7, 30.3, 27.3 (3 s, P(C₆H₁₁)₃), 23.0 (br. s,

H₂C=CH₂) ppm. C₃₉H₇₀NO₂P₂Re (833.13): calcd. C 56.22, H 8.47,

N 1.68; found C 56.67, H 8.59, N 1.61.

H_syn1H_anti1CRe, isomer B), 32.8 (d, J(CP) = 5 Hz, H_syn1H_anti1CRe,

isomer A), 30.5 (d, J(CP) = 6 Hz, H_anti2H_syn2CRe, isomer A), 29.7

(d, J(CP) = 6 Hz, H_anti2H_syn2CRe, isomer B), 29.2 (br. s, (CH₃)₂-

CHP, isomer A), 29.0 (br. s, (CH₃)₂CHP, isomer B), 25.5 (d, J(CP) NMR Tube Reactions. Synthesis of [Re(η²-H₂)(NO)(PR₃)₂Br₂] (R

= 8 Hz, (HHC=CHH)Re, isomer A), 24.9 (d, J(CP) = 8 Hz,

= iPr 5a, R = Cy 5b): A solution of 1a (0.010 g, 0.013 mmol) in

(HHC=CHH)Re, isomer B), 24.7 (d, J(CP) = 3 Hz, (HHC=CHH)- 0.5 mL C₆D₆or 1b (0.012 g, 0.012 mmol) in 0.5 mL of CD₂Cl₂was

5258

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

sealed under 1 bar of H₂. Formation of 5a,b takes place immedi-

ately upon shaking of the tubes. All NMR spectroscopic data are

in agreement with those previously reported.^[5]

1 mL of toluene was added dropwise with stirring to a solution of

8a (0.056 g, 0.094 mmol) in the same solvent. The mixture was

stirred further at room temperature for 22 h. The solvent was then

removed in vacuo and the residue extracted with pentane and

recrystallized from the same solvent at –30 °C to afford orange

Reaction of [Re(NO)(CO)(PiPr₃)₂(η²-C₂H₄)] (8a) with B(C₆F₅)₃to

Yield [Re(CO)₂{NOB(C₆F₅)₃}(PiPr₃)₂] (9a) and [Re(NO)(CO)₂- crystals of 9a; yield 0.039 g, 0.036 mmol, 38%. 9a and 10a were

(PCy₃)₂] (10a): A solution of B(C₆F₅)₃(0.048 g, 0.094 mmol) in identified by their IR spectra: 9a IR (ATR): ν = 1968 (s, νCO),

˜

Table 8. Summary of crystallographic data for complexes 4a, 4b, 6b, 8a, 9a, and 9b.

4a

4b

6b

Empirical formula

Molecular weight

C₅₄H₆₀BBrF₂₄N₃OP₂Re

1561.91

C₇₂H₈₄BBrF₂₄N₃OP₂Re·CH₂Cl₂C₂₄H₄₃NOPRe

1887.21

578.76

Color

Crystal system

Space group

yellow

monoclinic

C2/c

yellow

orthorhombic

Pna2₁

pale yellow

monoclinic

P2₁/c

a [Å]

b [Å]

c [Å]

α [°]

20.330(2)

15.0650(11)

21.790(2)

90

33.849(2)

12.3816(11)

19.0254(18)

90

14.1035(9)

11.3650(6)

17.4160(12)

90

β [°]

109.124(11)

90

6305.4(10)

4

1.645

2.722

90

121.264(7)

90

2386.2(3)

4

1.611

5.174

γ [°]

V [Å³]

7973.7(12)

4

1.572

Z

ρ_calcd.[g cm^–3]

µ [mm^–1]

2.232

Transmission range

Crystal size [mm]

hkl limiting indices

F(000)

0.7098–0.5185

0.15ϫ0.21ϫ0.27

–26/25, Ϯ19, Ϯ28

3096

2.15–28.05

26224

7594

415

0.0356, 0.0721

0.0829, 0.0883

0.831

0.7135–0.4660

0.18ϫ0.34ϫ0.43

–38/40, Ϯ15, Ϯ23

3792

1.75–25.92

53039

15066

960

0.0467, 0.0565

0.1182, 0.1237

0.987

0.5464–0.3401

0.22 ϫ 0.21 ϫ 0.13

–20/17, Ϯ16, Ϯ24

1168

2.96–30.43

51454

7139

272

0.0264, 0.0374

0.0744, 0.0752

0.971

θ limits [°]

Number of measd. reflections

Number of unique reflections

Number of parameters

[a]

R₁[I Ͼ 2σ(I)], all data

[b]

wR₂[I Ͼ σ(I)], all data

GOF (for F²)

∆ρ_max/min

–1.091, 1.576

–1.130, 1.686

–0.670, 2.011

8a

9a

9b

Empirical formula

Molecular weight

C₂₁H₄₆NO₂P₂Re

592.74

C₄₃H₅₄BF₁₅NO₃P₂Re

1176.83

C₅₆H₆₆BF₁₅NO₃P₂Re

1345.06

Color

Crystal system

Space group

orange

monoclinic

P2₁/n

orange

triclinic

¯

P1

a [Å]

b [Å]

c [Å]

α [°]

11.9641(13)

14.0278(10)

15.4531(16)

90.00

100.370(13)

90.00

2551.1(4)

4

1.543

11.2503(17)

12.818(2)

17.353(3)

80.100(19)

85.913(19)

75.505(18)

2385.6(7)

2

12.5747(8)

14.6063(10)

16.5491(10)

86.834(8)

88.538(8)

67.138(7)

2796.5(3)

2

β [°]

γ [°]

V [Å³]

Z

ρ_calcd.[g cm^–3]

1.638

2.712

1.597

2.324

µ [mm^–1]

4.904

Transmission range

Crystal size [mm]

hkl limiting indices

F(000)

0.1262–0.5018

0.14ϫ0.48ϫ0.57

Ϯ16, Ϯ19, Ϯ21

1200

3.20–30.39

29650

7583

260

0.0267, 0.0380

0.0685, 0.0706

1.061

0.652–0.407

0.27ϫ0.28ϫ0.44

–15/16, –17/18, Ϯ24

1176

2.79–30.39

51447

13094

575

0.0522, 0.0635

0.1412, 0.1478

1.081

0.666-0.845

0.17 ϫ 0.13 ϫ 0.08

–14/14, –17/17, Ϯ19

1356

2.83–30.39

43505

15287

712

0.0362, 0.0487

0.1262, 0.1277

1.067

θ limits [°]

Number of measd. reflections

Number of unique reflections

Number of parameters

[a]

R₁[I Ͼ 2σ(I)], all data

[b]

wR₂[I Ͼ σ(I)], all data

GOF (for F²)

∆ρ_max/min

–2.543, 1.111

–7.462, 4.098

1.788, –0.579

[a] R₁= ∑(F_o– F_c)/∑F_o; I Ͼ 2 σ(I). [b] wR₂= {∑w(F_o– F_c) /∑w(F_o²)²}^1/2

.

2

2 2

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5259

H. Berke et al.

FULL PAPER

1883 (s, νCO), 1458 (s, νNO) cm^–1. 10a IR (hexane): ν = 1940 (s,

ment details are given in the “_exptl_special_details” in the respec-

tive cif files.

˜

νCO), 1852 (s, νCO), 1605 (s, νNO) cm^–1.

Computational Details: DFT calculations were carried out using

the Amsterdam Density Functional program package ADF, release

2004.01.^[57]We used the Vosko–Wilk–Nusair^[66]local density

approximation (LDA) and the generalized gradient approximation

(GGA) with corrections for exchange and correlation according

to Becke^[67]and Lee–Yang–Parr,^[68]respectively (BLYP). The ADF

approach to DFT-GGA calculations is based on the use of Slater-

type orbitals (STO) as basis functions. The valence shells of all

atoms were described by double-ξ basis sets augmented by one po-

larization function (ADF database DZP), except for rhenium for

which a triple-ξ basis set was applied (ADF database TZP). The

frozen-core approximation was applied for the 1s electrons of car-

bon, oxygen, and nitrogen, for the 1s-2p electrons of phosphorus

and chlorine, for the 1s-3d electrons of bromine and for the 1s-4f

electrons of rhenium. Relativistic effects were taken into account

in all calculations using the zero order regular approximation

(ZORA).^[69]The relativistic atomic potentials necessary for the rel-

ativistic calculations for each atom were calculated using the auxil-

iary program DIRAC, which is supplied with the ADF program

package. The geometry optimizations were considered converged

when the change in the maximum gradient element was smaller

than 5.0ϫ10^–4HartreesÅ^–1. The other convergence criteria refer-

ring to changes in total energy, Cartesian or internal coordinates

had to converge to the default values proposed by the program.

Synthesis of [Re(CO)₂{NOB(C₆F₅)₃}(PiPr₃)₂] (9a): A solution of

B(C₆F₅)₃(0.064 g, 0.124 mmol) in 2 mL of toluene was added to a

solution of 8a (0.074 g, 0.124 mmol) in 15 mL of the same solvent

in a Young tap Schlenk tube. Without stirring the mixture was

frozen and kept under pressure of 1 bar of CO for 3 min. The mix-

ture was warmed to room temperature and stirred for 15 min. The

gas and the solvent were removed in vacuo and the orange residue

was extracted with CH₂Cl₂and washed several times using cold

pentane to afford pure 9a (0.126 g, 0.114 mmol, 92%). IR (ATR):

ν = 1968 (s, νCO), 1883 (s, νCO), 1458 (s, νNO) cm^–1. ¹H NMR

˜

(300.1 MHz, [D₈]toluene): δ = 0.92–0.80 (m, 6 H, (CH₃)₂CHP),

1.82–1.71 (m, 36 H, (CH₃)₂CHP) ppm. ³¹P{¹H} NMR

(121.5 MHz, [D₈]toluene): δ = 32.3 (s) ppm. ¹³C{¹H} NMR

(125.8 MHz, [D₈]toluene): δ = 215.9 (br., C=O), 24.0 (t, J(PC) =

11 Hz, (CH₃)₂CHP), 18.6 and 18.4 (2 s, (CH₃)₂CHP) ppm.

C₃₈H₄₂BF₁₅NO₃P₂Re (1104.68): calcd. C 41.31, H 3.83, N 1.27;

found C 41.75, H 3.78, N 1.30.

Synthesis of [Re(CO)₂{NOB(C₆F₅)₃}(PCy₃)₂] (9b): Complex 9b

(0.088 g, 0.066 mmol, 90%) was obtained using the same procedure

as described for 9a (method B) via the reaction of 8b (0.062 g,

0.074 mmol), B(C₆F₅) (0.038 g, 0.074 mmol) under 1 bar of CO. IR

(ATR): ν = 1974 (s, νCO), 1875 (s, νCO), 1457 (s, νN–OB) cm^–1.

˜

¹H NMR (200.0 MHz, C₆D₆, room temp.): δ = 2.20–0.81 (m,

P(C₆H₁₁)₃) ppm. ³¹P{¹H} NMR (80.9 MHz, C₆D₆): δ = 24.6

(s) ppm. ¹³C{¹H} NMR (75.5 MHz, C₆D₆): δ = 211.5 (br., C=O),

37.7 (t, J(PC) = 13 Hz, P(C₆H₁₁)₃, 30.5, 27.4 and 26.0 (3s,

P(C₆H₁₁)₃) ppm. C₅₆H₆₆BF₁₅NO₃P₂Re (1345.07): calcd. C 50.00, H

4.94, N 1.04; found C 50.09, H 5.03, N 1.00.

CCDC-297768 to -297771 (for 4a, 4b, 8a, 9a) and -609682

to -609683 (for 6b and 9b) contain the supplementary crystallo-

graphic data for this paper. These data can be obtained free of

charge from the Cambridge Crystallographic Data Centre via

www.ccdc.cam.ac.uk/data_request/cif.

X-ray Structure Analyses of Compounds 4a, 4b, 6b, 8a, 9a, and 9b:

All six crystals were protected in hydrocarbon oil and prepared for

the X-ray experiment by using a polarizing microscope. Selected Acknowledgments

crystals of 4a, 4b, 6b, 8a, 9a, and 9b (Table 8) were mounted on the

tip of a glass fiber and immediately transferred to the goniometer

of an imaging plate detector system (Stoe IPDS diffractometer).

The crystals were cooled to 183(2) K and 193(2) K (8a) in an Ox-

ford Cryogenic System. The crystal-to-image distances were set to

60, 70, 50, 50, 50, and 50 mm resulting in θ_maxvalues of 28.05,

25.92, 30.43, 30.39, 30.39, and 30.39°. For compounds 8a, 9a, and

9b rotation, and for 4a, 4b, and 6b oscillation scan modes were

applied. 8000 (5000 for 9a) reflections with IϾ6σ(I) were selected

for the cell parameter refinements. A total of 26224 (4a), 53039

(4b), 51454 (6b), 29650 (8a), 51447 (9a), and 43505 (9b) diffraction

intensities were collected of which 7594, 15066, 7139, 7583, 13094,

and 15287 were unique (R_int= 6.52, 8.59, 6.13, 5.01, 14.04, and

8.64%) after data reduction. Numerical absorption corrections

based on 12, 17, 8, 10, 15, and 7 crystal faces were applied with

FACEitVIDEO and XRED.^[60]The structures were solved by the

Patterson (for 4a, 6b, 8a, 9a, 9b) and by direct methods (for 4b)

using the SHELXS-97 program.^[61]Interpretation of the difference

electron density maps, preliminary plot generation, and checking

for higher symmetry were done with PLATON^[62](1990, 1991–

2005) and with the LEPAGE program.^[63]All structures were re-

fined with the program SHELXL-97^[61]using anisotropic displace-

ment parameters for heavy atoms, some exceptions in disordered

structures 4a and 4b are given in the CIF files (see CCDC deposi-

tion numbers, see below). Positions of H-atoms were calculated

after each refinement cycle (riding model). Structural plots were

generated using ORTEP III.^[64]For the achiral non-centrosymmet-

ric structure of 4b the correct space group choice (Pna2₁, no. 33)

was confirmed by Flack’s parameter x^[65]and PLATON.^[62]Refine-

This work has been supported by the Swiss National Science Foun-

dation and the Funds of the University of Zurich.

[1] a) W. R. Thiel, Angew. Chem. Int. Ed. 2003, 42, 5390; b) K. P.

Gable, J. J. J. Juliette, J. Am. Chem. Soc. 1996, 118, 2625; c)

A. M. LaPointe, R. R. Schrock, Organometallics 1995, 14,

1875.

[2] W. D. Jones, J. A. Maguire, G. P. Rosini, Inorg. Chim. Acta

1998, 270, 77.

[3] A. Kovacs, G. Frenking, Organometallics 2001, 20, 2510.

[4] a) J. K. Gibson, J. Organomet. Chem. 1998, 558, 51; b) L. A.

Friedman, S. H. Meiere, B. C. Brooks, W. D. Harman, Organo-

metallics 2001, 20, 1690; c) T. B. Gunnoe, W. D. Sabat, M. Har-

man, Organometallics 2000, 19, 728; d) A. S. Polo, M. K. Itok-

azu, K. M. Frin, A. O. D. Patrocinio, N. Y. M. Iha, Coord.

Chem. Rev. 2006, 250, 1669; e) O. V. Ozerov, J. C. Huffman,

L. A. Caulton, K. G. Watson, Organometallics 2003, 22, 2539;

f) O. V. Ozerov, J. C. Huffman, L. A. Watson, M. Pink, K. G.

Caulton, J. Am. Chem. Soc. 2007, 129, 6003.

[5] D. Gusev, A. Llamazares, G. Artus, H. Jacobsen, H. Berke,

Organometallics 1999, 18, 75.

[6] a) C. M. Frech, O. Blacque, H. Berke, Pure Appl. Chem. 2006,

78, 1877; b) C. M. Frech, O. Blacque, H. W. Schmalle, H.

Berke, C. Adlhart, P. Chen, Chem. Eur. J. 2006, 12, 3325.

[7] G. Kubas, J. Organomet. Chem. 2001, 635, 37.

[8] H. U. Hund, U. Ruppli, H. Berke, Helv. Chim. Acta 1993, 76,

963.

[9] G. Mercati, F. Morazzoni, F. Cariati, D. Giusto, Inorg. Chim.

Acta 1978, 29, 165.

[10] S. J. Mclain, R. R. Schrock, J. Am. Chem. Soc. 1978, 100, 1315.

5260

www.eurjic.org

Eur. J. Inorg. Chem. 2007, 5246–5261

Olefin Complexes of Low-Valent Rhenium

[11] S. J. Mclain, C. D. Wood, R. R. Schrock, J. Am. Chem. Soc.

1979, 101, 4558.

[12] F. Speiser, P. Braunstein, W. Saussine, Acc. Chem. Res. 2005,

38, 784.

[13] J. D. Fellmann, G. A. Rupprecht, R. R. Schrock, J. Am. Chem.

Soc. 1979, 101, 5099.

[14] J. T. Dixon, M. J. Green, F. M. Hess, D. H. Morgan, J. Or-

ganomet. Chem. 2004, 689, 3641.

[15] D. J. Marais, N. Sheppard, B. P. Stoicheff, Tetrahedron 1962,

17, 163.

[16] H. Yasuda, Y. Kajihara, K. Mashima, K. Nagasuna, K. Lee,

A. Nakamura, Organometallics 1982, 1, 388.

[17] C. Holtke, G. Erker, G. Kehr, R. Fröhlich, O. Kataeva, Eur. J.

Inorg. Chem. 2002, 2789.

[18] H. Yasuda, K. Tatsumi, T. Okamoto, K. Mashima, K. Lee, A.

Nakamura, Y. Kai, N. Kanehisa, N. Kasai, J. Am. Chem. Soc.

1985, 107, 2410.

[43] P. R. Sharp, Organometallics 1984, 3, 1217.

[44] F. M. Su, C. Cooper, S. J. Geib, A. L. Rheingold, J. M. Mayer,

J. Am. Chem. Soc. 1986, 108, 3545.

[45] G. R. Clark, A. J. Nielson, A. D. Rae, C. E. F. Rickard, J.

Chem. Soc. Chem. Commun. 1992, 1069.

[46] T. Aoshima, T. Tamura, Y. Mizobe, M. Hidai, J. Organomet.

Chem. 1992, 435, 85.

[47] M. D. Butts, J. C. Bryan, X. L. Luo, G. J. Kubas, Inorg. Chem.

1997, 36, 3341.

[48] S. Y. S. Wang, D. D. VanderLende, K. A. Abboud, J. M. Bon-

cella, Organometallics 1998, 17, 2628.

[49] H. Ishino, S. Tokunaga, H. Seino, Y. Ishii, M. Hidai, Inorg.

Chem. 1999, 38, 2489.

[50] E. Carmona, J. M. Marin, M. L. Poveda, J. L. Atwood, R. D.

Rogers, J. Am. Chem. Soc. 1983, 105, 3014.

[51] E. Carmona, A. Galindo, J. M. Marin, E. Gutierrez, A.

Monge, C. Ruiz, Polyhedron 1988, 7, 1831.

[19] T. A. Albright, J. K. Burdett, M. H. Whangbo, Orbital Interac-

tions in Chemistry, Wiley, New York, 1985.

[20] K. Bachmann, W. v. Philipsborn, Org. Magn. Reson. 1976, 8,

648.

[21] B. Machura, Coordin. Chem. Rev. 2005, 249, 2277.

[22] H. PtasiewiczBak, J. Leciejewicz, Pol. J. Chem. 1997, 71, 493.

[23] C. Pellecchia, A. Grassi, A. Immirzi, J. Am. Chem. Soc. 1993,

115, 1160.

[52] S. Komiya, A. Baba, Organometallics 1991, 10, 3105.

[53] J. Matthes, S. Grundemann, A. Toner, Y. Guari, B. Donnadieu,

J. Spandl, S. Sabo-Etienne, E. Clot, H. H. Limbach, B. Chaud-

ret, Organometallics 2004, 23, 1424.

[54] J. S. Dewar, Bull. Soc. Chim. Fr. 1951, 18, C71.

[55] J. Chatt, L. A. Duncanson, J. Chem. Soc. 1953, 2939.

[56] J. Uddin, S. Dapprich, G. Frenking, B. F. Yates, Organometal-

lics 1999, 18, 717 and references cited therein.

[24] G. Kehr, R. Roesmann, R. Fröhlich, C. Holst, G. Erker, Eur.

J. Inorg. Chem. 2001, 535.

[25] P. Yang, B. C. K. Chan, M. C. Baird, Organometallics 2004, 23,

2752.

[26] H. C. Strauch, G. Erker, R. Fröhlich, M. Nissinen, Eur. J. In-

org. Chem. 1999, 1453.

[27] Y. M. Sun, R. E. V. H. Spence, W. E. Piers, M. Parvez, G. P. A.

Yap, J. Am. Chem. Soc. 1997, 119, 5132.

[57] ADF2004.01. for an overview see: G. Frenking, N. Fröhlich,

The Nature of the Bonding in Transition-Metal Compounds,

Chem. Rev. 2000, 100, SCM, Theoretical Chemistry, Vrije Uni-

versiteit, Amsterdam, The Netherlands, http://ww w .scm.com;

a) F. M. Bickelhaupt, E. J. Baerends, C. F. Guerra, S. J. A.

Van Gisbergen, J. G. Snijders, T. Ziegler, J. Comput. Chem.

2001, 22, 931; b) C. F. Guerra, J. G. Snijders, G. te Velde, E. J.

Baerends, Theor. Chem. Acc. 1998, 99, 391.

[28] T. Wondimagegn, Z. T. Xu, K. Vanka, T. Ziegler, Organometal-

lics 2004, 23, 3847.

[29] H. Jacobsen, T. Brackemeyer, H. Berke, G. Erker, R. Fröhlich,

Eur. J. Inorg. Chem. 2000, 1423.

[30] Y. M. Sun, W. E. Piers, S. Rettig, J. Chem. Commun. 1998, 127.

[31] J. W. Strauch, G. Erker, G. Kehr, R. Fröhlich, Angew. Chem.

Int. Ed. 2002, 41, 2543.

[58] F. M. Bickelhaupt, E. J. Baerends in Reviews of Computational

Chemistry (Eds.: D. B. Boyd, K. B. Lipkowitz), Wiley-VCH,

New York, 2000, p. 1, vol. 15.

[59] P. Coppens, L. Leiserowitz, D. Rabinovich, Acta Crystallogr.

1965, 18, 1035.

[60] STOE-IPDS Software package, version 2.92/1999, STOE &

Cie, Darmstadt, Germany, 1999.

[32] H. Jacobsen, H. Berke, S. Doring, G. Kehr, G. Erker, R.

Fröhlich, O. Meyer, Organometallics 1999, 18, 1724.

[33] W. J. Huang, H. Berke, Chimia 2005, 59, 113.

[34] J. C. Peters, G. L. Hillhouse, Polyhedron 1994, 13, 1741.

[35] A. A. H. Van der Zeijden, C. Sontag, H. W. Bosch, V. Shklover,

H. Berke, D. Nanz, W. v. Philipsborn, Helv. Chim. Acta 1991,

74, 1194.

[61] G. M. Sheldrick, SHELX97, University of Göttingen,

Göttingen, Germany, 1998.

[62] a) A. L. Spek, Acta Crystallogr., Sect. A 1990, 46, C34; b) A. L.

Spek, PLATON,

A Multipurpose Crystallographic Tool,

Utrecht University, Utrecht, The Netherlands, 1998.

[63] Y. Lepage, J. Appl. Crystallogr. 1987, 20, 264.

[64] L. J. Farrugia, J. Appl. Cryst. 1997, 30, 565.

[36] K. E. Lee, A. M. Arif, J. A. Gladysz, Inorg. Chem. 1990, 29, [65] H. D. Flack, Acta Crystallogr., Sect. A 1983, 39, 876.

2885.

[66] S. H. Vosko, L. Wilk, M. Nusair, Can. J. Phys. 1980, 58, 1200.

[67] A. D. Becke, Phys. Rev. A 1988, 38, 3098.

[37] B. S. Tovrog, S. E. Diamond, F. Mares, A. Szalkiewicz, J. Am.

Chem. Soc. 1981, 103, 3522.

[38] S. D. Ittel, Adv. Organomet. Chem. 1976, 14, 33.

[39] M. R. Churchill, H. J. Wasserman, Inorg. Chem. 1981, 20,

4119.

[68] C. Yang, W. Parr, R. G. Lee, Phys. Rev. B 1988, 37, 785.

[69] a) E. van Lenthe, A. E. Ehlers, E. J. Baerends, J. Chem. Phys.

1999, 110, 8943; b) E. van Lenthe, E. J. Baerends, J. G. Snijders,

J. Chem. Phys. 1993, 99, 4597; c) E. van Lenthe, E. J. Baerends,

J. G. Snijders, J. Chem. Phys. 1994, 101, 9783; d) E. van Lenthe,

J. G. Snijders, E. J. Baerends, J. Chem. Phys. 1996, 105, 6505;

e) E. van Lenthe, R. van Leeuwen, E. J. Baerends, J. G.

Snijders, J. Quant. Chem. 1996, 57, 281.

[40] U. Radius, J. Sundermeyer, H. Pritzkow, Chem. Ber. 1994, 127,

1827.

[41] P. W. Dyer, V. C. Gibson, W. Clegg, J. Chem. Soc. Dalton Trans.

1995, 3313.

[42] L. D. Brown, C. F. J. Barnard, J. A. Daniels, R. J. Mawby, J. A.

Ibers, Inorg. Chem. 1978, 17, 2932.

Received: June 1, 2007

Published Online: October 1, 2007

Eur. J. Inorg. Chem. 2007, 5246–5261

www.eurjic.org

5261

Article Doi

Olefin complexes of low-valent rhenium

DOI: 10.1002/ejic.200700573

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.200700573

Full text of DOI:10.1002/ejic.200700573

Products guided by the article

R&D Labs maybe for 376350-36-6

Relevant to this article

Hot Product