Citation
Synthetic Development of C3-Symmetric Triphenoxymethane-Based Reagents for Selective Recognition and Sequestration of Lanthanides and Actinides

Material Information

Title:
Synthetic Development of C3-Symmetric Triphenoxymethane-Based Reagents for Selective Recognition and Sequestration of Lanthanides and Actinides
Copyright Date:
2008

Subjects

Subjects / Keywords:
Atoms ( jstor )
Chelates ( jstor )
Ethers ( jstor )
Extraction ( jstor )
Ions ( jstor )
Ligands ( jstor )
Metal ions ( jstor )
Nitrates ( jstor )
Oxygen ( jstor )
Solvents ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Embargo Date:
7/24/2006

Downloads

This item has the following downloads:


Full Text











SYNTHETIC DEVELOPMENT OF C3-SYMMETRIC TRIPHENOXYMETHANE
BASED REAGENTS FOR THE SELECTIVE RECOGNITION AND
SEQUESTRATION OF LANTHANIDES AND ACTINIDES














By

KORNELIA K. MATLOKA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2006

































Copyright 2006

by

Kornelia K. Matloka


































With my deepest love to Piotr, Nisia, and my entire wonderful family.
















ACKNOWLEDGMENTS

Words could never express my gratitude to my husband Piotr. I would have never

come this far without his love, courage, contagious optimism and genuine faith in me.

His dreams and bold aspirations inspired me to become the person I am today.

I could have never succeeded without the love and support of my wonderful family.

I thank my brave mom, Teresa, for motivating me to higher achievements and inspiring

my independence. I thank my grandma, Karolcia, and deceased sweet aunt Nisia for their

absolutely unconditional love, which continues to support me through difficult times.

I thank my great parents in law Marysia and Staszek, and my beloved sisters, Agnieszka

and Paulina. Even far away they have always managed to keep my spirit up when I

needed it the most. The sacrifices, patience and understanding of my wonderful family

have allowed me to pursue my dreams and become successful both in my career and life.

Eternal appreciation is given to people who made this great adventure possible,

especially Dr. Henryk Koroniak and Dr. Violetta Patroniak who encouraged me and Piotr

to undertake the challenge of studies abroad, Dr. James Deyrup, his wife Margaret and

Lori Clark who warmly welcomed us in Gainesville and the University of Florida.

The last five years of my life have been incredible thanks to my friends. Enough

thanks cannot be expressed to Iwona and Lukasz Koroniak for their hospitality and for

making the transition between Poland and the United States so smooth and easy. I cannot

imagine my life without the craziest of all Vicky Broadstreet, so French Flo Courchay

and the chocolate addict, gorgeous Merve Ertas. They have truly accepted and loved me









for who I am, and always have been there for me. I could not ask for better friends than

Travis Baughman, John Sworen, Sophie Bernard and Josh McClellan. I am deeply

grateful to all of them for sharing with me both the fun and the struggles of the graduate

school adventure.

I must also thank many extremely talented and wonderful people I have been

fortunate to be surrounded with over the years of my studies. I thank all past and present

members of the Scott group in particular Matt Peters who introduced me to the subj ect of

my research and taught me all the necessary laboratory techniques, my lab mate and

entertainer "Romeo" H. Gill, the beautiful, artistic soul of Cooper Dean, adorable Eric

Warner and rebellious Issac Finger. I thank my neighbour and lab mate who made me

feel very welcome when I first j oined the Scott group, and on whom I could have always

rely on, sweet and never rested recruiter Ivana Bozidarevic and her eccentric husband

Cira. A special acknowledgement is deserved by Ajaj Sah and Priya Srinivasan, who I

had the pleasure to share research projects with. Their hard work significantly contributed

to the research presented in my dissertation. Other special thanks go to my dear lab mate

Ranj an Mitra, who shared with me the great and the not so great days in the lab, patiently

listening to my excitements or complaints, which must have scared him for life. In

preparing this manuscript, I owe much to my colleagues Candace Zieleniuk, Eric Libra

and impressive synthetic chemist Melanie Veige. From the bottom of my heart I thank

them not only for putting up a great obstinate fight with the articles in editing my

dissertation, but also for being a wonderful vent for my frustrations. Finishing this

manuscript would not have been possible without them.









Tremendous gratitude goes to all the excellent scientists and educators, which I had

pleasure to meet and learn from at the University of Florida, especially members of my

supervisory committee: Dr. Daniel Talham, Dr. William Dolbier, Dr. Khalil Abboud and

Dr. James Tulenko, for the time and effort they invested in reading and discussing my

dissertation. Sincere thanks and affection is extended to an amazing man, both

professionally and personally, Dr. Tom Lyons for all the extraordinary conversations,

openness and friendship.

Many thanks go also to our research partners from Argonne National Laboratory,

Dr. Artem Gelis, Dr. Monica Regalbuto and Dr. George Vandegrift, and the Nuclear

Energy Research Initiative (Grant 02-98) of the Department of Energy National Nuclear

Security Administration for financial support.

In closing, I wish to express my special gratitude to a remarkable person and

passionate scientist, Dr. Mike Scott, whose tremendous help, encouragement, brilliant

creativity and enthusiasm has guided my research from its conception to its completion.

I will be eternally grateful to him for creating an absolutely incredible atmosphere of my

graduate studies. His flamboyant personality and the biggest heart in the world have

turned this initially terrifying experience into wonderful life changing j ourney. Through

his inhuman patience, care and understanding he gave me the comfort to find myself in

the new environment so I could finally start to grow as a scientist. His trust and support

have slowly built my confidence, and the freedom to explore the chemistry he offered,

inspired my true passion for science. I am deeply thankful for his support and the

attention he lavished on my scientific development. His thoughtful advice proved

invaluable, which cannot be repaid in words alone. I would like him to know how










grateful I am for not only having him as my research advisor, but simply for having

known him.

Mike, thank you.



















TABLE OF CONTENTS


IM Le

ACKNOWLEDGMENT S .............. .................... iv

LI ST OF T ABLE S ................. .............._ xi....___....

LI ST OF FIGURE S .............. .................... xii

AB S TRAC T ......_ ................. ............_........x

CHAPTER

1 INTRODUCTION ................. ...............1.......... ......

1.1 Benefits of Nuclear Fuel Reprocessing .............. ........ ...............
1.2 Liquid-Liquid Extraction Partitioning Processes: an Overview ................... ..........3
1.3 Characteristics of Trivalent Lanthanides and Actinides Valuable for
Separation ................ ...............7...
1.4 Actinides Binding Controversy .............. ... ...............8...
1.5 The Basics of Liquid Liquid Extraction Process ................. ................ ...._.9
1.5.1 Influence of the Organic Diluent on Extraction Process ................... .........10
1.5.2 Influence of the Aqueous Phase Composition on Extraction Process........ 11
1.5.3 Thermodynamics of Biphasic Complexation ............__.. .......__........12
1.5.4 Advantages of Large Extractants over Small Chelates .............. ..... ..........12
1.5.5 Types of Extraction Reactions ................. ...............13..............
1.6 Research Obj ectives ................. ...............14...............

2 CMPO FUNCTIONALIZED C3-SYMMETRIC TRIPODAL LIGANDS FOR
LANTHANIDES AND ACTINIDES SEPARATIONS IN THE NITRIC ACID
LIQUID/LIQUID EXTRACTION SYSTEM .............. ...............16....

2. 1 Introducti on ................. ........_ ........_ ...........1
2.1.1 Organophosphorous Extractants............... ..............1
2. 1.2 Development of the Tris-CMPO Chelate ................. ........................18
2.2 Results and Discussion ..................... .. ............................2
2.2. 1 Effect of the Structural Modification of Triphenoxymethane Platform
on the Tris-CMPO Extraction Profile ......____ ..... ... .__ ........__......21
2.2.2 Ligand Flexibility vs. Binding Profile............... .. ...............2
2.2.3 Complexation Studies with Bis-CMPO Compound .............. ................25
2.2.4 Attempts to Resolve the Tris-CMPO Solubility Issue .............. ................27











2.2.5 Plutonium (IV) and Americium(III) Extractions............... .............3
2.2.6 Comparison of Solid State Structures of Tris-CMPO Complexes of
Trivalent Metal lons............... ...............35..
2.3 Conclusions............... ..............4
2.4 Experimental Section............... ...............42
2.4. 1 General Consideration ...._.._.._ ..... .._._. ...._.._ ...........4
2.4.2 Metal lons Extractions ...._.._.._ ..... .._._. ...._.._ ...........4
2.4.3 Isotopes Stock Solutions............... ...............4
2.4.4 Synthesis............... .. ..............4
2.4.5 X-Ray Crystallography............... ............5

3 DESIGN, SYNTHESIS AND EVALUATION OF PHOSPHINE SULFIDE
BASED CHELATES FOR THE SEPARATION OF TRIVALENT
LANTHANIDES AND ACTINIDES .............. ...............62....

3 .1 Introducti on ................. ...............62.._......
3.2 Results and Discussion ................... ...............66..
3.2.1 Synthesis and Extraction Data............... ...............66..
3.2.2 Crystal Structure Analysis............... ...............69
3.3 Conclusions............... ..............7
3.4 Experimental Section............... ...............74

4 BINDING OF TRIVALENT F-ELEMENTS FROM ACIDIC MEDIA WITH A
C3-SYMMETRIC TRIPODAL LIGAND CONTAINING DIGLYCOLAMIDE
AND T HIO DIGLY COLAMIDE ARMS .............. ...............78....


4. 1 Introducti on ....._.................. ...............78.....
4.2 Results and Discussion .............. ...............80....
4.2.1 Ligand Synthesis .............. ...............80....
4.2.2 Extraction Experiments .............. .......... ........... ........81
4.2.2.1 Extraction properties of large chelate vs. small diglycolamide........81
4.2.2.2 Ligand flexibility vs. extraction performance .............. .... ........._..84
4.2.2.3 Solvent effect on ligand extraction profile ................. .... .............. .85
4.2.3 Investigation of Solid State Complexes of Trivalent Lanthanides..........._..89
4.2.4 Solution Structure of Extracted Species ............... ... ...... ..._. ... ........._.....94
4.2.5 Importance of the Etheric Oxygen of Tris-DGA in Metal Binding ...........96
4.3 Conclusions............... ..............9
4.4 Experimental Section............... ...............99
4.4.1 General Considerations .............. ...............99....
4.4.2 1H NMR Experiment ............ .....__ ...............99.
4.4.3 Synthetic Procedures ............ ..... .__ .....__ ...........10

5 PYRIDINE N-OXIDE FUNCTIONALIZED C3-SYMMETRIC CHELATES
FOR F-ELEMENS BINDING................ ...............10

5 .1 Introducti on ................. ...............107........... ...
5.2 Results and Discussion ................ ...............111........... ...












5.2. 1 Synthesis of Tris-PyNO Derivatives .........__... ........ ..................11 1
5.2.2 Extraction Experiments ....._._ ................ ............... 112 ....
5.2.3 Solid State Studies ................. ...............117........... ...
5.3 Conclusions............... ..............12

5.4. Experimental Section............... ...............121


6 SUM MARY .........._...._ ......... ...............126.....


LIST OF REFERENCES ............._... ...............128...._.........


BIOGRAPHICAL SKETCH ............._.. ...............139...._..........

















LIST OF TABLES


Table pg

2-1 Distribution coefficients (D) and extraction percentage (%E) for ligands 2-6a,
2-6b and 2-6c............... ...............23..

2-2 Distribution coefficients (D) and extraction percentage (%E) for ligands 2-1, 2-7
and 2-10a. .............. ...............27....

2-3 Distribution coefficients (D) and extraction percentage (%E) for ligands 2-14a and
2-14b............... ...............31.

2-4 Distribution coefficients (D) for the extraction of Pu(IV), U(VI), Am(III) and
Eu(III) by ligands 2-6b, 2-14a and 2-14b in methylene chloride and 1-octanol......33

2-5 Selected bond lengths (A+) for compounds 2-10b, 2-14a, [2-6a-TbNO3](NO3)2,
[2-14a-TbNO3](NO3)2 and [2-6c-BiNO3](N O3) 2 .................... ............... 4

2-6 X-ray data for the crystal structures of 2-10b, 2-14a and the complexes
[2-6a-TbNO3](NO3)2, [2-6c-BiNO3](NO3)2 and [2-14a-TbNO3](NO3)2 ................... 61

3-1 Extraction percentage (%E) for ligands 2-6a, 3-2a, 2-10 and 3-3. ........... ................69

3.2 Selected bond lengths (A+) for compounds: 3-3 and [3-3-Tb(NO3)3] COmplex. .........._72

3-3 X-ray data for the crystal structures of 3-3 and [3-3-Tb(NO3)3] COmplex. .................. 73

4-1 Extraction data (logD) for ligands 4-2 and 4-5 in dichloromethane...........................8

4-2 Extraction data (logD) for ligands 4-2 and 4-5 in octanol and dodecane. ........._._......87

4-3 Extraction data (logD) for ligands 4-5 and 4-7 in dichloromethane...........................8

4-4 X-ray data for the crystal structures of[4-5-Ce] [Ce(NO3)6], [4-6-Eu](NO3)3,
[3x4-2-Yb](NO3)3, [4-5-Yb](NO3)3 and [Yb-cage](NO3)3........ ...............10610


















LIST OF FIGURES


figure pg

1-1 Nuclear fuel cycle. ........._.._...... ..... ...............2...

1-2 Common pathways in spent fuel reprocessing. ............. ...............4.....

1-3 Popular neutral organophosphorus extractants ................. ..............................4

1-4 Fully incinerable extractants ................. ...............5............ ...

1-5 S ANE X extract ants............... ...............

1-6 SANEX nitrogen based extractants. ............. ...............6.....

1-7 TALSPEAK extractants. ............. ...............7.....

1-8 Illustration of the "all up" conformation of the oxygen atoms on the
triphenoxymethane platform. ............. ...............13.....

2-1 Acidic organophosphorous extractants. ........._._.. ....__.. ...._... ...........1

2-2 Schematic depiction of proposed solution structure of the americium (III) nitrato-
CMPO complex at high nitric acid concentration..........._.._.. ......._.._........._..19

2-3 Calix[4]arenas with CMPO functions at the narrow and wide rims............................20

2-4 Classic carbamoylmethylphosphine oxide (CMPO). ............. .....................2

2-5 Synthesis of tris-CMPO 2-4. ........._... ....___ ....___ ........_ ..............22

2-6 Synthesis of tris-CMPO 2-10.. ............ ...............24.....

2-7 Metal extraction percentages (%E) for the ligands 2-6a, and 2-10a using 10-4 M
metal nitrate in 1 M nitric acid and 10-3 M ligand in dichloromethane. ................25

2-8 Diagrams of the 10 coordinate +2 cationic thorium(IV) nitrate complex of 2-6
with two coordinated NO3~ COunterions. ............. ...............25.....

2-9 Bis-CMPO compound 2-7. .............. ...............26....

2-10 Metal extraction percentages (%E) for the ligands 2-6a, and 2-7. ............................26











2. 11 Fragment of the crystal structure of 2-6b molecules forming hydrogen bond
connected network. ........._...._ ...._._. ...............28.....

2-12 Diagram of the solid-state structure of 2-10b ........._......_ ....._... ........._.....29

2-13 Synthesis of tris-CMPO 2-14.. ............ ...............30.....

2-14 Diagram of the solid-state structure of 2-14a. ........._..._.._ ...._._. ......._..._.....3

2-15 Metal extraction percentages (%E) for the ligands 2-1, 2-6a, and 2-14a. ........._......32

2-16 Metal extraction percentages (%E) for the ligands 2-6b, 2-14a, and 2-14b. .............34

2-17 Diagrams of neodymium(III) complexes of 2-6b ....._____ ... .....___ ................36

2-18 Diagram of the structure of [2-6c-BiNO3](N O3)2 ......... ................. ...............37

2-19 Diagram of the structure of compound [2-6a-TbNO3](NO3)2.3 8

2-20 Diagram of the structure of compound [2-1 4a-TbNO3](NO3)2. ............. ..... ........._..3 9

3-1 Structures of sulfur based extractants ................. ...............62..............

3-2 Structure of the aromatic dithiophosphinic acids. ............. ...............64.....

3-3 Anticipated binding mode for tris-CMPS extractant .................... ............... 6

3-4 Synthesis of tris-CMPS extractants ................ .............. ......... ........ ....66

3-5 Comparison of metal binding by tris-CMPO (2-6a) and tris-CMPS (3-2a). ..............67

3-6 Comparison of metal binding by 3-2a and 3-3. ............. ...............68.....

3-7 Comparison of metal binding by tris-CMPO (2-10a) and tris-CMPS (3-3). ..............68

3 -8. Diagram of the solid-state structure of 3 -3 ................ ...............70...........

3-9 D iagram of the structures of [3-3-Tb(N O3 3]............... ...............71

4-1 Synthesis of C3-Symmtric tris-diglycolamides. ............. ...............80.....

4-2 Extraction of trivalent lanthanides with 4-2 and 4-5 in dichloromethane. ........._......82

4-3 Extraction of trivalent lanthanides with 4-5 in dichloromethane and 1-octanol. ........85

4-4 Extraction of trivalent lanthanides with 4-2 and 4-5 in 1-octanol and n-dodecane.....86

4-5 Extraction of trivalent lanthanides with 4-5 and 4-7 in dichloromethane. ........._......88

4-6 The tricapped trigonal prismatic (TTP) geometry around nine coordinate Yb(III). ...89










4-7 Diagram of the structures of [4-5-Yb]3+ and [4-2-Yb]3+ ............ ......_..._........90

4-8 Diagram of ytterbium encapsulated by a cage-like derivative of tris-DGA
com pound. .............. ...............94....

4-9. Superimposed 1H NMR spectra of ligand 4-5 ......____ .... ... .__ ........... ....95

5-1 Structures of the most extensively studied amine oxides. .............. ....................10

5-2. The electron distribution in pyridine N-oxide. ............. ...............108....

5-3 Resonance structures of pyridine N-oxide ................. ...............108..............

5-4 Synthesis of tris-pyridine N-oxides. ................ ......... ......... ............1

5-5 Structure on the hexachlorinated cobaltocarborane sandwich anion (CO SAN). ......113

5-6. Stacked 1H NMR spectra of ligand 5-4. ........... ..... .__ .. ....__.......15

5-7 Stacked 1H NMR spectra of ligand 5-1 ................ .......... ............ ........1

5-8 Two geometric extremes of metal binding by the substituted N-oxide. ................... .1 18

5-9 Diagram of the coordination environment of [5-4-Yb(NO3)2](NO3). ................... ...1 19

5-10. Coordination environment of ytterbium (III) in the complex with 5-4. .................1 19
















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

SYNTHETIC DEVELOPMENT OF C3-SYMMETRIC TRIPHENOXYMETHANE
BASED REAGENTS FOR THE SELECTIVE RECOGNITION AND
SEQUESTRATION OF LANTHANIDES AND ACTINIDES

By

Kornelia K. Matloka

May 2006

Chair: Michael J. Scott
Major Department: Chemistry

The prospective increase of global nuclear power utilization requires a significant

modification of nuclear waste management to help overcome related environmental,

economic, and political challenges. The collaborative effort with Argonne National

Laboratory has led to the development of reagents with the ability to selectively bind

lanthanides and actinides in conditions simulating nuclear waste solutions, and could

contribute to the advancement of the nuclear fuel and waste reprocessing strategies. The

focus has been placed on the fundamental chemistry of metal-ligand interactions both in

the solid state complexes and in solution. The study of these complexes assisted in the

design of improved systems for metal separation. A sequence of tripodal chelates bearing

a variety of binding moieties, diglycolamide (DGA), thiodiglycolamide (TDGA),

carbamoylmethylphosphine oxide (CMPO), carbamoylmethylphosphine sulfide (CMPS)

and pyridine N-oxide (PyNO), precisely arranged on a C3-Symmetric triphenoxymethane

molecular platform were synthesized. The impact of structural modifications of these










ligands on their affinity for f-element ions in 1 M nitric acid extraction system has been

evaluated.

The reorganization of three diglycolamide binding units on the triphenoxymethane

scaffold resulted in significant enhancement of the extraction efficiency and selectivity

within trivalent lanthanide ions series. Additionally, the tris-diglycolamide chelate has

been recognized as the first single, purely oxygen based donor capable of fully satisfying

the tricapped trigonal prismatic geometry favored by nine coordinated lanthanides. The

CMPO-based ligand has shown an excellent binding efficiency and a remarkable

selectivity for tetravalent actinides. The structural modifications of this chelate system

led to the development of one of the most efficient plutonophiles. Experiments with

CMPS, TDGA and PyNO compounds provided valuable information about the

coordination preferences of the tested cations in the solid state complexes and in solution.

The synthetic development and characterization of tripodal chelates and their metal

complexes are presented herein. The impact of the structural derivatization of ligands on

lanthanide and actinide ion extraction and separation is also discussed along with its

implication towards potential applications in waste reprocessing.















CHAPTER 1
INTTRODUCTION

According to the World Nuclear Association there are 441 nuclear power reactors

currently operating in 31 countries, generating over 368 gigawatts of electrical energy

worldwide.1,2 The United States alone houses 103 reactors that produce approximately

20% of total electricity. The International Energy Outlook projects a 57% increase in the

world energy consumption over the 2002 to 2024 time period.3 The most significant

energy demand increase is expected to arise among the emerging economies of Asia, in

particular China and India, as well as economies of Eastern Europe and the former Soviet

Union. Considering that the renewable sources of energy (biomass, solar, wind) alone

cannot produce enough energy to sustain global development, the nuclear power

expansion appears to be economically and environmentally reasonable alternative to the

fossil fuel energy. The expansion of the nuclear power would significantly reduce the

greenhouse gas emissions like carbon dioxide, sulphur dioxide and nitrogen oxides

generated from fossil fuel combustion (coal, oil and natural gas), and ultimately reduce

the problematic air pollution. Moreover, nuclear energy could be efficiently used to

produce large quantities of hydrogen gas, a potential maj or energy carrier that is clean

and environmentally friendly. Contrary to public opinion, the radiological hazard of

nuclear waste can be reduced, and the efficiency of nuclear power plants can be further

improved through the recycling of the spent nuclear fuel. Currently, the potential threat

of nuclear weapon proliferation and expensive reprocessing technologies prevent the










United States nuclear industry from waste reprocessing. In order to extend the

exploitation of nuclear power, waste treatment is inevitable.

1.1 Benefits of Nuclear Fuel Reprocessinn

In a typical light water reactor, the operational life-span of a fuel rods is only three

years, and since they still contain approximately 96% of their original energy potential,

the recycling of fuel rods offers both significant economic and environmental impacts.4

In the US, spent fuel rods currently remain in storage at the reactor site, awaiting eventual

transport for disposal at a government repository. Protracted litigation, however, may

keep the Yucca Mountain Repository closed for many years to come.' The partitioning

of radioactive waste followed by the transmutation of problematic long-lived radionuclei

would reduce uncertainties related to the long-term waste storage (several hundred

thousand years) in the geological repositories.

Uranium Ionversionl Uranium
Mining to UO Ercmn


Recovered Fuel
Uranium -r Fabrication














Waste
For
Burial


Figure 1-1. Nuclear fuel cycle.









After removal from a reactor, a spent fuel rod contains mainly a mixture of 235U

and 238U along with small amount of 239Pu intermixed with radioactive and non-

radioactive fission products. An efficient spent fuel reprocessing protocol would involve

the recovery of these two maj or components (U and P), followed by the separation of

long-lived transplutonium radionuclides from other relatively innocuous elements.

Uranium and plutonium can be reused as a fuel, while the long-lived actinides could

further be transmuted by neutron bombardment into short-lived and non-radioactive

elements, decreasing long-term radiotoxicity of waste and practically eliminating the

waste disposal problem. Unfortunately, the transmutation of actinides would be impeded

by even small amounts of lanthanide ions, and a very efficient protocol for their

separation must be developed for this process to work.

1.2 Liquid-Liquid Extraction Partitioninn Processes: an Overview

Nuclear waste reprocessing begins with the dissolution of used fuel rods or waste in

concentrated nitric acid. To date, the most successful technique used for the partitioning

of nuclear waste components is the liquid-liquid extraction separation, which

predominately employs organophosphorus extractants. Industrial uranium and plutonium

recovery utilizes neutral monodentate tri-n-butyl phosphate [TBP] in the PUREX

(Plutonium/URanium EXtraction) process.6,7 Other popular organophosphorus

compounds like bidentate CMPO [octyl(phenyl)-N,N-diisobutylcarbamoyl

methylphosphine oxi de] and DHDECMP [dihexyl -N,N-di ethyl carb amoylmethyl ene

phosphonate]s have also been widely studied as chelating agents for both actinides and

lanthanides.













r PUREX
recovery


TRUEX DIAMEX
1TUX [IAMX

Other Fission
Products An(lll) + Ln(lll)
TALSPEAK



rSANEX7
SANEX-S SANEXE-N)


CYANEX 301 BTsTMAHDPTZ
IU, I1 IB~r [IVInP




Figure 1-2. Common pathways in spent fuel reprocessing.

The extraction power of CMPO and TBP, originally studied by Horwitz and

coworkers, has found application in the commercially operating TRUEX (TRansUranium

elements EXtraction) process9-11 for the removal of both lanthanide and actinide ions

from high level liquid waste generated during PUREX reprocessing.12-14







TBP CMPO DHDECMP

Figure 1-3. Popular neutral organophosphorus extractants.









The demand to fully accomplish environmental mission of the waste treatment has

driven the development of purely C, H, N, and O based extractants that would not

generate secondary waste after final incineration. Significant attention has been focused

on diamide based extractantsl5-22 in particular N,N'-dimethyl-N,N'-dibutyl tetradecyl

malonamide (DMDBTDMA). Based on DMDBTDMA and later N,N'-dimethyl-N,N'-

dioctyl hexyloxyethyl malonamide (DMDOHEMA) the DIAMEX (DIAMide Extraction)

process has been developed and strongly promoted in France as an environmentally

friendly alternative to TRUEX.23-25



OO


DMDBTDMA DMDOHEMA

Figure 1-4. Fully incinerable extractants.

None of these hard-donor oxygen based extraction systems, offer a solution to the

most problematic partitioning of trivalent actinides from the chemically similar

lanthanides. Interestingly, some potential for minor actinides separation has been

presented by soft-donor ligands employed in the SANEX (Selective ActiNide Extraction)

process. The SANEX procedure uses either acidic sulfur bearing or neutral nitrogen

based extractants. The most promising separation has been achieved by bis(2,4,4-

trimethylpentyl)dithi ophosphini c aci d known as Cyanex-3 01 (Figure 1 -5).26-28

Nevertheless, the commercial application of this extraction system has been limited by

low radiation stability, sulfur and phosphorus containing degradation products and the

necessity to adjust the pH to 3-5. To partially overcome the problematic pH adjustment,

a modified solvent mixture of bis(chlorophenyl)dithiophosphinic acid and the hard donor










tri-n-octyl-phosphine oxide (TOPO) extractant has been used in the German process

ALINA,29 but the radiological stability and secondary waste generation problems remain.









Cyanex-301 C 1TOPO


Figure 1-5. SANEX extractants.

In the second variant of the SANEX concept, either bi s-5,6-sub stituted-bi s-

triazinyl-1 ,2,4-pyridines (BTPs, R1, R2 = H or alkyl group)30,31 Or a synergistic mixture of

2-(3,5,5 -trimethylhexanoyl amino)-4,6-di-(pyri dine-2-yl)-1,3,5 -triazine (TMAHDPTZ)32,

and octanoic acid are used (Figure 1-6). Hydrolytic instability of BTPs and, as in the

case of S-bearing compounds, the necessity of some pH adjustment for TMAHDPTZ

along with solubility issue, significantly limits the potential applications of the N-based

extraction system in the separation of trivalent lanthanides (Ln) and actinides (An).


/""-NH




BTPs
R R2 =H, alkyl TADT

Figure 1-6. SANEX nitrogen based extractants.

An alternative to the SANEX extraction system is the TALSPEAK process

(Trivalent Actinide/Lanthanide Separation by Phosphorus reagent Extraction of Aqueous

Komplexes) based on selective stripping of An(III) from a mixture of trivalent Ln and An

bound by diethylhexylphosphoric acid (HDEHP), with a aqueous solution of









diethylenetriaminopentaacetic (DTPA) and hydroxocarboxylic acids (lactic, glycolic or

citric).33,34 Although the process is relatively efficient, TALSPEAK suffers from

common drawbacks, namely the necessity of the pH adjustment, limited solvent loading

with metal ions and secondary phosphorus waste generation.35



OH Or



HDEHP DTPA

Figure 1-7. TALSPEAK extractants.

After years of intensive research An(III)/Ln(III) separation remains one of the key

problems facing the partitioning and transmutation of nuclear waste management.

Hopefully, the development of new or the improvement of historically proven

technologies will help raise public awareness and acceptability of nuclear power as the

most viable energy source to sustain global development without significant

environmental consequences.

1.3 Characteristics of Trivalent Lanthanides and Actinides Valuable for Separation

The chemical properties of lanthanides and actinide are very similar. Historically,

the close relationship between these two groups of elements helped predict the properties

of the transplutonium elements, which later resulted in their synthesis and isolation.

From a waste reprocessing perspective, however, such similarities are disadvantageous.

The identical oxidation states of trivalent actinides and lanthanides and an approximately

equal ionic radii, make the separation particularly difficult. Both trivalent lanthanides

and actinides are hard acids and form complexes preferentially with hard bases through

strong ionic interactions. Careful examination of their electronic structures and binding









with less compatible soft bases revealed small but important differences between these

two groups of cations.36

One of the maj or differences between lanthanides and actinides is the exceptional

stability of the lanthanides' trivalent oxidation state.36 This stability can be attributed to

the effective shielding of the 4f orbitals provided by the outer 6s and 5d orbitals. Across

the lanthanides series electrons successively populate the 4f orbitals. The radii of the

outer 6s and 5d orbitals are significantly bigger than the average radius of the 4f orbital,

which effectively shields the f electrons and stabilizes the third oxidation state.37 In the

case of actinides, more spatially extended 5f orbitals are no longer effectively shielded by

the outer 7s and 6d orbitals, which results in the decreased stability of the trivalent state.

The later actinides (transplutonium elements), however, show somewhat higher stability

of the third oxidation state with respect to the early members of the group. This effect

originates from slightly more pronounced decrease of the 5f orbital radius with respect to

the size of 7s and 6d orbitals as the elements become heavier. This provides some

shielding of f orbitals, although it is not as effective as in the case of 4f orbitals of

lanthanides.36,38

The separation between the lanthanide and actinide group can be well represented

by the partition of one isoelectronic pair of lanthanide and actinide ions, for instance

americium and europium. This representation is fairly accurate since the extent of

separation between these two groups depends mostly on the character and strength of the

interaction between the metal ion and ligand and not on the contraction of the ionic radii.

1.4 Actinides Bindinn Controversy

The interaction between the trivalent f-element cations and hard bases are ionic

(electrostatic) in nature. The lack of bonding orbital overlap constrains results in a wide









range of coordination numbers and geometries that are controlled by the electronic and

steric dynamics of the complex.36,39 In the 1950s, the concept of a small degree of

covalency in the f-element interactions with soft bases was proposed.40 The origin of

such phenomenon has become a subj ect of controversy ever since.

The trivalent An/Ln separation studies revealed that these ions can be separated by

a preferential elution of An from a cation exchange resin column using concentrated

hydrochloric acid.40 This phenomenon was attributed to the small degree of covalent

interactions between the actinide ion and chloride which was explained by the

participation of the 5f orbital in the binding. Also, a smaller energy difference between

the 5f and 6d orbitals of actinides with respect to the 4f and 5d orbitals of lanthanides

helped elucidate the greater sensitivity of actinides to their chemical environment.36,40

The effective shielding of the 4f orbital by the 6s and 5d orbitals and the energetic

mismatch among orbitals results in the weaker interactions between Ln(III) and the

ligands. According to the later literature analysis, however, this undeniable small

covalent contribution reflected by stronger complexation of trivalent actinides with soft

donor ligands involves more likely the 7s and/or 6d orbitals rather than 5f, which is also

consistent with the flexible metal coordination due to the spherical character of the s

orbital.37

1.5 The Basics of Liquid Liquid Extraction Process

The principle of the liquid liquid extraction states: "If two immiscible solvents

are placed in contact, any substance soluble in both of them will distribute or partition

itself between two phases in a definite proportion."41 The solvent extraction separation

process used in waste reprocessing is based on the transfer of a metal cation from an

aqueous phase (mineral acid) into an immiscible organic phase with simultaneous charge









neutralization.42,43 During the extraction, a variety of species are formed in combination

of metallic salt with water, mineral acid and organic solvent. The extent of extraction can

be expressed by the distribution ratio (D = C[Morg]/E[Maq@) Of the total metal ion

concentration in the organic phase (E[Morg]) againSt the total metal ion concentration in

the aqueous phase (E[Maq@), Or percentage wise (%E = [D/(D + 1)] 100). The separation

effectiveness of two different species can be assessed based on the separation factor

(SFA/B=DA/DB; the fraction of the individual distribution ratios of two extractable species

measured under the same conditions). For example, the successful separation of two

elements can be accomplished with separation factor of 100, which corresponds to 99%

of partitioning efficiency in one contact.42

In the liquid liquid extraction process the extractant is commonly

diluted/dissolved in a water immiscible organic solvent. Therefore, successful and

efficient separation is determined not only by the choice of proper ligand, but also by the

right choice of organic solvent and the content of the aqueous solution.

1.5.1 Influence of the Organic Diluent on Extraction Process

Choosing the organic diluent is perhaps as important as the ligand. The character

of the diluent directly influences the physical and chemical properties of the extractant.

The selection of solvent offers control over the organic phase density, viscosity, freezing,

boiling and flash points, separation ability of the mixed aqueous and organic phases,

ligand solubility in the aqueous phase, third phase formation (formation of two separate

organic phases), as well as, the distribution of the extractable species between two

immiscible phases, separation efficiency, kinetics of extraction, and even chemical and

radiolytic stability of the extractant.44 There are multiple literature examples where the









efficiency of extraction differs in different organic solvents by several orders of

magnitude.43 Depending on solvation properties of the solvent, the solubility and

therefore stability, as well as, stoichiometry of the extracted species are affected. For

example, as mentioned earlier both trivalent lanthanides and actinides are typical hard

acids therefore they preferentially react with hard bases. The slightly stronger interaction

between the hard metal ion and soft base can be facilitated only in weakly solvating

organic solvents. In the aqueous environment soft bases cannot compete with water for

the metal binding.36 Even with these common trends, the complexity of the extraction

process does not allow for the accurate prediction of the solvent effect on the separation,

which ultimately hampers fast progression in this field.43

1.5.2 Influence of the Aqueous Phase Composition on Extraction Process

Metal ions hydrolyze under low acid concentration, which may result in the

improvement or deterioration of the extraction efficiency and/or selectivity, depending on

the extraction mechanism.4,4 Higher acid concentration prevents metal hydrolysis by

decreasing the activity of water, but at the same time, it may also negatively influence

extraction by protonating the ligand. Since one of the energetic requirements for

successful phase transfer in the liquid-liquid extraction process is full or partial

dehydration of the metal ion upon complexation with a ligand, a lower water activity

results in a decreased net rate of the water exchange of the cation, improving the phase

transfer.42,43 A similar effect can be induced by adding the salt of small, extraction inert

cation with high charge density and a high degree of hydration. For example, lithium

nitrate is able to strongly compete with f-element cations over the water binding. Also,

an increased concentration of salt may affect the form of the extracted species and have

either a positive of negative effect on the extraction processes.









Ultimately, the addition of salt should be avoided if a similar enhancement of

extraction properties can be obtained using nitric acid as the "salting" out agent.

Opposed to inorganic salts, at the end of the extraction process, acid can be easily

evaporated from the system without increasing the volume of waste. 47

1.5.3 Thermodynamics of Biphasic Complexation

Thermodynamic changes in the complexation process of f-block cations are

predominately controlled by the changes in the solvation of the cation and the ligand.36

As extensively discussed by G. R. Choppin,36 upon complexation, these highly hydrated

metal ions release water molecules, which results in positive entropy change. The

dehydration process requires energy in order to break the interactions between the water

and the cation, as well as, other water molecules, which contributes to the positive change

of the net enthalpy of the complexation. The negative contribution to the net enthalpy

and entropy of complexation results from the binding of the cation and ligand, but it is

less significant than the contribution from the dehydration. It is believed that the slightly

higher degree of covalency in the actinide complexes with soft donor ligands may

provide some additional small thermodynamic stabilization to the complex by slightly

decreasing the net enthalpy of the process.48

1.5.4 Advantages of Larne Extractants over Small Chelates

The commercially operating partition processes commonly employ small organic

chelates. Typically two to four molecules are involved in the transfer of a metal cation

from an aqueous into the organic phase. With that in mind, a molecule with several

lighting moieties attached to some molecular support may have many advantages. The

new ligand may produce a significant change in the extraction properties due to the

entropic effect (release of water from highly hydrated f-element cations upon the










complexation) in addition to the modified composition of the complex produced by

multiple arms.49

We have found the C3-Symmetric triphenoxymethane molecule to have great utility

for the preparation of ligands with three extended arms,50,51 much like calix[4]arene can

serve as a base for four arm constructs.49,52-54

HO
R H

R


"all up" conformation

Figure 1-8. Illustration of the "all up" conformation of the oxygen atoms on the
triphenoxymethane platform relative to its central methane hydrogen.

Using simple techniques, we can incorporate a variety of binding moieties onto a

triphenoxymethane base through different alkyl linkers. Once the three phenols have

been derivatized, the triphenoxymethane molecule adopts exclusively an "all up"

conformation wherein the oxygens orient in line with the central methine hydrogen.so

The solubility properties of the final ligand can be easily modulated through simple

substitutions at the 2 and 4 positions of the phenols, as well as substitutions at the lighting

units. The following chapters report the synthesis of a series of C3-Symmetric

compounds along with a detailed investigation of binding properties of these new

chelates.

1.5.5 Types of Extraction Reactions

As summarized by K. L. Nash, 43 there are three types of solvent extraction

reactions: metal-complex solvation by an organic extractant, metal-ion complexation by

organic extractant and rare, ion-pair formation between an anionic aqueous complex and

a positively charged (protonated) organic extractant. In the first case, a neutral complex









of the metal ion with the conjugated base of the mineral acid is formed typically in the

aqueous phase, and upon solvation with the organic extractant is transferred to the

organic phase. Neutral organophosphorus compounds, such as phosphates or phosphine

oxides, as well as, ethers or ketones extract accordingly to this mechanism and

proportionally to their Lewis base strength. The second extraction mechanism is

observed in the case of acidic extractants, like phosphoric and carboxylic acids that form

complexes with water soluble metal ions in or near the interfacial zone. The combination

of acidic and neutral extractants often leads to the enhanced distribution ratio and is

commonly referred to as the synergistic effect. Even though the mechanism for this

effect is unknown, the enhanced stability of the extracted species is attributed to the

increased hydrophobicity of the complex that can be attained either by accommodating

the solvating molecule (synergist) in the expanded metal coordination sphere or by water

replacement.42'55

1.6 Research Objectives

The primary obj ective of the collaborative research effort with Argonne National

Laboratory has been the development of reagents with the ability to selectively bind

lanthanides and actinides in conditions simulating nuclear waste solutions that could

contribute to the advancement of the nuclear fuel and waste reprocessing strategies. The

focus has been placed on the fundamental chemistry of metal-ligand interactions both in

the solid state complexes and in solution. The study of these complexes assisted in the

design of improved systems for metal separation. A sequence of tripodal chelates bearing

a variety of binding moieties: diglycolamide (DGA), thiodiglycolamide (TDGA),

carbamoylmethylphosphine oxide (CMPO), carbamoylmethylphosphine sulfide (CMPS)

and pyridine N-oxide (PyNO) precisely arranged on a C3-Symmetric triphenoxymethane









molecular platform were synthesized. The impact of structural modifications of these

ligands on their affinity for f-element ions in 1 M nitric acid extraction system has been

evaluated with its implication towards potential applications in waste reprocessing.














CHAPTER 2
CMPO FUNCTIONALIZED C3-SYMMETRIC TRIPODAL LIGANDS FOR
LANTHANIDES AND ACTINIDES SEPARATIONS IN THE NITRIC ACID
LIQUID/LIQUID EXTRACTION SYSTEM

2.1 Introduction

As the world demand for energy increases, the use of nuclear power will continue

to grow in countries throughout the world. In order to sustain the world's development

many new reactors will be built and generate proportionally more radioactive waste.1-3

To maintain the low cost of nuclear energy and ensure environmental safety, the spent

nuclear fuel will need to be recycled worldwide. This process, however, will require a

significant advancement in the separation chemistry. The efficiency of separation

technologies developed in the 50's needs to be improved either through modification of

existing methods or/and the development of new procedures.

2.1.1 Ornanophosphorous Extractants

The most successful technique of nuclear waste partitioning is liquid-liquid

extraction, where elements of the irradiated material dissolved in nitric acid are

successively removed by a sequence of organic solvents. Initially, acidic

organophosphorous extractants like bis(2-ethylhexyl) phosphoric acid,56-58 diisodecyl

phosphoric acid,59 and bis(hexylethyl) phosphoric acid60 have been involved in waste

reprocessing procedures (Figure 2-1).9,61 These procedures, however, suffer from

significant drawbacks. In order to successfully extract trivalent elements the acidic

ligands require very low acidity and exhibit poor selectivity for trivalent actinides over

other fission products.9













HO, IO=P-O O=P-O




bis(2-ethylhexyl) phosphoric acid diisodecyl phosphoric acid bis(hexylethyl) phosphoric acid

Figure 2-1. Acidic organophosphorous extractants.

Along with acidic extractants, neutral organophosphorous compounds have also

been actively studied. Previous research has focused on compounds like tributyl

phosphate (TBP),5 a mixed trialkylphosphine oxides,62 dihexyl-N,N-

diethylcarbamoylmethyl phosphonate (DHDECMP),63-67 and octyl(phenyl)-N~,N-

diisobutylcarbamoylmethyl phosphine oxide (CMPO) (Figure 1-3, Chapter 1).68,69

Studies have shown that in order to afford efficient extraction by the monofunctional

extractants low acidities and occasional use of a salting out agents are required.9,57,62 In

contrast, bifunctional extractants such as DHDECMP and CMPO can effectively operate

at much higher acidities. It has been proposed that both the carbonyl and phosphoryl

oxygens of CMPO are directly involved in metal binding,6470 and the efficiency of this

bidentate ligand has been attributed to the chelate effect provided by the two donors.

According to Muscatello et al., in highly acidic media the carbonyl portion of ligands do

not participate in metal binding but rather protect the metal-phosphoryl bond from the

reaction with hydronium ions.71,72 Thus, in waste solutions typically of 1 to 3 M HNO3

the built-in buffering effect facilitates the extraction without necessitating an adjustment

in pH.









Industrial recovery of uranium and plutonium from spent nuclear fuel utilizes

neutral monodentate tri-n-butyl phosphate [TBP] in the PUREX (Plutonium/URanium

EXtraction) process.6,7 The mixture of CMPO and TBP extractants, originally studied by

Horwitz and coworkers, is applied in the commercially operating TRUEX

(TRansUranium elements EXtraction) process9-11 for the removal of both trivalent

lanthanide and actinide ions from the high level liquid waste generated during PUREX

reprocessing.12-14 Efficient extraction in these processes can be accomplished only with

high ligands concentrations (over 1 M), which results in the generation of significant

amount of unwanted secondary waste. Moreover, the lack of selectivity of CMPO and

DHDECMP creates a need for another separation step. Despite years of research studies,

the final partitioning of the trivalent actinides from the trivalent lanthanides as well as

selective separation of residual plutonium and uranium from the PUREX raffinate

remains a challenging task facing a new generation of separation chemists.

2. 1.2 Development of the Tris-CMPO Chelate

The structure of the classic CMPO has been modified in various ways to develop

new systems that would compensate for the lack of selectivity for trivalent actinides over

lanthanides in the TRUEX process.71,72 The solution structure of the Am(III) complex

with CMPO formed during the TRUEX procedure was examined by Horwitz et. al.73

Their work suggests that in the neutral complex the Am(III) ion is bound by three CMPO

molecules and three nitrate anions, and the additional nitric acid molecules are attached to

the CMPO carbonyl oxygens via hydrogen bonding interactions (Figure 2-2).49,73,74 The

resulting complex with tetravalent plutonium would involve two to four CMPO ligands,

while in the case of trivalent lanthanides more than one CMPO is bound. With these

properties in mind, a molecule with several CMPO groups attached to some molecular










support may have many advantages. The new ligand may produce a significant change in

the extraction properties due to the entropic effect (release of water from highly hydrated

f-element cations upon the complexation) in addition to the modified composition of the

complex produced by multiple arms.49


CMPO-HNO3 8



Am3
HNO3-CMPO CMPOHNO



Figure 2-2. Schematic depiction of proposed solution structure of the americium (III)
nitrato-CMPO complex at high nitric acid concentration adapted from F.
ArnaudNeu et. al. Perkin Trans. 2, 1996.49

Inspired by these ideas, a variety of calixarene and calixarene-like multi-CMPO

supported compounds have been developed,49,52,75-77 and indeed the reorganization of

several lighting units on the molecular platform significantly improved the binding

efficiency and/or selectivity of the ligand. Particularly interesting are cases of CMPO

moieties secured to resorcinarene, and the most extensively studied calix[4]arene

platforms.49,52-54,76,78,79 COmpounds with four CMPO moieties tethered to the narrow and

wide rims of calix[4]arenes (Figure 2-3) have shown an increased actinide affinity

relative to the mono-CMPO ligand, but their lanthanide extraction significantly varied

across the series.

Taking inspiration from the suggested coordination environment of the extracted

species in the TRUEX process and the calix[4]arene work, our research group has

developed chelate system with three precisely arranged carbamoylmethylphosphine oxide

moieties attached to a rigid, triphenoxymethane platform (tris-CMPO).S

















u L 0
s'O


4 ~Ph o


Figure 2-3. Calix[4]arenas with CMPO functions at the narrow and wide rims.

In the design of the tris-CMPO ligand, we hoped to achieve high distribution

coefficients for the selected metal ions, while maintaining high ion selectivity and great

stability toward hydrolysis in acidic media. The basicity of the phosphoryl oxygen

increases in the order: phosphate (RO)3PO < phosphonate (RO)2RPO < phosphinate

(RO)R2PO < phosphine oxide R3PO; were R=alkyl.43 An increase in the basicity

improves the extraction efficiency through a stronger interaction of the ligand with the

metal ion, but at the same time ion selectivity is decreased.72 Therefore, to afford both

high extraction efficiency and selectivity the substituents on the phosphorus should be

chosen very carefully. In the tris-CMPO chelate, the desired basicity of phosphoryl

oxygen was achieved via the replacement of the alkyl group present in the original

CMPO (Figure 2-4) with a phenyl group, thus decreasing the basicity of phosphine oxide

through an enhanced inductive effect of the second aromatic ring without the introduction

of easily hydrolyzable P-O-C bonds. In addition, the phenyl groups adj acent to the P=0

provide steric hindrance, an attribute possibly responsible for the higher selectivity for

Am(III) over Eu(III) in some CMPO derivatives.72














2-1O

Figure 2-4. Classic carbamoylmethylphosphine oxide (CMPO).

The extraction affinities of our ligand for a selected group of trivalent lanthanides

and tetravalent thorium have been compared to multi-CMPO calix[4]arene based

extractants. The tris-CMPO ligand system is superior in comparison to other CMPO

compounds in the highly selective binding of tetravalent thorium.51 The impact of further

structural ligand derivatization on the extraction selectivity and efficiency for tetravalent

actinides, with plutonium in particular is presented herein. The extraction behavior of the

tris-CMPO derivatives in comparison to the classic CMPO and other multi-CMPO

systems has been discussed, demonstrating an overall improvement in the development of

CMPO-based extractants for actinide/lanthanide separations. In addition, metal

complexes of tris-CMPO derivatives have been isolated and their structures were

elucidated by NMR, ICR-MS, and X-ray analysis, which provide potential rationalization

for the presented ligands extraction behavior.

2.2 Results and Discussion

2.2. 1 Effect of the Structural Modification of Triphenoxvmethane Platform on the Tris-
CMPO Extraction Profile

As mentioned above, a 3:1 CMPO to metal complex may form during the

extraction of americium from the acidic media in the TRUEX process,73,74,80 and to

facilitate the extraction we envisioned that a single ligand with three CMPO arms

extended out from a molecular platform could greatly enhance the binding and extraction

of the metal ions from an acidic solution. In our previous work with the










triphenoxymethane ligand, this base has been shown to favor a conformation with the

three phenolic oxygen atoms orientating themselves in an "all up" (Figure 1-8, Chapter 1)

conformation relative to the central methine hydrogen both in the solid-state and in

solution. So Tethering three CMPO moieties to this platform via these phenol oxygens

satisfies the requirement for proximate metal binding with CMPO groups. With the

triphenoxymethane platform, the alkyl group on the ortho-position of the phenol

moderates the solubility of the platform in organic solvents, as well as, exerts an

influence on the extended arms. Large, bulky groups such as tert-pentyl increase the

solubility but also restrict the flexibility of the three arms tethered to the phenolic

oxygens. In order to compare the properties of different variants of our ligand system,

two new tris-CMPO derivatives were synthesized (2-6a and 2-6c). The CMPO moieties

were tethered to the altered platform using well-established methodology developed for

the CMPO-calix[4]arene systems as outlined for the compounds 2-6a, b and c in Figure

2-5.49,75


o2N O




IIINH2 HN

OH H HH O H


SR1 3R3RR
R2 R2 R2 R2
2-2a, b, c 2-3a, b, c 2-4a, b, c 2-6a, b, c
a: R R2 =t-Pentyl
b: R R2 =t-Bu
c: R, = Me, R2 = t-Bu
Figure 2-5. Synthesis of tris-CMPO 2-4. (A) Nal, K2CO3, chloroacetonitrile, refluxing
acetone, 3 days; (B) LAH, diethyl ether; (C) p-nitrophenyl
(diphenylphosphoryl) acetate (2-5), 45-50 oC chloroform.









The results have shown a significant solubility variation within the studied t-pentyl,

t-Bu, and Me derivatives (2-6a, b, and c). All three compounds are readily soluble in

most common organic solvents such as dichloromethane, THF or methanol, but in less

polar solvents, the solubility of the methyl derivative (2-6c) is significantly limited. Of

the three, only t-pentyl derivative, 2-6a, is compatible with 1-octanol. Despite the

solubility differences and steric variations in the platforms, the modifications at the 2

position of the phenols do not significantly affect the affinity of tris-CMPO ligands for

Th(IV) as presented in Table 2-1. There does appear, however, to be a small increase in

the affinity for the lighter lanthanides as the size of the alkyl groups at the 2-position

decreases, and 2-6c exhibits a slightly higher affinity for La(III) in comparison to 2-6a

and 2-6b.

Table 2-1. Distribution coefficients (D)a and extraction percentage (%E)b,c for ligands:
2-6a, 2-6b and 2-6c. Aqueous phase: 10-4 metal nitrate in 1 M HNO3, Organic
phase: 10-3 M ligand in methylene chloride.
Ligand 2-6a 2-6b 2-6c
Cation D %E D %E D %E
Th (IV) >100 100 49.000 98 99.000 99
La (III) 0.031 3 0.042 4 0.111 10
Ce (III) 0.010 1 0.053 5 0.099 9
Nd (III) 0.042 4 0.031 3 0.087 8
Eu (III) 0.020 2 0.031 3 0.042 4
Yb (III) 0.042 4 0.111 10 0.063 6
aD calculated based on the E % values.
bE % = 100%([Mn+]org/[Mn+]total) after extraction as determined by Arsenazo(III) assay.
cMean value of at least four measurements. The precision: o(n-1) = + 1 or 2, where ocn-1) is
a standard deviation from the mean value.

2.2.2 Linand Flexibility vs. Binding Profile

Extensive work on "CMPO-like" molecules utilizing calix[4]arene as a platform

has found that there is a significant influence of ligand flexibility on its extraction

performance.76 Calix[4]arene extractants showed a strong increase of the extraction










percentage with increased length of the spacer between the amido and phenoxy group.

Thus, as proposed by the authors,76 One can anticipate a direct correlation between the

size of the cavity and the flexibility of the ligand, and its affinity for a particular cation.

The increased flexibility of the molecule should indeed allow for better accommodation

of metal ions due to the ability of the molecule to form a more appropriate cavity size.

To exercise this postulate, derivatives of the tris-CMPO system with an extended arm

length between the platform and the CMPO donors were synthesized. Methodology to

isolate a primary amine with three carbon spacer to the phenolic oxygen of the

triphenoxymethane was adapted from the preparation of calix[4]arene-based extractants

(Figure 2-6).76




'o

o H o
O N ONH2 H


OH H O H H O H


R 3B 3R 3
R R R R
2-2a, b 2-8a, b 2-9a, b 2-10a, b
a: R =t-Pentyl
b: R =t-Bu


Figure 2-6. Synthesis of tris-CMPO 2-10. (A) N-(3-bromopropyl)phthalimide, CS2CO3,
80-85 oC DMF, 6 days; (B) hydrazine mono hydrate, refluxing ethanol, 24h;
(C) p-nitrophenyl(diphenylphosphoryl)acetate (2-5), 45-50 oC chloroform, 3
days.

Products 2-10a and 2-10b were obtained via alkylation of the triphenoxymethane

platform with N-(3 -bromopropyl)phthalimide in the presence of cesium carbonate. The

subsequent treatment with hydrazine in ethanol to remove the phthalimide afforded 97%










of primary amine 2-9a and 92% of 2-9b, respectively. Final products were obtained in

70% for 2-10a and 49% for 2-10b yields.

The results of the extraction experiment revealed an anticipated increase in the

affinity of ligand 2-10a over more rigid 2-6a for the studied metal ions although, without

any expected significant decrease in Th(IV) selectivity (Figure 2-7).


100- 26
80-1 a 2-10a


40-,



Th(lV) La(lll) Ce(lll) Nd(lll) Eu(lll) vb(lll)

Figure 2-7. Metal extraction percentages (%E) for the ligands 2-6a, and 2-10a using 10-4
M metal nitrate in 1 M nitric acid and 10-3 M ligand in methylene chloride.

2.2.3 Complexation Studies with Bis-CMPO Compound

Ph PhN


Ph- OO C'

Figure ~ ~ ~ ~ 2-.Darm fte10codnt 2cto'nictoimIV irt omlxo-







[inur caseDaga of Th(V) nitrodiate ions geeatingdctonic hru(V iat complexes nve of th fc ta





31P NMR and FT-ICR-MS have confirmed the existence of these species in solution, the




































































L_~L _.


Th(lV) La(lll) Ce(lll) Nd(lll) Eu(lll) vb(lll)


Figure 2-10. Metal extraction percentages (%E) for the ligands 2-6a, and 2-7 using 10-4
M metal nitrate in 1 M nitric acid and 10-3 M ligand in methylene chloride.


r


affinity of the organic phase for these charged complexes might be limited. Therefore,


the bis-CMPO ligand (2-7) was synthesized using procedures similar to Figure 2-5.


Figure 2-9. Bis-CMPO compound 2-7.

With only two CMPO arms available for the metal, we envisioned that there should


be more space in the coordination sphere of the metal for the binding of an additional one


or two nitrates counterions. With the resulting reduction in positive charge of the


complex, the material would have enhanced solubility in the organic phase. The

extraction results revealed, however, that the reduction of the number of the binding units


to two severely diminished the effectiveness of 2-7 for the extraction (Figure 2-10).


O2-6a
a 2--7


I


100

80 -

60-



20-e









In fact, the bis-CMPO ligand showed very similar extraction behavior to the simple

CMPO extractant (2-1) (Table 2-2). Ligand 2-7 had very low affinity for Th(IV), as well

as, the series of lanthanides. Apparently, three preorganized CMPO arms are essential to

fulfill the geometry requirements around the metal center and afford an appreciable

extraction percentage.

Table 2-2. Distribution coefficients (D) and extraction percentage (%E) for ligands: 2-1,
2-7 and 2-10a. Aqlueous phase: 10-4 M metal nitrate in 1 M HNO3, Organic
phase: 3 x 10-3 M of 2-la, and 10-3 M of 2-7 and 2- 10a in methylene chloride
Ligand 2-7 2-1 2-10a
Cation D %E D (%E) D %E
Th (IV) 0.042 4 0.020 2 >100 100
La (III) 0.111 10 0.053 5 0.190 16
Ce (III) 0.064 6 0.064 6 0.190 16
Nd (III) 0.064 6 0.064 6 0.176 15
Eu (III) 0.064 6 0.075 7 0.163 14
Yb (III) 0.031 3 0.087 8 0.149 13
aThree times higher concentration of classical CMPO was used to keep the same
concentration of lighting units in the organic phase

2.2.4 Attempts to Resolve the Tris-CMPO Solubility Issue

The choice of solvents represents one of the most important factors in the liquid-

liquid separation science. To achieve effective phase separation, non-polar solvents

should be used. Due to safety concerns, the high boiling and flash points, and the low

toxicity of the solvent are as equally important as the polarity for waste clean-up

operations. Therefore to find application in the nuclear waste decontamination,

compatibility of the tris-CMPO ligand system with aliphatic solvents would be highly

desirable, and this trait can be achieved either by altering the structure of the ligand itself

or by the addition of a synergist. The solubility studies of 2-6b confirmed that the

addition of TBP as a synergist indeed induces a defined increase in solubility of the

extractant. Unfortunately, the effect is only temporary and the tris-CMPO ligand









eventually precipitates from the solution. This phenomenon can most likely be attributed

to the formation of both intra- and intermolecular hydrogen bonds.

In the previously published" solid-state structure of 2-6b there are two strong

hydrogen bonds present between the amide hydrogens and the phosphoryl and carbonyl

oxygens on adjacent arms (P=0'Namide: 2.801 A+, C=0 .Namide: 2.798 A+), as well as, the

intermolecular hydrogen bonds between phosphoryl oxygens on one ligand and the

remaining amide hydrogens on an other (Figure 2-11).





















-Intramolecular hydrogen bonds
-Intermolecular hydrogen bonds


Figure 2. 11. Fragment of the crystal structure of 2-6b molecules forming hydrogen bond
connected network. (Crystals obtained by slow diffusion of pentane into
saturated solution of 2-6b in methylene chloride).

The crystal structure of ligand 2-10b presented in Figure 2-12 significantly differs

from the more rigid compound 2-6b. In place of the extended hydrogen bond molecular

network, all three arms form only intramolecular hydrogen bonds between phosphoryl

oxygens and adj acent amide hydrogens (P=0 'N: 2.809 A+) constructing a rigorously C3-










symmetric structure in the solid-state. The aliphatic solvent may additionally force these

aromatic molecules to aggregate, further decreasing their solubility resulting in

precipitation.








03'




02" N1"













Figure 2-12. Diagram of the solid-state structure of 2-10b (30% probability ellipsoids for
N, O and P atoms; carbon atoms drawn with arbitrary radii). For clarity, all
hydrogen atoms have been omitted.

To prevent formation of the problematic hydrogen bonds, the amides were

alkylated. The synthesis of the tertiary amide derivative of the tris-CMPO ligand was

rather challenging. Even though preparation of the secondary amines 2-12 via acylation

of 2-4a and 2-9b, and LAH reduction of 2-1 1 was quite straightforward, the final

acylation with p-nitrophenyl (diphenylphosphoryl)-acetate (2-5) took approximately two

weeks to complete in the case of compound 2-14a, and 5 days in the case of 2-14b,

presumably due to the steric congestion of the three arms. A modified two-step

procedures was employed involving acylation of the secondary amine with chloroacetyl









chloride and subsequent Arbusov reactionsl presented in Figure 2-13. The reaction of

chloroacetyl chloride with 2-12a afforded an 85% yield of 2-13a, and the same reaction

with 2-12b resulted in a 66% yield of compound 2-13b. The final product of Arbusov

reaction between molecule 2-13a and ethyl diphenylphosphinite resulted in a 70% yield

of product 2-14a, while the same reaction with 2-13b gave product 2-14b with a 57%

yield.




O O

NH2 NH NH N- N-

Han O H HO H O H


R3R3R R R3

2-4a, 2-9b 2-11a, b 2-12a, b 2-13a, b 2-14a, b
a: n= 1, R = t-Pentyl
b: n =2, R =t-Bu

Figure 2-13. Synthesis of tris-CMPO 2-14. (A) ethylchloroformate, K2CO3,
dichloromethane, 3 days, rt; (B) LAH, THF, 5 days, rt; (C) chloroacetyl
chloride, K2CO3, 24h, refluxing dichloromethane; (D) ethyl
diphenylphosphinite, 150 oC, 40 h.

As shown in Figure 2-14, the polar cavity of the ligand was replaced by

hydrophobic interactions between three methyl groups located on the amidic nitrogens

which caused all three carbonyl oxygens to point outward. It was anticipated that such a

structural alteration would not only enhance the solubility of the extractant in non-polar

solvents, as desired, but perhaps even improve the extractability of Ln(III) and An(III),

due to increased basicity of the carbonyl oxygens.





























Figure 2-14. Diagram of the solid-state structure of 2-14a (30% probability ellipsoids for
N, O and P atoms; carbon atoms drawn with arbitrary radii). For clarity, all
hydrogen atoms have been omitted.

The alkylation of tris-CMPO has significant impact on the extraction properties of

the ligand. Surprisingly, both compounds 2-14a and 2-14b showed lower selectivity for

tetravalent thorium than their nonalkylated counterparts (Figure 2-15 and Table 2-3).

Table 2-3. Distribution coefficients (D) and extraction percentage (%E) for ligands: 2-14a
and 2-14b. Aqueous phase: 10-4 M metal nitrate in 1 M HNO3, Organic phase:
10-3 M of 2- 14a and b in methylene chloride
Ligand 2-14a 2-14b
Cation D %E D %E
Th (IV) 0.923 48 0.639 39
La (III) 0.064 6 0.064 6
Ce (III) 0.052 5 0.075 7
Nd (III) 0.064 6 0.087 8
Eu (III) 0.064 6 0.064 6
Yb (III) 0.075 7 0.075 7

The complexity of the extraction process does not allow for unambiguous

explanation of such behavior, but if the nitrate counterions are bound via hydrogen bond

interactions to the amidic hydrogens as it was observed in the solid state structures of the










metal complexes, presence of these hydrogens may be crucial for the transfer and

stability of the complex in the organic phase.

100r
m2--1
80J~ m 2-14a
0 2-6a

E%/


20C

0 agp -2-6a

Thl) e(lll) Nd(l) 2-1I) ( II


Figure 2-15. Metal extraction percentages (%E) for the ligands 2-1, 2-6a, and 2-14a using
10-4 M metal nitrate in 1 M nitric acid and 10- M ligand in methylene
chloride.

In the solid-state structure of the Th(IV) complex with 2-6a, the two nitrate anions

are positioned at distances of 2.91 1 A+ and 2.936 A+ from the amide nitrogens, suggesting

a weak hydrogen bonding interaction" The decrease in extraction ability for 2-14a was

especially pronounced in the case of Th(IV), for which the extraction was reduced over

50 percent with respect to the performance of the non-alkylated extractants (Figure 2-15).

In the case of the more flexible ligand 2-14b, a noticeable decrease of extraction (with

respect to non-alkylated 2-10a) along the entire series of studied metal ions was observed.

As a result, N-alkylated compounds 2-14a (shorter CMPO tripod linker) and 2-14b

(longer CMPO tripod linker) display a similar selectivity profile (Table 2-3).

A similar tendency for the decrease in affinity for the tetravalent ion by alkylated

extractant was observed in the case of Pu(IV). Upon amide alkylation, the D value drops

significantly from 43.60 to 2.60 at 10-3 M ligand concentration in dichloromethane.

However, a ten fold increase in ligand concentration restores the high distribution value









for Pu(IV) in the organic phase, also observed with the less rigid compound 2-14b (Table

2-4). In the case of the amide alkylated derivatives of CMPO bearing calixarenes, an

even more severe extraction decrease was reported. 53

Table 2-4. Distribution coefficients (D) for the extraction of Pu(IV), U(VI), Am(III) and
Eu(III) by ligands: 2-6b, 2-14a and 2-14b in methylene chloride and 1-octanol.
Ligand CL Solvent D Pu(IV) D U(VI) D Am(III) D Eu(IIII)
2-6b 10-3 M CH2 2, 43.60 -0.371a 0.412a
2-14ab 10-3 M CH2 2, 2.60 0.33
2-14bb 10-3 M CH2 2, 6.74 3.78
2-14b 10-2 M CH2 12 33.1 4.85 0.41 0.21
2-14b 10-2 M 1-octanol 6.1 1.54 0.058 0.025
alM HNO3/5M NaNO3
bPu(IV) 10-' M, U(IV) 10-4 M in 1 M HNO3

While alkylation of 2-6a and 2-10b resulted in the deteriorated solubility of their

derivatives 2-14a and 2-14b in non-polar solvents such as diethyl ether, the alkylation of

the more flexible compound (2-10b) made 2-14b compatible with 1-octanol. This

solubility improvement provided an opportunity to test the extraction potential of the tris-

CMPO ligand in diluent other than dichloromethane.

As it was observed in the case of "CMPO-like" calixarenesn7 the binding potential

of the tris-CMPO ligand was strongly decreased in 1-octanol. At the same molar

concentration of ligand 2-14b in both solvents, the affinity for metal ions was much lower

in 1-octanol than in dichloromethane (Table 2-4 and Figure 2-16). In 1-octanol the

hydrogen bonding interactions between the solvent and the phosphoryl and carbonyl

oxygens of the extractant may mitigate the extraction either by preventing ligand-metal

binding or by simply changing the stability of the complex in the organic phase. Efforts

to determine the structure of the extracted species with both alkylated and non-alkylated










ligands in solution, which would allow for the better understanding of the extraction

behavior of ligands, are in progress.

2.2.5 Plutonium (IV) and Americium(III) Extractions

The light actinides can be separated from the lanthanides due to the favored

extractability of their higher oxidation states. Since the results of extraction experiments

have proven the ability of ligands 2-6a, b and c to take advantage of differences in

oxidation states of tetravalent thorium and a series of trivalent lanthanides, we were

prompted to study the extraction of the Pu(IV) ion, to verify the ability of the tris-CMPO

molecule to preferentially extract tetravalent light actinides other than thorium.

45~
40 m 2-14a (1)
aI 2-14b (3)
35- m 2-14b (1)
30- m ----_ 2-14b (2)
25- 0 ----- 2-6b (1)
20C
15C
10C
5 2- 246bb 9))
2-14b (1)
O~ 2-14b (3)
(M U 2-14a (1)


Figure 2-16. Metal extraction percentages (%E) for the ligands 2-6b, 2-14a, and 2-14b
using 10-3 M ligand in methylene chloride (1), 10-2 M ligand in methylene
chloride (2), 10-2 M ligand in 1-octanol (3).


The tris-CMPO ligand was found to be significantly more effective for Pu(IV)

separation than an industrial mixture of mono-CMPO and TBP in the transuranium

elements extraction process (TRUEX).13 In fact, much lower concentrations of the tris-

CMPO ligands were required to achieve approximately the same distribution ratio (D) of

Pu(IV) as the mono-CMPO/TBP extraction system. The tris-CMPO ligands are also









much more selective than the CMPO/TBP mixture. After 24 hours of extraction, the

97.76 % (D = 43.60) of Pu(IV) was removed from the aqueous phase by ligand 2-6b at

concentration as low as 10-3M (Table 2-4) while the D values for all the lanthanides were

significantly below 1. To reach a similar distribution coefficient for the extraction of

Pu(IV) a solution of 0.2 M CMPO mixed with 1.0M TBPl3,82 WOuld be required and this

mixture would extract a significant amount of the trivalent lanthanides.

In the early fifties, Seaborg and coworkers40 found that trivalent lanthanides and

actinides could be separated using cation-exchange resin columns due to the ability of the

actinides to employ 5f orbitals in bonding. This idea was based on similar spatial

extensions of the 5f, 6d, 7s and 7p orbitals of the trivalent actinides, especially the

lightest ones.42 Since the radial distribution of 4f orbitals are severely limited, this subtle

difference could be exploited with the appropriate ligands to facilitate the separation. 42,83

With these issues in mind, the Am(III)/Eu(III) extraction experiment was performed to

test the ability of 2-6b to take advantage of such discrepancy between Am(III) and the

similarly size Eu(III) ion (Table 2-4). It was found however, that 2-6b has a very low

affinity for both ions (D below 0.02), and it is not able to distinguish between these two

elements. Upon enrichment of the aqueous phase with sodium nitrate, the distribution

coefficients (D) for Am and Eu slightly improved, as expected, but neither the D values

nor the separation factor were satisfactory.

2.2.6 Comparison of Solid State Structures of Tris-CMPO Complexes of Trivalent Metal
lons

The solvent extraction separation process is based on the transfer of a metal cation

from an aqueous phase into an immiscible organic phase with simultaneous charge

neutralization.42 As we have previously shown in organic solvents," the three CMPO










arms in ligand 2-6b tightly wrap around Ln(NO3)3 (Ln = Eu(III), Nd(III) ) in a bidentate

fashion forcing two of nitrate counterions out of the coordination sphere of the metal and

producing a complex with an overall 2+ charge. In the solid-state structure of the Nd(III)

species with 2-6b two distinct Nd(III) complexes co-crystallized. One metal center is

eight coordinate while the other Nd(III) contains an extra coordinated water molecule.

As expected, all of the metal oxygen bond lengths are slightly longer in the nine

coordinate species.

01 Ph Ph Ph Ph

Ph ONP
Ph PP,~O OP Ph O O" P\
O PPh Ph)P OP



Figure 2-17. Diagrams of neodymium(III) complexes of 2-6b: anhydrous 8 coordinated
+2 cationic complex (left) and water-containing 9 coordinated +2 cationic
complex (right).

Similar, although only 8 coordinate structures were obtained for Tb(III) complexes

with new derivatives of tris-CMPO chelates 2-6a and 2-14a and Bi(III) complex with

2-6c (Figures 2-18, 2-19, 2-20). Unlike Tb(III) complexes, the Bi(III) compound

contained two similar structures in the asymmetric unit. Our interest in the chemistry of

bismuth originates from the presence of significant amount of bismuth in waste generated

by the bismuth phosphate process popular in early 40's for plutonium and uranium

separation.84 In addition to solid state structure analysis, the affinity of ligands 2-6b and

2-6c toward trivalent bismuth was tested to ensure selectivity of the tris-CMPO chelate

over other than f-element trivalent metal ions present in waste. The extractability of

Bi(III) was found to be as low as the trivalent lanthanides. At the ten fold excess of

ligands 2-6b or 2-6c in dichloromethane only 8% of bismuth was transferred to the










organic phase, promising an extended application of our extraction system over high

bismuth content radioactive waste.














Figure 218. Diaram of te strucure of 12-cBN 3(0) 0 poait elsid
forBi NO ndP aom; crbn tom daw wih rbiray adi)11l
hyroenatmshaeben mite adth bsmthliad ons a07be
drawnwith olid in Of

In~~~ ~ 5h Boi tt tutr h ifrn aii o h tde ta (rvln ,

Tb,~~~~ ~ ~ 08 eesrnl elce ntevraiosi h odlntsi h eutn

comle. hesmllstan terfoe hemos eecroostie 8cordnae b(I2












Figure 2-18. Diagram of the structure of [2-6cTBiNO3](NO3)2: (30% ()231) prbbltyelpods

CN=8: fo Bi, N)2.59, O nd PN9 atoms;-.46(0) cabnaom rw ith) arbtray rdi)-.41() All

hydrogen3946) ato-ms h21ave ben0]N 2 omite ad te isut-lignd ons hveee










[2-8a-TbNO3](NO3)2: 2.347(4)-2.397(4), Nd CN=8: 2.358(9)-2.437(9), Nd CN=9:

2.430(10)-2.490(1 0), Bi(1): 2.444(7)-2.575(6), Bi(2): 2.464(6)-2.533(6)).

The nitrate coordination strength also follows the trend: Tb > Nd CN=8 > Nd

CN=9 > Bi ([2-6a-TbNO3](NO3)2: 2.468(4)-2.480(4), [2- 14a-TbNO3](NO3)2: 2.474(8)-

2.599(5), Nd CN=8: 2.53 5(10)-2.5 52(10), Nd CN=9: 2.615(10)-2.621(1 0), Bi(1):

2.444(6)-2.519(6), Bi(2): 2.458(5)-2.587(6)). The Bi-O separation in this 8 coordinate

compound was found to be very similar to the average Bi-O distance in the neutral 8 and

9 coordinate bismuth nitrate complexes with tridentate 2,6-bis(-CH2-P(=0)R2) pyridine

oxides [CN=8 (2.321 A+); CN=9 (2.340 A,)],86 and somewhat shorter from the distances in

the 9 coordinate nitrate complex with ('PrO)2(0)PCH2P(0)(O'Pr)2 [2,432(2)-2,544(2)

K]87













Figre2-9.Digra o te trctre f omoud 2-aTN0](001230
probably llpsid fr bN, 011 tos abo tm danw
arbitrary ~ ~ O rdi A hrOfe tm aebe mtedadtetrimlgn
bond hae ben dawnwit sold lne 5




































Figure 2-20. Diagram of the structure of compound [2-14a-TbNO3](NO3)2 (30%
probability ellipsoids for Tb, N, O and P atoms; carbon atoms drawn with
arbitrary radii). All hydrogen atoms have been omitted and the terbium-ligand
bonds have been drawn with solid lines.

The incorporation of the tertiary amide into the ligand scaffold seems to have very

small influence on the terbium coordination environment in the solid state complexes.

The distances between phosphoryl oxygens and terbium ions in [2-6a-TbNO3](NO3)2 and

[2-14a-TbNO3](NO3)2 are within the same range. The carbonyl oxygens and metal bond

lengths are only slightly shorter in [2-14a-TbNO3](NO3)2 COmplex, but the nitrate is

bound slightly weaker. Selected bond lengths of the molecular structures of 2-10b,

2-14a, [2-6a-TbNO3](NO3)2, [2-14a-TbNO3](NO3)2 and [2-6c-BiNO3](NO3)2 determined

by the single crystal X-ray diffraction analysis are summarized in Table 2-5.











[2-6c-BiNO3] [2-6c-BiNO3]
[2-6a-TbNO3] [2-14a-TbNO3]
2-10b 2-14a (NO3)2 (NO3)2
(NO3)2 (NO3)2
Bi(1) Bi(2)


Table 2-5. Selected bond lengths (A+) for compounds:
[2-6c-BiNO3](NO3)2.


2-10b, 2-14a, [2-6a-TbNO3](NO3)2, [2-14a-TbNO3](NO3)2 and


1.497(6)
1.502(7)
1.497(6)


1.506(6)
1.520(7)
1.498(7)


P(1)-O(5)
P(2)-O(7)
P(3)-O(9)
P(1)-O(3)
C(19)-O(2)
C(53)-O(4)
C(70)-O(6)
C(87)-O(8)
C(52)-O(4)
C(68)-O(6)
C(84)-O(8)
M-O(5)
M-O(7)
M-O(9)
M-O(4)
M-O(6)
M-O(8)
M-O(11)
M-O(10)


1.480(3)
1.485(3)
1.484(3)


1.502(5)
1.497(5)
1.503(5)


1.505(7)
1.510(7)
1.520(7)


1.477(3)
1.232(4)


1.230(4)
1.233(4)
1.231(4)


1.289(12)
1.260(11)
1.240(12)


1.261(7)
1.232(8)
1.248(7)
2.315(5)
2.308(5)
2.300(4)
2.347(4)
2.349(5)
2.397(4)
2.468(4)
2.480(4)


2.322(6)
2.315(7)
2.297(7)
2.326(8)
2.368(7)
2.3 73(7)
2.599(9)
2.474(8)


2.315(6)
2.334(6)
2.414(6)
2.450(6)
2.444(7)
2.575(6)
2.444(6)
2.519(6)


2.352(5)
2.330(6)
2.394(6)
2.464(6)
2.469(7)
2.533(6)
2.458(5)
2.587(6)









2.3 Conclusions

A series of molecules containing three precisely arranged

carbamoylmethylphosphine oxide (CMPO) moieties have been synthesized and their

ability to selectively extract actinides from simulated acidic nuclear waste streams has

been evaluated. The ligand system has shown an excellent binding efficiency for An(IV)

and Pu(IV) in particular.

As in the case of calix[4]arene and resorcinarene derivatives, the extraction

efficiency of classic (N,N-dii sobutylcarbamoylmethyl) octylphenylphosphineoxide

(CMPO) extractant was strongly improved by the attachment of CMPO-like functions on

the triphenoxymethane skeleton. The unique geometrical arrangement of the three

lighting groups, as opposed to four, dramatically changed the selectivity profie of this

multi-CMPO extractant. To the best of our knowledge, the tris-CMPO is the most

effective CMPO-based system for the selective recognition of tetravalent actinides,

plutonium in particular. The structural modifications oftris-CMPO have shown a

significant decrease in the extraction efficiency and selectivity with the introduction of a

tertiary amide into the ligand structure. On the other hand, elongation of the secondary

amide lighting arm slightly enhanced the extractability of all the studied ions without a

maj or reduction in selectivity. A reduction in the number of chelating groups from three

to two produced a very ineffective ligand demonstrating the significance of the presence

of exactly three preorganized chelating moieties for the efficient metal binding. This

remarkable attraction for tetravalent actinides can be attributed to the match between

coordination requirements of An(IV) and the geometrical environment presented by the

ligand. Moreover, the higher charge density of tetravalent actinides with respect to

trivalent f-elements may additionally account for the increased affinity of the tris-CMPO









for An(IV) and the complex stability. As a result, the ligand shows promise as an

improved extractant for Pu(IV) recoveries from high level liquid wastes, especially of

PUREX origin and in general actinide clean-up procedures. Immobilization of this

plutonophile on a solid support may offer a very efficient and cost-effective technique for

future plutonium separation and recycling. Efforts to produce these ligands are

underway.

2.4 Experimental Section

2.4.1 General Consideration

The lanthanide and actinide salts, La(NO3)3-H20 (Alpha Aesar), Ce(NO3)3-6H20,

Nd(NO3)3-6H20, Eu(NO3)3-5H20, Tb(NO3)3-6H20, Yb(NO3)3-5H20 (Aldrich),

Bi(NO3)3-5H20 (Acros) and Th(NO3)4-4H20 (Strem), were used as received. The

solutions were prepared from 18 MGZ Millipore deionized water, TraceMetal grade HNO3

(Fisher Scientific), and HPLC grade organic solvents. The Arsenazo(III) assay was

performed on a Varian Cary 50 UV/vis spectrophotometer while a 2500TR Packard

liquid scintillation analyzer was used for counting alpha-emitting Pu and U radionuclides.

A Canberra GammaTrac 1185 with Ge detector and AccuSpec-B multi-channel analyzer

was used for 241Am and 152Eu counting. All 1H, 13C and 31P NMR spectra were recorded

on a Varian VXR-300 or Mercury-300 spectrometer at 299.95 and 121.42 MHz for the

proton and phosphorus channels, respectively. Elemental analyses were performed by

Complete Analysis Laboratories, Inc. in Passipany, New Jersey. Mass spectrometry

samples were analyzed by Dr. Lidia Matveeva and Dr. Dave Powell at the University of

Florida on a Bruker Apex II 4.7T Fourier transform ion cyclotron resonance mass

spectrometer. Fast atom bombardment (FAB), ionization energy (IE) and liquid









secondary ion mass spectrometry (LSIMS) mass spectra were recorded on Finnigan

MAT95Q Hybrid Sector.

2.4.2 Metal lons Extractions

The lanthanides, bismuth and thorium extraction experiments followed previously

reported procedure. 51,76,88 The extractability of each cation was calculated as

%E = 100/(A1 A)/(A1 Ao), where A is the absorbance of the extracted aqueous phase

with the Arsenazo(III) indicator, Al is the absorbance of the aqueous phase before

extraction with the indicator, and Ao is the absorbance of metal-free 1 M nitric acid and

the indicator (hLn(III)= 655, hsi(III), Th(IV)= 665 nm). The errors, based on the precision of

the spectrophotometer and the standard deviation from the mean of at least three

measurements, were in most cases no higher than two percent. The distribution

coefficients (D) of cations defined by equation: D = E [Morg] / E [Maq] (where E [Morg ,

C [Maq] are the total concentration of the metal species in the organic and aqueous phase,

respectively) were calculated using formula: D = %E / (100 %E). In view of the errors

associated with the spectrophotometric technique, the maximum value that could be

measured for the extraction percentage and distribution ratio was 99.

In case of 239Pu, 238U, 241Am and 152Eu isotopes, a solution of ligand, 10-3-10-2M in

CH2 12, Or 1-octanol has been contacted with 1 M HNO3 COntaining radioactive nuclides.

Test tubes containing 1-2 mL of the aqueous phase and an equal volume of the organic

phase were placed into an orbital shaker at 20 OC. After a certain period of time, a test

tube was removed from the rotator, the layers were allowed to separate, and then were

transferred into the shell vials. Equal aliquots of the organic and the aqueous phases were

taken for counting and distribution coefficient determination.









Experiments with radioactive elements were performed by Dr. Artem Gelis at

Argonne National Laboratory.

2.4.3 Isotopes Stock Solutions

Plutonium stock solution The stock solution of tetravalent 239Pu was purified

using anion exchange method.89 The method is based on the retention of Pu(NO3)62- in

7.5 M HNO3 by the anion exchanger (Bio-Rad AG-1 x 8). All the cations were eluted

from the column with 7.5 M HNO3. Subsequently, Pu was stripped from the column with

a solution of 0.3 M HNO3 at 60 oC. The strip was taken to a wet salt with a few drops of

concentrated HCIO4 to destroy any organic impurities. The resulting Pu(VI) salt was

dissolved in 1 M HNO3 and was treated with 0. 1 mL of concentrated H202 to convert Pu

to tri- and tetravalent oxidation states.89 To oxidize Pu(III) to Pu(IV), 20 mg of solid

NaNO2 WAS added to the solution. An electron absorption spectrum, collected in 400-900

nm range, revealed no evidence of either Pu(III) or Pu(VI). The solution was diluted

with 1 M HNO3 to get 10-' M Pu(IV) and was used for the experiments.

Uranium stock solution A weighted amount of 238U30s of analytical purity was

dissolved in 14.7 M HNO3 and then diluted with water to 1.6 M UO2(NO3)2 in 1 M

HNO3. The solution was then diluted with 1 M HNO3 to 0.01 M UO2(NO3)2 and was

spiked with 233U(VI) (T%=1.59x105 y) to increase the effectiveness of the liquid

scintillation counting.

Americium stock solution The purity of the 241Am stock solution was determined

by the ICPMS analysis. The solution was evaporated to dryness and re-dissolved in 1 M

HNO3 .









Europium stock solution The 152Eustock was purchased from Isotope Products

Laboratories. The original 0.5 M HCI solution was taken to a wet salt with nitric acid

twice and was then redissolved in 1 M HNO3.

2.4.4 Synthesis

The synthetic methodology for the preparation of 2-6 and 2-10 has been adapted

from procedures developed in work with phenols and calix[n]arene platforms. 49,75,90,91

Detailed synthetic procedures were previously reported for preparation of molecule 2-4c

and 2-6b. 50,51,92 The p-nitrophenyl (diphenylphosphoryl)acetate 2-5 was prepared

according to literature directions. 49 COmpounds 2-6a, 2-10a, b, 2-12b, 2-14b used for

extraction experiments and their substrates were synthesized by Dr. Ajay Sah and Dr.

Priya Srinivasan.

Preparation of Tris(3,5-tert-pentyl-2-(cyanomethoxy)phenlmtae 2-3a.

Following the procedure described in reference" for 2-2b, a 10.70 g portion (15.00

mmol) of 2-2a was dissolved in dry acetone (200 mL) with 20.73 g (150.00 mmol) of

potassium carbonate, 22.48 g (150.00 mmol) of sodium iodide, and 7.59 g (120.0 mmol)

of chloroacetonitrile, and the solution was refluxed 48 hours under nitrogen. After the

solvent was removed in vacuo, the product was taken up in ether, dried with MgSO4,

filtered, and the solvent was removed. Recrystallization of the crude material from

ethanol afforded 8.91 g (72%) of product. 1H NMR (CDCl3): 6 = 0.55 (m, 18 H;

CH2CH3), 1.13 (s, 18 H; C-CH3), 1.36 (s, 18 H; C-CH3), 1.48 (q, J= 7.4, 6 H; CH2CH3),

1.74 (q, J= 7.5, 6 H; CH2CH3), 4.14 (s, 6 H; CH2CN), 6.17 (s, 1 H; CH), 7.04 (d, J= 2.6,

3 H; Ar-H), 7. 15 (d, J= 2.3, 3 H; Ar-H). 13C NMR (CDCl3): 6 = 9.4, 9.7, 28.6, 29.4,

35.5, 37.0, 38.1, 38.5, 39.5, aliphaticc); 57.8 (Ar-O-CH2); 115.0 (CN); 126.3, 127.5,









136.0, 141.5, 145.7, 151.8 (aromatic). Anal. Called for C55H79N303: C, 79.57; H, 9.59; N,

5.06. Found: C, 79.75; H, 10.07; N, 4.96.

Preparation of Tris(3 -methyl-5-tert-butyl-2-(cyanomethoxy)phenlmtae 2-3c.

Following the procedure described above, a 0.55 g (1.1 mmol) portion of 2-2c yielded

0.59g (89%) of pure product. 1H NMR (CDCl3): 6 = 1.18 (s, 27 H; Ar-C(CH3)3), 2.37 (s,

9 H; Ar-CH3), 4.14 (s, 6 H; Ar-O-CH2CN), 6.22(s, 1 H; C-H), 6.85 (b, 3 H; Ar-H), 7.12

(d, 3 H; Ar-H). 13C NMR (CDCl3): 6 = 17.1, 31.2, 34.3, 37.7, 57.2 aliphaticc); 115.5

(CN); 125.5, 127.4, 130.9, 135.0, 147.9, 151.4 (aromatic). Anal. Called for C40H49N303:

C, 77.51; H, 7.97; N, 6.78. Found: C, 77.68; H, 8.36; N, 6.71.

1,1'-bis(3 ,5-di-tert-butyl-2-(cynomethoxy)phenyl)etae As described for 2-3a, a

6.57 g (14.98 mmol) portion of 2,2'-ethylidenebis(4,6-di-tert-butylphenol to afford 4.20

g (54%) of product. 1H NMR (CDCl3): 6 = 1.18 (s, 18 H; CCH3), 1.34 (s, 18 H; CCH3),

1.60 (d, J= 6.9, 3 H; CHCH3), 4.52 (m, 4 H; CH2CN), 4.65 (q, J= 7.1i, 1 H; CHCH3),

7. 11 (d, J= 2.6, 2 H; Ar-H), 7. 18 (d, J= 2.6, 2 H; Ar-H). 13C NMR (CDCl3): 6 = 23.3,

31.6, 31.7, 32.3, 34.9, 35.7, aliphaticc); 58.9 (Ar-O-CH2), 115.7 (CN); 123.4, 124.2,

138.4, 142.5, 147.7, 152.2, (aromatic). Anal. Called for C34H48N202: C, 79.02; H, 9.36;

N, 5.42. Found: C, 80.39; H, 9.82; N, 5.44.

Preparation of Tris(3,5-di-tert-pentyl-2-(aminomethoxy)phnlmtae 2-4a As

outlined in reference," a diethyl ether solution of 2-3a (7. 12 g, 8.58 mmol) was added

dropwise over 30 min to a slurry of lithium aluminum hydride (4.36 g, 129.00 mmol) in

diethyl ether at 00 C. The mixture was allowed to warm to room temperature and stirred

for an additional 12-15 h. A 10 mL portion of 5% NaOH was slowly added to the slurry,

and the solution was allowed to stir for 30 minutes. The solution was dried with MgSO4,









filtered, and the solvent was removed in vacuo. The crude white solid was recrystallized

from acetonitrile to give to give 6.00 g (83%) of product. 1H NMR (CDCl3): 6 = 0.48 (m,

18 H; CH2CH3), 1.11 (s, 18 H; CCH3), 1.30 (s, 18 H; CCH3), 1.43 (q, J= 7.4, 6 H;

CH2CH3), 1.67 (q, J= 7.4, 6 H; CH2CH3), 2.89 (t, J= 5.0, 6 H; CH2CH2-NH2), 3.32

(t, J= 5.0, 6 H; O-CH2CH2), 6.38 (s, 1 H; CH), 7.00 (d, J= 2.6, 3 H; Ar-H), 7. 12 (d, J=

2.3, 3 H; Ar-H). 13C NMR (CDCl3): 6 = 9.3, 9.8, 28.8, 29.7, 35.3, 37.1, 37.9, 38.8, 39.4

aliphaticc); 42.8 (CH2- H2); 74.4 (O-CH2); 125.0, 128.2, 138.0, 140.1, 142.8, 152.9

(aromatic). Anal. Called for C55H91N303: C, 78.42; H, 10.89; N, 4.99. Found: C, 78.13;

H, 11.35; N, 4.66.

Preparation of Tris(3 -methyl-5-tert-butyl-2-(aminomethoxy)phenlmtae 2-4c.

Following the procedure outlined for 2-4a, a 6.46 g (10.0 mmol) portion of 2-3c gave 5.7

g (86%) of product. 1H NMR (CDCl3): 6 = 1.18 (s, 27 H; Ar-C(CH3)3), 2.26 (s, 9 H;

Ar-CH3) 2.87 (t, J= 5.1 Hz, 6 H; Ar-O-CH2), 3.31 (t, J= 5.2 Hz, 6 H; Ar-O-

CH2CH2 H2), 6.76 (s, 1 H; CH), 6.92 (d, J= 2.6 Hz, 3 H; Ar-H), 7.00 (d, J= 2.6 Hz, 3

H; Ar H). 13C NMR (CDCl3): 6 = 16.8, 31.3, 34.1i, 36.9, 42.4, 74.2 aliphaticc); 125.8,

125.9, 129.8, 136.3, 145.5, 152.3 (aromatic); LSI MS [M+H]' = 632.48

1,1'-bis(3 ,5-di-tert-butyl-2-(2-aminoethoxy)phenyleta. Lithium aluminum

hydride (1.49 g, 39.26 mmol) was suspended in dry ether (80 mL) and the reaction flask

was cooled to 0 oC. 1,1'-bis(3,5-di -tert-butyl -2-(cynomethoxy)phenyl)ethane (2.70 g,

5.23 mmol) was added in three portions with stirring. The reaction mixture was warmed

to room temperature and stirred overnight. The reaction was monitored by TLC,

(pentane:ether 80:20) and once completed, 5% sodium hydroxide solution (4.5 mL) was

added dropwise (ice bath) and the mixture was stirred until the suspension became milky









white. The resulting white solid was discarded through filtration and the organic layer

was dried over MgSO4. Removal of solvent under vacuum yielded analytically pure

product. Yield 2.30 g (84%). 1H NMR (CDCl3): 6 = 1.27 (s, 18 H, CH3), 1.42 (s, 18 H,

CH3), 1.67 (d, 3 H, J= 7.2 Hz, CHCH3), 3.15 (b t, 4 H, N-CH2CH2-O), 3.89 (m, 4 H, N-

CH2CH2-O), 4.70 (q, 1 H, J= 7.2 Hz, CH), 7.21 (s, 2 H, Ar-H), 7.24 (s, 2 H, Ar-H).

13C NMR (CDCl3): 6 = 23.95, 30.18, 31.57, 31.69, 32.33, 34.70, 35.64, 42.52 aliphaticc);

75.62 (O-CH2); 122.54, 124.69, 139.02, 141.80, 145.18, 153.66 (aromatic). Anal. Called

for C34H56N202: C, 77.81; H, 10.76; N, 5.34. Found: C, 78.32; H, 11.14; N, 5.08

Preparation of compound 2-6a. The synthesis of 2-6a followed the preparation

method for 2-6b51. A chloroform solution of 2-4a (2.92 g, 3.47 mmol) and p-

nitrophenyl(diphenylphosphoryl)acetate, 2-5, (4.16 g, 10.91 mmol) were stirred at 45oC

for three days. After cooling to room temperature, a 1 M solution of NaOH (100 mL)

was added and the mixture was stirred for 2 hours. The p-nitrophenol sodium salt was

extracted from the chloroform solution using 5% sodium carbonate (6 x 300 mL) and the

organic layer was further extracted with brine. The organic phase was dried with MgSO4,

filtered, and the solvent removed in vacuo to give 4.21 g (77%) of product as an off-white

solid material. 1H NMR (CDCl3): 6 = 0.46 (m, 18 H; CH2CH3), 1.06 (s, 18 H; CCH3),

1.23 (s, 18 H; CCH3), 1.40 (q, J= 7.3, 6H; CH2CH3), 1.57 (b, 6 H; CH2CH3), 3.41 (b, m,

18 H; O-CH2CH2-NHC(0)-CH2-P), 6.28 (s, 1 H; CH), 6.92 (d, J= 2.3, 3 H; Ar-H), 6.95

(d, J = 2.1i, 3 H; Ar-H), 7.38 (m, 18 H; P-Ar H), 7.72 (m, 12 H; P-Ar H), 7.89 (b, 3 H;

NH). 13C NMR (CDCl3): 6 = 9.3, 9.8, 28.7, 29.7, 35.5, 37.1, 37.8, 38.7, 39.3, 39.5

aliphaticc); 40.4 (CH2- H2); 70.4 (O-CH2); 125.0, 127.7, 128.7, 128.9, 131.2, 131.3,

132.1, 132.2, 137.8, 139.9, 142.8, 153.2 (aromatic); 165.4; 165.5 (C=0). 31P NMR










(CDCl3): 6 = 29.8. Anal. Called for C97H124N309P3: C, 74.26; H, 7.97; N, 2.68. Found:

C, 74.59; H, 8.14; N, 2.70.

Preparation of compound 2-6c. Following the procedure described above for 2-6a,

a 2.60 g (4. 11 mmol) portion of 2-4c was reacted with 4.90 g (12.85 mmol) of 2-5 to

afford 1.97g (35%) of product. 1H NMR (CDCl3): 6 = 1.14 (s, 27 H; Ar-C(CH3)3), 2. 12

(s, 9 H; Ar-CH3), 3.29 (b, 12 H; Ar-O-CH2CH2), 3.49 (d, J(H,P) = 13.9 Hz, 6 H; CH2-

POAr2), 6.65 (s, 1 H; CH), 6.84 (d, J= 2.4 Hz, 3 H; Ar-H), 6.94 (d, J= 2.5 Hz, 3 H; Ar-

H), 7.38 (m, 12 H; P-Ar H), 7.47 (m, 6 H; P-Ar H), 7.76 (m, 12 H; P-Ar H), 8.00 (b, 3 H;

N-H). 13C NMR [CDCl3]: 6 = 16.79, 31.2, 34.0, 38.7, 39.5, 40.0 aliphaticc); 70.6 (O-

CH2); 152.1, 145.6, 135.9, 132.7, 131.9, 131.4, 131.0, 130.9, 129.7, 128.6, 128.4,126.0,

125.5(aromatic); 165.1, 165.09 (C=0). 31P NMR (CD30D): 6 = 30.77. HR ESI-ICR MS

(sample injected as solution in 1% HNO3/MeOH): [M+H]+ = 1358.62. Anal. Called for

C82H94N309P3: C, 72.49; H, 6.97; N, 3.09. Found: C, 72.37; H, 6.97; N, 3.38.

Preparation of compound 2-7. The synthesis of 2-7 was adapted from that

described above for 2-6a, and a 1.15 g (2.19 mmol) portion of 1,1'-bis(3,5-di-tert-butyl-

2-(2-aminoethoxy)phenyl)ethane was treated with 1.76 g (4.62 mmol) of 7 to afford 2.00

g (90%) of product. 1H NMR (CDCl3): 6 = 1.25 (s, 18 H, CH3), 1.33 (s, 18 H, CH3), 1.43

(d, 3 H, J= 8.6 Hz, CHCH3), 3.42 3.71 (several multiplets, 8 H, P-CH2-C(0) + N-

CH2CH2-O), 3.99 (m, 4 H, N-CH2CH2-O), 4.70 (q, 1 H, J= 6.6 Hz, CH), 7. 15 (b s, 2 H,

Ar-H), 7.18 (b s, 2 H, Ar-H), 7.27 7.47 (m, 12 H, P-Ar), 7.78 (b m, 8 H, P-Ar), 8.27 (b,

3 H, NH). 13C NMR (CDCl3): 6 = 24.07, 31.59, 34.62, 35.51, 38.89, 39.69, 40.26

aliphaticc); 71.94 (O-CH2); 122.46, 124.10, 128.69, 128.85, 130.99, 131.12, 132.23,

132.61, 139.40, 141.65, 141.65, 145.36, 153.22 (aromatic); 165.36 (C=0). 31P NMR










(CDCl3): 6 = 30.41. Anal. Called for C62H78N206P2: C, 73.78; H, 7.79; N, 2.78. Found:

C, 73.85; H, 7.98; N, 2.73.

General procedure for synthesis of compounds 2-8. A stirring suspension of

triphenoxymethane molecule (2-2),5o N-(3-bromopropyl)phthalimide and cesium

carbonate in DMF was heated to 80-85 oC for six days. The reaction mixture was cooled

to room temperature and poured into cold water resulting in formation of a white solid

product. The mixture was transferred to a separation funnel and extracted with diethyl

ether. The solid product suspended in the ether layer and was collected by filtration and

dried.

Compound 2-8a. Using a 10.01 g (14.04 mmol) portion of tris(3,5-di-tert-pentyl-2-

hydroxyl)methane (2-2a) afforded 1 1.53 g (64%) of product. 1H NMR (CDCl3): 6 = 0.50

(m, 18 H; CH2CH3), 1.13 (s, 18 H; CCH3), 1.32 (s, 18 H; CCH3), 1.45 (q, J= 7.4, 6 H;

CH2CH3), 1.70 (q, J= 7.5, 6 H; CH2CH3), 2. 12 (b, 6 H; CH2CH2CH2), 3.50 (b, 6 H;

CH2CH2CH2-N), 4.01 (m, 6 H; O-CH2CH2CH2), 6.42 (s, 1 H; CH), 7.01 (d, J= 2.3, 3 H;

Ar-H), 7. 10 (d, J= 2.1i, 3 H; Ar-H), 7.45 (m, 6 H; Ar-H), 7.54 (m, 6 H; Ar-H). 13C NMR

(CDCl3): 6 = 9.3, 9.8, 28.8, 29.6, 29.7, 35.1, 36.3, 37.1, 37.9, 39.4 aliphaticc); 69.8 (O-

CH2); 122.9, 124.9, 128.1, 132.8, 132.9, 133.3, 138.3, 140.1, 142.5, 153.4 (aromatic);

168.3(C=0). FAB MS m/z = 1274.77 [M + H] .

Compound 2-8b. The material was obtained in 89% yield. 1H NMR (CDCl3): 6

1.19 (s, 27 H; CCH3), 1.34 (s, 27 H; CCH3), 2.19 (m, 6 H; CH2CH2CH2), 3.58 (t, J= 5.5,

6 H; CH2CH2-N), 4.03 (m, 6 H; O-CH2CH2), 6.48 (s, 1 H; CH), 7. 15 (d, J= 2.3, 2 H; Ar-

H), 7.23 (m, 2 H; Ar-H), 7.47 (m, 6 H; Ar-H), 7.54 (m, 6 H; Ar-H). 13C NMR (CDCl3): 6

= 29.6, 31.5, 31.7, 34.7, 35.7, 36.3 aliphaticc); 70.3 (O-CH2);1 22.4 122.9, 127.3, 132.8,









133.3, 138.0, 142.2, 144.6, 153.7 (aromatic); 168.3 (C=0). FAB MS m/z = 1212.66 [M

+ Na]

General procedure for synthesis of compounds 2-9a and 2-9b. To a suspension of

compound 2-8 in absolute ethanol, hydrazine mono hydrate (4 eq) was slowly added and

the mixture was refluxed for 24 h. The reaction was cooled to room temperature and the

solvent was partially evaporated under reduced pressure. The resulting residue was

poured into ice-cold water and a white precipitate quickly formed. The product was

collected by filtration.

Compound 2-9a. Yield 97%. 1H NMR (CDCl3): 6 = 0.40-0.49 (m, 18 H,

CH2CH3), 1.04 (s, 18 H, CH3), 1.25 (s, 18 H, CH3), 1.37 (q, J= 7.5 Hz, 6 H, CH2CH3),

1.62 (q, J= 7.5 Hz, 6 H, CH2CH3), 1.77 (quintet, ,J= 6.6 Hz, 6.9 Hz, 6 H, N-

CH2CH2CH2-O), 2.58 (b s, 6 H, NH2), 2.80 (t, J= 6.9 Hz, 6 H, N-CH2CH2CH2-O), 3.38

(b t, 6 H, N-CH2CH2CH2-O), 6.23 (s, 1 H, CH), 6.95 (s, 6 H, Ar-H). 13C NMR (CDCl3):

6 = 39.5, 39.4, 39.2, 37.7, 36.9, 35.0, 34.2, 29.5, 28.7, 9.7, 9.2 aliphaticc); 70.4 (O-CH2);

153.5, 147.4, 142.3, 139.7, 138.2, 128.0, 124.7 (aromatic). FAB MS m/z = 884.76 [M +

H] .

Compound 2-9b. Yield 92%. 1H NMR (CDCl3): 6 1.12 (s, 27 H, CH3), 1.25 (s, 27

H, CH3), 1.83 (quintet, J= 7.5 Hz, 6.6 Hz, 6 H, N-CH2CH2CH2-O), 2.87 (t, J= 7.5 Hz, 6

H, N-CH2CH2CH2-O), 3.38 (t, J= 6.6 Hz, 6 H, N-CH2CH2CH2-O), 6.25 (s, 1 H, CH),

7.10-7.16 (m, 6 H, Ar-H). 13C NMR (CDCl3): 6 = 31.32, 31.45, 33.17, 34.51, 35.46,

38.50, 38.98 aliphaticc); 70.50 (O-CH2); 122.47, 126.0, 127.38, 129.83, 131.79, 137.61,

141.79, 144.68, 153.35 (aromatic). FAB MS m/z = 800.67 [M + H] .









General procedure for synthesis of compounds 2-10a and 2-10b. A chloroform

solution of 0.75 g of amine (2-9) and 3.1 eq. of p-nitrophenyl(diphenylphosphoryl)acetat

(2-5) was stirred at 45-50 oC for 3 days. The reaction mixture was cooled to room

temperature; 1 M sodium hydroxide solution was added and stirred for 2 h. The organic

phase was extracted with 5% sodium carbonate followed by brine, and the solution was

dried over MgSO4. The solvent was removed in vacuo and acetonitrile (15 mL) was

added resulting in product precipitation. The solid was filtered, washed with acetonitrile,

and dried to afford pure product.

Compound 2-10a. Yield 70%. 1H NMR (CDCl3): 6 0.36-0.42 (m, 18 H, CH2CH3),

1.04 (s, 18 H, CH3), 1.15 (s, 18 H, CH3), 1.36 (q, J= 7.5 Hz, 12 H, CH2CH3), 1.50 (q, J=

7.5 Hz, 6 H, CH2CH3), 1.80 (b s, 6 H, N-CH2CH2CH2-O), 3.23 3.40 (several multiplets,

18 H, P(0) -CH2-C(0)N-CH2CH2CH2-O), 6. 15 (s, 1 H, CH), 6.88 (d, J= 1.8 Hz, 3 H,

Ar-H), 7.00 (d, J = 1.8 Hz, 3 H, Ar-H), 7.28-7.48 (m, 18 H, P-Ar H), 7.64-7.84 (m, 12 H,

P-Ar H). 13C NMR (CDCl3): 6 = 9.2, 9.7, 28.7, 29.6, 30.5, 35.1, 37.0, 37.7, 38.1, 39.2,

aliphaticc); 69.7 (O-CH2); 124.7, 127.9, 128.7, 128.9, 131.1, 131.6, 132.2, 133.0, 137.8,

139.7, 142.3, 147.4, 153.3 (aromatic); 165.22, 165.16 (C=0). 31P NMR: 6 = 30.3. ESI

FT -ICR MS m/z = 1633.88 [M + Na] Compound 2-10b. Yield 49% 1H NMR

(CDCl3): 6 = 1.16 (s, 27 H; CCH3), 1.21 (s, 27 H; CCH3), 1.91 (b, 6 H; CH2CH2CH2),

3.39 (several multiplets, 18 H; O-CH2CH2CH,-NH-C(0)CH,P(0)), 6.28 (s, 1 H; CH),

7.08 (d, J= 2.3, 3 H; Ar-H), 7.23 (d, J = 2.6, 3 H; Ar-H), 7.40 (m, 18 H; P-Ar H), 7.77

(m, 12 H; P-Ar H), 7.87 (t, J = 5.4, 3 H; NH). 13C NMR (CDCl3): 6 = 30.6 31.6, 31.7,

34.6, 35.6, 38.1, 38.9, 39.7 aliphaticc); 70.3 (O-CH2); 122.3, 127.1, 128.7, 128.9, 131.1,









131.2, 131.7, 132.17, 132.21, 133.1, 137.6, 141.9, 144.4, 153.6 (aromatic); 165.2,

165.1(C=0). 31P NMR (CDCl3): 6 = 29.8. ESI FT -ICR MS m/z = 1549.79 [M + Na] .

Compound 2-11a. To a mixture of 2-4a (2. 16 g, 2.56 mmol) and KOH (4.32 g) in

40 mL of dichloromethane, ethylchloroformate (1.6 mL, 1.82 g, 17 mmol) was added.

The mixture was stirred for 3 days at room temperature. The solution was then washed

with 200 mL of water and brine (50 mL), and dried over MgSO4. The solvent was

evaporated to afford 2.5 g of product in 92% yield. 1H NMR (CD30D): 6 = 0.56 (t, 9 H,

CH2CH3), 0.59 (t, J= 7.5, 9 H, CH2CH3), 1.18 (s, 18 H, CCH3), 1.26 (t, J= 7.5, 9 H,

OCH2CH3), 1.37 (s, 18 H, CCH3), 1.54 (q, J= 7.5, 6 H, CH2CH3), 1.76 (q, J= 7.5, 6 H,

CH2CH3), 3.39 3.65 (b m, 6 H, O-CH2CH2-N + 6 H, O-CH2CH2-N), 4. 12 (q, J= 7.2, 6

H, O-CH2CH3), ), 6.42 (s, 1 H, CH), 7.15 (b, 6 H, Ar-H). 13C NMR (CD30D): 6 = 9.8,

10.2, 15.2, 29.3, 30.3, 36.3, 38.0, 38.9, 40.4, 42.4 aliphaticc); 62.0 (CH2-O); 72.0 (CH2-

OAr); 126.4, 129.1, 139.3, 141.4, 144.2, 154.4 (aromatic); 159.1 (C=0). LSI MS: m/z =

1058.77 [M + H]. Anal. Calcd. for C64H103N309: C, 72.62; H, 9.81; N, 3.97; Found: C,

72.86; H, 10.30; N, 3.90.

Compound 2-11b. To a mixture of amine 2-9b (5 g, 6.3 mmol) and KOH (1 1.4 g)

in 20 mL of dichloromethane, ethylchloroformate (2.8 mL, 3.2 g, 29 mmol) was added.

The mixture was stirred for 3 days at room temperature. Subsequently solution was

washed with 250 mL of water and brine (50 mL), dried over MgSO4 and evaporated.

Yield: 75% (4.75 g). 1H NMR (CDCl3): 6 = 1.18 (s, 27 H, CH3), 1.23 (t, J= 6.90 Hz, 9

H, CH3), 1.33 (s, 27 H, CH3,), 1.98 (m, 6 H, N-CH2CH2CH2-O), 3.39 (m, 6 H, N-

CH2CH2CH2-O), 3.52 (m, 6 H, N-CH2CH2CH2-O), 4.11l(q, J= 6.20 Hz, 6 H, O-

CH2CH3), 5.39 (b s, 3 H, NH), 6.35 (s, 1 H, CH), 7.13-7.26 (m, 6 H, Ar-H). 13C NMR










(CDCl3): 6 =14.9, 30.9, 31.6, 34.6, 35.7, 39.1, 60.7, 71.0 aliphaticc); 122.6, 127.3, 144.7,

137.8, 141.9, 153.6 (aromatic); 157.1(C=0). FAB MS: m/z = 1016.73 [M + H] Anal.

Calcd. for C61H97N309: C, 72.08; H, 9.62; N, 4.13; Found: C, 72.07; H, 9.88; N, 4.06.

Compound 2-12a. To a stirred solution of lithium aluminium hydride (2.53 g,

0.067 mol) in tetrahydrofuran (500 mL) at 00C, 2.4 g (2.27mmol) of ester 2-11la was

added dropwise, and the reaction mixture was stirred at room temperature for 5 days. In

order to quench LAH, the solution was cooled to 00C, treated with 3 mL of water and

stirred for 5 minutes. A total of 3mL of 15% NaOH was then added dropwise, and after

additional 30 minutes, more water (9mL) was added (Steinhard's method).93 The solid

was separated, and the organic phase was dried over MgSO4. The solution was

evaporated to give crude product that was further purified by precipitation in acidified

pentane, dissolution in diethyl ether and extraction with IM NaOH. The organic phase

was dried with MgSO4 and evaporated to yield 2 g of product (95%) of pure product.

1H NMR (CDCl3): 6 = 0.55 (t, J= 7.2 Hz, 9 H, CH2CH3), 0.57 (t, J = 7.2 Hz, 9 H,

CH2CH3), 1.13 (s, 18 H, CH3,), 1.35 (s, 18 H, CH3,), 1.47 (q, J= 7.5 Hz, 6 H, CH2CH3),

1.73 (q, J= 7.5 Hz, 6 H, CH2CH3), 2.47 (s, 9 H, NCH3), 2.43 (b t, 6 H, N-CH2CH2-O),

3.65 (t, J= 5.7 Hz, 6 H, N-CH2CH2-O) 6.36 (s, 1 H, CH), 6.99 (b, 3 H, Ar-H), 7.05

(b, 3 H, Ar-H). 13C NMR (CDCl3): 6 = 9.3, 9.7, 28.8, 29.7, 35.35, 36.9, 37.1, 37.8, 39.3

aliphaticc); 52.2 (N-CH2); 71.7 (O-CH2); 124.9, 127.9, 138.0, 139.8, 142.6, 153.4

(aromatic). LSI MS: m/z = 884.76 [M + H] .

Compound 2-12b. To a stirring solution of ester 2-11Ib (7.0 g, 0.0069 mol) in

tetrahydrofuran (500 mL) at ice cold condition, lithium aluminium hydride (2.8 g,

0.074 moles) was slowly added. The reaction mixture was stirred at room temperature









for 6 days. The reaction mixture was cooled to ice cold temperature and IM NaOH (50

mL) was added and the stirring was continued for 15 minutes. Then water (100 mL) was

added and the content was transferred to separating funnel and extracted with diethyl

ether (4 x 50 mL). The organic phase was washed with brine (5 x 20 mL), dried over

MgSO4 and evaporated to give 5.04 g (87%) of pure product. 1H NMR (CDCl3): 6 = 1.18

(s, CH3, 27 H), 1.34 (s, CH3, 27 H), 1.95 (t, J= 7.35 Hz, 6 H, N-CH2CH2CH2-O), 2.43

(s, 9 H, CH3,), 2.73 (t, J= 7.35 Hz, 6 H, N-CH2CH2CH2-O), 3.54 (t, J= 6.30 Hz, 6 H,

N-CH2CH2CH2-O), 6.36 (s, 1 H, CH), 7.12-7.26 (m, 6 H, Ar-H). 13C NMR (CDCl3):

6 = 30.8 31.6, 31.7, 34.6, 35.6, 36.6, 38.8 aliphaticc); 49.3 (N-CH3); 71.0 (O-CH2);

122.6, 127.3, 138.0, 141.8, 144.3, 147.4, 153.9 (aromatic). FAB MS: m/z = 842.71

[M + H] .

Compound 2-13a. To a solution of the secondary amine 2-12a (2.42 g, 2.74 mmol)

and K2CO3 (4.50 g, 32.56 mmol) in CH2C 2 WAS added chloroacetyl chloride (1.40 mL,

17.60 mmol), and the reaction mixture was refluxed for 18 h. A second portion of

chloroacetyl chloride (0.70 mL, 8.80mmol) was added and refluxed for an additional

20 h. Subsequently, the solution was cooled down, and washed with 2 N NaOH, H20

and dried over MgSO4. The solvent was remover in vacuo, and the crude white solid was

recrystallized from dichloromethane/hexamethyl-di siloxane to give 2.70 (8 8%) g of pure

product. 1H NMR (CDCl3) aS well aS 13C NMR (CDCl3) Spectra are very complicated.

1H NMR (CDCl3): 6 = 0.51 0.60 (m, 18 H, CH2CH3), 1.12 (b, 18 H, CH3,), 1.30 (b, 18

H, CH3,), 1.44 1.73 (b m, 12 H, CH2CH3), 2.88 4.22 (several multiplets, 9 H N-CH3

+ 6 H N-CH2CH2-O + 6 H, N-CH2CH2-O), 4.06 (s, 6 H, CH2-C1), 6.33, 6.38, 6.43 (s, 1 H,

CH), 6.82 7. 11 (m, 6 H, Ar-H). LSI MS: m/z = 1112.67 [M + H] .









Compound 2-13b. To a solution of the secondary amine 2-12b (3.00 g, 3.60 mmol)

and K2CO3 (6.00 g, 43.40 mmol) in CH2C 2 (20 mL) was added chloroacetyl chloride

(2.08 mL, 26.15 mmol), and the reaction mixture was heated at 450C for 12 h. A second

portion of chloroacetyl chloride (1.04 mL, 13.07 mmol) was added and stirred for an

additional 20 h at 450C. Subsequently, the solution was cooled down, and washed with

2 N NaOH, H20 and dried over MgSO4. The solvent was remover in vacuo to give

2.50 g (66%) of pure product. 1H NMR (CDCl3): 6 = 1.18 (s, 27 H, CH3,), 1.36 (s, 27 H,

CH3,), 1.84 1.95 (m, 6 H O-CH2CH2CH2-N), 2.86 3.04 (m, 9 H, N-CH3), 3.37 3.65

(b m, 6 H, O-CH2CH2CH2-N + 6 H, CH2-C1), 4.04 4. 10 (m, 6 H, O-CH2CH2CH2-N),

6.33 (s, 1 H, CH), 7.20 7. 11 (m, 6 H, Ar-H). 13C NMR (CDCl3): 6 = 27.9, 29.3, 31.6,

33.7, 34.6, 35.7, 39.1, 41.1, 41.8, 46.347.9 aliphaticc); 70.4 (OCH2); 122.6, 127.5, 137.9,

141.8, 144.9, 153.5 (aromatic) 166.2 (C=0). El MS m/z = 1071.61 [M + H] Anal.

Calcd. for C61H94C 3N306: C, 68.36; H, 8.84; N, 3.92; Found: C, 68.58; H, 8.31; N, 3.71.

Compound 2-14a. Method A. The 2.50 g, 2.24 mmol of starting material (2-13a)

was dissolved in 9.00mL of ethyl diphenylphosphinite (9.59 g, 41.65 mmol) while the

temperature was gradually increased from 100 to 150 OC, and the mixture was stirred for

40 h. Subsequently, the reaction mixture was cooled down to rt, and the diisopropyl ether

was added till a white precipitate was formed. The solid was filtered and washed with

diisopropyl ether to afford 3.08 g (85%) of pure product. 1H NMR (CDCl3) aS well as

13C NMR (CDCl3) Spectra are very complicated. 1H NMR (CDCl3): 6 = 0.44 0.53

(m, 18 H, CH2CH3), 1.04 (s, 9 H, CH3,), 1.09 (s, 9 H, CH3,), [these two singlets merge

into one 6 = 1.8 at 55 oC], 1.21 (s, 9 H, CH3,), 1.27 (s, 9 H, CH3,), [these two singlets

merge into one 6 = 1.25 at 55 oC], 1.42 (b m, 6 H, CH3,), 1.62 (b m, 6 H, CH3,), 2.66 -









3.69 (several multiplets, 9 H, N-CH3 + 6 H N-CH2CH2-O + 6 H, N-CH2CH2-O), 6.21,

6.26, 6.32 (s, 1 H, CH), 6.83 7.02 (m, 6 H, Ar-H), 7.44 7.53 (m, 18 H, P-Ar H), 7.84

- 7.90 (m, 18 H, P-Ar H). ESI FT -ICR MS m/z = 827.94 [M + 2Na]2+, m/z = 1632.89

[M + Na] Anal. Calcd. for C1ooH130N309P3: C, 74.55; H, 8.13; N, 2.61; Found:

C, 74.34; H, 8.44; N, 2.61. Slow diffusion of pentane into solution of 2-14a in diethyl

ether/dichloromethane afforded crystals suitable for X-ray analysis.

Method B. A solution of secondary amine (0.47 g, 0.53 mmol) and p-nitrophenyl

(diphenylphosphoryl)acetate (1.10 g, 2.88 mmol) in dichloromethane, was stirred for

2 weeks at room temperature. Subsequently solution was treated with 1 M NaOH and

stirred for additional 2 hours. The p-nitrophenol salt was extracted from organic phase

using 5% sodium carbonate. Organic phase was dried over MgSO4, filtered and the

solvent removed in vacuo. The crude product was criticized by diffusion of pentane into

solution of product in diethyl ether/dichloromethane; yield 0.60 g (70%).

Compound 2-14b. Method A. The 0.50 g, 0.47 mmol of starting material (2-13b)

was suspended in 1.00mL of ethyl diphenylphosphinite (4.20 mmol) while the

temperature was gradually increased from 100 to 150 OC. Within first 3 h of reaction,

every 20 minutes the mixture was exposed for few seconds to the vacuum. The reaction

mixture was stirred at 150 oC for additional 20 h. Subsequently it was cooled down to

room temperature and the diethyl ether was added till a white precipitate was formed.

The solid was filtered and redissolved in diisopropyl ether to give pure product upon

crystallization. Yield 0.42 g (57%). 1H NMR (CDCl3) aS well aS 13C NMR (CDCl3)

spectra are very complicated. 1H NMR (CDCl3): 6 = 1.15 (s, 27 H, CH3,), 1.27 (s, 27 H,

CH3,), 1.78 1.95 (m, 6 H, N-CH2CH2CH2-O), 2.66 3.55 (several multiplets, 27 H: 9 H









N-CH3 + 6 H N-CH2CH2CH2-O + 6 H, N-CH2CH2CH2-O), 6.24 (b s, 1 H, CH), 6.93 (b s,

2 H, Ar-H), 7.06 (b s, 2 H, Ar-H), 7.38 (m, 18 H, P-Ar H), 7.74 (m, 12 H, P-Ar H). 13C

NMR (CDCl3) 6 = 27.9, 29.4, 31.5, 33.9, 34.5, 35.5, 36.9, 37.7, 38.6, 38.9, 46.0, 48.3

aliphaticc); 70.6 (O-CH2); 122.3, 127.2, 128.6, 128.7, 131.2, 131.3, 132.1, 133.3, 137.7,

141.7, 144.5, 144.7, 153.5 (aromatic); 164.8 (C=0). 31P NMR (CDCl3) 6 = 29.4, 29.3.

El MS m/z = 1567.86 [M + H] Method B. A solution of secondary amine (2.00 g, 2.40

mmol), p-nitrophenyl (diphenylphosphoryl)acetate (4.50 g, 11.80 mmol) and 1 mL of

Et3N in chloroform, was stirred for 5 days at 45 500C. After cooling down to the rt

solution was treated with 1 M NaOH and stirred for additional 2 hours. The p-

nitrophenol salt was extracted from organic phase using 5% sodium carbonate. Organic

phase was dried over MgSO4, filtered and the solvent removed in vacuo. The crude

product was washed with diethyl ether and dried yielding 2.50 g (67%) of clean product.

Tb-complex of 2-6a [2-6a-TbNO3](NO3)2. To a solution of 2-6a (0.200 g, 0. 127

mmol) in acetonitrile (8 mL), Tb(NO3)3.6H20 (0.058 g, 0.128 mmol) in methylene

chloride (4.5 mL) was added and reaction mixture was stirred overnight at room

temperature resulting in a white solid. The complex was isolated by filtration, washed

with acetonitrile and dried. Yield 0. 180 g (74%). Anal. Called for C97H124N6018P3Tb: C,

60.87; H, 6.53; N, 4.39. Found: C, 60.91; H, 6.66; N, 4.27. Slow diffusion of ether into a

concentrated solution of the complex in methanol, afforded crystals suitable for structural

analysis.

Tb-complex of 2-14a [2- 14a-TbNO3](NO3)2. A solution of Tb(NO3)3.6H20 (0.028

g, 0.062 mmol) in 1 mL of methanol was added to a solution of 2-14a (0. 100 g, 0.062

mmol) in methanol (1mL), and reaction mixture was stirred for one hour at room









temperature. A white precipitate formed within minutes, and the product was collected

by filtration, washed with cold methanol, and dried to obtain 0.080 g (62%) of product.

ESI FT-ICR MS m/z = 915.40 [2-14a-TbNO3 2+. Slow diffusion of ether into a

concentrated solution of the complex in mixture of methanol and dichloromethane

afforded crystals suitable for structural analysis.

Bi-complex of 2-6c [2-6c-BiNO3](NO3)2 A solution of Bi(NO3)3-5H20 (0.058 g,

0. 12 mmol) in 8 mL of 1:1 mixture of acetonitrile and methanol was added to a solution

of 2-6c (0.08g, 0.06mmol) in 2 mL of methanol, and the mixture was stirred for 1 hour.

Part of solvent was evaporated in-vacuo and the product was precipitated out by ether

diffusion. Yield 0.080 g (52%). Slow diffusion of ether into a saturated

acetonitrile/methanol solution of [2-6c-Bi(NO3)](NO3)2 affOrded crystals suitable for

structural analysis. 1H NMR (CD3CD), 550C: 6 = 1.15 (s, 27 H; Ar-C(CH3)3), 2.14

(s, 9 H; Ar-CH3), 3.17 (b, 12H; 3.96 Ar-O-CH2CH2), 6.74 (s, 1 H; C-H), 7.00 (d,

J = 2.05 Hz, 3 H; Ar H), 7. 10 (d, J = 2.56 Hz, 3 H; Ar H), 7.49 (m, 12 H; P-Ar H), 7.62

(m, 6 H; P-Ar H), 7.74 (m, 12 H; P-Ar H), 7.86 (s, 3 H; N-H). 31P NMR (CDCl3):

39.66. HR ESI-ICR MS (sample injected as solution in 1% HNO3/MeOH): m/z = 814.29

[2-6c-Bi(NO3 12+ and m/z =1691.56 [2-6c-Bi(NO3)2] Anal. Cald for

{ [2-6c-Bi(NO3)-MeOH-2H20] [NO3 2) 83H102BiN6021P3: C 54.73; H 5.64; N 4.61;

Found: C, 54.54; H, 5.32; N, 4.92.

2.4.5 X-Ray Crystallonraphy Unit cell dimensions and intensity data for all the

structures were obtained on a Siemens CCD SMART diffractometer at 173 K. The data

collections nominally covered over a hemisphere of reciprocal space, by a combination of

three sets of exposures; each set had a different $ angle for the crystal and each exposure










covered 0.30 in co. The crystal to detector distance was 5.0 cm. The data sets were

corrected empirically for absorption using SADABS.

All the structures were solved using the Bruker SHELXTL software package for

the PC, using either the direct methods or Patterson functions in SHELXS. The space

groups of the compounds were determined from an examination of the systematic

absences in the data, and the successful solution and refinement of the structures

confirmed these assignments. All hydrogen atoms were assigned idealized locations and

were given a thermal parameter equivalent to 1.2 or 1.5 times the thermal parameter of

the atom to which it was attached. For the methyl groups, where the location of the

hydrogen atoms is uncertain, the AFIX 137 card was used to allow the hydrogen atoms to

rotate to the maximum area of residual density, while Eixing their geometry. In cases of

extreme disorder or other problems, the non-hydrogen atoms were refined only

isotropically, and hydrogen atoms were not included in the model. Severely disordered

solvents were removed from the data for 2-10a, 2-14a, [2-6a-TbNO3](NO3)2, [2-

6c-BiNO3](NO3)2 and [2-14a-TbNO3](NO3)2 USing the SQUEEZE function in the Platon

for Windows software and the details are reported in the supporting information in the

CIF file for each structure. Structural and refinement data and selected bond lengths for

all the compounds are presented in the Tables 2-5 and 2-6.












[2-6a-TbNO3] [2-14a-TbNO3]
2-10b-CH30H 2-14a-3CHCl3 2Et20 [2-6c-BiNO3](NO3)2
S(NO3)2 Et20 (NO3)2


Table 2-6. X-ray data for the crystal structures of 2-10b, 2-14a and the complexes
and [2-14a-TbNO3](NO3)2


[2-6a-TbNO3](NO3)2, [2-6c-BiNO3](NO3)2


total reflections
unique reflections
Omax(0)
empirical formula
Mr
crystal system
space group

b (+)
c (A)
a (0)


Vc (83)
D, (g cm-3)
T (K)

pu(Mo-Koc) (mml)
R1 [I>2 20(I data]b
wR2 (all data)c
GoF


24103
5195
24.99
C96H1 19N3 010P3
1567.85
hexagonal
R-3
18.674(2)
18.674(2)
43.801(3)
90
90
120
13228(2)
1.181
173(2)

0.127
0.0816 [3938]
0.1980
1.140


32269
20440
25.00
C105H138C 9N309.5P3
2006.14
triclinic
P-1
17.250(2)
18.070(2)
21.557(3)
87.288(2)
66.965(2)
71.870(2)
5855.5(13)
1.138
173(2)

0.307
0.0826 [13367]
0.2673
1.115


31086
19674
25.00
C101H134N6019P3Tb
1987.97
triclinic
P-1
12.5597(9)
15.5209(12)
28.949(2)
93.700(2)
90.632(2)
90.532(2)
5630.9(7)
1.172
173(2)

0.732
0.0687 [11833]
0.1765
0.986


23270
14249
23.00
C1ooH124N6021P3Tb
1997.88
triclinic
P-1
12.4105(18)
13.713(2)
33.122(5)
88.353(3)
82.941(2)
66.989(2)
5148.0(13)
1.289
173(2)

0.803
0.0971 [11989]
0.2097
1.280


51156
30858
24.50
C84H102BiN6020P3
1817.61
triclinic
P-1
20.8891(13)
22.9583(15)
23.1141(15)
80.3080(10)
64.9140(10)
68.3520(10)
9330.4(10)
1.294
193(2)

2.008
0.0657 [15711]
0.1780
0.899


aObtained with monochromatic Mo Ke: radiation (h = 0.71073 A+)
bR1 E CIFo IFe /CFoI.
"wR2 = (E[w(Fo2 E2, 2 ICw Fo2 2 12














CHAPTER 3
DESIGN, SYNTHESIS AND EVALUATION OF PHOSPHINE SULFIDE BASED
CHELATES FOR THE SEPARATION OF TRIVALENT
LANTHANIDES AND ACTINIDES

3.1 Introduction

Recent challenges in the field of the partitioning and transmutation of highly acidic

nuclear waste involve the separation of minor actinides (neptunium, americium, and

curium) and long lived fission products. Over the years, several partition processes have

been proposed,94 but current protocols lack the unique ability to effectively separate

trivalent actinides from trivalent lanthanides. This fundamental problem in separation

science is due to the similarity in the ionic structure and radius of the trivalent f-elements

(especially Am, Eu and Nd).


H~S

SH O


HBTMPDTPHDDT
(Cyanex 301)

Figure 3-1. Structures of sulfur based extractants.

During the last couple of years, only several extraction systems employing soft

donor ligands such as sulfur and nitrogen have shown some selectivity for Am(III) over

Ln(III).95-98 For instance, a synergistic mixture of di-2-ethylhexyl dithiophosphoric acid

(HDEHDTP) (Figure 3-1) and tributylphosphate (TBP) has an observed separation factor

(SF = DA/,DEu) Of 60 for the partition of Am(III) over Eu(III).95









Work done by Zhu et al. have demonstrated that purified Cyanex 301 [bis(2,4,4-

trimethylpentyl)dithi ophosphini c aci d, HB TMPD TP] was able to separate Am(III) from

Eu(III) even more efficiently then the HDEHDTP/TBP mixture, with the separation

factor of 5900.27,99 Interestingly, several years earlier, these soft donor compounds have

not been considered as potential extractants for actinides and lanthanides separation due

to their hypothetical incompatibility with hard metal ions.

At the pH of 3 in nitric acid and IM NaNO3/kerosene extraction environment the

extraction enthalpy of Am(III) with purified 0.5 M Cyanex 301 was found to be

significantly less endothermic than that of Eu(III) (18.10 kJ/mol and 43.65 kJ/mol

respectively).27 The stronger affinity for trivalent americium was attributed to the higher

degree of covalency of the Am(III)-S bond compared to Eu(III)-S. Using theoretical

calculations Madic and coworkers have quantified the covalent effect and cited that the

covalent contribution for the Am-S 0-bond energy is higher by approximately 7.6 kJ/mol

than for Eu-S.48

Even though these acidic organophosphorous reagents are successful in the

differentiation between Am(III) and lanthanides, conditions required to maintain

selectivity during extraction have proven to complicate the partition process. The

extraction equilibrium strongly depends on the acidic nature of the extractant,99 and the

low acidity of Cyanex 301 (pKa = 2.6) limits the effectiveness of the compound to the

extraction systems with pH 3 or higher. Since the acidity of waste solutions is generally

in the range of 1 to 3 M HNO3, the extraction with Cyanex 301 is practically impossible

without prior acidity adjustment, which makes the separation process harder to manage.

To overcome the ligand dissociation problem that limits the effectiveness of Cyanex 301









at high acidity, the alkyl groups on the phosphorous were replaced by electron

withdrawing substituents (Figure 3-2).29 Such alteration significantly weaken the basicity

of the ligand and the new extractant was not able to bind any of the studied trivalent

metal ions (DAm,Eu < 10-3)



p" X =H, CH3, CI, F
x 'SH

Figure 3-2. Structure of the aromatic dithiophosphinic acids.29

The synergistic mixtures of this aromatic dithiophosphinic acid with TBP (tri-n-

butyl-phosphate) or TOPO (tri-n-octyl-phosphine oxide) showed slightly improved ion

binding.29,100-102 Even though these synergistic mixtures facilitate high selectivity for

An(III) over Ln(III) in a strong acidic medium (1.5 M HNO3), the system achieves still

very low distribution ratios that are insufficient for practical application.

The mechanism of M(III) extraction (M = Am, Eu) using mixtures of aromatic

dithiophosphinic acids with neutral O-bearing coextractants has been extensively

investigated from a theoretical chemistry standpoint.48 Madic et al. elucidated that the

formation of a hard/soft synergist extraction pair changes the nature of the M-S bonds

due to the ORM electron density transfer. It was determined that addition of a weak

oxygen donor strengthens the M-S bonds while a strong donor weakens them. The

mechanism has been confirmed by the 31P NMR chemical shifts studies of the neutral O-

bearing organophosphorus coextractants, where changes in the position of phosphorus

resonances were correlated with M(III) extractabilities.

Currently there are no extraction systems that would be free from maj or

shortcomings. Most of methods are limited by either low extent of ligand dissociation in










highly acidic medium, low efficiency in terms of values of distribution coefficients and

poor stability toward hydrolysis. These problems limit the application of presented

protocols such as Cyanex 301, HDEHDTP/TBP or aromatic dithiophosphinic acids in

large scale nuclear waste clean-up operations. The purpose of our research was to

develop a ligand that would combine the most advantageous qualities of previously

studied extractants, and create an extraction system that would be more suitable for

industrial process development. To avoid sensitivity of the ligand towards high acid

concentration and improve extraction efficiency the dithiophosphinic acid group was

replaced by phosphine sulfide and attached to the triphenoxymethane platform. As a

result a neutral hexadentate tri s-carbamoylmethylphosphine sulfide (tri s-CMP S) ligand

has been created. Another advantage expected from tris-CMPS was an enhanced ability

to bind metals through the three carbonyl oxygens (build in hard donor synergist) that

upon complexation could participate in the formation of a six-membered chelate ring and

improve the stability of the complex.



4 S6_:P c









Figure 3-3. Anticipated binding mode for tris-CMPS extractant.

The previous investigation of the affinity of tris-CMPO derivatives for trivalent f-

block elements in 1 M nitric acid/dichloromethane extraction systems has shown that

harder and more polar nitrates win the competition for binding ions over the neutral










chelate, which was established through low extraction efficiency (Chapter 2). Therefore,

in the case of tris-CMPS extractant, the rather low affinity for trivalent f-elements was

expected, yet slightly stronger interactions of ligand with the trivalent actinides were

anticipated. However, if in the environment rich in nitrates and water the phosphine

sulfide groups do not participate in metal binding, the ligand would not be able to

differentiate between these two groups of f-elements. Our observations are reported

herein.

3.2 Results and Discussion

3.2.1 Synthesis and Extraction Data

The synthetic methodology to obtain tris-CMPS compounds follows procedures

described for synthesis of tris-CMPO derivatives, and differs only in the last step where

p-nitrophenyl (diphenylphosphoryl)acetate (2-5) is replaced by (diphenyl-

phosphinothioyl)-acetic acid (3-1) as presented in Figure 3-4.





NH2

)n OSNH






R2
24a: n=1, R R2 =t-Pentyl 3-1 3-2a: n=1, R1, R2= t-Pentyl
24b: n=1, R R2 = t-Bu 3-2b: n=1, R1, R2 =t-Bu
24c: n=1, R = Me, R2 =t-Bu 3-2c: n=1, R = Me, R2 =t-Bu
2-9a: n=2, R R2 =t-Pentyl 3-3: n=2, R1, R2= t-Pentyl


Figure 3-4. Synthesis of tris-CMPS extractants. Conditions: mercaptothiazoline, DCC,
DMAP, methylene chloride, rt overnight.










Previously synthesized tris-CMPO compounds showed very good selectivity for

tetravalent actinides and lacked ability to efficiently bind Ln(III) and An(III) (Chapter 2).

The soft character of the basic phosphine sulfide groups in the new extractant may induce

slightly stronger attraction of ligand for trivalent actinides and afford some discrimination

between these two groups of elements in liquid-liquid extraction experiments.

Unfortunately extraction of 241Am(III) with tris-CMPS was found to be inefficient,

and expected selectivity for americium over europium was not observed. For the

comparison with the tris-CMPO ligand system, extraction experiments were performed

on a series of trivalent lanthanides and tetravalent thorium.



100 o6


60 NH NH
E%/
40r O' H OH
20C


Th(lV) La(lll) Ce(lll) Nd(lll) Eu(lll) vb(lll) 33
2-6a 3-2a

Figure 3-5. Comparison of metal binding by tris-CMPO (2-6a) and tris-CMPS (3-2a).

In the first experiment, binding properties of CMPO and CMPS derivatives with a

shorter (two carbon) spacer between the triphenoxymethane base and binding units were

compared (Figure 3-5). The results revealed very low extraction efficiency of 3-2a for all

studied cations. The tris-phosphine sulfide compound was no longer able to take

advantage of the difference in the oxidation states of f-elements cations. Soft, neutral

phosphine sulfide group was found to be incompatible with a hard acid such as

tetravalent thorium.










In order to test the influence of the flexibility of the lighting CMPS arm on the

extraction pattern, derivative 3-3, with elongated by an additional carbon atom arm was

synthesized. In comparison to the 3-2a, 3-3 has not shown any improvement in the

extraction (Figure 3-6).





15 S NH


E% 1 O OH HO H




T(lV) La(lll) Ce(lll) Nd(lll) Eu(lll) b(lll) 3 3
3-2a 3-3

Figure 3-6. Comparison of metal binding by tris-CMPS 3-2a and 3-3.

As highlighted in Figure 3-7, the 2-10a and 3-3 differ only in the nature of the

phosphine donor. The cavity size in both extractants. is similar and much like 2-10a, 3-3

should be able to easily adopt the geometry required by the metal ion and show some

binding enhancement.




100 ~ 2-10a $
OS O
80r/ 0~3NH NH
60
E%/
40 O H O H
20
20

T(lV) La(lll) Ce(lll) Nd(lll) Eu(lll) b(lll) 33
2-10a 3-3

Figure 3-7. Comparison of metal binding by tris-CMPO (2-10a) and tris-CMPS (3-3).









Exhibited low affinity of 3-3 for all tested ions has proven the importance of the

hard phosphine oxide in effective binding of any f-element ions, and negligible

involvement of the phosphine sulfide in the metal binding.

Table 3-1. Extraction percentage (%E) for ligands: 2-6a, 3-2a, 2-10 and 3-3. Aqueous
phase: 10-4 M metal nitrate in 1 M HNO3I, Organic phase: 10-3 M of ligand in
methylene chloride
Cation (10-4 M) Equivs of ligand
2-6a 3-2a 2-10a 3-3
in 1 M HNO3 18 Organic phase
Th (IV) 10 100 7 100 9
La (III) 10 3 9 16 9
Ce (III) 10 1 9 16 10
Nd (III) 10 5 10 15 10
Eu (III) 10 2 11 14 11
Yb (III) 10 4 9 13 9


3.2.2 Crystal Structure Analysis

Single crystals of ligand 3-3 were grown by slow diffusion of ether into the

concentrated solution of ligand in dichloromethane. In the crystal structure of 3-3

presented in Figure 3-8, the average length of the carbonyl bonds is similar to the length

of carbonyl bonds in the CMPO equivalent 2-10b [1.225(7) and 1.232(4) A+ respectively].

The average distances between the sulfur and phosphorous in phosphine sulfide moieties

(1.952(2) A+) are in the range of typical P=S bonds with phenyl substituents on the

phosphorus (Ph3PS, P=S: 1.951(2)-1.954 A+(4)),103-105 but almost 0.5 A+ longer than the

distance between phosphorous and oxygen in the phosphine oxide 2-10b [1.477(3) A+,

Chapter 2]. The P-C(Ph) mean bond length 1.810(6) A+ is also similar to the distances

found in Ph3PS (1.817(7) A+), as well as to those found in the tris-CMPO compound

[1.798(4) A+].


































Figure 3-8. Diagram of the solid-state structure of 3-3 (30% probability ellipsoids for N,
O, S and P atoms; carbon atoms drawn with arbitrary radii). For clarity, all
hydrogen atoms have been omitted.

In order to gain some understanding of the binding attributes of the ligand, a

complex of Tb(NO3)3 with ligand 3-3 was synthesized (Figure 3-9). [3-3-Tb(NO3)3]

compound contains two similar structures in the asymmetric unit. The potentially

hexadentate chelate 3-3, is coordinated to the metal center in a tris-monodentate manner

via carbonyl oxygens only. The interactions with terbium ions are similar in strength

(2.320(3) A+) to the interactions in the terbium complex with tris-CMPO 2-6a

(2.308(5) A+). Interestingly, none of the three sulfur atoms participate in the metal

binding. Instead, three nitrate ions are bound in a bidentate fashion, fully neutralizing the

charge of the nine coordinate complex.




















close-up view of
Tb coordination


Figure 3-9. Diagram of the structures of [3-3-Tb(NO3)3] (right) and close-up view of the
terbium coordination environment (left).

Some examples of the X-ray analyzed solid state complexes of lanthanides and

acidic organophosphorous ligands with sulfur directly bound to the metal required an

anhydrous environment during the synthesis due to their sensitivity to the moisture.106

If in the [3-3-Tb(NO3)3] COmplex prepared from hydrated salt of lanthanide nitrate

phosphine sulfides are not able to bind the cation, the aqueous extraction environment

must even more effectively prevent sulfurs from interactions with a metal, and leftover

three carbonyls cannot be expected to successfully complete with highly concentrated

nitrates and water for metal binding.









Table 3.2. Selected bond lengths (A+) for compounds: 3-3 and [3-3-Tb(NO3)3 -
3-3 [3-3-Tb(NO3)3] [3-3-Tb(NO3)3]
Tb(1) Tb(2)
P(1)-S(1) 1.956(2) 1.9590(14) 1.9432(19)
P(2)-S(2) 1.943(2) 1.9526(15) 1.9532(18)
P(3)-S(3) 1.958(2) 1.9485(15) 1.9536(17)
C(53)-O(4) 1.221(7) 1.241(4) 1.236(5)
C(70)-O(5) 1.223(7) 1.248(4) 1.233(4)
C(87)-O(6) 1.232(7) 1.251(4) 1.241(5)
P(1)-CPh(55) 1.802(6) 1.807(4) 1.847(6)
P(1)-CPh(61) 1.812(6) 1.810(4) 1.806(5)
P(2)-CPh(72) 1.806(6) 1.814(4) 1.828(5)
P(2)-CPh(78) 1.821(6) 1.812(4) 1.782(6)
P(3)-CPh(89) 1.802(6) 1.811(4) 1.814(4)
P(3)-CPh(95) 1.814(6) 1.807(5) 1.815(4)
M-O(4) -2.316(2) 2.298(3)
M-O(5) -2.317(3) 2.341(3)
M-O(6) -2.330(3) 2.318(3)
M-O(7) -2.493(3) 2.468(4)
M-O(8) -2.437(3) 2.426(4)
M-O(10) -2.524(3) 2.427(3)
M-O(11) -2.43 1(3) 2.505(3)
M-O(13) -2.492(3) 2.484(3)
M-O(14) -2.452(3) 2.465(3)









Table 3-3. X-ray data for the crystal structures of 3-3 and the [3-3-Tb(NO3)3] COmplex.
3-3-CH2C 2C4HloO [3-3 -Tb(NO3)3]-2CH3CN
total reflections 35243 73828
unique reflections 12730 48890
0max(0) 23.43 28.03
empirical formula C losH142C 2N307P3 S3 C104H136Ns015P3S3Tb
Mr 1818.21 2086.22
crystal system monoclinic triclinic
space group P21/n P-1
ar (A) 14.8551(10) 16.8425(7)
b (A) 13.6147(10) 25.0080(10)
c (A) 49.329(3) 27.0255(11)
a (0) 90 78.3410(10)
P (0) 96.5800(10) 74.6450(10)
y(0) 90 87.4780(10)
Vc (83) 9910.9(12) 10749.7(8)
D, (g cm-3) 1.219 1.289
T (K) 173(2) 173(2)
Z 4 4
pu(Mo-Koc) (mm l) 0.233 0.824
R1 [I>2 20(1 data]b 0.0858 [7722] 0.0558 [34928]
wR2 (all data)c 0.2182 0.1625
GoF 1.033 1.081
aObtained with monochromatic Mo Kee radiation (h = 0.71073 A+)
bR1 E CIFo IFe /CFoI.
"wR2 = (E[w(Fo2 Fe2 2 Cw Fo2 2] 1 2

3.3 Conclusions

Tripodal phosphine sulfide based compounds were synthesized as softer derivatives

of the tris-CMPO chelate. Their ability to differentiate between trivalent lanthanides and

actinides was tested using 41Am and 152Eu isotopes. Extraction experiments using

tris-CMPS compounds revealed that the ligands are not able to preferentially bind

trivalent Am over Eu. The soft nature of the phosphine sulfide donor group was found to

be incompatible with the hard tetravalent thorium, and in contrast to the tris-CMPO, the

tris-CMPS analog showed no selectivity for this metal ion.

The solid state structure of tris-CMPS with terbium nitrate showed that the

phosphine sulfide portion of the ligand is not involved in the metal biding. Although









with no structural studies on the tris-CMPS extractant in the solution, it can be assumed

that in the system with negatively charged nitrates, soft neutral sulfur atoms do not

participate in the coordination of hard metals.

3.4 Experimental Section

Amines 2-4 a, b, c and 2-9a were synthesized as described in Chapter 2. The

(diphenyl-phosphinothioyl)-acetic acid, 3-1, was prepared according to the literature

procedure.107 The final synthetic procedures of tris-CMPS compounds were developed in

the collaboration with Dr. Ajay Sah and Dr. Priya Srinivasan. The [3-3-Tb(NO3)3]

complex used in the discussion of this chapter was crystallized by Dr. Ajay Sah.

Compound 3 -2a. A mixture of (diphenyl -phosphinothi oyl)-aceti c aci d (3 -1 )

(5.40 g, 19.55 mmol), 2-mercaptothiazoline (2.51 g, 21.06 mmol) and

4-dimethylaminopyridine (0.60 g, 4.91 mmol) was stirred in dichloromethane (200 mL)

at room temperature for 30 min. N, N'-dicyclohexylcarbodiimide (4.36 g, 21.13 mmol)

was then added followed by additional 15 mL of dichloromethane. After 6 h, solid 2-4a

(4.44 g, 5.27 mmol) was added and the mixture was stirred for an additional 24 h at room

temperature. The slurry was filtered and the solvent was removed in vacuo. The product

was separated from byproducts by the dissolution in diethyl ether. Addition of methanol

to the concentrated solution of the compound resulted in precipitation of solid product

that was subsequently filtered and washed with cold methanol. Yield 5.57g (65%). 1H

NMR (CDCl3): 6 = 0.46 (t, J = 7.3, 18H; CH2CH3), 1.07 (s, 18H; CCH3), 1.23 (s, 18H;

CCH3), 1.41 (q, J = 7.3, 6H; CH2CH3), 1.59 (br, 6H; CH2CH3), 3.37 (br, 12H;

OCH2CH2NH), 3.64 (d, J = 14.1, 6H; C(0)CH2P(0)), 6.23 (s, 1H; CH), 6.92 (d, J = 2.1i,

3H; Ar), 6.96 (d, J = 2.1i, 3H; Ar), 7.38 (m, 18H; Ar), 7.64 (br, 3H; NH), 7.87 (m, 12H;










Ar). 13C NMR (CDCl3): 8 (C=0) = 165.13, 165.07; (Aromatic) = 153.1, 142.9, 140.0,

137.8, 131.90, 131.87, 131.7, 131.5, 128.9, 128.7, 127.7, 125.0; aliphaticc) = 70.3

(OCH2), 42.6 (CH2 2)>, 42.0, 40.2, 39.3, 37.8, 37.0, 35.5, 29.8, 28.7, 9.8, 9.3. 31P NMR

(CDCl3): 6 = 38.9. Anal. Called for C97H124N306P3S3: C, 72.04; H, 7.73; N, 2.60. Found:

C, 72.19; H, 7.85; N, 2.57.

C omp ound 3 -2b. A mixture of 3 -1 (2. 13 g, 7.7 1 mm ol), 2 -merc aptothi azol ine

(0.97 g, 8.14 mmol) and 4-dimethylaminopyridine (0.29 g, 2.37 mmol) was stirred in

dichloromethane (80 mL) for 30 min. The solid N, N'-dicyclohexylcarbodiimide (2.45 g,

1 1.87 mmol) was then added, followed by additional 20 mL of dichloromethane. After

6 h, solid 2-4b (1.69 g, 2.23 mmol) was added and the mixture was stirred for an

additional 24 h. The slurry was filtered and the solvent was removed in vacuo. The solid

was dissolved in diethyl ether and quickly filtered. Within several days upon slow

evaporation of solvent pure product precipitated from the ether solution. Yield 1.44g

(42%). 1H NMR (CDCl3): 6 = 1.16 (s, 27H; CCH3), 1.23 (s, 27H; CCH3), 1.85 (br, 6H;

CH2CH2CH2), 3.35 (br, m, 12H; OCH2CH2CH2NH), 3.54 (d, J= 14.1, 6H;

C(0)CH2P(0)), 6.24 (s, 1H; CH), 7.09 (d, J= 2.3, 3H; Ar), 7.20 (d, J= 2.6, 3H; Ar), 7.40

(m, 18H; Ar), 7.61 (t, J= 5.4, 3H; NH) 7.88 (m, 12H; Ar). 13C NMR (CDCl3): 8 (C=0)

= 164.82, 164.77; (Aromatic) = 153.6, 144.5, 142.0, 137.7, 132.8, 132.01, 131.97, 131.7,

131.5, 128.9, 128.8, 127.2, 122.4; aliphaticc) = 70.3 (OCH2), 37.9, 35.7, 34.7, 31.7, 30.6,

25.6. 31P NMR (CDCl3): 6 = 38.9. Anal. Called for C94H118N306P3S3: C, 71.68; H, 7.55;

N, 2.67. Found: C, 71.22; H, 7.88; N, 2.53.

Compound 3-2c. Method I: A mixture of 3-1 (0.30 g, 1.12 mmol),

4-dimethylaminopyridine (0. 13 g, 1.12 mmol) and EEDQ (1.11g, 4.48 mmol) were









dissolved in pyridine (10 mL). After stirring for 1 hour 2-4c (0.18 g, 0.28 mmol) was

added, and the reaction mixture was heated to 500C for 18 hours. After cooling to room

temperature, the solvent was removed in vacuo, and the residue was extracted with 9:1

CHCl3/MeOH solution followed by washing with 1N HC1. Organic phases were

collected, dried over MgSO4, and solvent was removed. The solid residue was dissolved

in diethyl ether and upon addition of pentane, the product precipitated out of the solution.

The compound was purified by column chromatography (SiO2, hexane/ether) to give

0.07 g of ligand 3-2c in form of a white solid (18 % yield). Method II: 3-1 (1.90 g,

6.90 mmol), 4-dimethylaminopyridine (0.31 g 2.53 mmol), mercaptothiazoline (0.85 g,

7. 13 mmol) and N, N'-dicyclohexylcarbodiimide (1.47 g, 7. 13 mmol) were dissolved in

dry dichloromethane (50 mL). After few minutes, the solution turned bright yellow and a

white solid separated out. The mixture was stirred for an additional 4 hours and 2-4c

(1.16 g, 1.84 mmol) was added. The resulting slurry was allowed to stir at room

temperature for 48 hours. The white solid of byproducts formed in the reaction was

filtered, and the solvent was removed in vacuo. Addition of ether to the condensed

reaction mixture dissolved the product leaving an amorphous mass of byproducts. The

organic solution was decanted, and within several days upon slow evaporation of solvent

pure product precipitated from the solution. Yield 0.8 g (31%). 1H NMR (CDCl3):

6 = 1.15 (s, 27 H; Ar-C(CH3)3), 2.13 (s, 9 H; Ar-CH3), 3.25 (t, 6 H; Ar-O-CH2CH2), 3.33

(t, 6 H; Ar-O-CH2CH2), 3.65 (d, J(H,P) = 14.1 Hz, 6 H CH2-POAr2), 6.67 (s, 1 H, C-H),

6.87 (b, 3 H; Ar-H), ), 6.67 (s, 1 H, C-H), 6.95 (b, 3 H; Ar-H), 7.42 (m, 18 H; P-Ar-H),

7.60 (t, 3 H; N-H) 7.89 (m, 12 H; P-Ar-H)). 31P NMR (CDCl3): 6 = 38.8.

MS [M+H] = 1406.5532 (Theoretical [M+H] = 1406.5596).









C omp ound 3 -3. A mixture of 3 -1 (4.7 3 g, 1 7. 12 mm ol), 2 -merc aptothi azol ine

(2.14 g, 17.95 mmol) and 4-dimethylaminopyridine (0.67 g, 5.48 mmol) was stirred in

dichloromethane (160 mL) for 30 min. N, N'-dicyclohexylcarbodiimide (5.34 g,

25.88 mmol) was added followed by additional portion dichloromethane (20 mL). After

6 h, solid 2-9a (4.10 g, 4.64 mmol) was added and the mixture was stirred for 24 h. The

slurry was filtered and the solvent was removed. The product was separated from the

reaction byproducts by dissolution in ether. The crude product was recrystallized from

methanol to give 6.70g (87%) of pure compound. 1H NMR (CDCl3): 6 = 0.45 (m, 18H;

CH2CH3), 1.10 (s, 18H; CCH3), 1.22 (s, 18H; CCH3), 1.42 (q, J= 7.4, 6H; CH2CH3),

1.57 (q, J= 7.1, 6H; CH2CH3), 1.82 (br, 6H; CH2CH2CH2), 3.30 (br, 12H;

OCH2CH2CH2), 3.54 (br, d, J= 13.6, 6H; C(0)CH2P(0)), 6. 18 (s, 1H; CH), 6.95 (d, J=

2.1i, 3H; Ar), 7.04 (d, J= 2.1i, 3H; Ar), 7.40 (m, 18H; Ar), 7.63 (t, J= 5.5, 3H; NH), 7.88

(m, 12H; Ar). 13C NMR (CDCl3): 8 (C=0) = 164.8, 164.7; (Aromatic) = 153.3, 142.4,

139.8, 137.9, 132.7, 131.94, 131.90, 131.6, 131.4, 128.9, 128.7, 127.9, 124.7; aliphaticc)

= 69.7 (OCH2), 42.7 (CH2 H2), 42.0, 39.3, 37.9, 37.7, 37.0, 35.3, 30.5, 29.7, 28.7, 9.7,

9.2. 31P NMR (CDCl3): 6 = 38.9. Anal. Called for C1ooH130N306P3S3: C, 72.39; H, 7.90;

N, 2.53. Found: C, 72.00; H, 8.10; N, 2.54.














CHAPTER 4
BINDING OF TRIVALENT F-ELEMENTS FROM ACIDIC MEDIA WITH A
C3-SYMMETRIC TRIPODAL LIGAND CONTAINING DIGLYCOLAMIDE AND
THIO DIGLYCOLAMIDE ARMS

4.1 Introduction

In the view of the ever-increasing use of nuclear power around the world,1-3 an

accelerated development of effective protocols for waste treatment increasingly becomes

imperative. Perhaps, one of the most significant obstacles faced in the separation science

is partitioning of minor actinides. Decades of work have been dedicated to the

development of amidic extractants for the f-element liquid-liquid waste separations.18,20-

22,108-116 Recently, a significant interest has been focused on one particular group of

amides diglycolamides (DGA).117-132 These completely incinerable tridentate, neutral

chelates are much more effective in coextraction of lanthanides and minor actinides than

commercially operating DIAMEX extractants.19,22-25,133-136 Moreover, unlike other

diamide-based compounds, DGAs exhibit significant selectivity within the lanthanide

series and track the increase in charge density .122,124, 125,129,130, 132 With this ability to

selectively bind metals within the series, DGAs offer many potential applications in

analytical chemistry in addition to waste partition. The further refinement of

diglycolamide based chelators and careful investigation of their binding potential may

help greatly improve efficiency of the lanthanide/actinide coextraction process, and a

fundamental understating of the remarkable selectivity of DGA may help in the

development of improved methods for partitioning of minor actinides.









In highly acidic nitric acid solutions common in nuclear waste reprocessing, two to

four molecules of diglycolamides appear to be involved in the coordination of trivalent f-

element ions during extractions, 122,126,129,132 and the reorganization of several lighting

DGA units onto a molecular platform could potentially improve the efficiency of the

extraction, as it was observed in some cases of carbamoylmethylphosphineoxide (CMPO)

- based extracts.49,51-54,76,78,79 A single ligand with three DGA arms would present the

metal with nine favourable donor groups, six of which are relatively hard amide oxygen

donors.

This chapter presents the synthesis of C3-Symmetric tripodal chelates bearing three

thio- and diglycolamide units precisely arranged on a triphenoxymethane platform along

with their evaluation as extractants in f-element separations. The ability of tripodal

ligands to extract trivalent lanthanides and actinides from nitric acid to the organic phase

was evaluated based on the comparison with the extraction data for the simple

[(Dii sopropylcarb amoyl)-methoxy]-N,N-dii sopropyl-acetamide (4-2). The influence of

the flexibility and lipophilicity of DGA arms in the ligand on the extraction profie has

been also investigated. To verify the contribution of the etheric oxygens to the tris-DGA

binding efficiency, a new C3-Symmetric derivative containing sulfur in the place of the

etheric oxygen has been synthesized, and the properties of both types of ligands have

been compared. Additionally, to determine the affect of immobilization of DGA on the

metal binding site geometry, three complexes of Ce(NO3)3, Eu(NO3)3 and Yb(NO3)3 with

tris-DGA ligands were synthesized. Solid structures of these complexes were elucidated

by ICR-MS and X-ray analysis and the results are discussed herein.










4.2 Results and Discussion

4.2.1 Linand Synthesis

The tri s-thio/diglycolamides (4-5 4-8) have been prepared by the reaction of

primary amines 2-4a or 2-9b with mono-substituted oxa/thio-pentaneamides (4-1, 4-3 and

4-4) and with the coupling agent benzotriazole-1 -yl-oxy-tri spyrrolidinophosphonium

hexafluorophosphate (PyBOP) as illustrated in Scheme 4-1. All final products have been

obtained in high yields and purity with relatively small synthetic effort.

The simple diglycolamide 4-2 was synthesized according to a modified literature

procedures and used as an extraction reference molecule. The synthetic pathways for

preparation of compounds 4-1 4-4 have been adapted from general procedures for

variety of diamides.137-139 Amines 2-4a51 and 2-9b (Chapter 2) were synthesized

according to previously developed procedures.



NH2 O?~

O ONR2 O n B N



Ri' O-


2-4a n =1
4-1X=0, R' =H, R=1Pr 2-b=
4-2 X= O, R' = N(1Pr)2, R =1Pr 3
4-3 X = O, R' = OH, R = nBu
4-4 X = S, R' = OH, R = 1Pr 4-5 R = 1Pr, X = O, n =1
4-6 R = 'Pr, X = O, n =2
4-7 R = nBu, X = O, n =1
4-8 R = 'Pr, X = S, n =1

Figure 4-1. Synthesis of C3-Symmtric tris-diglycolamides. (A) 1,4-dioxne, pyridine; (B)
PyBOP, diisopropylethylamine, DMF.









4.2.2 Extraction Experiments

The extraction experiments were performed on a series of eleven lanthanides,

152Eu, and 241Am radioisotopes. Solutions of 10-4 M metal nitrates in 1 M nitric acid were

mixed with equal volumes of 10-3 and 10-4 M organic solutions for approximately 20 h.

The reference molecule 4-2 was consistently used at a three times higher concentration

than the tris-diglycolamide derivatives for a fair comparison. The concentration of

lanthanide ions in the aqueous phase before and after the extraction were determined

spectrophotometrically (h = 655 nm), '"s or in the case of 241Am and 152Eu, they were

measured by a Canberra GammaTrac 1185 with Ge(Li) detector and AccuSpec-B multi-

channel analyzer. Extraction efficiencies were calculated using the formula:

%E = 100%(A1-A)/(A1-Ao), where A is the absorbance of the extracted aqueous phase

with the Arsenazo(III) indicator, Al is the absorbance of the aqueous phase before

extraction with the indicator, and Ao is the absorbance of metal-free 1 M nitric acid and

the indicator. The errors, based on the precision of the spectrophotometer and the

standard deviation from the mean of at least three measurements, were in most cases no

higher than two percent. The extraction percentage was further converted into

distribution ratios of the total metal ion concentration in the organic phase against the

total metal ion concentration in the aqueous phase (D = C[Morg]/E[Maq]) and in view of

the errors associated with the spectrophotometric technique, the maximum value that

could be measured for the extraction percentage and distribution ratio was 99.

4.2.2.1 Extraction properties of larne chelate vs. small dinlvcolamide

Typically, high distributions of trivalent f-elements in organic/acidic extraction

systems are reported with approximately ~100,000: 1 DGA to metal ion concentration

(e.g. 0.2 M N,N'-dimethyl-N,N' -diphenyl-4-oxapenanediamide DMDPhOPDA, 10-6 M









Ln(III)).122 With the tri-DGA ligand (4-5), comparable distribution ratios can be obtained

with a 10:1 ligand to metal ratio. To provide some context for the extraction of

efficiency of 4-5 under controlled experimental conditions, the properties of ligand have

been evaluated with respect to the performance the related DGA, 4-2, and since 4-5

contains three arms, the concentration of 4-2 was increased by factor of three. This

oversimplified comparison is used strictly to present the significant changes in the

efficiency and selectivity of studied diglycolamide chelates.





1 ~ o oa 4-2(1)
1 O 4 a o 4-5(1)
logD a a
oi d Cl g 5 5 4-2(2)
AO g 4-5(2)

-2 I

La Ce Pr Nd Eu Gd Tb Dy Er Tm vb

Figure 4-2. Extraction of 10-4 M solutions of trivalent lanthanides in 1 M HNO3 with
dichloromethane solutions containing ligand 4-2 at 3x10-3 M (1) or 3x10-4 M
(2) and solutions of 4-5 at concentrations of 10-3 M (1) and 10-4 M (2). Due to
the limitations of the spectrochemical assay, the error bars for D are quite
large at high extraction efficiency (>98%).

An experiment with the reference molecule 4-2 at 3x10-3 M (1) showed the typical

extraction pattern for diglycolamides, which gradually ascends across the lanthanide

series (Figure 4-2, Table 4-1). Once three DGA moieties were attached to the

triphenoxymethane platform (4-5), the efficiency of the ligand for the heaviest

lanthanides was remarkably improved. Only ten-fold excess of ligand 4-5 allowed for

quantitative removal of trivalent erbium, thulium and ytterbium from 1 M HNO3 Solution.

Even though the direct comparison of these two compounds (4-2 and 4-5) is impossible









due to the fundamental differences in the character of the extracted species, considering

only the concentrations required for high extractability, the efficiency and therefore

economy gain in the case of 4-5 is clearly evident. Moreover, this unique design of rather

flexible, nine oxygen donor cavitand allowed for much more sensitive ion size

recognition than in the case of any other DGA chelates.

Table 4-1. Extraction data (logD)* for ligands 4-2and 4-5 in dichloromethane. Extraction
of 10-4 M solutions of trivalent lanthanides in 1 M HNO3 with
dichloromethane solutions containing ligand 4-2 at 3x10-3 M (1) or 3x10-4 M
(2) and solutions of 4-5 at concentrations of 10-3 M (1) and 10-4 M (2).
Ligand 4-2 4-5
Cation 1 2 1 2
La (III) -0,43 -1,06 -1,06 -0,99
Ce (III) -0,09 -0,95 -0,66 -0,95
Pr (III) 0,08 -1,00 -0,30 -0,87
Nd (III) 0,16 -0,91 0,10 -0,72
Eu (III) 0,47 -1,00 1,16 -0,15
Gd (III) 0,51 -1,00 1,28 -0,08
Tb (III) 0,74 -1,00 1,69 0,09
Dy (III) 0,88 -0,91 1,69 0,21
Er (III) 1,02 -0,95 2,00 0,26
Tm (III) 1,00 -0,91 2,00 0,29
Yb (III) 0,99 -0,91 2,00 0,29
*D calculated based on the E % values.

Enhanced affinity of tris-DGA for the heaviest lanthanides and decreased attraction

for the lightest (La, Ce, Pr), resulted in improved separation factor between the elements

in the group (separation factor SFA/B=DA DB; the fraction of the individual distribution

ratios of two extractable solutes measured under the same conditions). For instance, the

value of the SF of Yb(III) and La(III) increased from SFYb/La=26 for 4-2 (1) to

SFYb/La=1 138 for 4-5 (1). Surprisingly, further dilution of the organic phase to a 1:1

metal to ligand ratio maintained a high extraction efficiency for the heaviest lanthanides;

over two thirds of Tm(III) and Yb(III) was transferred into the organic phase. At the

same time, with a three times higher concentration of 4-2 (2) (3x10-4 M), the extraction









was negligible and no separation was observed suggesting necessity of the significant

excess ofligand 4-2 to achieve appreciable extraction, and once again confirming the

efficiency gain through the DGA reorganization in the compound 4-5.

4.2.2.2 Linand flexibility vs. extraction performance

In work with calix[4]arenes and triphenoxymethane molecule appended with

CMPO arms, the extraction efficiency of the constructs was amplified by the increased

flexibility of the linker between the CMPO and the base skeleton (Chapter 2).76, This

trend can be attributed to the enhanced ability of ligand to satisfy the geometrical

requirements of the metal center, but often, the improvement in binding affinity comes at

the expense of selectivity. In 4-2, the DGA groups are tethered to the triphenoxymethane

platform by only two carbons, and to test the effect of the length of this arm linker on the

extraction ability, a new tris-substituted diglycolamide was synthesized with three

carbons linking the DGA arms to the triphenoxymethane base (4-6).

Considering the high efficiency of the "more rigid" chelate 4-5 at the 10: 1 ligand to

metal proportion, and expected performance enhancement of 4-6 over 4-5, the

experiment has been conducted with only 10-4 M concentration of 4-6 in the

dichloromethane (1:1 ligand to metal ratio). Interestingly, the new extractant did not

appear to be more effective or less selective than its "more rigid" equivalent. In fact, the

4-6 exhibits the same extraction behavior as compound 4-5, suggesting that the binding

environment in the two ligands is nearly identical. As it will be discussed later, the

coordination setting of the metal center in the complexes of 4-5 and 4-6 were found to be

nearly indistinguishable. 140 Apparently, both ligands are perfectly suited to fulfill the

coordination requirements of trivalent lanthanides in the solid state and in an organic

solution which explains identical extraction behavior.




Full Text

PAGE 1

SYNTHETIC DEVELOPMENT OF C3-SYMMETRIC TRIPHENOXYMETHANE BASED REAGENTS FOR THE SELECTIVE RECOGNITION AND SEQUESTRATION OF LANTHAN IDES AND ACTINIDES By KORNELIA K. MATLOKA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2006

PAGE 2

Copyright 2006 by Kornelia K. Matloka

PAGE 3

With my deepest love to Piotr, Nisi a, and my entire wonderful family.

PAGE 4

iv ACKNOWLEDGMENTS Words could never express my gratitude to my husband Piotr. I would have never come this far without his love, courage, c ontagious optimism and genuine faith in me. His dreams and bold aspirations inspired me to become the person I am today. I could have never succeeded without the love and support of my wonderful family. I thank my brave mom, Teresa, for motivati ng me to higher achievements and inspiring my independence. I thank my grandma, Karolc ia, and deceased sweet aunt Nisia for their absolutely unconditional love, which conti nues to support me through difficult times. I thank my great parents in law Marysia and Staszek, and my beloved sisters, Agnieszka and Paulina. Even far away they have alwa ys managed to keep my spirit up when I needed it the most. The sacrifices, patience and understanding of my wonderful family have allowed me to pursue my dreams and beco me successful both in my career and life. Eternal appreciation is given to people w ho made this great adventure possible, especially Dr. Henryk Koroniak and Dr. Violet ta Patroniak who encouraged me and Piotr to undertake the challenge of studies abroa d, Dr. James Deyrup, his wife Margaret and Lori Clark who warmly welcomed us in Ga inesville and the University of Florida. The last five years of my life have been incredible thanks to my friends. Enough thanks cannot be expressed to Iwona and Luka sz Koroniak for their hospitality and for making the transition between Poland and the United States so smooth and easy. I cannot imagine my life without the cr aziest of all Vicky Broadstree t, so French Flo Courchay and the chocolate addict, gorgeous Merve Erta s. They have truly accepted and loved me

PAGE 5

v for who I am, and always have been there fo r me. I could not ask for better friends than Travis Baughman, John Sworen, Sophie Bern ard and Josh McClellan. I am deeply grateful to all of them for sh aring with me both the fun and the struggles of the graduate school adventure. I must also thank many extremely talented and wonderful people I have been fortunate to be surrounded with over the years of my studies. I thank all past and present members of the Scott group in particular Matt Peters who intr oduced me to the subject of my research and taught me all the necessary laboratory techniques, my lab mate and entertainer Romeo H. Gill, the beautiful, ar tistic soul of Cooper Dean, adorable Eric Warner and rebellious Issac Finger. I thank my neighbour and lab mate who made me feel very welcome when I fi rst joined the Scott group, and on whom I could have always rely on, sweet and never rested recruiter Iv ana Bozidarevic and her eccentric husband Cira. A special acknowledgement is deserved by Ajaj Sah and Priya Srinivasan, who I had the pleasure to share research projects w ith. Their hard work si gnificantly contributed to the research presented in my dissertation. Other special thanks go to my dear lab mate Ranjan Mitra, who shared with me the great a nd the not so great days in the lab, patiently listening to my excitements or complaints, which must have scared him for life. In preparing this manuscript, I owe much to my colleagues Candace Zieleniuk, Eric Libra and impressive synthetic chemist Melanie Ve ige. From the bottom of my heart I thank them not only for putting up a great obstinate fight with the articles in editing my dissertation, but also for bei ng a wonderful vent for my frustrations. Finishing this manuscript would not have been possible without them.

PAGE 6

vi Tremendous gratitude goes to all the excellent sc ientists and educators, which I had pleasure to meet and learn from at the Univer sity of Florida, esp ecially members of my supervisory committee: Dr. Daniel Talham, Dr. William Dolbier, Dr. Khalil Abboud and Dr. James Tulenko, for the time and effort they invested in reading and discussing my dissertation. Sincere thanks and affecti on is extended to an amazing man, both professionally and personally, Dr. Tom Lyons for all the extraord inary conversations, openness and friendship. Many thanks go also to our research pa rtners from Argonne National Laboratory, Dr. Artem Gelis, Dr. Monica Regalbuto and Dr. George Vandegrift, and the Nuclear Energy Research Initiative (Grant 02-98) of the Department of Energy National Nuclear Security Administration for financial support. In closing, I wish to express my special gratitude to a remarkable person and passionate scientist, Dr. Mike Scott, whos e tremendous help, encouragement, brilliant creativity and enthusiasm has guided my resear ch from its conception to its completion. I will be eternally grateful to him for creating an absolutely incredible atmosphere of my graduate studies. His flamboyant personality and the biggest heart in the world have turned this initially terrifying experience into wonderful life ch anging journey. Through his inhuman patience, care and understanding he gave me the comfort to find myself in the new environment so I could finally start to grow as a scientist. His trust and support have slowly built my confidence, and the freedom to explore the chemistry he offered, inspired my true passion for science. I am deeply thankful for his support and the attention he lavished on my scientific development. His thoughtful advice proved invaluable, which cannot be repaid in wo rds alone. I would like him to know how

PAGE 7

vii grateful I am for not only having him as my research advisor, but simply for having known him. Mike, thank you.

PAGE 8

viii TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES.............................................................................................................xi LIST OF FIGURES..........................................................................................................xii ABSTRACT....................................................................................................................... xv CHAPTER 1 INTRODUCTION........................................................................................................1 1.1 Benefits of Nuclear Fuel Reprocessing..................................................................2 1.2 Liquid-Liquid Extraction Partitio ning Processes: an Overview.............................3 1.3 Characteristics of Trivalent Lant hanides and Actinides Valuable for Separation................................................................................................................7 1.4 Actinides Binding Controversy..............................................................................8 1.5 The Basics of Liquid Liquid Extraction Process..................................................9 1.5.1 Influence of the Organic Diluent on Extraction Process............................10 1.5.2 Influence of the Aqueous Phase Composition on Extraction Process........11 1.5.3 Thermodynamics of Biphasic Complexation.............................................12 1.5.4 Advantages of Large Extractants over Small Chelates..............................12 1.5.5 Types of Extraction Reactions....................................................................13 1.6 Research Objectives..............................................................................................14 2 CMPO FUNCTIONALIZED C3-SYMMETRIC TRIPODAL LIGANDS FOR LANTHANIDES AND ACTINIDES SEPA RATIONS IN THE NITRIC ACID LIQUID/LIQUID EXTRACTION SYSTEM............................................................16 2.1 Introduction...........................................................................................................16 2.1.1 Organophosphorous Extractants.................................................................16 2.1.2 Development of the Tris-CMPO Chelate...................................................18 2.2 Results and Discussion.........................................................................................21 2.2.1 Effect of the Structural Modifi cation of Triphenoxymethane Platform on the Tris-CMPO Extraction Profile..............................................................21 2.2.2 Ligand Flexibilit y vs. Binding Profile........................................................23 2.2.3 Complexation Studies with Bis-CMPO Compound...................................25 2.2.4 Attempts to Resolve the Tr is-CMPO Solubility Issue...............................27

PAGE 9

ix 2.2.5 Plutonium (IV) and Am ericium(III) Extractions........................................34 2.2.6 Comparison of Solid State Struct ures of Tris-CMPO Complexes of Trivalent Metal Ions.........................................................................................35 2.3 Conclusions...........................................................................................................41 2.4 Experimental Section............................................................................................42 2.4.1 General Consideration................................................................................42 2.4.2 Metal Ions Extractions................................................................................43 2.4.3 Isotopes Stock Solutions.............................................................................44 2.4.4 Synthesis.....................................................................................................45 2.4.5 X-Ray Crystallography...............................................................................59 3 DESIGN, SYNTHESIS AND EVALUAT ION OF PHOSPHINE SULFIDE BASED CHELATES FOR THE SEPA RATION OF TRIVALENT LANTHANIDES AN D ACTINIDES........................................................................62 3.1 Introduction...........................................................................................................62 3.2 Results and Discussion.........................................................................................66 3.2.1 Synthesis and Extraction Data....................................................................66 3.2.2 Crystal Structure Analysis..........................................................................69 3.3 Conclusions...........................................................................................................73 3.4 Experimental Section............................................................................................74 4 BINDING OF TRIVALENT F-ELEMENTS FROM ACIDIC MEDIA WITH A C3SYMMETRIC TRIPODAL LIGAND CONTAINING DIGLYCOLAMIDE AND THIO DIGLYCOLAMIDE ARMS..................................................................78 4.1 Introduction...........................................................................................................78 4.2 Results and Discussion.........................................................................................80 4.2.1 Ligand Synthesis........................................................................................80 4.2.2 Extraction Experiments..............................................................................81 4.2.2.1 Extraction properties of large ch elate vs. small diglycolamide........81 4.2.2.2 Ligand flexibility vs. extraction performance..................................84 4.2.2.3 Solvent effect on ligand extraction profile.......................................85 4.2.3 Investigation of Solid State Comp lexes of Trivalent Lanthanides.............89 4.2.4 Solution Structure of Extracted Species.....................................................94 4.2.5 Importance of the Etheric Oxyge n of Tris-DGA in Metal Binding...........96 4.3 Conclusions...........................................................................................................98 4.4 Experimental Section............................................................................................99 4.4.1 General Considerations..............................................................................99 4.4.2 1H NMR Experiment..................................................................................99 4.4.3 Synthetic Procedures................................................................................100 5 PYRIDINE N-OXIDE FUNCTIONALIZED C3-SYMMETRIC CHELATES FOR F-ELEMENS BINDING..................................................................................107 5.1 Introduction.........................................................................................................107 5.2 Results and Discussion.......................................................................................111

PAGE 10

x 5.2.1 Synthesis of Tris-PyNO Derivatives........................................................111 5.2.2 Extraction Experiments............................................................................112 5.2.3 Solid State Studies....................................................................................117 5.3 Conclusions.........................................................................................................120 5.4. Experimental Section.........................................................................................121 6 SUMMARY..............................................................................................................126 LIST OF REFERENCES.................................................................................................128 BIOGRAPHICAL SKETCH...........................................................................................139

PAGE 11

xi LIST OF TABLES Table page 2-1 Distribution coefficients ( D ) and extraction percentage (%E) for ligands 2-6a, 2-6b and 2-6c............................................................................................................23 2-2 Distribution coefficients ( D ) and extraction percentage (%E) for ligands 2-1, 2-7 and 2-10a..................................................................................................................27 2-3 Distribution coefficients ( D ) and extraction percentage (%E) for ligands 2-14a and 2-14b.........................................................................................................................3 1 2-4 Distribution coefficients ( D ) for the extraction of Pu(IV), U(VI), Am(III) and Eu(III) by ligands 2-6b, 2-14a and 2-14b in methylene chloride and 1-octanol......33 2-5 Selected bond lengths () fo r compounds 2-10b, 2-14a, [2-6aTbNO3](NO3)2, [2-14aTbNO3](NO3)2 and [2-6cBiNO3](NO3)2......................................................40 2-6 X-ray data for the crystal structur es of 2-10b, 2-14a and the complexes [2-6aTbNO3](NO3)2, [2-6cBiNO3](NO3)2 and [2-14aTbNO3](NO3)2...................61 3-1 Extraction percentage (%E) for lig ands 2-6a, 3-2a, 2-10 and 3-3. ............................69 3.2 Selected bond lengths () for compounds: 3-3 and [3-3Tb(NO3)3] complex............72 3-3 X-ray data for the crystal st ructures of 3-3 and [3-3Tb(NO3)3] complex...................73 4-1 Extraction data (log D ) for ligands 4-2 and 4-5 in dichloromethane............................83 4-2 Extraction data (log D ) for ligands 4-2 and 4-5 in octanol and dodecane....................87 4-3 Extraction data (log D ) for ligands 4-5 and 4-7 in dichloromethane............................88 4-4 X-ray data for the crystal structures of[4-5Ce][Ce(NO3)6], [4-6Eu](NO3)3, [3x4-2Yb](NO3)3, [4-5Yb](NO3)3 and [Yb-cage](NO3)3....................................106

PAGE 12

xii LIST OF FIGURES Figure page 1-1 Nuclear fuel cycle......................................................................................................... .2 1-2 Common pathways in sp ent fuel reprocessing..............................................................4 1-3 Popular neutral orga nophosphorus extractants..............................................................4 1-4 Fully incinerable extractants..........................................................................................5 1-5 SANEX extractants........................................................................................................6 1-6 SANEX nitrogen based extractants...............................................................................6 1-7 TALSPEAK extractants................................................................................................7 1-8 Illustration of the “all up” c onformation of the oxygen atoms on the triphenoxymethane platform....................................................................................13 2-1 Acidic organophosphorous extractants........................................................................17 2-2 Schematic depiction of proposed solution structure of the amer icium (III) nitratoCMPO complex at high nitric acid concentration....................................................19 2-3 Calix[4]arenas with CMPO functi ons at the narrow and wide rims............................20 2-4 Classic carbamoylmethyl phosphine oxide (CMPO)...................................................21 2-5 Synthesis of tris-CMPO 2-4........................................................................................22 2-6 Synthesis of tris-CMPO 2-10......................................................................................24 2-7 Metal extraction percentages (%E) for the liga nds 2-6a, and 2-10a using 10-4 M metal nitrate in 1 M nitric acid and 10-3 M ligand in dichloromethane...................25 2-8 Diagrams of the 10 coordinate +2 ca tionic thorium(IV) nitrate complex of 2-6 with two coordinated NO3 counterions...................................................................25 2-9 Bis-CMPO compound 2-7...........................................................................................26 2-10 Metal extraction per centages (%E) for the ligands 2-6a, and 2-7.............................26

PAGE 13

xiii 2.11 Fragment of the crystal structure of 2-6b molecules forming hydrogen bond connected network....................................................................................................28 2-12 Diagram of the solid-state structure of 2-10b............................................................29 2-13 Synthesis of tris-CMPO 2-14....................................................................................30 2-14 Diagram of the solid-state structure of 2-14a............................................................31 2-15 Metal extraction percen tages (%E) for the ligands 2-1, 2-6a, and 2-14a..................32 2-16 Metal extraction percen tages (%E) for the ligands 2-6b, 2-14a, and 2-14b..............34 2-17 Diagrams of neodymi um(III) complexe s of 2-6b......................................................36 2-18 Diagram of the structure of [2-6cBiNO3](NO3)2......................................................37 2-19 Diagram of the structure of compound [2-6aTbNO3](NO3)2.38 2-20 Diagram of the struct ure of compound [2-14aTbNO3](NO3)2.................................39 3-1 Structures of sulfur based extractants..........................................................................62 3-2 Structure of the aroma tic dithiophosphinic acids........................................................64 3-3 Anticipated binding mode for tris-CMPS extractant...................................................65 3-4 Synthesis of tris-CMPS extractants.............................................................................66 3-5 Comparison of metal binding by tris-C MPO (2-6a) and tris-CMPS (3-2a)................67 3-6 Comparison of metal binding by 3-2a and 3-3............................................................68 3-7 Comparison of metal binding by tris -CMPO (2-10a) and tris-CMPS (3-3)................68 3-8. Diagram of the solid-state structure of 3-3.................................................................70 3-9 Diagram of the structures of [3-3Tb(NO3)3]...............................................................71 4-1 Synthesis of C3-symmtric tris-diglycolamides............................................................80 4-2 Extraction of trivalent lanthanides with 4-2 and 4-5 in dichloromethane...................82 4-3 Extraction of trivalent lanthanides with 4-5 in dichloromethane and 1-octanol.........85 4-4 Extraction of trivalent lanthanides with 4-2 and 4-5 in 1-octanol and n-dodecane.....86 4-5 Extraction of trivalent lanthanides with 4-5 and 4-7 in dichloromethane...................88 4-6 The tricapped trigonal prismatic (TTP ) geometry around nine coordinate Yb(III)....89

PAGE 14

xiv 4-7 Diagram of the structures of [4-5Yb]3+ and [4-2Yb]3+..............................................90 4-8 Diagram of ytterbium encapsulated by a cage-like derivative of tris-DGA compound.................................................................................................................94 4-9. Superimposed 1H NMR spectra of ligand 4-5............................................................95 5-1 Structures of the most ex tensively studied amine oxides..........................................107 5-2. The electron distributi on in pyridine N-oxide..........................................................108 5-3 Resonance structures of pyridine N-oxide.................................................................108 5-4 Synthesis of tris-pyridine N-oxides...........................................................................111 5-5 Structure on the hexachlorinated c obaltocarborane sandwich anion (COSAN).......113 5-6. Stacked 1H NMR spectra of ligand 5-4....................................................................115 5-7 Stacked 1H NMR spectra of ligand 5-1.....................................................................117 5-8 Two geometric extremes of meta l binding by the substituted N-oxide.....................118 5-9 Diagram of the coordina tion environment of [5-4Yb(NO3)2](NO3).......................119 5-10. Coordination environment of ytterbium (III) in the complex with 5-4..................119

PAGE 15

xv Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHETIC DEVELOPMENT OF C3-SYMMETRIC TRIPHENOXYMETHANE BASED REAGENTS FOR THE SELECTIVE RECOGNITION AND SEQUESTRATION OF LANTHAN IDES AND ACTINIDES By Kornelia K. Matloka May 2006 Chair: Michael J. Scott Major Department: Chemistry The prospective increase of global nuclear power utiliz ation requires a significant modification of nuclear wast e management to help overcome related environmental, economic, and political challenges. The collaborative effort with Argonne National Laboratory has led to the development of reag ents with the ability to selectively bind lanthanides and actinides in conditions si mulating nuclear waste solutions, and could contribute to the advancement of the nuclear fuel and waste reproces sing strategies. The focus has been placed on the fundamental chem istry of metal-ligand interactions both in the solid state complexes and in solution. Th e study of these comple xes assisted in the design of improved systems for metal separation. A sequence of tripodal chelates bearing a variety of binding moieties, diglycol amide (DGA), thiodiglycolamide (TDGA), carbamoylmethylphosphine oxide (CMPO), car bamoylmethylphosphine sulfide (CMPS) and pyridine N-oxide (PyNO), precisely arranged on a C3-symmetric triphenoxymethane molecular platform were synthesized. The im pact of structural m odifications of these

PAGE 16

xvi ligands on their affinity for f-element ions in 1 M nitric acid extraction system has been evaluated. The preorganization of three diglycolam ide binding units on the triphenoxymethane scaffold resulted in significant enhancement of the extraction efficiency and selectivity within trivalent lanthanide ions series. A dditionally, the tris-diglycolamide chelate has been recognized as the first single, purely oxygen based donor capable of fully satisfying the tricapped trigonal prismatic geometry fa vored by nine coordinate d lanthanides. The CMPO-based ligand has shown an excellent binding efficiency and a remarkable selectivity for tetravalent actinides. The stru ctural modifications of this chelate system led to the development of one of the most efficient plutonophiles. Experiments with CMPS, TDGA and PyNO compou nds provided valuable information about the coordination preferences of the tested cations in the solid state complexes and in solution. The synthetic development and characteriza tion of tripodal chelat es and their metal complexes are presented herein. The impact of the structural deriva tization of ligands on lanthanide and actinide ion extraction and se paration is also disc ussed along with its implication towards potential appl ications in waste reprocessing.

PAGE 17

1 CHAPTER 1 INTRODUCTION According to the World Nuclear Associat ion there are 441 nucl ear power reactors currently operating in 31 c ountries, generating over 368 giga watts of electr ical energy worldwide.1,2 The United States alone houses 103 reactors that pr oduce approximately 20% of total electricity. The International Energy Outlook projects a 57% increase in the world energy consumptions over the 2002 to 2024 time period.3 The most significant energy demand increase is expected to aris e among the emerging economies of Asia, in particular China and India, as well as econom ies of Eastern Europe and the former Soviet Union. Considering that the renewable sour ces of energy (biomass, solar, wind) alone cannot produce enough energy to sustain gl obal development, the nuclear power expansion appears to be economically and e nvironmentally reasonable alternative to the fossil fuel energy. The expansion of the nuclear power would si gnificantly reduce the greenhouse gas emissions like carbon dioxide sulphur dioxide and nitrogen oxides generated from fossil fuel combustion (coal, oil and natural gas), and ultimately reduce the problematic air pollution. Moreover, nucl ear energy could be efficiently used to produce large quantities of hydroge n gas, a potential major en ergy carrier that is clean and environmentally friendly. Contrary to public opinion, the radi ological hazard of nuclear waste can be reduced, and the efficien cy of nuclear power pl ants can be further improved through the recycling of the spent nucl ear fuel. Currently, the potential threat of nuclear weapon proliferation and expens ive reprocessing technologies prevent the

PAGE 18

2 United States nuclear industry from wast e reprocessing. In order to extend the exploitation of nuclear power, waste treatment is inevitable. 1.1 Benefits of Nuclear Fuel Reprocessing In a typical light water reactor the operational lifespan of a fuel rods is only three years, and since they still c ontain approximately 96% of th eir original energy potential, the recycling of fuel rods offers both si gnificant economic and environmental impacts.4 In the US, spent fuel rods currently remain in storage at the reactor site, awaiting eventual transport for disposal at a government repository. Protract ed litigation, however, may keep the Yucca Mountain Repository closed for many years to come.5 The partitioning of radioactive waste followed by the transmut ation of problematic long-lived radionuclei would reduce uncertainties re lated to the long-term wast e storage (several hundred thousand years) in the geological repositories. Figure 1-1. Nuclear fuel cycle.

PAGE 19

3 After removal from a reactor, a spent fuel rod contains mainly a mixture of 235U and 238U along with small amount of 239Pu intermixed with radioactive and nonradioactive fission products. An efficient spent fuel repro cessing protocol would involve the recovery of these two major components (U and P), followed by the separation of long-lived transplutonium radionuclides from other relatively innocuous elements. Uranium and plutonium can be reused as a fuel, while the long-lived actinides could further be transmuted by neutron bombardm ent into short-lived and non-radioactive elements, decreasing long-term radiotoxicity of waste and practically eliminating the waste disposal problem. Unfortunately, the transmutation of actinides would be impeded by even small amounts of lanthanide ions, and a very efficient protocol for their separation must be developed for this process to work. 1.2 Liquid-Liquid Extraction Parti tioning Processes: an Overview Nuclear waste reprocessing begins with the di ssolution of used fuel rods or waste in concentrated nitric acid. To date, the most successful technique used for the partitioning of nuclear waste components is the li quid-liquid extraction separation, which predominately employs organophosphorus extracta nts. Industrial uranium and plutonium recovery utilizes neutral monodentate tr i-n-butyl phosphate [TBP] in the PUREX (Plutonium/URanium EXtraction) process.6,7 Other popular organophosphorus compounds like bidentate CMPO [oct yl(phenyl)-N,N-diisobutylcarbamoyl methylphosphine oxide] and DHDECMP [dih exyl-N,N-diethylcarbamoylmethylene phosphonate]8 have also been widely studied as chelating agents for both actinides and lanthanides.

PAGE 20

4 Figure 1-2. Common pathways in spent fuel reprocessing. The extraction power of CMPO and TBP, originally studied by Horwitz and coworkers, has found application in the co mmercially operating TRUEX (TRansUranium elements EXtraction) process9-11 for the removal of both la nthanide and actinide ions from high level liquid waste generated during PUREX reprocessing.12-14 O P N P O O O O O N O P O O O TBP CMPO DHDECMP Figure 1-3. Popular neutral organophosphorus extractants.

PAGE 21

5 The demand to fully accomplish environmental mission of the waste treatment has driven the development of purely C, H, N, and O based extractants that would not generate secondary waste after final incineration. Significan t attention has been focused on diamide based extractants15-22 in particular N,NÂ’-dimet hyl-N,NÂ’-dibutyl tetradecyl malonamide (DMDBTDMA). Based on DM DBTDMA and later N,NÂ’-dimethyl-N,NÂ’dioctyl hexyloxyethyl malonamide (DMDOH EMA) the DIAMEX (DIAMide Extraction) process has been developed and strongly pr omoted in France as an environmentally friendly alternative to TRUEX.23-25 O N O N O N O N O DMDBTDMA DMDOHEMA Figure 1-4. Fully incinerable extractants. None of these hard-donor oxygen based extr action systems, offer a solution to the most problematic partitioning of trivalen t actinides from the chemically similar lanthanides. Interestingly, some potentia l for minor actinides separation has been presented by soft-donor ligands employed in the SANEX (Selective ActiNide Extraction) process. The SANEX procedure uses either acidic sulfur bearing or neutral nitrogen based extractants. The most promisi ng separation has been achieved by bis(2,4,4trimethylpentyl)dithiophosphinic acid known as Cyanex-301 (Figure 1-5).26-28 Nevertheless, the commercial application of this extraction system has been limited by low radiation stability, sulfur and phosphorus containing degradati on products and the necessity to adjust the pH to 3-5. To part ially overcome the problematic pH adjustment, a modified solvent mixture of bis(chlorophe nyl)dithiophosphinic acid and the hard donor

PAGE 22

6 tri-n-octyl-phosphine oxide (TOPO) extracta nt has been used in the German process ALINA,29 but the radiological stab ility and secondary waste generation problems remain. P SH S P SH S P O Cl Cl Cyanex-301 Modified Cyanex-301 TOPO Figure 1-5. SANEX extractants. In the second variant of the SANEX concept, either bis-5,6-substituted-bistriazinyl-1,2,4-pyrid ines (BTPs, R1, R2 = H or alkyl group)30,31 or a synergistic mixture of 2-(3,5,5-trimethylhexanoylamino)-4,6-di-(pyr idine-2-yl)-1,3,5-tr iazine (TMAHDPTZ)32, and octanoic acid are used (Figure 1-6). Hydr olytic instabil ity of BTPs and, as in the case of S-bearing compounds, the necessity of some pH adjustment for TMAHDPTZ along with solubility issue, significantly limits the potential applica tions of the N-based extraction system in the separation of triv alent lanthanides (Ln) and actinides (An). N N N N N NH O N N N N N N N R1 R1 R1 R1 R2 BTPs R1, R2 = H, alkyl TMAHDPTZ Figure 1-6. SANEX nitroge n based extractants. An alternative to the SANEX extraction system is the TALSPEAK process (Trivalent Actinide/Lanthanide Separation by Phosphorus reagent Extraction of Aqueous Komplexes) based on selective stripping of An (III) from a mixture of trivalent Ln and An bound by diethylhexylphosphoric acid (HDEHP ), with a aqueous solution of

PAGE 23

7 diethylenetriaminopentaacetic (DTPA) and hydr oxocarboxylic acids (lactic, glycolic or citric).33,34 Although the process is relatively efficient, TALSPEAK suffers from common drawbacks, namely the necessity of the pH adjustment, limited solvent loading with metal ions and secondary phosphorus waste generation.35 P O O OH O N N N O HO O OH OH O O HO O HO HDEHP DTPA Figure 1-7. TALSPE AK extractants. After years of intensive re search An(III)/Ln(III) separati on remains one of the key problems facing the partitioning and transmut ation of nuclear waste management. Hopefully, the development of new or th e improvement of historically proven technologies will help raise public awareness and acceptability of nuclear power as the most viable energy source to sustain global development without significant environmental consequences. 1.3 Characteristics of Trivalent Lanthanide s and Actinides Valuable for Separation The chemical properties of lanthanides and actinide are very similar. Historically, the close relationship between these two groups of elements helped predict the properties of the transplutonium elements, which later resulted in their synt hesis and isolation. From a waste reprocessing perspective, however such similarities are disadvantageous. The identical oxidation states of trivalent ac tinides and lanthanides and an approximately equal ionic radii, make the separation particularly difficult. Both trivalent lanthanides and actinides are hard acids and form comple xes preferentially with hard bases through strong ionic interactions. Careful examination of their electronic structures and binding

PAGE 24

8 with less compatible soft bases revealed sm all but important differences between these two groups of cations.36 One of the major differences between lant hanides and actinides is the exceptional stability of the lanthanide sÂ’ trivalent oxidation state.36 This stability can be attributed to the effective shielding of the 4f orbitals pr ovided by the outer 6s and 5d orbitals. Across the lanthanides series electrons successively populate the 4f orbitals The radii of the outer 6s and 5d orbitals are sign ificantly bigger than the average radius of the 4f orbital, which effectively shields the f electrons and stabilizes the third oxidation state.37 In the case of actinides, more spatially extended 5f orbitals are no longer effectively shielded by the outer 7s and 6d orbitals, which results in th e decreased stability of the trivalent state. The later actinides (transplut onium elements), however, show somewhat higher stability of the third oxidation state with respect to the early member s of the group. This effect originates from slightly more pronounced decrease of the 5f orbital radius with respect to the size of 7s and 6d orbitals as the elements become heavier. This provides some shielding of f orbitals, although it is not as e ffective as in the case of 4f orbitals of lanthanides.36,38 The separation between the lanthanide a nd actinide group can be well represented by the partition of one isoelectronic pair of lanthanide and actinide ions, for instance americium and europium. This representati on is fairly accurate since the extent of separation between these two groups depends mo stly on the character and strength of the interaction between the metal ion and ligand an d not on the contraction of the ionic radii. 1.4 Actinides Binding Controversy The interaction between the trivalent f-el ement cations and hard bases are ionic (electrostatic) in nature. Th e lack of bonding orbital overlap constrains results in a wide

PAGE 25

9 range of coordination numbers and geometries that are controlled by the electronic and steric dynamics of the complex.36,39 In the 1950s, the concept of a small degree of covalency in the f-element interac tions with soft bases was proposed.40 The origin of such phenomenon has become a subject of controversy ever since. The trivalent An/Ln separation studies revealed that thes e ions can be separated by a preferential elution of An from a cation exchange resin column using concentrated hydrochloric acid.40 This phenomenon was attributed to the small degree of covalent interactions between the actinide ion a nd chloride which was explained by the participation of the 5f orb ital in the binding. Also, a sm aller energy difference between the 5f and 6d orbitals of actinides with resp ect to the 4f and 5d or bitals of lanthanides helped elucidate the greater sensitivity of actinides to their chemical environment.36,40 The effective shielding of the 4f orbital by the 6s and 5d orbitals and the energetic mismatch among orbitals results in the weaker interactio ns between Ln(III) and the ligands. According to the later literature analysis, however, this undeniable small covalent contribution reflected by stronger complexation of tr ivalent actinides with soft donor ligands involves more likely the 7s and/or 6d orbitals rather th an 5f, which is also consistent with the flexible metal coordination due to the spherical character of the s orbital.37 1.5 The Basics of Liquid Liquid Extraction Process The principle of the liquid – liquid extraction states: “I f two immiscible solvents are placed in contact, any substance soluble in both of them will di stribute or partition itself between two phases in a definite proportion.”41 The solvent extraction separation process used in waste reprocessing is based on the transfer of a metal cation from an aqueous phase (mineral acid) in to an immiscible organic ph ase with simultaneous charge

PAGE 26

10 neutralization.42,43 During the extraction, a variety of species are formed in combination of metallic salt with water, mineral acid and or ganic solvent. The extent of extraction can be expressed by the distribution ratio ( D = [Morg]/ [Maq]) of the total metal ion concentration in the organic phase ( [Morg]) against the total metal ion concentration in the aqueous phase ( [Maq]), or percentage wise (%E = [ D /( D + 1)] 100). The separation effectiveness of two different species can be assessed based on the separation factor (SFA/B= DA/ DB; the fraction of the individual distribut ion ratios of two extractable species measured under the same conditions). For ex ample, the successful separation of two elements can be accomplished with separati on factor of 100, which corresponds to 99% of partitioning efficiency in one contact.42 In the liquid – liquid extraction pr ocess the extractant is commonly diluted/dissolved in a water immiscible orga nic solvent. Therefore, successful and efficient separation is determined not only by the choice of proper ligand, but also by the right choice of organic solvent and the content of the aqueous solution. 1.5.1 Influence of the Organic Diluent on Extraction Process Choosing the organic diluent is perhaps as important as the ligand. The character of the diluent directly influe nces the physical and chemical properties of the extractant. The selection of solvent offers control over the organic phase dens ity, viscosity, freezing, boiling and flash points, separation ability of the mixed aqueous and organic phases, ligand solubility in the aque ous phase, third phase formati on (formation of two separate organic phases), as well as, the distribu tion of the extractable species between two immiscible phases, separation efficiency, ki netics of extraction, a nd even chemical and radiolytic stability of the extractant.44 There are multiple literature examples where the

PAGE 27

11 efficiency of extraction differs in different organic solvents by several orders of magnitude.43 Depending on solvation properties of the solvent, the solubility and therefore stability, as well as, stoichiometry of the extracted species are affected. For example, as mentioned earlier both trivalen t lanthanides and actinides are typical hard acids therefore they preferentially react with hard bases. The sligh tly stronger interaction between the hard metal ion and soft base can be facilitated only in weakly solvating organic solvents. In the aqueous environmen t soft bases cannot compete with water for the metal binding.36 Even with these common trends, the complexity of the extraction process does not allow for the accurate predic tion of the solvent effect on the separation, which ultimately hampers fast progression in this field.43 1.5.2 Influence of the Aqueous Phas e Composition on Extraction Process Metal ions hydrolyze unde r low acid concentration, which may result in the improvement or deterioration of the extracti on efficiency and/or selectivity, depending on the extraction mechanism.45,46 Higher acid concentration prevents metal hydrolysis by decreasing the activity of water, but at the same time, it may also negatively influence extraction by protonating the ligand. Sin ce one of the energetic requirements for successful phase transfer in the liquid-liquid extraction process is full or partial dehydration of the metal ion upon complexation with a ligand, a lower water activity results in a decreased net rate of the water exchange of the cation, improving the phase transfer.42,43 A similar effect can be induced by a dding the salt of small, extraction inert cation with high charge density and a high degree of hydra tion. For example, lithium nitrate is able to strongly compete with felement cations over the water binding. Also, an increased concentration of salt may affect the form of the extracted species and have either a positive of negative eff ect on the extraction processes.

PAGE 28

12 Ultimately, the addition of salt should be avoided if a similar enhancement of extraction properties can be obt ained using nitric acid as the “salting” out agent. Opposed to inorganic salts, at the end of the extraction process, acid can be easily evaporated from the system without increasing the volume of waste. 47 1.5.3 Thermodynamics of Biphasic Complexation Thermodynamic changes in the complexation process of f-block cations are predominately controlled by the changes in the solvation of the cation and the ligand.36 As extensively discussed by G. R. Choppin,36 upon complexation, these highly hydrated metal ions release water molecules, whic h results in positive entropy change. The dehydration process requires en ergy in order to break the in teractions between the water and the cation, as well as, other water molecule s, which contributes to the positive change of the net enthalpy of the complexation. Th e negative contribution to the net enthalpy and entropy of complexation results from th e binding of the cation and ligand, but it is less significant than the contri bution from the dehydration. It is believed that the slightly higher degree of covalency in the actinid e complexes with soft donor ligands may provide some additional small thermodynamic stabilization to the complex by slightly decreasing the net enthalpy of the process.48 1.5.4 Advantages of Large Extractants over Small Chelates The commercially operating partition pr ocesses commonly employ small organic chelates. Typically two to four molecules ar e involved in the transfer of a metal cation from an aqueous into the organic phase. With that in mind, a molecule with several ligating moieties attached to some molecula r support may have many advantages. The new ligand may produce a significant change in the extraction pr operties due to the entropic effect (release of water from highly hydrated f-element cations upon the

PAGE 29

13 complexation) in addition to the modified composition of the complex produced by multiple arms.49 We have found the C3-symmetric triphenoxymethane mol ecule to have great utility for the preparation of ligands with three extended arms,50,51 much like calix[4]arene can serve as a base for four arm constructs.49,52-54 HO OH HO R R R R R R H "all up" conformation Figure 1-8. Illustration of the “all up” conformation of the oxygen atoms on the triphenoxymethane platform relative to its central methane hydrogen. Using simple techniques, we can incorpor ate a variety of binding moieties onto a triphenoxymethane base through different alkyl linkers. On ce the three phenols have been derivatized, the triphenoxymethane mo lecule adopts exclusively an “all up” conformation wherein the oxyge ns orient in line with the central methine hydrogen.50 The solubility properties of the final liga nd can be easily modulated through simple substitutions at the 2 and 4 positions of the phenol s, as well as substitutions at the ligating units. The following chapters report the synthesis of a series of C3-symmetric compounds along with a detailed investiga tion of binding properties of these new chelates. 1.5.5 Types of Extraction Reactions As summarized by K. L. Nash, 43 there are three types of solvent extraction reactions: metal-complex solvation by an or ganic extractant, metal-ion complexation by organic extractant and rare, ion-pair forma tion between an anionic aqueous complex and a positively charged (protonated) organic extractant. In the first case, a neutral complex

PAGE 30

14 of the metal ion with the conj ugated base of the mineral acid is formed typically in the aqueous phase, and upon solvation with the or ganic extractant is transferred to the organic phase. Neutral organophosphorus compounds, such as phosphates or phosphine oxides, as well as, ethers or ketones ex tract accordingly to this mechanism and proportionally to their Lewis base strengt h. The second extraction mechanism is observed in the case of acidic extractants, like phosphoric an d carboxylic acids that form complexes with water soluble metal ions in or near the interfacial zone. The combination of acidic and neutral extractants often leads to the enhanced distribution ratio and is commonly referred to as the synergistic effect. Even though the mechanism for this effect is unknown, the enhanced stability of the extracted species is attributed to the increased hydrophobicity of the complex that can be attained either by accommodating the solvating molecule (synergist) in the e xpanded metal coordination sphere or by water replacement.42,55 1.6 Research Objectives The primary objective of the collaborativ e research effort with Argonne National Laboratory has been the development of reag ents with the ability to selectively bind lanthanides and actinides in conditions simu lating nuclear waste solutions that could contribute to the advancement of the nuclear fuel and waste reproces sing strategies. The focus has been placed on the fundamental chem istry of metal-ligand interactions both in the solid state complexes and in solution. Th e study of these comple xes assisted in the design of improved systems for metal separation. A sequence of tripodal chelates bearing a variety of binding moieties: diglycol amide (DGA), thiodiglycolamide (TDGA), carbamoylmethylphosphine oxide (CMPO), car bamoylmethylphosphine sulfide (CMPS) and pyridine N-oxide (PyNO) precisely arranged on a C3-symmetric triphenoxymethane

PAGE 31

15 molecular platform were synthesized. The im pact of structural m odifications of these ligands on their affinity for f-element ions in 1 M nitric acid extraction system has been evaluated with its implication towards potenti al applications in waste reprocessing.

PAGE 32

16 CHAPTER 2 CMPO FUNCTIONALIZED C3-SYMMETRIC TRIPODAL LIGANDS FOR LANTHANIDES AND ACTINIDES SEPA RATIONS IN THE NITRIC ACID LIQUID/LIQUID EXTR ACTION SYSTEM 2.1 Introduction As the world demand for energy increases, the use of nuclear power will continue to grow in countries throughout the world. In order to sustain the worldÂ’s development many new reactors will be bui lt and generate proportionally more radioactive waste.1-3 To maintain the low cost of nuclear energy and ensure environmen tal safety, the spent nuclear fuel will need to be recycled worldw ide. This process, however, will require a significant advancement in the separation ch emistry. The efficiency of separation technologies developed in the 50Â’s needs to be improved either th rough modification of existing methods or/and the development of new procedures. 2.1.1 Organophosphorous Extractants The most successful technique of nucle ar waste partitioning is liquid-liquid extraction, where elements of the irradiated material dissolved in nitric acid are successively removed by a sequence of organic solvents. Initially, acidic organophosphorous extractants like bi s(2-ethylhexyl) phosphoric acid,56-58 diisodecyl phosphoric acid,59 and bis(hexylethyl) phosphoric acid60 have been involved in waste reprocessing procedures (Figure 2-1).9,61 These procedures, however, suffer from significant drawbacks. In order to successf ully extract trivalent elements the acidic ligands require very low acidi ty and exhibit poor selectivity for trivalent actinides over other fission products.9

PAGE 33

17 P O O O HO bis(2-ethylhexyl) phosphoric aciddiisodecyl phosphoric acid P OO O OH bis(hexylethyl) phosphoric acid P OO O OH Figure 2-1. Acidic organophosphorous extractants. Along with acidic extractants, neutra l organophosphorous compounds have also been actively studied. Previous resear ch has focused on compounds like tributyl phosphate (TBP),57 a mixed trialkylphosphine oxides,62 dihexylN,N diethylcarbamoylmethyl phosphonate (DHDECMP),63-67 and octyl(phenyl)N,N diisobutylcarbamoylmethyl phosphine oxide (CMPO) (Figure 1-3, Chapter 1).68,69 Studies have shown that in order to afford efficient extraction by the monofunctional extractants low acidities and occasional us e of a salting out agents are required.9,57,62 In contrast, bifunctional extractants such as DHDECMP and CMPO can effectively operate at much higher acidities. It has been proposed that both the carbonyl and phosphoryl oxygens of CMPO are directly involved in metal binding,64,70 and the efficiency of this bidentate ligand has been attributed to th e chelate effect provided by the two donors. According to Muscatello et al. in highly acidic media the carbonyl portion of ligands do not participate in metal binding but rather protect the metal-phosphoryl bond from the reaction with hydronium ions.71,72 Thus, in waste solutions typically of 1 to 3 M HNO3 the built-in buffering effect f acilitates the extrac tion without necessitating an adjustment in pH.

PAGE 34

18 Industrial recovery of uranium and plutoni um from spent nuclear fuel utilizes neutral monodentate tri-n-butyl phosphate [TBP] in the PUREX (Plutonium/URanium EXtraction) process.6,7 The mixture of CMPO and TBP extractants, originally studied by Horwitz and coworkers, is applied in the commercially operating TRUEX (TRansUranium elements EXtraction) process9-11 for the removal of both trivalent lanthanide and actinide ions from the hi gh level liquid waste generated during PUREX reprocessing.12-14 Efficient extraction in these processes can be accomplished only with high ligands concentrations (over 1 M), whic h results in the generation of significant amount of unwanted secondary waste. Moreove r, the lack of sele ctivity of CMPO and DHDECMP creates a need for another separation step. Despite years of research studies, the final partitioning of the trivalent actinides from the trivalent lanthanides as well as selective separation of residual plutoniu m and uranium from the PUREX raffinate remains a challenging task facing a ne w generation of separation chemists. 2.1.2 Development of the Tris-CMPO Chelate The structure of the classic CMPO has been modified in various ways to develop new systems that would compensate for the lack of selectivity for tr ivalent actinides over lanthanides in the TRUEX process.71,72 The solution structure of the Am(III) complex with CMPO formed during the TRUEX procedure was examined by Horwitz et. al.73 Their work suggests that in the neutral complex the Am(III) ion is bound by three CMPO molecules and three nitrate anions, and the addi tional nitric acid molecules are attached to the CMPO carbonyl oxygens via hydroge n bonding interactions (Figure 2-2).49,73,74 The resulting complex with tetrav alent plutonium would involve two to four CMPO ligands, while in the case of trivalent lanthanides more than one CMPO is bound. With these properties in mind, a molecule with several CMPO groups at tached to some molecular

PAGE 35

19 support may have many advantages. The new lig and may produce a sign ificant change in the extraction properties due to the entropic effect (release of water from highly hydrated f-element cations upon the complexation) in a ddition to the modified composition of the complex produced by multiple arms.49 CMPOHNO3NO3NO3NO3HNO3CMPO CMPOHNO3Am3+ O P C8H17 N O Am3+O N O O HNO3 Figure 2-2. Schematic depiction of proposed solution structure of the americium (III) nitrato-CMPO complex at high nitric acid concentration adapted from F. ArnaudNeu et. al. Perkin Trans. 2 1996.49 Inspired by these ideas, a variety of calixarene and calixarene-like multi-CMPO supported compounds have been developed,49,52,75-77 and indeed the preorganization of several ligating units on the molecular pl atform significantly improved the binding efficiency and/or selectivity of the ligand. Particularly interesti ng are cases of CMPO moieties secured to resorcinarene, and the most extensively studied calix[4]arene platforms.49,52-54,76,78,79 Compounds with four CMPO moieti es tethered to the narrow and wide rims of calix[4]arenes (Figure 2-3) have shown an increased actinide affinity relative to the mono-CMPO ligand, but their la nthanide extraction si gnificantly varied across the series. Taking inspiration from the suggested c oordination environmen t of the extracted species in the TRUEX process and the cal ix[4]arene work, our research group has developed chelate system with three preci sely arranged carbamoylmethylphosphine oxide moieties attached to a rigid, triphe noxymethane platform (tris-CMPO).51

PAGE 36

20 O HN O P Ph Ph O 4 O HN O P Ph Ph O 4 C5H11 Figure 2-3. Calix[4]arenas with CMPO f unctions at the narrow and wide rims. In the design of the tris-CMPO ligand, we hoped to achieve high distribution coefficients for the selected metal ions, while maintaining high ion selectivity and great stability toward hydrolysis in acidic me dia. The basicity of the phosphoryl oxygen increases in the order: phosphate (RO)3PO < phosphonate (RO)2RPO < phosphinate (RO)R2PO < phosphine oxide R3PO; were R=alkyl.43 An increase in the basicity improves the extraction efficiency through a st ronger interaction of the ligand with the metal ion, but at the same time ion selectivity is decreased.72 Therefore, to afford both high extraction efficiency and selectivity the substituents on the phosphorus should be chosen very carefully. In the tris-CMPO chelate, the desired basicity of phosphoryl oxygen was achieved via the replacement of the alkyl group presen t in the original CMPO (Figure 2-4) with a phenyl group, thus decreasing the basicity of phosphine oxide through an enhanced inductive effect of the second aromatic ring without the introduction of easily hydrolyzable P-O-C bonds. In addi tion, the phenyl groups adjacent to the P=O provide steric hindrance, an attribute possibly responsible for the higher selectivity for Am(III) over Eu(III) in some CMPO derivatives.72

PAGE 37

21 O P N O 2-1 Figure 2-4. Classic carbamoylme thylphosphine oxide (CMPO). The extraction affinities of our ligand for a selected group of trivalent lanthanides and tetravalent thorium have been compared to multi-CMPO calix[4]arene based extractants. The tris-CMPO ligand system is superior in comparison to other CMPO compounds in the highly selective binding of tetravalent thorium.51 The impact of further structural ligand derivatization on the extraction selectivity and efficiency for tetravalent actinides, with plutonium in part icular is presented herein. The extraction behavior of the tris-CMPO derivatives in comparison to the classic CMPO and other multi-CMPO systems has been discussed, demonstrating an overall improvement in the development of CMPO-based extractants for actinide/lant hanide separations. In addition, metal complexes of tris-CMPO derivatives have b een isolated and their structures were elucidated by NMR, ICR-MS, and X-ray anal ysis, which provide pot ential rationalization for the presented ligands extraction behavior. 2.2 Results and Discussion 2.2.1 Effect of the Structural Modification of Triphenoxymethane Platform on the TrisCMPO Extraction Profile As mentioned above, a 3:1 CMPO to metal complex may form during the extraction of americium from the aci dic media in the TRUEX process,73,74,80 and to facilitate the extraction we envisioned th at a single ligand with three CMPO arms extended out from a molecular platform coul d greatly enhance the binding and extraction of the metal ions from an acidic solu tion. In our previous work with the

PAGE 38

22 triphenoxymethane ligand, this base has been shown to favor a conformation with the three phenolic oxygen atoms orientating themselv es in an “all up” (Figure 1-8, Chapter 1) conformation relative to the central methine hydrogen both in the solid-state and in solution. 50 Tethering three CMPO moieties to th is platform via these phenol oxygens satisfies the requirement for proximate me tal binding with CMPO groups. With the triphenoxymethane platform, the alkyl group on the ortho -position of the phenol moderates the solubility of the platform in organic solvents, as well as, exerts an influence on the extended arms. Large, bul ky groups such as tert-pentyl increase the solubility but also restrict the flexibility of the three arms tethered to the phenolic oxygens. In order to compare the properties of different variants of our ligand system, two new tris-CMPO derivatives were synthesi zed (2-6a and 2-6c). The CMPO moieties were tethered to the altered platform us ing well-established methodology developed for the CMPO-calix[4]arene systems as outlined for the compounds 2-6a, b and c in Figure 2-5.49,75 R2 R1 O H 3 C N R2 R1 O H 3 NH2 R2 R1 O H HN O P O O2N O O P O 2-2a, b, c2-3a, b, c2-4a, b, c 2-6a, b, c 2-5 R2 R1 OH H 3 3 a : R1, R2 = t-Pentyl b : R1, R2 = t-Bu c : R1 = Me, R2 = t-Bu AB C Figure 2-5. Synthesis of tris-CMPO 2-4. (A) NaI, K2CO3, chloroacetonitrile, refluxing acetone, 3 days; (B) LAH, diethyl ether; (C) p -nitrophenyl (diphenylphosphoryl) acetate (2-5), 45-50 oC chloroform.

PAGE 39

23 The results have shown a significant so lubility variation within the studied t -pentyl, t -Bu, and Me derivatives (2-6a, b, and c). All three compounds ar e readily soluble in most common organic solvents such as dich loromethane, THF or methanol, but in less polar solvents, the solubility of the methyl de rivative (2-6c) is significantly limited. Of the three, only t -pentyl derivative, 2-6a, is compatible with 1-octanol. Despite the solubility differences and ster ic variations in the platform s, the modifications at the 2 position of the phenols do not significantly aff ect the affinity of tris-CMPO ligands for Th(IV) as presented in Table 2-1. There does appear, however, to be a small increase in the affinity for the lighter lanthanides as the size of the alkyl gr oups at the 2-position decreases, and 2-6c exhibits a slightly highe r affinity for La(III) in comparison to 2-6a and 2-6b. Table 2-1. Distribution coefficients ( D )a and extraction percentage (%E)b,c for ligands: 2-6a, 2-6b and 2-6c. Aqueous phase: 10-4 metal nitrate in 1 M HNO3, organic phase: 10-3 M ligand in methylene chloride. Ligand 2-6a 2-6b 2-6c Cation D %E D %E D %E Th (IV) >100 100 49.000 98 99.000 99 La (III) 0.031 3 0.042 4 0.111 10 Ce (III) 0.010 1 0.053 5 0.099 9 Nd (III) 0.042 4 0.031 3 0.087 8 Eu (III) 0.020 2 0.031 3 0.042 4 Yb (III) 0.042 4 0.111 10 0.063 6 aD calculated based on the E % values. bE % = 100%([Mn+]org/[Mn+]total) after extraction as determined by Arsenazo(III) assay. cMean value of at least four measurements. The precision: (n-1) = 1 or 2, where (n-1) is a standard deviation from the mean value. 2.2.2 Ligand Flexibilit y vs. Binding Profile Extensive work on “CMPO-like” molecules utilizing calix[4]arene as a platform has found that there is a significant influe nce of ligand flexibi lity on its extraction performance.76 Calix[4]arene extractants showed a strong increase of the extraction

PAGE 40

24 percentage with increased le ngth of the spacer between the amido and phenoxy group. Thus, as proposed by the authors,76 one can anticipate a direct correlation between the size of the cavity and the flexibility of the lig and, and its affinity for a particular cation. The increased flexibility of the molecule should indeed allow fo r better accommodation of metal ions due to the ability of the molecu le to form a more appropriate cavity size. To exercise this postulate, derivatives of the tris-CMPO system with an extended arm length between the platform and the CMPO donors were synthesized. Methodology to isolate a primary amine with three car bon spacer to the phenolic oxygen of the triphenoxymethane was adapted from the prepar ation of calix[4]arene-based extractants (Figure 2-6).76 O O N R R OH H 3 R R O H 3 NH2 R R O H 3 HN R R O H 3 O P O 2-2a, b2-8a, b2-9a, b 2-10a, b a : R = t-Pentyl b : R = t-Bu A B C Figure 2-6. Synthesis of tris-CMPO 210. (A) N-(3-bromopropyl)phthalimide, Cs2CO3, 80-85 oC DMF, 6 days; (B) hydrazine mono hydrate, refluxing ethanol, 24h; (C) p -nitrophenyl(diphenylphosphor yl)acetate (2-5), 45-50 oC chloroform, 3 days. Products 2-10a and 2-10b were obtained via alkylation of the triphenoxymethane platform with N-(3-bromopropyl)phthalimide in the presence of cesium carbonate. The subsequent treatment with hydr azine in ethanol to remove the phthalimide afforded 97%

PAGE 41

25 of primary amine 2-9a and 92% of 2-9b, resp ectively. Final products were obtained in 70% for 2-10a and 49% for 2-10b yields. The results of the extraction experiment re vealed an anticipated increase in the affinity of ligand 2-10a over more rigid 2-6a for the studied meta l ions although, without any expected significant decrease in Th(IV) selectivity (Figure 2-7). 0 20 40 60 80 100 E% Th(IV)La(III)Ce(III)Nd(III)Eu(III)Yb(III) 2-6a 2-10a Figure 2-7. Metal extraction percentages (%E) for the ligands 2-6a, and 2-10a using 10-4 M metal nitrate in 1 M nitric acid and 10-3 M ligand in methylene chloride. 2.2.3 Complexation Studies with Bis-CMPO Compound O O P Ph Ph O O P Ph Ph O O P Ph Ph O N O O Th O N O O Figure 2-8. Diagrams of the 10 coordinate +2 cationic thorium(IV) n itrate complex of 2-6 with two coordinated NO3 counterions.The crystal st ructures of thorium and trivalent metal nitrates of 2-6b51 and 2-6c showed three CMPO arms tightly bound to the metal center, which allowed space on the meta ls for only one [in case of Ln(III)] and two [in case of Th(IV)] nitrate ions, generating dicationic complexes. In view of the fact that 31P NMR and FT-ICR-MS have confirmed the exis tence of these species in solution, the

PAGE 42

26 affinity of the organic phase for these char ged complexes might be limited. Therefore, the bis-CMPO ligand (2-7) was synthesized using procedures similar to Figure 2-5. O HN O P O O NH O P O 2-7 Figure 2-9. Bis-CMPO compound 2-7. With only two CMPO arms available for th e metal, we envisioned that there should be more space in the coordination sphere of the metal for the binding of an additional one or two nitrates counterions. With the re sulting reduction in po sitive charge of the complex, the material would have enhanced solubility in the or ganic phase. The extraction results revealed, however, that the reduction of the number of the binding units to two severely diminished the effectiven ess of 2-7 for the extraction (Figure 2-10). 0 20 40 60 80 100 E% Th(IV)La(III)Ce(III)Nd(III)Eu(III)Yb(III) 2-6a 2 7 Figure 2-10. Metal extraction percentages (% E) for the ligands 2-6a, and 2-7 using 10-4 M metal nitrate in 1 M nitric acid and 10-3 M ligand in methylene chloride.

PAGE 43

27 In fact, the bis-CMPO ligand showed very similar extraction behavior to the simple CMPO extractant (2-1) (Table 2-2). Ligand 27 had very low affinity for Th(IV), as well as, the series of lanthanides. Apparently, three preorganized CMPO arms are essential to fulfill the geometry requirements around th e metal center and afford an appreciable extraction percentage. Table 2-2. Distribution coefficients ( D ) and extraction percentage (%E) for ligands: 2-1, 2-7 and 2-10a. Aqueous phase: 10-4 M metal nitrate in 1 M HNO3, organic phase: 3 x 10-3 M of 2-1a, and 10-3 M of 2-7 and 2-10a in methylene chloride Ligand 2-7 2-1 2-10a Cation D %E D (%E) D %E Th (IV) 0.042 4 0.020 2 >100 100 La (III) 0.111 10 0.053 5 0.190 16 Ce (III) 0.064 6 0.064 6 0.190 16 Nd (III) 0.064 6 0.064 6 0.176 15 Eu (III) 0.064 6 0.075 7 0.163 14 Yb (III) 0.031 3 0.087 8 0.149 13 aThree times higher concentration of classi cal CMPO was used to keep the same concentration of ligating un its in the organic phase 2.2.4 Attempts to Resolve the Tris-CMPO Solubility Issue The choice of solvents repres ents one of the most impor tant factors in the liquidliquid separation science. To achieve e ffective phase separati on, non-polar solvents should be used. Due to safety concerns, the high boiling and flash points, and the low toxicity of the solvent are as equally im portant as the polarit y for waste clean-up operations. Therefore to find applicati on in the nuclear waste decontamination, compatibility of the tris-CMPO ligand system with aliphatic solvents would be highly desirable, and this trait can be achieved eith er by altering the struct ure of the ligand itself or by the addition of a synergis t. The solubility studies of 2-6b confirmed that the addition of TBP as a synergist indeed induces a defined increase in solubility of the extractant. Unfortunately, the effect is only temporary and the tris-CMPO ligand

PAGE 44

28 eventually precipitates from the solution. This phenomenon can most likely be attributed to the formation of both intraand intermolecular hydrogen bonds. In the previously published51 solid-state structure of 2-6b there are two strong hydrogen bonds present between the amide hydrogens and the phosphoryl and carbonyl oxygens on adjacent arms (P=OÂ…Namide: 2.801 , C=OÂ…Namide: 2.798 ), as well as, the intermolecular hydrogen bonds between phosphoryl oxygens on one ligand and the remaining amide hydrogens on an other (Figure 2-11). Figure 2.11. Fragment of the crystal structur e of 2-6b molecules forming hydrogen bond connected network. (Crystals obtained by slow diffusion of pentane into saturated solution of 2-6b in methylene chloride). The crystal structure of ligand 2-10b pres ented in Figure 2-12 significantly differs from the more rigid compound 2-6b. In pl ace of the extended hydrogen bond molecular network, all three arms form only intr amolecular hydrogen bonds between phosphoryl oxygens and adjacent amide hydrogens (P=OÂ…N: 2.809 ) constructing a rigorously C3-

PAGE 45

29 symmetric structure in the solid-state. The aliphatic solvent may additionally force these aromatic molecules to aggregate, further decreasing their solubility resulting in precipitation. Figure 2-12. Diagram of the solid-state struct ure of 2-10b (30% probabi lity ellipsoids for N, O and P atoms; carbon atoms drawn w ith arbitrary radii). For clarity, all hydrogen atoms have been omitted. To prevent formation of the proble matic hydrogen bonds, the amides were alkylated. The synthesis of the tertiary amide derivative of the tris-CMPO ligand was rather challenging. Even though prepar ation of the secondary amines 2-12 via acylation of 2-4a and 2-9b, and LAH reduction of 211 was quite straightforward, the final acylation with p -nitrophenyl (diphenylphosphoryl)-acet ate (2-5) took approximately two weeks to complete in the case of compound 2-14a, and 5 days in the case of 2-14b, presumably due to the steric congestion of the three arms. A modified two-step procedure52 was employed involving acylation of th e secondary amine with chloroacetyl

PAGE 46

30 chloride and subsequent Arbusov reaction81 presented in Figure 2-13. The reaction of chloroacetyl chloride with 2-12a afforded an 85% yield of 2-13a, and the same reaction with 2-12b resulted in a 66% yield of compound 2-13b. The final product of Arbusov reaction between molecule 2-13a and ethyl diphenylphosphinite resulted in a 70% yield of product 2-14a, while the same reaction with 2-13b gave product 2-14b with a 57% yield. R R O H 3 NH2 R R O H 3 NH O O R R O H 3 NH R R O H 3 N O Cl R R O H 3 N O P O n n n n n 2-4a, 2-9b2-11a, b2-12a, b 2-13a, b 2-14a, b a : n = 1, R = t -Pentyl b : n = 2, R = t -Bu ABCD Figure 2-13. Synthesis of tris-CMP O 2-14. (A) ethylchloroformate, K2CO3, dichloromethane, 3 days, rt; (B) LAH, THF, 5 days, rt; (C) chloroacetyl chloride, K2CO3, 24h, refluxing dichloromethane; (D) ethyl diphenylphosphinite, 150 C, 40 h. As shown in Figure 2-14, the polar cavity of the ligand was replaced by hydrophobic interactions between three methyl groups located on the amidic nitrogens which caused all three carbonyl oxygens to point outward. It was anticipated that such a structural alteration would not only enhance th e solubility of the extractant in non-polar solvents, as desired, but perhaps even improve the extractability of Ln(III) and An(III), due to increased basicity of the carbonyl oxygens.

PAGE 47

31 Figure 2-14. Diagram of the soli d-state structure of 2-14a (30% probability ellipsoids for N, O and P atoms; carbon atoms drawn w ith arbitrary radii). For clarity, all hydrogen atoms have been omitted. The alkylation of tris-CMPO has significan t impact on the extr action properties of the ligand. Surprisingly, both compounds 2-14a and 2-14b showed lower selectivity for tetravalent thorium than th eir nonalkylated counterparts (Figure 2-15 and Table 2-3). Table 2-3. Distribution coefficients ( D ) and extraction percentage (%E) for ligands: 2-14a and 2-14b. Aqueous phase: 10-4 M metal nitrate in 1 M HNO3, organic phase: 10-3 M of 2-14a and b in methylene chloride Ligand 2-14a 2-14b Cation D %E D %E Th (IV) 0.923 48 0.639 39 La (III) 0.064 6 0.064 6 Ce (III) 0.052 5 0.075 7 Nd (III) 0.064 6 0.087 8 Eu (III) 0.064 6 0.064 6 Yb (III) 0.075 7 0.075 7 The complexity of the extraction process does not allow for unambiguous explanation of such behavi or, but if the nitrate count erions are bound via hydrogen bond interactions to the amidic hydrogens as it was observed in the solid st ate structures of the

PAGE 48

32 metal complexes, presence of these hydrogens may be crucial for the transfer and stability of the complex in the organic phase. Th(IV) La(III) Ce(III) Nd(III) Eu(III) Yb(III) 2 1 2-14a 2-6a 0 20 40 60 80 100 E% 2 1 2-14a 2-6a Figure 2-15. Metal extraction percentages (%E) for the ligands 2-1, 2-6a, and 2-14a using 10-4 M metal nitrate in 1 M nitric acid and 10-3 M ligand in methylene chloride. In the solid-state structure of the Th(IV) complex with 2-6a, the two nitrate anions are positioned at distances of 2.911 and 2.936 from the amide nitrogens, suggesting a weak hydrogen bonding interaction51 The decrease in extrac tion ability for 2-14a was especially pronounced in the case of Th(IV), for which th e extraction was reduced over 50 percent with respect to the performance of the non-alkylated extractants (Figure 2-15). In the case of the more flexible ligand 2-14b, a noticeable decrease of extraction (with respect to non-alkylated 2-10a) along the entire series of st udied metal ions was observed. As a result, N-alkylated compounds 2-14a (shorter CM PO tripod linker) and 2-14b (longer CMPO tripod linker) display a simila r selectivity profile (Table 2-3). A similar tendency for the decrease in affi nity for the tetravalent ion by alkylated extractant was observed in the case of Pu(IV). Upon amide alkylation, the D value drops significantly from 43.60 to 2.60 at 10-3 M ligand concentration in dichloromethane. However, a ten fold increase in ligand concen tration restores the high distribution value

PAGE 49

33 for Pu(IV) in the organic phase, also observe d with the less rigid compound 2-14b (Table 2-4). In the case of the amide alkylated de rivatives of CMPO bearing calixarenes, an even more severe extraction decrease was reported. 53 Table 2-4. Distribution coefficients ( D ) for the extraction of Pu(IV), U(VI), Am(III) and Eu(III) by ligands: 2-6b, 2-14a and 2-14b in methylene chloride and 1-octanol. Ligand CL Solvent D Pu(IV) D U(VI) D Am(III) D Eu(IIII) 2-6b 10-3 M CH2Cl2 43.60 0.371a 0.412a 2-14ab 10-3 M CH2Cl2 2.60 0.33 2-14bb 10-3 M CH2Cl2 6.74 3.78 2-14b 10-2 M CH2Cl2 33.1 4.85 0.41 0.21 2-14b 10-2 M 1-octanol 6.1 1.54 0.058 0.025 a1M HNO3/5M NaNO3 bPu(IV) 10-5 M, U(IV) 10-4 M in 1 M HNO3 While alkylation of 2-6a and 2-10b resulted in the deteriorated solubility of their derivatives 2-14a and 2-14b in non-polar solvents such as diethyl ethe r, the alkylation of the more flexible compound (2-10b) made 2-14b compatible with 1-octanol. This solubility improvement provided an opportunity to test the extraction potential of the trisCMPO ligand in diluent other than dichloromethane. As it was observed in the case of “CMPO-like” calixarenes77 the binding potential of the tris-CMPO ligand was strongly decrea sed in 1-octanol. At the same molar concentration of ligand 2-14b in both solvents, the affinity for metal ions was much lower in 1-octanol than in dichloromethane (Tab le 2-4 and Figure 2-16). In 1-octanol the hydrogen bonding interactions between the solvent and the phosphoryl and carbonyl oxygens of the extractant may mitigate the extr action either by preventing ligand-metal binding or by simply changing the stability of the complex in the organic phase. Efforts to determine the structure of the extracted species with both alkylated and non-alkylated

PAGE 50

34 ligands in solution, which would allow for the better understanding of the extraction behavior of ligands, are in progress. 2.2.5 Plutonium (IV) and Americium(III) Extractions The light actinides can be separated fr om the lanthanides due to the favored extractability of their higher oxidation states. Since the results of extraction experiments have proven the ability of ligands 2-6a, b a nd c to take advantage of differences in oxidation states of tetravalent thorium and a series of triv alent lanthanides, we were prompted to study the extraction of the Pu(IV) ion, to verify the ability of the tris-CMPO molecule to preferentially extract tetravalent light actinides other than thorium. Pu(IV) U(VI) Am(III) Eu(IIII) 2-14a (1) 2-14b (3) 2-14b (1) 2-14b (2) 2-6b (1) 0 5 10 15 20 25 30 35 40 45 D 2-14a (1) 2-14b (3) 2-14b (1) 2-14b (2) 2-6b (1) Figure 2-16. Metal extraction percentages (% E) for the ligands 2-6b, 2-14a, and 2-14b using 10-3 M ligand in methylene chloride (1), 10-2 M ligand in methylene chloride (2), 10-2 M ligand in 1-octanol (3). The tris-CMPO ligand was found to be significantly more effective for Pu(IV) separation than an industrial mixture of mono-CMPO and TBP in the transuranium elements extraction process (TRUEX).13 In fact, much lower c oncentrations of the trisCMPO ligands were required to achieve a pproximately the same distribution ratio ( D ) of Pu(IV) as the mono-CMPO/TBP extraction sy stem. The tris-CMPO ligands are also

PAGE 51

35 much more selective than the CMPO/TBP mixture. After 24 hours of extraction, the 97.76 % ( D = 43.60) of Pu(IV) was removed from the aqueous phase by ligand 2-6b at concentration as low as 10-3M (Table 2-4) while the D values for all the lanthanides were significantly below 1. To reach a similar di stribution coefficient for the extraction of Pu(IV) a solution of 0.2 M CM PO mixed with 1.0M TBP13,82 would be required and this mixture would extract a si gnificant amount of the tr ivalent lanthanides. In the early fifties, Seaborg and coworkers40 found that trivalent lanthanides and actinides could be separated using cation-excha nge resin columns due to the ability of the actinides to employ 5f orbi tals in bonding. This idea was based on similar spatial extensions of the 5f, 6d, 7s a nd 7p orbitals of th e trivalent ac tinides, especially the lightest ones.42 Since the radial distribution of 4f or bitals are severely limited, this subtle difference could be exploited with the appropr iate ligands to facilitate the separation. 42,83 With these issues in mind, the Am(III)/Eu(III) extraction experiment was performed to test the ability of 2-6b to take advantage of such disc repancy between Am(III) and the similarly size Eu(III) ion (Table 2-4). It was found however, that 2-6b has a very low affinity for both ions ( D below 0.02), and it is not able to distinguish between these two elements. Upon enrichment of the aqueous pha se with sodium nitrate, the distribution coefficients ( D ) for Am and Eu slightly improved, as expected, but neither the D values nor the separation factor were satisfactory. 2.2.6 Comparison of Solid State Structures of Tris-CMPO Complexes of Trivalent Metal Ions The solvent extraction separation process is based on the transfer of a metal cation from an aqueous phase into an immiscible organic phase with simultaneous charge neutralization.42 As we have previously shown in organic solvents,51 the three CMPO

PAGE 52

36 arms in ligand 2-6b tightly wrap around Ln(NO3)3 (Ln = Eu(III), Nd(II I) ) in a bidentate fashion forcing two of nitrate counterions out of the coordination sphere of the metal and producing a complex with an overall 2+ charge. In the solid-state st ructure of the Nd(III) species with 2-6b two distinct Nd(III) comple xes co-crystallized. One metal center is eight coordinate while the other Nd(III) cont ains an extra coordinated water molecule. As expected, all of the metal oxygen bond le ngths are slightly longer in the nine coordinate species. O O P Ph Ph O O P Ph Ph O O P Ph Ph O N O O Nd OH2 O O P Ph Ph O O P Ph Ph O O P Ph Ph O N O O Nd Figure 2-17. Diagrams of neodymium(III) complexes of 2-6b: anhydrous 8 coordinated +2 cationic complex (left) and wate r-containing 9 coordi nated +2 cationic complex (right). Similar, although only 8 coordinate struct ures were obtained for Tb(III) complexes with new derivatives of tris-CMPO chelates 2-6a and 2-14a and Bi(III) complex with 2-6c (Figures 2-18, 2-19, 2-20). Unli ke Tb(III) complexes, the Bi(III) compound contained two similar structures in the asymmetr ic unit. Our interest in the chemistry of bismuth originates from the presence of signi ficant amount of bismuth in waste generated by the bismuth phosphate process popular in early 40Â’s for plutonium and uranium separation.84 In addition to solid stat e structure analysis, the af finity of ligands 2-6b and 2-6c toward trivalent bismuth was tested to ensure selectivity of the tris-CMPO chelate over other than f-element trivalent metal ions present in waste. The extractability of Bi(III) was found to be as low as the trivalen t lanthanides. At the ten fold excess of ligands 2-6b or 2-6c in dichloromethane only 8% of bismuth was transferred to the

PAGE 53

37 organic phase, promising an extended app lication of our extraction system over high bismuth content radioactive waste. Figure 2-18. Diagram of th e structure of [2-6cBiNO3](NO3)2 (30% probability ellipsoids for Bi, N, O and P atoms; carbon atom s drawn with arbitrary radii). All hydrogen atoms have been omitted and the bismuth-ligand bonds have been drawn with solid lines. In the solid state structure the different radii85 of the studied metals (trivalent Nd, Tb, Bi) were strongly reflected in the vari ations in the bond lengths in the resulting complex. The smallest, and therefore the mo st electropositive 8 c oordinate Tb(III) (r = 1.180 ) attracts significantly stronger bi nding atoms than the 8 (r = 1.249 ) and 9 (r = 1.303 ) coordinate Nd(III), and the 8 coor dinate Bi(III) (r = 1.310 ). In all cases the metal phosphoryl oxygen distance is s horter than the metal carbonyl (P=O-M: [2-14aTbNO3](NO3)2: 2.297(7)2.322(6), [2-6aTbNO3](NO3)2: 2.300(4)-2.315(5), Nd CN=8: 2.394(9)-2.452(9), Nd CN=9: 2.429(9)-2 .436(10), Bi(1): 2.315(6)-2.414(6), Bi(2): 2.330(6)-2.394(6); C=O-M: [2-14aTbNO3](NO3)2: 2.326(8)-2.373(7),

PAGE 54

38 [2-8aTbNO3](NO3)2: 2.347(4)-2.397(4), Nd CN=8: 2.358(9)-2.437(9), Nd CN=9: 2.430(10)-2.490(10), Bi(1): 2.444(7)-2.575( 6), Bi(2): 2.464(6)-2.533(6)). The nitrate coordination stre ngth also follows the tren d: Tb > Nd CN=8 > Nd CN=9 > Bi ([2-6aTbNO3](NO3)2: 2.468(4)-2.480(4), [2-14aTbNO3](NO3)2: 2.474(8)2.599(5), Nd CN=8: 2.535(10)-2.552(10), Nd CN=9: 2.615(10)-2.621(10), Bi(1): 2.444(6)-2.519(6), Bi(2): 2.458(5)-2.587(6)). Th e Bi-O separation in this 8 coordinate compound was found to be very similar to the average Bi-O distance in the neutral 8 and 9 coordinate bismuth nitrate comp lexes with tridentate 2,6-bis(-CH2-P(=O)R2) pyridine oxides [CN=8 (2.321 ); CN=9 (2.340 )],86 and somewhat shorter from the distances in the 9 coordinate nitrate complex with (iPrO)2(O)PCH2P(O)(OiPr)2 [2,432(2)-2,544(2) ].87 Figure 2-19. Diagram of the st ructure of compound [2-6aTbNO3](NO3)2 (30% probability ellipsoids for Tb, N, O an d P atoms; carbon atoms drawn with arbitrary radii). All hydrogen atoms have been omitted and the terbium-ligand bonds have been drawn with solid lines.

PAGE 55

39 Figure 2-20. Diagram of the stru cture of compound [2-14aTbNO3](NO3)2 (30% probability ellipsoids for Tb, N, O an d P atoms; carbon atoms drawn with arbitrary radii). All hydrogen atoms have been omitted and the terbium-ligand bonds have been drawn with solid lines. The incorporation of the tertiary amide into the ligand scaffold seems to have very small influence on the terbium coordination e nvironment in the solid state complexes. The distances between p hosphoryl oxygens and terbium ions in [2-6aTbNO3](NO3)2 and [2-14aTbNO3](NO3)2 are within the same range. The carbonyl oxygens and metal bond lengths are only slightly shorter in [2-14aTbNO3](NO3)2 complex, but the nitrate is bound slightly weaker. Selected bond lengths of the molecular structures of 2-10b, 2-14a, [2-6aTbNO3](NO3)2, [2-14aTbNO3](NO3)2 and [2-6cBiNO3](NO3)2 determined by the single crystal X-ray diffraction analysis are summarized in Table 2-5.

PAGE 56

40Table 2-5. Selected bond lengths () for compounds: 2-10b, 2-14a, [2-6aTbNO3](NO3)2, [2-14aTbNO3](NO3)2 and [2-6cBiNO3](NO3)2. 2-10b 2-14a [2-6aTbNO3] (NO3)2 [2-14aTbNO3] (NO3)2 [2-6cBiNO3] (NO3)2 Bi(1) [2-6cBiNO3] (NO3)2 Bi(2) P(1)-O(5) 1.480(3) 1.502(5) 1.505(7) 1.497(6) 1.506(6) P(2)-O(7) 1.485(3) 1.497(5) 1.510(7) 1.502(7) 1.520(7) P(3)-O(9) 1.484(3) 1.503(5) 1.520(7) 1.497(6) 1.498(7) P(1)-O(3) 1.477(3) C(19)-O(2) 1.232(4) C(53)-O(4) 1.230(4) 1.289(12) C(70)-O(6) 1.233(4) 1.260(11) C(87)-O(8) 1.231(4) 1.240(12) C(52)-O(4) 1.261(7) C(68)-O(6) 1.232(8) C(84)-O(8) 1.248(7) M-O(5) 2.315(5) 2.322(6) 2.315(6) 2.352(5) M-O(7) 2.308(5) 2.315(7) 2.334(6) 2.330(6) M-O(9) 2.300(4) 2.297(7) 2.414(6) 2.394(6) M-O(4) 2.347(4) 2.326(8) 2.450(6) 2.464(6) M-O(6) 2.349(5) 2.368(7) 2.444(7) 2.469(7) M-O(8) 2.397(4) 2.373(7) 2.575(6) 2.533(6) M-O(11) 2.468(4) 2.599(9) 2.444(6) 2.458(5) M-O(10) 2.480(4) 2.474(8) 2.519(6) 2.587(6)

PAGE 57

41 2.3 Conclusions A series of molecules containing three precisely arranged carbamoylmethylphosphine oxide (CMPO) moie ties have been synthesized and their ability to selectively extract actinides from simulated acidic nuclear waste streams has been evaluated. The ligand system has shown an excellent binding efficiency for An(IV) and Pu(IV) in particular. As in the case of calix[4]arene and resorcinarene derivatives, the extraction efficiency of classic ( N,N -diisobutylcarbamoylmethyl) octylphenylphosphineoxide (CMPO) extractant was strongly improved by th e attachment of CMPO-like functions on the triphenoxymethane skeleton. The unique geometrical arrangement of the three ligating groups, as opposed to f our, dramatically changed the selectivity profile of this multi-CMPO extractant. To the best of our knowledge, the tris-CMPO is the most effective CMPO-based system for the selec tive recognition of tetr avalent actinides, plutonium in particular. The structural modifications of tris-CMPO have shown a significant decrease in the extraction efficiency and sele ctivity with the introduction of a tertiary amide into the ligand structure. On the other hand, elongation of the secondary amide ligating arm slightly enhanced the extrac tability of all the st udied ions without a major reduction in selectivity. A reduction in the number of chelating groups from three to two produced a very ineffective ligand de monstrating the significance of the presence of exactly three preorganized chelating moieties for the efficient metal binding. This remarkable attraction for tetravalent actinides can be attributed to the match between coordination requirements of An(IV) and the geometrical environment presented by the ligand. Moreover, the higher charge density of tetravalent actinides with respect to trivalent f-elements may additionally account fo r the increased affinity of the tris-CMPO

PAGE 58

42 for An(IV) and the complex stability. As a result, the ligand shows promise as an improved extractant for Pu(IV) recoveries fr om high level liquid wastes, especially of PUREX origin and in general actinide clean -up procedures. Immobilization of this plutonophile on a solid support may offer a very efficient and cost-effective technique for future plutonium separation and recycli ng. Efforts to produce these ligands are underway. 2.4 Experimental Section 2.4.1 General Consideration The lanthanide and actinide salts, La(NO3)3H2O (Alpha Aesar), Ce(NO3)36H2O, Nd(NO3)36H2O, Eu(NO3)35H2O, Tb(NO3)36H2O, Yb(NO3)35H2O (Aldrich), Bi(NO3)35H2O (Acros) and Th(NO3)44H2O (Strem), were used as received. The solutions were prepared from 18 M Millipore deionized wate r, TraceMetal grade HNO3 (Fisher Scientific), and HPLC grade organic solvents. The Arsenazo(III) assay was performed on a Varian Cary 50 UV/vis spectrophotometer while a 2500TR Packard liquid scintillation analyzer was used for c ounting alpha-emitting Pu and U radionuclides. A Canberra GammaTrac 1185 with Ge detector and AccuSpec-B multi-channel analyzer was used for 241Am and 152Eu counting. All 1H, 13C and 31P NMR spectra were recorded on a Varian VXR-300 or Mercury-300 sp ectrometer at 299.95 and 121.42 MHz for the proton and phosphorus channels, respectively. Elemental analyses were performed by Complete Analysis Laboratories, Inc. in Passipany, New Jersey. Mass spectrometry samples were analyzed by Dr. Lidia Matveeva and Dr. Dave Powell at the University of Florida on a Bruker Apex II 4.7T Fourier transform ion cyclotron resonance mass spectrometer. Fast atom bombardment (FAB), ionization energy (IE) and liquid

PAGE 59

43 secondary ion mass spectrometry (LSIMS) mass spectra were recorded on Finnigan MAT95Q Hybrid Sector. 2.4.2 Metal Ions Extractions The lanthanides, bismuth and thorium ex traction experiments followed previously reported procedure. 51,76,88 The extractability of each cation was calculated as %E = 100/(A1 – A)/(A1 – A0), where A is the absorbance of the extracted aqueous phase with the Arsenazo(III) indicator, A1 is the absorbance of the aqueous phase before extraction with the indicator, and A0 is the absorbance of metal-free 1 M nitric acid and the indicator ( Ln(III) = 655, Bi(III), Th(IV) = 665 nm). The errors, ba sed on the precision of the spectrophotometer and the standard devi ation from the mean of at least three measurements, were in most cases no hi gher than two percent. The distribution coefficients ( D ) of cations defined by equation: D = [Morg] / [Maq] (where [Morg], [Maq] are the total concentration of the metal species in the organic and aqueous phase, respectively) were calculated using formula: D = %E / (100 %E). In view of the errors associated with the spectrophotometric tec hnique, the maximum value that could be measured for the extraction percen tage and distribution ratio was 99. In case of 239Pu, 238U, 241Am and 152Eu isotopes, a solution of ligand, 10-3-10-2M in CH2Cl2, or 1-octanol has been contacted with 1 M HNO3 containing radioactive nuclides. Test tubes containing 1-2 mL of the aqueous phase and an equal volume of the organic phase were placed into an orbital shaker at 20 After a certain period of time, a test tube was removed from the rotator, the layers were allowed to separate, and then were transferred into the shell vial s. Equal aliquots of the orga nic and the aqueous phases were taken for counting and distributi on coefficient determination.

PAGE 60

44 Experiments with radioactive elements were performed by Dr. Artem Gelis at Argonne National Laboratory. 2.4.3 Isotopes Stock Solutions Plutonium stock solution The stock so lution of tetravalen t 239Pu was purified using anion exchange method.89 The method is based on the retention of Pu(NO3)6 2in 7.5 M HNO3 by the anion exchanger (Bio-Rad AG-1 x 8). All the cations were eluted from the column with 7.5 M HNO3. Subsequently, Pu was stri pped from the column with a solution of 0.3 M HNO3 at 60 C. The strip was taken to a wet salt with a few drops of concentrated HClO4 to destroy any organic impurities The resulting Pu(VI) salt was dissolved in 1 M HNO3 and was treated with 0.1 mL of concentrated H2O2 to convert Pu to triand tetravalent oxidation states.89 To oxidize Pu(III) to Pu(IV), 20 mg of solid NaNO2 was added to the solution. An electron absorption spectrum, collected in 400-900 nm range, revealed no evidence of either Pu (III) or Pu(VI). The solution was diluted with 1 M HNO3 to get 10-5 M Pu(IV) and was used for the experiments. Uranium stock solution A weighted amount of 238U3O8 of analytical purity was dissolved in 14.7 M HNO3 and then diluted with water to 1.6 M UO2(NO3)2 in 1 M HNO3. The solution was then diluted with 1 M HNO3 to 0.01 M UO2(NO3)2 and was spiked with 233U(VI) (T=1.59x105 y) to increase the effectiveness of the liquid scintillation counting. Americium stock solutio n The purity of the 241Am stock solution was determined by the ICPMS analysis. The solution was evapor ated to dryness and re-dissolved in 1 M HNO3.

PAGE 61

45 Europium stock solution The 152Eustock was purchased from Isotope Products Laboratories. The original 0.5 M HCl solution was taken to a wet salt with nitric acid twice and was then redissolved in 1 M HNO3. 2.4.4 Synthesis The synthetic methodology for the preparati on of 2-6 and 2-10 has been adapted from procedures developed in work with phenols and calix[ n ]arene platforms. 49,75,90,91 Detailed synthetic procedures were previously reported for preparation of molecule 2-4c and 2-6b. 50,51,92 The p -nitrophenyl (diphenylphosphoryl )acetate 2-5 was prepared according to literature directions. 49 Compounds 2-6a, 2-10a b, 2-12b, 2-14b used for extraction experiments and their substrates were synthesized by Dr. Ajay Sah and Dr. Priya Srinivasan. Preparation of Tris(3,5tert -pentyl-2-(cyanomethoxy)phenyl)methane, 2-3a. Following the procedure described in reference51 for 2-2b, a 10.70 g portion (15.00 mmol) of 2-2a was dissolved in dry a cetone (200 mL) with 20.73 g (150.00 mmol) of potassium carbonate, 22.48 g (150.00 mmol) of sodium iodide, and 7.59 g (120.0 mmol) of chloroacetonitrile, and the solution wa s refluxed 48 hours under nitrogen. After the solvent was removed in vacuo the product was taken up in ether, dried with MgSO4, filtered, and the solvent was removed. Recrys tallization of the crude material from ethanol afforded 8.91 g (72%) of product. 1H NMR (CDCl3): = 0.55 (m, 18 H; CH2C H3), 1.13 (s, 18 H; C-C H3), 1.36 (s, 18 H; C-C H3), 1.48 (q, J = 7.4, 6 H; C H2CH3), 1.74 (q, J = 7.5, 6 H; C H2CH3), 4.14 (s, 6 H; C H2CN), 6.17 (s, 1 H; C H ), 7.04 (d, J = 2.6, 3 H; ArH ), 7.15 (d, J = 2.3, 3 H; ArH ). 13C NMR (CDCl3): = 9.4, 9.7, 28.6, 29.4, 35.5, 37.0, 38.1, 38.5, 39.5, (aliphatic); 57.8 (Ar-O-CH2); 115.0 (CN); 126.3, 127.5,

PAGE 62

46 136.0, 141.5, 145.7, 151.8 (aromatic). Anal. Calcd for C55H79N3O3: C, 79.57; H, 9.59; N, 5.06. Found: C, 79.75; H, 10.07; N, 4.96. Preparation of Tris(3-methyl-5tert -butyl-2-(cyanomethoxy)ph enyl)methane, 2-3c. Following the procedure described above, a 0.55 g (1.1 mmol) portion of 2-2c yielded 0.59g (89%) of pure product. 1H NMR (CDCl3): = 1.18 (s, 27 H; Ar-C(C H3)3), 2.37 (s, 9 H; Ar-C H3), 4.14 (s, 6 H; Ar-O-C H2CN), 6.22(s, 1 H; CH ), 6.85 (b, 3 H; ArH ), 7.12 (d, 3 H; ArH ). 13C NMR (CDCl3): = 17.1, 31.2, 34.3, 37.7, 57.2 (aliphatic); 115.5 (CN); 125.5, 127.4, 130.9, 135.0, 147.9, 151.4 (aromatic). Anal. Calcd for C40H49N3O3: C, 77.51; H, 7.97; N, 6.78. Found: C, 77.68; H, 8.36; N, 6.71. 1,1Â’-bis(3,5-ditert -butyl-2-(cynomethoxy)phenyl)ethane. As described for 2-3a, a 6.57 g (14.98 mmol) portion of 2,2Â’-ethylidenebis(4,6-ditert -butylphenol) to afford 4.20 g (54%) of product. 1H NMR (CDCl3): = 1.18 (s, 18 H; CC H3), 1.34 (s, 18 H; CC H3), 1.60 (d, J = 6.9, 3 H; CHC H3), 4.52 (m, 4 H; C H2CN), 4.65 (q, J = 7.1, 1 H; C H CH3), 7.11 (d, J = 2.6, 2 H; ArH ), 7.18 (d, J = 2.6, 2 H; ArH ). 13C NMR (CDCl3): = 23.3, 31.6, 31.7, 32.3, 34.9, 35.7, (aliphatic); 58.9 (Ar-O-CH2), 115.7 (CN); 123.4, 124.2, 138.4, 142.5, 147.7, 152.2, (aromatic). Anal. Calcd for C34H48N2O2: C, 79.02; H, 9.36; N, 5.42. Found: C, 80.39; H, 9.82; N, 5.44. Preparation of Tris(3,5-ditert -pentyl-2-(aminomethoxy)phenyl)methane, 2-4a As outlined in reference,51 a diethyl ether solution of 23a (7.12 g, 8.58 mmol) was added dropwise over 30 min to a slurry of li thium aluminum hydride (4.36 g, 129.00 mmol) in diethyl ether at 0 C. The mixture was allo wed to warm to room temperature and stirred for an additional 12-15 h. A 10 mL portion of 5% NaOH was slowly added to the slurry, and the solution was allowed to stir for 30 minutes. The solution was dried with MgSO4,

PAGE 63

47 filtered, and the solvent was removed in vacuo The crude white solid was recrystallized from acetonitrile to give to give 6.00 g (83%) of product. 1H NMR (CDCl3): = 0.48 (m, 18 H; CH2C H3), 1.11 (s, 18 H; CC H3), 1.30 (s, 18 H; CC H3), 1.43 (q, J = 7.4, 6 H; C H2CH3), 1.67 (q, J = 7.4, 6 H; C H2CH3), 2.89 (t, J = 5.0, 6 H; CH2C H2-NH2), 3.32 (t, J = 5.0, 6 H; O-C H2CH2), 6.38 (s, 1 H; C H ), 7.00 (d, J = 2.6, 3 H; ArH ), 7.12 (d, J = 2.3, 3 H; ArH ). 13C NMR (CDCl3): = 9.3, 9.8, 28.8, 29.7, 35.3, 37.1, 37.9, 38.8, 39.4 (aliphatic); 42.8 (CH2-NH2); 74.4 (O-CH2); 125.0, 128.2, 138.0, 140.1, 142.8, 152.9 (aromatic). Anal. Calcd for C55H91N3O3: C, 78.42; H, 10.89; N, 4.99. Found: C, 78.13; H, 11.35; N, 4.66. Preparation of Tris(3-methyl-5-tert-buty l-2-(aminomethoxy)phenyl)methane, 2-4c. Following the procedure outlined for 2-4a, a 6.46 g (10.0 mmol) portion of 2-3c gave 5.7 g (86%) of product. 1H NMR (CDCl3): = 1.18 (s, 27 H; Ar-C(CH3)3), 2.26 (s, 9 H; Ar-C H3) 2.87 (t, J = 5.1 Hz, 6 H; Ar-O-C H2), 3.31 (t, J = 5.2 Hz, 6 H; Ar-OCH2C H2NH2), 6.76 (s, 1 H; C H ), 6.92 (d, J = 2.6 Hz, 3 H; ArH ), 7.00 (d, J = 2.6 Hz, 3 H; Ar H ). 13C NMR (CDCl3): = 16.8, 31.3, 34.1, 36.9, 42.4, 74.2 (aliphatic); 125.8, 125.9, 129.8, 136.3, 145.5, 152.3 (aromatic); LSI MS [M+H]+ = 632.48 1,1Â’-bis(3,5-ditert -butyl-2-(2-aminoethoxy)phenyl)ethane. Lithium aluminum hydride (1.49 g, 39.26 mmol) was suspended in dry ether (80 mL) and the reaction flask was cooled to 0 oC. 1,1Â’-bis(3,5-ditert -butyl-2-(cynomethoxy)phenyl)ethane (2.70 g, 5.23 mmol) was added in three portions with stirring. The reaction mixture was warmed to room temperature and stirred overni ght. The reaction was monitored by TLC, (pentane:ether 80:20) and once completed, 5% sodium hydroxide solution (4.5 mL) was added dropwise (ice bath) and the mixture wa s stirred until the suspension became milky

PAGE 64

48 white. The resulting white solid was discar ded through filtration a nd the organic layer was dried over MgSO4. Removal of solvent under v acuum yielded analytically pure product. Yield 2.30 g (84%). 1H NMR (CDCl3): = 1.27 (s, 18 H, C H3), 1.42 (s, 18 H, C H3), 1.67 (d, 3 H, J = 7.2 Hz, CHC H3), 3.15 (b t, 4 H, N-C H2CH2-O), 3.89 (m, 4 H, NCH2C H2-O), 4.70 (q, 1 H, J = 7.2 Hz, C H ), 7.21 (s, 2 H, ArH ), 7.24 (s, 2 H, ArH ). 13C NMR (CDCl3): = 23.95, 30.18, 31.57, 31.69, 32.33, 34.70, 35.64, 42.52 (aliphatic); 75.62 (O-CH2); 122.54, 124.69, 139.02, 141.80, 145.18, 153.66 (aromatic). Anal. Calcd for C34H56N2O2: C, 77.81; H, 10.76; N, 5.34. Found: C, 78.32; H, 11.14; N, 5.08 Preparation of compound 2-6a The synthesis of 2-6a followed the preparation method for 2-6b51. A chloroform solution of 2-4a (2.92 g, 3.47 mmol) and pnitrophenyl(diphenylphosphoryl)acetate, 25, (4.16 g, 10.91 mmol) were stirred at 45 C for three days. After cooling to room te mperature, a 1 M solution of NaOH (100 mL) was added and the mixture was stirred for 2 hours. The p-nitrophenol sodium salt was extracted from the chloroform solution usi ng 5% sodium carbonate (6 x 300 mL) and the organic layer was further extracted with brin e. The organic phase was dried with MgSO4, filtered, and the solvent removed in vacuo to give 4.21 g (77%) of product as an off-white solid material. 1H NMR (CDCl3): = 0.46 (m, 18 H; CH2C H3), 1.06 (s, 18 H; CC H3), 1.23 (s, 18 H; CC H3), 1.40 (q, J = 7.3, 6H; C H2CH3), 1.57 (b, 6 H; C H2CH3), 3.41 (b, m, 18 H; O-C H2C H2-NHC(O)-C H2-P), 6.28 (s, 1 H; C H ), 6.92 (d, J = 2.3, 3 H; ArH ), 6.95 (d, J = 2.1, 3 H; ArH ), 7.38 (m, 18 H; P-Ar H ), 7.72 (m, 12 H; P-Ar H ), 7.89 (b, 3 H; N H ). 13C NMR (CDCl3): = 9.3, 9.8, 28.7, 29.7, 35.5, 37.1, 37.8, 38.7, 39.3, 39.5 (aliphatic); 40.4 (CH2-NH2); 70.4 (O-CH2); 125.0, 127.7, 128.7, 128.9, 131.2, 131.3, 132.1, 132.2, 137.8, 139.9, 142.8, 153.2 (aromatic); 165.4; 165.5 (C=O). 31P NMR

PAGE 65

49 (CDCl3): = 29.8. Anal. Calcd for C97H124N3O9P3: C, 74.26; H, 7.97; N, 2.68. Found: C, 74.59; H, 8.14; N, 2.70. Preparation of compound 2-6c Following the procedure described above for 2-6a, a 2.60 g (4.11 mmol) portion of 2-4c was r eacted with 4.90 g (12.85 mmol) of 2-5 to afford 1.97g (35%) of product. 1H NMR (CDCl3): = 1.14 (s, 27 H; Ar-C(CH3)3), 2.12 (s, 9 H; Ar-CH3), 3.29 (b, 12 H; Ar-O-C H2C H2), 3.49 (d, J (H,P) = 13.9 Hz, 6 H; C H2POAr2), 6.65 (s, 1 H; C H ), 6.84 (d, J = 2.4 Hz, 3 H; ArH ), 6.94 (d, J = 2.5 Hz, 3 H; ArH ), 7.38 (m, 12 H; P-Ar H ), 7.47 (m, 6 H; P-Ar H ), 7.76 (m, 12 H; P-Ar H ), 8.00 (b, 3 H; N -H ) 13C NMR [CDCl3]: = 16.79, 31.2, 34.0, 38.7, 39.5, 40.0 (aliphatic); 70.6 (OCH2); 152.1, 145.6, 135.9, 132.7, 131.9, 131.4, 131.0, 130.9, 129.7, 128.6, 128.4,126.0, 125.5(aromatic); 165.1, 165.09 (C=O). 31P NMR (CD3OD): = 30.77. HR ESI-ICR MS (sample injected as solution in 1% HNO3/MeOH): [M+H]+ = 1358.62. Anal. Calcd for C82H94N3O9P3: C, 72.49; H, 6.97; N, 3.09. Found: C, 72.37; H, 6.97; N, 3.38. Preparation of compound 2-7. The synt hesis of 2-7 was adapted from that described above for 2-6a, and a 1.15 g (2.19 mmol) portion of 1,1’-bis(3,5-ditert -butyl2-(2-aminoethoxy)phenyl)ethane was treated with 1.76 g (4.62 mmol) of 7 to afford 2.00 g (90%) of product. 1H NMR (CDCl3): = 1.25 (s, 18 H, C H3), 1.33 (s, 18 H, C H3), 1.43 (d, 3 H, J = 8.6 Hz, CHC H3), 3.42 – 3.71 (several multiplets, 8 H, P-C H2-C(O) + NC H2CH2-O), 3.99 (m, 4 H, N-CH2C H2-O), 4.70 (q, 1 H, J = 6.6 Hz, C H ), 7.15 (b s, 2 H, ArH ), 7.18 (b s, 2 H, ArH ), 7.27 – 7.47 (m, 12 H, P-Ar), 7.78 (b m, 8 H, P-Ar), 8.27 (b, 3 H, N H ). 13C NMR (CDCl3): = 24.07, 31.59, 34.62, 35.51, 38.89, 39.69, 40.26 (aliphatic); 71.94 (O-CH2); 122.46, 124.10, 128.69, 128.85, 130.99, 131.12, 132.23, 132.61, 139.40, 141.65, 141.65, 145.36, 153.22 (aromatic); 165.36 (C=O). 31P NMR

PAGE 66

50 (CDCl3): = 30.41. Anal. Calcd for C62H78N2O6P2: C, 73.78; H, 7.79; N, 2.78. Found: C, 73.85; H, 7.98; N, 2.73. General procedure for synthesis of com pounds 2-8. A stirring suspension of triphenoxymethane molecule (2-2),50 N-(3-bromopropyl)phthalimide and cesium carbonate in DMF was heated to 80-85 oC for six days. The reaction mixture was cooled to room temperature and poured into cold wa ter resulting in forma tion of a white solid product. The mixture was transferred to a separation funnel and extracted with diethyl ether. The solid product suspended in the et her layer and was collected by filtration and dried. Compound 2-8a. Using a 10.01 g (14.04 mmol) portion of tris(3,5-ditert -pentyl-2hydroxyl)methane (2-2a) afforded 11.53 g (64%) of product. 1H NMR (CDCl3): = 0.50 (m, 18 H; CH2C H3), 1.13 (s, 18 H; CC H3), 1.32 (s, 18 H; CC H3), 1.45 (q, J = 7.4, 6 H; C H2CH3), 1.70 (q, J = 7.5, 6 H; C H2CH3), 2.12 (b, 6 H; CH2C H2CH2), 3.50 (b, 6 H; CH2CH2C H2-N), 4.01 (m, 6 H; O-C H2CH2CH2), 6.42 (s, 1 H; C H ), 7.01 (d, J = 2.3, 3 H; ArH ), 7.10 (d, J = 2.1, 3 H; ArH ), 7.45 (m, 6 H; ArH ), 7.54 (m, 6 H; ArH ). 13C NMR (CDCl3): = 9.3, 9.8, 28.8, 29.6, 29.7, 35.1, 36.3, 37.1, 37.9, 39.4 (aliphatic); 69.8 (OCH2); 122.9, 124.9, 128.1, 132.8, 132.9, 133.3, 138.3, 140.1, 142.5, 153.4 (aromatic); 168.3(C=O). FAB MS m/z = 1274.77 [M + H]+. Compound 2-8b. The material wa s obtained in 89% yield. 1H NMR (CDCl3): = 1.19 (s, 27 H; CC H3), 1.34 (s, 27 H; CC H3), 2.19 (m, 6 H; CH2C H2CH2), 3.58 (t, J = 5.5, 6 H; CH2C H2-N), 4.03 (m, 6 H; O-C H2CH2), 6.48 (s, 1 H; C H ), 7.15 (d, J = 2.3, 2 H; ArH ), 7.23 (m, 2 H; ArH ), 7.47 (m, 6 H; ArH ), 7.54 (m, 6 H; ArH ). 13C NMR (CDCl3): = 29.6, 31.5, 31.7, 34.7, 35.7, 36.3 (ali phatic); 70.3 (O-CH2);1 22.4 122.9, 127.3, 132.8,

PAGE 67

51 133.3, 138.0, 142.2, 144.6, 153.7 (aromatic); 168.3 (C =O). FAB MS m/z = 1212.66 [M + Na]+ General procedure for synthesis of compounds 2-9a and 2-9b. To a suspension of compound 2-8 in absolute ethanol, hydrazine mono hydrate (4 eq) was slowly added and the mixture was refluxed for 24 h. The reactio n was cooled to room temperature and the solvent was partially evaporated under redu ced pressure. The resulting residue was poured into ice-cold water and a white precipitate quickly formed. The product was collected by filtration. Compound 2-9a. Yield 97%. 1H NMR (CDCl3): = 0.40-0.49 (m, 18 H, CH2C H3), 1.04 (s, 18 H, C H3), 1.25 (s, 18 H, C H3), 1.37 (q, J = 7.5 Hz, 6 H, C H2CH3), 1.62 (q, J = 7.5 Hz, 6 H, C H2CH3), 1.77 (quintet, J = 6.6 Hz, 6.9 Hz, 6 H, NCH2C H2CH2-O), 2.58 (b s, 6 H, N H2), 2.80 (t, J = 6.9 Hz, 6 H, N-C H2CH2CH2-O), 3.38 (b t, 6 H, N-CH2CH2C H2-O), 6.23 (s, 1 H, C H ), 6.95 (s, 6 H, ArH ). 13C NMR (CDCl3): = 39.5, 39.4, 39.2, 37.7, 36.9, 35.0, 34.2, 29.5, 28.7, 9.7, 9.2 (aliphatic); 70.4 (O-CH2); 153.5, 147.4, 142.3, 139.7, 138.2, 128.0, 124.7 (aromatic). FAB MS m/z = 884.76 [M + H]+. Compound 2-9b. Yield 92%. 1H NMR (CDCl3): 1.12 (s, 27 H, C H3), 1.25 (s, 27 H, C H3), 1.83 (quintet, J = 7.5 Hz, 6.6 Hz, 6 H, N-CH2C H2CH2-O), 2.87 (t, J = 7.5 Hz, 6 H, N-C H2CH2CH2-O), 3.38 (t, J = 6.6 Hz, 6 H, N-CH2CH2C H2-O), 6.25 (s, 1 H, C H ), 7.10-7.16 (m, 6 H, ArH ). 13C NMR (CDCl3): = 31.32, 31.45, 33.17, 34.51, 35.46, 38.50, 38.98 (aliphatic); 70.50 (O-CH2); 122.47, 126.0, 127.38, 129.83, 131.79, 137.61, 141.79, 144.68, 153.35 (aromatic). FAB MS m/z = 800.67 [M + H]+.

PAGE 68

52 General procedure for synthesis of co mpounds 2-10a and 2-10b. A chloroform solution of 0.75 g of amine (2-9) and 3.1 eq. of p -nitrophenyl(diphenylphosphoryl)acetate (2-5) was stirred at 45-50 oC for 3 days. The reaction mixture was cooled to room temperature; 1 M sodium hydroxide solution wa s added and stirred for 2 h. The organic phase was extracted with 5% sodium carbona te followed by brine, and the solution was dried over MgSO4. The solvent was removed in vacuo and acetonitrile (15 mL) was added resulting in product precipitation. The solid was filtered, washed with acetonitrile, and dried to afford pure product. Compound 2-10a. Yield 70%. 1H NMR (CDCl3): 0.36-0.42 (m, 18 H, CH2C H3), 1.04 (s, 18 H, C H3), 1.15 (s, 18 H, C H3), 1.36 (q, J = 7.5 Hz, 12 H, C H2CH3), 1.50 (q, J = 7.5 Hz, 6 H, C H2CH3), 1.80 (b s, 6 H, N-CH2C H2CH2-O), 3.23 3.40 (several multiplets, 18 H, P(O) -C H2-C(O)N-C H2CH2C H2-O), 6.15 (s, 1 H, C H ), 6.88 (d, J = 1.8 Hz, 3 H, ArH ), 7.00 (d, J = 1.8 Hz, 3 H, ArH ), 7.28-7.48 (m, 18 H, P-Ar H ), 7.64-7.84 (m, 12 H, P-Ar H ). 13C NMR (CDCl3): = 9.2, 9.7, 28.7, 29.6, 30.5, 35.1, 37.0, 37.7, 38.1, 39.2, (aliphatic); 69.7 (O-CH2); 124.7, 127.9, 128.7, 128.9, 131.1, 131.6, 132.2, 133.0, 137.8, 139.7, 142.3, 147.4, 153.3 (aromatic); 165.22, 165.16 (C=O). 31P NMR: = 30.3. ESI FT -ICR MS m/z = 1633.88 [M + Na]+. Compound 2-10b. Yield 49% 1H NMR (CDCl3): = 1.16 (s, 27 H; CC H3), 1.21 (s, 27 H; CC H3), 1.91 (b, 6 H; CH2C H2CH2), 3.39 (several multiplets, 18 H; O-C H2CH2C H2-NH-C(O)C H2P(O)), 6.28 (s, 1 H; C H ), 7.08 (d, J = 2.3, 3 H; ArH ), 7.23 (d, J = 2.6, 3 H; ArH ), 7.40 (m, 18 H; P-Ar H ), 7.77 (m, 12 H; P-Ar H ), 7.87 (t, J = 5.4, 3 H; N H ). 13C NMR (CDCl3): = 30.6 31.6, 31.7, 34.6, 35.6, 38.1, 38.9, 39.7 (aliphatic); 70.3 (O-CH2); 122.3, 127.1, 128.7, 128.9, 131.1,

PAGE 69

53 131.2, 131.7, 132.17, 132.21, 133.1, 137.6, 141.9, 144.4, 153.6 (aromatic); 165.2, 165.1(C=O). 31P NMR (CDCl3): = 29.8. ESI FT -ICR MS m/z = 1549.79 [M + Na]+. Compound 2-11a. To a mixture of 2-4a (2.16 g, 2.56 mmol) and KOH (4.32 g) in 40 mL of dichloromethane, ethylchloroform ate (1.6 mL, 1.82 g, 17 mmol) was added. The mixture was stirred for 3 days at room temperature. The solution was then washed with 200 mL of water and brine (50 mL), and dried over MgSO4. The solvent was evaporated to afford 2.5 g of product in 92% yield. 1H NMR (CD3OD): = 0.56 (t, 9 H, CH2C H3), 0.59 (t, J = 7.5, 9 H, CH2C H3), 1.18 (s, 18 H, CC H3), 1.26 (t, J = 7.5, 9 H, OCH2C H3), 1.37 (s, 18 H, CC H3), 1.54 (q, J = 7.5, 6 H, C H2CH3), 1.76 (q, J = 7.5, 6 H, C H2CH3), 3.39 – 3.65 (b m, 6 H, O-CH2C H2-N + 6 H, O-C H2CH2-N), 4.12 (q, J = 7.2, 6 H, O-C H2CH3), ), 6.42 (s, 1 H, C H ), 7.15 (b, 6 H, ArH ). 13C NMR (CD3OD): = 9.8, 10.2, 15.2, 29.3, 30.3, 36.3, 38.0, 38.9, 40.4, 42.4 (aliphatic); 62.0 (CH2-O); 72.0 (CH2OAr); 126.4, 129.1, 139.3, 141.4, 144.2, 154.4 (aromatic); 159.1 (C=O). LSI MS: m/z = 1058.77 [M + H]. Anal. Calcd. for C64H103N3O9: C, 72.62; H, 9.81; N, 3.97; Found: C, 72.86; H, 10.30; N, 3.90. Compound 2-11b. To a mixture of amin e 2-9b (5 g, 6.3 mmol) and KOH (11.4 g) in 20 mL of dichloromethane, ethylchloroformate (2.8 mL 3.2 g, 29 mmol) was added. The mixture was stirred for 3 days at room temperature. Subsequently solution was washed with 250 mL of water a nd brine (50 mL), dried over MgSO4 and evaporated. Yield: 75% (4.75 g). 1H NMR (CDCl3): = 1.18 (s, 27 H, C H3), 1.23 (t, J = 6.90 Hz, 9 H, C H3), 1.33 (s, 27 H, C H3,), 1.98 (m, 6 H, N-CH2C H2CH2-O), 3.39 (m, 6 H, NC H2CH2CH2-O), 3.52 (m, 6 H, N-CH2CH2C H2-O), 4.11(q, J = 6.20 Hz, 6 H, OC H2CH3), 5.39 (b s, 3 H, N H ), 6.35 (s, 1 H, C H ), 7.13-7.26 (m, 6 H, ArH ). 13C NMR

PAGE 70

54 (CDCl3): = 14.9, 30.9, 31.6, 34.6, 35.7, 39.1, 60.7, 71.0 (aliphatic); 122.6, 127.3, 144.7, 137.8, 141.9, 153.6 (aromatic); 157.1(C=O). FAB MS: m/z = 1016.73 [M + H]+. Anal. Calcd. for C61H97N3O9: C, 72.08; H, 9.62; N, 4.13; Found: C, 72.07; H, 9.88; N, 4.06. Compound 2-12a. To a stirred soluti on of lithium aluminium hydride (2.53 g, 0.067 mol) in tetrahydrofuran (500 mL) at 0C, 2.4 g (2.27mmol) of ester 2-11a was added dropwise, and the reaction mixture was st irred at room temperature for 5 days. In order to quench LAH, the solution was cooled to 0C, treated with 3 mL of water and stirred for 5 minutes. A tota l of 3mL of 15% NaOH was then added dropwise, and after additional 30 minutes, more water (9 mL) was added (SteinhardÂ’s method).93 The solid was separated, and the organi c phase was dried over MgSO4. The solution was evaporated to give crude product that was further purified by preci pitation in acidified pentane, dissolution in diet hyl ether and extrac tion with 1M NaOH. The organic phase was dried with MgSO4 and evaporated to yield 2 g of product (95%) of pure product. 1H NMR (CDCl3): = 0.55 (t, J = 7.2 Hz, 9 H, CH2C H3), 0.57 (t, J = 7.2 Hz, 9 H, CH2C H3), 1.13 (s, 18 H, C H3,), 1.35 (s, 18 H, C H3,), 1.47 (q, J = 7.5 Hz, 6 H, C H2CH3), 1.73 (q, J = 7.5 Hz, 6 H, C H2CH3), 2.47 (s, 9 H, NC H3), 2.43 (b t, 6 H, N-C H2CH2-O), 3.65 (t, J = 5.7 Hz, 6 H, N-CH2C H2-O) 6.36 (s, 1 H, C H ), 6.99 (b, 3 H, ArH ), 7.05 (b, 3 H, ArH ). 13C NMR (CDCl3): = 9.3, 9.7, 28.8, 29.7, 35.35, 36.9, 37.1, 37.8, 39.3 (aliphatic); 52.2 (N-CH2); 71.7 (O-CH2); 124.9, 127.9, 138.0, 139.8, 142.6, 153.4 (aromatic). LSI MS: m/z = 884.76 [M + H]+. Compound 2-12b. To a stirring solution of ester 2-11b (7.0 g, 0.0069 mol) in tetrahydrofuran (500 mL) at ice cold c ondition, lithium aluminium hydride (2.8 g, 0.074 moles) was slowly added. The reaction mixture was stirred at room temperature

PAGE 71

55 for 6 days. The reaction mixture was cooled to ice cold temperature and 1M NaOH (50 mL) was added and the stirring was continue d for 15 minutes. Then water (100 mL) was added and the content was transferred to se parating funnel and extracted with diethyl ether (4 x 50 mL). The organic phase was washed with brine (5 x 20 mL), dried over MgSO4 and evaporated to give 5.04 g (87%) of pure product. 1H NMR (CDCl3): = 1.18 (s, C H3, 27 H), 1.34 (s, C H3, 27 H), 1.95 (t, J = 7.35 Hz, 6 H, N-CH2C H2CH2-O), 2.43 (s, 9 H, C H3,), 2.73 (t, J = 7.35 Hz, 6 H, N-C H2CH2CH2-O), 3.54 (t, J = 6.30 Hz, 6 H, N-CH2CH2C H2-O), 6.36 (s, 1 H, C H ), 7.12-7.26 (m, 6 H, ArH ). 13C NMR (CDCl3): = 30.8 31.6, 31.7, 34.6, 35.6, 36.6, 38.8 (aliphatic); 49.3 (N-CH3); 71.0 (O-CH2); 122.6, 127.3, 138.0, 141.8, 144.3, 147.4, 153.9 (aromatic). FAB MS: m/z = 842.71 [M + H]+. Compound 2-13a. To a solution of the s econdary amine 2-12a (2.42 g, 2.74 mmol) and K2CO3 (4.50 g, 32.56 mmol) in CH2Cl2 was added chloroacetyl chloride (1.40 mL, 17.60 mmol), and the reaction mixture was refluxed for 18 h. A second portion of chloroacetyl chloride (0.70 mL, 8.80mmol) was added and refluxed for an additional 20 h. Subsequently, the solution was cool ed down, and washed with 2 N NaOH, H2O and dried over MgSO4. The solvent was remover in vacuo and the crude white solid was recrystallized from dichlorome thane/hexamethyl-disiloxane to give 2.70 (88%) g of pure product. 1H NMR (CDCl3) as well as 13C NMR (CDCl3) spectra are very complicated. 1H NMR (CDCl3): = 0.51 – 0.60 (m, 18 H, CH2C H3), 1.12 (b, 18 H, C H3,), 1.30 (b, 18 H, C H3,), 1.44 – 1.73 (b m, 12 H, C H2CH3), 2.88 – 4.22 (several multiplets, 9 H N-C H3 + 6 H N-C H2CH2-O + 6 H, N-CH2C H2-O), 4.06 (s, 6 H, C H2-Cl), 6.33, 6.38, 6.43 (s, 1 H, C H ), 6.82 – 7.11 (m, 6 H, ArH ). LSI MS: m/z = 1112.67 [M + H]+.

PAGE 72

56 Compound 2-13b. To a solution of the secondary amine 2-12b (3.00 g, 3.60 mmol) and K2CO3 (6.00 g, 43.40 mmol) in CH2Cl2 (20 mL) was added chloroacetyl chloride (2.08 mL, 26.15 mmol), and the reac tion mixture was heated at 45 C for 12 h. A second portion of chloroacetyl chloride (1.04 mL 13.07 mmol) was added and stirred for an additional 20 h at 45 C. Subsequently, the solution was cooled down, and washed with 2 N NaOH, H2O and dried over MgSO4. The solvent was remover in vacuo to give 2.50 g (66%) of pure product. 1H NMR (CDCl3): = 1.18 (s, 27 H, C H3,), 1.36 (s, 27 H, C H3,), 1.84 – 1.95 (m, 6 H O-CH2C H2CH2-N), 2.86 – 3.04 (m, 9 H, N-C H3), 3.37 3.65 (b m, 6 H, O-CH2CH2C H2-N + 6 H, C H2-Cl), 4.04 – 4.10 (m, 6 H, O-C H2CH2CH2-N), 6.33 (s, 1 H, C H ), 7.20 – 7.11 (m, 6 H, ArH ). 13C NMR (CDCl3): = 27.9, 29.3, 31.6, 33.7, 34.6, 35.7, 39.1, 41.1, 41.8, 46.347.9 (aliphatic); 70.4 (OCH2); 122.6, 127.5, 137.9, 141.8, 144.9, 153.5 (aromatic) 166.2 (C=O). EI MS m/z = 1071.61 [M + H]+. Anal. Calcd. for C61H94Cl3N3O6: C, 68.36; H, 8.84; N, 3.92; Found: C, 68.58; H, 8.31; N, 3.71. Compound 2-14a. Method A. The 2.50 g, 2.24 mmol of starting material (2-13a) was dissolved in 9.00mL of ethyl diphe nylphosphinite (9.59 g, 41.65 mmol) while the temperature was gradually increased from 100 to 150 C, and the mixture was stirred for 40 h. Subsequently, the reaction mixture was cooled down to rt, and the diisopropyl ether was added till a white precipitate was formed. The solid was filtered and washed with diisopropyl ether to afford 3.08 g (85%) of pure product 1H NMR (CDCl3) as well as 13C NMR (CDCl3) spectra are very complicated. 1H NMR (CDCl3): = 0.44 – 0.53 (m, 18 H, CH2C H3), 1.04 (s, 9 H, C H3,), 1.09 (s, 9 H, C H3,), [these two singlets merge into one = 1.8 at 55 C], 1.21 (s, 9 H, C H3,), 1.27 (s, 9 H, C H3,), [these two singlets merge into one = 1.25 at 55 C], 1.42 (b m, 6 H, C H3,), 1.62 (b m, 6 H, C H3,), 2.66 –

PAGE 73

57 3.69 (several multiplets, 9 H, N-C H3 + 6 H N-C H2CH2-O + 6 H, N-CH2C H2-O), 6.21, 6.26, 6.32 (s, 1 H, C H ), 6.83 – 7.02 (m, 6 H, ArH ), 7.44 – 7.53 (m, 18 H, P-Ar H ), 7.84 – 7.90 (m, 18 H, P-Ar H ). ESI FT -ICR MS m/z = 827.94 [M + 2Na]2+, m/z = 1632.89 [M + Na]+. Anal. Calcd. for C100H130N3O9P3: C, 74.55; H, 8.13; N, 2.61; Found: C, 74.34; H, 8.44; N, 2.61. Slow diffusion of pentane into solution of 2-14a in diethyl ether/dichloromethane afforded crystals suitable for X-ray analysis. Method B. A solution of secondary am ine (0.47 g, 0.53 mmol) and p-nitrophenyl (diphenylphosphoryl)acetate (1.10 g, 2.88 mmol) in dichloromethane, was stirred for 2 weeks at room temperature. Subsequently solution was treated with 1 M NaOH and stirred for additional 2 hours. The p-nitrophenol salt was extracted from organic phase using 5% sodium carbonate. Or ganic phase was dried over MgSO4, filtered and the solvent removed in vacuo The crude product was criticized by diffusion of pentane into solution of product in diethyl ether/di chloromethane; yield 0.60 g (70%). Compound 2-14b. Method A. The 0.50 g, 0.47 mmol of starting material (2-13b) was suspended in 1.00mL of ethyl diphenylphosphinite (4.20 mmol) while the temperature was gradually increased from 100 to 150 C. Within first 3 h of reaction, every 20 minutes the mixture was exposed for few seconds to the vacuum. The reaction mixture was stirred at 150 C for additiona l 20 h. Subsequently it was cooled down to room temperature and the diethyl ether was added till a white precipitate was formed. The solid was filtered and redissolved in diisopropyl ether to give pure product upon crystallization. Yield 0.42 g (57%). 1H NMR (CDCl3) as well as 13C NMR (CDCl3) spectra are very complicated. 1H NMR (CDCl3): = 1.15 (s, 27 H, C H3,), 1.27 (s, 27 H, C H3,), 1.78 – 1.95 (m, 6 H, N-CH2C H2CH2-O), 2.66 – 3.55 (several multiplets, 27 H: 9 H

PAGE 74

58 N-C H3 + 6 H N-C H2CH2CH2-O + 6 H, N-CH2CH2C H2-O), 6.24 (b s, 1 H, C H ), 6.93 (b s, 2 H, ArH ), 7.06 (b s, 2 H, ArH ), 7.38 (m, 18 H, P-Ar H ), 7.74 (m, 12 H, P-Ar H ). 13C NMR (CDCl3) = 27.9, 29.4, 31.5, 33.9, 34.5, 35.5, 36.9, 37.7, 38.6, 38.9, 46.0, 48.3 (aliphatic); 70.6 (O-CH2); 122.3, 127.2, 128.6, 128.7, 131.2, 131.3, 132.1, 133.3, 137.7, 141.7, 144.5, 144.7, 153.5 (aromatic); 164.8 (C=O). 31P NMR (CDCl3) = 29.4, 29.3. EI MS m/z = 1567.86 [M + H]+. Method B. A solution of secondary amine (2.00 g, 2.40 mmol), p-nitrophenyl (diphenylphosphoryl)a cetate (4.50 g, 11.80 mm ol) and 1 mL of Et3N in chloroform, was stirred for 5 days at 45 – 50 C. After cooling down to the rt solution was treated with 1 M NaOH and stirred for additional 2 hours. The pnitrophenol salt was extracted from organic phase using 5% sodium carbonate. Organic phase was dried over MgSO4, filtered and the solvent removed in vacuo The crude product was washed with diethy l ether and dried yielding 2.50 g (67%) of clean product. Tb-complex of 2-6a [2-6aTbNO3](NO3)2. To a solution of 2-6a (0.200 g, 0.127 mmol) in acetonitrile (8 mL), Tb(NO3)3.6H2O (0.058 g, 0.128 mmol) in methylene chloride (4.5 mL) was added and reaction mixture was stirred overnight at room temperature resulting in a wh ite solid. The complex was isolated by filtration, washed with acetonitrile and dried. Yield 0.180 g (74%). Anal. Calcd for C97H124N6O18P3Tb: C, 60.87; H, 6.53; N, 4.39. Found: C, 60.91; H, 6.66; N, 4.27. Slow diffusion of ether into a concentrated solution of the complex in methanol afforded crystals suitable for structural analysis. Tb-complex of 2-14a [2-14aTbNO3](NO3)2. A solution of Tb(NO3)3.6H2O (0.028 g, 0.062 mmol) in 1 mL of methanol was added to a solution of 2-14a (0.100 g, 0.062 mmol) in methanol (1mL), and reaction mi xture was stirred for one hour at room

PAGE 75

59 temperature. A white precipitate formed within minutes, and the product was collected by filtration, washed with cold methanol, and dried to obtain 0.080 g (62%) of product. ESI FT-ICR MS m/z = 915.40 [2-14aTbNO3]2+. Slow diffusion of ether into a concentrated solution of the complex in mixture of methanol and dichloromethane afforded crystals suitable for structural analysis. Bi-complex of 2-6c [2-6cBiNO3](NO3)2 A solution of Bi(NO3)35H2O (0.058 g, 0.12 mmol) in 8 mL of 1:1 mixture of acetonitr ile and methanol was added to a solution of 2-6c (0.08g, 0.06mmol) in 2 mL of methanol and the mixture was stirred for 1 hour. Part of solvent was evaporated in-vacuo and the product was preci pitated out by ether diffusion. Yield 0.080 g (52%). Slow diffusion of ether into a saturated acetonitrile/methanol solution of [2-6cBi(NO3)](NO3)2 afforded crystals suitable for structural analysis. 1H NMR (CD3CD), 55C: = 1.15 (s, 27 H; Ar-C(C H3)3), 2.14 (s, 9 H; Ar-C H3), 3.17 (b, 12H; 3.96 Ar-O-C H2C H2), 6.74 (s, 1 H; CH ), 7.00 (d, J = 2.05 Hz, 3 H; Ar H ), 7.10 (d, J = 2.56 Hz, 3 H; Ar H ), 7.49 (m, 12 H; P-Ar H ), 7.62 (m, 6 H; P-Ar H ), 7.74 (m, 12 H; P-Ar H ), 7.86 (s, 3 H; NH ). 31P NMR (CDCl3): = 39.66. HR ESI-ICR MS (sample injected as solution in 1% HNO3/MeOH): m/z = 814.29 [ 26cBi(NO3)]2+ and m/z =1691.56 [2-6cBi(NO3)2]+. Anal. Cald for {[2-6cBi(NO3)MeOH2H2O][NO3]2} C83H102BiN6O21P3: C 54.73; H 5.64; N 4.61; Found: C, 54.54; H, 5.32; N, 4.92. 2.4.5 X-Ray Crystallography Unit cell dimensions and intensity data for all the structures were obtained on a Siemens CCD SM ART diffractometer at 173 K. The data collections nominally covered over a hemisphere of reciprocal space, by a combination of three sets of exposures; each set had a different angle for the crystal and each exposure

PAGE 76

60 covered 0.3 in The crystal to detector distance was 5.0 cm. The data sets were corrected empirically for absorption using SADABS. All the structures were solved using the Bruker SHELXTL software package for the PC, using either the direct methods or Patterson functions in SHELXS. The space groups of the compounds were determined fr om an examination of the systematic absences in the data, and the successful solution and refinement of the structures confirmed these assignments. All hydrogen at oms were assigned idealized locations and were given a thermal parameter equivalent to 1.2 or 1.5 times the thermal parameter of the atom to which it was attached. For th e methyl groups, where the location of the hydrogen atoms is uncertain, the AFIX 137 card was used to allow the hydrogen atoms to rotate to the maximum area of residual densit y, while fixing their geometry. In cases of extreme disorder or other problems, the non-hydrogen atoms were refined only isotropically, and hydrogen atoms were not incl uded in the model. Severely disordered solvents were removed from the data for 2-10a, 2-14a, [2-6aTbNO3](NO3)2, [26cBiNO3](NO3)2 and [2-14aTbNO3](NO3)2 using the SQUEEZE function in the Platon for Windows software and the details are repor ted in the supporting information in the CIF file for each structure. Structural and refinement data and selected bond lengths for all the compounds are presented in the Tables 2-5 and 2-6.

PAGE 77

61Table 2-6. X-ray dataa for the crystal structures of 210b, 2-14a and the complexes [2-6aTbNO3](NO3)2, [2-6cBiNO3](NO3)2 and [2-14aTbNO3](NO3)2 2-10bCH3OH 2-14a3CHCl3Et2O [2-6aTbNO3] (NO3)2 Et2O [2-14aTbNO3] (NO3)2 [2-6cBiNO3](NO3)2 total reflections unique reflections max( ) empirical formula Mr crystal system space group a () b () c () ( ) ( ) ( ) Vc (3) Dc (g cm-3) T (K) Z (MoK ) (mm-1) R1 [ I 2 ( I ) data]b w R2 (all data)c GoF 24103 5195 24.99 C96H119N3O10P31567.85 hexagonal R3 18.674(2) 18.674(2) 43.801(3) 90 90 120 13228(2) 1.181 173(2) 6 0.127 0.0816 [3938] 0.1980 1.140 32269 20440 25.00 C105H138Cl9N3O9.5P3 2006.14 triclinic P -1 17.250(2) 18.070(2) 21.557(3) 87.288(2) 66.965(2) 71.870(2) 5855.5(13) 1.138 173(2) 2 0.307 0.0826 [13367] 0.2673 1.115 31086 19674 25.00 C101H134N6O19P3Tb 1987.97 triclinic P -1 12.5597(9) 15.5209(12) 28.949(2) 93.700(2) 90.632(2) 90.532(2) 5630.9(7) 1.172 173(2) 2 0.732 0.0687 [11833] 0.1765 0.986 23270 14249 23.00 C100H124N6O21P3Tb 1997.88 triclinic P -1 12.4105(18) 13.713(2) 33.122(5) 88.353(3) 82.941(2) 66.989(2) 5148.0(13) 1.289 173(2) 2 0.803 0.0971 [11989] 0.2097 1.280 51156 30858 24.50 C84H102BiN6O20P3 1817.61 triclinic P -1 20.8891(13) 22.9583(15) 23.1141(15) 80.3080(10) 64.9140(10) 68.3520(10) 9330.4(10) 1.294 193(2) 4 2.008 0.0657 [15711] 0.1780 0.899 aObtained with monochromatic Mo K radiation ( = 0.71073 ) bR1 = Fo Fc / Fo cw R2 = { [ w ( Fo 2 Fc 2)2/ [ w ( Fo 2)2]}1/2

PAGE 78

62 CHAPTER 3 DESIGN, SYNTHESIS AND EVALUATION OF PHOSPHINE SULFIDE BASED CHELATES FOR THE SEPARATION OF TRIVALENT LANTHANIDES AND ACTINIDES 3.1 Introduction Recent challenges in the field of the part itioning and transmutation of highly acidic nuclear waste involve the se paration of minor actinides (neptunium, americium, and curium) and long lived fission products. Over th e years, several partition processes have been proposed,94 but current protocols lack the uni que ability to effectively separate trivalent actinides from trival ent lanthanides. This funda mental problem in separation science is due to the similarity in the ionic st ructure and radius of the trivalent f-elements (especially Am, Eu and Nd). P SH S P O O HS S HDEHDTP HBTMPDTP (Cyanex 301) Figure 3-1. Structures of su lfur based extractants. During the last couple of years, only several extraction systems employing soft donor ligands such as sulfur and nitrogen have shown some selectivity for Am(III) over Ln(III).95-98 For instance, a synergistic mixture of di-2-ethylhexyl dithiophosphoric acid (HDEHDTP) (Figure 3-1) and tributylphosphate (TBP) has an observed separation factor (SF = DAm/ DEu) of 60 for the partiti on of Am(III) over Eu(III).95

PAGE 79

63 Work done by Zhu et al have demonstrated that purified Cyanex 301 [bis(2,4,4trimethylpentyl)dithiophosphinic acid, HBTMPDTP] was able to separate Am(III) from Eu(III) even more efficiently then the HDE HDTP/TBP mixture, with the separation factor of 5900.27,99 Interestingly, severa l years earlier, these soft donor compounds have not been considered as potential extractants for actinides and lanthanides separation due to their hypothetical incompatibil ity with hard metal ions. At the pH of 3 in ni tric acid and 1M NaNO3/kerosene extraction environment the extraction enthalpy of Am(III) with purif ied 0.5 M Cyanex 301 was found to be significantly less endo thermic than that of Eu(III) (18.10 kJ/mol and 43.65 kJ/mol respectively).27 The stronger affinity for trivalent americium was attributed to the higher degree of covalency of the Am(III)-S bond comp ared to Eu(III)-S. Using theoretical calculations Madic and coworkers have quantifie d the covalent effect and cited that the covalent contribution for the Am-S -bond energy is higher by approximately 7.6 kJ/mol than for Eu-S.48 Even though these acidic organophosphorous reagents are successful in the differentiation between Am(III) and lanthani des, conditions required to maintain selectivity during extraction have proven to complicate the partition process. The extraction equilibrium strongly depends on the acidic nature of the extractant,99 and the low acidity of Cyanex 301 (pKa = 2.6) limits the effectiveness of the compound to the extraction systems with pH 3 or higher. Sin ce the acidity of waste solutions is generally in the range of 1 to 3 M HNO3, the extraction with Cyanex 301 is practically impossible without prior acidity ad justment, which makes the separati on process harder to manage. To overcome the ligand dissociation problem th at limits the effectiveness of Cyanex 301

PAGE 80

64 at high acidity, the alkyl groups on th e phosphorous were replaced by electron withdrawing substituents (Figure 3-2).29 Such alteration significantly weaken the basicity of the ligand and the new extractant was not able to bind any of the studied trivalent metal ions ( DAm,Eu < 10-3). P S SH X X X = H, CH3, Cl, F Figure 3-2. Structure of the ar omatic dithiophosphinic acids.29 The synergistic mixtures of this aroma tic dithiophosphinic acid with TBP (tri-nbutyl-phosphate) or TOPO (t ri-n-octyl-phosphine oxide) sh owed slightly improved ion binding.29,100-102 Even though these synergistic mixtur es facilitate high selectivity for An(III) over Ln(III) in a str ong acidic medium (1.5 M HNO3), the system achieves still very low distribution ratio s that are insufficient fo r practical application. The mechanism of M(III) ex traction (M = Am, Eu) using mixtures of aromatic dithiophosphinic acids with neutral O-bearing coextracta nts has been extensively investigated from a theore tical chemistry standpoint.48 Madic et al. elucidated that the formation of a hard/soft syne rgist extraction pair changes the nature of the M-S bonds due to the O M electron density transfer. It was determined that addition of a weak oxygen donor strengthens the M-S bonds while a strong donor weakens them. The mechanism has been confirmed by the 31P NMR chemical shifts studies of the neutral Obearing organophosphorus coextractants, wher e changes in the position of phosphorus resonances were correlated with M(III) extractabilities. Currently there are no extraction syst ems that would be free from major shortcomings. Most of methods are limited by either low extent of ligand dissociation in

PAGE 81

65 highly acidic medium, low efficiency in terms of values of distribu tion coefficients and poor stability toward hydrolysis. These problems limit the application of presented protocols such as Cyanex 301, HDEHDTP/TBP or aromatic dithiophosphinic acids in large scale nuclear waste clean-up operations. The purpose of our research was to develop a ligand that would combine the mo st advantageous qualities of previously studied extractants, and create an extracti on system that would be more suitable for industrial process deve lopment. To avoid sensitivity of the ligand towards high acid concentration and improve extraction effi ciency the dithiophosphinic acid group was replaced by phosphine sulfide and attached to the triphenoxymethane platform. As a result a neutral hexadentate tris-carbam oylmethylphosphine sulfide (tris-CMPS) ligand has been created. Another advantage expect ed from tris-CMPS was an enhanced ability to bind metals through the three carbonyl oxygens (build in hard donor synergist) that upon complexation could participate in the form ation of a six-membered chelate ring and improve the stability of the complex. O H NH C3 O4 C2 P1 S6 3M5 Figure 3-3. Anticipated binding mo de for tris-CMPS extractant. The previous investigation of the affinity of tris-CMPO derivatives for trivalent fblock elements in 1 M nitric acid/dichloro methane extraction systems has shown that harder and more polar nitrates win the competition for binding ions over the neutral

PAGE 82

66 chelate, which was established through low extr action efficiency (Chapter 2). Therefore, in the case of tris-CMPS extractant, the rather low affinity for trivalent f-elements was expected, yet slightly stronger interactions of ligand with the trivalent actinides were anticipated. However, if in the environmen t rich in nitrates and water the phosphine sulfide groups do not participate in metal binding, the ligand would not be able to differentiate between these two groups of f-elements. Our observations are reported herein. 3.2 Results and Discussion 3.2.1 Synthesis and Extraction Data The synthetic methodology to obtain tris -CMPS compounds follows procedures described for synthesis of tris-CMPO derivativ es, and differs only in the last step where p -nitrophenyl (diphenylphosphoryl)acetate (2-5) is replaced by (diphenylphosphinothioyl)-acetic acid (3-1) as presented in Figure 3-4. HO O P S 3-1 2-4a : n=1, R1, R2 = t-Pentyl 2-4b : n=1, R1, R2 = t-Bu 2-4c : n=1, R1 = Me, R2 = t-Bu 2-9a : n=2, R1, R2 = t-Pentyl R2 R1 O H 3 NH2 n+ R2 R1 O H 3 NH O P S n 3-2a: n=1, R1, R2 = t-Pentyl 3-2b: n=1, R1, R2 = t-Bu 3-2c: n=1, R1 = Me, R2 = t-Bu 3-3: n=2, R1, R2 = t-Pentyl Figure 3-4. Synthesis of tris-C MPS extractants. Conditions: mercaptothiazoline, DCC, DMAP, methylene chloride, rt overnight.

PAGE 83

67 Previously synthesized tris-CMPO com pounds showed very good selectivity for tetravalent actinides and lacked ability to e fficiently bind Ln(III) and An(III) (Chapter 2). The soft character of the basic phosphine su lfide groups in the new extractant may induce slightly stronger attraction of ligand for trivalent actinides and afford some discrimination between these two groups of elements in liquid-liquid extracti on experiments. Unfortunately extraction of 241Am(III) with tris-CMPS was found to be inefficient, and expected selectivity for americium over europium was not observed. For the comparison with the tris-CMPO ligand system extraction experiments were performed on a series of trivalent lanthani des and tetravalent thorium. 0 20 40 60 80 100 E% Th(IV)La(III)Ce(III)Nd(III)Eu(III)Yb(III) 2-6a 3-2a O H NH O P S 3 O H NH O P O 32-6a 3-2a Figure 3-5. Comparison of metal binding by tris-CMPO (2-6a) and tris-CMPS (3-2a). In the first experiment, binding properties of CMPO and CMPS derivatives with a shorter (two carbon) spacer be tween the triphenoxymethane ba se and binding units were compared (Figure 3-5). The results revealed ve ry low extraction efficiency of 3-2a for all studied cations. The tris-phosphine sulf ide compound was no longer able to take advantage of the difference in the oxidation st ates of f-elements cations. Soft, neutral phosphine sulfide group was found to be inco mpatible with a hard acid such as tetravalent thorium.

PAGE 84

68 In order to test the influence of the flexibility of the ligating CMPS arm on the extraction pattern, derivative 3-3, with elonga ted by an additional carbon atom arm was synthesized. In comparison to the 3-2a, 3-3 has not shown any improvement in the extraction (Figure 3-6). 0 5 10 15 E% Th(IV)La(III)Ce(III)Nd(III)Eu(III)Yb(III) 3-2a 3 3 O H NH O P S 33-3 O H NH O P S 33-2a Figure 3-6. Comparison of metal bi nding by tris-CMPS 3-2a and 3-3. As highlighted in Figure 3-7, the 2-10a a nd 3-3 differ only in the nature of the phosphine donor. The cavity size in both extr actants is similar and much like 2-10a, 3-3 should be able to easily adopt the geomet ry required by the metal ion and show some binding enhancement. 0 20 40 60 80 100 E% Th(IV)La(III)Ce(III)Nd(III)Eu(III)Yb(III) 2-10a 3 3 O H NH O P O 32-10a O H NH O P S 33-3 Figure 3-7. Comparison of metal binding by tris-CMPO (2-10a) and tris-CMPS (3-3).

PAGE 85

69 Exhibited low affinity of 3-3 for all test ed ions has proven the importance of the hard phosphine oxide in effective binding of any f-element ions, and negligible involvement of the phosphine sulf ide in the metal binding. Table 3-1. Extraction percentage (%E) for lig ands: 2-6a, 3-2a, 2-10 and 3-3. Aqueous phase: 10-4 M metal nitrate in 1 M HNO3, organic phase: 10-3 M of ligand in methylene chloride Cation (10-4 M) in 1 M HNO3 Equivs of ligand in organic phase 2-6a 3-2a 2-10a 3-3 Th (IV) 10 100 7 100 9 La (III) 10 3 9 16 9 Ce (III) 10 1 9 16 10 Nd (III) 10 5 10 15 10 Eu (III) 10 2 11 14 11 Yb (III) 10 4 9 13 9 3.2.2 Crystal Structure Analysis Single crystals of ligand 3-3 were grow n by slow diffusion of ether into the concentrated solution of ligand in dichlorome thane. In the crystal structure of 3-3 presented in Figure 3-8, the av erage length of the carbonyl bonds is similar to the length of carbonyl bonds in the CMPO equivalent 2-10b [1.225(7) a nd 1.232(4) respectively]. The average distances between the sulfur and phosphorous in phosphine sulfide moieties (1.952(2) ) are in the range of typical P=S bonds with phenyl substituents on the phosphorus (Ph3PS, P=S: 1.951(2)-1.954 (4)),103-105 but almost 0.5 longer than the distance between phosphorous and oxygen in the phosphine oxide 2-10b [1.477(3) , Chapter 2]. The P-C(Ph) mean bond length 1.810 (6) is also similar to the distances found in Ph3PS (1.817(7) ), as well as to t hose found in the tris-CMPO compound [1.798(4) ].

PAGE 86

70 Figure 3-8. Diagram of the solid-state structur e of 3-3 (30% probability ellipsoids for N, O, S and P atoms; carbon atoms drawn with arbitrary radii). For clarity, all hydrogen atoms have been omitted. In order to gain some understanding of the binding attributes of the ligand, a complex of Tb(NO3)3 with ligand 3-3 was synthesi zed (Figure 3-9). [3-3Tb(NO3)3] compound contains two similar structures in the asymmetric unit. The potentially hexadentate chelate 3-3, is c oordinated to the metal center in a tris-monodentate manner via carbonyl oxygens only. The interactions w ith terbium ions are similar in strength (2.320(3) ) to the interactions in the terbium complex with tris-CMPO 2-6a (2.308(5) ). Interestingly, none of the thre e sulfur atoms participate in the metal binding. Instead, three nitrate ions are bound in a bidentate fa shion, fully neutralizing the charge of the nine coordinate complex.

PAGE 87

71 Figure 3-9. Diagram of the structures of [3-3Tb(NO3)3] (right) and close-up view of the terbium coordination environment (left). Some examples of the X-ray analyzed so lid state complexes of lanthanides and acidic organophosphorous ligands with sulfur directly bound to the metal required an anhydrous environment during the synthesis due to their sensitivity to the moisture.106 If in the [3-3Tb(NO3)3] complex prepared from hydrated salt of lanthanide nitrate phosphine sulfides are not able to bind the cation, the aqueous extraction environment must even more effectively prevent sulfurs from interactions with a metal, and leftover three carbonyls cannot be expected to successf ully complete with highly concentrated nitrates and water for metal binding.

PAGE 88

72 Table 3.2. Selected bond lengths () for compounds: 3-3 and [3-3Tb(NO3)3]. 3-3 [3-3Tb(NO3)3] Tb(1) [3-3Tb(NO3)3] Tb(2) P(1)-S(1) 1.956(2) 1.9590(14) 1.9432(19) P(2)-S(2) 1.943(2) 1.9526(15) 1.9532(18) P(3)-S(3) 1.958(2) 1.9485(15) 1.9536(17) C(53)-O(4) 1.221(7) 1.241(4) 1.236(5) C(70)-O(5) 1.223(7) 1.248(4) 1.233(4) C(87)-O(6) 1.232(7) 1.251(4) 1.241(5) P(1)-CPh(55) 1.802(6) 1.807(4) 1.847(6) P(1)-CPh(61) 1.812(6) 1.810(4) 1.806(5) P(2)-CPh(72) 1.806(6) 1.814(4) 1.828(5) P(2)-CPh(78) 1.821(6) 1.812(4) 1.782(6) P(3)-CPh(89) 1.802(6) 1.811(4) 1.814(4) P(3)-CPh(95) 1.814(6) 1.807(5) 1.815(4) M-O(4) 2.316(2) 2.298(3) M-O(5) 2.317(3) 2.341(3) M-O(6) 2.330(3) 2.318(3) M-O(7) 2.493(3) 2.468(4) M-O(8) 2.437(3) 2.426(4) M-O(10) 2.524(3) 2.427(3) M-O(11) 2.431(3) 2.505(3) M-O(13) 2.492(3) 2.484(3) M-O(14) 2.452(3) 2.465(3)

PAGE 89

73 Table 3-3. X-ray dataa for the crystal structures of 3-3 and the [3-3Tb(NO3)3] complex. 3-3CH2Cl2C4H10O [3-3Tb(NO3)3]2CH3CN total reflections unique reflections max( ) empirical formula Mr crystal system space group a () b () c () ( ) ( ) ( ) Vc (3) Dc (g cm-3) T (K) Z (MoK ) (mm-1) R1 [ I 2 ( I ) data]b w R2 (all data)c GoF 35243 12730 23.43 C105H142Cl2N3O7P3S3 1818.21 monoclinic P 21/n 14.8551(10) 13.6147(10) 49.329(3) 90 96.5800(10) 90 9910.9(12) 1.219 173(2) 4 0.233 0.0858 [7722] 0.2182 1.033 73828 48890 28.03 C104H136N8O15P3S3Tb 2086.22 triclinic P -1 16.8425(7) 25.0080(10) 27.0255(11) 78.3410(10) 74.6450(10) 87.4780(10) 10749.7(8) 1.289 173(2) 4 0.824 0.0558 [34928] 0.1625 1.081 aObtained with monochromatic Mo K radiation ( = 0.71073 ) bR1 = Fo Fc / Fo cw R2 = { [ w ( Fo 2 Fc 2)2/ [ w ( Fo 2)2]}1/2 3.3 Conclusions Tripodal phosphine sulfide based compounds we re synthesized as softer derivatives of the tris-CMPO chelate. Their ability to differentiate between trivalent lanthanides and actinides was tested using 41Am and 152Eu isotopes. Extraction experiments using tris-CMPS compounds revealed that the liga nds are not able to preferentially bind trivalent Am over Eu. The soft nature of the phosphine sulfide donor group was found to be incompatible with the hard tetravalent thor ium, and in contrast to the tris-CMPO, the tris-CMPS analog showed no selec tivity for this metal ion. The solid state structure of tris-CMPS with terbium nitrate showed that the phosphine sulfide portion of the ligand is not involved in the metal biding. Although

PAGE 90

74 with no structural studies on the tris-CMPS extractant in the solution, it can be assumed that in the system with negatively charged nitrates, soft neutral sulfur atoms do not participate in the coordination of hard metals. 3.4 Experimental Section Amines 2-4 a, b, c and 2-9a were synthe sized as described in Chapter 2. The (diphenyl-phosphinothioyl)-acetic acid, 3-1, wa s prepared according to the literature procedure.107 The final synthetic procedures of tris-CMPS compounds were developed in the collaboration with Dr. Ajay Sah a nd Dr. Priya Srinivasan. The [3-3Tb(NO3)3] complex used in the discussion of this chapter was crystallized by Dr. Ajay Sah. Compound 3-2a. A mixture of (diphenyl -phosphinothioyl)-a cetic acid (3-1) (5.40 g, 19.55 mmol), 2-mercaptothi azoline (2.51 g, 21.06 mmol) and 4-dimethylaminopyridine (0.60 g, 4.91 mmol) wa s stirred in dichloromethane (200 mL) at room temperature for 30 min. N, N Â’-dicyclohexylcarbodiimide (4.36 g, 21.13 mmol) was then added followed by additional 15 mL of dichloromethane. After 6 h, solid 2-4a (4.44 g, 5.27 mmol) was added and the mixture was stirred for an additional 24 h at room temperature. The slurry was filtered and the solvent was removed in vacuo The product was separated from byproducts by the dissolution in diethyl ether. Addition of methanol to the concentrated solution of the compound resulted in precipita tion of solid product that was subsequently filtered and washed w ith cold methanol. Yield 5.57g (65%). 1H NMR (CDCl3): = 0.46 (t, J = 7.3, 18H; CH2C H3), 1.07 (s, 18H; CC H3), 1.23 (s, 18H; CC H3), 1.41 (q, J = 7.3, 6H; C H2CH3), 1.59 (br, 6H; C H2CH3), 3.37 (br, 12H; OC H2C H2NH), 3.64 (d, J = 14.1, 6H; C(O)C H2P(O)), 6.23 (s, 1H; C H ), 6.92 (d, J = 2.1, 3H; Ar), 6.96 (d, J = 2.1, 3H; Ar), 7.38 (m, 18H; Ar), 7.64 (br, 3H; N H ), 7.87 (m, 12H;

PAGE 91

75 Ar). 13C NMR (CDCl3): ( C =O) = 165.13, 165.07; (Aromatic) = 153.1, 142.9, 140.0, 137.8, 131.90, 131.87, 131.7, 131.5, 128.9, 128.7, 127.7, 125.0; (aliphatic) = 70.3 (O C H2), 42.6 ( C H2NH2), 42.0, 40.2, 39.3, 37.8, 37.0, 35.5, 29.8, 28.7, 9.8, 9.3. 31P NMR (CDCl3): = 38.9. Anal. Calcd for C97H124N3O6P3S3: C, 72.04; H, 7.73; N, 2.60. Found: C, 72.19; H, 7.85; N, 2.57. Compound 3-2b. A mixture of 3-1 ( 2.13 g, 7.71 mmol), 2-mercaptothiazoline (0.97 g, 8.14 mmol) and 4-dimethylaminopyrid ine (0.29 g, 2.37 mmol) was stirred in dichloromethane (80 mL) for 30 min. The solid N, NÂ’-dicyclohexylcarbodiimide (2.45 g, 11.87 mmol) was then added, followed by additiona l 20 mL of dichloro methane. After 6 h, solid 2-4b (1.69 g, 2.23 mmol) was added and the mixture was stirred for an additional 24 h. The slurry was filtered and the solvent was removed in vacuo The solid was dissolved in diethyl ethe r and quickly filtered. With in several days upon slow evaporation of solvent pure product precip itated from the ether solution. Yield 1.44g (42%). 1H NMR (CDCl3): = 1.16 (s, 27H; CC H3), 1.23 (s, 27H; CC H3), 1.85 (br, 6H; CH2C H2CH2), 3.35 (br, m, 12H; OC H2CH2C H2NH), 3.54 (d, J = 14.1, 6H; C(O)C H2P(O)), 6.24 (s, 1H; C H ), 7.09 (d, J = 2.3, 3H; Ar), 7.20 (d, J = 2.6, 3H; Ar), 7.40 (m, 18H; Ar), 7.61 (t, J = 5.4, 3H; N H ) 7.88 (m, 12H; Ar). 13C NMR (CDCl3): ( C =O) = 164.82, 164.77; (Aromatic) = 153.6, 144.5, 142.0, 137.7, 132.8, 132.01, 131.97, 131.7, 131.5, 128.9, 128.8, 127.2, 122.4; (aliphatic) = 70.3 (O C H2), 37.9, 35.7, 34.7, 31.7, 30.6, 25.6. 31P NMR (CDCl3): = 38.9. Anal. Calcd for C94H118N3O6P3S3: C, 71.68; H, 7.55; N, 2.67. Found: C, 71.22; H, 7.88; N, 2.53. Compound 3-2c. Method I: A mi xture of 3-1 (0.30 g, 1.12 mmol), 4-dimethylaminopyridine (0.13 g, 1.12 mm ol) and EEDQ (1.11g, 4.48 mmol) were

PAGE 92

76 dissolved in pyridine (10 mL). After s tirring for 1 hour 2-4c (0.18 g, 0.28 mmol) was added, and the reaction mixture was heated to 50C for 18 hours. Af ter cooling to room temperature, the solvent was removed in vacuo and the residue was extracted with 9:1 CHCl3/MeOH solution followed by washing with 1N HCl. Organic phases were collected, dried over MgSO4, and solvent was removed. The solid residue was dissolved in diethyl ether and upon addition of pentane, the product precipitated out of the solution. The compound was purified by column chromatography (SiO2, hexane/ether) to give 0.07 g of ligand 3-2c in form of a white so lid (18 % yield). Me thod II: 3-1 (1.90 g, 6.90 mmol), 4-dimethylaminopyridine (0.31 g 2.53 mmol), mercaptothiazoline (0.85 g, 7.13 mmol) and N, NÂ’-dicyclohexylcarbodiim ide (1.47 g, 7.13 mmol) were dissolved in dry dichloromethane (50 mL). After few minutes, the solution turned bright yellow and a white solid separated out. The mixture was stirred for an additional 4 hours and 2-4c (1.16 g, 1.84 mmol) was added. The resulting slurry was allowed to stir at room temperature for 48 hours. The white solid of byproducts formed in the reaction was filtered, and the solvent was removed in vacuo Addition of ether to the condensed reaction mixture dissolved the product leav ing an amorphous mass of byproducts. The organic solution was decanted, and within se veral days upon slow evaporation of solvent pure product precipitated from the solution. Yield 0.8 g (31%). 1H NMR (CDCl3): = 1.15 (s, 27 H; Ar-C(CH3)3), 2.13 (s, 9 H; Ar-CH3), 3.25 (t, 6 H; Ar-O-CH2CH2), 3.33 (t, 6 H; Ar-O-CH2CH2), 3.65 (d, J (H,P) = 14.1 Hz, 6 H CH2-POAr2), 6.67 (s, 1 H, C-H), 6.87 (b, 3 H; Ar-H), ), 6.67 (s, 1 H, C-H), 6.95 (b, 3 H; Ar-H), 7.42 (m, 18 H; P-Ar-H), 7.60 (t, 3 H; NH ) 7.89 (m, 12 H; P-Ar-H)). 31P NMR (CDCl3): = 38.8. MS [M+H]+ = 1406.5532 (Theoretical [M+H]+ = 1406.5596).

PAGE 93

77 Compound 3-3. A mixture of 3-1 (4 .73 g, 17.12 mmol), 2-mercaptothiazoline (2.14 g, 17.95 mmol) and 4-dimethylaminopyrid ine (0.67 g, 5.48 mmol) was stirred in dichloromethane (160 mL) for 30 min. N, NÂ’-dicyclohexylcarbodiimide (5.34 g, 25.88 mmol) was added followed by additional po rtion dichloromethane (20 mL). After 6 h, solid 2-9a (4.10 g, 4.64 mmol) was added and the mixture was stirred for 24 h. The slurry was filtered and the solvent was re moved. The product was separated from the reaction byproducts by dissolution in ether. The crude product was recrystallized from methanol to give 6.70g (87%) of pure compound. 1H NMR (CDCl3): = 0.45 (m, 18H; CH2C H3), 1.10 (s, 18H; CC H3), 1.22 (s, 18H; CC H3), 1.42 (q, J = 7.4, 6H; C H2CH3), 1.57 (q, J = 7.1, 6H; C H2CH3), 1.82 (br, 6H; CH2C H2CH2), 3.30 (br, 12H; OC H2CH2C H2), 3.54 (br, d, J = 13.6, 6H; C(O)C H2P(O)), 6.18 (s, 1H; C H ), 6.95 (d, J = 2.1, 3H; Ar), 7.04 (d, J = 2.1, 3H; Ar), 7.40 (m, 18H; Ar), 7.63 (t, J = 5.5, 3H; N H ), 7.88 (m, 12H; Ar). 13C NMR (CDCl3): ( C =O) = 164.8, 164.7; (Aromatic) = 153.3, 142.4, 139.8, 137.9, 132.7, 131.94, 131.90, 131.6, 131.4, 128.9, 128.7, 127.9, 124.7; (aliphatic) = 69.7 (O C H2), 42.7 ( C H2NH2), 42.0, 39.3, 37.9, 37.7, 37.0, 35.3, 30.5, 29.7, 28.7, 9.7, 9.2. 31P NMR (CDCl3): = 38.9. Anal. Calcd for C100H130N3O6P3S3: C, 72.39; H, 7.90; N, 2.53. Found: C, 72.00; H, 8.10; N, 2.54.

PAGE 94

78 CHAPTER 4 BINDING OF TRIVALENT FELEMENTS FROM ACIDIC MEDIA WITH A C3SYMMETRIC TRIPODAL LIGAND C ONTAINING DIGLYCOLAMIDE AND THIO DIGLYCOLAMIDE ARMS 4.1 Introduction In the view of the ever-increasing use of nuclear power around the world,1-3 an accelerated development of effective protocol s for waste treatment increasingly becomes imperative. Perhaps, one of the most signifi cant obstacles faced in the separation science is partitioning of minor actinides. Decad es of work have been dedicated to the development of amidic extractants for th e f-element liquid-liquid waste separations.18,2022,108-116 Recently, a significant in terest has been focused on one particular group of amides diglycolamides (DGA).117-132 These completely incinerable tridentate, neutral chelates are much more effective in coextract ion of lanthanides and minor actinides than commercially operating DIAMEX extractants.19,22-25,133-136 Moreover, unlike other diamide-based compounds, DGAs exhibit signifi cant selectivity within the lanthanide series and track the increase in charge density .122,124,125,129,130,132 With this ability to selectively bind metals within the series DGAs offer many potential applications in analytical chemistry in addition to waste partition. The further refinement of diglycolamide based chelators and careful i nvestigation of their binding potential may help greatly improve efficiency of the lant hanide/actinide coextraction process, and a fundamental understating of the remarkable selectivity of DGA may help in the development of improved methods for partitioning of mi nor actinides.

PAGE 95

79 In highly acidic nitric acid solutions comm on in nuclear waste reprocessing, two to four molecules of diglycolamides appear to be involved in the coor dination of trivalent felement ions during extractions,122,126,129,132 and the preorganizati on of several ligating DGA units onto a molecular platform could pot entially improve the efficiency of the extraction, as it was observed in some cases of carbamoylmethylphosphineoxide (CMPO) based extracts.49,51-54,76,78,79 A single ligand with three DGA arms would present the metal with nine favourable donor groups, si x of which are relatively hard amide oxygen donors. This chapter presents the synthesis of C3-symmetric tripodal chelates bearing three thioand diglycolamide units precisely ar ranged on a triphenoxymethane platform along with their evaluation as extr actants in f-element separati ons. The ability of tripodal ligands to extract trivalent lanthanides and act inides from nitric aci d to the organic phase was evaluated based on the comparison with the extraction data for the simple [(Diisopropylcarbamoyl)-methoxy]-N,N-diisopropyl-a cetamide (4-2). The influence of the flexibility and lipophilicity of DGA arms in the ligand on the extraction profile has been also investigated. To verify the contribution of the etheric oxygens to the tris-DGA binding efficiency, a new C3-symmetric derivative containing sulfur in the place of the etheric oxygen has been synthesized, and the properties of both types of ligands have been compared. Additionally, to determine the affect of immobilization of DGA on the metal binding site geometry, three complexes of Ce(NO3)3, Eu(NO3)3 and Yb(NO3)3 with tris-DGA ligands were synthesized. Solid structures of these complexes were elucidated by ICR-MS and X-ray analysis and the results are discussed herein.

PAGE 96

80 4.2 Results and Discussion 4.2.1 Ligand Synthesis The tris-thio/diglycolamides (4-5 – 4-8) have been prepared by the reaction of primary amines 2-4a or 2-9b with mono-subs tituted oxa/thio-pentaneamides (4-1, 4-3 and 4-4) and with the coupling agent benz otriazole-1-yl-oxy-trispyrrolidinophosphonium hexafluorophosphate (PyBOP) as illustrated in Scheme 4-1. All final products have been obtained in high yields and purity with relatively small synthetic effort. The simple diglycolamide 4-2 was synthesized according to a modified literature procedure137 and used as an extraction reference molecule. The synthetic pathways for preparation of compounds 4-1 4-4 have been adapted from general procedures for variety of diamides.137-139 Amines 2-4a51 and 2-9b (Chapter 2) were synthesized according to previously developed procedures. HNR2XO O O X NR2 O R' O O NH2 n 3 O NH 3 X NR2 O O n B + 4-1 X = O, R' = OH, R = iPr 4-2 X = O, R' = N(iPr)2, R = iPr 4-3 X = O, R' = OH, R = nBu 4-4 X = S, R' = OH, R = iPr 4-5 R = iPr, X = O, n =1 4-6 R = iPr, X = O, n =2 4-7 R = nBu, X = O, n =1 4-8 R = iPr, X = S, n =1 2-4a n =1 2-9b n =2 A Figure 4-1. Synthesis of C3-symmtric tris-diglycolamides. (A) 1,4-dioxne, pyridine; (B) PyBOP, diisopropylethylamine, DMF.

PAGE 97

81 4.2.2 Extraction Experiments The extraction experiments were performed on a series of eleven lanthanides, 152Eu, and 241Am radioisotopes. Solutions of 10-4 M metal nitrates in 1 M nitric acid were mixed with equal volumes of 10-3 and 10-4 M organic solutions for approximately 20 h. The reference molecule 4-2 was consistently used at a three times higher concentration than the tris-diglycolamide derivatives fo r a fair comparison. The concentration of lanthanide ions in the aqueous phase before and after the extraction were determined spectrophotometrically ( = 655 nm),51,88 or in the case of 241Am and 152Eu, they were measured by a Canberra GammaTrac 1185 with Ge(Li) detector and AccuSpec-B multichannel analyzer. Extraction efficiencies were calculated using the formula: %E = 100%(A1-A)/(A1-A0), where A is the absorbance of the extracted aqueous phase with the Arsenazo(III) indicator, A1 is the absorbance of the aqueous phase before extraction with the indicator, and A0 is the absorbance of metal-free 1 M nitric acid and the indicator. The errors, based on the pr ecision of the spectrophotometer and the standard deviation from the mean of at leas t three measurements, were in most cases no higher than two percent. The extraction percentage was further converted into distribution ratios of the tota l metal ion concentration in the organic phase against the total metal ion concentrati on in the aqueous phase ( D = [Morg]/ [Maq]) and in view of the errors associated with the spectrophotometric techniqu e, the maximum value that could be measured for the extraction pe rcentage and distribution ratio was 99. 4.2.2.1 Extraction properties of large chelate vs. small diglycolamide Typically, high distributions of trivalent f-elements in organic/acidic extraction systems are reported with approximately ~100,000:1 DGA to metal ion concentration (e.g. 0.2 M N,NÂ’-dimethyl-N,NÂ’-diphenyl-4 -oxapenanediamide DMDPhOPDA, 10-6 M

PAGE 98

82 Ln(III)).122 With the tri-DGA ligand (4-5), compar able distribution ratios can be obtained with a 10:1 ligand to metal ratio. To pr ovide some context for the extraction of efficiency of 4-5 under controlled experiment al conditions, the prope rties of ligand have been evaluated with respect to the perfor mance the related DGA, 4-2, and since 4-5 contains three arms, the concentration of 4-2 was increased by factor of three. This oversimplified comparison is used strictly to present the significant changes in the efficiency and selectivity of studied diglycolamide chelates. -2 -1 0 1 2 3 LaCePrNdEuGdTbDyErTmYb log D 4-2(1) 4-5(1) 4-2(2) 4-5(2) Figure 4-2. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with dichloromethane solutions c ontaining ligand 4-2 at 3x10-3 M (1) or 3x10-4 M (2) and solutions of 4-5 at concentrations of 10-3 M (1) and 10-4 M (2). Due to the limitations of the spectroche mical assay, the error bars for D are quite large at high extracti on efficiency (>98%). An experiment with the re ference molecule 4-2 at 3x10-3 M (1) showed the typical extraction pattern for diglycolamides, whic h gradually ascends across the lanthanide series (Figure 4-2, Table 4-1). Once th ree DGA moieties were attached to the triphenoxymethane platform (4-5), the efficiency of the ligand for the heaviest lanthanides was remarkably improved. Only ten-fold excess of ligand 4-5 allowed for quantitative removal of trivalent erbi um, thulium and ytterbium from 1 M HNO3 solution. Even though the direct comparison of these two compounds (4-2 and 4-5) is impossible

PAGE 99

83 due to the fundamental differences in the char acter of the extracted species, considering only the concentrations required for high ex tractability, the efficiency and therefore economy gain in the case of 4-5 is clearly evident. Moreover, this unique design of rather flexible, nine oxygen donor cavitand allowe d for much more sensitive ion size recognition than in the case of any other DGA chelates. Table 4-1. Extraction data (log D )* for ligands 4-2and 4-5 in dichloromethane. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with dichloromethane solutions c ontaining ligand 4-2 at 3x10-3 M (1) or 3x10-4 M (2) and solutions of 4-5 at concentrations of 10-3 M (1) and 10-4 M (2). Ligand 4-2 4-5 Cation 1 2 1 2 La (III) -0,43 -1,06 -1,06 -0,99 Ce (III) -0,09 -0,95 -0,66 -0,95 Pr (III) 0,08 -1,00 -0,30 -0,87 Nd (III) 0,16 -0,91 0,10 -0,72 Eu (III) 0,47 -1,00 1,16 -0,15 Gd (III) 0,51 -1,00 1,28 -0,08 Tb (III) 0,74 -1,00 1,69 0,09 Dy (III) 0,88 -0,91 1,69 0,21 Er (III) 1,02 -0,95 2,00 0,26 Tm (III) 1,00 -0,91 2,00 0,29 Yb (III) 0,99 -0,91 2,00 0,29 *D calculated based on the E % values. Enhanced affinity of tris-DGA for the heav iest lanthanides and decreased attraction for the lightest (La, Ce, Pr), resulted in im proved separation factor between the elements in the group (separation factor SFA/B=DA/DB; the fraction of the individual distribution ratios of two extractable solutes measured under the same conditions). For instance, the value of the SF of Yb(III) and La(III) increased from SFYb/La=26 for 4-2 (1) to SFYb/La=1138 for 4-5 (1). Surprisingly, further dilution of the organic phase to a 1:1 metal to ligand ratio maintained a high extraction efficiency for the heaviest lanthanides; over two thirds of Tm(III) and Yb(III) was tran sferred into the organic phase. At the same time, with a three times higher concentration of 4-2 (2) (3x10-4 M), the extraction

PAGE 100

84 was negligible and no separation was observe d suggesting necessity of the significant excess of ligand 4-2 to achieve appreciabl e extraction, and once again confirming the efficiency gain through the DGA preorganization in the compound 4-5. 4.2.2.2 Ligand flexibility vs. extraction performance In work with calix[4]arenes and trip henoxymethane molecule appended with CMPO arms, the extraction efficiency of th e constructs was amplified by the increased flexibility of the linker between the CM PO and the base skeleton (Chapter 2).76, This trend can be attributed to the enhanced ab ility of ligand to satisfy the geometrical requirements of the metal center, but often, th e improvement in binding affinity comes at the expense of selectivity. In 4-2, the DGA groups are tethered to the triphenoxymethane platform by only two carbons, and to test the effect of the leng th of this arm linker on the extraction ability, a new tris-substituted di glycolamide was synthesized with three carbons linking the DGA arms to the triphenoxymethane base (4-6). Considering the high efficiency of the “more rigid” chelate 4-5 at the 10:1 ligand to metal proportion, and expected performan ce enhancement of 4-6 over 4-5, the experiment has been conducted with only 10-4 M concentration of 4-6 in the dichloromethane (1:1 ligand to metal ratio). Interestingly, the new extractant did not appear to be more effective or less selective than its “more ri gid” equivalent. In fact, the 4-6 exhibits the same extraction behavior as compound 4-5, suggesting that the binding environment in the two ligands is nearly iden tical. As it will be discussed later, the coordination setting of the meta l center in the complexes of 4-5 and 4-6 were found to be nearly indistinguishable.140 Apparently, both ligands are perfectly suited to fulfill the coordination requirements of tr ivalent lanthanides in the so lid state and in an organic solution which explains identic al extraction behavior.

PAGE 101

85 4.2.2.3 Solvent effect on ligand extraction profile There are many complex processes that influe nce the transfer of the metal ion from an acid layer into an organic phase, and in a very simplistic view, the three major factors dominate the extraction event: solubility, steric hindrance presented by a ligand, and the electronic effect.141 The properties of the organic me dium strongly influence not only the binding ability of an extractant, but also the st ability of the complex in the organic phase. Therefore, our investigation also involved the influence of solvents other than dichloromethane on the ability of extractan t 4-5 to bind trivalent lanthanides. -20 0 20 40 60 80 100 LaCePrNdEuGdTbDyErTmYb D DCM 1-octanol Figure 4-3. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with dichloromethane (DCM) and 1-octanol solutions containi ng ligand 4-5 at 10-3 M. In our studies with 4-5, a significant modulation in the extraction pattern was noted when 1-octanol was substituted for dichloromethane as the organic solvent (Figures 4-3 and 4-4).140 From a gradually ri sing affinity toward heavier lanthanides in dichloromethane, the extractant tends to fa vor the middle lanthanides in 1-octanol, in particular europium. The heav iest lanthanides are still more readily extracted than the lightest ones, although the sepa ration between Yb(III) and La( III) significantly decreases from SFYb/La=1138 to SFYb/La=27. The preference of Eu(III) binding arises apparently from combined effects of the best fit between metal and ligand, and the complex

PAGE 102

86 stabilization provided by 1octanol. Therefore, the origin of selectivity in dichloromethane may not be simply related to the careful recognition of size of the metal ion or an increased charge de nsity on the metal center. The selectivity must be strongly associated with the properties of an organic pha se and results from th e combined effect of many possible interactions between the ligand, metal and both aqueous and organic solutes. A similar tendency (within the expe rimental error) was observed at 1:1 ligand to metal ratio [Figure 4-4, 4-5 (2)]. -2 -1 0 1 2 LaCePrNdEuGdTbDyErTmYb log D 4-2(1) 4-5(1) 4-5(2) 4-5(3) Figure 4-4. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with solutions containing ligand 4-2 in octanol at 3x10-3 M (1) or ligand 4-5 in octanol at 10-3 M (1) or 10-4 M (2) or in n-dodecane at 10-4 M (3). Due to the limitations of the spectrochemical assay, the error bars for D are quite large at high extraction efficiency (>98%) The reference molecule 4-2 does not show analogous behavior and at 3x10-3 M concentration exhibits a smaller and constant affinity for most lanthanides except for La(III), which is extracted to the lowest exte nt (Figure 4-4, table 4-2). In the case of ndodecane, limited solubility of N,NÂ’-diisopropyl derivative 4-5 allowed an extraction with only one equivalent of th e ligand with respect to the metal ions [4-5 (3)]. The affinity of extractant for lanthanides was found to be ve ry strong, but the resulting

PAGE 103

87 complexes were not soluble in this str ongly nonpolar medium and precipitated upon extraction. Table 4-2. Extraction data (log D )* for ligands 4-2and 4-5 in octanol and dodecane. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with solutions containing ligand 4-2 in octanol at 3x10-3 M (1) or ligand 4-5 in octanol at 10-3 M (1) or 10-4 M (2) or in dodecane at 10-4 M (3). Ligand 4-2 4-5 Cation 1 1 2 3 La (III) -0,39 -0,52 -1,69 -0,23 Ce (III) -0,03 0,35 -0,75 0,16 Pr (III) 0,09 0,89 -0,37 0,21 Nd (III) 0,16 1,38 -0,18 0,31 Eu (III) 0,12 1,82 -0,03 0,39 Gd (III) 0,09 1,69 -0,12 0,39 Tb (III) 0,09 1,69 -0,18 0,43 Dy (III) 0,09 1,51 -0,25 0,45 Er (III) 0,09 1,16 -0,37 0,45 Tm (III) 0,07 0,98 -0,45 0,50 Yb (III) 0,03 0,91 -0,43 0,50 *D calculated based on the E % values The simplicity of structural modification of ligandÂ’s substituents allows for tuning the solubility of this bulky polar ligand, and therefore facilitates te sts in the industrial solvents such as 1-octanol and n-dodecane. In order to increase the lipophilicity of the tris-DGA chelate, a new derivative with two nbutyl substituents att ached to each of the terminal amidic nitrogens was synthesized (4-7 ). As expected, the new ligand is highly compatible with nonpolar solvents, but at the same time, the binding ability of 4-7 relative to 4-5 is cons iderably suppressed. A similar tr end has been noted in simple diglycolamides.129 In four different diluents: ch loroform, toluene, n-hexane and ndodecane, the distribution coefficients of th e ligands decreased when the length of the alkyl chains was extended.

PAGE 104

88 -1,5 -0,5 0,5 1,5 2,5 3,5 LaCePrNdEuGdTbDyErTmYb log D 4 5 4 7 Figure 4-5. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with dichloromethane solutions contai ning ligands 4-5 and 4-7 at 10-3 M. Due to the limitations of the spectroche mical assay, the error bars for D are quite large at high extracti on efficiency (>98%). Table 4-3. Extraction data (log D )* for ligands 4-5and 4-7 in dichloromethane. Extraction of 10-4 M solutions of trivalen t lanthanides in 1 M HNO3 with dichloromethane solutions contai ning ligands 4-5 and 4-7 at 10-3 M. Ligand 4-5 4-7 Cation La (III) -1,06 -1,03 Ce (III) -0,66 -1,06 Pr (III) -0,30 -1,00 Nd (III) 0,10 -0,98 Eu (III) 1,16 -0,54 Gd (III) 1,28 -0,51 Tb (III) 1,69 -0,24 Dy (III) 1,69 -0,09 Er (III) 2,00 0,07 Tm (III) 2,00 0,10 Yb (III) 2,00 0,12 *D calculated based on the E % values At a ratio of 10:1 ligand to meta l in the dichloromethane/1 M HNO3 system, the extraction efficiency of 4-7 re lative to 4-5 is reduced by n early 50%, and in the case of the heaviest lanthanides, the distribution ratio for the extraction with 4-7 ( D ~ 1) is significantly lower than the corresponding value for 4-5 ( D > 99) (Figure 2-5). Even though both ligands exhibit the same general trend for pref erential extraction of the heavier lanthanides, the selectivity of the tw o is notably different. For example, the

PAGE 105

89 separation between Yb(III) and La(III) drops from SFYb/La=1138 for 4-5, to SFYb/La=14 for 4-7, while between Yb(III) and Nd (III) the factor decreases from SFYb/Nd=78 to SFYb/Nd=13 respectively for 4-5 and 4-7. In addition, the extraction efficiency for the four lightest lanthanides is negligib le, and the limitations of th e spectrophotometric analysis complicate the analysis of this trend. In the attempt to fully delineate the tr is-DGAs extraction behavior experiments with 4-7 in 1-octanol and n-dodecane were pe rformed. Unfortunately, due to the phase separation problems and large errors asso ciated with measurements, ambiguous and irreproducible result s were obtained. 4.2.3 Investigation of Solid State Co mplexes of Trivalent Lanthanides The coordination number of trivalent lantha nide hydrated salts in solution vary from 9 to 8 across the series,142-147 and in the solid state, th e lanthanide aqua complexes adopt nearly rigorous tricapped trigonal prismatic (TTP) geometry(Figure 4-6). 142,146-148 Figure 4-6. The model of an ideal tricappe d trigonal prismatic (TTP) geometry around nine coordinate metal ion (left). The top and side views of a slightly distorted TTP coordination environment of the Yb(III) center in the complex with ligand 4-5 (nonbonding part of the ligand has been omitted for clarity). The crystal structure of Yb(III) comp lex with ligand 4-5 (right).

PAGE 106

90 Surprisingly, there are only few simple ligands compatible with highly acidic solutions that can fully satisfy the geometri cal preferences of Ln (III) and An(III) centers, but none of them is purely oxygen based. The tris-DGA ligand presents lanthanide s with nine oxygen donor groups, six of which are relatively hard amid e oxygens. The crystal structur es of the representatives of the heaviest, lightest, and th e middle lanthanides, Yb(NO3)3 and Ce(NO3)3 with 4-5, and Eu(NO3)3 with 4-6, demonstrate a good match betw een the size and th e tricapped trigonal prismatic coordination requirements of lant hanides and the binding pocket of tris-DGA ligands.140 A depiction of the structure of the cationic complex [4-5Ln]3+ [Ln = Ce(III), Yb(III)] is presented in Figure 4-7. Figure 4-7. Diagram of th e structures of [4-5Yb]3+ (left) and [4-2Yb]3+ (right) (30% probability ellipsoids for Yb, N, a nd O atoms; carbon atoms drawn with arbitrary radii). For clarity, all hydr ogen atoms and nitrates have been omitted. Primed and unprimed are related by a 2-fold symmetry operation. Ce(III) ligated by 4-5 adopts structure nearly identical to [4-5Yb]3+.

PAGE 107

91 All three DGA arms are equally involved in the tight metal binding in a tridentate fashion and form a slightly distorted TTP a rrangement about the metal center. The ligand fully saturates the coordination sphere of th e metal ion leaving no space for nitrates to bind. The two of the three charge neutralizi ng nitrates were found to be bonded to amide hydrogens of the ligand via hydrogen bonding intera ctions in the solid state structure of 4-5. This ability of the amides to hydrogen bond to the nitrate counterions may facilitate the extraction event by providing a suitable e nvironment for the ions in the organic solvent. Initially, only two nine coordinate co mplexes of ligand 4-5 with Ce(III) and Yb(IIII) have been synthesized and revealed th e same slightly distorted tricapped trigonal prismatic geometry. The triangular faces of the carbonyl oxygens O( 6), O(9), O(12) and O(4), O(7), O(10), are somewhat twisted about the three fold axis. The etheric oxygens (equatorial oxygens) O(5), O(8), and O(11) cap the faces of the trigonal prism. Consistent with the size differences in the two metals (ionic radius for CN=9, rCe(III)=1.336 , rYb(III)=1.182 ),85 the smaller, more electropositive Yb(III) binds tighter to the ligand. The bond lengths to the am idic oxygens {Ce: 2.396(9)-2.479(8) [mean distance of 2.434(9)]; Yb: 2.301(2)-2.351(3) [mean distance of 2.313(3)]} are shorter than to the etheric oxygens {Ce: 2.544(7 )-2.597(7) , [mean 2.566(7)]; Yb: 2.413(2)2.455(2) , [mean 2.429(2)]} by approximately 0.11 . Interestingly, in the only structurally characterized ex ample of a DGA-Ln(III) complex available in the literature, the bond distances to the etheric oxygens varied from 2.679 to 2.849 in a ten coordinate La(III) bis DGA complex.149 In order to ascertain the influence of the triphenoxymethane base on the metal center, th e structure of the ytterbium complex with

PAGE 108

92 three N,N,NÂ’,NÂ’-tetra-isopropyl-4-oxa-penta nediamide reference molecules (4-2) was determined (Figure 4-7), and the Yb(III) center in this species ([3(4-2)Yb]3+) maintains the same coordination environment. Th e distances between amidic oxygens 2.282(7)2.344(9) [mean 2.317(9)] and the etheri c oxygens and metal 2.421(8)-2.436(8) [mean 2.426(8)] are virtually indistinguish able from the structure of [4-5Yb]3+. The orientation of the oxygen atoms and th e metal oxygen bond distances in the DGA complexes are somewhat different from the aqua groups. For example, in the nonaaqualanthanoid(III) complex [M(OH2)9]3+ (M = Ce(III), Yb(III)) with trifluoromethanesulfonates anions, the distances to the pr ism oxygens are 2.489(2) and 2.302(2) , and to the equatorial oxygens ar e 2.594(2) and 2.532(3) for the Ce and Yb respectively.150 The difference is most apparent in the case of Yb, where the mean distance from the metal to equatorial oxygen is on average 0.103 shorter in the trisDGA complex than in the aqua complex, wh ile the metal-prism oxygen distances are only marginally longer by 0.011 . The Yb(III ) is held tightly by the ligand, and the twist angle of the trigonal prism is only 15.2 In the case of Ce(III), both prismatic and equatorial bonds are shorter than in the aqua complex by 0.055 and 0.028 , respectively, but the twist angle increases to 21.6. The change in distortion from idealized TPP geometry follows the trend observed in the dichloromethane extraction with 4-5, where Yb(III) is preferentially rem oved over Ce(III) and other light lanthanides, with a separation factor of SFYb/Ce=450. At this stage of investigation, however, the conclusive determination of the origin of such selectivity in the solution is premature. As revealed in the extrac tion experiments described a bove, the distance separating the metal binding groups from the triphenoxymet hane platform had virtually no influence

PAGE 109

93 on the selectivity or binding efficiency of the ligands. Therefore, Eu(III) complex with the “more flexible” ligand 4-6 has been synthesized and the metal coordination environment has been compared to the binding site of cerium and ytterbium complexes to verify anticipated similarities. The structural analysis of the europium complex revealed that the Eu is bound by the ligand with simila r strength as Ce(III) and Yb(III) by ligand 4-5. The bond strength sequence in these comple xes follows the trend in the ionic crystal radius (for CN=9, rCe(III)=1.336 , rEu(III)=1.260 , rYb(III)=1.182 ), and is consistent with an increase in the charge density of metal ions: Ce(III)
PAGE 110

94 shorted than in the case of [4-5Yb]3+ (2.296(7) and 2.313(3) respectively) while the mean length of the etheric O-M bonds is slightly longer (2.447(7) and 2.429(2) ). Figure 4-8. Diagram of ytterbium encapsula ted by a cage-like derivative of tris-DGA compound. 4.2.4 Solution Structure of Extracted Species Given the ability of one equivalent of 4-5 to extract ~70% of the heaviest lanthanides from the acidic layer, the reacti on can be easily monitored when deuterated dichloromethane is used for the extracti on experiment. Figure 4-8 presents three superimposed 1H NMR spectra of the ligand 4-5 equilibrated with 1 M nitric acid (a), the extracted species generated in upon comple xation of Lu(III) by ligand 4-5 in the 1 M nitric acid/D2-dichloromethane extraction system (b), and the Lu(NO3)3 complex with 4-5 formed directly in D2-dichloromethane (c ). The chemical shifts of protons in the

PAGE 111

95 close proximity to the coordination site si gnificantly change upon formation of complex. Protons that belong to the C H2C H2 tripod-DGA linker become more distinct. Figure 4-9. Superimposed 1H NMR spectra of ligand 4-5 equilibrated with 1 M nitric acid (a), Lu(III) extracted by ligand 4-5 from 1 M nitric acid into the D2-dichloromethane (b) and Lu(III) complex with 4-5 formed in D2-dichloromethane (c). Complexation of metal through amidic a nd etheric oxygens deshields the surrounding protons and results in noticeable downfield ch ange in their chemical shifts. Without a metal, the 1H NMR spectrum of 4-5 exhibits two sharp singlets for the DGA C H2OC H2 protons and these signals broaden, merge, a nd shift downfield when a metal is added. The resonances for the protons of secondary amides are the most affected by metal complexation and they shift from a value of 8.15 ppm for 4-5 to 9.75 ppm for the Lu(III) complex of 4-5. From a comparison of the spectrum of the extracted species and the isolated Lu(III) complex of 4-5 (structure completely analogous to the Yb(III) complex presented in Figure 4-7), the only significant difference is upfield shift of the amidic protons in the Lu-tris-DGA speci es extracted from the acid layer, and the presence of a significant amount of water and acid may influence the chemi cal shift of these protons

PAGE 112

96 causing a less significant change in their position relative to the metal complex formed in dry dichloromethane. Nevertheless, the spectrum demonstrates that a C3-symmetric complex is formed, and all three arms of th e ligand simultaneously bind the metal center during the extraction event. More over, the central methine hydrogen (C H ), aromatic (7.07 and 7.24 ppm) and C H2OC H2 regions in the b spectrum are completely superimposable with spectrum c suggesting the structures of tw o species are nearly if not completely identical. In agreement with the extraction data at a 1:1 metal to ligand ratio highlighted in Figure 4-2, some residual amount of a free ligand is evident in spectrum b. This simple experiment highlights that th e extraction event invol ves a single ligand binding to the Lu(III) center, and undoubtedly, the large increase in entropy produced by this process contributes to th e high extraction efficiency. Often, the removal or “stripping” of the metal center from highly efficient extractant molecules is problematic, but in th e case of 4-5, a single contact with either weak acidic solution or pure water is suffici ent to completely remove the metal center from the ligand. When a solution of the Lu(I II) extracted complex with 4-5 (Spectrum b in Figure 4-8) was shaken with 0.01M HNO3, all of the metal ions were released from 4-5 into the aqueous layer. Although this experi ment indicates that the metal ions can be easily extracted back from the organic to th e aqueous phase using weak acid, the most effective concentration of n itric acid for more practical stripping still needs to be determined. 4.2.5 Importance of the Etheric Oxyge n of Tris-DGA in Metal Binding The separation of An(III) from the chemi cally similar Ln(III) has been the most challenging task in nuclear waste partitioni ng. As is the case of most hard donor extractants, compound 4-5 is not particularly effective in discrimination of these two

PAGE 113

97 groups of elements. The separation factor of 6.16 obtained for trivalent europium and americium in dichloromethane was only sligh tly higher in comparison to the literature data for simple DGAs in chlorinated so lvents (e.g. 0.1 M N,N,NÂ’,NÂ’-tetraoctyl-4oxapenanediamide: 3.43 in dichloroethane, 0.81 in chloroform).129 Most successful extractants for lanthanide(III)/actinide(III) se parations contain soft donor atoms for metal binding,95-97,99,120,130,151-153 and the central oxygen atom in the DGA ligand can be replaced with a softer sulfur atom produc ing a thiodiglycolamide (TDGA). Based on a comparison of the extraction ability of TD GA and the related glutalamide (GLA) ligand with a CH2 group in place of the centr al oxygen, it appears that the central sulfur atom in TDGA may interact with the Am(III) center in a 1M NaClO4/nitrobenzene extraction system (pH = 3).129 Since the size and flexibility of the group linking the amide donors is very similar, the groups would form a n early identical 8-membered chelate with americium if the sulfur atom does not interact with the metal. The much higher extraction efficiency of the TDGA with respec t to the GLA suggests otherwise. At pH > 3 and/or in the absence of a synergistic ag ent (e.g., thenoyltrifluor oacetone) however, the thiodiglycolamides (TDGA) lose their ability to bind efficiently Am(III). Given that the mutual arrangement of ligating TDGA units could enhance binding much like the DGA derivative 4-5, a thiodiglycolamide derivativ e (4-8) was synthesized, and its extraction potential was tested on a series of trivalent lanthanides including 152Eu and actinide 241Am. It was anticipated that the softness of the sulfur as well as the significant difference in the size of sulfur and oxygen (atomic radius: S = 1.27 , O = 0.65 ) would strongly affect the interacti ons between the ligand and the metal, and possibly help differentiate trivalent lanthanides and actinides In spite of the a dvantageous entropic

PAGE 114

98 effect, experiments with the preorganized thiodiglycolamides showed negligibly low distribution coefficients for both la nthanides and americium in 1 M HNO3 (10-3M of 4-8 in dichloromethane DAm(III) < 0.001) suggesting a fundamental requirement of the etheric oxygen for metal binding under th ese particular conditions. The structural studies of N,NÂ’-dimethyl -N,NÂ’-diphenyl-diglycolamide complexes with Ln(III) in solution system via EXAFS (e xtended X-ray fine structure) spectroscopy revealed participation of the et heric oxygens in the metal binding.124 Consequently, the structure of tris-DGA complexes with lantha nides both in the solid state and in the solution could be very similar. If the ethe ric oxygen is involved in metal binding in the extracted species, replacing it with much bigger sulfur atom would cause a severe distortion from TTP geometry, and ultimately result in poor metal binding. In view of the crowded environment of the nonadentate ligand 4-5 and the inability of the ligand arms to rotate once placed on the triphenoxymethane platform, the extracted species would likely interact with several of th e donor groups presented by the ligand, and the ether oxygens appear to be very important in this event. 4.3 Conclusions New tripodal chelates bearing three diglycolamide and thiodiglycolamide units precisely arranged on a triphenoxymethane platfo rm have been synthesized to provide for highly efficient and selective extraction of trivalent f-elem ent cations from nitric acid media. Exploiting the preference of the me tal center for TPP geometry, the ligand has been designed to completely fill the coordination sphere of the metal and 1H NMR experiments suggest a single ligand binds th e smaller lanthanides during the extraction event. The ligand with n-butyl substituents on the diglycolamide arms was found to be a significantly weaker extractant in comparison to the di-isopropyl anal ogs. The distance

PAGE 115

99 separating the metal binding groups from th e triphenoxymethane platform had little influence on the selectivit y or binding efficiency of the ligands. The tristhiodiglycolamide derivative proved to be an in effective chelate for f-elements in the 1 M nitric acid extraction system and demonstrat ed the importance of the etheric oxygens in the metal binding. 4.4 Experimental Section 4.4.1 General Considerations The lanthanide and actinide salts were us ed as received. The solutions were prepared from 18M Millipore deionized water, TraceMetal grade HNO3 (Fisher Scientific), and HPLC grade dichloromethane (Fisher Scientific). A Varian Cary 50 UV/Vis spectrophotometer was used for the Arsenazo(III) assays. Elemental analyses were performed at the in-house facility of De partment of Chemistry at University of Florida. The 1H, 13C, and spectra were recorded on a Varian VXR-300 or Mercury-300 spectrometer at 299.95 and 75.4 MHz for the prot on and carbon channels respectively. Mass spectrometry samples were analyzed on a Bruker Apex II 4.7T Fourier transform ion cyclotron resonance mass spectrometer. 4.4.2 1H NMR Experiment Equal volumes of 10-2 M solutions of Lu(NO3)3 in 1 M nitric acid and Tris-DGA (4-5) in D2-dichloromethane were mixed for 2h. The second portion of the same 10-2 M solution of 4-5 in D2-dichloromethane was mixed with 1 M nitric acid. After phase separation, the organic layers were analyzed in the 1H NMR experiment. Another NMR sample of the Lu(III) complex with 4-5 was synthesized di rectly in the deuterated solvent. The 10-5 mol of Lu(NO3)35H2O salt was stirred with 1 mL of the 10-2 M solution of 4-5 in D2-dichloromethane for 2h. Even though the complex partially

PAGE 116

100 precipitated from the solution, the concentration of the sample was sufficient for the NMR analysis. 4.4.3 Synthetic Procedures [(Diisopropylcarbamoyl)-methoxy]-acetic acid 4-1. A mixture of 6.09 mL (43.00 mmol) of diisopropyl amine and 3.50 mL (0.043) of pyr idine was slowly added to a solution of 5.15 g (43.00 mmol) of diglycolic anhydride in 40 mL of 1,4dioxane, under an ice bath condition. After stirring the reaction mixture for approximately 20 h at room temperature, solvent was evaporated under re duced pressure, and 6 M hydrochloric acid was added. The organic phase was further ex tracted with dichloromethane, dried over magnesium sulfate, and partially evaporat ed. Clean product cr ystallized upon slow solvent evaporation. Yield: 4.50g (48 %). 1H NMR (CDCl3): = 1.26 (d, J = 6.7 Hz, 6H; NCHC H3), 1.38 (d, J =6.9 Hz, 6H; NCHC H3), 3.40 – 3.63 (two multiplets, 2H; NC H CH3), 4.10 (s, 2H; C H2CON), 4.30 (s, 2H; C H2COOH). 13C NMR (CDCl3): = 20.1, 20.3 (aliphatic); 46.7, 47.8 (CH-N); 72.0, 72.4 (O-CH2); 169.9, 172.3 (C=O). HR LSIMS [M+H]+ = 218.1392, (Theoretical [M+H]+ = 218.1392). 2-[(Diisopropylcarbamoyl)-methoxy]-N,N -diisopropyl-acetamide 4-2. A cold solution of mono-substituted amide 41 (2.00 g, 9.20 mmol) in 40 mL of dry dichloromethane was gradually treated with oxalyl chloride (1.46 mL, 18.40 mmol). After 4 h solvent was evaporated and the oran ge sticky residue was dissolved in 20 mL of dioxane. Subsequently, 6.45 mL (46.00 mmol) of diisopropyl am ine was slowly added to the solution, and stirred overnig ht at room temperature. Th e solvent was evaporated, and the residue was dissolved in a mixture of di ethyl ether and pentane, extracted with 1 M hydrochloric acid followed by the extraction with 1 M sodium hydroxide, and died over

PAGE 117

101 magnesium sulfide. Pure pr oduct crystallized up on slow evaporation of solvent with 43 % yield (1.10g). 1H NMR (CDCl3): = 1.18 (d, J = 6.4 Hz, 12H; NCHC H3), 1.40 (d, J = 5.4 Hz, 12H; NCHC H3), 3.44 (m, 2H; NC H CH3), 3.90 (m, 2H; NC H CH3), 4.22 (s, 4H; OC H2). 13C NMR (CDCl3): = 20.2, 20.5 (CH3); 46.0, 48.0 (CH-NH); 70.4 (OCH2); 168.1 (C=O). HR LSIMS m/z [M+H]+ = 301.2487, (Theoretical m/z [M+H]+ = 301.2491). 2-(2-(diisopropylamino)-2-oxoethylthio)aceti c acid (4-4). The methodology for the preparation of 4-4 followed the s ynthetic pathways reported for 4-1140 and 4-3138. A mixture of 3.54 mL (25.00 mm ol) of diisopropyl amine and 1.00 mL (0.18 mol) of pyridine was slowly added to a soluti on of 3.00 g (23.00 mmol) of 1,4-oxathiane-2,6dione in 40 mL of 1, 4-dioxane at 0C. After stirring the reaction mixture for approximately 20 h at room temperature, the solvent was evaporated under reduced pressure, and 3 M hydrochloric acid was adde d. The organic phase was further extracted with chloroform, dried over magnesium sulfat e, and partially evaporated. The product crystallized upon slow evaporation of the solvent to afford 1.50 g (30%) of product. 1H NMR (CDCl3): = 1.12 (d, J = 6.7 Hz, 6H; NCHC H3), 1.26 (d, J =6.9 Hz, 6H; NCHC H3), 3.25 (s, 2H; C H2CON), 3.37 (m, 1H; NC H CH3), 3.42 (s, 2H; C H2COOH), 3.89 (m, 1H; NC H CH3). 13C NMR (CDCl3): = 20.1, 20.4, 33.5, 34.5, 46.4, 50.1 (aliphatic); 168.8, 171.6 (C=O). HR LSIMS [M+H]+ = 234.1172, (Theoretical LSIMS [M+H]+ =234.1143). Compound 4-5. Method I. To a mixture of 0.85 g (7.12 mmol) of 2mercaptothiazoline, DCC (1.47g, 7.12 mmol) and DMAP (0.11, 0.89 mmol) in 60 mL of dichloromethane (DCM), 1.55g (7.12 mmol) of mono-substituted oxa -pentaneamide 4-1

PAGE 118

102 was added and stirred for 5 h. Subsequentl y, 1.50 g (1.78 mmol) of amine 2-4a dissolved in 10 mL of DCM was added dropwise, and the solution was stirred for 48 h. White precipitate was filtered away, and solvent wa s evaporated in vacuo. The residue was treated with diethyl ether followed by additi on of pentane. The crashed out solid was filtered away and remaining solution of produc t was evaporated. The crude material was dissolved in mixture of DCM and hexamethyl disiloxane and left for crystallization. Yield: 1.50 g (59 %). Method II. A mixture of 1.0 equiva lence of mono-substituted amide (4-1), 2.0 eq. of ethyl-diisopropyl-a mine, and 1.1 eq. PyBOP (Benzotriazole-1-yloxy-trispyrrolidinophosphonium hexafluor ophosphate) was stirred in DMF for approximately 30 min. Subsequently, 0.3 eq. of amine (2-4a) was added and stirred for 48h. Upon following treatment with 10% hydroc hloric acid, white solid crashed out of solution. A precipitate was ex tracted with diethyl ether, and the organic solution was further washed with 0.5 M sodium hydroxide and dried over magne sium sulfide. A solvent was evaporated under reduced pressu re leaving clean light yellowish product in 73% yield. 1H NMR (CDCl3): = 0.52 (m, 18H; CH2C H3), 1.13 (s, 18H; CC H3), 1.20 (d, J = 5.9 Hz, 18H; NCHC H3), 1.31 (s, 18H; CC H3), 1.39 (d, J = 6.1 Hz, 18H; NCHC H3), 1.20 – 1.70 (two broad multiplets, 6H + 6H; C H2CH3), 3.20 – 3.90 (broad multiplets: 6H, CH2C H2-NH2; 6H, NC H CH3; 6H, O-C H2CH2), 4.10 (s, 6H; OC H2), 4.22 (s, 6H; OC H2), 6.43 (s, 1H; C H ), 7.02 (s, 3 H; ArH ), 7.11 (s, 3 H; ArH ), 7.89 (bt, 3 H; NH ). 13C NMR (CDCl3): = 9.2, 9.7, 20.6, 21.0, 28.9, 29.7, 35.4, 37.0, 37.7, 38.6, 39.3, 39.5, 46.2, 47.8 (aliphatic); 70.2, 71.1 (O-CH2); 125.0, 127.9, 137.8, 139.9, 142.9, 153.0 (aromatic); 167.6, 170.2 (C=O). HR ESI-ICR MS [M+K+H]2+ = 739.5151, (Theoretical

PAGE 119

103 m/z [M+K+H]2+ = 739.5200). Anal. Found: C, 71.3; H, 10.2; N, 5.7. Calc. for C85H142N6O12: C, 70.9; H, 9.9; N, 5.8 %. Compound 4-6. A mixture of mono-substituted amide (4-1) (1.27 g, 5.87 mmol), ethyl-diisopropyl-amine (1.94 mL, 11.73 mm ol), and PyBOP (4.35 g, 6.45 mmol) was stirred in 40 mL of DMF for approximately 30 min. A 1.56 g (1.76 mmol) portion of amine 2-9b was added and stirred for 48 h. U pon treatment with 10% hydrochloric acid, a yellow solid precipitate d from the solution. The solid was extracted with diethyl ether, and the organic solution was further wash ed with 0.5 M sodium hydroxide, and dried over magnesium sulfate. The solvent was evaporated under reduced pressure leaving 1.90 g (73%) of clean, light yellow product. 1H NMR (CDCl3): = 0.51 (m, 18H; CH2C H3), 1.12 (s, 18H; CC H3), 1.19 (d, J =6.7 Hz, 18H; NCHC H3), 1.30 (s, 18H; CC H3), 1.38 (d, J = 6.7 Hz, 18H; NCHC H3), 1.67 (q, J =7.6 Hz, 6H; C H2CH3), 2,00 (b, 6 H; CH2C H2CH2), 3.47 (broad multiplet: 18 H; O-C H2CH2C H2-NHCO, 3H; NC H CH3), 3.77 (m, 3H; NC H CH3), 4.07 (s, 6H; OC H2), 4.20 (s, 6H; OC H2), 6.34 (s, 1H; C H ), 6.99 (d, J = 2.0 Hz, 3 H; ArH ), 7.04 (d, J = 2.0 Hz, 3 H; ArH ), 7.84 (bt, 3 H; NH ). 13C NMR (CDCl3): = 9.2, 9.7, 20.6, 20.8, 28.7, 29.6, 30.5, 35.2, 36.8, 37.0, 37.7, 38.9, 39.2, 46.1, 47.7, 70.0, 70.9, 71.1 (aliphatic); 124.7, 127.9, 138.0, 139.7, 142.4, 153.4 (aromatic); 167.6, 169.9 (C=O). HR ESI-ICR MS m/z [M+H+K]2+ = 760.5481, (Theoretical m/z [M+H+K]2+ = 760.5430). Anal. Calcd for C88H148N6O12: C, 71.31; H, 10.06; N, 5.67. Found: C, 71.68; H, 10.30; N, 5.61. Compound 4-7. A mixture of mono-substituted amide (4-3) (1.46 g, 5.95 mmol), ethyl-diisopropyl-amine (1.96 mL, 11.84 mm ol), and PyBOP (3.90 g, 7.49 mmol) was stirred in 40 mL of DMF fo r approximately 30 min. Subsequently, 1.50 g (1.78 mmol)

PAGE 120

104 of amine (2-4a) was added and stirred for 48 h. Upon treatment with 10% hydrochloric acid, a white solid precipitated from the solu tion. The solid was extracted with diethyl ether, and the organic solution was furthe r washed with 0.5 M sodium hydroxide, and dried over magnesium sulfate. The solvent was evaporated under reduced pressure to yield 2.00 g (74%) of product. 1H NMR (CDCl3): = 0.47 (m, 18H; CH2C H3), 0.80 – 1.70 (signals: 18H, NCH2CH2CH2C H3; 18H + 18H, CC H3; 12H, NCH2CH2C H2CH3; 12H, NCH2C H2CH2CH3; 6H + 6H, C H2CH3), 3.12 (t, J = 7.4 Hz, 6H, NC H2CH2CH2CH3), 3.27 (t: J = 7.2 Hz, 6H, NC H2CH2CH2CH3), 3.10 – 3.80 (broad signals: 6H, CH2C H2-NH2; 6H, O-C H2CH2), 4.06 (s, 6H; OC H2), 4.24 (s, 6H; OC H2), 6.37 (s, 1H; C H ), 6.98 (s, 3 H; ArH ), 7.06 (s, 3 H; ArH ), 7.86 (t, J = 5.7 Hz, 3 H; NH ). 13C NMR (CDCl3): = 9.3, 9.7, 13.9, 14.0, 20.2, 20.4, 29.7, 29.9, 31.2, 35.4, 37.0, 37.8, 38.6, 39.2, 39.5 (aliphatic); 45.8, 46.7, 69.8, 70.3 (O-CH2-CO); 71.4 (O-CH2); 124.9, 127.8, 137.8, 139.9, 142.8, 153.1 (aromatic), 168.3, 170.0 (C=O). HR LSIMS [M+H]+ = 1524.1699 (Theoretical LSIMS [M+H]+ = 1524.1703). Anal. Calcd for C91H154N6O12: C, 71.71; H, 10.18; N, 5.51. Found: C, 71.99; H, 10.35; N, 5.45 Compound 4-8. A mixture of mono-substituted amide 4-4 (1.79 g, 7.67 mmol), ethyl-diisopropyl-amine (2.54 mL, 15.37 mm ol), and PyBOP (4.40 g, 8.46 mmol) was stirred in 40 mL of DMF fo r approximately 30 min. Subsequently, 1.95 g (2.31 mmol) of amine (2-4a) was added and stirred for 48 h. Upon treatment with 10% hydrochloric acid, a white solid precipitated from the solu tion. A solid was collected, extracted with diethyl ether, and the organi c solution was further washed with 0.5 M sodium hydroxide, and dried over magnesium sulfide. The so lvent was evaporated under reduced pressure leaving 2.90 g (84%) of light yellow product. 1H NMR (CDCl3): = 0.51 (m, 18H;

PAGE 121

105 CH2C H3), 1.12 (s, 18H; CC H3), 1.21 (d, J =6.4 Hz, 18H; NCHC H3), 1.31 (s, 18H; CC H3), 1.37 (d, J = 6.2 Hz, 18H; NCHC H3), 1.20 – 1.70 (two broad multiplets, 6H + 6H; C H2CH3), 3.27 –4.10 (signals: 6H, O-C H2CH2; 6H, OC H2; 6H, OC H2; 6H, CH2C H2NH2; 6H, NC H CH3), 6.43 (s, 1H; C H ), 7.01 (s, 3 H; ArH ), 7.07 (s, 3 H; ArH ), 7.94 (bt, 3 H; NH ). 13C NMR (CDCl3): = 9.3, 9.7, 20.6, 20.9, 28.7, 29.7, 35.3, 35.5, 36.1, 37.0, 38.7, 39.3, 40.2, 46.2, 49, 7, 49, 9 (aliphatic); 70.3 (O-CH2); 125.0, 127.9, 137.8, 140.0, 142.9, 153.2 (aromatic); 167.7, 169.7 (C=O). HR ESI-ICR MS m/z [M+Na]+ = 1509.9872 (Theoretical m/z [M+Na]+ = 1509.9893). Anal. Calcd for C85H142N6O9S3: C, 68.60; H, 9.62; N, 5.65, Found: C, 68.94; H, 9.99; N, 5.51. General procedure for preparation of comp lexes. A solution of 0.7 equivalent of Ln(NO3)3 xH2O in methanol was added to a solution of 1 equivalent of 7 in methanol, and mixture was stirred at room temperatur e for approximately 4 h. The Ce complex, was formed in reaction of (NH4)2Ce(NO3)6 and ligand 4-5. Due to instability of Ce(IV) in the organic solvents, the crystalline material collected after diffusion of ether into the reaction mixture in methanol contained complex of Ce(III) exclusively. Upon slow diffusion of ether to a metha nol solution of [4-5Yb](NO3)3 complex, clear crystals were formed with 40% yield. Crystals of Eu co mplex with 4-6 and Yb-cage were obtained via diffusion of ether into the concentrated solution of the complex in methanol and dichloromethane (1:1). The PF6 anions found in [4-5Eu](NO3)3 complex were used in the synthesis of ligand 4-5. Compound [4-5 Ce][Ce(NO3)6]. HR ESI-ICR MS m/z [M+Ce(NO3)2]+ = 1702.9413, (Theoretical m/z [M+ Ce(NO3)2]+ =1702.9491). Compound [4-5 Yb](NO3)3. HR ESI-ICR MS m/z [M+Yb(NO3)2]+ =1736.9813, (Theoretical m/z [M+ Yb(NO3)2] + = 1736.9842).

PAGE 122

106Table 4-4. X-ray data for the crys tal structures of[4-5Ce][Ce(NO3)6], [4-6Eu](NO3)3, [3x4-2Yb](NO3)3, [4-5Yb](NO3)3 and [Yb-cage](NO3)3. [4-5Ce] [Ce(NO3)6] 4MeOHC4H10O [4-6Eu](NO3)2.5 MeOH1/2PF6 [3x4-2Yb] (NO3)3 [4-5Yb](NO3)32MeOH [Yb-cage](NO3)3 total reflections unique reflections max( ) empirical formula Mr crystal system space group a () b () c () ( ) ( ) ( ) Vc (3) Dc (g cm-3) T (K) Z (MoK ) (mm-1) R1 [ I 2 ( I ) data]b w R2 (all data)c GoF 33661 22398 28 C93H168Ce2N12O35 2294.63 monoclinic Pc 19.075(2) 13.2833(14) 24.916(3) 90 110.643(2) 90 5907.8(11) 1.290 173(2) 2 0.838 0.0761 [12522] 0.1852 0.924 24923 17264 25.61 C89H152EuF3N8.5O20.5P0.51907.66 triclinic P1 12.2719(14) 12.4189(14) 41.104(5) 85.623(2) 88.434(2) 61.098(2) 5467.8(11) 1.159 273(2) 2 0.6448 0.0578 [15717] 0.1406 1.100 38856 6165 25 C48H96N9O18Yb 1260.38 tetragonal P42/n 16.9828(7) 16.9828(7) 24.3051(15) 90 90 90 7010.0(6) 1.194 173(2) 4 1.396 0.0706 [4059] 0.1993 1.111 69932 24823 28 C87H150N9O23Yb 1863.20 monoclinic C2/c 71.172(5) 12.8866(9) 23.3365(16) 90 94.489(1) 90 21338(3) 1.160 173(2) 8 0.942 0.0480 [18157] 0.1471 1.066 36955 22802 24.25 C122H188N9O24Yb 2337.85 triclinic P1 12.3923(8) 22.0965(14) 27.4858(17) 71.9520(10) 82.3300(10) 89.9410(10) 7085.5(8) 1.096 273(2) 2 0.722 0.1016 [15923] 0.2840 1.648 aObtained with monochromatic Mo-K radiation ( = 0.71073 ) bR1 = Fo Fc / Fo cw R2 = { [ w ( Fo 2 Fc 2)2/ [ w ( Fo 2)2]}1/2

PAGE 123

107 CHAPTER 5 PYRIDINE N-OXIDE FUNCTIONALIZED C3-SYMMETRIC CHELATES FOR F-ELEMENS BINDING 5.1 Introduction The phosphorous oxide derivatives are among the most successful extractants for lanthanide and actinide separations. It th erefore seems reasonable to expect similar extraction properties from oxides of other el ements close to phosphor us in the periodic table.45 Indeed ketones, silicon oxides, sulphoxi des and arsine oxides have been studied as extraction solvents. Some of these system s, however, suffer practical drawbacks. The ease of hydrolysis of silicon oxides renders them unsuitable for highly acidic f-element separations, while the applic ation of arsine oxides is limited by their challenging synthesis.154-156 Only ketones41 and sulphoxides157-160 have found limited success in this field. Surprisingly, except for studies on the extraction potential of trioctylamine and 5(4-pyridyl)nonane N-oxides by Ejaz,45,46,161-171 little attention has been paid to the amine oxides.172-174 5-(4-pyridyl)nonane N-oxide trioctyl amine oxide N O N O Figure 5-1. Structures of the most extensively studied amine oxides.

PAGE 124

108 The excellent donor ability and stability of the N-oxide group in acidic media makes N-oxide derivatives promising reagents for An/Ln extractions. Additionally, the incinerability of these organic molecules is in accord with minimization of waste generated during nuclear waste treatment. Although aliphatic amine oxides attract meta l cations much more strongly than their aromatic counterparts, th eir application as partitioning r eagents is severely limited. Due to their high polarity, aliphatic amine oxi des effectively displace water molecules from the coordination sphere of cations and, as such, they are not particularly selective in cation binding. The different binding propertie s of aromatic amine oxides arise from the distribution of their electron density. In pyridine N-oxide, electrons can be delocalized over the aromatic ring, while in the alipha tic case the negative dipole is positioned directly on the oxygen resulting in higher di pole moment (trioctylamine oxide: 5.4 D, pyridine N-oxide: 4.2 D, and for comparison TBP: 3.05 D).163 N O 0.912 0.902 1.142 1.384 1.616 Figure 5-2. The electron dist ribution in pyridine N-oxide.175-178 The electronic configuration of pyrid ine N-oxide can be represented by the following resonance structures:179 N N N N N O O O O O III IIIIV V Figure 5-3. Resonance struct ures of pyridine N-oxide adapted from Ochiai, E. J. Org. Chem. 1953 18 534.179

PAGE 125

109 As a result of delocalization of electrons in resonance structures II-IV, the dipole moment of pyridine N-oxide is lower than predicted ( = 4.24 rather than 4.38 D).180 In fact, pyridine N-oxide is a quite soft oxygen donor and can form stable complexes not only with mercury (II) [Hg(C5H5NO)6](C1O4)2,181 but also with mercury (I) Hg2(C5H5NO)4(C1O4)2.182 Therefore pyridine N-oxide deri vatives could be successful in the problematic separation of trival ent actinides and lanthanides. Interestingly, in the liquid-liquid extr action, the pyridine N-oxide derivative 5-(4-pyridyl)nonane N-oxide shows very little affinity for trivalent f-elements over a whole range of acidity ( DCe(III) ~ 0.001 in solution of 0.1M ligand xylene).170 The compound seems to prefer intera ctions with more highly oxidized ions. For comparison, in the same extraction setting the distribution coefficient obtained for cerium (IV) was close to unity, and a relatively small dependen ce of the extraction efficiency on the acid concentration (~ D 0.15) was observed with a minimum at approximately 0.1 M and the maximum at 7 M HNO3.169 On the other hand, the extracti on efficiencies of other highly oxidized ions such as Th(IV) and U (VI) were strongly affected by acid concentration. The most effective, quantita tive extraction of Th(IV) wa s afforded by 0.1 M pyridine Noxide in xylene in the ra nge of 0.1 0.5M nitric acid.45 At higher acid concentration the extraction efficiency was signi ficantly decreased, possibly due to strong interactions between the extractant and acid, as well as th e formation of anionic thorium hexanitrate complex which is impossible to extract by Noxide. A similar situation was found in the case of hexavalent uranium, where the dist ribution coefficient reached its maximum at about 0.1 – 0.5 M HNO3 ( DU(VI)~10) and abruptly decr eased with higher acid concentration (at 1 M HNO3 D ~8, and at 10 M D ~0.1).46 As opposed to the aliphatic

PAGE 126

110 amine oxides, this relatively weaker aromatic base may not be able to adapt the ion pair mechanism to restore efficient metal binding in very acidic environments. As a function of polarity, higher basicity of aliphatic amine oxides causes stronger attraction of H+ which may initially lower the extraction e fficiency at the moderate to high acid concentration (pKa of conjuga ted acids of heterocyclic a nd aliphatic amines oxides are ~1.0 and ~5.2 respectively; for comparison pKa of TBP168 ~ 0.20).163 However, in highly acidic solution some aliphatic amine oxides can adapt an ion pair mechanism where the protonated form of ligand attracts negativel y charged metal complex if present (e. g. [R3NOH]+[UO2(NO3)3]-) and facilitate the transferred of a metal into the or ganic phase. Intrigued by the binding pr operties of heterocyclic amine N-oxides and the vast possibilities of structural m odifications that promise cont rol over the binding potential, tris-pyridine N-oxide ligands have been s ynthesized. The ligand design was inspired by our prior success in enhancement of the effici ency and selectivity of small extractants via attachment of three binding units onto the triphenoxymethane scaffold.51,140,183 The designed ligand could be attract ive for trivalent actinides in the liquid-liquid extraction systems not only through the entropic effect, but also by the chelate effect due to a second type of oxygen donor (amide) in cl ose proximity to each N-oxide. Although similar donor properties of pyridine N-oxi de (PyNO) and triphenyl phophine oxide (dipole moments of 4.24 D and 4.28 D respectively)180,184 predict moderate to low affinity for trivalent f-elements, it was envi sioned that the lower steric hindrance around the N-oxide donor atom may facilitate better binding. The selectivity for An(III) was expected to arise from the fairly so ft character of the aromatic N-oxide.

PAGE 127

111 5.2 Results and Discussion 5.2.1 Synthesis of Tris-PyNO Derivatives R R O HN O N O 3 R R O HN O N 3 R R O H2N n 3 O N Cl Et3N/THF or pyridine/dioxane 3-chloroperoxybenzoic acid CH2Cl25-1 R = t-Pentyl, n = 1 5-2 R = t-Pentyl, n = 2 5-3 R = t-Butyl, n = 2 2-4a R = t-Pentyl, n = 1 2-9a R = t-Pentyl, n = 2 2-9a R = t-Butyl, n = 2 5-4 R = t-Pentyl, n = 1 5-5 R = t-Pentyl, n = 2 5-6 R = t-Butyl, n = 2 n n Figure 5-4. Synthesis of tris-pyridine N-oxides. The synthetic methodology to obtain tris -pyridine N-oxides is summarized in Figure 5-4. In the first step the picolinamide derivatives were synt hesized by reaction of a primary amine185,186 with picolinic acid chloride in the presence of an organic base. The picolinic acid chloride was prepared fo llowing established procedures using either thionyl chloride186 or oxalyl chloride.186 According to the liter ature, the reaction of picolinic acid and thionyl chloride yields mixtures of compounds, most likely dimers, polymers and cyclic hexamers of acid chlori de hydrochloride rath er than a monomeric nonprotonated acid chloride, which can adversel y affect the yield and purity of the reaction.187,188 Indeed, in the case of a compoun d 5-1 a cleaner and more efficient reaction was realized when th e nonprotonated acid chloride wa s used. Also, changing the base and solvent from triethylamine and TH F to pyridine and dioxane improved the ease of reaction workup. In the second step the picolinamide was treated in portions with 70-75% m -chloroperbenzoic acid. This synthetic procedure was adapted from methodology

PAGE 128

112 reported for the oxidation of various pyridine derivatives.181 After approximately 3 days of stirring the reaction mixtur e at room temperature, the dichloromethane solution was washed with saturated sodium hydrogen carbona te and brine, and dried over magnesium sulfate. Solvent was partially evaporated and slow addition of pentane resulted in the precipitation of analy tically pure product. 5.2.2 Extraction Experiments To allow the direct comparison of chelate extraction power, the preliminary extraction experiments were performed under the same conditions as with all other previously described ligands. Compo unds 5-4 and 5-6 were tested at 10-3 M concentration in CH2Cl2. The extraction of a series of tr ivalent lanthanide and tetravalent thorium nitrates from 1 M nitric acid revealed a low affinity of ligands for those ions. The extraction percentages obtained with the more flexible compound 5-6 (elongated arm) were slightly higher than for the 5-4, however the highest ex traction value did not exceed 11 3%. Low affinities for tested me tal ions may not necessarily reflect poor intrinsic ability of the ligands to bind, but may instead arise from low stability of complexes in dichloromethane, or simply fr om the protonation of lig ands in the acidic extraction environment. Since the goal has been to identify a liga nd that could extract ions typical for the nuclear waste environment, the binding potential of the tris-PyNO was not evaluated at higher pH. Also, limited solubility of the liga nd restricted extraction options to only few organic solvents. To improve ligand performance, a counterion more capable then nitrate to stabil ize the charged complex in the organic phase was utilized. Recently, significant attention has been focused on CObaltocarborane SANdwich anion (COSAN) as an extraction supporti ng agent. The hexachlorinated cobalt dicarbollide along with polyet hylene glycol and carbamoyl phosphine oxide (CMPO) in

PAGE 129

113 mononitrotrifluorotoluene has been used as a synergistic mixture of the UNEX (UNiversal EXtraction) proce ss, the one stage solvent desi gn for a simultaneous recovery of all long-lived radionuclid es from high level waste.189-192 The chlorinated derivative of bis(1,2-dicarbollide) cobaltate is more resistan t to warm concentrated nitric acid and to intensive radiation (Figure5-5). Cl BH CH BHCH B BH B BH B H B B H B HB BH B BBHHC BH CH BHCo BHCs+[Co(C2B9H8Cl3)2]Cs hexachlorinated cesium bis(1,2-dicarbollide) cobaltateCl Cl Cl Cl Cl Figure 5-5. Structure on the hexachlorinate d cobaltocarborane sandwich anion (COSAN). The acid form of cobalt dicarbollid e has properties of a superacid,193 namely a compound with acidity greater than that of 100 wt. % sulf uric acid. A solution of COSAN in organic solvent forms an ion pair [H+nH2O][COSAN-] with n = 5.5 upon contact with water.194 The proton can be exchanged for a metal cation, enabling COSAN to work as a cation exchanger. More importa nt is the unique capacity of COSAN to act as a lipophilic counterion that through compen sation of the charge of the complex can facilitate the extraction event of a metal to the organic phase. As opposed to the very polar nitrate, this hydrophobi c species can very effectiv ely stabilize the extracted

PAGE 130

114 complex in the organic solvent. Additi on of COSAN as a synergist has improved extraction coefficients of many extraction procedures by two to three orders of magnitude.195-201 According to the results of mo lecular dynamic studies by G. Wipff et al. the highly surface active COSAN anion cr eates a film on the water and organic solvent interface. This negatively charged la yer attracts hard cations normally repelled by the interface therefore assisting in th e binding of a metal ion by the primary extractant.202 The major weakness of this bulky c ounterion is its limited solubility in organic diluents to the environmentally hazardous nitrobenzene and halogenated hydrocarbons derivatives. Dichloroethane and 2-nitrophenyl octyl et her were chosen for extraction studies with synergistic mixture of tris-PyNO and COSAN. Initially 10-3 M concentration of 5-4 and three fold higher concentration of COSA N in 2-nitrophenyl octyl ether mixture was used for extraction of trivalent La, Eu and Yb nitrates (10-4 M) from 1 M nitric acid. Unfortunately, no improvement of extracti on was observed. Similar results were obtained in the extraction of eleven lanthanides with a solution of 10-3 M 5-4 in dichloroethane with either 10-3M or 3x10-3M of COSAN. Extraction with unoxidized pyridine derivative 5-1 at10-3 M with 3x10-3M COSAN in C2H4Cl2 was also unsuccessful. Since most of published extr action procedures do not describe any special conditioning of COSAN prior to the extraction, the compound was used in the form of cesium salt, as received. This form of COSAN may not be the most suitable for extraction. Conversion of COS AN salt into the acid via multi ple treatment with sulfuric acid as described by Smirnov et al .203,204 could prove to be more useful and experiments using converted compound ar e currently underway.

PAGE 131

115 NMR extraction studies with a solution of 0.01 M lutetium n itrate in 1 M nitric acid and a molar equivalent of 5-4 in deuterated dichloromethane confirmed the inability of 5-4 to transfer the meta l into the organic phase. Figure 5-6. Stacked 1H NMR spectra of ligand 5-4 equilibra ted with 1 M nitric acid (a), Lu(III) extraction by ligand 5-4 from 1 M nitric acid into the D2-dichloromethane (b) and Lu(III) complex with 5-4 formed in D2-dichloromethane (c). Spectrum a shows 10-2M solution of ligand 5-4 equilibr ated with 1M nitric acid. Upon contact with 10-2 M lutetium nitrate solution in 1 M nitric acid (spectrum b) absolutely no complex has been formed in the organic phase. Spectra a and b are virtually identical and show no signs of extraction of lutetium into deuterated dichloromethane. It was determined, howev er, that ligand 5-4 is capable of binding lutetium, as documented in spectrum c. The complex of lutetium with tris-PyNO 5-4 was obtained via dissolution of 1 molar equivale nt of hydrated lutetium nitrate salt in 10-2 M solution of 5-4 in deuterated dichlorometh ane. Upon mixing, a substantial amount of

PAGE 132

116 complex precipitated out of solution, suggesti ng that dichloromethan e can neither solvate nor effectively stabilize the complex in the organic phase, preventing the extraction event. Fortunately, the concentration of so luble lutetium complex in the NMR solvent was sufficient for analysis. Protons from the two carbon spacer between the platform and the PyNO unit are shifted slightly upfield. The most significant difference in the ligand and the complex spectra can be observed in th e aromatic region. Resonances of the two protons in closest proximity to N-O and the amide are affected the most, and due to the deshilding effect of the me tal binding, they move ap art and shift downfield. Interestingly, amidic protons originally pr esent at 11.51 ppm (spectra a and b) vanished upon complexation. The NMR studies demonstrat e that ligand 5-4 is capable of metal binding, but apparently the acid concentra tion and instability of the complex in dichloromethane defer extraction. The binding ability of the unoxidized, soft er pyridine derivative 5-1 was also examined (Figure 5-7). As in the case of 5-4 no sign of Lu(III) tran sfer into the organic phase was observed (spectrum b). Signals from protons of the CH2CH2 linker split and move upfield. Again, as expected, the aromatic protons are most affected by complexation. The four multiplets of pyridin e shift downfield by approximately 0.1 ppm, and sensitive to the special or ientation protons of triphenoxymethane platform are shifted farther apart from each other. Amidic prot ons originally concealed by the pyridine peak at 8.53 ppm, shift upon complexation to 9.20 ppm.

PAGE 133

117 Figure 5-7. Stacked 1H NMR spectra of ligand 5-1 equilibra ted with 1 M nitric acid (a), Lu(III) extraction by ligand 5-1 from 1 M nitric acid into the D2dichloromethane (b) and Lu(III) co mplex with 5-4 formed in D2dichloromethane (c). Based on results of the NMR experiment it can be concluded that the soft donor 5-1 is also capable of metal binding. As opposed to Lu-PyNO, the complex of 5-1 is soluble in dichloromethane. Protonation of the liga nd is therefore the majo r reason for the ligand inability to bind and transport a metal ion into the organic phase. 5.2.3 Solid State Studies Pyridine N-oxide is a vers atile ligand capable of bindi ng a wide range of metal ions. The oxygen can donate elect rons from the highest occupied molecular orbital to the empty orbital of a metal ion of an appropriate symmetry, or act as a -electron density acceptor using the empty orbital of oxygen.175,176,205

PAGE 134

118 N O M N O M X X sp2 M-O-N angle ~120 sp3 M-O-N angle ~108 Figure 5-8. Two geometric extremes of metal binding by the substituted N-oxide. Figure adapted from Ragsdale, R. O. Coord. Chem. Rev. 1968, 3 375.178 The geometry around the oxygen atom the pyrid ine N-oxide is strongly influenced by the electronic configurati on of the metal and by substituents on the pyridyl ring.178 Figure 5-8 shows two extremes of metal bindi ng by pyridine N-oxide, but in most cases a compromised geometry around oxygen is obser ved, with M-O-N angles between 120 and 108. The analysis of the partially refined crys tal structure of ytterbium (III) complex of 5-4 revealed an unusual arrang ement around the metal center (Figure 5-9). The metal in the mono cationic complex is nine coordinate with two nitrates bound directly in the bidentate fashion. Interestingl y, only five of the six avai lable donor atoms of the ligand are involved in the binding. The less basi c carbonyl oxygen, rather than the N-oxide oxygen, is pointing out of the coordination s ite. The distances between the N-oxide oxygens of the bidentate bound arms and the ytte rbium are, as expected, shorter (2.239(6) and 2.296(7) ) than that in the monodentate complex (2.344(7) ). One of the two bidentate arms is bound more tightly to th e metal (one arm: NO-M 2.239(6) and CO-M 2.274(6), second arm: 2.296(7) and 2.351(7) respectively).

PAGE 135

119 Figure 5-9. Diagram of the coordi nation environment of [5-4Yb(NO3)2](NO3) (30% probability ellipsoids for Yb, N, a nd O atoms; carbon atoms drawn with arbitrary radii). The tr iphenoxymethane base, one nitrate and all hydrogen atoms have been removed for clarity. The distances between the oxi de and ytterbium as well as the N-O bond lengths are comparable with data published by Paine et al. for derivatives of phosphinopyridine N and P-oxide complexes.206 The charge of the complex is balanced by another nitrate connected to the complex via hydrogen bonding with one of the amidic protons of the bidentate coordinated arms (N-O 2.871 ). N O O 117.74 119.43 122.63N O O 117.30 119.60 123.10N O O 115.30 119.22 124.94 M M M 116.82 128.54 98.18 126.77 127.13 74.95 127.21 1.339 1.2592 2 9 6 2 3 5 1 1 4 6 0 1.334 1.2452 2 3 9 2 2 7 4 1 4 7 3 1.3411 2 1 1 1 4 5 0 2 3 4 4 Figure 5-10. Coordination environment of ytterbium (III) in the complex with 5-4.

PAGE 136

120 The N-O-M angle fits within the exp ected 108-120 range for only the more strongly bound arm. The other two arms bind under larger (1 26.7(3) and 127.2(3)) angles. This discrepancy may be caused by a steric constraint introduced by the bound nitrates, or by limited flexibility of the liga nd and its inability to adapt to the required conformation. In effect, one of the arms acts as a weaker monodentate binder. If similar species are formed during the extraction proces s, the structural constraint may directly affect the extraction of metal ion, especially in the highly competitive, water and nitrate rich extraction environment. Although the solid state structure may not reflect the composition of the complex in solution, it implies competition between the carbonyl oxygens and nitrites over metal binding. The formation of the six membered ring upon metal complexation with N-oxide and carbonyl oxygens should provide some stabilization to the extracted complex, but effective binding re quires a particular orientation of the oxygen donors on the arm, which seems to be impeded by the spatial arrangement of the two already bound N-oxide arms. In the absence of structural constraints one of the nitrates would need to bind in a monodentate fashion, to maintain the same coordination number in the comple xes which is rare and less energetically favored. Described coordination environment of the ytterbium ion will be verified upon the completion of crystallograp hic refinement of [5-4Yb(NO3)2](NO3) structure. 5.3 Conclusions Both the C3-symmertic pyridine N-oxide and un oxidized pyridine derivatives are capable of binding trivalent lanthanides but not in 1 M nitric acid/dichloromethane environment. The lack of extraction strength can be attributed to the combined effects of the protonation of ligand in a 1 M acid, instability of the complex in dichloromethane, as well as competition for metal binding in the highly polar extraction medium.

PAGE 137

121 5.4. Experimental Section The NMR extraction experiment was pe rformed as described in Chapter 4.183 Compounds 5-2, 5-3, and 5-6 were synt hesized by Dr. Priya Srinivasan. General procedures for the synthesis of picolinamides. Method A. A suspension of primary amin e and freshly prepared picolinic acid chloride hydrochloride in dry THF was treated with triethylamine, and the resultant deep blue colored mixture was stirred at room temperature for 1-3 days. Triethylamine hydrochloride was removed by filtration and so lvent was removed in vacuo. The residue was dissolved in dichloromethane and treate d with diluted acid and base to remove residual picolinic acid and triethylamine (1M HCl, followed by water, 1M NaOH and again water). The soluti on was then dried over MgSO4, filtered and concentrated. Slow diffusion of pentane into a diet hyl ether solution of crude product afforded clean material as a white precipitate. Method B. A suspension of freshly prepar ed picolinic acid chloride in dioxane was mixed with dioxane solution of primary amine. The mixture was tr eated with pyridine, and stirred at room temperature for approxi mately 20h. The solvent was removed and the residue was treated with diethy l ether. The resultant mixture was washed with 1M HCl, followed by water, 1M NaOH and again water. The ethereal solution was dried over MgSO4, filtered, and treated with activated charcoal for decolorization. Partial evaporation of ether and addi tion of pentane yielded the pur e product as a white solid. Compound 5-1. Method A. Using 2.0 g of primary amine 2-4a (2.37 mmol) and 3.0 g portion of picolinic acid chloride hydrochloride (16.90 mmol approximately, assuming no decomposition to picolinic acid ) in 100 mL of dry THF, and 9 mL of triethylamine (64.53 mmol) 1.4 g (51 %) of produ ct was obtained. Method B. Using

PAGE 138

122 2.0 g of primary amine 2-4a (2.37 mmol) in 10 mL of dioxane, 3.0 g portion of picolinic acid chloride (~21.00 mmol) in 20 mL of dioxane, and 13 mL of pyridine (12.71g, 0.16 mol) 2.1 g (77 %) of product was obtained. 1H NMR (CDCl3): = 0.52 (t, 18 H; ArCH2C H3), 1.16 (b s, 18 H; Ar-C-C H3), 1.28 (b s, 18 H; Ar-C-C H3), 1.49 (q, 6H; Ar-C-C H2), 1.56 (b q, 6H; Ar-C-C H2), 3.53 (b t, 6 H; N-C H2CH2), 3.86 (b t, 6 H; Ar-OC H2CH2), 6.58 (s, 1 H; C H ), 7.05 (s, 3 H; ArH ), 7.23 (s, 3 H; ArH ), 7.30 (t, 3 H; PyH ), 7.70 (t, 3 H; PyH ), 8.11 (d, 3 H; PyH ), 8.55 (b d 6 H; 3 H PyH and 3 H NH ) 13C NMR (CDCl3): = 9.30, 9.68, 29.43, 34.86, 37.09, 37.87, 38.57, 39.36, 40.06 (aliphatic); 70.43 (OC H2); 122.17, 125.08, 126.07, 128.07, 137.12, 137.87, 140.41, 143.09, 148.30, 150.07, 153.00 (aromatic); 164.49 (C=O ). HR EI MS m/z = 1156.7622 (Theoretical m/z = 1156.7704). Anal. Calcd for C73H100N6O6: C, 75.74; H, 8.71; N, 7.26, Found: C, 76.05; H, 9.09; N, 7.01. Slow diffusi on of pentane into a solution of 5-1 in diethyl ether afforded crystals suitable for X-ray analysis. Compound 5-2. Method A. Using 2.00 g of 2-9a (2.00 mmol), 1.96 g of picolyl chloride (14.10 mmol) and 6.0 mL of Et3N (4.03 g, 40.00 mmol) in 200 mL THF 1.89 g (70%) of product was obtained. 1H NMR (CDCl3): 0.53 (m, 18H; Ar-CH2C H3); 1.14 (s, 18H; Ar-C-C H3), 1.32 (s, 18H; Ar-C-C H3), 1.46 (q, 3JHH = 7.2 Hz, 6H; C H2CH3), 1.68 (q, 3JHH = 7.2 Hz, 6H; C H2CH3), 2.14 (br peak, 6H; NCH2C H2CH2O), 3.58 (m, 6H; NC H2CH2CH2O), 3.66 (m, 6H; NCH2CH2C H2O), 6.42 (s, 1H; C H ), 7.02 (m, 6H, ArH ), 7.30 (m, 3H; PyH ), 7.77 (m, 3H; PyH ), 8.21 (m, 3H; PyH ), 8.41 (m, 3H; PyH ), 8.45 (br s, 3H; N H D2O exchangeable,). 13C NMR (CDCl3): 9.3, 9.8, 28.2, 28.6, 30.7, 35.2, 37.1, 37.3, 37.8, 39.3 (aliphatic); 70.4 (O C H2); 122.4, 124.8, 126.0, 128.0, 137.2, 138.2,

PAGE 139

123 139.9, 142.4, 148.1, 150.5, 153.6 (aromatic); 164.6 (C=O). Anal. Calcd for C76H106N6O6: C, 76.09; H, 8.91; N, 7.01, Found: C, 76.40; H, 9.28; N, 6.79. Compound 5-3. Method A. Using 2.00 g of primary amine 2-9a (2.50 mmol) and 1.96 g picolyl chloride hydrochloride (14.10 mmol) in 200 mL of THF and 6.0 mL of Et3N (4.32 g, 42.70 mmol) 1.95 g (70%) of product was obtained. 1H NMR (CDCl3): 1.11 (s, 27H; C H3), 1.25 (s, 27H; C H3), 2.11 (t, 3JHH = 6.6 Hz, 6H; NCH2C H2CH2O); 3.55-3.66 ( m, 12H; NC H2CH2C H2O), 6.40 (s, 1H; C H ), 7.08 (s, 6H; ArH ), 7.25 (m, 3H; PyH ), 7.62 (m, 3H; PyH ), 8.13 (m, 3H; PyH ), 8.32 (m, 6H; PyH and N H D2O exchangeable). 13C NMR (CDCl3): 30.7, 31.7, 34.7, 35.7, 37.5 (aliphatic); 70.9 (O C H2); 122.4, 125.9, 127.3, 137.2, 148.1, 154.1 (aromatic), 165.2 (C=O). Anal. Calcd. for C70H94N6O6: C 75.37; H 8.49; N 7.53%. Found: C 74.89; H 8.70; N 7.33%. General procedures for the synt hesis of picolinamide N-oxides. A 0C solution of the primary amine in CH2Cl2 was treated with m-chloroperbenzoic acid in portions. The resu ltant orange-yellow solution was stirred at room temperature for 3 days. The yellow-pa le solution was subsequently washed with saturated NaHCO3, water and brine. The orga nic layer was dried over MgSO4 or Na2SO4, filtered and removed l eaving a pale-yellow solid. The product was further purified either by washing with diethyl et her or precipitation from a mixture of CH2Cl2/ether and pentane. Compound 5-4. A mixture of 2.00 g (1.73 mmol) of amine 5-1 and 2.00 g (70-75%) of m -chloroperbenzoic acid in 50 mL of CH2Cl2 afforded 1.20 g (58%) of pure product upon precipitation from Et2O/pentane. 1H NMR (CDCl3): = 0.52 (b m, 18 H; Ar-CH2C H3), 1.16 (b s, 18 H; Ar-C-C H3), 1.27 (b s, 18 H; Ar-C-C H3), 1.48 (b q, 6H;

PAGE 140

124 C H2C H3), 1.65 (b, 6H; C H2 C H3), 3.50 (b s, 6 H; N-C H2CH2), 3.83-3.97 (two broad peaks, 6 H; Ar-O-C H2CH2), 6.60 (s, 1 H; C H ), 7.02 (b s, 3 H; ArH ), 7.21 (b s, 3 H; ArH ), 7.27 (b m, 6 H; PyH ), 7.35 (m, 6 H; PyH ), 11.56 (b, 3 H; NH ) 13C NMR [CDCl3]: = 9.28, 9.66, 28.39, 29.36, 34.80, 37.07, 37.79, 38.58, 39.25, 40.60 (aliphatic); 70.16 (O-CH2); 124.85, 126.00, 126.89, 128.04, 128.81, 138.03, 140.30, 140.85, 140.94, 142.73, 153.31 (aromatic); 160.09 (C=O ). HR LSIMS m/z [M + H]+ = 1205.7684 (Theoretical m/z [M + H]+= 1205.7630). Anal. Calcd for C73H100N6O9: C, 72.73; H, 8.36; N, 6.97, Found: C, 73.10; H, 8.63; N, 6.96. Compound 5-5. A mixture of 2.00 g of 5-2 (1.70 mmol) and 2.50 g of m chloroperbenzoic ac id in 100 mL of CH2Cl2 afforded 1.25 g (60%) of pure product upon washing with diethyl ether. 1H NMR (CDCl3): 1H NMR (CDCl3): 0.52 (m, 18H; ArCH2C H3), 1.11 (s, 18H; C H3), 1.33 (s, 18H; C H3), 1.45 (m, 6H; C H2CH3), 1.68 (m, 6H; C H2CH3,), 2.14 (br s, 6H; NCH2C H2CH2O), 3.61 (m, 12H; NC H2CH2C H2O), 6.36 (s, 1H; C H ), 6.98 (d, J = 6.3 Hz, 6H; ArH ), 7.27 (m, 6H; PyH ), 8.24 (d, J = 6.3 Hz, 3H; PyH ), 8.43 (d, J = 7.8 Hz, 3H; PyH ), 11.20 (br s, 3H; N H D2O exchangeable,). 13C NMR (CDCl3): 9.2, 9.6, 28.6, 29.4, 30.1, 35.0, 36.9, 37.2, 37.6, 39.1 (aliphatic); 70.0 (OCH2); 124.6, 126.8, 127.0, 127.7, 128.9, 138.0, 139.7, 140.5, 140.9, 142.2, 153.5 (aromatic); 159.6 (C=O). HR MS (EI-positive): 1269.7912 [M+Na]+ (Theoretical m/z [M+Na]+= 1269.7919). Anal. Calcd for C76H106N6O9: C, 73.16; H, 8.56; N, 6.74%. Found: C, 74.65; H, 8.92; N, 6.37%. Compound 5-6. A mixture of 2.00 g of 5-3 (1.8 mmol) and 2.50 g of m -chloroperbenzoic acid (70-75%) in 100 mL of CH2Cl2 afforded 1.44 g (69%) of pure product upon washing with diethyl ether. 1H NMR (CDCl3): 1.18 (s, 27H; C H3), 1.33

PAGE 141

125 ((s, 27H; C H3), 2.19 (t, 3JHH = 6.6 Hz, 6H; NCH2C H2CH2O), 3.65 (m, 12H; NC H2CH2C H2O), 6.41 (s, 1H; C H ), 7.14 (s, 6H ArH ), 7.25 (m, 3H; PyH ), 7.29-7.39 (m, 6H; PyH ), 8.22 (m, 3H; PyH ), 8.42 (m, 3H; PyH ), 11.20 (br s, 3H; N H D2O exchangeable). 13C NMR (CDCl3): 30.3, 31.6, 34.7, 35.7, 37.4, 38.8 (aliphatic); 70.6 (O C H2); 122.4, 127.1, 129.2, 138.1, 140.7, 141.1, 141.9, 144.5, 153.9 (aromatic); 159.8 (C=O). HRMS (EI-positive): 1163.72 [M+H]+ (Theoretical m/z [M+H]+= 1163.7161). Anal. Calcd for C70H94N6O9: C 72.26; H 8.14; N 7.22%. Found: C 72.61; H 8.21; N 7.02%. Yb-complex of 5-4 [5-4Yb(NO3)2](NO3). A solution of Yb(NO3)36H2O (0.200 g, 0.166 mmol) in 1 mL of methanol was adde d to a solution of 5-4 (0.074 g, 0.166 mmol) in methanol (1 mL), and the reaction mi xture was stirred for one hour at room temperature. The solvent was partially removed in vacuo and the product was precipitated by diffusion of ethe r. The product was obtained in 50% yield (0.13 g). Slow diffusion of ether into a concen trated methanol solution of th e complex afforded crystals suitable for structural analysis.

PAGE 142

126 CHAPTER 6 SUMMARY Current key challenges in the field of th e partitioning and transmutation of nuclear waste involve the separation of the residual uranium and pl utonium and minor actinides (neptunium, americium, and curium) from large volume of nonradioactive waste elements. With the intent to contribute to the advancement of nuclear fuel and waste reprocessing, a series of reag ents for the selective recogniti on and sequestration of target metal ions has been developed. The ability of the ligands to selectively bind problematic waste components has been evaluated in the liquid-liquid extraction experiments performed under conditions simulating a nuc lear waste stream. The design of new extractants has been inspired by the structur es and properties of small molecules that have been used in waste reprocessing pro cedures. The goal of this research was to improve the efficiency of these binding species by their immobilization on the triphenoxymethane molecular platform. It wa s anticipated that such modification would enhance the binding ability of original extractants via an increase of entropy in the system. Enhanced binding could also l ead to the reduction of the organic waste generated after the extraction through the use of low concentrations of more efficient organic compounds. Also, preorganization of binding units on the molecular scaffold was expected to induce new properties in the ligands (e.g. sel ectivity) due to the changed steric environment and stoichio metry of the extracted species. In most metal extraction processes, two to four small lig ands are involved in the transf er of metal ions into organic phase; therefore the C3-symmetric tripodal platform presen ts itself as a perfect base for

PAGE 143

127 the attachment of binding moieties capable of mimicking the stoichiometry of extracted species. To address the issue of the selective rem oval of the residual plutonium from the PUREX rafinate, the tris-carbamoylmethylphos phine oxide (tris-CM PO) extractant has been developed. Tris-CMPO derivatives have shown excellent binding efficiency and a remarkable selectivity for tetravalent actin ides with target Pu(IV) in particular. Modification of the less popular diglycolamide has led to an enhancement of its unique selectivity within the lanthanides series, but al so resulted in the design of the first purely oxygen based ligand perfectly suited to fulfill the tricapped trigonal prismatic coordination requirements of these trivalent ions In the attempt to tackle the problematic trivalent lanthanides and actin ides separation a sequence of tripodal chelates bearing thiodiglycolamide (TDGA), carbamoylmethyl phosphine sulfide (CMPS) and pyridine Noxide (PyNO) were synthesized. The investig ation of the compositi on of their complexes provided valuable information about the coordi nation preferences of the tested cations in the solid state and in solution. The further refinement of the presented ch elates and careful in vestigation of their binding potential may contribute to a more efficient lanthanide/actinide extraction process, and a fundamental understanding of their selectivity may help in the development of more effective methods for the partitioning of minor actinides.

PAGE 144

128 LIST OF REFERENCES (1) World Nuclear Power Reactors 2004-06 and Uranium Requirements ( http://www.world-nuclear.org/info/reactors.htm ; last viewed on April 2006) (2) Freemantle, M. Chem. Eng. News 2004 82 31. (3) International Energy Outlook 2005, DOE/EIA-0484 ( http://www.eia.doe.gov/oiaf/ieo/pdf/0484(2005).pdf ; last viewed on April 2006) (4) Hore-Lacy, I. Nuclear Electricity ; 7th ed.; Uranium Information Centre Ltd. and World Nuclear Association, 2003. (5) Wald, M. L. In New York Times ; (Late Edition (East Coast)); New York: Jan. 31, 2005. (6) McKay, H. A. C.; Miles, J. H.; Swanson, J. L. Science and Technology of Tributyl Phosphate, Boca Raton, 1984; p 1. (7) Miles, J. H. Science and Technology of Tributyl Phosphate Boca Raton, 1984; p 11. (8) Marsh, S. F.; Simi, O. R. Anal. Chem Energy Technol., 25th, Oak Ridge National Laboratory,TN, 1981; p 69. (9) Horwitz, E. P.; Kalina, D. G.; Diamond, H.; Vandegrift, G. F.; Schulz, W. W. Solvent Extr. Ion Exch. 1985 3 75. (10) Schulz, W. W.; Horwitz, E. P. Sep. Sci. Technol. 1988 23 1191. (11) Chamberlain, D. B.; Leonard, R. A.; Hoh, J. C.; Gay, E. C.; Kalina, D. G.; Vandegrift, G. F. TRUEX Hot Dem onstration: Final Report, ANL-89/37. (12) Murali, M. S.; Mathur, J. N. Solvent Extr. Ion Exch. 2001 19 61. (13) Ozawa, M.; Nemoto, S.; Togashi, A.; Kawata, T.; Onishi, K. Solvent Extr. Ion Exch. 1992 10 829. (14) Mathur, J. N.; Murali, M. S.; Natarajan, P. R.; Badheka, L. P.; Banerji, A.; Ramanujam, A.; Dhami, P. S.; Gopalakrishnan, V.; Dhumwad, R. K.; Rao, M. K. Waste Management 1993 13 317. (15) Musikas, C.; Hubert, H. ISEC '83, Denver, Colorado, USA, 1983; p 449.

PAGE 145

129 (16) Musikas, C. Inorg. Chim. Acta 1987 140 197. (17) Cuillerdier, C.; Musikas, C.; Hoel, P. In New Separation Techniques for Radioactive Waste and Othe r Specific Applications ; Cecille, L., Cesarci, M., Pietrelli, L., Eds.; Elsevier App lied Science: London and New York, 1991. (18) Cuillerdier, C.; Musikas, C. ; Hoel, P.; Nigond, L.; Vitart, X. Sep. Sci. Technol. 1991 26 1229. (19) Cuillerdier, C.; Musikas, C.; Nigond, L. Sep. Sci. Technol. 1993 28 155. (20) Nigond, L.; Musikas, C.; Cuillerdier, C. Solvent Extr. Ion Exch. 1994 12 261. (21) Nigond, L.; Musikas, C.; Cuillerdier, C. Solvent Extr. Ion Exch. 1994 12 297. (22) Nigond, L.; Condamines, N.; Cordier, P. Y.; Livet, J.; Madic, C.; Cuillerdier, C.; Musikas, C.; Hudson, M. J. Sep. Sci. Technol. 1995 30 2075. (23) Madic, C.; Blanc, P.; Condamines, N. ; Baron, P.; Berthon, L.; Nicol, C.; Pozo, C.; Lecomte, M.; Philippe, M.; Masson, M.; Hequet, C.; Hudson, M. J. In Fourth International Conference of Nuclear Fu el Reprocessing and Waste Management, London, UK, 1994. (24) Charbonnel, M. C.; Nicol, C.; Bertho n, L.; Baron, P. International Conference GLOBALÂ’97, Yokohama, Japan, 1997. (25) Bisel, I.; Nicol, C.; Charbonnel, M. C. ; Blanc, P.; Baron, P.; Belnet, F. The Fifth Information Exchange Meeting on Actinides and Fission Products Partitioning and Transmutation, Mol, Belgium, 1998. (26) Zhu, Y. J.; Chen, J. F.; Choppin, G. R. Solvent Extr. Ion Exch. 1996 14 543. (27) Zhu, Y.; Chen, J.; Jiao, R. Solvent Extr. Ion Exch. 1996 14 61. (28) Hill, C.; Madic, C.; Baron, P.; Ozawa, M.; Tanaka, X. J. Alloy Compd. 1998 271273 159. (29) Modolo, G.; Odoj, R. Solvent Extr. Ion Exch. 1999 17 33. (30) Kolarik, Z.; Mullich, U.; Gassner, F. Solvent Extr. Ion Exch. 1999, 17 1155. (31) Kolarik, Z.; Mullich, U.; Gassner, F. Solvent Extr. Ion Exch. 1999 17 23. (32) Boubals, N.; Drew, M. G. B.; Hill, C.; Hudson, M. J.; Iveson, P. B.; Madic, C.; Russell, M. L.; Youngs, T. G. A. J. Chem. Soc. Dalton. 2002 55. (33) Baybarz, R. D. J. Inorg. Nucl. Chem. 1965 27 1831.

PAGE 146

130 (34) Persson, G. E.; Svantesson, S.; Wingefors, S.; Liljenzin, J. O. Solvent. Extr. Ion Exch. 1984 2 89. (35) Madic, C. Actinide and Fission Produc t Partitioning and Transmutation, Madrit, 2000. (36) Choppin, G. R. J Less-Common Met 1983 93 323. (37) Jorgensen, C. K. Isr. J. Chem. 1980 19 174. (38) Choppin, G. R. J. Alloy Compd. 1995 223 174. (39) Choppin, G. R. J. Alloy Compd. 2002 344 55. (40) Diamond, R. M.; Street, K.; Seaborg, G. T. J. Am. Chem. Soc. 1954 76 1461. (41) Irving, H.; Williams, R. J. P. In Treatise on Analytical Chemistry ; Kolthoff, I. M., Elving, P. J., Sandell, E. B., Eds.; Interscience Publishers: New York-London, 1959; Vol. 3. (42) Nash, K. L. Solvent Extr. Ion Exch. 1993 11 729. (43) Nash, K. L. In Handbook on the Physics and Chemistry of Rare Earths ; Gschneider, K. A., Eyring Jr., L., Choppin, G. R., Lander, G. H., Eds.; Elsevier Science: Amsterdam, 1994; Vol. 18. (44) Schulz, W. W.; Navratil, J. D.; Talbot, A. E. Science and Technology of Tributyl Phosphate ; CRC Press, Inc.: Boca Raton, Florida, 1984. (45) Ejaz, M. Talanta 1976 23 193. (46) Ejaz, M.; Carswell, D. J. J. Inorg. Nucl. Chem. 1975 37 233. (47) Siekierski, S. J. Inorg. Nucl. Chem. 1962 24 205. (48) Ionova, G.; Ionov, S.; Rabbe, C.; Hill, C.; Madic, C.; Guillaumont, R.; Krupa, J. C. Solvent Extr. Ion Exch. 2001 19 391. (49) ArnaudNeu, F.; Bohmer, V.; Dozol, J. F.; Gruttner, C.; Jakobi, R. A.; Kraft, D.; Mauprivez, O.; Rouquette, H.; SchwingW eill, M. J.; Simon, N.; Vogt, W. J. Chem. Soc., Perkin Trans. 2 1996 1175. (50) Dinger, M. B.; Scott, M. J. Eur. J. Org. Chem. 2000 2467. (51) Peters, M. W.; Werner, E. J.; Scott, M. J. Inorg. Chem. 2002 41 1707. (52) Boerrigter, H.; Ver boom, W.; Reinhoudt, D. N. J. Org. Chem. 1997 62 7148.

PAGE 147

131 (53) Schmidt, C.; Saadioui, M.; Bohmer, V.; Host, V.; Spirlet, M. R.; Desreux, J. F.; Brisach, F.; Arnaud-Neu, F.; Dozol, J. F. Org. Biomol. Chem. 2003 1 4089. (54) Arduini, A.; Bohmer, V.; Delmau, L.; Desr eux, J. F.; Dozol, J. F.; Carrera, M. A. G.; Lambert, B.; Musigmann, C.; Pochini, A.; Shivanyuk, A.; Ugozzoli, F. Chem. Eur. J. 2000 6 2135. (55) Kandil, A. T.; Aly, H. F.; Raieh, M.; Choppin, G. R. J. Inorg. Nucl. Chem. 1975 37 229. (56) Liljenzin, J. O.; Rydberg, J.; Skarnemark, G. Sep. Sci. Technol. 1980 15 799. (57) Cecille, L.; Dworschak, H.; Girardi, F.; Hunt, B. A.; Mannone, F.; Mousty, F. Actinide Separation, Washington D. C., 1980; p 427. (58) Koch, G. Actinide Separation, Washington D. C., 1980; p 411. (59) Ishimori, T. Actinide Separation, Washington D. C., 1980; p 343. (60) Horwitz, E. P.; Delphin, W. H.; Ma son, G. W.; Steindler, M. “Recovery Alternatives Applicable to Waste Streams in Actinide Partitioning and Transmutation Program Process,” U. S. Department of Energy, 1978. (61) Stary, J. Talanta 1966 13 421. (62) Zhu, Y. J.; Jiao, R.; Wang, S.; Fan, S.; Liu, B.; Zheng, H.; Zhou, S.; Chen, S. International Solvent Extraction Conf erence (ISEC'83), Denver, Colorado, 1983. (63) Schulz, W. W.; McIsaac, L. D. Inte rnational Solvent Extraction Conference, Toronto, Canada, 1979; p 619. (64) Schulz, W. W.; Navratil, J. D. Recen t Development in Separation Science, Boca Raton, Florida, 1981; p 31. (65) McIsaac, L. D.; Baker, J. D.; Krupa, J. F.; Meikrantz, D. H.; Schroeder, N. C. Actinide Separation, Washington D. C., 1980; p 395. (66) Bond, E. M.; Leuze, R. E. Actinid e Separation, Washingt on D. C., 1980; p 441. (67) Navratil, J. D.; Martel la, L. L. Actinide Recovery from Waste and Low-Grade Sources, New York, 1982; p 27. (68) Horwitz, E. P.; Diamond, H.; Kalina, D. G. Plutonium Chemistry, Washington D. C., 1983; p 433. (69) Horwitz, E. P.; Diamond, H.; Kalin a, D. G.; Kaplan, L.; Mason, G. W. International Solvent Extraction Conf erence (ISEC'83), Denver, Colorado, 1983. (70) O'Laughlin, J. W. Progress in Nuclear Energy. ; Pergamon Press.: New York, 1966.

PAGE 148

132 (71) Kalina, D. G.; Horwitz, E. P.; Kaplan, L.; Muscatello, A. C. Sep. Sci. Technol. 1981 16 1127. (72) Horwitz, E. P.; Kalina, D. G.; Kaplan, L.; Mason, G. W.; Diamond, H. Sep. Sci. Technol. 1982 17 1261. (73) Martin, K. A.; Horwitz, E. P.; Ferraro, J. R. Solvent Extr. Ion Exch. 1986 4 1149. (74) Baker, J. D.; Mincher, B. J.; Meikrantz, D. H.; Berreth, J. R. Solvent Extr. Ion Exch. 1988 6 1049. (75) Matthews, S. E.; Saadioui, M.; Bohmer, V.; Barboso, S.; Arnaud-Neu, F.; Schwing-Weill, M. J.; Carrera, A. G.; Dozol, J. F. J. Prakt. Chem./Chem.-Ztg. 1999 341 264. (76) Barboso, S.; Carrera, A. G.; Matthews, S. E.; Arnaud-Neu, F.; Bohmer, V.; Dozol, J. F.; Rouquette, H.; Schwing-Weill, M. J. J. Chem. Soc., Perkin Trans. 2 1999 719. (77) Delmau, L. H.; Simon, N.; Schwing-We ill, M. J.; Arnaud-Neu, F.; Dozol, J. F.; Eymard, S.; Tournois, B.; Gruttner, C.; Musigmann, C.; Tunayar, A.; Bohmer, V. Sep. Sci. Technol. 1999 34 863. (78) Boerrigter, H.; Ver boom, W.; Reinhoudt, D. H. Liebigs Ann.-Recl. 1997 2247. (79) Boerrigter, H.; Verboom, W. ; de Jong, F.; Reinhoudt, D. N. Radiochim. Acta 1998 81 39. (80) Horwitz, E. P.; Martin, K. A.; Diamond, H.; Kaplan, L. Solvent Extr. Ion Exch. 1986 4 449. (81) Bhattacharya, A. K.; Thyagarajan, G. Chem. Rev. 1981 81 415. (82) Ozawa, M.; Nemoto, S.; Nomura, K.; Korea, Y.; Togashi, A. International Information Exchange Program on Actin ide and Fission Product Separation and Transmutation, Argonne, Illinois, 1992. (83) Cotton, F. A.; Wilkinson, G. Advanced Inorganic Chemistry ; 5th ed.; Wiley: New York, 1988. (84) Seaborg, G. T.; Thompson, S. G.; Davidson, N. R. 1959 Patent US 2903335 (85) Shannon, R. D. Acta Crystalogr. A 1976 32 751. (86) Engelhardt, U.; Rapko, B. M.; Duesler, E. N.; Frutos, D.; Paine, R. T.; Smith, P. H. Polyhedron 1995 14 2361. (87) Mehring, M.; Mansfeld, D.; Schurmann, M. Z. Anorg. Allg. Chem. 2004 630 452.

PAGE 149

133 (88) Marczenko, Z. Separation and Spectrophotometric Determination of Elements ; E. Horwood; Halsted Press: Chichester, New York, 1986. (89) Cleveland, J. M. The Chemistry of Plutonium ; Gordon and Breach Science: New York, 1969. (90) Xu, J.; Kullgren, B.; Durbin, P. W.; Raymond, K. N. Journal of Medicinal Chemistry 1995 38 2606. (91) Smukste, I.; Smithrud, D. B. Tetrahedron 2001 57 9555. (92) Peters, M. W. Ph. D., University of Florida, 2002. (93) Fieser, L. F.; Fieser, M. Reagents for Organic Synthesis ; Wiley: New York, 1967. (94) Zhu, Y. J. Chin. J. Nucl. Radiochem. 1989 11 212. (95) Musicas, C. Actinide/Lanthanide Separation, Honolulu, Hawaii, 1984; p 19. (96) Ensor, D. D.; Jarvinen, G. D.; Smith, B. F. Solvent Extr. Ion Exch. 1988 6 439. (97) Smith, B. F.; Jarvinen, G. D.; Jones, M. M.; Hay, P. J. Solvent Extr. Ion Exch. 1989 7 749. (98) Kolarik, Z.; Mullich, U. Solvent Extr. Ion Exch. 1997 15 361. (99) Chen, J.; Zhu, Y. J.; Jiao, R. Z. Sep. Sci. Technol. 1996 31 2723. (100) Modolo, G.; Odoj, R. J. Alloy Compd. 1998 271 248. (101) Modolo, G.; Odoj, R. OECD NE A Workshop on Long-Lived Radionuclide Chemistry in Nuclear Waste Treatme nt, Villeneuve-les-Avignon, France, 1997. (102) Modolo, G.; Odoj, R. Actnid es'97, Baden-Baden, Germany, 1997. (103) Codding, P. W.; Kerr, K. A. Acta Crystallogr B 1978 34 3785. (104) Ziemer, B.; Rabis, A.; Steinberger, H. U. Acta Crystallogr C 2000 56 58. (105) Chekhlov, A. N. J. Struct. Chem. 2002, 43 364. (106) Pinkerton, A. A.; Schwarzenbach, D.; Spiliadis, S. Inorg. Chim. Acta 1987 128 283. (107) Malevann, R.; Tsvetkov, E. N.; Kabachni, M. Zh. Obshch. Khim. 1971 41 1426. (108) Fritz, J. S.; Orf, G. M. Anal. Chem. 1975 47 2043. (109) Thiollet, G.; Musikas, C. Solvent Extr. Ion Exch. 1989 7 813.

PAGE 150

134 (110) Condamines, N.; Musikas, C. Solvent Extr. Ion Exch. 1992 10 69. (111) Nair, G. M.; Prabhu, D. R.; Mahajan, G. R.; Shukla, J. P. Solvent Extr. Ion Exch. 1993 11 831. (112) Prabhu, D. R.; Mahajan, G. R.; Nair, G. M.; Subramanian, M. S. Radiochim. Acta 1993 60 109. (113) Cuillerdier, C.; Musikas, C. Sep. Sci. Technol. 1993 28 115. (114) Nakamura, T.; Miyake, C. Solvent Extr. Ion Exch. 1995 13 253. (115) Narita, H.; Yaita, T.; Tachimori, S. Solvent Extr. Ion Exch. 2004 22 135. (116) Manchanda, V.; Pathak, P. Sep. Purif. Technol. 2004 35 85. (117) Stephan, H.; Gloe, K.; Beger, J.; Muhl, P. Solvent Extr. Ion Exch. 1991 9 435. (118) Stephan, H.; Gloe, K.; Beger, J.; Muhl, P. Solvent Extr. Ion Exch. 1991 9 459. (119) Yao, J.; Wharf, R. M.; Choppin, G. R. In Separation of F-Elements ; Nash, K. L., Choppin, G. R., Eds.; Plenum Press: New York, 1995, 1995. (120) Sasaki, Y.; Choppin, G. R. J. Radioan. Nucl. Ch. Ar. 1996 207 383. (121) Sasaki, Y. I.; Choppin, G. R. Anal. Sci. 1996 12 225. (122) Narita, H.; Yaita, T.; Tamura, K.; Tachimori, S. Radiochim. Acta 1998 81 223. (123) Sasaki, Y.; Choppin, G. Radiochim. Acta 1998 80 85. (124) Narita, H.; Yaita, T.; Tachimori, S. ISEC'99, Barcelona, Spain, 1999; p 693. (125) Narita, H.; Yaita, T.; Tamura, K.; Tachimori, S. J. Radioanal. Nucl. Chem. 1999 239 381. (126) Sasaki, Y.; Sugo, Y.; Tachimori, S. International Conference ATALANTE 2000, Avignon, France, 2000. (127) Sasaki, Y.; Choppin, G. R. J. Radioanal. Nucl. Chem. 2000 246 267. (128) Tachimori, S.; Suzuki, S.; Sasaki, Y. J. Atom. Energ. Soc. Jpn. 2001 43 1235. (129) Sasaki, Y.; Sugo, Y.; Suzuki, S.; Tachimori, S. Solvent Extr. Ion Exch. 2001 19 91. (130) Sasaki, Y.; Tachimori, S. Solvent Extr. Ion Exch. 2002 20 21.

PAGE 151

135 (131) Yaita, T.; Herlinger, A. W.; Thiyagarajan, P.; Jensen, M. P. Solvent Extr. Ion Exch. 2004 22 553. (132) Zhu, Z. X.; Sasaki, Y.; Suz uki, H.; Suzuki, S.; Kimura, T. Anal. Chim. Acta 2004 527 163. (133) Nigond, L., Univesite de Clermont-Ferrand II, 1994. (134) Satmark, B.; Courson, O.; Malmbeck, R.; Pa gliosa, G.; Romer, K.; Glatz, J. P. The Sixth International Information Exchange Meeting on Actinide and Fission Product Partitioning and Transmutati on, Madrit, Spain, 2000; p 279. (135) Malmbeck, R.; Courson, O.; Pagliosa, G.; Romer, K.; Satmark, B.; Glatz, J. P.; Baron, P. Radiochim. Acta 2000 88 865. (136) Courson, O.; Lebrun, M.; Malmbeck, R. ; Pagliosa, G.; Romer, K.; Satmark, B.; Glatz, J. P. Radiochim. Acta 2000 88 857. (137) Pretsch, E.; Amma nn, D.; Osswald, H. F.; Guggi, M.; Simon, W. Helv. Chim. Acta 1980 63 191. (138) Ping, Z.; Jing, C.; Chunyu, L.; Guoxin, T. Chem. J. Internet 2003 5 52. (139) Nagao, Y.; Miyamoto, S.; Ha yashi, K.; Mihira, A.; Sano, S. Tetrahedron Letters 2002 43 1519. (140) Matloka, K.; Gelis, A.; Regalbuto, M.; Vandegrift, G.; Scott, M. J. J. Chem. Soc., Dalton Trans. 2005 3719. (141) Spjuth, L.; Liljenzin, J.; Skalberg, M. ; Hudson, M.; Chan, G.; Drew, M.; Feaviour, M.; Iveson, P.; Madic, C. Radiochim. Acta 1997 78 39. (142) Cossy, C.; Helm, L.; Merbach, A. E. Inorg. Chem. 1988 27 1973. (143) Cossy, C.; Barnes, A. C.; Enderby, J. E.; Merbach, A. E. J. Chem. Phys. 1989 90 3254. (144) Cossy, C.; Helm, L.; Merbach, A. E. Inorg. Chem. 1989, 28 2699. (145) Meier, W.; Bopp, P.; Probst, M. M.; Spohr, E.; Lin, J. I. Journal of Physical Chemistry 1990 94 4672. (146) Bandurkin, G. A.; Dzhurinskii, B. F. Zh. Neorg. Khim. 1997 42 1143. (147) Abbasi, A.; Lindqvist-Reis, P.; Erikss on, L.; Sandstrom, D.; Lidin, S.; Persson, I.; Sandstrom, M. Chem. Eur. J. 2005 11 4065. (148) Floris, F. M.; Tani, A. J. Chem. Phys. 2001 115 4750.

PAGE 152

136 (149) Zhang, Y.; Liu, W.; Wang, Y.; Tang, N.; Tan, M.; Yu, K. J. Coord. Chem. 2002 55 1293. (150) Chatterjee, A.; Maslen, E. N.; Watson, K. J. Acta Crystallogr B 1988 44 381. (151) Sasaki, Y.; Choppin, G. R. J. Radioanal. Nucl. Chem. 1997 222 271. (152) Sasaki, Y.; Adachi, T.; Choppin, G. R. J. Alloy Compd. 1998 271 799. (153) Sasaki, Y.; Choppin, G. J. Radioanal. Nucl. Chem. 2000 246 267. (154) Merijani.A; Zingaro, R. A. Inorg. Chem. 1966 5 187. (155) Irgolic, K.; Zingaro, R. A.; Linder, D. E. J. Inorg. Nucl. Chem. 1968 30 1941. (156) Linder, D. E.; Zingaro, R. A.; Irgolic, K. J. Inorg. Nucl. Chem. 1967 29 1999. (157) Gaur, P. K.; Mohanty, S. R. Naturwissenschaften 1963 50 614. (158) Kennedy, D. C.; Fritz, J. S. Talanta 1970 17 823. (159) Mohanty, S. R.; Nalini, P. Curr. Sci. 1967 36 175. (160) Nikolaev, A. V.; Torgov, V. G.; Andrievs, V. N.; Galtsova, E. A.; Gilbert, E. N.; Kotlyaro, I. L.; Mazalov, L. N.; Mikhailo, V. A.; Cheremis, I. M. Zh. Neorg. Khim. 1970 15 1336. (161) Ejaz, M. Anal. Chim. Acta 1974 71 383. (162) Ejaz, M. Radiochim. Acta 1975 22, 51. (163) Ejaz, M. Separ. Sci. 1975 10 425. (164) Ejaz, M. J. Radioanal. Chem. 1975 27 67. (165) Ejaz, M. Mikrochim. Acta 1976 2 265. (166) Ejaz, M. Mikrochim. Acta 1976 1 643. (167) Ejaz, M. Anal. Chem. 1976 48 1158. (168) Ejaz, M. Anal. Chim. Acta 1976 81 149. (169) Ejaz, M. J. Radioanal. Chem. 1977 36 409. (170) Ejaz, M. J. Radioanal. Chem. 1977 35 341. (171) Ejaz, M.; Carswell, D. J. J. Radioanal. Chem. 1976 29 259.

PAGE 153

137 (172) Torgov, V. G.; Mikhailo, V. A.; Startsev, E. A.; Nikolaev, A. V. Dok. Akad. Nauk SSSR 1966 168 836. (173) Gilbert, E. N.; Torgov, V. G.; Mikhail o, V. A.; Artyukhi, P. I.; Nikolaev, A. V. Dok. Akad. Nauk SSSR 1967 174 1329. (174) Maksimov, Z. B.; Puzic, R. G. J. Inorg. Nucl. Chem. 1972 34 1031. (175) Herlocke, D. W.; Drago, R. S.; Meek, V. I. Inorg. Chem. 1966 5 2009. (176) Dickson, F. E.; Gowling, E. W.; Bentley, F. F. Inorg. Chem. 1967 6 1099. (177) Garvey, R. G.; Ragsdale, R. O. J. Inorg. Nucl. Chem. 1967 29 1527. (178) Garvey, R. G.; Nelson, J. H.; Ragsdale, R. O. Coord. Chem. Rev. 1968 3 375. (179) Ochiai, E. J. Org. Chem. 1953 18 534. (180) Linton, E. P. J. Am. Chem. Soc. 1940 62 1945. (181) Carlin, R. L. J. Am. Chem. Soc. 1961 83 3773. (182) Potts, R. A.; Allred, A. L. Inorg. Chem. 1966 5 1066. (183) Matloka, K.; Gelis, A.; Regalbuto, M. ; Vandegrift, G. F.; Scott, M. J. Sep. Sci. Technol. 2006 (accepted for publication). (184) Smith, J. W. Electric Dipole Moments ; Butterworth's Scientific Publications: London, 1955. (185) Christensen, J. B. Molecules 2001 6 47. (186) Nattinen, K. I.; Rissanen, K. Cryst. Growth Des. 2003 3 339. (187) Nakajima, M.; Sasaki, Y.; Shiro, M.; Hashimoto, S. Tetrahedron-Asymmetry 1997 8 341. (188) Gan, X. M.; Binyamin, I.; Rapko, B. M.; Fox, J.; Duesler, E. N.; Paine, R. T. Inorg. Chem. 2004 43 2443. (189) Rais, J.; Selucky, P.; Kyrs, M. J. Inorg. Nucl. Chem. 1976 38 1376. (190) Makrlik, E.; Vanura, P. Talanta 1985 32 423. (191) Romanovskiy, V. N.; Smirnov, I. V.; Baba in, V. A.; Todd, T. A.; Herbst, R. S.; Law, J. D.; Brewer, K. N. Solvent Extr. Ion Exch. 2001 19 1.

PAGE 154

138 (192) Herbst, R. S.; Law, J. D.; Todd, T. A.; Romanovskii, V. N.; Babain, V. A.; Esimantovski, V. M.; Zaitsev, B. N.; Smirnov, I. V. Sep. Sci. Technol. 2002 37 1807. (193) Juhasz, M.; Hoffmann, S.; Stoyanov E.; Kim, K. C.; Reed, C. A. Angew. Chem. Int. Edit. 2004 43 5352. (194) Vanura, P.; Makrlik, E.; Rais, J.; Kyrs, M. Collect. Czech. Chem. Commun. 1982 47 1444. (195) Rais, J.; Tachimori, S. J. Radioan. Nucl. Ch. Le. 1994 188 157. (196) Rais, J.; Tachimori, S. Sep. Sci. Technol. 1994 29 1347. (197) Rais, J.; Tachimori, S.; Selucky, P.; Kadlecova, L. Sep. Sci. Technol. 1994 29 261. (198) Navratil, J. D.; Vanura, P.; Jedinakovakrizova, V. J. Radioan. Nucl. Ch. Le. 1995 201 81. (199) Vanura, P.; Stibor, I. Collect. Czech. Chem. Commun. 1998 63 2009. (200) Rais, J.; Selucky, P.; Sistkova, N. V.; Alexova, J. Sep. Sci. Technol. 1999 34 2865. (201) Kyrs, M.; Svoboda, K.; Lhotak, P.; Alexova, J. J. Radioanal. Nucl. Chem. 2002 254 455. (202) Coupez, B.; Wipff, G. Comptes Rendus Chimie 2004 7 1153. (203) Stoyanov, E. S.; Smirnov, I. V. J. Mol. Struct. 2005 740 9. (204) Smirnov, I. V.; Babain, V. A.; Shadrin, A. Y.; Efremova, T. I.; Bondarenko, N. A.; Herbst, R. S.; Peterman, D. R.; Todd, T. A. Solvent Extr. Ion Exch. 2005 23 1. (205) Garvey, R. G.; Ragsdale, R. O. J. Inorg. Nucl. Chem. 1967 29 745. (206) Rapko, B. M.; Duesler, E. N.; Smith, P. H.; Paine, R. T.; Ryan, R. R. Inorg. Chem. 1993 32 2164.

PAGE 155

139 BIOGRAPHICAL SKETCH Kornelia Karolina Matloka was born in Kamien Pomorski in Poland, on August 15, 1977. Shortly after, she moved to Gniezno where she resided for the next eighteen years. She began her undergraduate studies in 1996 at Adam Mickiewicz University in Poznan, majoring in inorganic/bioinorganic chemistry. Her research consisted of the synthetic development of supramolecular complexes containing rare earth metals under the supervision of Professor Wanda Radecka-Pary zek. While at college, she was a president of the Student Government of the Chemistry Department, a member of the University Senate, Academic Sport Association and a recipient of Adam Mickiewicz Foundation Scholarship in 2000. In December 2000, she graduated maxima cum laude with a Master of Science degree in chemistry. Soon afte r, in January 2001, she began her graduate career in inorganic/organic ch emistry at the University of Florida under the guidance of Dr. Michael J. Scott. She completed the requirements for the degree of Doctor in Philosophy in May 2006.