'1

  
 

   
  
 
  
   
 
      
   

 
 
 
   
   

    
      
    
  

    
       

 

'_ M “$333
w

  

 

 

   
 
     

      
 
   

"AM." - .. ¥

fl‘t' gh‘mwll‘nu‘”, Y! h 321“. 5P .144

Whifi“ T “3.1, :9. "1’! mm Yb??? V
a“ ukkwa '"Imr £"w

.
>. O
.
4.9 v u —
1 -
'1‘}-

:‘u'.
Ji-‘u‘ .
.~ .
_.’-v‘v—w—
4....—

Wing: .
‘1 up“ ’2‘“

..
F
{'3’ ..
.'.._.."

01“

     
  

“ 413‘ h,

"I §

Li. H “
1-' lig‘.“hlu'p.,§:k“vuh :Mzg' “5':

 

.333?
5? gm.-

         
   

. i
.
‘-
:3:-
j‘

  

EVEN sfilfi- bit} 1,“ gig?

     
   
   
       
     
 
  
   
 

.2 ' .
73:..53 ,,......,"'.
;_.-<:

    

    
    

  
   
  
  

    

“ «“9 i-iu a” b 1? =. AL":

.1 ‘ nigh‘.‘ 14.12. is“; "1 mile ‘ .fii‘l. jfi'fik {'12:th pig-LN-Ifk “Fagin 33$ ‘
fiwwwfi””§ ' “%~fi“‘ ¥%ge $§£fi§*.
11.39.. » ' 1"}:1¥Ia'~3‘§ ‘. .mfi: :..r“:-t:.v<€s'fis' .

kafl

.- . fl 13:;
V 15' 13E? 93

5‘" £113. . :i .‘4: y:
_1¥5§sfia;fw ..». .5. 1.

.. max-- -

‘ .I

   

   
 
 
     
 

In

  
 
 

:5. q

1 [WW My my

$325211 :1 1‘1::I.1 . '1: JER3" Iii: uu‘L My“? (it; “fig; ’ilglil‘u
h

“Mnd%tg NWu a
W? ”"51! i" ‘1:- “E? W!

 

év c'. tit—5?.” kg“: 1‘30""?
5‘“: mm: Mm
-.!:.-:,‘, Ila-I? tilfil

”—-
4

 
  
 
 
 

 
  
 

u m)“ r}, 1?: "h,“ (Kit: “wig: ‘vj 3i
{:3}:

2;th .. '...

’l..

 

      

      
     
   
  
   
  

 
 
 
 

     

  

'-'r.-
W'-
._._- :3:

KW}? o'.‘ uh“,
"7W

   
   

““5?131‘1“*!E!‘E13;;1‘“Ht12!‘JJ'N' :.
E}; “HM 1' $8,? I rt} ‘ 3L»!

 
 
  
 
   
     
   
 
   
    
  
 
  

 

! $121. ,
. It‘dvgnu '.. .

. 3 i Vidifi'ru' '4; .
22 h kw: ”‘2’; ,_
V ‘ e . .
E '1: .1 "h? 1..
: 2 51‘ w
-I v:

iii 12%"

. 5, 7:!

{

   

"'l

     
  
  
    
   
 
 
   
    
  
  
  

‘1:th .‘1

-..rv.-v-

_._
0- -.CU> .
7'» ,..- .

‘ ;":'&5i£*3§

:: ‘. . -;'

 
  
  
   
  

~l~
I”:
' [lug-1"" “2'1 My.

"hit:

 
 
 
  
 

 

  
 
   
       
 
  
 
  
 

K v
-Ax .
3&9

't‘;.:
‘3-
.2}

11”}
y“: cl} 21:4,“.
92;:‘4‘3‘21‘1‘

'1:

  
    
  

. \ .:.
‘C
\S
'9‘

    
  
      

 

”I

  
   
     
 

 

 

. Id:
"5411'”

'...
.-...-..~_._

     
    

    

 
 

‘. . a_
-‘-:| .1.“ )I.
m."- J fling; '7
, .. h
4w. .
{Mi 9 n
(1"! ‘-
‘ .l .
{9' ‘

      
  

 

WE.
.fihm

 
 
   

‘ 1 )
‘57,.f‘lgrg-Jl.
“WW “ ”’3.er
’ 'Nu‘

 

Fl. I
‘ "fur: ,

 

 

SIYY LIBRARIES

IMHWIWIH“WilliNI‘IiitliHlltili I

3129301682

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

PhotOphysics of Coordination Compounds
Intercalated into Layered Metal Phosphates

presented by

Eric Alan Saari

has been accepted towards fulfillment
of the requirements for

Ph.D. Chemistry

degree in

MG New

Major professor

 

 

Date 3 Nth iqci 8

MSU is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

LIBRARY

Michigan State

Unlverslty

 

 

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

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 mumps-p.14

 

PHOTOPHYSICS OF COORDINATION COMPOUNDS INTERCALATED INTO
LAYERED METAL PHOSPHATES

By

Eric Alan Saari

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

DOCTOR OF PHILOSOPHY

Department of Chemistry

1998

ABSTRACT
Photophysics of Coordination Compounds Intercalated into Layered Metal Phosphates
By

Eric Alan Saari

Organization of the solid state offers control of reactions through the use of
size/shape selectivity, guest/host interactions, and reactant pair proximity. Luminescence
techniques provide a sensitive probe of microheterogeneous environments. This work
will focus on three complementary probes used to explore the galleries of layered metal
phosphates (LMPs) toward the end of assembling a photocatalytic system.

Clever use of the intracrystalline environment of host structures provides an
alternative to the solution phase in the construction of a photocatalytic system. LMPs
(specifically or-M(HPO4)2-H20 where M=Zr,Ti,Sn) were the layered species studied as
target hosts for such a system. Recent work in the Nocera group has identified a class of
compounds that can undergo multielectron redox chemistry upon irradiation with one
photon. Such compounds could become a multielectron photoactive center (MPC),
whose structure offers, potential for simplified photocatalytic schemes. A Layered
Integrated Photochemical System (LIPS) is proposed and three probes examined toward
the end of combining the advantages of the solid state and MPC chemistry into an

integrated photocatalytic system.

The first probe, the ReOzpyf cation, will reveal the protic environment of the
galleries and a proposed intercalation mechanism. The second probe, Ru(bpy)32+,
suggests the redox behavior of the LMPs can be adjusted by varying the metal identity,
potentially enhancing control over a LIPS. The third, Mozdmhp4 (Hdmhp=2,4-dimethyl-
6-hydroxypyrimidine), is an MPC model compound. It has been incorporated using the
Lewis basic sites on its ligand to provide the intercalative driving force. The effect of
protonation and differing incorporation environments on the excited state of will be
explored.

All complexes are used to ascertain the role of guest/host electron transfer and
host perturbations on the excited state behavior; generally the complex was not
deactivated by the LMP suggesting a LIPS is feasible.

Although progress has been made toward the construction of a LIPS, fundamental
limitations of the component materials will likely preclude success without additional
advances and changes in the given scheme. These limitations include lack of MPC
chemistry generality, no successful demonstration of MPC regeneration, the small
quantity of available energy in one green photon, and future difficulties with

reactant/product mass transport through the system.

Copyright by
ERIC ALAN SAARI
1998

To the glory of God

ACKNOWLEDGEMENTS

Certain parts of my work were unmistakably built upon the labors of students who
preceded me in the Nocera group. For example, Mark Newsharn and Claudia Turré both
built and improved upon the emission instrument before it was my responsibility; Zoe
initially taught me how to use it. Yeung-gyo Shin got me started in the layered metal
phosphates and certain pieces of instrumentation. Colleen, Claudia, Jeff, and Janice
provided inspiration and encouragement for more than one prank. Other more
contemporaneous students, such as Ann, Sara, and Carolyn, worked on various aspects of
the M—‘LM chemistry, which had bearing on the motivation for my entire project. Mark,
Wanda, Dimitris, and Al Barney provided other scientific and personal interaction.
Carolyn Hsu acquired FAB-MS data for me on occasion. J.P. Kirby was a friend and
also assisted in acquiring some NMR data. Jeff Zaleski and James Roberts graciously
collected about twenty picosecond emission lifetime measurements for the molybdenum
work. Dan Engebretson has been a consistent and knowledgeable colleague with respect
to the practical aspects of laser spectroscopy. I came to appreciate his practical bent as he
assisted me in getting started with photon counting and we worked together on planning
the Nocera renovations at MIT.

The support staff here in the Chemistry Department were important in many
aspects of my career here. Ron Haas, Scott Sanderson, and Tom Clarke (retired) of the

MSU Chemistry Instrument Services patiently taught me practical aspects of electronics

vi

and instrumental repair and diagnosis while putting forth frequently valiant efforts to
keep the Nocera group hardware running in top form. Their absence would have made
the instrument maintenance and upgrades considerably tougher, more stressful, and
probably not as successful as they turned out to be. Martin Rabb (retired) and John
Rugis, Chemistry Dept. Electronics Designers designed a few little black boxes to fit
specific needs for the emission instrument and other laboratory equipment, saving our
group considerable effort and Prof. Nocera money. Dr. Tom Carter, Dr. Tom Atkinson,
and Paul Reed provided valuable advice and assistance with respect to my responsibilities
in upgrading and troubleshooting the Nocera group computers. Although I infrequently
interacted with the glassblowers Keki Mistry, Manfred Langer, and Scott Bankroff, I
regularly used the fruits of their labors were in the form of Schlenk lines, Ar manifolds,
and sample holders. Their excellent workmanship has all too often been taken for
granted, as we easily forget how good we have it here. Torn Geissenger has been a
consistent model of courteous and efficient service in the subbasement stocker which
he has run very well, despite Deluge ‘96, funding limitations, and bureaucratic
hindrances. The secretarial support staff here in the Department have helped smooth the
inevitable bureaucratic hassles associated with research and being a graduate student.
Beth Thomas superbly ran the main office with a consistently cheerful attitude. Linda
Krause, Dan’s best secretary during his stay here, kept us informed as to Dan's
whereabouts as well as anyone could.

Financial assistance from NSF, CFMR, MRSEC, MSU, NIH, and Air Force
directly and indirectly provided funds for salaries, supplies, and major equipment used in

my research.

vii

Prof. Dan Nocera, who will read this section with more interest than the rest of
the dissertation, provided me the opportunity to develop scientific, chemical, and
technical skills. He provided the initial direction and parameters for the project, as well
as funding, support, patience, and guidance. His high standards in the areas of
organization, communication, and presentation of scientific material have always been a
challenge (not usually enjoyed at the time) and I appreciated this opportunity to improve
my abilities while working in his labs.

My parents have always encouraged the development of my mind and character.
Only in adulthood have I begun to understand and appreciate their example, sacrifices,
love and encouragement.

My deepest gratitude to my wife, Ruth, for her patience and love and respect
during our courtship and our first months of marriage during which the bulk of this
dissertation was written. She is truly a wonderful lady, sent from God.

To all these people and organizations, I offer a grateful "Thank you."

Lastly, my indebtedness to Jesus Christ is far beyond anything I could express.

viii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ xv
LIST OF FIGURES .......................................................................................................... xvi
LIST OF ABBREVIATIONS .......................................................................................... xix
CHAPTER 1 ........................................................................................................................ 1
INTRODUCTION ............................................................................................................... 1
Multielectron Photoactive Center (MPC) Design ........................................................... 4
Combining the Advantages of Solid State and MPC Chemistry: a Layered Integrated
Photochemical System .................................................................................................... 6
Layered Metal Phosphates (LMP’s) as LIPS Host Structures ......................................... 9
Photochemical Probes of LIPS ...................................................................................... 13
Thesis Outline ............................................................................................................... 14
CHAPTER 2 ...................................................................................................................... l8
EXPERIMENTAL METHODS ........................................................................................ 18
Instrumental Methods .................................................................................................... 18
Powder X-ray Diffraction .......................................................................................... 18
Elemental Analysis by XPS ...................................................................................... 19
Solution UV-Vis-N IR ............................................................................................... 20
Diffuse Reflectance UV—Vis-NIR ............................................................................. 21
Steady state emission experiment: sample preparation ............................................. 22
Time resolved emission: nanosecond lifetime apparatus .......................................... 22

ix

Time resolved emission: picosecond lifetime apparatus ........................................... 22

Emission Instrument Update ..................................................................................... 23
Steady State Emission Instrument, Previous Configuration ................................. 23
Steady State Emission Instrument, Current Configuration ................................... 23

Description of current hardware configuration ................................................. 23
Experiments Possible on the Instrument ........................................................... 26
Description of current computer and electronic components ............................ 26
Photon counting detection system and data collection options ......................... 28
Description of Current Software setup and organization .................................. 31
Correction of Data for Instrument Response- Hardware and Software ............ 33
Software Utilities ............................................................................................... 35
Layered Metal Phosphate Hosts .................................................................................... 36

Synthesis of Layered Metal Phosphates .................................................................... 36
Starting Materials .................................................................................................. 36
a-Zr(HPO4)2°H20 (ZrP) ....................................................................................... 36
a-Ti(HPO4)2°H20 (TiP) ....................................................................................... 37
a-Sn(HPO4)2°H20 (SnP) ...................................................................................... 37

Swelling of Layered Metal Phosphates ..................................................................... 4O
Swelling Strategies ................................................................................................ 40
Preparation of Amine Swelled Layered Metal Phosphates ................................... 41

Tert-butylamine swelled ZrP (TBA-ZrP) .......................................................... 43
Tert-octylamine swelled ZrP (TOA-ZrP) .......................................................... 43
Tert-butylamine swelled TiP (TBA-TiP) .......................................................... 44
Tert-octylarnine swelled TiP (TOA-TiP) .......................................................... 44
Tert-butylamine swelled SnP (TBA-SnP) ......................................................... 44

X

Tert-octylamine swelled SnP (TOA-SnP) ......................................................... 45

Structure and Analysis of the Amine Swelled Layered Metal Phosphates ........... 45
EtOH-ZrP .............................................................................................................. 48
Preparation of Complexes ............................................................................................. 50
Preparation of trans-dioxotetrakis(pyridine)rhenium(V) iodide [ReOzpy4I] .......... 50
Preparation of M(bpy)32+ Complexes ........................................................................ 51
Preparation of M02(dmhp)4 ....................................................................................... 52
CHAPTER 3 ...................................................................................................................... 53
OReO INCORPORATION ............................................................................................... 53
Introduction ................................................................................................................... 53
Review Of ReOzpy4+ Literature .................................................................................... 53
Trans-dioxotetrakis(pyridine)rhenium(V) iodide [ReOzpy4+] Results .......................... 62
Solvent Dependency Experiments ................................................................................ 64
Incorporation, Experimental ...................................................................................... 64
Results and Characterization of the Incorporated Layers- Solvent Dependency ...... 64
ESCA / Loading / Powder XRD ........................................................................... 64
Emission, Lifetime ................................................................................................ 65
Discussion- Solvent Dependency .............................................................................. 69
Loading .................................................................................................................. 69
XRD ...................................................................................................................... 7O
Inferred Orientation(s) ........................................................................................... 71
Emission ................................................................................................................ 71
Emissive Lifetime ................................................................................................. 72
LMP Comparison Experiments ..................................................................................... 73
Incorporation, Experimental ...................................................................................... 73

xi

Results and Characterization of the Incorporated Layers- LMP Comparisons ......... 73

Loading / Powder XRD ......................................................................................... 73
Emission / Lifetime / Absorbance ......................................................................... 74
Discussion— LMP Comparisons ................................................................................. 74
Loading .................................................................................................................. 74
XRD ...................................................................................................................... 78
Inferred Orientation(s) ........................................................................................... 78
Emission ................................................................................................................ 78
Emissive Lifetime ................................................................................................. 78
Conclusions ................................................................................................................... 79
Possible Diffusion / Insertion Mechanism ................................................................ 79
Presence Of Protons .................................................................................................. 79
Amine Packing .......................................................................................................... 80
Electron and Energy Transfer .................................................................................... 80
CHAPTER 4 ...................................................................................................................... 81
Ru(pr)32+ INCORPORATION ...................................................................................... 81
Introduction ................................................................................................................... 8 1
Review of the Ru(bpy)32+ Probe .................................................................................... 82
Synthesis and Incorporation of Layers .......................................................................... 87
Results ........................................................................................................................... 87
Discussion ..................................................................................................................... 91
XRD .......................................................................................................................... 91
Diffuse Reflectance ................................................................................................... 93
Emission .................................................................................................................... 93
Lifetime ..................................................................................................................... 93

xii

Conclusions ................................................................................................................... 94

CHAPTER 5 ...................................................................................................................... 95
M02 INCORPORATION .................................................................................................. 95
Introduction ................................................................................................................... 95
Quadruply Bonded Dimetallic Complex as an MPC ................................................ 95
Three methods to getting MAM inside the Host ..................................................... 100
Selection of M02(dmhp)4 as MAM Intercalant ........................................................ 101
Characterization Of M02(dmhp)4 ................................................................................ 102
Acid Titration Studies ............................................................................................. 108
Experimental- Titrations ...................................................................................... 108
Results ................................................................................................................. 109
Emission, Absorption ...................................................................................... 109
Lifetimes .......................................................................................................... 113
Discussion of Titrations ...................................................................................... 113
Incorporation of Moz(dmhp)4 into LMP Layers ......................................................... 115
Procedures ............................................................................................................... 116
Incorporation with EtOH-swelled ZrP ................................................................ 116
Incorporation with amine swelled LMPs ........................................................... 116
Characterization ....................................................................................................... 1 l6
Powder XRD ....................................................................................................... 1 l6
Absorption, Emission, ps Lifetime ...................................................................... 117
Discussion ............................................................................................................... 122
Problems .............................................................................................................. 122
Inferred Orientation(s) ......................................................................................... 122
Absorption, Emission .......................................................................................... 123

xiii

pr n

Lifetime ............................................................................................................... 124

Conclusions ................................................................................................................. 125
Significance of Results with Respect to LIPS Construction ....................................... 126
LIST OF REFERENCES ................................................................................................ 130

xiv

Table 1.
Table 2.
Table 3.
Table 4.
Table 5.
Table 6.
Table 7.

LIST OF TABLES

Loading data for solvent dependency experiments. ........................................... 66

Fitted emission lifetimes and intensities for solvent dependency experiments. 66

Summary of lifetime data for Ru(bpy)32+ and Zn(bpy)32+ in TBA-LMPS. ........ 92
Mozdmhp4 Titration I data (HCl in MeOH solution). ...................................... 114
Mozdmhp4 Titration II data (triflic acid in MeOH solution). ........................... 114
Mozdmhp4 Titration HI data (triflic acid in CH3CN solution). ........................ 115
Absorption and Emission Data Summary for Moz(dmhp)4. ............................ 119

XV

LIST OF FIGURES

Figure 1. Typical Excited State Sequence of Events .......................................................... 2
Figure 2. Mallouk system- Solid State Charge Separated Assemblies. ............................. 5
Figure 3. Generic photoinitiated multielectron redox scheme using an MPC. .................. 5
Figure 4. Operational Schematic of a Layered Integrated Photochemical System. ........... 7
Figure 5. Top view of one layer of a-ZI‘(HPO4)2' H20; apical O atoms lie above or
below P atoms and are not shown. ............................................................................ 11
Figure 6. Oblique view of a-Zr(HPO4)2-HZO. ................................................................. 12
Figure 7. Emission instrument schematic- key components. ........................................... 25
Figure 8. XRD Spectrum of a-Zr(HPO4)2°HzO. .............................................................. 38
Figure 9. Diffuse reflectance spectra Of pure ZrP, TiP, and SnP. .................................... 39
Figure 10. Swelling schematic of ZrP with TOA and EtOH ............................................ 42
Figure 11. XRD Spectra of TBA-ZrP and TOA-ZrP. ...................................................... 46
Figure 12. Top view of LMP showing association of amine with apical oxygens. ......... 49

Figure 13. The synthesis and structure of trans-dioxotetrakispyridinerheniumW) iodide.

................................................................................................................................... 55
Figure 14. Diagram of Re(V) d orbitals in D4}. symmetry in ReOzpy4+. .......................... 56
Figure 15. Diagram of the ReOzpy4+ electronic states. .................................................... 57
Figure 16. Room temperature and 77K emission spectra Of crystalline ReOzpy4I. ......... 59
Figure 17. Excitation Spectrum of ReOzpy4I. .................................................................. 60
Figure 18. Diffuse reflectance UV-Vis-NIR spectrum of crystalline ReOzpy4I. ............. 63

xvi

Figure 19. Typical powder XRD pattern of ReOzpy4+ in TOA-ZrP. ............................... 67
Figure 20. ReOzpy4I in TOA-ZrP emission spectra with varying intercalation solvents. 68
Figure 21. XRD of ReOzpy4+ in TBA-ZrP and TBA-TiP. ............................................... 75
Figure 22. Emission spectra of ReOzpy4I in TBA-ZrP and TBA-TiP. ............................ 76
Figure 23. Diffuse Reflectance UV-Vis-NIR of ReO;;_py4+ in TBA-ZrP and TBA-TiP... 77
Figure 24. Coupling of LMP metal atoms to MPC. ......................................................... 83
Figure 25. Structure Of 4d6 Ru(bpy)3,2+ ............................................................................. 84
Figure 26. Schematic of Ru(bpy)32+ dilution with Zn(bpy)32+. ........................................ 86
Figure 27. Powder XRD spectra of Ru(bpy)32+ and Zn(bpy)32+ in TBA swelled LMPs. 88
Figure 28. Diffuse reflectance spectra of Ru(bpy)32+ and Zn(bpy)32+ in TBA-LMPs ...... 89
Figure 29. Emission spectra of Ru(bpy)32+ in TBA swelled LMPs. ................................ 90
Figure 30. MO Diagram for MAM Species. ................................................................... 96
Figure 31. Photoinitiated multielectron redox scheme in MAM system. ....................... 98
Figure 32. Three Schemes to MPC Incorporation ............................................................ 99
Figure 33. Synthesis of Moz(dmhp)4. ............................................................................. 103
Figure 34. Absorption spectra of Mozdmhp4 in MeOH. ................................................ 104
Figure 35. Diffuse reflectance spectra of crystalline Mozdmhp4. .................................. 105
Figure 36. Emission Spectra of Mozdmhp4 as crystals and in MeOH solution. ............ 106
Figure 37. Excitation spectrum of crystalline Mozdmhp4. ............................................. 107
Figure 38. Emission Spectra of Mozdmhp4 in MeOH solution titrated with HCl. ......... 110
Figure 39. Emission Spectra of Mozdmhp4 in MeOH solution titrated with triflic acid.l
................................................................................................................................. 11
Figure 40. Emission Spectra of Mozdmhp4 in CH3CN solution titrated with triflic acid.
................................................................................................................................. 1 12
Figure 4]. XRD spectrum of Mozdmhp4 in EtOH-ZrP. ................................................. 118

xvii

Figure 42. Diffuse reflectance spectra of Mozdmhp4 in EtOH-ZrP. .............................. 118

Figure 43. Diffuse Reflectance of Mozdmhp4 in TBA-LMP. ........................................ 119
Figure 44. Emission Spectra Of Mozdmhp4 in EtOH-ZrP. ............................................. 120
Figure 45. Emission Spectra of Mozdmhp4 in amine-swelled LMPs. ............................ 121

xviii

BA
BA-ZrP
bpy

CLO

CT

DAQ
DIO
dppm
dmhp
ESCA (XPS)
EtOH-ZrP
EtOH

eV

GPIB
GUI

H-bond

LIST OF ABBREVIATIONS

Analog to Digital Conversion

n-butylamine (CH3CH2CH2CH2NH2)

ZrP swelled with n-butylamine
2,2'-bipyridine

Complex Layered Oxide

Charge Transfer

Data AcQuisition

Digital Input/Output
bis(diphenylphosphino)methane
2,4-dimethyl-6-hydroxypyrimidine (deprotonated)
X-ray Photoelectron Spectroscopy
Zr(HPO4)2-xEtOH (Ethanol swelled ZrP)
ethanol

electron Volt

General Purpose Interface Bus, IEEE-488
Graphical User Interface

Hydrogen bond

xix

Hdmhp
Hxl
Hxl-ZrP
Hmhp

I vs. t
IEEE-488
LabVIEW

Max

’l-EM

KMAx
LIPS
LMP

LMPR

MSU

0.5Na-ZrP

2,4-dimethyl-6-hydroxypyrimidine

n-hexylamine (CH3CH2CH2CH2CH2CH2NH2)
ZrP swelled with n-hexylamine
6-methyl-2-hydroxypyridine

Intensity as a function of time

General Purpose Interface Bus, GPIB

Laboratory Virtual Instrument Environment Workbench
Excitation wavelength

Emission wavelength

Wavelength of maximum intensity

Layered Integrated Photochemical System

Layered Metal Phosphate

Layered Metal Phosphonate

Large Step Driver

Low power Schottky Transistor to Transistor Logic
quadruply bonded compound

methanol

Metal to Ligand Charge Transfer

Multielectron Photoactive Center

Michigan State University

Zr(NaPO4)(HPO4)'5H20

XX

NMR

NI

OLIS
OReO+
PMT

Py
ReOzpy4+

RT

Ruwpy)?”

S/N

SnP

TBA

TBA-TiP

TBA-SnP

TBA-ZrP

TiP

TOA

TOA-SnP

TOA-TiP

TOA-ZrP

triflic acid

nuclear magnetic resonance

National Instruments, Inc.

On-Line Instrument Systems, Inc.
dioxotetrakispyridinerhenium(V)+ (ReOzpy4+)
PhotOMultiplier Tube

pyridine

dioxotetrakis(pyridine)rhenium(V)+ (OReO+)
room temperature
tris(2,2’-bipyridine)ruthenium(II) cation
Signal to Noise ratio

Tin Phosphate (a-Sn(HPO4)2-H20)
Tert-butylamine (CH3)3CNH2
Tert—butylamine swelled TiP

Tert-butylamine swelled SnP

Tert-butylamine swelled ZrP

Titanium Phosphate (a-Ti(HPO4)2-H20)
Tert-octylamine (CH3)3CCH2C(CH3)2NH2
Tert-octylamine swelled SnP

Tert-octylamine swelled TiP

Tert-octylamine swelled ZrP

trifluoromethanesulfonic acid

xxi

TTL
UV-Vis-NIR
VI

XPS (ESCA)
XRD

21103133932+

ZrP

Transistor-Transistor Logic
UltraViOlet-Visible-Near InfraRed
Virtual Instrument

X-ray Photoelectron Spectroscopy
X-ray Diffraction
tris(2,2’-bipyridine)zinc(II) cation

Zirconium Phosphate (or-Zr(HPO4)2-HZO)

xxii

CHAPTER 1

INTRODUCTION

The goal of this thesis is to build an understanding of energy and electron
transfer1 in lamellar materials between a guest coordination molecule and sites in the
layered host. Such an integrated photochemical system serves to transform reactant(s)
into product(s) of higher enthalpy using light to supply the energy required to drive the
reaction. By definition the photocatalytic components in the system must be either
unchanged or regenerated in the process of the cycle, and thus their chemistry must be
integrated. The knowledge acquired fi'om this thesis should provide the underpinning for
the design and construction of new materials capable of promoting light induced catalytic
processes.

One or more electrons not residing in a ground (lowest energy) state molecular
orbital gives rise to a molecule in an electronic excited state. A photointegrated process
built around an excited state process is shown schematically in Figure 1. The steps
outlined in this figure, and related concerns, are a challenge to the design of
photocatalytic systems. The obvious first step is collecting or harnessing a UV or visible
photon; this is accomplished with a light harvesting complex (LHC). The LHC should
ideally absorb a relatively high proportion of the photons passing through or near it since

light energy not absorbed cannot be utilized. Common examples of LHC’s include

1

aromatic rings, conjugated organics and in photosynthesis, chlorophylls and carotenoids.
UV-visible absorption is of little value if the collected energy is dissipated into molecular
vibrations and rotations, and thus transduction (receiving energy in one form from a
source and retransmitting it to another in a possibly different form) is necessary if useful
chemistry is to be achieved. In nature, charge separation of an electron/hole pair is the
most common transduction step. The potential energy of the charge separated state must
be stored in some manner until it can be utilized for a useful chemical reaction; if not the
charges will recombine and the energy harnessed will be dissipated as heat or useless
work. For this reason, effective prevention of charge recombination of electron/hole
pairs in a myriad of donor/acceptor complexes has been an objective of a substantial

research effort in recent years.2

 

 

 

transduction chemical reactionl->Lregeneration I

Figure 1. Typical Excited State Sequence of Events.

Photosynthesis is of interest to those designing integrated photochemical systems
because it serves as a benchmark for photocatalysis. In photosynthesis, photon
absorption by the light harvesting complex promotes efficient transmembrane
electron/hole separation, which is ultimately manifested in the multielectron chemistry of
water oxidation at the Oxygen-Evolving Complex of Photosystem II and proton reduction
in Photosystem I.3

Numerous attempts to mimic photosynthesis have focussed on the charge
separation event and its storage at a remote catalytic site. An inherent problem with

implementing this approach in solution is that the donor and acceptor of different

molecules diffuse toward one another allowing charge recombination to readily occur.
Accordingly, the design of biomirnetic assemblies have relied on the synthesis of
intramolecular charge separating networks that propagate efficient electron/hole
separation on different time scales. The basic tenet of the approach involves the transport
of charge separated electrons and holes away from a LHC by covalently attaching
electron donors and electron acceptors to it to form artificial systems called ‘triads.’2""10

Charge separation in such systems occurs by the vectorial transport of an electron/hole
pair away from a light harvesting center (i.e. DP*A —) D+P'A -) D+PA‘ or DP*A —)
DP+A' —> D+PA'). Even though the charge separated states of triads can persist into the
microsecond range, the quantum yield for their production may be low, owing to simple
charge recombination of the electron/hole pair.ll Moreover, multielectron photocatalysis
based on one electron charge separation requires that the initial electron/hole pair be
stored in catalytic centers at the terminus Of the network, and the overall process must be
repeated to build up the necessary equivalents of holes and electrons needed for reaction.
Despite much work, such an integrated process has yet to be achieved.

The intracrystalline environment of host structures provide an alternative
architectural framework to assemble photochemical subunits for light-to-energy
conversion processes as covalent bonds are not always needed. The solid state
potentially offers control of spatial issues, size/shape selectivity, guest/host interactions,
and reactant pair proximity. For instance, when the electron transfer components of
triads are organized within solid state architectures, their complexity is greatly
reduced.'2’l3’I4 Mallouk et. al. have catenated a metalloporphyrinls or Ru(bpy)32+

photosensitizer“5 to a methylviologen charge transfer/Pt catalyst self-assembly organized

within the channels of Zeolite-L as shown in Figure 2. The dimensions Of the channels
confine the large photosensitizer to the exterior of the zeolite surface while juxtaposed to
the long, cylindrical zeolytic channels occupied by viologen acceptors. Excitation
prompts electron transfer from the photosensitizer to the viologen, which in turn relays
the electron to metallic Pt sites in the interior of the zeolite where proton reduction to
hydrogen occurs. In this manner, a triad (sacrificial electron donor, photosensitizer,
electron transfer carrier) is spatially organized at the zeolite/solution interface without the

need for covalent attachment of a donor-photoreceptor-acceptor assembly.

Multielectron Photoactive Center (MPC) Design

The structural complexity demanded for efficient charge separation and storage
may be relaxed if more than one electron can be moved from a discrete excited state.
Moreover, multielectron reaction from an electronically excited core obviates the
necessity for charge storage coupled to catalytic redox centers. Though such
multielectron reactions are rare, work in the Nocera labsl7 indicate that multielectron
reactions can occur with properly designed excited state molecules.

The operational schematic of a multielectron photoactive center is shown in
Figure 3. One photon excites the MPC, which then performs a multielectron redox
reaction at a substrate. A charge storage and relay system are unnecessary because the
photoreaction is accomplished in a discrete step. Several factors must be considered in
designing or selecting a photocatalyst for an MPC. Fundamentally a reasonable

absorption cross section (8) at the Aexcimfion is needed to generate a discrete excited state

from which more than one electron can be moved. The ligand set and metal must be

hv

PC

 

 

 

 

 

e e e
Multi
Zeolite L

 

Figure 2. Mallouk system- Solid State Charge Separated Assemblies.

hv possible hv

MPC ——> MPC OQC' MPC“

 

Regeneration

Figure 3. Generic photoinitiated multielectron redox scheme using an MPC.

 

complementary so that only oxidation states more than one electron apart are favored. In
addition, a redox substrate must bind weakly enough so that can be removed at a later
time, but bind with an adequate association constant for efficient reaction. After
excitation and reaction, the photoreactant must be regenerated from a redox active
species. The model on which this thesis is based is to have the redox equivalents for

regeneration provided from remote sites within a layered material.

Combining the Advantages of Solid State and MPC Chemistry: at Layered
Integrated Photochemical System

The goal of the research discussed in this dissertation is to combine the
advantages of the solid state (organization and control) and MPC chemistry (simplified
multielectron redox) into an integrated photocatalytic system. This strategy is reflected
in our general conceptual plan for constructing Layered Integrated Photochemical
Systems (LIPS) as shown in Figure 4. A LIPS consists of incorporating an MPC into a
layered inorganic host species containing redox active metal atoms. Ultraviolet or visible
light activates an MPC, which is incorporated in the layered host species. An
appropriately shaped and sized molecule diffusing toward the MPC is available for
multielectron reaction at the MPC. The now oxidized MPC must be returned to its
original ground state for catalysis to be effected. In principle, regeneration of the
oxidized photoactive core can be achieved with auxiliary redox-active metal centers in a
solid state support, by the transfer of electrons or holes from redox active metal atoms in
the layer to the MPC. The surface of the lamellar host would catalyze the reaction of a

redox complementary species to close the catalytic cycle.

 

 

 

Figure 4. Operational Schematic of a Layered Integrated Photochemical System.

In this approach, multielectron activation occurs in a discrete step at the MPC and
thus does not rely on charge propagation via one-electron intermediates Of a charge
separating network; intricate structural engineering of photoactive solids is no longer
needed to prevent charge recombination. Thus utilization of the multielectron
photoreactions made possible by the MPC in a LIPS eliminates the need for elaborate
charge transport and storage assemblies near a light harvesting center.

Though the structural complexity required for a LIPS is substantially reduced, the
design of a LIPS is far from trivial. Because the catalytic cycle in Figure 4 presumes that
the MPC can selectively effect the multielectron activation Of a substrate and then be
regenerated, interactions of the MPC and its host are central to the design of a LIPS
strategy. Physical, chemical, electronic and redox properties require compatibility
between supporting host structures and MPCs. If a LIPS is to be prepared, an appropriate
layered host must be carefully selected after considering several criteria. The material
must be readily characterized; crystallinity and reproducibility in its preparation are
expected. The interlayer environment must not be hostile to any guest molecules that
will be present, particularly the MPC. Intercalation activity must permit incorporation of
appropriate guests, and the mass transport of reactants and products. The host must be
stable in the presence of and transparent to light that excites the MPC. Finally, a most
vital attribute is the ease of modification of the host to permit facile tunability of redox

processes to ensure LIPS function.

Layered Metal Phosphates (LMP's) as LIPS Host Structures

Early studies at MSU concentrated on complex layered oxides (CLOs) as layered
supports,18 but we soon realized that CLOs are not readily adapted to Figure 4 because
the metals composing the layers of CLO structures are generally redox-inactive,'9 the
basicity of CLOs is not compatible with some coordination compounds forming instead
hydroxo and 0x0 complexes,20 and the chemical properties of CLOs, for the most part,
are too irreproducible for reliable photochemical study. For these reasons, new layered
materials were sought for LIPS design; layered metal phosphates appeared to satisfy the
necessary criteria for LIPS design.”25 These include facile and reproducible preparation
in high yield and owing to their crystallinity nature they are well characterized. LMPs
display high stability toward temperature, irradiation, and most organic and inorganic
environments.24 These metal phosphates are very stable in neutral and acidic
environments and are compatible with the acidic properties of many transition metal
cores. Moreover, the redox properties of the layers may be tuned because a variety of
metals (some redox-active) including Zr, Ti, Sn, Hf,26 V,27 and Nb28 may be incorporated
into the host structure. The intercalation behavior as well as the layer’s electronic and
redox properties are functions of the metal’s identity. A number of intercalation
strategies may be employed that permit versatile intercalation chemistry. Additional
modification by replacement of the P atoms with As or Sb is sometimes possible and
results in subtle differences in steric effects and chemistry. Furthermore, the solid state
coordination environment can be modified by using layered metal phosphonatesa’zg’30

which have an organic group (of many possible types) covalently bound to the layer and

projecting into the gallery (interlayer) space. Thus in principle an appropriate

10

coordination environment essential for multielectron reactivity of the MPC cores can be
designed into the solid.

Of the various classes of LMPs, we have selected the Ot-Zr(HPO4)2-H20 structure

type for study. Zirconium phosphate (ZrP) consists of layers 7.6 A apart formed from the
octahedral coordination Of Zr4+ by oxygens of the tripodal bases of six different
phosphate tetrahedra in a D3,: arrangement“ as shown in Figure 5. Figure 6 illustrates
how Zr atoms reside coplanar to each other in parallel sheets. Three of the phosphate
oxygens are bound to different metal centers leaving an apical oxygen bearing an
exchangeable proton directed into the gallery region, giving rise to an ion exchange
capacity of 2W per Zr atom. Individual layers are not hydrogen bonded to each other
and the hydration water is hydrogen bonded in an intralayer fashion.32 The apical
oxygen atom gives rise to the ion exchange intercalation chemistry where incorporation
of guests occurs by BrOnsted acid/base chemistry or simple cation exchange. Many
classes of compounds have been intercalated, including alkali metal ions,”222324
amines,”222324 alcohols,32 coordination compounds,33 and appropriately functionalized
organometallic complexes.33 Compounds that will hydrogen bond to the exchange sites
are also energetically favored as guest molecules.

The layered phosphates present a diverse redox chemistry. The Zr(IV) atoms
composing the layers are do and redox inactive; ZrP is transparent to visible light and is
white in color. Thus the layers are photochemically inert inasmuch as electron and
energy transfer between the layer and an excited guest cannot occur. (It has been

demonstrated that energy transfer can occur between two species through multiple ZrP

layers”) Consequently this material provides a reference host structure for excited state

O

l

/P\

O 0\ g ’0
\\\\\\ III”:
0’00. .\\\‘\O ‘
0 P.‘ i
6

O\§’O

.\\\\\M/”’u

OIIIIIIH'U

)K

O

l

/P\

Olll’h

CHIN-"'0

O\§’O

o\‘\\\ Ill”.

O\é’0

o\\\\\\o.

Olllllu-‘D

\‘\\\\ Ila". . “\\\ [/Ii’o.
‘ 0,0” “($0 0
(‘3 .. l?" (K
P (E) P

O. \\\\\\O. \O.

.\\\\\M/”’:.
.‘\\\\ ‘

Ol/I”' O/I”" O””,.

0

Figure 5. Top view of one layer of Ot-Zr(HPO4)2- H20; apical O atoms lie above or
below P atoms and are not shown.

 

 

12

 

 

 

O Zr, Ti, or Sn

ew Of a-Zr(HPO4)2-H20.

Figure 6. Oblique vi

13

studies Of LMP intercalates. Single and multielectron redox active species are achieved
with respective titanium (TiP) and tin (SnP) substitution. These materials are
isostructural to ZrP, with similar intercalation behavior, but titanium has common
oxidation states of IV and III ((10 and (1') whereas tin has common oxidation states Of IV
and II (dlo and szdlo). This one—electron and two-electron redox activity provides a

rational design for effecting the type of catalytic chemistry described in Figure 4.

Photochemical Probes of LIPS

The study of MPC / LMP interactions requires judiciously designed excited state
probe molecules. This thesis presents three types of coordination complexes to probe
different facets of guest/host relationships critical to the design Of materials described by
Figure 4. Trans-dioxotetrakispyridinerhenium(V) (OReO+) is a cationic d2 complex. The
oxo's are proton acceptors and upon excitation the longest lived emissive excited state Of
the complex is readily quenched. This property makes it a useful probe for the
characterization of the protic environment of galleries in which information on spatial
location and orientation is deduced.'8 The ubiquitous Ru(bpy)32+ complex is a useful
probe of the coupling of redox equivalents to the interlayer environments Of MPCs. It is
extremely susceptible to electron transfer quenching and hence an ideal probe of one
electron and two electron equivalent layered materials. In the studies described herein, it
is an ideal probe of the one electron and two electron chemistry of the TiP and SnP
respectively. The final probe of Figure 4 is the quadruply bonded (Mil-M) compounds.
These species are multielectron photoreagents. The excited state properties important to

multielectron chemistry are retained. The quadruply bonded dimolybdenum complex

l4

M02(dmhp)4 serves as a model MPC for LIPS structures. Proton presence, ligand basic
site interactions, and photophysics of MAM complexes are ideally modeled in this

system.

Thesis Outline

A myriad of details must be addressed for the successful construction of a LIPS.
This thesis is concerned with the first steps of LIPS design: incorporation of MPC's into
layered phosphates and the parameters that affect the rate of energy and electron transfer
between a guest and host.

To this end, construction of LIPS based on layered metal phosphates poses many
interesting issues in regard to the design of photocatalytic solids described by Figure 4.
The following basic questions must be resolved for the goals depicted by Figure 4 to be
realized, and consequently are the focus of my research program. 1) Can the MPC core
be intercalated while retaining its excited state properties? 2) Even if the excited state
properties Of the intercalated MPC core are retained, will it be deactivated by energy
transfer with the layer before chemistry at the MPC can be achieved? 3) What about
electron transfer deactivation? 4) Assuming that the mixed valence MPC core is
achieved, will the photochemistry that has been seen in solution be retained? 5) Can the
MPC be regenerated to its original state using electrons from (11 metal atoms in the layer
at the end of the proposed photocatalytic cycle? This issue has two aspects: is it possible
to transfer electrons from the host to the guest, and if so, will the mere addition of
electrons be sufficient to induce any necessary molecular changes to regenerate the

original form?

15

The first issue is obviously vital, but also nontrivial as it involves intercalation of
an intact luminescent core into a potentially unfiiendly environment. Conceptually, a
general pre-swelling strategy (vide infra) and a hydrogen bonding strategy have been
used by several researchers.35 Strategies and procedures applied for the preparation Of
LIPS structures will be discussed in Chapter 2. Chapters 3-5 each demonstrate a complex
can be successfully intercalated while retaining its excited state properties.

With respect to the second issue, energy transfer in the solid state between a LMP
host and an intercalated guest has been addressed to a limited extent by Ellis and co-
workers.”37 They discovered that energy could readily and efficiently transfer from a
layered uranyl phosphate (generic formula UOzPO4) to guest molecules. For example,
the host’s luminescent U022+ center, an integral component of the layer, was found to
transfer energy to an intercalated Eu3+ ion. In contrast, it has been demonstrated34 that
energy transfer can occur between two species through multiple ZrP layers. The
differentiating factor here is that energy transfer requires overlapping absorption bands,
so care taken in selecting both hosts and guests minimizes this problem. The selected
LMPs (ZrP, TiP, SnP) are nonabsorbing in the near UV and visible regions of the
spectrum, so energy transfer will not be an issue. Experimental demonstration of this
point is given with the ReOzpy4+ (OReO+) probe in Chapter 3. The third, fourth, and fifth
issues will be variously addressed in Chapters 3, 4, and 5.

Chapter 2 is dedicated to the general experimental details of this work and the
preparation and swelling of layered metal phosphates. All instrumentation and common
procedures are described; particular attention is given to the detailed documentation of

the emission instrument upgrade. LMP characterization and amine and ethanol

l6

incorporation chemistry will also be explored in Chapter 2. Preparations of the three
probes used in Chapters 3-5 are given.

Chapter 3 focuses on the first probe, the ReOzpy4+ complex. Studies in Chapter 3
will hint at intercalation mechanisms while demonstrating that subtleties in incorporation
methodologies can strongly affect the excited state behavior of a complex in the LMP
galleries. The gallery protons were also determined to be located away from the OReO+
core. More importantly, it was discovered that the excited state of incorporated ReOzpy4+
was not deactivated by means other than those found in aprotic solution; hence issues 2
and 3 (deactivation by energy or electron transfer) are not necessarily a problem.

Chapter 4 focuses on the second probe, the Ru(bpy)32+ complex. A partial answer
on the fifth (regeneration) issue is that some electron transfer can occur between the guest
complex and LMP. This chapter focuses on delineating the electronic differences in the
LMP series with respect to redox quenching by using Ru(bpy)3,2+ as a probe. The data
indicate that the one electron redox probe couples better to a particular LMP with similar
one electron redox properties than to the others. This apparent behavior, if generally
extensible, could permit tuning of the LMP to appropriately complement the desired
regenerative redox behavior of the MPC and/or hinder redox quenching of the MPC
excited state. Regeneration of the MPC has not been demonstrated in this work, nor by
others in solution phase.

Chapter 5 focuses on the third probe, M02(dmhp)4, which serves as a model MPC
inside the LMP structure. Regarding the fourth issue of retention of photoactivity,
scattered reports on the photochemistry of a few simple coordination compounds

intercalated in layered metal phosphates existrm’39 Upon irradiation, reactions typical of

17

the solution phase are often observed. Although these reactions are relatively
uninteresting photosubstitutions, they suggest that the photochemical scheme of Figure 4
would be possible in lamellar solids. In this work, the excited state behavior Of
Moz(dmhp)4 in the presence of titrated acid, its incorporation into an LMP, and its initial
photochemistry are explored. As the M02(dmhp)4 complex is neutral, it does not readily
incorporate into the anionic LMP unless it is protonated, making it cationic like the
complexes examined in Chapters 3 and 4. This intuitive discovery has significant bearing
on issue 1. The final section in Chapter 5 delineates the strengths and weaknesses of
attempting to construct a LIPS in the general manner described herein. Although
significant progress has been made, other fundamental limitations of the component
materials will likely preclude successful construction of a LIPS without additional

advances and some changes in the given scheme.

CHAPTER 2

EXPERIMENTAL METHODS

Instrumental Methods
Powder X-ray Diffraction

Powder X-ray diffraction patterns of samples were recorded on a Rotaflex system
from Rigaku with Bragg-Brentano geometry. The Cu-Ka line was obtained from a
rotating Cu anode (45 kV, typically 100 mA) and directed toward the sample using a l/6°
divergence slit and a 1/6° receiving slit. The diffracted X-ray beam was further refined
by a curved graphite single crystal monochromator (1.05° scatter slit and 1/6°
monochromator receiving slit), which was set for detection of the secondary X-ray
diffraction line. Typically, spectra were collected in the 2-theta range Of 1° or 2° to 30°
or 40°. The resulting data were recorded and processed using the manufacturer-provided
software DMAXB on a microVAX computing system (earlier data) or on an IBM
compatible PC.

The samples, normally dried powders, were mounted by one of two methods.
Earlier samples were mounted by pressing the powder onto a piece of double-sided tape
attached to a 1" x 2" slide glass. Later samples were mounted in glass sample holders

available from Rigaku. Rigaku glass sample holders consisted of 35 mm x 49 mm plates

of glass 2 mm thick with a rectangular recess in the surface to hold the sample. The

18

l9

recess in the slides used was 16 mm x 20 mm in area and ground to a depth of 0.2 mm.

For all powdered samples, compounds were placed on the double sided tape or in the
recess of the sample holder; they were spread and packed tightly into place by sliding a
clean glass microscope slide over a small mound of sample.
Elemental Analysis by XPS

The analyses of elements by X-ray Photoelectron Spectroscopy (XPS) in layered
materials were performed with a Perkin Elmer PHI 4500 ESCA System, which is housed
in the Composite Materials and Structure Center at Michigan State University. The
X-ray source was monochromated Al or Mg Kc lines (A1(Ka) = 1486.6 eV, 600 W /
15 kV; Mg(Ka) = 1253.6 eV, 400 W / 15 kV). Samples to be analyzed were pressed onto
double-sided tape adhered on one side to a stage, which was placed at an oblique angle to
the incident X-ray beam (65° for the Al source and 45° for the Mg source) under ultra-
high vacuum (10'8 - 10'9 torr). The sample stage was positioned so that the signal
intensity for oxygen (binding energy of 531 eV) was maximized, as oxygen is the most
abundant element in layered metal phosphates. Elements were identified by survey scans
from binding energies of 1200 eV to 0 eV. The strongest photoelectron line fi'om each
element was used to Obtain relative atomic concentrations by integrating the area of the
peak and multiplying by an appropriate response factor. All data manipulation and
output were accomplished on an Apollo Workstation using XPS ESCA software provided

by Perkin Elmer.

20

Solution UV- Vis-NIR

Solution electronic absorption spectra were collected by using a Cary 2300 or a
Cary 17D UV-Vis-NIR spectrophotometer, typically with 1 cm path length UV grade
fused quartz cuvettes. The spectrophotometers were typically purged with house
nitrogen.

The Nocera group Cary 17D, a superb UV-Vis—NIR spectrophotometer, was built
in the 1970's to the highest optical standards. Its original serial number is 232 and is
equipped with a wavelength range Of 185 to 2650 nm. The Cary 17’s sample chamber
was modified by the machine shop in the Chemistry Department at MSU. A 30 cm tall
chamber extension was constructed to provide room for tall cells of the long-neck and
high vacuum varieties. The original sliding sample holders, previously an occasional
source of positional irreproducibility, were fixed into place with provision for removing
them when needed to accommodate the cryostat head. Another removable modification
was an adjustable rod and cradle that supported the ground glass adapter on the hi gh-
vacuum cells which permits firm and reproducible positioning of the cell in the chamber.
The original Cary jacketed cuvette holders with their capability for temperature
regulation were retained.

The Cary 17D was modernized and computerized in June 1994 by On-Line
Instrument Systems, Inc. (OLIS) of Bogart, GA. Troublesome analog electronics and
certain mechanical components were replaced with dramatically simplified modern
systems, while the premium quality Optics and Optical drive mechanisms were retained.
(The Optical bench of the Cary has the potential for many decades of reliable service.)

Excitation is provided by an air-cooled 650 W tungsten halogen [type DWY] Vis—NIR

21

lamp, and a D2 UV lamp and power supply supplied by OLIS (replacing the original
hand-blown D2 UV larnp). Hardware and software for instrument control and data
acquisition were provided by OLIS. The computer is a Gateway 2000 running MS-DOS
6.21 with a 80486DX2 66 MHz microprocessor, 16 MB RAM, a 540 MB hard disk, a
15” SVGA monitor, and an Intel Ethernet card. Spectra collected before modemization
were recorded on chart paper, while more recent spectra were recorded digitally.

The Nocera group Cary 2300 with 185 to 3152 nm wavelength range was
equipped with a IEEE-488 interface board and a DS-15 data station running on an 8085
microprocessor with software provided by Varian. The Cary 2300 equipment included a
five position turret sample holder and a 30 cm tall chamber extension. Spectral
measurements reported in this work were recorded on chart paper.

Diffuse Reflectance U V- Vis-NIR

Diffuse reflectance UV-Vis-NIR spectra of solid state samples were collected
using a Shirnadzu UV-3101PC double beam scanning spectrophotometer in the
Kanatzidis laboratories at MSU.40 The instrument is equipped with an ISR—3100 UV-
Vis-NIR Integrating Sphere Attachment and controlled by an IBM compatible computer
using Microsoft Windows operating system with instrument software provided by
Shimadzu. Each sample was prepared by spreading a thin layer of powder onto the
surface of a BaSO4 cake (Wako Pure Industries), which was pressed into the standard
sample holder provided by the manufacturer. Spectra were collected at room temperature
in reflectance mode, where the 100% reflectance baseline was referenced to a holder with

the white background of pure BaSO4 powder.

22

Steady state emission experiment: sample preparation
Powdered samples were pressed into a standard glass XRD holder or placed in an
evacuated 4 mm diameter quartz tube equipped with a high vacuum stopcock and
standard taper ground glass adapter. The XRD holder was oriented 15-20° from normal
to the excitation beam. Spectra at 77 K were Obtained by immersing the quartz tube into
an unsilvered quartz dewar filled with liquid nitrogen. Solution emission spectra were
typically obtained from standard 1 cm square UV-grade fused quartz cuvettes. Most
emission spectra reported in this work were recorded on the Nocera group emission
instrument before it was modified and updated.
Time resolved emission: nanosecond lifetime apparatus
All time resolved measurements Of emission intensity of the rhenium and
ruthenium systems were made in the LASER Laboratory in the Chemistry Department at
MSU. Emission lifetime measurements were obtained on an instrument constructed at
Michigan State University.41 The measurement apparatus connected to the PMT detector
was a CAMAC crate housing a transient digitizer in the early rhenium experiments, and a
storage oscilloscope for the other measurements. The powder samples were placed in an
evacuated quartz tube as indicated above for both RT and 77K measurements; data were
typically collected for 1000 pulses where Lx=532 nm. The data were fit to a
monoexponential or biexponential decay using Kinfito or Kaleidagraphc software.
Time resolved emission: picosecond lifetime apparatus
All time resolved measurements of emission intensity of the molybdenum systems
were made in the LASER Laboratory in the Chemistry Department at MSU. Emission

lifetime measurements were obtained on an instrument constructed at Michigan State

23

University.42'43 The powder samples were placed in an evacuated quartz tube as indicated
above for both RT and 77 K measurements. Liquid samples were placed in a 1 cm
pathlength long-neck cuvette fitted with a septum and bubbled with Ar gas before
measurements were made. The data were fitted using Kaleidagraph‘D software to a
monoexponential or biexponential decay.
Emission Instrument Update

Steady State Emission Instrument, Previous Configuration

The Nocera group emission instrument has been described previously.44 A
complete description of the current instrument is given below. The modifications and
updates made minimal changes on the optical bench itself; the improvements were
focused on the user interface, data handling, task automation, component control, error
prevention, and detection system. From a spectroscopic viewpoint, the most important
change was the detection system upgrade from lock-in to photon counting.

Steady State Emission Instrument, Current Configuration

Description of current hardware configuration

Emission and excitation spectra were collected on an instrument constructed at
Michigan State University, schematically illustrated in Figure 7. The excitation beam of
the spectrometer originates from either a 200 W Xe/Hg or 150 W Xe arc lamp, mounted
in a Spex 1909 lamp housing (f/4), powered by a Oriel 8500 are lamp power supply, and
is focused onto the entrance slit of a Spex 1680B double monochromator (0.22 m, f/4).
The wavelength-selected excitation light, chopped with a DigiRad model C980 Optical
Chopper, collimated by an Oriel f/4 lens, passes through a 2” Oriel interference filter and

is focused onto the sample with a f/l .5 lens. A quartz bearnsplitter between the chopper

24

and the sample reflects some of the excitation light into a powered photodiode connected
to the input of a PARC EG&G 5209 lock-in amplifier, phase matched to the reference
signal of the chopper. Light emitted from the sample is collected at 90° to the excitation
beam with a f/ 1.5 collimating lens and then is focused by a second lens (f/8) onto the
entrance slit of a Spex 0.5 m 1870B monochromator (f/8). All four lenses are Oriel 2”
UV grade fused silica mounted in standard holders. One or more 2” Schott type colored
glass long pass cutoff filters (series KV, WG, GG, CO, or RG) are employed to reduce
the quantity Of excitation light scattered into the detection system. The sample holder
Offers translational adjustment in three axes, and may be rotated in the axis normal to the
excitation and detection optics’ axes. Standard solution cuvettes, front face powder
holders, and 4 mm quartz tubes can be readily accommodated. The dispersed emission is
detected by a Hamamatsu R943-02 (or occasionally R316—02) head-on photomultiplier
tube (PMT), which is cooled with dry ice to -55°C in a Products for Research (model
TE241RF) housing and powered by a Bertan Associates model 215 high voltage power
supply set to negative polarity. The signal from the photomultiplier tube is passed into a
Stanford Research Systems SR4OO Two Charmel Gated Photon Counter; no
preamplification of the signal is performed unless the R316-02 is used. The lock-in
output is collected in digital format as the observed excitation intensity; output from the
photon counter is collected as the Observed uncorrected emission intensity. Instrumental
control, digital data collection, and manipulation are performed through a Graphical User
Interface (GUI) using software written at Michigan State University and run on a
Pentium® computer equipped with a National Instruments® (N I) Lab-PC+ (DAQ) board

and an AT-GPIB/TNT(PnP) board.

25

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Arc Lamp
CPU

F DAQ GPIB

~ board board
Excitation Tl'L l g
Monochromator E
F reference fre uenc g
[ Chopper i ----------------------- 9 ------ YR g

* s r gag

* : Lock-In :3 -

Photodiodei : I Amplifieri lllllllllg CD
’ .MMMM Emission l E TIL E
Eample Monochromator] 5 g
, i E
PMTI- ---------- j Photon Countgr-l

 

 

Figure 7. Emission instrument schematic- key components.

 

26

Experiments Possible on the Instrument

The three experiments currently possible on the emission instrument are emission,
excitation, and intensity as a function Of time (I vs. t). Emission is the most common
experiment in this work and will be assumed unless specified otherwise. In an emission
experiment, the sample is excited at a constant wavelength and the emitted light intensity
is measured as a function of wavelength. In an excitation experiment, the sample is
excited at a varying wavelength and the emitted light intensity (at some specific
wavelength) is measured as a function of excitation wavelength. In the I vs. t experiment,
the sample is excited at a constant wavelength and the emitted light intensity is also
observed at some different constant wavelength as a function of elapsed time. Due to
instrumental limitations, the I vs. t experiment is not a lifetime measurement; but an
observation Of the steady state emission intensity. It is useful for long averaging at one
wavelength and also for a series of intensity comparison experiments as one can have all
the data in one file.

Excitation spectra may be obtained with minimal hardware changes from the
standard emission configuration. The excitation lamp must be the 150 W Xe arc lamp,
chosen for its relatively smooth output intensity as a function of wavelength.

Description of current computer and electronic components

The computer for the emission instrument is a Gateway 2000® running under the
Windows 95® operating system. The computer is equipped with a 75 MHz Pentium®
microprocessor, PCI Bus, 32 MB physical RAM, a Vivatron® 1776 color monitor, a
National Instruments (NI) Lab-PC+ (DAQ) board, and a N1 AT-GPIB/I'NT(PnP) (IEEE

488.2) board.

27

The GPIB (General Purpose Interface Bus) and DAQ (Data AcQuisition) cards
are the computer’s hardware interface tO the instrument. The GPIB board is cabled
directly to a PARC EG&G 5209 Lock-In amplifier and a Stanford Research Systems
Model SR4OO two channel gated photon counter, for setting parameters and reading data
and parameters by using standard digital communication protocols. The DAQ electrical
connections are provided in an aluminum enclosure assembled by the MSU Chemistry
Instrument Services and John Rugis, Chemistry Department Electronics Designer.

An NI CB-SO connecting block in the enclosure is cabled directly to the Lab-PC+
board. External BNC and 25-pin D connectors connect to cables from both
monochromators and the analog output of the EG&G 5209 lock-in. Inside the enclosure,
the lock-in’s output is connected to ADC (Analog to Digital Conversion) channel 1,
which is set to the O-IOV range and protected by zener diodes against potentials above
20 V. The lock-in’s output is proportional to the signal from the photodiode monitoring
the excitation light’s intensity. The LSD (Large Step Driver) Spex monochromators are
connected to a LSTTL digital input port and a digital output port on the Lab-PC+. The
8-bit digital input port is connected to the four monochromator limit switches and is
regularly monitored while the monochromators are moving; any of these lines going to
logical high indicates that one of the monochromators has reached a mechanical limit.
The remaining four unused inputs on the input port are pulled low (to ground) through
resistors to eliminate false TTL high readings. Four lines on the Lab-PC+’s digital output
port are connected to the monochromators’ TTL direction and step inputs. Toggling a
step input high and low once will move a monochromator one step, which corresponds to

a specific change in center wavelength. Control of direction of the monochromator’s

28

movement is determined by whether a high or low signal is sent to the direction input.
The digital output port’s lines are connected through intermediate CMOS inverters
(74AC14) as the Lab-PC+ board cannot sink enough current to ensure reliable
monochromator operation. (TTL and LSTTL are slightly different electrical logic
standards.) A 100 kHz TTL timebase, powered by the Lab-PC+, is also in the enclosure
for historical reasons but is now unused.

During data collection, the Lab-PC+ D10 (Digital Input / Output ports) moves the
proper monochromator a set number Of steps corresponding to a fixed change in
wavelength, the Lab-PC+ ADC reads channel 1, and the SR400 reports counts for each
datum. The raw data are temporarily stored in RAM in an array, and are periodically
plotted on-screen to display them in approximately real time. The data interval is user
selectable in five resolutions: 1 nm, 5 A, 2 A, 1 A, and 0.2 A. Routine spectra are taken
at 1 nm intervals by default. The NI AT-GPIB/TNT(PnP) board is used to configure the
EG&G 5209 lock-in, read the chopper frequency, configure the SR400, and read the
number of counts.

The original Spex 1673C Mini-Drives are no longer used for monochromator
control and are incompatible with the LabVIEW GUI as they can only be controlled
through manual keypad inputs. As no modifications were made to the monochromators
themselves, these Mini-Drives are still fully functional and could manually Operate the
monochromators by simply reconnecting the cables.

Photon counting detection system and data collection options
Photon counting is an excellent light detection method with high SfN ratio for

situations with low intensity and reasonably high duty cycle. A photon counting

29

detection system is comprised of the photomultiplier tube, its high voltage power supply,
and a photon counter. A PMT is a sensitive, high gain, low noise detector that is capable
of detecting single photons. Upon striking the photocathode, the photon causes the
ejection of an electron by the photoelectric effect. A series of dynodes multiplies this
small electrical pulse by many orders of magnitude, so that it can easily be detected. The
assumption is made that most output pulses caused by photons will be of a minimum
voltage that can be roughly calculated from known detection system values, and that most
noise pulses will be of lesser voltage. These electrical pulses are input to a photon
counter, which uses a discriminator to measure the voltage of each pulse; those above a
certain threshold are counted while those below it are not.

An additional method used to increase the S/N ratio is to chop the excitation beam
and gate the photon counter. If the emissive lifetime Of the sample is fast relative to the
duration Of excitation, the emission will occur (roughly) in time only during excitation;
by extension no emission should occur between excitation intervals. Gating the photon
counter means counting only during specified time windows. In practice, the chopper
blocks the beam 50% Of the time at a frequency of 337 Hz or 13 Hz. For
synchronization, the reference signal of the chopper is connected to the photon counter’s
trigger input. Gates A and B on the SR400 Two Channel Gated Photon Counter are
programmed with time parameters that correspond to full illumination and no
illumination of the sample. These gate delays and widths were determined empirically
using a fast storage oscilloscope. Pulses exceeding the discriminator threshold during the
full illumination time window (Gate A) are summed as “light counts,” and the pulses

during the no illumination time window are summed as “dark counts.” In most cases

30

subtracting dark counts from light counts yields net counts, the number for relative
observed emission intensity. (Mode 3 in the next paragraph describes the exception.)

Three different data collection modes are available on the instrument. In the first
mode the monochromator is moved before photon counting begins for each datum,
yielding the best resolution. In the second (default) mode photon counting and mono
movement occur simultaneously, resulting in faster data collection with some loss of
resolution. Both modes subtract dark counts from light counts to arrive at net counts
reported. The third mode is for samples whose lifetimes are long (greater than 1 ms).
Photon counting and mono movement occur simultaneously as in the second mode, but
the dark counts are not subtracted from the signal. For long lifetimes (relative to
chopping speed), a substantial quantity of real emission is also occurring during the dark
cycle. If the dark counts were subtracted as noise, then the Observed signal could be quite
weak.

The second value (sometimes) measured at each datum is the observed excitation
intensity, or photodiode channel. A quartz beamsplitter between the chopper and the
sample reflects some of the excitation light into a powered photodiode connected to the
input of a PARC EG&G 5209 lock-in amplifier, phase matched to the chopper’s
reference signal. The lock-in’s output is passed as a voltage to the Lab-PC+ for ADC
(Analog to Digital Conversion). This system for measuring excitation intensity was
implemented for historical reasons and remains adequately fast and accurate. However
there is now no hardware barrier to programming the software to have the lock-in
perform the ADC and report the reading digitally. The manipulation of the Observed data

is described below.

31

Description of Current Software setup and organization

National Instrument’s NI-DAQ® driver software was used for both boards in
conjunction with software written at Michigan State University in the LabVIEW® (G)
language. NI’s LabVIEW for Windows v3.1.1. is a graphical programming environment
which is well suited for instrument interfacing and control. A Virtual Instrument (VI),
analogous to a program, was created for the GUI. Dozens of sub-VI’s, which are
analogous to subroutines, were also created to perform specific functions. The computer
controls both monochromators, the 5209 lock-in, and the SR400 photon counter. The
GUI controls the instrument and collects, corrects, scales, displays, and saves the data
and experimental information. On-line documentation was provided for all controls and
indicators with which users interact. Substantial additional documentation was provided
in the information section of the VI’s as well as in the coding and commenting of each
Vl’s diagram, i.e. the source code.

The V1 containing the main front panel is the main VI that runs the instrument. It
is structured as two loops that run concurrently; one is primarily used to update
instrument status variables and indicators and iterates at several Hz. The other loop
contains all the control code for the instrument and iterates once per data acquisition; it
consists of a nine frame sequence structure. The first frame is mainly for configuration of
the instrument. The user inputs the current monochromator positions and the installed
PMT type if their identity is unknown to the software. The lock-in is tested if powered
on and queried for the chopper frequency; the frequency and PMT identity are later used
in setting the correct photon counter configuration constants. Most controls are activated

here. In the second frame the user selects parameters for the next scan and may interact

32

with and control various parts of the instrument. From this frame modules can be called
that will manually slew or recalibrate the monochromators, change parameters on the
lock-in, or Observe a live display of the photon counter’s data. The third frame tests
values selected for the monochromator parameters to ensure that they are reasonable and
will not damage the instrument. Most controls are grayed out at this point. The fourth
frame positions the monochromators and builds file header strings. The monos are
moved to the positions required to initiate scanning so that backlash is eliminated and so
that they never overlap in wavelength. The user is prompted to select the filename for the
data file. A 30-1ine file header is composed at this point from both user input and
computer deduced information. The fifth frame sends configuration information to the
photon counter and displays the Go! dialog button; the intent is to provide an opportunity
to make last minute checks or adjustments, such as refilling a nitrogen dewar or turning
off room lights, before initiating the scan. The sixth frame is where the data are
collected, corrected, written to file, and displayed on screen. The appropriate data
collection module appears on screen during the scan, showing current data as they are
collected. The seventh frame updates some indicators and locks the data file to prevent
accidental overwrite. The eighth frame slews the monochromator back to its starting
position. The ninth frame handles issues related to shutting down the software, such as
returning the photon counter to local control and setting/resetting some internal variables.

The ASCII text format data file is locked to reduce the chance of an accidental
overwrite. Lines are delimited in DOS/Windows format (CR then LF) for maximum
compatibility with other platforms and with software for data manipulation and display.

A 30 line header is devoted to recording experimental information, including user,

33

time/date stamp, instrumental parameters, and comments; header information is derived
with a minimum of user input or duplication to reduce errors. The data begins on line 31

l or sec), uncorrected PMT channel data,

in four tab-delimited columns: nm (or cm'
corrected/scaled final data, and photodiode channel voltage (0-10 V).

The monochromator stepper motor drivers were completely written from scratch,
with timing, backlash compensation, and error checking features as already described and
fully documented in the software. The LabVIEW VI/driver for the EG&G 5209 was
almost completely rewritten from the example driver supplied by NI; their driver could
not be used as several necessary functions were lacking and it was written for a similar
lock-in (EG&G 5210) which has a different GPIB command set. The rewritten driver
reads the current settings, can reset all parameters to a default, and implements most of
the available functions, many of which are not needed by the user. The low level drivers
for the SR400 were generally slightly modified from those supplied by N1; the high level
drivers are entirely new or substantially rewritten.

Correction of Data for Instrument Response- Hardware and Software

Correction of data for instrument response is necessary as certain instrument
components perform differently as wavelength is varied, leading to substantial distortions
in the observed spectra. The variations are in both the excitation and emission sections of
the instrument. The only case in which these variances are irrelevant is for an I vs. t
experiment as neither Max nor ABM are varied.

With respect to excitation, the lamp output, monochromator throughput, and

photodiode response all vary as AEX changes. For an emission spectrum, these variances

are not important as the excitation wavelength kgx is held constant during the scan;

34

problems arise only when comparison of absolute intensities is attempted between scans
of differing 15x. For excitation spectra, the intensity of the excitation varies with
wavelength as the excitation monochromator is scanned, requiring a correction to the
observed emission intensity. An accurate profile of lamp intensity incident on the sample
is unavailable as it is difficult to directly measure the profile and as the lamp output
changes with power and age. In practice a reasonable approximation is made by
measuring part of the beam’s intensity with a photodiode and correcting the Observed
intensity by the published photodiode response as a function of Max-

With respect to the emission side, the monochromator throughput and PMT
response both vary with ABM. For an excitation spectrum, these variances are not
important as the emission wavelength 3.5M is held constant during the scan; problems
arise only when comparison Of absolute intensities is attempted between scans of
differing 3.5M. The variances are important when an emission spectrum is collected. In
practice a reasonable approximation is made by comparing the Observed spectrum of a
NBS-traceable lamp to the accepted spectrum of that lamp and calculating a ABM
dependent set of correction factors as a function of 3.5M. Specifically, the spectral
responses of the monochromator and PMT were calibrated with a standard of spectral
irradiance (45 W tungsten halogen lamp, serial L-374) from Optronic Laboratories; the
lamp spectrum over the wavelength range of interest was recorded. Estimated irradiance
values at wavelengths between standard listed points were approximated by fitting the
data to a polynomial. The raw Observed lamp spectrum was recorded under standard data
collection conditions by using a precision constant current source of 6.500 i 0.002 A at

6.7 V designed and constructed by Martin Rabb at MSU, and fitted to a function of

35

several Gaussian bands. The data for the correction function are generated by software in
situ by dividing point by point (as a function of wavelength) the accepted irradiance value
function by the function of the observed signal of the standard lamp. All collected
spectra are thus corrected as a function of wavelength for PMT and instrument response
by using point by point multiplication of the sample’s observed data and a correction data
function. A correction function was generated for each PMT type. Data collection may
begin and end anywhere within the valid range of the correction function. The valid
wavelength range of the correction function is determined by the instrument response,
PMT response, and range of the standard lamp used. An alternate standard lamp for the
200-400 nm wavelength range for performing a similar (UV only) calibration uses a
UV-40 standard lamp and a Model 45D Precision Constant Current Source of
500 i 0.1 mA at 10 V, both provided by Optronic Laboratories, Inc.
Software Utilities

Several utilities were written for common instrument tasks. A simple print utility
can recall spectra and print screen dumps with header information for notebook purposes.
A multiple print utility will display and print multiple spectra. A utility for moving
monochromators is intended for when the detection system is Off line; i.e. a limited
version of the full software package that has no code for the detection system or data

collection / manipulation.

36

Layered Metal Phosphate Hosts
Synthesis of Layered Metal Phosphates

Starting Materials

The starting materials were Obtained commercially and normally used as received
without further purification. All water was from the departmental distilled source.

a-Zr(HPO4)2°I~I20 (ZrP)

Numerous methods of ZIP preparation have been reported.”45 70 g of ZrCl4 were
dissolved with 300 mL of distilled water in a 1000 mL round bottom flask. 2 g of ZrP
seed crystals, prepared in a reaction like this one without seed crystals, were added.
400 mL of 85% H3PO4 (diluted to 600 mL) was added to the vigorously stirring solution;
an extremely viscous slurry immediately formed. The slurry was refluxed for two weeks
then vacuum filtered with a large fine-porosity fiit, washed with water, and air dried
under aspirator vacuum. The powder was twice resuspended, filtered, washed with water
and air dried; the second time it was also washed well, first with ethanol then diethyl
ether. The white powder was air dried and stored in ordinary containers. The fine fluffy
white powder was highly crystalline, yielding a standard X-ray diffraction
pattemzpzzz‘i’zfm’46 for Zr(HPO4)2 with a layer spacing of 7.6 A as shown in Figure 8. The
XRD data of other LMP's were similar. Yield was almost quantitative, with losses
mostly due to handling and spillage during the purification process. The diffuse
reflectance UV-Vis-NIR spectrum is shown in Figure 9. Note there is essentially no

absorbance except in the shorter UV.

37

a-Ti(HPO4)2-H2O (TiP)

Titanium phosphate was prepared as described47 by using anhydrous TiCl4 as the
Ti source. The product was a fine white crystalline powder with a layer spacing of 7.6 A.
The diffuse reflectance UV-Vis-NIR spectrum is shown in Figure 9. Note the absorbance
edge is substantially to the red of ZrP and SnP, although still well in the UV.

a-Sn(I-IPO4)2-H2O (SnP)

Tin phosphate was prepared similarly to the literature method.48 192 mL
concentrated HN03 and 543 mL 85% H3PO4 were diluted to l L in a 2 L beaker. 93.5 g
SnCl4-5H2O were added to the stirring solution over 5 minutes; no precipitation was
observed and heating was begun. Initially red/yellow gas was evolved and precipitation
started after 1-2 hours as a milky suspension. Heating continued at a slow reflux for
eight days with constant stirring; water was added to maintain the l L volume. The
sample was vacuum filtered by using a 350 mL fine hit; the filtrate came through slowly
and was clear. The crystals were stirred and washed with 300 mL water in portions and
then air dried. The powder was resuspended and washed as above twice more; the XRD
pattern improved with the resuspensions. The product was a very fine white crystalline
powder with a layer spacing of 7.8 A. The diffuse reflectance UV-Vis-NIR spectrum is

shown in Figure 9; it is similar to that of ZrP.

38

 

ITII[IIIIIIIII[IIIIIIIIIIIIIIIIIIITIITI

d001=7.6 A

 

 

 

IlllillllillllillllilILJiJlllillIiiLlll

 

   

intensity
IIIIIITIIIlllllllllflllll[lllllllll[IIll

 

N.“

.‘ _. A AA
ILIIILLLIIIILIJIlllilllliilll

5 10 15 20 25 30 35 4O
2 theta

 

 

 

 

0

Figure 8. XRD Spectrum of a-Zr(HPO4)2-H2O.

39

 

120

 

100

80

 

 

 

 

 

 

 

8 i
C

E .1

8 60 :- a—Sn(HPO4)2 —

c a

9 ~— Ot-Ti(HPO4)2 a

t 40 l a-Zr(HPO4)2 L

20 '_‘ 1

t -

0 i: ' ‘ .1- - I I I I I 4 I I l I I I I L I I L I 4

200 400 600 800 1000 1200
nm

Figure 9. Diffuse reflectance spectra of pure ZrP, TiP, and SnP.

40

Swelling of Layered Metal Phosphates
Swelling Strategies

The Ot—Zr(HPO4)2-H2O layers are held together by Van der Waals forces and not

hydrogen bonding,46 as might be expected from the presence of protons on the apical
oxygens. Simple exposure of the ZrP to solvated guest molecules is frequently
inadequate to force intercalation since the ZrP requires modest quantities of energy to
overcome the Van der Waals forces and spread its layers to permit the entry of a solvated
ion.26 For this reason, sterically bulky molecules are Often difficult or impossible to
directly incorporate into the gallery regions. The intimate details of intercalation are not
known for the LMPs, but presumably the layers must be bent at some time if a sizeable
guest is to enter.

The flexibility of several layered systems has been estimated from the crystal
structure data using a computational procedure. The calculation was predicated upon the
assumption that the layer's rigidity is proportional to bond length changes induced during
the bending of a unit surface area in a specific way. The calculation did not account for
varying bond force constants. In comparing six classes of layered materials, it was noted
that the increasing calculated flexibility correlated to easier intercalation (as
experimentally observed) of molecular species. The authors also suggested that "the
flexibility of layers can exhibit anisotropy" resulting in the intercalation reaction zones
moving only in preferred directions in a crystal.32

One strategy that has been used to overcome intercalation difficulties is to

“pre-swell” the LMP with other guest molecules to partially overcome the attractions and

then introduce the desired guest molecules, as shown schematically in Figure 10.

41

Primary n-alkyl monoamines21 and primary n-alkyl alcohols32 have been reported as
successful swelling agents in the literature, forming bilayers in the galleries at or near
100% exchange capacity. Once inside, the alcohols are readily removed but the amines
are not. In the experiments reported here similar strategies have been used to exchange
bulky coordination compounds into the lamellar galleries.

ZrP swelled with n-butylamine (BA-ZrP) was used to attempt intercalation of
coordination compounds as early attempts by direct solvation techniques failed.
Although literature reports indicate BA-ZrP has been successfully used as a swelling
agent49 for the incorporation Of other coordination compounds,35 attempted exchange of
compounds of interest (vide infra) with BA-ZrP were largely unsuccessful, usually
resulting in total failure and/or Obvious decomposition of the complex. On the
presumption that longer chains would more readily permit intercalation, similar attempts
using ZrP swelled with n-hexylamine (Hxl-ZrP) were also unsuccessful. Reasonably
consistent success was achieved by using the bulky primary alkyl amines tert-butylamine
(TBA) and tert—octylamine (CH3)3CCH2C(CH3)2NH2 (TOA) as swelling agents. The
highly branched aliphatic chains of TBA and TOA are less efficiently packed than
unbranched n-alkyl chains; presumably easier intercalation Of large cationic complexes is
promoted by loosely packed chains as they are more readily moved aside and separated.

Preparation of Amine Swelled Layered Metal Phosphates

The swelling of layered metal phosphates by various primary alkyl amines is a
general reaction, with most primary n-alkylamines and many common secondary amines
reported in the literature.21 I have not located any references specifically describing the

intercalation Of tert-octylamine (CH3)3CCH2C(CH3)2NH2 (TOA) into any layered metal

Figi

42

LMP swelling with TOA Incorporation of complex

s H0 H“ /' 'j
1156‘? EC 0859333?" ficgpygetx ;
V??? .. Tiaeif

 

 

 

 

 

20.0A

 

 

 

L

 

TOA

Zr(HPO.,)2 H20 HC'

/ y
T L
l I I l
H0 H0 H.0 H.O

m
-l- l 3 1. 0.5 eq. NaOH

2. EtOH I HCIO4

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Tl/ "a H [/1 l I If
/I :I j if )C C%ordin tion "’° "° "° "’°
14.211 (Comp ex’ C%ordinpetion
K K ( K °"‘"
L ?
LMP stelling with EtOH Incorporation of complex

Figure 10. Swelling schematic Of ZrP with TOA and EtOH.

43

phosphate. Tert-butylammonium has been reported as an exfoliant Of ZrP, in essence to
"peel off" individual layers, suspending them in solution.50 Exfoliation is normally not a
significant issue in the following preparations, as quaternary ammonium salts were not
present and the swelling was frequently done quickly to minimize exfoliation. Less than
quantitative yields can be explained by loss during filtering and by exfoliation.

T ert-butylamine swelled ZrP (T BA-ZrP)

100 mL of an aqueous 0.2 M HCl and 0.4 M TOA solution was prepared in a
beaker. l g of Zr(HPO4)2 was added while stirring; the white ZrP flocculated within a
few minutes to form a sticky white paste that readily adhered to the sides of the beaker.
The paste was scraped out of the beaker after 20 to 30 minutes and placed into a fine-
porosity fiit and filtered (along with the solution and its suspended paste) under aspirator
vacuum, and repeatedly washed with acetone and water. The white product was air dried
and had an amine smell. Powder X-ray diffraction of the product demonstrated a layer
spacing 16.7 A (sometimes less) with weak higher order peaks. TBA-ZrP has been
reported26 and the reported d-spacing is not consistent with that observed here. These
inconsistencies can be attributed to variations in hydration, loading, and resultant packing
of the amines in the gallery.

T ert-octylamine swelled ZrP (TOA-ZrP)

100 mL of an aqueous 0.2 M HCl and 0.4 M TOA solution was prepared in a
beaker. 1 g of Zr(HPO4)2 was added while stirring; the white ZrP flocculated within a few
minutes to form a sticky white paste that readily adhered to the sides of the beaker. The
paste was scraped out of the beaker after 20 to 30 minutes and placed into a fme-porosity

fiit and filtered (along with the solution and its suspended paste) under aspirator vacuum,

44

and repeatedly washed with acetone and water. The white product was air dried and had
an amine smell. Powder X-ray diffraction indicated a layer spacing of 19.0 A with weak
higher order peaks.

T ert-butylamine swelled T I? ( T BA-T I?)

5.2 g TiP was placed in 25 mL neat TBA and refluxed overnight. The white
solution and suspension were placed into a fine-porosity frit and filtered under aspirator
vacuum. The white product was air dried and had an amine smell. Powder X-ray
diffraction revealed a layer spacing 15.2 A with weak higher order peaks.

T ert-octylamine swelled T iP (T OA-T iP)

Despite numerous and varied attempts, a pure TOA-TiP phase was never
prepared. 15 mL of neat TOA was placed in a test tube. 5 g of TiP was added and the
tube was stoppered, shaken and left to stand overnight. The contents were placed into a
fme-porosity frit and filtered under aspirator vacuum. The white product was air dried
for several days and had an amine smell. Powder X-ray diffraction: layer spacing 17.8 to
20.1 A with peaks corresponding to unswelled TiP.

T ert-butylamine swelled SnP (T BA-SnP)

15 mL of neat TOA was placed in a test tube. 6.2 g of SnP was added and the
tube was stoppered and shaken; the temperature noticeably increased. The tube was
periodically vented to release built up pressure. The contents were placed in a fine-
porosity fiit and filtered under aspirator vacuum; no washing was done. The white
product was air dried for two days and had an amine smell. Powder X-ray diffraction:

layer spacing 15.3 A with one weak higher order peak.

45

T ert-octylamine swelled SnP ( T 0A -SnP)

15 mL of neat TOA was placed in a test tube. 5 g of SnP was added and the tube
was stoppered, shaken and left to stand overnight. The contents were placed into a fine-
porosity fiit and filtered under aspirator vacuum. The white product was air dried for
several days and had an amine smell. Powder X-ray diffraction: layer spacing 19.4 A
with weak higher order peaks.

Structure and Analysis of the Amine Swelled Layered Metal Phosphates

Powder diffraction data shown in Figure 8 indicate that the original

a-Zr(HPO4)2-H2O is quite crystalline as evidenced by narrow diffraction maxima with a

minimal amount of baseline noise; the dam spacing of 7.6 A corresponds to the literature

2"222324’46 and the other expected maxima are present. In contrast, Figure 11 shows

value
that n-butylamine, n-hexylamine, TBA, and TOA-exchanged ZrP have evenly spaced
layers with substantial stacking disorder. The dam spacings of the BA-ZrP, Hxl-ZrP,
TBA-ZrP, and TOA-ZrP are 19.0 A, 24 A, 16.7 A, and 19.8 A respectively, confirming
that the amines form bilayers in the galleries. Previous studies summarized in the
literature indicate that n-alkylamines are incorporated at 100% of theoretical capacity
(one amine per exchangeable site) and that simple bond length and angle calculations
show the amines will tilt about 30° from normal to the lamellar plane to pack with
greatest efficiency.” Alternate explanations of amine monolayers and interdigitated
amine groups are not consistent with powder diffraction or loading data. The presence of

only d001, d002, and (1003 peaks in the spectra are evidence for a "turbostratic" structure for

the alkylamine swelled ZrP’s, which means that the layers are located parallel to each

46

TBA-ZrP

IIIITIITIIIITFTIIIIIfIIII'IIllllIrrrIIrr

 

16.7 A

 

 

TI

In

IIIIIIIIIIIIIJIIIIIIIIIIIILIIIIIIIIILII

5 10 15 20 25 30 35 40
2 theta

intensity
IITIITIIIIIITFW—IIIIIIIITill—IIIT

I

 

 

 

O

TOA-ZrP

TIII[IDIIIIIIIrITTrIITIITHTTTTMIIIIIIII

 

20A

intensity

IIIIFIIIIIIIIIIIIIIIIIIITII

 

 

IIIIIIIIIIIIlIlIIIIIlIIIiII

 

 

 

6”?
i

i

i

i

it

i

i

 

20
2 theta

C
01
...s
O
_L
01
N
01
00
O
00
01
A
0

Figure 11. XRD Spectra of TBA-ZrP and TOA-ZrP.

47

other a consistent distance apart, but with substantial stacking disorder such as
irregularities in relative layer translational position and/or from rotation of the layers
about an axis normal to the lamellar planes. This is consistent with the above amine
packing model.

Amine loadings of BA-ZrP and TOA-ZrP were estimated by elemental analysis
with the ESCA (XPS) technique. Mole atomic percentages of the products were obtained
using a computer algorithm, and the results used to calculate the absolute loadings of
amine (and later the intercalated coordination compounds) in the layered metal phosphate
complexes. The measured loadings were 70% for BA-ZrP and 55% for TOA-ZrP. The
well established literature value of BA-ZrP loading (100%)2"222324 has been determined
by elemental analysis, potentiometric titration, and other methods, so my figure indicates
a systematic error. As XPS analyzes near the surface, results may reflect the surface
loading instead of the bulk loading. An additional source of error may be evaporation of
amine into the high vacuum (minimum 1x10'7torr) environment of the XPS apparatus,
lowering the actual amine loading level on the surface and/or interior. As amine swelled
phosphates have some residual amine odor after washing and drying, offgassing under
high vacuum is likely. Drying of the amine under mechanical vacuum has resulted in
slightly contracted (0.1-0.5 A) d-spacings. These two observations support the above
loss of amine hypothesis. In light of the above, I estimate that the TOA-ZrP loading is
actually 60-80%.

The differences in intercalation capabilities between ZrP treated with n-alkyl
amines and TOA can reasonably be traced to the nature Of the gallery alkyl environment.

The n-alkyl “tails” of BA-ZrP and Hxl-ZrP are arranged in a well ordered and tightly

48

2‘ The highly branched alkyl groups in the gallery of

packed manner in their bilayers.
TOA-ZrP are sterically incapable of tight Van der Waals packing (like the straight chain
n-alkyl groups) due to their varying cross section along the chain. Also for steric reasons,
TOA is also incorporated into ZrP at only about 60-80% of ion exchange capacity
whereas the others are at about 100%. Since there is much more empty space in the
TOA-ZrP gallery and the hydrocarbon tails are presumably more easily moved aside than
in the other systems, incorporation of coordination compounds in these galleries is
relatively easy compared to the BA-ZrP and Hxl-ZrP galleries.

In general, interlayer coordination of the amine to the ZrP occurs via hydrogen
bonding. The ZrP apical O atoms are cation exchange sites 5.3 A apart in a trigonal
pattern which are occupied by protons in the hydrogen exchanged form.26 These oxygen
atoms are arranged such that a primary amine might H-bond to 3 0'3 using its 2 protons
and one of the layer's protons; each apical O atom could also H-bond to 3 separate
primary amines. Figure 12 shows a schematic top view of the complementary association
of the amine groups, apical oxygens, and protons. Such an arrangement would readily
satisfy charge balance and intercalation would formally be a net swelling, not an ion
exchange. As the pKa of ZrP is estimated to be 3 to 5, all three protons would be
formally "owned" by the more basic amine.

EtOH-ZrP

Ethanol swelled Zr(HPO4)2 (EtOH-ZrP) was prepared using literature
methods.32’5 ' 10.0 g of Zr(HPO4)2 added with stirring to 800 mL of l M NaCl formed a

white slurry, and a pH probe was allowed to equilibrate in the solution. The cation

exchange was conducted by a slow potentiometric titration of the solution with 1.0 M

49

ZFI
2.1

H H H H H H
\g. \9.‘ \g.
H H H
I I I
’ .‘ / .' ’ .’
H H H H H H
‘0'. \9, \O.
H H H
I I I
’ .' ’ .° , .’
H H H H H H
\9'. \g. \9'.
H H H
I I I
H H H H H H
\O. \9. \O.
H H H
I I I
H H H H H H
\O. \9. \O..

Nitrogen atoms are from primary amine
R group extends up out of page (not shown)

Oxygen atoms are apical 0 from phosphate group with proton
Remainder of LMP not shown

Figure 12. Top view of LMP showing association of amine with apical oxygens.

50

NaOH. The titration endpoint (based on the published titration curve) was chosen at
pH=5 when the intermediate Zr(NaPO4)(HPOa)-5H2O [0.5 Na-ZrP] is formed. The
pentahydration of this intermediate expands the layers considerably, permitting the
incorporation of alcohols. The 0.5 Na-ZrP was vacuum filtered with a fine 150 mL frit,
washed with distilled water, kept wet and stored over water to prevent dehydration.
Ethanolic perchloric acid solution was made by using 3 mL of 70% HC104 in 200 mL of
100% EtOH. Into the stirred solution was added some of the 0.5 Na-ZrP paste. After a
few minutes, the suspension of Zr(HPOa)2-xEtOH was vacuum filtered with a fine 60 mL
fiit and washed with 100% EtOH. EtOH-ZrP is a white paste that will readily lose EtOH
upon drying in air at room temperature; EtOH will not reenter and spread the layers upon
direct contact. EtOH-ZrP stored over 100% EtOH has been successfully stored for over a

year. The layer spacing is about 14 A.

Preparation of Complexes
Preparation of trans-dioxotetrakismyridine)rhenium(V) iodide [Re02py4l]

The preparation of ReO2py41 was reported in 1965.52 The procedure was used
through the 1980's, but has been superseded. Recent work53 provided a substantially
improved general synthesis (see
Figure 13) of trans-ReO2L4+ and trans-ReOzL’2+ (L=monodentate basic ligand such as
pyridine and L’=bidentate ligand such as ethylenediamine or bipyridine) by using the
intermediate ReO2(PPh3)2I. (Another efficient general method has been reported“)
ReO2(PPh3)21 was readily prepared in good yield from KReO4 via ReO(OEt)(PPh3)212

under slightly modified conditions.55 In a typical reaction, 50 g PPh; and 60 mL 48%

51

aqueous HI were dissolved in 350 mL of hot 100% ethanol and then 10.66 g KReO4 was
added while stirring. The PPh3 must be in excess and in high concentration or yields are
drastically reduced. Military green crystals of ReO(OEt)(PPh3)2I2 are formed during one
hour of reflux. The solution was vacuum filtered while hot on a fine frit and washed well
with hot ethanol; substantial quantities of PPh3 that may remain after reaction necessitate
repeated washing. The product is used without further purification to generate the
ReO2(PPh3)21 intermediate. Typically all of the ReO(OEt)(PPh3)2I2 was added while
stirring to a beaker containing 400 mL reagent grade acetone, 8 mL distilled water, and
2 mL 1 M NaOH. (The solution must be made basic or yields will be substantially
reduced.) After stirring overnight at RT, the opaque purple solution was filtered on a fine
fiit and yielded 31.36 g (97.9% from KReO4) of purple crystals of ReO2(PPh3)2I, which
were washed with dry acetone. All of the unpurified ReO2(PPh3)2I was refluxed in
300 mL neat pyridine56 with stirring and slowly cooled to RT producing orange crystals
that were filtered on a fine fiit and washed with benzene and hexanes and air dried. Yield
20.3 g (83% from KReO4). If the pyridine is saturated with product from a previous
preparation before addition of ReO2(PPh3)2I, the yield is increased further. ReO2py41 has
been reported as very soluble in acetone;52 actually it is sparingly soluble in dry acetone
but quite soluble in wet acetone. The ReO2py4+ molecular weight is 661.52 g/mol.
Preparation of M (bpy) 32 + Complexes

Ru(bpy)32+ and Zn(bpy)32+ were prepared similarly to the literature methods.”58
2.06g of ZnCl2 were placed in a 100 mL roundbottom flask with 6.88 g bipyridine
(stoichiometric ratio) and 50 mL of 100% ethanol. The mixture was refluxed for two

days, vacuum filtered, washed with methanol, and air dried. The crystalline powder

52

exhibited a faint pink color; it should be colorless as Zn2+ is d”). The color is attributed to
trace iron impurities in the source ZnCl2.
Preparation of Mo 2(dmhp)4

MO2(dmhp)4 was prepared using a modified literature59 method as shown in
Figure 33. MO(CO)6 and Hdmhp (2,4-dimethyl-6-hydroxypyrimidine) were Obtained
from Aldrich and used as received. Diglyme (bis-2-methoxyethyl ether) (EM Science or
Aldrich) and hexanes were freshly distilled over sodium and P205 respectively. 100 ml
diglyme, 10 ml hexanes, and 6.0 g each Of M0(CO)6 and Hdmhp (slight excess of ligand)
were refluxed under Ar ovemight, then cooled. If the mixture were left to reflux for
several days, its purity decreased due to formation of a silty dark precipitate, presumably
resulting from trace water contamination. The mixture was filtered on a medium fiit by
using aspirator vacuum, washed with dry hexanes, and briefly dried under mechanical
vacuum. Attempts at further bulk purification were unsuccessful. Sublimation with heat
under mechanical vacuum of an early impure sample resulted in collection of only free
ligand as determined by NMR, and chromatography attempts were hampered by wet
solvents, packings, scale up problems, and poor solubility.

Analysis by FAB-MS yielded a parent ion peak at 685 (calculated 684.4 g/mol)
with a pattern closely resembling the calculated one. The yellow crystalline powder is
reasonably stable with a diglyme odor and is modestly soluble in MeOH and EtOH and
almost insoluble in hexane. The reported crystal structure has one equivalent of diglyme
co-crystallized; integration by ’H NMR of diglyme and ligand methyl resonances showed

nearly a 1:1 ratio.

CHAPTER 3

OREO INCORPORATION

Introduction

In order to ascertain whether electronic excited properties of coordination
compounds are retained in phosphate layers, I began studies with the trans-
dioxotetrakis(pyridine)rhenium(V)+ cation (OReO+) as the guest molecule. ReO2py4+
was initially selected because it is a relatively stable cation that can exchange with the
proton in ZrP to form a stable intercalate, and it is a lumophore whose emissive
properties are highly dependent upon its environment. It can be used as a probe of
intercalation chemistry and positioning of the complex in the gallery. Two series of
experiments were conducted. The first series, Solvent Dependency, was intended to
ascertain the effect of intercalation conditions on the product formed and photophysics.
The second series, LMP Comparisons, was designed to determine differences of LMP

interaction with the excited states.

Review Of Re02py4+ Literature

The trans-ReOsz cation has a linear O=Re=O+ core (
Figure 13) and a ground state Re=O bond length of 1.76 A.60 The Re coordination
environment is of distorted D4}. symmetry, with the four pyridine ligands in the equatorial

plane twisted along Re—N bonds 10-20° from the OReO axis due to steric and packing

53

54

effects.61 In solution the pyridine ligands can exchange via a dissociative mechanism as
determined by NMR.62

The d-orbital manifold of d2 ReO2py4+ complexes is shown in Figure 14. The
coordinate axes are defined as having the z-axis along the O=Re=O bonds and the x and
y axes coincident with the Re-N bonds. In the diamagnetic 'Alg ground state, two
electrons reside in the b2g (dxy) orbital. The lowest energy ligand field electronic
transition arises from the promotion of a nonbonding b2g electron to an eg (dmdyz) Re-O
rt" orbital, resulting in 3Eg and lEg excited states. Population of the dtt“ (O) orbital
results in a stretching distortion of 0.07 A along the Re=O bond.63 This transition is
Laporte forbidden (as g<—>g), acquiring its intensity from vibronic and spin-orbit (for
triplet only) mechanisms. It is possible that intensity stealing from proximate charge
transfer (CT) states occurs also. The 3Eg state in Dah symmetry is sixfold degenerate and
decomposes, in the presence of strong spin-orbit coupling, to the states shown in Figure
15 in the D4,, double group.63 The near-UV transition is metal to ligand charge transfer
(MLCT) d(Re)—>1t*(py) on the basis of resonance Raman studies.64

Literature values for the UV-Vis spectrum of d2 ReO2py4+ are [Nnm (€/M“'cm")]
= 331 (19400); 445 (1200) in H2O solvent};5 The 331 nm MLCT band is dependent upon
solvent polarity, shifting to lower energy in less polar solvents.66 Very similar to d2
ReO2py4+, the UV-Vis spectrum of d1 ReO2py42+ also has two peaks [Nnm (e/M‘lcm")]
= 337 (28000); 440 (1400) 67 in methanol/water solVent.

Substitution of a 2,2’-bipyridine (bpy) for two pyridine ligands results in
substantially different chemistry and photophysics compared to ReO2py4+. The chemical

stability declines, thermodynamic destabilization of the Re(V) relative to both Re(VI) and

55

 

 

 

 

 

 

 

 

 

 

PPh3
HI, H20, PPh3 go“
KReO4 y EtO Re=
EtOH
I
PPh3
acetone, H2O
heat
_ O _ + PPh3
l'
I ‘0‘“ py heat 03"";
Re‘“ Re I
py / py pyridine
PY l o
— O - PPh3

Figure 13. The synthesis and structure of trans-dioxotetrakispyridinerhenium(V) iodide.

 

 

a1, (dB)
b1, (dffl
eg (dmdyz) #
bzg (dxy) +
1Eg
singlet

56

 

 

Ill

1

ground state

 

til

3
Es
triplet

Figure 14. Diagram Of Re(V) d orbitals in Dar, symmetry in ReO2py4+.

57

04h] double group

_ pyridine MLCT

U
I l .5
3'

 

 

 

1
1E9 E9 ( E9)
3 A19 . A29

59 E (3E )
319 . 329
highly
emissive
1 1
A19 A19 ( A19)
9 to g forbidden with spin-orbit
perturbations .
”3:33;?ch (calculated) D4“ (04")
mechanisms
also promoted by S l O
(for triplet)

Figure 15. Diagram of the ReO2py4+ electronic states.

intense a

modest e

buned
weak a
buned

ground state

58

Re(III) drastically alters the electrochemistry, UV-Vis absorptions shift 40 to 50 nm, and
phosphorescence completely disappears.68 These changes result from the distortion of the
O=Re=O unit from a trans arrangement Of 180° to a cis arrangement of 121°.69 The trans
form of d2 ReO2py4+ is thermodynamically preferred as the d electrons are confined in the
dxy orbital orthogonal to the rhenium oxo core whereas the cis form is destabilized by clxy
electrons interacting with filled ptt“ oxygen orbitals.70 These findings are significant for
the present study in that bpy substitution cannot be used for stabilization purposes and
substantial distortion of the ReO2py4+ complex in the host will be easily observed
spectroscopically. Unless stated otherwise, all references to ReO2py4+ (OReO+) in this
work concern the trans form.

Exciting the lowest energy electronic transition produces a 3B8 luminescent
excited state with typical (unquenched) lifetimes of 10-17 us in solution and 32 us in

crystals at room temperature.63 In nonaqueous solution, the emission spectrum is
featureless with a 7cm = 640 nm. In homogenous nonaqueous solution, only one
emissive state is Observed.66 The lemma is dependent on solvent polarity; more polar
microheterogeneous environments lead to emission at longer wavelengths, suggesting
that the complex is not relaxed in the surrounding environmentw’71 The characteristic
solid state emission spectrum is shown in Figure 16, and an excitation spectrum in
Figure 17. At room temperature, ReO2py4+ luminescence consists of a broad emission
profile between 550 nm and 800 nm with a 900 cm'1 vibrational progression arising from
the Re-O symmetric stretch. At 9 K, the structure is more pronounced, and an additional
210 cm'] vibronic progression is evident due to the symmetric Re-N stretching vibration.

. . . . + . .
These progressrons In the emrssron band are observed when ReO2py4 rs In an ordered

59

Emission Spectra of Crystalline ReO 2py4|
300 lllll

 

250

 

N
O
O

 

 

_|.
o
o

Relativelntensity
A
01 01
O O
TTITIITIIIIIITTIIIITTjTIIIIlI

 

 

 

rIll1IIII]IIIILLIIIILJIIJIIIIII

0-l"’lJilIllllLllllllLlllL '.'l"

 

550 600 650 700 750 800 850
nm

Figure 16. Room temperature and 77K emission spectra of crystalline ReO2py41.

60

 

500 T l I I I l T l I I l l l I I V Ifi l I T l T T I I T I f
364 nm
400

300

200

[ITIIIIIIIIIIIIIIIII

100 435 nm

IIIIILllIlIIIIlllIIlllJ

 

 

[III

1 l l 1 I l l l l I l I I J I l J I I 1 I I I

0

 

300 350 400 450 500 550 600
nm

Adetection = 630 "m

dilute CHZCI2 solution

Figure 17. Excitation Spectrum of ReO2py41.

61

environment;“ additionally a viscous environment is likely correlated to longer emissive
1ifetimes.18’“

The luminescence is readily quenched by hydrogen bonding to the O atom.“ The
excited state of ReO2py4+ is quenched by protons of organic acids in acetonitrile; trends
in quenching rate constants are not solvent dependent.“’72 The luminescence quantum
yield of the excited state is 0.02-0.04 in aprotic solutions“ and is negligible in protic

environments. Small quantities of water in the solvent can shift the Am from 640 (dry) to

675 nm."3 Emission lifetime varies over four orders of magnitude upon shifting fiom
aqueous to nonaqueous regions.“ The ground state pKa is -0.3 67 and the excited state
pKa (pKa‘) is 11.3. Along these lines, a large H/D effect on the lifetimes is observed in
aqueous solution.“

The excited state of ReO2py4+ is a good reductant, efficiently reducing acceptors
such as methylviologen and olefins. In addition to Br6nsted basic properties, ReOzpy4+
may be reversibly reduced to Re(III) electrochemically in aqueous solution;“’73 the one-
electron intermediate was inferred to disproportionate. Re(II) can catalyze the reduction
of IT“ to H2, and oxidation of the Re complex to other oxidation states will result in

74 The redox properties of the ReO2py4+

oxidation of secondary alcohols to ketones.
complex are extremely sensitive to the O-Re-O angle. Substitution of bpy for two py
ligands results in the O-Re-O angle shifting from 180° to 121°, causing the reversible
oxidation potential to shift by 600 mV;“ lesser shifts can be induced by varying the
electron withdrawing abilities of substituted py and bpy ligands.

The trans- ReO2py4+ cation itself is a poor choice for an MPC, as it is known to

undergo only single electron photoredox chemistry. However it is sensitive to its

62

environment and thus is a useful probe of heterogeneous environments, including
micelles and (most relevant to this work) lamellar solids.“ The work Of Newsham
et a1.18 is especially relevant in allowing orientation of the OReO+ core within layered
solids to be probed. Photophysical studies show that the cation can "key" into specific
sites in the gallery of CLOs leading to varied excited state behavior.18 Accordingly,
studies with Re02py4+ were undertaken to probe the TOA-ZrP gallery environment and
its dependence upon various intercalation solvents and to compare the photophysics of
TBA-ZrP and TBA-TiP with the intent of determining guest/host interactions with

respect to redox properties.

Trans-dioxotetrakis(pyridine)rhenium(V) iodide [ReO2py4+] Results

Substantial characterization has been reported in the literature and summarized
above, so primarily supplemental data are reported below. The UV-Vis-NIR diffuse
reflectance spectrum Of crystalline ReO2py4+ is shown in Figure 18; absorption maxima
are at 333 and 430 nm with a shoulder at approximately 520 nm.

For 1H NMR, the only expected resonances will arise from the pyridine protons.
In CDCl3 solution, 1H NMR yields three multiplets at 9.06, 7.54, and 7.80 ppm with a
peak area integration ratio of approximately 222:]. Additional peaks consistent with free
pyridine are Observed in the spectrum; this is understandable as it was the crystallization
solvent. Literature values are 9.14, 7.51, and 7.73 ppm in CD3N02 solvent62
corresponding to the C2/C6, C3/C5, and C4 protons respectively.

The emission spectrum of crystalline ReO2py4+ is shown in Figure 16. Note that

fine structure is seen at RT and clearly resolved at 77 K. The emission maxima were

63

Diffuse Reflectance UV-Vis-NIR of Crystalline ReOzpy4l

 

        
 

 

     
 

 

 

 

_ I I I I I l I I I I I ‘I I I I I I I
100 j
80 :— j
8 Z Z
5 _ 3
*5 6° - _
o
G: '- _
9 _ _
e\° 40 r —
: 520(?) nm shoulder :
20 j :
*_‘ lMAx (crystalline) = 333, 430 nm ‘
I l l I I l 1 I I I I I 1 1 l I I I I q
0
200 400 600 800 1000 1200

nm

Figure 18. Diffuse reflectance UV-Vis-NIR spectrum of crystalline ReO2py41.

64

636 nm at RT and 633 nm at 77 K. Vibrational progressions were observed in the
crystalline powder at approximately 800 cm’1 at room temperature and at 900 cm'1 and

200 cm“ at 77 K.

Solvent Dependency Experiments
ReO2py4+ exchanged TOA-ZrP was prepared under different intercalation solvent
conditions to Observe the effect upon the intercalation chemistry and the ReO2py4+
excited state properties.
Incorporation, Experimental
0.20 g TOA-ZrP, 20 mL of alcohol, 1.5 mL H20 and 0.70 g ReO2py4+ were
stirred at ambient temperature for 18 days, then filtered with fine fiits under aspirator
vacuum and washed well with water, acetone and hexanes. For 1-Octanol, l-heptanol,
and 1-hexanol, a light yellow powder was isolated; for l-butanol and l-propanol, the
powder was medium yellow in color. OReO+ was sparingly soluble in the mixtures
containing hexanol, heptanol, and octanol. To 65 mL isopropanol (2-propanol), 5 mL
H2O, 0.3 g TOA-ZrP and 1.5 g ReO2py4+ were added and stirred at room temperature for
four days; then vacuum filtered with a fine fiit and washed well with acetone and water
until the wash solvent was clear. 60 mL of 95% ethanol was used in a similar
preparation, except that no water was added. Both resulted in yellow powders.
Results and Characterization of the Incorporated Layers- Solvent Dependency
ESCA I Loading / Powder XRD
Elemental analysis of the ReO2py4+ intercalated TOA-ZrP by using ESCA yielded

relative mole ratios of the elements Zr, Re, and N of the layered products from the

65

various solvent mixtures. The zirconium phosphate sheets are chemically stable and
provide a reference from which the loadings and various ratios are computed. The Re/Zr
mole ratios range from .07 to .17 (3.5% to 8%) of theoretical exchange capacity and are
summarized in Table 1. Note that the color saturation of the samples roughly follows the
same trend as the loadings. An analysis on pure TOA-ZrP indicated a loading of
1.15 moles of TOA per mole Zr (58%). A full loading of 2.00 would result from every
site in the galleries being occupied. Upon ReOzpy4+ incorporation the measured TOA
loading was typically reduced to about 25%.

Powder X-ray diffraction data of ReO2py4+ exchanged TOA-ZrP typically have
only (1001, d002, and d003, peaks observed as shown in Figure 19. The doc] spacing of about
18 A for ReO2py4+ exchanged layers is observed for most solvent systems used.

Emission, Lifetime

Steady state emission (phosphorescence) spectra were obtained by excitation at
435 nm. The emission spectra for the various products are shown in Figure 20. Table 2
includes approximate relative observed intensities of the emission spectra.

The emissive lifetimes (Acx=532 nm) for the different products are summarized in

Table 1. After reexamination of the original decay curves and consultation with Dan
Engebretson in the Nocera group at MSU, the values shown were determined to be
questionable. For some samples the PMT saturated, preventing accurate measurement of
the early part of the decay curve. Secondly, the CAMAC crate's accuracy in the first
samples are systematically questionable. Thirdly, the above hardware's bandwidth was
potentially inadequate for measuring the fast components. For these reasons the values

must be considered only approximate, with unknown and potentially varying errors.

66

Table 1. Loading data for solvent dependency experiments.

 

 

intercalation color Reer TOA/Zr loss of spacing (A)
solvents ratio ratio TOA
1-octanol/ H2O light yellow 0.10 0.57 50% 17.9
1-heptanoll H2O light yellow 0.07 0.69 40% 18.0
1-hexanoll H2O light yellow 0.07 0.52 55% 17.9
1-butanol/ H2O yellow 0.15 0.51 55% 18.1 ,15.7
1-propanol/ H2O yellow 0.15 0.18 84% 18.1,15.7,14.3
ethanol/H2O yellow 0.15 0.48 59% 16.0
isopropanol/H2O yellow 0.17 0.53 55% 18.9
"pure" TOA-ZrP white N I A 1 .15 N / A 19.8

 

 

 

 

 

 

 

Table 2. Fitted emission lifetimes and intensities for solvent dependency experiments.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

emission lifetime data
synthesis Re/Zr TOA/Zr d- relative Temp % r1 % t2
solvent ratio ratio space emission (K) t1 (115) t2 (us)
(A) intensity
1-octanoll 0.10 0.57 17.9 10x 295 56.7 29.7 43.3 5.1
water (29%) 77 54.4 42.9 45.6 3.2
1-heptanol/ 0.07 0.69 18.0 10x 295 54.8 30.8 45.2 6.1
water (35%) 77 65.9 55.5 34.1 10.5
1-hexanoll 0.07 0.52 17.9 1x 295 42.8 25.1 57.2 2.7
water (26%) 77 46.6 26.6 53.4 1.9
1-butanol/ 0.15 0.51 18.1, 1x 295 65.2 25.4 34.8 4.2
water (26%) 15.7 77 51.0 324 49.0 5.4
1-propanol/ 0.15 0.18 18.1 1x 295 75.7 16.7 24.5 2.1
water (9%) 15.7 77 67.6 33.0 32.4 3.0
14.3
ethanol] 0.15 0.48 16.0 10x 295 87.3 42.4 12.7 5.5
water (24%) 77
isopropanol/ 0.17 0.53 18.9 1000x 295 85.4 42.6 14.6 13.7
water (26%) 77 92.5 67.7 7.5 10.4
OReO+ ---- --—- 295 74.7 35.9 25.3 8.1
cgstals 77 48.6 82.1 51.4 48.9
TOA-ZrP —--— 1.15 20.0
(58%)

 

 

 

 

 

 

67

X-ray Diffraction Pattern of ReOZpy4 in TOA-ZrP

TfirlIIIIIIIIITIIIfiIIIIIfiIIIIIIfiIIHIIIII

 

typically about 18 A

 

IIIlIllIIIllJIIlIllIlllIIlIIIIIIllIIIII

Relative Intensity
IIIIIIIIIIIITIIIITTIIIIIIITIIIIIIIIII

 

 

 

AAA Lg. A L‘AJ A LklAl

10 15 20 25 30 35 40
2 theta

O
01

Figure 19. Typical powder XRD pattern of ReO2py4+ in TOA-ZrP.

68

 

 

 

 

 

 

 

 

 

 

_ 4F I I I I7] I I F7 I I I I I I I l I l l I I I l '_

g; : " :
a) r t
C : I
0) — _
*3 —- ..
C ; I
'5 C :
.2 T T—
H ’- _,
<_u : :
a) r i
n: 3, :
" lllllllIllllllll "min"

550 650 750 850
l-Octanol/H2O

:II II II I II II I II T II II II II II I II:

.E‘ 2— '2
g i j
'2 a 5
fl : :
0 i. j
m _ _
”I l 1 I IJ_J,I l I l I I I I l I L I I I I l l I l l l I“

550 650 750 850
l-hexanol / H2O

k.l II II II II II II II II II II II II I ll:
g: —:
a) i j
E 2 3
g g 3
E a —
0 T . ‘1
n: :..a —
I I I I I I I I,I I L I I l I I I ILLJ,I I_LJ, VILIT

 

 

 

550 650 750 850

 

 

 

 

 

l-propanol / H2O
El I I II I I HI I I I l I l l I II I I I l I l
E‘: :
I _ _
E - _
9 — , _
'5 . :
Q r ,' —:
05 ' _
r: r: ,
:rl‘il lllllllllllll I"J‘L‘L:

55 750

 

I

  

Relative Intensity

‘IIIII llIIII lllIl IIIIIII

 

 

II II II I II II II II II II II II II

III

I

XJJIIIIIIIIIIIIIIIIIIIII

 

 

‘1. llllllllll
550 650

1 l 17510
1-heptanol/ H20

00
(11
O

 

IIII

“I
.J.'

Relative Intensity

IIlllIIllllIllllIllllIIIIII

 

 

I‘T I7 I I I l I I l I I l I I l I T I I l I I l [I
n‘.
g :- \,." n

:11 II II II II ll II II 1 II III II II 11

lIllllIllllIlIIIIllllIllllill

 

 

550 650 750
l-butanol / H20

850

 

H

l

Relative Intensity
lIIIllIIlIIIllIIlIlIIllII

 

 

II III III III III III III III III II

.‘ '
_I II II 11 11 I1 11 II II II II 111 II Id

LL

lIlIlIlllIIIIIlllIlllIlIlII

 

 

550 650
ethanol / H20

750

 

dotted line = RT
solid line = 77K

 

 

 

isopropanol / H20

850

Figure 20. ReOzpy4I in TOA-ZrP emission spectra with varying intercalation solvents.

69

With these factors considered, the observed lifetimes are similar to those of solid
ReOzpy4I. Additionally, there are no significant trends observed in the lifetimes. The
diffuse reflectance instrument was not yet purchased at the time of the solvent
dependency experiments, so no measurements were made.

Discussion- Solvent Dependency

Loading

There are three limitations to the XPS method of elemental analysis. First, the
high vacuum in the sample chamber will cause some loss of any volatile component such
as the intercalation solvents or TOA. Secondly, it is sensitive to surface contamination.
Finally, the X-rays will penetrate only a few layers into the sample, resulting in the
results being skewed toward the surface (rather than bulk) composition.

The TOA loadings of the ReOzpy4+ intercalates were calculated using the nitrogen
data and summarized in Table 1; four N’s per Re were subtracted from the total moles of
N and the difference was attributed to the TOA N atoms. Values measured for C were
not considered as it is a common surface contaminant. Note that for most intercalates
there is an observed loss of approximately half of the TOA. This TOA loss is expected
on the basis of sterics, as bulky ReOzpy4+ is intercalated while the layer d-spacing
decreases; the loss is also expected upon ion exchange as the primary alkyl amine has
fewer protons to bond to in the interlayer region.

Sample variations and error in the measured quantities of Re present both
contribute to the wide range (9-3 5%) of measured TOA loading values. An analysis on

pure TOA-ZrP indicated a loading of 1.15 moles of TOA per Zr (5 8%). A full loading of

70

2.00 would result from every site in the galleries being occupied, but this is not the case
here due to steric hindrance.

Significantly, there were no observed correlations of Re loading to emission
intensity, emission lifetimes, or ReOzpy4+ solubility in the solvent. The loadings were
within a factor of two of each other, with the lightly loaded samples having the weakest
color. This colorimetric observation supports the general validity of the Re loading data.

XRD

Powder X-ray diffraction data (Figure 19) demonstrate that ReOzpy4+ exchanged
TOA-ZrP typically have evenly spaced layers with only d001, d002, and does Peaks
observed. This dam spacing of about 18 A for ReOzpy4+ exchanged layers is observed no
matter what solvent system is used. Such a pattern is consistent with a turbostratic
structure, where there is long range order only in the crystallographic c-direction. The
layers are parallel at regular spacings, but with significant disorder (translation and/or
rotation) with respect to each other.

The slight contraction from the 20 A TOA-ZrP dam spacing is probably due to the
decrease in TOA loading and slightly increased interlayer attractions resulting from
incorporation of the ReOzpyf cation. The estimated d-spacing for ReOzpy4+ exchanged
into ZrP with no swelling agent is 11 A to 16 A, depending upon the orientation of the
Re02py4+ cores relative to the layers and the amount of "keying" present. The X-ray and
ESCA data indicate that the interlayer alkyl amine environment is retained upon

intercalation.

 

IV

 

71

Inferred Orientation(s)

Previous work in another systeml8 has demonstrated that the orientation of
ReOzpyf in a layered system may be established. In these experiments, the amines pillar
the ZrP layers enough to keep a spacing of about 18 A. As the estimated d-spacing for
ReOzpy4+ exchanged into unpillared ZrP is 11 A to 16 A, none of the possible
orientations may be excluded by the XRD data.

Emission

Whereas the structure of the products made by ReOzpy4+ in ZrP under different
conditions are similar, the excited state properties depend markedly upon the alcohol
used, as shown in Table 2 and Figure 20. The existence of emission spectra indicate the
intact ReOzpy4+ unit is present and not significantly bent. The observed maxima of the
more strongly emitting samples was 630 to 640 nm at RT and 77 K. This means the
excited ReOzpy4+ is not relaxed well in its environment. Note the samples with weak
intensity were correlated to the disappearance of the progressions. This disappearance
may be real or possibly an artifact of the instrument working at the edge of its detection
limits.

More specifically, the emission spectrum for the intercalation of ReOzpy4+ from a
l-octanol/water solvent mixture is shown in Figure 20. Retention of vibrational structure
and intense luminescence at room temperature indicate that the ReOzpy4+ ion is virtually
unquenched by hydrogen bonding. This is not the case as the alcohol’s n-alkyl chain is
shortened. For example, the weak emission intensity and loss of vibrational fine structure
when l-butanol/water is used to exchange the ion indicate the OReO+ core is in a

hydrophilic interlayer environment. The notable exception is ethanol, which might result

72

from its higher vapor pressure and possible evaporative loss. Isopropanol was examined
along with the n-alkyl alcohols above as an intercalation solvent. The room temperature
emission spectra of the OReO+ exchanged (with isopropanol) product was discovered to
be very intense and almost identical to the crystalline ReOzpy4+ emission spectra.

Differences in emission spectra and intensity indicate a substantial difference in
the protic environments in the galleries of the various layered compounds. These results
suggest that the ReOzpy.fr luminescence is readily quenched with shorter alcohols by
hydrogen bonding to the oxygen atoms of ReOzpyf. When fine structure is present, it is
additional evidence the ReOzpy4+ is cradled in an organized alkyl environment, as a very
structured emission band is observed when the complex is in an ordered cluster.66

Emissive Lifetime

As mentioned above, the uncertainty in the actual lifetime values prevents their
detailed dissection. The lifetimes are similar to or longer than literature values from
other systems. Based on the generous layer spacing and existence of alkyls in the
galleries, the environment is viscous with minimal or nonexistent tumbling of ReOzpy4+
ion. The roughly comparable lifetimes of these samples suggests a similar interlayer
environment for all, with an unquenched ReOzpy4+ accounting for most of the emission
intensity. This does not fully correspond to the weak observed emission intensity from
certain samples. Perhaps these samples have a significant quantity of nonemissive
complexes.

The other significant finding is that the ReOzpy4+ excited state is not effectively

coupled to the layered ZrP. If the coupling were significant, the emission lifetime would

Irwin" I
. . I

 

73

be noticeably attenuated in all samples due to the presence of new nonradiative decay

pathways or quenching mechanisms.

LMP Comparison Experiments
In a second series of experiments, ReOzpy4+ exchanged TBA-ZrP and TBA-TiP
were prepared under the same intercalation solvent conditions to observe the effect upon
the ReOzpy4+ excited state pr0perties. As the difference between the systems is the
identity of the LMP metal, any significant difference in excited state properties is likely
due to redox activity of the Ti atoms.
Incorporation, Experimental
For the LMP comparisons, 0.20 g TBA-ZrP or TBA-TiP, 5 mL of isopropanol,
5 mL H20 and 0.30 g ReOzpy4I were stirred together overnight at ambient temperature,
then filtered with fine fi'its under aspirator vacuum and washed well with water. An
off-white powder was isolated for both. For unknown reasons the SnP products could not
be prepared as ReOzpy4+ decomposed, therefore no analysis was completed.
Results and Characterization of the Incorporated Layers- LMP Comparisons
Loading I Powder XRD
The loadings were not directly measured by XPS. A colorimetric estimate by eye
on the basis of yellow color saturation in comparison to the solvent dependency samples
indicated the loading was less than 1%. Powder X-ray diffraction of the products yielded
patterns with prominent sharp peaks at 15.7 A and 14.9 A for the TBA-ZrP product and

14.9 A for the TBA-TiP product as shown in Figure 21.

74

Emission / Lifetime / Absorbance

The steady state emission (phosphorescence) spectra were obtained by excitation
at 435 nm. The emission spectra for the two products are shown in Figure 22. The
observed intensities of the emission were quite weak, presumably due to low loading.
The weak intensities and poor S/N preclude accurate measurement of peak maxima.

The lifetime (KEX=532 nm) for TBA-ZrP was a biexponential decay of 17.8 us
(18%) and 1.0 us (82%). The lifetime for TBA-TiP was fitted to a biexponential decay
of 17.1 us (46%) and 1.5 us (54%).

The diffuse reflectance absorption spectra are shown in Figure 23. The maxima
were similar. The absorption edge in the low 300’s in TBA-TiP is of the TiP itself, as
can be seen in Figure 9.

Discussion- LMP Comparisons

Loading

The original mass of the swelled layers that reacted (0.20 g) was reduced by about
25%; the reduction could be readily explained by loss of TBA to the solvent and in
filtration; a 1% ReOzpy4+ loading would add no substantial mass to the collected powder.
Quantitative measurements of solution ReOzpy4+ before and after incorporation would
have errors greater than the estimated loading. For purposes of this chapter, the loading

should be considered to be a maximum of 1%.

75

 

 

 

 

 

 

 

I I I I I T I I I I I I I I I I I r I I I I I I I I I I r I rT T I I I I I I
— X-ray Diffraction Pattern of ReOzpy“+ in TBA-ZrP -
'_ ‘1
,3 ~ 15.7 A -
U) .- a
C
.93 - ..
,5 — a
o ~ .
.2 _ a
a
5! _ a
o
m '- ‘l
'- 'l
~ 4
r 1 1 .LI-I.MML5AAW~A-+wa Jim». .I
0 5 10 15 20 25 30 35 40

2 theta

 

TIITIIIIIITjIIIIIIIIIIIIIIIIIIIIIIIIIIT

I

X-ray Diffraction Pattern of ReOzpy4+ in TBA-TiP

14.9 A

Relative Intensity

 

I11I1lIIIIJLIIIIIIJIIIIIIIIIIIILI

IIITIIIIIIIITIrIIIIIIIIIITITIIIII

 

 

 

 

LLIHAALIAI+AWLiimiIIIAII

0 5 1O 15 20 25 30 35
2 theta

r—

 

.b
C

Figure 21. XRD of ReOzpy4+ in TBA-ZrP and TBA-TiP.

500

400

300

200

Relative Intensity

100

76

 

 

 

—- OReO in TBA-ZrP
--------- OReO in TBA-TiP

 

 

 

 

coo-ggauocucouugoog
wuov"

-
.-.-.-----.--
-co----------
--.--.--.-..-
fiQO‘OCUV'
.Uoo

 

 

 

I 1 J I l I 1 I

T’I'FT ’ '

'.-i. a. e.
I

L 1 1. O L I ills-.171. '.‘.1°".i

 

 

 

 

nm

Figure 22. Emission spectra of ReOzpy4I in TBA-ZrP and TBA-TiP.

800

h ..

5

 

77

 

 

 

 

 

 

120 _ I I I I I I I I I I I I I I I I I T I _

F- I -<

_ 3\ «I -

a , I ‘h

100 T ,,~-a;~~u“u'u'|”, VIM-f} I'.‘ NW1

~ . ' ’ I ’ 1

P— l -1

I I

a, 80 r , nouse —

o b “' s , l d

C " ’ "

E I— d

g 60 [PI ~I‘ :
c l I I \ I

e *4 l I I‘ I -I

..\° 7 ‘J . ’ ‘

40 f x ,‘ j

L t,’ """ OReO in TBA-Zr P a

20 ; — OReO in TBA-TIP _:

— —I

0 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1 I 1 1 1
200 400 600 800 1000 1200

Figure 23. Diffuse Reflectance UV-Vis-NIR of ReOzpy4+ in TBA-ZrP and TBA-TiP.

78

XRD

Powder X-ray diffraction data (Figure 21) demonstrate that ReOzpy4+ exchanged
TBA-ZrP and TBA-TiP typically have evenly spaced layers as dom and dooz peaks are
observed. Such a pattern is consistent with a turbostratic structure, but with slightly less
long range order than in the solvent dependency TOA-ZrP systems. Similarly, the
contractions from the TBA swelled LMP dom spacings are probably due to a decrease in
TBA loading and slightly increased interlayer attractions resulting from incorporation of
the ReOzpy4+ cation. The X-ray data indicate that the interlayer alkyl amine environment
is retained upon intercalation.

Inferred Orientation(s)

Previous work in another system18 has demonstrated that the orientation of
ReOzpyf in a layered system may be established. In the LMP dependency experiments,
the amines pillar the ZrP and TiP layers enough to keep a spacing of about 15 A. As the
estimated d-spacing for ReOzpy4+ exchanged into unpillared ZrP is 11 A to 16 A, none of
the possible orientations may be excluded by the XRD data.

Emission

The existence of emission spectra (Figure 22) indicate the ReOzpy4+ unit is
present; unfortunately the weak intensity precludes proper assessment of its maxima and
fine structure, if any.

Emissive Lifetime

The quality of the lifetime data for these experiments is considerably higher than
the solvent dependency data. The time constant for the long component of the decay is

17-18 us, identical to the literature lifetime values in dry pyridine solvent.63 These nearly

79

identical data between TBA-ZrP and TBA-TiP indicate the ReOzpy4+ is photophysically
in a nearly identical environment. This means that there are no significantly differing

deactivation pathways in these LMP hosts.

Conclusions
Possible Diffusion / Insertion Mechanism

The results of the solvent dependency studies imply the existence of different
alcohol intercalation mechanisms. I propose that the shorter n-alcohols hydrogen bond to
the ReOzpy4+ and are carried into the hydrophobic layer upon intercalation; quenching is
substantial because the ReOzpy4+ ion tends to remain hydrogen bonded to the alcohol. In
the case of longer alcohols, the steric congestion within the TOA interlayer precludes the
OReOValcohol hydrogen bond from surviving incorporation into the layer frequently.

If steric factors play a significant role in precluding co-intercalation, then
branched alcohols would tend to be absent from the gallery region and the room
temperature emission spectra would not show quenching due to the hydrogen bonding.
The room temperature emission spectra of the ReOzpyI;+ exchanged (using isopropanol)
product was discovered to be almost identical to the crystalline ReOzpy4I emission
spectra, in fact with more visible fine structure and greater intensity than the other
samples. These observations support the above proposal.

Presence 0f Protons
The existence of emission indicates that ReOzpy4+ is incorporated intact. Absence

of quenching in the intensely luminescent samples indicates that the exchangeable

 

8O

protons in the layer are bound up away from the OReO+ core. Their likely location is
between the amines and the apical oxygen atoms on the ZrP.
Amine Packing

As was stated in Chapter 2, primary n-alkyl amines packed very efficiently,
hindering the incorporation of bulky complexes even though the layers were adequately
separated. The highly branched chains of tert-butylamine and tert-octylamine cannot
pack tightly, permitting easier intercalation of the ReOzpy4+ complex. The loosely
packed hydrophobic branches provide an ordered environment for the complex, leading
to structured emission spectra.

Electron and Energy Transfer

In the LMP comparison studies, the observed emissive lifetimes of both the long
and short components are nearly identical, so the various deactivation pathways are
approximately the same. As the lifetime long components are the same as those found in
nonaqueous solution, the complex is not deactivated by means other than those in
solution. No energy transfer is present to ZrP or TiP, consistent with the observation of
energy transfer between two species through multiple layers of ZrP.34 This finding is not
surprising as the ReOzpyf emission band does not overlap with any LMP absorption
bands. No electron transfer is present either, as the lifetimes are not attenuated. The TiP
is thus not reduced by excited ReOzpy4+. The significance of these findings is that an
MPC will not necessarily be deactivated by the host LMP layers; such a deactivation

would reduce the efficiency of a LIPS or preclude its function.

CHAPTER 4

RU(BPY)32+ INCORPORATION

Introduction

No energy transfer was found between ReOzpy4+ and ZrP or TiP. No electron
transfer is present either, as the lifetimes are not attenuated. The TiP is thus not reduced
by excited ReOzpy4+. The significance of these findings is the implication that an MPC
will not necessarily be deactivated by the host LMP layers; such a deactivation would
reduce a LIPS efficiency or preclude its function. It is important that the LMP be able to
transfer redox equivalents (i.e. electrons) when appropriate to or from the MPC for
regeneration purposes. Such regeneration of the MPC will be a critical step in
completing the catalytic cycle. The ideal host would by design preferentially transfer 1
or 2 electrons to help complete the catalytic cycle; the number would be dependent upon
the specific MPC. For example, the LIPS efficiency might substantially improve if the
LMP was incapable of redox quenching the MPC by a one electron mechanism.
Consequently, for studies in the pursuit of a functional LIPS, a series of chemically
similar hosts with varying redox capability are needed.

The purpose of this Chapter is to probe the transfer of electrons between the host
and guest. One aspect to be addressed is whether there are differences for one or two
electron transfers. The ubiquitous tris(2,2’-bipyridine)ruthenium(II) cation

(Ru(bpy)32+)75 will be used to confirm the LMP capability for the purpose of coupling

81

 

82

redox equivalents in the layer to the MPC for regeneration. Ru(bpy)32+ provides a
complementary probe to ReOzpy4+ on the coupling of electrons to the LMP as their
sensitivity to surroundings differ. Ru(bpy)32+ will be used for one electron quenching
studies with each LMP.

The LMP host series has a redox inactive species (ZrP) along with single (TiP)
and multielectron (SnP) redox active species. The redox inactive d0 Zr(IV) atoms
composing the layers are photochemically inert inasmuch as electron transfer between the
layer and an excited guest can not occur; consequently this material provides a reference
host structure for excited state studies of LMP intercalates. The common titanium
oxidation states of IV and III (d° and d') and tin of Iv and II (d'° and szdlo) could in
principle couple to a catalytic site in the galleries with one-electron and two-electron
redox activity, respectively; Figure 24 summarizes the MPC regeneration scheme as a
function of host metal identity. Comparison of Ru(bpy)32+ one electron quenching

studies along this LMP series will provide clues to their potential for constructing a LIPS.

Review of the Ru(bpy)32+ Probe

Ru(bpy)32+ combines chemical stability, 3 long excited state lifetime, and useful
redox properties.75 Its structure is shown in Figure 25; the immediate coordination sphere
is of Oh (octahedral) symmetry, although the complex symmetry is actually D3.

In aqueous solution the lowest energy absorption band (d —> n', MLCT) exhibits a
maxima at 452 nm. Excitation into any of the absorption bands yields a single emission

band at 602 nm at RT in degassed aqueous solution that is reasonably intense (¢ = 0.042)

and long lived (I = 600 ns).75 Emission properties are fairly temperature and medium

 

 

83

 

 

 

 

 

 

Zr4+ o 1
Ti“ 1
Sn4+ 2

Figure 24. Coupling of LMP metal atoms to MPC.

 

84

 

Figure 25. Structure of 4d6 Ru(bpy)32+.

 

85

dependent; this sensitivity to the local microenvironment provides a useful probe Of

intracrystalline environments. Indeed it has been used as a probe in studies of various

78,79,80 81,82

clays,76’77 layered metal phosphates, and phosphonates.

Excited Ru(bpy)32+ can act as a photosensitizer; upon excitation, it can strongly
drive either a reduction or an oxidation dependent upon whether an electron acceptor or
donor is present. As the ZrP, TiP, and SnP metal atoms are all (10 or d"), the LMPs will
only be able receive electrons, so any redox change in the Ru(bpy)32+ will be to
Ru(bpy)33+.

Excited Ru(bpy)32+ will readily self quench under easily obtainable conditions.
As the lifetime in dilute solution is about 600 us, at even modest concentrations an
excited complex can frequently be quenched by another cation. If individual Ru(bpy)32+
cations can be separated, self quenching can be minimized. The d10 Zn2+ in Zn(bpy)32+ is
not readily redox active and is colorless in solution. It is isostructural to Ru(bpy)32+ and
will thus afford similar intercalation behavior. Figure 26 shows a schematic of
Ru(bpy)3.2+ dilution with Zn(bpy)32+ in a ratio of 1:10. Dilution experiments will be
described later in this chapter to delineate the effect of self-quenching.

Self quenching of Ru(bpy)32+ is especially important in lamellar solids.76
Lowering Of the Ru(bpy)32+ loading will thus have no effect on the presence of
aggregation, only change its distribution; self quenching will still exist but rates will
change in an unknown way. This self-quenching has mechanistically been explained in
terms of the “antenna effect” similar to that found in natural systems.“84 Zn(bpy)32+

dilution is necessary to maintain a sterically similar environment in the galleries as the

number of guest molecules will remain approximately constant. Zn(bpy)32+ dilution was

Tl

1T-.. .
1. .

‘.

 

86

 

O Rumpus/>3”

O Zn(bpy)a2+

Figure 26. Schematic of Ru(bpy)32+ dilution with Zn(bpy)32+.

87

demonstrated to substantially reduce the self-quenching rate.76 No work has been
reported involving Ru(bpy)32+ dilution with Zn(bpy)32+ in any layered metal

phosph(on)ate system.

Synthesis and Incorporation of Layers

All operations were done under Ar with minimal light present; solvents were
previously degassed with Ar. Stock solutions of Ru(bpy)32+ or Ru(bpy)32+ with
Zn(bpy)32+ diluent in MeOH/1420 solutions were prepared. The Ru/Zn mole ratios were
1:0, 1:9.4, and 1:117. 10 mL of solution containing approximately 0.1 g combined
Ru(bpy)32+ and Zn(bpy)32+ were added to a test tube containing about 0.1 g TBA-ZrP,
TBA-TiP, or TBA-SnP and stirred for several hows. The suspension was filtered on a
fine fi’it and washed with water and methanol. The powder was loaded into an quartz
tube as described in Chapter 2 and evacuated to preclude 02 quenching. The time
resolved emission measurements were made, followed by characterization by steady state
emission, diffuse reflectance absorption, and then XRD. It is assumed that Ru(bpy)32+
and Zn(bpy)32+ incorporate similarly, resulting in the same proportion both in and out of

the galleries.

Results

Figure 27 shows the XRD data for each of the samples, with the baseline Offset
for clarity. The spectra are similar, varying somewhat in d-spacing. Figure 28 shows the
diffuse reflectance data for each Of the samples. Note the trend toward disappearance of

the MLCT band at about 450 nm as the Ru(bpy)3,2+ is diluted with Zn(bpy)32+. Figure 29

 

88

 

 

  
    

 

        
  

 

 

 

 

 

 

 

 

 

 

RulZn
ratio rYIrI'lIII'Il'[I'TIIIIITITIIIITI
1:117 l l TBA-ZrP
lllk15.5,13.8A
19.4 M
1:0
I l l l 1 L4
I I l I I I I I I I I I I
TBA-TiP
1:117
119.4
1:0
TBA-SnP
1:117 'L‘M’J
1:9.4 WWW
1:0
15.8A
l l l I l l l l L 1 l
0 5 1o 15 20 25 30

2 theta

Figure 27. Powder XRD spectra of Ru(bpy)32+ and Zn(bpy)3,2+ in TBA swelled LMPs.

89

 

 

 

 

I1 I I r I I I I I T T T I r I T I fi I I
l 1:117 TBA-ZrP
Irv/To.
I‘ "
l\ l 1 :o
\l
\\ Ru I.Zn
I ratio
\J'
L l l L L L I l l l I l I 1 l l I l I
I I I I I I I I I I I I I I I I I I I
1:117 TBA-TiP

 

 

 

 

 

 

Wfi i 1 I n L 1 L I . l I I I l I I I
T T I I I I I I I T I T I I I I I I I
1:117 TBA-SnP
3 f Ira-av - .. .--..-.-
I ,rxx 1:9.4 *
I l. 1:0
. l
l “’
\l
l
.1.
xx 1
l L l l J 1 i i l 4 L l I l l l l l l
200 400 600

 

800 1 000

1200
nm

Figure 28. Diffuse reflectance spectra of Ru(bpy)32+ and Zn(bpy)32+ in TBA-LMPs.

 

I I I I I I I I I I I I I I I I I l l I I I I I I I I I r

TBA-ZrP

   

Ru IIZn
ratio

1:0 Amax= 619 nm
1:9.4 Max: 608 nm
1:117 Am“: 602 nm

     
   
 

  
  
 

 

 
 

TBA-TiP

  
     
 

1:9.4 Amax = 624 nm
1:117 Amax= 612 nm

  
   
 
   

 

TBA-SnP

1:0 kmax=631nm
1:9.4 kmax=611nm
1:117 Amax=605nm

 

  
  
  

 

 

1 l l l l 1 1 PL 1 L l g l l_l l L l J J_L 1 I; l I l l
500 550 600 650 700 750 800
nm Max = 450 nm

Figure 29. Emission spectra of Ru(bpy)32+ in TBA swelled LMPs.

91

shows the steady state emission data (Max = 450 nm) for each of the samples. Note that
the XMAX blue shifis as the Zn(bpy)32+ dilutes the Ru(bpy)32+. Table 3 shows the fitted
lifetime data for each of the samples; several general trends should be noted. First, the
proportion of the long component increased with dilution. Second, the value of the long

component rapidly plateaued at 1.4-1.5 us for both ZrP and SnP but not TiP. Third, for

both long and short components, the TiP samples had shorter lifetimes than the ZrP and

SnP samples at the same dilution.

Discussion
XRD
The samples display turbostratic layering. The crystallinity varies, and is never
very high. The d-spacing for the undiluted Ru(bpy)32+ samples is 15-16 A, reasonably
close to literature values for Ruaapy)32+ directly incorporated in ZrP.78 Some of the
diluted samples display contracted d-spacings as low as 13 A. These smaller d-spacings
are consistent with complexes tightly keyed into the LMP. Uncoordinated aromatic
compounds such as bpy should have a lesser spacing of about 10.9 A. Note that the 2+
charge of the cations provides substantial electrostatic attraction to the anionic LMP.
Additional evidence for the reasonableness of the intercalation interpretation is
the incorporation of Ru(bpy)32+ (including Zna)py)32+ dilutions) into EtOH-ZrP. As the
EtOH will readily leave the galleries and not return, the end products will likely be pure
ZrP or ZrP with complex inside. In a series comparable to the one above, the d-spacings

started at 14.1 A and decreased to 12.3 A while the powders were strongly emissive.

 

* ‘7’
— J

 

92

Table 3. Summary of lifetime data for Ru(bpy)32+ and Zn(bpy)32+ in TBA-LMPS.

 

 

 

 

 

 

 

 

TBA-LMP [ Ru lZn xs | 13 (ns) xL | n (ns)
Zr 1:0 .43 320 .57 870
Zr 1:9.4 .32 770 .68 1500
Zr 1:117 .42 170 .58 1400
Ti 1:0 .83 53 .17 280
Ti 1:9.4 .39 200 .51 950
Ti 1:117 .42 200 .58 1200
Sn 1:0 .70 100 .30 370
Sn 1:9.4 .26 320 .74 1400
Sn 1:117 .23 350 .77 1500

TBA-LMP I Ru lZn x5 | 15 (ns) xL | n (ns)
Zr 1:0 .43 320 .57 870
Ti 1:0 .83 53 .17 280
Sn 1:0 .70 100 .30 370
Zr 1 19.4 .32 770 .68 1500
Ti 1:9.4 .39 200 .61 950
Sn 1:9.4 .25 320 .74 1400
Zr 1:117 .42 170 .58 1400
Ti 1:117 .42 200 .58 1200
Sn 1:117 .23 350 .77 1500

 

Conditions: 532nm excitation, 610nm detection
xs = fraction of short component
15 = lifetime of short component

 

XL = fraction of long component
11 = lifetime of long component

 

93

Diffuse Reflectance
The noted trend toward disappearance of the MLCT band reflects the decreasing
quantity of Ru(bpy)3.2+ in the host as Zn(bpy)32+ is transparent in the visible region; ligand
centered bands of both species remain in the UV. A quantitative comparison of absolute
intensities between samples is not advisable as sample preparation is not reproducible and
limited quantities of product were available.
Emission
The existence of emission proves the lumophore is intact. The emission blue shift
noted above indicates the Ru(bpy)32+ is less effectively relaxed by its environment as the
dilution with Zn(bpy)32+ progresses. This is consistent with the increasing lifetimes noted
in Table 3.
Lifetime
As noted in Table 3, the proportion of the long component increased with
dilution. This is consistent with self-quenching of Ruaapy)32+. For this reason, the
dilution experiments were necessary to address and overcome the issue of self-quenching.
The value of the long component rapidly plateaued at 1.4-1.5 us for both ZrP and
SnP but not TiP. The similar plateau for SnP and ZrP indicates that the self-quenching is
no longer a major factor in deactivation, and that the microenvironments are similar.
The TiP samples had shorter lifetimes for both long and short components than
the ZrP and SnP samples at the same dilution. This suggests a fundamental difference in
the microenvironment of swelled TiP that increases the quenching rate. Specifically, TiP

appears to have an additional quenching mechanism.

 

94

Conclusions

TiP appears to quench the excited state of Ru(bpy)32+, whereas ZrP and SnP do
not. This quenching can be rationalized by comparing the redox properties of the metals
composing the LMP. As Ru(bpy)32+ is a single electron reducer, it would couple more
effectively to a site that could undergo one electron reduction as opposed to two electron
reduction or no reduction. This would mean that Ru(bpy)32+ transferred an electron to the
TiP, although it may not have remained there for long.

Implications for this finding with respect to a LIPS are twofold. First, the transfer
of an electron from a complex to the layer indicates that transfer in the opposite direction
is also possible. With respect to the LIPS scheme shown in Figure 4, such a reverse
transfer will be required to regenerate the layers. It should be noted that such
regeneration will require the redox equivalent(s) to be made available to the layer
externally as the metals are already in a closed shell configuration.

The second implication is that as one electron redox coupling has been found for
TiP, by analogy two electron redox coupling may exist for SnP. With respect to LIPS,
the LMP could be selected for its ability to promote or hinder one or two electron
processes. For example, SnP could slow one electron redox quenching of an excited

MPC or promote a two electron regeneration.

CHAPTER 5

M02 INCORPORATION

Introduction

As mentioned in Chapter 3, the ReOzpyf system is not appropriate for our
desired LIPS design since the metal oxo core cannot function as an MPC. Similarly, the
Ru(bpy)32+ complex, although a powerful photoredox sensitizer, can only do single
electron chemistry unless a charge storage and catalytic site are provided, which is
contrary to the direction of this project. As the environment of the interlayer gallery has
been characterized and the redox interactions between guest and host partly delineated, it
is now appropriate to incorporate an MPC into a LMP and explore its initial
photochemistry.

Quadruply Bonded Dimetallic Complex as an MPC

Since construction of a LIPS will require an MPC, an appropriate class of
complexes must be selected. Work in the Nocera laboratories has identified candidates
for the role of an MPC; certain quadruply bonded dimetallic complexes have been
demonstrated to perform a multielectron reduction of substrate upon irradiation with one
photon. This photochemistry is rationalized by the MO diagram for a typical quadruply
bonded (14 compound in Figure 30. The lowest energy electronic transition is typically

l(132 -> 85*), a metal to metal charge transfer (MMCT) where a delta electron is promoted

95

 

 

 

 

 

96
L L

" WN- " MN-
/ | / I
L L

 

 

 

 

 

 

 

eg "4
5* (D10) 8
III 8 (D29) 8
\D
Q , .

 

Figure 30. MO Diagram for MAM Species.

* (am rat .3.

 

dx2-y2 - L

dvz+dyz

dxy+dxy

dxz"dxz

dYZ‘dyz

dzz + dzz

 

97

to an antibonding orbital.85 The electrons in this excited state are weakly coupled
between the two metal atoms, leading to a state has been recently proven to be
zwitterionic.86 It is ironic that such weak coupling would occur between metal atoms
bound by a quadruple bond a short distance apart. Excitation of the relatively weak delta
bond can thus provide an excited state that is adequately long lived to maintain an
oxidation state difference of two between the metal atoms, permitting redox chemistry to
occur.

A limited number of quadruply-bonded metal-metal (MA-M) systems have been
extensively characterized with regard to their reactions with small molecule substrates,
and shown to carry out two electron photoreactions at the electron rich and coordinatively

1.” The photoreaction is initiated upon the

unsaturated metal atom as shown in Figure 3
metal localized excitation of a M"Mll quadruple bond with visible light to form a MmMl
intermediate. Stabilization of the excited state is enhanced by the asymmetry in the
molecule, which can be achieved by ligand foldover. The reduced metal is
coordinatively unsaturated, and can readily undergo two electron oxidation (MmMm).
Presumably a two electron reduction at the MI" site is also possible. Regeneration of the
MAM complex has not been achieved in solution, so construction of a completed LIPS is
not yet achievable.

As the above M-4-M chemistry in the solution phase has now been demonstrated,
the question of positioning it between layers and its resultant initial photochemistry
comes to the fore. As the goal is to obtain multielectron photochemistry in a layered

system, the next logical extension of my research was incorporation of MAM cores into

the LMP gallery with retention of their emissive properties.

98

CI
“ch " C' (\i ("max

hvm' 'M=M '_,IIIMEMIXY CHIN—W

CI’LD 'L) 'L)

M = M0, W
I I = dppm = bis(diphenylphosphino)methane

XY = CH3|

Figure 31. Photoinitiated multielectron redox scheme in M—4-M system.

 

99

 

M E M“+ Ill

 

 

 

LL L LXLX
|/_|/ Ig_l/
L/ITL/T x/lTx/ITI
L L L

 

 

 

 

./ LT" ¢

 

 

l 1
G3 (7
I/EL I/L “I15: I/x

LLCIIII/L/‘I x/Il x/T

I:

Figure 32. Three Schemes to MPC Incorporation.

100

Three methods to getting MiM inside the Host

Construction of a LIPS will require incorporation of an MPC into a host; three
approaches summarized in Figure 32 have been elaborated specifically for the
introduction of photoactive MAM cores into layered phosphate host structures by the
Nocera Group. These schemes are: (1) direct intercalation of solvated MAM cores into
layered phosphates wherein the phosphate groups of the layers form the ligation sphere
for the bimetallic core; (2) acid-base reaction of specially fimctionalized ligands on the
bimetallic core with protons from the layers; and (3) replacement of the phosphate groups
with functionalized phosphonates that offer well-defined coordination sites for the MAM
core. Benefits and drawbacks of each of these methodologies are outlined below.

The first approach, by Dr. Yeung-gyo Shin,”88 was the direct intercalation of
solvated M-4-M cores into the native layered phosphate host structure. For example, a
cationic dimolybdenum core ligated by highly labile acetonitrile (M02(CH3CN)34+) can be
introduced into the galleries of reduced vanadium phosphate (N axVOPO4). It was shown
that the phosphates from the LMP displaced the CH3CN ligands and provided the
coordination environment for the multiply bonded bimetallic core, which keyed into the
LMP with the metal-metal bond perpendicular to the layers. This scheme is
fundamentally limited in that the ligand sphere is the host itself, which cannot be
substantially modified. Additionally, phosphate ligands are not suitable for multielectron
photochemistry. In summary the approach is straightforward but cannot be used for
construction of a LIPS because the incorporated complexes cannot be MPCs.

The second approach, the one pursued in this work, was the incorporation of a

separately prepared M-‘LM complex with ancillary ligands containing an uncoordinated

W‘W

:u

 

101

Lewis basic site. Intercalation occurs because of the attraction of the base to the protic
sites in the galleries. The advantage over the previous method is far greater flexibility
with respect to the MAM ligand set, including the possibility of a mixed set. A drawback
is that the periphery of the ligands (or the entire complex) must be sufficiently attractive
to the host to provide an adequate driving force for incorporation. Some of this
disadvantage can be compensated for by host pre-swelling.

”’89 was the coordination of MAM

The third approach, by Dr. Mark Torgerson,
cores with ligands provided from modified phosphate layers. This layered metal
phosphonate (LMPR) approach involves ligands covalently bonded to the host. A
layered metal phosphonate is prepared by using (in principle) almost any organic ligand
attached to the phosphonate. Great ligand variety is possible. A drawback to this
approach is that some of the target ligands may be nontrivial to prepare in substantial
quantity. More limiting is that not all phosphonates readily form highly crystalline hosts
in high yields. Lastly, as there is no control over the exact manner in which the ligands
later coordinate to the MAM cores, characterization of the exact nature of coordination is
difficult at best. It may not be possible to prepare consistent mixed ligand sets around the
metal cores.

Selection of Moz(dmhp)4 as MiM Intercalant

As mentioned in chapter 2, alkyl and aryl amines readily intercalate into
Zr(HPO4)2-H2021'26 to form bilayers. The driving force for intercalation is the formation
of alkylammonium ions. Metal complexes have been introduced within the layers of

LMPs when one or more of the ligands of the complex has an ancillary amine.33 Thus the

strategy for introducing MAM cores into layered phosphates described by Figure 32 can

 

 

102

be developed by including amine or pyridine functionalities directly onto the ligands of
an MJ-M core. The uncoordinated heterocyclic nitrogen of the pyrimidine ligand of
neutral Moz(dmhp)4 (Hdmhp = 2,4-dimethyl-6-hydroxypyrimidine)59 makes this complex
an intercalant of LMPS. The crystal structure of M02(dmhp)4 was first reported in 1979
along with those of Cr and W analogs with the same ligand set.59 No other
characterization was reported. The symmetry of the molecule is D2d with pairs of N and
O eclipsing each other as shown in Figure 33. The asymmetric ligands alternate
orientation, with opposite ligands in the same orientation. The molecule has a quadruple
bond 2.07 A long between the molybdenum d4 atoms. The ligands bind equatorially; the
O and one of the N atoms coordinate in a bridging fashion with alternating orientation.
The uncoordinated N atom is spatially distant from the M-‘LM core, unprotonated, and
may act as a Lewis basic site.
Characterization Of Moz(dmhp)4

In methanol solution M02(dmhp)4 exhibits absorption bands at 493.5 nm (8 = 880
M'lcm'l) and 370 nm (8 = 15200 M'lcm") as shown in Figure 34; absorption bands of
Moz(mhp)4 (mhp = 6-methyl-2-hydroxypyridine) are similar at 495 nm (e = 2450
M'lcm") and 405 nm (8 = 13270 M'lcm'l). Fine structure clearly visible in the lowest
energy absorption corresponds to the well-known 1(52 -—> 55*) transition85 at room
temperature, with a progression of about 350 cm". Such fine structure has been reported
in M02(mhp)4.85 In crystalline form (Figure 35), the bands are at 394 nm and 494 nm.

Excitation of the 52 —> 55* absorption of the compound in MeOH solution leads

to red luminescence (lmax=575 nm) which is comparable in energy and intensity to that

 

 

 

 

103
N \
co A
I ’00 N \ til/+3
OC7M°_CO 1' I diglyme reflu) I 4.0— ,N
00 I / /MO=MO
co N CH ‘0 | ,N I
N\ ’0
Mo(CO)6 + Hdmhp > M02(dmhp)4
Cotton,1979

Figure 33. Synthesis of Moz(dmhp)4.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

104
3 h I I I I I I I I I I I I I I I l I I I I l I I I I I I I I .4
2 5 1 Absorption Spectra of Mozdmhp4 in MeOH j
2 E €
§ 15 :— €
to n a
e I -
8 :. —‘
n 1 _ « i
< : I
. - r
0.5 r ’1 I
0 E i 1?“
-0.5 I- J l l I 1 1 l l L 4 L 1 1 I I L l 1 L 1 l l l I l l l l d
200 300 400 500 600 700 800
nm
1.5 *- l I I I I I I I r I I I I I I I I I I I I I I I I I I _‘
C Absorption Spectra Detail of Mozdmhp4 in MeOH I
8 1 — i
C T 4
(U I- 4
e - I
o _ I
(D " -1
.0 ~ 4
< : 493.5 nm j
0.5 f j
0 l l L 1 l i l l l l l 1 l l l 1 l l 1 t L l l I 1 _‘
440 460 480 500 520 540 560
nm

Figure 34. Absorption spectra of Mozdmhp4 in MeOH.

105

 

80 90 100
TIIIIIIITTIIIIIIIIII
.I

70

60

 

50

IIIIIIfiI

494 nm

 

3194mm

1 1 l l l l l 1 1

l

l

 

L111llllllllllllllllllllllll

 

O l l l l
‘200 400 600 800 1000
nm

Figure 35. Diffuse reflectance spectra of crystalline Mozdmhp4.

1200

 

 

106

 

IIIIIIIIIITIIIITIIIITII—IIIIIIIIIII

)VMAX (crystalline) = 548 nm

  

IllllllllILllllllllllill

 

I

.‘j
II‘ I
’1 III

I

‘-_ -

III IIII IIII IIII IITT IIII
I I I I I

MW "
‘ ' "A' "I' 'MM
‘I

 

 

,,/ I'I -
_ ' IA .1 a
: AEX = 406, 435, 450, 470 nm l a
>— 1 l 1 1 l 1 l 1 l l L L J L L L l l l l 1 l l 1 I L LJ_L 1 1 l 1 1 5

 

450 500 550 600 650 700 750 800
nm

 

fIIIIIIIIIIrTIIIIIIITTIIIIITIIIIII

I

Am (solution) = 577 nm

1111111111111[1111111111111

 

 

   

AEX = 366,406,435 nm

lllllllLLlLJJLLglMLILL114141111ll

450 500 550 600 650 700 750 800
nm

 

IIIT IIIrIIIIIrTTITIIIIIIIIIIIFTTIIIII

 

Figure 36. Emission Spectra of Mozdmhp4 as crystals and in MeOH solution.

Figure 37.

107

 

 

IIIIIIIIIIIIIIIIIIIIIIIII

 

llllllllllllllllllllglll

 

 

 

E Adetection = 550 "m
JL 1 l I l 1— i A L l J i l l l l l l l l l l l l
300 350 400 450 500
nm

Excitation spectrum of crystalline Mozdmhp4.

 

108

observed in Mozmhp4. The steady-state emission spectra of crystalline and dissolved
(MeOH solution) Moz(dmhp)4 are shown in Figure 36; AM are at 548 nm and 577 nm,

respectively. An excitation spectrum of the crystalline form is shown in Figure 37; peaks
occur at 370 nm and 480-500 nm when the detection wavelength is 550 nm.

Time resolved measurements of the emission intensity of crystalline and dissolved
(MeOH solution) M02(dmhp)4 result in monoexponential decays that were fitted to 1.9 ns
and 3.7 ns respectively.

Acid Titration Studies

The presence of basic sites on the ligands provides an opportunity for an acid to
perturb the excited states of Moz(dmhp)4. As the driving force for intercalation will be
attraction of ligand Lewis basic sites to LMP, interactions would likely be present. Such
perturbations could affect the desired photoredox chemistry of the Moz(dmhp)4, or as will
be shown, provide some indication as to the binding of the M02(dmbp)4 in the galleries.
Early tests of M02(dmhp)4 solutions acidified with HCl solution exhibited a yellow-
orange color that turned red. For these reasons certain properties of the compound were
examined in free solution as a function of equivalents of acid titrated.

Experimental- Titrations

Three titration experiments were done; these will be referred to as Titrations I, II,
and III. For Titrations I and II, the emission spectra were measured as a function of acid
titrated into the methanol solution. Both HCl and trifluoromethanesulfonic (triflic) acid
were used with concentrations of 1.2 M and 1.0 M in methanol, respectively. 100 mL of
1.4x10‘4 M Moz(dmhp)4 methanolic solution was first prepared. Acid was added one

equivalent at a time with a 10 uL syringe with stirring. One equivalent is defined here as

 

 

 

 

109

one Lewis basic site; therefore four equivalents are required to fully protonate the
Moz(dmhp)4 molecule if a high association constant is assumed. The absorption and
emission spectra were recorded at each point. Titration III was a triflic acid titration in
acetonitrile solution. Picosecond emission lifetime measurements were made only for
Titration I; the solution in the septa-covered cuvette was degassed by bubbling with
argon for 10 minutes.

Results

Emission, Absorption

In Titration I, the luminescence is significantly red-shifted and attenuated upon
addition of HCl equivalents as shown in Figure 38. The Max and integrated peak areas
are summarized in Table 4. The normalized quantum yield (defined as equal to one for
the unprotonated complex) consistently decreased, even when more than the
stoichiometric four equivalents of acid are added. Both UV-Vis bands shifted to the red
with no isosbestic points.

For Titration II, the absorbance spectrum also red shifted with no isosbestic
points. The emission spectrum in Figure 39 was different from the former in that the A...“
quickly red shifted to a definite limit of about 600 run while decreasing in intensity, then
substantially increased in intensity while retaining a km“ = 600 nm. The 24m and
integrated peak areas are summarized in Table 5; note the normalized quantum yield

initially decreased before substantially increasing to nearly original values.

 

 

 

110

 

I

I r I F I I

T

T

 

Figure 38. Emission Spectra of Mozdmhp4 in MeOH solution titrated with HCl.

 

Titration I

 

 

700

750

 

 

  

 

 
 

 

   

fl I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I
— -I
t Titration ll 1 -
._ IA", ..
— 1' ".‘ _
_ I '\ _
I— . ‘ —4
~-

L I \ a
_ - ‘, 4
__ I "

. ‘5 —l
1— ‘ -I

I b
b- . ‘ -1
)— ’ JR —4
I— I t —
I— “ .1
b- , |\' —4 I
p ’ K :
r- . ‘1 fl
_ , , _ I
r 5 'v a _.i
D— I .'¢'O'\.‘I “ j l
_ ' ‘q -1
a J

1 1 1 1 l 1 1 1 1 l 1 1 1 1 L 1 M 1 J 1 1 1 L I L 1 1 L l 1 1 1 1

 

 

 

450 500 550 600 650 700 750 800
nm

Figure 39. Emission Spectra of Mozdmhp4 in MeOH solution titrated with triflic acid.

 

112

 

FITIIIIIIIIIIIIIIIIIIIT—FTTI'TIiIIIII

II }
Titration III 1" 3L 3

 

 

I I I I I I I I I I I I I I I I I' r I I I I I r I I I I I

 

 

lllllllllllllllIIIIlIIlIlllllllllJ

450 500 550 600 650 700 750 800
nm

Figure 40. Emission Spectra of Mozdmhp4 in CH3CN solution titrated with triflic acid.

 

113

Titration 111 (Figure 40) was qualitatively similar to II, as the emission spectrum
lam rapidly reached a very different limit of 670 nm, remaining there for many
equivalents. The intensities (Table 6) rose steadily and then suddenly fell after about four
equivalents of acid were added.

Lifetimes

The results for Titration I are summarized in Table 4. The lifetime is initially
monoexponential at 3.7 ns. One equivalent of acid titrated into the solution yielded a
new fast (<70 ps) decay component. Additional acid equivalents resulted in results best
fitted by additional decay components; a fast component (40-60 ps) and a slow
component (1.4-4.4 us) were always present.

Discussion of Titrations

The absence of isosbestic points in Titration I and 11’s absorbance spectra
indicates more than two species are present in each. These species may be either the
complex at differing protonation levels or significantly different species.

Comparison of the emission spectra of Titrations II and III demonstrates a solvent
dependence. One possibility is the MeOH acting as a base for the triflic acid, but this is
not likely as the uncoordinated nitrogen on the ligands should be more basic than the O
atom in MeOH. More plausible is the MeOH solvent associating with the Moz(dmhp)4
complex far more significantly than CH3CN; the MeOH may thus interfere with the triflic
acid binding whereas the acetonitrile does not. Thus the Titration III emission spectra
differences may solely reflect the results of protonation of the complex. The Titration III

emission maxima (around 670 nm) will reappear later in a specific LMP system.

I1". .- . £9 ‘1 am. -'
i;
f

 

 

 

 

114

Table 4. Mozdmhp4 Titration I data (HCl in MeOH solution).

 

 

 

 

 

 

 

 

 

 

Equiv. acid KEN. Max Abs at integrated relative ps
added 435nm EM peak quantum lifetime
area yield

0 575 0.10 71754 1 Q 3.7 ns 100 %
1 580 0.22 92724 0.59 Q 66 ps 2.6 %
4.4 ns 97.4 %
2 620 0.42 72915 0.24 Q 53 ps 15 %
379 ps 55 %
1.4 ns 30 %
3 625 0.61 57415 0.13 Q 63 ps 15 %
444 ps 70 %
1.6 ns 15 %
4 650 0.74 37367 0.070 Q 58 psi 9 %
438 ps 74 %
2.2 ns 18 %
5 675 0.80 31796 0.055 Q 45 ps 7 %
410 ps 60 %
2.2 ns 32 %
10 695 0.83 14667 0.025 Q 40 ps 31 %
298 ps 32 %
1.4 ns 37 %

20 700+ 0.79 7600 0.013 Q

 

 

 

 

 

 

Table 5. Mozdmhp4 Titration II data (triflic acid in MeOH solution).

 

 

 

 

 

 

Equiv. 25M Max Abs at integrated relative
acid 435nm EM peak quantum
added area yield
0 572 0.15 23700 1 Q

1 576 0.54 10971 0.13 Q

2 597 0.84 9706 0.073 Q

3 601 0.98 14237 0.092 Q

4 598 0.98 24510 0.16 Q

5 597 0.98 36470 0.24 Q
10 593 0.70 93485 0.85 Q

 

 

"“1 “-

 

 

115

Table 6. Mozdmhp4 Titration 111 data (triflic acid in CH3CN solution).

 

 

 

Equiv. 2.5M Max EM peak
acid area

added
0 570 22520
1 661 57788
2 670 337868
3 667 760178
4 667 516491
5 657 41202

 

 

 

Fits of the Titration I time resolved emission data rapidly descend into a
multiexponential morass, precluding meaningful interpretation beyond the inferred
existence of many possible compounds and states. It is likely that in Titration I the
chloride ion (and some hydroxide ions) undergo chemical reactions such as ligand

substitution on the complex.

Incorporation of M02(dmhp)4 into LMP Layers

The results of the above titration experiments help in the interpretation of the
following results. This proton dependence of the Moz(dmhp)4 luminescence provides a
convenient experimental handle in the study of the intercalation chemistry of this
compound with LMPs. As with the ReOzpy4+ and Ru(bpy)32+ systems, intercalation was
not accomplished without swelling the layered phosphate. Both ethanol and amine
swelling methodologies were used. As Moz(dmhp)4 is a neutral compound, it does not
have a large electrostatic driving force for amine intercalation; consequently it is more

difficult to introduce this compound than others into the galleries at high loadings.

 

 

 

116

Procedures

Incorporation with EtOH-swelled ZrP

The ethanol method uses ZrP swelled with ethanol. As ethanol is readily and
irreversibly removed by evaporation at RT, any species between the sheets will be
trapped in “bare” ZrP layers. This useful property only exists with ZrP, as TiP and SnP
do not have a stable EtOH swelled phase. In a typical incorporation, Moz(dmhp)4 is
dissolved in EtOH, and EtOH-ZrP paste is added. Sometimes a drop or two of acid is
added to the solution with the rationale of protonating the Moz(dmhp)4 so it will attach to
the anionic sheets more easily. The product is vacuum filtered, washed well with solvent,
and dried.

Incorporation with amine swelled LMPs

In a typical incorporation, Moz(dmhp)4 is dissolved in MeOH or EtOH, an amine-
LMP powder is added, and the mixture is stirred for hours. The product is vacuum
filtered, washed well with solvent, and dried. As indicated earlier, a difficulty is the
absence of an overwhelming driving force to promote intercalation of neutral
Moz(dmhp)4. This same feature can result in some deintercalation upon washing.
Consequently reproducibility with Moz(dmhp)4 is difficult. Studies on TOA-ZrP,
TBA-ZrP, TBA-TiP, and TBA-SnP will be reported.

Characterization

Powder XRD

The EtOH-ZrP intercalate has a (1001 spacing of ~14 A, which significantly
contracts to 10 A upon reaction with acidified Moz(dmhp)4. A typical trace is in Figure

41. The amine pre-treated LMPs turn pale yellow upon reaction with Moz(dmhp)4 and

 

117

the final product exhibits a slightly contracted d-spacing of 17.9 A for TOA-ZrP. Other
amine swelled LMPs behave in an analogous manner, but with lesser spacings.

Absorption, Emission, ps Lifetime

The diffuse reflectance spectrum of acidified M02(dmhp)4 intercalated EtOH-ZrP
is shown in Figure 42. Figure 43 displays traces of amine swelled LMPs, with the data
summarized in Table 7.

Under illumination the M02(dmhp)4 intercalated EtOH-ZrP is emissive. Figure 44
shows spectra with maxima at about 670 nm. The M02(dmhp)4 emission inside amine
swelled phosphates is shown in Figure 45, and the emission maxima are listed in Table 7.
Emission was not detected for TBA-TiP; only scattered light was detected by the
instrument. The emission lifetimes for M02(dmhp)4 inside various amine swelled

phosphates are compiled in Table 7.

 

 

 

118

 

VIII IIII IIII IIII IIII IIIT IIII IIIr
I I I I I I I

9.9 A

 

 

lllillllilllliliilillllillll

IITIIITIIIIIIIIIIIIIIIIIIIIII

   

 

 

 

JlLlll

0 5 10 15 20 25 30 35 4O
2 theta

Figure 41. XRD spectrum of Mozdmhp4 in EtOH-ZrP.

 

 

100 I I I I I T I I I I I I I I I I I I I

   

.—
y—
u-
80—
b.—
b
—

O)
O

   

% Reflectance

 

J l l i l l l l l l 1 i l l l i 1

 

 

 

4O —
- 580 nm
“ 473 nm
20 r
O I- l I 1 i l l l I L l 1 L l 1 l i l I l
200 400 600 800 1000 1200
nm

Figure 42. Diffuse reflectance spectra of Mozdmhp4 in EtOH—ZrP.

 

119

 

‘7 .7” IE4 “Jk‘fi/‘MIM NA A M W“

  
 

 

,. . , )‘vMAX (TBA-SnP) = 460-470 nm
I/‘x/“I‘xf/ 1W (TOA-ZrP) = 392 nm, 493 nm
I;
I
I;
I
/

lMAx (T BA-ZrP) = 397 nm, 493 nm

 

 

 

 

200 400 600 800 1000 1200

nm

Figure 43. Diffuse Reflectance of Mozdmhp4 in TBA-LMP.

Table 7. Absorption and Emission Data Summary for Moz(dmhp)4.

 

 

 

 

 

 

 

 

 

 

 

 

 

Mus Max Mus Max Mm max ps lifetime

TOA-ZrP 392 493 560 4.1 ns 100 %
TBA-ZrP 397 493 550 4.2 ns 100 %
TBA-TiP ----- ---- ---- 800 ps 65 %

4.7 ns 35 %

TBA-SnP ----- 460-470 ? 540 1.8 ns 79 %

7.6 ns 21 %

EtOH-ZrP 473 580 668 ~ I
Mozdmhp4 crystals 394 494 548 1.9 ns 100 %
Mozdmhm solution 370 493 577 3.7 ns 100 °/o

 

 

 

 

 

120

 

    
   

 

     
 
     
 

 

I T I r I I I I I I I I I fT r I f I I I I I I I I I I I I I I I I
F 1".IUVI ‘
* )‘vMAX = 668 nm .e L. -
>— " .4 -1
’ u
_ v 5“ —
I— 'v —1
'~

'— ’ \3‘ —1
g ‘ s
r- ~. ‘
.93 ~ ,: K", -

c , .
. . _.
E‘ r fa ‘

On
9 P g "
1‘: r {I . -
e _ I: —
< P .. . 7
.. " {J 25:
."‘.or:,":
u- v “1.:
_ Max = 406, 435 nm
- x JurLu‘i‘.‘ - r I J 1 l 1 1 1 1 I 1 1 l 1 I 1 I 1 I

 

 

450 500 550 600 650 700 750 800
nm

Figure 44. Emission Spectra of Mozdmhp4 in EtOH-ZrP.

121

 

I I I I I I I I I I I I I I I I I I I I I I I I I I

IMAX '(TOA-ZrPI = 560 mi

KMAX (TBA-ZrP ‘

.50 nm
II 412%
WWW fI VIVIW

I “I

III IMMWI KMAx(TBA-SnP I 40 nm

I

 

IW
M MIA/\-
/ InfuI‘WIIIII

M/ AMAX(TBA-TiP) =too weak M M WW
...IIII..I.IIII..I I‘l'flii

500 550 600

Wm
WW II I

 

 

IIII I III III

 

 

nm
Max = 450 nm

Figure 45. Emission Spectra of Mozdmhp4 in amine-swelled LMPs.

122

Discussion

Problems

There were many difficulties associated with these experiments. Specifically, the
intercalation behavior of M02(dmhp)4 (particularly in EtOH-ZrP) was not always
reproducible. Two forms were noted; the data are supplied for one; properties for the
other are similar to those of the amine swelled systems. The first form normally required
inclusion of acid in the incorporation procedure. The added acid on occasion
decomposed the Moz(dmhp)4.

The small driving force associated with intercalating a neutral complex into an
amine swelled LMP also makes low loadings common and deintercalation possible. The
emission from Moz(dmhp)4 is weak and the white LMP layers scatter light, creating
artifacts in the emission spectra.

Inferred Orientation(s)

The short interlayer spacing of the ethanol pre-treated material necessarily
demands that the M02(dmhp)4 complex be tilted end-on with its metal-metal axis normal
to the plane of the layers; in contrast, accommodation of Moz(dmhp)4 with the Moi-Mo
core parallel to the layers would require a d-spacing of ~1 3 A with keying into the lattice
and ~15 A without keying. The highly protonated Moz(dmhp)4 would have a substantial
positive charge, increasing its attraction to the anionic layers, which could provide the
driving force for such tight packing. Further solidifying the case for intercalation,
elemental analysis of a Moz(dmhp)4 / EtOH-ZrP sample revealed a Mo/Zr mole ratio of

0.4; such high loading cannot be explained by Mo species only attached on the surface.

123

However, the orientation of the M02(dmhp)4 in the gallery is not known for any of
the amine swelled LMPs, as the greater interlayer spacing permits greater freedom of
positioning.

Absorption, Emission

The diffuse reflectance data for the amine systems exhibit bands characteristic of
crystalline M02(dmhp)4. In contrast, the EtOH-ZrP system’s bands (Figure 44) are red-
shifted 80-90 nm.

Under illumination OLEX = 450 nm) most solids are emissive. For the amine
swelled LMPs, Figure 45 shows that the emission arises mostly from the deprotonated
form of Moz(dmhp)4, which typically has KMAx = 550-580 nm as noted earlier in the
chapter. This behavior is not too surprising inasmuch as pre-treatment of LMP with TOA
or TBA renders the proton unavailable for reaction with the Moz(dmhp)4 by using it to
H-bond to the LMP. The TBA-TiP emission was not conclusively detected by the
emission instrument, but the picosecond time resolved instrument was able to detect a
decay. The reason for the weak emission is probably the presence of small quantities of
complex, as shown by the nearly flat absorbance line in the region where the Moz(dmhp)4
bands are expected to be.

The EtOH-ZrP intercalated product is luminescent, and as shown in Figure 44, the
emission spectrum (KM/xx = 668 nm) is consistent with a Moz(dmhp)4 complex whose
pyrimidine ligands are protonated as in Titration III. This observation likely indicates
that (at least) some protons are exchanged on the ligands’ basic sites. Note that the
protonation occurs either before or during incorporation as it is necessary to add acid to

prepare these samples. The presence of emission maxima at 668 nm, similar to those

 

 

 

124

observed for Titration III, despite use of conditions from Titration 1, supports the simple
protonation hypothesis. The red shift indicates that protonation has stabilized the
Moz(dmhp)4 excited state.

More specifically, if in free solution (Titrations I, II) in addition to the protons the
chloride, hydroxide, and/or methanol significantly associated with the M02(dmhp)4
complex, the resulting perturbation of the excited state could significantly change the
emission spectra. This reasonably presumes that the triflate and acetonitrile
(Titration III) are effectively noncoordinating and that bound protons would cause the
emission to shifi 100 nm to the red. A possible reason for the intercalation behavior of
the EtOH-ZrP host is that anionic ZrP layers will not associate with the chloride,
hydroxide, and methanol, but will be strongly attracted to the protonated (now cationic)
complex and readily tightly sandwich it because of electrostatic forces, in the process
forcing away any weakly associated chloride, hydroxide, and methanol ligands. The end
result would be a protonated Moz(dmhp)4 complex inside simple ZrP layers.

The difficulty in consistently reproducing this type of sample indicates the
properties are dependent upon the intercalation specifics, consistent with the above
model.

Lifetime

Unfortunately no lifetime data exist for any red-shifted M02(dmhp)4 inside
EtOH-ZrP. The weak emission of the TBA-TiP and TBA-SnP products was detected
with the ps lifetime apparatus. The monoexponential lifetimes for the amine swelled ZrP

products were almost the same as each other (4.1 ns and 4.2 ns) and close to that of

l I
‘

 

 

 

125

crystalline Moz(dmhp)4 (3.7 ns). An excited state of an M-4—M compound can thus be

maintained in a LMP system as the lifetimes are not substantially attenuated.

Conclusions

Of the driving forces for incorporation reactions, an acid-base association strategy
is substantially weaker than an electrostatic one. In the EtOH-ZrP incorporation case,
high loading was obtained because the complex was protonated (receiving positive
charge) before exposure to the layers, so the intercalation was driven by electrostatic
forces. The amine swelled cases relied on entropy, some association with the amine tails,
and possible acid/base attractions to gain intercalation, but at lower loading levels. These
observations suggest that a LIPS might be easier to assemble for an anionic layer if the
MPC were cationic.

The data suggest that the protons in these amine swelled LMPs are localized and
do not readily protonate guests as the amines efl'ectively bind up layer protons.

The retention of emissive lifetimes and properties suggests that the LMP is not
quenching the excited state of the M-‘LM. Lack of such quenching, as was reported in
Chapter 3, indicates promise for the development of LIPS photochemistry.

The M02(dmhp)4 complex is oriented with the MAM axis perpendicular to the
layers in the EtOH-ZrP material; no conclusions can be drawn about the orientation in the
other individual materials. The tight packing of this M-i-M material would create
problems for a practical MPC for two reasons. First, reactant and product diffusion
would become problematic. Second, highly sterically constrained locations would tend

to inhibit any necessary MPC ligand rearrangements.

 

 

 

126

Significance of Results with Respect to LIPS Construction

The conclusions of Chapters 3, 4, and 5 in this work address a number of the
issues related to constructing a Layered Integrated Photochemical System using M—4—M
complexes and LMP's as components. Despite several encouraging findings, significant
progress must be made in several areas for a LIPS to be viable.

The discovery that three different types of coordination complexes intercalate into
LMPs while retaining their excited state properties is significant. This apparent
generality suggests that the identity of the complexes is not important as long as they are
chemically compatible with the host LMP. As the LMP does not generally quench the
excited state, the LMP is not involved in energy transfer with the complex, so a LIPS is
possible. Such a deactivation would preclude a priori any MPC operation.

Ability to preferentially promote or hinder two electron vs. one electron reactions
would be important. Chapter Four's indication of possible excited state deactivation by
single electron transfer to TiP is encouraging. The possibility of two electron redox
reactions coupling with SnP and the resultant tuning capability by LMP variation could
be powerful tools if the observed phenomena are correctly interpreted and general. Such
synergy between the LMP and MPC electronic states would be vital.

The tight binding of IF by amines at the LMP / amine interface simplifies the
understanding of chemistry in the galleries of amine swelled LMP's. This binding should
be checked for compatibility with the redox chemistry desired in the galleries.

The incorporation strategy used will need to be considered. The intuitive general

finding that the electrostatic method has a much stronger driving force than the acid /

 

 

127

base method was exemplified by the discovery that the cationic MLM MPC model
compound was far easier to incorporate in significant quantity. A drawback to the
electrostatic incorporation strategy is that the highly positive charge on MPC can cause
tight packing of the complex in the LMP galleries. Tight packing generates significant
problems with respect to diffusion of substrate in and out of layers; i.e. mass transport of
reactants and products becomes impractically slow. Additionally, the MAM MPC
described herein requires the ability of the ligand set to rearrange for the photochemistry
to occur.'7 Tight packing creating steric hindrances might thus prevent MPC operation
due to lack of ligand rearrangement. Overcoming the difficulties posed by tightly packed
components may require changes in strategy, such as pillaring the LMP's, using layered
metal phosphonates as hosts, coating a porous support with a thin layer of a LMP, or
abandoning LMP's altogether. Unfortunately each of these alternatives would introduce
new problems.

Another fundamental weakness of constructing a LIPS using an MAM as an MPC
is thermodynamic. If a 1(82 —9 58*) transition in an M—4—M complex falls at 500 nm,
2.48 eV of energy is absorbed per photon. This energy, when divided between two redox
equivalents, is merely 1.24 eV each. When allowance is made for inevitable energy
losses at each step of the catalytic cycle, potentially much less than 1 eV of energy will
be available to do useful redox chemistry. Such low levels of energy (1 eV 5 1240 nm)
would provide access to very few redox reactions of catalytic interest. Additionally,
difficulty in extending the M—4—M multielectron photochemistry in solution (and

regenerating the starting complex) in the Nocera laboratories does not bode well for it

It} 235. firm—3'

i

 

 

128

becoming a viable MPC. For a LIPS to be successfully constructed, an alternative MPC

will need to be developed.

 

 

LIST OF REFERENCES

 

 

LIST OF REFERENCES

' Kavamos, (3.1.; Turro, MJ. Chem. Rev. 1986, 86, 401.

N

(a) Meyer, T.J. Acc. Chem. Res. 1989, 22, 164. (b) Photoinduced Electron Transfer
11; Mattay, J ., Ed.; Springer-Verlag: New York, 1990. (c) Vogler, A.; Kunkely,
H. Comm. Inorg. Chem. 1990, 9, 210. (d) Gust, D.; Moore, T.A. In
Supramolecular Photochemistry; Balzani, V., Ed.; Reidel: Boston, 1987.
(e) Wasielewski, M.R. In Photoinduced Electron Transfer; Fox, M.A., Chanon,
M., Eds; Elsevier: Amsterdam, 1988, Part A, p161.

(a) Babcock, G.T. In New Comprehensive Biochemistry: Photosynthesis; Amesz, J .,
Ed.; Elseiver: Amsterdam, 1987. (b) Allen, J.P.; Feher, G.; Yeates, T.O.; Rees,
D.C.; Deisenhofer, J .; Michel, H.; Huber, R. Proc. Natl. Acad. Sci. USA 1986, 83,
8589. (c) Christou, G. Acc. Chem. Res. 1989, 22, 328. (d) Kirmaier, C.; Holten,
D. In The Photosynthetic Bacterial Reaction C enter-Structure and Dynamics;
Breton, J ., Vermeglio, A. Eds.; Plenum: New York, 1988; p 219.

3

Meyer, T. J. Acc. Chem. Res. 1989, 22, 164.
5 Duesing, R.; Tapolsky, G.; Meyer, T. J. J. Am. Chem. Soc. 1990, 112, 5378.
6 Perkins, T. A.; Humer, W.; Netzel, T. L.; Schanze, K. S. J. Phys. Chem. 1990, 94, 2229

7 Schmehl, R. H.; Ryu, C. K.; Elliott, C. M.; Headford, C. L. E.; Ferrere, S. In Electron
Transfer in Biology and the Solid State; Johnson, M. K.; King, R. B.; Kurtz, Jr.,
D. M.; Kutal, C.; Norton, M. L.; Scott, R. A., Eds.; Advances in Chemistry Series
No. 226; American Chemical Society: Washington, DC, 1990; p 211.

W

Photoinduced Electron Transfer 11; Mattay, J ., Ed.; Springer-Verlag: New York, 1990.

9 Gust, D.; Moore, T. A.; Moore, A. L.; Gao, F.; Luttrull, D.; DeGraziano, J. M.; Ma, X.
C.; Making, L. R.; Lee, S.-J.; Trier, T. T.; Bittersmann, E.; Seely, G. R.;
Woodward, S.; Bensasson, R. V.; Rougée, M.; De Schryver, F. C.; Van der
Auweraer, M. J. Am. Chem. Soc. 1991, 113, 3638.

'0 Wasielewski, M. R.; Gaines 111, G. L.; O’Neil, M. P.; Svec, W. A.; Niemczyk, M. P. J.
Am. Chem. Soc. 1990, 112, 4559.

130

 

131

11 Gust, D.; Moore, T.A.; Moore, A.L.; Barrett, D.; Makings, L.R.; Liddell, P.A.;
DeSchryver, F.C.; Van der Auweraer, M.; Benssasson, R.V.; Rougee, M.J. J. Am.
Chem. Soc. 1988, 110, 321.

‘2 Rarnamurthy, V. In Photochemistry in Organized and Constrained Media;
Ramamurthy, V., Ed.; VCH: New York, 1991; Chapter 10.

Gratzel, M. Heterogeneous Photochemical Electron Transfer; CRC Press: Boca
Raton, FL, 1989.

Kalyanasundaram, K. Photochemistry in Microheterogeneous Systems; Academic
Press: Orlando, FL, 1987; Chapter 10.

'5 Persaud, L.; Bard, A. J.; Campion, A.; Fox, M. A.; Mallouk, T. E.; Webber, S. E.;
White, J. M. J. Am. Chem. Soc. 1987, 109, 7309.

‘6 Krueger, J. S.; Mayer, J. 13.; Mallouk, T. E. J. Am. Chem. Soc. 1988, 110, 8232.
'7 Nocera, D.G. Acc. Chem. Res. 1995, 28, 209, and references therein.

'8 Newsham, M. D.; Giannelis, E. P.; Pinnavaia, T. J.; Nocera, D. G. J. Am. Chem. Soc.
1988, 110, 3885.

'9 King, R. D.; Nocera, D. G.; Pinnavaia, T. J. J. Electroanal. Chem. 1987, 236, 43.
2° Dulebohn, J. 1.; Ward, D. L.; Nocera, D. G. Polyhedron 1991, 10, 2813.

2' Alberti, G.; Costatino, U. In Intercalation Chemistry; Whittingham, M.S., Jacobson,
A.J., Eds.; Academic Press: New York, 1982.

22 Clearfield, A.; Nancollas, G.H.; Blessing, R.H. In [an Exchange and Solvent
Extraction; vol. 5; Marinsky, J .A., Marcus, Y., Eds.; Marcel Dekker: New York,
1973; PP 2-120.

23 Alberti, G.; Costantino, U.; Marmottini, F. In Recent Developments in Ion Exchange;

Williams, P.A., Hudson, M.J., Eds; Elsevier: London, 1987; pp 233-256.
2“ Clearfield, A. Chem. Rev. 1988, 88, 125.
25 Clearfield, A. Comm. Inorg. Chem. 1990, 10, 89.

26 Clearfield A. In Inorganic Ion-Exchange Materials; Clearfield, A., Ed.; CRC Press:
Boca Raton, FL 1982.

27 Ladwig, G. Z. Anorg. Allg. Chem. 1965, 338, 266.

 

28

29

30

31

32

33

34

35

36

37

38

39

132

Garcia-Ponce, A.L.; Moreno-Real, L.; Jimenez-Lopez, A. Inorg. Chem. 1988, 27,
3372.

Dines, M.B.; DiGiacomo, P.M. Inorg. Chem. 1981, 20, 92-97.
Burwell, D.A.; Thompson, M.E. Chem. Mater. 1991, 3, 730.
Clearfield, A.; Smith, G.D. Inorg. Chem. 1969, 8, 431.

Costantino, U. J. Chem. Soc., Dalton Trans. 1979, 402.

Lee, C.F.; Thompson, M.E. Inorg. Chem. 1991, 30, 4.

Kaschak, D.M.; Mallouk, T.E. J. Amer. Chem. Soc. 1996, 118, 4222.
Clearfield, A.; Tindwa, R.M. Inorg. Nucl. Chem. Lett. 1979, 15, 251.
Hunsberger, L.R.; Ellis, A.B. Coord. Chem. Rev. 1990, 97, 209.
Olken, M.M.; Verschoor, C.M.; Ellis, A.B.1norg. Chem. 1986, 25, 80.
Rosenthal, G.L.; Caruso, J. J. Sol. State Chem. 1991, 93, 128.

Olken, M.M.; Ellis, A.B. J. Am. Chem. Soc. 1984, 106, 7468.

40 McCarthy, T.J.; Ngeyi, S.-P.; Liao, J.-H.; DeGroot, D.; Hogan, T.; Kannewurf, CR;

41

42

43

44

45

46

47

48

Kanatzidis, M.G. Chem. Mater. 1993, 5, 331.
Newsham, M. Ph.D. Dissertation, Michigan State University, East Lansing, MI, 1988.
Bowman, L.E.; Berglund, K.A.; Nocera, D.G. Rev. Sci. Instr., 1993, 64, 338.
Zaleski, J. Ph.D. Dissertation, Michigan State University, East Lansing, MI, 1993.
Mussel], R.D.; Nocera, D.G. J. Am. Chem. Soc. 1988, 110, 2764.

Benhamza, H.; Barboux, P.; Bouhaouss, A.; Josien, F.; Livage, J. J. Mater. Chem.
1991, 1, 681.

Troup, J .M.; Clearfield, A. Inorg. Chem. 1977, 16, 3311.

Alberti, G.; Cardini-Galli, P.; Costantino, U.; Torracca, E. J. Inorg. Nucl. Chem. 1967,
29, 571.

Rodriguez-Castellon, E.; Rodriguez-Garcia, A.; Bruque, S. Inorg. Chem. 1985, 24,
1187.

 

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

65

66

67

68

133

Yamanaka, S.; Horibe, Y.; Tanaka, M. J. Inorg. Nucl. Chem. 1976, 38, 323.
Alberti, G.; Casciola, M.; Costantino, U. J. Colloid Interface Sci. 1985, 107, 256.
Alberti, G. Acc. Chem. Res. 1978, 11, 163.

Beard, J. H.; Casey, J.; Murrnann, R. K. Inorg. Chem. 1965, 4, 797.

Brewer, J. C.; Gray, H. B. Inorg. Chem. 1989, 28, 3334.

Ram, M. S.; Hupp, J. T. Inorg. Chem. 1991, 30, 130.

Ciani, G. F.; D'Alfonso, G.; Romiti, P. F.; Sironi, A.; Freni, M. Inorganica Chimica
Acta, 1983, 72, 29.

F reni, M.; Giusto, D.; Romiti, P.; Minghetti, G. Gazz. Chim. Ital. 1969, 99, 286.

Sprintschnik, G.; Sprintschnik, H.W.; Kirsch, P.P.; Whitten, D.G. J. Amer. Chem. Soc.
1977, 99, 4947.

Inskeep, R.G. J. Inorg. Nucl. Chem. 1962, 24, 763.

Cotton, F. A.; Niswander, R. H.; Sekutowski, J. C. Inorg. Chem. 1979, I8, 1152.
Lock, C. J. L.; Turner, G. Acta Cryst. 1978, B34, 923.

Calvo, C.; Krishnamachari, N.; Lock, C. J. L. J. Cryst. Mol. Struct. 1971, 1, 161.

Helm, L.; Deutsch, K.; Deutsch, E. A.; Merbach, A. E. Helv. Chim. Acta 1992, 75,
210.

Winkler, J. R.; Gray, H. B. Inorg. Chem. 1985, 24, 346.
Johnson, C. S.; Mottley, C.; Hupp, J. T.; Danzer, G. D. Inorg. Chem. 1992, 31, 5143.
Pipes, D. W.; Meyer, T. J. Inorg. Chem. 1986, 25, 3256.

Thorp, H. H.; Kumar, C. V.; Turro, N. J.; Gray, H. B. J. Am. Chem. Soc. 1989, 111,
4364.

Ram, M. S.; Jones, L. M.; Ward, H. J .; Wong, Y.-H.; Johnson, C. S.; Subramanian, P.;
Hupp, J. T. Inorg. Chem. 1991, 30, 2928.

Ram, M. S.; Johnson, C. S.; Blackboum, R. L.; Hupp, J. T. Inorg. Chem. 1990, 29,
238.

 

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

134

Blackboum, R. L.; Jones, L. M.; Ram, M. S.; Sabat, M.; Hupp, J. T. Inorg. Chem.
1990, 29, 1791.

Jones-Skeens, L. M.; Zhang, X. L.; Hupp, J. T. Inorg. Chem. 1992, 31, 3879.
Liu, W.; Welch, T. W.; Thorp, H. H. Inorg. Chem. 1992, 31, 4044.

Liu, W.; Thorp, H. H. Inorg. Chem. 1994, 33, 1026.

Pipes, D. W.; Meyer, T. J. J. Am. Chem. Soc. 1985, 107, 7201.

Thorp, H. H.; Van Houten, J.; Gray, H. B. Inorg. Chem. 1989, 28, 889.

Kalyanasundaram, K. Photochemistry of Polypyridine and Porphyrin Complexes;
Academic Press: London, 1992.

Ghosh, P.K.; Bard, A]. J. Phys. Chem. 1984, 88, 5519.

(a) Krenske, D.; Abdo, S.; Van Damme, H.; Cruz, M.; Fripiat, J.J. J. Phys. Chem.
1980, 84, 2447. (b) Abdo, S.; Canesson, P.; Cruz, M.; Fripiat, J .J .; Van Damme,
H. J. Phys. Chem. 1981, 85, 797. (c) DellaGuardia, R.A.; Thomas, J.K. J. Phys.
Chem. 1983, 87, 990. (d) Schoonheydt, R.A.; De Pauw, P.; Vliers, D.; De
Schrijver, F.C. J. Phys. Chem. 1984, 88, 5113. (e) Nakamura, T.; Thomas, J.K.
Langmuir, 1985, 1, 568. (f) Joshi, V.; Kotkar, D.; Ghosh, P.K.; J. Am. Chem.
Soc. 1986, 108, 4650. (g) Giannelis, E.P., Nocera, D.G., Pinnavaia, T.J. Inorg.
Chem. 1987, 26, 203. (h) Joshi, V.; Ghosh, P.K.; J. Chem. Soc., Chem. Commun.
1987, 789. (i) Kuykendall, V., Thomas, J.K. Langmuir, 1990, 6, 1346.

Yeates, R.C.; Kuznicki, S.M.; Lloyd, L.B.; Eyring, E.M. J. Inorg. Nucl. Chem. 1981,
43, 2355.

Vliers, D.P.; Schoonheydt, R.A.; De Schrijver, F .C. J. Chem. Soc., Faraday Trans. 1,
1985, 81, 2009.

Vliers, D.P.; Collin, D.; Schoonheydt, R.A.; De Schryver, F.C. Langmuir 1986, 2,
165.

Colon, J .L.; Yang, C.-Y.; Clearfield, A.; Martin, C.R. J. Phys. Chem. 1988, 92, 5777.
Colon, J .L.; Yang, C.-Y.; Clearfield, A.; Martin, C.R. J. Phys. Chem. 1990, 94, 874.
Kumar, C.V., Williams, Z.J. J. Phys. Chem. 1995, 99, 17632.

Kumar, C.V., Chudhari, A., Rosenthal, G.L. J. Am. Chem. Soc. 1994, 116, 403.

Manning, M.C.; Trogler, W.C. J. Am. Chem. Soc. 1983, 105, 5311-5320.

 

135

86 Engebretson, D.S.; Zaleski, J.M.; Leroi, 0.13.; Nocera, D.G. Science 1994, 265, 759.
87 Shin, Y.K.; Nocera, D.G. J. Am. Chem. Soc. 1992, 114, 1264.

88 Shin, Y.K.; Saari, E.A.; Torgerson, M.R.; Nocera, D.G. Selectivity in Catalysis; Davis,
M., Suib, S., Eds; ACS Symp. Ser. 1993, 517, 244.
89 Torgerson, M.R. Ph.D. Dissertation, Michigan State University, East Lansing, MI,
1996.

 

 

 

"‘1611111111111“