University of Groningen Chemistry of vanadium-carbon single ...In this chapter the synthesis and reactivity of vanadium(III) complexes, containing sterically demanding alkyl groups

University of Groningen

Chemistry of vanadium-carbon single and double bondsBuijink, Jan Karel Frederik

IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite fromit. Please check the document version below.

Document VersionPublisher's PDF, also known as Version of record

Publication date:1995

Link to publication in University of Groningen/UMCG research database

Citation for published version (APA):Buijink, J. K. F. (1995). Chemistry of vanadium-carbon single and double bonds. s.n.

CopyrightOther than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of theauthor(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).

Take-down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons thenumber of authors shown on this cover page is limited to 10 maximum.

Download date: 14-06-2021

https://research.rug.nl/en/publications/chemistry-of-vanadiumcarbon-single-and-double-bonds(8b4d3620-28b4-425b-b90b-0be269d322eb).html

* This work has been performed in collaboration with R. Zijlstra. Part of this work has been performed at the

California Institute of Technology under supervision of Prof. R. H. Grubbs. Part of this chapter has been

communicated: Buijink, J.-K. F.; Meetsma, A.; Teuben, J. H. Organometallics 1993, 12, 2004-2005.

49

Chapter 2__________________________________________________________

Homoleptic Alkyl Complexes of Vanadium(III)*__________________________________________________________

2.1 Introduction

Sterically demanding alkyl groups lacking ß-hydrogens (e.g. neopentyl, neophyl) have

played an important role in the development of transition metal olefin metathesis1 and ring-

opening metathesis polymerization2 (ROMP) catalysts. Transition metal complexes containing

these alkyl groups are known to decompose thermally through α-hydrogen abstraction to givemetal alkylidenes.3 A relatively small number of vanadium complexes incorporating these alkyl

groups are known,4 with V[CH(SiMe3)2]34c being the only example of a homoleptic V(III)

alkyl complex. The thermal decomposition of these compounds has been investigated only in

the case of the monocyclopentadienyl bis(neopentyl) complex CpV(CH2-t-Bu)2PMe3.4m The

vanadium alkylidene species produced in the thermal decomposition of this compound are

active as catalysts for ROMP of norbornene, although the activity of the isolated alkylidene

CpV(CH-t-Bu)dmpe is low (vide infra).

In this chapter the synthesis and reactivity of vanadium(III) complexes, containing

sterically demanding alkyl groups lacking β-hydrogens, but without stabilizingcyclopentadienyl ligands is described. A special section is reserved for the activity of these

compounds in olefin metathesis reactions.

2.2 Attempted synthesis of homoleptic alkyl complexes of vanadium(III).

Reaction of VCl3(THF)3 with three equivalents of t-BuCH2Li in diethyl ether under

nitrogen does not yield the expected homoleptic tris(neopentyl) vanadium complex, but

instead, the bridged dinitrogen complex [(t-BuCH2)3V]2(µ-N2) (1), which was isolated in

45% yield (eq 1). Compound 1 is a red-brown crystalline solid, poorly soluble in aliphatic and

aromatic hydrocarbons and extremely air sensitive (pyrophoric). To determine the molecular

structure of 1 an X-ray structure determination was carried out. The crystal structure of 1

involves the packing of 3 molecules in the hexagonal unit cell. The asymmetric unit contains

Chapter 2: Homoleptic alkyl complexes of vanadium(III).

49

one neopentyl fragment and a V and a N atom at a 3-fold axis. The molecular structure of 1 is

shown in Fig. 1. Selected bond lengths and angles are given in Table I.

(1)V N N V

t-Bu

t-But-Bu t-Bu

t-But-Bu

1

VCl3(THF)3 + 3 t-BuCH2Li - 3 LiCl, 3 THFN2, Et2O

1/2

Table I. Selected geometrical data for [(t-BuCH2)3V]2(µ-N2) (1).

Bond lengths (Å)

V(1)-N(1) 1.7248(18)

V(1)-C(1) 2.0262(16)

N(1)-N(1)c 1.250(3)

C(1)-C(2) 1.532(2)

C(2)-C(3) 1.525(2)

C(2)-C(4) 1.529(2)

C(2)-C(5) 1.531(2)

Bond angles (°)

N(1)-V(1)-C(1) 109.70(4)

C(1)-V(1)-C(1)a 109.24(6)

V(1)-N1)-N(1)c 180(-)

V(1)-C(1)-C(2) 129.57(9)The label a indicates the symmetry operation 1-y, 1+x-y, z; label c 2/3-x, 4/3-y, 1/3-z.

Molecule 1 consists of two identical tris(neopentyl) vanadium fragments linked by a

bridging dinitrogen ligand, the molecule possessing 3 symmetry. The vanadium atoms have a

tetrahedral environment, three of the coordination sites are occupied by neopentyl groups, the

remaining position by the µ-N2 ligand. The orientation of the two tris(neopentyl) vanadium

fragments is staggered, with all the t-Bu groups pointing inward, an effect that might be due

to crystal packing or could result from maximalization of dispersion forces. The closest

contacts between the t-Bu hydrogens of the two tris(neopentyl) vanadium units are in the

order of 3 Å, similar to those found in liquid apolar solvents. The V-C bond distances, all

2.0262 (16) Å, are among the shortest reported for V-C single bonds: for comparison, the

vanadium-carbon distances in Mes3VO5 are in the range 2.022(4)-2.079(3) Å, those in Li[( t-

Bu3SiN)2VMe24j are 2.043(5) and 2.057(8) Å. The V-N distances [1.7248(18) Å] are shorter

than those found in the V(II)-(µ-N2) complex (µ-N2){[( o-Me2NCH2)C6H4]2V(Py)}2(THF)2[1.833(3)/ 1.832(3) Å].6 These V-N bond lengths come close to V=N bond lengths observed

in imido complexes of V(IV) and V(V) (usually 1.60-1.68 Å).7


50

Figure 1. Molecular structure of [(t-BuCH2)3V]2(µ-N2) (1), with adopted numbering scheme, front view (top)

and side view (bottom).


51

This observation is in agreement with the perfectly linear arrangement of the V-(µ-

N2)-V unit and the rather long N-N distance [1.250(3) Å], which is comparable to the mean

nitrogen-nitrogen bond length of 1.24 Å observed for several diazo compounds RN=NR (R =

F, Me, Ph).8

Like for other transition metals, the isolable molecular nitrogen complexes for

vanadium can be divided into two classes on the basis of the oxidation state of the metal. In

low oxidation state complexes like [Na(THF)][V(N2)2(Ph2PCH2CH2PPh2)2]9 [V(-I)] the

dinitrogen ligands have bond orders of ~3 and bond distances only slightly longer than that in

free N2 (1.0976 Å).10 The high oxidation state complexes (µ-N2){[( o-

Me2NCH2)C6H4]2V(Py)}2(THF)26 [V(II)] and {Na[O(CH2CH2OMe)2]2}[Na(VMes3)2(µ-

N2)]11 [V(II)] contain bridging dinitrogen ligands in which the bond order is more reduced

(~2) as evidenced by the relatively long N-N distance of 1.228(4) Å and 1.280(21) Å,

respectively. Complex 1 represents the first example of a V(III) molecular dinitrogen complex

and is in that sense comparable to the bridging dinitrogen complexes of Nb(III) and

Ta(III).12,13

The bonding between metal and nitrogen in µ-dinitrogen transition metal complexeshas been evaluated mainly with the help of the simple qualitative molecular orbital scheme

pictured in Fig. 2.

MM N N

MM N N

MM N N

MM N N 1e

2e

3e

4e

Figure 2. Qualitative four center molecular orbital scheme for binuclear dinitrogen complexes.

This scheme describes the four center M-N-N-M interactions in an idealized four-fold

symmetry.14 The molecular orbitals arise from linear combinations of Mdxz and Npx orbitals.

Because of the presence of an equivalent set of molecular orbitals arising from combinations


52

of Mdyz and Npy the energy levels of the molecular orbitals are degenerate. The model has

been used to explain the differences between N-N bond lengths in µ-N2 complexes. In

complexes possessing a high number of d-electrons available for M-N bonding, such as

{[Ru(NH3)5]2(µ-N2)} 4+, the first µ-N2 complex that was characterized by X-ray

crystallography,15 the 3e MO, which is bonding in respect to N2, is filled, leading to a short N-

N bond (1.124 Å). This N-N bond is only slightly longer than the N-N bond in free

dinitrogen.10 In complexes with fewer d-electrons available for M-N bonding, such as the

niobium and tantalum bridging dinitrogen complexes [M(PMe3)(CH-t-Bu)(CH2-t-Bu)]2(µ-

N2),12,13 only the 2e MO, which is anti-bonding in respect to N2, is filled. In these complexes

the N-N bond distance is increased to approximately 1.30 Å. Although no bonding scheme is

available for the present vanadium µ-N2 complex, which has a three-fold symmetry, the

bonding is expected to be similar to that in the niobium and tantalum complexes. The low

number of d-electrons available for M-N bonding in 1 would allow only filling of a molecular

orbital which is similar to the 2e MO in Fig. 2, thus explaining the observed elongated N-N

bond in 1. Molecular orbital calculations, which could provide a more quantitative insight in

the bonding in µ-N2 complex 1, are in progress. The niobium and tantalum µ-N2 complexes

of Schrock et al. have been the subject of both crystallographic studies13 and Fenske-Hall

molecular orbital calculations.16 The MO calculations show that the actual bonding in these µ-

N2 complexes is more complicated than assumed above, whereas on the basis of the X-ray

data a simple, easily understood description of the bonding was devised by Schrock and

Churchill.12,13 In this description, the four d-electrons available for M-N bonding in the

complexes [M(PMe3)(CH-t-Bu)(CH2-t-Bu)]2(µ-N2), are fully transferred from the metal to

the more electronegative element nitrogen. The bonding is then discussed in terms of d0 metal

centers and the bridging N2 ligand formally acting as a hydrazido N24- (or diimido) group. The

bond order in the N-N linkage of an N24- group should be 1, with an estimated bond length of

1.45 Å, explaining the elongated N-N bonds observed in the niobium and tantalum µ-N2complexes. For vanadium, which is more difficult to oxidize than the heavier group 5

members, this description seems less appropriate.

1 is diamagnetic (by NMR spectroscopy), displaying a simple 1H NMR spectrum with

one resonance for the hydrogens of the methyl groups and one resonance for the methylene

protons, which is broadened (∆ν1/2 = 33) by a combination of unresolved coupling to the I =7/2 vanadium nucleus and quadrupolar relaxation effects.4e,f The broadening is even more

pronounced for the methylene carbon resonance in the 13C NMR spectrum, which is observed

as a plateau-form resonance with a half-width of approximately 3000 Hz. Tetrahedral

vanadium complexes tend to be high-spin and therefore paramagnetic. However,

delocalization of unpaired d-electrons on the tetrahedral vanadium centers in 1 through the


53

bridging dinitrogen ligand might reduce the barrier for spin-pairing, thus allowing the complex

to become diamagnetic.

The 51V NMR spectrum of 1 shows a moderately-resolved quintet at 1237 ppm. The

low field chemical shift reflects the electron deficiency of the metal atom.4i The quintet results

from coupling of vanadium to both 14N (I = 1) atoms, with the one- and two-bond coupling

constants appearing to be equal (48 Hz), although the width of the resonances precludes

accurate determination. For the labelled complex [(t-BuCH2)3V]2(µ-15N2) (1-15N2) (vide

infra) a moderately resolved triplet (15N; I = 1/2) with an apparent V-15N coupling constant

of 76 Hz is obtained. Nearly equal one- and two-bond coupling constants of metal to 15N in

µ-N2 complexes have been observed before by Schrock et al. for [C5Me5WMe3](µ-15N2).17

This behavior is in agreement with a significant degree of delocalization throughout the

MNNM system.

In agreement with observations made by Schrock et al. on M-(µ-N2)-M systems (M =

Mo, W)17,18 an absorption of medium intensity at 858 cm-1 in the IR spectrum of 1 is

tentatively assigned to a V=N stretch. It shows the expected shift to lower energy when 14N2is replaced by 15N2 (839 cm-1 in 1-15N2). Due to the fact that 1 is centrosymmetric, the N-N

stretching vibration is not IR active.

Reaction of VCl3(THF)3 with three equivalents of PhMe2CCH2Li in diethyl ether

afforded the tetrahydrofuran adduct of tris(neophyl) vanadium(III), (PhMe2CCH2)3V.THF

(2), which could be isolated in 20% yield (eq 2).

(2)VCl3(THF)3 + 3 PhMe2CCH2Li V THF

Me2CPh

Me2CPhMe2CPh

Et2O

- 3 LiCl, 2 THF

2

2 is a dark-blue, crystalline paramagnetic (by NMR spectroscopy) complex, readily

soluble in aliphatic and aromatic hydrocarbons. The compound decomposes rapidly (hours) at

ambient temperatures in solution and slowly (days) in the solid state, but can be kept

indefinitely at -20 °C.An X-ray structure determination of 2 was carried out. The crystal structure of 2

involves the packing of 4 molecules in the unit cell. The molecular structure of 2 is depicted in

Fig. 3, selected geometrical data are given in Table II.


54

Figure 3. Molecular structure of (PhMe2CCH2)3V.THF (2), with adopted numbering scheme.

Table II. Selected geometrical data for (PhMe2CCH2)3V.THF (2).

Bond lengths (Å)

V-C(1) 2.103(4)

V-C(11) 2.080(4)

V-C(21) 2.080(5)

V-O 2.025(3)

C(1)-C(2) 1.556(6)

C(11)-C(12) 1.540(6)

C(21)-C(22) 1.546(6)

Bond angles (°)

O-V-C(1) 109.6(1)

O-V-C(11) 108.1(1)

O-V-C(21) 113.7(1)

C(1)-V-C(11) 109.8(2)

C(1)-V-C(21) 109.0(2)

C(11)-V-C(21) 106.5(2)

V-C(1)-C(2) 129.4(3)

V-C(11)-C(12) 130.1(3)

V-C(21)-C(22) 128.0(3)

The molecular structure of 2 shows a tetrahedral arrangement of the three neophyl

groups and the THF ligand around the vanadium atom, with bond angles varying between

106.5(2)° and 113.7(1)°. The three V-C bond distances are identical within experimentalorder and range from 2.080(5) to 2.103(4) Å. They compare well to those found in the

isostructural VMes3.THF19 (2.099(6)-2.116(7) Å), while they are significantly longer than

those found in the dinitrogen bridged vanadium(III) tris(neopentyl) dimer (1) (vide supra). In


55

this molecule the size of vanadium is reduced by partial donation of electrons to the nitrogen

ligand. Another interesting feature of 2 is the way in which the phenyl groups of the neophyl

ligands are arranged. They are all positioned at the same side of the molecule and seem to

form a pocket in which the THF molecule lies. Two of the phenyl rings are arranged parallel

to the THF ring, one is arranged perpendicular. The same behavior is found in the neopentyl

compound 1 where all the t-Bu groups point to the center of the molecule, thereby shielding

the dinitrogen ligand. It is unclear whether this situation exists only in the solid state, as a

result of crystal packing, or is also present in solution, possibly due to London dispersion

forces.

The reaction leading to 2 can be performed both under argon and nitrogen. In the last

case, the formation of a bridging dinitrogen complex as observed for the corresponding

neopentyl derivative, is not observed. Cooling of pentane solutions of 1 under nitrogen to -80

°C did not lead to the formation of a bridging dinitrogen complex, contrasting the behaviorobserved for the tris(neopentyl) species (t-BuCH2)3V.THF (vide infra) where upon cooling

THF is readily displaced by dinitrogen even in THF solutions. It is likely that the stability of

the dinitrogen complex 1 depends largely on the shielding of the dinitrogen moiety by the

neopentyl ligands, thus precluding easy displacement of the dinitrogen ligand by stronger

Lewis bases. The structure of 2 indicates that the neophyl groups also tend to shield the

remaining coordination site on vanadium. However, given the short inter atomic distances

between the t-Bu groups of the two tris(neopentyl)vanadium centers in 1, replacing neopentyl

by neophyl to give [(PhMe2CCH2)3V]2(µ-N2) would lead to increased steric strain in thishypothetical molecule compared to 1. The steric strain would probably induce a more open

structure, thereby making the dinitrogen ligand more susceptible to displacement by Lewis

bases and preventing isolation of a stable dinitrogen complex in the neophyl case.

2.3 Reactivity of the dinitrogen complex [(t-BuCH2)3V]2(µ-N2) (1).

Dinitrogen activation by early transition metals has become a well-studied subject over

the past decades.20 This research has mainly been initiated by the discovery of early transition

metal containing nitrogenases (V,21 Mo,22 W23) in dinitrogen-fixing bacteria such as Azobacter

and Anabaena. A number of dinitrogen-fixing model systems based on vanadium have been

published,6,9,11,24,25 varying from V(II) catechol systems which catalytically reduce dinitrogen

to ammonia in the pH range 9-11,24b to the V(-I) system [V(N2)2(Me2PCH2CH2PMe2)2]-

which produces approximately 4/3 mol of ammonia upon protonation with HCl.25 The

composition of the products (NH3, N2H4) of dinitrogen reduction by these systems seems to

be largely dependent on the number of electrons26 that can be released by the metal:20 in the

latter system 4 electrons are involved and the metal is oxidized from V(-I) to V(III), whereas


56

protonation of (µ-N2){[( o-Me2NCH2)C6H4]2V(Py)}2(THF)2 produces 2/3 mol of ammonia

with the involvement of 1 electron per V [V(II) → V(III)]. 27 Leigh analyzed these and otherresults in a thermodynamic context by assuming that the availability of electrons from a metal

atom is an expression of the redox potential exhibited for oxidation states selected.20 His

conclusions support the observed product formation of the V(II) and V(-I) dinitrogen systems

upon protonation.

Neither ammonia nor hydrazine is produced when 1 is treated with excess HCl in

diethyl ether under inert atmosphere. Instead neopentane is produced while the complexed

dinitrogen is liberated quantitatively (Töpler pump determination) (eq 3).

(3)V N N V

t-Bu

t-But-Bu t-Bu

t-But-Bu

1

HCl, Et2O 2 VCl3 + 6 CMe4 + N2

It appears that, despite the fact that the dinitrogen ligand in 1 is considerably reduced

and therefore expected to be readily protonated, proton attack occurs only at the peripheral

methylene groups of the neopentyl ligands. This phenomenon might be explained by assuming

that the molecular structure in the solid state of 1 is also the dominant structure in solution. In

this case, the HOMO of 1, which is likely to be localized mainly on the VNNV moiety, is

shielded by the neopentyl ligands. Therefore, the HOMO is less susceptible to attack by

electrophiles, and electrophilic attack is likely to occur at the next best site of electron density,

the V-C bonding orbitals. Because of the bending of the neopentyl ligands these molecular

orbitals are easily accessible for small electrophiles such as H+.

An extension of Leigh's analysis for the higher oxidation states of group 5 metals also

predicts that dinitrogen is not reduced by V(III), but that Nb(III) and Ta(III) should be able

to reduce N2 to ammonia. Indeed, the M(III) dinitrogen-bridged complexes

[{M(S 2CNEt2)3} 2N2] [M = Nb (N-N = 1.252(16) Å), Ta) produce hydrazine quantitatively

upon protonation.28 It is, however, questionable to what extent Leigh's approach, which is

based upon the comparison of oxidation potentials for a series of redox couples in aqueous

acid solution, can be extended to organometallic compounds such as 1. The approach would

be more meaningful if oxidation potentials of organometallic systems in non-aqueous

solutions could be used, but these are available only for a limited number of redox couples.


57

The rest of the reactivity of 1 is dominated by loss of dinitrogen. With Lewis bases L 1

reacts to give dinitrogen and the corresponding V(III) adducts (Me3CCH2)3V.L [3, L = PMe3(a), pyridine (b), t-BuCN (c)] (eq 4).

t-But-Bu

t-But-But-Bu

t-Bu

V N N V

t-Bu

t-But-Bu

2 L2 V L (4)

3 a: L = PMe3 b: L = pyridine c: L = t-BuCN

- N2

1

The adducts 3 are paramagnetic oils, which are very soluble in pentane. They have

colors typical for V(III) species,19 varying from blue-green (3a) to purple (3b), and are

characterized by their shifted, broad resonances for the methyl protons of the neopentyl

ligands. No loss of dinitrogen is observed when 1 is dissolved in diethyl ether, a weak Lewis

base. Reversible coordination of dinitrogen is observed in tetrahydrofuran (Scheme 1).

t-But-Bu

t-But-But-Bu

t-Bu

V N N V

1

2 THF- N2

t-Bu

t-But-Bu

2 V THFN2

15

cooling

t-But-Bu

t-But-But-Bu

t-Bu

V N N V 1515

1- N215

-2 THF

Scheme 1. Reversible coordination of dinitrogen by 1.

Upon dissolving 1 in THF dinitrogen is liberated quantitatively (Töpler pump

determination) over a period of several minutes, while the color of the solution changes from


58

red-brown via green to intense blue. Cooling of tetrahydrofuran solutions of (t-

BuCH2)3V.THF (3d) under dinitrogen to -80 °C shows the reverse color change and theformation of crystalline 1. This behavior can be used to produce [(t-BuCH2)3V]2(µ-15N2) (1-15N2) when the nitrogen atmosphere is replaced by 15N2. Similar color changes are observed

in the salt metathesis reaction producing 1. The initial blue color of the reaction product in

diethyl ether (3d) changes to green upon replacement of the solvent by pentane and then to

red-brown with the formation of 1 upon cooling. The green color is thought to arise from the

mixing of the red-brown color of 1, which might be partially formed, and the blue color of

3d.

2.4 Oxidation and insertion reactions with 1 and 2.

As described in chapter 1, two electron oxidation provides a valuable tool in the

characterization of paramagnetic V(III) complexes, since it produces diamagnetic V(V)

complexes which can be analyzed by a range of NMR spectroscopical techniques. For the

complexes 1 and 2 two electron oxidation would yield rare V(V) tris(alkyl) species.

Attempted synthesis of these species by alkylation of vanadium(V) precursors was successful

only in two cases,4i,l due to the easy reduction of vanadium(V) to lower oxidation states by

the alkylating reagent. On the other hand, oxidation of homoleptic V(III) complexes to V(V)

might be hampered by the fact that in this kind of complexes the metal-carbon σ bond isnormally the most reactive center, thus leading to insertion of the oxidizing agent in a metal-

carbon bond instead of oxidation. Direct oxidation of the metal center by oxidizing agents like

oxygen or oxygen transfer reagents has thus far been reported only for alkyl/aryl chromium29

and aryl vanadium5,30 complexes.

Reaction of 1 (eq 5) and 2 (eq 6) with styrene oxide produces yellow, crystalline

OV(CH2CMe2R)3 (4a, R = Me; 4b, R = Ph), rare examples of vanadium(V) oxo complexes

OVR3 (R = CH2SiMe3,4a R = Mes5,30).

(5)V N N V

t-Bu

t-But-Bu t-Bu

t-But-Bu

1

O

Ph- N2, 2

Ph

2

t-But-Bu

t-Bu

V O

4a


59

(6)V THF

Me2CPh

Me2CPhMe2CPh

2

O

Ph- THF,

Ph

Me2CPhMe2CPh

Me2CPh

V O

4b

The complexes 4 a and b are very soluble in all common organic solvents, and are

also observed to be the major products of air oxidation of 1 and 2. They crystallize in the

form of very long and thin needles, which prevented their analysis by X-ray diffraction. Their

thermal stability is low, both in solution and in the solid state, but they can be stored for

indefinite time at -25 °C.The complexes 4a and 4b are characterized by their infrared spectra, showing strong

absorptions for the V=O bonds at 984 and 1001 cm-1, respectively. These absorptions are

found on the high energy end of the range reported for monooxo vanadium complexes (1035-

875 cm-1),7 implicating relatively strong vanadium-oxo bonds in 4a and 4b. The 1H and 13C

NMR spectra of diamagnetic 4 show broad, plateau-form resonances for both the methylene

protons and the methylene carbons, characteristic for vanadium(V) alkyl complexes.4j The

resonances of 4 in the 51V NMR spectra are found at the low field end of the 51V chemical

shift range.4i A very common feature in metal NMR spectroscopy of transition metal

complexes in the d0 configuration is an increase of shielding with increasing electronegativity

χ of the ligands attached to the coordination center.31 This trend, the "inverse" χ dependenceof metal shielding has been observed, among others, in a number of vanadium systems like

OVX3 (X = Br, Cl, F),32 t-BuNVX3 (X = Cl, Br, OR),

33 and p-tolylNVX 3 (X = Cl, OR,

CH2SiMe3),4i and seems to be determined mainly by the difference in energy between the

HOMO and LUMO orbitals. The energy difference increases with increasing electronegativity

of the ligand X, thus leading to a decrease in the overall paramagnetic shielding experienced

by the vanadium atom, and resulting in a high-field shift for the 51V NMR resonances.4i,34

Comparing the 51V chemical shifts within a series of OVR3 complexes could provide

information upon the relative electronegativities of the R groups. Unfortunately, the number

of OVR3 complexes, other than those reported here, for which 51V NMR data are available is

limited to one (e.i. OV(CH2SiMe3)3).35 Nevertheless, the shielding in OVR3 complexes is

observed to increase in the order OV(CH2-t-Bu)3 (δ 1212) < OV(CH2SiMe3)3 (δ 1205) <OV(CH2CMe2Ph)3 (δ 1191), thus suggesting the electronegativity of the alkyl group R toincrease in the same order.

Reaction of 1 with the substituted diazomethane Ph2CN2 also proceeds through

liberation of nitrogen and oxidation of (t-BuCH2)3V(III) to vanadium(V), but in this case


60

insertion of the diazomethane in V-C bonds is an important side reaction. The reactions that

occur are shown in Scheme 2.

V N N V

t-Bu

t-But-Bu t-Bu

t-But-Bu

1

1/2Ph2CN2- 1/2 N2

Ph2CN2

Ph2CN2

t-But-Bu

t-Bu

V NN CPh2

CPh2NN

t-But-Bu

Vt-Bu

N

NCPh2

t-Bu

N

NCPh2

CPh2NN

t-But-Bu

V

N

N CPh2

56

Scheme 2. Reaction of 1 with diphenyldiazomethane.

The first step in this reaction is thought to be coordination of diphenyldiazomethane to

1, similar to the reactions of 1 with Lewis bases discussed above, followed by a break-up of

the dimeric structure and the formation of two molecules of (t-BuCH2)3V=N-N=CPh2. This

molecule is believed to be a vanadium(V) tris(neopentyl) imido complex. Two electron

oxidation of V(III) to form V(V) imido complexes upon reaction with diazomethanes has

been observed before by Schrock et al. in the reaction of V[N(CH2CH2NSiMe3)3] with

(trimethylsilyl)diazomethane.36 In general, early transition metals not in their highest oxidation

state form complexes that contain metal-nitrogen multiple bonds with substituted

diazomethanes,37 whereas late transition metals tend to produce complexes containing metal-

carbon double bonds under evolution of nitrogen.1a The imido vanadium(V) tris(neopentyl)

complex could neither be isolated nor observed by 1H NMR spectroscopy during low-

temperature studies, due to its fast reaction with diphenyldiazomethane to form the mono and

bis insertion products (t-BuCH2)2V(N(CH2-t-Bu)N=CPh2)NNCPh2 (5) and (t-

BuCH2)V(N(CH2-t-Bu)N=CPh2)2NNCPh2 (6), respectively. Complexes 5 and 6 are both

crystalline materials, but only 6 could be obtained reproducibly in good yield. The mono

insertion product 5 was obtained pure on one occasion in a poor yield from the reaction of 1


61

with two equivalents of diphenyldiazomethane (V:N = 1:1). All other attempts at varying

temperatures and with varying stoichiometries led to mixtures of either 1 and 5 (V:N < 1) or

5 and 6 (V:N > 1), the latter presumably due to fast insertion of diphenyldiazomethane in 5.

Insertion of diphenyldiazomethane in 6 was not observed.

Complexes 5 and 6 are diamagnetic (by NMR spectroscopy) and are characterized as

vanadium(V) imido complexes by the low-field 13C resonances (δ 163.0 and 158.2 ppm,respectively) for the imine-like, sp2-hybridized carbon of the non-inserted diazomethane

molecule. The inserted diazomethane molecules form amido ligands N(CH2-t-Bu)NCPh2,

with the CH2 group displaying characteristic low-field shifted triplets (JCH = 135 Hz) in the13C NMR spectra. The vanadium-bound CH2 groups in 5 give rise to a broad resonance at δ100 ppm in the 13C NMR spectrum, and two doublets in the 1H NMR spectrum. It is assumed

that both 5 and 6 have a tetrahedral structure about vanadium and that the imido-

diazomethane lies in plane with the inserted diazomethane in 5 and with the remaining

neopentyl group in 6. This would render the vanadium-bound CH2 groups in 5 diastereotopic,

and the nitrogen-bound CH2 groups in 6 as well. The 1H NMR spectrum of 6 displays the

expected two doublets for the nitrogen-bound CH2 groups and a singlet for the vanadium-

bound CH2 group. Furthermore, in each 1H NMR spectrum two resonances for the neopentyl

methyl groups can be found, in a 2:1 ratio. The 51V NMR spectra of the two complexes show

resonances which are too broad to detect any V-N coupling, consistent with a highly

unsymmetrical environment.4i The resonance for 5 at δ 541 ppm shows the expected high-field chemical shift upon replacement of a neopentyl group by a more electronegative amido

ligand (6, δ -176 ppm).4i Moderate to strong resonances in the IR spectra of 5 and 6 at 972cm-1 are tentatively assigned to V=N stretches.17

Attempts to verify the proposed and rather unique structures of 5 and 6 by means of

X-ray diffraction failed because no single crystals of sufficient quality could be obtained.

Structures in which the second nitrogen of the inserted diazomethane is interacting with

vanadium, a bonding situation similar to those observed for early transition metal acyls,

cannot be excluded a priori on the basis of the present spectroscopic data.

Exploratory experiments, performed on micromolar scale in benzene-d6 at 25 °C,suggest that 1 and 2 also undergo two electron oxidation in reaction with 2-methylaziridine

(giving R3V=NH) and sulfur (R3V=S), albeit that the reactions are not clean. Research in this

field is still in progress. No reactions of 1 and 2 were observed with Se, Me3SiN3,

PhCH=PPh3 or PhN=PPh3 (one equivalent per V, benzene-d6, 25 °C).

2.5 Olefin metathesis reactions.


62

Until now no reports have appeared on olefin metathesis chemistry of d0 vanadium

complexes. This might be mainly due to the absence of d0 alkylidene species for vanadium,

although several suitable precursors for d0 vanadium alkylidenes have been reported.4a,i,j,l To

get a better understanding of the activity of d0 vanadium in olefin metathesis reactions, the

activity of two possible precursors for d0 vanadium alkylidenes was examined in an

exploratory fashion. As possible precursors were taken the bridged dinitrogen complex [(t-

BuCH2)3V]2(µ-N2) (1) and the oxo-vanadium complex OV(CH2-t-Bu)3 (4a).

The thermal decomposition of these two high-valent vanadium tris(neopentyl)

compounds was studied first, in order to establish their suitability as precursors for d0

vanadium alkylidenes. The products of α-hydrogen abstraction, the assumed pathway for theformation of metal alkylidenes, should be neopentane (CMe4).

Thermal decomposition of 1 (benzene, 25 °C, 16 h) produces large amounts of CMe4(> 2 equivalents) but no dinitrogen (Töpler pump experiment). Therefore, it is likely that 1

decomposes through one or more α-hydrogen abstractions (Fig. 4), leading to vanadiumalkylidenes and/or alkylidynes as dimers which still contain the bridging N2-ligand, and are

therefore likely to be d0 compounds.

V N N V

t-Bu

t-But-Bu t-Bu

t-But-Bu

α-H abstr.

- CMe4

α-H abstr.

- CMe4

α-H abstr.

- CMe4

t-But-But-Bu

t-Bu

V N N V

t-Bu

t-Bu

V N N V

t-Bu

t-Bu

t-Bu t-Bu

t-Bu

t-Bu

V N N V

Figure 4. Possible α-hydrogen abstractions in 1.

Attempts to stabilize these species by performing the thermal decomposition in the

presence of Lewis-bases (L) like PMe3 or THF lead only to loss of dinitrogen and formation

of relatively stable vanadium(III) Lewis-base adducts (t-BuCH2)3V.L (vide infra). The

organometallic products of the consequent thermolysis of these adducts are virtually


63

insoluble, black materials, which give no NMR signals, probably as a result of the low

solubility.

Like 1, 4a is thermally not stable at room temperature, decomposing in solution with a

half-live time of 4-6 hours at 25 °C. Thermal decomposition gives CMe4 and intractable,insoluble black materials, tentatively formulated as vanadium(V) alkylidenes/alkylidynes.

Thermolysis of 4a in the presence of Lewis-bases yielded neopentane and soluble

decomposition products, but these compounds are NMR silent, suggesting homolytic splitting

of vanadium-carbon bonds and the formation of lower valent vanadium species.

In the presence of norbornene, thermal decomposition of 1 and 4a induces the ring-

opening metathesis polymerization of norbornene, but not of less strained cyclic olefins such

as cyclopentene and cyclooctene. Likewise, no metathesis of acyclic olefins was observed by

thermal decomposition products of 1 and 4a. Activity in olefin metathesis was checked by

studying the isomerization of cis-3-hexene to the thermodynamically more stable trans-3-

hexene, which is catalyzed by metal-alkylidenes (eq 7). The isomerization can easily be

followed by 1H NMR spectroscopy (olefinic protons).

Et Et Et

Et

M=CHR(7)

The polymerization of norbornene by thermal decomposition products of 1 and 4a was

studied in more detail. The polymerizations were first performed on a small scale and studied

by 1H NMR spectroscopy to get a qualitative idea about the rate of polymerization, then on a

larger scale to be able to isolate and study the physical properties of the polymers.

At room temperature polymerization was found to be slow, with conversions after 24

h not higher than 20%. At 60 °C higher conversions were obtained, ranging from 65% for 1to 98% for 4a after 24 h. After 4-5 h at 60 °C about 80% of the final conversion is reached,and the rate of polymerization begins to slow down, probably due to thermal inactivation of

the catalyst. An effort was made to see the propagating species in the NMR spectrum by

looking at the nucleophilic alkylidene hydrogen region (δ ≈ 15-10 ppm), but no resonanceswere found, not even at very high concentrations of catalyst precursor (up to the solubility

limits). This could mean that the propagating species is only present in very low

concentrations, or is a paramagnetic V(III) or V(IV) species.

For the polymer obtained from the polymerization experiment of 1 with norbornene at

25 °C it was first established that the polymer was indeed polynorbornene formed by ROMPof norbornene by comparison of its 1H NMR spectrum with that of an authentic sample. Then

the cis- and trans-content was calculated from the integrations of the 13C and 1H NMR

spectra.38 In both cases values of 27% cis and 73% trans were obtained. The values for Mn,


64

Mw and the polydispersity as obtained by gel permeation chromatography (GPC) can be

found in Table III (see Experimental). The values given are corrected for the differences in

physical properties between polynorbornene and polystyrene.39 From the value for Mn(81000) it can be concluded that the average number of norbornene units in a polymer is

about 900. Since the ratio of olefin/1 was 100 at the start of the polymerization, this must

mean that no more than about 10% of the catalyst precursor produces an active catalyst,

assuming that all polymer chains grow at a similar rate. The calculated polydispersity of 2.14

is high. A value close to unity would mean a polymer with a very narrow molecular weight

distribution, and therefore well defined physical properties.40 The molecular weight

distribution observed here reflects the formation of catalytically active species in the

thermolysis of 1.

The cis:trans ratio of the polymers obtained from ROMP of norbornene induced by 4a

at 25 °C and 1 and 4a at 60 °C does not deviate dramatically from the statistical 1:1 in any ofthe experiments. The GPC data for these polymers show several (2 to 3) overlapping broad

curves (estimated polydispersities higher than 2.5). This so-called multimodal behavior can be

explained by assuming that several catalytic species producing polymers with different

molecular weight distributions are active at the same time. Since the catalyst precursors are

multi-alkyl species repeated α-hydrogen abstractions could account for the formation ofseveral catalytically active species.

2.6 Concluding remarks.

Salt metathesis of VCl3(THF)3 with three equivalents of t-BuCH2Li or PhMe2CCH2Li

produces homoleptic vanadium(III) alkyl fragments VR3 (R = t-BuCH2, PhMe2CCH2), which

can be isolated as the dinitrogen complex [(t-BuCH2)3V]2(µ-N2) and the THF adduct

(PhMe2CCH2)3V.THF, respectively. The dinitrogen complex represents the first dinitrogen

complex of vanadium(III). Based upon the molecular structure the dinitrogen ligand can

considered to be highly reduced, but this is not reflected in the reaction with HCl, where

dinitrogen is liberated quantitatively instead of being reduced to hydrazine or ammonia.

Reaction with Lewis bases L also proceeds with loss of dinitrogen to produce the Lewis base

stabilized complexes (t-BuCH2)3V.L.

Oxidation of the VR3 fragments provides in some cases routes to vanadium(V)

tris(alkyl) complexes. Oxygen transfer from styrene oxide produces the oxo vanadium(V)

complexes R3VO, whereas the dinitrogen complex reacts with diphenyldiazomethane through

oxidation and insertion in V-C bonds to form vanadium(V) imido amido alkyl complexes.

The thermal decomposition of two of the high-valent vanadium alkyl complexes

presented here, and the activity of the decomposition products in olefin metathesis reactions


65

were investigated. Although the dinitrogen complex and the tris(neopentyl) oxo complex are

rather poor catalyst precursors for ring-opening metathesis polymerization of norbornene, and

are not active in the polymerization of less strained cyclic olefins or the metathesis of non-

cyclic olefins, it can be concluded from these investigations that the thermal decomposition of

these alkyl complexes produces high-valent vanadium alkylidenes, which are active as

catalysts for ROMP of norbornene. The activity of these alkylidenes prepared in situ is

markedly higher than the activity of the only isolated nucleophilic alkylidene for vanadium

CpV(CH-t-Bu)dmpe (chapter 1), which is a d2 alkylidene. On the basis of these results it can

therefore be expected that well-defined, d0 vanadium alkylidenes are active in ROMP of

strained cyclic olefins.

2.7 Experimental.

General details. All manipulations were performed under nitrogen or argon (when stated)

using Schlenk techniques or a glove box, or using vacuum line techniques. Solvents [diethyl

ether, tetrahydrofuran (THF), pentane (mixed isomers), benzene, and deuterated solvents]

were distilled from Na/K alloy before use. NMR spectra were recorded on a Varian VXR-300

(1H, 300 MHz; 13C, 75.4 MHz; 51V, 78.9 MHz) spectrometer in benzene-d6 at 20 °C (unless

stated otherwise), chemical shifts in ppm, downfield from TMS (δ 0.00, 1H, 13C) or VOCl3 (δ0.00, 51V) positive. Half-width values and coupling constants are reported in Hz. IR spectra

were recorded on a Mattson-4020 Galaxy FT-IR spectrophotometer from Nujol mulls

between KBr discs (unless stated otherwise), wave numbers in cm-1. Gel permeation

chromatography was performed on a Waters GPC-120C instrument, molecular weights are

reported relative to narrow molecular weight polystyrene. Elemental analyses were performed

at the Microanalytical Department of the University of Groningen. Values given are the

average of at least two independent determinations.

Materials. VCl3(THF)341 and Ph2CN2

42 were prepared according to published procedures. t-

BuCH2Li and PhMe2CCH2Li were prepared from the corresponding alkyl chloride by

refluxing with two equivalents of lithium in hexane, followed by filtration and crystallization

at -25 °C. Styrene oxide (Janssen) was dried over molecular sieves (4 Å) and distilled beforeuse. Norbornene, cyclooctadiene and cyclopentene (Aldrich) were dried over sodium,

distilled and stored under nitrogen. Cis-3-hexene (Trans World), t-BuCN, pyridine (Janssen)

were used as received. PMe3 was prepared according to an adapted literature procedure,43

using MeMgI instead of MeMgBr.

[(t-BuCH2)3V]2(µ-N2) (1). Onto a mixture of VCl3(THF)3 (2.67 g, 7.14 mmol) and t-

BuCH2Li (1.67 g, 21.44 mmol), 40 mL of diethyl ether was condensed at -196 °C. The


66

mixture was thawed out and warmed to 20 °C while being stirred. After 1 h the solvent was

removed in vacuo. The residue was extracted twice with 25 mL of pentane. Concentrating

and cooling the combined extracts to -25 °C yielded 0.90 g (45 % based on V) of red-brown

crystals: 1H NMR δ 2.13 (∆ν1/2 = 33, 12H, CH2), 1.21 (s, 54H, CMe3); 13C NMR δ 118-108(CH2), 37.6 (CMe3), 34.0 (CMe3); 51V NMR δ 1237 (quintet, average 51V-14N couplingconstant 48); IR 2775(w), 2688(w), 2681(w), 1363(m), 1257(w), 1234(s), 1076(m),

1058(m), 931(w), 858(m), 752(m), 563(m), 490(m). Anal. Calcd for C30H66V2N2: C, 64.72;

H, 11.95; V, 18.29. Found: C, 62.49; H, 11.51; V, 18.49. Carbon and hydrogen data are low

due to explosive burning of the compound with oxygen.

(PhMe2CCH2)3V.THF (2). Onto a mixture of VCl3(THF)3 (1.66 g, 4.44 mmol) and

PhMe2CCH2Li (1.87 g, 13.33 mmol), 40 mL of diethyl ether was condensed at -196 °C. Themixture was thawed out and allowed to warm up to 25 °C while being stirred. After 1 h thesolvent was removed in vacuo. The residue was extracted twice with 20 mL of pentane.

Concentrating and cooling of the combined blue extracts to -25 °C yielded 0.47 g (0.90mmol, 20%) of dark-blue crystals: 1H NMR δ 18 (∆ν1/2 = 800), 8.4 (∆ν1/2 = 600), 7.5 (∆ν1/2= 50), 6.9 (∆ν1/2 = 50), 5.6 (∆ν1/2 = 300), -3.4 (∆ν1/2 = 650). IR 1792(w), 1690(w),1597(m), 1493(m), 1358(m), 1300(w), 1269(m), 1207(w), 1184(m), 1167(m), 1109(w),

1078(w), 1030(m), 1010(m), 918(w), 902(w), 856(s), 763(s), 721(m), 700(s), 594(m),

567(w), 532(m). Anal. Calcd for C34H47VO: C, 78.13; H, 9.06; V, 9.75. Found: C, 77.80; H,

9.02; V, 9.92.

Protonation of [(t-BuCH2)3V]2(µ-N2) (1). Under argon, 1 (129 mg, 0.23 mmol) was

dissolved in 30 mL of diethyl ether. To the clear solution 6.8 mL of a 0.68 M solution of HCl

in diethyl ether (4.6 mmol) was added. For the determination of hydrazine 25,0 mL of

distilled water was added. No hydrazine could be detected by a spectrophotometric method.44

In a second run the organic volatiles were removed in vacuo, and a modified Kjehldahl

distillation was performed.45 No ammonia could be detected in the distillate by a

spectrophotometric method using Nessler's reagent.46

(t-BuCH2)3V.L (3a-d). For 3a-c 2 equivalents of L were added with a syringe to 0.1-0.2

mmol of 1 in 0.4 mL of benzene-d6. For 3d 0.1 mmol of 1 was dissolved in 0.4 mL of THF-

d8. 1H NMR spectra were recorded after 15 minutes at 25 °C over the range +100 to -100

ppm: 3a (L = PMe3, blue-green) δ 0.1 (∆ν1/2 = 120, 27H, t-Bu), -1.5 (∆ν1/2 = 600, 9H,PMe3); 3b (L = pyridine, blue-purple) δ -2.6 (∆ν1/2 = 300, t-Bu); 3c (L = t-BuCN, green) δ3.9 (∆ν1/2 = 120, 27H, t-Bu), -1.5 (∆ν1/2 = 410, 9H, t-BuCN); 3d (L = THF, dark-blue) δ3.0 (∆ν1/2 = 360, t-Bu).


67

[(t-BuCH2)3V]2(µ-15N2) (1-15N2). Onto 1 (31 mg, 0.055 mmol) was condensed

approximately 4 mL of THF at -196 °C in a 60 mL Schlenk vessel connected to a vacuum

line equipped with a Töpler pump. After thawing out, the mixture was stirred for 15 minutes

at 25 °C, yielding a blue solution, from which 0.037 mmol of N2 (68%) could be pumped off

in 2 freeze/thaw cycles. 15N2 was added (1 atm.), and the vessel was kept at -80 °C for 16 h,

causing a color change to red-brown and the deposit of a small amount of red-brown

microcrystalline material: 51V NMR δ 1237 (triplet, average 51V-14N coupling constant 76);IR 2775(w), 2688(w), 2681(w), 1363(m), 1257(w), 1234(s), 1076(m), 1058(m), 931(w),

839(m), 752(m), 563(m), 490(m).

(t-BuCH2)3VO (4a). Styrene oxide (0.24 g, 2.01 mmol) was added to a solution of 1 (0.56

g, 1.00 mmol) in 20 mL of benzene. After stirring for 3/4 h at 20 °C the solvent was removed

in vacuo and the residue extracted with 40 mL of pentane. Cooling the extract to -25 °C

yielded 0.38 g (1.36 mmol, 68 % based on V) of yellow crystals: 1H NMR (toluene-d8): δ1.73 (∆ν1/2 = 92, 6H, CH2), 1.08 (s, 27H, CMe3); 13C NMR δ 120-107 (CH2), 37.6 (CMe3),34.0 (CMe3); 51V NMR δ 1212 (∆ν1/2 = 50); IR 2737(vw), 2702(vw), 1361(vs), 1255(m),1234(s), 1170(vw), 1121(vw), 1076(m), 1062(s), 1016(m), 984(vs), 933(m), 912(m), 748(s),

594(s), 573(m), 509(s). Anal. Calcd for C15H33VO: C, 64.26; H, 11.86; V, 18.17. Found: C,

63.39; H, 11.63; V, 18.57. Carbon and hydrogen data are low due to explosive burning of the

compound with oxygen.

(PhMe2CCH2)3VO (4b). To a blue solution of 2 (0.19 g, 0.36 mmol) in 10 mL of benzene

was added styrene oxide (41 µL, 0.36 mmol) at 25 °C. After stirring for 15 minutes at thistemperature the organic volatiles were removed in vacuo and the oily residue was extracted

with 20 mL of pentane. Concentrating and cooling of the yellow extract to -80 °C gave 0.10g (0.21 mmol, 50 %) of yellow needles in two crops: 1H NMR δ 7.2 (m, Ar H), 7.05 (m, ArH), 1.64 (CH2, ∆ν1/2 = 80), 1.38 (s, CMe2); 13C NMR (toluene-d8, -10 °C) δ 151.0 (Ar C),128.5 (Ar C), 125.9 (Ar C), 125.7 (Ar C), 120-108 (CH2), 41.6 (CMe2Ph), 31.1 (CMe2Ph);51V NMR δ 1191 (∆ν1/2 = 63). IR 1784(w), 1599(w), 1493(m), 1304(w), 1269(w), 1207(w),1186(m), 1111(w), 1080(m), 1028(w), 1014(w), 1001(s), 763(s), 723(m), 696(s), 594(w),

542(w). Anal. Calcd for C30H39VO: C, 77.23; H, 8.43; V, 10.92. Found: C, 76.77; H, 8.46;

V, 11.27.

(t-BuCH2)2V(N(CH2-t-Bu)N=CPh2)NNCPh2 (5). Diphenyldiazomethane (0.25 g, 1.30

mmol) was added to a solution of 1 (0.36 g, 0.65 mmol) in 20 mL of benzene at 25 °C. Afterthe evolution of N2 had ceased (10 min) the organic volatiles were removed in vacuo. The

brown oily residue was extracted with 30 mL of pentane. Concentrating and cooling of the

extract to -25 °C gave 69 mg (0.106 mmol, 8%) of shiny brown-green crystals: 1H NMR δ


68

8.1, 7.8, 7.1 (3 m, 20H, Ar H), 2.91 (s, 2H, NCH2), 2.53 (d, 2H, JHH = 9, CHH), 1.71 (d,

2H, JHH = 9, CHH), 1.10 (s, 18H, CMe3), 0.89 (s, 9H, NCH2CMe3); 13C NMR δ 163.0(NN=CPh2), 140.4, 139.4, 137.6, 137.5, 137.1, 130.8, 130.5, 130.1, 129.3, 128.8, 128.6,

128.5, 128.3, 128.1, 127.9, 127.5 (Ar C), 100 (∆ν1/2 = 400, VCH2), 70.6 (t, JCH = 135,NCH2), 38.5 (CMe3), 33.8 (CMe3), 33.2 (NCH2CMe3), 28.0 (NCH2CMe3); 51V NMR δ 541(∆ν1/2 = 1000). IR 1263(m), 1232(w), 1199(w), 1153(m), 1126(w), 1074(m), 1026(m),972(m), 935(w), 918(w), 891(w), 763(s), 723(s), 702(s).

(t-BuCH2)V(N(CH 2-t-Bu)N=CPh2)2NNCPh2 (6). Diphenyldiazomethane (0.39 g, 2.01

mmol) was added to a solution of 1 (0.28 g, 0.50 mmol) in 15 mL of benzene at 25 °C. Afterthe evolution of N2 had ceased (10 min) the solution was stirred for 10 minutes. Removal of

the organic volatiles in vacuo gave a brown oily residue that was extracted with two times 20

mL of pentane. Concentrating and cooling of the combined extracts to -25 °C gave 0.32 g(0.37 mmol, 56% based on Ph2CN2) of orange needles: 1H NMR δ 8.1, 7.9, 7.6, 7.3, 7.1 (5m, 30H, Ar H), 2.63 (s, 2H, CH2), 3.31 (d, 2H, JHH = 14, NCHH), 2.85 (d, 2H, JHH = 14,

NCHH), 1.23 (s, 9H, CMe3), 1.12 (s, 18H, NCH2CMe3); 13C NMR δ 158.2 (NNCPh2),140.4, 139.4, 137.6, 137.5, 137.1, 130.8, 130.5, 130.1, 129.3, 128.8, 128.6, 128.5, 128.3,

128.1, 127.9, 127.5 (Ar C), VCH2 not observed, 70.0 (t, JCH = 135, NCH2), 37.7 (CMe3),

33.8 (CMe3), 32.5 (NCH2CMe3), 28.9 (NCH2CMe3); 51V NMR δ -176 (∆ν1/2 = 1060). IR1305(w), 1263(m), 1230(w), 1201(w), 1174(w), 1151(m), 1130(w), 1074(m), 1028(m),

1001(w), 972(s), 916(w), 889(w), 763(s), 723(s), 698(s), 677(w), 650(w), 603(m), 570(w),

536(w), 482(m). Anal. Calcd for C54H63N6V: C, 76.57; H, 7.50; V, 6.01. Found: C, 75.93;

H, 7.43; V, 6.32.

Thermal decomposition of 1 and 4a. NMR: Solutions of 1 (10 mg, 0.018 mmol) and 4a (15

mg, 0.054 mmol) in 0.4 mL of benzene-d6 were monitored by 1H NMR periodically over a

period of 24, both at 25 °C, and in a second run at 60 °C. The resonances due to the staringmaterial disappeared, and the formation of a dark precipitate was observed. The only new

resonance was due to neopentane (δ 0.90 ppm). Töpler pump: Under high-vacuum conditionsapproximately 4 mL of benzene was condensed onto 1 (38 mg, 0.068 mmol) at -196 °C in avessel equipped with a Young valve. The valve was closed and the vessel allowed to warm up

to 25 °C. After 24 h of stirring at 25 °C the gaseous products produced (0.148 mmol) werepumped off with a Töpler pump by repeated freeze(-80 °C)/thaw cycles. In a secondexperiment with 1 (40 mg, 0.072 mmol) 0.186 mmol of gas was obtained. In both cases, all of

the gas could be condensed at -196 °C, indicating that no N2 had formed.ROMP of cyclic olefins with 1. In a dry box 100-400 equivalents of a cyclic olefin (see

Table III) were added to a 0.01 mM solution of 1 (13 mg, 2.3 x 10-5 mol) in benzene (1.47

g) in a 10 mL vial. After mixing the vial was closed and left standing for 17-100 h (see Table


69

III) at 25 °C with occasional shaking. The resulting mixtures were worked up according tothe general workup procedure for polynorbornene (see below).

Table III. Survey of the activity of 1 in ROMP of strained cyclic olefins ([1] = 0.01 mmol/L).________________________________________________________________________________________

cyclic olefin ratio olefin:1 reaction time polymera GPC-datab

________________________________________________________________________________________

norbornene 100 17 h 20 % Mn = 81000Mw = 170000PDI = 2.14

cyclooctadiene 100 100 h 0 % -cyclopentene 400 100 h 0 % -________________________________________________________________________________________

a percentage of recovered cyclic olefin as a polymer after isolation (vide infra) is given. b Gel permeation

chromatography data recorded in CH2Cl2, PDI is polydispersity (Mw/Mn).

Attempted metathesis of cis-3-hexene with 1 and 4a. A typical experiment using 4a as a

catalyst precursor is described. In a dry box a solution of 4a (2.4 mg, 8.7 µmol) in 0.5 mL ofbenzene-d6 was placed in an NMR tube fitted with a septum. Cis-3-hexene (10 µL, 80 µmol)was added with a syringe. No isomerization of cis-3-hexene (olefinic protons at 5.37 ppm) to

trans-3-hexene (olefinic protons at 5.47 ppm) at room temperature was observed over a 24 h

period.

ROMP of norbornene with 1 and 4a. NMR: Solutions of 1 (2 mg, 3.6 µmol) and 4a (1.6mg, 5.7 µmol) were dissolved in 0.4 mL of benzene-d6 together with a 50-100 fold excess ofnorbornene. The solutions were kept at 25 °C, and in a second run at 60 °C. The reactionswere monitored periodically by 1H NMR spectroscopy over a period of 24 h. The formation

of polynorbornene and decomposition of the organometallic starting materials were observed,

but no vanadium alkylidene species could be detected. The conversions of norbornene were

estimated on the basis of the integrals of the resonances in the olefinic region. Preparative

scale: In a dry box 100 equivalents of norbornene were added to a 0.01 mM solution of

catalyst precursor in benzene in a reagent tube. After mixing, the tube was closed with a

septum. The tubes were placed in a 60 °C oil bath outside the dry box for 15 h and thenworked up according to the general procedure for polynorbornene (see below).

Isolation of polynorbornene. The polymer/catalyst/solvent gel was dissolved in CH2Cl2containing 5 weight percent of butylated hydroxytoluene (BHT) and a few drops of methanol

(for quenching any still active catalyst). The amount of solvent required was about 0.5 mL for

20 mg of expected polynorbornene. The obtained solution was filtered through glasswool to

remove any precipitate and through silica to remove decomposition products of the catalyst.

The silica column was eluted with more CH2Cl2/BHT solution to make sure that all


70

polynorbornene had come off. For small amounts of polynorbornene these filtrations can

conveniently be performed in Pasteur pipettes. The now clear and colorless polynorbornene

solution was added dropwise to about ten times its volume of methanol under rapid stirring.

The finely divided white polynorbornene obtained in this manner was isolated by

centrifugation. After the solvent had been decanted, the solid polynorbornene was dried

immediately in vacuo and handled under inert atmosphere to prevent cross linking of the

polymer induced by oxygen.

Table IV. Crystallographic Data for 1 and 2.

1 2

formula

fw

cryst syst

space group

a, Å

b, Å

c, Å

β, deg

V, Å3

Z

dcalcd, g cm-3

F(000), e

µ (Mo Kα), cm-1

cryst size, mm

T, K

θ limits, deg

no. of data collcd

no. unique data

no. reflns obsd

no of params refined

R(F)

Rw(F)

w

C30H66N2V2

556.75

trigonal

R3

10.072(1)

29.494(1)

2591.2(4)

3

1.0703(2)

918.0

5.4

0.22x0.22x0.12

130

1.00 < θ < 27.0

2033

1264

1131

97

0.027

0.030

1/σ2(F0)

C34H47OV

522.7

monoclinic

P21/n

8.845(5)

29.496(4)

12.289(4)

110.28(3)

3007(2)

4

1.15

1128

3.40

0.30x0.30x0.30

172

2.5 < θ < 25

5411

2512

325

0.043

0.049

1/σ2(F0)

References and notes.


71

1 (a) Grubbs, R. H. in Comprehensive Organometallic Chemistry; Wilkinson, G.; Stone, F. G. A.;

Abel, E. W., Eds.; Pergamon Press: Oxford, 1982; Vol. 8, Chapter 54. (b) Schrock, R. R. J. Organomet.

Chem. 1986, 300, 249.2 Feldman, J.; Schrock, R. R. in Progress in Inorganic Chemistry; Lippard, S. J., Ed.; Wiley: New

York, 1991, Vol. 39, 1.3 (a) Schrock, R. R. J. Am. Chem. Soc. 1974, 96, 6796. (b) Schrock, R. R.; Fellmann, J. D. J. Am.

Chem. Soc. 1978, 100, 3359.4 (a) Mowat, W.; Shortland, A.; Yagupsky, G.; Hill, N. J.; Yagupsky, M.; Wilkinson, G. J. Chem. Soc.,

Dalton Trans. 1972, 533. (b) Razuvaev, G. A.; Latyaeva, V. N.; Vyshinskaja, L. I.; Linoyova, A. N.;

Drobotenko, V. V; Cherkasov, V. K. J. Organomet. Chem. 1975, 93, 113. (c) Barker, G. K.; Lappert, M. F.;

Howard, J. A. K. J. Chem. Soc., Dalton Trans. 1978, 734. (d) Bagdasaryan, A. K.; Gorelik, V. M.;

Bondarenko, G. N.; Dolgoplosk, B. A. Dokl. Akad. Nauk. SSSR 1977, 236, 877. (e) Preuss, F.; Ogger, L. Z. Z.

Naturforsch. 1982, 37B, 957. (f) Preuss, F.; Becker, H. Z. Naturforsch. 1986, 41B, 185. (g) Preuss, F.; Becker,

H.; Häusler, H. J. Z. Naturforsch. 1987, 42B, 881. (h) Preuss, F.; Becker, H.; Wieland, T. Z. Naturforsch.

1990, 45B, 191. (i) Devore, D. D.; Lichtenhan, J. J.; Takusagawa, F.; Maatta, E. A. J. Am. Chem. Soc. 1987,

109, 7408. (j) De With, J.; Horton, A. D.; Orpen, A. G. Organometallics 1990, 9, 2207. (k) De With, J.;

Horton, A. D. Angew. Chem., Int. Ed. Engl. 1993, 32, 903. (l) Preuss, F.; Overhoff, G.; Becker, H.; Häusler,

H.; Frank, W.; Reiss, G. Z. Anorg. Allg. Chem. 1993, 619, 1827. (m) Hessen, B.; Meetsma, A.; Teuben, J. H.

J. Am. Chem. Soc. 1989, 111, 5977.5 Ruiz, J.; Vivanco, M.; Floriani, C.; Chiesi-Villa, A.; Guastini, C. J. Chem. Soc., Chem. Commun.

1991, 762.6 Edema, J. J. H.; Meetsma, A.; Gambarotta, S. J. Am. Chem. Soc. 1989, 111, 6878.7 Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley: New York, 1988.8 Churchill, M. R.; Lin, K.-K. G. Inorg. Chem. 1975, 14, 1133.9 Rehder, D.; Woitha, C.; Priebsch, W.; Gailus, H. J. Chem. Soc., Chem. Commun. 1992, 364.10 Wilkinson, P. G.; Houk, N. B. J. Chem. Phys. 1956, 24, 528.11 Ferguson, R.; Solari, E.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Angew. Chem. Int. Ed. Engl. 1993,

32, 396.12 (a) Turner, H. W.; Fellmann, J. D.; Rocklage, S. M.; Schrock, R. R.; Churchill, M. R.; Wasserman, J. J.

Am. Chem. Soc. 1980, 102, 7809. (b) Rocklage, S. M.; Turner, H. W.; Fellmann, J. D.; Schrock, R. R.

Organometallics 1982, 1, 703. (c) Rocklage, S. M.; Schrock, R. R. J. Am. Chem. Soc. 1982, 104, 3077. (d)

Churchill, M. R.; Wasserman, H. J. Inorg. Chem. 1982, 21, 218.13 Churchill, M. R.; Wasserman, H. J. Inorg. Chem. 1981, 20, 2899.14 (a) Chatt, J.; Dilworth, J. R.; Richards, R. L. Chem. Rev. 1978, 78, 589. (b) New Trends in the

Chemistry of Nitrogen Fixation, Chatt, J.; General Da Câmara Pina, L. M.; Richards, R. L., Eds.; Academic

Press: London, 1980. (c) Sellman, D.; Angew. Chem. 1974, 86, 692.


72

15 (a) Treitel, I. M.; Flood, M. T.; Marsh, R. E.; Gray, H. B. J. Am. Chem. Soc. 1969, 91, 6512. (b)

Fergusson, J. E.; Love, J. L.; Robinson, W. T. Inorg. Chem. 1972, 11, 1662.16 Powell, C. B.; Hall, M. B. Inorg. Chem. 1984, 23, 4619.17 O'Regan, M. B.; Liu, A. H.; Finch, W. C.; Schrock, R. R.; Davis, W. M. J. Am. Chem. Soc. 1990, 112,

4331.18 Schrock, R. R.; Kolodziej, R. M.; Liu, A. W.; Davis, W. M.; Vale, M. G. J. Am. Chem. Soc. 1990, 112,

4338.19 Gambarotta, S.; Floriani, C.; Chiesi-Villa, A.; Guastini, C. J. Chem. Soc., Chem. Commun. 1984,

886.20 Leigh, G. J. Acc. Chem. Res. 1992, 25, 177.21 (a) Morningstar, J. E.; Hales, B. J. J. Am. Chem. Soc. 1987, 109, 6854. (b) Hales, B. J.; Case, E. E.;

Morningstar, J. E.; Dzeba, M. F.; Mauterer, M. A. Biochemistry 1986, 25, 7521. (c) Eady, R. R. Biofactors

1988, 1, 111. (d) Yakunin, A. F.; Ni, C. V.; Gogotov, I. N. Dokl. Akad. Nauk. SSR, 1989, 307, 1269. (e)

Kentemich, T.; Danneberg, G.; Hundeshagen, B.; Bothe, H. FEMS Microbiol. Lett. 1988, 51, 19.22 Leigh, G. J. J. Mol. Catal. 1988, 47, 363 and references cited therein.23 Hales, B. J.; Case, E. E. J. Biol. Chem. 1987, 262, 16205.24 (a) Denisov, N. T.; Efimov, O. N.; Shuvalova, N. I.; Shilova, A. K.; Shilov, A. E. Zh. Fiz. Khim.

1970, 44, 2694. (b) Luneva, N. P.; Nikonova, L. A.; Shilov, A. E. Kinet. Catal. 1980, 21, 1041.25 Woitha, C.; Rehder, D. Angew. Chem., Int. Ed. Engl. 1990, 29, 1438.26 Reduction of N2 to two mol of ammonia requires 6 electrons, to hydrazine 4.27 Leigh, G. J.; Prieto-Alcón, R.; Sanders, J. R. J. Chem. Soc., Chem. Commun. 1991, 921.28 (a) Dilworth, J. R.; Henderson, R. A.; Hills, A.; Hughes, D. L.; Macdonald, C.; Stephens, A. N.;

Walton, D. R. M. J. Chem. Soc., Dalton Trans. 1990, 1077. (b) Henderson, R. A.; Morgan, S. H.; Stephens,

A. N. J. Chem. Soc., Dalton Trans. 1990, 1101.29 Noth, S. K.; Heintz, R. A.; Haggerty, B. S.; Rheingold, A. L.; Theopold, K. H. J. Am. Chem. Soc.

1992, 114, 1892.30 Vivanco, M.; Ruiz, J.; Floriani, C.; Chiesi-Villa, A.; Rizzoli, C. Organometallics 1993, 12, 1802-

1810.31 Rehder, D. Magn. Reson. Rev. 1984, 9, 125.32 Rehder, D. Z. Naturforsch. 1977, 32B, 771.33 Preuss, F.; Towae, W.; Kruppa, V.; Fuchslocher, E. Z. Naturforsch. 1984, 39B, 1510.34 Priebsch, W.; Rehder, D. Inorg. Chem. 1985, 24, 3058.35 Feher, F. J.; Blanski, R. L. Organometallics, 1993, 12, 958.36 Cummins, C. C.; Schrock, R. R.; Davis, W. D. Inorg. Chem. 1994, 33, 1448.37 (a) Roland, E.; Walborsky, E. C.; Dewan, J. C.; Schrock, R. R. J. Am. Chem. Soc. 1985, 107, 5795.

(b) Chisholm, M. H.; Folting, K.; Huffman, J. C.; Raterman, A. L. Inorg. Chem. 1984, 23, 2303. (c)

Hillhouse, G. L.; Haymore, B. L. J. Am. Chem. Soc. 1982, 104, 1537.


73

38 (a) Ivin, K. J.; Laverty, D. T.; Rooney, J. J. Makromol. Chem. 1977, 178, 1545. (b) Ivin, K. J.;

Laverty, D. T.; Rooney, J. J. Makromol. Chem. 1978, 179, 253.39 A correction factor of 2.2 for polystyrene vs polynorbornene has been proposed: Schrock, R. R.;

Feldman, J.; Cannizzo, L. F.; Grubbs, R. H. Macromolecules 1987, 20, 1169.40 Schrock, R. R. Acc. Chem. Res. 1990, 23, 158.41 Kurras, E. Naturwissenschaften 1959, 5, 171.42 Miller, J. B. J. Org. Chem. 1959, 24, 560.43 Luetkens, M. L., Jr.; Sattelberger, A. P.; Murray, H. H.; Basil, J. D.; Fackler, J. P., Jr. Inorg. Synth.

1984, 26, 7.44 Watt, G. W.; Crisp, J. D. Anal. Chem. 1952, 24, 2006.45 Richardson, G. T.; Davies, J. A.; Edwards, J. G. Fresenius J. Anal. Chem.1991, 340, 392.46 Boltz, D. Colorimetric determination of non-metals, Wiley; New York, 1978, 205.

Documents

University of Groningen Chemistry of vanadium-carbon single ...In this chapter the synthesis and reactivity of vanadium(III) complexes, containing sterically demanding alkyl groups