Halogen Substituted Bis(arylimino)Pyridine Transition Metal

Download PDF
3 downloads 0 Views 2MB Size Report
Comment
Feb 4, 2008 - The X-ray analyses were performed by Dr. Wolfgang Milius (University ...... 5. VCl3. 6. CrCl3. 7. MnCl2. 8. FeCl2. 9. FeCl3. 10. CoCl2. 11. NiCl2.
JJC

Jordan Journal of Chemistry Vol. 3 No.2, 2008, pp. 109-145

Halogen Substituted Bis(arylimino)Pyridine Transition Metal Complexes as Catalysts for the Oligomerization and Polymerization of Ethylene

Marcus Seitz, Christian Görl, Wolfgang Milius, Helmut G. Alt* Laboratorium für Anorganische Chemie, Universität Bayreuth, Postfach 10 12 51, Universitätsstraße 30, D95440 Bayreuth, Germany Received on Oct. 16, 2007

Accepted on Feb. 4, 2008

Abstract A series of 15 complexes containing 3d transition metals ranging from titanium to nickel and halogen functionalized bis(arylimino)pyridine ligands was synthesized and characterized. After activation with methylalumoxane (MAO), these catalysts oligomerized ethylene to give αolefins with 4-40 carbon atoms. The influence of the metal center, the halogen substituents, and the reaction parameters on the product compositions are discussed. Some of the described catalyst precursors showed the potential to isomerize α-olefins and to generate olefins with uneven numbers of carbon atoms. Quantum mechanical calculations (DFT, B88LYP) helped explaining the isomerization behaviour of 5-halide-2-methyl substituted transition metal compounds. Relationships between the structure of the catalyst precursors or the parameters of the reactions and the activity or the product compositions were confirmed by experiments.

Keywords:

Bis(arylimino)pyridine;

α-Olefins;

Ethylene;

Polymerization;

Oligomerization; Polyethylene

1. Introduction In the past few years, late transition metal complexes became more and more interesting as catalyst precursors for the oligomerization and polymerization of αolefins. For example, α-diimine nickel complexes developed by Brookhart et al.[1] were used for the polymerization of ethylene. These complexes also showed the potential to incorporate polar monomers like acrylates. In 1998, Gibson[2,3] and Brookhart[4-6] applied bis(imino)pyridine iron and cobalt complexes for the polymerization of ethylene. Numerous transition metal complexes containing 2,6-bis(arylimino)pyridine ligands are known in the literature since the 1960’s using especially the first row transition

metals

iron[1-23],

cobalt[7,11,13,15,19,22-26],

nickel[19,27,28],

zinc[15,19,29-30],

vanadium[19,31-33] and chromium[19,34-37] as central metals. The early work also focused on other late transition metals like copper[38-41], ruthenium[42,43], rhodium[44-47], and iridium[46] while some actual work deals with the synthesis of titanium, zirconium, hafnium[48-50], and manganese[51-56] complexes. Especially bis(imino)pyridine iron * Corresponding author: Tel.: +49-921-552555; fax: +49-921-552044. E-mail-address: [email protected] (H.G. Alt)

complexes proved to be very active catalysts after activation with a suitable co-catalyst like MAO. 2,6-Bis(arylimino)pyridine iron complexes bearing substituents both at positions 2 and 6 of the iminophenyl rings are known to produce only polyethylene. If only one of the ortho positions of the imino nitrogens is substituted, these complexes produce oligomer/polymer mixtures or pure oligomer mixtures depending on the size of the substituent and the reaction conditions[4]. These oligomer mixtures consist of αolefins with 6-24 carbon atoms. α-Olefins are important resources for the synthesis of a variety of fine chemicals or can be used as comonomers for polymerization reactions. A great deal of theoretical work has been performed[7,25,57-60] investigating mechanistic

aspects

of

oligomerization

and

polymerization

reactions

with

bis(imino)pyridine complexes. The iron complex bis(5-chloro-2-methylphenyl-1ethylimino)pyridine iron(II) chloride[61] proved to be very active for the oligomerization of ethylene (3660 kg prod./g Fe · h). Since bis(imino)pyridine complexes are suitable candidates for structure-property relationships, this complex was chosen for further investigations. Furthermore, a series of bis(imino)pyridine complexes containing the 3d transition metals titanium, vanadium, chromium, manganese, iron, cobalt, and nickel is prepared whereby the bis(imino)pyridine ligand is kept the same. With this series of catalyst precursors the influences of different metal centers both on catalyst activities and product compositions are analyzed. Additionally, the influence of different halogen substituents at the ligand framework and at the metal center on the oligomerization and polymerization behavior is investigated. A couple of highly active halogen substituted bis(arylimino)pyridine iron complexes were already described in the literature[8,9,12,13,62].

A

new

aspect

concerning

the

ability

of

some

of

the

bis(imino)pyridine complexes to produce olefins with uneven numbers of carbon atoms is discussed. To our knowledge, the described catalysts are the first ones which are able to produce 1-alkenes with uneven carbon numbers from ethylene. Finally, optimized catalysts are presented with regard to their activity and their potential to produce olefins with uneven numbers of carbon atoms.

2. Materials and methods All experimental work was routinely carried out using Schlenk technique. Dried and purified argon was used as inert gas. n-Pentane, diethyl ether, toluene und tetrahydrofuran were purified by distillation over Na/K alloy. Diethyl ether was additionally distilled over lithium aluminum hydride. Methylene chloride was dried with phosphorus pentoxide and calcium hydride. Methanol and ethanol were dried over molecular sieves. 1-Butanol (p. a.) was purchased from Merck and used without prior distillation. Deuterated solvents (CDCl3, CD2Cl2) for NMR spectroscopy were stored over molecular sieves (3Ǻ). Methylalumoxane

(30

%

in

toluene)

was

purchased

from

Crompton

(Bergkamen) and Albemarle (Baton Rouge, USA / Louvain – La Neuve, Belgium). Ethylene (3.0) und argon (4.8/5.0) were supplied by Rießner Company (Lichtenfels).

110

All other starting materials were commercially available and were used without further purification. 2.1 NMR spectroscopy The spectrometer Bruker ARX 250 was available for recording the NMR spectra. The samples were prepared under inert atmosphere (argon) and routinely recorded at 25 °C. The chemical shifts in the 1H NMR spectra are referred to the residual proton signal of the solvent (δ = 7.24 ppm for CDCl3, δ = 5.32 ppm for CD2Cl2) and in

13

C

NMR spectra to the solvent signal (δ = 77.0 ppm for CDCl3, δ = 53.5 ppm for CD2Cl2). 2.2 Mass spectrometry Mass spectra were routinely recorded at the Zentrale Analytik of the University of Bayreuth with a VARIAN MAT CH-7 instrument (direct inlet, EI, E = 70 eV) and a VARIAN MAT 8500 spectrometer. Post-processing and data analyses were performed using the software “Maspec II32 Data System”. 2.3 GC/MS GC/MS spectra were recorded with a HP 5890 gas chromatograph in combination with a HP 5971A mass detector. A 12 m J&W Scientific fused silica column (DB1, diameter 0.25 mm, film 0.33 µm, flow 1ml/min) respectively 25 m J&W Scientific fused silica column (DB5ms, diameter 0.25 mm, film 0.33 µm, flow 1ml/min) were used, helium (4.6) was applied as carrier gas. Using a 12 m column, the routinely performed temperature program started at 70 °C (2 min). After a heating phase of eleven minutes (20K/min, final temperatur 290 °C) the end temperature was held for a variable time (plateau phase). At the Zentrale Analytik of the University of Bayreuth, GC/MS spectra were routinely recorded with a HP5890 gas chromatograph in combination with a MAT 95 mass detector. 2.4 Gas chromatography For the analysis of organic compounds, especially oligomer mixtures, a PERKIN ELMER Auto System gas chromatograph (column: HP1, 28 m, diameter 0.32 mm / carrier gas helium, flow 5.7 ml/min, split 3.5 ml/min) was used. The standard temperature program contained a starting phase at 50 °C (3 min), a heating phase of 50 minutes (heating rate 4 K/min, final temperatur 250 °C) and a plateau phase at 250 °C (37 min). 2.5 IR spectroscopy For the recording of IR spectra, the compounds were levigated with dried cesium iodide. Thereof, thin pellets were prepared applying a pressure of 10 bar. The pellets

111

were introduced into a PERKIN ELMER Spectrum 2000 FT-IR instrument containing a He-Ne-laser. The maximum resolution was 0.15 cm-1. IR absorptions in the range of 400-4000 cm-1 were recorded in steps of 1.0 cm-1. Fourier transformations and postprocessing was performed using the software “Spectrum for Windows” (PERKIN ELMER). 2.6 X-ray analysis The X-ray analyses were performed by Dr. Wolfgang Milius (University Bayreuth) using a Siemens P4 diffractometer (radiation source: MoKα, λ = 0.71073 Å). Crystal data for compound 1: Empirical formula: C23 H21 Cl2 N3; Formula weight: 410.33; yellow prisms from diethylether; monoclinic; space group P 21/c; a = 11.1358(22) Å; b = 15.7676(18) Å; c = 12.3067(16) Å; α = 90°, β = 95.88(1)°, χ = 90°; Volume = 2149.5(6) Å3; Z = 4; d(calc) = 1.268 g/cm3; absorption coeffizient 0.315 mm-1; F(000) = 856; Theta range for data collection 1.84-22.50°; Index ranges: -1011, -416, -1313; reflections

collected:

3429;

independent reflections

2649

[R(int) =

0.0211];

completeness to theta = 22.50°: 94.4 %; refinement method: full-matrix least-squares on F2; Goodness-of-fit 1.021; R1 [l>2σ(I)] = 5.24%, wR2 = 0.1312; R1 (all data) = 8.70 %, wR2 = 0.1531; extinction coefficient = 0.0097(15); largest diff. peak and hole: 0.259 and -0.174 e. Å3. The crystal was sealed in a glass capillary and measured at 293 K. Table 1. Atom coordinates (·10-4) and equivalent isotropic shift parameters (Å2 ·103) for compound 1. x

y

z

U(eq)a)

Cl(1)

1165(1)

3844(1)

1902(1)

116(1)

Cl(2)

-1717(2)

-5024(1)

2601(1)

122(1)

N(1)

-2775(3)

-295(2)

786(2)

53(1)

N(2)

-2455(3)

1931(2)

712(2)

56(1)

N(3)

-3869(3)

-2378(2)

1021(3)

63(1)

C(1)

-3575(3)

-898(2)

969(3)

51(1)

C(2)

-4668(4)

-719(2)

1348(3)

58(1)

C(3)

-4941(4)

115(2)

1553(3)

61(1)

C(4)

-4129(3)

740(2)

1361(3)

55(1)

C(5)

-3059(3)

515(2)

970(3)

46(1)

C(6)

-2141(3)

1156(2)

705(3)

51(1)

C(7)

-953(4)

839(3)

419(4)

88(2)

C(8)

-1636(3)

2572(2)

439(3)

54(1)

C(9)

-1800(4)

2932(2)

-596(3)

62(1)

C(10)

-2798(4)

2622(3)

-1427(4)

90(2)

C(11)

-1003(4)

3567(3)

-837(4)

77(1)

112

x

y

z

U(eq)a)

C(12)

-91(4)

3852(3)

-91(4)

75(1)

C(13)

34(4)

3495(3)

933(4)

72(1)

C(14)

-736(3)

2863(2)

1204(3)

63(1)

C(15)

-3199(4)

-1788(2)

736(3)

58(1)

C(16)

-2066(4)

-1905(3)

202(4)

90(2)

C(17)

-3553(4)

-3240(2)

848(4)

61(1)

C(18)

-3993(4)

-3664(2)

-92(4)

67(1)

C(19)

-4768(5)

-3219(3)

-990(4)

100(2)

C(20)

-3686(4)

-4513(3)

-176(4)

81(1)

C(21)

-2985(4)

-4935(3)

630(5)

83(2)

C(22)

-2587(4)

-4506(3)

1555(4)

76(1)

C(23)

-2860(4)

-3652(3)

1684(4)

70(1)

ij

a) U(eq) is defined as 1/3 of the track of the orthogonalized U tensor.

Table 2. Bond lengths [Å] and angles [°] in compound 1. Bond lengths [Å]

Angles [°]

Cl(1)-C(13)

1.733(5)

C(1)-N(1)-C(5)

118.4(3)

Cl(2)-C(22)

1.734(5)

C(6)-N(2)-C(8)

119.7(3)

N(1)-C(1)

1.337(4)

C(15)-N(3)-C(17)

119.9(3)

N(1)-C(5)

1.340(4)

N(1)-C(1)-C(2)

122.6(3)

N(2)-C(6)

1.271(4)

N(1)-C(1)-C(15)

115.4(3)

N(2)-C(8)

1.425(4)

C(2)-C(1)-C(15)

122.0(3)

N(3)-C(15)

1.264(4)

C(1)-C(2)-C(3)

118.5(3)

N(3)-C(17)

1.426(5)

C(4)-C(3)-C(2)

119.4(4)

C(1)-C(2)

1.377(5)

C(3)-C(4)-C(5)

118.9(3)

C(1)-C(15)

1.500(5)

N(1)-C(5)-C(4)

122.1(3)

C(2)-C(3)

1.378(5)

N(1)-C(5)-C(6)

115.4(3)

C(3)-C(4)

1.374(5)

C(4)-C(5)-C(6)

122.4(3)

C(4)-C(5)

1.377(5)

N(2)-C(6)-C(7)

125.2(3)

C(5)-C(6)

1.497(5)

N(2)-C(6)-C(5)

116.9(3)

C(6)-C(7)

1.489(5)

C(7)-C(6)-C(5)

117.8(3)

C(8)-C(14)

1.381(5)

C(14)-C(8)-C(9)

120.5(4)

C(8)-C(9)

1.390(5)

C(14)-C(8)-N(2)

120.8(4)

C(9)-C(11)

1.389(5)

C(9)-C(8)-N(2)

118.6(4)

C(9)-C(10)

1.512(6)

C(11)-C(9)-C(8)

117.6(4)

C(11)-C(12)

1.373(6)

C(11)-C(9)-C(10)

122.0(4)

C(12)-C(13)

1.374(6)

C(8)-C(9)-C(10)

120.4(4)

C(13)-C(14)

1.379(5)

C(12)-C(11)-C(9)

122.6(4)

C(15)-C(16)

1.493(5)

C(11)-C(12)-C(13)

118.5(4)

C(17)-C(18)

1.383(6)

C(12)-C(13)-C(14)

120.7(4)

C(17)-C(23)

1.384(6)

C(12)-C(13)-Cl(1)

120.0(4)

C(18)-C(20)

1.387(6)

C(14)-C(13)-Cl(1)

119.3(4)

C(18)-C(19)

1.504(6)

C(13)-C(14)-C(8)

120.1(4)

113

Bond lengths [Å]

Angles [°]

C(20)-C(21)

1.371(6)

N(3)-C(15)-C(16)

125.5(3)

C(21)-C(22)

1.359(6)

N(3)-C(15)-C(1)

116.8(3)

C(22)-C(23)

1.393(6)

C(16)-C(15)-C(1)

117.7(3)

C(18)-C(17)-C(23)

121.2(4)

C(18)-C(17)-N(3)

120.9(4)

C(23)-C(17)-N(3)

117.7(4)

C(17)-C(18)-C(20)

117.4(4)

C(17)-C(18)-C(19)

121.1(4)

C(20)-C(18)-C(19)

121.5(4)

C(21)-C(20)-C(18)

122.8(5)

C(22)-C(21)-C(20)

118.5(4)

C(21)-C(22)-C(23)

121.4(5)

C(21)-C(22)-Cl(2)

119.8(4)

C(23)-C(22)-Cl(2)

118.8(4)

C(17)-C(23)-C(22)

118.7(4)

2.7 Synthesis of the 2,6-bis(arylimino)pyridine compounds 1-3 To a solution of 0.82 g (5 mmol) 2,6-diacetylpyridine in 150 ml of toluene were added 12,5 mmol (2,5 equivs.) of a substituted aniline and a few milligrams of paratoluenesulfonic acid. The reaction mixture was heated under reflux for 8-24 hours applying a Dean-Stark-trap. After cooling to room temperature, 200 ml of a saturated sodium hydrogencarbonate solution were added, the organic phase was separated and filtered over sodium sulfate and silica. The solvent was removed and 20 ml of methanol were added. The imino compounds precipitated when stored at - 20 °C for some days. After filtration and washing with cold methanol, the products were dried in vacuo. 2.8 Synthesis of the mono(imino)pyridine compound 4 To a solution of 0.82 g (5 mmol) 2,6-diacetylpyridine in 150 ml of toluene were added 5 mmol (1 equiv.) of 5-chloro-2-methylaniline and a few milligrams of paratoluenesulfonic acid. The reaction mixture was heated under reflux for 8 hours applying a Dean-Stark-trap. After cooling to room temperature, 200 ml of a saturated sodium hydrogencarbonate solution were added, the organic phase was separated and filtered over sodium sulfate and silica. The solvent was removed and 20 ml of methanol were added. The imino compound precipitated when stored at - 20 °C over night. After filtration and washing with cold methanol, the product was dried in vacuo.

114

Table 3. NMR and MS data for compounds 1-4. Com-

1

H NMR

pound

a)

13

C NMR

3

8.39 d (2H, JHH = 7.94 Hz, PyH3), 3

7.90 t (1H, JHH = 7.94 Hz, PyH4), 3

7.15 d (2H, JHH = 7.92 Hz), 7.01 1

3

4

dd (2H, JHH = 7.92Hz, JHH = 3.17 4

Hz), 6.72 d (2H, JHH = 3.17 Hz),

b)

MS [m/z]

c)

°+

167.5 (Cq, C=N), 154.9 (Cq, PyC2/6),

409 M

150.8 (Cq, C-N), 136.9 (CH), 131.6

394 M - Me (100)

(Cq, C-Cl), 131.4 (CH), 125.5 (Cq, C-

284 M – (Cl-Ph-

Me), 123.4 (CH), 122.5 (CH), 118.0

Me) (12)

(CH), 17.2 (Ar-CH3), 16.4 (N=C-CH3)

125 (Cl-PhMe)(40)

2.36 s (6H, N=C-CH3), 2.08 s (6H, Ar-CH3) 3

8.35 d (2H, JHH = 7.71 Hz, PyH3), 3

7.85 t (1H, JHH = 7.71 Hz, PyH4), 3

7.11 d (2H, JHH = 7.63 Hz), 6.99 2

3

4

dd (2H, JHH = 7.63 Hz, JHH = 3.02 4

Hz), 6.69 d (2H, JHH = 3.02 Hz),

°+

167.1 (Cq, C=N), 161.3 (Cq, PyC2/6),

499 M

150.7 (Cq, C-N), 137.1 (CH), 132.6

484 M - Me (37)

(Cq, C-Me), 131.5 (CH), 125.9 (Cq, C-

328 M – (Br-Ph-

Br), 123.7 (CH), 121.0 (CH), 113.8

Me) (10)

(CH), 17.8 (Ar-CH3), 16.4 (N=C-CH3)

169 (Br-PhMe)(38)

2.28 s (6H, N=C-CH3), 2.04 s (6H, Ar-CH3) 3

8.41 d (2H, JHH = 8.36 Hz, PyH3),

377 M

155.0 (Cq, PyC2/6), 151.0 (Cq, C-N),

362 M - Me (100)

136.9 (CH), 131.2 (CH), 122.5 (CH),

268 M – (F-Ph-

dd (2H, JHH = 8.85 Hz, JHH = 3.42

122.3 (Cq, C-Me), 110.1 (CH), 105.2

Me) (10)

4

(CH), 17.0 (Ar-CH3), 16.4 (N=C-CH3)

109 (F-Ph-Me)(76)

199.3 (Cq, C=O), 169.6 (Cq, C=N),

286 M

155.3 (Cq, PyC2/6), 151.2 (Cq, C-N),

271 M - Me (100)

135.7 (CH, PyC4), 131.8 (Cq, C-Cl),

251 M - Cl (16)

131.0 (CH), 129.1 (CH), 125.5 (Cq, C-

166 (37)

7.92 t (1H, JHH = 8.36 Hz, PyH4), 3

7.17 d (2H, JHH = 8.85 Hz), 6.75 3

°+

167.5 (Cq, C=N), 163.4 (Cq, C-F),

3

3

4

Hz), 6.45 d (2H, JHH = 3.42 Hz), 2.35 s (6H, N=C-CH3), 2.15 s (6H, Ar-CH3) 3

8.41 d (2H, JHH = 8.05 Hz, PyH3), 3

7.87 t (1H, JHH = 8.05 Hz, PyH4), 3

7.15 d (2H, JHH = 7.87Hz), 7.03 4

3

4

dd (2H, JHH = 7.87 Hz, JHH =3.02 4

Hz), 6.71 d (2H, JHH = 3.02 Hz),

°+

Me), 123.4 (CH), 122.7 (CH), 118.1

2.77 s (3H, O=C-CH3), 2.24 s (3H,

(CH), 25.5 (CH3), 17.6 (O=C-CH3),

N=C-CH3), 2.03 s (3H)

16.4 (N=C-CH3)

a) 25 °C, in CDCl3, rel. CHCl3, δ = 7.24 ppm b) 25 °C, in CDCl3, rel. CDCl3, δ = 77.0 ppm c) in brackets: intensity of the ion peak in relation to the base peak

2.9 General synthesis of 2,6-bis(arylimino)pyridine transition metal complexes 5-19 An amount of 0.5 mmol of the 2,6-bis(arylimino)pyridine compound was dissolved in 20 ml 1-butanol or 20 ml THF and reacted with 0.5 mmol of the desired water free metal salt mostly resulting in an immediate colour change. The mixture was stirred for 3-5 hours at room temperature whereby the complexes precipitated. In case of the nickel complexes, it was necessary to keep the reaction mixture under reflux. n-Pentane (10 ml) was added for complete precipitation. The complexes were filtered over a glass frit, washed three times with 15 ml n-pentane, and dried in vacuo. If necessary, the complexes were recrystallized from methanol or methylene chloride.

115

Table 4. MS, IR, and elemental analysis data of the complexes 5-19. Nr.

MS m/z (%)

complex

Cl

Cl

Cl

N

N

Cl

N

Cl

Cl

Cl

N

N

Cl

N

Cl

Cl

Cl

N

N

Cl

N

Cl

Cl

N

N

9

N

Cl

Cl

N

N

Cl

N

Cl

11

N

N

Cl

Cl

539 M • 504 M-Cl (2) 409 M-NiCl2 (35) 394 M-NiCl2-Me (75)

Cl

N

N N

48.69 48.93

3.61

3.75

7.30

7.44

1578

48.31 48.67

3.75

3.73

7.29

7.40

1654

48.41 48.58

3.79

3.72

7.24

7.39

1633, 1593

51.23 51.52

3.88

3.95

7.69

7.84

1626, 1593

51.28 51.43

3.99

3.94

7.76

7.82

1633, 1593

48.06 48.25

3.67

3.70

7.22

7.34

n.d.

51.02 51.14

3.85

3.92

7.63

7.78

n.d.

51.05 51.16

3.78

3.92

7.67

7.78

+

Cl Ni

12

n.d.

+

N

Cl

571 M • 536 M-Cl (20) 500 M-2Cl (25) 409 M-FeCl3 (45) 394 M-FeCl3-Me (100) 540 M • 505 M-Cl (6) 409 M-CoCl2 (37) 394 M-CoCl2-Me (100)

Cl Co

Cl

Ntheor [%]

+

Cl

Fe

10

Nexp [%]

+

N

Cl

568 M • 533 M-Cl (3) 498 M-2Cl (7) 409 M-CrCl3 (21) 394 M-CrCl3Me(100)

537 M • 502 M-Cl (1) 409 M-FeCl2 (21) 394 M-FeCl2-Me (100)

Cl

N

Cl

567 M • 532 M-Cl (2) 497 M-2Cl (3) 409 M-VCl3 (37) 394 M-VCl3-Me (100)

Cl

Fe Cl

Htheor [%]

+

N

Cl

564 M • 529 M-Cl (3) 409 M-TiCl3 (29) 394 M- TiCl3-Me (100)

536 M • 501 M-Cl (2) 409 M-MnCl2 (19) 394 M-MnCl2-Me (100)

Cl Mn

8

Hexp [%]

+

Cl

Cr

7

Ctheor [%]

+

Cl

V

6

Cexp [%]

+

Cl

Ti

5

IR ν(C=N) -1 [cm ]

116

Nr.

MS m/z (%)

complex Br

Cl

N

N

Cl

N

Br

Br

Cl

N

N

Cl

N

Br

15

N

N

Br

Br

625 M • 590 M-Cl (1) 546 M-Br (20) 499 M-FeCl2 (59) 484 M-FeCl2-Me (94)

N N

Cl

F

N

N

F

N

Cl

Cl

N

504 M • 469 M-Cl (1) 378 M-FeCl2 (10) 363 M-FeCl2-Me (100)

Ntheor [%]

1593

43.98 44.13

3.29

3.38

6.64

6.71

1591

39.05 39.13

3.06

3.00

5.84

5.95

n.d.

43.66 43.93

3.30

3.37

6.55

6.68

n.d.

43.97 44.13

3.32

3.38

6.54

6.71

1605

54.34 54.79

4.17

4.20

8.19

8.33

n.d.

46.22 46.47

3.70

3.66

6.59

6.77

n.d.

41.64 41.76

3.18

3.20

6.22

6.35

+

413 M • 378 M-Cl (3) 286 M-FeCl2 (42) 271 M-FeCl2-Me (100)

Cl Fe

18

Nexp [%]

+

Cl Fe

17

Htheor [%]

+

Cl

N

705 M • 625 M-Br (20) 546 M-2Br (35) 409 M-FeBr3 (50) 394 M-FeBr3-Me (100)

Cl

Fe

16

Hexp [%]

+

N

Cl

625 M • 546 M-Br (40) 409 M-FeBr2 (50) 394 M-FeBr2-Me (100)

628 M • (not visible) 549 M-Br (5) 409 M-NiBr2 (46) 394 M-NiBr2-Me (100)

Br Ni

Cl

Ctheor [%]

+

Br

Fe

14

Cexp [%]

+

Br Fe

13

IR ν(C=N) -1 [cm ]

O N

+

Cl

Cl

Cl

Fe

19

Br

N

N N

Br

660 M • (not visible) 625 M-Cl (4) 590 M-2Cl (3) 546 M-Cl-Br (19) 499 M-FeCl3 (67) 484 M-FeCl3-Me (100)

2.10 Oligomerization of ethylene at low pressure An amount of 0.1 – 0.3 mmol of the desired complex was placed in a Schlenk tube and suspended in 100 ml of toluene or n-pentane. After activation with methyl alumoxane (30% in toluene), an ethylene pressure of 0.5 bar or 1.0 bar was applied and the mixture was stirred for one hour at room temperature. The reaction was stopped by releasing the pressure. The mixture was carefully poured into 100 ml of diluted hydrochloric acid. When a polymer was obtained, it was separated by filtration

117

using a Büchner funnel. The polymer was washed with water and acetone and finally dried in vacuo. The liquid organic phase was washed twice with 50 ml of water and dried over sodium sulfate. The resulting solutions were analyzed by gas chromatography. Including the weight increase of the solutions, the activities were calculated by integration of the GC peaks.

3. Results and discussion 3.1 Synthesis and characterization of 2,6-bis(arylimino)pyridine compounds Condensation reactions of 2,6-diacetylpyridine with 5-halogen-2-methyl substituted anilines yielded the 2,6-bis(arylimino)pyridine compounds 1 – 3 (see Scheme 1). R

R O

R

O

toluene p-TosOH

N 2

NH2

N

N

reflux - 2 H 2O

N

R = Hal Compound

R

Yield

1

Cl

78

2

Br

63

3

F

49

Scheme 1. Synthesis of 5-halogen-2-methyl substituted 2,6-bis(arylimino)pyridine compounds. Additionally, the monosubstituted compound 4 was prepared from 2,6-diacetylpyridine and 5-chloro-2-methylaniline (Scheme 2) in a 61% yield.

Cl

O

N N

Scheme 2. Monosubstituted compound 4. The compounds 1 – 4 were characterized by GC/MS, 1H NMR and spectroscopy. The spectra of compound 1 are discussed representatively.

118

13

C NMR

394 284 N

Cl

125

243

N

Cl

N

M+

166 M=409

Scheme 3. Mass spectrum of 2,6-bis(5-chloro-2-methylphenyl-1-ethaneimino)pyridine (1). The molecule ion at m/z = 409 is clearly visible in the mass spectrum of 1 (Scheme 3). Due to the chlorine substituents, the peak shows a characteristic isotope pattern which fits excellently to the theoretically calculated distribution (see Scheme 4).

measured

calculated

Scheme 4. Isotope pattern of the molecular ion of compound 1. The base peak at m/z = 394 results from the loss of one iminomethyl group. Again, the characteristic isotope pattern can be observed. The peaks at m/z = 266 and m/z = 243 can be explained by α-cleavage reactions starting from the imino nitrogen atom and the nitrogen atom of the pyridine ring. The loss of one of the substituted phenyl rings gives a peak at m/z = 284.

119

19

Cl

20

18

21

22

23

17 16

2,4 7

N 6

N 5 4

N

1

8

11 10

12

15

13 14

Cl

22,23

9

15,21

2 3

12,18 13,19 3 7,9 CHCl3

2,4

13,19 12,18

15,21

3 CHCl3

Scheme 5. 1H NMR spectrum of compound 1. The 1H NMR spectrum of 1 (see Scheme 5) shows a dublet at δ = 8.39 ppm that can be assigned to the meta protons (2,4) at the pyridine ring. The corresponding triplet at δ = 7.90 ppm stems from the para proton (3) of the pyridine ring. The phenyl protons appear at δ = 7.15 ppm (12,18), δ = 7.01 ppm (13,19), and δ = 6.72 ppm (15,21). Finally, the two singlets at δ = 2.36 ppm and δ = 2.08 ppm can be assigned to the iminomethyl groups (7,9) and the methyl groups at the phenyl rings (22,23).

120

15,21

2,4

12,18 13,19 7,9

22,23

3 19

Cl

20

18

21

22

23

17 16

7

N 6

N 5 4

1,5

1

8

10

15

13 14

Cl

9

2 3

CDCl3

11,17 6,8

N

11

12

10,16 14,20

Scheme 6. J-modulated 13C NMR spectrum of compound 1. A J-modulated

13

C NMR spectrum (Scheme 6) was recorded from compound 1.

The resonance signal at δ = 167.5 ppm can be assigned to the imino carbon atoms (6,8). The quaternary carbons of the pyridine ring (1,5) give the peak at δ = 154.9 ppm followed by the signal for the nitrogen-bonded carbon atoms of the phenyl rings (10,16) at δ = 150.8 ppm. The para carbon atom of the pyridine ring (3) yields the signal at δ = 136.9 ppm. The signal for the chloro substituted carbon atoms (14,20) appears at δ = 131.6 ppm. At δ = 131.4 ppm, the signal for the carbon atoms 12 and 18 can be found. The methyl substituted quaternary carbon atoms of the phenyl rings (11,17) give the signal at δ = 125.5 ppm, while the meta-standing carbon atoms of the pyridine ring (2,4) produce the signal at δ = 123.4 ppm. The unsubstituted ortho-carbon atoms of the phenyl rings (15,21) appear at δ = 122.5 ppm, followed by the signal for the para-CH groups (13,19) at δ = 118.0 ppm. The methyl groups at the phenyl rings (22,23) give the signal at δ = 17.2 ppm, while the iminomethyl groups (7,9) yield the signal at δ = 16.4 ppm. After crystallization from diethylether, single crystals of 1 were obtained which were subjected to X-ray analysis.

121

Scheme 7. X-ray structure of compound 1. The structure of 1 is analogous to already published structures showing other substitution patterns at the iminophenyl rings. The crystal data can be found in the Experimental part. 3.2 Synthesis of 2,6-bis(arylimino)pyridine transition metal complexes Using the bis(arylimino)pyridine compounds 1-3 and the mono(imino)pyridine compound 4, a series of coordination compounds including the 3d transition metals from titanium to nickel was prepared (see Scheme 8 and Table 5). Titanium, vanadium, and chromium were applied in the oxidation state +III, while manganese, cobalt, and nickel were used in the oxidation state +II. In case of iron, both iron(II) and iron(III) complexes were prepared. After dissolving the 2,6-bis(arylimino)pyridine compound in 1-butanol, THF, or diethylether, the corresponding metal salt was added resulting in an immediate color change. In most cases, the complexation reactions were completed within three hours. The complexes could be isolated in very high yields (80-95 %). For the nickel complexes, (dme)NiBr2 and (dme)NiCl2 were prepared as starting materials

according

to

Nylander[63].

THF

adducts

of

titanium(III)chloride,

vanadium(III)chloride, and chromium(III)chloride were prepared following a general procedure[64,65]. All complexes were characterized by mass spectrometry, IR, and elemental analysis. Additionally, the magnetic moments of the complexes were determined using the Evans NMR method[66-68].

122

R

R

R MXn

N

M

1-BuOH, Et2O or THF r.t., 3-5 h (reflux for Ni complexes)

N N

R

Xn

N

N N

R = Hal X = Hal n = 2; 3 Scheme 8. Synthesis of 2,6-bis(arylimino)pyridine transition metal complexes. Table 5. Synthesized 2,6-bis(arylimino)pyridine transition metal complexes. Nr.

complex

Cl

Cl

R

M

n

X

Cl

N

N

Cl

Cl

Ti

3

Cl

N

Cl

Cl

V

3

Cl

N

Cl

Cl

Cr

3

Cl

N

Cl

Cl

Mn

2

Cl

Cl

Fe

2

Cl

N

Cl

Cl

Cl

N N

Cl

Cl

Cl

N N

Cl

THF

black

91

Et2O

red

96

[%]

TiCl3• 3 THF

VCl3• 3 THF

Cl

Cr

7

colour

Cl

V

6

yield

solvent

Cl

Ti

5

educt

CrCl3• 3 THF

THF

dark green

87

Cl Mn

8

Cl

N N

Cl

MnCl2• 2 THF

THF

yellow

89

n-BuOH

blue

92

Cl Fe

9

Cl

N

N N

Cl

123

FeCl2

Nr.

complex

Cl

Cl

R

M

n

X

educt

solvent

colour

yield [%]

Cl

Fe

10

Cl

N

N

Cl

Cl

Fe

3

Cl

FeCl3

n-BuOH

orange

93

N

Cl

Cl

Co

2

Cl

CoCl2

n-BuOH

green

90

N

Cl

Cl

Ni

2

Cl

NiCl2•

THF

DME

(boi-ling)

orange

91

blue

88

N

Cl

Cl Co

11

Cl

N N

Cl

Cl Ni

12

Cl

N N

Br

Br Fe

13

Cl

N

N

Cl

Cl

Fe

2

Br

FeBr2

n-BuOH

N

Cl

Cl

Fe

3

Br

FeBr3

n-BuOH

N

Cl

Cl

Ni

2

Br

NiBr2•

THF

DME

(boi-ling)

N

Br

Br

Br

Fe

14

Cl

N N

Br

dark brown

86

Br Ni

15

Cl

N N

Cl

orange

95

Cl Fe

16

Br

N

N

Br

Br

Fe

2

Cl

FeCl2

n-BuOH

blue

94

N

F

F

Fe

2

Cl

FeCl2

n-BuOH

blue

85

N

Cl

Cl Fe

17

F

N N

124

Nr.

complex Cl

R

M

n

X

educt

solvent

colour

Cl

Fe

2

Cl

FeCl2

n-BuOH

blue

Br

Fe

3

Cl

FeCl3

n-BuOH

yield [%]

Cl Fe

Cl

18

N

O N

Cl

Cl

Cl

Fe

19

Br

N

90

N

Br

N

Cl

dark brown

87

Cl Fe

Cl

N

N

Cl

N

Scheme 9. IR spectrum of complex 9. The spectrum shows the charateristic ν (C=N) band at 1626 cm-1. Due to the coordination of iron(II)chloride, the band is shifted to lower energy compared with the ligand precursor (ν = 1638 cm-1). The bands at 1593 cm-1, 1484 cm-1, 1267 cm-1, and 811 cm-1 are characteristic for the substituted phenyl rings. Scheme 10 shows the mass spectrum of complex 9. The molecular ion appears at m/z = 537. The loss of FeCl2 results in the formation of the peak at m/z = 409

125

corresponding to the bis(arylimino)pyridine ligand. Peaks below this value can be explained in analogy to ligand precursor 1.

Cl

Cl Fe

Cl

N

N

Cl

N

C23H21Cl4FeN3 537.10

M+

Scheme 10. Mass spectrum of 9. Analogously to the bis(arylimino)pyridine compound, the isotope pattern agrees very well with the theoretically calculated distribution (Scheme 11).

126

measured:

Calculated:

Scheme 11. Isotope pattern for the molecule ion of complex 9. The magnetic moments of the transition metal complexes were determined applying the Evans NMR method. Since the effective magnetic moments µeff directly correspond with the electronic configuration, the electronic ground states of the 2,6bis(arylimino)pyridine metal complexes can be obtained[69] (see Table 6). Table 6. Magnetic moments µeff and number of unpaired electrons in complexes derived from ligand precursor 1. complex

metal center

µeff

unpaired electrons

6

V(III)

2.97

2

7

Cr(III)

4.77

3

8

Mn(II)

5.92

5

9

Fe(II)

5.40

4

11

Co(II)

4.09

3

The knowledge of the electronic ground states plays an important role for “ab initio” calculations concerning the theoretical investigation of the oligomerization reactions.

127

3.3 Results of the homogeneous ethylene oligomerization and polymerization After activation with methylalumoxane (MAO), the transition metal complexes 519 were used as catalyst precursors for the homogeneous oligomerization of ethylene. The influences of different reaction parameters (metal, substituents at the ligand framework, ethylene pressure, temperature, Al:M ratio) were investigated (Table 7). To confirm the stability of the halogenated bis(arylimino)pyridine ligands against trimethylaluminum/methylalumoxane, samples of the complexes were activated with MAO. The mixtures were hydrolyzed after five minutes, worked up, and analyzed by GC/MS revealing that the ligand systems remained unchanged. The oligomerization products were characterized by GC and GC/MS and the Schulz-Flory coefficient α was determined[70-73]. Data analysis was performed using a computer program which was developed for this special purpose[74]. Table 7. Oligomerization and polymerization results for the complexes 5-19 (solvent: 250 ml toluene, activator: MAO, 1h). uneven olefins [%]

Schulz-Flory coefficient α

82

-

polymer

22.0

615

-

0.71

72

3.7

105

-

0.88

250

35

1.9

54

-

polymer

25

250

504

28.1

782

4.4

0.77

0.5

0

250

1505

84.0

2337

2.7

0.80

9

0.5

25

250

473

26.4

734

3.6

0.83

9

0.5

50

250

433

24.2

672

3.4

0.73

9

0.5

75

250

25

1.4

39

-

0.77

10

1

25

250

783

43.7

1216

3.8

0.77

10

0.5

25

150

390

21.8

605

1.8

0.76

10

0.5

25

250

705

39.3

1095

2.5

0.78

10

0.5

25

350

584

32.6

907

3.9

0.82

10

0.5

25

500

462

25.8

717

6.8

0.76

10

0.5

25

750

1363

76.1

2116

6.0

0.82

10

0.5

25

1000

658

36.7

1022

5.3

0.81

10

0.5

25

1500

678

37.8

1053

3.4

0.83

Complex

p [bar]

T [°C]

Al:M

Activity [g/g M·h]

Activity [kg/mol M·h]

5

0.5

25

250

61

2.9

6

0.5

25

250

431

7

0.5

25

250

8

0.5

25

9

1

9

128

TOF [mol C2H4/ -1 mol Cat·h ]

uneven olefins [%]

Schulz-Flory coefficient α

952

2.4

0.83

0.8

21

-

polymer

80

4.7

131

-

polymer

250

126

7.4

207

-

polymer

25

250

297

16.6

461

7.3

0.80

1

25

500

198

11.0

307

19.2

0.78

14

0.5

25

250

342

19.1

531

5.2

0.80

15

0.5

25

250

71

4.2

117

-

polymer

16

0.5

25

250

1690

94.3

2624

-

0.85

17

0.5

25

250

890

49.7

1382

2.8

0.82

18

0.5

25

250

3720

207.6

5775

-

0.90

19

1

0

750

2970

165.7

4611

-

0.80

Complex

p [bar]

T [°C]

Al:M

Activity [g/g M·h]

Activity [kg/mol M·h]

10

0.5

25

2000

613

34.2

11

0.5

25

250

13

12

0.5

25

250

12

1

25

13

0.5

13

TOF [mol C2H4/ -1 mol Cat·h ]

The following mechanism is proposed for oligomerization and polymerization reactions using bis(imino)pyridine transition metal complexes[57,58]:

N

N X

N

X N

Me

MAO

M

N

N

M

methylation methyl abstraction

N MAO-Me

N

Me N

coordination

M N

free coordination site

MAO-Me

N

N

Me N

M N

insertion

N

M

N

Me N

MAO-Me

M N

MAO-Me

MAO-Me

Scheme 12. Proposed mechanism of ethylene oligomerization and polymerization with bis(imino)pyridine transition metal complexes. As the main chain termination reactions, β-hydrogen elimination, β-hydrogen transfer, or chain transfer to aluminum centers can be considered. The GC spectrum of an oligomer mixture obtained with 9/MAO is shown in Scheme 13.

129

C-10

Intensität [mV]

Intensität [mV]

C-10

C-12

C-12 C-9

C-14

C-11 τ [min]

C-18 C-22

C-26

τ [min] Cl

Cl Fe

Cl

N

Polymerization conditions: 250ml toluene; MAO (Al : Fe = 250 : 1); 0.5 bar ethylene; 25°C; 1h.

Cl

N N

Scheme 13. GC spectrum of an oligomer mixture obtained with 9/MAO. The mixture consists of olefins with carbon numbers between 6 and 34. Since the GC integrals are proportional to the molar amount of each component, the logarithm of the GC integrals can be plotted against the carbon number to check whether the obtained distribution matches an Anderson-Schulz-Flory distribution[75]. 6.3 6.2

log [CnH2n]

log[CnH2n]

6.1 6 5.9 5.8 5.7 5.6

y = -0.0426x + 6.6744 R2 = 0.9973

5.5 8

10

12

14

16

18

20

22

24

26

28

C-Atom -Zahl

Number of carbon atoms

Scheme 14. Plot of the logarithmic GC integrals against the carbon numbers.

130

The coefficient of determination R2 is an indicator for the accordance with the Anderson-Schulz-Flory theory. In Scheme 14, R2 is 99.7 %. For each distribution, the characteristic Schulz-Flory coefficient α can be determined: α =

kpropagation kpropagation + ktermination

=

mol (Cn+2) mol (Cn)

A higher coefficient α directly corresponds to an increased propagation probability resulting in higher molecular weight products. The upper limit α = 1 is never reached. Interestingly, small amounts of olefins with uneven numbers of carbon atoms could be detected by GC/MS analyses in some of the oligomer mixtures (see enlarged part of Scheme 13). This result was proved by comparing the GC/MS data with the results obtained for uneven numbered α-olefins used as references (1-nonene, 1undecene). The GC retention times are identical and the fragmentation patterns in the mass spectra agree very well. 3.3.1 Influence of the metal center on the oligomerization activities and the isomerization potentials The influence of different metal centers in complexes bearing one and the same bis(imino)pyridine ligand on the ethylene oligomerization and polymerization activities is shown in Scheme 15. 800 705

-1

Activity [g(PE)/g(M)*h ]

700 600 500

473

431

400 300 200 100

72

61

80

35

13

0 TiCl3 5

VCl3 6

CrCl3 7

MnCl2 8

FeCl2 9

FeCl3 10

CoCl2 11

NiCl2 12

Scheme 15. Ethylene oligomerization activities of complexes 5-12 bearing bis(imino)pyridine compound 1 as chelating ligand. Polymerization conditions: 250 ml of toluene; (Al:M = 250:1); 0.5 bar ethylene; 25 °C; 1h. The highest activities were obtained with the vanadium complex 6 and the iron complexes 9 and 10 while the cobalt complex 11 showed the lowest activity in this series. Only the iron catalysts 9 and 10 gave small amounts of uneven numbered

131

olefins. The overall contents of these uneven numbered olefins in the obtained mixtures are 3.6 % for 9/MAO and 2.5 % for 10, respectively. In case of the complexes 5 (Ti), 8 (Mn), 11 (Co), and 12 (Ni), only polymeric products were produced. Differences in their potential to isomerize α-olefins can be found for the vanadium complex 6, the chromium complex 7 and both iron complexes (Scheme 16).

Cl

Cl

Cl

Cl

N

N

Cl

Cl

N

6

7

τ [min]

τ [min]

Cl

Cl

Cl

N

Intensity [mV]

N

9

Cl

Intensity [mV]

N

Cl

Cl

Fe

Fe Cl

Cl

N N

Intensity [mV]

N

Intensity [mV]

Cl

Cr

V Cl

Cl

τ [min]

Cl

N

N

Cl

N

46

τ [min]

Scheme 16. Parts of the GC spectra obtained for the oligomer mixtures using 6/MAO, 7/MAO, 9/MAO, and 10/MAO. The pictures show the isomers in the region of C-10 up to C-12. The chromium complex 7 showed the highest selectivity for α-olefins while both iron complexes yielded also traces of other isomers of the even numbered alkenes and small amounts of uneven numbered olefins. In contrast, the mono(imino)pyridine iron complex 18 did not produce any uneven numbered oligomers but showed a quite high activity (3720 g/g Fe · h-1). The vanadium complex 6 produced a whole series of

132

isomers of even numbered olefins but no uneven numbered olefins were observed. The following mechanism is assumed for the isomerization of α-olefins:

N N

M

n

H

β-hydrogen elimination

N

N

N

n

M

H

N

M

hydrogen transfer to the coordinated olefin

N

N

H

N

N

n

N

N M

H

n

N

N

β-hydrogen elimination

M N

ethylene insertion

n

2-olefins

H

further isomerization

internal olefins ethylene insertion

methyl side chains ethyl, propyl side chains

Scheme 17. Proposed mechanism for the isomerization reactions of α-olefins. According to Brookhart[2,24,29,30] and Ziegler[57,58], a so-called “chain-running” mechanism is supposed. The decisive reaction step is the β-hydrogen elimination, since the interaction of the metal center and a β-hydrogen atom affects the degree of isomerization. Using quantum chemical calculations, the degree of these β-agostic interactions can be determined. Therefore, two transition structures were calculated for the proposed active species of the complexes 6, 7, and 9, whereby the first structure is always described with β-hydrogen interaction, the second one without β-hydrogen interaction. The energy differences between the two species can be explained as the degree of interaction. For the calculations, cationic catalyst species bearing a propyl substituent at the metal center were used and the potential energies were minimized using B88LYP. The structures of the catalyst precursors were optimized with MM3. Substituents were introduced or exchanged keeping the main geometry constant. In case of the vanadium complex 6, the active species described by Gambarotta[31] was used (Scheme 18, I, addition of a methyl group to C-2 of the pyridine ring), while the free coordination site in the chromium complex 7 is occupied by a methyl group according to the active species presented by Esteruelas[34] (Scheme 18, II). Calculations for the activated iron complex 9 were performed applying the structure parameters used by Gibson[76] and Ziegler[58] (Scheme 18, III). To rule out geometric effects, an iron complex with an analogous structure to the vanadium complex was used (Scheme 18, IV).

133

H V

V

I

Cl

N

N

Cl

Cl

N

N

Cl

N

Cl

N

Cl

N

Cl

N

N Me

Me

H Cr

Cr

II

Cl

N

N

Cl

Cl

N N

N

H Fe

Fe

III

Cl

N

N

Cl

Cl

N N

N

H Fe

Fe

IV

Cl

N

N

Cl

Cl

N N

N Me

Me

Scheme 18. Structures of proposed active species (left: without β-agostic interaction; right: with β-agostic interaction) of the complexes 6 (I), 7 (II), and 9 (III) used for the calculations of β-agostic interaction energies. All structures were optimized with MM3 to reduce the calculation times. The energy calculations were performed starting the dGauss algorithm (CaChe 6.1)[77,78] applying DFT on the basis of SCF optimizations. In dGauss, geometry optimizations on the basis of the calculated energy gradients are carried out applying the BroydenFletcher-Goldfarb-Shanno method[79]. For all SCF and gradient calculations the local exchange potentials were used as described by Vosko, Wilk, and Nusair[80]. Non local corrections were then computed on the basis of the resulting geometries (local spin density geometry) and electron densities. For correlation, the „Becke ´88 Functional”[81]

134

is used including the additional specifications from Lee and Miehlich[82,83] for exchange interactions. DZVP[84] was used as a basic set for orbital calculations. For the corresponding atoms the following orbitals were included into the calculations: Table 8. Electron configurations for „ab initio“ calculations. atom H C, N Cl V, Cr, Fe

orbitals 1s,2s 1s, 2s, 2p, 3s, 3p, 3d 1s, 2s, 2p, 3s, 3p, 3d, 4s, 4p 1s, 2s, 2p, 3s, 3p, 3d, 4s, 4p, 4d, 5s

The energy contents of each structure were computed and the energies of the βhydride interactions were calculated as the energy differences between the structures without and with β-agostic interactions (see Scheme 19 and Table 9). I

Chromium complex 6

II

Vanadium complex 7

135

III

Iron(II) complex 9

IV

Iron(II) complex 9 (modified in analogy to the vanadium complex 7[45]

Scheme 19. Energy differences between the structures without (left) and with (right) βagostic interactions for the catalysts 7-9. Table 9. Energy differences between the structures with and without β-agostic interactions. metal center

∆E kcal/mol

vanadium

±0

chromium

+ 168.7

iron

+ 62.8

iron (V structure)

+ 133.5

It is clearly visible that an agostic interaction between the metal center and a hydrogen atom at the β-position of the growing chain affects the geometry of the whole molecule. These structural changes are responsible for the resulting energy differences which are in good agreement with the oligomerization behavior of the catalyst systems 6/MAO, 7/MAO, and 9/MAO. In case of the chromium complex 7, the largest energy difference was found corresponding to a high selectivity towards αolefins, since β-hydrogen elimination and isomerization are energetically unfavorable. Both calculated structures of the iron complex 9 also show an increased total energy

136

when β-agostic interactions are assumed. In contrast to the chromium complex 7, isomerization reactions are not completely prevented but occur at a very low level. In case of the vanadium complex 6, both structures are energetically equivalent. Therefore, isomerization reactions can proceed on a large scale. 3.3.2 Formation of olefins with uneven numbers of carbon atoms Among the synthesized transition metal complexes only iron complexes showed the potential to produce olefins with uneven numbers of carbon atoms. Different reaction pathways can be assumed for the formation of these unusual products. Siedle et al.[85] proposed the transfer of methyl groups from MAO to the catalytically active species, while the growing chain is transferred to an aluminum center (Scheme 20). This kind of reaction would comply with a repeated activation step resulting in the formation of uneven numbered olefins.

N N

N n MeAlR2 N

Fe N

N

n

Fe

Me

AlR2

N

Fe

Me

N

N n AlR2

Scheme 20. Reaction pathway proposed by Siedle et al. for the formation of olefins with uneven numbers of carbon atoms. To verify this mechanism, oligomerization reactions were carried out using the catalyst systems 9/MAO and 10/MAO and 1-octene as the monomer. Both catalysts produced dimers (hexadecene isomers) and trimers (tetracosene isomers) of 1-octene, but there was no evidence for the formation of uneven numbered oligomers. Also, saturated hydrocarbons which should be formed by hydrolysis of the alkylaluminum species, could not be detected. Another possible mechanism known in the literature[86,87] is the β-carbon elimination pathway (Scheme 21). Usually this reaction is preferred by electron lacking d0 complexes[88].

N isomerization

N

Fe

N Me

n

N

β-carbon elimination

N

Fe N

Scheme 21. Mechanism of the β-carbon elimination.

137

n Me

Since only the iron complexes produced oligomers with uneven carbon numbers, this pathway does not seem to be relevant. A plausible mechanism including the experimental results is proposed in Scheme 22.

N

N

N isomerization N

Fe

n

N

n

Fe

metathesis

N

N

N

N

n

Fe

Scheme 22. Metathesis reaction generating oligomers with uneven numbers of carbon atoms. At the beginning, 1-olefins are isomerized to give the corresponding 2-olefins. The 2-olefins remain in the coordination sphere of the metal and undergo a metathesis reaction with another coordinated ethylene molecule resulting in olefins with uneven numbers of carbon atoms. Since the “chain-running” mechanism is energetically hindered for the investigated iron complexes, 2-olefins must be the main products of isomerization reactions. Their concentration is evidently high enough to undergo metathesis reactions. 3.3.3 Influence of the halogen substituents at the metal center For the preparation of complexes 13-15 metal bromides were applied instead of the metal chlorides. Compared to the chloride complexes, the bromide complexes showed lower activities (Scheme 23).

-1

Activity [g(PE)/g(Fe)*h ]

800

705

700 600 500

473

400

342

297

300 200 100

80

71

NiCl2 12

NiBr2 15

0 FeCl2 9

FeBr2 13

FeCl3 10

FeBr3 14

Scheme 23. Comparison of the ethylene oligomerization and polymerization activities of bis(imino)pyridine metal chloride and bromide complexes. Polymerization conditions: 250 ml of toluene; (Al:M = 250:1); 0.5 bar ethylene; 25 °C; 1h.

138

Contrarily, the overall contents of uneven numbered α-olefins increase when changing the metal halide from chlorine to bromine (Table 10): Table 10. Overall contents of uneven numbered α-olefins in the mixtures produced with 9/MAO, 10/MAO, 13/MAO and 14/MAO. Complex

Content of uneven numbered olefins [%]

9

3.6

10

2.5

13

7.3

14

5.2

This result can be explained by steric effects. If a bromide ligand is transferred to a MAO cage, the counter ions are better separated. Consequently, there is more space around the catalytically active center leading to an increased rate of metathesis reactions. The oligomer distributions are little influenced by the change from a metal chloride to the corresponding metal bromide. The α values only vary in the range 0.780.80. 3.3.4 Effect of different halogen substituents on the ligand frameworks The influence of different halogen substituents on the ligand framework was investigated with the iron(II) chloride complexes 9, 16 and 17 (Scheme 24).

1800

1690

-1

Activity [g(PE)/g(Fe)*h ]

1600 1400 1200 1000

890

800 600

473

400 200 0

F

Cl 17

Br 9

16

Scheme 24. Polymerization activities of iron(II) chloride complexes bearing different halogen substituents in their ligand frameworks. Polymerization conditions: 250 ml of toluene; (Al:M = 250:1); 0.5 bar ethylene; 25 °C; 1h.

139

Complex 16 with a bromo substituted ligand framework shows the highest activity among these three complexes. The activities of the fluoro and chloro substituted complexes 17 and 9 are apparently smaller. While the bromo substituted complex 16 did not give any uneven numbered olefins, the content of uneven numbered olefins in the mixture produced with 17/MAO is 2.8 %. In case of the bromo substituted complex 16, there is not enough space around the metal center for metathesis reactions due to the big halogen substituent. Since β-hydrogen elimination reactions or chain transfer reactions to aluminum centers are also hindered, an increased activity and a higher Schulz-Flory coefficient are observed. The Schulz-Flory coefficients α increase in the row F (α = 0.82) < Cl (α = 0.83) < Br (α = 0.85) providing higher molecular weight products by increasing the size of the halogen substituent. 3.3.5 Influence of the ethylene pressure Ethylene was oligomerized and polymerized at 0.5 bar and 1.0 bar applying the catalysts 9/MAO, 10/MAO, and 12/MAO. As can be seen in scheme 25, the activities increase with increasing pressure.

900 783

-1

Activity [g(PE)/g(Fe)*h ]

800 705 700 600 504

473

500 400 300 200

126

80

100 0 0.5 bar

1.0 bar

Cl

0.5 bar

Cl

Cl

Fe Cl

N

0.5 bar

1.0 bar Cl

Cl

Cl

Cl

N

N N

9

10

Cl Ni

Fe N

N

1.0 bar Cl

Cl

Cl

N

N

Cl

N

12

Scheme 25. Activities of the catalyst systems 9/MAO, 10/MAO, and 12/MAO at different ethylene pressures. Polymerization conditions: 250 ml of toluene; (Al:M = 250:1); 25 °C; 1h.

140

In case of the iron catalysts, the increased ethylene pressure led to a higher content of uneven numbered olefins. For 9/MAO, the amount increased from 3.6 % at 0.5 bar to 4.4 %, while for 10/MAO 3.8 % of uneven numbered olefins were detected by GC compared with an amount of 2.5 % at 0.5 bar ethylene pressure. In contrast, the Schulz-Flory coefficients α decrease with increasing pressure resulting in lower molecular weight olefins (see Table 3). At higher ethylene pressure, there is a higher probability for metathesis reactions, since the concentrations of both required educts are increased. 3.3.6 Influence of the polymerization temperature The catalyst 9/MAO was chosen to investigate the influence of the reaction

-1

Activity [g(PE)/g(Fe)*h ]

temperature on the oligomerization activity and the product composition. 1600 1400 1200 1000 800 600 400 200 0

1505

473

433 25

0°

25°C

50°C

75°C

Scheme 26. Activities of the system 9/MAO at different reaction temperatures. Polymerization conditions: 250 ml of toluene; (Al:M = 250:1); 0.5 bar ethylene; 1h. At 0°C, the catalyst 9/MAO shows the highest activity (1505 g/g M · h). The strong decrease in activity at higher temperatures can be explained with a faster deactivation of the catalytically active species. In the range from 0°C to 50°C, the amounts of uneven numbered olefins are quite similar. In contrast, at 75°C no uneven numbered olefins were observed. The temperature dependence of these reactions can be explained with different reaction orders for oligomerization (first order to monomer) and metathesis reaction (first order to monomer and additionally a dependence from the formation rate of the 2-olefins). Therefore, an optimum equilibrium between the formation of 2-olefins and the metathesis reaction is reached at 25°C. 3.3.7 Influence of the aluminum/metal ratio The ratio of co-catalyst to catalyst precursor is a very important parameter for all oligomerization and polymerization reactions. Using the catalyst 10/MAO, the influence of different aluminum/metal ratios on the activity and the product composition was investigated.

141

1600 1363

Activity [g(PE)/g(Fe)*h-1 ]

1400 1200 1000 705

800 600 400

658

584

678

613

462

390

200 0 150:1

250:1

350:1

500:1 750:1 Ratio Al:Fe

1000:1 1500:1 2000:1

Scheme 27. Activities of the catalyst 10/MAO at different Al:Fe ratios. Polymerization conditions: 250 ml of toluene; 0.5 bar ethylene; 25 °C; 1h. At a ratio Al:Fe = 750:1, the activity reaches its maximum value. Applying lower values, there are not enough suitable aluminum centers available in the mixture (possibly “free” TMA molecules), while at higher Al:Fe ratios side reactions become more dominant resulting in deactivation of the catalytically active species. Similarly, the overall content of uneven numbered olefins first increases and reaches a maximum at a ratio Al:Fe = 500:1, while at higher ratios a gradual decrease can be observed (see Scheme 28).

Overall content of uneven numbered olefins [%]

8 7 6 5 4 3 2 1 0 150:1

250:1

350:1

500:1

750:1

1000:1

1500:1

2000:1

Ratio Al : Fe

Scheme 28. Overall contents of uneven numbered oligomers at different Al:Fe ratios applying the catalyst 10/MAO. Polymerization conditions: 250 ml of toluene; 0.5 bar ethylene; 25 °C; 1h.

142

3.3.8 Optimization of the catalysts Taking into account the structure-property relationships derived from the experimental data, the catalysts were optimized with regard to higher activity and an increased formation rate of uneven numbered olefins. To achieve higher activities, the iron(III) chloride complex 19 was prepared containing bromo substituents at the ligand framework. Under optimized conditions (Al:Fe = 750:1; 1 bar ethylene; 0°C), an activity of 2970 [g products/g Fe · h] was obtained. As expected, the catalyst 19/MAO did not produce uneven numbered olefins.

Cl

Cl

Cl

Fe Br

N

Br

Br Fe N

Br

Cl

N

N

N

N

19

13

Cl

Scheme 29. Optimized catalyst precursors. The highest amount of olefins with uneven numbers of carbon atoms was found applying 13/MAO (Al:Fe = 500:1; 1 bar ethylene; 25°C). For this catalyst, the overall content in the reaction mixture increased to 19.2 %.

4. Conclusion A series of bis(arylimino)pyridine transition metal complexes with halogen containing ligand frameworks was prepared, characterized, and applied for catalytic ethylene oligomerization and polymerization reactions. Some of these catalysts surprisingly produced α-olefins with odd carbon numbers. As a possible explanation of this unprecedented behavior, a combined isomerization/metathesis reaction pathway is proposed. Optimized bis(arylimino)pyridine iron catalyst precursors are presented.

Acknowledgements We thank Saudi Basic Industries Corporation (SABIC, Riyadh, Saudi Arabia) for the financial support.

References [1] [2] [3] [4] [5] [6]

Johnson, L. K.; Christopher, M.; Brookhart, M., J. Am. Chem. Soc., 1995, 117, 6414-6415. Britovsek, G. J. P.; Gibson, V.; Kimberley, B. S.; Maddox, P. J.; McTavish, S. J.; Solan, G. A.; White, A. J. P.; Williams, D. J., Chem. Commun., 1998, 849-850. Britovsek, G. J. P.; Gibson, V. C.; Kimberley, B. S.; Maddox, P.J.; McTavish, S. J.; Solan, G. A.; White, A. J. P.; Williams, D. J., J. Am. Chem. Soc., 1999, 121, 8728-8740. Small, B. L.; Brookhart, M.; Bennett, A. M. A., J. Am. Chem. Soc., 1998, 120, 4049-4050. Small, B. L.; Brookhart, M., J. Am. Chem. Soc., 1998, 120, 7143-7144. Small, B. L.; Brookhart, M., Polym. Prepr. (Am. Chem. Soc., Div. Polym. Chem.), 1998, 39, 213.

143

[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56]

Britovsek, G. J. P.; Mastroianni, S.; Solan, G. A.; Daugh, S. P. D.; Redshaw, C.; Gibson, V. C.; White, A. J. P.; Williams, D. J.; Elsegood, M. R. J., Chem. Eur. J., 2000, 6, 2221-2231. Chen, Y.; Qian, C.; Sun, J., Organometallics, 2003, 22, 1231-1236. Chen, Y.; Chen, R.; Qian, C.; Dong, X.; Sun, J., Organometallics, 2003, 22, 4312-4321. Kim, I.; Han, B. H.; Ha, C. S.; Kim, J. K.; Suh, H., Macromolecules, 2003, 36, 6689-6691. Sun, W. H.; Tang, X.; Gao, T.; Wu, B.; Zhang, W.; Ma, H., Organometallics, 2004, 5037-5047. Zhang, Z.; Zou, J.; Cui, N.; Ke, Y.; Hu, Y., J. Mol. Cat. A: Chem., 2004, 219, 249-254. Zhang, Z.; Chen, S.; Zhang, X.; Li, H.; Ke, Y.; Lu, Y.; Hu, Y., J. Mol. Cat. A: Chem., 2005, 230, 1-8. Sun, W. H.; Jie, S.; Zhang, S.; Zhang, W.; Song, Y.; Ma, H.; Chen, Y.; Wedeking, K.; Fröhlich, R., Organometallics, 2006, 25, 666-677. Chen, Y.; Huang, Y.; Li, Z.; Zhang, Z.; Wei, C.; Lan, T.; Zhang, W., J. Mol. Cat. A: Chem., 2006, 259, 133-141. Gibson, V. C.; Long, J. L.; Oxford, P. J.; White, A. J. P.; Williams, D. J., Organometallics, 2006, 25, 1932-1939. Ionkin, A. S.; Marshall, W. J.; Adelman, D. J.; Shoe, A. L.; Spence, R. E.; Xie, T., J. Polym. Sci. A: Polym. Chem., 2006, 44, 2615-2635. Ionkin, A. S.; Marshall, W. J.; Adelman, D. J.; Fones, B. B.; Fish, B. M.; Schiffhauer, M. F., Organometallics, 2006, 25, 2978-2992. Gibson, V. C.; Redshaw, C.; Solan, G. A., Chem. Rev., 2007, 107, 1745-1776. Ionkin, A. S.; Marshall, W. J.; Adelman, D. J.; Fones, B. B.; Fish, B. M.; Schiffhauer, M. F.; Soper, P. D.; Waterland, R. L.; Spence, R. E.; Xie, T., J. Polym. Sci. A: Polym. Chem., 2008, 46, 585-611. McTavish, S.; Britovsek, G. J. P.; Smit, T. M.; Gibson, V. C.; White, A. J. P.; Williams, D. J., J. Mol. Cat. A: Chem., 2007, 261, 293-300. Kaul, F. A. R.; Puchta, G. T.; Frey, G. D.; Herdtweck, E.; Herrmann, W. A., Organometallics, 2007, 26, 988-999. Sun, W. H.; Hao, P.; Zhang, S.; Shi, Q.; Zuo, W.; Tang, X., Organometallics, 2007, 26, 2720-2734. Lions, F.; Martin, K. V., J. Am. Chem. Soc., 1957, 79, 2733-2738. Small, B. L., Organometallics, 2003, 22, 3178-3183. Kleigrewe, N.; Steffen, W.; Blömker, T.; Kehr, G.; Fröhlich, R.; Wibbeling, B.; Erker, G.; Wasilke, J. C.; Wu, G.; Bazan, G. C., J. Am. Chem. Soc., 2005, 127, 13955-13968. Sacconi, L.; Morassi, R.; Midollini, S., J. Chem Soc. A, 1968, 1510-1515. Alyea, E. C.; Merrell, P. H., Inorg. Chim. Acta, 1978, 28, 91-97. Alyea, E. C.; Merrell, P. H., Syn. React. Inorg. Metal-Org. Chem., 1974, 4, 535-544. Merrell, P. H.; Alyea, E. C.; Ecott, L., Inorg. Chim. Acta, 1982, 59, 25-32. Reardon, D.; Conan, F.; Gambarotta, S.; Yap, G.; Wang, Q., J. Am. Chem. Soc., 1999, 121, 93189325. Schmidt, R.; Welch, M. B.; Knudsen, R. D.; Gottfried, S.; Alt, H. G., J. Mol. Cat. A: Chem., 2004, 222, 17-25. Milione, S.; Cavallo, G.; Tedesco, C.; Grassi, A.; J. Chem. Soc., Dalton Trans., 2002, 1839-1846. Esteruelas, M. A.; Lopez, A. M.; Mendez, L.; Olivan, M. ; Onate, E., Organometallics, 2003, 22, 395406. Small, B. L.; Carney, M. J.; Holman, D. M.; O’Rourke, C. E.; Halfen, J. A., Macromolecules, 2004, 37, 4375-4386. Nakayama, Y.; Sogo, K.; Yasuda, H.; Shiono, T., J. Polym. Sci. A: Polym. Chem., 2005, 43, 33683375. Vidyaratne, I.; Scott, J.; Gambarotta, S.; Duchateau, R., Organometallics, 2007, 26, 3201-3211. Black, D., Austr. J. Chem., 1968, 21, 803-805. Restivo, R. J.; Ferguson, G.,. J. Chem. Soc., Dalton Trans., 1976, 518-521. Balamurugan, R.; Palaniandavar, M.; Halcrow, M. A., Polyhedron, 2006, 25, 1077-1088. Xu, X.; Xi, Z.; Chen, W.; Wang, D., J. Coord. Chem., 2007, 60, 2297-2308. Cetinkaya, B.; Cetinkaya, E.; Brookhart, M.; White, P. S., J. Mol. Cat. A: Chem., 1999, 142, 101-112. Dayan, O.; Cetinkaya, B., J. Mol. Cat. A: Chem., 2007, 271, 134-141. Haarman, H. F.; Bregman, F. R.; van Leeuwen, P. W. N. M.; Vrieze K., Organometallics, 1997, 16, 979-985. Dias, E. L.; Brookhart, M.; White, P. S. Organometallics, 2000, 19, 4995-5004. Nueckel, S.; Burger, P., Organometallics, 2001, 20, 4345-4359. Kooistra, T. M.; Hetterscheid, D. G. H.; Schwartz, E.; Knijnenburg, Q.; Budzelaar, P. H. M.; Gal, A. W., Inorg. Chim. Acta, 2004, 357, 2945-2952. Dorer, B. A., WO 0047586, 2000. Schmidt, R.; Hammon, U.; Welch, M. B.; Alt, H. G.; Hauger, B. E.; Knudsen, R. D., US 6458905, 2002. Calderazzo, F.; Englert, U.; Pampaloni, G.; Santi, R.; Sommazzi A.; Zinna, M., J. Chem. Soc., Dalton Trans., 2005, 914-922. Lavery, A.; Nelson, S. M., J. Chem. Soc., Dalton Trans., 1984, 615-620. Lavery, A.; Nelson, S. M., J. Chem. Soc., Dalton Trans., 1985, 1053-1055. Nelson, S. M.; Lavery, A.; Drew, M. G. B., J. Chem. Soc., Dalton Trans., 1986, 911-920. Edwards, D. A.; Mahon, M. F.; Martin, W. R.; Molloy, K. C.; Fanwick, P. E.; Walton, R. A., J. Chem. Soc., Dalton Trans., 1990, 3161-3168. Reardon, D.; Aharonian, G.; Gambarotta, S.; Yap, G. P. A., Organometallics, 2002, 21, 786-788. Gibson, V. C.; McTavish, S.; Redshaw, C.; Solan, G. A.; White, A. J. P.; Williams, D. J., J. Chem. Soc., Dalton Trans., 2003, 221-226.

144

[57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88]

Margl, P.; Deng, L.; Ziegler, T., Organometallics, 1999, 18, 5701-5708. Deng, L.; Margl, P.; Ziegler, T., J. Am. Chem. Soc., 1999, 121, 6479-6487. Wang, H.; Yan, W. D.; Jiang, T.; Liu, B. B.; Xu, W. Q.; Ma, J. J.; Hu, Y., Chin. Chem. Lett., 2003, 14, 257-258. Small, B. L.; Marcucci, A. J., Organometallics, 2001, 20, 5738-5744. Schmidt, R.; Hammon, U.; Gottfried, S.; Welch, M. B.; Alt, H. G., J. Appl. Polym. Sci., 2003, 88, 476482. Görl, C.; Alt, H. G., J. Organomet. Chem., 2007, 692, 4580-4592. Nylander, L. R.; Pavkovic, S. F., Inorg. Chem., 1970, 9, 1959-1960. Shamir, J., Inorg. Chim. Acta, 1989, 156, 163-164. Dilworth, J. R.; Richards, R. L., Inorg. Synth., 1980, 20, 119-127. Evans, D. F., J. Chem. Soc., 1959, 2003-2005. Deutsch, D. L.; Poling, S. M., J. Chem. Ed., 1969, 46, 167-168. Gerlach, D. H.; Holm, R. H., Inorg. Chem., 1969, 11, 2292-2297. Cotton, F. A.; Wilkinson, G., Advanced Inorganic Chemistry, John Wiley and Sons, New York, 1980. Schulz, G. V., Z. phys. Chem. B, 1935, 30, 379-398. Schulz, G. V., Z. phys. Chem. B, 1939, 43, 25-46. Flory, P. J., J. Am. Chem. Soc., 1940, 62, 1561-1565. Kehlen, H.; Rätzsch, M. T., Z. phys. Chem., 1984, 256, 1049-1060. Seitz, M., Diploma thesis, University of Bayreuth, Germany, 2001. Svejda, S. A.; Brookhart, M., Organometallics, 1999, 18, 65-74. Griffiths, E. A. H.; Britovsek, G. J. P.; Gibson, V. C.; Gould, I. R., Chem. Commun., 1999, 13331334. Andzelm, J.; Wimmer, E., Physica B, 1991, 172, 307-317. Andzelm, J.; Wimmer, E., J. Chem. Phys., 1991, 96, 1280-1303. Schlegel, H. B., Ab Initio Method in Quantum Chemistry I, Wiley, New York, 1987. Vosko, S. H.; Wilk, L.; Nusair, M., Can. J. Phys., 1980, 58, 1200-1211. Becke, A. D., Phys. Rev. A, 1988, 38, 3098-3100. Lee, C.; Yang, W.; Parr, R. G., Phys. Rev. B., 1988, 37, 785-789. Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H., Chem. Phys. Lett., 1989, 157, 200-206. Godbout, C. N.; Salahub, D. R.; Andzelm, J.; Wimmer, E., Can. J. Chem., 1992, 70, 560-571. Siedle, A. R.; Lamanna, W. M.; Newmark, R. A.; Schroepfer, J. N., J. Mol. Cat. A: Chem., 1998, 128, 257-271. Yang, X.; Marks, T. J., J. Am. Chem. Soc., 1993, 115, 3392-3393. Bunel, E.; Burger, B. J.; Bercaw, J. E., J. Am. Chem. Soc., 1988, 110, 976-978. Sini, G.; Macgregor, S. A.; Eisenstein, O.; Teuben, J. H., Organometallics, 1994, 13, 1049-1051.

145