-v -.

 

 

w "I'V‘Wr'wr-Iu'

P
l'
,.
'3

 

 

 

 

 

 

 

 

 

 

 

1
I. “NI—a —i: “-5. -x. -.-.,‘ :gL

4 n-

 

 

 

 

 

 

 

 

 

 

l
m.” ."., .‘ .‘
‘11.” I II
'II
5, n: 0
”1,1,4,

1
w

-I u ,«
“Isiah.

2,. 1 .
”damn;
, x
In H:
.‘41'4 ‘1
'rul)q mu l
‘ 1"!

l 4 I
.
ill‘lt-‘vg'fl‘l7 I

)~ I H
v. i , 41.2.}, ,
,. “l > V.“
"ng ‘ l
l 1 1 r

 

 

"'L":'.-....I..T..‘\a.\.
. . .kLV.-.—.l
- -uw 1,1...

 

. .

.l p."

rrfltz'r:
,

11‘.:‘hj~"-~
I x I». ...«. L-:..
"1..- . anus“... .I--

.u I
‘ 1;... --. ...
1K'\..l'..:\l.{ li‘\.l|:lu 4..-
-_‘I«.‘1.-v-n.'.x. h-l-u
..4- (1;;

a.

"f“ 3.4.1
.I won-.. ugm ..
w . .L' .Is‘z 4A4
ul»L\
_...-\.I\._z.—s.4:..'u m“, _

.w. 6. I- «I.

. nay: :

'u .u.
f!“

acu-

I

l n. E.
..l.‘?_;?;’.|f,

”a

«r.

, .

I “1 VI
, .. .m ,
1,=~,.z.k..fl.,u\

4

"I11 I -u\ 'l I I
I

l-U'll“ VJ
w- II‘“
("I ‘F -

t
"5} 7‘35.-

I rwv r m l l-IJ- Ivrru N um “’l'l' ("f
‘7'"‘1‘ph‘m 7M." h;ij

., I
l ‘I‘ d
w [>11le .
I I
’1

, .,
1- «can: rm : Fm m;
”.17....

.r .u- -..
1a,)“ 1:,
c.1111...
-...,.

mr

1L
rv‘vneulrrxvuan ‘1
p. “ml“
0‘

I r- m
uvw- Ian-rm - l
v'

r I u ..- v -
1. . '2‘”; ‘,.‘ wk" 1:3,"... . .
”1...”.-1. Hi 1:“, ..
mm
Ir» I-Ia-MI
,.. .

I I:
P V “'
I" ,1 '1‘ 4.41:1; ,
nu)»:- ww I ; 3.1.1., ,., .4... .a'u 'J"l“'
a v4
‘0

”a! fay:
.1 uni-:1

5.1.71.3:
mmq- ; ”.11; Inn-IE‘II" ,
H" " " "2.1:...1‘an‘r-u '5
”Fl”. I “
1—1:“ '3': ... . .133...” .V........,....
u“ .1- ..

2v.-
11-:
.3.» 1“”. M,
11-41 i'v‘r

1

:-:1I'~c::n{
,1,. imw :33.

a»
. 1. .1, .
r, . '1’: .’;"‘ u,’ .

m Ik- - worm-fl”

m v tmwmwmgw 1

m wan—pp
. (:3, véwpa Jan-mum:-
_ a» 53-... rang-«~55, F'

 

 

 

THE‘V-ZLK

 

 

 

 

 

 

 

 

 

llllillllllllllllllllllllllllll
3 1293 00877 1069

This is to certify that the

dissertation entitled

Rafi/1611A Speowsoofy 0‘3: m Odi‘FCOi Foyrifirwh
[Mac/vocifde .- forfhycene Cw. J firefly/ong_
presented by

lQTw/ Yo cm '<\/Dc

has been accepted towards fulfillment
of the requirements for

Mdegreeinmm Ole‘zqflimé‘j

kwé‘f/[MZWK

Major professor

Date 3' 16 / ’2-

MSUMM” ,. . . ". m” .1. 042771

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

'mozm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

czicirc‘idatedue. pm&p.1

 

RAMAN SPECTROSCOPY OF MODIFIED PORPHYRIN MACROCYCLES
- PORPHYCENE AND CHLORIN -

BY

YOUNKYOO KIM

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

 

Department of Chemistry

1992

 

Abstract

4?2¢"%0a’

RAMAN SPECTROSCOPY OF MODIFIED PORPHYRIN MACROCYCLES
- PORPHYCENE AND CHLORIN -

BY

YOUNKYOO KIM

The electronic and vibronic properties of
metalloporphycenes (MPPC) which are isomers of the
extensively studied metalloporphyrins, and the effect of
lower symmetry in metallooctaethylchlorins (MOEC) were
studied by using UV-visible, IR, Raman spectroscopy and
normal coordinate calculations.

Close investigation of absorption band positions,
spectral shifts caused by metal substitution, ring
bromination and oscillator strengths reveal that the w-+ w*
states of MPPC are adequately defined by one-electron wave
functions unlike those of metalloporphyrins. Thus the «—
electron configuration interaction, which dominates the
spectroscopy of the metalloporphyrins, is weak in the MPPC.
Raman spectra of CuPPC with either Soret or visible
excitation do not show anomalously polarized bands, which
are commonly observed in metalloporphyrins. This means that

Herzberg - Teller mechanisms for vibronic coupling are

extremely weak in the MPPC. Several vibrational bands show
dependence on the central metal in the MPPC that is
analogous to the metal dependent frequencies of the
structure sensitive bands in the metalloporphyrins. Like
hemes, both UV-visible and resonance Raman spectra of FePPC
derivatives demonstrate sensitivity to iron spin and
oxidation states. This suggests an overlap of the iron dmfl
‘%m and the porphycene w orbitals.

Normal coordinate calculations based on the IR and
resonance Raman spectra of selectively meso-deuterated CuOEC
(CuOEC-h,“ CuOEC-afi-dz, CuOEC-yb-dz, CuOEC-d4) are reported.
In general, the calculations show that various force fields
satisfactorily reproduce the experimental data of CuOEC.
However, none of these force fields reproduce the
frequencies and meso-deuteration shift patterns of all of
the modes. This suggests that 1,4 and higher interaction
constants could be important to the valence force field. All
of the force fields indicate that the eigenvectors of
certain skeletal mode are localized on sectors of the
macrocycle. A detailed vibrational spectroscopic
characterization of ZnOEC was also completed. The meso-
deuteration shift patterns of CuOEC and ZnOEC show that mode
localization phenomena are not dependent on the identity of
the central metal but are inherent properties of

metallochlorins that results from their reduced symmetry.

To my mother and in memory of my father

iv

ACKNOWLEDGMENTS

I am deeply grateful to Dr. Gerald T. Babcock for his
guidance and financial support. I wish to also thank Dr.
George E. Leroi for serving as second leader. Tony, Juan and
Dr. T. Carter deserve a special thanks for their technical
assistance with Raman spectroscopy. Special thanks is
extended to W. Wu and Einhard for their sincerity,
enthusiasm and perseverance in the preparation of
porphycenes and chlorine. Thanks also go to Mike, Hak-Hyun
and Alex for their discussions and help with the set of the
normal coordinate analysis programs. I also express my
thankfulness to all our group member for their friendship. I
also appreciate the encouragement and assistance of Mr. Dae
W. Yun, all of my friends and all Korean students in the
chemistry department. Without their assistance, this work
would have been much more difficult.

Finally, I would like to thank to my wife (Yunmi), my
daughter (Yeaji), and my family, including my parents-in-
law, for their patience, love, help, and constant

encouragement.

TABLE OF CONTENTS

Page
LIST OF TABLES ........... . ........ ... ..... . .............. vfii
LIST OF FIGURES ....... ... ................................. ix
Chapter 1
General Introduction..................... ..... ... ......... 1
1.1) The Structure of Macrocycles studied. ...............1
1.1.1) Metalloporphyrin .................. .......... ..... 1
1.1.2) Metalloporphycene ......OOOOOOOOOOO0.0.00.00.00.00 3
1.1.3) MetalIOChlorj-n......OOOOOOOOOOOOOOOOOOOOOOOOOOOOO3
1.2) Electronic Absorption Spectroscopy ........ ........ .. 6
1.3) Infrared and Raman Spectroscopy ...... ........ . ..... 11
Chapter 2
Materials and Methods....................................20
2.1) Preparation of Porphycenes ........................ .20
2.2) Preparation Of Chlorins ...OOOOOOOOOOOO00.00.0000... 21
2.3) Spectroscopic Studies ..... ......................... 22
Chapter 3
Electronic and Vibronic Properties of
Metalloporphycenes................................ ..... ..24
3.1) Introduction .................... ..... .. ....... ..... 24
3.2) Results ............OOOOOOOOOOO... ........ I. ....... .26
3.2.1) Electronic Absorption Spectra ...................26
3.2.2) Resonance Raman Spectra ......... ...... . ....... .. 34
3.3) DiscuSSionsOOOOOOOOOOOOOOOOO000.000....00.0...O....47
3.3.1) Electronic Properties ............ ..... .. ...... .. 47
3.3.2) Symmetry and the Resonance Raman
Depolarization Ratio ............. .............. . 55
3.3.3) Vibronic Coupling ...............................56

vi

Page

3.3.4) Effects of Selective Deuteration and Metal

Substitution on Raman Frequencies .............. 57

3.3.5) Structural-sensitive Vibrations .......... ..... .. 60

3.3.6) Iron Complexes ................. ................ . 62
Chapter 4

Normal Coordinate Analysis and Effects of Lower Symmetry in

Metallochlorins .........................................66
4.1) Introduction ................... ................ .... 66
4.2) Theory of Normal Vibrations ...................... . 69
4.2.1) Normal Modes .......................... ....... ... 69
4.2.2) Internal Coordinates ......................... ..70
4.2.3) The Transformation from Cartesian to
Internal Coordinates .... ........... ........ ..... 71
4.2.4) The Kinetic Energy in Internal Coordinates ......75
4.2.5) The Potential Energy in Internal Coordinates ....77
4.2.6) The Secular Equation ............................78
4.2.7) The Potential Energy Distribution ............... 79
4.3) Normal Coordinate Calculations ............ ..... . 79
4.4) Results ................................... ...... 94
4.4.1) Vibrational Spectra of CuOEC and ZnOEC . ....... 94
4.4. 2) Vibrational Assignments for CuOEC ............ 105
4.5) Discussions ................................ ..... 113
4.5.1) Vibrational Assignments .. ..... ................. 113
4. 5.1.1) CRT,Vibrations............. ............... .118
4. 5.1.2) CR1,Vibrations.................. ......... ..121
4. 5.1.3) CaCb and CaN Vibrations 122
4. 5.1.4) C%H Vibrations ............................. 123
4.5.2) Comparison of the Force Fields ................ 125
4 5.2.1) General Feature of the Calculations ....... 125
4 5.2.2) Relative Merit of Individual Force Fields .139
4.6) Summary and Conclusions .......................... 142
Chapter 5
Conclusions and Future Work ........ .. ...... ..... ....... 143

List of References ........... .... ...................... 147

vfi

List of Tables

Page
UV-visible Absorptions (nm) of Metallo-
porphycenes and Metalloporphyrins and
Assignment of Electronic Transitions of MPPC. ..... 33
Resonance Raman Depolarization Ratios,
p=Il/I”’ for cuPPCOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOO 42
Raman Vibrational Frequencies (cm4) of
MHPPC Complexes......OOOOOOOOOOOOOOOO0.... ..... 00.45
(A) Core Sizes of PPC and Analogous
Porphyrin Complexes..... ........ ...... ............ 63
(B) Estimates of Structural Parameters for
MPPC and MOEP Vibrational Frequencies............. 63

Geometric Parameters for the CuOEC Calculation.... 81

In-Plane C5 Symmetry Coordinates for CuOEC. ...... 85

Refined Diagonal Force Constants for CuOEC. ....... 90
Fixed Diagonal Force Constants for CuOEC... ....... 91
Fixed Off Diagonal Force Constants for CuOEC. ..... 92

Vibrational Frequencies (cmq), Isotope
Shifts (cmd), and Symmetries for the High-
Frequency Modes of CuOEC and ZnOEC............... 104

Observed and Calculated Frequencies and
Isotope Shifts (cm4) for the High-Frequency
Skeletal Modes of CuOEC.......................... 110

Comparison of the Frequencies and Deuteration
Shifts (cm4) for the Various Force Fields........ 114

Average Errors between the Observed and
Calculated Frequencies and Deuteration Shifts
for the Various Forces Fields. ................... 116

The Correlation of D4h and C2 In-Plane Modes. . . .. . 117

\liii

List of Figures

Page

Structure of metallooctaethylporphyrin... .......... 2

Structure of metallo-2,7,12,17-tetrapropyl-
porphycene.........OOOOOOOOOOOOOOOOO0.0.00. ..... .0004

Structure of metallooctaethylchlorin. .............. 5

Molecular structure of (a) FeHOEP and

(b) FeHOEC as determined by X-ray crystallo-
grath(fromrefo17).ooooooooooooooooooooeoooooooo7

Electronic absorption spectrum of CuOEP in
CH2C12 SOlution.......OOOOOOOOOOOOOOOOOOO. 0000000000 8

Porphyrin M.O.s comprising the Gouterman
four orbital model (from ref 20).................. 10

Energy diagram for metalloporphyrin absorption
spectra. The porphyrin orbitals are indicated

at the center of the figure. The mixed-excited

states (alulegl) and (a2u1egl) interact via
configuration interaction (C.I.) to give the a-

band and Soret, respectively. This is shown on

the right hand side of the figure. Interaction

of porphyrin orbitals (any inn) with the metal
orbitals of the appropriate symmetry (d),z and dyz)

is shown on the left hand side of the figure. .....12

Molecular process associated with a) Infrared,
and b) Raman spectroscopy. ........................ 13

UV-visible absorption spectra of CuPPC and
CuOEP (inset) in CHRIQ solution............. ...... 27

UV-visible absorption (dashed curve) and first
derivative (solid curve) absorption spectra of
CuPPC in CH2C12 solution.................. ......... 29

ix

Page

UV-visible absorption (dashed curve) and first
derivative (solid curve) absorption spectra of
NiPPC in CH2C12 SOlution.......OOOOOOOOOOOOOOOOOOO.31

UV-visible absorption spectra of FePPC derivative

in CH2C12 solution. The ferrous sample was prepared
using hydrazine hydrate as a reductant as described
in MATERIALS and METHODS. The absorption at 280 nm

in the spectrum of the (Im)2FemPPC is from excess
N-methylimidazole.................................35

Resonance Raman spectra of CuOEP and CuPPC

in CHZCl2 solution at concentration of

approximately 0.1 and 0.2 mM, respectively.

Laser powers at 363.8 nm were 20 and 10 mW,
respectively. Asterisks(*) in Figures 3-5, 6, 7

and 8 represent solvent vibrations. ............... 37

Polarized resonance Raman spectra of CuPPC

in CHZCI2 solution at 406.7 nm and 613.0 nm.
Concentrations were approximately 0.5 mM and

0.3 mM and laser powers 40 and 100 mW,

respectively. In each spectrum, the top

trace represents I” and the bottom trace

represents Ii..... ...... .. ...... .... .............. 40

Resonance Raman spectra of NiPPC derivatives

in CH2C12 solution at concentration of

approximately 0.2 mM. Laser power at

363.8 nm was 30 mW............... ................. 43

Resonance Raman spectra of FeIII and FeII

complexes of PPC. Concentrations in CHRIQ

solution were approximately 0.1 mM and

laser power typically 20 mW. Spectrum a

was measured with a different diode array

detector than spectra b-d. ................ ........ 48

Schematic orbital energy level diagrams for

CuOEP(D4h) and CuPPC(D2h). 51
Bond Stretch: Arij ................................72
Valence Angle Bend: Aaijk ..........................72
Out-of—Plane Wag: A0 ............................. 73

Linear Valence Angle Bend: A¢ ............ ........ 73
TorSion: AT ......OOOOOOOOOOOOOOOOOO0...... ....... 74
Structure and Atom labeling scheme for CuOEC. .....82

Internal coordinate representation of the

In-plane metallochlorin bond stretching

deformations. The numbers on the bonds refer

to the numbers used in the computer program of
Schachtschneider..................................83

Internal coordinate representation of the
metallochlorin angle bending deformations.

The numbers on the bonds refer to the numbers

used in the computer program of Schachtschneider.
Some of internal coordinates could not be

represented in plane. In number series,

left hand side numbers represent internal

coordinates and right hand side number series
represent atom labeling in Figure 4-6. ..... ....... 84

Resonance Raman spectra of ZnOEC, ZnOEC-a,B-d2,
ZnOEC-'y,r5-d2 and ZnOEC-d4 in CH2C12 solution

obtained with Soret excitation at 406.7 nm.

Laser power: 7 mW, concentrations: ~70 pM.
Asterisks(*) in Figures 4-9, 10, 11, 12, 13

and 14 represent solvent vibrations... ............ 95

IR spectra of saturated CHCl3 solution of
ZnOEC, ZnOEC-a,fl-d2, ZnOEC-'y,6-d2 and ZnOEC-d4. ..... 97

Resonance Raman spectra of CuOEC,

CuOEC-oz,B-d2, CuOEC-7,6-d2 and CuOEC-d4

in CH2C12 solution obtained with visible

excitation at 406.7 nm. Laser power: 7 mW,
concentrations: ~40 uM. .......... ................. 99

Resonance Raman spectra of CuOEC,

CuOEC-a,B-d2, CuOEC-'y,<5-d2 and CuOEC-d4

in CHZC:l2 solution obtained with Soret

excitation at 615.0 nm. Laser power: 40 mW,
concentrations: ~70 uM. ................. ......... 101

xi

Page

Polarized resonance Raman spectra of ZnOEC

and ZnOE:C-o:,B-d2 in CH2C12 solution

obtained with Soret excitation at 406.7 nm.

Laser power: 7 mW, concentrations: ~40 pM. . ...... 106

Polarized resonance Raman spectra of
ZnOEC—'y,¢3-d2 and ZnOEC-—d4 in CHZCl2 solution
obtained with Soret excitation at 406.7 nm.
Laser power: 7 mW, concentrations: ~40 pM. .......108
Vibrational eigenvectors of CuOEC which

contain substantial contributions from CaCn1
stretching vibrations Displacements are

shown only for those atoms whose motions
contribute significantly to the normal mode

(10% or greater of the maximum atomic

displacement of a given mode)......... ........... 127

Vibrational eigenvectors of CuOEC which

contain substantial contributions from

CbCb stretching motions. Displacements are

shown only for those atoms whose motions

contribute significantly to the normal mode

(10% or greater of the maximum atomic

displacement of a given mode)...... ......... ..... 131

Vibrational eigenvectors of CuOEC which

contain substantial contributions from Ogi
deformations. Displacements are shown only

for those atoms whose motions contribute
significantly to the normal mode (10% or

greater of the maximum atomic displacement

of a given mode). .............. . ...... ... ........ 133

Vibrational eigenvectors of CuOEC which

contain substantial contributions from both

CaCb and CaN stretching vibrations.

Displacements are shown only for those atoms

whose motions contribute significantly to the

normal mode (10% or greater of the maximum

atomic displacement of a given mode).... ......... 135

xfi

CHAPTER 1

GENERNRL INTRODUCTION

1.1) The structure of the macrocycles studied

1.1.1) notalloporphyrin

The metalloporphyrins, referred to as the pigments of
life owing to their biological significance, are
characterized by a tetrapyrrole macrocycle with an
extensively delocalized r electron system.1
Metalloporphyrins play important roles in such diverse
biochemical functions as oxygen transport, storage and
utilization (hemoglobin, myoglobin, cytochrome oxidase),
electron transfer (the cytochromes) and catalysis
(cytochrome P450, horseradish peroxidase, catalase).2 Figure
1-1 shows the basic structure of a typical synthetic
metallooctaethylporphyrin. The four pyrrole rings are joined
at the methine bridge or meso positions, which are labelled
a,B,7,6 in Figure 1-1, and various peripheral substituents
can be linked to the b-carbon positions, which are usually
labelled 1-8. The remaining carbon atoms adjacent to the

pyrrole nitrogens are called a-carbons.

Figure 1-1 Structure of Metallooctaethylporphyrin

1.1.2) Metalloporphycene

Metalloporphycene, a tetrapyrrole macrocycle recently
synthesized by Vogel et al.3, represents a class of
molecules that are structural isomers of metalloporphyrins.
The limited solubility of porphycene and metalloporphycene
led Vogel and co-workers‘f‘to develop the synthesis of 2, 7,
12, 17-tetrapropylporphycene (Figure 1-2). This compound is
expected to assume a role in porphycene investigations
similar to that of octaethylporphyrin (OEP) in studies of

model porphyrin complexes.

1.1.3) Metallochlorin

Metallochlorin is a metalloporphyrin derivative in
which a Cbe bond of one of the pyrrole rings has been
reduced.9 The structure of metallooctaethylchlorin is shown
in Figure 1-3. The reduced Cg;,bond can have the cis or
trans configurations. Metallochlorins are ubiquitous in
nature, occurring as the chlorophyll pigments in
photosynthetic systems”, in marine worm pigments“, and in
the green heme proteins, leucocyte myeloperoxidase”,
sulfmyoglobin”, sulfhemoglobin14 and bacterial heme-d
containing enzymes.”16 Reduction of a Cbcb bond in a
metalloporphyrin to form a metallochlorin lowers the
molecular symmetry from D4h to C2 and has a profound effect
on the entire macrocycle. A comparative study of FeOEP and
FeOEC by X-ray crystallography17 revealed that the average

iron-nitrogen distance is very similar in the two

 

 

 

 

 

Figure 1—2 Structure of Metallo-2,7,12,l7—tetra-
propylporphycene

 

Figure 1-3 Structure of Metallooctaethylchlorin

complexes(1.996i for FeOEP and 1.986i for FeOEC). However,
the macrocycle of iron octaethylporphyrin is planar but
that of iron octaethylchlorin is S4 ruffled (Figure 1-4) .
This distortion, which is due to the decreased aromaticity
of the chlorin, is produced by the increased flexibility of

the macrocycle.

1.2) Electronic absorption spectroscopy

A great deal of structural information can be obtained
from the absorption spectra of modified porphyrin
macrocycles. The object of several spectroscopic theories
has been to explain the optical spectra of porphyrin and
modified porphyrin macrocycles. Figure 1-5 shows the
electronic absorption spectrum of a typical metalloporphyrin
of D4h symmetry, copper octaethylporphyrin, in CH2C12
solution. Two visible bands are seen between 500 and 600 nm,
and the two are separated by ~1250 cm4» The lower-energy
band (called a) is the electronic origin Q(0,0) of the
lowest-energy excited singlet state;18 The higher energy
band (called 8) associated with the Q(0,0) transition
includes one mode of vibrational excitation and is denoted
Q(0,1).318 An exceedingly intense band (called the Soret)
appears around 400 nm. It is the origin B(0,0) of the second
excited singlet state.18 Better resolved Soret spectra
sometimes show another band ~1250 cm4-to the blue, which is
attributed to the optical transition that includes one

quantum of vibrational excitation in addition to the B(0,0)

(a)
FeHOEP

(b)
FeHOEC

 

Figure 1-4 Molecular structure of a) FeII OEP b) Fen one
as determined by x-ray crystallography
(from ref.l7)

Absorbance

 

 

 

001 00°

 

300

l I fi I r T

r
400 500 600
WAVELENGTH, nm

i
700

Figure 1—5 Electronic absorption spectrum of CuOEP
in C82C12 solution

energy and is denoted B(1,0).18 In Dulsymmetry these Q and B
electronic transitions are of ER symmetry and therefore
degenerate and x,y polarized in the plane of the ring. In
order to understand the porphyrin electronic transitions,

Longuet-Higgins et al.19 employed molecular orbital

calculations. Under Dfllsymmetry, they obtained two top-
filled orbitals, an and ah1(with am of lower energy) and
two lowest empty degenerate orbitals of egsymmetry (Figure
1-6). This identifies two electronic transitions, (aux-seq)
and (aml-seg), corresponding to the Q and B bands,
respectively. However, this development could not predict
the correct intensities for the two transitions. The
weakness of this method comes from the neglect of
interaction between the two electronic transitions. Based on
molecular orbital calculations, Goutermanmkm-recognized the
necessity of configuration interaction in order to describe
the porphyrin electronic transitions. The singly excited
configurations (amfeg) and (ahfeg) have the same symmetry
and undergo configuration interaction to produce the excited
states that correspond to the B and Q transitions. The
transition dipoles for the B state are additive and allowed;
hence, the strong Soret absorption. On the other hand, the
transition dipoles in the Q state are forbidden, which
accounts for the weak visible absorption. The energy and
intensity of the B and Q transition in the porphyrins are
also affected by other electronic factors such as: the

central metal ion, metal oxidation state, spin state and

10

 

 

 

 

b1 (32 u) b2(a1u)

Figure 1-6 Porphyrin M.O.s comprising the Gouterman
four oribtal model (from re£.20)

ll

axial-coordination. The metal dw orbitals can conjugate with
the w electron system of the porphyrin macrocycle

(Figure 1—7).

1.3) Infrared and Raman Spectroscopy22a

Molecular vibrational spectroscopy deals with
interactions of electromagnetic radiation with matter that
produce a variation in the vibrational motions of the
molecules. The study of these vibrations is important for
the understanding of several physical properties of
molecular compounds that depend upon the vibrational states
of the molecules, and represents a powerful tool for
molecular-level investigations. The interaction of
electromagnetic radiation with an assembly of molecules can
be analyzed by measuring either the absorption of infrared
radiation or the scattering of light. Although the two
phenomena are controlled by completely different interaction
mechanisms, they are determined by the same types of
transitions between vibrational states. In principle the
infrared spectrum of molecules can be observed either in
absorption or in emission, because molecules can emit the
same radiation that they can absorb. In practice, however,
the absorption spectrum is much more convenient for
laboratory measurements. Figure 1-8(a) shows the molecular
processes involved in infrared spectroscopy. Here, |m> and
|n> represent different quantum states of a vibrational

mode in the ground electronic state of the molecule.

12

.565 E at Ba 5.3 92:8? Batcokaa «5 .8 £8

.3mm «2 :55 8:»: 05 co «Ba 25. co. 2: :o

as are 2.. as. 15.25 25:8 @398 t

cones—35 6ch 2: .8 02¢. 2:2. Em... 2: so 556% r 3:. 563888. Show was @538 05 oz» 3
2.0V cocoa—85 :fiaflsmccoo a? 8895 $3.53 98 A3323 823 365-358 2:35»... 2: .8

9:8 95 .a 38%.: 9! 235.6 stain—on 2:. 3.8% corn—83¢ stag—RUE 8.. Enema—o .335 a. - a games

 

renown
3.39“: M av ....—

.m. a

 

3v

venom

.41.:

E

I
I
I
o

I
O
O

\

oOsuu

I
I

ZHEH Whom

153‘

.52.:

auuegmiau

.i:- 1.5.".

s

s

\

s

. 0 no»:

\

s

\

0

03M

 

0h

 
  

couauanomus.x

 

 

 

"Lu

 

 

nonauaaomus

  

 

 

menu

 

l3

cmfidm

81:5 52:3.
3:258:

o

H
N

 

H

 

E

 

 

 

 

 

3 can

olhi.

ou><

actozaom

guess.

fl III

N

 

 

>5 PE

 

 

---.F

 

>< + ¢>
causes...
8:655...“
o H
d III-ll

N Ill

 

 

 

 

 

.amoonouuoemu

.UowcumcH Am nu“: oeuMAOOnmm mneooum Haasovoz oia Gunman

 

 

 

 

 

 

 

><i¢>
Exam—5.:
8.35 A...
o cE>
H I o r A...
«HI
.2. 83
{11:4... 1.3
Hesuua>
o
.m
eeusum
caisson
3 ca 2..

hmoomouuoomm decounueufls

l4

Transitions between these energy levels occur with
absorption or emission.

According to Bohr's rule,
um = (En - Em) / h (eq. 1-1)

where h is Planck's constant andlfiw Eh are the energy state
of |m> and |n>. In quantum mechanics a radiative transition

can take place only if the integral

S 41m Iii/«n dr (eq. 1-2)

is not zero, where ¢m,rhlare the wavefunctions associated
with the two states (m and n) and p is the dipole moment
operator. It turns out that, when YR and $5 are harmonic
oscillator wavefunctions, the integral (eq. 1-2) is always
zero unless m - n = :14 Additionally, the vibrational mode
is infrared inactive if there is no change in the molecular
dipole moment during the vibration.

The Raman effect is a phenomenon describing the
inelastic scattering of electromagnetic radiation by matter.
When light scatters off a molecule, the scattered frequency
of light will normally be the same as the incident frequency
(elastic or Rayleigh scattering). With Raman scattering,
incident light of frequency %,will result in a gain or loss
of energy by the scattered molecules. A simplified diagram
describing the Raman scattering process is depicted in

Figure 1-8(b). The designations "Stokes" and "anti Stokes"

15

indicate that the scattered light is of lower or higher
frequency, respectively, than the incident light. At room
temperature or lower, the Stokes Raman scattering will
dominate for most modes, as it originates from the ground
vibrational state. The resonance Raman effect is also
illustrated as a scattering phenomenon in which the incident
photons are of sufficient energy to bridge the gap between
the ground state and the electronic excited states. The
intensity of the Raman scattering process is given by

I. = (12815/9c4nowsi“ E..I<a,.>..l2 (eq. 1-3)

where I0 is the incident radiation intensity, vs:represents

the frequency of the scattered radiation, and q” is a
component of the polarizability tensor (p,a = x,y and z) .
The polarizability tensor is connected with quantum theory
by the Kramers-Heisenberg-Dirac dispersion formula;23 The
polarizability tensor elements in the dispersion formula can

be described as:

<G|up|E><E|pa|F> <G|po|E><E|up|G>
(0%)“ : (l/hmfi { v86 - V0 + iI‘E + VEF + V0 + i1"E (eq. 1-4)

where up and u, are dipole moment transition operators of

 

 

polarization direction p and a. |G>, |E> and |F> are the
wave functions for the ground, excited and final state, and

mm and mm are the frequencies for the transition between
the subscripted states. PE is the transition halfwidth,

which is a function of the lifetime of the excited state

16

|E>. The summation runs over all intermediate states, |E>,
exclusive of IS) and |F>. The first term in eq. 1-4 is
important for the resonance Raman effect. When the incident
frequency (a) is very close to the frequency of the
electronic transition (mm), the intensity will be dominated
by the first term as the energy separation le;"Vo| is
minimized. This resonance condition results in a substantial
enhancement of the intensity of Raman scattering compared to
non-resonance (or normal) Raman scattering.

In eq. 1-4 |G>, |F> and |E> represent the wave
functions associated with the total (vibrational and
electronic) Hamiltonian of the system. By making the Born-
Oppenheimer approximation, the vibronic states can be
represented as products of pure vibrational states and pure
electronic states. As a result, the expression for the

polarizability tensoru'is given by eq. 1-5.

 

 

(aw) = A + B + ~ - - (eq. 1-5)
<n|k><k|m> ]
A — IMGE(RO) '2 2" {Vac ' ”o ' iI‘12:
_ 6&5 <an|k><k|m> + <n|k><lflle>i
B - MGE(R°){<9R Ek{ ”ac - yo - iI‘E

MGE(R0) = (Gil‘aiE>
where |n>, |k> and |m> represent the vibrational levels in
the initial, excited and final state, respectively. A in the

polarizability expression involves no vibronic coupling

17

between electronic manifolds and is known as the Albrecht A-
term. Since the vibrational integrals in the A-term
scattering are simple overlap integrals, this mechanism is
sometimes referred to as Franck-Condon scattering.
Franck-Condon scattering intensity is determined by
three factors: the transition moments may the Franck-Condon
overlap intergals, and the frequency dependence of the
resonance denominator. The Franck-Condon factors are only
non-zero if the orthogonality is removed between a ground
state vibrational level and an excited state vibrational
level. The removal of orthogonality can be achieved by a
difference of vibrational frequency between the ground and
excited states or a displacement of the potential energy
mininum in the excited state. Because it is usual to assume
a common frequency for ground and excited states,
vibrational frequency differences between ground and excited
states are generally thought to be ineffective in producing
non-zero Franck-Condon factors. Accordingly, displacement of
the potential energy mininum in the excited state is the
predominant cause of A-term intensity and therefore only
totally symmetric modes give rise to A-term scattering.%‘
The B term is called the Albrecht B-term or Herzberg-
Teller scattering term. Scattering via this mechanism
results from vibronic borrowing of intensity between the
resonant electronic transition and a nearby electronic

6 .
transition through {gfim} . Both symmetric and nonsymmetric
k R0

18

modes can be enhanced by B-term mechanisms. But the
magnitude of B-term enhancement of symmetric modes is small
compared with that of A-term enhancement. In the B-term
mechanism, the harmonic oscillator matrix elements <n|R|k>
6mkfl_, i.e., modes along which there is no displacement of
the potential energy mininum in the excited state will have
non-zero integrals. Therefore, non-totally symmetric modes
are enhanced by the B-term mechanism;25

Resonance Raman studies of porphyrins have revealed an
important correlation between the spin and the oxidation
states of the central metal. The resonance Raman scattering
for porphyrins obtained with Soret excitation is dominated
by the Franck-Condom mechanism. The vibrational modes of
metalloporphyrins between 1450 and 1700 cmd-represent
mainly C-C and C-N stretching modes of the
metalloporphyrins. The frequencies of modes in this region
that involve the CaCm and the Cbcb bonds have been shown to
exhibit a linear correlation with core-size (distance
between the central metal and the nitrogen of pyrrole).26"29
The core-size expansion produces a deformation of the C5155
bond angles while the pyrrole rings keep their structural
integrity. The increase of the C5;,bond length is reflected
by a lowering of the corresponding vibrational frequency.
Therefore, modes with cghlstretching character are more
sensitive to core-size change than are the modes with

primarily C§;,stretching character. Determination of the

core-size sensitivity by using resonance Raman provides a

19

useful means of assigning vibrational modes in the
metalloporphyrin systems. For the CaN symmetric vibration in
the 1355 - 1375 cm4~region, the frequency is sensitive to
electron density in the porphyrin 1” orbitals. Therefore,
the frequency of the CaN symmetric vibration is determined
to a significant extent by the oxidation state of the metal
and the presence of axial ligands.

The resonance Raman scattering for porphyrins obtained
with visible excitation are dominated by the Herzerg-Teller
mechanism owing to mixing and borrowing of intensity between
the resonant electronic transition (Q band) and the Soret
bandfi”*Therefore, the totally symmetric modes, which give
rise to polarized bands, are weak or absent in visible
excitation spectra. On the other hand, non-totally symmetric
modes, which show depolarized or anomalously polarized

bands, are strongly enhanced.25

CHAPTER 2

MATERIALS AND METHODS

2.1) Preparation of Porphycenes

All of the porphycenes used in the studies reported
were prepared by Weishih Wu (Michigan State Univ.) according
to the procedures below.

The free base 2 , 7 , 12 , 17-tetrapropylporphycene (HZPPC)
was made according to the method of Vogel et al.4 The metal

insertion procedures for NiII and Cup were described
earlier.30 Selective deuteration at the Ce position (Figure
1-2), NiPPC-9,10,19,20-d4 (abbreviated NiPPC-d4(et)), was
synthesized according to the published method.-30 Deuteration
at the Cb. positions, NiPPC-3,6,13,16-d4 (NiPPC-d4(py)) was
accomplished by refluxing HZPPC deuterated acetic acid in
the presence of nickel acetate and sodium acetate under
nitrogen atmosphere for 8 hours.
3,6,13,16-tetrabromo-2,7,12,17-tetrapropylporphycene,
which involves bromination at the CW positions (Figure 1-
2), was achieved by the following method. A solution of
bromine in chloroform was slowly added to a stirred
solution of the free base porphycene. The yield of mono-,
di-, tri-, and tetrabromoporphycene depended on the amount

20

21

of bromine added. The reaction was easily monitored by TLC
developed with hexane/CHZClz. The Ni(H) and Cu(II) complexes
were prepared in the same manner as described above.

In order to synthesize ironflfl)'tetrapropylporphycene
chloride (ClFemPPC), ferric chloride (30 mg) and sodium
acetate (5 mg) were added to a solution of the free base
porphycene (30 mg) in acetic acid (20 ml) and refluxed for
10 hours. The organic layer was separated, washed with water
and evaporated. The residue was run on a silica gel column
and eluted with methylene chloride to first remove by-
products and then with MeOH(2 %)/CH2C12 to collect the dark-
green iron(III) complex. (Im)2FemPPCCl was produced by
titration of the ferric chloride complex with N-
methylimidazole (Im). To reduce the (ImnzFemPPCCl
chemically, a CHZCl2 solution of (Im)2FemPPCCl was made
anaerobic by several freeze-pump-thaw cycles and titrated

with hydrazine hydrate.

2.2) Preparation of Chlorine

All of the chlorine used in the studies reported below
were prepared by Einhard Schmidt (Michigan State Univ.).
Chlorine were prepared by the method of Whitlock et al.“:
following the reduction reaction, residual porphyrin was
removed by alumina column chromatography with benzene as

eluent. HZOEC-d4 was prepared by treatment of HZOEC with
DZSO4/D20 (9:1, v/v) . Reduction of the DZSO4/DZO ratio to 6:1

(v/v) afforded HZOEC-7,6-d2. Re-exchange at the 'y and 6

22

positions of HZOEC-d4 with stO4/H20 (6:1, v/v) produced
HzoEc-afl-dz. The deuteration process was followed and
confirmed by'IHNMR by using a Varian VXR-300s NMR
spectrometer. Metal insertion was achieved by dissolving the
chlorin in CI*12Cl2 and adding an appropriate volume of
saturated zinc or copper acetate in MeOH. The process was
followed by UV-vis spectroscopy with a Shimadzu UV-160

spectrometer.

2.3) Spectroscopic Studies

UV-visible absorbance measurements were recorded on a
Perkin Elmer Lamda 5 UV/Vis Spectrometer. Optical spectra
were recorded before and after the Raman experiments to
confirm the stability of each particular complex to the
laser irradiation employed.

Spectrophotometric grade CHCl3 and CH2C12 were purchased
from Aldrich company. Both solvents were purified by the
method of Perrin & Perrin32 in which the solvent of interest
is stirred with concentrated Hgflh overnight. The organic
layer is washed with H20 twice, 10 % NaHCO3 once, and again
washed with H20 twice. The solvent is dried over CaClz, and
then distilled from ng.

Infrared spectra at room temperature were recorded on a
Nicolet IR/42 FTIR spectrometer. Samples were prepared by
saturating CHCl3 with the various metal chlorins, then
running the spectra in a son pathlength NaCl cavity cell

obtained from Spectra Tech, Inc.

23

Resonance Raman spectra were obtained on a computer-
controlled Spex 1401 Ramalog with PMT detection, a Spex 1877
Triplemate with EG&G Model 1420 detector and OMAII
electronics, and a Spex 1877 B outfitted with a EG&G Model
1421 detector and OMAIH computer. The laser systems include
a Spectra Physics Model 164 argon ion and model 375 dye
laser (to provide wavelengths of 630.0 and 613.0 nm), a
Coherent Innova 100 argon (363.8 nm) and an Innova 90
krypton(568.2 and 406.7 nm) ion laser, and a Quanta Ray DCR-
2A Nd:YAG and POL-2 dye laser combination (390.0 and
355.0 nm) . Spectra were obtained for CHzCl2 solution of the
metallochlorins or porphycenes, which were contained in a
cylindrical quartz spinning cell. Long exposure (more than
20 min) of the metallochlorins to high power laser beams
(e.g., 2: 15 mW at 406.7 nm) caused sample degradation.
Therefore, the Raman spectra of the metallochlorins were
collected within 1 minute of initiating low power laser
irradiation (e.g., - 7 mW at 406.7 nm) with a Spex 1877
Triplemate with a EG&G Model 1420 diode array detector in
order to avoid sample degradation. All of Raman spectra were
taken under aerobic conditions at room temperature.

Laser powers and concentrations used for specific

spectra are noted in the figure captions.

CHAPTER 3

ELECTRONIC AND VIBRONIC PROPERTIES OF
METALLOPORPHYCENES

3.1) Introduction

Porphycene is a structural isomer of porphine, the
parent compound of porphyrins. Porphycenes qualify as
potential agents for tumor marking and photodynamic therapy
owing to their high absorption intensity in the wavelength
region above 620 nm, in which body tissue is translucent,
and to their stability towards photooxidation and singlet
oxygen sensitization.33'3S For these reasons the electronic
properties of porphycenes are of interest. Figure 1—2
represents the structure of 2,7,12,17-tetrapropylporphycene.

Several groups have executed systematic studies of the
effects of symmetry lowering on the macrocyclic vibrational
properties of metallooctaethylchlorins(MOEC)fihmrm and
metallopheophorbides39 to understand vibrational spectra of
naturally occurring pigments (e.g. chlorophyll) possessing
tetrapyrrole-derived structures of symmetry lower than that
of metallooctaethylporphyrin. Similarly, the spectroscopic
properties of metal complexes of bacteriochlorins and

bacteriopheophorbides“*“~should progressively approach

24

25

those of bacteriochlorophyll. In this spirit,
metalloporphycene represents an example of a macrocycle of
true Dfilsymmetry which is approximated by heme a42 and
metallobacteriochlorins.41 Furthermore, the bipyrrole cha.
bond is a structural feature of the corrin macrocycle; thus,
cobalt complexes of porphycene derivatives may model some
aspects of vitamin Bu. Renner et al.330 have characterized
the NiPPC anion radical and compared the redox properties of
this compound to other nickel porphinoid models of coenzyme
F430 from methanogenic bacteria.

Resonance Raman spectroscopy (RR) has been applied
extensively to synthetic MOEP“, MOEC%”“*”,
metallobacteriochlorins“*“, and metallocorphinoids.“-It has
been used to study redox-altered states of MOEP and MOEC
including 1 cationflmw and anionso’52 radicals. The studies of
these model compounds have contributed to our understanding
of resonance Raman results from a number of important
biological systems including heme proteins”, reaction
center chlorophyll and bacteriochlorophy115+55, coenzyme
Egy§957 and vitamin Bup5959.A range of information regarding
structure and electronic properties can be offered by the
resonance Raman technique. Moreover time-resolved resonance
Raman techniques have provided substantial insight in the
study of dynamic processes. Interpretation of results from
complex biological systems and from sophisticated techniques
requires thorough grounding in basic spectroscopic study.

For these reasons the UV visible and resonance Raman spectra

26

of metal-substituted and selectively deuterated porphycene
compounds are investigated and compared to the well known

spectra of analogous octaethylporphyrin species.

3.2). Results

3.2.1) Electronic Absorption Spectra
Figure 3-1 shows the electronic absorption spectrum of CuPPC
and, for comparison, the spectrum of the analogous porphyrin
compound, CuOEP (inset). Superficially, the optical spectrum
of CuPPC is similar to that of the porphyrin series.
However, the oscillator strength ratio, which was determined
by measuring the areas under the Soret and visible bands for
the absorption spectra plotted in a wavenumbers scale, of
CuPPC is much greater (more than 5 times) than that of
CuOEP. Additionally, the near-UV band of CuPPC is noticeably
split into two components. Compared to that of CuOEP, this
band is blue shifted and the visible band is red-shifted in
the CuPPC spectrum. Figures 3-2 and 3-3 show absorption
spectra of NiPPC and CuPPC. In order to demonstrate the
spectral features of these compounds more clearly, the first
derivative of each spectrum is superimposed. Table 3-1
collects the band positions determined by inspection of
expanded spectra, as well as of first and second derivative

spectra of NiPPC and CuPPC. Also included are hand positions
for the CuII complex of 3,6,13,16-BryPPC (i.e. Br is

substituted for H at the CV positions, Figure 1-2). These

are compared to absorption maxima for the following

27

Figure 3-1 UV-visible absorption spectra of CuPPC
and CuOEP (inset) in CH2C12 solution.

 

 

28

ABSORBANCE

 

now

«me

 

995m

 

 

 

 

 

Octto

 

OcOmn

0..

a;

 

 

 

4 d

«a:

1

4

I

d

a
5:0

1

J 1“ q

as
<<><mrm2042

 

29

Figure 3-2 UV-visible absorption (dashed curve) and first
derivative (solid curve) absorption spectra of
CuPPC in CH2C12 solution.

Emma. Om3_<).=<m
. O J

 

 

 

30

CuPPC

600

 

 

 

 

 

 

_""

 

 

 

 

 

 

 

moz<m¢Omm<

700

(nm)

500

WAVELENGTH

400

300

 

Figure 3-3

31

UV-visible absorption (dashed curve) and first
derivative absorption (solid curve) spectra of
NiPPC in CH2C12 solution.

 

32

Emma. Om2<>fl<m

 

 

 

NiPPC

 

500

WAVELENGTH

 

 

 

 

 

 

 

 

moz<m¢0mm<

 

600

400

("I“)

33

Table 3-1. UV-Visible Absorptions (nm) of Metalloporphycenes
and Metalloporphyins and Assignment of Electronic
Transitions of MPPC

 

 

NiPPCa CuPPC Cu-BrpRPC one electron w symmetry
601.5 613 631 blub2g B3u(x)

563 (sh) 574 586 bmb3g B2u(y)

388 387.5 388 aub2g Bzu(Y)

369 (sh) 368 (sh) 369 (sh) aub3g B3u(x)
NiOEPb CuOEP NiDPDMEC Ni-2,4-Br5DPDME
551 561 549.5 553

516 525 514 518

391 397 390.5 395

 

a*Band positions of CH2C12 solution were determined by

inspection of expanded spectra, as well as from first
and second derivative spectra.

bBand positions of CHZCl2 solution were determined from
multiple runs of each absorption spectrum.

CBand positions of CHCl3 solution as reported in
ref. 60, 61.

34

porphyrin complexes: NiOEP, CuOEP, NiDPDME, and Ni-2,4-
BrDPDME (DPDME=deuteroporphyrin EX dimethyl ester). Table 3-
1 illustrates that a change in the central metal or
bromination at the ring periphery significantly shifts both
the near-UV (Soret) and visible bands in the porphyrin
complexes; however, these substitutions in the porphycene
complexes shift only the visible bands and leave the near-UV
band positions virtually unchanged. This illustrates a basic
difference in the electronic properties of
metalloporphycenes and metalloporphyrins and will be
discussed later in this chapter.

Figure 3-4 illustrates the UV-visible spectral changes
manifested during conversion of the five-coordinate ferric
chloride complex to a 6-coordinate ferric bis(N-methyl
imidazole), (Im)2, complex. The spectrum of (Im)§FePPCCl
(Figure 3-4) is quite similar to that of CuPPC (Figure 3-1),
however, the structure in the near-UV band is less well
resolved. The ClFemPPC spectrum is more complex than that
of (Im)2FemPPCCl. Reduction of the (Im)2FemPPC to (Im)2FeHPPC
with hydrazine hydrate generates the UV-visible spectrum

inset in Fig.3-4.

3.2.2) Resonance Raman Spectra

Fig.3-5 represents the Raman spectra obtained in
resonance at 363.8 nm of CuOEP and CuPPC. Like the optical
absorption spectra, there are superficial similarities. Both

are dominated by intense features in the 1300-1650 cm”-

Figure 3-4

35

UV-visible absorption spectra of FePPC
derivative in CHZCl2 solution. The ferrous
sample was prepared using hydrazine hydrate
as a reductant as described in MATERIALS

and METHODS. The absorption at 280 nm in the
spectrum of the (Im)?FemPPC is from excess
N-methylimidazole.

H19N313AVM

36

ABSORSANCE

 

 

 

 

 

 

Figure 3-5

37

Resonance Raman spectra of CuOEP and CuPPC
in CH2C12 solution at concentration of
approximately 0.1 and 0.2 mM, respectively.
Laser powers at 363.8 nm were 20 and 10 mW,
respectively. Asterisks(*) in Figures 3-5,
6, 7 and 8 represent solvent vibrations.

38

Aex = 363.8 nm

RSI 3.... ML
8.... .i .... 33! can—....
33! eon—l
e nae—l o—v—l.
22 i 39!
Snpl m8.

FSPIIIIHH

 

 

inch-lulu

 

 

 

 

snow IHH

 

 

 

SSI I
II to: II
P 3: l C 2: ....
E P
0 Re. I P
U U can I:
c ... .. c ... ..
) \I
a b a; .I
s 6'5 .I. O '
C 'A
.5 i
son I:
can H
8.... 3.
can ....
Sn I
sill-Ill”
hum ..l

@

>._._mzm._.z. 2<3<m

I

r

f

r

‘—

r

U

200 400 600 800 1000 1200 1400 1600

 

RAMAN SHIFT (cm-1)

39

region. In the CuOEP complex these normal modes involve C-C
and C-N stretching coordinates. Their large intensity is
attributed to large displacement along these normal
coordinates in the 1 arm} excited states of the porphyrin
macrocycle. The frequencies of these vibrations are
structure sensitive, i.e. they are dependent on the size and
redox state of the central metal. The normal coordinate
analyses‘50"64 of NiOEP and CuOEP indicate that the dominant
Raman-active high frequency normal mode of CuOEP (Figure 3-
5) include two primarily'uCJIIStretches, mfl(1637 cm4y tag)
and V3( 1503 cm‘l, alg) , and two primarily VCbe stretching
motions, v2(1591 cm‘l, alg) and V11(1568 cm'l, blg). The V4(1379
cm‘l, alg) mode of metalloporphyrins is predominantly a vCaN,
6CJE‘"breathing" motion of the macrocycle.

Figure 3-6 shows polarized RR spectra from CuPPC that
were obtained by using laser excitation at 406.7 and
613.0 nm. Similar spectra were measured with Am,==568.3 and
630 nm, and depolarization ratios for the most intense bands
are collected in Table 3-2 in order to determine the
symmetry of the principle vibrations of the CuPPC spectrum.

Figure 3-7 shows the RR spectra of NiPPC, NiPPC-d,(py),
and NiPPC-d,(et). The latter two complexes have deuterium
substituted for hydrogen at the Ca pyrrole positions and at
the C; ethylene positions, respectively. The frequencies of
the Raman bands are collected in Table 3-3 along with those

measured for CuPPC and CoPPC.

 

Figure 3-6

40

Polarized resonance Raman spectra of CuPPC
in CH2C12 solution at 406.7 nm and 613.0 nm.
Concentrations were approximately 0.5 mM and
0.3 mM and laser powers 40 and 100 mW,
respectively. In each spectrum, the top
trace represents I” and the bottom trace
represents Ii.

41

can. I

02:. I

 

CuPPC

 

....-ic

Com—I

0.9— I
86— I

.8. I I!

ovu— I

A033 406.7 "m

 

 

 

 

 

>._._mzm._.z. z<3<¢

 

m an"
..m 32I
m hoopll
1| «3"
gpl.
rm m
Opp-I r o.
m
em 6
1 I .
H
r O
«A
a.
rm 5
v 000'
J 1m
6

 

RAMAN SHIFT (cm-1)

42

Table 3-2 Resonance Raman Depolarization Ratios, p=Ii/IH,

 

for CuPPC.
Raman shift Excitation Wavelengths, Aw (nm)
(cmqj 406.7 568.2 613.0 630.0
1590 0.3 0.2 0.2 0.3
1559 0.4 0.5 0.5 0.4
1506 0.4 0.3 0.3 0.3
1417 0.5* 0.4 0.5 0.5
1363(w) 0.1 (w) 0.4 (w)
1326 (w) (w) 0.4 (w)
1301 0.5 0.4 0.5 0.4
1240 0.2 (w) 0.2 0.3
1165 0.6* (w) (w) (w)
1116 0.3 0.3 0.4 0.4
1065 0.5 (w) (w) (w)
996 0.2 0.3 0.3 0.3
936 (w) 0.6 1.0(w,?) 1.0(w,?)

 

*; Indicates that the p value may be anomalously high due to
contribution from an underlying depolarized solvent band.

w; Weak intensity making measurement of p unreliable.

43

Figure 3-7 Resonance Raman spectra of NiPPC derivatives
in CH2C12 solution at concentration of

approximately 0.2 mM. Laser power at
363.8 nm was 30 mW.

44

 
      

 

 

 

 

 

 

 

 

 

 

 

 

 

0
0
25
1
an: I «an. I
:9 I at: I - moo— I rm
3...? 1
m . 8! II
n 83 .
3 82 I 3.... i 2....— 32 I 1.
m w ...... ) ,m
3 D. .. Stl .... 2
= l\ «N: l\ 1
C 4 33.... 4 O
O. . 0
1 I .3 I C C 1
N P maul P
3. I W W r
e ask I N e I N e
e nan I Hill e0 Hi I H
08 I N8- 8......

T

trial ..I , eI

 

200 400 600 800

r

>2m2mhz. z<3<¢

RAMAN SHIFT (cm-1)

 

 

45

Table 3-3 Raman Vibrational Frequencies (cm4) of
NWPPC Complexes.

aThe CoPPC spectra (not shown) indicated some
residual free-base present in the sample.
These contributions may skew the frequencies
presented here despite efforts to substract
them from the data.

bThese frequencies most likely represent
normal mode of significantly different
character than those of the natural
abundance sample.

SC means small .

IF means large.

46

 

CuPPC CoPPCal NiPPC NiPPC-d4(py) NiPPC-d4(et) Assignment

1590 1612 1625 1620 1625 uCC(py)
1559 1571 1579 1574 1565 uCC(et,py)
1506 1515 1520 1509 1518 VCC(py)
1417 1407 1410 1403 1357 yogi” 6CJ1
1363 1368 1368 1368(w) 1368(w) ucgq
1326 1327 1327 1327 ucgq
1301 1299 1305 1302 1305 VCatN
1270b
1240 1231 1227 1187 5CeH(SC %)
1165 1164 1167 925 1167 5Cb.H(Ld %)
1116 1120 1122 1118
1065(w) 1064(W)
1085b
1016b
996 995 996 996 Vcb_propyl
936 944 944 931
848 846 ~850(w) 813(w)
607 607 608 602 606
528
494(w)

227

 

47

Figure 3-8 compares RR spectra of the five-coordinate

ClFemPPC and six-coordinate (IM)2FeHPPC. The R spectrum
excited at 390 nm of the reduced product appears in Figure
3-8c, while the spectrum at kw:= 355 nm of the same
compound in the presence of excess reductant appears in
Figure 3-8d. The spectra are consistent with metal-centered

reduction.

3.3) Discussion

3.3.1) Electronic Properties

In order to understand the electronic properties of the
porphycene macrocycle and to compare them to the porphyrins,
the UV-visible absorption spectra of four coordinate,
divalent, transition-metal complexes that exist in a single
spin state and that do not exhibit metal-macrocycle w
bonding will be discussed.

The UV-visible absorption of the metalloporphyrins have
been the subject of extensive study for many years. They
arise for the most part from in plane I e'w’ transitions of
the macrocycle which have been described by the four-orbital
model of Gouterman.20’21 His model came from earlier attempts
to explain the w-electron spectra of large conjugated ring
molecules.1$65‘8'The model adequately explains the effects
of variation of the central metal and peripheral
substituents on the electronic structure of the macrocyclic

complexes. Comparisons of these effects in metalloporphyrin

48

Figure 3-8 Resonance Raman spectra of FeIII and Fe11
complexes of PPC. Concentrations in
CH2C12 solution were approximately 0.1 mM and
laser powers typically 20 mW. Spectrum a was
measured with a different diode array detector
than spectra b-d.

 

 

 

49

«an. I

    

....V

 

 

 

 

 

C
n. » ...
.. m
u.
..3. ”m m 3
.. ) a
on Mu mm hm
...... I I M
b m: fl

>h.m2mh,= 2(31!

 

HAMAN SHIFT (cm'I)

50

spectra to the corresponding effects in metalloporphycene
reveal fundamental differences.

The differences in the electronic properties of
metalloporphycene from those of metalloporphyrin are the
consequence of the difference in point group symmetry.
Figure 3-9 depicts an orbital energy level diagram for both
metalloporphyrin (Dag and metalloporphycene (Dug. For
metalloporphyrin four essentially degenerate one electron
excited state wave functions are heavily mixed by
configuration interaction (CI) resulting in linear
combinations usually referred to as the Q (visible) and
B (Soret) states.20‘21 These are of Eu(x,y) symmetry. The Q
state transition is formally forbidden but gains intensity
through vibronic coupling to the B state, which represents
an allowed transition. In Dfllmetalloporphyrins the x and y
components of the Q and B states are degenerate, and each
total wave function involves all four one-electron wave
functions. The configuration interaction comes about
primarily because the HOMO energy splitting is small
compared to the electron interaction energy. The latter can
be estimated by the observed energy difference between the B

and Q transition energies and, from the data in Table 3-1,

is ~7300 cm4-for CuOEP. Shelnutt69‘7O estimates the energy

difference between the alu and an orbitals of CuOEP to be

~1700 cm4a As a consequence of the electronic

configuration, the spectral effects of substitution at the

metal center or at the pyrrole periphery change both the Q

51

Figure 3-9 Schematic orbital energy level diagrams
for CuOEP(D4h) and CuPPC(D2h)

52

 

 

 

 

 

 

 

 

 

e. e, b .
. LUMo ”7‘? “250cm“
I bJ‘T '
E x Y Y x x Y Y x 463000111"
mu
:2
In
b..
17000!!!" {J a; HOMO ' 9600 cm"
a...
a. J
CuOEP

CuPPC

 

53

and B transition energies. On going from NiOEP to CuOEP

there is an expansion of the porphyrin core that results in

an increase in the energy of theIam_orbital relative to the
am_and LUMO orbitals““m; thus, a red-shift occurs.
On the other hand, the degeneracy of the LUMOs and the

near degeneracy of the HOMOS are lifted for

metalloporphycene. For CuPPC it was estimated that a ~1250

cm"1 gap occurs between the b2g and b3g orbitals and a

 

splitting of ~9600 cm'1 between the au and blLJl orbitals based

on the optical spectrum (Figure 3-1 and Table 3-1) and
schematic orbital energy level diagrams (Figure 3-9). This
results in four allowed transitions, each involving
essentially one-electron excited-state wave functions. Table
3-1 assigns these transitions to the metalloporphycene
absorption spectra. According to this model, the effects of

substitution are easily understood. Table 3-1 and Figure 3-
9 show that the energy gaps between bluandb3g orbitals for
NiPPC and CuPPC are ~1776O cm"1 and ~17550 cm4,

respectively, and that the energy gaps betweenb1U and b29

orbitals for NiPPC and CuPPC are ~16630 cm"1 and ~16300 cm4,

respectively. These mean that NiII substitution for CuII

lowers the relative energy of thelbm_orbital by ~300 cm”-
and causes a blue-shift of both the anx0 and Bmxy)
transitions that correspond to the bhgbg and bhghg one-

electron excited-state wave functions, respectively.

Comparison of CuPPC with Cu-BrQRPC reveals that

tetrabromination at the CW positions raises the bwlenergy

 

54

level by ~400 (210‘1 relative to that of the au orbital while

causing no change in the energy level of the LUMOs,
accounting for the observed red-shifts. Thus, the relative
energy level of theblu orbital (which corresponds to the a2u
orbital in Dalthe symmetry)“~increases in the series NiPPC,
CuPPC, and Cu-BrgPPC. This trend most likely originates in
the expansion of the macrocycle.

Increased core size on going from NiPPC to CuPPC is
analogous to the OEP complexes. Also, tetrabromination
increases the macrocycle core size because the bulky Cb
substituents interact sterically.8.Although metal
substitution and bromination of the ring may well change the
absolute energies of all orbitals, the relative energies of
the au or the LUMOs are not influenced by substitution. As a
result the near-UV transition energies do not change.72
Accordingly, in the absence of strong configuration
interaction, the MO energy level diagrams can be estimated
directly from the absorption spectra. The large splitting
(~9600 cm4) of the HOMO levels cause the absence of
configuration interaction (CI) between the near-UV and
visible transitions of this symmetry. The gap between the
LUMOs is smaller(~1250 cm4); however, the orthogonality of
each of the components of the visible and near-UV
transitions also precludes strong CI effects.

Earlier, Pla’c't'iw'66 and co-workers pointed out examples
of what they termed "long-field" molecules in which CI was

minimized. These include pyrene, biphenyl, and

55

tetrahydroporphine. Although Platt61 speculated that there
is most likely no pure long-field molecule,
metalloporphycene may represent the best example of such

thus far encountered.

3.3.2) Symmetry and the Resonance Raman Depolarization
Ratio

The skeletal structure of metalloporphycene belongs to

 

the 02h point group (Figure 1-2) . The x and y symmetry axes
are oriented through the bridging positions, not through the
pyrrole groups, as in the metalloporphyrin (Dug. For the
37-member structure in which all substituents are considered
point masses, 51 Raman-active and 45 IR-active fundamental
vibrations should occur. In the resonance Raman spectrum,
only in-plane w e’w” electronic transitions of the
macrocycle are expected to be observed. Thus the number of
in-plane Raman-active fundamentals should be 35 modes (18 ag
and 17 big) . These modes, ag and big, can be readily
distinguished by measurement of the Raman depolarization
ratio, p. For totally symmetric ag vibrations, p will be
between 0 and 3/4. For blgr modes, p 2 3/4 ; that is, the
latter modes can assume anomalous polarization if the
scattering tensor is a mixture of symmetric and
antisymmetric contributions.”73 Table 3-2 shows that,
although some differences in the relative intensities occur,
the p values of the major RR bands observed are between 0.2

and 0.5. Thus, only polarized modes may be observed for the

56

metalloporphycene and the distinctive inverse or anomalous
polarization phenomenon present in the RR scattering of the
metalloporphyrin74 is not observed. Based on the
depolarization ratios, all of these vibrations are assigned

to the totally symmetric representations ag.

3.3.3) Vibronic Coupling

Although the one-electron w e'm” states of the

 

metalloporphycene do not show strong electronic coupling,
i.e., they exhibit weak configuration interaction, vibronic
coupling is still conceivable. Under D2h symmetry B2u x B3u =

b Thus the b1g modes can couple these electronic

lg°
transitions via a Herzberg-Teller (HT) mechanism. Electronic
transitions of like symmetry can couple via ag modes; thus,
it is possible for totally symmetric modes to exhibit HT
activity. Indeed, vibronic activity of‘am,modes has been
reported for metalloporphyrins.75‘78 The relative oscillator
strengths of the visible and near-UV bands apparent from the
absorption spectrum in Figure 3-1, however, imply an absence
of HT coupling in MPPC compounds. In the case of
metalloporphyrins, HT coupling of a formally forbidden
transition (the Q band) to an allowed one (the Soret band)
results in a smaller ratio of visible to near-UV oscillator
strengths.“*“-But resonance Raman depolarization ratios for
metalloporphycene do not exhibit any anomalously polarized
modes (Table 3-2). Thus there is no conclusive evidence of

HT coupling and the scattering derives primarily from an

57

Albrecht A-term (Franck-Condon) mechanism. Since no

degenerate representations are contained in the D2h point

group, Jahn-Teller (JT) coupling is not expected. In view of
the above, all of the 18 vibrations listed for CuPPC in
Table 3-3 are assigned to the totally symmetric

representation, a. Owing to the apparent sparseness of

g.
vibronic coupling, the spectra of NiPPC and CuPPC are

remarkably simple.

 

This situation in metalloporphycene is in contrast to the
case of the metalloporphyrins, where the CI-induced excited
states are extensively affected by both JT and HT couplings.
As a result, the RR spectra of the MOEP class are
exceedingly rich in vibronic information.”i7*”8

3.3.4) Effects of Selective Deuteration and Metal

Substitution on Raman Frequencies

In the region above 1400 cmq-the intense RR bands of
Figure 3-7 are clearly isotope sensitive and easily
correlated. The 5-11 cm"1 shifts upon d4 (py) but not d4(et)
substitution obviously indicate that the 1625 and 1520 cm—1
vibrations of NiPPC involve the CW atom and not the
ethylene bridge carbons(Ce).‘Thus, they are most likely
localized on the pyrrole rings and/or across the Caxg, bond.
On the other hand, the features at 1579 and 1410 cm4-shift
in both isotopically substituted species. This suggests
delocalized motions involving both pyrrole and ethylene

bridge carbons. The mode at 1410 cmA-shifts more

dramatically upon d40et) substitution and is hence assigned

58

as a large percentage of uCeCe and possibly some 6CeH
character. The bands in the 1300-1400 cm4-region shift
little upon deuteration at the outer ring carbons and are
assigned to C-N stretches.

Below 1300 cm4-the Raman features are less intense and
the normal modes likely undergo changes in composition upon
deuteration. This makes it difficult to correlate features
and to ascertain the internal coordinates involved based
purely on isotopic shifts. In the d,(py) derivative,
features appear at 1270, 1085, 1016 and 925 ch-that are
not easily associated with vibrations of the natural
abundance sample. This is also true for the 1187 cm4-
feature in the dqoet) sample. Some suggestions are presented
in Table 3-3 but these must be considered tentative. For
example, the 925 ch-feature in the NiPPC-d,(py) spectrum is
most likely due to a 5CbD motion. In NiOEP, NiTPP, and CuTPP
(TPP=tetraphenylporphyrin) the 6CmH and 6CbH motions,
respectively, can cause shifts upon deuteration ranging from
30 to 400 cm4, depending on the potential energy
distribution(P.E.D.) of the normal mode.60"64 Hence,
correlating the 925 cm'1 feature with the 1167 cm"1 band
present in both NiPPC and NiPPC-d4(et) is plausible. The
1410 cm4-vibration of NiPPC can be compared to a mode at

1504 cm"1 of both CuPPC and NiTPP that has been assigned by

Atamian et al.63 and Li et al.64 to a mixture of “15% and
6CbH motions. The deuterium shifts of 44 and 47 cm'1 reported

for these porphyrin vibrations63"64 are similar to the 53 cm”-

59

which is observed for NiPPC upon deuteration at the ethylene
bridge carbons. CuTPP and NiTPP also have a mode around 1375
cm"1 displaying approximately a ~40 ch-deuterium shift.
This mode is likely a mixture of the vCaCb and CaN or 6CbH
coordinates.“”“. By analogy to these assignments, based on

the frequency and isotopic shift, the 1410 cm4-feature of
NiPPC is assigned to the smaller ~7 cm"1 shift on d4(py)

substitution, possibly some ”Ca.Cb. or uCa.N motion. Finally,
the 996 cmq-mode probably involves the pyrrole substituents
similar to the 1025 cm4-vibration in NiOEP.“62
Incorporation of deuterium at the CV position most likely
causes a change in normal mode composition resulting in a
new vibration at 1016 cmfl-(and possibly also at 1085 cm4).
This could well result from coupling of the &;,D vibration,
which has shifted into the 900 - 1000 cmfi-region, with the
996 cm‘1 mode. The 0Cb.D could then derive some intensity
from the 996 cm4-vibration resulting in the features at 925
and 1016 cm‘1 observed in the spectrum of NiPPC-d4 (py) .
Similar coupling may also occur for the 944 cm4-vibration,
as it cannot be found in the NiPPC-dApr) spectrum. This
vibration undergoes an isotopic shift upon d,(et)
substitution, whereas the 996 cm4-vibration does not,
indicating that they are localized in different areas of the

macrocycle.

60

3.3.5) Structure Sensitive Vibrations

In metalloporphyrins the vibrations in the 1400 - 1700
cm4-range are distinctly dependent on the central metal.
This is due to variations in the size of the porphyrin core
in response to metal substitution. Spaulding et al.79
identified the relevant structural parameter to be the

center to pyrrole nitrogen distance, dQN, Huong and

PommierM’quantified the relationship between vibrational

 

frequency and daN, and subsequent work extended the
correlation to a large number of vibrations of a variety of
metalloporphyrin?h“*“-complexes. Because the relationship
between frequency and core size is dependent on the P.E.D.
of the vibration, such analysis has been used to support
vibrational assignments in metallochlorins?”38 and in the «-
cation radicals of metalloporphyrins and metallochlorins.46
Comparison of the crystal structures of NiII and MgII
porphyrinsaz‘83 shows that the CaCm and Cbe bond lengths
change by 0.044 and 0.014 A upon expansion of the porphyrin
dc,N from 1.958 to 2.055 A. The lengthening of these bonds
decreases vcacm and vaCb frequencies. Thus, those normal
coordinates involving uCaCm motion, followed by those
involving “QIH motion, show the steepest inverse slopes in
the relationship between frequency and core size V = K(A-
01810.8

Table 3—3 shows that the metal dependence for the

O

vibrational frequencies of PPC is similar to that of the

analogous OEP complexes. The vibrations at 1625, 1579, and

61

1520 cmq-in NiPPC decrease by 35, 20, and 14 cm4,
respectively, in CuPPC. This suggests that these vibrational
frequencies are structure sensitive. The structural
parameter that determines the vibrational frequencies cannot
be identified with certainty until more crystal structures
are determined. Correlation of vibrational frequency to
metalloporphycene structure will be complicated by the
apparent dependence of the core skeletal structure on the
peripheral substituents.8'This is in contrast to the case
for metalloporphyrins“, where the relative invariance of
the core size of different porphyrins complexed to the same
metal has been exploited extensively in establishing
correlations of vibrational frequencies to dCtN.79

Given the above cautions, a preliminary analysis of the
available structural and vibrational data suggests that the
three highest RR frequencies of the PPC macrocycle are,
nevertheless, approximately linearly dependent on the
center-to-nitrogen distance. Table 3-4 collects the
structural data for PPC compounds and compares them to those
of the analogous porphyrins. Measurements of HZPPC, HZPPC-
d4(py) , and HZPPC-d4(et) reveal that the three highest RR
frequencies of the free base PPC at 1606, 1559, and
1505 cmq-derive from normal modes with qualitatively
similar P.E.D.'s to those of the metalloporphycene
complexes. Thus, these points can be used to estimate K and

A parameters for a structural correlation of porphycene

62

vibrational frequency to core size for each of these modes
(Table 3-4).

Comparison of these estimated structural parameters to
those of porphyrins (Table 3-4 and ref. 27, 79-84 and
references therein) suggests interesting generalizations.
The porphycene macrocycle is in general smaller, more rigid,
and less able to expand than the porphyrin ring. This is
reflected by the fairly high K values(inverse slope)
estimated for PPC vibrations compared to those of typical
high frequency porphyrin modes.27r‘“5t8c"81 Also, the core size
of the free base porphycene falls within the range of these
metal complexes, whereas the free base porphyrin has a
larger core size than the metalloporphyrin complexes
considered here. This relatively contracted core size for
IHPPC is the result of exceptionally strong internal
hydrogen bonds between the imine pyrrole nitrogen atoms and

the free base protons.-°~8

3.3.6) Iron Complexes

In Figure 3-8 the ClFemPPC spectrum is more complex
than that of (ImnzFemPPCCl, suggesting that the w e’m”
excited states of the ring (B2L1 and B3u in D2h symmetry)

interact with w-+ d”, d charge-transfer (CT) excited

X2
states (also BmIand Bag. These CT states are likely
produced from blu(porphycene) -> b3g,bzg( iron) transitions

analogous to theIam(porphyrin)-+ eg(iron) transitions of D4h

hemes~85fi6.As the latter do not occur in low-spin FeIn

Table 3-4

63

(A) Core Sizes of PPC and Analogous Porphyrin Complexes

 

dew?)
PPC ref. Porphyrin ref.
Ni 1.896 4 1.958 79, 82
Cu 1.94 i 0.01 5 2.000 79
H2 1.927 4 2.065 83

(B) Estimates of Structural Parameters for MPPC
and MOEP-38 Vibrational Frequencies

 

 

NiPPC vibration K A
(cm-1) (CW1, ii) (i)
1625 760 4.03
1579 490 5.12
1520 350 6.26

NiOEP vibration K A
(cm-1) (cm-l/ 3.) (2‘1)
1655 405 6.05
1602 236 8.74
1576 203 9.73

64

systems, their presence in the absorption spectrum5“68<mf
ClFemPPC is consistent with a spin state higher than S =
1/2. Analogous ferric porphyrin complexes exhibit spectral
trends qualitatively similar to those in Figure 3-8.

By analogy to ClFeIII porphyrins, a pure 8 = 5/2 high
spin state is expected for ClFemPPC; however, magnetic
susceptibility measurements indicated that substantial S =
3/2 intermediate spin character exists.87 The smaller core

size of porphycene presumably increases the splitting of the

dxadfl level from the other iron orbitals making the S = 5/2

state less accessible than in porphyrin complexes.

In iron porphyrins the vibrational frequencies of the
macrocycle are known to be sensitive to the ligation, spin,
and oxidation state of the Fe. This situation has been
extensively exploited by RR study of heme proteins.53 It is
of interest to determine if a similar situation exists for
iron porphycenes. The EPR spectrum of the six-coordinate
ferric complex confirms an S = 1/2 spin state for the iron
atom. The features of (ImnzFemPPCCl at 1298, 1513, 1567, and
1605 cm“1 in Figure 3-8 are clearly higher in frequency than
the analogous features in the ferric chloride complex. This
indicates that the conformation of the porphycene macrocycle
is more contracted in the.(IIm)2FeIII complex than in the ClFeIII
adduct.

Figure 3-8 shows that the metal center in porphycene is
reduced. The 13 cm"1 decrease from 1298 to 1285 ch-upon

reduction is larger than the 7-9 cm4-decrease exhibited by

65

the vibrations in the 1400-1700 cm“1 range. While the latter
reflect the macrocycle expansion attributable to the larger
ionic radius of FeII than Fem, the decrease in frequency of
the mode around 1300 cm”1 is much larger than expected by
comparison to the NiII and Cu11 complexes. Consequently, this
feature appears analogous to the v4CaN mode of hemes, the
"oxidation state marker". The porphycene mode is likely to
be a VCaN or vCa.N motion, and these data suggest overlap
between the porphycene b3g(1r*) and bzg(1r*) orbitals and the
iron b3g(dyz) and bzg(dxz) orbitals, analogous to the
situation in iron porphyrins.53'88 Reduction of the metal to

FeII most likely leads to population of the macrocycle 1r*

orbitals and weakening of the CaN and/or Ca.N bonds.

CHAPTER 4

EFFECTS O? LOWER SYMMETRY AND NORMAL COORDINATE
ANALYSES OP METALLOCHLORIN

4.1) Introduction

Hydroporphyrins serve as the prosthetic groups in a
variety of proteins that participate in electron transfer
and catalysis.1‘2’14'89‘98 The structural and electronic
properties of these macrocycles determine the detailed
aspects of their functional nature. The vibrational
characteristics of the ring are a particularly sensitive
indicator of these properties.99 Accordingly, many groups
have been using vibrational spectroscopy to probe
hydroporphyrins and related macrocycles.263640100‘123 At
present, a substantial body of vibrational data is available
for reduced pyrrole macrocycles. Most of these data have
been acquired for metallochlorins. These macrocycles are
structurally similar to metalloporphyrins except that one of
the four pyrrole rings is reduced (Figure 1-3).

The availability of a considerable amount of
vibrational data for metallochlorins makes these rings the
most amenable to vibrational analysis. The first

interpretations of the vibrational spectra of these systems

66

 

67

were made exclusively by analogy to those of
metalloporphyrins.”5'39!98'100‘105 This type of analysis presumes
that the vibrational eigenvectors of the two different
classes of macrocycles are similar. Because there is no a
priori basis for this assumption, normal coordinate analyses
are required to identify the exact forms of the normal modes
of the reduced pyrrole systems. In this regard, normal
coordinate calculations have been reported for a variety of
metallochlorins, including NiOECNLHW, CuaD
tetraphenylchlorinu“, and several NiflD pheophorbides.36
Bocian et al.36'108 have examined all of these species by
using semiempirical quantum force field and/or empirical
force field (GF matrix) methods. Several other groupsnmfll3
have also examined MOEC complexes by using empirical force
field methods. Bocian et al.'s calculations (both empirical
and semiempirical) predict that the forms of the normal
modes of metallochlorins are in general different from those
of porphyrins.”108 In contrast, empirical calculations on
NiOEC by Prendergast and Spiro109 predict that the
vibrational eigenvectors of metallochlorins are similar to
those of metalloporphyrins.

The origin of the disagreements between the empirical
and semiempirical normal coordinate calculations on NiOEC is
not clear, although Spiro and co-workers have suggested that
they are due to intrinsic inaccuracies in the semiempirical
approach.109 While it is true that semiempirical vibrational

analysis methods are less accurate than empirical

68

approaches, the development of an accurate empirical force
field requires data from a variety of isotopomers. However,
Spiro et a1. did not use the vibrational data from a series
of meso-deuterated NiOEC complexes (NiOEC, NiOEC-a,B-d2,
NiOEC-7,0-d2, NiOEC-d4) . They utilized only the vibrational
data from NiOEC-ny-d2 and the pyrrole-”N isotopomer.1°9 The
inclusion of additional meso-deuteration data (NiOEC-aJiwg,
NiOEC-d4) is essential for the development of an accurate
force field, because the deuteration in NiOiE:C--'y,<5-d2 samples
only one of the two symmetry inequivalent methine bridges in
the macrocycle (Figure 1-3).

Recently, Fonda et al.38 obtained detailed IR and
resonance Raman data for a series of meso-deuterated CuOEC
complexes (CuOEC, CuOEC-02,8-d2, CuOEC-7,0-d2’ and CuOEC-d4) .
The deuteration in these isotopomers systematically perturbs
each of the symmetry inequivalent methine bridges of the OEC
ring. Now it is possible to develop an empirical force field
for CuOEC based on the vibrational data from a series of
meso-deuterated CuOEC complexes. Furthermore, the detailed
IR and resonance Raman data were also obtained from the
various meso-deuterated ZnOEC complexes (ZnOEC, ZnOEC-aJi—
d2, ZnOEC-7,6-d2, ZnOEC-d4) . These data allow the examination
of the effects of the central metal ion on the forms of the
vibrational eigenvectors. Additionally, this new information
provides a basis for a comparison of the various force
fields for MOEC complexes and the factors that influence the

forms of the vibrational eigenvectors.

69

4.2) Theory of Normal Vibrationsn‘l'125

4.2.1) Normal Modes

Each normal coordinate is a linear combination of basis
coordinates. Prior to the development of the Wilson method,
the Cartesian coordinates were used as the basis set. In
contrast, the Wilson method uses internal coordinates as the
basis set. In fact, only five different types of internal

coordinates are needed to represent a given normal

 

coordinate completely.

For a general polyatomic molecule, there are 3N - 6
normal coordinates (3N - 5 for a linear molecule), where N
is the number of atoms in the polyatomic molecule. In a
diatomic molecule, the internal coordinate (bond stretch) is
the normal coordinate. The restoring force for the diatomic
molecule will then be:

F = -kq (eq. 4-1)
where q represents the internal coordinate.

The potential energy V will be given by:

2V = qu (eq. 4‘2)

The kinetic energy T will be given by:

2T = 62 (eq. 4-3)
where q = dq/dt. If the normal modes of a general polyatomic
molecule are to behave in the same manner as the normal mode
of the diatomic molecule, the expressions for the potential
and kinetic energies will not have cross terms. That means

the normal modes are defined to be mutually orthogonal,

70

hence the name "normal“ modes. In terms of the normal
coordinates, Q, the kinetic and potential energy expressions

for a general polyatomic molecule can be written as:

2T = E ('2: 2v = 8 x10: (eq. 4-4)

1

where 01 = in/dt, and the summations over i run from 1 to
3N - 6 (or 3N - 5 for linear molecules).

The kinetic and potential energies take a diagonal form
in their matrix representations. The expressions in eq. 4-4
can be rewritten as:

2T = Q'Q 2v = 90.9 (eq. 4-5)
where Q is the matrix (column vector) of the normal
coordinates, Q is its time derivative and A is the diagonal
matrix of the Avalues. Q' and Q' are the transpose matrices

of Q and Q, respectively.

4.2.2) Internal Coordinates

There are five kinds of internal coordinates that are
used to describe the motion of a molecule. Except for the

linear bend, all the internal coordinates' § vectors are

defined in terms of unit vectors, 6”,

directed along the
bonds. The expressions for the § vectors also incorporate

information on the bond lengths and valence angles. The an

are defined as:
eij = [(xj - xi)1 + (yj - yi)j + (zj - zi)k]/rij where

(xi,yi,zi) and (xj,yj,zj) are the Cartesian coordinates of

71

atoms i and j, respectively, and rij is the distance between

them. The five types of internal coordinates are:

(1) Bond Stretch : Arij (Figure 4-1)
(2) Valence Angle Bend : AaUk (Figure 4-2)
(3) Out-of-Plane Wag : A0 (Figure 4-3)
(4) Linear Valence Angle Bend : A¢ (Figure 4-4)

(5) Torsion : Ar (Figure 4-5)

Decius proved that this set of basis vectors is sufficient
to describe the normal coordinates of any molecule
completely.
4.2.3) The Transformation from Cartesian Coordinates to
Internal Coordinates
The expressions for the kinetic and potential energies
in the Cartesian coordinates can be written as:

X

2T = x'gg 2v = 2y: x (eq. 4-6)

where x is the matrix of the Cartesian coordinates. The
kinetic energy, when expressed in terms of Cartesian basis
vectors, takes the form of the diagonal matrix M, formed
from the masses of the constituent atoms. This formulation
parallels that of normal coordinates where the kinetic
energy also takes a diagonal form.

The case for the potential energy, however, will not be
as simple. The Ef'matrix of eq. 4-6 will be non-diagonal.

Except for the case of a linear molecule, the elements of

72

 

pm
II
I

 

Figure 4-1 Bond Stretch: Arij

   

0

ozin at 180
si = (cosozijk eji - ejk) /rij Sine:ijk
sk = (cosozijk ejk - eji) /rjk Sinaijk

sj= - (si + sk)

Figure 4—2 Valence Angle Bend: Aaijk

73

0.1:: 180°
0=¢90°

('39

 

 

 

* 1 ékxéjl
s.= - L7 . - tande”
I :8 cosfisinoi fl
_. 1 a. Xé. cos¢cos¢ - COS¢>
S = 1k. II I k. 2 I
k rjk Sind>i cos§s1n oi

 

i [é‘kXé‘l] [cosqsicombl - cosok]

cosdIsinqui

Figure 4-3 Out-Of-Plane Wag: Ad

 

 

A
U
4’
(9 J- <9

0 is a unit vector along j - A, perpendicular to i-j-k.

Figure 4-4 linear Valence Angle Bend: A¢

74

 

am: , 03k, ¢1so°

For each i-type atom:

g = 1 ei.><e.k

For each l-type atom:

n1 1 r r sinzau (eii J'k
n fijk W
1 cosaikl

+ —- (5 x 5.)
n1 ' 2. m R

J

8* = nl E r r sinza (elk x ekj)
1 cosm-k
-e -o
+ E -————JL-'(eu X e )

I'll ’ 2 ”k

ni = the number of i-type atoms
n1 = the number of l-type atoms

Figure 4-5 Torsion: Ar

 

75

the.§f matrix will be difficult to interpret physically. For
a non-linear molecule, combinations of atomic motions along
the x, y and 2 directions will in general be mixtures of
stretches and bends. Thus, interpretations of the physical
meaning of the elements of the.§f'matrix, in terms of
conventional force constants, will be very difficult. These
difficulties can be bypassed if internal coordinates are
chosen for the basis set rather than Cartesian displacement
vectors. Then, the kinetic and potential energies can be
expressed as:

2T = 8'8 3. 2v = 3'1“. .8. (eq. 4-7)
where g is the matrix (column vector) of the internal
coordinates and K and E are the kinetic and potential
energies matrices, respectively. Both the K and 5 matrices
will be non-diagonal. Constructing a non-diagonal kinetic
energy matrix is complicated matter. However, the
interpretation of the elements of the new E matrix in terms

of primary and interaction force constants is now relatively

easy.

4.2.4) The Kinetic Energy in Internal Coordinates

By definition, the momentum va conjugate with xi,

is given by:

6T .
PXi = 3x1 = mixi (eq. 4-8)

 

The kinetic energy will be given by

76

2T = g; ml 2,, (eq. 4-9)
The column vectorlgxcan be expressed as:

Eh = M 3 (eq. 4-10)
A set of conjugate momenta, P, can be defined with the

internal coordinates, q:

P = a? k = 1,2,°-°°-s (eq. 4-11)
qk aqk

 

The momentum, P2! in terms of Cartesian coordinates, can be

1

written in terms of the coordinates corresponding to the

translational and rotational motions of the molecule by q:

and their conjugate momenta by Pg;

 

p = 6?
fl 6xi
= Efi-i;-§3£-+ E,_g;_§g§
aqk 3X1 3 69; 3X1
= E: qu Bk,Xi (eq. 4-12)

The second term becomes zero since the momenta corresponding
to the translational and rotational motions are zero. In
matrix form

2.. = 9'2 (eq. 4-13)
From eq. 4-10 and eq. 4-13

M X = 3'2 (eq. 4-14)
The B is the transforming matrix defining internal
coordinates in terms of Cartesian coordinates:

3 = g 2; (eq. 4-15) 3 = B x (eq. 4-16)

From eq. 4-14, 2.; = m-lg'g. Inserting this into eq. 4-16

:13 = g M'lB'P (eq. 4-17)

 

77

So, the new expression for kinetic energy (eq. 4-9) will be
given by:
2T = B'B m-lg'g (eq. 4-18)

If a new matrix G is defined as

g = B M’lg' (eq. 4-20)

Then eq. 4-20 becomes

2T = g'g P (eq. 4-21)

Combining eq. 4-21 with eq. 4-17

B. = 9 2 or P = 913 (eq. 4-22)

The kinetic energy can then be rewritten again as:

”=292
= (91.8) wuss-18)
= étg-lé (eq. 4-23)
Comparing eq. 4-23 with eq. 4-7, it is clear that
K. = 9‘1

Actually, once the B matrix is known, the G matrix follows

automatically via the relationship given in eq. 4-20.

4.2.5) The Potential Energy in Internal Coordinates
The potential energy was already defined in eq. 4-7.

The elements of the F matrix will be defined as:

32V 32V
F. . = = . 4-24
1] [aqiaqj] e [aqjaqi] e (eq )

The subscript "e" indicates that the partial derivative is

 

 

to be evaluated at the equilibrium geometry. The

78

interpretation of the elements of the F matrix is now very

simple.

4.2.6) The Secular Equation
The matrix Q is the transformation matrix, which
relates the internal coordinates to the normal coordinates
as:
R = L 9 (eq. 4-25)
The potential energy can then be written as:
W = 3': 8. = (19) was»
= Q'L'E A 2 (eq. 4-26)
Comparison of eq. 4-26 with eq. 4-5 yields the important
identity:
9'2 1. = 1 (eq. 4-27)
The kinetic energy can now be written as:
2T = 89-18 = (Lé>'9-1<1§2>
= Q'L'Q‘léfz (eq. 4-28)

Comparison of eq. 4-28 with eq. 4-5 yields the second

identity:
1'94; = 1 (eq. 4-29)
where 1 is the identity or unit matrix. Therefore,
L'E L = .I. L
= 1'91; 1 (eq. 4-30)
Left-multiplying by (g')d,
E L = 911 1 (eq. 4-31)

Left-multiplying by Q I

9 E L. = 1 2». (eq. 4-32)

79

This is the Secular Equation to be solved for the
vibrational analyses. The vibrational Hamiltonian matrices
are the G and F matrices and the solutions of the Secular
Equation (X values) are related to the vibrational

frequencies.

4.2.7) The Potential Energy Distribution

Morino et al.”5jproposed that the potential energy

distribution of a normal coordinate, QN, be defined by:

1
V(QN) = 3 Q; g FijLiNLjN (eq. 4-33)

In general, the value of Fulmfl%n is large when i = j. Thus,

the Ftflti terms are the most important in determining the

distribution of the potential energy and the magnitudes of

the Fig11 terms provide a measure of the relative

contribution of each internal coordinate, qi,'to the normal

coordinate QN.‘Therefore the potential energy distribution

is defined by the following expression.

PEI)ij = (Fiiij)/(231F..ij) (eq. 4-34)

11

4.3) Normal Coordinate Calculations

Normal coordinate calculations for CuOEC were performed
by using the empirical GF matrix method of Wilson.124

The G matrix was constructed by using the
crystallographic data for FeOEC.17 This structure was also
used by Prendergast and Spiro109 in their normal coordinate

calculations on NiOEC. The crystallographically determined

80

structure of the FeOEC ring skeleton is non-planar and
deviates somewhat from Czsymmetry. In this calculation, the
geometry was adjusted to give a planar, C5 structure. The
ethyl groups were treated with the ethylene carbons and
hydrogens and the terminal methyl groups positioned above
and below the plane of the macrocycle such that C5 symmetry
was preserved. The bond lengths and bond angles of the ring
skeleton in this idealized structure are given in Table 4-1.
Figure 4-6 gives the pyrrole ring designations and atom
numbering scheme that were used in the calculations. Figures
4-7 and 4-8 display the internal coordinate numbering scheme
for CuOEC. Table 4-2 shows the C5 symmetry coordinates that
were used in the normal coordinate calculations.

The normal modes of CuOEC were calculated with an all-
valence force field by using the vibrational analysis
programs of Schachtschneider.127 The normal coordinate
calculations for CuOEC were initiated by constructing the F
matrix reported for NiOEC by Spiro et al.109 This force field
for NiOEC was developed via a selective refinement of the
valence force field previously developed for NiOEP.128 The
magnitude of the diagonal Ogglforce constants used by Spiro
et al.109 follows a pattern that corresponds to a long-short-
long-short alteration of the cgh1bond lengths (starting
from the reduced ring) rather than the actual short-long-
short-long pattern.17'129'130 The initial calculations on CuOEC
utilized the F matrix constructed for NiOEC with the CJ;1

force constants ordered consistent with the bond lengths. As

81
Table 4-1 Geometric Parameters for the CuOEC Calculation

Bond Lengths(A)

 

Inter. Coor. I 11:, Ni IIj , IVj III
CuN 1.979 1.973 1.973 1.956
Na 1.327 1.370 1.389 1.379
ab 1.502 1.437 1.446 1.433
am 1.363 1.401 1.363 1.401
bb 1.433 1.351 1.351 1.346
mH 1.100
b1 1.488 1.505 1.501 1.506
12 1.502 1.529 1.529 1.529
Anglesgdegree)
NCuN 90 90 90 9o
CuNa 131.2 127.6 127.4 127.6
Nab 119.5 110.6 110.6 110.6
aNa 100.5 105.0 105.0 104.7
Nam 122.9 124.4 124.4 124.4
abb 100.9 107.7 106.0 107.0
bam 117.6 125.0 125.0 125.0
ama 125.4 125.4 125.4 125.4
abl 112.9 124.6 125.1 124.8
bb1 130.2 127.6 128.8 128.2
b12 117.5 128.3 128.3 128.3

Labels: a,b and m are the Ca, Cb and Cm atoms; 1 and 2 refer
to the first and second C atoms of the ethyl groups.

82

 

 

32 24 25 33 53

52

44 6O 61 45

Figure 4-6 Structure and atom labeling scheme for
CuOEC

83

36 35 14‘ 147

 

 

150 148
149 122 121 46
Y 24 23 5
43 4
16 8 7 15
13 35 17 14 132
37 2 22 34 130
1 4 2
3 9 6 13
31 3 1 29
137 38 0 5 3312
26 4 21 127
136 38 18 13 12
44 19 1 12 20
I3 27 28 01
139141 144 142
Figure 4-7 Internal coordinates representation of

the in-plane metallochlorin bond
stretching deformations. The numbers on
the bonds refer to the numbers used in
the computer program of Schachtschneider.

1

84

197 90

98 20110 109 25
117 18.
12‘“8 123

194
.195

 

 

”Y 8
94 862674 73 81 93
176 95 88 52953 87 92
175 159 83 15.17
158 103 7 67 6478; 100 5 172
111 116108
60 55 48 47 58
12 56 4546 51 115
178 104 7663 6371 9910152
17.162 84 79 15 169
85 49 50 170
9689 61 57 62 86 91
77 69 70 78 90
(X
164 105 106 167
18 165 16.184
18
185
101 = 8-20-28 102 8 9-21-29 125 =38-20-21
126 =39-21-20 151 =40-26-18 154 =41-27-19
157 =42-30-22 160 843-26-18 163 =44-32-24
166 =45-32-24 171 =56-26-48 174 =57-27-49
177 =50-30-58 180 =59-31-51 183 =60-32-52
186 =61-33-53 189 =20-38-62 192 =21-39-63
193 =46-38-54 196 347-39-55 199 =26-18-6-19

200827-19-7-18
203832-24-12-25

201:30-22-10-23
204=33-23-13-24

202=31-23-11-22

Figure 4-8 Internal coordinates representation of
the metallochlorin angle bending

deformations.The numbers on the bonds
refer to the numbers used in the
computer program of Schachtschneider.
Some of internal coordinates could not
represent in plane. In number series,
left hand side numbers could be
represented internal coordinates and
right hand side number series represent
atom labeling in Figure 4-6.

85

Table 4-2 In-Plane C5 Symmetry Coordinates for CuOEC

A
: 1 + 3 2: 2
3: 4 4: 5 + 10

: 6 + 9 : 7 + 8
7. 11 + 12 8: 13 + 18

: 14 + 17 10: 15 + 16
11: 19 + 20 12: 21 + 26
13: 22 + 25 14: 23 + 24
15: 27 + 28 16: 29 + 31
17: 30 18: 32
19: 33 + 38 20: 34 + 37
21: 39 + 40 22: 41 + 44
23: 42 + 43 24: 45 + 46
25: 47 + 48 26: 49 + 50 - 2(57)
27: 51 + 52 - 2(58) + 55 + 56 - 2(60)
28: 53 + 54 - 2(59)
29: 61 + 69 - 2(77) + 62 + 70 — 2(78)
30: 63 + 71 - 2(79) + 68 + 76 - 2(84)
31: 64 + 72 - 2(80) + 67 + 75 - 2(83)
32: 65 + 73 - 2(81) + 66 + 74 - 2(82)
33: 90 + 91 - 2(86) + 89 + 96 - 2(85)
34: 92 + 93 - 2(87) + 94 + 95 - 2(88)
35: 97 + 105 - 2(113) + 98 + 106 - 2(114)
36: 99 + 107 - 2(115) + 104 + 112 - 2(120)

37: 100 + 108 - 2(116) + 103 + 111 - 2(119)
38: 61 - 69 + 62 - 70

39: 63 - 71 + 68 - 76

40: 64 - 72 + 67 - 75

41: 65 - 73 + 66 - 74

42: 97 -105 + 98 - 106

43: 99 - 107 + 104 - 112

44: 100 -108 + 103 - 111

45: 89 - 96 + 90 - 91

46: 92 - 93 - 94 + 95

86

Table 4-2 (cont.) In-Plane C5 Symmetry Coordinates for
CuOEC

A (cont.)
47: 35 + 36

48: 121 + 122
49: 101 - 109 + 102 - 110

50: 124 - 126 - 125 + 123
51: 117 + 118 52: 127 + 136
53: 128 + 137 54: 129 + 138
55: 130 + 133 56: 131 + 134
57: 132 + 135 58: 139 + 142
59: 140 + 143 60: 141 + 144
61: 145 + 148 62: 146 + 149
63: 147 + 150 64: 151 + 160
65: 152 + 161 66: 153 + 162
67: 154 + 157 68: 155 + 158
69: 156 + 159 70: 163 + 166
71: 164 + 167 72: 165 + 168
73: 169 + 178 74: 170 + 179
75: 171 + 180 76: 172 + 175
77: 173 + 176 78: 174 + 177
79: 181 + 184 80: 182 + 185
81: 183 + 186 82: 187 + 190
83: 188 + 191 84: 189 + 192
85: 193 + 196 86: 194 + 197
87: 195 + 198 88: 199 + 202
89: 200 + 201 90: 203 + 204
91: 51 - 52 + 56 - 55 92: 205 + 206
B
93: 5 - 10 94: 6 - 9
95: 7 - 8 96: 11 - 12

97: 13 - 18 98: 14 - 17

87

Table 4-2 (cont.) In-Plane C5 Symmetry Coordinates for

CuOEC
B (cont.)
99: 15 - 16 100: 19 - 20
101: 21 - 26 102: 22 - 25
103: 23 - 24 104: 27 - 28
105: 33 - 38 106: 34 - 37
107: 39 - 40 108: 49 - 50
109: 51 - 52 + 55 - 56 110: 54 - 53
111: 89 - 96 + 91 - 90 112: 92 - 93 + 94 - 95

113: 97 + 105 — 2(113) - 98 - 106 + 2(114)
114: 99 + 107 - 2(115) - 104 -112 + 2(120)
115: 100 + 108 - 2(116) - 103 -111 + 2(119)
116: 97 - 105 + 106 - 98

117: 99 - 107 + 112 - 104

118: 100 - 108 + 111 - 103

119: 41 - 44 120: 42 - 43
121: 45 - 46 122: 48 - 47
123: 61 + 69 - 2(77) - 62 - 70 + 2(78)
124: 63 + 71 - 2(79) - 68 - 76 + 2(84)
125: 65 + 72 - 2(80) - 67 - 75 + 2(83)
126: 65 + 73 - 2(81) - 66 — 74 + 2(82)

127: 61 - 69 + 70 - 62

128: 63 - 71 + 76 - 68

129: 64 - 72 + 75 - 67

130: 65 - 73 - 66 + 74

131: 1 - 3 132: 29 - 31
133: 51 + 52 - 2(58) - 55 - 56 + 2(60)
134: 35 - 36

135: 121 - 122

136: 101 — 109 - 102 + 110

137: 124 - 126 + 125 - 123

138: 117 - 118

139: 127 - 136 140: 128 - 137

88

Table 4-2 (cont.) In-Plane C5 Symmetry Coordinates for

CuOEC
B (cont.)
141: 129 - 138 142: 130 - 133
143: 131 - 134 144: 132 - 135
145: 139 - 142 146: 140 - 143
147: 141 - 144 148: 145 - 148
149: 146 - 149 150: 147 - 150
151: 151 - 160 152: 152 - 161
153: 153 - 162 154: 154 - 157
155: 155 - 158 156: 156 - 159
157: 163 - 166 158: 164 - 167
159: 165 - 168 160: 169 - 178
161: 170 - 179 162: 171 - 180
163: 172 - 175 164: 173 - 176
165: 174 - 177 166: 181 - 184
167: 182 - 185 168: 183 - 186
169: 187 - 190 170: 188 - 191
171: 189 - 192 172: 194 - 197
173: 193 - 196 174: 195 - 198
175: 199 - 202 176: 200 - 201

177: 203 - 204
178: 90 + 91 - 2(86) - 89 - 96 + 2(85)
179: 92 + 93 - 2(87) - 94 - 95 + 2(88)
180: 205 - 206

89

expected, this force field overestimates the vibrational
frequencies for the skeletal modes of CuOEC (the skeletal
mode frequencies of NiOEC are in general higher than those
of CuOEC.“i"%“”). Consequently, the magnitude of all the
diagonal stretching force constants were systematically
reduced before attempting any perturbative refinement of the
force field. This procedure brought the calculated skeletal
frequencies of CuOEC into better accord with those observed
experimentally. This force field is designated FT‘I in Table
4-3. Only those force constants that are different from
those reported by Spiro et al.109 are listed in the Table 4-
3. Tables 4-4 and 4-5 show force constants that are the same
as those reported by Spiro et al.109 For reference, the force
field obtained by reversing the C411stretching force
constants of FT'I is also listed in the Table 4-3 and is
designated FF H. For this force field and all others
described below, only those force constants that are
different from those of FFI are included in Table 4-3.
During the course of the vibrational analyses, many
strategies were employed in attempts to improve the fit
between the observed and calculated vibrational frequencies
and isotope shifts. These approaches were based on
systematically varying selected diagonal and off-diagonal
force constants of the ring skeleton. In all cases the force
constants for the reduced ring, the ethyl groups and the
meso-hydrogens were held fixed. Two additional force fields

that give reasonable fits to the data are also given in

90

 

Table 4-3 Refined Diagonal Force Constantsa for CuOEC
internal
coord . I III, IV1 112, IV2 III
FFIb K(aN’)l 5.459 5.411 5.127 5.325
K(ab)2 4.018 5.335 5.431 5.527
x(bb)3 3.752 6.703 6.703 6.703
x(am)4 6.819 6.390 6.819 6.390
x(CuN)S 1.443 1.571 1.571 1.499
FFIIc K(am) 6.390 6.819 6.390 6.819
FFHP3 K(aN) 5.743 7.168 5.888 6.151
x(ab) 4.018 5.555 5.368 5.373
K(bb) 3.752 6.660 6.660 6.961
x(am) 5.138 6.964 6.776 6.470
K(CuN) 1.503 1.636 1.636 1.561
1,2 x-x(am-am) 0.2 0.4
FFFW=
1,2 x-x(am-am) 0.435 0.412
1,4 x-x(am-am) -0.362 -0.749 0.129

 

aIn mdyn/A for stretches (x) and stretch-stretch
interaction (x-x).

bForce field I is numbered for using in Table 4-5.

CAll other force constants are the same as in force field I

91

 

 

Table 4-4a Fixed Diagonal Force Constantsb for CuOEC

internal coord. II 11th1 112,1V2 IH
x(mI-I)6 4.560 4.560 4.560 4.560
K(b1)7 3.688 4.064 4.064 4.064
K(12)8 4.600 4.600 4.600 4.600
x(lH)9 4.560 4.560 4.560 4.560
K(bH)10 4.560
H(aNa)11 1.620 1.620 1.620 1.620
H(Nab)12 1.370 1.370 1.370 1.370
H(abb)13 1.170 1.370 1.370 1.370
woman,1 0.300 0.300 0.300 0.300
H(Nam)15 0.830 0.830 0.830 0.830
11(1na16)16 0.730 0.830 0.830 0.830
H(ama)n 1.100 1.100 1.100 1.100
H(ab1)18 0.900 1.200 1.200 1.200
H(bb1)19 0.900 1.200 1.200 1.200
H(b12)20 1.200 1.200 1.200 1.200
chuN)21 0.250 0.250 0.250 0.250
H(amH)2;Z 0.500 0.500 0.500 0.500
H(blH)23 0.625 0.625 0.625 0.625
H(21H)2,4 0.625 0.625 0.625 0.625
H(H1HI)25 0.520 0.520 0.520 0.520
11(a1611)26 0.400
H(be)2.7 0.400
H(1bHZ)28 0.400
7(b1)29 0.400 0.400 0.400

aref. 109

bIn mdyn/A for stretches (k), mdyn A/rad? for in-plane
bends (H) and mdyn A/rad? for out-of-plane bends (7).

 

92

Table 4-5

aref. 109

bIn mdyn/A for (x-x), (x-H) and mdyn A /rad2 for (x-H).

ci-j, interaction between internal coordinates i and j,

labeled according to the numbering scheme in Tables
4-3 and 4-4.

(K-K)L2 : stretch-stretch interaction between two bonds
sharing one atom

(Io-101,3 : stretch-stretch interaction between two bonds
separated by one atom.

(x-H)2 : stretch-bend interaction between bond and angle
sharing two common atoms.
(x-H)1 : stretch-bend interaction between bond and angle

sharing one common atom.

(H-H)2 : bend-bend interaction between angles sharing
two common atoms

93

Table 4-5a Fixed Off Diagonal Force Constantsb for CuOEC

 

internal coord.C I H,HLIV
1-1, 1-4, 1-5, 2-3, 0.430 0.430
2-7, 4-4, 4-7, 7-8
2-4 0.040 0.430
5-5 0.150 0.150
1-2, -2 0.250 0.250
1-4, -5, 2-4, 2-7 -0.250 -0.250
1-7

(K'H)2
4-17, 1-11, 2-12 0.245 0.245
2-16, 4-16, 2-18
2-13 0.160 0.160
3-13 0.020 0.320
3-19, 7-18, 6-19 0.320 0.320
1-12. 1-15, 1-14 0.100 0.100
4-22 0.080 0.150
5-23, 8-24 0.110 0.110

(K‘HI1
1-15, 1-12, 2-11, 7-12 -0.120 -0.120
1-12, 2-15, 4-12, 2-19, -0.245 -0.245
1-22

(H'H)2
15-17 0.060 0.060
23-24 -0.070 -0.070

 

94

Table 4-3. One of the force fields was obtained by first
refining all of the diagonal skeletal constants. This

refinement was followed by a second iteration in which only

the diagonal force constants of the CaCm and Cbe bonds were
allowed to vary. Finally, the 1,2 CaCm-Cacm interaction
constants for the a,B- and 7,6smethine bridges were given
different values because this improved the fit in the
observed versus calculated meso-deuteration shifts. This
force field is designated.FT‘IH in Table 4-3. The force
field designated FF IV was obtained by adding 1,4 CaCm-Cacm

interaction constants to FFI and then refining both these

and 1,2 Cacm'CaCm interaction constants.

4.4) Results

4.4.1) Vibrational Spectra of CuOEC and ZnOEC

Figure 4-9 shows resonance Raman spectra obtained with
Soret excitation for ZnOEC, ZnOEC-a,B-d2, ZnOEC-'y,<5-d2 and
ZnOEC-d4 in CHZCl2 solutions. The IR spectra for this set of
ZnOEC derivatives in CHC’l3 solution are shown in Figure 4-
10. The total number of bands observed by IR spectroscopy in
these measurements is greater than the number detected by
resonance Raman spectroscopy owing to the IR activity of the
internal vibrations of the ethyl groups. Soret excitation
Raman spectra of CuOEC, CuOEC-afi-dZ, CuOEC-7,6-d2, and
CuOEC-d5, which complement the visible excitation Raman
spectra (Figure 4-11) in earlier workfihfll, are shown in

Figure 4-12. Comparison of the vibrational data for ZnOEC

 

Figure 4-9

95

Resonance Raman spectra of ZnOEC, ZnOEC-afl-dz,
ZnOEC-vxS-d2 and ZnOEC-d4 in CH2C12 solution
obtained with Soret excitation at 406.7 nm.
Laser power: 7 mW, concentrations: ~40 pM.

Asterisks(*) in Figures 4-10, 11, 12, 13 and 14
represent solvent vibrations.

96

ppm—1

chap!

ZnOEC

awn—1

—Q-_I

nan—I
Gav—l
vow-1
.
Nun—I
own—1
now—1
can—1
.NQ—I

  

ZnOEC-1113412

 

 

 

ZnOEC-78412

   

ZnOEC-d4

 

 

 

 

97

Figure 4-10 IR spectra of saturated CI-ICl3 solution of
ZnOEC, ZnOEC-a,B-d2, ZnOEC-7,0-d2 and ZnOEC-d4

98

 

1.09. 1.00.
IIhCONOO I baa
I .3 In: I :o
11".
1 on. use 13.
I no. I no. I o:
I 22 I :2 1 «S.
.«a—
n 3.: u 82 "3.2
Coop v8. .2:
I app Igpp Inc:
....m:22 I a: in"?
I hva luv—p IND-"Op”.
I3: I «a: Ina—p
I heap I «nap 169N—
I Crap Ih—np Impa—
.1009-
I Oanhn— I .50- IMOhnh-np
I non. Iconp
I p09— a:
I 3:, I Cov— fl On.-
I 15:32 I on!
I pavp
I hum—
.uanm-
I.aomp
I 08.

I ONO—

 

ZnOEC-d4

I 3.
I we.
I pNO

apa—

InONp

luau
I0.»-

13-

1&3—

«pup

' I

U‘TTY—V'r'ri

V Y 7 ‘ I I V

V

I

. 71 2+ . .
1300 1200

‘ T I

1 400

.T.fi.
1500

V T I V V

Y

 

 

1 000 900 800

1100

1 600

1700

cm

 

99

Figure 4-11 Resonance Raman spectra of CuOEC,
CuOEC-afi-dz, CuOEC-7,6-d2 and CuOEC-d4
in CH2C12 solution obtained with visible
excitation at 615.0 nm. Laser power: 40 mW,
concentrations: ~70 pH (from ref.38).

100

CuOEC-a. B-d;

I. ..= 615.0061

nun— I

 

.s—pI

 

 

69:1

 

 

 

 

.2: I

 

.3. I III .11

 

CuOECm 8-6:

 

CUOEC'CA

1700

   

1300
~I
CM

 

 

 

 

 

1500

 

 

101

Figure 4-12 Resonance Raman spectra of CuOEC,
CuOEC-a,B-d2, CuOEC-7,0-d2 and CuOEC-d4
in CH2C12 solution obtained with Soret
excitation at 406.7 nm. Laser power: 7 mW,
concentrations: ~40 uM.

102

‘4... = 406.7nm

 

 

 

 

CuOEC-cB-dz

 

 

CuOEC-yG-dz

   

«on. -

 

1100

 

 

900

103

(Figures 4-9 and 4-10) with those for CuOEC in Figure 4-11
and 4-12 shows a close correspondence in the modes observed
and in the pattern of vibrational frequency shifts upon
selective deuteration at the methine bridge positions (Table
4-6). Thus, a detailed discussion of the mode assignments is
deferred to the normal coordinate analysis section below.
Here, only some general observations and comments on the
behavior of several of the modes in the high frequency
region will be made.

The metal-dependent frequency differences between the
correlated modes in Table 4-6 arise from the larger core
size, and correspondingly lower macrocycle mode frequencies,
for ZnOEC relative to CuOEC. These metal dependencies have
been observed for metallochlorins by several
authorsZ‘SI-mvlool101 and correspond to the well-known core size
dependence of metalloporphyrin macrocycle modes.’79 In the
original work with CuOEC, our visible excitation Raman and
IR data showed two prominent modes in the 1580 - 1605 cm4-
region. The ZnOEC data presented here and the normal
coordinate calculations discussed below indicate that a
third band should be apparent in this frequency region (for
ZnOEC the corresponding region is 1565 - 1590 cm“1 and these
three modes occur at 1569, 1571 and 1586 cm'1 in the h4
species (Table 4-6)). Accordingly, higher resolution IR data
for CuOEC and its specifically deuterated derivatives were
recorded and the occurrence of a third mode in this region

at 1583 cmd-(Table 4-6) was revealed. Inspection of

104

 

 

 

 

cowuoufioxm manwmw> ha umocsssm " O

xuoa 4H4» " a

an .umu H m

4 6H1 s: «I oomvH

4 an s: 61 om4H HH1 61 on 664H

m cm: om: HI smmH 6H1 6H1 6 nva

4 m1 «I HI 4an HI 6 HI sva

4 o o o HsmH HI HI HI mmmH

m mHI HI HHI mmmH HHI 41 6H- 44mH

4 5H1 HI 4H1 ommH nHI NI ~HI momH

4 «HI 61 41 nmoH 6H1 s: n: 446H

a>m 464 484 484 4: 464. 484 484 4:
mammal quaqul

.nomocu 4:6 naomoao 46 wave: socosaoumuonz on» you

mmHHuoaaam can .6Taov muuflzm onouomH .6Taov mmwocmsvmum Hocoauaunw> .oI4 manna

105

polarized resonance Raman (Figures 4-13 and 4-14) and IR
spectra (Figure 4-10) of ZnOEC also revealed that the 1535
cm"1 band in the Soret excitation Raman spectrum (Figure 4-
9) consists of a 1538 cmfl-polarized mode and a 1527 cm“-
anomalously polarized mode. The properties and frequency
shift patterns of these modes correspond closely to those of
the 1547 cm4-and 1543 cm4-modes in CuOEC.

The compilation of mode frequencies and deuteration

shifts for CuOEC and ZnOEC in Table 4-6 shows that the

 

frequency shift pattern observed for CuOEC upon selective
methine deuteration is paralleled by that observed for the
ZnOEC derivatives. The shift magnitudes for the Zn species

are greater than those of the Cu complexes, however, owing

to the lower CHAImode frequencies and hence better mixing
with.C%H deformations that occurs in ZnOEC. For several
correlated modes in the two metallochlorins, the frequency
shift patterns upon a,B-deuteration are distinct from those
upon'yfi-deuteration, which indicates that mode localization
phenomena in the metallochlorins do not depend on the
central metal but rather on the macrocycle symmetry, as

suggested earlier.”38

4.4.2) Vibrational Assignments for CuOEC

The vibrational assignments for the high-frequency
(above 1000 cm'l) skeletal modes of CuOEC are given in Table
4-7. These assignments were made on the basis of the

frequencies, isotope shifts, and resonance Raman band

 

 

106

Figure 4-13 Polarized resonance Raman spectra of ZnOEC
and ZnOEC---cr,B-d2 in CI-12Cl2 solution
obtained with Soret excitation at 406.7 nm.
Laser power: 7 mW, concentrations: ~40 uM.

1107

can—

kex = 406.7nm

 

 

 

 

 

 

 

 

 

108

Figure 4-14 Polarized resonance Raman spectra of
ZnOEC-7,6-d2 and ZnOEC-d4 in CHZCl2 solution
obtained with Soret excitation at 406.7 nm.
Laser power: 7 mW, concentrations: ~40 FM.

109

2.6..

406.7nm

A” =

 

 

 

 

ZnOEC '0 4

 

 

 

900

0
1
1

 

£560.98a 43828524989 6 6 m N N 4 $3 30 N64H

66988 .E.E._0.986N H 6 N N H N H m 4 663 65H 3. 26H

_0..986H .9H.E..0.986N H 6 4 o 6 m N m 66.3 68H mmmH

6.982 5560.988 H H m 4. o 4 4 o m NmmH E60 83

952.982 468080.988 N 4. H 4 m H64H 36 c 664H

9452.989 .8.6v80.986H 6.380.983 N 6H 6H 8 4. m 4 4 23 464H 3 . 664H

945606988 6.380.983 466080.983 H H 6 NH 4 6 m 6 4 $3 86H EC 83

Her—552.986 6.580.983 m 8 8H 6 8H N o m 34H N46H Em 6 N43

386.380.986 4:05.988 .E.E606984N H o N m H o o H 4 63 4.43 30 4.43
6.580.984 448080.986 9506069864 H 6 H 6 m 36H

8.880.983 .E.E._0698HH 55606988 o N H o H m N 4 HSH EC 83

3.380.988 6.580498% 6 0 6H H H 4H 4 6 0H 6 $3 $3 36 6 463

8.658088 466580.988 493804988 6 H 4H NH 6 N HH NH 4 6H6H N66H EC 33
8.380.988 6580.988 0 6H 8 m m 6H6H

88580683 6580.988 6 6 4H 3 HH 8 H m 4 646H 446 H 6 E 6 446H

vucmacowmma 0460 0630 came Umno 0460 page name Umno Shm 0460 mH mm
zmH< 404.435 NUI©>< NUInc< Umno

64680 66 66662 Hmummem mocmzwmumuanm
020 now $53 muufinm wmouowH 0cm mmflocmsvmhm nvmpwazoamo can mom>ummno 434 0.369

111

 

88580683 494452.988 N N8H+ o 68H+ 4 NOHH
854480688 $2.989 4560.983 6 NH NN 6 N N o H 6 m 84HH NNHH ES NNHH
408982 63.524988 4 NH H N+ NN+ 4N+ NN+ o o 4 NHHH 45 H4HH
~0..98NN .AEVZ.98§ 6 3 6+ 4+ 0 o m oNHH 33 66HH
A_>4.H=.4060.98NH 44804480488 N 0 N6 N NNN N+ o 4 0H NH 8NHH 43 NNHH
. £560.98: 4680:8888 H 66N o 0 N6 N m 6H NH HH NH 36 0 6HNH
94828869988 .8.3=80%8ON 6+ N6N NNN H 4H NoN NNN 4 66 NH 4&6NNH
E45 Nm08N6 468580488 0 NNH 84 o N 8NH mm 4 66NH 43 68NH
842858089840 .82898NH N 6H+ 8N+ 8H+ 6H+ o H H+ 6 mm NH 33 88NH

E43 NE088 .6249 84H N 8 8 o 4 6 N NH

6843 NE0848 4685804846 H 8 NH 6 6 NH 4 NoNH
8.538048% .824982 4 6 4mN 4N4 6+ 8 + m HmNH oHNH

94.524988 448608984 444628988 4 N N H 4 NHNH
2.580.988 .ccdvza986N N H 6 N N H N H m 6H NH 6HNH EENHNH
9580.986 45828528982 N 4 6 N N 4 N4NH 43 N6NH

vucmecmflmmfi Uamo vmno vamo Umno name nmno camo Umno aha vao mH mm
26 H4 4012.35 Nvlmb< Nunnad. vwno

.0m050 mo mmvoz Hmumamxm mocmzvamI508m may you
$-53 muufiam mmouomH new mmflocmsvmum nvmuwasoamo can 46045093 Tyson; 8:4 manna

112

.mua musmwh ca czonm mcoHuHmom owaoaoouome map on ummmu mmmmnucoumm ca
mumuomumso ms» 086 E0 460 460 .coaumauowmv manamlcw u a van soumuum

a «m3OHHOM mm mum GOHHQHMOflGU QUOZU

.Umnaudaomwv u mu «vmuwumHom hamsoaosocm H mm «vmuflumaom u m ”mcoau68>0uan¢o

.OOH .mwu Eoum mum mundnm 26H msa .mm .mmH

Huh SHHB UmHMHSOHMUQ

EOHM Gmxmu ¢H03 mumeomouomfl UmUMHHmufimvlomme 05H UGM UWOSU HON mwflocmfluwhw NH UGM CMEMM 00GMCONOH 039m

 

 

8608983 .846080688 N o o o 4 68H
N0498HH 4840604983 8 6+ 8 o 4 666H
954628988 46.8.6.3:80688 6 HH+ 6H+ 6+ 6 663
86.34480686 46804480686 480608988 4 4H+ 6H+ 6+ 4 N66H
406984N 6 N6H+ N+ N6H+ 6 3:
40698NH AN>48524988H N 68+ o 66+ 6 66HH
_0._98NH .A_>4._:Vza98NH 6 4H H+ 4HH+ N 6 8HHH
uucwecmwmmd namo Umno name Umno namo vmno vamo vmno EMm 0Hmo mH
26H< 4UI©>QG< anm>< Ncunafl Umno
.0m090 mo mmvoz Hmumawxm mocwzvmumlnvwm map How
$-83 muuflnm mmouomH 656 mmflocmfivmnm nvmumHsoamu 0cm mvm>nmmno 70:03 4.14 manna

113

polarizations of the various modes. The calculated
frequencies and potential energy distributions listed in the
table were obtained with.FT1. Table 4-8 compares the
frequencies and meso-deuteration shifts for eight high-
frequency skeletal modes calculated by using four different
force fields (FF I-FT‘FV). The average errors between the
observed and calculated frequencies and meso-deuteration
shifts are compared for all four force fields in Table 4-9.

These errors were determined on the basis of the 16 highest

 

frequency observed resonance Raman bands and 37 meso-

deuteration shifts. These bands correspond to CaCm, Cbcb, CaN

and CaCb stretching motions.

4.5) Discussion

4.5.1) Vibrational Assignments

The molecular symmetry of metallochlorin is lowered to
<% from Dfllfor a planar metalloporphyrin macrocycle. The
resonance Raman active and IR active vibrational modes
belong to the A and B symmetry species of’Cb molecular
symmetry. Table 4-10 represents the correlation of Duland C2
in-plane modes. The selective meso-deuteration data, in

conjuction with the normal coordinate calculations, have

 

allowed us to refine the previous interpretation of the
vibrational spectra of CuOEC.“Nm These refinements are
readily extended to the spectra of other MOEC complexes.

Most of the modes above 1000 ch-can now be assigned with a

114

Table 4-8 Comparison of the Frequencies and Deuteration
Shifts (cmdj for the Various Force Fields

 

Acxfi-d2 A'w‘S-d2 Aaz,B,'y,¢‘5-d4

obfii 1644 3 7 10
FFI 1645 1 11 14
FFII 16M) 2 8 14
FFIH ZHM6 3 7 13
FFIV' 1642 1 12 15
obsd. 1602 12 2 13
FFI 1616 11 5 14
FF II 1617 10 8 15
FF III 1613 8 6 12
FF IV 1604 6 12 15
obsd. 1584 10 4 11
FFI 1582 8 14 18
FF II 1578 15 6 18
FF III 1590 12 4 14
FF IV 1581 13 7 18
obsd. 1583 2 1 1
FFI 1571 3 O 3
FF II 1576 3 1 3
FF III 1590 3 0 3
FF IV 1572 2 0 2

115

Table 4-8 (cont.) Comparison of the Frequencies and
Deuteration Shifts (cmdd for the Various
Force Fields

 

Aozfi-d2 Amé-d2 Aa,B,'y,5-d4
obsd. 1547 l 0 3
FF I 1561 0 1 2
FF H 1561 0 2 2
FF III 1556 1 2 4
FF IV 1552 0 1 1
obsd. 1543 0 l7 17
FF I 1453 2 6 7
FF II 1450 0 24 25
FF ID 1474 4 18 21
FF IV 1466 2 6 10
obsd. 1507 6 6 13
FF I 1491 3 4 9
FF II 1492 1 6 8
FF III 1511 3 6 12
FF IV 1494 6 4 11
obsd. 1486 4 7 10
FF I 1470 5 7 10
FF II 1469 7 5 10
FF III 1486 7 11 16
FF IV 1474 3 7 10

 

 

116

Table 4-9 Average Errors Between the Observed and Calculated

Frequencies and Deuteration Shifts for the Various
Force Fieldsa

FF Acm‘1 AAd (cm’l) description

 

I 13.5 (8.40)b 2.89 (2.47) force field for NiOEC with sealed
diagonal force constants

II 14.2 (9.00) 2.43 (2.20) same as I with C‘fiCm force constant
reversed (see ref. 109)

III 11.6 (7.73) 2.48 (2.43) skeletal diagonal force constant refined,
1,2 C.Cm'C.Cm interaction constant adjusted

IV 12.1 (7.86) 2.69 (2.35) same as I but only 1,2 CaCm-C‘Cm and
1,4 CsCm-Cacm interaction constant adjusted

 

aValues tabulated from the 16 observed bands and 37 observed
deuteration shifts in the 1650 - 1300 cm‘1 region.

bValues in parentheses do not include the observed 1543 cm4-
band and its corresponding deuteration shifts.

117

Table 4-10 The correlation of D4h and C2 In-plane Modes

 

D4h C:2

Algm) -’ MP)

Blg(dp) ‘-' Mp)

A2g(ap) --’ B(ap. dp)
Bzg(dp) --’ B(dp)

BLAIR) -’ Mp) + B(dp)

 

p = polarized

ap
dP

anomalously polarized

depolarized

118

reasonable degree of confidence. Many of the vibrational
assignments reported in Table 4-7 are consistent with
previous qualitative interpretations of the spectra of
CuOEC. However, the new information reported herein
necessitates the reevaluation of certain assignments. The
specific assignments for selected high-frequency skeletal

modes are discussed in more detail below.

4 . 5. 1. 1) O‘Cn Vibrations

There are eight cgg1vibrations expected in the high-
frequency region (4 A and 4 B). Table 4-7 shows that seven
of the eight modes are observed in the vibrational spectra
(1644, 1602, 1584, 1543, 1507, 1486, 1466 cmrl). The
corresponding resonance Raman and IR bands of ZnOEC are
tabulated in Table 4-6 and ref. 38 (The 1486 cm"1 band of
CuOEC is only enhanced by visible excitation; for ZnOEC
visible excitation measurements are not feasible owing to
the huge fluorescence from the lowest excited singlet
state.).

The only missing vibrational band is the highest
frequency B-symmetry mode that is calculated at 1618 cm4u
The skeletal mode frequencies of MOEC are observed to be
metal dependent, and hence core-size dependent. The
empirical expressionm’that relates vibrational frequency
and core-size for a metalloporphyrin is:

V = K ( A - d )

119

where v is the vibrational frequency in cm'l, K(cm‘1/A) and
A(A) are parameters characteristic of the macrocycle, and d
is the center to nitrogen distance, or core size, of the
metalloporphyrin in X. The physically useful parameter is
the K value, which is proportional to the amount of CJ%1
stretching character in the vibrational mode. Fonda etral.38
reported the core-size correlation parameters, K and A, for
the high frequency modes of the Ni, Cu, and Zn complexes of
OEC and OEP, respectively. Theyuassumed the same core-size
for porphyrin and chlorin complexes, namely Ni(1.958 K)”,
Cu(2.000 &)132 and Zn(2.047 it) .133 In the study of NiOEC by
Bocian et al.'36 the corresponding band for the mode
calculated at 1618 cm4-in CuOEC (the highest B symmetry
mode) was observed near 1644 cm”; this vibration could be
the analog of the missing cghlmode of CuOEC. Because the
core-size correlation parameters (K and A) are positive, the
corresponding resonance Raman band of CuOEC compared to
NiOEC would be expected at lower frequency. These
expectations are supported by observed and calculated
frequencies.

In a previous resonance Raman study of CuOEC, Ozaki et
a1.100 assigned the 1602 cm"1 band to a totally symmetry cga,
stretching motion. However, the meso-deuteration data
indicate that this band is associated with a CaCm mode
localized on the a,B-methine bridges (ref.38 and this work).

The careful investigation of IR and resonance Raman

spectra shows that the composite 1583 cm"1 band in CuOEC is

120

composed of 1584 and 1583 car1 bands. The measurement of the
polarization ratio reveals that the 1584 cm“1 band is the
depolarized B symmetry mode and the 1583 cm"1 band is the
polarized A symmetry mode. Furthermore, the 1584 cm“1 band
shows strong isotope dependence upon deuteration of the
methine chlorins. 0n the other hand, the frequency shifts of
the 1583 cm'1 band for the methine deuteriated chlorins is
not great. The 1584 cm"1 band reported by Fonda et al.38
showed isotope shifts for meso-deuteration and frequency
shifts on change of peripheral substituents from CuOEC to
copper etiochlorin I (CuECI). Although these researchers
were not aware that this spectral feature is actually

composed of two modes, it is now clear that the 1584 cm'3L

band is a mixture of two modes (CaCm and Cbcb) . Therefore,
the 1584 and 1583 cm‘1 modes are assigned to CalCm and Cbe
stretching modes, respectively. The normal coordinate
calculation presented in this work also supports this
interpretation (see Table 4-7).

Compared to CuOEC, the corresponding mode frequencies
of ZnOEC are generally lower owing to the core-size effect.
Moreover, the frequency shift magnitudes of meso-deuterated
ZnOEC are greater than those of CuOEC because the lower
frequency CalCm modes in ZnOEC can mix better with the CmH
deformations. However, the investigation of IR, resonance
Raman and polarization ratios for ZnOEC also reveals that
the deuteration shift patterns are virtually the same as

those for CuOEC (see Table 4-6). This means that the

121

vibrational mode localization phenomenon in chlorin
complexes does not depend on the central metal in the
chlorin, but rather the symmetry of chlorin.

One other cgklvibration is evident in the CuOEC
spectrum; that is the mode observed at 1543 cm4-in.both the
IR and Soret excited resonance Raman. This mode is not
enhanced by visible excitation and is assigned as a B-
symmetry, 7,5-methine bridge localized CaCm vibration on the
basis of its resonance Raman polarization (anomalous) and
meso-deuteration shift pattern (0 cm4-in CuOEC-aJiwg; 17
cm“1 in both CuOEC-7,6-d2 and CuOEC-d4) . In addition, an
analog of this mode is observed at 1527 cm4-in.ZnOEC. The
IR spectrum of ZnOEC shows that the 1535 cm-1 resonance
Raman band of ZnOEC is a mixture of the 1538 and 1527 cm4-
vibrational modes. Similar to the 1543 cm"1 band of CuOEC,
this band shows the anomalous polarization and a similar
frequency shift pattern to that for the meso—deuterated
chlorins (1 cm‘1 in ZnOEC-a,B-d2; 20 cm'1 in both ZnOEC-7,5-d2
and ZnOEC-d4). However, the frequency of this band is
anomalously high compared to the calculated frequency and is
not reproduced in any force field calculations reported

herein (Tables 4-3 and 4-8).

4.5.1.2) cbcb Vibrations
There are three cg%,vibrations expected in the high—
frequency region (2 A and 1 B). The fourth C5;,stretch is

associated with the reduced ring and expected at lower

122

frequency. 0f the three high-frequency cga,vibrations of
CuOEC, the two A-symmetry modes are observed in the
vibrational spectra (1583 and 1546 cm4). For ZnOEC, the
corresponding bands are observed at 1571 and 1538 cm4y
respectively. The methine deuteration behavior and normal
coordinate analysis of these two modes demonstrate that
there is mixing of the cgk‘and C9%,internal coordinates.
But the contribution of the C5%‘internal coordinate is not
great. The B-symmetry band is not observed for CuOEC but is
calculated at 1557 cm4-(FT‘I). In NiOEcuhuhuw, the analog of
the B-symmetry band is observed at 1572 cm*» Accordingly,
the B-symmetry vibration of CuOEC would most likely fall in

the 1560 - 1570 cm‘1 region.

4.5.1.3) 9.36 and 0.x Vibrations

Under C2 symmetry, there are eight CaN and six CaCb
modes expected in the high-frequency region (4 A and 4 B; 3
A and 3 B, respectively). Two other cg%,modes are
associated with the reduced ring and expected at lower
frequency. The CaCb and CaN vibrations are extensively mixed
with one another and with other motions of the macrocycle
(Table 4-7). Consequently, it is not generally appropriate
to classify the observed bands as a specific type of mode.
Inspection of the vibrational data for CuOEC reveals that
there are eight prominent bands that can be associated with

each and can vibrations (1403, 1388, 1373, 1362, 1318, 1155,

1141 and 1129 cm”). All of these bands have analogs in the

123

spectra of ZnOEC and NiOEC.”38 Most of them have been
identified on the basis of their 15N'shifts. Furthermore,

the vibrational frequencies of ZnOEC, CuOEC and NiOEC that

can be associated with CaLCb and C311 modes are very similar.
This means that the CaCb and CaN modes in chlorin complexes

do not depend on the central metal in the chlorin.

4 . 5 . 1. 4) can Vibrations

There are four'C5H deformations (2 A and 2 B). These

modes are extensively mixed with other vibrations of the
macrocycles as is evidenced by the complicated meso-
deuteration shift patterns observed in the 1100 - 1300 cm”-
spectral region. On the basis of the spectral data for CuOEC
and the normal coordinate calculations, the bands at 1310,
1238, 1215 and 1198 cm‘1 (Figure 4-11) are assigned as the
four CmH deformations. Boldt et al.36 assigned the 1310 cm-1
band as a mode localized on the a,B-methine bridges.

However, normal coordinate calculations presented here and
careful investigation of spectra indicate that the motion is
actually localized on the‘mfi-methine bridges. These
assignments for the CmH deformation were derived in the
following manner. In case of the B-symmetry 1310 cm"1 band
based on the normal coordinate calculations, a large meso-
deuteration shift is observed in CuOEC-7,5-d2 and CuOEC-d4.
This is not immediately apparent because the spectra of both
of these isotopomers exhibit an anomalously polarized band

in the 1320 - 1325 cm‘lregion.38 However, this latter band

124

is actually the analog of a band observed at 1318 cm"1 in
the normal isotopomer. In CuOEC, this band is very weak, but
its intensity increases in CuOEC-'y,<5--d2 and CuOEC-d4. The
fact that the 1318 cm4-band is a different mode not
associated with a.C%H deformation is clearly revealed in
both the IR and resonance Raman spectra of CuOEC-d,,'which
exhibit a strong band in this region. The apparent absence
of the 1310 cm‘1 band in the spectrum of CuOEC-a,B-d2 is due
to the fact that this mode upshifts approximately 10 cm4-131
this isotopomer and becomes a shoulder on the low-energy
side of the 1318 cm"1 band. This upshift is predicted by the
normal coordinate calculations (Table 4-7). The band
associated with the A-symmetry analog of the 1310 cm4-
'vibration is more difficult to identify. The 1198 cm-1 band
is attributed to this mode on the basis of its selective
shifts in the spectra of CuOEC-'y,<S-d2 and CuOEC-d4. The
assignment of the 1238 and 1215 (3111‘1 bands as the A and B
symmetry CmH deformations localized on the a,B-methine
bridges was also based on the selective meso-deuteration
shifts of these bands. Both of these modes are observed in

the spectra of CuOEC-7,5-d2 and both are absent in the
spectra of CuOEC-ozflnd2 and CuOEC-d4.33

In addition to the four bands discussed above, several
other modes are quite sensitive to meso-deuteration. In
particular, the 1275 cm“1 band appears to be comprised of
two meso-deuteration sensitive bands (one polarized and the

other anomalously polarized). In CuOEC-a,B-d2, the

125

anomalously polarized band upshifts to 1288 cm*, while the

polarized band downshifts to 1178 cm4. This reveals a band

at 1263 cm‘1 which is attributable to a CH2 deformation of
the ethyl group125- In CuOEC-7,6-d2, the anomalously polarized
band upshifts to approximately 1293 cm4, while the
polarized band does not shift. Deuteration sensitivity is
also observed for the 1141 cm‘1 band. In CuOEC-a,B-d2, this
mode does not shift, whereas in CuOEC-'y,<3-d2 this mode
upshifts to 1173 cm‘l. In CuOEC-d4, the 1141 cm‘1 band
upshifts to 1180 cmqu This band falls at nearly the same
frequency as the 1178 cm’1 band of CuOlE:C-<::,B-d2 (1275 cm”1 of
the normal isotopomer). Consequently, the 1180 cm"1 band of
CuOEC-d4 is probably a superposition of the upshifted 1141

cm“1 and downshifted 1275 cm4-bands.

4.5.2) comparison of the Force Fields

4.5.2.1) General Features of the Calculations

The objectives of performing normal coordinate
calculations on CuOEC were to develop a satisfactory force
field for MOEC complexes and to determine the factors that
influence the vibrational eigenvectors. The forms of the
eigenvectors were obtained from the normal coordinate
calculations to confirm the mode localization effect. The
semiempirical calculations on MOEC by Boldt et al.331 predict
that the normal modes of the metallochlorins are different

from those of metalloporphyrins. In particular, the

126

semiempirical method predicts that the vibrational
eigenvectors of a number of the skeletal modes are localized
on specific sectors of the macrocycle. The localization was
most pronounced in the vibrations involving the methine
bridges. This localization phenomenon was attributed to the
length alteration pattern that occurs in the CaCm bonds of
the OEC macrocycle.”:n"!130 Indeed, the vibrational data
obtained for selectively meso-deuterated CuOEC and ZnOEC
support the idea of localized eigenvectors for certain
modes.38 On the other hand, the the normal coordinate
calculations on NiOEC by Prendergast and Spiro109 predict
that the vibrational eigenvectors of this complex are
delocalized and are generally similar to those of NiOEP.
However, their calculations are not based on sufficient
experimental data (they did not use the experimental data
for NiOEC-ozfl-d.2 and NiOEC-d”. limited experimental data can
produce incorrect calculated results in a normal coordinate
analyses.

Normal coordinate calculations based on a series of
meso-deuterated CuOEC(CuOEC, CuOECwLB-dz, CuOEC-7,5-d2 and
CuOEC-d4) predict that certain vibrational eigenvectors are
localized. The localization phenomenon is manifested
regardless of the quality of fit to the experimental data.
The localization is particularly apparent in modes
associated with the CJ% stretches and CEH deformations.

These are illustrated in Figures 4—15-18, which depict

selected vibrational eigenvectors for CuOEC. These

 

Figure 4-15

127

Vibrational eigenvectors of CuOEC which
contain substantial contributions from

CaCm stretching vibrations. Displacements
are shown only for those atoms whose motions
contribute significantly to the normal mode
(10% or greater of the maximum atomic
displacement of a given mode).

128

 

 

calc. 1645 calc. 1616
obs. 1644 obs. 1602

’

88

Q 0

 

calc. 1491 calc. 1470
obs. 1507 obs. 1486

129

Figure 4-15 (cont.) Vibrational eigenvectors of CuOEC
which contain substantial contributions
from C5% stretching vibrations.
Displacements are shown only for those
atoms whose motions contribute significantly
to the normal mode (10% or greater of the
maximum atomic displacement of a given mode).

130

calc. 1618 calc. 1582
obs. ---- obs. 1584

0

‘

 

 

 

 

 

calc. 1453 calc. 1461
obs. 1543 obs. 1466

Figure 4-16

131

Vibrational eigenvectors of CuOEC which
contain substantial contributions from

CbCb stretching motions. Displacements are
shown only for those atoms whose motions
contribute significantly to the normal mode
(10% or greater of the maximum atomic
displacement of a given mode).

132

 

calc. 1571 calc. 1557
obs. 1583 obs,

 

calc. 1561
obs. 1547

Figure 4-17

133

Vibrational eigenvectors of CuOEC which
contain substantial contributions from CHI
deformations. Displacements are shown

only for those atoms whose motions
contribute significantly to the normal
mode (10% or greater of the maximum atomic
displacement of a given mode).

134

calc. 1291 calc. 1218
obs. 1307 obs. 1238

 

calc. 1218 calc. 1210
obs. 1215 obs. 1198

Figure 4-18

135

Vibrational eigenvectors of CuOEC which
contain substantial contributions from both
CaCb and CaN stretching vibrations.
Displacements are shown only for those atoms
whose motions contribute significantly to the
normal mode (10% or greater of the maximum
atomic displacement of a given mode).

136

 

calc. 1392 calc. 1385
obs. 1403 obs. 1388

 

calc. 1386 calc. 1355
obs. 1373 obs. 1362

137

(cont.) Vibrational eigenvectors of CuOEC
which contain substantial contributions

from both CaCb and CaN stretching vibrations.
Displacements are shown only for those atoms
whose motions contribute significantly to
the normal mode (10% or greater of the
maximum atomic displacement of a given mode).

Figure 4-18

calc. 1343 calc. 1315
obs. 1352 obs. 1318

calc. 1313
obs. ----

 

139

eigenvectors were calculated by using'FT’I. On the basis of
these results, it is clear that mode localization is a
general consequence of symmetry lowering and does not depend
on the fine details of the force field or the central metal.
This conclusion is further supported by the observation that
the isotope-shift patterns observed for selectively meso-
deuteriated ZnOEC are virtually identical to those observed
for CuOEC. Furthermore, Li and Zgierskili34 recently reported
detailed normal coordinate calculations for free base and
NiOD porphine. Similar localization occurs as a consequence
of the bond-length disparities that occur in the low
symmetry free base. The localization is apparent in the
eigenvectors of free base porphyrin despite the fact that
the bond length disparities82 are significantly smaller than

those in MOEC complexes.“3fl9430

4.5.2.2) Relative Merits of Individual Force Fields

Inspection of Table 4-8 reveals that FF I ~ FF IV are
approximately comparable in their quality as judged by the
calculated average errors in the frequencies and meso-
deuteration shifts. However, none of these force fields
adequately reproduces all of the detailed spectral features.
For example, the mode observed at 1602 cm4-downshifts
12 cm’1 upon a,B-methine deuteration, but only 2 cm'1 upon
7,6-methine deuteration. FF I predicts shifts of 11 cm“1 and
5 cm*, respectively, which is a reasonable representation

of the observed shift pattern. In contrast, FF H predicts

140

shifts of 10 cm4-and 8 cm”; FT‘IH predicts shifts of 8 cm-1
and 6 cm4; FT‘IV predicts shifts of 6 cmd-and 12 cm4» These
predicted shift patterns are considerably less satisfactory
than those obtained with.FT‘I. On the other hand, FT‘I fails
to predict the shift pattern observed for the 1584 cm4-
band, whereas the other three force fields give a
satisfactory representation of the observed pattern. As
Table 4-8 shows, the 1584 cm‘1 band downshifts 10 cmd-upon
(LB-methine deuteration and 4 cm“1 upon yfi-methine
deuteration. FF I predicts shifts of 8 cm‘1 and 14 cm‘l,
respectively. FF H predicts shifts of 15 cm4-and 6 cm”; FF
IH predicts shifts of 12 cm-1 and 4 cm47 FT‘IV predicts
shifts of 13 ch-and 7 cm4. In the case of the 1543 cm‘1
band, although FF H and.FT’IH do give a satisfactory
representation of the observed shift pattern, all of the
force fields significantly underestimate the frequency of
this mode. This mode alone accounts for approximately 30%
of the average error in the frequencies (Table 4-9). This
may suggest that higher order interaction, not accounted for
in the valence force fields, is important in the MOEC
macrocycle. Even if none of these force fields exactly
reproduce all of the observed shift patterns, these force
fields give a reasonable accounting of the vibrational
characteristics of MOEC complexes.

One of the criteria that has been used in deriving a
satisfactory force field is the requirement that the force

constants scale with the bond lengths.109r128 Of the

141

calculations reported here, this requirement is met by FF I
and.FT'IV. In contrast, the force constants used for the
cgcmjbonds in FF H are in reverse order from those that

should be used based on the bond lengths. In the case of FF

1H, the refinement increases the Ca(IIl)-N(II) stretching force

constant from 5.411 mdyn/fl to 7.168 mdyn/X. Simultaneously,

the CaCm force constant for the bond adjacent to ring I is

reduced from 6.819 mdyn/K to 5.138 mdyn/K. Both of these

 

refined values are unreasonable based on the lengths of the
CaCm and CaN bonds in tetrapyrrolic macrocycles.135
Regardless, both FF H.and.FT‘IH give reasonable average
errors in both the frequencies and meso-deuteration shifts.
Indeed, by these criteria, FF H1 is the "best". Accordingly,
the bond-length criterion, while appealing, is not
necessarily a satisfactory indictor of the merit of a force
field. Li and Zgierski134 found that the inclusion of certain
1,4 interaction constants was important in obtaining an
accurate force field for porphines. In the calculations on
CuOEC, the effects of 1,4 interactions were explored only in
a limited fashion. A detailed assessment of the effects of
these (and higher order) interactions is not possible
because there are a large number of these types of
interactions and the experimental information is too limited

to allow an evaluation of their importance.

142

4.6) Summary and Conclusion

The valence force field calculations reported herein
indicate that various force fields satisfactorily reproduce
the general features of the vibrational spectra of CuOEC.
However, none of these force fields reproduces the
frequencies and meso-deuteration shift patterns of all of
the modes. This suggests that 1,4 and higher interaction

constants make substantial contributions to the valence

 

force field.

Meso-deuteration shift patterns of CuOEC and ZnOEC show
that mode localization phenomena are not properties of the
central metal but those of metallochlorin symmetry. All of
the empirical force fields predict that the eigenvectors of
certain skeletal modes are localized on sectors of the
macrocycle. This result is in general agreement with the

prediction of the QCFF/PI calculation by Boldt et al.36

CHAPTER 5

CONCLUSIONS AND FUTURE WORK

The goal of this research project was to explore the
effects of lower symmetry in modified macrocycles by means
of investigating the electronic and vibronic properties of
metalloporphycenes and metallochlorins.

Metalloporphycenes, which are isomers of the
extensively studied metalloporphyrins, qualify as potential
agents for tumor marking and photodynamic therapy owing to
their high absorption intensity in the wavelength region
above 620nm.33‘35 Furthermore, metalloporphycene represents
an example of a macrocycle of true Dfllsymmetry, which is
approximated by heme a and metallobacteriochlorins.41‘42 For
these reasons the electronic and vibronic properties of
porphycenes are extremely interesting. The lower symmetry
(D60 Of metalloporphycene compared to the Dfllsymmetry of
metalloporphyrin lifts x, y degeneracy and reduces
configuration interactions in the electronic transitions of
metalloporphycene. Close investigation of absorption band
positions demonstrates that the w e'w” states of
metalloporphycenes are adequately defined by one-electron

wave functions, unlike those of metalloporphyrins.

143

144

Vibrational analyses of resonance Raman spectra of NiPPC,
NiPPC-d1(py) and NiPPC-d4(et) give some insight into the
normal mode composition. However, more complete vibrational
analyses should be executed by normal coordinate
calculations based on the vibrational data of additional
isotopically substituted compounds (for example 15N
substitutied metalloporphycene). Vibrational eigenvectors
based on the normal coordinate calculations will help to
understand the consequences of symmetry-lowering in
metalloporphycene. The vibrational properties of
metalloporphyrins in biological systems are influenced by
the central metal, axial ligation, peripheral substitution,
ring oxidation, reduction etc. Therefore, the vibrational
properties of metalloporphycenes for CO or CN' bounding as
axial ligands and for porphycene cations or anions will be
one of the interesting fields to investigate in the future.
Metallochlorin is a metalloporphyrin derivative in
which a ng,bond of one of the pyrrole rings has been
reduced.9 Metallochlorins are ubiquitous in nature,
occurring as the chlorophyll pigments in photosynthetic
systemsuh in.marine worm pigments11 etc. Reduction of a C61)
bond in a metalloporphyrin lowers the molecular symmetry
from D4h to C2 and has a profound effect on the entire
macrocycle. Therefore, the vibrational characteristics of
the metallochlorins are particularly important for gaining
an understanding of symmetry-lowering effects in modified

macrocycles. Accordingly, many groups3‘5v38'109 have studied the

145

vibrational data. There remain, however, some controversies
in the interpretation of the vibrational modes of
metallochlorins. As discussed in chapter 4, the main
disagreement involves vibrational mode localization effects
in metallochlorins. Bocian et al.-36 have examined NiOEC by
using a semiempirical quantum force field (QCFF/PI) method.
Their calculations predict that the forms of the vibrational
eigenvectors in metallochlorins are in general different
from those of porphyrins. In contrast, empirical
calculations (GF matrix method) on NiOEC by Prendergast and
Spiro109 predict that the vibrational eigenvectors of
metallochlorins are similar to those of metalloporphyrins.
The origin of the disagreements between the empirical and
semiempirical normal coordinate calculations on NiOEC is not
clear. One possible reason for disagreement is limited
experimental data. For the development of an accurate
empirical force field it is necessary to obtain data from a
variety of isotopomers. As a result of normal coordinate
calculations for CuOEC, a good start in understanding the
controversy for mode localization effects in chlorin has
been made and the vibrational mode properties in chlorins
can be understood more clearly.

As described in this dissertation, normal coordinate
calculations based on the IR and resonance Raman spectra of
selectively meso-deuterated CuOEC (CuOEC-h), CuOEC-aB-dz,

CuOEC-yd-dz, CuOEC-d4) show that the eigenvectors of certain.

skeletal modes are localized on sectors of the macrocycle.

146

The meso-deuteration shift patterns of CuOEC and ZnOEC show
that mode localization phenomena are not dependent on the
identity of the central metal but are inherent properties of
metallochlorins that result from their reduced symmetry.
Therefore, normal coordinate calculations by using
selectively meso-deuterated ZnOEC vibrational spectroscopic
data will be useful in testing and exploring the effect of
mode localization in lower symmetry macrocycles. Like
metalloporphyrins, the vibrational properties of
metallochlorins will be influenced by the central metal,
axial ligation, peripheral substitution, ring oxidation,
reduction etc. Therefore, an analysis of the vibrational
properties of metallochlorins within the centext of these
factors will help in providing an understanding of the
reduced macrocycle. Once the mode assignments for
metallochlorins have been firmly established, the
vibrational properties of metallochlorins in biological

systems will be an interesting topic for investigation.

LIST OF REFERENCES

10.

11.

12.

147

LIST OF REFERENCES

Dolphin,D., Ed. The Porphyrin; Academic Press: New York,
1978; Vol. I

Dolphin,D., Ed. The Porphyrin; Academic Press: New York,
1978; VOl.‘VH

Vogel,E.; Kocher,M.; Schmickler,H.; Lex,J. Angew. Chem.,
Int. Ed. Engl. 1986, 25, 257-259.

Vogel,E.; Balci,M.; Pramod,K.; Koch,P.; Lex,J.;
Ermer,O.; Angew. Chem., Int. Ed. Engl. 1987, 26,
928-931.

a) Vogel,E.; Kocher,M.; Balci,M.; Teichler,I.; Lex,J.;
Schmickler,H.; Ermer,O. Angew. Chem., Int. Ed. Engl.
1987, 26, 931-934. b) Furenlid,L.R.; Renner,M.R.;
Smith,K.M.; Fajer,J. J. Am. Chem. Soc. 1990, 112,
1634-1635.

Wehrle,B.; Limbach,H.-H.; Kocher,M.; Ermer,O.; Vogel,E.
Angew. Chem., Int. Ed. Engl. 1987, 26, 934-936.

Schulpmann,J.; Huber,M.; Toporowicz,M.; Kocher,M.;
Vogel.E.; Levanon,H.; Mobius,K. J. Am. Chem. Soc. 1988,
110, 8566—8567.

Vogel,E.; Kocher,M.; Lex,J.; Ermer,O. Irs. J. Chem.
1989, 29, 257-266.

Scheer,H. in The Porphyrin; Dolphin,D., Ed.; Academic
Press: New York, 1978; Vol. VII, pp 1-44.

Svev,W.A. in The Porphyrin; Dolphin,D., Ed.; Academic
Press: New York, 1978; Vol. V, pp 341-399.

Ballantine,J.A.; Psaila,A.F.; Pelter,A.; Murray-Rust,P.;
Ferrito,V.; Schembri,P.; Jacaarini,V. J. Chem. Soc.,
Perkin Trans. I, 1980, 1080-1089.

Babcock,G.T.; Ingle,R.T.; Oertling,W.A.; Davis,J.C.;
Averill,B.A.; Husle,C.L.; Stufkens,D.J.;
Bolscher,B.G.J.M.; Wever,R. Biochem. Biophys. Acta 1985,
828, 58-66.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

148

Andersson,L.A.; Loehr,T.M.; Lim,A.R.; Mauk,A.G. J. Biol.
Chem. 1984, 259, 15340-15349.

Brittian,T.; Greenwood,C.; Barher,D. Biochim. Biophys.
Acta 1982, 705, 26-32.

Chang,C.K.; Wu,W. J. Biol. Chem. 1986, 261, 8593-8596.

Chang,C.K.; Barkigia,K.M.; Hanson,L.K.; Fajer,J. J. Am.
Chem. SOC. 1986, 108, 1352-1354.

Strauss,S.H.; Silver,M.E.; Long,K.M.; Thompson,R.G.;
Hudgens,R.A.; Spartalian,K.; Ibers,J.A. J. Am. Chem.
Soc. 1985, 107, 4207-4215.

P1att,J.R. in Radiation Biology; Hollaender,A. Ed.;
Mcgraw-Hill, New York, Vol. III, pp

Longuet-Higgins,H.; Rector,C.W.; P1att,J.R. J. Chem.
Phys. 1950, 18, 1174-1181.

Gouterman,M. J. Chem. Phys. 1959, 30, 1139-1161.
Gouterman,M. J. Mol. Spec. 1961, 6, 138-163.

a) Califano,M. Vibrational States; John Wiley & Sons,
New York, 1-15. b) Centeno,J.A. Dissertation (1987),
Michigan State Univ. East Lansing, MI, 48824

Heilter,W. in The Quantum Theory of Radiation, 3rd ed.
Oxford University Press, New York, 1954.

Albrecht,A.C. J. Chem. Phys. 1960, 33, 156-169.
Clark,R.J.H.; Stewart,B. Struct. Bond. 1979, 36, 1-80.

Ozaki,Y.; Iriyama,K.; Ogoshi,H.; Ochiai,T.; Kitagawa,T.
J. Phys. Chem. 1986, 90, 6105-6112.

Callahan,P.M.; Babcock,G.T. Biochemistry 1981, 20,
952-958.

Spiro,T.G.; Stong, J. D.; Stein, P. J. Am. Chem. Soc.
1979, 101, 2684-2655.

Choi, S.; Spiro, T. G.; Langry, K. C.; Smith, K. M.;
Budd, D. L.; La Mar, G. N. J. Am. Chem. Soc. 1982,
104, 4345-4351.

Renner,M.W; Forman,A.; Wu.W.; Chang,C.K.; Fajer,J. J.
Chem. SOC. 1989, 111, 8619-8621.

 

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

149

Whitlock, H. W. Jr.; Hanauer, R.; Oester, M. Y.; Bower,
B. K. J. Am. Chem. SOC. 1969, 91, 7485-7489.

Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. In the
purification of Laboratory Chemicals 2nd Ed. Pergamon
Press; New York, 1980.

Aramendia,P.F.; Redmond,R.W.; Novell,S.; Schuster,W.;
Braslavsky,S.E.; Schaffner,K.; Vogel,E. Photochem.
Photobiol. 1986, 44, 555-559.

Levanon,H.; Toporowicz,M.; Ofir,H.; Fessenden,R.W.;
Paritosh,K.D.; Vogel,E.; Kocher,M.; Parmod,K. J. Phys.
Chem. 1988, 92, 2429-2433.

Redmond,R.W.; Valduga,G.; Novell,S.; Braslavsky,S.E.;

Schaffner,K.; Vogel,E.; Pramond,K.; Kocher,M. J.
Photochem. Photobiol. B 1989, 3, 193-207.

Boldt,N.J.; Donohoe,R.T.; Birge, R.R.; Bocian,D.F. J.
Am. Chem. Soc. 1987, 109, 2284-2298.

Andersson, L.A.; Loehr,T.N.; Stershic,M.T.;Stolzenberg,
A.M. Inorg. Chem. 1990, 29, 2278-2285.

Fonda,H.N.; Oertling,W.A.;Salehi,A.; Chang,C.K.;
Babcock,G.T. J. Am. Chem. Soc. 1990, 112, 9497-9507.

Andersson,L.A.; Loehr,T.M. Cotton,T.M.; Simpson,D.J.;
Smith,K.M. Biochim. Biophys. Acta. 1989, 974, 163-179.

Donohoe,R.J.; Frank,H.A.; Bocian,D.F.; Photochem.
Photobiol. 1988, 48, 531-537.

Schick,A.G.; Bocian,D.F. Biophys. Biochem. Acta. 1987,
895, 127-154.

Woodruff,W.H.; Kossler,R.J.; Ferris,N.S.; Dallinger,R.F.
Adv. Chem. Ser. 1982, 201, 625-659.

Kitagawa,T.; Ozaki,Y. Struct. Bonding 1987, 109, 71-114.
Shelnutt,J.A. J. Am. Chem. SOC. 1987, 109, 4169-4173.
Oertling,W.A.; Salehi,A.; Chung,Y.C.; Leroi,G.E.;
Chang,C.K.; Babcock,G.T. J. Phys. Chem. 1987, 91,
5587-5898.

Salehi,A.; Oertling,W.A.; Fonda,H.N.; Chang,C.K.;
Babcock,G.T. Photochem. Photobiol. 1988, 48, 525-530.

 

47.

48.

49.

50.

51.

52.

53.

54.

55.

56

57.

58.

59.

60.

61.

62.

63.

64.

150

Donohoe,R.T.; Duchowski,J.K.; Bocian,D.F. J. AM. Chem.
Soc. 1988, 110, 6119-6124.

Oertling,W.A.; Salehi,A.; Chang,C.K.; Babcock,G.T. J.
Phys. Chem. 1989, 93, 1311-1319.

Czernuszewicz,R.S.; Macor,K.A.; Li,X.-Y.; Kincaid,J.K.;
Spiro,T.G. J. AM. Chem. SOC. 1989, 111, 3860-3869.

Ksenofontova,N.M.; Maslov,V.G.; Sidorov,A.N.; Bobovich,
Y.S. Opt. Spectrosc. 1976, 40, 452-465.

Teraoka,J.; Hashimoto,S.; Sugimoto,H.; KitagaWa,T. J.
Am. Chem. Soc. 1987, 109, 180-184.

Donohoe,R.T.; Atamian,M.; Bocian,D.F. J. Am. Chem. Soc.
1987, 109, 5593-5599.

Spiro,T.G. in Iron Porphyrins; Lever,A.B.P., Gray,H.B.,
Eds.; Addison-Wesley, Reading,MA, 1983; Part2, 89-159.

Lutz,M.; Robert,B. in Biological Applications of Raman
Spectrocsopy; Spiro,T.G. Ed.; John Wiley & Sons, New
York, 1988, 347-411.

Callahan,P.M.; Cotton,T.M. J. Am. Chem. Soc. 1987, 109,
7001-7007.

Schiemke,A.K.; Eirich,L.D.; Loehr,T.M. Biochim. Biophys.
Acta. 1983, 748, 143-150.

Schiemke,A.K.; Scott,R.A.; Shelnutt,J.A. J. Am. Chem.
SOC. 1988, 110, 1645-1646.

Mayer,E.; Gardiner,D.J.; Hester,R.E. Mol. Phys. 1973,
26, 783-787.

Salama,S.; Spiro,T.G. J. Raman Spectrosc. 1977, 6,
57-60.

Abe,M.; Kitagawa,T.; Kyogoku,Y. J. Chem. Phys. 1978, 69,
4526-4534.

Gladkov,L.L.; Solovyov,K.N. Spectrochim, Acta 1986, 42A,
1-10.

Li,X.-Y.; Czernuszewicz,R.S.; Kincaid,J.A.; Stein,P.;
Spiro,T.G. J. Phys. Chem. 1990, 94, 47-61.

Atamian,M.; Donohoe,R.J.; Lindsey,J.R.; Bocian,D.F. J.
Phys. Chem. 1989, 93, 2236-2243.

Li,X.-Y.; Czernuszewicz,R.S.; Kincaid,J.A.; Su,O.Y.;
Spiro,T.G. J. Phys. Chem. 1990, 94, 31-47.

65.

66.

67.

68.

69.

70.

71.

72.

73.

74.

75.

76.

77.

78.

79.

80.

81.

82.

151

Simpson,W.T. J. Chem. Phys. 1949, 17, 1218-1221.
P1att,J.R. J. Chem. Phys. 1950, 18, 1168-1173.
Pariser,R.; Parr,R.G. J. Chem. Phys. 1953, 21, 767-776.
Moffitt,W. J. Chem. Phys. 1954, 22, 1820-1829.
Shelnutt,J.A. J. Phys. Chem. 1984, 88, 4988-4992.

Shelnutt,J.A.; Prtiz,V. J. Phys. Chem. 1985, 89,
4733-4739.

Obtained from the correlation tables in: Wilson,E.B.;
Decius,J.C.; Cross, P.C. In Molecular Vibrational: The
Theory of Infrared and Raman Vibrational Spectra; Dover
Publications, 1955; Appendix X. Note that the C2" axes

of MOEP are retained in MPPC.
The effects of ring bromination and metal substitution
on the individual MO energies will be more readily

understood when calculations yielding orbital
coefficients become available.

Bernstein,H.T. Phil. Trans. R. Soc. London A 1979, 293,
287-302.

Spiro,T.G.; Strekas,T.C. Proc. Natl. Acad. Sci. U.S.A.
1972, 69, 2622-2626.

Shelnutt,J.A.; O'Shea,D.C.; Yu,N.-T.; Cheung,L.D.;
Felton,R.H. J. Chem. Phys. 1976, 64, 1156-1165.

Shelnutt,J.A.; Cheung,L.D.; Chang,R.C.C.; Yu,N.-T.;
Felton,R.H. J. Chem. Phys. 1977, 66, 3387-3398.

Nishimura,Y.; Hirakawa,A.Y.; Tsuboi,M.J. Mol. Spectrosc.
1977, 68, 335-358.

Shelnutt,J.A. J. Chem. Phys. 1981, 74, 6644-6657.

Spaulding,L.D.; Chang,C.C.; Yu,N.-T.; Felton,R.H. J. Am.
Chem. Soc. 1975, 97, 2517-2525.

Huong,P.V.; Pommier,J.-C.C.R. Acad. Sci. Ser. C 1977,
285, 519-522.

Parthasarathi,N.; Masen,C.; Yamaguchi,S.; Spiro,T.G. J.
Am. Chem. Soc. 1987, 109, 3865-3871.

Cullen,D.L.; Mayer,E.F. J. Am. Chem. Soc. 1974, 96,
2095-2102.

83.

84.

85.

86.

87.

88.

89.

90.

91.

92.

93.

94.

95.

96.

97.

98.

99.

152

a) Timkovich,R.; Tulinsky,A. J. Am. Chem. Soc. 1969, 91,
4430-4432. b) Silvers,S.; Tulinsky,A. J. Am. Chem. Soc.
1990, 112, 1634-1635.

Scheidt,W.R.; Lee,Y.T. Struct. Bonding 1987, 64, 1-70.
and references therein.

Smith,D.W.; Williams,R.T.P. Struct. Bonding 1969, 7,
1-45.

Adar,F. in The Porphyrins; Dolphin,D., Ed.; Academic:
New York, 1979; Vol.III, 167-209.

Neya,S.; Chang,C.K.; Vogel,E. manuscript in
preparation.

Kitagawa,T.; Abe,M.; Kyogoku,Y.; Ogoshi,H.; Sugimoto,H.;
Yoshida,Z. Chem. Phys. Lett. 1977, 48, 55-58.

Svec,W.A. In The Porphyrins; Dolphin, D., Ed.;
Academic Press: New York, 1978; Vol. 5, 341-349

Eglington,D.G.; Barber,D.; Thomson,A.J.; Greenwood,C.;
Segel,A.W. Biochim. Biophys. Acta 1982, 703, 187-195.

Okamura,M.Y.; Feher,G.; Nelson,N. In Photosynthesis;
Godvindjee, Ed.; Academic Press, 1982; Vol. I, 195-272.

Siegel,J.M.; Rueger,D.C.; Barber,M.J.; Krueger,R.J;
Orme-Johnson,N.R.; Orme-Johnson,W.H.J. Biol. Chem.
1982, 257, 6343-6350.

Ellefson,W,L.; Whitman,W.B.; Wolfe,R.S. Proc.
Natl. Sci. USA 1982, 79, 3707-3710.

Sibbet,S.S.; Hurst,J.K. Biochemistry 1984, 23,
3007-3013.

Bondoc, L. L.; Chan, M. H.; Price, M. A.; Timkovich,
R. Biochemistry 1986, 25, 8458-8466.

Chatfield,M.J.; La Mar,G N.; Lecomte,J.T.J.;
Balch, A.L.; Smith,K.M.; Langry,K.C. J. Am. Chem.
SOC. 1986, 108, 7108-7110.

Mcree,D.E.; Richardson,D.C.; Richardson,J.S.;
Siegel,L.M. J. Biol. Chem. 1986, 261, 10277-10281.

Andersson,L.A.; Sotiriou,C.; Chang,C.K.; Loher,T.M.
J. Am. Chem. Soc. 1987, 109, 258-264.

Schick,G.A.; Bocian,D.F. Biochim. Biophys. Acta
1987, 895, 127-154.

153

100. Ozaki,Y.; Kitagawa,T.; Ogoshi,H. Inorg. Chem. 1979,
18, 1772-1776.

101. Ozaki,Y.; Iriyama,K.; Ogoshi,H.; Ochiai,T.;
Kitagawa,T. J. phys. Chem. 1986, 90, 6113-6118.

102. Fujiwara,M.; Tasumi,M. J. Phys. Chem. 1986, 90,
250-255.

103. Fujiwara,M.; Tasumi,M. J. Phys. Chem. 1986, 90,
5646- 5650.

104. Andersson,L.A.; Loehr,T.M.; Chang,C.K.; Mauk, A.G.
J. Am. Chem. Soc. 1985, 107, 182-191.

105. Andersson,L.A.; Loehr,T.M.; Sotiriou,C.; Wu,W.;
Chang,C.K. J. Am. Chem. SOC. 1986, 108, 2908-2916.

106. Boldt,N.J.; Bocian,D.F. J. Phys. Chem. 1988, 92,
581-586.

107. Ogoshi,H.; Watanabe.E.; Yoshida,Z.; Kincaid,J.;
Nakamoto,K. Inorg. Chem. 1975, 14, 1344-1350.

108. Donohoe,R.J.; Atamian,M.; Bocian,D.F. J. Phys.
Chem. 1989, 93, 2244-2252.

109. Prendergast,K.; Spiro,T.G. J. Phys. Chem. 1991, 95,
1555-1563.

110. Lutz,M. In Advances in Infrared and Raman
Spectroscopy; Clark,R.J.H.; Hester,R.E., Eds.;
Wiley: New York, 1984; Vol. 11, 211-300.

111. Cotton,T.M.; Van Duyne,R.P. J. Am. Chem. Soc. 1987,
1981, 103, 6020-6026.

112. Callahan,P.M.; Cotton,T.M. J. Am. Chem. Soc. 1987,
109, 7001-7007.

113. Cotton,T.M.; Timkovich,R.; Cork,M.S. FEBS. Lett.
1981, 133, 39-44.

114. Ching,Y.; Ondrias,M.R.; Rousseau,D.L.; Muhoberac,
B.B.; Wharton,D.C. FEBS. Lett. 1982, 138, 239-244.

115. Andersson, L.A.; Loehr,T.M.; Weishih,W.; Chang,C.K;
Timkovich,R. FEBS. Lett. 1990, 267, 285-288.

116. Andersson,L.A.; Loehr,T.M.; Thompson,R.G.;
Strauss,S.H. Inorg. Chem. 1990, 29, 2142-2147.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

127.

128.

129.

130.

131.

132.

154

Ondrias,M.R.; Carson,S.D.; Hirasawa,M.; Knaff,D.B.
Biochim. Biophys. Acta 1985, 830, 159-163.

Han,S.; Madden,J.F.; Thompson,R.G.; Strauss,S.H.;
Siegel,L.M.; Spiro,T.G. Biochemistry 1989, 28,
5461-5470.

Melamed,D.; Sullivan,E.P.; Prendergast,K.; Strauss,
S.H.; SpirO,T G. Inorg. Chem. 1991, 30, 1308-1319.

Lai,K.K.; Yue,K.T. J. Raman Spectrosc. 1990, 21, 21-26.

Lai,K.K.; Moura,I.; Liu,M.Y.; Legall,J.; Yue,K.Y.
Biochim. Biophys. Acta 1991, 1060, 25-27.

Procyk,A.D.; Bocian,D.F. J. Am. Chem. Soc. 1991,
113, 3765-3773.

Gladkov,L.L.; Starukhin,A.S.; Shulga,A.M.
Spectrochim. Acta 1987, 43A, 1125-1134.

Wilson,E.B.; Desius,J.C.; Cross,P.C. In Molecular
Vibrations, McGraw-Hill, New York, 1955.

Mclntoch,D.F.; Michaelian,K.M. Can. J. Spectrosc.
1979, 24, 1-10. and references therein.

Morino,Y.; Kuchitsu,K. J. Chem. Phys. 1952, 20, 1809-
1810.

Strauss,S.H.; Silver,M.E.; Long,K.M.; Thompson,

R.G.; Hudgens,R.A.; Spartalian,K.; Ibers,J.A.

J. Am. Chem. Soc. 1985, 107, 4207-4215.

Schachtschneider,J.H. Vibrational Analysis of
Polyatomic Molecules; Shell Development Co.;

Emeryville, Ca, 1962; Vol. LHI.

Li,X.-Y.; Czernuszewicz,R.S.; Kincaid,J.R.; Stein,P;
Spiro,T.G. J. Phys. Chem. 1990, 94, 47-61.

Spaulding,L.D.; Andrews,L.C.; Williams,G.J.D.
J. Am. Chem. SOC. 1977, 99, 6918-6923.

Gallucci,J.C.; Swepston,P.N.; Ibers,J.A. Acta
Crystallogr. Sect. B: Struct. Crystallogr. Chem. 1982,
383, 2134-2139.

Because of its longer singlet lifetime and higher
fluoresence yield, it was impossible to record visible
excitation raman spectra for ZnOEC and its selectively
deuterated compounds.

155

132. Moustakali,I.; Tulinsky,A. J. Am. Chem. Soc. 1973,
95, 6811-6815.

133. Collins,D.M.; Hoard,J.L. J. Am. Chem. Soc. 1970,
92, 3761-3771.

134. Li,X.-Y.; Zgierski,M.Z. J. Phys. Chem. 1991, 95,
4268-4287.

135. Scheidt,W.R. In the Porphyrins; Dorphin, D. Ed.;
Academic Press; New York, 1978; Vol HI. 463-512.

 

 

 

 

 

 

 

 

 

 

...
,‘4 .4.» v... . ....

”-6 - v
“4.4.4.6”...
. ...“.

 

+4446~wa~w ~
. ... ...... '." .4.
. 6

‘Ipiunl\ll . -
I.\4'I‘VI\‘\1Q.VIII\-’l‘!'l\
‘ ......

'u

w'; "$.12...

:66
‘w. 1 .m- - M'-

. mu. 4-.-
4

. .6»
...” v—uA- ~ - .-....,..._, 6......" ."... >I'I'l‘l‘ll'il'
8.3,». :64 N. .:..:.:u.wv~m ' ' 'f " ' '.' m
t . r

.. um
.63....“ t "‘ ' ‘ " ' ' "
3v 3 “HI.” nv::-IIL‘=H~W"~' 5 "
... or -v -u...'...m-uvn-\v-t #- ..-I--~_-"““" "- ' .i‘ "

"..f-....5.1'. . 4.4....“ ,. rm»: .y-nu_"W\W"’"""""""1"" "7 ' "'
.- rw- mv v-I'I vm-I;""'l|l ; J. ' ‘ v ' "‘ kz.‘ ‘ 43.4.1. I U

“a W I ~ _ y). .6..- -:-;!A“"‘" y . ‘f 'mb.-.....'..:I,
.. n r... .... u‘vln ‘ ' ' "H”
- :71. t ”In" .... H “6164'." . ,1..,unvuu'\!14-

h.m.nmm4:mm "”1““! > " “3:"

,wyn'hvgrw I ‘ I H I“.

HIGQN STQTE UNIV.

LIBRQR

‘9' 29030987199

I ll