"1.

"-

 

 
 

"III-"i

    

I; Milly- I
{,II 'I 1 :IxI'II“ 111,:‘62 ‘,,

”’11‘,I,,;I
‘11 1‘1

    
   
 
  

 

   

     

”‘II‘VI?

     
      
 
 
 

    
   

'W‘

     
    

.—

W»

.g‘u"." w.-—-‘
w

      
         
   

   

‘M.u— .
....‘_ .«-—-.a.-———-
vow—‘—

. .
~.-'.Q—.€l. '
..L

“<-
.fl‘5
1:.
. .__._.hv
.
-_ —.-._.
#

  
    
  
 

_.-
w
—_—v~— —v-A-
..

    

v.0 ml

 

" ‘flfl
i: w ;
"L. 5%-: -
‘ -
Mg
w. .- ‘1
.. - u ._
* " was.
.0... ..
:i . .1‘_.."’ ‘~‘— --
M ‘
' 7&7 ~
' M‘
I A
. can _ ¢~'
., .—
. . mm”
.. " "
' m

r--.

  

M
.4 h...‘
r --

a—_«
.g- nu.-
. .- 'I'
‘r

  

.1. __
A

V ;:: ”~—
W‘Qu-
_"‘, ”12.?“
" ' W
o

u
c
‘_
a
”.4 <
.o—hy

c-..
4

 
 

.-

rfiiq’.’

Git.
‘3

m:
W
“fun

g

.9

 
 

..
4.
-
“323.14.
..- .,

 
 

.
a -.‘o'oui‘ ¢
‘2'! ..

“W 5:32..
.. . -
«a?» .r‘“

A. " .2‘.
....
'iéhxig“

:rm

Us?
..

..r

    
   
   
 
 

  
    
 
  

 

.
v’ -.:“w
- -'c-".v-\~ ..

: ¢-v-v-v
4’".~"2.

'5. ~ -
-‘

  

" [2115.14 .- .. '5 _
'I‘I’D'H . ...:1, e;...,§,.tf ,
1115-1311519 . ,1, 1; ..1‘.I_ 1112,15: 61:115.. '35,,31, 11..

“111’ -s Ir, 12133111?“ I.
111. pf. 17., "”135'1'1.LI1T’-1 1111 I
II' ' 1,
11“

   

- ~' 3‘1':
11119
’ -.‘

 
 
 

  

132..

- I31“ oly
121'41'1 ”7:11-71
. 1‘- -‘1 71L.

    

yI-

 
 
 
   
   
   
     
   

    
  
   

. {.1 ”p‘,‘
-. 315:“,qu
I)5,:L;!:

M" 1N"?
3731311“?

-..4 .u. -
- «4
s ‘--o- ,_. -
outrun-—
n . .

.1
. A - -Lw;.
coo...
..1.

.

   
   
  
 

      

     

. . .
L.
.1!

 

{£3.13
,In" alliifl rig:
\

«:1-1' I, ”Fahd“ ‘

     
    
  
    

   

. ; :,:2. . 5.5131 -
. . ’1'“ 1111:5111} 1.. $.11 ..:1,, v1
' ,_ "’;‘.‘.,.,tI,l‘Iu1'1m L2,, ”‘1." :,I:'I,...,.- ff‘é'.
. I A i,1;“f:'l [1:61 lg“ ”ll” I ,\‘III \i'1lit‘g?u1;;f.s- “,1
'171‘,,1"1“1I111'I”11191-1“ "" 1.1 '
w, H,I,I1II|1
‘1‘1!,If‘,11

”11,,
1111111’1I1I 11 l1

       
        
 
  

  
  

  

. jam €11
1I111~1l,.,.'3=1l'1.,1 1%,”,
1H1‘11111’!

I “1 11'“
‘1" 1'1; 111*

I

   
 

1’ -.
.. 1119,.

1|. ' II; ‘, ‘-'

,IEI, I. l. .‘ . I_'i,1|,,“,

l

  
 
 

.I. _.. 1.

1, ('7‘. ”-1,-1.7"...‘IL‘LI’ E'..3";l,1_-:E . 11:53..
44-1-1 31"." ‘ "~ "
flu"! I: I '|!'_.\,,

 

'Fév‘flchY ..
:l' .._._-
:22“
...
m.“
..o
v
2'.
_.
wt.
3':
“‘2.
..—
J.
g.
*5;—
.._.
b..."
b:
*1..-
M.—
F-
-;

 

  

 

THLSIS «a

This is to certify that the

dissertation entitled
METAL VAPOR SYNTHESIS, CHARACTERIZATION, ELECTRON-
TRANSFER KINETICS, CATALYSIS, ELECTROCHEMISTRY AND
SPECTROSCOPIC STUDIES OF BIS(ARENE)CHROMIUM COMPOUNDS
presented by

Tomi T. Li

has been accepted towards fulfillment
of the requirements for

Ph.D. Chemistry

degree in

 

 

 

 

 

‘60.»... H \ WM ..
Major professor ‘ ‘

Date February 4, 1982

 

MS U is an Affirmative Action/Equal Opportunity lrun'nm'on 0- 12771

 

 

MSU

LIBRARIES
A“.

 

 

RETURNING MATERIALS:
PIace in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

METAL VAPOR SYNTHESIS, CHARACTERIZATION, ELECTRON-
TRANSFER KINETICS, CATALYSTS, ELECTROCHEMISTRY AND
SPECTROSCOPIC STUDIES OF BIS(ARENE)CHROMIUM COMPOUNDS

BY

TOMI TING-TUNG LI

A DISSERTATION

Submitted to
MICHIGAN STATE UNIVERSITY
in partial fulfillment of the requirements

for the degree of

DOCTOR OF PHILOSOPHY

Department of Chemistry

1982

ABSTRACT

METAL VAPOR SYNTHESIS. CHARACTERIZATION, ELECTRON-TRANSFER
KINETICS, CATALYSIS. ELECTROCHEMISTRY AND SPECTROSCOPIC
STUDIES OF BIS(ARENE)CHROMIUM COMPOUNDS

By

Tomi Ting-Tung Li

X-ray crystallography and dynamic NMR have been
applied to determine the molecular structure of bis-
(1,3,5-trii30propylbenzene)chromium(I) iodide and
bis(l,3,5-triis0propylbenzene)chromium(O) respectively.
The staggered conformation was found to be stablized
by steric factors in both solid and solution states.For
a series of monosubstituted bis(arene)chromium cations,
it has been found that UV ligand-metal charge transfer
(LMCT) bands and ESR parameters have a lack of dependence
on the nature of ring substituents. The electrochemical
investigation of substituted bis(arene)chromium(I)
cations revealed an excellent correlation between half-
potential (E%) in DMSO and meta-substituent constants.

A correlation of the oxidation stability of sand-
wich compounds with their gas-phase ionization potentials
is evident from a linear relationship between the

oxidation half-potentials E; (or sum of Hammett<3m values)
2

and the gas-phase ionization potentials.

Comparisons have been made between kinetics para-
meters for the self exchange of (n6-arene)20r(I)/(O)
(where arene = benzene, toluene, methoxybenzene, biphenyl,
ethylbenzoate and chlorobenzene) measured by ESR line
broadening in dimethylsulfoxide, with the predictions
from contemporary electron-transfer theory. The biphenyl
system was additionally studied in a number of other
solvents. These reactions provide especially tractable
systems with which to test theories of outer-sphere
electron transfer since the work terms should be essenti—
ally zero, and the inner-shell contributions AG;n to
the free energy barrier are small and can be estimated
from infra-red spectrOSCOpy combined with crystallo-
graphic data. The solvent dependence Of the rate con-
stants for (06H5- C6H5)20r(I)/(O) self exchange was found
to be in reasonable agreement with the predictions of
the dielectric continuum model. The "experimental"
frequency factors are compared with the estimates
Obtained from the "reactive collision" and the "ion-pair

pre-equilibrium" model.

To My Wife and My Family

ii

ACKNOWLEDGMENTS

I would like to express my sincere gratitude to my
research preceptor, Professor Carl H. Brubaker, Jr., for
his guidance, patience, support and inspiration during
the course of this study.

I would also like to thank Professor Michael J. Weaver
and my fellow graduate students for all the help, discussions
and friendship.

Finally, I wish to express my deepest appreciation

to my wife, Teresa, for encouragement and love.

iii

TABLE OF CONTENTS

Chapter Page
LIST OF TABIIE S o I o o o o o o I o o o o a o o c o o o v
LIST OF FIGURES a o o o o o o o o o o o o o o o o u o Vii 1

PART I. SUBSTITUENT EFFECTS IN CATICNIC BIS(ARENE)-
CHROMIUM COMPLEXES. X-RAY, ESR, ELECTRONIC
SPECTRA AND CYCLO VOLTAMMETRY MEASUREMENTS . 1

IntrOdUCtion I I I I I I I I I I I I I I I I 3
Experimental . . . . . . . . . . . . . . . . 10
Results and Discussion . . . . . . . . . . . 22

PART II.ELECTROCHEMICAL AND ESR KINETICS STUDIES OF
ELECTRON TRANSFER IN THE SYSTEM BIS(ARENE)-
CHROMIUM(O)/BIS(ARENE)CHROMIUM(I). . . . . . 66

Introduction . . . . . . . . . . . . . . . . 67
Experimental I I I I I I I I I I I I I I I I 72

Results and Discussion . . . . . . . . . . . 77

PARTIII.ELECTRON EXCHANGE BETWEEN BIS(rF-ARENE)-
'CHROMIUM(I) AND BIS(nR-ARENE)CHROMIUM(O).
COMPARISONS BETWEEN EXPERIMENTAL AND CALCU-
LATED KINETICS PARAMETERS. . . . . . . . . . 104

IntrOduCtion I I I I I I I I I I I I I I I I 106
Experimental I I I I I I I I I I I I I I I I 11 0
Results and Discussion . . . . . . . . . . . 111

iv

Table

10

11

LIST OF TABLES

ESR parameters for Cr(arene)2+ in DMSO solu-

Page

tion 25°C IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIQII

Anisotropic ESR data for Cr(arene)2+ in DMF/

CHC13(

Cyclic voltammetry of Cr(I) complexes in

1 ;1 V/V) glassy SOlution ”140°C 0 I o o o o o o o

DNISO/Ool M BuuNClOu at 250C cacoouoooooooocoooo

Atomic coordinates and standard deviations in

bis(l,3,5-triisopropylbenzene)chromium iodide..

Anisotropic thermal parameters and standard

deviations in bis(l,3,5-triisopr0pylbenzene)-

Chromium iOdide IIIIIIIIIIIIIIIIIIIIIIIIIII

0
Bond distance (A ) Of (1,3,5-triisopropyl-

benzene)2Cr+ in cation 1 and cation 2 .........

Bond angle (deg) Of (1,3,5-triisopropyl-

benzene)ZCr+ ....... ....... ......... .......
Distance to the ring plane from the atoms..
Rate constants and activation parameters in

system ArZCro/ArZCr+ electron transfer

reaction in DMSO ......... ............ ........

Solvent effects on rate constants and

activation parameters in the system,

<biph)20r°/(biph)20r+ ......... ............

Cyclic voltammetry Of Cr(I) complexes in

24

29

37

48

52

53

55

87

88

Table

12

13

14

15

16

17

18

19

DMSO/0.1 M BuuNClOQ at 25°C ................
Cross-reaction constants kij for electron
transfer reaction ......... ............. ....
Electrochemical parameters and ionization
potentials for the chromium sandwich com-
pounds. .......... ................. ........ ..
Summary of ionization potentials and 20m

of metal sandwich compounds.. ....... . .......
Calculated reorganization energies and
related parameters for the self-exchange

of bis(né-arene)Cr(I)/Cr(0) in DMSO at 2200
Comparison of experimental and calculated
enthalpies Of activation for the self-
exchange of bis(né-arene)Cr(I)/Cr(0) in
DMSO at 22°C...............................
Comparison of calculated and "experimental"
frequency factors (M'lsec'l) for the self-
exchange of bis(né-arene)Cr(I)/(O) in DMSO
at 22°C..... ........ . ..................... .
Comparison Of "experimental" and calculated
free energies and entropies of activation
for the self-exchange of bis(Yé-arene)-
Cr(I)/(O) in DMSO at 22°C ... ............. .
Rate constants. activation parameters and

experimental frequency factors A(M-lsec-l)

vi

Page
89

95

100

113

119

123

125

Table Page
for the self-exchange Of bis(C6H5)2Cr(I)/

(O) in various solvents at 220C............ 127

vii

Figure

\O(D\lo\

10

11

12

13

LIST OF FIGURES

Metal vapor apparatus.............. ...... .. 19
UV L-M charge transfer bands for (arene)2Cr+

in methanol.... ............ . ..... .......u.. 23
The number Of substituents effect on the com-
ponents Of hyperfine structure for (arene)2-
Cr+........................................ 26
The number of substituents effect on the
components Of hyperfine structure for
(arene)2Cr+. ..... . .............. . ......... . 27
(a) Isotropic ESR spectrum (b) Anisotropic

ESR spectrum of (arene)ZCr+................ 28
AnisotrOpic ESR spectra for (arene)2Cr+.... 31
Anisotropic ESR spectra for (arene)ZCr+.... 32
100 mV/s cyclo voltammograms of (arene)2Cr+ 35
The E% values plotted against Kym of the
reduction OfCArene)2Cr+ ................ .... 39
Cyclic voltammogram of (C6H5COOCZH5)ZCr+

at 25°C ..... ........... ...... .......... ... no
The structure of (1,3,5-triisopropylbenzeneé-
Cr+ for cation 1 and cation 2 . .......... .. 43
Stereoview of the unit cell of (1,3,5—triiso-

propylbenaene)2CrI ........... .............. nu

Stereoview Of (1,3,S—triisopropylbenzene)2-

viii

Figure . Page
Cr+ cation .......... ..................... .. 46
14 ORTEP drawing of (1,3,5-triisopropylben-
zene)2Cr+ cation.. .......................... 51
15 The temperature dependence of 1H NMR spectra
in C7D8(toluene) for (1,3,5-triisopropyl-
benzene)2Cr(O) .......... . ................ ... 58
16 The temperature dependence of 13C NMR

spectra in C7D8(toluene) for (1,3,5-triiso-,

propylbenzene)20r(0). ...................... . 59
17 ESR spectrum of bis(diphenylacetylene)-
chromium cation in toluene at 25°C.......... 63
18 A possible mechanism for generalized
Oligomerization. ......................... ... 65
19 ESR spectra of 5.4 x 10-4 M (06HSCl)ZCr+
in DMSO at 22°C ...................... ....... 78
(a) No added (C6H5Cl)20r° ................. . 79
(b) Computer Simulation .................... 79
(c) Added 4 x Io‘ZM (C6H5CI)2Cr°. ......... .. 80
(d) Computer simulation... ................. . 80
(e) Added 8 x 10'2m (06H5C1)2Cr0 ........... . 81
(f) Computer simulation . ................... 81
(g) Added 1.5 x 10‘1M (C6H5Cl)2Cro. ......... 82
(h) Computer simulation ................. ... 82
(i) Added 2 x 10’1M (C6H501)20r°.. .......... 83
(j) Computer simulation . ................... 83

ix

Figure
20

21

22

23

24

25

26

27

28

29
30

Page
Broadening vs. concentration of bis-

(chlorobenzene)chromium(O) in DMSO at

220CIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 84
O/+

Log k Kg. 1/T for (C6HSOCH3MECr

electron transfer reaction in DMSO......... 85

The compensation effect: plots Of TAST

for (a) changes of substituents on

electron exchange reaction (b) changes

of solvents on electron exchange reaction.. 96
Plot of E% for the oxidation Of chromium
sandwich compounds XE; their lowest

ionization potentials (IP)................. 98
Correlation of the sum of Hammett parameter.
cm, with ionization potential Of (arene)2M

or ferrocenes . ........ . ......... . ....... .. 101
Generalized M. O. scheme of (n6-arene)2M

and (n6-arene)2M+, M=Cr . ................. . 102
The salt effect on electron transfer rate
constant of (CéHscl)ZCrO/+...... ...... ..... 120
Plot of logarithm of rate constant k for

self exchange Of Cr(C6H5'C6H5) °/+ in

various solvents versus (l/DOp-l/Ds)....... 131
NMR and Mass spectra of (06H501)2Cr(0)..... 141
NMR and Mass spectra of (C6HSCHO)2Cr(O).... 142

IR spectra of (C6H OCH3)2CrO and (06H -

5 5
OCH3)2Cr+..... ........................... .. 143

N

Figure Page
31 IR spectra of (C6H5Cl)2Cr(O) and
(06HSCl)2Cr(I). ........................... 144

xi

PART I

SUBSTITUENT EFFECTS IN CATIONIC BIS(ARENE)CHROMIUM
COMPLEXES. X-RAY, ESR, ELECTRONIC SPECTRA AND CYCLO
VOLTAMMETRY MEASUREMENTS

SUMMARY

X-ray crystallography and dynamic NMR have been
applied to determine the molecular structure Of bis-(1,
3,5-triisopr0pylbenzene)chromium(I)iodide and bis(1,3,5-
triisopropylbenzene)chromium(O) respectively. The stagger-
ed conformation was found to be stablized by steric factors
in both solid and solution states. For a series of mono-
substituted bis(arene)chromium cations, it has been found
that UV ligand-metal charge transfer (LMCT) bands and
ESR parameters do not depend, on the nature 0f
Of ring substituents. However, the inductive effects
appear to be more significant in terms of increasing
of number of electron-donating groups on the arene ring.
The electrochemical investigation of substituted bis-
(arene)chromium(I) cations revealed an excellent cor-
relation between half-potential (E%) in DMSO and meta-
substituent constants. This result indicates an increase
to more negative potential (i.e., increasing difficulty
of reduction) for electron donating substituents and

conversely a decrease for electron withdrawing

substituents.

The condensation of diphenylacetylene with chromium
atoms causes catalytic Oligomerization to give a mixture
of dimers and trimers of diphenylacetylene together with
bis(diphenylacetylene)chromium. A possible mechanism

for a generalized Oligomerization scheme is proposed.

INTRODUCTION

The inception of metal vapor synthesis, marked by
the intial paper Of Timms,1 which reported the formation
of bis(arene)metal complexes by treating transition-

2

metal vapors with various arenes at -196°C, provides

3

a useful alternative to conventional synthetic methods.

0..
@.

Many bisarene1tcomplexes of transition metal are

Mm + .x ————*_‘96.c

 

unobtainable by classical methods due to side reactions
that occur with substituted arenes bearing halogens or
functional groups such as amines or methoxy groups posess-
ing lone pair of electrons. For example when halogen-
substituted arenes (e.g.,chlorobenzene,14,4-dichloro-
biphenyl) are used in the Friedel-Crafts syntheses, the
final products consist of the desired bis(halogeneoarene)-
chromium complexes together with appreciable quantities

of the corresponding halogen-free bis(arene)chromium 'n-.
complexes.“ Therefore, the metal vapor synthesis has

provided a powerful entry into a variety Of difficultly

accessible bis(arene)chromium complexes.

The main advantage in metal vapor synthesis is first
that a pure bis(arene)chromium complexe is obtained by
direct one-step synthesis. Second, there are clearly
direct routes for the synthesis of bis(arene)metal
compOunds with specific groups (e.g. COOR, CHO, OR, Cl
etc.), since the conventional Friedel-Crafts synthesis
requires that the arenes components be inert to aluminum
chloride at Ioo°-2OO°C.

In the course of this work some bis(arene)chromium
complexes have been prepared by condensation of chromium
vapor with arenes in order to investigate mechanisms and
kinetics Of electron transfer between (arene)2Cro and
(arene)20r+. (See part II)

The development Of techniques for metalation Of
(benzene)2M, M=Cr, V, has provided routes for making the
derivative of bis(benzene)meta15"9 and in particular
binuclear metalarenes.9 A rigid, side-on dinuclear
arenechromium complex. Cr2(06H5-06H5)2.has been made by

10 Hence, attempts

vapor synthesis in low yield of 0.3%.
to synthesize various dinuclear, sandwich arenechromium
compounds have been also described from both metalation
of dibenzenechromium (see appendix) and condensation of
chromium vapor with arenes. The schemes Of making the

substituted bis(arene)chromium complexes and binuclear

metalarenes are shown as follows:

(A) Bis(arene)metal compound

Q“, <9?"

n. .0'.‘ A M M

TMEDA, 70 C '
Li X

(B) Intersandwich compound

- R ©"

~
-'

mm, 70': ! -Io'c I

M

 

@*@

 

 

@r*—©

+42):

o)
<8

( 4 )

1H and 13C NMR studies have been conducted for a

series of bis(arene)chromium(O)compounds by Graves and
Lagowski.630 They concluded from an analysis of 13C

C-4 chemical shift of monoarenechromium(0) complexes
that a significant reduction in ring aromaticity occurs
upon ring complexation to chromium. Further, the 13C
0-1 chemical shift in monoarenechromium(0) compounds has
been found to be a function of electronegativity Of
substituent.

Electrochemical studies on the reactivity of the
organometallic compounds as a function of the substituents
have led to the possibility of estimating the nature of
substituent electronic effects on the metal center.11
This approach has proved fruitful in the chemistry of

12’13 and arenecyclopentadienyl iron cations.1LL

ferrocene
In this work, the first electrochemical investigation
on bis(arene)chromium cations is also interesting.

The dibenzenechromium(0) compound has been used as a
catalyst in polymerization of ethylene under 250°C and
2800 lb/in2 pressure. This polymerization may be proceed
by atomic level catalysis since the reaction begins at
the temperature close to the decomposition point of
dibenzenechromium. From the preceeding Observation, it
is not surprising that in the chromium vapor synthesis.

the vaporized chromium atoms can be used as catalysts

for various Oligomerization and polymerization of the

18-21 and quadricyclane?2 It is Of

Olifins. alkynes
interest to note that catalytic activity towards ethylene
polymerization Of a binary system consisting Of bis(arene)-
chromium cation and alkylaluminum has been explored under

26 Recently Huang and coworkers

a variety of conditions.
used bis(arene)chromium(O) compounds as catalysts for
oligmerization of perfluoropropylene under mild conditions
(at room temperature).25 In addition to bis(arene)chromium-
(0) complexes, bis(arene)titanium(0) complexes have also
‘been reported to be effective catalysts for Oligomerization
of butadiene under various conditions.24

Substituent effect on the ESR data in cationic bis-
(arene)chromium complexes has been the subject of several
reports.31+"36 primarily with alkylsubstituted dibenzene-
chromium compounds.35’36 It is instructive to examine
these results in the light of ESR data relating to the
nature of various substituents in the arene ring.

The cO-condensation of alkynes with transition-metal
atoms causes catalytic Oligomerization of the alkyne.18-21
Some alkynes were trimerized to benzene by reacting with
chromium vapor, but no bis(arene)chromium compounds have
been detected among the reaction products. Co-condensation
of chromium vapor with benzonitrile Obviously tends to
bis(benzonitrile)chromium, but this major process is
accompanied by cyclotrimerization of benzonitrile to form

a small amount of 2,4,6-triphenyltriazine.21

The reductive Friedel-Crafts method affords a con-
venient route to Obtain certain simple bis(arene)chromium
complexes, e.g., bis(benzene)chromium and bis(toluene)-
chromium. Thus, when the complexed alkylbenzenes (e.g.
ethylbenzene, isopropylbenzene) are used in conventional
Friedel-Crafts synthesis a mixture of bis(arene)chromium
complexes is Obtained. The reason that individual bis-
(arene)chromium complexes cannot be obtained is that
transalkylation and isomerization of the original alkyl-
benzene occur due to the action of aluminum chloride at
Ibo-200°C.

In order to avoid the impurities formed in the Friedel-
Crafts method, bis(l,3,5-triisopropylbenzene)chromium has
been synthesized by using the co-condensation of chromium
vapor with 1,3,5-triisopropylbenzene at -196°C. Oxidation of
the complex in air in presence of KI aqueous solution
gives the iodide of bis(l,3,5-triisoprcpylbenzene)chromium-
(13 which wasobtained in the crystalline state and used
for the X-ray structure study. The structure of bis-
(benzene)chromium(I) iodide,27 bis(toluene)chromium(I)

28 bis(ethylbenzene)chromium iodide,29 n12 -[3,3]-

iodide,
(paracyclophane)chromium(I) triiodide30 and 1,1.-
tetramethylenedibenzenechromium(I) iodide31 have been
studied. The studies Of effects of the substituents

on structures, stereochemistry and bonding of bis(arene)-

chromium complexes has been investigated recently by

32

Eyring, Zuerner and Radonovich. There has been only one
comformational study so far in the bis(arene)chromium com-
plexes by means of dynamic NMR spectroscopy.33 One inter-
esting feature of bis(l,3,5-triisopropylbenzene)chromium(I)
iodide is an interpretation Of conformational studies by
the steric effect of three isopropyl groups on the arene
ring. Hence, X-ray crystalography and dynamic NMR have
been applied to determine the molecular structure of
bis(l,3,5-triisopropylbenzene)chromium iodide and bis-
(1,3,5-triisopropylbenzene)chromium(0) repectively.

In this work the metal vapor synthesis, characteri-
zation, electron-transfer kinetics (part II and part III),

catalysis, electrochemistry and spectroscopic studies of

some bis(arene)chromium compounds have been described.

EXPERIMENTAL SECTION

I, General Information

Airless ware, a glove box, and a vacuum line were
used for this work. All the solvents used for ESR or
cyclic voltammetry measurements were purified by standard

67

techniques.

II. Materials and Syntheses Of Bis(arene)chromium(0) com-

pounds

The (n6-arene)zchromium(0) complexes, which were
synthesized by metal-ligand condensation techniques,63
contained the functional groups CH3, H, COOC2H5’ Cl and

65

OCH The compounds prepared by Friedel-Crafts syntheses

3.
contained H, CH3 and Ph.

As an example of the vapor synthesis of chromarenes,
(06H50H3)20r, was prepared by the following procedure.
In a 1 L. flask (-196°C), chromium vapor was produced by
electrical heating (8 V, 60 A) in a tightly wound conical
tungsten basket at a pressure of <10-LF torr(Fig. 1). In
a typical experiment, toluene (8 ml) was condensed with
chromium vapor (600 mg) over 1 h. The reaction flask
was allowed to warm to room temperature and the excess

unreacted toluene was removed at reduced pressure. A

Cold finger was placed in the flask and the product sublimed

10

11

sublimed at 700C in a vacuum. The black solid was removed

form the cold finger and stored_in an argon-filled dry box.

These compounds were identified by mass and NMR

spectra.63C Mass spectral data: m/e 276, C H

12 10°12°r(5'°%)'
360, CzuHZOCr(4.6%), 208, 012H12cr(63'6%)’ 236, Clquécr-
(60.2%), 268, C14H160Cr(22.1%), 352, C18H2004Cr(20.2%),
292, 018H240r(100%), 376, 024H36Cr(31.0%). Examples of
NMR and Mass spectra for bis(arene)chromium(o) are shown
in Figure 28 and 29.

The bis(né-arene)chromium(I) cations were Obtained
by oxidizing these chromium(O) compounds in air in the
presence of water to give the hydroxide of bis(né-arene)-
chromium(I). By adding sodium tetraphenylborate or
potassium iodide to these aqueous solutions, the precipi-
tate of the yellow tetraphenylborate or iodide salt of
bis(né-arene)chromium(I) was collected, filtered, washed
with water, then ether, and dried in a vacuum at room
temperature. The purity of the compounds was verified
by IR, UV-visible spectra and microanalyses. The IR
bands in the 300-500 cm-1 region are characteristic Of

66

bis(arene)metal sandwich compounds. The functional

groups, -CH3, -Ph, -i-C H -H, -0CH -COOC H -Cl and

3 79 3! 2 5’
-CHO were confirmed by IR. Among these CrI complexes,

(C6HSOCH3)2

Cr203, and the compound had to be stored in an argon-

CrBPhu is easily oxidized further,perhaps to

filled dry—box.

12

Tetrabutylammonium hexafluorophosphate (BuuNPFé)
was prepared by mixing tetrabutylammonium hydroxide and
ammonium hexafluorophosphate in acetone. The addition
of water gave a precipitate and the precipitate was
recrystallized three times from 95% ethanol. Tetrabutyl-
ammonium perchlorate (BuuNClon) was Obtained from G.F.
Smith Chemical CO. and recrystallized three times from

95% ethanol.

III. Prepartion of Benzyl(penzaldehydelchromium and Bis-

(penzaldehyde)chromium8

A three necked round-bottom flask (250 mL) was fitted
with a septum cap, a water cooled condenser and a nitrogen
inlet tube. To 2.08 g (10 mmole) of Cr(06H6)2 was added
31 mL of BuLi (1.6 Molar in hexane, 50 mmole), 7.0 mL
(50 mmole)N,N,N',N'-Tetramethylethylenediamine(TMEDA) and
30 mL dry hexane. The mixture was refluxed for 3 h and
then allowed to stand overnight. Then it was cooled in
a dry ice-acetone mixture and 4.9 mL(0.03 mmole) Of DMF
was added (-800C) in 10 minutes and the color changed from
red-brown to yellow-brown. The solution was stirred for
2 hours and then allowed to warm gradually to room
temperature. It was cooled to -20°C and then a solution
of 3.70 g (0.1 mmole) of NHuCl in 10 mL of water was added
in 1 h. The temperature of the mixture was gradually

brought to room temperature and 20 mL Of water and 66 mL

of benzene were added to the flask. The mixture was
filtered and the layers were separated. The aqueous
solution was extracted twice with benzene (150 mL). The
benzene solutions were combined; diluted with petroleum
ether (66 mL), and chromalographed on.A1203/10% H20 (150g
in petroleum ether in a 2.5 x 40 cm column).

Elution of the lower pale brown band with a 1:1
benzene-petroleum ether mixture gave 0.5 g (12%) of a
dark red substance, which sublimed in vacuo at 80°C
(2 x 10-3mm) to give a bright red petroleum jelly-like
substance. Cr(CéHSCH0)(C6H6). The NMR spectrum exhibits
a singlet at 4.15 ppm for the benzene ring protons; a
singlet at 10.08 ppm for the aldehyde proton, two poorly
resolved peaks at 4.91 ppm(ortho) and 4.32 ppm (meta and
para) for ring protons of benzaldehyde. Mass spectrum
gives m/e, 236(M, 44.1%), 158(CrC6H CHO, 38.1%),
130(CrC6H6, 73.6%), 52(Cr, 100%).

Elution of the dark brown band with benzene gave

5

2.3 g (63%) of Cr(C6H5CH0)2. The substance sublimes in
vacuo at loo-120°C (4 x 10"3 mm) and crystallizes from
benzene-petroleum ether as dark cherry red crystals.
The NMR spectrum exhibits 9.82 ppm for aldehyde proton
and two poorly resolved peaks at 4.78 ppm (ortho) and
4.24 ppm (meta and para) for ring protons. Mass

spectrum: m/e 264(M.2T%). 158(Cr06H CH0. 34.9%).

5
105(C6HSCHO-H, 24%), 52(Cr, 100%).

14

IV. Reaction of 1,3,5—triisopropybenzene with Chromium

Vapor

Seven tenths (0.7) g of chromium was condensed with
30 mL 1,3,5—triisopropylbenzene at—196OC anddelo'u torr
as described previously.63 The mixture was allowed to warm
to room temperature. The excess 1,3,5-triisopropylbenzene

4 torr) at 100°C, and a cold

was removed in a vacuum (<10‘
finger was placed in the flask. The bis(1.3,5-triisopro-
pylbenzene)chromium was sublimed at 130°C and 10"1+ torr
(5% yield). The orange-brown solid was removed from the
cold finger and stored in the argon-filled dry box. Mass
spectrum at 70 eV: m/e, 460(M, 41.8%), 250(CrL, 72.5%),

204(L, 48.5%). 189(L—CH3.100%), 52(Cr, 19.9%). NMR: 1H

NMR in C6D5CD3(25°C), one doublet of methyl protons
centered at51.32 ppm, one septet of isopropyl protons
centered at62-77 PPm. one Singlet of ring protons at
64.32 ppm. 130 NMR in C6D5CD§25°C): one singlet of methyl
carbons at 5 25.09 ppm, one singlet Of isopropyl carbons
at $33.19 ppm. One singlet of unsubstituted ring carbons
at 573.48 ppm and one singlet of substituted ring carbons
at 599.08 ppm.

The ("6—1,3,5-triisopropylbenzene)20hromium(I) was
Obtained by oxidizing the chromium(o) compound by bubbling

air into the benzene-water solution. The aqueous layer

was collected and potassium iodide was added to form a

15

yellow precipitate. The yellow salt of (n6-1.3,5-triiso-
propylbenzene)chromium(I) was filtered and recrystallized
from 95% ethanol. ESR in methanol solution (-60°C), g=
1.9866, A(HAr)=3'5 gauss, A(530r)=18.0 gauss; in glassy
solution of CHClB/DMF = 1/1, v/v(-14O°C), gL =2.0014,

g"= 1.9801, AL(53Cr)=25.3. The UV spectrum of L-M charge

transfer at 359 nm is characteristic band for(arene)20r+.

V. Reaction of Diphenylacetylene with Chromium Vapor

Method A

Chromium (0.7 g) was condensed at -196°C and 10‘”

torr with a decane solution Of biphenylacetylene (ca 5g
in 20 ml) over 1.5 hours from a closely-wound, conical
tungsten basket in a 1 L. flask. The mixture was allowed
to warm to room temperature and the excess decane was
removed in a vacuum at 50°C. The unreacted biphenylacetylene
was sublimed onto a cold finger in a vacuum at 40°C. The
residue was analyzed by mass spectrometry. The residue
contained bis(diphenylacetylene)chromium(O)(m/e=408)
together with a mixture of dimers (m/e=356) and trimer
(m/e=534) of biphenylacetylene.

The residue was precipitated from light petroleum
(-20°C) to give dark crystahsof bis(diphenylacetylene)-
chromium(0). The mass spectrum has m/e 408(M), 231(ML)

and 52(Cr). The ESR spectrum Of the cationic analogue

 

16

bis(diphenylacetylene)chromium(I) has g = 1.9878,
A(HAr)=3.75 gauss.

The mixture of organic products was repeatedly
chromatographed on A1203. The trimer of bisphenylacetylene-
(hexaphenylbenzene) was isolated and then identified by

1H NMR spectrum has a singlet at 6.81

MS and NMR. The
ppm. The mass spectrum has a parent peak at m/e=535.
The reddish-orange dimer(tetraphenylcyclobutadiene) was
not isolated as a pure compound. However, the mass

spectrum exhibits m/e, 336(M), 279(M—C6H5). 202(M-206H5),

126(M-3C6H5) and confirms the existence Of the compound.

Method B

Chromium (0.6 g) was evaporated over 2 hours and
condensed with diphenylacetylene vapor (10 g) at -196°C
by introducing the latter simultanously through a heatable
inlet (+500C). After removing the excess unreacted
diphenylacetylene, the residue was analyzed by mass
spectroscopy. Only the dimer and trimer Of diphenylace-

tylene could be found in the reaction products.

VI. Physical Measurements

NMR spectra were Obtained by use of a Varian T-60
or a Bruker WM 250 spectrometer. Benzene-d6 and toluene

-d8 were used as solvents for all the bis(arene)2chromium(0)

17

compounds in the measurements. Chemical shifts are
relative to internal TMS. Variable temperature 130 and
1H NMR spectra were obtained by use of the Bruker WM 250
spectrometer with TMS as an internal standard. ESR spectra
were recorded on a E-4 ESR spectrophotometer with a
variable temperature controller and DPPH was used as an
external standard. Infrared spectra Of KBr pellet sampler
were obtained by use of a Perkin-Elmer 457 spectrophoto-
meter at room temperature. Mass spectra were recorded by
use of Finnigan 4000 mass spectrometer (70 eV) with an
Incos data system. UV-visible spectra were recorded by
use of a Cary model 14 spectrophotometer. Microanalyses

were carried out by Schwarzkopf Laboratory, Woodside,

New York.

VII. Crystallographic Data and X-ray_Structure Analysis

A single crystal Of (1,3,5-triisopropylbenzene)ZCr
was prepared by slow evaporation of a methanol-chloroform
solution at 25°C. Crystal data: Crystals Of CBOHQSCrI
are triclinic; space group PI; a = 9.988(3), b=16.199(3),
c=9.479(2)A°, a: 91.08(2), 3: 106.92(2), y: 96.33(2);

1

2:2; M= 587.61;,%3= 1.340 g cm' ; p = 1.330 g cm‘3.

o
Lattice dimensions were determined by using a Picker FACS-
I automatic diffractometer and MoKa1 ( A= 0.70926AO)
radiation.

Intensity data were measured by using MOKa1radiation

18

(29 = 600) yielding 8422 totally unique reflections

max
and, based on I>3o (I), 5358 Observed data. The structure
was solved by Patterson methods and the refinement Of

the structure was by full-maxtrix least-square techniques.

The final R value was 0.037,

VIII. Metal Vapor Apparatus

A View of the metal vapor apparatus utilized in this
study is provided in Figure 1. It is based upon designs
previously described by Skell, Timms and Klabunde.64

The l L flask containing the evaporator assembly
is connected to the pumping system through an 8 mm high-
vacuum stopcock and a demountable 18/9 ball joint. By con-
necting this ball joint with a 3 mm high-vacuum stopcock
on the side arm for introduction of argon during the
product workup, another 8 mm right angle high-vacuum
stopcock is connected right on the top Of the ball joint.
The 3 mm stopcock is located between the ball joint and
8 mm right angle stopcock.

The evaporator assembly fits into the flask by
a 50 mm glass 0-ring joint with stainless steel plate
or inserted through 60/50 glass joint with stainless
stell plate stuck with expoxy cement to the upper part.

Two water-cooled 1/4 in copper tube act as electrodes

leads for the evaporator coil. One Of the copper tubes

19

@839 Scamsmwflc.lb

 

mesa owcmso

a

mapmnmmam yomm> Hopes H .wflm \\

me 0p

v

Mflv \\11 poemuogm>m

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a
anammmm _
sowsonco.
.
shade!
v a a...
.\\. -sooao

 

 

pounce g

.a

nmsuoh
Income op

 

\

how>hommh
shaved

20

is soldered to the stainless stell plate, the other is
insulated from the plate by a ceramic washer stuck with
epoxy cement to the plate and the tube. A terminal
block for the evaporator is soldered to the end of each
tube. A third 1/4 in. copper tube, soldered to the
stainless steel plate and to the appropriate terminal
block, admits the ligand vapor. It ends with two holes
(1/8 in) for spreading the vapor stream.

For simplicity, the three copper tubes can be
secured with expoxy cement to the plate directly.

The evaporator coil has five turns of 1 mm or 1.2 mm
diameter tungsten wire wound in a spiral with a 450 apex
angle. The bottom turn is bent close to the hole at
the apex of the core.

The ligand reservoir (10-50 ml) with 5 mm Teflon
stopcock is attached to the top by a 1/4 in. copper
inlet tube by a 18/9 glass-copper ball joint.

The vacuum pumping system consists of a mechanical
pump backing an Oil diffusion pump, which connects to a
pump trap, ion guage and a trap for excess ligand.
Electrical power for heating the coil is supplied from
a transformer with an output range 20 V and 100 amp,
controlled by a Rheoste .

A cold finger, with either an 0-ring or a 60/50 glass

joint is put in place of the evaporator as shown as

follows:

21

Vacumm System

I
1

Liquid N 2 "

 

 

 

 

"90r- shows the cold finger for sublimation.

RESULTS AND DISCUSSION

Characterization and Identification
A. Electronic Spectra

A characteristic UV absorption band at 340 nm for
(benzene)2Cr+ has been assigned as being associated
with the ligand to metal (elfiI'a1g) charge transfer.5Ob
The analogous bands, assigned by analogy, are listed
in Table 1 for a series of substituted (arene)2Cr cations.
The results Show that the variation of monosubstituents
in the ring have little effect on the UV ligand to metal
bands. It is of interest to note that those bands of the
complexes (1,3,5-trimethylbenzene)2Cr+, (1,3,5-triiso-
propylbenzene)20r+ and (hexamethylbenzene)2Cr+ induced
a red-shift of 16-26 nm compared with the bis(benzene)-
chromium cation (Figure 2). Thus, this shift might be con-
sidered as being due to a significant inductive effects
in terms of increasing the number of electron-donating
groups in the arene ring, since the electronic and
resonance effects differ only slightly in monosubstituted
bis(arene)chromium cations.

B. ESR Spectra
(a) Measured in DMSO at 25°C

Figure 3 and 4 Show that cationic bis(arene)chromium

complexes ShOW’a decrease in the components Of hyperfine

22

23

Ci? {1'}
A -— Cr I C”

¢> {3:}-

©0043

.....- + F _____ 0+

<3)— (b.0013

€3> Q“
C"- a‘ G” ....... + I~

CHO / \ c

ANA
\ \
\
/
/

Absorption

 

 

 

l l J J i l L J 1
BIG 320 330 340 350 360 370 380 390 400

Wavelength in nm

+
'Fig. 2 UV L-M charge transfer bands for (arene)2Cr

in methanol.

24

Table l-ESR Parametersa for Cr(né-arene)2+ in DMSO

solution 25°C

 

 

06H6 3.42 18.1 1.9860 340
06H50H3 3.46 18.0 1.9865 341
C6H50CH3 3.72 17.7 1.9872 346
C6H5C6H5 3.44 18.3 1.9853 346
C6H5000C2H5 3.30 18.3 ‘ 1.9858 345
06H5CH0 3.25 17.3 1.9863 339
1,3,5-C6H3(CH3)3 3.54 18.0 1.9860 356
C6H5CH0,06H6 3.23 18.1 1.9856 342
C6H5Cl 3.58 18.1 1.9867 342

.50 18.0 1.9866 359

KI.)

19395'C6H3(i-03H7)3
-- 18.2 1.9867 366
06(0113)6

I

 

a The experimental error of aI0.02 guass and gi0.0005.
b Measured in methanol, error in i 1 nm, loge:= 3.7-4.1

for the absortion coeficient.

structure with an increase the number of substituents due
to the number of ring protons (S=%) coupled with the
unpaired electrons. The results in Table 1 indicate the
effects of arene ring substituents on the ESR parameters
for a series of monosubstituted (né-arene)20r+ in DMSO
are less sensitive to the nature of substituents. These
results are quite consistant with those of UV bands which
show no significant substituent effects on L-M charge
transfer in monosubstituted bis(arene)chromium complexes.
In light of data Obtained from ESR spectra one may suppose
that a substituent attached to a complexed arene induces
little change in the electron-density distribution in

the ring. It is not as effective as that of the uncom-
plexed free arene ring in terms of the build up effective
charge densities on the ring protons. However, the
substituent effects are essentially inductive because
more alkylation in the arene ring is noted in the UV
ligand—metal transfer band. An analogous Observation is
obtained for ESR spectra in which the proton hyperfine
constants, A(HAr) increase with increasing number of
alkyl substuents on the arene ring.35’36 A possible
rationalization for this observation can be made by
considering that the electronic effect is greater when
more electron-donating groups are transmitted to fewer
ring protons in comparison with the five ring protons

in monosubstituted bis(arene)chromium cations. The

26

.+Hombocohmv pom chapozppm

ocflmmomh: mo mpcocomsoo one :0 pooewo mpCOSPfipmnsm Mo popes: one m .mflm

on:

an:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

gauzenazo c. 6.3 3 c
.1
n ..I<
IUOIQV 31a 5 Pmm
co
...
.5?va : AD
oBAWWV . I L
TII u .
_ t d nfim<m
u

.+poNAocommv pom whoposzm

onwwpoahs ho mesocOQEoo on» so poommo mQQOSPHpmnsm mo hopes: one :.wwm

00.

27

 

 

 

 

 

 

 

 

cm :0 5 PM.

 

I
0 O.
.631 um

35502 c. uoom-

 

 

 

 

 

 

ozufiv g T l.
.031 M .

288v I

 

A

28

 

 

 

 

 

 

 

9
9.50 l 1 @Ho
* 7’1, Hi @ao
(a) i H 25°Cin DMSO
: '1
.11
.i !

 

 

 

 

 

@0043
Cr+
(b) OCH3

 
  
       

-I40‘c In CHcIB/DMH'II’I.)
A; (53C!)

P___——————IO

Fig. 5 (a). Isotropic ESR spectrum (b) Anisotropic ESR

spectrum of (arene)20r+.

29

.mooo.ouw .mmmsm mo.ou< mo noppo HopcoEwnomxo one

A4< N + =<vm\a "la sore cocaseoo mosas> <

D

.m

 

 

Hema.a aaoo.m m.mm a.m -- -- slamme-avmsoo-m.m.a
oomo.H mmoo.m m.mm o.m pc.m s.m Hommoo
mmao.a amoo.m a.mm o.m s.m o.m cmoo .omommoo
mono.a maoo.m m.mm e.m .. .. movmmooum.m.H
moao.a omoo.m m.mm m.a e.m so.m omommco
asao.a osoo.m s.mm H.: om.m o.m : oooommoo
mono.a mmoo.m a.mm a.s .. I- mmcommco
oomo.a mmoo.m 8.:m m.s no.m m.m moommco
ammo.a Hmoo.m a.mm c.m a.m no.m mommco
omao.fi Hmoo.m o.mm o.m o.m H.m cmco
as =u Asommvae Asommvms Locales Asesv=<

 

coflpzaom

zmmmaw A>\$ H.vaaomo\msn Ga +oncohMICCVHo Mom moron mmm camohPOMfic< I N canoe

3O

satellites in the spectra of bis(arene)chromium cations
arise from the 53Cr isotope with a spin I=3/2. The
hyperfine splittings are calculated from the observed
satellites and are about 18 G for all the bis(arene)-
chromium cations (Table 1). All the g values are close

to 1.986 which is less than g values CV2.003) Of uncomplex-
38

It is generally recog-

37.50

ed substituted benzene radicals.
nized from several M0 calculations that overlap of
the metal (3dz2) with arene(o, alg) results in the
delocalization of the unpaired spin over the ring protons.
This lack of sensitivity of ESR.A(HAr),.A(53Cr) and g
factors in the monosubstituted bis(arene)chromium(I)
cations leads to the conclusion that the spin delocali-
zation from Cr(3d22) to arene(<L alg) is approximately
equal and is less effective than the delocalization of

38

the unpaired C2 2 electron to the uncomplexed free arene.
The substituent merely effects a redistribution of magneti-
zation among these arene protons.

The substituent effects on the ESR hyperfine con-
stants are not large in the bis(arene)chromium(I) cations.

However, these data can be related to paramagnetic NMR

measurements which might yield much more useful infor-

mation.50

(b) ESR Measured in CHCl -DMF(1:1 v/v) at -140°c

3

31

-I40‘c ln-CHCls/DMH". X.)

ALPS Cr)

 

 

 

 

440': h CHCl3/DMF(i( 1'4)

 
 

AL (5%”
r———-«
4‘1

    

Fig. 6 Anisotropic ESR spectra for (arene)2Cr+

32

@040
Cr+

I
440‘: in CHCISIDMFl/l T’v)

H“

 
   

 

E - I40°c in CHCl3/DMF ('42.)
+

$94 } Ad” Cr)

 

 

 

l ' ‘

 

206

..—

Fig. 7 Anisotropic ESR spectra for (arene)2Cr+

33

The anisotropic ESR spectra of all the substituted
bis(arene)chromium cations studied in a glassy dimethyl~
formanidezchloroform (1:1) solution consisted of two
lines described by coupling of the unpaired electron
with the nuclear spin of proton ( =%) and with aniso-
tropic g-tensors (g" and g‘L ),(see Table 2). However,
the observed hyperfine structure was well resolved only
for the bis(benzene)chromium cation and increasingly
less well resolved on passing from (benzene)(benzaldehyde)-
chromium to bis(benzaldehyde)chromium to bis(1,3,5—triiso-
propylbenzene)chromium cations (Figure 5,6,7). This implies
that, in the glassy solution, the molecular rotation will
be prevented from averaging out the anisotropy by the
increase in the number of substituents. Such hindered
rotation has been suggested by the lack of resolution
in the ESR spectra of alkylsubstituted bis(arene)chromium
complexes.36’37

In the bis(arene)chromiufll)compounds with small
substituent groups, such as H, CHO, Cl, CCH3 and CH3.
the values of AN(HAr) and/or A1(HAr) can be observed
since the rotation averages out the anisotropy around
2 and/or x,y axes. However, the compounds posess bulky
substitutents such as Ph or i-C3H7 or have more substi-
tuents on the ring cause the hyperfine lines to be‘

broadened and no hyperfine structure can be obtained on

z and x,y peaks. All the AH(HAr) and AL(HAr) are listed

34

in Table 2.

In the ESR spectra of substituted bis(arene)chromium
cations, the Al(53Cr) can be evaluated from the observed
53Cr line around the x,y peak. No axial 53Cr hyperfine
structure can be found because it is hidden under the
z peak,37 but A“(530r) still can be estimated from the
formalism of A = 1/3(A"+ 2A,) and are listed in Table 2.

In summary, the lack of any significant dependence
of ESR parameters on the ring substituent is not sur-
prising in view of the small effects of ring substituents

on ligand-metal charge transfer bands.

Reduction Potentials

One electron reduction potentials of bis(né-arene)-
chromium cations were measured in DMSO by cyclic voltamme-
try. Typical cyclic voltammograms are shown in Figure 8.
The separation of the peak potential and the height of
peaks is close to unity. (Table 3) All the half-wave
potentials, (E%)ArZCrO/Ar20r+, for the complexes are
given in Table 3.

The observed variation in the E% for the complexes
shows that the inductive effects of substituents on the
arene system are transimitted to the chromium atoms. The
Cr(I) oxidation state is destablized by increasing

electron-withdrawing strength of substituents (more

35

Figure 8.

100 mV/s cyclo voltammograms of m10-3M

(a) (C6H5Cl)2Cr+ (b) (06H .06H6)2Cr+ (c) (C6H50CH Cr+

5 3)2
in DMSO/0.1_M BuuNClon at Pt electrode vs. SCE(25°C).

36

made»

 

 

 

To- To- To- 8.6... To- ...c- n6- «no-
3 .8 3
\1N\III-l
o: .
4am
.6 3. 3

 

 

 

 

 

 

acouuso

.ommdzm a4.o\omsn SH .m .AHvesfiEounvocOQMVMHn no mammoQHzCTQQMMPop

ohm pump esp .AHVESHEopsvo:OQMVMflQ Ho padm oncOH .o ma.oo.mu:o:pfipmn3m PCoMOMMfio

how mumposmnmm pposemm yo 83m one .6 .mo>m3 Awav owoocm dam AOflv cacoSHMO 059 how
mpcopnzo goon one mo owptg .o .xmom caconme 6cm owoocm a“ monopothc one we mm<,n

.Hmwpsopom goon oHUoSPMO can owoocm one no ommuo>m one no oopMHsoamo who; mosam> mm m

 

37

 

 

cm.o- mo.a on mmm.o +soxcmcUVAomommoov 6H
ms.o- Ho.H am cam.o o +soonmommcov o
oo.o+ Ho.“ on mmo.o . +sommmfismmomfivmmcoum.m.aL m
35.6- mo.a ms mmm.o e o +soNAHommcov a
ma.o- oo.H mm mmm.o +somnmmmoooommoov 8
No.6- me.a so mao.e +aomnmmcomscev m
mm.o- cc.a or mms.o +somammoommoov a
o mo.H ma ooa.o +somAcmcov m
aa.o+ mo.H we o~m.o +somfimmommoov m
«3.0+ oo.~ om cao.o o+som mammovmmoo H
amaow.. o ”a nfi>svawo Amom .m>->v ccsoasoo oz

u-I

.oomm es ooaozasm a.a.o\omzn cs moxoaasoo AHvso mo Rheoseoeao> oefloso . m canoe

38

positive potentials) and stablized by an increasing
electron—donating strength of substituents (more negative
potentials). Similar observation have been reported by
using a rotating disk electrode16 and in acetonitrile.17
Hammett's parameters 20m for substituents are listed
together with the E% values in Table 3. In Figure 9, the
sum of Ohm values for the substituents in the complexes

are plotted against the E; values. The straight line
2

obeys the following equation (slope‘3= -.58)

- E = 0,73 - 0.58 fifi................(4)

NIH

The reduction of bis(ethylbenzoate)chromium(I) in
DMSO is abnormal (Figure 10). Two reduction waves
which correspond to one electron process are seen in a
height ratio of 1:0.6. It is suggested that a structure
5)2Cr+, possibly leading to the
69

"slipped sandwich" structure,

change for (C6H5000C2H
occurs at the second
wave at E% = -0.558 V, AEp = 71 mV, iC/ia = 1.03.
It has been shown that the structure of (n6-p-06H4F2)2V
has an interesting distortion on the arene ring (deviation
from the planarity of arene 0.06 A0).70 Thus, electro-
chemical evidence suggests that such a "slipped sandwich"
structure, just like (né-p-C6H4F2)2V, occurs to relieve
the electron density around the metal. In fact, it is

very similar to that reported for the reduction of

cyclooctatetracne, in which the molecule undergoes a

39

 

009 u-

0.8 _

j

-E§(V)

0.71

0.6»

0’I5 I

00H P

0.3

V

b

0.2 [4 A

 

A

 

A g A .A— A A

 

 

-006 "00,4v -002 O 002 00"“ 006

20m

Fig.5? Hammett plot of the reduction of AreneZCr+ at a
Pt electrode. The E2 values plotted against 20m.

The numbers correspond to the compounds

listed in Table 3.

40

Figure 10.

Cyclic voltammogram of (06H 000C Cr+(r\.10-3 M)

5 2H5)2
at 25°C in DMSO/0.1 M BuuNClou. The scan rate was 100

mV/sec at Pt electrode.

41

uu~o>

5.0

O

H

 

 

 

 

 

o

 

 

acouuao

42

profound structural change, i.e., from a tub to a planar

72

system. The first wave at -0.355 V is apparently due

to the symmetric coplanar structure.

X-ray Structural Study of Bis(1,3,5-triisopropylbenzene)-

chromium(I) Iodide

The crystal structure consists of two independent
cations (CrC30H48+) and two symmetry related anions (1‘)
in each cell. The stereoscopic view of two unit cells of
Cr E1,3,5-06H3(i-CBH7)§E+I' along c is shown in Fig. 12
Two chromium atoms are situated in special pOsitionS,
(0,0,0) and (5,3,5) and the iodines are situated in a
general position. The four Cr-I distance are 6.178, 6.266,
6.357 and 6.592 A0. Each cation consists of a chromium(I)
atom situated at an inversion center. For example, cation
1 contains Crl, C1 to C15 and C1. to C15 , cation 2 con-
tains Cr2, C16 to C30 and 016' to 030'. The normal
vectors of the two sets of arene rings form an angle of
~700. The detailed comformations of two cations are very
similar. The steroview and ORTEP drawing of cation 1 is
shown in Fig. 13 and 14 respectively. Hydrogen atoms have
been omitted for clarity. The atomic coordinate and the
isotropic thermal parameters are listed in Table 4 and 5
respectively.

Ring carbons are planar with the maximum devivation

43

Am oanmevmpopoemmmm anemone 6cm A: canoe Vmopmcfiopooo

owsovm Op mummom esosom mcflaaoan one .m Coflme 6cm

H coame how HMO

N coarse

Nam

A

mmm

ouavmzcoum.m.au mo osseosson one .HH osswaa

a coaeso

 

44

 

Fig. 12 Stereoview of the unit cell of

[1 13v 5_06H3(l—CBH7)3] 2CI‘I -

45

 

r‘ ‘6
‘-" ‘9 is
, ,‘p» 5‘.
- ' x
v: e "I .. \v

‘_ .. -
«ll 5' 4‘ fig
. ‘hl\_ ‘fl'b ‘
s it. '4 -
\.V ‘
If. :‘\
\ “y "/
Ami-

+

Fig. 13 Stereoview of [1,3,5—C6H3(i-C3H7)3]2Cr cation.

Hydrogen atoms are omitted for clarity.

46

_

Table 4- Atomic coordinateéiand standard deviations in
. bis(l,3,5-triisopropylbenzene)chromium iodide.

ATOM

X Y Z
C1 .0320(3) .1357(1) .0017(3)
C2 .1627(3) .1053(2) .0651I3)
C3 .1815(3) .0519C2) o1838(3)
C4 .3644C3) .0276(2) .2362I3)
C5 -.9682(3) .0552(2) .1746l3)
C6 -.0819(3) .1105l2) .0571($)
C7 .0216I3) .ZOOSCZ) .1142C3)
C8 .0728‘6) .2849(2) .0347(5)
C9 -.1237(5) .2005(3) .2227(5)
C10 o3253(3) .0273(2) .2664(3)
C11 .4018(4) .0973(3) .3807(4)
C12 .5154C4) .0077(3) .1695C5)
C13 -.1889(3) .0375(2) .2397(3)
CIQ '01937‘4) 00449(2) 03122(4)
C15 -.1826(5) .1093(3) .3480(5)
C16 .5651I3) .4958t2) .7419(3)
C17 .5043(3) .4158(2) .6786(3)
C18 .3686(3) .4023K2) .5751(3)
C19 .2962(3) .4723C2) .5326(3)
C20 o3516(3) .5531(2) .5945C3)
C21 .4871(3) .5634(2) .6984(3)
C22 .707413) .510712) .8608(3)
C23 .SOOSCQ) .4422l3) .8653(5)
C24 .6821t5) .5226I3) 1.0114(43'
C25 .2989C3) .3142(2) .5272(4)
C26 _ .2176(4) .2994(2) .3652(4)
C27 .2022(5) .2909t3) .6251(5)
C28 .2623(4) .6238(2) .5572(4)
C29 .3460(5) .7085(2) .5665(5)
C30 o1680(5) .6235i3) .6585(7)
H(2) .237(3) o121(2) .034i3)
HIQ) .070(3) .010(2) .310l3)
H(6) -.165(4) .127(2) .01713)
Ht?) .092(3) .190(2) o165l3)
H(8A) .163(5) .288t3) .024l5)
HIBB) ~o002(5) .301(3) .030(5)
H(8C) .359(5) .324(3) .101(5)
H(9A) -.197(5) o214(3) .160(S)
H198) -o149(4) .149C3) .282(5)
”(9C) ‘0123(4) 0242(2) 0293(9’
HIIO) .307l3) .025(2) .315(3)
H(11A) .345(5) .120(5) .435(6)
H(118) .433(5) .148(3) .337i5)
H(11C) .492(4) .082(2) .442i4)
HIIZA) .446t5) .058(3) o121l5)

Table 4- Continued-

ATOM

H(128Y
'H(12C)

H(13)

H(14A)
H(148)
H(14C)
H(15A)
H(158)
5515c»
H(17)

H(19)

H(21)

H(22)

H(23A)
H(238)
chsc)
H(24A)
H(248)
thac»
4125)

H(26A)
H(260)
H(26C)
H(27A)
H(27B)
H(27C)
H(28)

H(29A)
H(298)
H(29C)
H(30A)
H(308)
H(30C)
I

CR1

ca:

aCalculated standard deviations are indicated

X

 

.0369(6)

0500(5)
0263(4)
0106(5)
0191(5)
0274(4)

‘0183(4)

0082(6)
0255(4)
0554(3)
0213(4)
0519(3)
0760(3)
0825(5)
0760(5)
0893(4)
0634(4)
0633(5)
0771(5)
0375(3)
0276(4)
0151(4)
0172(4)
0121(6)
0257(6)
0163(5)
0208(4)
0402(5)
0407(4)
0278(4)
0219(6)

0111(6)

..10394)
.47521(3)

0
0500

in parentheses.

47

Y

 

“0043(4)
“0005(3)

0037(2)

-0051(3)
‘0090(3)
-0050(2)

~015°(3)
0111(3)
0101(2)
0372(2)
0465(2)

0614(2).

.563(2)
0438(3)
.392131
.461(2)
.474(3)
0572(5)
0534(3)
0276(2)
.30512)
.337x2)
0241(3)
.325c3)
.293(3)
.237c3)
0613(2)
.710<3)
0730(2)
.7sn(2)
0638(3)
.sszca)
0661(2)

076$96(2)

0
0500

Z
0097(6)

.223(5)
0166(4)
.397(5)
.2socs»
.351t4)
0306(5)
0436(6)
0383(4)
.71013)
0466(4)
.737c3)
0839(3)
0773(5)
0885(5)
0948(4)

10029(4)
10012(5)
10088(5)

0546(3)
0305(4)
0332(4)
0340(4)
0608(6)
0720(6)
0598(5)
0463(4)
0495(6)
0677(4)
0532(4)
0756(6)
0660(6)
0622(4)

002456(3)

0
0500

48

.~.s~.
S2.
~«.nc0~
.NVN0.
Auvc0
aa~N0I
AN.N0~
“awn-I
auum0
auvo0l
Amun0l

..mvon0
AQVHC0I

Auvnm0

A~v~m0
3:2;

.N9h0m
amvmoa
aawn0
.mvn0u
Auvmo
.mvw0
.an0
.~ua0a

.o.mo0
.o.sm.

..~.o~.

A~.wu.

cam“.
.~.¢.~
.fi.~.~
.H.~.H

amvm0NI

A—vm0l
guvn0l
«Neo0l
.mec0a

amvoO0I
acvmm0

onm~0
.®.NN0
.mvmm0
.mvnm0
Ameow0

,awvmm0

amvnw0
.mvwh0
Advom0
oncm0

.mvoo0
acvmuo

.mvmo0l

.mwmn0
.mumm0

.an0l
Anvc0
anvm0
.an0l
aavh0t

gmvwoc
A~VQ0N

Am.NO0«

anew.
Aueu0l

nmvw0n
.~.¢0

.mvmm0

.«vmod
.~.c0l

auvh0

.«va0d

.~.00
.ovum0
.mvem0

amefio0l

Amuwa0
.mku0
.mva0
.hecno

mum

aovmn0

.mvm0u

.NVN0

AHVN0H

.¢.mo.
.w.~c.
.x.mo.
.m.oc.
.m.¢~.
.m.mo.

nAm

Am~N¢0
.fivc0

aavu0l

.mvmc0
.mvhm0
.mvhc0
acuwm0
6m~Nn0

.mvmc0l

abvqn0
Nam

.350.
6505
3.26
.n.~.m
.~.0.¢
.~.m.n
.~.m.c
.~.~.e
.H.o.c
.~.m.~
.~.m.¢
.~.¢.~
.~.¢.~
.~.~.~
.~.o.n
.~.m.~
.fl.s.~
.~.e.~
.~.m.m
.~.¢.e
.H.».~
.~.o.m
.~.m.n
.H.m.n
.~.¢.c
¢~.c.m
.~.o.~
.~.m.~
.H.n.~
.~.~.~
.H.m.~
.~.~.~
.~.e.m
nnm

.23 .m
5?;
3.25
.m.m.m
.~.¢.n
.~.o.n
.~.¢.m
.d.m.m
.~.¢.~
.~.m.m
.~.m.m
.H.~.n
.~.m.~
.~.c.n
.~.~.n
.~.m.~

e.g.00m

.Avs0m
.N.O0m
.mvo00
.uvm0c
.mvm0m
.mvo0m
Aav¢0n
.mwm0e
eave0~
Aavm0m
adve0~
auwm0~
Auvm0~
auvm0m
aaew0m

aoecc0m

Nmm

$wa
A3n~.fi
fivnm 0o
.m.w.¢
.~.o.¢
.H.~.n
.u.~.m
.N.n.¢
.H.o.m
.~.m.¢
.N.~.n
.~.o.n
.~.u.n
.~.o.m
.~.~.~
...m.m
.a.o.m
.~.~.m
.~.n.m
.m.a.¢
.~.n.~
._.~.~
._.~.n

Advo0m.

.m.0.m
.n.~.h
.a.w.n
.avn.~
.H.m.~
afivn0~
.~.H.m
.~.o.H
.a.o0m
dam

mcflcOM ESHEopsoAmCmNCmQHhQOMQomwwhpum.m.vafin

qu
"mu

one
mNu
mNu
~NU
mmu
mNu
emu
nmu
NNU
“mu
emu
mac
mau
hau
0H0
mac
emu
nuu
mau
duo
emu

mu

mu

50

cu

mu

cu

no

Nu

mu

zoh<

Ca mcowpmH>mU cumuCMPm vamMWMmmeMMmm Adahmnp canonpomwc< um manme

49

.m swsopsp H cmaoho mum b ccm H can .nprmH
Hamo Hmoopmflomg wnp ma m¢9m¢ .meCw nmaafis map ma m 090:?

B

:\A Asvm<em<0AHvm<em¢05Hm0Asvm0Ava vzam-

«890% 0:9 mm: “OPOM% mMSPMMmQEmp was
.mmmePMMMQ :fl cmpwowccw mum mcowpwfi>mv upmvcum cmPMHSOHmom

newscfipcoo um mange

50

Figure 14. ORTEP drawing of (1,3,5-triisopropyl—

benzene)2Cr+ cation

51

 

52

Table 6- Bond Distance (A0)a of (1,3,5-triisopropyl-
benzene)2Cr+ in Cation 1 and Cation 2

 

 

Cation 1 Cation 2
01-Cr1 2.185(0) C16-Cr2 2.199(1)
C2-Cr1 2.167(1) 017- Cr2 2.188(0)
03-0r1 2.191(1) C18-Cr2 2.204(1)
C4-Cr1 2.165(1) 019- or2 2.147(1)
CS-Crl 2.184(0) 020-0r2 2.183(1)
C6-Cr1 2.170(0) C21—Cr2 2.169(0)
02-01 1.416(4) o 7- 016 1.412(4)
C6‘C1 1.410(3) 021- C16 1.412(4)
07- 1 1.523(2) 022- C16 1.526(3)
03-02 1.414(2) 018-0:7 1.414(3)
04- 3 1.419(3) C1fl9 1.414(4)
Clo-C3 1.522(3) C2 H5 1.518(4)
05-04 1.416(4) 020- 0:: 1.409(4)
C6‘C5 1.417(2) 021- 020 1.413(4)
013-05 1.511(3) 028-020 1.512(4)
08-07 1.512(4) 023-022 1.518(5)
09-07 1.515(4) egg-€22 1.532(5)
011-010 1.524(4) 026‘025 1.514(2)
012-010 1.512(2) 027-025 1.531(3)
C14‘C13 1.516(4) 029-028 1.513(5)
015-013 1.520(3) 030-028 1.528(5)

 

aEstimated standard deviations are in parentheses

units of least significant digits.

in the

53

m a - +
1able 7- Bond angle(deg) of’[1,3,5-06H3(1—C6H7)3]2Cr

 

 

Cation 1 Cation 2
020106 118.8(2) 017016021 118.2(2)
020107 119.3(2) 017016022 122.5(2)
060107 121.6(3) 021016022 119.3(2)
010203 121.3(2) 016017018 121.7(2)
CZCBCLL 118.2(2) 017018019 117.9(2)
0203010 122.2(2) 017018025 119.7(2)
0403010 119.3(1) 019018025 122.0(2)
030405 122.0(2) 018019020 122.4(2)
C4C5C6 118.0(2) 019020021 117.6(3)
0405013 122.1(1) 019020028 119.9(3)
0605013 119.6(2) 021020028 122.4(2)
010605 121.7(2) 016021020 122.1(2)
010708 108.0(1) 016022023 114.2(2)
010709 115.0(2) 016022024 108.8(2)
0807c9 110.4(4) 023022024 109.9(2)
03010011 107.9(2) 018025026 115.7(2)
03010012 114.8(1) 018025027 107.4(2)
011010012 110.5(1) C26C25C27 110.3(3)
05013014 115.2(2) 020028029 114.3(3)
05013015 107.8(2) 020028030 108.8(2)
014013015 110.9(0) 029028030 110.2(3)

 

a . . . . .
Estimated standard dev1ations are in parentheses in

the units of least significant digits.

54

from the ring plane being O.OO8AO (cation 1) and 0.01 A0
(cation 2)(Table 8) and the two ring planes within the
cation are parallel. The average C-C bond distance in
the ring is 1.416 A0 (cation 1) and 1.413AO (cation 2).
The average Cr-C bond distance 1.655AO (cation 1) and
1.664Ao (cation 2) and are close to the corresponding
values of other known bis(arene)chromium(O) and bis-

28-31 The bond distance

(arene)chromium(I) complexes.
and bond angles for both cation 1 and cation 2 are listed
in Table 6 and 7 respectively. Inspection of Table 8
reveals the orientation of the is0propyl groups in cation
1 is somewhat different at carbon 3 than at carbons 1 and
5. The deviation from the ring plane for the attached
isopropyl group at carbon 3 is larger, [C10(0.17A0),
011(1 .6 A0), ClZ(-O.6O.AO)] compared with values at carbon
1 [07(0.14A°), 08.(1.61A°), C9(-O.47AO)] and carbon 5
[013(0.13A°) C15(1.6OAO), C14(-O.48A°)]. The interplanar
distance between the two rings in cation 1 is 3.31AO
larger than the corresponding value in bis(benezene)-
chromium iodide (3.18 A0)?7 Therefore there must be
considerable steric hindrance between the isoprOpyl
groups, taking the bond between the ring carbon and the
attached isopropyl carbon'h6O out of the ring plane.
There is a similar conformation for cation 2 (Table 8).
Because of the steric interaction of six isopropyl

groups between two rings, the oritation of cationic

55

Table 8— Distancea to the ring plane from the atoms

 

 

 

 

Cation 1 Cation 2

Atom Distance to the plane Atom Distance to the plane
01 .00369 C16 -.00352
C2 -.00746 C17 -.00302
C3 .00622 018 .01180
C4 -.00135 C19 -.01415
CS -.00232 C20 . .00733
C6 .00123 C21 .00155
Cr1 -1.65396 Cr2 '-1.66290
C7 .14378 022 .06728
C8 1.61376 024 1.49764
C

9 -.46559 023 -.36884
Clo .17336 C25 .21749
C11 1.66229 C27 1.70750
C12 -.60026 026 -.61658
013 .1290? C28 .13270
C14 -.46783 C29 -.48621
C15 1.60307 C30 1.61114
a Distance (A0), sign "-" means atoms under the plane,

sign "+" means atoms above the plane.

56

bis(l,3,5-triisopr0pylbenzene)chromium with respect to
the isopropyl groups favors the staggered conformer and
that one is formed. The stablization of the staggered
conformer is dominated by steric factors due to the
three bulky isoprOpyl substituents in the ring and is
also supported by dynamic NMR studies.

It is of interest to know that how much the internal
ring angles at a substituted carbon change due to the
three electron-donating isoprOpyl groups on the ring
in bis(l,3,5-triisopropylbenzene)chromium(I) iodide.
The effects of various substituents on internal ring
angle of uncomplexed benzene rings has been reviewed by

39-00 They gave a general rule

Domenicano and co-worker.
in which the ring angles at the substituted carbon
appears to be larger than 1200 for electron withdrawing
groups and less than 120 for electron-donating groups.
Accordingly, since there are three isopropyl-substituted
electron donating groups, then one would expect the

ring angles to be <1200. This effect is also expected to
have a cumulative effect of three isopropyl groups in
meta positions.39 The X-ray determination reveals that
the ring angle at the substituted carbon has the average
value of 118.60, consistent with values observed in
uncomplexed aromatics and consistent with the electron-

donating nature of isopr0pyl groups. Similar results

have been found with electron-withdrawing substituents

57

32

in bis(arene)chromium(o) compounds.

NMR Studies on [1,3,5-06H3(i-03H7512Cr(0)

 

The observed temperature dependence of the 1H and

130 NMR signals in C7D8(toluene) for the bis(1,3,5-triiso-
propylbenzene)chromium complex indicates a restricted
rotation around chromium-arene bond (Figure 15 and 16).
The 1H NMR spectrum at -90°C exhibits two resonances

at 4.21 ppm and 4.39 ppm for ring protons and they are
of unequal intensity and thus the line difference is
due to the exchange of two different site populations.
Similarly, the 130 NMR spectrum shows absorptions at
72.63 ppm and 73.41 ppm for unsubstituted ring carbons.
From the distinct pair of chemical shifts at -90°C, it
is possible to estimate an approximate barrier to
rotation of 10.5 t 1 kcal/mol for the staggered confor-

mation by using the convenient equation,101

k=n(Wé-WO).
where WS is the half-height line-width of the staggered
conformer resonance line due to the exchange process, and
W0 is the half-height line-width of the sta ered con-
former for a zero exchange rate. It is reasonable to
assume that the barrier to rotation in bis(1,3,5-triiso-
propylbenzene)chromium is mainly due to excess steric

interaction of the three bulky isopropyl groups and is

supported by the X-ray crystallography study.

58

-3o°c

 

 

 

 

 

-60‘c
-90°c
\ x
1 1 1 1 1 1

 

1
Fig. 15 The temperaturedependence of “H NMR spectra
in C7D8(toluene) for (1.3.S-triisopropyl-

benzene)2Cr(0).

59

 

-2.1°c
-3o°c
-60°C
T T T
80 70 60 0pm

Fig. 16 The temperaturedependence of 130 NMR spectra
in C7D8(toluene) for 1,3,5-triisopropyl-

benzene)2Cr(0).

60

Since the rotation of 1,3,5-triis0proPylbenzene
ligand of bis(l,3,5-triisopropylbenzene)chromium(O) is
restricted in solutiin, there might be two different
rotameric conformers in this process. The kinetics of
this process is described in terms of two unequally
populated ring proton sites between the sterically
preferred staggered conformer(1) and the electronically

preferred eclipsed conformer(2)
Ck '-—————— ::§C?
(2)

(1)

Steric factors appear to be prevalent in the crystal
as the X-ray structure determination shows. If this
preference for a staggered conformation holds in solution,
then the more intense peak will be identified from 1H NMR
spectrum at -900C. (Figure 15) The population difference
between the staggered and eclipsaiconformations, ratio
60:1, allows calculation of an 1.5 kcal/mol energy
difference between conformers. Since the isopropyl

groups retain rotation in solution, the eclipsed con-

formation can exist only in small amounts in solution

61

due to the steric factors being overcome to some extent

by electronic factors.* Hence, the population of the
staggered conformer is larger than that of the eclipsed
conformer.

* The electronic factors permit a maximum overlap between
the filled orbital of the arene and the metal free orbitals.

Redox Potential and ESR Spectra of[1,3,5- C6H3(i-C3H7)3]2Cr+

 

The half-wave potential of the reversible one
electron reduction of the bis(l,3,5-triisopropylbenzene)-
chromium cation in DMSO/0.1 M BuuNPFé at 25°C is -0.925 v
vs. SCE and is 0.135 V more negative than that of bis-
(benzene)chromium(I) under the same conditions. This
supports the previous studies15 that the inductive
effects of substituents on the arene system is transmitted
to the chromium atom and is stablized by the in increasing
electron donating strength of the substituents (more
negative potentials).

The ESR spectrum in methanol solution (-6OOC) gives

g = 1.9866, A( = 3.5 gauss and A(53Cr)=18.0 gauss

HAr)
which confirms the previous conclusion that ESR parameters
do not depend on the presence of the substituents on the
arene ring related to bis(arene)chromium(I) compounds.

The ESR spectrum in glassy solution (CHClB/DMF = 1/1, -14OOC
) gives gL=2.0014, gu=1.9801 and AL(53Cr)=25.3 gauss.

62

No hyperfine structure can be observed since the isopropyl

groups prevent rotation at low temperatures. The steric

factor of three bulkyl isopropyl group on the ring will

be predicted to affect substantially the frequency factor
. o/+

for self-exchange of [1,3,5-06H3(1-03H7)3]20r and

has been discussed in ref. 42.

Mechanism of Catalytic Oligomerization on Reaction of

Diphenylacetylene with Chromium Vapor

The condensation of alkynes with transition-metal
atoms causes catalytic Oligomerization and high oligomers.
18-21 Some alkynes have been trimerized to benzene by
reacting with chromium vapor, but no bis(arene)chromium
compounds have been detected among the reaction products.

On condensation of chromium vapor with benzonitrile,
bis(benzonitrile)chromium forms, but this major process

is accompanied by cyclotrimerization of benzonitrile to

form a small amount of 2,4,6-triphenyltriazine.21 Reported
here is the reaction of chromium atomswith diphenylacetylene
to a give bis(diphenyacetylene)chromium, 1%(I), together
with a mixture of dimers 34%(II) and trimers 65%(III) of
diphenylacetylene as determined by mass spectrometry.

An analysis of the ESR spectrum of bis(phenylacetylene)-
chromium(I)(Figure 17), gives g = 1.9878 and exhibits

hyperfine structure with A(HAr) = 3.75 gauss, which is

d)
6

.oomm aw 0:0:H0p cw cowHMO 55HEOusoAccodzpoomahcoanvyawn 60 83690090 mmm ma .wflm

GO—

 

 

0:033-u ... 00mm

0......va

@000

64

similar to a series of bis(arene)chromium(I) compounds15

with a sandwich structure.

 

The reddish-orange dimer, tetraphenylcyclobutadiene,

may have been isomerized to an air stable colorless compound
of tetraphenyltetrahedrane, identified from its mass spectrum

“1 that tetra-tert-butyl-

and is similar to the reported
cyclobutadiene can. 'be converted to tetra-tert-butyl-
tetrahedrane. The amount of trimer, hexaphenybenzene, has
been isolated and identified by comparison of its MS, NMR
and IR with those an authentic sample.

0n the basis of mass spectroscopic product analysis,

a possible mechanism for a generalized Oligomerization

scheme is proposed and shown in Figure 18.

 

 

 

 

 

Fig. 18 A possible mechanism for generalized oligomer-

ization.

PART II

ELECTROCHEMICAL AND ESR KINETICS STUDIES OF ELECTRON
TRANSFER IN THE SYSTEM BIS(ARENE)CHRCMIUM(0)/EIS(ARENE)-
CHROMIUM(I)

SUMMARY

The self-exchange rates and activation parameters
in the (né-arene)20r(0)/O16-arene)2Cr(I) systems (where
arene = toluene, benzene, methoxybenzene , biphenyl,
ethylbenzoate and chlorobenzene) have been measured by
the ESR line broadening technique in DMSO solvent.

For a series of solvents, a linear relationship
between log k (the exchange rate constant of (biph)2Cr°/+)
and (l/n2 - l/DS) where n2 and DS are the optical and
static dielectric constants of the solvent medium
respectively, has been observed. This linearity is
predicted by Marcus treatment for outer-sphere electron
transfer process.

A correlation of the oxidation stability of sandwich
compounds with their gas-phase ionization potentials is
evident from a linear relationship between the oxidation
half-potentials E% (or sum of Hammett (3m values) and the

gas-phase ionization potentials.

66

INTRODUCTION

Electron exchange between various sandwich compounds

and the corresponding cations have been investigated by

43’4“ Typical of this class is the reaction

44-46

several groups.

between ferrocene and ferricinium ion.

@ @ 4‘?

+- IRE :::::: lag 4. (We

@ @

'Among those methods for measuring the rates of

(20

exchange reactions, the ESR method has been applied to
electron transfer reactions between aromatic hydrocarbons

”7'48 The ESR line

and their anion or cation radicals.
broadening method cannot be applied to the ferrocene
systems because of the short relaxation time character-
istic of the ferricinium cation49 so further corroboration
of results seems unlikely.

The isoelectronic and isostructural bis016-arene)-
chromium(O) complexes (abbreviated Ar2Cr) and the cor-
responding chromium(I) cation (abbreviated Ar20r(I)) can
be studied by ESR line broadening. Their molecular orbital

bonding schemes are well known.35’50 Polarograms of cation

and neutral species yield similar E; values, indicating
2

67

68

the reversibility of the reduction-oxidation couple.
Methyl derivatives, like bis(toluene)chromium, have more
negative reduction potentials, whereas phenyl derivatives
like bis(biphenyl)chromium, have more positive reduction
potentials.52 A good correlation is obtained between

the reduction potential and localization of the highest
occupied molecular orbital (HOMO) predicted by substituent
effects.53 Therefore, a correlation with exchange rates
would be interesting. The core sandwich structure remains
essentially the same throughout the series of derivative§¥+
so the free energy change related to electronic reorgani-
zation and inner sphere contributions can be eliminated
for theoretical treatment. The fact that one of the
reactants is neutral,allows further simplification in
theoretical calculations of the exchange rate, since
coulombic interactions are absent. Finally, the electron
exchange would necessarily be simple, because the
reaction of the ligand, ion transfer, or atom-transfer

are unlikely.

For these reasons, the electron exchange between
ArZCr(0) and ArZCr(I) was studied by ESR linewidth
measurements. The effects of substituents upon the exchange
rate were studied, and the effect of solvent variation on
the exchange rate was determinated in relation to the
predictions of the Marcus theory.55

The theory of electron transfer is a straightforward

69

example of a jump process involving many Sites. For the
slow limit,56 the electron-exchange sites interchange so
slowly that the resonance lines are only slightly broadened,
. are the resonance frequencies

J
of the ith and jth line, and Tp’ the mean life of the

paramagnetic species. Assuming Lorentzian lines Shapes,

second-order rate constant k is

1.52 x 107 AH
k = C ......(1)

 

where AH is the peak-to-peak increase in line width
increase due to exchange, and C is concentration of the
neutral molecule (diamagnetic form) in moles/liter. For
the fast limit,57 when the transfer is so fast that an
averaging of the individual resonance frequencies occurs
'and only one coalescenaatime is obtained at the center
of the original spectrum (fwi-letpx>l). The second
order rate constant k is given

2.03 x 1077
H(C)

......(2)

 

where‘V is the second moment (in gaussz)58 of exchange
of the ESR spectrum in the absence of exchange, AH is
the peak-to-peak line width of the derivative curve

corrected for natural line width, and C is the concen-
tration of the neutral molecule (diamagnetic form) in

moles/liter. Adams el at 59 have made a quantitative

7O

observation of slow exchange (low-concentration regions
on neutral molecules) and fast exchange (high-concentration
on neutral molecules) in ESR of organic radicals.

The chemistry and air stability of the symmetric
(ArZCr) compounds79 have been correlated with the sum of
the Hammett 20m parameters of the substituents?0 The
existence of the series of sandwich compounds depends on
two factors: (a) The thermal stability of the compound
which reflects the strength of the ligand-metal bonds.

(b) The kinetic stability of the compound which reflects
the one-electron oxidation of the compounds.. For example,
the (“é-arene)2M compound of later transition metals is
destabilized due to the extra electron being accommodated

*-
in antibonding e orbitals (Figure 25) and weakening

18
the metal-ligand bonds. There is also an evidence for
14e', 15e', and 16e' species (n6-arene)1M (M = Fe, Co, Ni)
in arene matrices at 770K, which decompose at ca-5000.6l’66
Whatever their thermal stability, the (n6-halogenoarene)2M
complexes (M = V, Cr, Mo) are less sensitive to air. Other
electron-withdrawing substituents have the same effect.
The relative air stability is due to the inductive effect
of electron-withdrawing substituents on the arene rings.

This effect is also seen with substituented ferrocenes.62’

12'1“ The kinetic inertness of the chromarenes is re-
flected in their resistance to air oxidation. The

presence of electronegative substituents in the arenes

71

63

should retard the one-electron oxidation to ArZCr+ as
supported by polarographic studies on these systems63a
and the activation energy barrier should correlate well

with oxidation potential or ionization potential in the

gas-phase.

EXPERIMENTAL SECTION

Materials and Syntheses

The (n6-arene)2chromium(0) complexes which were

synthesized by the metal-ligand condensation techniques21b’

22 contain functional groups of CH3, H, COOCZHS’ and by
Friedel-Crafts syntheses23 contain H and Ph. Bis(chloro-
benzene)chromium, (06H5Cl)2Cr, Bis(methoxybenzene)chromium,
(C6HSOCH3)2Cr and bis(l,3,5-trimethylbenzene)chromium
[(C6H3(CH3)3 JZCr were kindly provided by professor K. L.
Klabunde, Kansas State University.

As an example of the vapor syntheses of chromarenes,
(06H5CH3)2Cr, was prepared by the following procedure. In
a 1 L flask (-196°C), chromium vapor was produced by
electrical heating (8V, 60A) in a close-wound, conical
tungsten basket at a pressure of <10-LL Torr. In a typical
experiment, toluene (8ml) was co-condensed with chromium
vapor (600 mg) over 1 hour. The reaction flask was
allowed to warm to room temperature and the excess unreact-
ed toluene was removed in a vacuum. A cold finger was
placed in the flask and the product sublimed at 80°C
in a vacuum. The black solid was removed from the cold
finger and stored in an argon-filled dry box.

These compounds were identified by mass and NMR

spectra. Mass spectral data: m/e 276, C12H100120r (5.8),

72

 

73

m/e 360, C24HZOCr (4.6), m/e 208, 012H12Cr (63.6),

m/e 236, Clquécr (60.2), m/e 268, ClquéCr (22.1),

m/e 352, C18H2004Cr (20.2), m/e 292, C18H24Cr (100).
The bis(né-arene)chromium(l) cations were obtained

by oxidizing these chromium(O) compounds in air in

the presence of water, giving the hydroxide of bis(né—

arene)chromium(I). By adding sodium tetraphenylborate

or potassium iodide to these aqueous solutions, the

precipitate of yellow tetraphenylborate or iodide salt

of bis(n6-arene)chromium(I) was collected, filtered,

washed with water, then ether, and dried in a vacuum

at room temperature. The purity of the compounds was

verified by IR, UV-visible spectra and microanalyses.
The IR bands in the 300-500 cm'1 region are char-

acteristic of diarene sandwich compounds.66 The func-

tional groups, -CH3, -Ph, -H, -OCH3, -COOC2H5, and -Cl

were confirmed by IR. Among these Cr(I) complexes,

(C H OCH CrBPh is easily oxidized and had to be
6 4

5 3’2
stored in an argon-filled dry-box.

All the solvents used for ESR or cyclic voltammetry
measurements were purified by standard techniques.67
Tetrabutylammonium hexafluorophosphate (BuuNPFé)

was prepared by mixing tetrabutylammonium hydroxide and
ammonium hexafluorophosphate in acetone, followed by
precipitation by the addition of water and the precipitate

was recrystallized 3 times from 95% ethanol. Tetrabutyl-

7

74

ammomium perchlorate (BuuNClOu) was obtained from G. F.
Smith Chemical Co. and recrystallized three times from

95% ethanol.

ESR Kinetics Measurements

ESR spectra were recorded by use of a Varian E-4 ESR
spectrophotometer with a variable temperature controller.
The exchange solutions were prepared in the argon-filled
dry-box by preparing a solution of the complex cation

0M) with varying concentrations of zerova-

(typically 10'
lent complex and a constant cation concentration. Samples
were added, in the dry-box, to a 3 mm ESR tube by means

of a glass syringe and the tubes were sealed on a

standard vacuum line. The rates of the rapid electron-
transfer reactions were determined from the increase in
line widths after the adding of known amounts of zero-
valent complexes. The slow exchange limits were used

for all cases in DMSO and other solvents. The calculation
of the rate constants were made according to the procedures

described elsewhere.LL7b

Cyclic Voltammetry

The standard reduction potentials of all the Cr(0)/
Cr(I) couples were measured by cyclic voltammetry with
a PAR 174 polarograph, coupled to a Hewlett-Packard

Model 7045A fast X-Y recorder.

75

Two-compartment glass cells (solution ca. 10ml)
were employed for the electrochemical measurements.
Contact of the solution with the reference electrode
was through a glass frit of very fine grade. The solu-
tion was deaerated by a stream of dry nitrogen before
the potential was scanned. The experiments were carried
out in DMSO (MIG-3M) at room temperature with 0.1M
tetrabutylammonium perchlorate as supporting electrolyte,
by using a platinum "flag" working electrode68 and a
saturated calomel reference electrode (SCE). Although
the separation of cathodic and anodic peaks (Table III)
was larger than the theoretical separation of 59 mV,
the unit ratio of the peak current iC/ia gives evidence
for reversible one-electron processes.

Oxidation potentials of the three couples (06H6)20r0/
(C6H6)20r+, (C6H5CH3)2Cr°/(06H5CH3)2Cr+, and [C6H3(CH3)3] 2-
Cro/[C6H3(CH3)3];Cr were measured in dichloromethane
(~10'3M solution) at room temperature with 0.1M tetra—
butylammonium hexafluorophosphate against SCE. The

scanning voltage was 100 mV/s and the current range 10011A.

Other Measurements

Infrared spectra were obtained by means of a Perkin-
Elmer 457 spectrophotometer in KBr pellets at room

temperature. Mass spectra were recorded by use of

76

Finnigan 4000 mass spectrometer (70 eV) with an Incos
data system.

NMR spectra were obtained by use of a Varian T-60
or a Bruker WM 250 spectrometer. UV-Visible spectra

were recorded by use of Cary Model 17 spectrophotometer.

ESR Spectra Simulation

The ESR spectra were Simulated by using the computer
program ESRSIM01The g values of the chromium electron,
the hyperfine coupling constants of the hydrogen and
chromium, and the line widths were adjusted to obtain
the best fit to the observed spectra.

Good agreement within experiment uncertainty
between the calculated concentration and the concen-

tration obtained from ESR spectra was obtained.

RESULTS AND DISCUSSION

ESR Measurement

Data indicating a monotonic decrease in the intensity
(or increase in the linewidth) of the ESR spectrum of
the bis(arene)chromium cation with the addition of the
respective zerovalent complex have been obtained. An
example (Figure 19) shows the effect on the spectra of

[(06H501)20r]BPh,+ (5.4 x 10"“5 in DMSO) of the addition

2 1

of (C6H Cl)20r to give concentration of O, 10' , 10' M

5
respectively, at 22°C.
The representative spectra of six solutions of

(C6H OCH Cro/(C6H OCH3)2Cr+ in DMSO at 22°C various

5 3)2 5
concentration of (CH3006H5)CrO were examined for broaden-
ing of the hyperfine components. It was observed that
the broadening of the hyperfine components was a linear
function of concentration, as shown in Figure20, and this
allowed the calculation of the exchange rate constant,
k = 2.0 x 108
Errors of 10-20% are introduced in these rate con-
stants from the concentration and line-width determi-
nation due to both the intrinsic lack of precision in
area measurements and the changing line shape at these
concentrations.
The activation energy was obtained from a plot
of log k vs. 1/T;e.g"(C6H

0CH3)20r°/ (C6H 0CH3)2C1~+ in

5 5

77

78

Figure 19.

ESR spectra of 5.4x10-u

M (06H5Cl)2Cr+ in DMS0(22°C)
(a) No added (06H501)2Cr°. (b) Computer simulation.

(o) Added 4x10-2M (C6H5Cl)20r°. (d) Computer simulation.
(e) Added 8x10-2M (06H5Cl)20r°. (f) Computer simulation.
(g) Added 1.5x10-1M (06H5Cl)2CrO. (h) Computer simulation.

(i) Added 2x10-1M (06H5Cl)20r°. (j) Computer simulation.

 

 

 

(0.)

 

 

 

 

 

 

 

 

 

 

(b)

 

 

 

 

X\/\ U
. 1

 

8 :2

(g)

IOG

(h.)

 

(i.)

83

0 D _ l‘.
..H

84

 

2.5

200 ’

1.5

A pr

(guass)

1.0 -

0.5 .

 

 

 

0 ‘ . . 1 1. .
0 .05 .10 .15 .20

Concentration of (06H501)20r+, M

Fig.2N) Broadening Kg. concentration of bis(chlorobenzene)-

chromium(0) in DMSO at 22°C.

85

 

owns 5 53600.. 000053 53030 {ououanmoonmcov non +34. 3 08 3.03

n3 2+

mun

...n

n.n .n

.— mod

 

 

86

DMSO, (Figure 21) and is 3.6 I 0.4 Kcal/mole. Activation
parameters were evaluated from equation 3.

KT AS*_ 611*)

k: E exp (T RT .0.......0(3)

 

Results of self-exchange rate measurements and activation
parameters are collected in Table S9-
The effects of SOIVent variation on the exchange

rate of bis(biphenyl)chromium°/+ were listed on TablelCh

Redox Potentials

One electron reduction potentials of bis(né-arene)-
chromium cation were measured in DMSO by cyclic voltamme-
try. Typical cyclic voltammograms are shown in Figure 8.

The separation of the peak potential and the
height of peaks is close to unity. (Tablell)

All the half-wave potentials, (E%) ArZCro/ArZCr+,
for the complexes are given in Table 11.

The first electrochemical investigation on bisarene-
chromium reveals an excellent correlation between the
measured E% values and the sum of Hammett parameter,

am as discussed in Part I.

Discussion

5

In terms of the Marcus theory,5 the rate constant

 

87

comm as .m

 

 

m.m H m.NH- 4.6 H m.m m.o H m.H H.N H 0.0m Hommoo c
cum H mama- mac H std m.o H H.~ «.6 H m.HH memoooommoo m
m.m H :.mH- o.o H o.m :.o H o.m H.m H m.mm mmoommoo :
m.m H m.mH- o.o H o.m :.o H o.m om.o H mm.s mmoommoo m
:.N H «.mH- m.o H :.m m.o H 0.: ms.o H mm.m omoo m
o.m H 6.00- m.o H 6.: 6.6 H 6.: 04.6 H mm.m mmommoo H

Asov4m< Aoaoeiwmoxvam< AoaoE\H60Mvmm now a OHM: 98.3 02

6H .0. s.

 

+Homn<\ohomp< Bowman «3 whoposmHmm :oHpm>Hpom ER mesmpmsoo opmm I m 35.69

OWSQ C...” COfiPOMQH .HwhumHMMQu. COHPomHG

88

mpMConhmommCmthonm .H

 

 

o0HsmepoHHsspochuz .z .o o0onHHsMHssposH0 .0 oHprHcoNcon .o
pammeoo oHHpomHmHu OHPMHm 0:0 aonpmo a 0:0 N: .n 0000 .0
0.H u
0Hm.0 0.0 H 0.0H1 0.0 H 0.0 0H.0 H H0.H Hocmspoeuoccaeom
000.0 0.0 H 0.0H- 0.0 H 0.0 Hm.0 H 00.H 000
01H u
H00.0 0.0 H 0.0H- 0.0 H H.m 00.0 H 00.H H000000210000000
000.0 0.0 H 0.0H- 0.0 H 0.0 00.0 H 0H.0 0020
000.0 0.0 H 0.0H- 0.0 H 0.0 H0.0 H 00.0 00020
00m.0 0.0 H 0.0H- 0.0 H H.0 00.0 H 00.0 0200
Hum n
00.0 H.0 H.0.0H- 0.0 H 0.H 00.0 H 00.0 Hocmgpoe.mcouccm
ma 0: I
IMI 1 IA! Azov0m<. AmHoE\Hmon0m< 0HuommH12muonx pam>aom
n

 

.+000A00an\o000A00H00 .200000 0:0 0H

mHmpoEMHmm :oHpm>Hpom 0:0 macawmzoo mHMH so mpomHHm pam>aom .. 0H manme

89

.mQCmSPHPmnsm HCmHmHHHU How mumpmsmpma

ppmeemm Ho 5:0 0:9 .mm>03 AMHV OHUocm 0:0 AOHV 0Hvo£pwo 059 How magmpnso xmmm

6
0:9 Ho 0prmo .xmmm oHconpmo 0:0 0Hcoc0 CH mocmHmHHHo map mH amen .HmeCmpoQ

xmmm 0H00£pmo 0:0 oHvocm man we mwmumwm 0:0 mm cmPMHSoamo 0H0; mmsam> mmm

 

 

 

00.0- 00.H 00 000.0 0000000A000000v
00.0- 00.H H0 000.0 00000000000000000000
0H.0- 00.H 00 000.0 H000000000003
00.0- 00.H 00 000.0 00000000000000000
0 00.H 00 000.0 H600A0000v
0H.0+ 00.H 00 0H0.0 0000000A00000000
0 0 0 0 0 0
E 00 Q A m .m> >v
ow o. A>Ev m< mo N\H ccsomEoo
H. ml
0 o . n 0

 

.0o00 00 00Hoz0sm a H.0\0020 :H mmonQEoo Atho Ho zuposempHo> oHH000 -HH 0H000

90

for an outer sphere electron-transfer reaction is

_AC*/RT
k : Ze 0000000000000000000(5)

*
where AG is the free energy of activation and Z is

11

collision number of about 10 1/mole°sec. The free

energy of activation is given by

O

F
*
AG = wr + (1 + 91§-)2/4 .........(6)
F; donotes F0 + Wp - Wr, AF0 is the standard free

energy change for the reaction, Wr (or WP) is the work

required to bring the reactants (or products) together.

A = A + A. ..................(7)

where Ac andlki are the outer and inner sphere reorgani-
zation contribution to)».
1 )( 1

_ 2 1 1 _
1 o - (Ae )(Zal + 2a r

1
2 n2 Ds

where Ae is the charge transferred, a1 and a2 are the
radii of the two reactants, r, the distance between the
two centers of the reactants in the activated complex;

n and DS are the refractive index and static dielectric

91

constant respectively.

For the type of electron-transfer reactions we
have studied, the standard free energies of reaction
are zero and the work terms are small and can be ne-
glected, since one of the reactants is neutral. If
it is also assumed that a1 = a2, r = a1 + a2, a simple

relation given eq. 6 was derived as follows,

A ' 1 1
AG-T-‘E+T’T*RET(?”—D;)H(9)

*
-A -
and k=ZeXP(Rg )=ZeXp (fi)00000000(10)

 

OI‘ l0gk=10gZ--‘§—;'R'T—' 00000000000000000000(11)

A plot of log k Kg. (1/n2 - 1/DS) should be linear
with lepe -Ae2/18.42RTa1 and intercept log Z - 11/9.2RT.

The plot of log k Kg. (1/n2 - 1/DS), the electron
exchange for bis(biphenyl)chromium(O)/bis(biphenyl)-
chromium(I) in seven different solvents shows the expected
linearity. From the slope, a1 was calculated to be 6.4AO.
It means the separation of cation and parent in the
activated complex is about 12.8Ao. The head-head and

72
side-side distance of Cr-Cr+ in Cr(biph)2 are 8.0AO

73! 28-32

and 11.8A° respectively. From the distance

point of view, it seems that the more reasonable agree-

92

ment suggesting a "side-by-side" positioning of chromarene
derivatives in the transition state rather than a "sitting
on top" arrangement. This proposal for a "side-by-side"
mechanism also support in kinetic studies of ferrocenes

44a Elschenbroich and Zenneck43

and carbollyl derivatives.
have investigated the electron exchange in chromarene
systems, postulated a "sitting on top" mechanism because
there is no significant influence of the degree of

methylation on the rate of electron exchange.

 

Their suggestion is based on the calculated rate

1H NMR line broadening data, by using

constants from
the method of Johnson7LL under certain conditions, which
depend on the hyperfine splitting constants for two

of the equivalent sets of nuclei differing significantly,
and the lines not exhibiting the same limiting case

of pulse height. The error in this calculation is

great if both lines approach the weak limit, meanwhile,
it is difficult to find an ideal concentration to suit
both limits in the same system.

The kinetics data for the six "self-exchange" rates

for the type of reaction are given by reaction 12.

Cr(06H5x)Z+ Cr(C6H5X)2+—0 Cr(C6HSX)2++ Cr(cé'st)2 ..(12)

where X = -CH3, -H, -OCH3, -C6H5, -COOC2 5, -Cl.

93

The solvent dependence of the rate constants for
(C6H5'C6H5) Cr(I)((O) self exchange was found to be
in reasonable agreement with the predictions of the
dielectric continuum model. From the intercept, i/4
was calculated to be 0.6 kcal mole-1. This result
implies only a small inner-sphere reorganization con-
tribution to the total free energy of activation from
the core sandwich structural rearrangement in the
transition state.

If an outer-sphere mechanism is assumed, the

cross reaction is given by reaction 13.

+ Cr ZI-L— Cr + Cr (13)

The rate constant kij can be expressed as

1
2

kij = (kii kjj K11 fij

V

................(14)

where kii and k. are the self-exchange rate constants

33'
for each chromarene(Ar2Cr) and its oxidized cation(Ar20r+)

corresponding to the process Shown in reaction 12,

94

K is the one electron equilibrium constant defined

1.1
by the relevant AEO value for the cross reaction,

reaction 13, and where fi. is given by the expression

J
log f .=(log K .)2/4 log (k k /ZZ) (15)
13 13 ii jj ......
. . . 11 -1 -1
Z 18 the COlllSlOn frequency (10 M sec ).

Good estimates of K1 can be obtained fromA E%

J
values of various reactant couples and the assumption
of the validity of eq. 14. The cross-reaction rate
constants kij can be calculated by using the five
self-exchange rate constants(Table 9) together with the
known equilibrium constants(Table 12).

The effect of solvent and substituents on rates
are demonstrated in plots of TASt against AH*. These
effects lead to the result that they will generally give
rise to a fairly exact compensation between AH*

75

and TAS*. This compensation effect is found and the
plots of AH* vs. TAS* are good straight lines (Figure
22).

Replacement of electron-withdrawing group substi-
tuents in the arene ring may lead to a higher positive
charge building up the metal and retard the one-
electron oxidation to Ar2Cr+. If the changes in the
lowest ionization potentials for neutral sandwich

complexes of chromium are related to a change in

electron density about the Cr(O) center, it is reason-

 

95

Table 12-Cross-reaction constants kij for electron

transfer reaction

 

 

 

Reaction 11tx10-8‘1_VI_"1sec-1 Reaction chIO-8y1"]'sec'1
11 0.33 34 6.21
12 1.49 35 28.8
13 2.34 41 . 0.026
14 15.6 42 0.14
15 59.8 43 0.24
21 0.12 44 2.33
22 0.60 45 13-3
23 0.97 51 0.003
24 7.45 52 0.018
25 29.2 53 0.036
31 0.093 54 0.43
32 0.48 55 3.15

33 0-77

 

96

 

4.5

O Substituent

 

A Solvent

AH
( kcal 4

mole
2e5 ’

/
/

 

 

 

4.3 ‘ A 4.0 A A 3.5 A 3.0
-ms‘ (Keel/mole)

Fig. 22 The coupon-“ion effect: plots of 068‘ against AH‘
for (a) change: of substituents on electron exchange
reaction (1)) changes of eoiventc on electron exchange

reaction

97

zable to assume that the same variation in electron
ciensity should be reflected in E% for oxidation of
Ineutral chromium sandwich complexes (Figure 23). A plot
of E% of Cr(I/O) correlates well with the gas phase
ionization potentials of chromium aromatic sandwich

_ complexes. The trend of the air stability of the chromium
compounds is 17e-> 169-> 18e‘. The E% values and
ionization potential values for sandwich complexes are
collected in Table 13.

The increase in electron density on the central
metal is due to the electron donating methyl substitu-
tion which causes a general decrease in the ionization
energy also observed in the M076, Nb77, and Fe78 com-
plexes, whereas in Fe(Cle)2 the electron-withdrawing
group produces the opposite effect. An electron with-
drawing group has a kinetic stabilizing effect: CF3 and
F are the best substituents.63 Hao and McGlinchey79

1H and 130 NMR chemical shifts for a series

showed that
of unsymmetrical bis016-arene)chromium derivatives

could be correlated with mum and the oxidative stability
of the complexes. We have tried to correlate for this
substituent effect on the air stability of sandwich
complexes, by using the sum of the Hammett Oh parameters
of substituents Kg. gas-phase ionization potentials:

straight lines are found (FigureZflL). The ionization

Potentials are listed in Table 14.

98

 

 

 

 

 

'1'1 L O Cr(mes)2
-1.0 ’
o Cr(tol)2
Cr(ba)
.0e9 b O 2
«0.8 ,
o Cr(Cp)(Cht)
-0.7 . ° 0"”)2
10!)
2
-0e6 b
“cos b
.0.“ ’
'°'3 * Cr(Cprz) 0
-0.2 ‘ ‘ ‘ ** ‘ ‘ ' L ‘ ‘
5‘0 6.0 7eo
IP(eV)
Fig.23 Plot of 31/2 for the oxidation of chronim sandwich

compound: to. their lowest ionization potentials (11)).
has benzene. tol: toluene. nee: neaitylene ,
Ops cyclopentadiemrl. Cht: cycloheptatrienyl

99

Table 13 Electrochemical parameters and ionization

potentials for the chromium sandwich compounds

 

 

Compounde No.of e- -E%(V)a IP (eV)d
Cr(mes)2 180' 0.982b 5.01
Cr(tol)2 180’ 0.867b 5.24
Cr(bz)2 180‘ 0.813b 5.40
Cr(cp)(cht) 18e' 0.650 5.59
Cr(cp)2 16e‘ 0.61c 5.7
Cr(bz)(cp) 17e' 0.19c 6.20

 

a. Measure in CHZClZ/O.1 BuuNPF6 at Pt electrode of
cyclic voltammetry at 25°C vs. SCE.

b. This work.

c. U. Koelle, Z. Naturforsch., 1979, 34b, 759.

d. S. Evans, M. L. H. Green, B. Jewitt, G. H. King and
A. F. Orchard, J. Chem. Soc. Faraday Trans.II.,
1974. 29. 356-

e. bz: benzene, tol: toluene, mes: mesitylene, cp:

cyclopentadienyl, cht: cycloheptatrienyl

100

Table 14 - Summary of ionization potentialsa and

tom of metal sandwich compounds

 

Compound Ip(eV) - £0m Compound Ip(eV) to

 

Cr(mes)2 5.01 0.42 Nb(mes)2 5.18b 0.42

 

Cr(tol)2 5.24 0.14 Nb(tol)2 5.49b 0.14
Cr(bz)2 5.40 0 Nb(bz)2 5.57b 0
Mo(mes)2 5.13 0.42 (CHBCp)2Fe 6.88 0.14
Mo(tol)2 5.32 0.14 (Cp)2Fe 6.72 O
Mo(bz)2 5.52 O (Cle)2Fe 7.03 - 0.74

 

a. Ref. 78
b. Ref. 76

50 (Figure 25) is given

The generalized Mo scheme
for 17e' and 18e- complexes such (bz)2Cr+. J. C. Green
and M. L. H. Green77'8O assigned the photoelectron
spectra of such (ré-arene)2M (M = Ti, Nb, Cr, Mo, and

W) species in terms of the M0 model. The a orbital

1%
is essentially non—bonding and explains why it may

contain 0, 1 and 2 electrons with little effect on

thermal stability.

101

 

O ArZCr

[3 Ar Nb
0.4 . D
on
\o\

. CpZFe
D C

20

-0.2 ,

 

 

 

5.0 6.0 7.0
Ip(eV)
Fig. 24 Correlation of the au- of Hanett pare-eter, 20 h,

with ionization potential of (arene)! or ferrocenes
o ArZCr A Arzlo CI Arzfl'b O Opzl’e

102

 

 

 

 

\
. \\.\ x \
s . \ \
e \\ \ x,
s \
x \ \ \
ss \ \ \ \
s \ \ \ \
s \ \ \ \
\ \V \ \ \
\ \\ \ \ \
\ \ \ \
\ \ \ \
a \
‘ \ \ \ \

 

 

 

 

 

 

llil

[Eli

 

1141

[VIP

 

 

 

 

 

 

 

 

 

 

 

 

.....u 1..
IV 4....
4' a...“
.‘\ \
\\ \\ \\\\\ x
\ \\ \\\. \
\\ \
\\\ \
\\\ .\
\\\\ \
\\\ \
\ \ \
a' it
' a'
A.

 

+
I

6

.313. 25 Generalized 11.0. scheme 0: (n -arene)zl

=Cr

...

-arene)zl . M

and (06

103

The a (metal dz2 orbital) ionization energies

1%
increase for the compounds (mes)2M, (tol)2M and (bz)2M

(M = Cr, Mo, Nb). The plots of 20m vs. IP are parallel
lines similar to those of substituted ferrocene compounds
and reflects the chemical similarity (Figure 24 ).

In conclusion, it appears that the electron exchange
reactions proceed by the outer sphere mechanism and
probably by the side to side transition state. Sub-
stituent effects and solvent effects on the reactions
correlate well with the Marcus theory and supported
by electrochemical results, ionization potentials,
and Hammett ‘Jm parameters point to electron densities

about chromium being affected by inductive effects and

account for variations in exchange rates .

 

PART III

ELECTRON EXCHANGE BETWEEN BI8016-ARENE)CHROMIUM(I) AND
BIS(n6-ARENE)CHROMIUM(0). COMPARISONS BETWEEN EXPERIMENTAL
AND CALCULATED KINETICS PARAMETERS.

SUMMARY

Comparisons have been made between kinetics para-
meters for the self exchange of O16-arene)20r(I)/(O)
(where arene = benzene, toluene, methoxybenzene , biphenyl,
ethylbenzoate and chlorobenzene) measured by ESR line
broadening in dimethylsulfoxide, with the predictions
from contemporary electron-transfer theory. The biphenyl
system was additionally studied in a number of other
solvents. These reactions provide especially tractable
systems with which to test theories of outer-sphere
electron transfer since the work terms should be essenti-
ally zero, and the inner-shell contributionsAG;n to
the free energy barrier are small and can be estimated
from infra-red spectroscopy combined with crystallo-
graphic data. The solvent dependence of the rate con-
stants for (C6H5-C6H5)2Cr(I)/(O) self exchange was found
to be in reasonable agreement with the predictions of
the dielectric continuum model. Frequency factors were
derived from the experimental rate constants coupled with

the estimates of AGIn and values of the outer-shell

104

 

105

contribution AG:ut obtained by using the dielectric con-
tinuum model. These were found to be somewhat (ca. 5-20
fold) larger than the corresponding frequency factors
derived from the experimental activation enthalpies
combined with the dielectric continuum estimates of the
activation entropies. These "experimental" frequency
factors are compared with the estimates obtained from

the "reactive collision" and the "ion-pair pre-equilibrium"
model. The majority of the experimental frequency factors

were found to be numerically closer to the predictions of

 

the former model. However, the experimental values were
found to decrease substantially with the addition of
substituents on the arene ring, indicating the importance
of steric effects. It is suggested that the discrepancies
between the experimental results and the predictions of
the pre-equilibrium model observed for these and other
systems may be due to a combination of steric and non-

adiabatic effects.

INTRODUCTION

Outer-sphere electron exchange reactions constitute
an especially interesting class of chemical reactions in
solution in that it is anticipated that quantitative
theoretical descriptions of the reaction dynamics can be
provided in many cases. Theoretical treatments of vary-
ing levels of complexity (and usefulness) have been
developed that allow detailed insight into the physical

processes involved.81

Some recent treatments outline a
useful framework within which various aspects of the
theories may be tested by comparison with experimental
data.82-84
One question that has arisen is the choice of the

frequency factor for outer-sphere reactions.82’83 Two

alternative models have been proposed.82 The "Reactive
collision" model considers that electron transfer occurs
upon collision between appropriately energetic reactants,

leading to the formulation
k = Krn Z[exp(-w/RT)exp(-AG*/RT)]......(16)

where k is the observed (second-order) rate constant
for electron exchange,K is the electronic transmission
coefficient, Pn is a nuclear tunneling factor, Z is the
collision frequency, w is the work required to form

the collision complex from the separated reactants,

106

107

and 40* is the free energy of activation for the elementary
electron-transfer step.82’83 The alternative "pre-equili-
brium" model considers that reaction occurs by activation
within a previously formed bimolecular assembly.82’83

This leads to the expression

k = KFnKSVp [exp(-w/RT)exp(-AG*/RT)]....(17)

where K; is the equilibrium constant for formation of _‘

the "precursor complex" in the absence of work terms

(i.e. when w = 0), up is the frequency factor for acti- »-

 

vation within the precursor complex, and the other terms
have the same significance as in Eqn. (16). These two
formulations therefore differ in the overall preexponential
factor in that the "classical" frequency factor A equals
the collision frequency Z in Eqn. (16), but is replaced
by the composite term ngp in Eqn. (17). Although the
collision formulation is most commonly employed for
bimolecular solution reactions including electron transfer
processes, the pre-equilibrium model may provide a more
appropriate description of outer-sphere electron transfer
when the reaction is expected to take place via electron
tunneling between weakly interacting species.

Comparisons between the predictions of Eqns. (16)
and (17) with experimental data can be made by obtaining
estimates of the frequency factor A in two ways. First,

the observed rate constant at a given temperature can

 

108

be combined with theoretical values of AG*, w,n:, and Pn
to yield estimates of A [method (a)]. Second, the
Arrhenius activation energy, Ea’ derived from the observed
temperature dependence of k can be corrected for the
temperature dependences of’K, Pn, and A to yeild an
"experimental value" of the enthalpic component of AG*,
AH*. Values of A can then be found by combining this
value of AH* and a value of k at a given temperature
with theoretical estimates of AS*, w,:<, and Tn [Method
(bu . Since both methods depend critically upon the
assumed values of AG* or AS*, w, K, and Tn, it is clearly
important to select systems for which these parameters
can be estimated with confidence. Suitable redox couples
will therefore be those for which the structural differ-
ences between the oxidized and the reduced form are
known, and preferably small, so that the inner-shell
contribution to AG* can be calculated and.Fn will be
close to unity.8LIL Additionally, it is desirable that the
reactants be spherical, or nearly so, and the product
of the reactant charges be small, so that the work term,
w, is small and the outer-shell (solvent) contribution
to (fl* is most likely to conform to the prediction of
the conventional dielectric continuum model.

Several tests of Eqns. (16) and (17) have recently

82

been made. Brown and Sutin studied the self exchange

of Ru(III)/(II) couples containing polypyridine and/or

 

109

ammine ligandsin aqueous solution. Better agreement
between the experimental rate constants and activation
parameters was obtained by using the reactive collision
rather than the preequilibrium formulation. Similar
findings were reported by Meyer 23 a; in a study of
electron exchange of [Ru30(CH3C02)6(py)jr/° in dichlo-
romethane.85 Systems of this latter type, where one
form of the redox couple is uncharged, have the important I.
advantage that the electrostatic work required to form

the encounter complex should be essentially zero. _‘

 

We have recently reported rate constants and
activation parameters for the self exchange of various
bis(n6-arene)Cr+/° couples, Cr(C6H5X)2+/°, where X =
-H, -CH3 -0CH

-06H -C00C2H5, or -Cl, in dimethylsul-

3’ 5'
foxide (DMSO), by using the ESR line-broadening techniquéé
The self exchange of Cr(CéHS-C61-15)2+/O was additionally
studied in six other solvents. These reactants provide
especially suitable systems with which to test models

of outer-sphere electron transfer. Besides the likely
absence of electrostatic work terms, they have the
advantage that the inner-shell components AGIn are

small and can be calculated from appropriate structural

data. The comparison of rate constants and activation

parameters for such a series of redox couples with

 

various ring substituents can therefore provide a

means of exploring the influence of reactant asymmetry

110

on the frequency factors for electron exchange. Such
comparisons between the experimental rate parameters
and the predictions of Eqns. 16 and 17 are given in

the present report.

Experimental Section

The synthesis of all bis(n6-arene)chromium com-
pounds were described in Ref. 15. Infrared spectra
were obtained by means of a Perkin-Elmer 457 spectro-
meter, by using either KBr pellets or Nujol mulls at
room temperature. The reported IR frequencies are
accurate to 10.5 cm'l. The Nujol mulls of the air
sensitive Cr(O) complexes were prepared in an argon-
filled dry box and the spectra were measured in an
argon atmosphere. Examples of IR spectra for (arene)2Cro

and (arene)20r+ are shown in Figure 30 and 31.(See

Appendix)

 

RESULTS AND DISCUSSION

Reorganization Energies

The contribution to AG* arising from inner-shell

reorganization, AG;n(T), for each self-exchange reaction

 

was calculated by means of the expression86
2
002nm) ... :foffia 1‘1); ......(18)
0Y1 1V0

0 and f1 are the force constants for the

metal-arene bonds in the Cr0 and CrI oxidation states,

In Eqn. (18), f

Aa is the difference in the corresponding equilibrium
bond distances, n is the number of metal-ligand bonds
involved and yi = hvicoth(hui/4kBT), wherevi are the
observed Cr-arene stretching frequencies for the oxid-
ation state i. [Equation (18) takes into account the
nuclear tunneling factor Tn at a given temperature;

thus Ang (T) is related to the classical inner-shell
reorganization energyliGIn by AG:n(T) = AG:n - RTlnI'n.84
However, Pn is close to unity (1.0 - 1.2) for the present
systems.] The normal vibrations of Cr(C6H6)2 and Cr(CBHé)+
have been assigned to be of D6h symmetry from the

87

infrared spectra, so that n can be set equal to two.
The force constants f0 and f1 in Eqn. (18) refer to
the symmetrical stretching vibration of the two arene

rings with respect to the chromium atom. The frequencies

111

 

112

of these symmetrical vibrations, v0 and\)1, have been

determined for the Cr(C6H6)2+/° couple to be about 270cm"1

87c allowing the force constants

2 .
vm where m 18 the
y’ y

and 279cm-1, respectively,
to be determined from f = 5.89x10-
mass of one arene ring. Values of V0 and V1 for the
other arenes studied here are unavailable. However,
they are unlikely to be greatly different then for

Cr(C6H6)2+/O; thus the asymmetrical X-Cr-Y stretching

frequencies v: andrv? are within 5-10 cm"1 for all the

Cr(I)- and Cr(O) arenes (Table 15). The values of f0

and f1 were therefore estimated assuming that v =

o
1 and v1= 279 cm-1.
. X +/o
The value of Aa 18 0.07 for the Cr(06H6)2

couple as found from X-ray crystallographic data.88 For

270 cm-

the other Cr(I)/(0) couples, estimates were obtained

from the corresponding difference in infrared stretching

frequencies by assuming that Aa a(f% - f?), where f:

and f? are the force constants for the asymmetric vibr-

ations obtained from \% and‘v? using the formula fa =
2(va)2

5.89 x 10 mCrmy/(mCr + Zmy) , where mCr 18 the

87a,c

mass of the chromium atom. The required proportion-

ality constant was determined from the experimental data
for Cr(C6H6)2+/°. Justification for this procedure is

91

available for metallocenes. The estimates of AG;n(T)

resulting from inserting the appropriate values of f0,

 

 

I UGZCHPCOO I m
e e m g
.90 mm M E Q

.ooH mocmnmhmm Q .mHCMpmcoo ownpooamwc 0o9 mmousom .Amav .cvm wcfims omwmasoamo

O

.Amav .cvm wcwmz ompmasoamo n

.mamcos amazomaos

800m 009080900 mm A+po|ouov momumwu awoaoscpmvcfl :mmelznumon= mmmnm>m 0:9 0

 

113

oado

 

moOI meOI 000' NoHI 0.0 I Oafil mAHIHOElemUonOV *m<
0.: 0.0 0.0 0.0 0.0 0.0 000-000.00000 000m00
0.0 0.0 H.m 0.0 m.0 0.0 00000.00000 pem0<
00.0 00.0 00.0 00.0 00.0 00.0 0A0-000.00000 Aevcmua
. . A 80v 0? .H?

000.000 000.000 000.000 000.000 000 000 000 000 0- 0 0
00.0 00.0 00.0 00.0 00.0 00.0 A0000
0.0 0.0 0.00 0.0 0.0 0.0 , 00000

000000 00000000000 0000000 00000000 0000000 0000 00000

 

.0000 00 0000 00 00000\AH000A0000<-000000 00 00000000

IHHmm 03p 00% mumpoSMMmm umpmawm 0cm mmwmhmcm :owpmuwcmmhomm oopMHSono ma manna

114

 

.on m9oC9oow :9 Cm>9w mmohsom Eop9

950

00090990 90\mac 0:0 90\momc 9o mosam> mcfim: AHNV .cvm 8000 0090930900 ma
.2.

m<
0 0 0 0 -00
«HAN >_+ N >0 \ m > m 9mg I 9 0009

.Ano9mp9m 0909oeehmv .CO9908 Hamsm meCa 90 augmsvmpw mwmum>m 0:9 09 :09 000:3

...... .0..- 0 .....I. 1'01- . 00..
.90 $0 :0 ...00 2.; ...000
.300?
CH .950 CH
90 o>99m>9000 00390009809 0:9 800% £0009 m9 Aev .mAV *mc 0cm ABV mmq mo saw 0:9 0
.3.
.00..”00 000 A000M0<00 000 000 0

- 000000000 - 00 00000

115

f1 and Aa into Eqn.(18) are given in Table 15. It is
seen that the values of AG;n(T) are mostly below 1-1.5

kcal-mol-1

, indicating that inner-shell reorganization
provides only a small contribution to the electron-
transfer barrier. This is expected since the orbital
into which the electron is transferred, a1g having dz2
symmetry, is essentially nonbondingsoso that it may con-
tain one or two electrons [as in Cr(I) and Cr(O), res-
pectively] with little change in the metal-arene bonding.
The other contribution to AG* is provided by the

"outer—shell" reorganization energy of the surrounding

solvent Aqut. This component was calculated by using

the dielectric continuum expression91
AG* = _EE (.l. + _l_ _ l_)(_l_._ l.» (19)
out 2a1 2a2 r DOP DS
where Dop and D8 are the optical and static dielectric

constants. respectively, a1 and a2 are the radii of the
two (spherical) reactants, r is the distance between
the two reacting centers in the activated state, and e
is the electronic charge. For the reactions considered

here, it is reasonable to assume that approximately

r = 2a1 = 2a2 so that Eqn. (19) reduces to
AG* = _§E_ (_l_ _ _l_) (20)
out r DOp DS

The values of r for each reaction were calculated by

assuming that the transition state is formed with the

 

116

two reactants placed "side-by-side" (i.e. with the arene
rings lying in the same plane),15 by using the reported
bond distances and van der Waals radii.9o’92 These
estimates of r and the resulting values of AGgut obtained

from Eqn. (19) by using known values of Do and DS are

P
given in Table 15 (see footnotes for data sources). Also
listed in Table 15 are the calculated values of AG*,
00:81 o” obtained for each reaction from the sum of the
corresponding values ofNAng and AGZut.
In addition to the free energies of reorganization,
it is also necessary to calculate values of the enthalpic
and entropic components of AG*, AH* and AS*, respectively.
The outer-shell component of AS*, Asout' was determined

from the temperature derivative of Aqut’

‘1-
AS* = _ (E58222) = 92 _l_.3222_1.lL.EB§) (21)
out dT r D2 dT D2 dT "
op s

The inner-shell component of AS*. AS , was calculated

*-
in
from the temperature derivative of AG;n(T) [Eqn.(18)],

thereby taking into account the (small) temper

84 . *
of 1h. The resulting values of Asin (-O.3 to -O.6 e.u.)

were then summed with the corresponding estimates of

As*
out
'1'

calc

to yield the calculated activation entropies

AS listed in Table 15.

 

 

 

 

117

Comparison of Experimental and Theoretical Kinetics
Parameters

As noted in the Introduction, tests of the appli-
cability of Eqns.(15) or (16) to describe the experimental
kinetics parameters can be made either by combining the
experimental rate constant at a given temperature with
a theoretical estimate of AG*[ method (an , or by employ-
ing both the rate constant and its temperature derivative
along with a theoretical estimate of AS*[method (bfl .

The resulting "experimental" frequency factors can be
compared with the values of Z and K;\b calculated by
using various models. Alternatively, the calculated
frequency factors can be combined with the observed

rate constant and activation parameters to yield "experi-
mental" values of AG* and AS* which can be compared with
the corresponding theoretical estimates. Clearly, the
reliability with which a given model for calculating
frequency factors can be tested depends sensitively on
the accuracy with which AG* or AS* can be calculated,

and zigg'zggsa. However, the experimental enthalpies

of activation AH* obtained from the temperature dependence
of the rate constants are insensitive to the particular
model employed for the frequency factor. Therefore

the comparison between the experimental and calculated
values of AH* provides a useful independent test of the

likely validity of such theoretical reorganization

 

 

118

parameters.

Table 16 contains a comparison of experimental and
calculated values offlAH* for the various Cr(I)/(O) self-
exchange reactions in DMSO. Two "experimental" values
of AH* are listed for each reaction. The first type,

labelled Ang, are equal to the Arrhenius activation

energy, Ea = R alnk/a(1/T); these values are consistent

with the preequilibrium formulation since the preexponen-

II—l

tial factor in this model[.<I‘nKo . Eqn.(17n is expected

pvp
to be essentially independent of temperature. The second

 

 

type, labelled AH;C, are uniformly 0.3 kcal°mol-1(= RT/2)
smaller than AHpe‘ these correspond to the use of the
reactive collision formulation since the collision
frequency Z appearing in Eqn.(16) is expected to be
proportiona1828to Tit. The calculated values of AH*,

labelled AH* in Table 16, were obtained from the

calc
*
corresponding values of Aczalc and Ascalc given in
. ‘I' _ * *
Table 15 by us1ngAHcalc - AGcalc + TAScalc'

It is seen that both AH;e andAH;c are uniformly
smaller than the corresponding values ofAH':alc by
amounts varying from ca. 0.5 - 2 kcal-mol'l. Since the
estimates of AHin are likely to be correct within at
least ca. 0.5 kcal'mol-l, it seems likely that these
discrepancies are due chiefly to theoretical estimates

of the outer-shell component £8qut that are somewhat

too large.

 

119

 

 

 

 

.mH 0H0.0.H. a“ :0>Hm 0H0mm<0c0 od0mo< mo m0za0> mcwmz

0H0mm<e + oa0mw< u 0H0mm0.souh 00:009no .Cow90>w9o0 ho m0wma0:9:0 0090H30H0o0

.00 0.0 -.00 u 0M00 0000 000000000

.zmh020 £0090>fl9o0 050:0:ph0 0:9 00 0m 090:3 .0m u 0m00 thm 00200990:

.ma0oc0u0h0m Scum :0x09 .9:09m:oo 090: H09C0Ewhmnxm0

0.0 0.0 0.0 00 H00000

0.0 0.0 m.m mm 00000000000

0.: 0.0 0.0 mm 0:00000

0.0 0.0 0.m 0.0 m0000000

0.0 4 m.0 0.0 0.0 m000000

0.0 0.m 0.0 0.0 0000

H..Hosé0ox H..Hosé0ox Huaoe.a0ox H.0007: mnoH
0 000w: 0 0&2 0 0.000 0 x 00000
0090H30H0MI. uuuuuu H09C0Ewh0nxm uuuuuu

 

.0000 p0 0020 :0 A0000\AH000A0000<-000000 00 000000xm-0a0m 0:» 000

:ofl90>w9o< we m0wma0:9:m 0090H30H0o 0:0 H09cmsflu0mxm Mo comfih0maoo 0H 0HQ0B

 

 

 

 

anion conc entration. M

l 20
3.0 *
Tm 2.5 [
m2- 200 l . KCLO4
I I
9 o
o '5 o KPF
EC I o L ..0 6
L0
.5 . - 4 L .
' O .l .2 .3 .4 .S

Fig.26 The salt effect on electron transfer rate constant

of (C6H5Cl)2Cr°/+.

 

121

Nonetheless, assuming for the moment that the
- *

8*
calc or A

theoretical reorganization parameters AG calc

given in Table 15 are correct, these quantities can be
combined with the experimental rate constants k given
in Table 1693 to yield "experimental" estimates of

the frequency factors Aa and Ab by using the relations

W
ll

*
Aa exp(-AGcalc/RT) ............(22a)

or
{-
cal

k Ab exp(-‘AH*/RT) exp(AS c/R)..(22b)

where AH* in Eqn.(22b) is obtained from the experimental
Arrhenius slope (most simply by assuming that Ab is

independent of temperature, i .e. AH* = A ng) . The

resulting values of Aa and Ab are listed in Table 17.

. Pe rc
along With the frequency factors Acalc and Acalc

calculated from the conventional forms of the preequi-

librium and reactive collision models given by Eqns.(23)

and (24), respectivelysaz’83

A2210: \fingxp(-w/RT)

' AGin % URN 3
= ( in l ) 366%— exp(-w/RT)............(23)
caC

rc = _
Acalc Z exp( w/RT)

= (N/103)[83kBT(m1+ m2)/m1m2]%r2-exp(-w/RT)..(24)

where N is Avogadro's number,n& andm2 are the massescfi‘the

 

 

 

122

two reactants. w is the electrostatic work required

to form the collision complex from the separated re-
actants, and Vin(s'1) is the effective frequency of

the inner-shell motion (metal-ligand vibration); this
was obtained from the symmetrical stretching frequencies
1 84

, respectively) using ‘v. =

v v '
0 and 1 (270 and 279 cm in

2 2 2 2 % . . .
C[2vov1/Tvo +'v1)] , where C 1s the veloc1ty of light.

Since one of the reactants is uncharged, it is assumed

In v.7

that w = 0.93

Inspection of Table 1? reveals that the "experimen-

“1’ JV"?
.. I

 

tal" frequency factors Aa are, in the majority of cases,

rc
calc’

although ca. 5-20 fold smaller than A2210. The frequency

close to (within a factor of two of).the values of A

factors Ab derived from the experimental rate constants
{-

calc [Eqn.(22b)]

and the activation energies along with AS

are seen to be significantly smaller, ca. 1/20 of the

 

corresponding values of Aa(Table 17). These latter
discrepancies result from the differences between the
experimental and calculated values of AH* (Table 16),

since the determination of Aa utilizes AGzalc’ whereas

the determination of Ab employs only the entropic

*-

calc [Eqns.(22a) and (22b)]- On the ba31s

component AS
of the present evidence, it is difficult to decide
whether Aa or Ab more closely approximates the true
frequency factors; clearly the choice depends upon

* . .
or AS are con31dered more reliable.

*
whether AG calc

calc

123

 

 

 

 

 

 

 

O I O 0 m0
oHoH N.: HHoH H 0 HHoH o m NHoH x m 0 Ho m o
. . . . 0 0 0 0
oHoH : N HHoH m : HHoH m m 0000 x H m m 0000 m o
. . . . 000
oHoH H m HHOH m : H0oH 0 0 MHoH x 0 H A 0 0V
. . . . 0 0 0
0000 0 0 0000 0 0 0000 0 0 0000 x 0 0 000 0 0
. . . . 0 0 0
HHoH m H HHOH 0 m HHoH H m NHoH x m 0 mu m o
. . . . 00
0000 000 0000 0 0 0000 0 0 0000000.0 0 0
0000 .000000 0000 .000000 000 .000000mm0 A00 .000000mm0 00000
00opo0m mpopo0m 00:030000 0090030H0o
zoC0zv00m =H0PC08000QxM:
.0000 00 0000 :0 000\0000000000<-0cvm00 00 0000;000:0000 000 000
AHno0mHuzv mHOpo0m 00:030000 =H0PC08000mxm= 0:0 0000030H0o ho £00000Q800 0H 00909

 

124

obtained from the observed rate constants by using fre-
quency factors calculated from the preequilibrium and
collision models, respectively [Eqns.(23) and (24)] .

Listed alongside are the theoretical estimates AGcalc

from Table 15. It is seen that the corresponding values

of AG; c:and AG* for most of the reactions agree

calc
closely (within 0.3 kcal-mol-l), parelleling agreement

between Aa and Ar , whereas the corresponding values I

calc

of AGpe are significantly (1.5-2 kcal-mol'l) larger. 1

Similar results have been obtained previously for

several other outer-sphere self-exchange reactions.82-85 [

 

Table 18 also contains estimates of activation entropies

A530 andAs;c by combining the observed rate constants
. . . . pe re
and activation energies along With Acalc and Acalc'

 

It is seen that the values of As;c are uniformly 3—6 e.u.

more negative than AS* , which follows since Ab < Aa ,

calc
Arc (Table 17). The values of A856 (- 9 to - 13 e. u. )

calc

seem unreasonably negative; a largeP increase in the
extent of solvent ordering in the transition state
compared to that for the separated reactants seems
unlikely for the present systems. Consequently, it

appears that the values of Ar listed in Table 17

calc
are too large.

Another method of estimating the frequency factor
involves monitoring the rate constant for a given

exchange reaction in several solvents. Since we can

 

125

 

 

 

 

 

.00 00000 0000 00000 .0000M00 000 00H00 00 000.0

.00 00000 00.0.0 00000 0m00 .e\00-000 - 0M0: u 0.000 00.0.0 000000000
.00 00000 0000 00000 0m00 .e\00 m00.- 0m00v u 0m00 0000 000000000

.00 0000.0 00.0.0 00000. 25000 0000 00 ””00 .00 0000
00000\000000- u 0000 0000 000000000

00000\000000- u 0m00 0000 000000000

0,0- 0.0- 0.00- 0.0 0.0 0.0 000000
0.0- 0.0- 0.00- 0.0 0.0 0.0 00000000000
0.0- 0.0- 0.00- 0.0 0.0 0.0 0000000
0.0- 0.0- 0.00- 0.0 0.0 0.0 00000000
0.0- 0.0- 0.0- 0.0 0.0 0.0 0000000
0.0! $.31 H.ml 0.0 o.m 3.0 wmmo

.0 oa0om< 0 00m< 0 0mm< ooa0wo< pl *3 0 0mo< 0:050
HIHoE.HIw00. H00 HIHoE.H0ox

.000moup:m +:oap0>0po< .000w00:m,00nm :owp0>MwO<

 

.0000 00 0000 00 00y\00000000000-000000 00 00000000-0000 000 000 00000>0000

mo 00wmo00:m 0:0 000w00:m 000m 0000H30H0o 0:0 :H0p:0EH00QxM: mo :o0000QEoo 00 00909

 

126

write
-A A* + 00* /
1". exp -( Gin out) RT o.ooooooooooooo.0(25)

if AGZut is given by the dielectric continuum model

[Eqn.(20)1- then

 

2
- * e ._£___l_.
log k - log A - AGin/2.303RT - 2.303x4rRT(DOp Ds)..(26)
Therefore, a plot of log k versus (l/DOp - l/DS) should ‘
‘

allow A to be obtained from the intercept, and r from

the slope. Such a plot is shown for the Cr(06H5-06H5)2+/°

 

self-exchange reaction in seven solvents in Figure 24.
The rate data are taken from Ref. 15, and were corrected

for the effects of diffusion in the conventional manner

1 -1 -1 .
app - kdiff’ where kapp is the measured

(apparent) rate constant, and kdiff = 8RT/3000n, where

by using k- = k

 

n is the solvent viscosity. (This correction turned out
to be small, yet significant in several solvents.) The
values of k are also listed in Table 19.

It is seen that the variation of k among most of
the solvents studied is approximately in accord with
Eqn.(15), suggesting that the dielectric continuum
model provides a reasonable description of the solvent
influence upon AG*. The straight line drawn through
the points has a slope of 3.9 and a y-intercept of

10.1. The resulting value of r, 15.5 g, is roughly

0
comparable to the value, 11.6 A (Table 15). deduced

zpwmoomfl>

 

 

: vmzcwpcoo I
.HN mozmpmmmm CH cm>flw mpmv

wcflmz .Apxmp mmmv Cowmsmmwc Mom ompomppoo #demcoo mpmn HMPCwEHmexmm

 

1237

 

 

H>\> “\flu
HHcH x m.m B.e- ~.oa- m.n m.n N.“ zon=u\o=ou
dag” x h.~ “Ho” x o.m m.c- h.n- . m.m - m.q o.n a.m o.H uuaaoauau oaoflxaoum
fi>\> e\HH
-- “flea x ”.0 e.¢- ~.c~- ¢.m h.m o.H =o~=u\o=ou
afiofi x c.~ HH¢~ x m.m m.~- «.m- m.HH- H.¢ m.~ H.m N.~ ouwamapomfixguoefia
ado“ x ~.n aficfi x m.» m.o- v.5- ~.mH- ~.v n.~ o.~ m.~ oufixostmflxguoawn
ego“ x 5.5 Hag” x ~.. ¢.H- H.o- m.HH- m.m ¢.~ h.~ H.v oflfluuwcoucom
H>\> H\mu
--- Hag“ x H.N v.0- o.-- o.~ m.~ v.0 zon=u\o=ou
on a
u 3&2 m uqu o .2 u 3592 u av: n was 7037:
.77: £de n< nag. .ccmv m< xxx: u:o>Hom
H 23.7...» Tag—Two“. Hao 7H9: Hmux m
nuouuum 525330.; ..Huucoawuoaxm: .muaobcm coflum>3u< £33935 scanning

 

.oomm pm mpcm>aom

macapm> CH on\AthoNAmmoovmfln Ho wmcmnoxMIMHmm mzp pom Aauomm Huav<

myopomm mesmsvmpm Hapsmswhmmxm Uzm mhmvmsmumm soapm>fipo< .mpcumcoo mpmm ma manna

128

 

 

n cmscfipCOO u

owppomamfln Mo wands: .Spfism .m.m .ppohpmz .<.< n meMPMCONCmm .Hm moCmpmwmm
I mufixOMHSmH>£PmSHU .HocMSPmE .mcmNSmn .mn mom mmopsom .Aommav mdda .MN .Emso

.m»:& .h .zme< .A .m .Hmnmsflm .Q Ca Cm>flm .mpMCopumO mamammopm mom pmmoxm .mm

mocmhmmmm Song cmcHMpno pzm>aom comm MOM 900 mo mmsam> .mpCm>Hom mpzm pom mmsam>

mo Coapmaomhmpcfl hamaH hp omcflMPno mmnszwe H0QM£P¢E\QON:mQ how we vcm moo

.mcHSOHHow Scum umcHMvnQ ”Anmv.cvma

oammchccmHAmmv.cva oamwu<_wcfipmasoamO pom cmnflAUmu apmc paupmcoo cappomamfimn

.3 myocpoow CH Cm>flm mmohsom 8099 mPCMmeOO owhpomamfiv wcwm: .mH manna CH mm cmcHMPnow

.ma manme op mmpocpoow

oamo

OH< UCM PCNPMCOO mummh HMPCmEflthNm EOHH UGCHMPQO

cw umpflpommc mm HAmv.cumH m

.wH wands Op mmpocpoog

CH omnwhomwc mm ”Amv.cum_ oamm< cam pampmcoo mpdp Hancosflnmmxm Eoph chHMpnom

.3 wponoom Ca Cm>fim mmoh30m 809M mPCMszoo owhpomamwc mafia: .QH manna CH mm chHMPno

o
.m oh N
.am m.o . m u *mc was» wcfiezmmm m 50pm omcfimpnoo

.mm u mmmq wasp wcHesmmm mm zmhmcm coapm>wwom HapCmsflhmmxm Scum umSHMPno

* n

. umzcfipcoo . 0H magma

129

 

 

.Ammmav mmmm .mN

.Emgo .mzsm .h .zopocomse .m .H .mcmmm .m I mpMCoDhmo mamazmohm .Adomav omm .flw
.Emso .m»:& .h .maoo .m .m .cmswflmz .2 .m .cmnpmz .H .3 .mmmm .h .m I mvwemsuom
Iazspmsfin .Aammav :Hm .ogwo .m.= .62Mpm .hsm .Hpmz :.mcHsuHA mhzm Mo mvcumCOO

I omscflpsoo I mH magma

 

130

from the reactant structure by assuming a side-by-side

configuration in the transition state. Assuming that

AGzn 1.3 kcal-mol-1(Table 15), a frequency factor A
of ca. 9 x 1010 M.1 sec-1 is obtained. Considering
the lengthy extrapolation of Figurezy necessary to

determine A, this value is in reasonable agreement with

rc . .
Aa’ Ab and also Acalc’ but again 1s markedly smaller
pe
than Acalc (Table 1?). r1.

Table 19 also contains values of Aa and Ab’ deter-

mined for the self exchange of (C6H5-06H5)20r(I)/(0)

 

in each solvent, which were obtained from the the ii
experimental data as in Table 17. As expected, the
values of Aa are largely independent of the solvent,
although the values of Ab do vary somewhat and are 1/3
- 1/15 of those for Aa' This variation of Ab with
the solvent most likely arises from a solvent depen-
dence of the outer-shell entropy term AS:ut which
is larger than calculated by using the dielectric con-
tinuum model. However, it is seen from the values of
AH;e and AH;c also listed in Table 19 that the major
part of the solvent dependence of k arises from vari-
ations in the enthalpic component, as predicted from
theory.

Since the collision model yields frequency factors

that are in closer agreement with the experimental

results than are the frequency factors obtained from

 

 

131

Figure 27— Plot of logarithm of rate constant k for
o/+ . .
self exchange of Cr(C6H5.CéH5)2 in various
solvents versus (l/Dop- l/DS), where DOp and
DS are the optical and static dielectric
constants for each solvent.
. CéHé/CHBOH (9/1 v/v);

. benzonitrile:

 

. dimethylsulfoxide;

1

2

3

4. dimethylformamide;

5. 06H6/CH30H (I/u v/v);
6. propylene carbonate;
7

. C6H6/CH30H (1/7 v/v).

values of k given in Table 19; sources for

 

DOp and DS given in footnote h to Table 19.

 

132

 

88

 

. log k,
M" sec."

8.6
8.4

82

 

 

 

l I -l‘ I l

‘ 03 Q4 0.5
woo, - VD.)

 

 

133

the simple preequilibrium model, it might be inferred
that the former model provides a more accurate descrip-
tion of the physical events leading to electron transfer;
i.e., that outer-sphere electron transfer is consummated
by collision between suitability activated reactants
rather than by vibronic excitation within a preformed
bimolecular assembly. However, such a conclusion may

be unwarranted. Both Agile and A2210 were calculated
by assuming that the reactions are adiabatic, i.e. that

K = 1. Some recent calculations84’95 indicate that
substantially nonadiabatic pathways (K'hlo-3 -10-2)
may predominate even for Fe(OH2)63+/2+. Values of K

below unity for the present system would tend to bring

rc
Acalc

point is that a steric factor arising from the non-

into closer agreement with Aa and Ab. A related

spherical shape of the reactants should probably be

included in the overall frequency factor. This con-
tention is supported by the observation that both Aa
and Ab fall significantly as substituents are added
to the arene rings, to an extent that is greater for
larger substituents (Table 17). Thus the values of
both Aa and Ab for Cr(CéHS'C6H5)2+/° self exchange

are about fourfold less than for Cr(C6H6)2+/°, even

though the values of Ape and Arc

calc calc are slightly

greater for the latter reaction. Even the parent

arene reactants Cr(06H6)2+/° are not spherical;

 

 

134

reaction may occur preferentially, for example, with the
arene rings in a "side-by-side", rather than "vertically

15.

stacked" configuration. 43 Such steric selectivity

. pe rc
would yield smaller values of Acalc and Acalc than
obtained by using the conventional formulas [Eqns.(23)
and (24)] which refer to spherical structureless re-
actants.

. . 82-85

Further expreSSion the prev1ously used for

the precursor "stability constant" Kp = Qn'NrB/BOOOexp
(-w/RT), appearing in Eqn.(23) may yield incorrectly
large estimates of K and hence APe . This expression

p calc
refers to the formation of contact pairs between two
spherical species. It seems more appropriate to
visualize Kp as being the probability of one reactant
being within a given inclusion volume surrounding the
other reactant within which electron transfer can occur.
Most simply, the magnitude of this volume can be

determined by

KP = [4“N(dg - d3)/30001 exp(-w/RT).......(27)

where d1 is the minimum (contact) distance between the
reactants, and d2 is the maximum distance on which
electron tunneling can effectively occur. If, for
example, d2-d1 = 1 R, Eqn.(16) leads to values of

K and hence Agile for the present systems that are

P
factors of 2-3 smaller than using Eqn.(23). However,

 

135

in reality Kp will continously decrease as the inter-
nuclear distance increases, rather than exhibit the
discontinuity that is presumed in Eqn. (27).97 Sub-
stantially smaller (3. 10-fold) values of Kp can also
be deduced using this model by taking into account
the nonsphericity of the reactants and assuming that
only certain precursor structures, such as a "side-
by-side" configuration, result in electron transfer.

Consequently the values of A2210 obtained by

 

taking such steric and tunneling factors into account

could become comparable to the experimental estimates

of Aa and Ab. Also, the apparently good agreement
pe . . .
seen between Acalc and Aa and Ab is probably misleading

since the inclusion of reasonable steric factors for
the present reactants into the collisional model

[Eqn.(24X1 would almost certainly yield values of

AP9

calc which are markedly too small.

Conclusions

By and large, the experimental kinetics parameters
are in reasonable agreement with the predictions from
conventional electron-transfer theory, at least if
the collision model is used to provide estimates of
the effective frequency factors. However, it is

interesting to note that the deviations observed

 

136

82

It has been suggested that the activation entropyIAS*

can be taken as zero for self-exchange reactions since

*-

out is also predicted to be approximately

As;n a. o and AS
zero from the dielectric continuum model. If indeed
ES*¢B 0, then the smaller frequency factors Ab will be
approximately correct. However, since the experimental
activation enthalpies are smaller than the calculated
values, it is likely that there would be a correspond-

ing discrepancy between the actual and calculated

activation entropies via a "compensation effect",

‘I'

*
so that AS < Ascalc'

For example, the effect could
arise from an increase in specific solvent polarization
required to form the transition-state from the separated
reactants, yielding an unexpected negative contribution
to both AH* andlSS*. Consequently, the actual frequency
factors may well be larger than Ab' and close to Aa'

The latter corresponds to the situation in which the
discrepancies between the actual and calculated en-

thalpic and entropic components of'AG* cancel, so that

the actual free energies of activation approximately

3(-

equal AGcalc'

The comparison between the predictions of the pre-
equilibrium and collision models and the experimental
kinetic parameters can equivalently be presented in

terms of reorganization energies. Table 18 contains

*-

*-
pe and AGr

values of free energies of activation AG c

 

137

between experiment and theory are qualitatively similar
to those seen previously for other self-exchange reac-

tions. Thus it has been found that AH* < AH:a and

lc

* * o o o 0
AS <IAS for ferriCinium/ferrocene in a number of

calc

98 82

solvents, and for Ru(NH3)4bpy3+/2+ in aqueous media.
Most likely, these discrepancies reflect a limitation
of the dielectric continuum model, possibly arising
from changes in short-range reactant-solvent inter-
actions required to form the encounter complex "solvent
cage" prior to electron transfer. The requirement of
forming a particular encounter geometry, with the two
reactants essentially in contact so to maximize the
transmission coefficient, may partly be responsible

for the experimental frequency factors being markedly
smaller than those calculated than are obtained using
a simple model involving activation of a precursor
complex formed in a prior equilibrium step. However,
the simple collision model appears to have greater

practical utility for outer-sphere processes, at least

for the purpose of making numerical calculations.

138 .m.

APPENDIX

There exist a considerable interest in binuclear

102’103 and chromarene.10

mixed valence metallocenes
Although a multitude of spectrosc0pic techniques have
been applied in the study of the metallocene systems,
certain controversial points concerning intervalence
transfer in the monocations and the nature spin-spin
interaction in the dication remain. Several attempts
have been made to synthesized the binuclear chromarenes
but these attempts have failed. The following procedures
104

are described by using Ullmann coupling reaction

from halochromarenes.

(1)

 

Cr .1: Cr

A mixture of 0.276 g (06HSCl)2Cr, 1.2 g cupper
bronze and 50 g biphyl was heated under nitrogen at 180°-
19o°c for 19 h. The mixture was filtered hot and diluted
to volume of 2 L with hexane to precipitate the product.
The dark residue was extracted well with boiling benzene
and the benzene was evaporated. The dark crystal residue
was examined by mass spectrometry and only (C6H6)20r

was found.

 

 

139

(2)

@u. . coo Q)

2 Cu

Cr N Cr T Cr + Cl’
<5)" 6 <8 <6
As described procedure on pagelji, a solution con-
taining 21.2 g(0.06 mole) of iodine dissolved in 150 mL
of dry diethyl ether was rapidly added to the cooled
solution flask at -78°C and the mixture was stirred at
-78°C 1h, it was allowed to warm slowly to 0°C, and then
copper bronze(20 g) was added and the mixture was heated
1&000 14 h and then cooled. It was taken up in benzene
and chromatographed on dry column of A1203/10% H20. Elution
of the column yield a band contained Cr(C6H6)2 and
‘ Cr(06H6)(06H5.06H5).
(3)

@" @‘"° <59 @

Cr + Cf ———-¢- Cf + Cr

en em e e

IK5OH:

140

A three-necked round-botto flask fitted with a
septum cap, a water cooled condenser and a nitrogen
inlet tube. To 0.055 g (0.55 mmole) of Cr(C6H6)2 in
200 mL dried hexane (2.75 mmole) and 1.27 mL TMEDA were
added. The mixture was refluxed for 1 h and allowed to
room temperature. A 0.46 g (0.143 mmole) of Cr(C6H5CHO)2
was slowly 1 h and stirred in -78°C 30 min, the solution
was allowed to room temperature. After removing the
solvent, the residue was sublimed at 90°C/0.2 mm Hg to
give dark compound. The compound was exaimed by mass
spectrometry and found Cr(CéH6)2 and Cr(C6H6)(C6H5CH(0H)-
C6H5).

141

a» 12

<€>-CL 1

I i)
s . s I “x
I.J__ :LJ L..." _ .... 1L MJ’ V\\_

..--.....-..4-.§‘--‘---..oasouu.

a s z”

 

 

. 52
I... .- 147455
I???

 

 

 

 

 

 

 

1 L
-,
4
233
)3, 69 I92. 21' it

Fig.28 NMR and Mass spectra of (C6HSCl)2Cr(0).

142

 

 

 

 

 

 

 

 

 

. -_'- ---

L...—._.. .7...- -0—7..-

 

I s
W
’L _______,_/ ¥’/ — ...-—
_, __.-...__.._._ ______,_ -- _-_- -- -
7;" Z 3 ' '~ fifi: I 1 fi Y I V "

Ina-u ’2

'1

QCHO
fi
138

1
39.84

1

4

-- :4
135
* l
19 129
l 5‘ 1‘ - L A 2‘36 ,
h 1 T TVTVI r Y I TjI' fi'r'TI *j—f f f V

'1 E '4‘ m 1% 2m en

Fig.29 NMR and Mass spectra of (C6H CH0)2Cr(0).

5

143

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

"II III

 

 

I z : l ' I
wumsn IDA -)

 

; .
INO

 

 

 

 

WAVENUMBEIHCM ') _ 7

and (C 6H50CH3) 201-"

O

31 IR spectra of (C(3H5OCH3)2CI‘

Fig.

 

144

 

 

 

 

 

 

 

 

 

 

 

 

 

I 400

 

 

Mt!)

 

 

 

 

 

 

 

+

Fig. 30 IR spectra of (06H5c1)20r° and (C6H5Cl)ZCr

145
REFERENCE

1. P. L. Timms, Chem. Commun., 1968,1525.

2. For comprehensive reviews and leading references to the
chemistry of metal synthesis, see ref. 63.

3. Reviews to conventional synthetic method for the chemistry

of bis(arene)chromium, see R. P. E. Sneeden, 'Organo-
chromium compounds", Academic Press, New York, 1975.
4. R. W. Bush and H. R. Snyder, J. Org. Chem., 1960, 2 , 1240.
5. Ch. Elschenbroich, J. Organomet. Chem., 1970, 22, 677.
6. H. Burdorf and Ch. Elschenbroich, Z. Naturforsch., 1981,
£26, 94.
7. Ch. Elschenbroich and F. Stohler., Chemia., 1974, 28, 730.
J. Organomet. Chem., 1974, 6 , C31.
8. A. N. Nesmeyanov et a1, lzv. Akad. Nauk. SSSR. Se;.,
1974, 2865. igig, 1976, 922.
9. Ch. Elschenbroich and J. Heck, Angew.Chem. lnt. Ed. Engl.,
1981, g9, 267, ib_ig, 1977. g, 479.
10. Ch. Elschenbroich and J. Heck, J. Am. Chem. Soc., 1979,
191. 6773.
11. C. Senoff. Coordination Chem. Rev., 1980, 32, 111.
12. G. L. K. Hoh, W. E. McEwen and J. Kleinberg, J. Am. Chem.
_s_o_c., 1961, _8_3, 3149.
13. E. G. Perevalova gt a1, DoklJ Akad. Nauk. SSSR., 1964,
iii! 857.
14. A. N. Nesmeyanov gt al, J. Organomet. Chem., 1969, 20,
169.

15. T. T-T. Li and C. H. Brubaker, Jr, J. Organomet. Chem.,

 

16.

17.

18.

190

20.

21.

22.
23.

24.

25.

26.

27.
28.

146

1981, 226, 223.

L. P. Yureva gt 2}, J. Organomet. Chem., 1981, 229,

43.

P. M. Treichel, G. P. Essenmacher and K. J. Klabunder,
ingrg. Chem. Acta., 1981, 48, 41.

P. S. Skell, D. L. Williams-Smith and M. J. McGlinchey,
J. Am. Chem. Soc., 1973, 25, 3337

J. A. Gladyse, J. G. Fulcher, S. J. Lee and A. B.
Bocarsley, Tetrahedron Lett., 1977, 32, 3421.

L. H. Simons and J. J. Lagowski, The reaction of
alkynes with transition metal atoms, "Fundamental
Research in Homogeneous Catalysis", Tsutsui Ed, Plenum
Press, NY 1977, P. 73.

A. N. Nesmeyanov gt 2;, g;_grganomet. Chem., 1981, _;Z,
195.

A. J. L. Hanlan et al, inorgLChem., 1980, 29, 543.

M. Tsutsui, J. Aryoshi, T. Koyamo and M. N. Levy, 2g!
Chem. Ser. 1968, 29, 266.

V. M. Akhmedov, M. T. Anthony, M. L. H. Green and D.
Young, J. Chem, Soc, Dalton, 1975,1412.

Y. Huang et al, g;_grganomet. Chem., 1981, 228, 169,
1981, 225, 185.

Y. Tazima and S. Yuguchi, Bull. Chem. Soc. Jap3, 1966,
‘32, 2534.

B. Morosin, Acta. Cryst., 1974, 239, 838.

 

0. V. Starovskii and Yu. T. Struchkov, Zh. Strukt. Khim.,
1961, 2, 161.

290

30-

310

320

33-

34.

35-

36.

37.

38.

39-

40.

41.

147

V. A. Lebedev, V. P. Golovachev and E. A. Kuzmin,
Zh. Strukt. Khim., 1977, 28, 1073.

M. Ziegler gt gt, Angew. Chem.;nt. Ed. Engl., 1980,
22, 44. Acta. Cryst., 1980, 826, 2054.

T. N. Salnikova, V. G. Andrianov, L. P. Yureva and
N. N. Zaitseva, Koord. Khim., 1978, 4, 288.

M. W. Eyring, E. C. Zuerner and L. J. Radonovich,

Inorg. Chem., 1981, 2Q, 3405.

U. Zenneck, Ch. Elschenbrich and R. Mockel, J. Organo-
met. Chem., 1981, 229, 177.

H. Burdorf and Ch. Elschenbroich, Z. Natgtfgtggg.,
1981. 368, 94.

Ch. Elschenbroich, R. Mockel, U. Zenneck and D. W.
Clack, Ber. Bunsenges. 2hys.Chem., 1979, 83, 1008.

B. G. Gribov, B. I. Kozyrkin, A. D. Krivospitskii

and G. K. Chirkin, Pro. Acad. Sci. USSR. Chem. Sect.,
1970, £23, 457.

R. Prins and F. J. Reinders, Chem. Phys.Lette§.,
1969: 2A 45-

E. T. Kaiser and L. Kevan, "Radical ion" Interscience,
John Wiley & Son, New York 1968, page 211 and 151.

A. Domenicano, A. Vaciago and C. A. Coulson. Agtg.
Cryst. 1975. §3_. 221.

A. Domenicano and A. Vaciago, Acta. Cryst., 1979, 835,
1382.

G. Maier, S. Pfrien, U. Schafer and R. Matusch, Agggt.

Chem. Int. Ed. Engl., 1978, Z, 520.

42.

“'3.

44.

148

T. T-T Li, M. J. Weaver and C. H. Brubaker, Jr., J. Am.

Chem. Soc., in press (1982).

Ch. Elschenbroich and U. Zenneck, J. Organomet. Chem.,

1978, 169, 125. .

(a) J. H. Espenson and J. R. Pladziewicz, J. Am. Chem.
§2§;. 1973: 25! 56-

(b) E. S. Yang, M-S Chan and A. C. Wahl, J. Phys. Chem.

 

1980, g3, 3094.

45. D. R. Stranks, Dis. Faraday Soc., 1960, 29, 73.

46.

“'70

48.

(a) I. Ruff, V. J. Friedrick, K. Demeter and K.
Csillag, J. Phygg Chem., 1975, 25, 3303.

(b) A. C. Wahl and M. W. Dietrick, J. Chem. Phys.,
1963, 3g, 1591.

(c) A. F. Voigt and S. Katz. Dissertation abstracts.,
1959, 19, 2242.

(a) R. L. Ward and S. I. Weissman, J.5Am;5Chem. Soc.,
1954, 26, 3612.

(b) M. T. Johns and S. I. Weissman, J. Am. Chem. Soc.,
1962, 83, 4269.

(c) P. J. Zandstra and S. I. Weissman, J. Am. Chem.
Soc., 1962, gt, 4408.

_(d) s. I. Weissman, gt a;..11_9heml_Ehxs-. 1960. 22.

626.

(a) T. A. Miller and R. N. Adams, 14_Amt Chem. 889.,
1968, 99, 3611.

(b) Exch React., Prclemp 1965, 1, 1965, 151.

49.

50.

51.

52.

(o)

(d)

(e)

(a)

(b)

(a)

(b)

(C)

(a)

(b)

(a)

(b)

149

N. Hirota, R. Garraway and W. Schook, J. Am. Chem.
.Sgg., 1968, 0, 3611.

G. L. Malinoski, Jr., W. H. Bruning and R. G.
Griffin, J. Am. Chem. Soc., 1970, 92, 2665.

B. A. Kowert, L. Marcoux and A. J. Bard, J. Am.
Chem. Soc., 1972, 99, 5338.

F. A. Prins and F. J. Reinder, J. Am. Chem. Soc.,
1969, 92, 4929.

M. F. Rettig, "Spin Distribution in Organometallic

 

Compounds",NMR of Paramagnetic Molecules,Princi-
ples and Applications, Ed by G. N. LaMar, W. D.
Horock, Jr., RI H. Holm Academic press, N.Y. 1973,
p 217.

I. H. Hillier and R. M. Canadine, Discuss Faraday

SQCO, 19699 89 27'

J. Weber, M. Geoffroy, A. Goursot and E. Penigault,
_JJ Am. Chem. Soc., 1978, 288, 3995.

S. E. Anderson, Jr., and R. S. Drago, Inorg. Chem.
1972, it, 1564.

C. Furlani and E. 0. Fischer, Z. Electrochem.,
1957, 62, 481.

C. Furlani, A. Furlani and L. Sestili, J. electro-
anal Chem., 1965, 99, 104.

A. Korshunou, gt g;., Dokl, Akad. Nauk, SSSR.,
1958, 222, 769.

I. A. Suskina gt 62., idid., 1971, 2, 425

53-

54.

55-

56.

57-

58.

59.

60.

61.

62.

63.

150

(c) L. N. Vertyulina, t 1., Zh-Obschch. Khim.,

1963. 33, 285.
A. A. Vlcek, Z. Anorg. Allg. Chem. 1960, 104, 109.

 

 

A. A. Vlcek, Z. Anorg. Allg. Chem. 1964, 334, 205.
R. A. Marcus, J. Chem. Phyg,, 1965,‘93, 679 and

 

references

R. L. Ward and S. I. Weissman, J. Am. Chem., Soc.,
1957, 29, 2068. .

C. Chang and C. S. Johnson, J. AE- Chem.g§gc., 1966,
22. 2338.

C. P. Poole, Jr., "Electron Spin Resonance" Interscience,
New York, N. Y., 1967. pp 807-811.

T. A. Miller and R. N. Adams, J. Chem. Phy§,, 1966,
33, 4022.

P. Zuman, "Substituent effect in organic polarography"
Plenum Press, N.Y. 1967, pp 46-48.

R. G. Gastinger and K. J. Klabunde, Transition Mg .

Qhfimn 1979.l9, 1.

S. P. Gubin and E. G. Perevalova, Dle‘_AkaQJ_Nauk,

$553,, 1962, 293, 1351.

(a) K. J. Klabunde, Acc,_chem1_3es,, 1975. §. 393.

(b) J. R. Blackborow, D. Young, "Metal Vapor Synthesis
In Organometallic Chemistry" Springer-Verlag,

Berlin Heidelberg, N.Y. 1979.
(c) V. Graves and J. J. Lagowski, Ingrg_ghgm,, 1976,

12. 577-

151

(d) K. J. Klabunde, "Chemistry of Free Atom and

Particles", Academic Press, N.Y. 1980.

(e) M. J. McGlinchey, gt g2.. Qrganomet. Chem., 1977,

1.4.1.. 185-

64.(a) P. L. Timms _t'22., Inorg. gxgth., 1979, 29; 59.

(b) M. Moskovits and G. A. Ozin,"Cryochemistry"

Wiley, New York, 1976.

65.(a) G. Brauer., Ed., "Handbook of preparative Inorganic

66.

67c

68.

69.

700

71.

Chemistry", vol. 2, Academic Press, N.Y. pp. 1396.

(b) E. 0. Fischer., "Inorganic Syntheses", 1960, 6, 132.

H. F. Efner, D. E. Tevault, W. B. Fox and R. R. S.
Mardzewski, J. Organomet. Chem. 1978, 296, 45.

D. D. Perrin, W. L. F. Armaregu and D. R. Perrin,
"Purification of laboratory chemicals", N.Y. Pergamon
Press, 1966.

M. J. Weaver, gt g2, J. Am. Chem. Soc., 1979, 292, 1131.
J. D. L. Holloway and W. E. Geiger, Jr...J4_Aml_Qh§m.
'§gg., 1979, 292, 2038.

L. J. Radonovich, E. C. Zuerner, H. F. Efner and K. J.
Klabunde,_2norg, Chem., 1979, 25, 2976.

 

W. Waller, Ph. D. Dissertation, Michigan State Uni.,
E. Lansing, Mi., 1973.

72.(a) J. Bruhin, F. Gerson and H. Phya-Nishiguchi, Helv.

allimo ACt3-1977, .6_O_, 2470

(b) 0. W. Horwarth and G. K. Fraenkel, J. Am. Chem.

 

M09 19669 §§9 4514'

73-

74.

75-

76c

77-

78.
79-

80.

81.

152

(c) H. P. Fritz et al, Z. Naturforsch., 1978, 33_, 498.

(a) K. Keulen and F. Jellinek, J. Organomet. Chem.

 

1966, 5, 490.
(b) P. E. Riley and R. E. Davis, Inorg. Chem., 1976,
$5. 2735-

(a) B. J. Hevbert and D. E. Smith, J,_Elggtrgggal.
Chgmfli 1971I llu 333-

(b) R. D. Allendoerfer and P. H. Rieger, J. Am. Chem.
Eqn., 1965I §ZI 2336-

(c) C. S. Johnson, Jr., J. Chem. Phys., 1963, 39, 2111.

K. J. Laidler, fChemical Kinetics", McGraw-Hill, N.Y.

1965, pp. 251-253.

S. Evans, J. C. Green and S. E. Jackson, J. Chem. Soc.,

 

Dalton., 2919, 304.

F. G. N. Cloke, M. L. H. Green and D. H. Price, 8.
CM” Dalton, 1978, 532.

K. D. Warren, Structure and Bonding., 1976, 22, 45.

N. Hao and M. J. McGlinchey, J. Organomet. Chem. 1978
E, 381.

F. G. N. Cloke, M. L. H. Green and G. E. Morris, 8.
thm. Soc., Chem. Commun., 1975, 1417, 1978, 431, 1978,
72.

For recent reviews, see (a) Schmidt, P.P. in "Electro-
chemistry - A Specialist Periodical Report",_2gg

.Qheminal_SQQi§3¥ (PUb) V01- 5. Chapter 2. 19753 V01- 6.
Chapter 4, 1977.

82.

830

84.

85.

86.

87c

88.

89.
90.

153

(b) Ulstrup, J., "Charge Transfer Processes in Con-
densed Media", Springer Verlag, West Berlin, 1979.

(c) Cannon, R. D., "Electron Transfer Reactions",
Butterworths, London, 1980.

(a) Brown, G. M.; Sutin, N., J. Am. Chem. Soc., 1979,
282, 883.

(b) R. A. Marcus, Int, J, Qhem. Kinet, 1981, 23, 865.

Sutin, N. in "Tunneling in Biological Systems",

(Chance, B.: Devault, D. C.: Frauenfelder, H.; Marcus,

R. A.; Schrieffer, J. R.; Sutin, N. eds), Academic

Press, New York, 1979, p. 201.

Brunschwig, B. 8.; Logan, J.; Newton, M. D.; Sutin,

N., J. Am. Chem. Soc. 1980, 282, 5798.

Walsh, J. L.: Baumann, J. A.; Meyer, T. J., 2norg, Chem.,

1980, 29, 2145.

Bukhs, E.; Bixon, M.; Jortner. J.; Navon, G., Inorg.

88gt., 1979I 28. 2014.

Johns, L. H., "Inorganic Vibrational Spectroscopy",

Marcel Dekker, New York, 1971, p. 175.

Maslowsky, Jr., E., "Vibrational Spectra of Organo-

metallic Compounds", Wiley-Interscience, New York,

1977I P- 301-

Fritz, H. P., Adv. Organometal. Chem., 1964, 2, 239.

(a) Kenlen, E.; Jellinek, E., J. Organomet. Chem.,

1966, 5, 490.

154

(b) Starovskii, O.V.; Strichkov, Y. T., Acad. Nauk.
SSSR. 1961, 35, 620.
91. Haaland, A., Acc. Chem. Res., 1979, 22, 415.

 

92. Gordon, A. J.; Ford, R.A., "The Chemists Companion
- A Handbook of Practical Data Techniques and Refer-
ences", Wiley-Interscience, New York, 1972, pp. 107-
112. h

93. The rate constants k reported in ref.15 which were
used in Table16Iand elsewhere in the present paper
were obtained in the absence of a background electrolyte
(reactant concentrations 10-4 - 10'3 M). The addition
of KClOu had only a negligible influence upon k(<10%)
even up to 0.5 M, although small yet significant
decreases in k1$20%) were observed upon the addition
of KPF6. This latter influence may be due to slight
ion association. These results support the contention
that the work terms w, at least those arising from.
simple electrostatic interactions, are negligible for
these systems. (See Figure 26).

94. See for example, Lumry, R.; Rajender, S., Biopolymers,
1970, 9, 1125.

95. Newton, M. D., 2nt. J. Quantum Chem., Symp., 1980,
12.. 363.

96. Fuoss, R. M., J. Am. Chem. Soc., 1958, 88, 5059.

97. A related pre-equilibrium formulation for nonadiabatic

reactions with a detailed consideration of inclusion