41 4| . .I:\‘.' ,
IfIIIIsE:_1~.e;!cI—:;*:. 5, ‘2,"

 

 

———w‘

 

9471“) u-‘\‘ HI . um -;I.
dcifi'ségé'dfiéIIWNxI:.‘vzv‘5“:

PART ONE
INTERCHANGE OF MONOHAPTO- AND
PENTAHAPTOCYCLOPENTADIENYL RINGS IN EARLY
TRANSITION METAL METALLOCENE SYSTEMS

PART TWO
A NEW ROUTE TO PREPARING POLYMER-ATTACHED
METALLOCENE DERIVATIVES

PART THREE
GYGLOPENTADIENYL LIGAND EXCHANGE REACTIONS
IN SELECTED SYSTEMS

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
JOHN GOO-SHUH LE
1977

 

LIBRARY

II. Ecliigan Stan)
University

 

 

This is to certify that the

thesis entitled

(1) INTEROHANOE OF MONOHAPTO- AND PENTAHAPTO
CYCLORENTADIENYL RINGS IN SOME EARLY TRANSITION
METAL METALLOCENE SYSTEMS (2) A NEw ROUTE TO
PREPARING POLYMER-ATTACHED METALLOCENE DERIVATIVES

(3) CYCLOPENTADIENYL BRggNQ1§¥CHANGE REACTIONS IN
SELECTED SYSTEMS

John Guo-shuh Lee

has been accepted towards fulfillment
of the requirements for

Ph. D. . CHEMISTRY
degree m

 

Major professor

Date 5190’?)

0-7 639

 

 

 

ABSTRACT
PART ONE

INTERCHANGE OF MONOHAPTO- AND PENTAHAPTOCYCLOPENTADIENYL
RINGS IN SOME EARLY TRANSITION METAL METALLOCENE SYSTEMS

PART TWO
A NEW ROUTE TO PREPARING POLYMER-ATTACHED METALLOCENE DERIVATIVES
PART THREE
CYCLOPENTADIENYL LIGAND EXCHANGE REACTIONS IN SELECTED SYSTEMS
BY

John Guo—shuh Lee

PART ONE

PMR and mass spectral analysis have been used to study the inter-
change of pentahapto-bonded cyclopentadienyl rings with monohapto-bonded
cyclopentadienyl rings in the compounds (CSHS)4M (M - Ti, Zr, Hf, Nb,
Ta, Mo, and W) and (C5H5)3V or monohapto-bonded benzylcyclopentadienyl
rings in the compounds (C6H5CH205H4)(CSHS)2MC1 (M - Ti, Zr, Hf, Nb, Ta,
Mo, and W). As soon as the CpaM (or CpBMCI) species are generated (in-
dicated by a color change), the exchange occurs and the equilibrium is

established. As reported, no such interchange was observed in (C5H5)4Mo

on the PMR time scale; however, it does occur after a longer time.

John Guo-shuh Lee
PART TWO

New methods of attaching transition metals to polymers have been
studied. The metallocene dichlorides of Ti, Zr, Hf, V, Nb, Ta, Mo, and
W have been attached to polystyrene divinylbenzene beads by facile ex-
change reactions instead of the previous synthetic routes involving

synthesis of monocyclopentadienyl metal halides.
PART THREE

Cyclopentadienyl ligands have been exchanged between molecules of
metallocene dichloride (M - V and Hf) in benzene solely by photochemical
process. The exchange of cyclopentadienyl ligands between molecules of
vanadocene monochloride (vanadocene or chromocene) occurs by both ther-

mal and photochemical processes.

PART ONE
INTERCHANGE OF MONOHAPTO- AND PENTAHAPTOCYCLOPENTADIENYL
RINGS IN SOME EARLY TRANSITION METAL METALLOCENE SYSTEMS
PART TWO
A NEW ROUTE TO PREPARING POLYMER-ATTACHED METALLOCENE DERIVATIVES
PART THREE

CYCLOPENTADIENYL LIGAND EXCHANGE REACTIONS IN SELECTED SYSTEMS
BY

John Guo—shuh Lee

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1977

To My Parents

ii

ACKNOWLEDGEMENTS

I am very pleased to thank Professor Carl H. Brubaker, Jr., for
his guidance and encouragement throughout this investigation.

Dr. Larry W. Shive has also provided many hours of helpful
discussions for which I am very grateful.

I wish to acknowledge the generous financial support for this
research.by the National Science Foundation.

I wish to express my sincere gratitude to my parents and my wife's
parents for their immeasurable encouragement during my years of graduate
study.

I express my deepest appreciation and thanks to my wife, Lu-chuang,

for her love and devotion throughout my graduate education.

iii

TABLE OF CONTENTS

PART ONE

INTERCHANGE OF MONOHAPTO- AND PENTAHAPTOCYCLOPENTADIENYL RINGS IN SOME

EARLY TRANSITION METAL METALLOCENE SYSTEMS

INTRODUCTIONOOOCC.COOOOOOOOOOOOOOIOOOOOOOOOOIOOOIOOOOOOOOOOCUOO0...... 2

RESULTSANDDISCUSSIONOOOOOOO....00....0...0.0...OOOIOOOOOOOOOOOOOOOO. 3

A.

B.

C.

Interchange of Monohapto- and Pentahaptocyclopentadienyl
Rings in Cp4M (M = Ti, Zr, Hf, Nb, Ta, Mo, and W) System..... 3

Interchange of Monohapto- and Pentahaptocyclopentadienyl
Rings in Cp4V, Cp3V, and Cp2V Systems........................ 8

'Interchange of Monohapto- and Pentahaptocyclopentadienyl

Rings in Cp3MC1 (M - Ti, Zr, Hf, Nb, Ta, Mo, and W) Systems.. 8

EDERIWNTAL.OOOOOOOOOOOOO0.0.0.0....0.00.00.00.00.00000000000000000CO13'

A.

B.

GeneraIOOOCOOOOOOOOOO00......0.0.0.000...OOOOOOOOOOOOOOOOOOOO13

Preparation of (C5H5)2(C5D5)2M (M - Ti, Zr, Hf, and V) and
Their Reactions with Gaseous HCl.............................l3

Preparation of (C5H5)2(CSDS)V and Its Thermal Decomposition

PtOdUCtSoooooooooooocooooooooooooooooooooooooooooooo00000000014

CYC10pentadiene Ring Exchange between Vanadocene and NaCSDS..14

Preparation of (CSH5)4-n(CSDS)nM (M = Nb and Ta) and Their

Reaction With Gaseous HCIOOOOOOOOOOOOOIOOCOCOOOOOOOIOOOOOOOOO14

Preparation of (CSHS)2(CSD5)2M (M.= Mo and W) and Their

Reactions with Gaseous HCl...................................15

Preparation of (C6H5CHZCSH4)(C5H5)2MC1 (M.- Ti, Zr, Hf, Nb,

Ta, Mo, and W) and Their Reactions with Gaseous HCl..........15

iv

TABLE OF CONTENTS (Cont.)

PART TWO

A NEW ROUTE TO PREPARING POLYMER-ATTACHED METALLOCENE DERIVATIVES

INTRODUCTIONIOOOOOO0.00.00.00.00.000000000000000000000000000000 17
RESULTSANDDISCUSSIONOOOOOOOOOOI0.0.0.0.000...OOOOOOOOOOOOOOOI 20

A. A New Route to Preparing Polymer Attached Metallocene
DerivativeSoooo00.00.000.000.coco-0000000000000.000000 20

B. Polymer Supported Methylene Bridged Titanocene Dich-

lorideOOOOOOOO0.0.0.0...0..OOOOOOOIOOCOOOOOOOOOOOOI... 24

EHERIENTALOOOOOOOOOOOOOOOO00.0.0...O...OOCOOOOOOOOOOOOOOOOOOO 27
A. General 0 O O O O O O O I O O C O O C O O O O O C O O O O O O O O O O O O C O O O O O C O O O C O C O 29
B. Preparation of Anion Beads............................ 29

C. Preparation of the Polymer Supported Metallocene
DiChlorides 0f Ti, zr’ and HfOOOOOOOOOOOOOOO00......O. 30

D. Preparation of the Polymer Supported Metallocene
Dichlorides of Nb, Ta, Mo, and W...................... 30

E. Preparation of the Polymer Supported Vanadocene
DiChlorideOOOOOOOOO00......OOOOOOOOOOOIOOOOOOOOO0.0... 30

F. Preparation of the Polymer Supported Methylene
Bridged Titanocene Dichloride......................... 31

PART THREE

CYCLOPENTADIENYL LIGAND EXCHANGE REACTIONS IN SELECTED SYSTEMS

INTRODUCTIONOOOO0.0......OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO 33
RESULTS AND DISCUSSION 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 38
A. Photolytic Cyclopentadienyl Ligand Exchange between

Vanadocene Monochloride and Perdeuterovanadocene
MonOChlorideOOOOOOOOOOOO...OOOCOOOOOOOOOOOO0.0.0.0.... 38

TABLE OF CONTENTS (Cont.)

B.

C.

Cyc10pentadienyl Ring Exchanges - Miscellaneous.....

MeChanismSOOOOOOOOOOOOOOOOOOOOOOOO0.0...0.0.0.000...

EDERIMNTALOOOOOOOOOOOOOOOO0.0000000000000000000...000......

A.

B.

G.

GeneraIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO000......

Photolytic Cyclopentadienyl Ligand Exchange between
Vanadocene Monochloride and (D-lO)Vanadocene
MonOChloride0.000000000000000IOOOOOOOOOOOOOOOOCOOOOO

Photolytic Cyclopentadienyl Ligand Exchange between
Hafnocene Dichloride and (D-10)Hafnocene Dichloride.

Thermal and Photolytic Cyclopentadienyl Ligand
Exchange between the 3-10 and D-lO analogues of
szv and Of szVClz...000......OCOOOOOCOOOQOOCOOOOOO

Thermal and Photolytic Cyclopentadienyl Ligand
Exchange in the Three Possible Binary Combinations
Of cpzv’ szVCl’ and szVClz.OOIOOOOOOOOOOOOOOOOOOOO

Thermal and Photolytic Cyc10pentadienyl Ligand
Exchange in Mixed Metallocene Systems of (C5H5)2Cr/
(C5D5)2V, (C5H5)2Cr/(CH3C5H4)2VC1, and
(CHBCSH4)ZVC1/((C5H5)2T1C1)2........................

quantum Yield Determination.OOOOOOOOOOOOOOOOOOOOOOOO

BIBLIOGRAPHYOOOOO...0.0000....OOOOOOOOOOOOOOOOCOOO0.00.0.0...

vi

41

42

43

43

43

44

44

44

45

45

50

LIST OF TABLES

TABLE PAGE

1. PMR Results of Ti(A), Ti(B), Zr(A), Zr(B), Hf(A), and Hf(B).. 3

2. Mass Spectrographic Data for Ti(A), Ti(B), Zr(A), Zr(B),
Hf(A), and Hf<B>OOCCOOOOOCOOOC00......OOOOIOOCOOOOCOCOOOCOCCO 5.

3. Mass Spectrographic Data for Nb(A), Nb(B), Ta(A), and Ta(B).. 6
4. Mass Spectrographic Data for M(A) and M(B) (M 8 Mo and W)... 7
5. Mass Spectrographic Data for V(A) and V(B)................... 10
6. Mass Data of Ti(C), Zr(C), and Hf(C)......................... 10

7. Mass Spectrographic Data for Ti(A), Ti(C), Zr(A), Zr(C),
Hf(A), and Hf(C)ooooocoo-ococoo00000000000cocoooooooooooooooo 1]-

8. Mass Spectrographic Data for Nb(A), Nb(C), Ta(A), and Ta(C).. 11
9. Mass Spectrographic Data for Mo(A), Mo(C), W(A), and W(C).... 12

10. Analytical Results for the Polymer Supported Metallocene

DiChlorj-des...OOOOOOOOOOOIOOOOOOOOOODOOOOOOOOOOOOOOCOOOOOOOOO 23

11. Mass Spectrographic Data for the Perdeuterovanadocene

Monochloride/Vanadocene Monochloride Ligand Exchange......... 39

12. CyclOpentadienyl Ligand Exchange between Vanadocene Come
pomds Of Different OXidation States I O O O O O C O O O O O O O O O O O I O O O O O O 39

vii

LIST OF FIGURES

FIGURE PAGE
1. Scheme for Sigma-pi Ring Interchanges in Cp3MCl Systems... 9
2. Scheme for the Preparation of Polymer Supported Me-

tallocene Dichloride by a New Route....................... 21
3. Scheme for the Preparation of Polymer Supported
Titanocene DiChlorideCIOOC0.0000IOOCCOOCOCOOOOOOOOOOOOOOOO 22
4. ESR Spectrum of Supported Vanadocene Dichloride........... 25
5. ESR Spectrum of Supported Niobocene Dichloride............ 26
6. Scheme for the Preparation of Polymer Supported
Methylene Bridged Titanocene Dichloride (a)............... 27
7. Scheme for the Preparation of Polymer Supported
Methylene Bridged Titanocene Dichloride (b)............... 28
8. Proposed Mechanism for the Photolytic Ligand Exchange
Reaction Of Titanocene DiChlorideOCOOOOOOOOOCOOOOIOCOOOOI. 34
9. Reaction Scheme for the CpTiCl2 Intermediate.............. 35
10. PrOposed Mechanism for the Photolytic Exchange Reactions
of Titanocene Dichloride.................................. 36

ll. McKay Plot for Perdeuterovanadocene Monochloride/

Vanadocene Monochloride................................... 40

12. Mass Spectrum of Vanadocene Monochloride.................. 46

13. Mass Spectrum of (D-10)Vanadocene Monochloride............ 47

14. Mass Spectrum of Exchange Mixtures........................ 48

viii

PART ONE
INTERCHANGE OF MONOHAPTO- AND PENTAHAPTOCYCLOPENTADIENYL RINGS IN SOME

EARLY TRANSITION METAL METALLOCENE SYSTEMS

INTRODUCTION

Ligand sigma-pi rearrangements have been observed for almost all
of the transition metals with a variety of ligands. The rearrangements
are not only of interest in the field of organometallic chemistry but
are also involved in such industrially important processes as the
Ziegler-Natta polymerization of olefins, the oxo process (hydroformy-
lation of olefins), and in homogeneous catalysis in general.

Interchange ofhl-CSH5 andITS-CSHS rings was first observed in
temperature-dependent PMR studies of 011-C535)2015-CSHS)2T1 1 . This
work and subsequent investigations of three other CpéM (M - Zr, Hf,
and Mo) 2—4 systems demonstrated that only the molybenum complex does
not undergo rapid ring interchange at ambient temperatures. Through

the use of labelled Cp compounds (C in Cp4M and C H CH C H in

51’s 65254
ijMCl systems), the interchange reactions in some do to d2 metallo-
cenes were investigated. The results show that the interchange of

monohapto- and pentahaptocyclopentadienyl rings occurs in all the

systems studied.

RESULTS AND DISCUSSION

A. Interchange of Monohapto- and Pentahaptocyclopentadienyl Rings in
9P4M (M - Ti, Zr,4Hf, Nb,iTa, Mo, and W) Systems

 

The compounds (CSH5)2-n(CSDS)2+nM’ where n - 0, 1, 2, and M = Ti,
Zr, Hf, Nb, Ta, Mo, and W; were prepared from the corresponding

(C5H5)2MC12 [abbreviated M(A)] and two moles of NaC (or TlC D for

SDS 5 5

Mo system). Excess NaCSDS was used for the Nb and Ta compounds. When
the resultant products were treated with gaseous HCl, 0-bonded Cp- was
replaced by C1-, and a mixture of (CSHS)2-n(CSD5)nMC12 [abbreviated
M(B)] was isolated. The PMR data for the cyclopentadienyl protons of
M(A) and M(B) (M - Ti, Zr, and Hf) are listed in Table 1.

Table 1. PMR Results of Ti(A), Ti(B), Zr(A), Zr(B), Hf(A), and Hf(B)
[in THF with Equal Concentrations of M(A) and M(B)]

 

 

 

*
Metallocene Dichloride C H (ppm) Peak Ratio M(A)/M(B)
5 5
Expected Found
Titanocene Dichloride 6.6 2 2.12
Zirconocene Dichloride 6.62 2 1.89
Hafnocene Dichloride 6.52 2 1.95

 

*M(A) = (C5H5)2MIC12

M(B) - (CSH ) MC12, (c5H5)(c5D5)Mc12, and (C5D5)2MC1

5 2 2°

4
The ratio of the peak areas M(A)/M(B) [M = Ti, Zr, and Hf] for
two different equimolar solutions based on absolute 1H nmr inter-
gration is £1. 2 and suggests that essentially half of the CSHS ring
were replaced by CSDS rings. The mass Spectrographic data of M(A)
and M(B) (M - Ti, Zr, Hf, Nb, Ta, Mo, and W; Table 2, 3, and 4) are
consistent with there being a mixture of C D MCl C D H MCl

10 10 2’ 10 5 5 2’

1
10H10M012 as expected if initially n -bonded C

interchanged with nS-bonded C

and C rings were

st
5H5 rings. My results confirm the fast
exchange expected in d0 systems. With the intermediately bonded

ring concept proposed by Cotton 3’4 fast ring interchange in d1
systems can also be explained. As reported 4 , no such inter-

change was observed in (C5H5)4Mo on the PMR time scale, since it has
an eighteen electron configuration in its ground state and would have
to go through a high-energy, intermediately bonded state in order to
exchange the ring. However, this interchange is observed in my
experiments which encompass a longer period of time. The PMR spectrum
of (C5H5)2(C5D5)2Mo (prepared from either (CSH5)2MoI2 and two moles

of T1C5D5 or from (C5D5)2MoI2 and two moles of T1C5H5), was measured.
The spectrum (toluene-d8) consisted of two very sharp singlets at

6 3.83 and 5 3.94 with equal intensities. With no interchange, only
one of these two singlets should be observed in the PMR spectra 4 .
Attempts to determine the rate of interchange were unsuccessful since
by the time the Cp4Mo species were generated (indicated by a violet

color), the interchange had occurred and the equilibrium had been

established.

Table 2. Mass Spectrographic Data for Ti(A), Ti(B), Zr(A), Zr(B),
Hf(A), and Hf(B).
(based on Ti,48; Zr,90; Hf,180; 01,35)

 

 

 

Ti(A) Ti(B) Zr(A) Zr(B) Hf(A) Hf(B) Ion
m/e Rel. int.(Z) m/e Rel. int.(Z) m/e Rel. int.(%) M = Ti, Zr, Hf
258 O 14 300 O 28 390 O 24 clonlonclz+
253 o 30 295 O 50 385 O 49 ClODSHSMC12+
248 86 17 290 51 25 380 35 25 CIOHIOMC12+
223 O 9 265 O 9 355 O 12 C10D10M01+
218 O 20 260 O 15 350 O 23 C10D5H5M0I+
213 75 12 255 20 8 345 10 16 clofilouci+
188 O 90 230 O 100 320 O 95 csnsuc12+
183 100 100 225 100 95 315 100 100 CSHSMO12'+
160 O o 202 O 28 292 O 23 C3D3MC12+
157 O o 199 15 26 289 27 35 C3H3MC12+
153 O 45 195 O 20 285 O 5 CSDSMCI+
148 80 50 190 10 20 280 5 3 csnsuci+

 

NOte: M(A) - (C5H5)2MC12

M(B) - (C5H5)(C5D5)MC12, (C5H5)2MC12, and (C5D MCI

5)2 2.

Table 3 Q

(based on Nb,93; Ta,181; C1,35)

Mass Spectrographic Data for Nb(A), Nb(B), Ta(A), and Ta(B).

 

 

 

Nb(A) Nb(B) Ta(A) Ta(B) Ion
m/e Rel. int.(Z) m/e Rel. int.(Z) M - Nb, Ta
303 O 60 391 O 63 ClODloMC12+
298 O 42 386 0 20 clonsusnmz+
293 87 10 381 90 O C10H10MC12+
284 O 11 372 O 38 clonlouom+
279 O 8 367 o 18 ClODSH5M0C1+
274 13 3 362 10 o CloH10M0C1+
268 O 20 356 0 3O 61011101161+
263 O 15 351 o 25 clonsusum+
258 30 5 346 16 5 cloamnci+
233 O 100 321 O 100 C5D5”C12+
228 100 35 316 100 17 CSHSMC12+
198 o 18 286 CSDSMC1+
193 10 5 281 csnsuci+
NOte:v Excess NaC D was used in this study.

5 5

Table 4.

Mass Spectrographic Data for M(A) and M(B) (M = M0 and W).
(based on Mo,95; W,184; 01,35)

 

 

Mo(A) Mo(B)*

W(A) W(B)

Ion

 

 

m/e Rel. int.(Z) m/e Rel Int.(Z) M . Mo, W
305 1 9 394 0 45 01001011012+
300 2 15 389 0 80 clonsusuzclz+
295 25 7 384 85 42 010111011012+
270 O 55 359 0 27 clonlouci+
265 1 100 354 0 55 010051151101+
260 100 44 349 53 25 0108101101+
235 2 27 324 2 100 c505M012+ , 01001011+
230 32 52 319 ' 100 97 0585Mc12+ , C10D5H5M+
225 33 20 314 8 5 010111011+
200 3 65 65051101+
195 95 70 05851101+
and TIC D

*From the reaction of (C5H5)2MOI

2

5 5

B. Interchange of Monohapto- and Pentahaptocyclopentadienyl Ringg:in
94V, Cp3V, and Cp2_Y_

Since it appears that Cpav cannot be prepared 5 , the interchange
reactions of CpZV and Cp3V were chosen for study. The compound,
(C5H5)2(CSD5)V, was prepared from (C5H5)2VC1 and one mole of NaCSDS'
When the mixture of compounds was heated, o-bonded Cp was eliminated
and exchange products (C5H5)2V, (CSH5)(CSD5)V, and (C5D5)2V were
isolated by sublimation. Mass spectrographic data confirmed the ex-
change (Table 5.). When vanadocene was treated with NaCSD5 in excess,
ring substitution to give mostly (C5D5)2V was observed by the mass
spectra of the final products (Table 5.).

C. Interchange of Monohapto- and Pentahaptocyclopentadienyl Rings in
gpaMCl (M a T1, Zr, Hf, Nb, Ta,_Mo, and W) systems

(CSHS)2(C5H4CH2C6H5)MC1 (M - Ti, Zr, Hf, Nb, Ta, Mo, and W) were
prepared from the corresponding metallocene dichloride and one mole of
sodium (or lithium) benzylcyclopentadienide. The interchange follows
the reaction scheme in Figure 1.

Exchange products of (C6H5CH2C5H4)n(C5H5)2-nM012 (n a 0, 1, and 2)
were isolated and the PMR spectral data for M - Ti, Zr, and Hf are
listed in Table 6. The presence of C635- and -CH2- protons are good
indications of the ring exchange reactions.

The mass spectrographic data for M - Ti, Zr, Hf, Nb, Ta, Mo, and
W reveal the presence of either the parent peaks of
(C5H5)(C5H4CH2C6H5)MC12, or their fragments. (Table 7-9). These data
suggest the interchange of the rings. By the same scheme vanadocene
dichloride was tested and no ring interchange was detected from the PMR

and mass spectra, nor was any starting material, vanadocene dichloride,

recovered.

 

 

 

Figure 1. Scheme for Sigma-pi Ring Interchanges in Cp3M01 Systems

10

Table 5. Mass Spectrographic Data for V(A) and V(B).
(based on v.51; C1,35)

 

 

 

V(A) V(Bl) V(BZ) V(B3)
m/e Rel. int(Z) Rel. int.(Z) Ion
191 0 53 100 100 c 0 v+
10 10
+
186 0 95 15 4 clonsasv
+
181 84 100 5 2 C10310V
121 0 65 85 8O csnsv‘+
116 100 97 12 4 CSH5V+

 

Note: V(Bl): sublimation product from (C5H5)2VCl and NaCSDS.

 

 

 

V(BZ): product isolated from the reaction between (C5H5)2V
and excess NaCSD5 with 15 minutes stirring.
V(B3): stirring for 4 days, all others same as B.
Table 6. PMR Data for Ti(C), Zr(C), and Hf(C),
M(C) - (CSHS)2_n(C6HSCH2C5H4)nMC12
Rel. Int.
M(C) Cp- C6H5- -CH2-
Ti(C) 6.48(18) 7.2(9) 4.1(1)
Zr(C) 6.2(3.2) 7.05(1) 3.95(0)

Hf(C) 6.35(12) 7.15(7) 4.05(1)

 

11

Table 7. Mass Spectrographic Data for Ti(A), Ti(C), Zr(A), Zr(C),
Hf(A), and Hf(C).

[(CSH5)(C6HSCH2C5H4)MC12 - P, (M I Ti, Zr, and Hf)]

(based on Ti,48; Zr,90; Hf,180; C1,35)

 

 

Ti(A) Ti(C) Zr(A) Zr(C) Hf(A) Hf(C)
m/e Rel. int.(Z) m/e Rel. int(Z) m/e Rel. int(Z) Ion

 

338 0 0 380 0 7 470 0 40 P+
302 0 5 344 0 6 434 0 0 (P-Hc1)+
273 0 9 315 0 14 405 0 100 (ix-05115)+
+

248 86 29 290 51 44 380 35 25 010111011012
237 0 4 279 0 0 369 0 0 (Ia-110145115)+

+
213 75 30 255 20 50 345 10 53 ClOH10MC1
183 100 100 225 100 100 315 100 50 0585M012+
148 80 65 190 0 0 280 0 0 051151101+

 

Table 8. Mass Spectrographic Data for Nb(A), Nb(C), Ta(A), and Ta(C).

[(C5H5)(C6H5CH2CSH4)MC12 ' P, (M - Nb and Ta)]

(based on Nb,93; Ta,181; C1,35)

 

 

 

Nb(A) Nb(C) Ta(A) Ta(C)
m/e Rel. int.(Z) m/e Rel. int.(Z) Ion
+
383 O 11 471 0 27 p
347 0 11 435 0 12 (P-HC1)+
318 0 O 406 0 7 (p-csus)+
+.
293 87 80 381 90 65 C10H1OMClz
+.
274 13 55 362 10 13 cloulouocl
4+
258 30 30 346 16 12 clonlouc1
228 100 100 316 100 100 c H MCl +

5 5 2

 

12

Table 9. Mass Spectrographic Data for Mo(A), Mo(C), W(A), and W(C).

[(CSH5)(C6H5CH2C5H4)MC12 - P, (M a Mo and W)]

(based on Mo,95; W,184; C1,35)

 

 

 

Mo(A) Mo(C) W(A) w(0)
m/e Rel. int.(%) m/e Rel. int.(%) Ion
‘ +
385 0 0 474 0 3 p
349 0 7 428 0 7 (P-Hcl)+
320 0 0 409 0 0 (13-05115)+
315 0 3 404 0 3 (P—C12)+
+
295 25 15 384 85 80 0108168012
+
260 100 100 349 53 88 ClouldMC1
230 20 21 319 100 100 058511012+
+
225 35 35 314 8 15 C10310M
4.
204 9 11 293 50 55 0311311012
195 95 100 284 15 20 051151101+

 

Note: M(C) is a mixture of (CSHS)2-n(C6H5CH2C5H4)nMC12’ n = O, 1, 2.

EXPERIMENTAL

A. General

Oxygen and moisture were excluded from the reaction mixtures by
thoroughly drying the glassware and reagents, and by manipulating the
reagents under dry argon or in a vacuum in Schlenk-type apparatus.

D20 was purchased from J. T. Baker Chemical Co. T12504 and the

metal chlorides (TiCl ZrCl HfCl VC13, VC1 NbCl TaCl MoCl

4’ 4’ 4’
and W016) were purchased from Alfa Products.

4’ 5’ 5’ 5’

Vanadocene, vanadocene monchloride 6 , and the metallocene dichlo-

rides of Ti, Zr, Hf, V, Nb, Ta, Mo, and W were prepared as previously

7-12 13 14
reported . CSD6 , C6H5CH2CSH5 and their lithium salts,

(CSH5)2MoI2 4 and TICSD5 4 were also prepared by known methods.

PMR spectra were obtained by using a Varian T-60 NMR spectrometer

and by using TMS as a reference. The mass spectra were obtained by

use of a Perkin Elmer model RMU-6 mass spectrometer.

5)2_(_c_:505)211 (M - Ti, ZrLHf, and V) and Their

Reactions with Gaseous HCl

 

 

B. Preparation of (CSH

 

(CSHS)2MC12 (0.01 mole) was mixed with the benzene suspension of

NaCSD5 (0.02 moles). The reaction mixture was stirred for 15 minutes,

and the filtrate obtained after separation of the sodium chloride was

evaporated to dryness. This residue of (C5H5)2(CSD5)2M was then re-

dissolved in toluene, the flask partially evacuated and then refilled
with gaseous 301. After stirring for 30 minutes, the solvent was re-
moved under reduced pressure, and the products, M(B), were purified by

2

vacuum sublimation (1450/10- torr).

13

14

C. Preparation of (C5§5)2(§525)V and Its Thermal Decomposition Products

VCl (3.4 x 10-3 moles) was mixed with 60 m1 of THF at 0°,
3

(C535)2

and 11.5 ml of NaCSDS (1 ml a 0.39 x 10‘

minutes. The solvent was removed in a vacuum and then 100 m1 of cold

moles) in THF and stirred for 15

ether (0°) was added. The mixture was stirred for 15 minutes, filtered
and concentrated to 40 ml. On slow cooling to -78°, (CSH5)2(C5D5)V se-
parated as black crystals. When the compound was heated to 75°, the mix-

ture of (C5H5)2V, (C5H5)(CSD5)V, and (C5D5)2V were sublimed.

D. Cyclopentadiene Ring Exchange between Vanadocene and NaC5_l_)_5

(CSH5)2V in THF was treated with NaCSD5 in large excess for 15

minutes (another trial was allowed to react for 4 days). The reaction
mixture was evaporated to dryness and blue crystals of (C5H5)2V, (C5D5)2V'

and (C5H5)(C5D5)V were isolated upon sublimation (700/10.2 torr).

E. Preparation of (C

(0525)fiM (M - Nb and Ta) and Their Reactions
with Gaseous HCl

5E5)4-n

 

(C5H5)2MC12 was treated with excess NaCSD5 in THF. The mixtures,
which became red in four hours, were stirred an additional 12 hours, the
solvent removed at reduced pressure and the residue evaporated to dryness.
Dry ether was added and the mixture stirred for an additional hour. The
red filtrate of (C5H5)4-n(C5D5)fiM was separated from the salt residue,
the flask was partially evacuated and refilled with gaseous HCl, after
which the mixture immediately became dark colored. Two hours later, the
supernatant, clear, etheral solution was decanted, and the residue was

evaporated to dryness. The M(B) component was sublimed at 2800/10.2 torr.

15

F. Preparation of (C H ) (C D ) M (M = M0 and W) and Their Reactions
5 5 2 5 5 2
with Gaseous HCl

 

 

(C5H5)2M0I2 [or (C5D5)2 5D5

[or TlCSHS] (0.02 moles) in THF. After 12 hours, a yellow precipitate

M012] (0.01 mole) was mixed with T1C

of thallium(I) iodide had formed and the color of the solution changed
from green to red. The resultant violet solution that formed after an
additional 60 hours was reduced to dryness in a vacuum and the dry re-
sidue extracted with 200 m1 of toluene. The red filtrate of
(05H5)2(C5D5)2Mo was separated and the reaction with HCl carried out
as above. The green residues of Mo(B) were filtered and dried in

vacuum.

G. Preparation of (C6HSCH2CSH4)(CSH5)2MCI (M = Ti, Zr, Hf, Nb, Ta, Mo,

and W) and Their Reactions with Gaseous HCl

(C5H5)2‘MC12 (0.5 g) was suspended in toluene, and then a stoichio-
metric amount of sodium (or lithium) benzylcyclopentadienide was added.
The reaction times varied: 30 minutes for T1, Zr, and Hf; 6 hours for
Nb and Ta; 4 days for Mo and W. The filtrate was separated from the
residues, and it was treated with HCl. The mixtures of M(C) (M - Mo
and W) were precipitated and were separated and dried in vacuum, while
the others (M - Ti, Zr, Hf, Nb, and Ta) were simply dried in vacuum

and purified by vacuum sublimation.

PART TWO

A NEW ROUTE TO PREPARING POLYMER-ATTACHED METALLOCENE DERIVATIVES

l6

INTRODUCTION

Conventional heterogeneous catalysts, although widely used indus-
trially, generally suffer from the disadvantages that (1) design and
improvement are difficult because the active sites are not welldefined,
(2) only a small percentage of the active components are accessible and
effective, and (3) the control that can be exerted over the composition
and structure of the active sites is relatively limited. In contrast,
homogeneous transition metal catalysts are not only structurally better
defined, but the steric and electronic environment of the catalytic
active site can often be varied widely and in a systematic manner so
that the course and rate of the reaction may be precisely controlled.

However, homogeneous catalysts have so far found only limited use,
chiefly because of the difficulty of their separation from the reaction
products. Making homogeneous transition metal catalysts insoluble by

attachment to immobile supports such as silica 15-18 , clay 19 , or

polymeric material 20-41 is a significant step in improving their

industrial applicability. While the solid-support technique obviously
prevents the loss of possibly expensive materials and contamination of
reaction products, the method also offers the opportunity to prepare a
new class of catalyst systems with other desirable properties. Several
different types of transition metal compounds have been bound to a va-

rious polymeric material, mostly polystyrene-divinyl benzene copolymers.

'Manassen 42 has suggested that, when a catalyst is attached to a poly-

mer chain, there may be adventitious changes due to the introduction of

17

l8

preferred orientations, a change stereochemistry about the metal atom,
steric crowding, or a change in the equilibrium between the catalyst
metal and its ligands. Patchornik and Krans 43 have pointed out that
when a catalyst is bound to a cross-linked insoluble polymer, its mo—
tion is restricted, and if the molecules of catalyst are bound at an
appropriate distance, a situation approaching infinite dilution may be
reached. Reaction on such a catalyst may be more rapid than one in
which the catalyst molecules are free. For instance, the polymer-sup-
ported catalysts have been demonstrated to have selectivity towards
molecules of different bulk and polarity. Also, the attachment of a
saturated complex that is a potential catalyst to a rigid support,
followed by reductive elimination of a ligand, should produce higher
concentration of monomeric coordinatively unsaturated species than is
obtained in solution. This would then be reflected in the increase in
catalytic activity of that complex as compared to a similar non-attached
complex under the same conditions as have been reported by Brubaker
and co~workers 39—41 .

In this research, a new route to preparing polymer-attached
metallocene derivatives has been developed (Figure 2). By using this
route, metallocene dichlorides of Ti, Zr, Hf, V, Nb, Ta, Mo, and W have
been attached to polystyrenedivinylbenzene beads.

If the catalyst is heterogenized by linking it to a polymer, the
catalytic as well as the support part must be able to stand whatever
temperature is necessary for the reaction which is to be conducted, so
it is important to use ligands which coordinate strongly and give
thermally stable complexes. Putting methylene bridged titanocene di-

chloride on the support takes advantage of the fact that chelate

19

complexes are more stable, inert, and less labile. This species would

be expected to react like titanocene dichloride, but to be more stable.

RESULTS AND DISCUSSION

A. A New Route to Preparing Polymer Attached Metallocene Derivatives
The metallocene dichlorides of Ti, Zr, Hf, Nb, Ta, Mo, and W

have been attached to the polymer according to a new route (Figure 2).

It has been determined that there is sigma-pi interchange of ligands

in (CSH5)3MC1 (PART ONE) by use of C H CH C H - for one of the ligands.

6 5 2 5 4
Following that observation, polymer-attached (C5H5)3MCl was synthesized
from (C5H5)2MC12 and polymer-attached CSH -, its Cp rings allowed to

interchange, and then converted it to polymer attached metallocene
dichloride by treatment with HCl. This procedure leads to the polymer
supported metallocene dichloride at two thirds of the original con-
centration of the (C5H5)3MC1, but provides a much more facile route to
its preparation than the use of (C5H5)MC13 and the polymer-attached
CSHS- CFigure 3). Further, it is very difficult to prepare monocyclo-
pentadienyl halide derivatives of many metals - zirconium, hafnium,
niobium, tantalum, molybdenum, and tungsten monocyc10pentadienyl halides
are all very difficult to prepare and purify - and so this reaction
provides a means of making the polymer-attached metallocene dichlorides
for those metals.

Because no direct ring exchange was detected (page 8) between
benzylcyclOpentadienide anion and vanadocene dichloride, the attachment
procedure is different for vanadocene dichloride. Since we do know
that the cyclopentadienide anion will displace a bound ring in

20

 

 

Figure 2. Scheme for the Preparation of Polymer Supported Metallocene
Dichloride by a New Route

22

{CH-CH2 )- {CH-CH2}-

+ 01011200115 >

 

-(CH-CH2-)—
CpTiCl
3 \ H20

H C \ .
2 . 0 T1012
L1.

 

\

A)

Figure 3. Scheme for the Preparation of Polymer Supported Titanocene
Dichloride (ref. 14)

23

vanadocene, this behavior can be used to attach the complex to the
beads. After this product was treated with HCl vanadocene monochloride
was generated it was then further oxidized by air/HCl mixture to vana-
docene dichloride 6 .

The analytical results of the metals and chloride concentrations

of the supported metallocene dichlorides are listed in Table 10.

Table 10.

Analytical Results for The Polymer Supported Metallocene Dichloride

 

 

 

 

Concentration (mmole/ggbeads) 33512
Attached Species Metal Chloride Expected Found
Ti(D) 0.1 0.21 2 2.1
Zr(D) 0.19 0.45 2 2.4
Hf(D) 0.11 0.26 2 2.3
V(D) 0.048 0.109 2 2.3
Nb(D) 0.11 0.22 2 2.0
Ta(D) 0.4 0.92 2 2.3
Mo(D) 0.1 0.24 2 2.4
W(D) 0.12 0.23 2 1.9
BTi(a) 0.15 0.33 2 2.2
BTi(b) 0.12 0.28 2 2.3

 

Note: M(D) = (-C5H4)M(C5H5)C12

In general, they are in reasonable agreement with the expected
results. Since the beads are not totally inert to the gaseous HCl, the
metal to chloride ratio may increase, if the reaction time is too long.

However; usually the replacement reactions are fast, and keeping the

24

reaction times less than 2 hours leads to the desired products. ESR
spectral data of attached vanadocene and niobocene dichloride (Figures

4 and 5) were identical with those reported for nonattached species 7 .

B. Polymer Supported Methylene Bridged Titanocene Dichloride

 

In the supported methylene bridged titanocene dichloride, charac-
teristics of the supported catalyst which prevent the loss of metal,
avoid contamination of product, etc., were imporved (compared with
regular supported titanocene dichloride), for the following reasons:
(1) Higher thermal and photolytic stability.

The methylene bridged cyclopentadienyl ligands prefer pentahapto
bonding to the metal 44 , and usually pentahapto bonded rings are
difficult to displace.

(2) Retention of higher catalytic activity.

Even if cleavage of one of the Cp rings occurs in Cp2T1012 and
CpTiClzR (in which R has substituded a Cp ring) is generated, the metal
will still be attached on the polymer and act as an active catalyst in
the next cycle.

Supported methylene bridged titanocene dichlorides were prepared
according to Figures 6 and 7.

There are at least two advantages to these synthetic routes.
First is the faster reaction rate. As soon as the first chlorine atom
in the metal chloride derivatives is replaced by an incoming Cp anion
the second chlorine atom is rapidly replaced by the other linked anion
near by. Second is the retention of the loading in the reaction between
Cp2T1C12 and Cp anion beads. After treatment with HCl, the concentra-

tion of the originally attached CpaTi is less likely to decrease, since

25

oowuoanoun osmooomcm> emuuoneom mo abuuuonm “mm .c shaman

comm ouom mm=<o 88

‘P

>I P
1 a

 

db

N
1 4

«II‘

CD

 

 

 

26

00mm

owauoasowo msooonoaz vouuommsm mo asuuomnm «mm

.m onomwm

ooum cmpm mma<o comm

‘ 1

 

 

 

 

 

 

1

 

27

the polymer attached methylene bridged Cp ligands prefer to be pi—
bonded. That makes (B) more stable than (A) in the sigma-pi equilbrium,
and therefore only the free Cp ligands in (B) will be substituted by

C1. (Figure 6 and 7).

As” O —»~°¢° “O

 

cEjn Q «fin—cu, ucu,

.©r@

 

 

 

,., O m. 1.6253

Figure 6. Scheme for the Preparation of Polymer Supported Methylene
Bridged Titanocene Dichloride (a)

28

 

Li" CH0: Q
m. ’ ‘97 rig O
OH,Li
R
It" “E “W O a,

II)

 

,L'LSZR 1.... @435...
@638

Figure 7. Scheme for the Preparation of Polymer Supported Methylene
Bridged Titanocene Dichloride (b)

EXPERIMENTAL

A. General

General experimental techniques, sources of common chemicals and
analytical instruments used in this work have been described in the
EXPERIMENTAL section of PART ONE.

Cp beads (20% crosslinked polystyrene-divinylbenzene copolymer)
were supplied by C. P. Lau, B. H. Chang and J. Ray of this Department.

The metal and chloride analysis of the polymer-attached complexes
was obtained as previously described 39-41 . Titanium, hafnium, vana-
dium, niobium, tantalum, molybdenum and tungsten were determined by
decomposition of the polymer (0.5 g) with three drOps of concentrated
H2304 and heating over a very low flame, and then by ignition of the
metal containing polymer at 9000 (6000 for Mo and W) for 12 hours. The
residue was weighed as the oxide. Metal complexed chloride was removed
by digestion of the polymer samples in 2 N KOH solution at 100°C for
12 hours. Chloride was determined by the Volhard method following

acidification of the aqueous supernatant. The ESR spectra were re-

cdrded by use of a Varian E-4 ESR spectrometer.

B. Preparation of Anion Beads

The polymer attached cyclopentadienide beads (1 g) in THF were
treated with excess CH3Li in ether for 3 days. Excess CHaLi was re-
moved by thorough washing with THF (40 m1 x 5 times). For the attach-

ment of Ti and Hf, the lithium salt of the anion beads was kept in THF,

29

30

but for all the others, the dry beads were stored in an argon filled

dry box.

C. Preparation of the Polymer Supported Metallocene Dichlorides of
T1, Zr and Hf

 

 

A toluene solution of the metallocene dichloride was added to the
anion beads (THF suspension), stirred for 2 days, and the beads were
then treated with HCl. The color of the beads immediately changed
from dark brown to red for the Ti species and yellow to cream for Hf,Zr
species. After stirring for 5 hours, the solvent was removed and the
beads were washed with CHCl3 and THF until the washings were freed of
metallocene dichloride and lithium chloride. The beads were then
dried in vacuum.

D. Preparation of the Polymer Supported Metallocene Dichlorides of
Nb, Ta, Mo, and W

 

Excess metallocene dichlorides of niobium, tantalum, molybdenum
and tungsten were added to anion beads in dry box, and then 100 m1 of
THF was added to the mixture. After the flask was removed from the
dry box, the mixture was stirred for 6 days, and then the flask was
partially evacuated and refilled with H01 as above. The color of the
beads changed from.red to brown-black for niobium and tantalum and to
green for molybdenum.and tungsten species. Through the same washing
procedures, the beads were dried in vacuum. (Ta beads are air sen-

sitive).

E. Preparation of the Polymer Supported Vanadocene Dichloride
Excess vanadocene was mixed with the anion beads in 100 m1 of THF,

for 6 days. The atmosphere over the beads was then replaced with HCl

31

gas as above. The color of the beads changed from dark blue to light
blue in 10 minutes, as expected for the attached vanadocene monochlo—
ride. Five hours later, the whole mixture was exposed to the air and
the attached species was oxidized to green vanadocene dichloride beads.
The solution was decanted, the beads were washed with THF as above,

and dried in vacuum.

F. Preparation of the Polymer Supported Methylene Bridged Titanocene

 

 

Dichloride
Cyclopentadienide anion beads were treated with either (a) 0112012
or (b) CHClB, and yellow beads of -CpCHC12, or -CpCHZCl were formed

immediately. The resulting beads were washed with THF and then treated
with excess Nan. After being stirred for 24 hours, NaCl and the

excess Nan were removed by thorough washing with THF. CHBLi was

added, and the mixture stirred for two days. Excess CH3Li was removed
with THF washes. Then the anion beads [product from bath (a) and (b)]

were treated with szTiC1 in THF and then treated with gaseous HCl.

2
Bridged titanocene dichlorides were formed, washed with THF and then

dried.

PART THREE

CYCLOPENTADIENYL LIGAND EXCHANGE REACTIONS IN SELECTED SYSTEMS

32

INTRODUCTION

The photochemistry of organometallic compounds of early transition
metals has received considerable attention. Hunt and Winter 45 and
McFarlane and Tindall 46 reported the observation of photoreduction of
titanium(IV) alkoxides. Harrigan, Hammond, and Gray 47 found that the
irradiation of cyclopentadienyl derivatives of titanium(IV) in solvents
containing halogen atoms leads to photolysis and abstraction of halo-
gens. The photochemical reactions of dimethyl derivatives of titano-
cene, Zirconocene, and hafnocene were studied by Ald and Rausch 48 .
Laine and Ford 49 reported that photolysis of neutral bimetallic come
plexes such as (OS-CpW(CO)3)2 and (OS-CpMio(C0)3)2 leads to homolytic
metal-metal bond cleavage to give reactive organometallic radical
species. The interest in such species derives in part from their po-

tential roles in the synthesis of new organometallic complexes 50-52

and from possible usage in certain homogeneous catalytic processes 53’

54

In our laboratory, the first photoexchange of cyclopentadienyl
rings has been reported for d0 systems 55’57 . Irradiation of a
1 : l benzene solution of titanocene dichloride](D-10)titanocene di-
chloride with 313 nm wavelength light produced the exchange of the
cyc10pentadienyl ligand. Later it was expanded to the other metal
systems 56’58 . The mechanism proposed by Vitz the photoexchange re-

actions of the cyclopentadienyl ligands between titanocene dichloride

33

34

co

omwuoanuan msooooouwa mo msoauumom owomnoxm msmwag oaumaouonm How amwsmsowz venomoum

 

 

 

.m ouswam

©0\ ,, 7 I v . .1, N p
o o , a m n o . ._/
a w o o .ofi A o 8 O 6 x6 QB _o

35

molecules on the basis of fundamental group theoretical and chemical
considerations is given in Figure 8.

At the same time, Harrigan, £5, 21. propose a mechanism for the
photocleavage of Cp2T1012 to generate CpTiCl2 in neat benzene 47
which may well explain the Cp exchange reaction between (C5H5)2T1012

and (C5D5)2TiC12. Harrigan said that the formation of CpTiCl radical

2
as a distinct intermediate prior to reaction with solvent is supported
by neglibible effect of solvent concentration on the quantum yield and
the order of magnitude increase in quantum yield upon permethylation of
the cyclopentadienide ring. This latter effect was seen as a decreased
probability of ring-metal bond reformation due to greatly increased
steric hindrance. Additionally, he suggested that the formation of
CpTiCl2 in neat benzene in very low quantum yield indicates that reduc-
tive dissociation is an open reaction pathway and that recombination of
the cyclopentadienyl radical with CpTiCl2 is as expected, quite effi-
cient.

In order to clarify the role of CpTiClz, a series of reactions
under the conditions where the presence of CpTiCl2 intermediate seems

probable was reviewed (Figure 9).

 

 

 

61
Br2 (C12) CpTiClZBr (CpTiC13)
CHCl (CCl , HCl)---—-—-CpTiCl 47
3 4 3
CpTiCl2 + 60
MeOH CpTiClz(0CH3) , CpTiCl3
CBr4 CpTiClzBr, CpTiCl3

Figure 9. Reaction Scheme for the CpTiCl2 Intermediate

36

In the case of Br2 (or C12), the Br (or Cl) radical which is ther-
mally generated initiates the formation of CpTiCl2 intermediate by the
abstraction of one Cp ring (giving CpX, X = Cl, Br, CpX may further re—
act with X2 to give C H X as side product) and then reacted with Br

5 5 5
(or C12) to give CpTiCl

2

2X. In all the other cases, the CpTiCl which

2
is generated by the photolysis of titanocene dichloride solutions ab-
stracts a halogen atom to give the desired product. All these results
are favor a CpTiCl2 intermediate which was proposed by Harrigan and

t. 1..

Base on CpTiCl2 intermediate, another photolytic ligand exchange

pathway can be prOposed (Figure 10).

 

 

 

hv .

(05115)2T1012 ‘hv (c585) + (C5H5)TiCl2
(05D5)2T1012 (0505) + (0505mm2
(C5H5)° + (05D5)2T1012 (0585)(c505)T1012 + (0505)

 

(CSDS) + (CSH ) TiCl

5 2 2 (05115)(0505)T1012 + (6585)

. fast
(CSHS) + (05D5)TiC12 (C5H5)(C5D5)TiCl

2
, fast
(CSDS) + (C5H5)TiCl2 (CSHS)(C5D5)T1C12

 

 

Figure 10. PrOposed Mechanism for the Photolytic Exchange Reactions
of Titanocene Dichloride
(based on Photolytic cleavage to give cyclopentadienyl
radicals)

37

All the possible radical abstraction or combination process other
than these are associated with decomposition. In the abscence of such
decomposition, the recombination of the CpTiCl2 and Cp radical has to
be faster than any other processes. The exact nature of the CpTiCl2
is not fully understood at this moment. A comprehensive study of the
formation of CpTiCl2 and the different types of reactions it can under-
go is under investigation in our laboratory.

substitution of pi-bonded ligands in general has received much
attention. Information on ligand substitution mechanisms should enable
us more profoundly to understand homogeneous catalysis by transition
metal complexes. In 1956, Wilkinson, Cotton, and Birmingham 59 re-
ported the first pi-ligand transfer reaction in which ferrocene was
prepared from chromocene and ferrous chloride. In 1974 Rettig, g3,

a1. 13 , reported pi-ligand transfer reactions in (a) the reaction of

LiCSD5 with MCp2 in THF, M = V, Cr, Mn, Fe, Co, and Ni; and (b) the
reaction of Ni(C5D5)2 with MCp2 (M = V, Cr, Co) in benzene. Here I
wish to report my results of cyclopentadienyl ligands exchange in
various systems. Cp rings are exchanged between molecules of vana-
docene, chromocene or vanadocene monochloride and their analogs by
a thermal process; however, this exchange can be enhanced by uv
irradiation. In hafnocene dichloride and vanadocene dichloride and

their perdeutoanalogs, the exchange occurs exclusively by photo-

chemical process.

RESULTS AND DISCUSSION

A. Photolytic Cyglopentadienyl Ligand Exchange between Vanadocene
Monochloride and Perdeutero Vanadocene Monochloride

 

 

When benzene solutions containing a vanadocene monochloride]
(D-10)vanadocene monochloride mixture were irradicated with 313 nm
light (Intensity equal to 0.0025 Ei/liter-hour) intermolecular ex-
change of the cyclopentadienyl rings occurs. Table 11 shows the per-
tinent peaks in the mass spectrum of samples irradiated for varying
times. The peak at m/e = 221, corresponding to (D-5)vanadocene mono-
chloride, increases relative to the peaks at m/e = 216 and 226, the D-0
and D-10 species, respectively, as photolysis time increases.

The possibility of hydrogen-deuterium exchange rather that cyclo-

pentadienyl exchange is ruled out because there is a constant ratio of
all peaks between m/e - 60 and 71 in both photolyzed and unphotolyzed
samples. The ratio of the intensities of the m/e - 221 and 216 peaks
1221/1216’ is computed for each irradiation time. The fraction of ex-
change, F, is calculated on the basis of an equilbrium value of 1.88.
For each sample, a plot of 1n(1-F) vs. photolysis time was made, the
most probable slope calculated by a standard least squares treatment of
data, and the rate calculated by using the McKay equation 62 (Figure 11).
The quantum yield, calculated by dividing the exchange rate (1.1 x 10-4
M/hr.) by the light intensity (2.5 x 10-3 Ei/liter-hr) is 0.044 mol/Ei.

The equilibrium constant for the reaction:

(05-0 H) vc1 + (NS-c D) v01 3 2015-0 H)(I15-C D )V01

5 5 2 5 5 2 5 5 5 5

38

Table 11.

39

Mass Spectrographic Data for The Perdeuterovanadocene

Monochloride/Vanadocene Monochloride Ligand Exchange.

 

 

 

 

 

 

 

Time(hr) m/e = 216 221 226 65 66 67 68 69 70 71 I/ZZl/I226 F
0 85 1.5 100 85 100 32 20 32 60 58 0.015 0.008
10 90 23 100 90 100 32 16 40 70 64 0.23 0.12
20 92 45 100 81 100 32 21 32 63 57 0.45 0.24
30 89 70 100 78 100 39 24 46 58 70 0.70 0.37
40 90 88 100 82 100 35 20 41 62 66 0.88 0.47
50 94 110 100 84 100 34 18 40 65 62 1.10 0.59
60 92 128 100 85 100 33 19 36 68 64 1.28 0.68
70 87 141 100 87 100 36 20 34 68 66 1.41 0.74
80 90 157 100 82 100 34 18 36 70 65 1.57 0.83
90 88 178 100 86 100 36 17 44 65 60 1.78 0.94
120 91 189 100 82 100 37 18 43 69 64 1.88 1.0
Note: 2 thermal-exchange has been substracted from normalized intensity.
Table 12. Cyclopentadienyl Ligand Exchange between Vanadocene Compounds
of Different Oxidation States.
Compounds Exchange Results
3* b*
1% 5
(05H 5)2v01/(cn3c 5 H4)2VC12 Z
(05H5 )ZV/(CHBC 5 H4)2vc12 3% 30%
(CSH 5)zvl(cn3csaa)2vc1 52 50%

 

*a : room temperature, 24 hours

b : 70°, 24 hours

4O

 

'UI ba‘slbio'o.

15..

 

l-F

 

 

1 1 1 1 1 1
LC 20 30 4.0 5.0 6.0

TIME (IO HOURS)

 

Figure 11. McKay Plot for Perdeuterovanadocene Monochloride/
Vanadocene Monochloride

41

calculated from the intensities of the exchange peaks in the mass
spectrum at time t was x - (1221)2/(1226)(1216) - 4. (c .. a).

The exchange of cyclopentadienyl ligands is not solely a photo-
chemical process. 102 thermal exchange is observed when the sample
solution is stirred in the dark at room temperature for one week (about
0.5% for every 12 hours). At higher temperatures, the thermal exchange

is even faster.

B. chlopentadienyl Ring Exchanges - Miscellaneous

Thermal cyclopentadienyl ligand exchange has also been observed
in vanadocene/(D-10)vanadocene and chromocene/(D-10)chromocene. It
can be enhanced by either increasing the temperature or by irradiating
with uv light.

Like all the metallocene dichlorides, ligand exchange in hafnocene
didhloride/(D-10)hafnocene dichloride and vanadocene](D-10)vanadocene
dichloride occurs only by a photochemical process.

Rapid photolytic Cp ring exchange was found as expected in the
vanadocene/(D-10)vanadocene system.

we did observe side reactions in the following compounds. More
than 202 of the vanadocene dichloride or hafnocene dichloride photoly-
tically decomposed during the 24 hours photolysis at 313 nm required
to reach.exchange equilbrium.

The exchange of the cyclopentadienyl ligand in the three possible
binary combinations of CpZV, Cp2V01, and CpZVCl2 were studied. The
results are listed on Table 12. Irradiating the reaction mixtures by
all wavelength light always enhances the reactions, and Cp exchange

rates increase with decreasing metal oxidation states.

42

The photolytic ring exchange reactions were not able to be
detected in the mixed metallocene systems of (C5H5)2/(CH3C5H4)2VC1

or ((C TiC1)2/(CH C H ) VCl. The thermal exchange does occur in

535)2 3 s 4 2

(C5H5)2Cr/(CH3C5H4)2VC1 and (CSHS)2Cr/(CSD5)2V systems.

C. Possible Mechanism

 

There are two possible mechanisms which can be used to explain
the photolytic cyclopentadienyl ligand exchange reactions between
H-lO and D-10 metallocene dichlorides. First, the bimolecular bridg—
ing mechanism which was prOposed by Vitz on the basis of fundamental
group theoretical and chemical considerations, is shown in Figure 8.

Second, the photodissociation and recombination mechanism
suggested by Harrigan's experiments, in which CpTiCl2 in neat benzene,
can be generated from the direct photocleavage of the Cp2T1C12, then
rapidly recombine with a cyclopentadiene radical to regenerate tita-
nocene dichloride, is shown in Figure 10. In this pathway, however,
there is an equal chance for CpTiCl2 to recombine with a perdeutero-
cyclopentadiene radical to generate the exchange species.

Certainly more study is needed to evaluate the relative im-

portance of these two possible pathways.

 

43

EXPERIMENTAL

A. General

General experimental techniques, sources of common chemicals and
analytical instruments used in this work have been described in the
EXPERIMENTAL section of PART ONE. CrCl3 was purchased from Alfa Pro-
duct. Chromocene was prepared as previous reported 63 , and its cy-
clopentadienyl-d10 metallocene analog was similarly prepared.

The two filter solutions for the isolation of 313 nm light were:
(1) 0.002 E K CrO

2 4
cm) 60 .

in 0.07 M K2C03 (1 cm) and (2) 0.96 M C0804 (1.9

Each benzene solution, containing a metallocene compound (10-3 M)
and its analog (10-.3 MD, was irradiated by 313 nm light in merry-go-
round photolysis apparatus. The irradiated samples were analyzed by
mass spectrometry (Perkin Elmer RMU—6 Mass Spectrometer). Valerophe-
none actinomers were used to determine the lamp intensity 60 . These
solutions, irradiated under essentially the same condition as the ex-
change solutions, were analyzed by gas chromatography.

B. Photolytic Cyclopentadienyl Ligand Exchange between Vanadocene
Monochloride and (D-10)Vanadocene Monochloride

Weighed amounts of vanadocene monochloride and perdeuterovanado-
cene monochloride were stirred with benzene in 50 m1 volumetric flasks
in a glove box under argon. The solutions were stirred thoroughly and

diluted to 200 ml to give 1.1 x 10-2Ԥ;vanadocene monochloride and

44

1.2 x 10-2 M (D-10)vanadocene monochloride. After mixing the two in

dim light 3 m1 aliquots were added to the 13 mm culture tubes to make
about 60 samples. The tubes were securely stOppered under argon and
sealed with an oxypropane torch. Twenty-four exchange samples and six
actinomers were mounted in the merry-go-round photolysis apparatus
equipped with a medium-pressure mecury lamp centered with 313 nm light
filter solution. The temperature of photolysis was kept at 25°. The
equipment has been described previously 60 . After measured periods
of irradiation, four samples and one actinometer were withdrawn. The
samples were pooled, evaporated to dryness, and sublimed at 800 before
analyzing by mass spectrometry. The actinometer was analyzed by gas
chromatography.

C. Photolytic Cycgpentadienyl Ligand Exchange between Hafnocene
Dichloride and (D-10)Hafnocene Dichloride

 

 

The procedure was the same as described as above. A mixture of
1.5 x 10-2 Miin hafnocene dichloride and 1.2 x lO-Z‘M’in (D-10)hafno-
cene dichloride was used.

D. Thermal and Photolytic Cyclopentadienyl Ligand Exchange between
the H-10 and D—10 analogues of CpZV, Cp2Cr and of szvglz

 

The procedure was the same as described above. The concentration
of vanadocene dichloride and its analog is about 2 x 10-4 M, The con-
centration of vanadocene, chromocene and their analogues were close to

10'2 g.

E. Thermal and Photolytic Cyclopentadienyl Ligand Exchange in The
Three Possible Binary Combinations of CpZV,ACp2VCI and QR»V§;2

Weighed amounts of vanadocene and bis-(methylcyclopentadienyl)

vanadium monochloride were stirred with benzene in 100 m1 volumetric

45

flasks in a glove box under argon. The solutions were then mixed tho-

roughly to give 5.2 x 10-2‘M_vanadocene and 3.7 x 10-2

M bis (me thylcy-
clopentadienyl) vanadium monochloride. Half of the solution was stirred
in the dark, while the other half was stirred and heated to 70°. After
24 hours, the solvent was removed and the samples were analyzed by mass
spectrometry. Similar procedures were carried out in the study of the
(C5H5)2V/(CH3C5H4)2VC12, and (C5H5)2VC1/(CH3CSH4)2V012 systems.

F. Thermal and Photolytic Cyclopentadienyl Ligand Exchange in the
Mixed Metallocene Systems of (C5H5)20r/(C5D5)2VI SC5H5)2QEL

(QR3C5H4)2V01, and (CH3C5H4)2VC1/((C5H5)2Tigl)2

 

The procedure was the same as described above.

G. (Quantum,Yie1d Determination

 

The quantum yield for each ligand exchange process was determined
by applying the approach of McKay analysis 62 on the mass spectrogra-
phic data. It was calculated from the exchange rate and light inten-
sity.

The mass spectrum of vanadocene monochloride has a well characte-
rized mass spectrum.with a parent ion peak at m/e - 251. The mass
spectrum of the perdeutero- compound is, of course, identical except
that the peaks corresponding to fragments with one or two cyclopenta-
dienyl moieties appear 5 or 10 units higher, respectively. Figure 12
and 13 show the mass spectra of the two derivatives. In the mass spec-
trum of an unphotolyzed mixture of vanadocene monochloride and its per-
deutero-analog there are peaks at 216 and 226 with associated mani-
folds due to the isotopes 37Cl, 35Cl, 51V, 12C, and 13C. The peak at
m/e - 221 is zero for the unphotolyzed mixture. Photolyzed samples

show increased intensity of the 221 peak (Figure 14), which corresponds

opwuoasooaoz ocoUOVmem> mo asuuommm mom: .NH «Hanan

m\E Hm ow oHH mNH HmH HwH can

 

46

: __ _ .: ._-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

47

mpwuoaeuoeoz oeooopmnm>AOHIQV mo abuuuomm mum: Ana ouawwm

o\a Hm om HNH onH Had cNN

 

 

é . E _ _ _ _f.

 

 

 

 

 

 

 

 

 

48

 

 

 

 

 

 

 

 

 

, pl. 111.11.] II

 

 

 

 

 

 

 

22 186 156151 m/e
A. At to,
. Ill .‘1 “II In
226 216 191 181 156151 m
B. At to

Figure 14. Mass Spectra of Exchange Mixtures

 

49

to (CSHS)(CSD5)VC1+. The ratio of the peaks at m/e - 221, and 226 for
various periods irradiation reflects the extent of exchange.
The logarithmic form of the McKay equation 62 , as applied to the

exchange system, is as follows:

1n(1-F) , _ R (reactant A) + (reactant B) t
(reactant A)(reactant B)

where R is the exchange rate and the bracketed quantities are the con-
centrations of A and B. F represents the fraction of exchange in time

t to exchange at equilibrimn (t + 0°) :

(1221/226)t
F:

 

(1221/2264.
where I221/226 is the ratio of the peaks at m/e - 221 and 226.

The plot of 1n(1-F) vs photolysis time (Figure 11), had a slope
of -0.019, which was determined by a linear squares treatment of the

data.

2 M)

From the slope and from the concentration of CpZVCl (1.1 x 10-

2

and (C5D5)ZVC1 (1.2 x 10- MD, the exchange rate, R, was computed as

1.1 x 10-4 gig/hr (or 3 x 10-8 M/sec).

The quantum yield, computed directly from the experimentally

5

observed exchange rate R (1.1 x 10- M/hr) and corresponding absorbed

intensity I (0.0025 Ei/liter-hr), can be generally expressed as 60 :

R.(mol/1iter-hr)

 

0 =
I (Ei/liter-hr)

and has the value of 0.044 mol/Ei.

I

BIBLIOGRAPHY

l.

2.

10.

11.

12.

13.

14.

15.

BIBLIOGRAPHY

J. L. Calderon, F. A. Cotton, and J. Takats, J. Amer. Chem. Soc.,
.9; (1971) 3587.

J. L. Calderon, F. A. Cotton, B. G. Deboer, and J. Takats, J. Amer.
Chem. Soc., 93 (1971) 3592.

 

F. A. Cotton, "Dynamic Nuclear Magnetic Resonance Spectroscopy",
Academic Press, New York (1975) p.419.

J. L. Calderon and F. A. Cotton, J. Organometal. Chem., 30 (1971)
377.

 

F. W. Siegert and H. J. De Liefde Meijer, J. Organometal. Chem.,
29_(1969) 141.

 

E. 0. Fischer, S. Vigoureus, and P. Kuzel, Chem. Ber., 2§_(l960)
701.

 

C. P. Stewart and A. L. Porte, J. Chem. Soc., Dalton, (1973) 722.

 

G. Wilkinson and J. M. Birmingham, J. Amer. Chem. Soc., Z§_(1954)
4281.

P. M. Durce, B. M. Kingston, M. F. Lappert, T. R. Spalding, and
R. C. Srivastave, J. Chem. Soc. (A), (1969) 2106.

 

F. W. Siegert and H. J. De Liefde Meijer, J. Organometal. Chem.,
23_(1970) 177.

A. Van. Baalen, C. J. Groenenboom, and H. J. De Liefde Meijer,
J. Organometal. Chem., Zfi_(1974) 245.

 

R. L. Cooper and M. L. H. Green, J. Chem. Soc. (A), (1967) 1155.

 

M. E. Switzer and M. F. Rettig, Inorg. Chem., 13 (1974) 1976.
C. L. Gibbons, Ph.D. Thesis, Michigan State University (1973) 34.

R. Jackson, J. Ruddlesden, D. J. Thompson, and R. Whelan, J. Orga-
nometal. Chem., 125 (1977) 57.

 

50

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.
26.

27.

28.

29.

30.

31.

32.

51

J. P. Candlin and H. Thomas, "Adv. Chem. Ser., v. 132", American
Chemical Society, Washington D. C. (1974) 213.

K. G. Allum, R. D. Hancock, 1. V. Howell, S. McKenzie, R. C. Pit-
kethly, and P. J. Robinson, J. Organometal. Chem., 81 (1975) 203.

K. G. Allum, R. D. Hancock, S. McKenzie and R. C. Pitkethly,
"Catalysis", v. 1, J. W. Hightower (Ed.), North-Holland,
Amsterdam, (1977) 477.

T. J. Pinnavaia and P. K. Welty, J. Amer. Chem. Soc., 91_(1975)
3819.

 

J. C. Bailar, Jr., Cat. Rev. Sci. and Eng., 19_(l974) 17.

C. U. Pittman, Jr. and G. 0. Evans, Chemical Technology, (1973)
560.

 

Z. M. Michalska and D. E. erster, Platinum Metals Review, 18
(1974) 65. .

C. H. Brubaker, Jr., "Encyclopedia of Polymer Science and
Technology", Supplement v. 1, N. M. Bikales, Ed., John Wiley
and Son, New York (1976).

C. U. Pittman, Jr., L. R. Smith, and R. M. Hanes, J. Amer. Chem.
Soc., 21 (1975) 1742.

C. U. Pittman, Jr. and L. R. Smith, ibid., 21_(1975) 1749.
C. U. Pittman, Jr. and L. R. Smith, ibid., 21’(1975) 341.

C. U. Pittman, Jr., S. E. Jacobson, and H. Hiramoto, ibid., 91
(1975) 4774.

C. U. Pittman, Jr. and L. R. Smith, J. Ogganometal. Chem., 29_
(1975) 203.

C. U. Pittman, Jr. and R. F. Ellis, ibid., 22_(l974) 289.

C. U. Pittman, Jr. and R. M. Hanes, "Horizons in Organometallic
Chemistry", v. 239, Ivan Bernal, Ed., New York Academy of
Sciences, (1974) 77.

C. U. Pittman, Jr. and L. R. Smith, "Organotransition—metal
Chemistry", Yoshio Ishii and Minoru Tsutsui Eds., Plenum Press,
New York (1975) 143.

L. D. Rollman, J. Amerggphem. Soc., 21_(1975) 2132.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

4'8.
49.
50.

51.

52
L. G. Allum, R. D. Hancock, 1. V. Howell, R. C. Pitkethly, and
P. J. Robinson, J. Catal., 4§_(1976) 322.

K. G. Allum, R. D. Hancock, I. V. Howell, T. E. Lester, S. McKen-
zie, R. C. Pitkethly, and P. J. Robinson, ibid., 43 (1976) 331.

K. G. Allum, R. D. Hancock, 1. V. Howell, R. C. Pitkethly, and
P. J. Robinson, J. Organometal. Chem., 81 (1975) 189.

K. G. Allum, R. D. Hancock, 1. V. Howell, S. McKenzie, R. C. Pit-
kethly, and P. J. Robinson, ibid., §Z_(1975) 203.

J. P. Collman, L. S. Hegedus, M. P. Cooke, J. R. Norton, G. Dol-
cetti, and D. N. Marquant, J. Amer. Chem. Soc., 24_(1972) 1789.

J. P. Collman and C. A. Reed, ibid., 2§_(1973) 2048.

R. H. Grubbs, C. Gibbons, L. C. Kroll, W. D. Bonds, Jr., and
C. H. Brubaker. Jr., J. Amer. Chem. Soc., 2§_(1973) 2373.

W. D. Bonds, Jr., C. H. Brubaker, Jr., Chandrasekaran, E. 8.,
C. Gibbons, R. H. Grubbs, and L. C. Kroll, J. Amer. Chem. Soc.,
21_(1975) 2128.

 

E. S. Chandrasekaran, R. H. Grubbs, and C. H. Brubaker, Jr.,
J. Organometal. Chem., 120 (1976) 49.

J. Manassen, Platinum Metals Rev., 15 (1971) 142.

A. Patchornik and M. A. Krans, J. Amer. Chem. Soc., 23.(1970)
7587.

A. Dormond, Ou-khan, and J. Tirouflet, J. Qgganometal Chem.,
110 (1976) 321.

M. R. Hunt and G. Winter, Inorg. Nucl. Chem. Lett., §_(l970) 529.

F. E. McFarlane and G. W. Tindall, Inorg. Nucl. Chem. Lett., 2
(1973) 907.

R. W. Harrigan, George S. Hammond, and Harry B. Gray, J. Or-
‘ggnometal. Chem., 81 (1974) 79.

H. Ald and M. D. Rausch, J. Amer. Chem. Soc., 2§_(1974) 5936.
R. M. Laine and P. C. Ford, Inogg;,Chem.,.l§ (1977) 388.

M. S. Wrighton and D. S. Ginley, J. Amer. Chem. Soc., 21
(1975) 2065.

M. S. Wrighton and D. S. Ginley, ibid., 21.(l975) 4908.

 

52.

53.

54.

55.

56.

57.
58.

59.

60.
61.
62.

63.

53

M. F. Lappert and P. W. Lednor, "Adv. in Organometal. Chem."
v. 14, F. G. A. Stone and Robert West, Eds., Academic Press,
(1976) 345.

B. H. Byers and T. L. Brown, J. Amer. Chem. Soc., 21_(1975)
3260.

E. L. Mutterties, B. A. Sosinsky, and K. I. Zamaraev, ibid., 21
(1975) 5299.

Ed.Vitz and C. H. Brubaker, Jr., J. Organometal. Chem., 84
(1974) C16.

EdsVitz, P. T. Wagner and C. H. Brubaker. Jr., ibid., 107 (1976)
301.

Ed.Vitz and C. H. Brubaker, Jr., ibid., 104 (1976) C33.

 

Mein Peng and C. H. Brubaker, Jr., unpublished results.

G. Wilkison, F. A. Cotton, and J. M. Birmingham, J. InongLiNucl.
Chem., 2 (1956) 95.

 

Ed. Vitz, Ph.D. Thesis, Michigan State University (1974).
R. D. Gorsich, J. Amer. Chem. Soc., §2_(1960) 4211.

H. A. C. McKay, Nature, 148 (1938) 997.

 

E. 0. Fischer, W. Hafner, and H. O. Stahl, Z. AnorgégAllgem.
Chem., 282 (1955) 47.

 

111111111111111111111

03145 3149

1NillWlHHlllUlillll

     

  

    

3 1293