UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Chelate complexes of rhenium and technetium: toward their potential applications as radiopharmaceuticals Luo, Hongyan 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-060063.pdf [ 6.92MB ]
Metadata
JSON: 831-1.0059490.json
JSON-LD: 831-1.0059490-ld.json
RDF/XML (Pretty): 831-1.0059490-rdf.xml
RDF/JSON: 831-1.0059490-rdf.json
Turtle: 831-1.0059490-turtle.txt
N-Triples: 831-1.0059490-rdf-ntriples.txt
Original Record: 831-1.0059490-source.json
Full Text
831-1.0059490-fulltext.txt
Citation
831-1.0059490.ris

Full Text

CHELATE COMPLEXES OF RHENIUM AND TECHNETIUM: TOWARD THEIR POTENTIAL APPLICATIONS AS RADIOPHARMACEUTICALS by Hongyan Luo B. Sc. (Hons.), Central South Institute of Mining and Metallurgy, China, 1982 M. Sc, Central South University of Technology, China, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1995 © HONGYAN LUO, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. (Signature) Department of The University of British Columbia Vancouver, Canada Date Chemistry DE-6 (2/88) Abstract Technetium-99m is the major isotope used in diagnostic radiopharmaceuticals, and it will continue to dominate in the future. Macroscopic studies of stable technetium-99 complexes are very important in searching for new diagnostic radiopharmaceuticals based on known structural information. Possessing similar chemistry but being non-radioactive, rhenium is a good model for technetium. In addition, two isotopes of this element: rhenium-186 and rhenium-188, are good potential radionuclides for therapeutic radiopharmaceuticals. Therefore, complexes of both metals were synthesized with naturally occurring or easily synthesized chelates, and with new ligands designed to stabilize both metals in intermediate oxidation states. Technetium(V) and rhenium(V) complexes of the form [MOXL2] and [MOX3(ma)]-, where L is a bidentate (0,0) monoprotic ligand, 2-methyl-3-oxy-4-pyronate (maltolate, ma") or l,2-dimethyl-3-oxy-4-pyridinonate (dpp-) and X is a halo ligand, X = CI (M = Tc) or Br (M = Re), have been synthesized by ligand substitution reactions on [TcOCU]" or [ReOBr- anions. The structures of [ReOBr(ma)2], [(n-Bu)4N][ReOBr3(ma)] and [(n-Bu)4N][TcOCl3(ma)] were determined by X-ray crystallography. Conversion of [Re0Br3(ma)]" to [ReOBr(ma)2] was achieved by addition of excess Hma or standing. The rhenium(V) and technetium (V) complexes: [ReO(apa)], [{ReO(epa)}20], and [TcOCl(epa)] from a potentially pentadentate (N3O2) ligand precursor, N,N'-3-azapentane-l,5-diylbis(3-(l-iminoethyl)-6-methyl-2H-pyran-2,4(3H)-dione) (H3apa), or a potentially tetradentate (N2O2) ligand precursor, N,N'-ethylene-diylbis(3-(l-iniinoethyl)-6-methyl-2i7-pyran-2,4(3if)-dione) (H2erpa) are described. There was evidence indicating that H2ppa, N,N'-propylene-diylbis(3-(l-iminoethyl)-6-methyl-2H-pyran-2,4(3H)-dione), hydrolyzed in the course of coordination, forming a rhenium complex of the bidentate monoprotic (0 ,0) dehydroacetate. A modified preparation of the hydrochloride salt of a potentially tridentate ligand precursor, bis(o-hydroxyphenyl)phenylphosphine (abbreviatedH2PO2HCI) is described. From ii H2PO2HCI, and a potentially bidentate analog, o-hydroxyphenyldiphenylphosphine (HPO), POxx" (x = 1,2) complexes of rhenium(V) and technetium(V) were prepared by metathesis reactions with the appropriate metal(V) precursors, and/or by reduction/coordination reactions with ammonium perrhenate or pertechnetate. The new POx complexes fall into four categories: bis(PO) complexes [(MOCl(PO)]2 (M = Re or Tc), and [ReN(PO)2(PPh3)]; mono(P02) complexes [ReZCl(PPh3)(P02)] ( Z = O or NPh); mixed(Po/PC>2) complexes [ReZ(PO)(P02)] (Z= O or NPh); and mixed(P02HP02) complexes [MO(P02)(HPC>2)] (M = Re or Tc). Geometrical isomerism was observed for several complexes. The structures of fac-cis-(P,P)-[Re(NPh)Cl(PPh3)(P02)] and/ac-cw-(P,P)-[ReO(PO)(P02)] have been solved by X-ray crystallography. Preparation of the hydrochloride salt of a new potentially hexadentate ligand precursor P,P,P',P'-tetrakis(o-hydroxyphenyl)diphosphinoethane (abbreviatedH4P2O42HCI) is described. From H4P2O42HCI, two dinuclear complexes, [Re2C>2Cl2(PPh3)2(u-P204)], and [(«-Bu)4N]2[Re202Br4(|i-P204)], and two mononuclear complexes, [M(HP2C>4)]nS (M = Tc, nS = 2EtOH0.5PhMe; M = Fe, nS = 2H2O), were synthesized. The Tc complex was synthesized rapidly by a reduction/coordination route from the reaction of ammonium pertechnetate with H4P2O42HCI, in which the functionalized phosphine is a reducing as well as a ligating agent. Cyclic voltammetric study of the complex [Tc(HP2O4)]-2EtOH0.5PhMe revealed that the Tc(III) complex is well stabilized by the coordination environment. Labeling H2P02 HC1 andH4P2O42HCI with 99mTc was also pursued. The TLC results suggested that both ligands are labeled easily and rapidly in EtOH, in high radiochemical yield. The "mTc labeled complexes were found to be hydrolytically stable. The labeling procedures, without an added reducing agent, are simple and kit-amenable for potential clinical application. Two isomers are present in [99mTc(P02)(HP02)], consistent with the macroscopic chemistry. The preliminary in vivo biodistribution studies of the radiolabeled [99mTc(HP2C>4)] complex in mice showed that the complex has high liver affinity, and is worthy of further investigation into its use as a potential liver function agent. i i i Table of Contents Abstract Table of Contents List of Figures List of Tables List of Abbreviations Dedication Chapter One General Introduction Background The Project References Chapter Two Rhenium(V) and Technetium(V) Complexes of Bidentate Monoprotic (0,0) Ligands Introduction Experimental Results and Discussion Conclusion References Chapter Three Rhenium and Technetium Complexes from Pentadentate (N3O2) and Tetradentate (N2O2) Schiff Bases Introduction Experimental Results and Discussion Conclusion References ii iv vi ix X xiv 1 1 13 20 24 24 25 29 42 44 47 47 48 52 64 66 iv Chapter Four Rhenium(V) and Technetium(V) Complexes Incorporating the Deprotonated Forms of (o-hydroxyphenyl)diphenylphosphine (HPO) and Bis(o-hydroxyphenyl)phenylphosphine (H2PO2) 68 Introduction 68 Experimental 69 Results and Discussion 77 Conclusion 100 References 103 Chapter Five Metal Complexes Incorporating the Deprotonated Forms of P,P,P',P'-Tetrakis(o-hydroxyphenyl)diphosphinoethane (H4P2O4) 106 Introduction 106 Experimental 107 Results and Discussion 111 Conclusion 124 References 126 Chapter Six Radiolabeling Functionalized Phosphines with 99mTc an (j Preliminary Biodistribution Studies of [99mTc(HP204>] 128 Introduction 128 Experimental 130 Results and Discussion 133 Conclusion 142 References 143 Chapter Seven General Conclusions and Suggestions for the Future work 144 General Conclusions 144 Suggestions for the Future work 145 References 147 Acknowledgments ^ y Appendices 149 List of Figures Figure 1-1. Marketed 99mTc radiopharmaceuticals with known structures 4 Figure 1-2. 99mTc-radiopharmaceutical studies and Tc and Re coordination chemistry 14 Figure 1-3. Bidentate 3-hydroxy-4-pyrone/4-pyridinone (left) and multidentate Schiff base (right) ligands 16 Figure 1-4. Proposed neutral metal(IV and V) complexes 18 Figure 1-5. Multifunctionalized phosphine and diphosphine ligands 18 Figure 2-1. *H NMR spectra of the characteristic pair of ring hydrogens of the bis(maltolato) and mono(maltolato) complexes in CD3CN 31 Figure 2-2. ORTEP drawing of one of the two independent molecules in the asymmetric unit of [ReOBr(ma)2] 34 Figure 2-3. ORTEP drawing of the anions in [(n-Bu)4N][ReOBr3(ma)] (left) and[(n-Bu)4N][TcOCl3(ma)] (right) 36 Figure 2-4. *H NMR spectra (a. 300 MHz and b. 200 MHz) showing the conversion of [ReOBr3(ma)]- to [ReOBr(ma)2] in CD3CN 41 Figure 3-1. Dha and dha-based Schiff base ligands 48 Figure 3-2. Tautomerism of the Schiff base ligands 52 Figure 3-3. !H- *H COSY (400 MHz, ethylene region) for [ReO(apa)] 58 Figure 3-4. Ethylene region of the !H NMR spectrum (top, 400 MHz) and a computer-simulated AA'BB' system for [ {ReO(epa)} 2O] 60 Figure 3-5. ORTEP drawing of [Re(dha)Cl2(OPPh3)(PPh3)]EtOH 62 vi Figure 4-1. *H NMR (400 MHz) of [MOCl(PO)2] (M = Re, Tc) in CDCI3 84 Figure 4-2. Variable temperature 31P{ !H} NMR spectra (121 MHz) of [ReN(PO)2(PPh3)] in CDCI3 85 Figure 4-3. 31P{ !H} NMR (121 MHz) of [ReO(P02)(HP02)] in DMSO-d6 at room temperature and high temperature, and the proposed structures for the two diastereomers 89 Figure 4-4. *H NMR (200 MHz for M = Re, 400 MHz for M = Tc) of [MO(P02)(HP02)] in DMSO-d6 90 Figure 4-5. Selectively 3iP-decoupled lU{31P} NMR spectra (500 MHz, partial) of c«-(P,P)-[ReO(P02)(HP02)] in DMSO-</6 91 Figure 4-6. ORTEP drawing of/ac-cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) (with the solvent molecules omitted) 93 Figure 4-7. ORTEP drawing of cw-(P,P)-[ReO(PO)(P02)] (2) 94 Figure 4-8. 31P{ *H} NMR (81 MHz) of some complexes in CDCI3 showing their reactivity toward pyridine 99 Figure 5-1. lH NMR spectra for [ R e ^ C ^ P P l ^ M u - P ^ ) ] (top, 300 MHz) and [(n-Bu)4N]2[Re202Br4(u-P204)] (bottom, 400 MHz) 115 Figure 5-2. Diastereomers of the dinuclear complexes showing the AA'BB'XX' systems of the ethylene backbone, due to chirality of the metal center 116 Figure 5-3. The proposed isomers and the JH^H COSY spectrum (400 MHz) of [(n-Bu)4N]2[Re202Br4(n-P204)] (aromatic region); the correlated H resonances are designated by the same number as the PO phenyl ring 118 Figure 5-4. *H (top) and iHpiP} (bottom) NMR spectra (500 MHz) of [(«-Bu)4N]2[Re202Br4(|Li-P204)] (aromatic and ethylene regions) 119 Figure 5-5. 31P{ lH} NMR spectra (81 MHz) showing reactivity of the dinuclear complexes to pyridine 123 vii Figure 6-1. The highest reading of activity in the segments indicated by small circles corresponds to the Rf = 6.5/12.0 = 0.54; and radiochemical purity = counts of the three segments/total counts 131 Figure 6-2. Dilution scheme of the "mTc labeled complexes 132 Figure 6-3. Time dependence of labeling yield for the 99mTc-complexes fromH2P02 HC1 (left) andH4P204 2HC1 (right) 134 Figure 6-4. Radiochromatographic plots of [99mTcO(P02)(HP02)] (left) and or [99mTc(HP204)] (right) after dilution from 100 % EtOH to 20% EtOH in H 20 135 Figure 6-5. The radiochromatographic plot of the [99mTc(HP204)] final solution (10 (iCi/mL after dilution to 0.5 % EtOH in H20) 137 Figure 6-6. Tissue distribution diagram of [99mTc(HP204)] in mice (per organ) 139 Figure 6-7. Tissue distribution diagram of [99mTc(HP204)] in mice (per gram -data for liver, stomach, heart, bone, brain omitted) 140 Figure 6-8. Liver uptake of [99mTc(HP204)] relative to blood (top) and kidney (bottom) 141 viii List of Tables Table 2-1. Selected Bond Lengths (A) for one of the two independent molecules in [ReOBr(ma)2], and for the anions in [(n-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)] 35 Table 2-2. Selected Bond Angles (deg) for one of the two independent molecules in [ReOBr(ma)2], and for the anions in [(n-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)] 37 Table 3-1. Selected Bond Lengths (A) for [Re(dha)Cl2(OPPh3)(PPh3)]HOEt 61 Table 3-2. Selected Bond Angles (deg) for [Re(dha)Cl2(OPPh3)(PPh3)]HOEt 63 Table 4-1. Selected Bond Lengths (A) for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and ReO(PO)(P02) (2) 95 Table 4-2. Selected Bond Angles (deg) for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2) 96 Table 5-1. Results of Thermogravimetric Study on [Fe(HP204)(H20)] H 2 0 121 ix List of Abbreviations Abbreviation A A" a B B' b.p. br °C Calcd chiraphos Ci cis cm cm*1 COSY CV 5 d D d dd dec dha dien dipamp DMF DMPE DMSO Meaning radioactivity; a nucleus (NMR) a nucleus chemically equivalent to nucleus A, but magnetically non-equivalent axial a nucleus (NMR) a nucleus chemically equivalent to B, but magnetically non-equivalent boiling point broad (spectral) degrees Celsius calculated 1,2-dimethyl-1,2-bis(diphenylphosphino)ethane Curie (= 3.7 x 1010 disintegrations per second) Latin, "on the side", or the same side centimeter wave number(s) (IR) correlated spectroscopy (NMR) cyclic voltammetry chemical shift in parts per million (ppm) downfield from tetramethylsilane OR NMR) or 85% H3PO4 (31P NMR) day(s); doublet (NMR) deuterium deuterium doublet of doublets decomposed dehydroacetic acid; dehydroacetate diamine; diamine backbone l,2-bis[(o-methoxyphenyl)phenylphosphino]ethane N,N'-dimetyl formamide 1,2-bis(dimethylphosphino)ethane dimethyl sulfoxide X DPPE dt DTP DTPA E ECD EDTA edta4" EI en eq. Et eV FAB fac g h Hma Hdpp H3apa H2epa HDTP HMPAO H2ppa HPO H2P02 H4P2O4 HIDA HPLC HSAB Hz IDA 1,2-bis(diphenylphosphino)ethane doublet of triplets methylene diphosphonate N,N,N',N",N"-diethylenetriaminepentaacetic acid potential equatorial peak potential N,N-2-ethylenediylbis-L-cysteine diethyl ester N,N,N',N'-ethylenediaminetetraacetic acid N,N,N',N'-ethylenediaminetetraacetate electron impact bombardment (in mass spectrometry) ethylenediamine equivalent(s) ethyl, -C2H5 electron volt(s); equivalent to 1.60 x 10~19 Joule fast-atom-bombardment (in mass spectrometry) facial geometric isomer gram(s) hour(s) maltol, or 2-methyl-3-hydroxy-4-pyrone l,2-dimethyl-3-hydroxy-4-pyridinone N,N'-3-azapentane-1,5-diylbis(3-( 1 -iminoethyl)-6-methyl-2#-pyran-2,4(3#)-dione) N,N'-ethylene-diylbis(3-( 1 -iminoethyl)-6-methyl-2i/-pyran-2,4(3#)-dione) hydroxymethylene diphosphonate hexamethylpropylene amine oxime N,N'-propylene-diylbis(3-( 1 -iminoethyl)-6-methyl-2//-pyran-2,4(3//>dione) (o-hydroxyphenyl)diphenylphosphine bis(o-hydroxyphenyl)phenylphosphine P,P,P',P'-tetrakis(o-hydroxyphenyl)diphosphinoethane hepatobiliary iminodiacetic acid; N-[N'-(2,6-dimethylphenyl)-carbamoylmethyl] iminodiacetic acid high-performance layer chromatography Hard and Soft Acids and Bases hertz iminodiacetic acid xi 1 i.v. IR i-Pr J K k L LSIMS m M m m-|A MAG3 Me mer MIBI min mol mom m.p. MS m/z n-Bu v NMR Oketo ^oxy ORTEP it Ph current intraveneously infrared wo-propyl, -CH(CH3)2 coupling constant (NMR) Kelvin (=°C + 273.15) kilo (103); rate constant liter(s); ligand liquid secondary ion mass spectrometry milli (IO-3); multiplet (NMR); moderate (IR) metal (center); moles per liter (concentration unit); mega (106); a nucleus (NMR) metastable (isotope) meta, signifies that two substituents in a disubstitued benzene are 1.3 to each other micro (10-6); prefix for a ligand bridging over two metal centers mercaptoacetylglycylglycylglycine methyl, -CH3 meridional geometric isomer 2-methoxy-2-isobutyl isonitrile minute(s) mole(s) methoxymethyl melting point mass spectrometry mass-to-charge ratio (in mass spectrometry) normal butyl stretch vibration nuclear magnetic resonance ortho, signifies that two substituents in a disubstitued benzene are 1,2 to each other ketonic oxygen enolate oxygen Oak Ridge Thermal Ellipsoid Program para, signifies that two substituents in a disubstitued benzene are 1.4 to each other side-by-side bonding phenyl, -C6H5 Xl l ppm py R Rf s a SD t tl/2 TEAP TLC TMEDA trans w X X' y z parts per million pyridine organic group: alkyl or aryl group retention factor singlet (NMR); strong (IR); second(s) head-to-head bonding standard deviation triplet (NMR); time half-life tetraethylammonium perchlorate thin-layer chromatography N,N,N',N'-tetramethylethylenediamine Latin, "across", or the opposite side weak (IR) for halogen: a halide anion or a halo ligand; for donor atom; a nucleus (NMR) a nucleus chemically equivalent to nucleus X, but magnetically non-equivalent to it (NMR) year(s) an oxo ligand; an organoimido ligand xiii To my parents and to my wife, with much love xiv Acknowledgments It is my pleasure to thank Professor Chris Orvig for his guidance, support, encouragement, and patience throughout the course of this project, as well as for his accommodation to independent thoughts and freedom of individual interests in his group. Thanks are extended to Professor Don Lyster and his assistants, especially Terri, Anne, Can and Magie, at VGH, for their valuable efforts regarding the labeling and biodistribution studies. Thanks to all the members of this group, the past, especially Zaihui, Gordon, Hahn, and Martha, for their help in getting me into the state of "doing chemistry in English", and to Shuang and Veranja, for our discussions about chemistry, to Lucio for chemistry and his cheerful friendship, to the present members, the "harmonious coffee break" crowd, especially to Mark, Yan, Pete, Li-Wei, Ernest, Ed, Ika and Marco for their help with this thesis and/or helpful discussions. For their time, careful proofreading and suggestions, I would like to thank Professor R. Thompson, Professor A. Bree, and Professor C. Fyfe. Acknowledgment is made to Dr. Steven Rettig for his determination of all of the crystal structures reported herein, to the department support staff, most notably Marietta, Liane, Mr. Peter Borda, and technical staff of the mass spectrometry laboratory. Thanks to Dr. Xiaoliang Gao for introducing me to Canada, and for his help in NMR. Thanks to Guy for providing his ethylene backbone to my P2O4 chemistry, and for help in NMR, thanks to Drs. Nick Burlinson and Orson Chan for the NMR courses. Thanks to Shihua (Xia) for running TGA with my samples. I also thank Professor B. R. James and his group, Professor M. D. Fryzuk and his group, and Professor D. J. Berg for helpful discussions and suggestions. Thanks invariably go to my father, who encouraged me to stand higher and see further, to my mother, who always provides her children with her love and care, and ultimately to my wife, for her understanding, encouragement, support, and much more. XV Chapter One General Introduction 1.1 Background A radiopharmaceutical, is "a chemical substance, that contains radioactive atoms within its structure and is suitable for administration to humans for diagnosis or treatment of disease."13 Along with contrast agents for magnetic resonance imaging, radiopharmaceuticals belong to the discipline of nuclear medicine.* In practice, a formulated radiopharmaceutical is administered to a patient (usually by intravenous injection) with the goal of having the radionuclide localize within a specific organ or target tissue. Subsequently, with the diagnostic pharmaceutical localized, the emission of the penetrating y radiation from the organ can help to visualize the organ with an externally positioned detector (imaging); with the targeted therapeutic pharmaceutical, the emitted a or |3 particles can kill the diseased cells in the organ within a short range. Presently, the majority of available radiopharmaceuticals (95%) are for diagnosis.2 For diagnostic imaging, the properties of the radionuclide to be considered are the decay mode, photon energy, half-life and chemical activity.lb The desired decay mode should yield photons only — emitted a and P particles being completely absorbed by tissue, rendering a high radiation burden to the patient. The energy of the photon is ideally between 100 to 200 keV, high enough for the photon to readily penetrate and escape from the body, but low enough for the photon to be efficiently collimated. The half-life of the radionuclide should be long The Board of Trustees of the Society of Nuclear Medicine defines nuclear medicine as "the medical specialty which utilizes the nuclear properties of radioactive and stable nuclides for diagnostic evaluation of the anatomic and/or physical conditions of the body and provides therapy with unsealed radioactive sources."18 1 enough to allow labeling, administration, localization, and scintigraphic measurement without significant losses in activity, but should be relatively short to minimize the burden. The ideal radionuclide should have chemical properties that allow it to form different chemical compounds that localize in individual target organs selectively.lb-3 In addition, the radionuclide should be readily available, preferably, from a generator, since this is much more accessible to clinics or hospitals, in contrast to a cyclotron or a reactor. There are more than twenty elements that have radionuclides with useful properties for application in diagnostic nuclear medicine.lc A radiopharmaceutical which is an organic compound may be preferable, due to the similarities in composition and structural features to those in biological systems. Therefore, non-metal radionuclides, such as 14C, 3 2P, 1 2 3 I , 1 2 5 I , 1 3 1 I , etc., which are easily incorporated into organic molecules, are used in the clinic. However, there are some restrictions which limit their diagnostic applications.* Also, there are many metal isotopes bearing the essential physical properties of radiation for diagnosis or therapy as well. The metal isotopes used in diagnosis include 57Co, 67Ga, 68Ga, 99mTc, ^ I n , 113mIn, 197Hg, 203Hg, 20lTl. l f These isotopes can be used in radiopharmaceuticals because they can be appropriately coordinated by organic ligands, or tagged to a carrier molecule, such as an antibody. In other words, they can be "dressed in organic clothes" — such modification imparts the properties of the organic compounds to the metal complexes containing these isotopes. With a half-life of 6.02 h and radiation energy of 141 keV, "mTc j s a n idea\ isotope among all isotopes used in y-ray imaging. The relatively short half-life allows large doses (up to 30 mCi**) to be applied, so as to attain high resolution in imaging without creating too much burden to the patients. The decay of "mTc g i v e s 0ff a monoenergetic y-ray, with no accompanying a or B particle emission. Its daughter 99Tc, in turn, gives a negligible dose of While iodine isotopes either give additional undesired high radiation dose from p-emission (131I), or emit only a weak photon (125I), or are limited by availability (l23I),l° the isotopes 14C and 32P have no value as diagnostic imaging agents, but are very useful for studying the metabolic and pharmacokinetic pathways of drugs.le Radioactivity is expressed as disintegrations per second (dps) or becquerels (Bq), or as Curie (Ci). The conversion equations are: ^ 1 Bq = 1 dps l C i = 3.7xl01 0dps(Bq) lBq = 2 .7x lO- n Ci 2 radiation.* The energy of the radiation is sufficiently high for y-rays to reach the external camera, and sufficiently low to allow collimation by lead.4 However, 99mTc n a ( j n o t foun(i application in routine diagnostic use until the 9 9Mo/"mTc generator1 e and the Anger scintillation camera were developed in the late 1950s; the former made the isotope easily available, while the latter made y-ray imaging possible. Nowadays, obtaining 99mxc from m e generator, (as 99mTc04_ in saline, vide infra) is often referred to as "milking", since it is as easy as "milking a cow". Therefore, 99nrrc has been the preeminent radionuclide in the diagnostic radiopharmaceutical industry and will become more important;5 in 1982, it accounted for more than 80%,6 while in 1990, more than 90%,7 of all diagnostic radionuclide use in U.S.A. The chemical properties of technetium are advantageous, too. Positioned in the middle of the transition elements, technetium, the second row member of group 7, exhibits a wide range of chemistry. Oxidation states from -I to VII are all known for this element, giving diverse options to incorporate the 99HITC y-emitter into a variety of chemical forms, i.e., to dress the nuclide in various organic clothes. With a low tendency to bind organic ligands, 99mTc04~ itself has very limited application in imaging some types of lesions, e.g. brain and thyroid.8 It is a mild oxidant, easily reduced by moderately weak reducing agents, such as ascorbic acid or hydrochloric acid. A technetium center in a reduced oxidation state is able to be bound, and subsequently stabilized, by an appropriate organic ligand** (vide infra). Furthermore, the "cloth" of the Tc center in terms of ligated organic compounds, plays an important role in stabilizing and constructing the complex, since it determines if the 99mTc complex can function as a radiopharmaceutical (vide infra). There are a number of FDA***-approved 99mTc-radiopharmaceuticals on the market.3,7 However, most of them are " m i c complexes with unknown structures, such as agents for kidney imaging (Tc—DTP A), bone scanning (Tc—diphosphonates), and heptobiliary function Due to its long half-life (2.12 x 105 y), "Tc has very low specific radioactivity; 20 mCi of 99mTc is equivalent to 6.5xlO- 8mCiof"Tc. 4 * There are reported Tc(VII) complexes, such as Tc03Cl(bipy), bipy = 2,2'-bipyridine.9 The Food and Drug Administration of the United States. 3 FL ,-N-N ^T C C ^ C ^ I C*. n + N. N R -R=-CH2C(CH3)2(OCH3) Tc-MIBI a T^S^I \ CH3 Tc-Cl(CDO)3MeB EtOOC A j o l ^- .e :€r COOEt Tc-ECD 0 0 H Tc-HMPAO r-t° n 2-coo-TC-MAG3 Figure 1-1. Marketed 99mTc-radiopharmaceuticals with known structures.10 measuring (Tc-HIDA).* A few agents with known structures have been introduced since the early 1980s. These include Tc-MIBI (Cardiolite™) and TcCl(CDO)3MeB (Cardiotec™) for evaluating myocardial perfusion, Tc-HMPAO (Ceretec™) and Tc-ECD (Neurolite™) for cerebral perfusion, and TC-MAG3 (TechneScan MAG3™) for kidney function (Figure l-l).10 Several reviews present more detailed discussions about these clinically used, and the other potential, 99mTc-radiopharmaceuticals.3'4'7'11,12 There are only empirical opinions so far about Structures for DTPA (N,N,N',N",N"-diethylenetriaminepentaacetic acid), diphosphonates, HIDA (hepatobiliary iminoacetic acid) are as follows:7 R .H HOOC-HOOC— -COOH ^—COOH DTPA 1 ° R , ' J V ' R , ° '-COOH OH OH MDP (R = H) HMDP (R = OH) 4 R2 HIDA the in vivo distribution of a pharmaceutical based on its structural features. For instance, lipophilic monocationic species can accumulate in the myocardial tissue; small, neutral, lipophilic molecules may be able to cross the blood-brain barrier.3,11 Radiopharmaceuticals, foreign to the human body, are eventually excreted; it is the lipophilicity of the compound that mainly controls whether the excretion is through liver or kidney.13 Even though a number of imaging agents for various targets already exist, the demands from nuclear medicine practitioners are not entirely fulfilled and the potential of "mTc m radiopharmaceuticals is far from completely exploited.14 Therefore, Tc-related research for better 99mTc-radiopharmaceuticals is in full swing worldwide. The criteria for an ideal radiopharmaceutical include: a high activity ratio of target organ to non-target areas of the body; rapid elimination of the radiopharmaceutical from the bloodstream; optimum retention time in the organ and complete eventual elimination from the body.1,3,15,16 None of the existing 99mTc-radiopharmaceuticals meets all the requirements without limitations.3 The biological behavior of a radiopharmaceutical, resulting from the chemical/physical interactions between the molecules of the radiopharmaceutical and those of the biosystems (bloodstream, target organ, etc.), is determined by the structural features of radiopharmaceutical, i.e., there are structure-activity relationships involved in the in vivo localization and clearance of " m T c -chelate complexes.3,4'11"13 For a perfusion agent, the biodistribution of the radiopharmaceutical is determined by its shape, net charge, size, redox properties and lipophilicity.17,18 Alternatively, for the so-called bifunctional approach, the localization of the complex molecule depends on the properties of the essential functional group.13,17* However, the mechanism through which the structure of the radiopharmaceutical governs the in vivo distribution and other biological behavior is not yet well understood.** Nevertheless, the development of new and more efficacious ""^-radiopharmaceuticals will depend on structural characterizations of the Tc complexes used or to be used in the clinic,19 since the better understanding of The ligand associated with this approach is a bifunctional chelating agent, which contains a metal binding donor-set as well as a function group to bind a receptor in vivo. It still remains very difficult to predict the biodistribution of a new 99mTc radiopharmaceutical containing small ligands due to inadequate information. Relatively minor variations in the ligands can often lead to an empirical relationship between the properties of the whole complex and its biodistribution. From such relationships, predictions can be made and tested.13 5 structure-activity relationships will be possibly pursued. Needless to say, coordination chemistry of Tc is crucial in the development of 99mTc-radiopharmaceuticals which are usually coordination complexes of 99111X0 in various oxidation states stabilized by organic ligands. The chemistry of technetium is less developed, compared to its closest neighbors, Mo, Ru, Mn, Re. The reasons for this are the following: (1) the element, discovered in 1937 as the first artificial element, is still young; (2) all the isotopes of this element are radioactive; (3) the price of the most available isotope used to be very high.* As a result, there was little interest in the research relevant to this element until I960,19 academic curiosity being the only driving force,8 even though the first gram quantities of 99Tc were made available in 1952.21 This situation remained unchanged until the metastable isotope, 99mrrC) found the first medical application in 1962.13 The total number of reports on Tc chemistry and 99nrr/c_racij0pharmaceuticals in Chemical Abstracts expanded rapidly from 1950 to 1977, but dropped from then to 1980.19 The drop was due to a decline in reports on 99mTc .radiopharmaceuticals, presumably as uncharacterized 99mrpc compounds, rather than on relevant Tc chemistry. It was then realized that significant progress in 99nrr;c-radiopharmaceuticals must rely on advances in Tc coordination chemistry.20 Since then the synergistic interaction between inorganic chemistry and nuclear medicine has become so strong that regular dialogues between chemists and nuclear medicine practitioners all over the world take place for the development of new 99mjc. radiopharmaceuticals. Four such dialogues have been held at approximately four-year intervals since 1982,22~24** and each side now has a better understanding of the problems that face the other, i.e., of knowing what inorganic chemists are capable of doing for nuclear medicine practitioners, and realizing what the specific requirements of radiopharmaceuticals are.22 While 99mxc-radiopharmaceutical studies have progressed "from the application of crude, largely uncharacterized chemistry to the development of 99mjc . r a£}j0pj i a rn iaceu ticai s based on a solid structural chemistry foundation,"25 the basic chemistry of Tc has become much richer, * In the U.S.A., the price of " T c was $ 2800 per gram in 1960 but $ 55 per gram in 1982. Currently " T c is produced in kilogram amounts by fission of uranium?0 The Fourth International Symposium on Technetium in Chemistry and Nuclear Medicine was held in September 12-14, 1994, in Bressanone, Bolzano, Italy. 6 resulting in a more profound basis for further 99mTc-radiopharmaceutical studies. Technetium chemistry has been reviewed frequently since 1980.6,8'9'11'20'26"30 Understanding the 9 9Mo/"mTc generator points to the requirements for the preparation of ""^-radiopharmaceuticals and the reason for the use of " T c in chemical studies. The generator, like any other radionuclide generator, consists of a long-lived parent radionuclide (99Mo) that decays to yield short-lived daughter (99mTc). The parent and the daughter are not isotopes of the same element, therefore separation is possible. The parent nuclide in the 9 9Mo/"mTc generator is present as "M0O4 2 - dianion, which is loaded on to an alumina (AI2O3) column. Upon decay, the molybdate changes to pertechnetate (99mTcC>4_ and "TCO4" ) (Scheme I), which is uninegative and eluted by saline (0.9% NaCl in H2O). The molybdate dianion is absorbed by the surface of alumina due to its higher charge. The total concentration of the pertechnetate in the eluant is usually in the range 10'8 to 10-6 M.4 99mTc04" P" emission / ^v isomeric transition ti/2 = 66.02 h / ^ 7 6 % \ t i /2 = 6.02 h "Moo/ !24S .- ^ccy Pemissioc , * » R U O 4 P" emission \.V1 = 2.12 x 105 y Scheme 1-1. The decay for the mother and daughter nuclides of the 9 9Mo/"mTc generator.28 As the concentration of " m f c from a generator is very low, it is impossible to structurally characterize, or even identify the nature of the products from 99mTc04-, without any knowledge of the corresponding macroscopic Tc chemistry. The solution is obviously to develop a macroscopic Tc chemistry beforehand, where we can take advantage of the many physical facilities available to fully characterize the compounds. Macroscopic Tc chemistry is always investigated with the isotope 99Tc, a P-particle emitter (ti/2 = 2.12 x 105 y, E = 292 keV), due to its availability (vide supra) and much longer half-life; while the concentration of 7 99mTco4- eluted from the generator is only nanomolar, macroscopic amounts of " T c are commercially available. If macroscopic amounts of 9 9 mTc04" were available, it would be extremely dangerous to handle; it is much safer to handle macroscopic amount of " T c samples.* The P-emission from milligram quantities of " T c is effectively stopped by solutions and regular glassware,11 and the Bremsstrahlung is negligible due to the low energy of the P-emission.4 As an isotope of the same element, " T c should be chemically identical to " m T c : any disparity in practice would be most likely due to the difference in the concentrations available for the two isotopes.4 Comparison of the properties of 99mrj;c a n ( j 99j c compounds makes it possible to understand better the relationship between chemical structure, physicochemical properties and biological behavior after administration to humans or experimental animals.18 The element rhenium, the congener of Tc in the third transition metal series, is often covered by the Tc chemistry groups, since this element possesses similar chemical properties to those of technetium due to the "Lanthanide contraction", and Re is an environmentally preferable, non-radioactive element; in addition, rhenium chemistry has a related interest as its isotopes 186Re and 188Re are suitable candidates in radiotherapeutic applications.31'32** The benefits from such similarity are mutual. On one hand, Re complexes are excellent models for Tc complexes because Tc and Re complexes with same ligands have identical coordination parameters and the reaction conditions suitable for synthesis of Re complexes are often feasible for synthesis of the analogous Tc complexes. On the other hand, based on the use of a 99mjc imaging agent, it is possible to formulate an analogous 186/188R6 therapeutic To illustrate this, one can look at the definition equation of activity: ^ A = -dN/dt = kN = 0.693 N/tm BqCs"1) here k is the decay rate constant, t is the decay time, and N is the number of the atoms. Since 1 Ci of activity is equivalent to 3.7 x 1010Bq, 10 mg of 99mTc has the activity of A = [0.693 x 6.02 x 1023 (1/mol) x 0.010g/(99g/mol)]/[6.02 h x 3600 (s/h) x 3.7 x 1010 (Bq/Ci)] = 5.26xl04Ci while 10 mg of "Tc has the activity of A = [0.693 x 6.02 x 1023 (1/mol) x 0.010g/(99g/mol)]/[2.12 x 105 (y) x 365 (d/y) x 24 (h/d) x 3600 (s/h) x 3.7 xlO 1 0 (Bq/Ci)] = 1.70xlO"4Ci ** 186Re is especially attractive: it emits P particles (E = 1.07, 0.93 MeV) as well as y-rays (E = 137 keV) with half-life 90 h. The biodistributions of 186Re radiopharmaceuticals can be readily monitored by the same y-camera instrumentation employed for 99mTc agents.31 188Re could be useful in treating larger tumors as the (3-particles emitted can penetrate as far as 8 mm; in addition, a 188W/188Re generator has been developed.32 8 radiopharmaceutical that is rapidly eliminated from the blood stream and extracted by a target organ.* In using P-emitting therapeutic radiopharmaceuticals, it is particularly important to minimize the radiation dose to nontarget tissues.33 Differences do arise between the two elements, especially in redox chemistry. One fact we should be aware of when modeling Tc chemistry on that of Re, is that rhenium complexes are more thermodynamically stable in higher oxidation states than are their technetium analogs31. Review articles on rhenium coordination chemistry have appeared periodically.34"36 There are specific tasks for chemists in macroscopic Tc/Re chemistry. The first is to find or design ligands to stabilize Tc/Re centers in various oxidation states. High thermodynamic stability or kinetic inertness of a 99nrr/c o r 186/188R6 radiopharmaceutical is of paramount importance in quality control,111'17** especially when the relationship between structure and biodistribution is to be studied. Consequently, we want to synthesize Tc/Re complexes that are stable enough to resist hydrolysis and/or dissociation. Of course, for a given metal center, the stability of a complex will depend on the bound ligands. Existing ligands have provided chemists a wide selection from which to stabilize Tc/Re centers, while new ligands are constantly being investigated for the same purposes. There are some fairly general guidelines37'38 that we can follow in choosing or designing ligands to form stable Tc/Re complexes. For instance, the ligands should have an appropriate donor-set for the metal centers. The HSAB Principle*** is useful in this regard since the hardness of a metal center (Lewis acid), varies with the oxidation states. Ligands with hard donors (O, N) stabilize metal centers in higher oxidation states, whilst ligands with soft donors (S, P, As, C) stabilize metal centers in lower oxidation states. Furthermore, the chelate effect**** can be taken into account, i.e., ligands containing more than one donor are preferred in the effort to stabilize the given * Of course, a 99mTc radiopharmaceutical that possesses such properties should be developed first. The analogous 99mTc and 186/188Re complexes are identical and cannot be distinguished by a biological system. For instance, based on the fact found in the 1970s that 99mTc diphosphonates accumulate in the skeletal metastases of cancer patients, an agent 186Re-HEDP (hydroxyethylidene diphosphonate) has been prepared for palliation of bone cancer pain.10 * Unstable complexes would dissociate and then hydrolyze/disproportion to form MO2 and/or MO4" which show different distributions. The Principle of Hard and Soft Acids and Bases (HSAB Principle) states that hard acids prefer to bind to hard bases, and soft acids prefer to bind to soft bases.39 Complexes containing chelate rings are more stable than those that have same donors but without the chelate rings. This is the chelate effect. 9 metal centers. In addition, the positions of the donors in the ligands should be arranged so that five- or six-membered chelate rings will be formed upon coordination.* Finally, the complexes should be coordinatively saturated.17 Of course, to know how much the ligand will stabilize a given metal center, we must first synthesize the complex. There are two main synthetic routes to Tc complexes: the reduction/coordination route,** through which pertechnetate is reduced and then the reduced metal center is coordinated in the presence of both a reducing agent and a ligand; and the ligand-exchange route, in which an intermediate species of correct oxidation state, usually a complex of monodentate ligands, is prepared first, followed by substitution with new ligands.20 The ligand-exchange route dominates in macroscopic Tc chemistry since it offers a cleaner reaction due to a pre-specified oxidation state, and starting materials of Tc in various oxidation states are easily made ([TcOX4]_, [TcX6]2" or others). Although the reduction/coordination route is not the most widely used synthetic route in macroscopic synthesis, it is preferable here since it is this route that is applied in the preparation of 99mTc-radiopharmaceuticals. Rhenium complexes have been synthesized through the same two routes, with the ligand-exchange route being the most widely used. Although pertechnetate is easily reduced, the product varies depending on the nature of the reducing agent, the ligand available, the reaction medium and physical conditions such as temperature and pH, etc.9 Thus the second task for chemists is to find good reducing agents. The unwanted brown precipitate TcC>2 XH2O, for instance, is the most common co- or by-product when TCO4" is reduced in an aqueous solution without a suitable ligand or at high pH. The reduction of TCO4" with the same reducing agent, e.g., Na2S204, can yield products of different oxidation states depending on the presence or absence of a ligand: TCO2XH2O without a ligand, [TcO(SCH2CH2S)2]" with ethane-1,2-dithiol, Tc(acac)3 with pentane-2,4-dione, and [Tc(CNC4H9)6]+ with f-butylisonotrile. In addition, the reduction of TcCV with different reducing agents but the same ligand can generate different products.9 A few articles Complexes that contain five- or six-membered chelate rings (including the metal ion ) are usually more stable than those with same donors but less-than-five- or more-than-six-membered chelate rings due to less strain.37'38 It is often referred to as reduction/ligand-exchange or reduction/substitution route in other references.20 10 have been published dealing with the influences and efficiency of reducing agents for macroscopic synthesis40 or for 99mTc-radiolabeling.41 Commercial 99mTc-radiopharmaceutical kits almost invariably use Sn(II) as the reductant, as it is rapid and effective at room temperature. However, Sn(II) is easily oxidized to stannic ion (Sn(IV)), which is either rapidly hydrolyzed,41 or forms heteronuclear complexes with " m j c and the ligands.3 Sn(IV) (or Sn(II)) also competes with " m T c for ligands since tin is in large excess;42 it has been noticed that Sn(IV) and Tc(IV) are physically very similar.20 Electrolytic reduction of 99mTc04_ may potentially avoid the sources of impurities caused by addition of reducing agents;43 however, this method has not yet found application in the clinic due to requirement of additional equipment.3 Efficient chemical reducing agents, with the capability to reduce the 99mTc04" anion quickly in high yield without the above-mentioned problems, are desired in the preparation of radiopharmaceuticals so as to minimize the radiochemical impurities of the unbound 99mTc04" anion and/or the hydrolysis product 99mTc02xH20. The third task for chemists is the need to modify the ligands that form stable Tc/Re complexes. The function of a good ligand is more than just stabilizing the metal center; the formulation and structural features of the radiopharmaceutical, such as the size, shape, net charge, and lipophilicity, are all determined by the ligand. Thus, a ligand which is easily modified, without a significant loss of affinity to the metal, may result in an easy adjustment of the properties of the derivative complexes, the potential radiopharmaceuticals, to optimize the in vivo biodistribution. For instance, addition of pendant hydrophilic or lipophilic groups may influence the accumulation, retention, and excretion pathway in vivo. The choice of these property-controlling groups or any other biologically significant groups is based on the knowledge of the developments in organic chemistry, biochemistry, and pharmacology, as well as on the experience of previous studies. The fourth and most important task for chemists is to provide more structures of the Tc/Re complexes potentially important in radiopharmaceutical studies. The combination of this structural information with the results of in vivo biodistribution studies of the "nvT/c analogs, will definitely increase our understanding of structure-function relationships, which in 11 turn, would benefit the design of future pharmaceuticals. Three reviews have appeared on the structures of Tc complexes,19'44'45 while new structurally characterized Tc complexes are reported more frequently now than ever before. The final task is to develop fast, simple procedures for the in vitro microsynthesis of 99mTc complexes so that a nuclear medicine practitioner, without too much training in chemistry, will be able to prepare the "mTc complex in a radiopharmaceutical kit* for further in vivo evaluation. The preparation of the 99mTc-radiopharmaceutical is preferred to proceed in one easy step, directly from the reaction of 99mTc04", the reducing agent and the ligand.3 Therefore, searching for new 99mTc-radiopharmaceuticals is an interdisciplinary challenge, in which chemists are heavily involved: "nuclear chemists to provide isotopes, organic chemists to provide new synthetic procedures in assembling the donor-set and backbone, as well as the methods of ligand modification, inorganic chemists to furnish the methodology required for the new metal-based radiopharmaceuticals, physical chemists to provide structural information related to the radiopharmaceuticals, biochemists to give insight into uptake and metabolism of the radiopharmaceuticals, radiopharmacists to certify the potential of the radiopharmaceuticals".2 Cooperation is crucial between the researchers in all chemical sub-disciplines and in radiopharmaceutical divisions.14 A radiopharmaceutical kit is simply a sterile reaction vial containing the nonradioactive chemicals required to produce a specific radiopharmaceutical after reaction with 99mTc generate elute. The primary contents in the kit are a complexing agent (ligand) and a reducing agent, usually stannous chloride.1*1 12 1.2 The Project In early studies of ""^-radiopharmaceuticals, only nuclear medicine practitioners were involved, and they focused on the two steps, in vitro labeling and in vivo distribution, leading to the final product from 99mTc04- (the only convenient source of 99mTc, Figure 1-2). The structure, or even the nature, of the 99mTc-labeled complex was unknown. However, knowledge of structure is a must in understanding the relationship between structure and biodistribution of radiopharmaceuticals. Basic coordination chemistry studies of Tc (99Tc) and Re with the same ligand is important since one may possibly construct a clear structural picture of the 99mTc-radiopharmaceutical, in addition to an effective labeling method. The importance of Tc and Re coordination chemistry in the development of 99mTc-radiopharmaceuticals is like the pier and base of a bridge (Figure 1-2); therefore, the basic coordination chemistry of these two elements is investigated in this project. The project, a joint program in the Department of Chemistry at UBC and the Division of Nuclear Medicine at the Vancouver Hospital, Oak Street Site, for the development of new 99mTc-radiopharmaceuticals involves the following strategic steps: 1. Macroscopic synthesis and characterization of Re complexes with both known and new ligands, following selection or design and synthesis of the ligands. 2. Macroscopic synthesis and characterization of the analogous 99Tc complexes with the same ligands. 3. In vitro microscopic synthesis of " T c complexes from 99mTc04" (reduction/ coordination route) to label the ligands with "mTc. 4. In vivo biodistribution studies of the "mTc complexes starting from 99mTc04". 13 Figure 1-2. 99mTc-radiopharmaceutical studies and Tc and Re coordination chemistry. 14 The project started from the base and the pier of the bridge: macroscopic Re and Tc chemistry, of which the core of this thesis is composed. We are interested in synthesis and characterization of new, stable Re and Tc complexes and in development of new modifiable ligands. In doing this, we hope that we can make our contribution to the basic Re/Tc coordination chemistry based on our observations. As for the goal of application in radiopharmaceutical design, it is our intention to develop neutral or cationic complexes suitable for potential brain or heart imaging (vide supra). Investigating macroscopic pentavalent Re/Tc chemistry through substitution is worth pursuing, with this oxidation state having found unanticipated importance in radiopharmaceutical chemistry.6 The precursors, as complexes of the cores M=03+ ,9 '27 M=N2+,46"48 M=NR3+,48,49 with monodentate ligands, are all readily available. For example, the anion [TcOCU]" is formed when the TcC>4_ is reacted with cold hydrochloric acid, and the most commonly used starting material [(rc-Bu)4N][TcOCl4] is isolated almost quantitatively upon addition of [(n-Bu)4N]Cl.50 Substitution of chloro ligands with chelates generally proceeds under mild conditions, yielding 5-, 6-, or 7-coordinate complexes in which the Tc0 3 + core remains intact.20 The complexes containing these cores are usually found to be diamagnetic; therefore the two d-electrons are paired in the non-binding dxy orbital.9'35 Thus the d2 configuration behaves as a closed shell.9 The order of the terminal metal-oxo bond in monooxo complexes exceeds two since both the oxygen p x and p y orbitals overlap with the Tt-symmetric dyz and dzx orbitals of the metal. The oxotechnetium(V) and oxorhenium(V) centers can be stabilized by a large variety of ligands. Bidentate, tridentate, tetradentate, or other multidentate ligands containing hard (O, N) or soft (S, P, As) donors are used frequently.9,27 '30 The 99mTc(V) radiopharmaceuticals are commonly found in clinical studies. Chapter 2 presents an effort to extend the properties of two bidentate ligands, Hma and Hdpp (Figure 1-3), the representatives of common chelates used by the Orvig group, 3-hydroxy-4-pyrones and 3-hydroxy-4-pyridinones respectively, to the pentavalent Re/Tc oxo complexes. Unique combinations of biologically significant properties (water solubility, 15 hydrolytic stability, and lipophilicity) have led to the wide use of tris(maltolato)aluminum(III) in the study of Al neurotoxicity,51 and to the potential application of a 67Ga complex, tris(l-p-methoxyphenyl-2-methyl-3-oxy-4-pyridinonato)gallium(III), in heart imaging.52 The chelates are either naturally occurring (Hma), or easily synthesized (Hdpp) and easily modifiable.53 Hma, X = O H3apa, Y = CH2NHCH2 Hdpp X = NCH3 H2epa, Y = none H2ppa, Y = CH2 Figure 1-3. Bidentate 3-hydroxy-4-pyrone/4-pyridinone (left) and multidentate Schiff base (right) ligands. Chapter 3 describes attempts to utilize multidentate Schiff base* ligands (Figure 1-3), to stabilize Re/Tc centers. The formation of many chelate rings would be expected to enhance the stability of the complexes. The ease of synthesis of these types of ligands has led to extensive investigation of their technetium chemistry.26 In developing other novel ligand systems, the difficulty in choosing reducing agents for 99mTc-radiopharmaceuticals was taken into account along with the stability requirement. The existence of some organic compounds that contain donor atoms and are capable of reducing provides a comprehensive solution. For instance, triphenylphosphine, is a good ligand as well as a reducing agent. The phosphine phosphorus, a good a-donor and also a weak n acceptor, is subject to oxidation. In the synthesis of Tc complexes from TCO4" (e.g., [TcCl4(PPh3)2],54 [TcNCl2(PPh3)2],46 [Tc(NPh)Cl3(PPh3)2)],49 the phosphine functions as a ligand as well as a reducing agent. The complexes are all air/moisture stable. These complexes have a combination of hard and soft donors with relatively hard metal centers. However, there are no A Schiff base is a substituted imine formed by condensation reaction of a primary amine (RNH2) and a ketone or an aldehyde. 16 reports on their possible application in radiopharmaceutical studies, presumably due to the hydrolyzable monodentate ligands (e.g. CI).55 The only way to form a chelated complex and to reduce the number of the chloro ligands is to modify a phosphine by functionalization of the alkyl or aryl arms with anionic groups. Davison and coworkers synthesized three Tc(III) complexes with an umbrella tetradentate triprotic phosphinotrithiol (PS3) ligand.56,57 Refosco et al. have obtained neutral Tc(III) complexes of [Tc(PX)3] with functional group X being thiol S or phenol O,58 and cationic Tc(III) complex [Tc(PNH)2(PNH2)]+ from o-phosphinophenylamine.59 The PN ligand formes Tc(V) and Re(V) oxo complexes, [MOX(PNH)2] and [MO(OR)(PNH)2] (M = Tc/Re, X = CI/Br, R = H, alkyl or aryl group) as well.60 The results reveal that, as ligands, these functionalized phosphines are all able to stabilize the metal center of intermediate oxidation states, while they also function as reducing agents in the preparation of some of the complexes. Among the above functionalized phosphines, the PN and PO systems each has a combination of one soft phosphorus and one hard amine nitrogen or phenolic oxygen donor. It is interesting that the PN ligand can stabilize metal centers in the oxidation states of both three and five,59'60 while the PS ligands form M(III) complexes only.58 That is to say, the ligands with a combination of hard and soft donors show more flexibility in stabilizing metal centers. This is not surprising when considering the existence of so many Re and Tc complexes of various oxidation states that incorporate both P and CI ligands (vide supra). Fryzuk has developed a hybrid ligand system (PNP) that incorporates phosphine (soft) and amide (hard) into one array, and the PNP ligands were used to form complexes with both early transition metal centers and late transition metal ions.61 There is more evidence showing the harmonious coexistence of soft donors and hard donors bound to Tc centers in the intermediate oxidation states. Deutsch's group has synthesized cationic diphos and diars complexes from the reaction of pertechnetate and excess chelate phosphine and arsine ligands. The stable complexes of metal(I, III, V) are in the forms of [Tc(L-L)3]+, [TcCl2(L-L)2]+ or [Tc02(L-L)2]+ (L = P or As). Synthesis and characterization of a number of stable complexes of Tc(V), Tc(III), Tc(II) and Tc(I) with DMPE (l,2-bis(dimethylphosphino)ethane), and biodistribution studies of their 17 99mTc analogs were reported.62 Neither CI nor the DMPE were lost under in vivo conditions. This indicates that at least kinetic inertness of the diphosphine complexes can be enhanced by the chelate effect. >^X Z Z N X = O/N; Y = Cl/OR; Z = O/NR; L = PR3/py Figure 1-4. Proposed neutral metal(IV and V) complexes. Based on this idea of hard-soft combination, a new type of functionalized phosphines PX2 (X is a hard O or N donor), with potential tridenticity and diproticity, has been considered. A number of structural models for neutral metal(IV and V) complexes, incorporating PX2, PX with or without other ligands can be easily built up as shown in Figure 1-4. OH ^ ^ O H H O ^ ^ HPO, x = 1 H4P2O4 H2P02, x = 2 Figure 1-5. Multifunctionalized phosphine and diphosphine ligands. Phenolic oxygen has been chosen as the anionic functional group, so a modified ligand based on triphenylphosphine, a potentially tridentate dianionic H2PO2 ligand, bis(o-hydroxyphenyl)phenyphosphine, has been prepared (Figure 1-5). The double-functionalized phosphine, with a combination of soft phosphorus and two hard oxygen donors, should also stabilize Tc or Re centers in their intermediate oxidation states. Two phenol hydroxyl groups, each easily deprotonated in the presence of a metal ion and/or a base, provides two hard monoanionic donors to bind the metal center and to neutralize its positive charge; its 18 introduction retains the air stability of the functionalized phosphines. Furthermore, the lipophilic phenyl rings are easily modified; therefore, the biodistribution of the "mTc complexes is potentially adjustable. The chemistry of triphenylphosphine derivatives, HPO and H2PO2 systems is demonstrated in Chapter 4. The same idea applied to the diphosphine, P,P'-bis(diphenylphosphino)ethylene, led to another new ligand, H4P2O4 (Figure 1-5). This ligand resembles EDTA in potentially denticity (6) and proticity (4), and formation of a hydrohalide salt. Both ligands form five-membered chelate rings on coordination. Tc(IV) and Tc(V) complexes of EDTA are known, [(H2edta)Tc(|i-0)2Tc(H2edta)]63 and [TcO(edta)]-,20 the former is a six-coordinate binuclear complex and the latter is an anionic seven-coordinate complex. The differences between the two ligands are marked, besides that of reducibility. The H4P2O4 ligand has two phosphine P donors instead of two amine N donors, and it has four phenol groups instead of carboxylic acid groups. Thus, coordination chemistry of this ligand may differ from that of EDTA in binding mode, preference and deprotonation ability. The chemistry of this H4P2O4 ligand, P,P,P',P'-tetra(o-hydroxyphenyl)diphosphinoethane ligand is explored in Chapter 5. As the ultimate goal of this work is to develop new 99mTc-radiopharmaceuticals, we were interesting in seeing how much further we could go with our ligands in their potential application in imaging. As soon as macroscopic Re/Tc complexes were synthesized, the chemistry was translated to the microscopic scale with 99mTc, and the biological studies of the 99mTc complexes were carried out. Chapter 6 gives the preliminary results of the radiolabeling and the in vivo biodistribution studies with selected ligand systems. 19 References (1) Kowalsky, R. J.; Perry, J. R. Radiopharmaceuticals in Nuclear Medicine Practice; Appleton & Lange: Norwalk, 1987, a. p 1; b. p 76; c. p 78; d. p 90;e. p 94; f. p27; g. p 29; h. p 82. (2) McCarthy, T. J.; Schwarz, S. W.; Welch, M. J. /. Chem. Ed. 1994, 71, 830. (3) Schwochau, K. Angew. Chem. Int. Ed. Engl. 1994, 33, 2258. (4) Deutsch, E.; Libson, K. Comments Inorg. Chem. 1984, 3, 83. (5) Carretta, R. F. J. Nucl. Med. 1994, 35, 24N. (6) Jones, A. G.; Davision, A. J. Nucl. Med. 1982, 23, 1041. (7) Jurisson, S.; Berning, D.; Jia, W.; Ma, D. Chem. Rev. 1993, 93, 1137. (8) Clarke, M. J.; Fackler, P. H. Structure and Bonding 1982, 50, 57. (9) Davison, A. In Technetium in Chemistry and Nuclear Medicine; Deutsch, E., Nicolini, M. and Wagner, H. N., Eds.; Cortina International: Verona, 1983; p 3. (10) Abrams, M. J.; Murrer, B. A. Science 1993,261, 725. (11) Clarke, M. J.; Podbielski, L. Coord. Chem. Rev. 1987, 78, 253. (12) Eckelman, W. C; Volkert, W. A. Int. J. Appl. Radiat. hot. 1982, 33, 945. (13) MarziUi, L. G.; Kramer, A. V.; Burns, H. D.; Epps, L. A. In Technetium in Chemistry and Nuclear Medicine; Deutsch, E., Nicolini, M. and Wagner, H. N., Eds.; Cortina International: Verona, 1983; p 63. (14) Lever, S. Z.; Wagner, H. N. In Technetium and Rhenium in Chemistry and Nuclear Medicine 3; Nicolini, M., Bandoli, G. and Mazzi, U., Eds.; Cortina Internatioal: Verona, 1989; p 649. (15) Kronauge, J. F.; Davison, A.; Roseberry, A. M.; Costello, C. E.; Maleknia, S.; Jones, A. G. Inorg. Chem. 1991, 30, 4265. 20 (16) Eckelman, W. C; Levenson, S. M. Int. J. Appl. Radiat. hot. 1977, 28, 67. (17) Parker, D. Chem. in Britain 1994, 30, 818. (18) Wagner, H. N.; Kramer, A. V. In Technetium in Chemistry and Nuclear Medicine; Deutsch, E., Nicolini, M. and Wagner, H. N., Eds.; Cortina International: Verona, 1983; p 161. (19) Bandoli, G.; Mazzi, U.; Roncari, E.; Deutsch, E. Coord. Chem. Rev. 1982,44, 191. (20) Deutsch, E.; Libson, K.; Jurisson, S.; Lindoy, L. F. Prog. Inorg. Chem. 1983, 30, 75. (21) Pinkerton, T. C ; Desilets, C. P.; Hoch, D. J.; Mikelsons, M. V.; Wilson, G. M. /. Chem. Ed. 1985, 62, 965. (22) Technetium in Chemistry and Nuclear Medicine; Deutsch, E.; Nicolini, M.; Wagner, H. N., Eds.; Cortina International: Verona, 1983. (23) Technetium in Chemistry and Nuclear Medicine 2; Nicolini, M.; Bandoli, G.; Mazzi, U., Eds.; Cortina International: Verona, 1986. (24) Technetium and Rhenium in Chemistry and Nuclear Medicine 3; Nicolini, M.; Bandoli, G.; Mazzi, U., Eds.; Cortina International: Verona, 1990. (25) Fritzberg, A. R. In Radiopharmaceuticals: Progress and Clinical Perspectives; Fritzberg, A. R., Ed.; CRC Press, Inc.: Boca Baton, Florida, 1986; Vol. 1; preface. (26) Dilworth, J. R. Trans. Met. Chem. 1990,15, 411. (27) Mazzi, U. Polyhedron 1989, 8, 1683. (28) Seiler, H. G. Metal Ions Biol. Syst. 1983,16, 317. (29) Jones, A. G.; Davison, A. Int. J. Appl. Radiat. hot. 1982, 33, 867. (30) Davison, A.; Jones, A. G. Int. J. Appl. Radiat. hot. 1982, 33, 875. (31) Deutsch, E.; Libson, K.; Vanderheyden, J.-L.; Ketring, A. R.; Maxon, H. R. Nucl. Med. Biol. 1986,13,465. (32) Chem. & Eng. News 1992, September 7, 33. 21 (33) Marzilli, L. G.; Banaszczyk, M. G.; Hansen, L.; Kuklenyik, Z.; Cini, R.; Taylor, A. Inorg. Chem. 1994, 33, 4850. (34) Conner, K. A.; Walton, R. A. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A., Eds.; Pergamon: 1987; Vol. 4; p 125. (35) Rouschias, G. Chem. Rev. 1974, 74, 531. (36) Fergusson, J. E. Coord. Chem. Rev. 1966,1,459. (37) Martell, A. E.; Hancock, R. D.; Motekaitis, R. J. Coord. Chem. Rev. 1994,133, 39. (38) Hancock, R. D.; Martell, A. E. Chem. Rev. 1989, 89, 1875. (39) Pearson, R. G. /. Chem. Ed. 1968, 45, 581. (40) Jones, A. G.; Orvig, C; Trop, H. S.; Davison, A.; Davis, M. A. J. Nucl. Med. 1980,21, 279. (41) Fritzberg, A. R.; Lyster, D. M.; Dolphin, D. H. J. Nucl. Med. 1977,18, 553. (42) John, C. S.; Schlemper, E. O.; Hosain, P.; Paik, C. H.; Reba, R. C. Nucl. Med. Biol. 1992, 269, 269. (43) Russell, C. D.; Majerik, J. Int. J. Appl. Radiat. Iso. 1978, 29, 109. (44) Tisato, F.; Refosco, F.; Bandoli, G. Coord. Chem. Rev. 1994,135/136, 325. (45) Melnik, M.; van Lier, J. E. Coord. Chem. Rev. 1987, 77, 275. (46) Kaden, L.; Lorenz, B.; Schmidt, K.; Sprinz, H.; Wahren, M. Isotopenpraxis 1981,17, 174. (47) Chatt, J.; Rowe, G. A. J. Chem. Soc. 1962,4019. (48) Chatt, J.; Garforth, J. D.; Johnson, N. P.; Rowe, G. A. /. Chem. Soc. 1964, 1012. (49) Nicholson, T.; Davison, A.; Jones, A. G. Inorg. Chim. Acta 1991,187, 51. (50) Davison, A.; Trop, H. S.; Depamphilis, B. V.; Jones, A. G. Inorg. Synth. 1982, 21, 160. (51) Hewitt, C. D.; Herman, M. M.; Lopes, M. B. S.; Savory, J.; Wills, M. R. Neuropath. Appl. Neurobiol. 1991,17, 47 and references therein. 22 (52) Zhang, Z.; Lyster, D. M.; Webb, G. A.; Orvig, C. Nucl. Med. Biol. 1992,19, 327. (53) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1992, 31, 598. (54) Mazzi, U.; de Paoli, G.; di Bernardo, P.; Magon, L. /. Inorg. Nucl. Chem. 1976, 38, 111. (55) Johnson, N. P.; Lock, C. J. L.; Wilkinson, G. /. Chem. Soc. 1964, 1054. (56) Nicholson, T.; Cook, J.; Davison, A.; Jones, A. G. Inorg. Chim. Acta 1994, 218, 97. (57) de Vries, N.; Cook, J.; Jones, A. G.; Davison, A. Inorg. Chem. 1991, 30, 2662. (58) Bolzati, C ; Refosco, F.; Tisato, F.; Bandoli, G.; Dolmella, A. Inorg. Chim. Acta 1992, 201,1. (59) Refosco, F.; Bolzati, C; Moresco, A.; Bandoli, G.; Dolmella, A.; Mazzi, U.; Nicolini, M. J. Chem. Soc, Dalton Trans. 1991, 3043. (60) Refosco, F.; Tisato, F.; Bandoli, G.; Bolzati, C ; Dolmella, A.; Moresco, A.; Nicolini, M. J. Chem. Soc, Dalton Trans. 1993, 605. (61) Fryzuk, M. D. Can. J. Chem. 1992, 70, 2839. (62) Deutsch, E.; Ketring, A. R.; Libson, K.; Vanderheyden, J.-L.; Hirth, W. W. Nucl. Med. Biol. 1989,16, 191 and the references therein. (63) Burgi, H. B.; Anderegg, G.; Blauenstein, P. Inorg. Chem. 1981, 20, 3829. 23 Chapter Two Rhenium(V) and Technetium(V) Complexes of Bidentate (0,0) Monoprotic Ligands 2.1 Introduction We have focused on modifiable ligands, and their complexes of medicinally-significant metal ions. The Hpophilicity, polarity and overall charge of the complex may be readily altered by simple substitutions in the molecular framework of the ligand in order to optimize the in vivo biodistribution. The derivatives of 3-hydroxy-4-pyrones and 3-hydroxy-4-pyridinones, two intensively studied ligand systems in our research group, are such modifiable ligands, forming i n stable complexes of Group 13 metal ions. Unique combinations of biologically significant properties (water solubility, hydrolytic stability, and Hpophilicity) have led to the wide use of Q tris(maltolato)aluminum in the study of Al neurotoxicity, and to the potential application of a 67Ga complex, tris(l-/?-methoxyphenyl-2-methyl-3-oxy-4-pyridinonato)gallium(III), for heart imaging. In addition, these ligands are non-toxic, and either naturally occurring or easily synthesized. In an effort to extend these properties to technetium complexes, we have synthesized and characterized neutral bis-ligand complexes of technetium(V) and rhenium(V) with 2-methyl-3-oxy-4-pyronate (maltolate, ma") and l,2-dimethyl-3-oxy-4-pyridinonate (dpp-), and have isolated the two mono-ligand intermediates. To our knowledge, these complexes are the first oxotechnetium(V) and oxorhenium(V) complexes containing bidentate (0,0) monoprotic ligands; some stoichiometrically and structurally similar complexes of bidentate (N,0) monoprotic ligands have been reported. As this work was in progress, Archer et al. reported the synthesis and crystal structure 24 of a cationic imidorhenium(V) complex of ma", [Re(ma)2(NPh)(PPh3)][BPh4], and Kanvinde et alP presented biodistribution studies on purported cationic 99mTc(IV) complexes of ma" and ka-.* Preliminary results of Kanvinde's work suggested that Tc derivatives of the 3-oxy-4-pyronate skeleton might be a new class of cationic complexes with potential as heart imaging radiopharmaceuticals. Me I ,CL .Me M^ ^Me O O ma- dpp-2.2 Experimental Materials. All chemicals were reagent grade and were used as received: [NH4]["Tc04] (a gift from the Du Pont Merck Pharmaceutical Company), NH4Re04 (Aldrich), and Hma (Sigma). Hdpp,18 [(«-Bu)4N][TcOCl4],19 [(n-Bu)4N][ReOBr4],20 [Ph4As][ReOBr4]20were prepared according to published procedures. Caution! " T c is a low energy (0.292 MeV) P~ emitter with a half-life of 2.12 x 105 years. All manipulations of solutions and solids were performed in a laboratory approved for the handling of low-level radioactivity, and normal safety procedures were followed at all times to prevent contamination. Instrumentation. NMR spectra (200 MHz, 300 MHz and 400MHz) were recorded on Bruker AC-200E, Varian XL 300 (VT !H NMR) or Bruker WH-400 ( iH-iH COSY) spectrometers, with 8 referenced to external TMS and assignments based on those for the unbound ligand. Chemical shifts are reported as 8 in ppm: 8 (multiplicity, number of hydrogen nuclei, coupling constant). Mass spectra were obtained with either an AEIMS-9 instrument (fast atom bombardment ionization, FAB), or DELSI-NERMAG R10-10c (desorption chemical Ka_ is the deprotonated form of kojic acid, or 5-hydroxy-2-(hydroxy-methyl)-4-pyrone.3 25 ionization, DCI). Only the most intense peak is given where consistent isotopic patterns were observed. Infrared spectra were recorded as KBr pellets in the range 4000-400 cm"1 on a Perkin-Elmer PE 783 spectrophotometer previously calibrated against the spectrum of polystyrene. Melting points were measured on a Mel-Temp apparatus and were uncorrected. Microanalyses were performed by Mr. P. Borda in this department or by Canadian Microanalytical Services Ltd. (Tc complexes). [ReOBr(ma)2]. Hma (58 mg, 0.46 mmol) was added to a solution of [(«-Bu)4N][ReOBr4] (152 mg, 0.2 mmol) in 5 mL of ethanol:acetone (1:1). The resulting brown-green solution was heated for 15 min, and was clarified by filtration. Slow evaporation of the solvent at room temperature resulted in green crystals. The supernatant was removed and the crystals were washed with diethyl ether, collected by centrifugation, and dried in air. The yield was 68 mg (64 %), m.p. 228-23 VC (dec). The product is soluble in acetone, acetonitrile, chloroform, methanol, water, and benzene, but insoluble in diethyl ether. Anal. Calcd (found) for Ci2Hi0BrO7Re: C 27.08 (27.31), H 1.89 (2.00), Br 15.01 (14.89). Mass spectrum (FAB): m/z = 532 ([ReOBr(ma)2]+), 453 ([ReO(ma)2]+). lH NMR (acetonitrile-^): 8.22 (d, 1H, 3/HH = 5.1 Hz), 7.73 (d, 1H, 3/HH = 5.4 Hz), 7.20 (d, 1H, 3/HH = 5.1 Hz), 7.06 (d, 1H, 3 /HH = 5.4 Hz), 2.73 (s, 3H), 2.37 (s, 3H). IR (cm"1, KBr disk): 1615, 1580, 1565, 1540, 1510, 1470 (all vs, mixed Vc=o and vc=c); 975 (s, VRe=o), 350 (m, VRe-Br)-[TcOCl(ma)2]. A procedure similar to that for [ReOBr(ma)2] was followed using [(n-Bu)4N][TcOCl4] (61 mg, 0.12 mmol) and Hma (28 mg, 0.22 mmol), except that the volume of solvent used was 10 mL and that five drops of 0.05 M NaOMe in methanol were added before the green solution was heated. The yield of the greenish yellow crystalline product was 21 mg (43 %). The product is soluble in acetone and slightly soluble in diethyl ether. Anal. Calcd (found) for Ci2Hi0ClO7Tc: C 36.06 (36.31), H 2.52 (2.69). Mass spectrum (DCI): m/z = 401 ([TcOCl(ma)2 + 1]+), 365 ([TcO(ma)2]+). lH NMR (acetonitrile-^): 8.30 (d, 1H, 3/HH = 5.2 Hz), 7.99 (d, 1H, 37HH = 5.0 Hz), 7.10 (d, 1H, 37HH = 5.1 Hz), 7.08 (d, 1H, 3 /HH = 5.1 Hz), 2.71 (s, 3H), 2.20 (s, 3H). IR (cm"1, KBr disk): 1610, 1575, 1560, 1540, 1510, 1470 (all vs, mixed Vc=o and Vc=c); 965 (s, VTC=O); 355, 340 (m, VTC-X)-26 [ReOBr(dpp)2]. To a solution of [(n-Bu)4N][ReOBr4] (73 mg, 0.096 mmol) in acetonitrile (5 mL) was added Hdpp (28 mg, 0.20 mmol). A blue-green solution formed, which subsequently changed to green upon heating for 10 min. The solution was reduced to ca. 2 mL in volume before it was clarified by filtration. Diethyl ether was added dropwise while shaking until the solution was close to cloudiness. A green solid deposited after storing at -4 °C overnight. The supernatant was removed and the green solid was washed with acetonitrile then diethyl ether, collected by centrifugation, and dried in air. The yield was 21 mg (39 %), m.p. 272-275 °C (dec). The product is soluble in acetone, acetonitrile, methylene chloride, methanol and water, but insoluble in diethyl ether. Anal. Calcd (found) for Ci4Hi6BrN2C<5Re: C 30.11 (30.51), H 2.89 (3.12), N 5.02 (5.24), Br 14.14 (14.08). Mass spectrum (FAB): m/z = 496 ([HRe02(dpp)2]+), 479 ([ReO(dpp)2]+). !H NMR (methanol-^): 8.02 (d, 1H, 3/HH = 6.8), 8.01 (d, 1H, 3 / H H = 6.8), 7.29 (d, 1H, 3 / H H = 6.8), 7.27 (d, 1H, 3 / H H = 6.8), 4.10 (two overlapped s, 6H), 2.75 (s, 3H), 2.72 (s, 3H). IR (cm'1, KBr disk): 1615, 1555, 1515, 1495, 1485, 1455 (all vs, mixed Vc=o and Vc=c); 955 (s, VRe=o); 375 (m, VRe-Br)-[TcOCl(dpp)2]-0.5H2O. A procedure similar to that for [ReOBr(dpp)2], using [(n-Bu)4N][TcOCU] (47 mg, 0.094 mmol) and Hdpp (30 mg, 0.22 mmol), was followed except that the solvent was ethanol:acetone (1:1) and that the resulting solution was brown-green. Recrystallization from acetonitrile/diethyl ether yielded a red powder (20 mg, 49 %). The product is soluble in methanol and water, but insoluble in diethyl ether. Anal. Calcd (found) for C14H17CIN2O5.5TC: C 38.68 (39.01), H 3.94 (4.29), N 6.44 (6.29). Mass spectrum (FAB): m/z = 391 ([TcO(dpp)2]+). ! H NMR (methanol-d4): 8.08 (d, 2H, 3 /HH = 6.8 Hz), 7.20 (d, 1H, 3/HH = 6.8 Hz), 7.18 (d, 1H, 3/HH = 6.8 Hz), 4.10 (two overlapped s, 6H), 2.76 (s, 3H), 2.73 (s, 3H). IR (cm-1, KBr disk): 1615, 1555, 1505, 1485, 1460 (all vs, mixed v c=o and vC=c); 950 (s, vTc=o); 375, 335 (m, vTc-x)-[(#i-Bu)4N][ReOBr3(ma)]. To [(n-Bu)4N][ReOBr4] (300 mg, 0.39 mmol) in acetonitrile (5 mL) was added Hma (69 mg, 0.55 mmol). A brown solution resulted and this was heated for 10 min. The solution was reduced to ca. 3 mL in volume and then clarified by filtration. Diethyl ether was added dropwise while shaking until the solution was close to 27 cloudiness. Brown crystals deposited after storing at -4 °C overnight. The crystals were collected by filtration, washed with diethyl ether, and dried in air. The yield was 214 mg (68 %). The product is soluble in acetone, acetonitrile, chloroform, methanol, and water, but insoluble in diethyl ether. Anal. Calcd (found) for C22H4iBr3N04Re: C, 32.64 (32.58); H, 5.11 (5.19); N, 1.73 (1.75); Br, 29.61 (29.63). Mass spectrum, FAB cation detection: m/z = 564 ([ReOBr3(ma) - 3]+), 453 ([ReO(ma)2]+), 344 ([Re02(ma)]+), 242 ([(n-Bu)4N]+); FAB anion detection: m/z = 567 ([ReOBr3(ma)]-), 487 ([ReOBr2(ma) - 1]-) , 79 (Br). lK NMR (acetonitruW3): 7.67 (d, 1H, 3 / H H = 5.4 Hz), 7.07 (d, 1H, 3/HH = 5.4 Hz), 3.08 (t, 8H), 2.41 (s, 3H), 1.59 (q, 8H), 1.35 (h, 8H), 0.96 (t, 12H). Infrared spectrum (cm"1, KBr disk): 1615, 1585, 1520, 1475 (all vs, mixed vc=o and vc=c); 970 (s, vRe=o); 355 (m, vRe-Br)-[Ph4As][ReOBr3(ma)]. To [Ph4As][ReOBr4] (93 mg, 0.10 mmol) in ethanol: acetone (1:1,5 mL), was added Hma (27 mg, 0.21 mmol). A brown-green solution resulted, which was clarified by filtration. Diethyl ether was added dropwise to the swirling solution until it was close to cloudiness. Brown crystals deposited after storing at -4 °C overnight. The crystals were washed with diethyl ether, collected by filtration, and dried in air. The yield was 36 mg (37 %). The product is soluble in chloroform, acetonitrile, and acetone, but insoluble in diethyl ether. Anal. Calcd (found) for C30H25AsBr3O4Re: C 37.91 (37.57), H 2.65 (2.55), Br 25.22 (24.95). Mass spectrum (FAB): m/z = 453 ([ReO(ma)2]+), 383 ([Ph4As]+), 344 ([Re02(ma)]+). !H NMR (acetonitrile-^): 7.90 -7.62 (m, 21H), 7.07 (d, 1H, 3/HH = 5.3 Hz), 2.40 (s, 3H). Infrared spectrum (cm"1, KBr disk): 1610, 1580, 1565, 1520, 1480, 1435 (all vs, mixed vc=0 and vc=c); 965 (s, v R e = 0 ) ; 350 (m, vRe-Br)-[(/i-Bu)4N][TcOCl3(ma)]. A procedure similar to that for [(n-Bu)4N][ReOBr3(ma)] was followed using [(n-Bu)4N][TcOCl4] (50 mg, 0.10 mmol) and Hma (41 mg, 0.33 mmol), except that seven drops of 0.2 M NaOH in methanol was added. The product was collected by centrifugation and dried in vacuo overnight. Yield was 46 mg (78 %). The product is soluble in methanol, water, and acetonitrile, but insoluble in diethyl ether. Anal. Calcd (found) for C2 2H4 iCl3N04Tc: C 44.94 (44.56), H 7.03 (6.90), N, 2.38 (2.37). Mass spectrum (FAB): m/z = 365 ([TcO(ma)2]+), 275 ([TcOCl(ma)]+), 242 ([(«-Bu)4N]+). !H NMR (acetonitrile-^): 28 7.92 (d, 1H, 3 / H H = 5.3 Hz), 7.06 (d, 1H, 3 / H H = 5.2 Hz), 3.07 (t, 8H), 2.21 (s, 3H), 1.59 (q, 8H), 1.33 (h, 8H), 0.95 (t, 12H). Infrared spectrum (cm"1, KBr disk): 1610, 1580, 1510, 1470 (all vs, mixed Vc=o and Vc=c); 965 (s, VTC=O); 360, 315 (m, VTC-X)-X-ray crystallographic analyses. All the crystal structures reported in this thesis were determined by Dr. Steven J. Rettig of the UBC Structural Chemistry Laboratory. Selected crystallographic data and selected coordinates for [ReOBr(ma)2], [(n-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)] appear in Appendix I. 2.3 Results and Discussion The [TcOCU]" or [ReOBr4]_ anions react with 2-methyl-3-hydroxy-4-pyrone (Hma) to give both mono- and bis-ligand complexes [MOX3(ma)]" and [MOX(ma)2] (M = Tc, X = CI; M = Re, X = Br), depending on the reaction conditions (solvent, counter cation of the starting material, etc.). For instance, acetonerethanol (1:1) as the reaction solvent (with subsequent slow evaporation of the solvent) was effective in depositing either [TcOCl(ma)2] or [ReOBr(ma)2] in crystalline form, but addition of diethyl ether as a precipitant to these systems produced [(n-Bu)4N][TcOCl3(ma)] or [Ph4As][ReOBr3(ma)]. Substitution in acetonitrile yielded [(«-Bu)4N][TcOCl3(ma)] or [(n-Bu)4N][ReOBr3(ma)]. Reaction of [(n-Bu)4N][ReOBr3(ma)] with Hma in acetonerethanol (1:1) produced [ReOBr(ma)2]. Using tetraphenylarsonium tetrabromooxorhenate as the starting material tended to produce mono-ligand complexes with Hma. With Hdpp, however, only bis-ligand products, [MOX(dpp)2], were obtained from different solvents (e.g. acetonitrile, methanol, or 1:1 acetone:ethanol) and either counter cation. All products are air stable in the solid state and in solution. However, it was observed that upon standing, [(n-Bu)4N][ReOBr3(ma)] converted to [ReOBr(ma)2] in acetonitrile or water (vide infra). All complexes were characterized by elemental analyses, infrared, mass, and *H NMR spectra, and in some cases by, X-ray crystallography. Analytical data established the formulations of all the complexes, while IR measurements confirmed that the ligands were 29 coordinated as evidenced by the disappearance of the O-H stretch and the shifting of the mixed Vc=o and Vc=c bands to lower wave numbers. Bis-ligand Complexes. The mass spectra are diagnostic of the complex formulations. In all cases, loss of one halo ligand from a [MOXL2] unit to give [MOL2]+ was observed as the base peak, indicating that the bidentate ligands are more strongly bound than the monodentate halo ligands. The IR spectra of the technetium complexes are almost superimposable on those of the rhenium complexes. The Tc=0 bond stretching vibrations were found at 965 cnr1 in [TcOCl(ma)2] and 950 cm"1 in [TcOCl(dpp)2], much lower values than the 1020 cm'1 observed for the starting material [(n-Bu)4N][TcOCl4].21 Similarly, the Re=0 bond stretching vibrations at 975 cm-1 for [ReOBr(ma)2] and 955 cm"1 for [ReOBr(dpp>2] are lower than 1010 cm"1 for [(n-Bu)4N][ReOBr4] or 1000 cm"1 for [Ph4As][ReOBr4].20 These observations suggested that a sixth bond was formed trans to the M=0 group.15 The four-band infrared spectral pattern (mixed Vc=o a nd ring Vc=c) between 1660 and 1450 cm-1, which is characteristic of 4-pyrones and 4-00 pyridinones, was preserved in all the complexes although the energy ordering was changed upon coordination. ' In the bis-ligand complexes, there are one or two additional absorptions in that region, as might be expected for the two non-equivalent environments of the two ligands (axial and equatorial relative to the M=0 linkage). A cw-(halo, oxo) structure was proposed for these complexes, as lH NMR and ^ ^ H COSY spectra revealed two sets of ligand hydrogen signals in a 1:1 ratio, corresponding to two non-equivalent deprotonated ligands in each complex. This structure is consistent with other reported oxometal(V) complexes of bidentate monoprotic ligands,10"15 and was confirmed by the X-ray structure of [ReOBr(ma)2] (vide infra). Two characteristic pairs of ring hydrogen doublets (H5-H6 and H5-H6, see the scheme under Table 2-1) was observed in the !H NMR spectra of all the complexes (those for the two bis(maltolato) complexes are shown in Figure 2-1). The values of the vicinal coupling constants (5.0 - 5.4 for ma- complexes, 6.8 for dpp- complexes) are consistent with those of the Al(III), Ga(III),3 and Zn(II)23 complexes of ma% and those of the Al(III), Ga(III),5 and Si(IV)24 complexes of dpp\ For [ReOBr(ma)2], the set of hydrogen signals less deshielded, and with similar chemical shifts and identical 7HH coupling constants to those of [ReOBr3(ma)]_ (where the 30 JC H I H4 J H r r i 1 1 1 1 1 1 1 1 1 1 1 8.6 8.4 8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 6.6 6.4 PPM Figure 2-1. lH NMR spectra of the characteristic pairs of ring hydrogens of the bis(maltolato) and mono(maltolato) complexes in CD3CN (H5, H6 from the axial ligand, H5', H6' from the equatorial ligand, see the scheme under Table 2-1). a. [ReOBr(ma)2] (400 MHz); b. [TcOCl(ma)2] (200 MHz); c. [ReOBr3(ma)]- (300 MHz); d. [TcOCl3(ma)]- (400 MHz); * free maltol. 31 ligand was known to chelate axially, vide infra), was assigned to the ligand in the axial position. Further credence was lent to this assignment by the fact that this ma- ligand is structurally very similar to the axial ma- of [ReOBr3(ma)]" relative to Re=0 in the solid state (Figures 2-1 and 2-2). The spectrum of [TcOCl(ma)2] was similarly assigned by comparison with that of [TcOCl3(ma)]_ (Figure 2-1). In the assignment of the spectra for [ReOBr(dpp)2] or [TcOCl(dpp)2], it is also assumed that the *H NMR signals of the axial ligand would be deshielded less than those of the equatorial ligand. The two doublets of H5 and H5' in [TcOCl(ma)2], which partially overlapped at room temperature (Figure 2-1), were observed to split completely upon cooling to -30 °C, and to collapse fully to one doublet at 65 °C in acetonitrile-^. Such a collapse occurred at 58 °C in [TcOCl(dpp)2] (methanol-f/4) but for [ReOBr(ma)2] there was no collapse to 120 °C (DMF-^7). Thus, the bis-ligand Tc complexes are more fluxional than their Re analogs. Mono-ligand Complexes. Substitution of bidentate monoprotic ligands onto the [TcOCU]" or [ReOBr4]" anion is thought to proceed in steps, in which a mono-ligand intermediate complex is involved. The isolated intermediates in the above substitutions were trihalo(maltolato)oxometalate(V) anions [MOX3(ma)]- (M = Tc, X = CI; M = Re, X = Br) as tetra(n-butyl)ammonium (M = Tc and Re) and tetraphenylarsonium salts (M = Re). Intense peaks for the [(n-Bu)4N]+ and [Ph4As]+ cations are present in the positive ion detection mass spectra, while the cationization of the anionic moiety is very weak and the pattern is different for different compounds. For instance, the spectrum of [(n-Bu)4N][ReOBr3(ma)] contained a peak corresponding to [ReOB^ma) - 3H]+, while [Ph4As][ReOBr3(ma)] showed a [Re02(ma)]+ peak, and [(n-Bu)4N][TcOCl3(ma)] showed a [TcOCl(ma)]+ peak. All of them had [MO(ma)2]+ peaks, also seen in the bis-ligand analogs, indicating that under ionizing conditions the mono-ligand intermediate molecules reacted with themselves to form the more stable bis-ligand complexes. The negative ion detection FAB mass spectrum of [(n-Bu)4N][ReOBr3(ma)] showed an intense [ReOBr3(ma)]_ peak. The IR spectra showed strong vc-H stretches for [(n-Bu)4N]+ or medium Vc-H f° r [Ph4As]+ in each compound. The spectrum of [(n-Bu)4N][TcOCl3(ma)] is almost superimposable on that of [(n-Bu)4N][ReOBr3(ma)]. The M=0 bond stretching vibrations, were 32 found to be at the same frequency for these tetra(n-butyl)ammonium salts as in each corresponding bis-ligand complex. The VRe=o for the tetraphenylarsonium salt is 10 cnr1 lower than that for its tetra(n-butyl)ammonium analog (the same difference in VRe=o between [(n-Bu)4N][ReOBr4] and [Ph4As][ReOBr4]).20 The four-band infrared spectral pattern between 1660 and 1450 cm-1, shifted to lower energy upon coordination, is preserved in the two [(n-Bu)4N]+ salts, while in the [Pb4As]+ salts this pattern is augmented by aromatic Vc=c bands. lH NMR spectra showed the presence of the cations and only one set of lH signals for the one bound ligand in each complex (Figure 2-1). Structure of ReOBr(ma)2. Single crystals of [ReOBr(ma)2] and [TcOCl(ma)2] were obtained from acetone:ethanol (1:1) solvent mixtures by slow evaporation. Crystals of [ReOBr(dpp)2] were obtained from acetonitrile by slow evaporation. All the spectral data suggested that these bis-ligand complexes have the same cis-(ha\o, oxo) structure so only the crystal structure of [ReOBr(ma)2] was solved. An ORTEP diagram of one of the enantiomers of one of the two independent molecules is shown in Figure 2-2. The asymetric unit consists of two molecules each of which has two enantiomers which appear side by side without significant interactions; Figure 2-2 shows a A stereoisomer. Bond distances and angles quoted are for the molecule containing Re(l); those for the molecule containing Re(2) are very similar. The overall geometry around the rhenium atom is best described as highly distorted octahedral with cw-(bromo, oxo) and cis maltolato ligands. The Re atom is 0.2 A out of the equatorial BrC>3 plane. The two maltolato moieties, each of which acts as a bidentate (0,0) donor ligand, are close to mutually orthogonal (dihedral angle 94.6°), with one ligand sitting in an equatorial plane and the other in an axial plane relative to Re=0. The axial ligand chelates with its anionic oxy O-donor trans to the Re=0 group, while the equatorial ligand coordinates such that the bromo ligand and an oxy O-donor are each trans to a neutral ketone O-donor, consistent with known examples of this [MOXL2] type of complexes, where L is a bidentate monoprotic ligand with neutral N and anionic O donors. The distortions from octahedral geometry are mainly caused by the acute bite angles of the bidentate maltolato ligand to the metal (0(3)-Re(l)-0(4) = 76.1(3)° for the axial ligand, and 0(6)-Re(l)-0(7) = 80.7(3)° for the equatorial ligand), and by 33 B r l Figure 2-2. ORTEP drawing of one of the two independent molecules in the asymmetric unit of [ReOBr(ma)2]. 34 Table 2-1. Selected Bond Lengths (A) for one of the two independent molecules in [ReOBr(ma)2], and for the anions in [(n-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)]. M-X(l) M-X(2) M-X(3) M-O(l) M-0(3) M-0(4) M-0(6) M-0(7) Oketo-C(4) Ooxy-C(3) 0(1)-C(2) 0(1)-C(6) C(2)-C(3) C(3)-C(4) C(4)-C(5) C(5)-C(6) [ReOBr(ma)2] M = Re, X = Br 2.510(1) 1.673 (8) 2.028 (7) 2.114(7) 1.961 (7) 2.060 (7) i 1.27 (l)b, 1.29 (l)c 1.35 (l)b , 1.36 (l)c 1.37 (l)b, 1.36 (2)c 1.35 (2)b, 1.34 (2)c 1.36 (2)b, 1.38(l)c 1.42(l)b, 1.40 (2)c 1.39 (l)b , 1.42 (2)c 1.32 (2)b, 1.32 (2)c [(n-Bu)4N][ReOBr3(ma)] M = Re, X = Br 2.545 (1) 2.530 (1) 2.473 (1) 1.664 (5) 2.103 (5) 2.034 (5) Coordinated liganda 1.264 (8)b 1.349 (8)b 1.366 (9)b 1.34 (l)b 1.37 (l)b 1.41 (l)b 1.42 (l)b 1.35 (l)b [(n-Bu)4N] [TcOCl3(ma)] M = Tc, X = CI 2.398 (1) 2.3121 (9) 2.378 (1) 1.647 (2) 2.019 (2) 2.152 (2) 1.269 (4)b 1.327 (4)b 1.363 (4)b 1.331 (5)b 1.360 (4)b 1.416 (4)b 1.432 (5)b 1.338 (5)b a Numbers corresponding to the scheme shown on the right b Axial ligand (with respect to M=0). r l ^ °°*y c Equatorial ligand (with respect to M=0). 35 Figure 2-3. ORTEP drawing of the anions in [(n-Bu)4N][ReOBr3(ma)] (left) and [(n-Bu)4N][TcOCl3(ma)] (right). 36 Table 2-2. Selected Bond Angles (deg) for one of the two independent molecules in [ReOBr(ma)2], and for the anions in [(«-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)]. [ReOBr(ma)2] [(n-Bu)4N][ReOBr3(ma)] [(n-Bu)4N][TcOCl3(ma)] (Xl)-M-X(l) 0(1)-M-X(2) 0(1)-M-X(3) 0(l)-M-0(3) 0(l)-M-0(4) 0(l)-M-0(6) 0(l)-M-0(7) X(l)-M-0(3) X(l)-M-0(4) X(l)-M-0(6) X(l)-M-0(7) 0(3)-M-0(4) 0(3)-M-0(6) 0(3)-M-0(7) 0(6)-M-0(7) 00Xya-M-X(l) Ooxya-M-X(2) 0oxya-M-X(3) Oketo-C(4)-C(3)a 0oxy-C(3)-C(4)a 0(1)-C(2)-C(3)a C(2)-C(3)-C(4)a C(3)-C(4)-C(5)a C(4)-C(5)-C(6)a C(5)-C(6)-0(l)a C(6)-0(1)-C(2)a M = Re, X = Br 97.4 (3) 166.8 (3) 91.5 (3) 106.9 (4) 93.5 (4) 86.7 (2) 87.3 (2) 93.7 (2) 168.8 (2) 76.1 (3) 85.2 (3) 83.2 (3) 80.7 (3) 86.7 (2) 117.2 (9)b, 116 (l)c 114 (l)b, 117.8 (8)c 118 (l)b, 118 (l)c 121 (l)b, 120(l)c 119 (l)b, 120 (l)c 118 (l)b, 116 (l)c 124 (l)b, 125 (l)c 121 (l)b, 121 (l)c M = Re, X = Br 93.5 (2) 95.3 (2) 102.9 (2) 89.8 (2) 166.1 (2) 76.4 (2) 84.6 (1) 85.9 (1) 91.0(1) 117.4 (7)b 114.9 (7)b 118.0 (8)b 122.4 (8)b 117.3 (8)b 117.7 (8)b 123.9 (8)b 120.6 (7)b M = Tc, X = CI 93.38 (8) 104.84 (8) 94.40 (8) 162.3 (1) 86.2 (1) 76.08 (8) 86.37 (7) 92.84 (7) 85.41 (7) 117.7 (3)b 115.2 (3)b 119.3 (4)b 120.7 (3)b 117.8 (3)b 117.3 (4)b 124.3 (4)b 120.6 (3)b a, b, c please refer to the footnotes under Table 2-1. 37 repulsions between the Re=0 linkage and the negatively charged cw-donors (either bromo or oxy oxygen of the equatorial maltolato ligand). The 0oxo-Re-Br and 0oxo-Re-0OXy angles are 97.4(3)° and 106.9(4)° respectively, in contrast with the 0oxo-Re-Oketo,eq and Ooxo-Re-Oketo,ax angles at 93.5(4)° and 91.5(3)°, respectively. The Re-Oketo and Re-0oxy bonds are significantly different in each maltolato moiety; as in maltolato complexes of Al(IH), B(IH), and Fe(III), the M-0o x y bonds are shorter than M-Oketo-The Re=0 bond length of 1.673(8) A is consistent with the values found in other oxo complexes of rhenium(V), indicating the retention of a multiple bond. As in Tc(V)-0oxo complexes, the oxo group exerts a profound trans effect/influence on the sixth coordination site kinetically/structurally. As a result, the linkage between the central atom and the sixth donor is usually lengthened. However, exceptions with the trans bond being not significantly 10 19 98 lengthened have been observed in complexes containing six-membered chelate ring. ' ' In [ReOBr(ma)2], the trans Re-0o x y bond (2.028(7) A) is significantly longer than the cis bond (1.961(7) A); in [Re(ma)2(NPh)(PPh3)]+ values of 1.996(7) and 1.987(7) A were found for Re-0 0 x y bonds trans and cis to the Re=NPh group, respectively. This indicates that the oxo O-donor has a stronger trans influence than the phenyl imido N-donor. The Re-Br distance (2.510(1) A) is close to the average value in [ReOBr4]- (2.48 A).29 Each maltolato ligand was found to be quite planar, as in its complexes of Al(III), and 9 f\ Fe(III). Upon coordination, an increase in aromaticity of the ligand skeleton may be expected, since chelating to a Lewis acid like 0=Tc(V) or 0=Re(V) favors forming a delocalized 7t system as is shown in Scheme 2-1. In fact, significant bond averaging and increasing planarity in comparison with free ligand were observed in Al(III) and Ga(III) complexes of dpp", an analog of ma". There has been no report on the crystal structure of maltol itself, nevertheless, comparison can still be made with other maltolate complexes 1>25 '26 '31 as well as with 4-pyrone. The averaging of bond lengths within the ligand skeleton was observed: the single C-C bonds are shortened to 1.39 - 1.42 A from 1.440 A in 4-pyrone whereas the double C=C bonds are slightly lengthened to 1.36 - 1.38 A from the reported value of 1.356 A, with the exception of C(5)-C(6), where the bonds are shortened instead (Table 2-3). 38 Considering the longer 00Xy-C(3) and Oketo-C(4) bond lengths as well as the shorter Re-0 bond lengths for the equatorial ligand versus the axial ligand, one may conclude that the equatorial ligand donates electron density to the central atom more than the axial ligand. Me O Q^V \ © Scheme 2-1 Structures of [(n-Bu)4N][ReOBr3(ma)] and [(n-Bu) 4N][TcOCl3(ma)] . Single crystals of [(n-Bu)4N][ReOBr3(ma)] suitable for X-ray studies were obtained from acetonitrile with diethyl ether as precipitant. Crystals of [(n-Bu)4N][TcOCl3(ma)] were grown by cooling an acetone:ethanol (1:1) solution with diethyl ether as the precipitant. The X-ray structures of the anionic complexes are shown in Figure 2-3, where the atom-numbering schemes are also defined. The two [MOX3(ma)]- anions are isostructural; each unit cell contains pairs of [MOX3(ma)]_ anions and [(n-Bu)4N]+ cations, with no significant interactions between them. The coordination environments around the metal atom in each complex are both highly distorted octahedral. The oxo oxygen of the M=0 bond and the negatively charged oxy oxygen donor of the ma- ligand, occupy apical trans positions subtending an angle at M of 162.3(1)° (M = Tc) or 166.1(2)° (M = Re); the three halo ligands and the ketonic oxygen donor of the ma" ligand form an equatorial plane out of which M is displaced about 0.2 A towards the oxo O (Figure 2-3). The distortions from ideal octahedral geometry are primarily due to the acute bite angle of the bidentate maltolato ligand to the metal, and due to repulsions between the electronically demanding M=0 linkage and the negatively charged equatorial halo ligands. The bite angles are 76.08(8)° and 76.4(2)° for M = Tc and for M = Re respectively; the 0oxo-M-X angles range from 93° to 105° while the Ooxo-M-Oketo are 86.2(1)° for M = Tc and 89.8(2)° for M = Re. Tc and Re have a 39 similar size due to the lanthanide contraction, the covalent radii being 0.74 A and 0.72 A (ionic radii 0.60, 0.58 A) for six-coordinated Tc(V) and Re(V), respectively. In these two complexes, however, the Tc-0 bonds are significantly different than the Re-O bonds. The Tc-Ooxo (1.647(2) A) and Tc-0OXy (2.019(2) A) bonds are shorter than the corresponding Re bonds (1.664(5) A and 2.034(5) A, respectively), whereas Tc-Oketo (2.152(2) A) is longer than Re-Oketo (2.103(5) A), indicating that Tc(V) has a greater affinity for the negatively charged donors than does Re(V). The mean M-X distances (2.359 A, for M = Tc and 2.515 A for M = Re ) are longer than those in the corresponding halo complexes (2.305 A in [TcOCU] , and 2.48 A in [ReOBr4] , ). In comparison to [ReOBr(ma)2], similar bond lengths are found for Re-0OXo> Re-Ooxy(axial), and Re-Oketo(axial) in [ReOBr3(ma)]-; in addition, the two molecules have very similar 0oxo-Re-00xy angles. The similarity in chemical environments of the axial ma" ligands in each complex was also supported by the very similar *H NMR chemical shifts (Figures 2-3 and 2-4). The bond lengths and bond angles in the coordinated maltolato of [ReOBr3(ma)]_ and [TcOCl3(ma)]_ parallel, for the most part, those in the axial ligand of [ReOBr(ma)2] (Tables 2-1 and 2-2). Conversions between the mono- and bis-ligand complexes. A slow conversion of [ReOBr3(ma)]~ ([(n-Bu)4N]+ salt) to [ReOBr(ma)2] was observed by *H NMR in acetonitrile-c?3 (Figure 2-4); a proposed equation for the overall conversion is shown in Scheme 2-2. After prolonged standing (ca. one year) the solution was found to be a 3:2 mixture of [ReOBr(ma)2] and [ReOBr3(ma)]~. For the technetium analog, however, no evidence for this conversion was found under the same conditions after standing for over four months, indicating that an even slower conversion reaction is involved, if one even takes place at all. The reverse conversion of bis-ligand to mono-ligand complexes was not observed in this solvent. CH,CN 2[ReOBr3(ma)]- • [ReOBr(ma)2] + [ReOBr4]" +Br Scheme 2-2 40 a yV. ~i i i i 1 1 1 1 1— 8.4 8.2 8.0 7.8 7.6 7.4 7.2 7.0 6.8 PPM Figure 2-4. *H NMR spectra (a. 300 MHz and b. 200 MHz) showing the conversion of [ReOBr3(ma)]- to [ReOBr(ma)2] in CD3CN. a. [ReOBr3(ma)]", a brown solution of two-week old; b. a green solution after five months. 2[MOX3(ma)r H20 fast •»• [MOX4] • + [MOX(ma)2]+X 1/3MCV + 2/3M02 + 4X" H20 1/3M04- + 2/3M02 + 2Hma +X" H 2 Q slow M = Re, X = Br; M = Tc, X = CI Scheme 2-3 41 Reactivity to water. While stirred in deionized water, [(n-Bu)4N][ReOBr3(ma)] slowly dissolved, resulting in a brown solution, which became green after approximately 10 minutes. This darkened, but, after stirring overnight, changed to greenish yellow with white crystals deposited; upon evaporating the solvent a black residue was isolated. The former product was identified as [(n-Bu)4N][Re04]* and the latter assumed to be Re02 (Scheme 2-3), 29 the known products of the hydrolytic disproportionation reaction of [ReOBr4]. All four of the rhenium and technetium maltolato complexes in acetonitrile-d3 were treated with D2O. All the mono-ligand and bis-ligand complexes were found to hydrolyze (Scheme 2-3). !H NMR showed that after being treated with D2O, the mono-ligand complexes converted to bis-ligand complexes relatively quickly; subsequently the concentration of bis-ligand complexes decreased slowly while that of the free Hma increased. In contrast to the expected lability of Tc > Re, ' [TcOCl3(ma)]- and [TcOCl(ma)2] were found to form Hma more slowly than their Re/Br analogs. This may be explained by the fact that the reactions are disproportion reactions and not simple ligand exchange or racemization processes. 2.4 Conclusion Neutral and anionic technetium(V) and rhenium(V) complexes of the form [MOXL2], and [MOX3(ma)]", respectively, where M = Tc, X = CI or M = Re, X = Br, and L is a bidentate (0,0) monoprotic ligand, 2-methyl-3-oxy-4-pyronate (maltolato, ma-) or l,2-dimethyl-3-oxy-4-pyridinonate (dpp"), have been synthesized and characterized. The bidentate (0,0) monoprotic ligands form stable oxotechnetium(V) and oxorhenium(V) complexes, which are structurally analogous to those complexes of the bidentate (0,N) monoprotic ligands. The mono-ligand complexes are intermediates in the formation of the bis-ligand complexes (Scheme 2-4). These compounds are important since they provide fully characterized models for the structures of intermediates in the synthesis of potential radiopharmaceuticals. * FABMS (both + and -) was performed to identify this white crystal: m/z - 242 ([(n-Bu)4N]+), 251(ReC>4"). 42 /Y . . Me O Hma, Y = O Hdpp, Y = NCH3 Hma or Hdpp Hma [MOX4]-/ M = Re,X = Br\ ^M = Tc,X = C y [MOX4]_ / M = Re, X = Br\ \ M = Tc,X = C l / M = Re Hma or standing [MOX3(ma)r / M = Re,X = Br\ \ M = Tc,X = Clj [MOXL2] L = ma" or dpp"\ M = Re,X = Br ] ,M = Tc,X = C l / Scheme 2-4 43 References (1) Finnegan, M. M.; Rettig, S. J.; Orvig, C. J. Am. Chem. Soc. 1986,108, 5033. (2) Nelson, W. O.; Rettig, S. J.; Orvig, C. J. Am. Chem. Soc. 1987,109, 4121. (3) Finnegan, M. M.; Lutz, T. G.; Nelson, W. O.; Smith, A.; Orvig, C. Inorg. Chem. 1987,26, 2171. (4) Matsuba, C. A.; Nelson, W. O.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1988, 27, 3935. (5) Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1988, 27, 1045. (6) Nelson, W. O.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1989, 28, 3153. (7) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 509. (8) Hewitt, C. D.; Herman, M. M.; Lopes, M. B. S.; Savory, J.; Wills, M. R. Neuropath. Appl. Neurobiol. 1991,17, 47 and references therein. (9) Zhang, Z.; Lyster, D. M.; Webb, G. A.; Orvig, C. Nucl. Med. Biol. 1992,19, 327. (10) Bandoli, G.; Mazzi, U.; Clemente, D. A.; Roncari, E. J. Chem. Soc, Dalton Trans. 1982, 2455. (11) Wilcox, B. E.; Heeg, M. J.; Deutsch, E. Inorg. Chem. 1984, 23, 2962. (12) Wilcox, B. E.; Cooper, J. N.; Elder, R. C ; Deutsch, E. Inorg. Chim. Acuta 1988,142, 55. (13) Duatti, A.; Marchi, A.; Rossi, R.; Magon, L.; Deutsch, E.; Bertolasi, V.; Bellucci, F. Inorg. Chem. 1988, 23, 4208. (14) Mazzi, U.; Roncari, E.; Rossi, R.; Bertolasi, V.; Traverso, O.; Magon, L. Trans. Met. Chem. 1980, 5, 289. 44 (15) Marchi, A.; Duatti, A.; Rossi, R.; Magon, L.; Mazzi, U.; Pasquetto, A. Inorg. Chim. Acta 1984, 81, 15. (16) Archer, C. M.; Dilworth, J. R.; Jobanputra, P.; Harman, M. E.; Hursthouse, M. B.; Karaulov, A. Polyhedron 1991,10, 1539. (17) Kanvinde, M. B.; Basmadjian, G. P.; Mills, S. L.; Kale, N. J. /. Nucl. Med. 1990, 31, 908 (abstract). (18) Kontoghiorghes, G. J.; Sheppard, L. Inorg. Chim. Acta 1987,136, L l l . (19) Davison, A.; Trop, H. S.; Depamphilis, B. V.; Jones, A. G. Inorg. Synth. 1982, 21, 160. (20) Cotton, F. A.; Lippard, S. J. Inorg. Chem. 1966, 5, 9. (21) Cotton, F. A.; Davison, A.; Day, V. W.; Gage, L. D.; Trop, H. S. Inorg. Chem. 1979, 18, 3024. (22) Katritzky, A. R.; Jones, R. A. J. Chem. Soc. 1960, 2947; Spectrochim. Acta 1961. (23) Morita, H.; Shimomura, S.; Kawaguchi, S. Bull. Chem. Soc. Jpn. 1976, 49, 2461. (24) Evans, D. F.; Wong, C. Y. Polyhedron 1991,10, 1131. (25) Orvig, C ; Rettig, S. J.; Trotter, J. Can. J. Chem. 1987, 65, 590. (26) Ahmet, M. T.; Frampton, C. S.; Silver, J. J. Chem. Soc, Dalton Trans. 1988, 1159. (27) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley-Interscience: New York, 1988; p 175. (28) Gilli, G.; Sacerdoti, M.; Bertolasi, V.; Rossi, R. Acta Cryst. 1982, B38, 100. (29) Cotton, F. A.; Lippard, S. J. Inorg. Chem. 1966, 5, 416. (30) Smitherman, H. C ; Ferguson, L. N. Tetrahedon 1968, 24, 923. (31) Annan, T. A.; Peppe, G; Tuck, D. G. Can. J. Chem. 1990, 68, 1598. (32) Norris, C. L.; Benson, R. C ; Beak, P.; Flygare, W. H. /. Am. Chem. Soc. 1973, 95, 2766. 45 (33) Shannon, R. D. Acta Cryst. 1976, A32, 751. (34) Helm, L.; Deutsch, K.; Deutsch, E. A.; Merbach, A. E. Helv. Chim. Acta 1992, 75, 210. (35) Johnson, D. L.; Fritzberg, A. R.; Hawkins, B. L.; Kasina, S.; Eshima, D. Inorg. Chem. 1984, 23, 4204. 46 Chapter Three Rhenium and Technetium Complexes from Pentadentate (N3O2) and Tetradentate (N2O2) Schiff Bases 3.1 Introduction In searching for new chelates from which to form stable Re/Tc complexes, in addition to the bidentate (0,0) monoprotic ligands, we have recently focused on some multidentate Schiff base ligands. Coordination of these multidentate Schiff bases can potentially result in the formation of numerous chelate rings, which may be useful in stabilizing Re/Tc complexes for use as radiopharmaceuticals. The ease of synthesis of multidentate Schiff base ligands has led to extensive 1 0 investigation of their coordination chemistry with Tc and many other metal centers. Some 'X f\ 7 8 Re/Tc complexes of tetradentate and pentadentate ' Schiff base ligands have been studied. Most of the ligands are derived from salicylaldehyde, thus they have the disadvantage of producing lipophilic Tc complexes which tend to localize preferentially in the liver.1 To avoid this problem, Liu et al have prepared a dha* based, potentially pentadentate (N3O2) ligand, H3apa, that contains both lipophilic hydrocarbon chains and polar carbonyl groups (Figure 3-1). With H3apa, a neutral oxotechnetium(V) complex was synthesized in good yield; thus the feasibility of preparing Tc(V) complexes of H3apa has led us to investigate further of Re/Tc coordination chemistry with this ligand and other dha based, N2O2 Schiff base ligands (H2epa and H2ppa, Figure 3-1). The N3O2 ligand H3apa, and N2O2 ligands H2epa and H2ppa, are derived from the condensation of two equivalents of dehydroacetic acid (dha) and one equivalent of diamine, i.e., diethylenetriamine (dien), ethylenediamine (en) or Dha stands for dehydroacetic acid, 3-acetyl-6-methyl-2//-pyran-2,4(3//)-dione. 47 propylenediamine (pn), respectively. This chapter summarizes the attempts to synthesize new Re/Tc complexes of the dha-based Schiff base ligands. Me Me Me [ ] Me o >=o o = / o Me' Me ivie O-0 Dehydroacetic acid (dha) H3apa = (dha)2dien, Y = CH2NHCH2 H2epa = (dha)2en, Y = none H2ppa = (dha)2pn, Y = CH2 Figure 3-1. Dha and dha-based Schiff base ligands. 3.2 Experimental Materials. [NH4][99Tc04] was obtained from the Du Pont Merck Pharmaceutical Company. Dehydroacetic acid, diethylenetriamine, ethylenediamine and propylenediamine (all from Aldrich) were used without further purification. [(n-Bu)4N][TcOCl4], [ReOCl3(PPh3)2],11 and [ReCl4(PPh3)2],12 were prepared according to published procedures. Caution! " T c is a low energy (0.292 MeV) P - emitter with a half-life of 2.12 x 105 years. All manipulations of solutions and solids were performed in a laboratory approved for the handling of low-level radioactivity, and normal safety procedures must be used at all times to prevent contamination. Instrumentation. *H NMR spectra were recorded on a Bruker AC-200E ^H, iH^H COSY, 200 MHz), or a Varian XL 300 (lH, 13C, 300 MHz), a Bruker WH-400 ( ^ H COSY), or a Bruker AMX-500 (lH, 500 MHz) spectrometer. Chemical shifts are reported as 8 in ppm referenced to external TMS: 8 (multiplicity, coupling constant,* number of hydrogen nuclei, assignment). Mass spectra were obtained with either a Kratos MS 50 (electron impact The constant is given only when necessary and possible. 48 ionization, EIMS) or a Kratos Concept II H32Q instrument (Cs+-LSIMS with positive or negative ion detection). Infrared spectra were recorded as KBr pellets in the range 4000-400 cm-1 on a Perkin-Elmer PE 783 spectrophotometer previously calibrated against the spectrum of polystyrene. Analyses for C, H, N were performed by Mr. Peter Borda (for ligands and the Re complexes) in this department or by Canadian Microanalytical Services Ltd. (for the Tc compounds). N,N'-3-azapentane-l,5-diylbis(3-(l-iminoethyl)-6-methyl-2^T-pyran-2,4(3f0-dione) (H3apa).9 To a hot solution of dehydroacetic acid (8.4 g, 50 mmol) in 150 mL methanol was added diethylenetriamine (2.5 g, 25 mmol). The solution was refluxed for 30 minutes and then concentrated to about 50 mL under reduced pressure as a white solid was formed. The mixture was cooled to room temperature and the solid was collected by filtration, washed with methanol and diethyl ether, and dried in air. Anal. Calcd (found) for C20H25N3O6: C, 59.54 (59.71); H, 6.25 (6.36); N, 10.42 (10.47). Mass spectrum (EI): m/z = 403 (molecular ion [C20H25N3O6]+). !H NMR (CDCI3): 14.1 (br s, 2H, OH), 5.5 (s, 2H, ring-//), 3.5 (m, 4H, imine-C//2), 2.9 (t, 4H, amine-C//2), 2.6 (s, 6H, imine-C//3), 2.0 (s, 6H, hng-CH3), 1.4 (br s, amine N-//). IR (cm"1, KBr disk): 3600-3300 (V0-H); 3000-2800 (VC-H); 1695, 1660, 1600, 1570 (all vs, mixed Vc=o and Vc=c)-N,N'-ethylene-diylbis(3-(l-iminoethyl)-6-methyl-2^-pyran-2,4(3fl)-dione) (H2epa). This material was prepared from the condensation reaction of dha and en, via a similar procedure as for H3apa. Anal. Calcd (found) for Ci8H2oN206: C, 59.97 (60.02); H, 5.60 (5.70); N, 7.78 (7.77). EIMS: m/z = 360 (molecular ion [Ci8H2oN206]+). lH NMR (CDCI3): 14.6 (br s, 2H, OH), 5.6 (s, 2H, ring-//), 3.8 (d, 4H, imine-C//2), 2.6 (s, 6H, imine-C//3), 2.1 (s, 6H, ring-C//3). IR (cm'1, KBr disk): 3600-3300 (V0-H); 3000-2800 (VC-H); 1710, 1660, 1580 (all vs, mixed Vc=o and Vc=c)-N,N'-propylene-diyIbis(3-(l-iminoethyl)-6-methyl-2H-pyran-2,4(3£/)-dione) semi-solvated methanol (H2Ppa*0.5MeOH). This material was prepared from the condensation reaction of dha and pn, via a similar procedure as for H3apa. Anal. Calcd (found) for C19.5H24N2O6.5: C, 59.99 (60.00); H, 6.20 (6.16); N, 7.18 (7.22). EIMS: m/z = 374 49 (molecular ion [Ci9H22N206]+). !H NMR (CDCI3): 14.4 (br s, 2H, OH), 5.6 (s, 2H, ring-//), 3.6 (m, 4H, imine-C//2), 3.3 (s, -1.5H, C//3OH), 2.6 (s, 6H, imine-C//3), 2.1 (m, 8H, central-e s and ring-C//3). IR (cm"1, KBr disk): 3600-3300 (V 0-H); 3000-2800 (VC-H); 1700, 1655, 1580 (all vs, mixed Vc=o and Vc=c)-[ReO(apa)]. Method A. To a mixture of [ReOCl3(PPh3)2] (167 mg, 0.200 mmol) and H3apa (103 mg, 0.256 mmol) was added 6 mL of ethanol. A green solution resulted after boiling. Six drops of triethylamine were added, and the solution was refluxed for a further 30 min. The solution was kept overnight at -4 °C, and the resulting brownish green precipitate was filtered off. The precipitate was recrystallized from acetone; brownish green crystals deposited as the solvent evaporated. These were collected by filtration, washed with diethyl ether, and dried in air to yield 85.8 mg (71 % based on Re). The product is soluble in chloroform and dichloromethane, but is insoluble in diethyl ether and water. Anal. Calcd (found) for C2oH22N307Re: C, 39.86 (39.91); H, 3.68 (3.75); N, 6.97 (6.77). LSIMS: m/z = 604 ([ReO(apa)+l]+). !H NMR (500 MHz, CDCI3): 5.73 (d, 47 = 0.9 Hz; IH, ring-//*), 5.05 (d, 47 = 0.8 Hz; IH, ring-//'*), 5.03 (dd, 2J = 11.6, 37 = 4.4 Hz, IH, amido -CHa//'e),** 4.77 (dd, 2J = 12.8, 37 = 6.0 Hz, IH, amido -CHa//e), 4.33 (dd, 2J = 12.5, 37 = 6.9 Hz, IH, amido -CHe//a), 4.09 (dd, 2 / = 12.2, 3 / = 5.8 Hz, IH, amido -CH'e//'fl), 4.07 (m, IH, imino -CHa//e), 3.96 (dt, 2 / = 11.5, 37 = 5.8 Hz, IH, imino -CH'a//'e), 3.82 (dt, 2J = 12.2,3J = 4.6 Hz, IH, imino -CH'e//'a), 3.51 (m, IH, imino -CHe//a), 3.04 (s, 3H, imino-C//3), 2.91 (s, 3H, imino-CH'i), 2.13 (d, 47 = 0.8 Hz, 3H, ring-C//3), 1.98 (d, 47 = 0.7 Hz, 3H, ring-C//'3). IR (cm"1, KBr disk): 1710, 1660, 1644, 1570, 1540 (all vs, mixed vc=o and Vc=c), 945 (s, vRe=o)-Method B. To the [ReCl4(PPh3)2] (168 mg, 0.197 mmol) and the H3apa (86 mg, 0.21 mmol) was added 7 mL of ethanol. Fourteen drops of triethylamine were added, and the mixture was refluxed for 2.5 h. The dark brown solution was clarified by filtration, and the Please refer to the geometry of the complex (Scheme 3-3). Joined at the amido N atom, the two halves of the bound apa3" are chemically non-equivalent with one enolic oxygen donor remaining in the equatorial plane and the other oxygen donor in the apical position (trans to Re=0). Thus, the corresponding half of the ligand is referred to as 'equatorial' or 'apical' (arbitrary), related to the Re=0 unit, and the hydrogen nuclei are designated as H for that of the 'equatorial' half, or as H 'for that of the 'apical' half. Please refer to the ring conformation (Figure 3-3). In the chelate ring, the methylene hydrogens are designated as Hg o r Ha to refer its (pseudo)equatorial or (pseudo)axial position relative to the ring conformation. 50 solvent allowed to slowly evaporate at room temperature. Brown crystals were collected and found to be identical to the product obtained via Method A. A yield of 37 mg (31% based on Re) was obtained. [{ReO(epa)hO]. To a mixture of [ReOCl3(PPh3)2] (166 mg, 0.200 mmol), H2epa (81.7 mg, 0.23 mmol) and Na2CC>3 (70 mg, 0.66 mmol) was added 30 mL of ethanol. The mixture was refluxed for 3 h, resulting in a grey-green solution and white precipitate (undissolved Na2C03). A green precipitate was deposited on the white precipitate while filtering. The precipitate mixture was recrystallized from acetone-chloroform mixture; green crystals and white crystals deposited as the solvents evaporated. The green crystals were mechanically separated from the white crystals, and dried in air. The yield of the green crystals (15 mg) was 13 % based on Re. The product is soluble in chloroform and dichloromethane, but is insoluble in diethyl ether and water. Anal. Calcd (found) for C36H36N40i5Re2i C, 38.03 (38.10); H, 3.19 (3.36); N, 4.93 (4.75). LSIMS: m/z = 561 ([ReO(epa) + 1]+), 1137 ([{ReO(epa)}20 + 1]+). lH NMR (CDCI3): 5.95 (s, 2H, ring-tf), 4.14 (m, 2H, ethylene-//), 3.90 (m, 2H, ethylene-H), 2.72 (s, 6H, imino-C#3), 2.28 (s, 6H, ring-C#3). IR (cm"1, KBr disk): 675 (s, vRe-o)-[TcOCl(epa)]-0.5H2O. To a mixture of [(n-Bu)4N][TcOCl4] (49.7 mg, 0.10 mmol) and H2epa (40.1 mg, 0.11 mmol) was added 25 mL of ethanol. The suspension became clear after refluxing for one hour. The solution was refluxed for two additional hours. The green yellow solution was then kept at -4 °C for days. A brown precipitate was filtered out, washed with Et20 and dried in air to yield 12 mg (23 % based on Tc). The product is slightly soluble in ethanol, chloroform, dichloromethane, and insoluble in Et20 and H2O. Anal. Calcd (found) for C18H19CIN2O7.5TC: C, 41.76 (41.89); H, 3.70 (3.81); N, 5.41 (5.58). LSIMS: m/z = 473 ([TcO(epa)]+). IR (cm"1, KBr disk): 960 (w, vTc=o)-[Re(dha)Cl2(OPPh3)(PPh3)]-EtOH. To a mixture of [ReOCl3(PPh3)2] (84.5 mg, 0.10 mmol) and H2ppa (39.1 mg, 0.10 mmol) was added 25 mL of ethanol. The suspension was refluxed overnight. The solution was clarified by filtration, then stored at -4 °C. Red crystals (X-ray quality) deposited from the solution. The supernatant solution was removed, then the 51 crystals were washed with diethyl ether, and dried in air. The product is soluble in chloroform and dichloromethane, and slightly soluble in diethyl ether. IR (cm-1, KBr disk): 695 (s, VRe-o)-X-ray Crystallography. The crystal structure of [Re(dha)Cl2(OPPh3)(PPh3)]-EtOH was determined by Dr. Steven J. Rettig of the UBC Structural Chemistry Laboratory. A summary of crystal data and final atomic coordinates and equivalent isotropic thermal parameters, are given in the Appendix II. The bond lengths and bond angles appear in Tables 3-1, 3-2, respectively. 3.3 Results and Discussion Ligands. The N2O2 Schiff base ligands, (H2epa and H2ppa) were prepared in a manner similar to that for the N3O2 Schiff base ligand, H3apa, i.e., the Schiff base condensation of ethylenediamine (en) or propylenediamine (pn), with two equivalents of dehydroacetic acid (dha). Each of these potentially tetradentate ligands contains an N2O2 donor set which may form three chelate rings. As in Hsapa, since the infrared spectra of the free ligands show very broad bands at 3500 cm"1 due to O-H stretching vibrations, the molecule probably exists mainly as the imine-enol form (Figure 3-2). This is also supported by the enol !H NMR signal, which lies downfield (8 > 14 ppm, as a broad singlet) due to intramolecular hydrogen bonding.13a Me p i Me 04 ) = N N={ o Me H3apa = (dha)2dien, Y = CH2NHCH2 H2epa = (dha)2en, Y = none H2ppa = (dha)2pn, Y = CH2 Figure 3-2. Tautomerism of the Schiff base ligands. 52 Synthesis of Complexes. [TcO(apa)] was synthesized via the ligand exchange route in moderate yields (49% and 54% from [TcOCU]' and [TcO(eg)2]", respectively; eg = ethylene glycol).9 [ReO(apa)] was prepared from the reaction of [ReOCl3(PPh3)2] or [ReCl4(PPh3>2] with the ligand in ethanol in the presence of excess triethylamine. With a N3O2 Schiff base ligand H3L, [MOL] was the only product. Upon coordination, the central amine group becomes very acidic due to the strongly electrophilic character of the [M=0]3+ core, and is deprotonated to give a neutral metal(V) complex. The formation of a Re(V) product from the reaction of [ReCl4(PPh3)2] and H3apa resulted from aerial oxidation.* Oxotechnetium(V) complexes have been shown to form in a similar fashion from the reactions of [TcCl4(PPh3)2] and some N2S2 ligands. In contrast to the N3O2 ligand H3apa, the coordination of the N2O2 Schiff base ligands to Tc(V) and Re(V) metal centers was found to be much less favorable, and more complicate. The reaction of H2epa with [ReOCl3(PPh3)2] produced the dinuclear complex [{ReO(epa)}20] in low yield, and attempts to increase this yield failed. The reaction of H2epa with [TCOCI4]-gave the complex [TcOCl(epa)], also in low yield. It is common for the yields of tetradentate N2O2 Schiff base metal complexes to be low. The low yields in our reactions with H2epa may be due to the diversity of the products (vide infra). In addition, the desired Schiff base complexes may be difficult to isolate due to reduction of the metal center (in the presence of PPI13) and/or hydrolysis of the Schiff base ligand during the reactions, as was seen for H2ppa (vide infra). Comparison of N3O2 and N2O2 ligands. It has been reported that substitution reactions onto the Tc0 3 + core with N2O2 Schiff base ligands might lead to three types of complexes, [TcOX(N202)], [{TcO(N202)}20] and [TcO(OH2)(N202)]+,4'6 in addition to an isolable intermediate [TcOCh(ON-NO)1~.4 Bandoli et al. assumed that all types of the complexes are formed but that the isolated product is the most stable; this assumption is supported by other investigations wherein several bands were observed when the reaction * In the reactions of [ReCl4(PPh3)2] and H2PO2HCI, [ReOCl(PPh3)(P02)] was obtained due to aerial oxidation of the starting material. Please refer to Chapter 4. 53 mixture was loaded onto a column. Product purification is the crucial step in most of the N2O2 cases; chromatography is often required. ' On the other hand, an N3O2 Schiff base ligand was found to form a [MO(N302)] complex in the only mode for the five donor atoms to surround the M=0 core, that in which the ligand is completely coordinated and there was only "1 Q one complex (two enantiomers) that can be formed. This sole pathway in formation of the complex rationalizes the higher yield in synthesis for N3O2 vs N2O2 ligands. It should also be mentioned that all the donor atoms in the N2O2 Schiff base complexes were found in the equatorial (relative to M=0) plane, in contrast with the behavior of the bidentate NO Schiff base ligands in the oxo complexes. Two NO Schiff base molecules in [MOX(NO)2] are never found to be co-planar, i.e., one NO bidentate ligand chelates in the plane equatorial to, the other in the plane axial to, the M=0 unit with the anionic donor atom trans to M=0. This non-coplanar configuration is thus considered to be thermodynamically more stable than a planar configuration. ' Based on their observations and other results, Refosco et al and Tisato et al have concluded that tetradentate N2O2 Schiff base ligands could form a twisted configuration only when there are five or more atoms in the central chain between the two imine N donor atoms. The inability to adopt the thermodynamically more favored twisted configuration dictates the inherent instability of the N2O2 complexes, and therefore the low yield in the synthesis. Meanwhile, side-reactions, such as reduction of the metal centers and/or hydrolysis of the bound ligands are enhanced. In the N3O2 complexes, such a twisted configuration is feasible, as is seen in all the three solved structures; high bond strains have not been found in the constructed models of such a configuration.8 The high yield of the N3O2 Schiff base complexes result from the only coordination mode for the five donor atoms in surrounding the M=0 core in a more stable configuration without high strain. In contrast, the N2O2 Schiff base complexes are in a much less stable configuration with a significant trans influence, when the four donor atoms of the ligand are bound to the metal in the equatorial plane and the oxo atom occupies the apical position. There are usually many anionic species competing for the sixth coordination site that is trans to, and labilized by, the M=0 group (trans effect), resulting in a low yield in isolation of any product. 54 Metal enhanced hydrolysis of the N2O2 Schiff base ligands also seriously hampers the synthesis of the complexes (vide infra). Formation of [Re(OPPh3)(dha)(PPh3)Cl2]EtOH. The trials with [ReOCl3(PPh3)2] and H2ppa in ethanol resulted in an unexpected Re complex, with one PPh3 migrating to the oxo oxygen, with the resultant vacant site filled, and one chloro ligand being replaced, by two oxygen donor atoms of dehydroacetate. The Re center is formally reduced to an oxidation state of three (Scheme 3-1). The Re=03+ core is known to react with phosphines and other oxygen acceptors, yielding Re(HI) complexes and phosphine oxides.153-16'17 In examining the RevO + X = Rera + XO transformation, Corny and Mayer observed that oxygen atoms were transferred to the substrate X when the X-0 bond strength exceeded 127 kcal/mol (X = PMe3 or PPh3), and the reverse reaction proceeded when the bond strength fell below 110 kcal/mol. Thus, the fact that the isolated complex is intermediate between the extremes of the expected oxygen transfer indicates that the oxygen transfer is not only related to the O-X bond energy, but also to the ancillary ligands. Jurisson et al observed that the reactions of acetylacetone based Schiff base complexes [TcO(OH2)(N202)]+ with three equivalents of phosphines PR3 yielded Tc(UI) complexes [Tc(PR3)2(N2C>2)] with the N2O2 Schiff base intact.5 •o* 6+0 Rev 8-Re v 3+ :PPh3 6+ PPh3 • • / ) (\V * Rev 5-n 3+ I PPh3 5+ Re111 8-3+ 3+ PPh3 >t l Re1" 5- J Scheme 3-1 The bound dha" is obviously the hydrolyzed product of the original Schiff base. Ligated Schiff bases are known to be susceptible to metal enhanced hydrolysis, as is known with some 19 lanthanide complexes. The hydrolysis of Schiff bases coordinated to Tc(V) is important in the development of 99mjc radiopharmaceuticals. While this hydrolysis may be detrimental in 55 many cases, it is postulated also to be advantageous in the design of 9 9nrr/c 20 radiopharmaceuticals that would decompose under certain conditions in vivo. Thus it is necessary to characterize the decomposed products. The isolation of this reduced- and hydrolyzed- Re species demonstrates one aspect of this type of decomposition. Characterization. N3O2 Complexes. The infrared spectrum of the [ReO(apa)] exhibits an VRe=o vibration at 945, higher than the VTC=0 band at 915 in [TcO(apa)];9 both are in the expected range for six-coordinated oxo complexes.15b The VTC=0 absorption is usually found at lower frequencies than VRe=o in analogous oxometal complexes.* LSIMS of the complex showed a molecular ion peak at m/z 603, confirming that the complex is monomeric [ReO(apa)]. The stoichiometry of [C2oH22N307Re] for the complex was confirmed by elemental analysis and is consistent with the [TcO(apa)] complex, the structure of which has been determined by X-ray diffraction methods. N2O2 Complexes. The N2O2 donor set is in the equatorial plane relative to the Tc=0 group in all the three types of complexes reported(vz"cfe supra). ' The infrared spectrum of [{ReO(epa)}20] exhibits two VRe_o vibrations at 675 and 645 cm"1, agreeing with the literature values of VM-0 for other [{MO(N202)hO] (M=Tc, Re) complexes.3'4'21,22 No absorption bands could be assigned to VRe=o; the two weak absorption bands at 950 and 910 cm-1 coincide with two moderate bands in the free ligands. A similar phenomenon was observed for other [{MO(N202)}20] complexes, and was not rationalized: there was no absorption which could be attributed to VTC=0 in an N2O2 Schiff base complex, and an N2O2 amine phenol complex; the VRe=o bands were found in only moderate or weak intensity for two N2O2 21 amine phenol complexes. LSIMS shows [ReO(epa) + 1]+ and [{ReO(epa)}20 + 1]+ at m/z 561 and 1037, respectively, suggesting that the complex is dinuclear with an oxygen atom bridging two ReO(epa) units. For [TcOCl(epa)], a weak IR absorption is present around 960 cm-1 and is tentatively assigned to Tc=0 stretching vibration and the m/z values corresponding to [TcO(epa)]+ and [Tc(epa)]+ are shown in the LSIMS, indicating a mononuclear oxo complex. The absence of the molecular ion in the mass spectrum may be explained by * Please refer to Chapters 2 and 4. 56 weakened bonding of Tc-Cl due to the trans influence of the oxo group. In a similar complex, [TcO{(sal)2en}Cl], this trans influence was found to weaken the Tc-Cl bond significantly.6 The stoichiometrics for [{ReO(epa)}20] and [TcOCl(epa)]0.5H2O were confirmed by elemental analyses. lH NMR Studies.* [ReO(apa)]. The resonances in the *H NMR spectrum (500 MHz) of [ReO(apa)] were assigned on the basis of the ^ H COSY (400 MHz, Figure 3-3), and by comparison with the *H NMR spectra for H3apa and [TcO(apa)].9 The [ReO(apa)] showed a spectrum similar to that of [TcO(apa)], except that resonances for the eight hydrogen nuclei of the two NCH2CH2N units were more distinguishable from one another. Thus it was possible to make a full assignment for the Re complex. For the free ligand, the seven types of resonances were assignable to enol H (14.00 ppm), ring H (5.50 ppm), imine CH2 (3.48 ppm), amine CH2 (2.92 ppm), imine CH3 (2.56 ppm), ring CH3 (2.00 ppm), and amine NH (1.45 ppm). As in [TcO(apa)], upon coordination the resonances corresponding to the enol H and amine NH disappeared due to deprotonation. Furthermore, the symmetry of the ligand is lost upon coordination; the two halves of the bound ligand, defined by the central nitrogen atom, became non-equivalent (as in [TcO(apa)]) as did the remaining hydrogens. The two halves were arbitrarily designated as "apical" and "equatorial", depending on the binding site of the enolate oxygen donor (Scheme 3-3, page 65). Thus two sets of the ring H, imine CH3 and ring CH3 hydrogen nuclei were observed, and were assigned to the two arms based on the fact that the signals of the nuclei (H) of the equatorial portion are more deshielded than those (H') of the apical.** The ^U-^H COSY of this complex showed the correlation between the ring H and ring CH3 of either upfield or downfield set, or of the assumed equatorial or apical arm. The same two sets of signals were also observed in the *H NMR spectrum of [TcO(apa)], which is known to have two non-equivalent halves of the bound ligand. Obviously, a similar geometry to that of [TcO(apa)] holds true for this [ReO(apa)] complex. * Spectra for [TcOCl(epa)]and [Re(OPPh3)(dha)(PPh3)Cl2] could not be obtained due to poor solubility and paramagnetism, respectively. As in [MOXL2], where the two bidentate L ligands are each bound equatorially and axially, the hydrogens in the equatorial ligand are usually more deshielded than those in the axial ligand. Please refer to Chapters 2 and 4. 57 Namido amido' CH'eH'a amido CHeHa imino' CH'eH'a imino CHeHa j i - 4.2 < . a %.l 5.2 4.8 4.4 4.0 3.6 ppm Figure 3-3. *H-JH COSY (400 MHz, ethylene region) for [ReO(apa)].* See footnotes in page 50. 58 The resonances for the two NCH2CH2N units, fall into eight almost distinguishable multiplets when observed on the 500 MHz spectrometer, indicating the non-equivalence of the two units. These hydrogen atoms were assigned based on ^ ^ H COSY (Figure 3-3) and on their similarity to the spectrum of [TcO(apa)]. The eight hydrogen signals were easily grouped into two sets, corresponding to the two NCH2CH2N units. Two of the four H signals in each unit, both appearing as a doublet of doublets, were assigned to the pseudo-equatorial H nuclei,2 as only these two could define a dihedral angle of close to 90° (Figure 3-3).* This assignment was supported by the fact that ring hydrogen atoms in pseudo-axial environments are usually more shielded than those in pseudoequatorial environments. b The other two in each unit were more complicated, either doublet of triplets or multiplets, due to unequal coupling to three hydrogen nuclei. This assignment is consistent with that of [TcO(apa)], which was based on ifl-1!! COSY, APT and !H-13C heteronuclear correlation spectra and those of the free ligand. [{ReO(epa)}20]. The *H NMR spectrum was easily assigned by comparison with that of the free ligand. For the free ligand, the five types of resonances were assigned to enol H (14.6 ppm), ring H (5.6 ppm), imine CH2 (3.8 ppm), imine CH3 (2.6 ppm), and ring CH3 (2.1 ppm). As expected, upon coordination, the enol H resonance disappeared while all other hydrogen resonances were shifted downfield from those of the free ligand. The ring H (5.95 ppm), imino CH3 (2.72 ppm), and ring CH3 (2.28 ppm) of the two halves of the ligand were indistinguishable, reflecting a symmetric molecule. The hydrogen resonances for the NCH2CH2N unit were split into two symmetrical multiplets, as an AA'BB' system, which was close enough, eventhough not exactly the same, to a computed simulation of such a system (Figure 3-4). The *£! NMR results suggested a structure for the complex with co-planar coordinated ligands, consistent with a proposed dinuclear structure with i symmetry (known for other dinuclear Re(V) or Tc(V) complexes with two N2O2 ligands.3'4'21'22 * According to the Karplus Equation, 3J = 4.2 - 0.5cos6 - 4.5cos26 a very small coupling constant (less than 1 Hz) is expected when 9 is close to 90°. 59 *S**ftw»r**^ 4.29 4.16 — I — 4.03 3.90 — I — 3.77 ppm Figure 3-4. Ethylene Hydrogen Region of the *H NMR spectrum (top, 400 MHz) and a computer-simulated AA'BB' system (8A = 4.14, 5B = 3.90; JAB =13.2, JAB' = 6.5) for [{ReO(epa)}20]. * = unidentified species; # = impurities 60 Structure of [Re(dha)Cl2(OPPh3)(PPh3)]EtOH. The ORTEP drawing of the complex is given in Figure 3-5. The unit cell contains four discrete [Re(dha)Cl2(OPPh3)(PPh3)]-EtOH molecules (two of each enantiomer), each of which has the Re(III) center coordinated by the two O donor atoms of the dha- ligand, two CI donor atoms, one triphenylphosphine P donor atom and one triphenylphosphine oxide O donor atom. The coordination geometry is a distorted octahedron with the enolic oxygen of the dha" located in the position trans to the Re-0=PPh3 unit and the remaining donor atoms occupying equatorial sites relative to the 0=PPh3 group such that the two chloro ligands are cis to each other. The distortion from an ideal octahedron is mainly caused by the variety of donor atoms (Table 3-1). In the absence of the sterically demanding oxo group, all the cis angles are close to 90°, with deviations of less than 4° (Table 3-2). The Re-Cl(l) bond that is trans to the phosphine P is long (2.430(1) A), while the Re-Cl(2) cis to the phosphine P is normal (2.362(1) A) in comparison with the Re-Cl distances of 2.35 and 2.36 A in ?ran5-(P,P)-[ReCl3(PPh3)2(CH3CN)],25 indicating that the phosphine P exerts a significant trans influence. The Re-P distance (2.404(1) A) is short in comparison to Table 3-1. Selected Bond Lengths (A) for [Re(dha)Cl2(OPPh3)(PPh3)]EtOH. Re(l)-Cl(l) Re(l)-P(l) Re(l)-0(4) 0(1)-C(1) P(2)-0(5) 0(4)-C(6) C(2)-C(3) C(3)-C(4) 2.430(1) 2.404(1) 2.061(3) 1.407(5) 1.502(3) 1.276(5) 1.420(5) 1.433(5) Re(l)-Cl(2) Re(l)-0(3) Re(l)-0(5) 0(1)-C(5) 0(3)-C(3) C(l)-C(2) C(2)-C(6) C(4)-C(5) 2.3621(1) 1.987(2) 2.051(2) 1.375(5) 1.278(4) 1.437(6) 1.421(6) 1.334(6) 61 C46 C24 C25 C30 C23 C18 C19 C42< Figure 3-5. ORTEP drawing of [Re(dha)Cl2(OPPh3)(PPh3)]EtOH. 62 Table 3-2. Selected Bond Angles (deg) for [Re(dha)Cl2(OPPh3)(PPh3)]EtOH. Cl(l)-Re(l)-Cl(2) Cl(l)-Re(l)-0(3) Cl(l)-Re(l)-0(5) Cl(2)-Re(l)-0(3) Cl(2)-Re(l)-0(5) P(l)-Re(l)-0(4) 0(3)-Re(l)-0(4) 0(4)-Re(l)-0(5) Re(l)-0(4)-C(6) 88.87(4) 87.91(8) 90.84(8) 93.38(7) 92.05(8) 92.02(8) 86.8(1) 87.6(1) 131.9(3) Cl(l)-Re(l)-P(l) Cl(l)-Re(l)-0(4) Cl(2)-Re(l)-P(l) Cl(2)-Re(l)-0(4) P(l)-Re(l)-0(3) P(l)-Re(l)-0(5) 0(3)-Re(l)-0(5) Re(l)-0(3)-C(3) Re(l)-0(5)-P(2) 176.59(4) 86.72(8) 92.40(4) 175.58(8) 88.86(7) 92.28(8) 174.4(1) 129.6(2) 158.9(2) the Re-P bonds of 2.47 and 2.48 A in ?rans-(P,P)-[ReCl3(PPh3)2(CH3CN)]. This can be explained by the absence of a trans influence attributable to this PPh3 group. The two Re-O bonds are both close to the lengths found for other Re-0 single bonds, with the Re-0(3) bond being 1.987(2) A, and the Re-0(4) bond being 2.016(3) A, indicating the two oxygen donor atoms are involved in a 7t-delocalization within the ligand (Scheme 3-2); in fact, an averaging of bond lengths is seen for this chelate ring (Table 3-1). The distance for Re to phosphine oxide oxygen is longer (2.051(2) A), but still comparable to the Re-O single bond. In contrast, the same oxygen is strongly bound to the phosphorus (O-P = 1.502(3) A), only about 0.02 A longer than the values found in the unligated OPPh3 (O-P = 1.483(2) A),27 where the P-O is believed to be multiply bound. The maintenance of multiple character of the P-0 bond* * ? 28 A typical P-0 single bond is 1.60 A in length. 63 suggests that the O-Re is a coordinate bond. Given the bonding features, it is analogous to a Re(III) complex of a phosphine oxide.* Scheme 3-2 3.3 Conclusion The N3O2 dehydroacetic acid based Schiff base ligand precursor H3apa has greater potential for use in radiopharmaceutical studies than the N2O2 Schiff base analogs, as stable oxotechnetium and oxorhenium complexes of H3apa can be prepared in good yields. Complexes of N 2 0 2 Schiff bases may be obtained ([MO(N202)X], [{MO(N202)}20], etc., Scheme 3-3); however, their formation is much less favorable, due to a variety of reaction pathways and a less stable coordination configuration. This severely limits the use of these tetradentate ligands in the design of radiopharmaceuticals. Based on the stoichiometry and geometry, the complex could also be viewed as an adduct of an oxorhenium(V) complex with a PPh3 bound to the oxo group. 64 H3apa = (dha)2dien, Y = CH2NHCH2 H2epa = (dha)2en, Y = none H2ppa = (dha)2pn, Y = CH2 H3apa tReOCl3(PPh3)2] or [ReCl4(PPh3)2] Me I \ o y-o N „ \ — / I \V Me Me °YST° V Me [ReO(apa)] Hjppa [ReOCl3(PPh3)2] H2epa \rReOCl3(PPh3)2] Me-0 Ph3P o-3-o-PPh3 II 0 : R < C I 0 Me [Re(dha)Cl2(OPPh3)(PPh3)] Me f~\ Me HKk Me 0 Me Me \ Me KVtR o > = N " N = < O Me \ / Me [TcOCl(epa)] Scheme 3-3 [{ReO(epa)}20] 65 References (1) Dilworth, J. R. Transition Met. Chem. 1990,15, 411. (2) Calligaris, M.; Randaccio, L. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A., Eds.; Pergamon: Oxford, 1987; Vol. 2; p 715. (3) Tisato, F.; Refosco, F.; Mazzi, U.; Bandoli, G.; Dolmella, A. Inorg. Chim. Acta 1989, 164, 127. (4) Bandoli, G.; Niconili, M.; Mazzi, U.; Refosco, F. J. Chem. Soc, Dalton Trans. 1984, 2505. (5) Jurisson, S.; Dancey, K. P.; McPartlin, M.; Tasker, P. A.; Deutsch, E. Inorg. Chem. 1984, 23,4743. (6) Jurisson, S.; Lindoy, L. F.; Dancey, K. P.; McPartlin, M.; Tasker, P. A.; Uppal, D. K.; Deutsch, E. Inorg. Chem. 1984, 23, 227. (7) Tisato, F.; Refosco, F.; Mazzi, U.; Bandoli, G.; Nicolini, M. Inorg. Chim. Acta 1991, 189, 97. (8) Refosco, F.; Tisato, F.; Mazzi, U.; Bandoli, G.; Nicolini, M. /. Chem. Soc, Dalton Trans. 1988,611. (9) Liu, S.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 4915. (10) Davison, A.; Trop, H. S.; Depamphilis, B. V.; Jones, A. G. Inorg. Synth. 1982, 21, 160. (11) Chatt, J.; Rowe, G. A. /. Chem. Soc. 1962,4019. (12) Rouschias, G.; Wilkinson, G. J. Chem. Soc. (Sect. A) 1966,465. (13) Silverstein, R. M.; Bassler, G. C ; Morrill, T. C. In Spectrometric Identification of Organic Compounds; 4 ed.; John Wiley & Sons: New York, 1981; a. p 196; b. p 189. (14) Marchi, A.; Marvelli, L.; Rossi, R.; Magon, L.; Bertolasi, V.; Ferretti, V.; Gilli, P. J. Chem. Soc, Dalton Trans. 1992, 1485. 66 (15) Nugent, W. A.; Mayer, J. M. Metal-LigandMultiple Bonds; Wiley-Interscience: New York, 1988; a. 241 ; b. 116. (16) Holm, R. H.; Donahue, J. P. Polyhedron 1993,12, 571. (17) Holm, R. H. Chem. Rev. 1987, 87, 1401. (18) Corny, R. R.; Mayer, J. M. Inorg. Chem. 1990, 29, 4862. (19) Liu, S.; Gelmini, L.; Rettig, S. J.; Thompson, R. C; Orvig, C. /. Am. Chem. Soc. 1992, 114, 6081 and the references therein. (20) Duatti, A.; Marchi, A.; Magon, L.; Deutsch, E.; Bertolasi, V. Inorg. Chem. 1987, 26, 2182. (21) Pillai, M. R. A.; Barnes, C. L.; Schlemper, E. O. Polyhedron 1994,13, 701. (22) Pillai, M. R. A.; John, C. S.; Schlemper, E. O.; Troutner, D. E. Inorg. Chem. 1990, 29, 1850. (23) Cole, E.; Copley, R. C. B.; Howard, J. A. K.; Parker, D.; Ferguson, G.; Gallagher, J. F.; Kaitner, B.; Harrison, A.; Royle, L. /. Chem. Soc, Dalton Trans. 1994, 1619. (24) Creswell, C. J.; Runquist, O. A.; Campbell, M. M. In Spectral Analysis of Organic Compounds; 2 ed.; Longman: New York, 1972; p 141. (25) Drew, G. B.; Tisley, D. G.; Walton, R. A. J. Chem. Soc, Chem. Commu. 1970, 600. (26) Conner, K. A.; Walton, R. A. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A., Eds.; Pergamon: Oxford, 1987; Vol. 4; p 125. (27) Goggin, P. L. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A., Eds.; Pergamon: Oxford, 1987; Vol. 2; p 487. (28) Gilheany, D. G. In The Chemistry of Organophosphorus Compounds; Hartley, F. R., Ed.; John Wiley & Sons: Chichester, 1992; Vol. 2, Chapter 1 (Structure and Bonding in Tertiary Phosphine Chalcogenides); p 1. 67 Chapter Four Rhenium(V) and Technetium(V) Complexes Incorporating the Deprotonated Forms of (o-hydroxyphenyl)diphenylphosphine (HPO) and Bis(0-nydroxyphenyl)phenylphosphine (H2PO2) 4.1 Introduction The necessity of finding a suitable reducing agent is very important in the preparation of 99mjc-radiopharmaceuticals. Stannous chloride, the most commonly used reductant in the "radiopharmaceutical kits" is effective; however, it is also easily oxidized by oxygen, and is rapidly hydrolyzed.1 With the metal ion reductant and the ligand both in large excess compared to the nanomolar scale of the radionuclide, the reactions involved can be very complex. Triphenylphosphine is known to form complexes with Re and Tc in various oxidation states, and neutral complexes are formed by incorporating halo and/or oxo ligands. These complexes are usually prepared from perrhenate or pertechnetate by reaction with the phosphine in the presence of a hydrohalic acid, the phosphine functioning as both ligand and reductant. The ease of synthesis makes many of these complexes good starting materials (via ligand-exchange); however, there are no reports on their possible application in radiopharmaceutical studies, presumably due to the hydrolyzable monodentate ligands (e.g. CI). Phosphines functionalized with anionic groups have appeared in Tc/Re chemistry. Functionalization at an ortho position of one or more phenyl groups on triphenylphosphine leads to potentially multidentate and, upon deprotonation, quite basic phosphine ligands, which may be good for the preparation of new hydrolytically stable Tc/Re complexes. Davison and coworkers have synthesized three Tc(IH) complexes with an umbrella tetradentate triprotic phosphinotrithiol (PS3) ligand. ' Refosco et al. have studied Tc(III) complexes with bidentate monoprotic (PX) 68 ligands (X = N,8 S, or O,9) and Tc(V) and Re(V) complexes with the PN ligand.10 In all the above cases, the anionic functional groups neutralized all or part of the positive charge at the metal centers while forming neutral Tc/Re complexes. •t> PO-There are, to our knowledge, no reports of Tc/Re complexes with potentially tridentate diprotic PX2 phosphine ligands. To initiate these studies, we have focused on bis(o-hydroxyphenyl)phenylphosphine (H2PO2). To the best of our knowledge, H2PO2 as a ligand has not been investigated; however, it has been used as a precursor in the synthesis of macrocyclic ligands. (Very recently, Fe and Co complexes of PO22" and P033~ ligands were 1 9 reported at a meeting. ) This functionalized phosphine ligand H2PO2 was prepared by a convenient large scale route, and its coordination chemistry as well as that of the monoprotic (o-hydroxyphenyl)diphenylphosphine (HPO), with pentavalent Tc and Re was investigated. This chapter presents the results of the studies. 4.2 Experimental Materials. All chemicals were reagent grade and were used as received: phenol, PPI13, Ph2PCl, PhPCl2, (MeO)2CH2 (dimethoxymethane), n-BuLi, and TMEDA (N,N,N'N'-tetramethylethylenediamine) were from Aldrich; NH4Re04 was a gift of Johnson-Matthey, Inc.; HC1 gas was from Matheson; [NH4][99TcC>4] was a gift from the Du Pont Merck Pharmaceutical Company. PhOCH 2 OCH 3 (mom-protected phenol),1 3 [ (n -Bu) 4 N][TcOCl 4 ] , 1 4 [ReOCl3(PPh3)2], [ReNCl2(PPh3)2],15'16and (o-hydroxyphenyl)diphenylphosphine (HPO)17 69 were prepared according to published procedures. [Re(NPh)Cl3(PPh3)2] was prepared by following a preparation for [Re(NPh)Cl3(PPhEt2)2].16 Caution! " T c is a low energy (0.292 MeV) p - emitter with a half-life of 2.12 x 105 years. All manipulations of solutions and solids were performed in a laboratory approved for the handling of low-level radioactivity, and normal safety procedures were followed at all times to prevent contamination. Instrumentation. Mass spectra were obtained with either a Kratos MS 50 (electron impact ionization, EIMS) or a Kratos Concept IIH32Q instrument (Cs+-LSIMS with positive or negative ion detection). Only the most intense peaks are given where consistent isotopic patterns were observed. Infrared spectra were recorded as KBr pellets in the range 4000-400 cm-1 on a Perkin-Elmer PE 783 spectrophotometer and were referenced to polystyrene. Microanalyses were performed by Mr. P. Borda in this department or by Canadian Microanalytical Services Ltd. (Tc complexes). *H NMR spectra (200, 400 or 500 MHz) were recorded on Bruker AC-200E, Bruker WH-400 ^ H ^ H COSY), or Bruker AMX-500 (!H{31P}) spectrometers with 5 referenced to external TMS. The 31P{1H} NMR spectra (81 or 121 MHz) were recorded on Bruker AC-200E or Varian XL 300 spectrometers, respectively, with 8 referenced to external phosphoric acid. The assignments were based on those for the unbound ligand and those for analogous complexes. PhP(o-C6H4OCH2OCH3)2 ((mom)2P02 , mom = CH 2OCH 3) . This was prepared from the mom-protected phenol according to a procedure for Ph2P(o-C6H4OCH2OCH3) ((mom)PO) with some modifications.18 To an ice-cooled solution of methoxymethyl phenyl ether (20.6 g, 149 mmol) in ca. 200 mL of petroleum ether (b.p. 35-65 °C, dried with anhydrous Na2S04 overnight) was added a suspension of 100 mL of 1.6 M (n-Bu)Li in hexanes and 17.5 g TMEDA in 50 mL of petroleum ether under N2. The mixture was stirred overnight at room temperature. A yellow precipitate formed from the orange solution, and the mixture was subsequently heated to about 40 °C under stirring. After the mixture was cooled to 0 °C, PhPCl2 (21.7g, 121 mmol) was added via a syringe. The resultant mixture was stirred for another 10 h, during which time it warmed to room temperature. The solvents were removed 70 by rotary evaporation, and to the residue was added Na2HPC>4 (0.1 M, 100 mL). The reaction mixture was then extracted with Et20 (2 x 200 mL) followed by CHCI3 (2 x 100 mL). All the organic layers were combined, concentrated to a reddish oil under low pressure, diluted with Et20 (ca. 20 mL), and stored at -4 °C overnight. A crystalline product was filtered out, washed with cold methanol (2 x 10 mL), and dried in vacuo. The yield was 15.6 g (55% based on mom-protected phenol). Anal. Calcd (found) for C22H23O4P: C 69.1 (68.8), H 6.1 (6.1). EIMS: m/z= 382 ([(mom)2P02]+), 367 ([(mom)2P02 - CH3]+). *H NMR (CDCI3): 7.35-7.25 (overlapped multiplets, 7H); 7.10-7.00 (multiplets, 2H); 6.84 (t, 2H); 6.75-6.65 (multiplets, 2H); 5.04 (d, 2H, CH2, 2 / H H ' = 8 Hz), 5.00 (d, 2H, CH'2, 2 /HH' = 8 Hz); 3.1 (s, 6H, CH3). 31P{lU} NMR (CDCI3): -26.1 (s). IR (cm"1, KBr disk): 3020 (m, VC -H); 3000-2800 (m, methyl and methene VC-H)-Bis(o-hydroxyphenyl)phenylphosphine hydrochloride, H2P02'HC1. This was prepared from the mom-protected phenol phosphine ((mom)2P02) according to a procedure 18 for HPO, with some modifications. Into a solution of (mom)2P02 (10. lg, 26.4 mmol) in 400 mL of anhydrous methanol (or ethanol) was bubbled anhydrous HCl gas via a dispersion tube for ca. 6 h with stirring. The mixture was further stirred overnight. The solution was then concentrated to off-white solids, which were washed with methanol (3 x 15 mL) and dried in vacuo. The yield was 5.8 g (67% based on phosphine); no recrystallization was necessary to obtain an analytically pure sample. Anal. Calcd (found) for C18H16CIO2P: C 65.4 (65.0), H 4.9 (4.9), CI 10.7 (10.9). EIMS: m/z= 294 ([H2P02]+), 199 ([H2P02 - C6H70]+). !H NMR (DMSO-^6): 10.2 (broad s, 2H), 7.8-7.3 (overlapped multiplets, 9H), 7.0-6.8 (multiplet, 4H). ^ P ^ H J N M R : 34.6 (s, DMSO-^6); -30.6 (s, py-d5). IR (cm"1, KBr disk): 3020 (vs, b, Vc-H)-m-(P,P)-[ReOCl(PO)2]-0.5H2O. To a mixture of [ReOCl3(PPh3)2] (84 mg, 0.1 mmol) and HPO (60 mg, 0.22 mmol) was added 10 mL of ethanol. The mixture was brought to reflux for 1/2 h, three drops of triethylamine were added, and the reaction mixture was refluxed for a further hour. After the mixture was cooled to room temperature, green solids were filtered out and recrystallized from CH2Cl2/Et20. The final product, yellowish green crystals, was 71 washed with Et20 and dried in vacuo overnight. The yield was 54 mg (67%). The product is soluble in acetone, acetonitrile, chloroform and dichloromethane, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for C36H29C103.5P2Re: C 54.0 (54.0), H 3.7 (3.5), CI 4.4 (4.4). LSIMS: m/z= 792 ([ReOCl(PO)2]+), 757 ([ReO(PO)2]+). *H NMR (CDCI3): 7.70-7.28 (overlapped multiplets, 16H), 7.20-7.04 (overlapped multiplets, 4H), 6.9 (overlapped multiplets, 4H, including t, 1H, p-H on the equatorial PO-phenyl ring, t, 1H, m'-H on the axial PO-phenyl ring, dd, 2H, m-Ph-H), 6.64 (t, 1H, p-H on the axial PO-phenyl ring), 6.56 (dd, 2H, o-Ph-H), 6.01 (dd, 1H, o'-H on the axial PO-phenyl ring). 3i?{lH} NMR (CDCI3): 15.4 (d), 2.2 (d); 2 7 P P = 10.1 Hz. IR (cm-1, KBr disk): 3420 (br, VO-H); 3060 (m, VC-H); 965 (s, vRe=o)-c i s - ( P , P ) - [ T c O C l ( P O ) 2 K C H C l 3 h / 6 . A procedure similar to that for [ReOCl(PO)2] was followed using [(n-Bu)4N][TcOCl4] (51 mg, 0.1 mmol) and HPO (59 mg, 0.21 mmol), except that no base was added and the brown precipitate was recrystallized from CHCl3/Et20. The yield of the purple crystalline product was 26 mg (36 %). The product is soluble in chloroform and dichloromethane, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for C36H28C103P2Tc-(CHCl3)i/6: C 59.9 (59.9), H 3.9 (4.2), CI 7.3 (7.5). LSIMS: m/z= 704 ([TcOCl(PO)2]+), 669 ([TcO(PO)2]+). *H NMR (CDCI3): 7.75-7.25 (overlapped multiplets, 16H), 7.18-7.04 (overlapped multiplets, 4H), 6.9 (overlapped multiplets, 4H, including p-H on the equatorial PO-phenyl ring, m'-H on the axial PO-phenyl ring, m-Ph-H as in Re analog), 6.64 (t, 1H, p-H on the axial PO-phenyl ring), 6.60 (dd, 2H, o-Ph-H), 6.0 (dd, 1H, o'-H on the axial PO-phenyl ring) IR (cm"1, KBr disk): 3060 (m, V C -H); 940 (s, VTC=O)-m*r-(P,P,P)-[ReN(PO)2(PPh3)]-H20. To a mixture of [ReNCl2(PPh3)2] (87 mg, 0.11 mmol) and HPO (77 mg, 0.28 mmol) was added 50 mL of ethanol, and the mixture was brought to reflux overnight. To the orange red solution was added 10 mL of cyclohexane, then the solution was clarified by filtration, and stored at 5 °C for slow evaporation. Green yellow crystals were filtered out, washed with cyclohexane, and dried in vacuo overnight. The yield was 42 mg (38%). The product is soluble in chloroform and dichloromethane, moderately soluble in ethanol, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for 72 C 5 4H45N0 3 P3Re: C 62.7 (62.8), H 4.4 (4.2), N 1.4 (1.2). LSIMS: m/z= 1018 ([ReN(PO)2(PPh3) + 1]+), 755 ([ReN(PO)2]+). *H NMR (CD2C12): 8.08 (dd, 2H), 7.98 (dd, 2H), 7.55-6.68 (overlapped multiplets, 33H), 6.53 (t, IH), 6.31 (overlapped multiplets, 3H), 6.22 (t, IH, p-U on the axial PO-phenyl ring), 5.87 (dd, IH, o'-H on the axial PO-phenyl ring). ^ P ^ H } NMR (CDC13): 34.7 (dd, PA), 23.0 (dd, PB), 13.9 (multiple!, Px) ; 2 / A B = 222.3, 2 / A X = 11-5, 2 / B X = 5.7 Hz. IR (cm-1, KBr disk): 3430 (br, V 0 -H) ; 3060 (m, VC-H); 1045 (m, VReEN)-/ac-cis-(P,P)-[ReOCl(PPh3)(P02)]. To [ReOCl3(PPh3)2] (85 mg, 0.1 mmol) and H2P02HC1 (64 mg, 0.19 mmol) was added 10 mL of ethanol. The mixture was brought to reflux for 1.5 h and then cooled to room temperature, whereupon the solvent was removed from the green solution by rotary evaporation. A golden crystalline product was obtained after the residue was recrystallized from CHC13/Et20. The crystals were filtered out, washed with cyclohexane, and dried in vacuo overnight. The yield was 31 mg (39% based on H2P02HC1). The product is soluble in chloroform and dichloromethane, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for C36H28C103P2Re: C 54.6 (54.2), H 3.6 (3.7), CI 4.5 (4.6). LSIMS: m/z= 793 ([ReOCl(PPh3)(P02) + 1]+), 757 ([ReO(P02)(PPh3)]+), 530 ([ReOCl(P02)]+), 495 ([ReO(P02)]+). !H NMR (CDCI3): 7.48 (overlapped multiplets, 8H), 7.40-7.26 (overlapped multiplets, 9H), 7.17 (overlapped multiplets, 7H), 6.84 (t, IH, m'-H on the axial PO-phenyl ring), 6.73 (t, IH), 6.61 (t, lH,p-H on the axial PO-phenyl ring), 5.76 (dd, IH, o'-H on the axial PO-phenyl ring). 31P{!H} NMR (CD2C12): 19.5 (d), -2.1 (d); 2 / P P = 7.2 Hz. IR (cm-1, KBr disk): 3060 (m, VC-H); 965 (s, VRe=o)-/ac-cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)]-2CHCl3. To [Re(NPh)Cl3(PPh3)2] (93 mg, 0.11 mmol) and H2P02HC1 (31 mg, 0.095 mmol) was added 10 mL of benzene and 3 mL of ethanol. The mixture was brought to reflux overnight. After cooling to room temperature, a dark green product was filtered out and recrystallized from chloroform/hexane. The crystals were washed with cyclohexane, and dried in vacuo overnight. The yield was 54 mg (65% based on H2P02HC1). The product is soluble in chloroform and dichloromethane, moderately soluble in benzene, methanol, ethanol, and acetone, but insoluble in diethyl ether, water, or cyclohexane. 73 Anal. Calcd (found) for C44H35Cl7N02P2Re: C 47.8 (47.8), H 3.2 (3.1), N 1.3 (1.3). LSIMS: m/z= 867 ([Re(NPh)Cl(PPh3)(P02)]+), 832 ([Re(NPh)(PPh3)(P02)]+), 605 ([Re(NPh)Cl(P02)]+), 570 ([Re(NPh)(P02)]+). *H NMR (CDCI3): 7.50 (dd, 6H, o-H on PPh3), 7.42-7.10 (overlapped multiplets, 18H), 7.01 (t, 1H), 6.87 (t, 1H, m'-H on the axial PO-phenyl ring), 6.79 (t, 2H, m-H on NPh), 6.58 (t, 1H), 6.55 (t, 1H, p-H on the axial PO-phenyl ring), 6.43 (d, 2H, o-H on NPh), 6.04 (dd, 1H, o'-H on the axial PO-phenyl ring). 31P{lH} NMR (CDCI3): 28.3 (s),* 15.4 (s). IR (cm'1, KBr disk): 3060, 2990 (m, V C -H) ; 1030 (m, vRe=Nph). /ac-/rans-(P,P)-[Re(NPh)Cl(PPh3)(P02)]-H20. To a solution of H 2 P 0 2 H C 1 (41.1 mg, 0.12 mmol) in 8 mL of methanol was added [Re(NPh)Cl3(PPh3)2] (90 mg, 0.10 mmol). The mixture was brought to reflux under N2 for 1 h and then cooled to room temperature. A brown precipitate was filtered out, washed with cold methanol and diethyl ether, and dried in vacuo overnight. The yield was 54 mg (63%). The product is soluble in chloroform, dichloromethane and 1,2-dichloropropane, moderately soluble in acetonitrile, and insoluble in cold methanol, diethyl ether or cyclohexane. Anal. Calcd (found) for C 4 2 H 3 5 C l N 0 3 P 2 R e : C 57.0 (56.8), H 4.0 (3.7), N 1.6 (1.7). LSIMS: m/z= 867 ([Re(NPh)Cl(PPh3)(P02)]+), 832 ([Re(NPh)(PPh3)(P02)]+), 605 ([Re(NPh)Cl(P02)]+), 570 ([Re(NPh)(P02)]+). *H NMR (CDCI3): 7.85 (dd, 6H, o-H on PPh3), 7.70(dd, 2H, o-H on PPh), 7.50 (t, 1H, m-H on the equatorial PO-phenyl ring), 7.46-7.18 (overlapped multiplets, 15H), 7.08 (t, 1H, m'-H on the axial PO-phenyl ring), 7.02 (dd, 1H, o'-H on the equatorial PO-phenyl ring), 6.78 (t, 2H, m-H on NPh), 6.68 (t, 1H, p-U on the equatorial PO-phenyl ring), 6.62 (d, 2H, o-H on NPh), 6.60 (t, 1H, p-U on the axial PO-phenyl ring), 6.32 (dd, 1H, o'-H on the axial PO-phenyl ring). 31P{lH} NMR (CDCI3): 34.5 (d), 2.9 (d); 2 / P P = 241.2 Hz. IR (cm-1, KBr disk): 3420 (br, V 0 -H); 3060 (m, VC-H); 1030 (m, vRe=NPh)-/ac-c/s-(P,P)-[ReO(PO)(P02)]-1.5H20. To a solution of HPO (16 mg, 0.060 mmol) in 11 mL of ethanol was added/ac-cw-(P,P)-[ReOCl(PPh3)(P02)] (36 mg, 0.046 mmol). The mixture was brought to reflux under N2 for 3.5 h and to it was subsequently added NaOAc This signal became a doublet at -20 °C or lower temperatures, with the coupling constant varying with temperature (5.2 Hz at -20 °C, 16.4 Hz at -70 °C). 74 (4.1 mg, 0.050 mmol) in 2 mL of ethanol. After one additional hour of refluxing, the solution was stored at 5 °C for slow evaporation. A brownish green crystalline product was filtered out, washed with cold ethanol and diethyl ether, then dried in vacuo overnight. The yield was 24 mg (62%). The product is soluble in chloroform, but insoluble in ethanol or diethyl ether. Anal. Calcd (found) for C36H3o05.5P2Re: C 54.1 (53.9), H 3.8 (3.7). LSIMS: m/z= 773 ([ReO(PO)(P02) + 1]+), 495 ([ReO(P02)]+). *H NMR (CDC13): 7.85 (dd, 2H), 7.65-7.00 (overlapped multiplets, 21H), 6.77 (t, 1H), 6.62 (t, 1H, ra'-H on the axial PO-phenyl ring), 6.42 (t, lH,/?-H on axial PO-phenyl ring), 5.68 (dd, 1H, o'-H on the axial PO-phenyl ring). 31P{ !H} NMR (CDCI3): 20.6 (d), 16.0 (d); 2 7 P P = 4.0 Hz. IR (cnr1, KBr disk): 3420 (br, VO-H); 3060 (m, VC-H); 965 (s, vRe=o)-frans-(P,P)-[Re(NPh)(PO)(P02)]. To a solution of HPO (17 mg, 0.060 mmol) in 10 mL of methanol was added frans-(P,P)-[Re(NPh)Cl(PPh3)(P02)]H20 (42 mg, 0.048 mmol). The mixture was stirred for 10 min; then NaOAc (4.7 mg, 0.05 mmol) in 2 mL of methanol was added dropwise. The mixture was refluxed under N2 for 4.5 h, then the solution was clarified by filtration, and stored at 5 °C for slow evaporation. The supernatant was removed and brownish green crystals were collected, dried in vacuo overnight. The yield was 15 mg (35%). The product is soluble in chloroform, acetonitrile, acetone, but insoluble in ethanol or diethyl ether. Anal. Calcd (found) for C42H32N03P2Re: C 59.6 (59.4), H 3.8 (3.9), N 1.7 (1.6). LSIMS: m/z= 848 ([Re(NPh)(PO)(P02) + 1]+), 570 ([Re(NPh)(P02)]+). ! H NMR (CDCI3): 7.95-6.40 (overlapped multiplets, 31H), 5.97 (dd, 1H, o'-H on axial PO-phenyl ring). 31P{1H} NMR (CDCI3): 35.7 (d), 15.4 (d); 2 / P P = 229.6 Hz. IR (cm"1, KBr disk): 3060 (m, VC-H); 1030 (m, VRe=NPh)-cw-(P,P)-[ReO(P02)(HP02)]. To NH4Re04 (27 mg, 0.10 mmol) and H2P02HC1 (119 mg, 0.36 mmol) was added 40 mL of ethanol. The mixture was refluxed overnight. After cooling, a green crystalline product was filtered out, washed with methanol and diethyl ether, then dried in vacuo. The yield was 66 mg (84%). The product is soluble in dimethyl sulfoxide and N,N'-dimethylformamide, slightly soluble in chloroform, dichloromethane, ethanol and methanol, but insoluble in benzene, diethyl ether, water or cyclohexane. Anal. Calcd (found) for 75 C36H2705P2Re: C 54.9 (54.7), H 3.5 (3.5). LSIMS: m/z= 789 ([ReO(P02)(HP02) + 1]+), 771 ([Re(P02)2]+), 495 ([ReO(P02)]+); 787 ([ReO(P02)2]-). ! H NMR (DMSO-J6): 9.95 (broad s, OH), 8.0-6.30 (overlapped multiplets), 5.45 (dd, 1H of the minor isomer, o'-H on the axial PO-phenyl ring), 5.23 (dd, 1H of the major isomer, o'-H on the axial PO-phenyl ring). 31P{ !H} NMR (DMSO-d6): 20.9 (d), 11.6 (d); 2 / P P = 4.1 Hz; a minor set also observed: 22.8 (d), 17.5 (broad, s); 2 / P P = 5.0 Hz). IR (cm'1, KBr disk): 3060 (m, V C -H) ; 990, 970 (m, VRe=o)-cw-(P,P)-[TcO(P02)(HP02)]. Method A. To NH 4Tc0 4 (17.5 mg, 0.10 mmol) and H2P02HC1 (107 mg, 0.32 mmol) was added 20 mL of ethanol. The solution changed to a brownish red color immediately, and then dark brown upon heating. The mixture was refluxed overnight. After cooling a brown powder was filtered out, washed with cold ethanol and diethyl ether, then dried in vacuo. The yield was 61 mg (87%). The product is soluble in dimethyl sulfoxide and N,N'-dimethylformamide, very slightly soluble in chloroform, dichloromethane, methanol and ethanol, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for C36H2705P2Tc: C 61.7 (61.4), H 3.9 (4.0). LSIMS (sample ground with KBr): m/z= 739 ([TcO(P02)(HP02) + K]+), 701 ([TcO(P02)(HP02) + 1]+). *H NMR (DMSO-d6): 10.15 (broad, s, OH), 8.0-6.30 (overlapped multiplets), 5.45 (dd, 1H of the minor isomer, o'-H on the axial PO-phenyl ring), 5.23 (dd, 1H of the major isomer, o'-H on the axial PO-phenyl ring). IR (cm-1, KBr disk): 3050 (m, VC-H); 965 (s, vTc=o)-Method B. To [(n-Bu)4N][TcOCl4] (52 mg, 0.10 mmol) and H2P02HC1 (64 mg, 0.19 mmol) was added 10 mL of ethanol. The mixture was brought to reflux for 4 h, then it was cooled to -4 °C. A brown powder was filtered out, washed with methanol and diethyl ether, and then dried in air. The yield was 44 mg (66%). This product is identical to that from Method A, as supported by IR and microanalysis data. X-ray Crystallographic Analyses of / a o c « - ( P , P ) - [ R e ( N P h ) C l ( P P h 3 ) -(P0 2)]-2CHC1 3 (1) and m-(P,P)-[ReO(PO)(P0 2 ) ] (2). All the crystal structures reported in this thesis were determined by Dr. Steven J. Rettig of the UBC Structural Chemistry Laboratory. Selected bond distances and bond angles for fac-cis-(P,V)-76 [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and ds-(P,P)-[ReO(PO)(P02)] (2) appear in Table 4-1 and Table 4-2, respectively. For the tables of selected crystallographic data and coordinates, please refer to Appendix HI. 4.3 Results and Discussion HPO, H2PO2 Ligands. A few procedures for the synthesis of (0 -hydroxyphenyl)diphenylphosphine (HPO) are known.17'18 The preparation of H2PO2, starting from anisole, has been reported by von Zon as an intermediate in the synthesis of macrocyclic monophospha-crown ether ligands from phenyl methyl ether. Based on our experience in the synthesis of HPO, we realized that the methyl protection is difficult to undo so we adapted a 1 Q method used for HPO to our preparation of the H2PO2 ligand, in which the methoxymethyl, or mom, group was used to protect the phenol OH. It is known that the mom group is easy to cleave with mineral acids. The mom-protected phenol was ortho-lithiated and then reacted with dichlorophenylphosphine to give the mom-protected intermediate, (mom)2P02. Upon treatment with anhydrous HC1 gas, the expected bis(o-hydroxyphenyl)phenylphosphine was obtained as the hydrochloride salt (Scheme 4-1). OH 1. CH3OCH2OCH3 2. n-BuLi/TMEDA Li O O PhPCl2 HCl(g) MeOHorEtOH(anhy.) (mom)2P02 HC1 H2P02HC1 Scheme 4-1 77 For the intermediate, (mom)2P02, the mass spectrometric parent ion peak (m/z= 382, [(mom)2PC>2]+) is present, consistent with the elemental analysis. In the 31P{1H} NMR spectrum in CDCI3, a singlet at -26 ppm is present and upfield from triphenyl phosphine (at -5 ppm). The *H NMR spectrum shows that aromatic hydrogen atoms are present in the range 6.6 to 7.4 ppm, an AB quartet centered at 5.02 ppm and a singlet at 3.1 ppm. The 13 aromatic hydrogen atoms were easily sub-grouped into four, with the integral ratio of 7:2:2:2. The last three subgroups were assigned to the hydrogen atoms on the two phenolic arms by the iH-1!! COSY spectrum. Peaks for the other hydrogen atoms on each phenolic arm are overlapped by the phenyl hydrogen atoms. The AB quartet centered at 5.0 ppm (integrating for four hydrogen atoms) were assigned to the methylene hydrogen atoms. The nonequivalence of the two methylene hydrogen atoms in each mom group is due to prochirality.* The formulation of H2PO2 HC1 is supported by the elemental analysis and by the 90 chemical shift characteristic for phosphonium P at +34.6 ppm in DMSO-fl?6> even though the EIMS shows only H2PO2. In the synthesis of HPO, an analytically pure sample was obtained by 1 Q sublimation. (Given that the author did not neutralize his product with a base, a routine practice 17 91 for free phosphines, and that phosphine hydrochloride salts dissociate at high temperature, it is quite possible that the crude HPO product cited could have been HPOHC1.) In this work, no attempt was made to convert the salt H2PO2HCI to the free phosphine, since the salt dissociated in the ligand-exchange reactions with it (vide infra) and in basic solution, as is indicated by the chemical shift of -30.6 ppm in py-i/5. Synthesis of the Complexes. The reaction of HPO with oxo or nitrido metal(V) starting materials in a 2:1 ratio gave the bis(PO) complexes, [MOCl(PO)2] (M = Tc or Re), or [ReN(PO)2(PPh3)]. With the bidentate monoprotic HPO, the formation of bis-ligand chloro In compounds such as OPPh(OPF2)2 (see below), even though the molecule is not chiral, (the two -OPF2 are equivalent), there is no symmetry relationship between the two F nuclei within one such group. Thus consequences are exactly the same as for chiral molecules. The two -CH2- H nuclei in each mom group of (mom)2PC>2 could be considered similarly. O II F 2 PO-^ P \ n / P V'"F 78 oxometal complexes, upon ligand exchange, is not surprising since the formulation [MOXL2] is common for the substitution of Re03 + and Tc03 + cores with other bidentate monoprotic ligands. A Tc(III) [Tc(PO)3] complex has been recently reported9 suggesting that the combination of hard (O) and soft (P) donors in the chelate will be suitable for stabilizing various intermediate oxidation states of Tc or Re. The formation of [ReN(PO)2(PPh3)] from [ReNCl2(PPh3)2] is, however, mildly surprising. It is a six-coordinate complex whereas a five-coordinate complex might have been expected because of the strongly trans labilizing nitrido group22a>b and the sterically demanding ligand phenyl groups; however the latter effect may actually be small. To our knowledge, this is the first example of three triarylphosphine phosphorus donors coordinating the same Re(V) center, suggesting that the bonding of the phenolato oxygen atom of each PO" ligand to the Re center may actually offset the steric repulsion among the bulky phenyl groups around the metal center. The other evidence for the reduced steric repulsion is that the anchored phosphines are found to bind the metal more tightly than the unanchored phosphine PPI13 {vide infra). The phosphine hydrochloride salt, H2PO2HCI, reacted directly with [ReZCl3(PPh3)2], resulting in the mono(P02) complexes [ReZCl(PPh3)(P02)] (Z = O, NPh), suggesting the direct dissociation of the salt to a potentially tridentate diprotic double-functionalized phosphine ligand. The mono(P02) complex [ReOCl(PPh3)(P02)] formed even though a 2:1 L:M ratio was used in the ligand exchange reaction with [ReOCl3(PPh3)2]. In contrast, with [(n-Bu)4N][TcOCU] as the starting material under the same conditions, the ligand exchange reaction led to the formation of the mixed(P02/HP02) complex [TcO(P02)(HP02)] (vide infra). These observations, taken together, suggest that the chelation of a PO22" is an easy first step and the bonding of the second PO22" is hindered by the presence of PPI13. With the phenylimido starting material, the isolated product was of either cis- or trans-(P,P) geometry depending on the reaction solvent (Scheme 4-2). The starting material, [Re(NPh)Cl3(PPh3)2], has the trans-(P,P) configuration.23 In polar methanol, the franj-configuration was retained after one PO2 ligand replaced one PPI13 and two adjacent CI ligands in the starting material; however, in the less polar benzene:ethanol (10:3) mixture, the cis-(P,P) product was formed. No conversion was observed from trans-(T>,V)-79 CI — Ph3P' NPh II ^ P P h 3 :Re CI CI H2P02HC1 MeOH H2P02HC1 Bz/EtOH T = 338 K O-NPh II ^ PPh3 :Re CI T = 355 K 1,2-dichloropropane/Bz(3:1) T=349K .0 NPh II ^ C l O - Re - PPh3 .0 Scheme 4-2 [Re(NPh)Cl(PPh3)(P02)] to cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)], after the former was heated at 80-85 °C in a l,2-dichloropropane:benzene-c?6 (3:1) mixture for 40 min. Clearly, the product geometry is not determined by temperature only; with oxo starting materials, the cis product was obtained in ethanol while attempts to prepare the trans product in methanol led to a mixture. The mixed(PO/P02) complexes [ReZ(PO)(P02)], which incorporate both PO22- and P O ligands, were synthesized via ligand exchange reactions of [ReZCl(PPh3)(P02)] (Z = O, NPh) with HPO in the presence of NaOAc. Direct ligand exchange of [Re(NPh)Cl3(PPh3)2] with H2P02HC1 and HPO (M:L:L' =1:1:1) yielded mostly [Re(NPh)Cl(PPh3)(P02)] with traces of fra>w-(P,P)-[Re(NPh)(PO)(P02)], suggesting that the binding of the PO22- is the first step, and that the substitution of the PO" ligand for the remaining CI and PPI13 ligands is unfavorable, as one HPO is competing against the three CI" and one PPI13 in the solution. An attempt with [(n-Bu)4N][ReOBr4] and H 2 P 0 2 HC1 and HPO (M:L:L' = 1:1:1) produced a mixture of [ReO(PO)(P02)], [ReO(P02)(HP02)], and [ReOBr(PO)2], as indicated by mass spectrometry (LSIMS) and 31P{ lU] NMR spectroscopy. Recrystallization of this mixture in EtOH yielded good crystals of [ReO(PO)(P02)], one of which was suitable for X-ray analysis. 80 An alternative route for the incorporation of two PO2 ligands on to one metal center, a reduction/coordination reaction with ammonium perrhenate or pertechnetate and an excess of H2PO2HCI, was explored. The hydrochloride salt dissociates to the corresponding phosphine which functions as reducing agent as well as ligand. Most reductions of MO4" (M = Re, Tc) with phosphines have been done in acidic environments. ' ' The HC1 salt, however, reacted with NH4MO4, at a molar ratio of L:M = -3.5:1 without any extra acid required, leading to the mixed(P02/HP02) complexes, [MO(P02)(HPC>2)] (M = Tc, Re) in high yields. Similar conditions applied to o-phosphinoamine, o-phosphinothiophenol and o-phosphinophenol (HPO) ligands formed [Tc(PX)3]-type complexes (X = N, P, or O) from NH4TCO4.9'10 These ligands can both coordinate and reduce the metal center, a distinct advantage when formulating potential 99nrr/c-radiopharmaceuticals, where the only convenient source is 99mTc04" from a "Mo/" m Tc generator. In an attempt to synthesize the binary complex [Tc(P02)2] through the reaction of [TCO4]" and H2PO2HCI, despite an excess of ligand (L:M = 6:1); the only product was [TcO(P02)(HP02)], which was also synthesized via a ligand exchange reaction with [(n-Bu)4N][TcOCl4]. Reaction of [ReCl4(PPh3)2] with two equivalents of H2PO2HCI produced only [ReOCl(PPh3)(P02)] presumably by a process which included aerial oxidation of the metal center. Characterization of the Complexes. All the complexes are air stable in the solid state and in solution. They were characterized by elemental analysis, infrared spectroscopy, mass spectrometry, 1H/31P{ lH} NMR spectroscopy, and by X-ray crystallography in the cases offac-cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)] and/ac-cw-(P,P)-[ReO(PO)(P02)]. Mass spectrometric data confirmed the formation of the complexes when the expected parent ions and/or their fragments were found, while microanalysis established the formulation of all the compounds. IR measurements confirmed the existence of metal-ligand multiple bonds and showed that the multidentate ligands were coordinated as evidenced by the absorptions shifted from those of the free ligands. Due to the overlapping of multiplets in the aromatic hydrogen region (5.0-8.5 ppm), full assignments of the ^K NMR spectra were not possible; nevertheless, the lH NMR spectra were still useful in verifying similarities between the analogous Re and Tc complexes. On 81 the other hand, 31P{1H} NMR played a crucial role in both identifying and determining the geometries of the Re complexes (vide infra). Bis(PO) Complexes. In the mass spectra, the loss of one chloro ligand from a [MOCl(PO)2] unit or the loss of the triphenylphosphine ligand from the [ReN(PO)2(PPh3)] unit produced [MO(PO)2]+ or [ReN(PO)2]+ base peaks, respectively, indicating that the bidentate PO_ ligands are more strongly bound than the monodentate CI or PPI13. The parent ion peaks are all present, but weaker in intensity. The IR spectrum of [TcOCl(PO)2] is almost superimposible on that of its ReO congener except for the M=0 stretching vibrations (VTC=0 = 940 cm-1, VRe=o = 975 cm-1), but quite different from that of the nitrido complex (VR^N = 1045 cm-1) reflecting the structures proposed (vide infra). All these absorption bands are in the normal range.22c Scheme 4-3. Proposed structures for [MOCl(PO)2] (left) and [ReN(PO)2(PPh3)] (right). For the oxorhenium complex [ReOCl(PO)2], an "equatorial" structure where two PO~ ligands are bound in the equatorial plane (relative to the axial Re=0) was ruled out by the 31P{1H} NMR, which shows two doublets with a coupling constant of 10 Hz, consistent with two nonequivalent phosphorus donors being cis to each other. A cis-(P,P) "twisted" structure is thus proposed for this complex - each of the two PO" moieties acts as a bidentate (P, O) monoprotic ligand, with the phenolate oxygen donor of the axial PO- ligand trans to the Re=0 group (Scheme 4-3). The exclusion of a P donor trans to the oxo ligand is based on other reported bis(ligand)halooxometal(V) complexes of bidentate monoprotic ligands, where the neutral donors are always cis to the oxo group.10* Upon coordination of HPO, an increase in Refer to Chapter 2. 82 chemical shift was observed from that of -28.7 ppm for the free ligand, as is generally seen for most phosphines (except PX3 and P(OR)3), due to the resultant deshielding effect.25 The upfield resonance at 2.2 ppm in the 31P{ 1H} NMR was assigned to the P donor of the axial ligand (vide infra), as it is more shielded since the neutral donor of the axial ligand usually donates less strongly than that of the equatorial one, consistent with the previously reported [ReOBr(ma)2] species (ma = maltolate anion) and as reflected by the bond distances (Chapter 2). For the cis-(P,P) "twisted" [ReOCl(PN)2], the Re-P bonds of the equatorial ligand were found to be significantly shorter than those of the axial ligand in each of two solved structures. For a series of five- or six-coordinate Ru(II) complexes, a linear correlation of P-Ru distances with 3 1P chemical shifts has been noted with the strongly bound phosphines giving downfield resonances. This cis-(P,P) "twisted" structure for [MOCl(PO)2] (M = Tc, Re) is consistent with the !H NMR spectra (Figure 4-1). The low symmetry produces rather complicated spectra since there are six different types of phenyl rings in each complex; however, the eight hydrogen atoms of the two phenolic arms are easily correlated in the lH-*H COSY spectrum (not shown), the upfield set of four hydrogen atoms being those of the axial ring as in [ReOBr(ma)2] (Chapter 2), and analogous complexes of the [ReOCl(PN)2] formulation. For the axial PO-phenyl ring set, the most upfield resonance at 5 6.01, a doublet of doublets resulting from coupling to the neighboring phosphorus and hydrogen nuclei, was assigned to the hydrogen atom on the other oriho position to P. This doublet of doublets becomes a simple doublet upon decoupling the upfield P at 2.2 ppm, indicating that the P resonance at 2.2 ppm belongs to the axial PO phosphorus atom. The *H NMR spectrum of the Tc analog (Figure 4-1) is identical to that of [ReOCl(PO)2], indicating the same arrangement. Consistent with this, the Tc complex shows two broad "waves" in the range of 10 to 50 ppm in its 31P{ lU} NMR spectrum; these "waves" became broad peaks upon cooling to -35 °C, at about 27 and 40 ppm respectively, suggesting that there are two non-equivalent P atoms in the complex. This resonance broadening is not uncommon in Tc-phosphine complexes10 and is probably caused by a scalar coupling of the 31P nuclei with the quadrupolar " T c nucleus.27'2 83 M = Re A. M = Tc A t > i -f—y . r 7.5 — j — i — i — i — i — r — P 7.0 6.5 6.0 ppm Figure 4-1. *H NMR (400 MHz) of [MOCl(PO)2] (M = Re, Tc) in CDC13. (* = impurities) In the 31P{ !H} NMR spectrum of the nitrido complex [ReN(PO)2(PPh3)] in CDCI3, an ABX system was observed (Figures 4-2 and 4-8A), indicating that there are three non-equivalent P atoms in a meridional arrangement. The downfield doublet of doublets (34.7 ppm) with coupling constants of 222 (27AB) and 12 Hz (27AX)» corresponds to one of the two trans P atoms, while the upfield doublet of doublets (23.0 ppm) with coupling constants of 222 (27BA) and 6 Hz (2/BX)> corresponds to the other trans P atom. The unique P atom (resonance at 13.9 84 22°C(after) * ^ « — +mm* 52 °C JL J i i n jUL^-JU__JL * • * * ^ ^ ^ ^ • i l — i n • I • M i " »«M • " ' i •' .tfUM i'" /V— I I I I I I I I | I I I I [ I I I I | I I I ' | ' ' ' ' J ' ' ' ' | ' ' 35.0 25.0 15.0 5.0 -5.0 ppm Figure 4-2. Variable temperature 31P{ !H} NMR spectra (121 MHz) of [ReN(PO)2(PPh3)] in CDC13 (heating to 52 °C and cooling back to 22 °C). * = unidentified species. ppm) is cis to both of the above P atoms, giving a multiplet resulting from coupling to the two cis P nuclei. This unique phosphorus (Px) was assigned to the bound PPI13 ligand on the basis of the replacement of the PPI13 ligand with pyridine (vide infra); thus the two trans P nuclei (PA and PB) belong to the PO" ligands. As in [ReOCl(PO)2], the downfield resonance at 34.7 ppm (PA) was assigned to the P donor of the equatorial hgand and that at 23.0 ppm (PB) to the axial P. The crowded environment around the Re center, which includes three triarylphosphines, led to the detectable dissociation of the PPI13 ligand in CDCI3, as revealed by the presence of a small peak at about -5 ppm in the 31P{1H} NMR. This peak is free PPI13; its intensity increased 85 significantly upon addition of PPI13 to the NMR sample in a separate test. The PPI13 dissociation is enhanced at a higher temperature, as shown in Figure 4-2. M o n o ( P 0 2 ) Re Complexes. The parent ions [ReOCl(PPh3)(PC>2) + 1]+ or [Re(NPh)Cl(PPh3)(P02)]+ and three fragment cations (loss of PPI13 and/or CI) are present in about the same intensities in the positive ion detection LSIMS. In all cases, only Re-containing fragments incorporated PO22", showing that the [ReZ(PC>2)] (Z = O, NPh) moiety is retained as a highly stable unit. The IR spectra of all complexes show one Vc-H stretch at about 3050 cm"1, while the cw-(P,P)-phenylimido complex shows an additional band at 2980 cm-1. The Re=Z bond stretching vibrations were found at 970 cnr1 (Z = O), or 1030 cm-1 (Z = NPh), normal for the respective moieties.22c>d No dissociation to give PPI13 in solution was observed for the three mono(P02) complexes at room temperature, presumably due to their being less steric repulsion than in [ReN(PO)2(PPh3)]. The oxo complex [ReOCl(PPh3)(PC>2)] shows an AX system in the 31P{1H} NMR spectrum, with the small coupling constant of 7.2 Hz indicating a cis-(P,P) geometry. The peak at 19.5 ppm was assigned to the PO22" ligand and the one at -2.1 ppm to the PPI13 ligand, as suggested by its displacement by pyridine (vide infra). This assignment is consistent with the crystal structure of the ds-(P,P)-phenylimido analog, where the PO22" ligand Re-P distance is shorter than the PPI13 Re-P distance (vide infra), indicating the P nucleus of the former ligand donates more electron density, thus is less shielded, than the latter. For the two isomers of the phenylimido complex [Re(NPh)Cl(PPh3)(P02)], the cw-(P,P) isomer is distinctly different from the trans-(P,P) isomer in its 31P{ !H} NMR spectrum. For the trans isomer, a typical AB quartet pattern demonstrates a trans-(P,P) structure where the coupling constant is large (calculated 8A and 8B being 34.5 and 2.9 ppm respectively, with observed 27pp = 241 Hz). The 31P{ lH} NMR spectrum of the cis-(P,P) isomer is consistent with two nonequivalent cis P nuclei (observed values being 28.3 and 15.4 ppm respectively). Again, the downfield peak (28.3 ppm for the cis or 34.5 ppm for the trans isomer) was assigned to the PO22" phosphorus atom while the upfield peak (15.4 ppm for the cis or 2.9 ppm for the trans isomer) was assigned to the bound PPh3 phosphorus atom as was seen in [ReOCl(PPh3)(P02)]. 86 Mixed(PO/PC>2) Complexes. In the LSIMS, the peaks for the parent ions [ReZ(P02)(PO) + 1]+ are stronger than those for the ions [ReZ(P02)]+, indicating that the parent ions are stabilized by replacement of monodentate ligands with the PO" chelate. In the IR spectra of both the complexes, there is one Vc-H at -3050 cm-1 and Re=Z stretching vibrations are at 965 cm-1 (Z = O) or 1030 cm-1 (Z = NPh), in the normal range for either type of linkage.22c'd A cis-(P,P) structure for the oxo complex is revealed by a small coupling constant of about 4 Hz in an AX system in the 31P{ *H} NMR spectrum. For the phenylimido complex, an AB quartet with a large 2/pp = 230 Hz coupling constant demonstrated a trans-(P,V) structure. These results are not surprising considering the geometry of the mono(P02) precursors for each mixed(PO/P02) complex. As discussed before, the downfield resonances were assigned to the PO22", while the other one to the PO" phosphorus atom in each case. These assignments are also consistent with the crystal structure of/ac-cw-(P,P)-[ReO(PO)(P02)], in which the Re-P distance for PO22" is significantly shorter than that for PO" (vide infra). Compared to that of the corresponding mono(P02) precursor, the chemical shift for the PO22" phosphorus atom in each mixed(PO/P02) complex does not change much, (19.5 and 34.5 ppm in the each precursor, 20.6 and 35.7 ppm in the mixed oxo and phenylimido complexes, respectively) while the PO" phosphorus atom is further deshielded (16.0 and 15.4 ppm in the mixed oxo and phenylimido complexes, respectively) with respect to the bound PPI13 phosphorus in the precursor (-2.1 and 2.9 ppm in the oxo and phenylimido complexes, respectively). This can be rationalized by stronger Re-P binding for PO" than for PPI13 due to increased anchoring, as illustrated by the structural data (Table 4-1). Mixed(P02/HP02) Complexes. The formulations [MO(P02)(HP02)] (M = Re, Tc) are based on elemental analyses as well as on spectral data. In the +LSIMS for [ReO(P02)(HP02)], the sample in DMSO gives peaks corresponding to molecular ion plus proton (M + 1), molecular ion losing one ligand (M - HPO2), and molecular ion losing one oxygen and one hydrogen atom (M - O - H, or [Re(P02)2]); the last is thought to be formed under the ionization conditions rather than coming from the original sample. When the samples were ground with KBr, prominent peaks were observed for [MO(P02)(HP02) + K]+ and 87 [MO(P02)(HP02) + 1]+ (M = Re, Tc) in the +LSIMS; the -LSIMS of [ReO(P02)(HP02)] revealed [ReO(P02)2]-. The IR spectra shows medium-intense Vc-H stretches at about 3060 cm-1 for each compound. The spectrum of [TcO(P02)(HP02)] is almost superimposable on that of [ReO(P02)(HP02)] except for the M=0 stretching vibrations; the Tc analog shows a strong VTC=0 stretch at 965 cm-1, while the Re=0 vibration is weaker and a doublet at 970 and 975 cnr1 with a sharp absorption at 990 cm-1, possibly a solid state effect. Two sets of AX patterns of relative intensities 3:1, were observed in the 31P{1H} NMR spectrum of [ReO(P02)(HP02)J (Figure 4-3), suggesting two isomers in the product. cis-(P,P) structures are proposed for each isomer, as the coupling constants involved are small. 2/pp for the major set of doublets (20.9, 11.6 ppm) is 4.1 Hz, while for the minor pair, one peak is a doublet of 5.0 Hz (22.8 ppm) and the other is broad (17.5 ppm). The isomerism is consistent with the two possible diastereomers, where the free phenol group is oriented up or down (Figure 4-3). As has already been discussed, the phosphorus on the fully bound (tridentate) P022~ is more deshielded than the PO" or PPI13 phosphorus, thus the upfield peak was assigned to the bidentate HP02" ligand, while the downfield peaks were assigned to the tridentate P022" ligands in each pair. The major set was assigned to the diastereomer with the up-orientation of the free phenol group, while the minor pair was assigned to the diastereomer with the down-orientation. Thus, the broadness of the HPO" phosphorus peak in the minor pair is explained by intermolecular proton exchange between the free phenolic hydroxyl group and the solvent. For the major diastereomer, hydrogen-bonding to the oxo group would hinder such a proton exchange, ensuring a normal narrow 31P resonance for the P atom in the bidentate HPO- ligand. There is no evidence for the interconversion of these two diastereomers as shown in high temperature 31P{ *H} NMR (Figure 4-3). In the 31P{ lH} NMR spectrum, the Tc analog shows two broad shouldered peaks at 34.5 and 45.0 ppm, indicating the presence of two diastereomers as in the Re complex. As monitored by the 31P{ JH} NMR spectrum, the Tc complex started to decompose at about 80 °C in DMSO-cfe, giving a yellow species and phosphine oxide. 88 (minor set) 150 °C H... O II °N^ ^ i c ? 3 0 N* *+tm9i*+ K******* *+*~*f ^ ^ w (major set) VM^^«*a^^rt»*«N«l'«*M*^«^MM*W^*>M*«*MMMMA<"' ^ - " * I > , . 20 °C • ^ • • • ^ • • ( • ^ ^ ^ ^ ^ • " ^ I I I I I I I I I 16.0 22.0 20.0 18.0 14.0 12.0 ppm Figure 4-3. 31P{ !H} NMR (121 MHz) of [ReO(P02)(HP02)] in DMSO-^6 at room temperature and high temperature, and the proposed structures for the two diastereomers. In the lH NMR spectrum of [ReO(PC>2)(HP02)], there are two upfield aromatic hydrogen signals, one major and one minor at about a 3:1 ratio, consistent with two isomers (Figure 4-4). These two signals, as doublets of doublets, were assigned to the hydrogen adjacent to the P atom in the axial phenolato ring of each isomer, as in the assignment for [ReOCl(PO)2] (vide supra). In consistence with this, the H signal (5.23 for the major or 5.45 ppm for the minor) was found 89 to be coupled to the downfield P signal (20.9 ppm for the major or 22.8 for the minor, which correspond to the PO22" P nuclei), as shown in the selectively decoupled ^ { ^ P } spectra 90 (Figure 4-5). Figure 4-4 also shows the presence of a hydrogen bonded phenolic hydrogen atom at about 10 ppm for each complex. The Tc analog shows a similar *H NMR spectrum, indicating similar structures, at least in solution (Figure 4-4). -A. M = Re JL^vA, M = Tc •MirtMtNinr^fMM JT^J\M- u,i mid i i i iwi? I I I I I I I I1! 5.0 ppm I 1 i 1 • 1 > 1 ' 1 f i 1 i i 1 1 1 1 i • 1 1 1 1 i ' 10.0 9.0 8.0 7.0 6.0 Figure 4-4. *H NMR (200 MHz for M = Re, 400 MHz for M = Tc) of [MO(P02)(HP02)] in DMSO-d6. 90 IT ° H 0 -(minor set) •P SA^Sk Re: I o (major set) ppm B _A 5.5 5.3 5.1 JL 5.5 5.3 ~¥ * 5.1 ppm Figure 4-5. Selectively 31P-decoupled 2H{31P} NMR spectra (500 MHz, partial) of cw-(P,P)-[ReO(P02)(HP02)] in DMS0-4-A. ! H NMR signals for the equatorial o -H nuclei of the two isomers; B. !H{31P} NMR signals with irradiation at 20.9 ppm; C. !H{31P} NMR signals with broadband irradiation. 91 X-ray Structures of /ac-c/s-(P,P)-[Re(NPh)Cl(PPh 3 ) (P0 2 ) ] -2CHCl 3 (1) and/ac-c«-(P,P)-[ReO(PO)(P02)] (2). Crystals of [ReOCl(PO)2], [TcOCl(PO)2], [ReN(PO)2(PPh3)], and [ReOCl(PPh3)(P02)] were easily grown to suitable size, while [ReO(P02)(HP02)] tended to form small crystals and [TcO(P02)(HP02)] always precipitated as a powder. Single crystals of/ac-cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) were obtained from chloroform/cyclohexane solvent mixtures, while those of/aocw-(P,P)-[ReO(PO)(P02)] (2) were obtained from ethanol, both by slow evaporation and the structures of both complexes were solved. The ORTEP diagrams of the molecules 1 and 2 are shown in Figures 4-6 and 4-7, respectively, while selected bond distances and selected angles are listed in Tables 4-1 and 4-2, respectively. The overall geometry around the rhenium atom in either 1 or 2 is best described as a highly distorted facial octahedron ("pinched") with cis phosphine ligands. In both molecules, the P022" ligand is bound in a facial manner, the meridional mode being sterically unavailable. Each of the two phenolato moieties in the P022" ligand functions as an anchor for the phosphine, with one PO phenyl ring sitting in an equatorial plane and the other axial to the Re=NPh (1) or Re=0 (2) unit. The PPI13 (1) or PO" (2) ligand coordinates such that its P donor is cis to the P02 2" P atom. The "pinched" distortion from octahedral geometry is mainly caused by the acute bite angles between the phenolate O atom and the P atom of the P022" or PO- ligands (P(l)-Re(l)-0(1) = 80.96(8)° and P(l)-Re(l)-0(2) = 79.98(8)° for (1) (P(l)-Re(l)-0(2) = 80.54(8)°, P(l)-Re(l)-0(2) = 79.98(8)° and P(2)-Re(l)-0(4) = 81.99(8)° for (2)). The repulsions between the two bulky triarylphosphine P donors, and between the Re=Z (Z=NPh 1, Z=0 2) linkage and the negatively charged cw-donors relative to Re=Z account for this distortion as well. The P(l)-Re(l)-P(2) angles are 105.64(4)° and 107.18(4)°, for 1 and 2 respectively. The Cl(l)-Re(l)-N(l) and 0(1)-Re-N(l) angles are 102.5(1)° and 102.2(1)° (1), while 0(l)-Re(l)-0(2) and 0(l)-Re-0(4) angles are 96.9(1)° and 108.3(1)° (2), respectively. In contrast, P(l)-Re(l)-N(l) and P(2)-Re(l)-N(l) are 93.0(1)° and 93.1(1)° (1), and P(l)-Re(l)-0(1) and P(2)-Re(l)-0(1) are 90.4(1)° and 94.59(10)° (2), respectively, indicating less repulsion between the Re=Z group and the neutral cw-donors. The Re atom is 0.296 (1) or 0.273 A (2) out of the equatorial plane 92 C4 C5 C40 (1) Figure 4-6. ORTEP drawing of/ac-cw-(P,P)-[Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) (with the solvent molecules omitted); 33% probability thermal ellipsoids are shown. 93 C4 C3 C29 C23 C22 (2) Figure 4-7. ORTEP drawing of cw-(P,P)-[ReO(PO)(P02)] (2); 33% probability thermal ellipsoids are shown. 94 Table 4-1. Selected Bond Lengths (A) for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and ReO(PO)(P02) (2). [Re(NPh)Cl(PPh3)(P02)]-2CHCl3(l) [ReO(PO)(P02)] (2) Re(l)-N(l) Re(l)-Cl(l) Re(l)-P(l) Re(l)-P(2) Red)-O(l) Re(l)-0(2) Re(l)-0(3) Re(l)-0(4) P(l)-C(l) P(l)-C(7) P(l)-C(13) P(2)-C(19) P(2)-C(25) P(2)-C(31) 0(1)-C(2) 0(2)-C(2) 0(2)-C(8) 0(3)-C(8) O(4)-C(20) C(l)-C(2) C(7)-C(8) C(19)-C(20) 1.728(3) 2.409(1) 2.387(1) 2.454(1) 2.050(3) 2.050(3) 1.800(4) 1.780(4) 1.811(4) 1.821(4) 1.834(4) 1.824(4) 1.342(5) 1.340(5) 1.398(6) 1.406(6) 2.391(1) 2.428(1) 1.692(3) 2.050(3) 2.026(3) 1.994(3) 1.795(4) 1.785(4) 1.802(4) 1.810(4) 1.805(4) 1.825(4) 1.339(5) 1.339(5) 1.353(5) 1.414(6) 1.413(6) 1.396(5) 95 Table 4-2. Selected Bond Angles (deg) for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2). Re(l)-N(l)-C(37) Cl(l)-Re(l)-N(l) P(l)-Re(l)-N(l) P(2)-Re(l)-N(l) 0(1)-Re(l)-N(l) 0(2)-Re(l)-N(l) P(l)-Re(l)-0(1) P(2)-Re(l)-0(1) 0(l)-Re(l)-0(2) 0(l)-Re(l)-0(3) 0(l)-Re(l)-0(4) 0(2)-Re(l)-0(3) P(l)-Re(l)-P(2) P(l)-Re(l)-0(2) P(l)-Re(l)-0(3) Cl(l)-Re(l)-P(2) P(2)-Re(l)-0(4) [Re(NPh)Cl(PPh3)(P02)]-2CHCl3(l) 171.7(3) 102.5(1) 93.0(1) 93.1(1) 102.2(1) 165.4(1) 80.96(8) 86.6(1) 105.64(4) 76.65(8) 88.25(4) [ReO(PO)(P02 90.4(1) 94.6(1) 96.9(1) 165.5(1) 108.3(1) 89.1(1) 107.18(4) 80.54(8) 77.59(8) 81.99(8) formed by the four cw-donors. It has been suggested that the repulsion exerted by the oxo group on the ligands cis to it increases in the order of increasing hardness of the ligands.30 For 1, the Re(l)-N(l) bond length of 1.728(3) A, which is the same as that in the starting material [Re(NPh)Cl3(PPh3) 2 ] 2 3 and not significantly different from that in [Re(ma)2(NPh)(PPh3)]+ (ma = maltolate anion) (1.709(8) A),31 indicates the retention of a multiple bond. The Re(l)-N(l)-C(37) angle is 171.7(3)°, equal to the analogous angles of 172.6(6)° in [Re(NPh)Cl3(PPh3)2],23 and 171.8(4)° in [Tc(NPh)Cl3(PPh3)2],32 within 96 experimental error. The small deviations from 180° in these cases suggest that the imido nitrogen is essentially sp-hybridized, that the M-N linkage is a real triple bond.22e In 2, with the length of 1.692(3) A, the Re(l)-0(1) bond is longer than that in [ReOCl3(PPh3)2] (1.663(5) A),33 but still in the range of values found in other oxo complexes of rhenium(V),22f indicating the retention of a multiple bond. The trans influence of these multiple bonds is insignificant in 1 and 2, however; the trans Re(l)-0(2) bond (2.050(3) A) is identical to the cis Re(l)-0(1) (1), while the trans Re(l)-0(3) bond (2.026(3) A) is actually shorter than one cis bond Re(l)-0(2) (2.050(3) A), but longer than the other Re(l)-0(4) bond (1.994(3) A) (2). Similar observations were made for [Re(ma)2(NPh)(PPh3)]+, where close values of 1.996(7) and 1.987(7) A were found for Re-31 O bonds trans and cis to the Re=NPh group, respectively. Furthermore, in [Re(NPh)Cl3(PPh3)2], the trans Re-Cl bond (2.402(2) A) is shorter than any of the cis Re-Cl bonds (2.415(2) A and 2.411(2) A),23 and in [(CH3)4N][ReO(02C6H4)2(PPh3)], the trans Re-O bond (2.041(7) A) is shorter than the cis Re-O bond (2.062(9) A), indicating a non-existent trans influence. In the Tc analog, [Tc(NPh)Cl3(PPh3)2], the trans influence is not observed either. All these factors suggest that the trans influence of multiple metal-ligand bonds is insignificant in these complexes. The cis ancillary ligands may play a role as well. The Re(l)-Cl(l) distance (2.409(1) A) in 1 equals the average Re-Cl value in [Re(NPh)Cl3(PPh3)2].23 The double phenolato anchored P donor in tridentate PO22" (1 and 2) is bound to the Re center significantly more closely than the single phenolato anchored P atom in bidentate PO" (2), which is in turn more closely bound than the non-anchored phosphorus atom in monodentate PPh3 (1). The Re-P022- distance, Re(l)-P(l) = 2.387(1) A in 1, is the shortest Re(V)-P bond to the author's knowledge (contrast it with 2.454(1) A for Re(l)-P(2) in 1). On the other hand, Re(l)-P(l) and Re(l)-P(2) are found to be 2.391(1) and 2.428(1) A, respectively, in 2. The average Re-P distance is 2.496 A in [Re(NPh)Cl3(PPh3)2]23 and is 2.519(1) A in [ReOCl3(PPh3)2].33 Singly aminophenylphosphine-anchored Re-P bonds are in the range 2.42 to 2.50 A. The strengthening of the Re-P bond with increased anchoring can be explained by the synergistic electronic effect (c + it) as well as reduced steric effects35 due to the reorientation and bonding of the anchoring arms (vide infra). 97 Whereas the phenyl groups in the bound PPI13 (1) are found in a syn conformation, as in free PPI13,36 the anchoring o-oxyphenyl groups are reoriented by the rotation of the P-C bond to fit the required octahedral geometry in both 1 and 2. This reorientation makes it possible for the o-oxyphenyl groups to bind the metal, forming chelate rings. Upon anchoring, there is no repulsive contribution (i.e. large cone angle) from the o-oxyphenyl groups since these groups are not bulky phenyl rings which keep P away from the metal; in fact they are holding the metal. In addition, the orientation of the o-oxyphenyl rings is coplanar or perpendicular to the equatorial plane, possibly locating the P atom in a position that favors transferring K donation from the metal center to the o-oxyphenyl rings. Thus, the steric repulsion decreases, and the K back-bonding (and a-donating) increases, in the order of PPh3, PO\ PO22", resulting in Re-P distances that decrease in the same order. While the C-0 and C-C bonds in the equatorial PO22" chelate rings are the same as those in the axial rings, within experimental error, the extreme P-C bond lengths are significantly different (o(A) = 0.006). The axial ring P-C bonds are shorter (P(l)-C(7) = 1.780(4) A (1), 1.785(4) A (2)) than their equatorial analogs (P(l)-C(l) = 1.800(4) A (1), 1.795(4) A (2)) (Table 4-1). The average corresponding distances are 1.826 A in the bound PPI13 moiety (1) and 1.831 A in free PPh3. This may indicate that the K-TZ interaction between P and C decreases in the order: axial anchoring rings, equatorial anchoring rings, bound PPI13, free PPI13. All the anchoring phenyl rings of the POxx" ligands in 1 and 2 were found to be quite planar, as are the other phenyl rings attached to either phosphine, indicating that upon coordination the aromaticity of the phenyl rings is retained. Reactivity to water/pyridine. To the complexes [ReO(P02)(HP02)] or [TcO(P02)(HP02)] in DMSO-cfe, was added 5 drops of D2O. A fluffy greenish precipitate or a brownish cloudy material, respectively, formed between the two solution layers and disappeared upon shaking the tube. No detectable change was observed in the 31P{ ^ H} NMR measurement in each test; however, in the lH NMR spectra the phenol hydrogen resonance at 10 ppm disappeared, consistent with an inter-molecular H-D exchange. The addition of water to [ReO(P02)(HP02)] led to no visible reaction after weeks. The mixed(P02/HP02) complexes are 98 tm*Lptt'"' •*n»i*»Mi»uhM»y Vpii%iwm< $> >|ii»i*a'i|ii'i«w* ReN(P0)2(PPh3) + py. B ReOa(PPh3)(P02) + py i^ W^wU'W^ ReN(POh(PPh3) ^V^V-^V^^^W^U*"^ ^ f . i ^ i ^ i i i i i i i i i 35.0 25.0 15.0 5.0 -5.0 ppm ReOa(PPh3)(P02) IWiiMW h^i.^*^*/ I ' l l ! 25.0 15.0 5.0 -5.0 ppm Figure 4-8. 31P{ !H} NMR (81 MHz) of some complexes in CDCI3 showing their reactivity toward pyridine. A. [ReN(PO)2(PPh3)]; B. [ReOCl(PPh3)(P02)]. stable in the presence of water. The complexes m<?r-(P,P,P)-[ReN(PO)2(PPh3)] or cw-(P,P)-[ReOCl(PPh3)(P02)], in chloroform-di (NMR sample) were treated with about five drops of pyridine, and a color change from brownish green to green for the former and from brownish red to brown for the latter was observed. In each case, replacement of PPh3 with pyridine is seen in the 31P{1H} NMR 99 spectrum by the increase or appearance of a singlet at ~ -5 ppm, and the disappearance of the upfield P signal in the original spectrum (Figure 4-8). Clearly, the PPI13 ligands are labile in the two complexes and the assignment of the bound PPI13 in the original 31P{ lU} NMR spectra can be made. Although dissociation of PPI13 in mer-(P,P,P)-[ReN(PO)2(PPh3)] was observed in CDCI3 (vide supra), an associative interchange mechanism, as opposed to a dissociative interchange, is believed to be involved in the py-exchange reaction because of the large excess of py, leading to a cw-(P,P)-pyridino species (Figure 4-8). Pyridine exchange was not found for the complex fran.s-(P,P)-[Re(NPh)Cl(PPh3)(P02)] at room temperature, although the exchange with HPO was observed in the presence of base when heated. 4.4 Conclusion Neutral rhenium(V) and technetium(V) complexes of the form [MOCl(PO)2], [ReN(PO)2(PPh3)],/ac-[ReZCl(PPh3)(P02)],/flc-[ReZ(PO)(P02)], and [MO(P02)(HP02)], where M is Tc or Re, Z is O or NPh, and POxx" is the o-(diphenylphosphino)phenolate (x = 1) anion or the o,o'-(phenylphosphino)diphenolate (x = 2) dianion, have been synthesized and characterized (Scheme 4-3). The bidentate (P.O) monoprotic ligands form oxo metal(V) complexes which are structurally analogous to the complexes of the two metals with other bidentate (0,0) or (N,0) monoprotic ligands. The mono(PC>2) phenylimido complexes were isolated as cis- or trans-(P,P) geometrical isomers, depending on the reaction conditions, while only a cw-(P,P) product was prepared for the mono(P02) oxo analog. Mixed-ligand complexes incorporating both PO22" and PO_ ligands with cis- or trans-(P,P) geometry were synthesized from appropriate mono(P02) complexes. Mixed(P02/HP02) complexes were obtained as mixtures of two diastereomers with one ligand protonated (HPO2); these diastereomers were not interconvertible and showed good resistance to hydrolysis. They could also be prepared by reduction directly from perrhenate or pertechnetate. Mixed(PO/PC«2) complexes or mixed(PC<2/HP02) complexes could not be prepared from triphenylphosphine-containing 100 [ReOCl3(PPh3)2] or [TBA][TcOCl4] [ReNCl2(PPh3)2] Bis(PO) Complexes Mixed(PO/P02) Complexes Mono(P02) Complexes " Z = NPh: [Re(NPh)Cl3(PPh3)2] Z = 0 : [ReOCl3(PPh3)2] [ReCl4(PPh3)2] [Re(NPh)Cl3(PPh3)2] ' ^ - O - ^ M — O ' - P I , 0 NH4MO4 or [ReCl4(PPh3)2] ^t Mixed(P02/HP02) Complex [TBA][TcOCl4] orNH4M04 (M = Re or Tc) Scheme 4-3 101 precursors via ligand-exchange reactions. The double anchoring phenolate rings in PO22" strengthen the Re-P bonds even more significantly than one phenolate does in PO". These compounds are important since they provide fully characterized models for the structures of intermediates in the synthesis of potential radiopharmaceuticals. Work on 99nrr/c complex of H2PO2 is presented in Chapter 6. 102 References (1) Fritzberg, A. R.; Lyster, D. M.; Dolphin, D. H. J. Nucl. Med. 1977,18, 553. (2) Conner, K. A.; Walton, R. A. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, K. A., Eds.; Pergamon: Oxford, 1987; Vol. 4; p 125. (3) Deutsch, E.; Libson, K.; Jurisson, S.; Lindoy, L. F. Prog. Inorg. Chem. 1983, 30, 75. (4) Fergusson, J. E. Coord. Chem. Rev. 1966,1, 459. (5) Johnson, N. P.; Lock, C. J. L.; Wilkinson, G. J. Chem. Soc. 1964, 1054. (6) Nicholson, T.; Cook, J.; Davison, A.; Jones, A. G. Inorg. Chim. Acta 1994, 218, 97. (7) de Vries, N.; Cook, J.; Jones, A. G.; Davison, A. Inorg. Chem. 1991, 30, 2662. (8) Refosco, F.; Bolzati, C; Moresco, A.; Bandoli, G.; Dolmella, A.; Mazzi, U.; Nicolini, M. J. Chem. Soc, Dalton Trans. 1991, 3043. (9) Bolzati, C ; Refosco, F.; Tisato, F.; Bandoli, G.; Dolmella, A. Inorg. Chim. Acta 1992, 201,1. (10) Refosco, F.; Tisato, F.; Bandoli, G.; Bolzati, C ; Dolmella, A.; Moresco, A.; Nicolini, M. J. Chem. Soc, Dalton Trans. 1993, 605. (11) von Zon, A.; Torny, G. J.; Frijns, J. H. G. Reel. Trav. Chim. Pays-Bas 1983,102, 326. (12) Gutierrez, E.; Koch, S. A. Abstracts of Papers, 208th National Meeting of the American Chemical Society, August 1994; American Chemical Society: Washington, DC. (13) Yardley, J. P.; Fletcher, H. 3rd. Synthesis 1975, 244. (14) Davison, A.; Trop, H. S.; Depamphilis, B. V.; Jones, A. G. Inorg. Synth. 1982, 21, 160. (15) Chatt, J.; Garforth, J. D.; Johnson, N. P.; Rowe, G. A. J. Chem. Soc. 1964, 1012. 103 (16) Chatt, J.; Rowe, G. A. /. Chem. Soc. 1962, 4019. (17) Empsall, H. D.; Shaw, B. L.; Turtle, B. L. / . Chem. Soc, Dalton Trans. 1976, 1500. (18) Rauchfuss, T. B. Inorg. Chem. 1977,16, 2966. (19) Ebsworth, E. A. V.; Rankin, D. W. H.; Cradock, S. Structural Methods in Inorganic Chemistry; Blackwel Scientific Publications: Oxford, 1987; p 58. (20) Tebby, J. C. In Phosphorus-31 NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G. and Quin, L. D., Eds.; VCH: Deerfield Beach, FL, 1987; p 1. (21) Senear, A. E.; Valient, W.; Wirth, J. /. Org. Chem. 1960, 25, 2001. (22) Nugent, W. A.; Mayer, J. M. Metal-Ligand Multiple Bonds; Wiley-Interscience: New York, 1988; a. p 156; b. p 179; c. pi 16; d. p 123; e. p 154; f. p 175. (23) Forsellini, E.; Casellato, U.; Graziani, R.; Carletti, M. C ; Magon, L. Acta Cryst. 1984, C40, 1795. (24) Mazzi, U.; de Paoli, G.; di Bernardo, P.; Magon, L. J. Inorg. Nucl. Chem. 1976, 38, 721. (25) Parish, R. V. NMR, NQR, EPR, and Mossbauer Spectroscopy in Inorganic Chemistry; 1st ed.; Ellis Horwood: New York, 1990, p 64. (26) Jessop, P. G.; Rettig, S. J.; Lee, C.-L.; James, B. R. Inorg. Chem. 1991, 30, 4617. (27) Abram, U.; Lorenz, B.; Kaden, L.; Scheller, D. Polyhedron 1988, 7, 285. (28) Dilworth, J. R.; Griffiths, D. V.; Hughes, J. M.; Morton, S.; Archer, C. M.; Kelly, J. D. Inorg. Chim. Acta 1992,195, 145. (29) Canestrari, M.; Chaudret, B.; Dahan, F.; Huang, Y.-S.; Poilblanc, R.; Kim, T.-C; Sanchez, M. J. Chem. Soc, Dalton Trans. 1990, 1179. (30) Bertolasi, V.; Sacerdoti, M.; Gilli, G.; Mazzi, U. Acta Cryst. 1982, B38, 426. (31) Archer, C. M.; Dilworth, J. R.; Jobanputra, P.; Harman, M. E.; Hursthouse, M. B.; Karaulov, A. Polyhedron 1991,10, 1539. (32) Nicholson, T.; Davison, A.; Jones, A. G. Inorg. Chim. Acta 1991,187, 51. 104 (33) Lebuis, A.-M.; Beauchamp, A. L. Can. J. Chem. 1993, 71, 441. (34) Kettler, P. B.; Chang, Y.-D.; Zubieta, J.; Abrams, M. J. Inorg. Chim. Acta 1994, 218, 157. (35) Levason, W. In The Chemistry of Organophosphorus Compounds; Hartley, F. R., Ed.; Wiley-Interscience: Chichester, 1990; Vol. 1, Chapter 15 (Phosphine Complexes of Transition Metals); p 572. (36) Dunne, B. J.; Orpen, A. G. Acta Cryst. 1991, C47, 345. 105 Chapter Five Metal Complexes Incorporating the Deprotonated Forms of P, P, P', P'-tetrakis(o-hydroxyphenyl)diphosphinoethane (H4P2O4) 5 .1 . Introduction The easy preparation of mixed(P02/HPC>2) complexes, [MO(PC>2)(HP02)] (M = Re or Tc), from the direct interaction of a salt of bis(o-hydroxyphenyl)phosphine (H2PO2HCI) with MO4", suggests that this functionalized phosphine is effective both as a reducing and a ligating agent, and that the combination of anchoring phenolate oxygen atoms (upon deprotonation) and the softer phosphine P atoms stabilizes Re and Tc metal(V) centers. The ligand can both coordinate and reduce the metal center, a distinct advantage when considering potential 99mrr/c_ radiopharmaceuticals where the only convenient source of 99mxc j s 99nrrc04- from a 99Mo/99mxc generator. In an extended study of functionalized phosphines, a novel, potentially hexadentate tetraprotic ligand precursor, the hydrochloride salt of P, P, P', P'-tetrakis(o-hydroxyphenyl)-diphosphinoethane, or H4P2O4 2HC1, which contains two soft phosphine phosphorous donor atoms and four hard phenolate oxygen atoms, was synthesized. It was of interest to investigate the coordination chemistry of Tc/Re with this functionalized diphosphine ligand with a P2O4 donor set before pursuing the labeling of this ligand with 99nrrc and evaluating its potential as an imaging agent. The results of these studies are presented in this chapter. H4P204-2HC1 106 5.2. Experimental Materials. All chemicals were reagent grade and were used as received: phenol, PPI13, dimethoxymethane, n-butyllithium, TMEDA (N,N,N'N'-tetramethylethylenediamine) (all above from Aldrich), CI2PCH2CH2PCI2 (a gift from Mr. G. Clentsmith and Professor M. D. Fryzuk),2 NH4Re04 (a gift from Johnson-Matthey, Inc.), HCl(g) (Matheson), and [NH4][99Tc04] (a gift from the Du Pont Merck Pharmaceutical Company). PhOCH20CH3 (mom-protected phenol) and [n-Bu^HTcOC^], were prepared according to published procedures. Caution! " T c is a low energy (0.292 MeV) (3~ emitter with a half-life of 2.12 x 105 years. All manipulations of solutions and solids were performed in a laboratory approved for the handling of low-level radioactivity, and normal safety procedures were followed at all times to prevent contamination. Instrumentation. Mass spectra were obtained with either a Kratos MS 50 (electron impact ionization, EIMS) or a Kratos Concept IIH32Q instrument (Cs+-LSIMS with positive or negative ion detection). Only the most intense peaks are given where consistent isotopic patterns were observed. Infrared spectra were recorded as KBr pellets in the range 4000-400 cm-1 on a Perkin-Elmer PE 783 spectrophotometer and were referenced to polystyrene. Microanalyses were performed by Mr. P. Borda in this department (all compounds other than the Tc complex) or by Canadian Microanalytical Services Ltd. (Tc complex). *H NMR spectra (200 MHz, 300 MHz, 400 MHz or 500 MHz) were recorded on Bruker AC-200E, Varian XL 300, Bruker WH-400 (JH-iH COSY), or Bruker AMX-500 ^Hf 3 1 ?}) spectrometers, respectively, with 8 referenced to external TMS. The 31P{ JH} NMR spectra (81 MHz or 121 MHz) were recorded on Bruker AC-200E or Varian XL 300 spectrometers, respectively, with 8 referenced to external H3P04 . The assignments were based on those for the unbound ligand and by comparison between analogous complexes. Thermogravimetric study was carried out on a TGA 51 107 Thermogravimetric Analyzer (TA Instruments), with N2 purging at 100 mL/rnin, at the heating rate of 10 °C/min to 600 °C. Electrochemistry. Cyclic voltammetry was carried out with a PARC (Princeton Applied Research Co.) Model 264 polarographic analyzer/stripping voltammeter. Cyclic voltammograms were recorded by using a PARC Model RE0089 X-Y recorder. Electrochemical measurements were carried out under an argon atmosphere at room temperature. Solution concentrations were approximately 10-3 M in complex and 0.1 M in the supporting electrolyte TEAP (tetraethylammonium perchlorate). A platinum working electrode, a platinum wire counter electrode, and an Ag/AgCl reference electrode were used, and checked periodically relative to a 5 mM solution of ferrocene in acetonitrile containing 0.1 M TEAP for which the ferrocenium/ferrocene reduction potential was 0.42 V and A£p =110 mV at a scan rate of 100 mV s-1. (0-C6H4OCH2OCH3)2PCH2CH2P(0-C6H4OCH2OCH3)2-O.5H2O ( (mom)4 -P2O4'0.5H2O, mom = CH2OCH3) This was prepared from the mom-protected phenol according to a procedure for Ph2P(o-C6H4-OCH20CH3) ((mom)PO) with some modifications. To an ice-cooled solution of methoxymethyl phenyl ether (21.75 g, 157 mmol) in ca. 200 mL of petroleum ether (b.p. 35 - 65 °C, dried with anhydrous Na2S04 overnight) was added a solution of 100 mL of 1.6 M n-BuLi in hexanes and 18.05 g TMEDA in 50 mL of petroleum ether under N2. The mixture was stirred for 4 h at room temperature. A yellow precipitate formed from the orange solution, which was subsequently heated to ca. 40 °C with stirring. After cooling to 0 °C, CI2PCH2CH2PCI2 (9.30 g, 40 mmol) in 20 mL of petroleum ether was then added via a syringe. The mixture was stirred overnight, during which time it warmed to room temperature. The solvents were removed by rotary evaporation and to the residue was added Na2HP04 (0.5 M, 200 mL). The reaction mixture was then extracted with Et20 (2 x 200 mL) followed by CHCI3 (2 x 100 mL). All the organic layers were combined, concentrated to a reddish oil under low pressure, diluted with MeOH (ca. 25 mL), and stored at -4 °C overnight. A yellowish crude product was filtered off, washed with cold methanol (2 x 10 mL), and dried in vacuo. The yield of the off-white product was 11.5 g (45 % based on mom-protected phenol). Anal. Calcd 108 (found) for C34H4i08.5P2: C 63.06 (63.19), H 6.38 (6.24). Mass spectrum (EI): m/e = 638 ([(mom)4P204]+), 623 ([(mom)4P204 - CH3]+). : H NMR (CDC13): 7.24 (t, 40, p-H), 7.14 (dd, 4H, o'-H), 7.03 (d, 4H, m-H), 6.91 (t, 4H, m'-H), 5.04 (s, 8H, OCH20), 3.17 (s, 12H, OCH3), 2.28 (dd, 4H, backbone-C#2). 31P{1H} NMR (CDCI3): -31.7 (s). IR (cm"1, KBr disk): 3060 (m, aromatic VC-H); 3000-2800 (m, vc-H, methyl and methylene of mom group). P, P, P', P'-tetra(o-hydroxyphenyl)diphosphinoethane Dihydrochloride Salt (H4P204-2HC1). This was prepared from the mom-protected phenolic phosphine (mom) 4 P20 4 according to a procedure5 for Ph2P(o-C6H4-OH) or HPO, with some modifications. Anhydrous HC1 gas was bubbled overnight with stirring, via a dispersion tube, into a solution of (mom)4P204 (7.36g, 11.5 mmol) in 350 mL of anhydrous methanol. (The white suspension dissolved to give a clear solution after 1 h, but became cloudy again overnight.) A fine white solid was filtered off, washed with ethanol (2 x 20 mL), and dried in vacuo. The yield was 3.95 g (74 % based on phosphine). No recrystallization was necessary to obtain an analytically pure sample. Anal. Calcd (found) for C26H26Cl204P2: C 58.33 (58.20), H 4.90 (4.95), CI 13.24 (13.07). EIMS: m/e = 462 ([H4P204]+), 368 ([H4P204 - C6H60]+). *H NMR (DMSO-^6): 10.8 (broad s, 4H, OH), 7.6-7.3 (overlapped multiplets, 8H), 7.0-6.8 (multiplet, 8H), 2.6 (broad s, 4H, backbone-C//2). 31P{!H} NMR: 43.5 (s, DMSO-d6); -36.4 (s, py-J5). IR (cm-1, KBr disk): 3100-2800 (vs, b, VC-H)-[Re202Cl2(PPh3)2(|Li-P204)]. A mixture of ReCl4(PPh3)2 (168.2 mg, 0.20 mmol), H4P204-2HC1 (109.2 mg, 0.20 mmol) and 8 drops of Et3N in 15 mL of ethanol was brought to reflux overnight. After the mixture was cooled to room temperature, a greenish yellow precipitate was isolated by centrifugation, washed with Et20 and dried in vacuo for 4 h. The yield was 40.1 mg (67%). The product was found to be soluble in pyridine, dimethyl sulfoxide (decomposes), slightly soluble in chloroform, but insoluble in diethyl ether. Anal. Calcd (found) for C62H5oCl206P4Re2: C 51.07 (50.76), H 3.46 (3.59), CI 4.86 (4.65). LSIMS: m/z= 1459 ([M + 1]+), 1196 ([M - PPh3)]+), 1161 ([M - CI - PPh3]+), 1124 ([M - 2C1 - PPh3]+), 645 ([Re(P204)]+), 263 (PPh3+ 1). !H NMR (py-d5): 7.70 (dd, 12H, o-H on PPh3), 7.50 (dd, 2H), 7.40-7.20 (overlapped multiplets, 18H, m and p-H on PPh3), 7.20 (d, 2H), 7.10 (d, 2H), 109 7.75 (m, 2H), 6.65 (t, 2H), 6.40 (t, 2H), 6.30 (t, 2H), 5.80 (dd, 2H), 2.85 (m, 2H, ethylene-HA and -#A'X 2.00 (d, 2H, ethylene-//B and -//B0- 31P{1H} NMR (py-dsY 15.7, -11.3. IR (cm-1, KBr disk): 3060 (m, VC-H); 965 (s, vRe=o). [(n-Bu)4N]2[Re202Br4(n-P204)]-Me2CO. To a mixture of [(n-Bu)4N][ReOBr4] (77.6 mg, 0.10 mmol), [(n-Bu)4N]Br (151.3 mg, 0.45 mmol) and H4P204-2HC1 (55.5 mg, 0.10 mmol) was added 20 mL of toluene, and the mixture was refluxed for 2 h. From the resultant green oil a green solid was precipitated with z'-PrOH. Recrystallization from acetone gave emerald green crystals which were filtered, washed with cyclohexane, and dried in vacuo overnight. The yield was 31 mg (35%). The product was found to be soluble in ethanol, acetone, chloroform, dichloromethane, but insoluble in diethyl ether or cyclohexane. Anal. Calcd (found) for C6iH98Br4N207P2Re2: C 42.46 (42.62), H 5.72 (5.47), N 1.62 (1.71). +LSIMS: m/z= 242 ([(n-Bu)4N]+), 645 ([Re(P204)]+), 662 ([ReO(HP204)]+); -LSIMS: 79 (Br), 645 ([Re(P204)]-), 661 ([ReO(P204)]-), 769 ([ReO(P204) + 108]-). !H NMR (acetone-^): 7.9-7.6 (overlapped multiplets, 4H), 7.22 (t, 1H), 7.16 (t, 1H), 7.1-6.9 (overlapped multiplets, 4H), 6.78 (t, 1H), 6.7-6.6 (overlapped multiplets, 2H), 6.55 (t, 1H), 6.2 (overlapped multiplets, 2H), 4.14 (m, 2H, ethlyne-#A and -HA'), 3.45 (t, 16H, a-H of the n-Bu group), 2.65 (d, 2H, ethylene-//B and -//B-), 1-70 (quintet, 16H, B-H of the n-Bu group), 3.45 (sextet, 16H, y-H of the n-Bu group), 0.95 (t, 24H, Me-// of the n-Bu group). 31P{ lH} NMR (acetone-d6): 18.0 (major); 17.0 (minor). IR (cm"1, KBr disk): 3060 (m, VC-H); 2980 and 2780 (s and m, Vc-H of the n-Bu group), 960 (s, VRe=o)-[Tc(HP2O4)]-2EtOH0.5PhMe. To a mixture of NH4Tc04 (18.7 mg, 0.10 mmol) and H4P204-2HC1 (109.2 mg, 0.20 mmol) was added 5 mL of ethanol and 20 mL of toluene. The color of the solution immediately became pale purple. Upon refluxing for 1 h, a dark crystalline precipitate formed from a dark purple solution. The supernatant was removed, and the crystalline solid was washed with toluene, hexanes, and diethyl ether (3 mL each). The precipitate was recrystallized from EtOH/Et20 and dried in air. The yield of the purple crystalline product was 57 mg (79 %). The product was found to be soluble in methanol, ethanol, acetone, moderately soluble in chloroform and dichloromethane, but insoluble in diethyl ether or 110 cyclohexane. Anal. Calcd (found) for C33.5H37O6P2TC: C 57.77 (57.97), H 5.35 (5.03), P 8.89 (9.12). LSIMS+: m/z= 559 ([Tc(HP204) + 1]+), 1021 ([Tc(H3P204)2]+), 1127 ([Tc(H3P204)2 + 106]+); LSIMS-: m/z= 557 ([Tc(P204)]-), 1019 ([Tc(H2P204)2]-), 1125 ([Tc(H2P204)2 + 106]"). IR (cm-1, KBr disk): 3230 (br, V 0 -H); 3060 (w, VC-H)-[Fe(HP 2 04) ] -2H 2 0 . To a mixture of Fe(acac)3 (35.3 mg, 0.10 mmol) and H4P204-2HC1 (54.0 mg, 0.10 mmol) was added 5 mL of methanol. Upon refluxing for 3 h, a green crystalline solid formed in a brownish solution. The supernatant was removed, and the crystalline solid was washed with Et20 ( 3 x 3 mL) and dried in air. The yield of the green crystalline product was 30.8 mg (56%). The product was found to be soluble in dimethyl sulfoxide, moderately soluble in methanol and ethanol, slightly soluble in acetone, acetonitrile, benzene and dichloromethane, but insoluble in diethyl ether or water. Anal. Calcd (found) for C26H25Fe06P2: C 56.33 (56.36), H 4.73 (4.61). LSIMS+: m/z= 516 ([Fe(HP204) + 1]+), 978 ([Fe(H3P204)2]+), 1084 ([Fe(H3P204)2 + 106]+); LSIMS-: m/z= 515 ([Fe(HP204)]-), 976 ([Fe(H2P204)2]-), 1082 ([Fe(H2P204)2 + 106]"). IR (cm"1, KBr disk): 3200 (br, VO-H), 3060 (w, VC-H)-5.3. Results and Discussion Ligand. The new ligand precursor H4P204-2HC1 was synthesized in a manner similar to that used to obtain H 2 P0 2 H O (Chapter 4). After protecting the hydroxyl group as the methoxymethyl (mom) ether, this mom-protected phenol was o/t/10-lithiated at low temperature under a nitrogen atmosphere, and 4 equivalents of it was then reacted with CI2PCH2CH2PCI2 to give the mom-protected intermediate, (mom)4P204. Upon treatment with anhydrous HC1 gas in methanol or ethanol, the expected functionalized diphosphine was obtained as the hydrochloride salt (Scheme 5-1). For the intermediate, (mom)4P204, the EIMS shows the parent ion peak (m/z= 638, [(mom)4P204]+), along with fragments formed from of the parent ion by the loss of -CH3, 111 OH 1. CH3OCH2OCH3 2. «-BuLi/TMEDA Li ^ O ^ O CI2PCH2CH2PCI2 HCl(g) MeOHorEtOH(anhy.) / \ •2HC1 (mom)4P204 H4P2042HC1 Scheme 5-1 -OCH3, -CH2OCH3, -OCH2OCH3 groups. In the 31P{!H} NMR spectrum in CDCI3, a singlet at -31.7 ppm, typical for a phosphine compound, is present. The *H NMR spectrum showes four hydrogen resonances at 7.24, 7.14, 7.03 and 6.91 ppm in the aromatic range, corresponding to p-, o'-, m- and m'-H of the phenyl ring,* assigned on the basis of the iH-1!! COSY spectrum. The resonances corresponding to methyl and methylene H atoms of the mom group, appear as singlets at 5.04 and 3.17, respectively. The two methylene H atoms in each mom group of (mom)2P02 give an AB quartet, indicating non-equivalence of the two methylene hydrogens due to prochirality (Chapter 4, page 78). This was not the case with (mom)4P204. The four chemically equivalent backbone H atoms give a doublet of doublets due to unequal coupling to the two P nuclei. As for H2PO2 HC1, the formulation of a hydrochloride salt This is designated as in the following diagram: m1 O ^ p m 112 H4P2O42HCI is supported by the elemental analysis and by the 31P chemical shift characteristic of a phosphonium P at +41.5 ppm in DMSO-^6- The salt dissociated in its ligand-exchange reactions (vide infra) and in basic solution, as is indicated by the chemical shift of -36.4 ppm in the 31P{ !H} NMR spectrum in py-d5. EIMS shows the presence of [H4P204]+. The *H NMR spectrum in DMSO-afo shows four overlapped H resonances in the aromatic region and a broad backbone H resonance at 2.6 ppm with the expected integral ratio. Synthesis of the Complexes. The phosphine hydrochloride salt H4P2O42HCI, when deprotonated by a base, reacted with ReCl4(PPh3)2 forming in a dinuclear complex [Re202Cl2(PPh3)2(|i-P204)]. Aerial oxidation was responsible for oxidizing Re(IV) to Re(V). In another trial without added base, the ligand did not react with the Re precursor; red crystals obtained from this system were identified by X-ray crystallography as a partially oxidized mixture of ?ra«5-[ReCl4(PPh3)2] and trans-iReOCl^iPPh^] (Appendix IV). A dinuclear dianionic complex [Re2C>2Br4(|i-P204)]2- ([(n-Bu)4N]+ salt) formed from the reaction of [{n-Bu)4N][ReOBr4] with the ligand, with added [(n-Bu)4N]Br but no added base. These observations, taken together, suggest that an extra driving force is necessary to push forward the substitution reaction of the H4P2O4 ligand with a precursor that contains PPh3. This is consistent with observations made in Chapter 4 with H2PO2. In addition, the chelation of the P2C>44_ ligand to a single Re center is hindered. The rapid reaction of H4P2O42HCI with TCO4- in a toluenerEtOH (4:1) mixture was shown by the immediate appearance of a purple color upon mixing the two materials. Upon heating for 1 h, a dark purple solid deposited. Here the free phosphine functions as a reducing agent as well as a ligand. EtOH added to the system (EtOH:PhMe = 1:4) aided the reaction by dissolving both starting materials and thus initiating the reaction; however, reactions carried out in absolute EtOH resulted in the same product, but in lower yields, as the product is more soluble in ethanol than in the solvent mixture (EtOH:PhMe = 1:4). A green product was isolated after 3 h refluxing of a mixture of H4P2O42HCI and [Fe(acac)3] (maroon) (L:M = 1:1) in MeOH, indicating that substitution is easily accomplished 113 with this ligand. Characterization of the Complexes. All the complexes are air stable in the solid state. The complexes were characterized by elemental analysis, infrared spectroscopy, mass spectrometry, and ^W^Pl lH} NMR spectroscopy. IR measurements confirmed the existence or absence of metal-oxo multiple bonds and ensured that the multidentate ligands were coordinated as evidenced by the absorptions shifted in comparison with the free ligands. Both *H NMR and 3 1 P { 1 H } spectra were very useful in verifying the structures of [Re202Cl2(PPh3)2(|H-P204)] and [(n-Bu)4N]2[Re202Br4(^-P204)]. Cyclic voltammetry presented additional information about the electrochemical activities of the Tc and Fe complexes (vide infra). Dinuclear (P2O4) Re Complexes. Mass spectrometric data confirmed the formation of the dinuclear complexes when the expected parent ions and/or their fragments were found, while microanalysis established the formulation of both compounds. The parent ion [Re202Cl2(PPh3)2(H-P2C>4) + 1]+ (m/z = 1459) is present in the +LSIMS of [Re202Cl2(PPh3)2(u-P204)]; however, the stronger peaks are at 263 (PPh3), 645 ([Re(P204)]+), and 1161 ([M - CI - PPh3]+ or [Re202Cl(PPh3)(u-P204)]+), respectively, indicating that the monodentate CI" or PPI13 ligands are subject to dissociation from the complex under the ionization conditions. For [(«-Bu)4N]2[Re202Br4(|i-P204] positive detection mode LSIMS, 242 ([(n-Bu)4N]+) is intense, along with a much weaker peak for [Re(P204)]+, while in the negative mode, 79 (Br") is the most intense peak, followed by fragments at 661 ([ReO(P204)]-) and 769 ([ReO(P204) + 108]-). The mass of 108 is that of the matrix, thioglycerol (C3H8O2S). The dianionic parent ion [Re202Br4(^i-P204)]2" did not show in the negative mode, presumably due to too strong an interaction between the dianion and the very polar matrix, thioglycerol, to allow the dianion to volatize. There are similarities in the IR spectra of the two dinuclear complexes. Strong vibrational bands at 2960 and 2880 cm-1 were observed for [(«-Bu)4N]2[Re202Br4((i-P204)], and are indicative of [(n-Bu)4N]+. Below 2000 cm-1, almost all the absorption bands for [(«-Bu)4N]2[Re202Br4(u-P204)] were observed in that of [Re202Cl2(PPh3)2(n-P204)]. The Re=0 114 stretches for [ ( n - B u ^ h ^ C ^ B ^ u ^ C ^ ) ] ) and [Re202Cl2(PPh3)2(^-P204)]), were found at 960 and 965 cm-1, respectively, being in the normal range for six-coordinate oxorhenium complexes.7 The additional bands in [Re202Cl2(PPh3)2(|i-P2C>4)] are attributed to PPh3 bound to the Re(V) center A seven-coordinate technetium complex, as formed with edta4*, seemed unfeasible with this P2O44- ligand. In [TcO(edta)]", the two neutral tertiary amine N atoms are approximately opposite the oxo ligand, the Tc-N bonds being somewhat lengthened by the trans influence. Thus, one possibility in the case of P2O44" is that the phosphine P atom is much softer than the amine N atom, and soft, neutral donors are usually in the cis positions relative to the oxo group to avoid the trans influence. Phenolate rings are also less flexible than the carboxylate arms. i—r 8.0 1 r 7.0 6.0 5.0 4.0 3.0 2.0 1.0 ppm # = solvent or HOD; * = impurities Figure 5-1. lH NMR spectra for [Re202Cl2(PPh3)2(u-P204)] (top, 300 MHz) and [0i-Bu)4N]2[Re2O2Br4(u-P2O4)] (bottom, 400 MHz). 115 In the 31P{!H} NMR spectrum of [Re202Cl2(PPh3)2(u-P204)], two peaks of equal intensity, indicating two nonequivalent phosphorus centers, are present (Figure 5-5, page 123). This suggested that the two 11-P2O44" P nuclei are equivalent, as are the two PPI13 P nuclei. In accord with this, in the *H NMR spectrum (Figure 5-1), there are eight types of hydrogen nuclei for the PO-phenyl rings of the complex in the aromatic region, indicating that the four phenolate rings of the ILI-P2O44" ligand are of two different types, each including two equivalent rings. In this aromatic region, a doublet of doublets (7.70 ppm, o-H) and overlapped multiplets (-7.3 ppm, m-H andp-H) are also present, corresponding to the phenyl H nuclei of PPI13 (Figure 5-1). The four H nuclei on the ethylene backbone are in an AA'BB'XX' spin system, due to chirality of the metal center (Figure 5-2, anti). The syn diastereomer is not present for this neutral dinuclear complex (vide infra). The appearance of two resonances (one triplet-like, the other doublet-like) corresponding to HA and HB nuclei were found to be complicated because of multiple couplings. Figure 5-2. Diastereomers of the dinuclear complexes showing the AA'BB'XX' systems of the ethylene backbone, due to chirality of the metal center. 116 For [(«-Bu)4N]2[Re202Br4((i-P204)], one major and one minor peak were observed in the 31P{1H} spectrum at 18.0 and 17.0 ppm, respectively, (Figure 5-5, page 123). In the lH NMR spectrum, there are more than eight signals for the PO-phenyl rings in the complex. In fact, sixteen H signals (some of them overlapped) in the aromatic region (6.0 - 8.0 ppm) are correlated into four groups by a iH-1!! COSY experiment (Figure 5-3), indicating that four types of PO-phenyl rings are present (numbered from ring 1 to ring 4). The H resonances corresponding to the o'-H of the four PO-phenyl rings are identified in the !H{31P} NMR spectrum (each of them is simplified to a doublet from a multiplet upon decoupling, Figure 5-4). The H signals of rings 1 and 4 are stronger in intensity than those of rings 2 and 3, and the P signal at 18.0 ppm is more intense than that at 17.0 ppm. These, taken together, indicate that there are two isomers, one major and one minor, present in the [(n-Bu)4N]2[Re202Br4(^,-P204)] sample. In the ^ C ^ H } NMR spectrum, only eight signals were observed in the aromatic region, tentatively assigned to the major isomer eight PO-phenyl ring carbon atoms which had hydrogen atoms attached; those of the minor isomer, and quaternary carbons in both isomers, were not observed because the sample was too dilute. For the ethylene H nuclei, a similar AA'BB'XX' spin system was observed for [(n-Bu)4N]2[Re202Br4(|i-P204)], and HA and HB were found to be coupled strongly to each other. The chemical shifts for ethylene C2H4 H nuclei of the backbone of each isomer coincide since only two resonances are present. Upon decoupling of the phosphorus, both HA and HB signals appear as doublets. Two models are proposed for the dinuclear complexes, as shown in Figure 5-2. Assumptions with adequate precedent are made to build these models: neutral P donor(s) are cis to the Re=0, and the PO2 donors of one end of the P2044 ' ligand facially coordinated. Both assumptions are based on the known complexes of PO22" ligand, [ReZCl(P02)(PPh3)] (Z = O or NPh, see Chapter 4). In fact, P trans to M=0, M=N, or M=NR has not been observed for complexes of this type (M = Re or Tc) incorporating phosphine or functionalized phosphine ' ' ligands; the P donors are always cis. 117 Br | Ring 2 j 4 32 1 7.8 7.4 7.0 6.6 6.2 ppm Figure 5-3. The proposed isomers and the W H COSY spectrum (400 MHz) of [(w-Bu)4N]2[Re202Br4(n,-P204)] (aromatic region); the correlated H resonances are designated by the same number as the PO phenyl ring. 118 u. 1—I J. i—r—i—I i '—r—i—I t i i r ) i " i " i i ^ f 8.0 7.5 7.0 6.5 5.0 t | i i l l J l l i l j l l ' l l 4.0 3.0 ppm Figure 5-4. *H (top) and lH{31P} (bottom) NMR spectra (500 MHz) of [(«-Bu)4N]2[Re202Br4(|i-P204)] (aromatic and ethylene regions). The two models might be described as syn and anti with respect to the two oxo groups, each of which possesses C2 or / symmetry, respectively (Figure 5-2). With aid of a model, it is easy to reason why there is only one isomer for [Re202Cl2(PPh3)2( i^-P204)j\ while there are two isomers for [(«-Bu)4N]2[Re202Br4(ji-P204)], as shown in the 31P NMR spectra (Figures 5-5a and e, page 123). [Re2C>2Cl2(PPh3)2(|i-P204)] exists, presumably as the anti isomer (i symmetry): the syn isomer is not feasible because of the two sterically bulky PPI13 groups. For this complex, a proposal with the two P donors trans to each other is also rejected by the fact that the mutual P-P coupling is quite small. In modeling [Re202Cl2(PPh3)2(^-P204)], both syn and 119 anti isomers are found to be feasible as B r is much smaller, and this might be the case as indicated by the NMR results. The two P signals at 15.7 and -11.3 ppm in the 31P{!H} NMR spectrum (Figure 5-5a) for [Re202Cl2(PPh3)2(u-P204)], are assigned to the P nuclei of U-P2O44- and PPI13 ligands, respectively. It was observed that phenolato-anchored phosphine P nuclei experience a greater deshielding effect than the PPI13 ligand in such complexes as [ReZCl(P02)(PPh3)] (Z = O or NPh, in Chapter 4), due to stronger donation. This U.-P2044' P chemical shift was found to be quite close to that of [(n-Bu)4N]2[Re202Br4(|Li-P204)], at 18.0 (major) and 17.0 (minor) ppm (Figure 5-5e). Further support to this assignment is depicted by Figures 5-5c, where the PPI13 P signal is identified (it is replaced by py-^/5, vide infra). Mononuclear (P2O4) Complexes. The empirical formulations [M(HP2C«4)]nS (M = Fe, nS = 2H2O; M = Tc, nS = 2EtOH0.5PhMe) are based on elemental analyses. As one of the solvent molecules is believed to be in the inner sphere of the complex in each case {vide infra), [Fe(HP204)(H20)] and [Tc(HP2C>4)(EtOH)] will be used. This hypothesis is supported by thermal analysis of the Fe complex (Table 5-1). The formulations are also consistent with spectral data. In the +LSIMS/-LSIMS for [Tc(HP204)(EtOH)], the parent ion is not present, suggesting that the ethanol is weakly bound, as is the water in [Fe(HP204)(H20)]. However, the fragment [Tc(H2P204)]+ (m/z = 559) is present in the +LSIMS, while in -LSIMS, [Tc(P204)]- (m/z = 557) is observed. As for [Fe(HP204)(H20)], [Fe(H2P204)]+ (m/z = 516) and [Fe(HP2C>4)]- (m/z = 515) are seen in +LSIMS and -LSIMS, respectively. An oxidation state of three is thus assigned to both complexes. Although there is not enough information to draw a clear picture of the structure of these complexes, a six-coordinate structure with one of the solvent molecules in the inner sphere and a pendent protonated phenol is proposed (Scheme 5-2). A similar structure [M(Hedta)(H20)] is known for many metal ions. The crystal structures of this type of complex reveal that the molecule contains pentadentate Hedta3" with water occupying the sixth position and one carboxylate arm free. It has been postulated that the ionic radius of the metal ion has a great influence on the denticity of the edta4- ligand. When a C.N. of 6 is preferred by ligand field stabilization and the ionic radius exceeds the "critical" 120 Table 5-1. Results of Thermogravimetric Study on [Fe(HP204)(H20)]H2Q. Temperature Range (°C) Volatile Product Weight Loss (%) Weight Loss (%) Calcd Found 50 — 234 H 2 0 3.27 3.26 234 — 250 H20 3.27 3.49 262— (decomposes) radius, i.e., 0.645 A for Fe(III), this structural arrangement is preferred since edta4- cannot span the large ion as a hexadentate ligand. The P2O44- ligand is expected to form five-membered chelate rings as does edta4". These chelate rings for this ligand would be expected to be more rigid as the phenolate arm may be more rigid than the acetate arm of edta4". Thus, such a six-coordinate structure for [M(HP2C>4)(S)] (M = Tc, S = EtOH; M = Fe, S = H20) would not be too surprising. The effective ionic radius of Tc(III) is estimated to be -0.685 A.* Therefore a similar structure for the complexes [M(HP2C>4)(S)] (M = Tc, S = EtOH; M = Fe, S = H20) with the P2C>44" ligand is proposed (Scheme 5-2). The IR spectra of these two complexes are superimposable for the major absorption bands in the range of 1600 - 700 cm-1 (there was no VTC=0 observed for the Tc complex), indicating a similar structural feature for both complexes. Shifts of these major bands from those for the free ligand were observed upon coordination. Electrochemical Behavior of the Mononuclear (P2O4) Complexes. The electrochemistry of [Tc(HP204)(EtOH)] and [Fe(HP204)(H20)] was investigated by cyclic voltammetry. For comparison, the supporting electrolyte, 0.1 M TEAP, in CH3CN was also studied. A huge reduction wave of the supporting electrolyte (C104-) and an oxidation wave were observed initiating at - -1.70 and ~ +2.75 V, respectively. After three drops of concentrated HCIO4 were added, a reduction wave was found at - 0.15 V vs Ag/AgCl (Appendix IV). This is The average of the effective ionic radii of the six- coordinate Mo(HI) (0.69 A) and Ru(III) (0.68 A).: 121 due to reduction of H+, as it is the only reducible species, and the potential found is close to the standard reduction potential (£Ag/AgCl = + 0.197 V vs NHE).26 No significant redox wave was detected in the range of +2.00 to -2.00 V vs. Ag/AgCl for [Tc(HP204)(EtOH)]. However, a similar reduction wave was found at approximately - 0.15 V after three drops of concentrated HCIO4 were added (Appendix IV). As was previously observed with HCIO4 in 0.1 M TEAP in MeCN, this is attributed to the reduction of H+. Upon addition of base, there were no significant changes in the cyclic voltammogram for [Tc(HP2C>4)(EtOH)]. For [Fe(HP204)(H20)], no significant redox wave was detected in the range of+2.00 ~ -2.00 V vs. Ag/AgCl (Appendix IV). The fact that no redox processes associated with the complexes over a very wide range of potentials were observed, clearly indicates that the oxidation states of the Tc(III) and Fe(III) complexes are preferentially stabilized kinetically* by the coordination environment of two P and four O donor atoms. In a previous study, it was found that Tc(IV) is stabilized over a wide range by six O donors from three bidentate monoanionic N-substituted-3-oxy-4-pyridinonate ligands. Preference of the P2O44- Ligand. The rigidity of the ligand frame and/or trans influence make it impossible for all the donor atoms to bind a single center of Re, Tc, or Fe. However, this ligand, can bridge two Re(V) centers, incorporating monodentate oxo, chloro, and PPI13 ligands to complete the coordination sphere of each Re(V). It does, however, greatly stablize Tc(III) and Fe(III) centers. Reactivity to pyridine. A sample of [Re202Cl2(PPh3)2(|i-P204)] in py-ds was used to examine the reactivity of the complex. The replacement of PPI13 with pyridine was observed in the 31P{1H} NMR spectrum by the appearance of a singlet at ~ -9.1 ppm (free PPI13), and the intensity decrease of the upfield P signal (-11.3 ppm, bound PPI13) in the original spectrum (Figure 5-5). After heating to 100 °C then cooling to room temperature, the replacement went to There must be a thermodynamic redox potential in the range of -2.0 to -2.0 V. See Lu, J.; Yamano, A.; Clarke, M. J. Inorg. Chem. 1990, 29, 3483. 122 a W|^A**^«**^ WVw«Wv*S»,>«—«w— I... .,•, ^ i ^ ) ^ « y ^ ^V»M>*V»»* ppm VftVW*q*v^^^*j«»f%WW d W •%*>»•»*»»•»*» OWMwiN—^^tw***1 AJv *«Nw^»«>i W ,* fn*» #•. r 50 g 40 30 T T 20 1 iw I 10 ppm a. [Re202Cl2(PPh3)2(u-P204)] in py-ds; b . a after standing for 6 months; c . b after heated to 100 °C then cooled to room temperature. e . [(w-Bu)4N]2[Re202Br4(^-P204)] in acetone-^; f. after 3 drops of py-^5 were added to e; g . after f was heated to ca. 60 °C then cooled to room temperature. Figure 5-5. 31P{ lH] NMR spectra (81 MHz) showing reactivity of the dinuclear complexes to pyridine. (* = unknown decomposition products.) 123 completion, whereupon the bound PPh3 P signal (-11.3 ppm) disappeared completely, and the downfield P signal (15.7 ppm) was replaced by two singlets (14.3 and 14.0 ppm). Clearly, the PPI13 ligands are labile in the complex and the assignment of the bound PPI13 in the original 31P{ if!} NMR spectrum is verified. No attempt was made to fully characterize the new complex formed with pyridine, however, syn and anti isomers are believed to be responsible for the two singlets at 14.3 and 14.0 ppm. As for [(n-Bu)4N]2[Re202Br4(u,-P204)], 3 drops of py-c?5 were added to a sample of the complex in acetone-^- Changes in the P resonances were also observed in the 31P{ ^K} NMR spectrum: the original peaks were reduced in intensity, while two new singlets appeared (Figure 5-4). After heating at ca 60 °C for 10 min and then cooling to room temperature (the solution changed to yellowish green from the original emerald green), this change went to completion, whereupon the original P signals (18.0 and 17.0 ppm) disappeared completely, giving rise to two singlets at 22.1 and 22.6 ppm (Figure 5-4). The dianionic complex is also labile to substitution. Again, similar syn and anti isomers could be assigned to the substitute products. 5.4. Conclusion The hydrochloride salt of a new ligand, the potentially hexadentate and tetraprotic H4P2O42HCI, was synthesized and characterized. With this salt, a neutral [Re202Cl2(PPh3)2(^-P204)] and a dianionic [Re202Br4(u-P204)]2- dinuclear rhenium(V) complexes, and two neutral mononuclear complexes [Tc(HP2O4)(EtOH)]EtOH0.5PhMe and [Fe(HP204)(H20)]H20, were synthesized and characterized (Scheme 5-2). The hexadentate (P2O4) tetraprotic ligand also functioned as a reducing agent; the synthesis of [Tc(HP2O4)(EtOH)]EtOH0.5PhMe was rapidly accomplished via a reduction/coordination route directly from pertechnetate. More study is necessary to investigate further the structural chemistry and the magnetic properties of the mononuclear complexes. These compounds are important since they provide useful information about the coordination preference of the ligand, 124 Dinuclear (P2O4) Re Complexes [Re202Cl2(n-P204)(PPh3)2] <CKL° v r_> -[Re202Br4(^-P204)] o2 ReCl4(PPh3)2 ReCl4(PPh3)2 * Mononuclear (P2O4) Complexes Scheme 5-2 [(n-Bu)4N]Br & [(n-Bu)4N][Re0Br4] p p. OH HO H4P204-2HC1 2HC1 NH4Tc04 or [Fe(acac)3] [M(P204)(S)] M = Fe, S = H 20 M = Tc, S = EtOH and for [Tc(HP2O4)(EtOH)]EtOH0.5PhMe especially, the effective synthesis is very important for the preparation of a 99mrrc complex for in vivo studies. Work on the 99mTc complex of this ligand, and its radiopharmaceutical chemistry is described in Chapter 6. 125 References (1) Luo, H.; Setyawati, I.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1995, 34, 2287. (2) Burt, R. J.; Chatt, J.; Hussain, W.; Leigh, G. J. J. Organomet. Chem. 1979,182, 203. (3) Yardley, J. P.; Fletcher, H. 3rd. Synthesis 1975, 244. (4) Davison, A.; Trop, H. S.; Depamphilis, B. V.; Jones, A. G. Inorg. Synth. 1982, 21, 160. (5) Rauchfuss, T. B. Inorg. Chem. 1977,16, 2966. (6) Tebby, J. C. In Phosphorus-31 NMR Spectroscopy in Stereochemical Analysis; Verkade, J. G. and Quin, L. D., Eds.; VCH: Deerfield Beach, FL, 1987; p 1. (7) Nugent, W. A.; Mayer, J. M. Metal-LigandMultiple Bonds; John Wiley & Sons: 1988, p 116. (8) Bandoli, G.; Mazzi, U.; Roncari, E.; Deutsch, E. Coord. Chem. Rev. 1982, 44, 191. (9) Clarke, M. J.; Fackler, P. H. Structure and Bonding 1982, 50, 57. (10) Abram, U.; Lorenz, B.; Kaden, L.; Scheller, D. Polyhedron 1988, 7, 285. (11) Abrams, M.; Larsen, S. K.; Shaikh, S. N.; Zubieta, J. Inorg. Chim. Acta 1991,185,1. (12) Chatt, J.; Garforth, J. D.; Johnson, N. P.; Rowe, G. A. /. Chem. Soc. 1964, 1012. (13) Conner, K. A.; Walton, R. A. In Comprehensive Coordination Chemistry; Wilkinson, G., Gillard, R. D. and McCleverty, J. A., Eds.; Pergamon: 1987; Vol. 4; p 125. (14) Forsellini, E.; Casellato, U.; Graziani, R.; Carletti, M. C ; Magon, L. Acta Cryst. 1984, C40, 1795. (15) Kaden, L.; Lorenz, B.; Schmidt, K.; Sprinz, H.; Wahren, M. Isotopenpraxis 1981,17, \1A. (16) Lebuis, A.-M.; Beauchamp, A. L. Can. J. Chem. 1993, 71, 441. (17) Rouschias, G. Chem. Rev. 1974, 74, 531. 126 (18) Nicholson, T.; Davison, A.; Zubieta, J. A.; Chen, Q.; Jones, A. G. Inorg. Chim, Acta 1995, 230, 205. (19) Tisato, F.; Refosco, F.; Moresco, A.; Bandoli, G.; Dolmella, A.; Bolzati, C. Inorg. Chem. 1995, 34, 1779. (20) Refosco, F.; Tisato, F.; Bandoli, G.; Bolzati, C ; Dolmella, A.; Moresco, A.; Nicolini, M. J. Chem. Soc, Dalton Trans. 1993, 605. (21) Parish, R. V. NMR, NQR, EPR, and Mossbauer Spectroscopy in Inorganic Chemistry; 1st ed.; Ellis Horwood: New York, 1990, p 64. (22) Gerdom, L. E.; Baenziger, N. A.; Goff, H. M. Inorg. Chem. 1981, 20, 1606. (23) Lin, G. H. Y.; Leggett, J. D.; Wing, R. M. Acta Cryst. 1973, B29, 1023. (24) Shannon, R. D. Acta Cryst. 1976, A32, 751. (25) Douglas, B. E.; Radanovic, D. J. Coord. Chem. Rew. 1993,128, 139. (26) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; John Wiley & Sons: New York, 1980, appendix. (27) Edwards, D. S.; Liu, S.; Poirier, M. J.; Zhang, Z.; Webb, G. A.; Orvig, C. Inorg. Chem. 1994, 33, 5607. Chapter Six Radiolabeling Functionalized Phosphines with 99mxc and Preliminary Biodistribution Studies of [99mxc(HP204)] 6.1 Introduction Labeling a ligand with 99mjc j s quite different from the synthesis of a 99Tc complex. Isolation of an analytically pure 99nrr/c compound is not possible, and the most important issue in the preparation is the quality. The safe and efficacious use of radiopharmaceuticals is ascertained by quality control.* Radiochemical purity is most significant because radiochemical impurities have biodistribution patterns different from the desired radiopharmaceutical. The radiochemical purity of 99irrr/c_iabeieci compounds is required to satisfy the general criterion of > 95% radiochemical purity, the "reasonable minimum standard." Other concerns regarding quality are the radioassay, sterility, apyrogenicity and toxicity, e t c 2b** 0f the radiopharmaceutical, which is usually a 99mjc complex in solution along with excess ligand, reducing agent, etc. Since radiopharmaceuticals are administered in extremely small amounts, chemical toxicity is not of the same importace that it is in traditional pharmaceuticals.115 In addition, because of the short half-life of 99mTc, the entire preparation for the final solution must be conducted within one or two hours, and preferably within 30 minutes. In practice, the preparation of a radiopharmaceutical is carried out in a "kit", i.e., a vial which contains ligand, reducing agent, and other additives if necessary, to which is added the required Quality control is defined as "a series of tests, observations, and analyses that will indicate beyond reasonable doubt the identity, quality and quantity of all ingredients present in a product and which will demonstrate that the technology employed in its formulation or manufacture will yield a dosage form of the highest possible safety, purity, and efficacy"13 In the hospital nuclear pharmacy, quality control procedures fall into three categories: radiation consideration (radionuclidic purity, radiochemical purity and radioassay), pharmaceutical consideration (pH and chemical purity), and biological consideration (sterility, pyrogens and toxicity).la-2b ** Supposing the radionuclidic purity and chemical purity of the 9 9 mTc04" in the eluate are satisfactory. 128 activity of [99mTc04]- in saline (0.9% NaCl). As mentioned in Chapter One, the choice of ligands and reducing agents is very important to the preparation. In the previous chapters, the coordination chemistry of Tc and Re with three types of ligand systems was studied: bidentate monoprotic (0,0) ligands, Schiff base (NxOy) ligands and functionalized phosphine (PxOy) ligands. Bidentate (0,0) ligands were investigated in a few radiopharmaceutical studies ' and the results showed that cationic tris-ligand complexes, [Tc(L)3]+, were easily synthesized from the reaction of TcCV (using 99nrrc or 99Tc) with an excess of N-substituted-3-hydroxy-4-pyridinones in the presence of a reducing agent such as Na2S2C>5 or Na2S204. In vivo studies of rabbits and mice revealed that several of these cationic complexes might be useful as morphologic kidney imaging agents, with [Tc(pap)3]+ showing elevated uptake in the kidneys even after 24 hours. One of the Schiff bases, H3apa (N3O2), was found to form a stable complex, ["TcO(apa)], from the ligand-exchange route in good yield. Thus H3apa was investigated thoroughly for in vitro radiolabeling with 99mrr,c; however, it was found to be very difficult to adapt this labeling to a "kit". The problem associated with this difficulty is finding the conditions under which the desired product is formed in a short period of time in nanomolar solution. Two new functionalized phosphines (as hydrochloride salts): H2PO2HCI and H4P2O42HCI were synthesized. From these two ligands, neutral complexes, [99TcO(P02)(HP02)] and [99Tc(HP2O4)]-2EtOH0.5PhMe were synthesized rapidly by direct interaction of 99Tc04" with the functionalized phosphines, which function as both reducing and ligating agents (see previous chapters). The successful preparation of these two 99Tc complexes prompted us to investigate the preparation of 99nrr,c analogs and the biodistribution of the 99mTc complexes. In this chapter are presented the results of radiolabeling (using 99mrpc) m e functionalized phosphine ligands (as H2PO2HCI and H4P2O42HCI), and of the preliminary biodistribution studies of the radiolabeled complex [99mXc(HP204)]. Hpap is the abbreviation for 3-hydroxy-l-(p-methoxyphenyl)-2-methyl-4-pyridinone. 129 (Qrh OH H2P02HC1 H4P2042HC1 6.2 Materials and Methods Materials. The phosphine hydrochloride salts H2PO2HCI and H4P2O42HCI were synthesized as described in Chapters 4 and 5, respectively. The [99mTc04]~ in saline (0.9% NaCl) was obtained from a Du Pont 9 9Mo/"mTc generator. Water for labeling was deionized (Barnstead D8902 and D8904 cartridges) and distilled (Corning MP-1 Megapure still). All other chemicals and solvents were obtained from commercial sources and were used as received. Caution! All handling of the y-emitting 99mTc an£j p-emitting " T c compounds was carried out in strict compliance with the UBC "Radionuclide Safety and Methodology" reference manual. Instrumentation. The radioactivity of the segments from developed TLC plates was measured either by a Cobra II (Auto-Gamma) (99mTc) or a Canberra Packard 1900 TR Liquid Scintillation Analyzer (99Tc). The extracted radioactivity of the excised organ in mice was determined by the Cobra II (Auto-Gamma). 99mjc Labeling. 1 mL of a saturated solution of the appropriate phosphine hydrochloride salt in absolute ethanol (0.014 and 0.016 M for H2PO2 HC1 and H4P2O42HCI, respectively) was added to a sterile, pyrogen-free, 10 mL vial. The vial with the solution was purged with nitrogen at a rate of approximately 0.1 LPM for at least 10 min. The desired amount of [99mTcC«4]~ in saline solution (usually 1 mCi; volumes varied from 0.01 to 0.04 mL) was injected into the vial. The syringe was then rinsed with this bulk solution, and the mixture heated in a 100 °C water-bath for 15 min. The vial was cooled and a sample was taken for TLC 130 HC1 •2HC1 evaluation of the radiochemical purity. Chromatography. TLC plates used are Whatman reversed-phase MK Ci8 glass plates, with CH3CN:MeOH:0.5 M NH4OAc:THF = 4:3:2:1 as the eluent. The TLC glass plate (7.6 cm long, 2.5 cm wide) was marked into 10, 12 or 15 roughly equal segments with a pencil, allowing for an extra 1.0 cm piece below the origin and 0.6 cm above the elution line (corrected for in Rf calculations, Figure 6-1). As shown in Figure 6-1, the larger the number of increments, the more accurate the Rf measurement. I O M o I O o 3 o I 5 Solvent front line 6.5 "^ 12.0 H o o o c D B J Origin Figure 6-1. The highest reading of activity in the segments indicated by small circles corresponds to Rf = 6.5/12.0 = 0.54; and radiochemical purity = counts of the three segments/total counts. 99mxc Complexes. After heating, the vial with the labeling mixture was cooled to room temperature, and from it a sample was taken with a 0.5 mL syringe and spotted at the origin. After the plate was developed, each individual segment of the stationary phase was scraped off the plate with a blade and transferred into a test tube which was appropriately numbered. The counts from each individual segment were measured using a y-counter. Subsequently, the Rf values for radioactive species were calculated, and the radiochemical purity of the desired 131 product was calculated based on the counts of the product (in three segments) over the total counts as shown in Figure 6-1. " T c Complexes. The complexes, brown [99TcO(P02)(HP02)] and purple [99Tc(HP2O4)]-2EtOH0.5PhMe were dissolved in DMSO and MeOH, respectively, TLC was run as described above, except that in this instance a liquid scintillation analyzer or visible inspection of the developed spot for [99Tc(HP2O4)]-2EtOH0.5PhMe was used to calculate Rf values. Dilution of the ^mjc Labeled Complexes in Ethanol. When TLC showed the desired labeling, 4.0 mL of distilled water was added to the ethanolic solution to examine the hydrolytic stability of the complexes. White cloudiness, similar to that detained on dilution of saturated ligand solutions with water, was usually observed. Samples were taken for TLC analyses and the results are shown in Figure 6-4. To prepare a final solution (100 uCi in 10.0 mL < 5 % EtOH/H20) of [99mTc(HP204)] complex for in vivo study, the above labeling procedure was followed starting with 1.62 mCi of [99mTcC«4]" and 0.8 mL of the saturated H4P2C«4-2HC1 solution in a two-step dilution procedure, as shown in Figure 6-2. A part (0.6 mL) of the resultant suspension in vial A was withdrawn, filtered, and transferred with a millipore filter attached syringe to vial B which contained 9.4 mL of distilled water. The radiochemical purity of the labeled complex in this final solution was also retained (vide infra). 9.2 mL H20 C ) was added to P C Vial A 0.6 mL was added to C. . ) 9.4mLH20 Q Z Z D p *~\ in Vial B r* ^ Vial A 1.62mCi/0.8mL 100 % EtOH Vial A 1.62mCi/10mL 8 % EtOH VialB O.lmCi/lOmL 0.5 % EtOH Figure 6-2. Dilution scheme of the 99mjc labeled complexes. 132 Biodistribution. Male mice (CD-I/UBC, 37-46 g each) were used for all experiments. All mice were injected with 0.1 mL (1 uCi) of the final [99mTc(HP204)] solution. The mice were anesthetized and sacrificed at 0.25, 0.5, 1, 4, and 24 h post injection. Five animals were used for each time point and the results are expressed as the average percent uptake of the injected dose per gram of tissue. The blood, kidneys, spleen, muscle, stomach, lungs, brain, bone, heart, and liver were excised and the extracted radioactivity was determined using a y-counter. 6.3 Results and Discussion 99mrpc radiopharmaceutical chemistry is different from 99 j c coordination chemistry because of huge variations of concentration and this difference in concentration may influence the rate of formation of a 99mrpc complex greatly if the order of the complex formation reaction is higher than one. The translation from macroscopic 99Tc chemistry to microscopic 99nrr/c chemistry is a step of paramount importace in 99mjc radiopharmaceutical studies. The neutral complexes, [99TcO(P02)(HP02)L and [99Tc(HP204)], were synthesized in high yield by direct interaction of ["TcO^- with the hydrochloride salts of the functionalized phosphines (see Chapters 4 and 5). Despite the fact that the 99Tc chemistry is well described, conditions for the efficacious microsyntheses of the 99m,pc analogs had to be worked out separately. Labeling Conditions. In practice, the concentrations of reactants other than ["mTc04]" are usually in great excess in order to maintain a high reaction rate (vide supra). Conditions which favor rapid formation of a 99mTc complex (i.e., higher temperature, large ligand excess As long as the concentrations of other reagents are all much greater than the initial concentration of [Tc]0, the time necessary for 99% of reaction to occur: t99 = 4.6/k for the pseudo first-order reaction; t99 = 99/k'[Tc]0 for the pseudo second-order reaction, Here k and k' is the conditional rate constant for the two types of reactions, respectively. 133 (L:M > 106:1)), are applied in radiolabeling; therefore, heating in a boiling water-bath and a saturated ligand solution were used in the labeling. An ethanolic solution was used as the labeling medium since [99TcO(P02)(HP02)] and [99Tc(HP204)] were synthesized in ethanol and both the phosphine hydrchloride salts are soluble in ethanol. The toxicity of EtOH may be ameliorated by dilution with water. When using less concentrated solutions of the salts (ca 0.1 mM H4P2O42HCI or 0.2 mM H2PO2HCI), the radiochemical purity dropped consisderably. The ligand solutions were purged with N2 to avoid aerial oxidation of the phosphines or the subsequently formed complexes. Bubbling nitrogen gas through the solvent readily dispels dissolved oxygen within minutes, and such purging is a common practice in radiopharmaceutical kit preparation. 100 T 90 --60 --50-I 1 1 1 1 1 1 0 20 40 60 80 100 120 Time / min Figure 6-3. Labeling yield for the 99mTc-complexes from H2P02HC1 (left) and H4P2O42HCI (right) vs. heating time. The influence of heating time on the labeling yield was examined. To find out the minimum reaction time for the formation of the desired species, samples were taken from the vial at intervals for TLC analysis. It turned out that a longer heating time period did not increase the radiolabeling yield significantly (Figure 6-3, vide infra) in either case, resulting in more counts corresponding to other, unknown species. Although base aids deprotonation of the phosphinophenol ligands, addition of Na2C03 resulted in a higher count at Rf = 0, presumably 134 0 20 40 60 80 100 120 Time / min 99mT/c02 (vide infra). An additional reducing agent, Na2S2C>4, caused lower radiochemical purity in labeling with either phosphine hydrchloride salts. Therefore, a simple procedure, consisting of N2-purging the saturated ligand solution in EtOH and adding the generator-eluate followed by heating for 15 minutes was a simple and rapid radiolabeling procedure amenable to clinical application. With this procedure, the labeling efficiency of [99mTc(HP204)], was consistently very high (> 95%, Rf = 0.43) and reproducible. The major species (Rf = 0.43) and a very minor component (Rf = 0.0) were found in labeling H2PO2, with an overall radiochemical yield of 80 ~ 90%. 3 0 i 2 5 -2 0 -V3 -4-t a 3 O u Total 01 4 J O # 1 0 -5 -0 -[ 1 I i • 1 • M BJ g , - ^- [^ O CO T— 1— Section Number 8 0 -7 0 -6 0 -- 5 0 -c 3 O u 3 4 0 -«4—1 O * 3 0 -2 0 -1 0 -0 -1 - •^- 1^ O CO * - •, _ Section Number Figure 6-4. Radiochromatographic plots of [99mTcO(P02)(HP02)] (left) and [99mTc(Hp2o4)] ( r ight) a f t e r dilution from 100 % EtOH to 20% EtOH in H2O. 135 Dilution. For in vivo biodistribution studies, the prepared nanomolar 99mxc complex is injected, along with its carrier solution, into the bloodstream (i.v.) of an animal. Absolute ethanol, the labeling solvent, is not suitable for i.v. injection; the labeled complex in this solvent must be diluted with water to a maximum of 10% (EtOH % in water, preferably less than 5 %). The complex, therefore, must be hydrolytically stable. The chemical stability of [99mxcO(P02)(HP02)] and [ " mTc(HP204)] in aqueous solution was emphatically demonstrated in the dilution of 99mjc complexes in EtOH with water to 20 % EtOH, as no detectable 99mTc02 or [99nvj/c04]" was found (Figure 6-4); there were no significant changes in the Rf values of the hottest spots or in the pattern of the radiochromatographic plots. This is not surprising, as it is known that the both H2PO2 HC1 and H4P2O42HCI ligands react with macroscopic amounts of pertechnetate easily, forming the corresponding " T c complexes which do not react with water (Chapters 4 and 5). Quality Control. Because the labeling of H2PO2 or H4P2O4 with 99nrr/c was conducted in absolute EtOH at high temperature, the remaining quality concern was radiochemical purity.* The radiochemical purity is usually determined by in vitro analytic methods, such as thin-layer chromatography (TLC), or high performance liquid chromatography (HPLC). The former is the simplest and most rapid method for routine quality control of radiopharmaceuticals. The Rf values for the various species on the strip were calculated based on radioactivity counts, and the Rf values were compared to those of known substances in the same system. The radiochemical purity of desired species was thus determined by the corresponding fraction of the total counts. The reversed-phase Cis glass plate with an eluent of a solvent mixture of CH3CN:MeOH:0.5 M NH40Ac:THF (4:3:2:1) was found to be superior to alumina or silica plates (with various eluents) in distinguishing the complexes and the impurities. The most common impurities in 99mjc labeling are 99mTc04" (free Tc, the unreacted starting material) and 99mTc02 (due to hydrolysis of the reduced 99mjc species). Under the given development conditions, the Rf values of [99inTc04]~ and "mTc02 were found to be 1.0 A sterile and pyrogen-free product is reasonable under the labeling conditions. 136 and 0.0, respectively. 99mTc02 was prepared by mixing [99mTcC>4]-, Na2S2C>4 and NaOH in water, and heating for 25 min. Two Rf values of 0.0 and 0.43 were obtained for [99jc(p02)(HP02)], consistent with NMR results which indicated the presence of two diastereomers (Chapter 4). The two isomers are different in polarity. For [99Tc(HP204)], a value of 0.43, was found by visible inspection or by P-scintillation counter analysis. One spot(Rf = 0.43) was found in the developed TLC plate for [99mxc(HP204)]; the same Rf value was found for p9Tc(HP204)]. Two spots were found when labeling PO2 (Rf = 0.0, 0.43), as for [99TcO(P02)(HP02)]. Although the first (Rf = 0.0) is, unfortunately, indistinguishable from 99111X002 (Rf = 0.0), this spot must be due to one of the two isomers of [99mxcO(P02)(HP02)]. It is known that [99Tc04]- is reduced easily by H 2 P 0 2 and forms [99TcO(P02)(HP02)] in good yield, and that the latter is known not to react with water. Thus, the identities of the complexes of 99Tc and 99mxc 0f b o m the PO2 and P2O4 ligands were verified by TLC. 7 0 n 6 0 -« 5 0 • c 3 U 4 0 • la £ 3 0 • O * 2 0 -10 -0 - — 1 1 1 - CO • 1 1 1 • • 1 PrWrri  1 1 1 1 1 1 1 1 U5 N- O) *~ Section Number • • CO • • 10 Figure 6-5. The radiochromatographic plot of the [99mTc(HP204)] final solution (10 uCi/mL after dilution to 0.5 % EtOH in H2O). 137 Biodistribution. Only in labeling H4P2O4, was the necessary 95% radiochemical purity achieved; therefore only [99mTc(HP204)] was investigated in biodistribution studies. High radiochemical purity of the labeled complex in the final pre-injection solution (100 |j,Ci in 10.0 mL solution of 0.5 % EtOH in H2O) was retained as shown in Figure 6-5. At this stage, [99mTcO(P02)(HP02)] is not suitable for biodistribution studies; however, separation of the two isomers by means of HPLC or other methodologies should be possible due to the difference in their polarities. The tissue distribution data (as percentage of radioactivity per organ and per gram of organ in mice, with the standard deviations) of the complex versus time post injection are listed in Appendix V. The data (per organ) of the complex in mice are shown graphically in Figure 6-6. Data are quoted as per organ (to see overall distribution) and as per gram of organ (to see the specific uptake). The clearance of [99mTc(HP204)] from the bloodstream was found to be rapid after intravenous injection; in 15 min, the level dropped to < 5% of the injected dose. The diagram clearly indicates that approximately 70% of the total injected dose in the form of [99mTc(Hp204)] was localized in the liver after a short period of time (i.e., 15 min or less), and that this elevated level was retained for 4 hours after injection. The level in all other organs or tissues was found to be low (less than 10%; Figure 6-6), with the second highest count found in the muscle. The relatively large count in the muscle might be due to its mass (-45% of a mouse). The low, but significant, level of uptake in organs other than the liver is shown on per gram basis in Figure 6-7. The percent radioactivity per gram in the spleen was second only to the liver, and it declined constantly against time. The uptake in the brain, heart, stomach, and bone is insignificant (Appendix V). Three mechanisms may account for liver localization: particulate agents (radiocolloids) can become entrapped in the liver for a prolonged time (thus permitting the evaluation of liver morphology); lipophilic complexes can be actively cleared from the blood by the hepatocytes (thus allowing an evaluation of hepatobiliary function); and clearance catalyzed decomposition of the agent can lead to it being "stuck" in the liver. ld>9 High deposition in the lungs, where there are very fine capillaries, is usually observed for particulate deposition. On the basis of the 138 Figure 6-6. Tissue distribution diagram of [99mTc(HP204)] in mice (per organ). fact that the radioactivity in the liver decreased after 24 hours and that the lung activity was found to be very low, lipophilicity is believed to be the mechanism for liver uptake. This is attributed to the aromatic rings of the ligand. 139 Figure 6-7. Tissue distribution diagram of [99mTc(HP204)] in mice (per gram -data for liver, stomach, heart, bone, brain omitted). [99mjc(jjp2o4)] as a Hepatobiliary Agent. Criteria for an ideal hepatobiliary agent include: (1) rapid extraction from the plasma by the hepatocytes, (2) effective competition with bilirubin excretion, (3) rapid transit through hepatocytes, (4) high biliary concentration, (5) 140 minimal renal excretion, and (6) ready availability in kit form with high 99mTc-labeling yields.le As shown in Figure 6-8, the ratio of liver/blood uptake remained > 25 beyond 4 h after injection, with a maximum about 1 h after injection. The ratio of liver/kidney remained > 10 for at least 4 h. With the dual functions of tL^C^-as both a ligand precursor and a reducing agent, a distinct advantage of this procedure is its simplicity due to the absence of an added reducing agent.* No other chemicals, such as pH buffer, stabilizer, etc. are required to conduct the labeling. The labeling solvent EtOH itself can function as a bacteriocidal agent. The liver kinetics appeared unique and may represent a different mechanism than that seen for the 99mTc_ labeled iminodiacetic acid (IDA) derivertives.** [99mTc(HP204)] therefore warrants further investigation. Figure 6-8. Liver uptake of [99mTc(HP204)] relative to blood (top) and kidney (bottom). For instance, the most commonly used reducing agent, SnCl2, may hydrolyze. This problem is circumvented by adding enough chelating agent.2d Reducing agent itself is usually in great excess to the radionuclide. 99mTc-labeled IDA derivatives are used to evaluate liver function, and biliary duct patency, and also in cholescintigraphy. The liver activity is seen within a few minutes post-injection, but it is cleared quickly and appears in the gallbladder and then in the intestine. The blood clearance half-time of 99mTc-HIDA (HIDA, or 2,6-dimethylphenylcarbamoylmethyl iminodiacetic acid) is only a few minutes, and urinary excretion is about 15% 90 minutes after injection.2e 141 6.4 Conclusion The "kit" procedures for preparation of the two complexes, [99mTcO(P02)(HP02)] and [99mTc(HP204)] have been developed and evaluated for their potential application in nuclear medicine. The in vitro labeling experiments with H2PO2HCI and H4P2O42HCI showed that the radiolabeled complexes, [99mTcO(P02)(HPC>2)] and [99mTc(HP204)], can be prepared in high radiochemical yield in ethanol within 30 min. The identities of the two " ^ T c complexes were proven to be [99mTcO(P02)(HP02)] and [99mTc(H2P2C>4)], respectively, on the basis of TLC. [99mTcO(P02)(HP02)] is present as two isomers, as in the 99Tc analog. Separation of the two isomers should be performed before in vivo studies of [99mTcO(P02)(HPC>2)] are undertaken. The simple, rapid labeling procedure developed for [99mTc(HP204)], provides a kit amenable for clinical application. The complex, [99mTc(HP2C«4)], showed resistance to hydrolysis. A biological uptake experiment in mice showed that [99mTc(HP2C<4)] has a high affinity for the liver, and this result is worthy of further investigation into the use of [99mTc(HP204)] as a potential liver function agent. 142 References (1) Kowalsky, R. J.; Perry, J. R. Radiopharmaceuticals in Nuclear Medicine Practice; Appleton & Lange: Norwalk, 1987, a. pl23; b. p 7; c. p 83; d. p 271; e. p 290. (2) Saha, G. B. Fundamentals of Nuclear Pharmacy; 3rd ed.; Springer-Verlag: New York, 1992, a. p 146; b. p 158; c. p 259; d. p 98; e. p. 253. (3) Krohn, K. A.; Jansholt, A.-L. Int. J. Appl. Radiat. hot. 1977, 28, 925. (4) Deutsch, E.; Libson, K. Comments Inorg. Chem. 1984, 3, 83. (5) Edwards, D. S.; Liu, S.; Lyster, D. M.; Poirier, M. J.; Vo, C ; Webb, G. A.; Zhang, Z.; Orvig, C. Nucl. Med. Biol. 1993, 20, 857. (6) Kanvinde, M. H.; Basmadjian, G. P.; Mills, S. L.; Kale, N. J. J. Nucl. Med. 1990, 31, 908 (abstract). (7) Edwards, D. S.; Liu, S.; Poirier, M. J.; Zhang, Z.; Webb, G. A.; Orvig, C. Inorg. Chem. 1994,33, 5607. (8) Liu, S.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30,4915. (9) Anderson, C. J.; John, C. S.; Li, Y. J.; Hancock, R. D.; McCarthy, T. J.; Martell, A. E.; Welch, M. J. Nucl. Med. Biol. 1995,22, 165. 143 Chapter Seven General Conclusions and Suggestions for Future Work 7.1 General Conclusions The objectives of this project were the investigation of the coordination chemistry of rhenium and technetium via the synthesis and characterization of new complexes of the two metals on the macroscopic scale, and, proceeding from this, the application of the new chemistry in the preparation of 99nrr;c complexes on the microscopic scale followed by in vivo biodistribution studies. These goals have been achieved. The methodology applied to this project proved correct. A series of rhenium(V) and technetium(V) complexes of bidentate monoprotic (0,0) ligands, ma", and dpp" were synthesized and characterized. Some of these complexes hydrolyze in water. Studies with multidentate Schiff base (N2O2) ligands showed that metal(V) complexes could be formed in low yields although hydrolysis of the ligands hindered further investigation in radiopharmaceutical studies. One pentadentate Schiff base (N3O2) ligand, H3apa, formed a stable metal(V) complex in good yield in macroscopic amounts; however, translation of this chemistry to a nanomolar scale was unsuccessful because of difficulties in defining suitable reaction conditions for labeling. Phosphines functionalized with two or more phenolic groups may provide a comprehensive solution to the development of new 99mrpc complexes, since they possess both chelating and reducing abilities, in addition to their easy modifiability. The hydrochloride salts of the potentially tridentate ligand precursor bis(o-hydroxyphenyl)phenylphosphine (abbreviated H2PO2HCI) and of the potentially hexadentate ligand precursor P,P,P',P'-tetrakis(o-hydroxyphenyl)diphosphinoethane (abbreviated H4P2O42HCI) have been 144 developed, and upon deprotonation, they function as both ligands and reducing agents simultaneously. Their properties are favorable for rapidly forming Re and Tc complexes via reduction/coordination, and for stabilizing metal centers in oxidation states of five or three. H4P2O42HCI was rapidly labeled with 99mjc a nd m e 99mTc_compiex w a s found t 0 have high liver uptake in mice. The design and synthesis of new ligand precursors, the macroscopic chemistry of Tc and Re complexes with characterization by physical techniques, and in vivo studies have been accomplished in this project. 7.2 Suggestions for Future Work PO2 System. One study that remains to be done is the separation of the two diastereomers of [99mrrco(P02)(HP02)], for further in vivo studies of this complex. The two isomers have different Rf values on a Cis reversed phase plate, indicating that the two isomers can be separated by chromatographic methods. There may be a way to avoid the formation of the above two isomers by using an appropriate mixture of HPO and HPO2 for labeling. Presumably, conditions which favor rapid formation of [99mrrco(PO)(P02)] could be ultimately set down. If good labeling were accomplished, the biodistribution of the complex could be tuned by modification of the phenyl and/or phenolic rings of the ligands. Investigation of a complex such as [Tc(NR)(PO)(P02)] (R = alkyl or aryl group) provides more possibilities for this type of complex for radiopharmaceutical purposes.1'2 P2O4 System. The rapid, high yield labeling of 99mjc t0 H4P2O4 is only the first step in the rational design of a radiopharmaceutical. To achieve targetted brain or heart imaging, further functionalization of the phenyl rings must be done. The lipophilicity can be modified by introducing water soluble functional groups, such as -SC^Na, -NMe3+X_, etc. to the phenyl rings.3 145 The introduction of biologically significant groups is a more active approach to rational design. This could be done by borrowing from pharmacology, where distributions of the drugs with known structures are studied. For instance, it is known that nitroimidazoles tend to localize in hypoxic regions in the body, including tumors. Some ligands suitable for binding Tc have been linked to these organic compounds.4'5 There are also possibilities to increase the cavity size of the P2O4 ligand in order to obtain fully bound complexes, M(P204) or [M(P2C>4)]". One possible way to do this is to use ortho carboxylate functional groups instead of ortho hydroxy groups: the chelate rings expand from five to six membered, and the new P2O4 ligand would be more easily deprotonated. An alternative might be to expand the backbone by using a propylene instead of an ethylene backbone. New PxOy Systems. Based on the ligand systems presented here, HPO, H2PO2, and H4P2O4, new ligands like H3P2O3, H2P2O2 and HP2O could also be made. While the H3P2O3 ligand, presumably, might form neutral complexes of M=03 + cores, the HP2O ligand might be useful for forming cationic metal(III) complexes. The H2P2O2 ligand might form tetrahedral complexes (with transition metal ions or even some main group metals). Again, ligands with both soft and hard donors would form very stable complexes, which might find applications in hydrometallurgy (in extraction or electroplating), or even in material sciences.6 With chiral functionalized phosphines, applications might be found in asymmetric hydrogenation {e.g. chiraphos, dipamp, etc)? (o-C6H4-OH)PhP^^ P(o-C6H4-OH)2 P h 2 P ^ ^ P(o-C6H4-OH)2 Ph2p'"~^ PPh(o-C6H4-OH) H3P2O3 H 2 P 2 0 2 H P 2 0 Although the PxOy systems were developed originally for Re and Tc chemistry because of their reducing and chelating abilities, their potential utility is not necessarily limited to these two metals. For instance, the combination of hard and soft donors makes it possible for them to form complexes with various other transition metals, as well as some post-transition metals.8 146 References (1) Rochon, F., D.; Melanson, R.; Kong, P.-C. Inorg. Chem. 1995, 34, 2273. (2) Nicholson, T.; Davison, A.; Zubieta, J. A.; Chen, Q.; Jones, A. G. Inorg. Chim. Acta 1995,230, 205. (3) Herrmann, W. A.; Kohlpaintner, C. W. Angew Chem. Int. Ed. Engl. 1993, 32, 1524. (4) Linder, K. E.; Chan, Y.-W.; Cyr, J. E.; Malley, M. F.; Nowotnik, D. P.; Nunn, A. D. J. Med. Chem. 1994, 37, 9. (5) Raju, N.; Ramalingam, K.; Nowotnik, D. P. Tetrahedron 1992, 48, 10233. (6) Dagani, R. In Chemical and Engineering News; Oct. 3,1994; p 31. (7) Kagan, H. B. In Comprehensive Organometallic Chemistry; G. Wilkinson, F. G. A. Stone and E. W. Abel, Ed.; Pergamon: Oxford, 1982; Vol. 8; p 463. (8) Fryzuk, M. D. Can. J. Chem. 1992, 70, 2839. 147 m Appendices Appendix I. Table 1-1. Selected Crystallographic Data for [ReOBr(ma)2], [(n-Bu)4N][ReOBr3(ma)], and [(n-Bu)4N][TcOCl3(ma)]. Complex formula fw cryst syst space group a, A b, A c, A a, deg Meg y, deg y,A3 z pc, g/cm3 T, °C radiation (k, A) \i, cm-1 Trans, factors R Rw [ReOBr(ma)2] Ci2HioBr07Re 532.32 triclinic PI 20.846 (2) 8.484 (2) 8.4809 (8) 85.89 (1) 89.831 (9) 89.80(1) 1496.0 (4) 4 2.363 21+1 C u ^ a (1.54178) 187.63 0.47-1.00 0.043 0.050 [(n-Bu)4N] [ReOBr3(ma)] C22H4iBr3N04Re 809.49 monoclinic Pllln 11.074(3) 15.713 (4) 17.370 (4) 90 104.77 (2) 90 2923 (2) 4 1.839 21+1 Mo Ka (0.71069) 83.01 0.48-1.00 0.033 0.027 [(n-Bu)4N] [TcOCl3(ma)] C22H4iCl3N04Tc 586.93 monoclinic P2\lc 10.724 (2) 11.327(2) 23.651 (2) 90 101.46(1) 90 2815.6(8) 4 1.384 21±1 Mo Ka (0.71069) 8.02 0.84-1.00 0.034 0.030 149 Table 1-2. Selected Final Atomic Coordinates (fractional) and Equivalent Isotropic Thermal Parameters Beq (A2)a for [ReOBr(ma)2],b [n-Bu4N][ReOBr3(ma)], and [n-Bu4N][TcOCl3(ma)]. atom Re(l) Br(l) 0(1) 0(3) 0(4) 0(6) 0(7) Re(l) 0(1) 0(3) 0(4) Br(l) Br(2) Br(3) Tc(l) O(l) 0(3) 0(4) Cl(l) Cl(2) Cl(3) X 0.86545 (2) 0.94249 (6) 0.8890 (4) 0.8474 (3) 0.9354 (3) 0.7910 (3) 0.7935 (3) [n 0.35701 (1) 0.2851 (5) 0.3164(4) 0.4375 (4) 0.56647 (9) 0.16227(9) 0.42468 (9) [n 0.28086 (3) 0.3664 (2) 0.2149 (2) 0.4087 (2) 0.41344 (8) 0.12549(8) 0.1389(1) y [ReOBr(ma)2] 0.25710(6) 0.4323 (2) 0.303 (1) 0.1542(8) 0.076 (1) 0.391 (1) 0.1095(9) -Bu4N][ReOBr3(ma)] 0.20849 (2) 0.1781 (3) 0.0898 (3) 0.2158 (3) 0.14743 (6) 0.26506 (6) 0.35516(6) -Bu4N][TcOCl3(ma)] 0.09103 (2) 0.1656(2) 0.0221 (2) 0.1592(2) -0.08075 (8) -0.00110(8) 0.24987 (8) z 0.39283 (6) 0.2350 (2) 0.572 (1) 0.1890(9) 0.389(1) 0.3359 (9) 0.4807 (9) 0.15185(2) 0.0598 (3) 0.1952(3) 0.2708 (3) 0.14407(6) 0.18490(6) 0.13237(5) 0.15172(1) 0.20554 (9) 0.07274 (9) 0.1001 (1) 0.16973(4) 0.19103 (4) 0.12044(4) #eq 3.03 (2) 4.84 (7) 4.4 (4) 3.4 (3) 3.9 (4) 4.0 (4) 3.8 (4) 3.30(1) 5.0 (3) 4.0 (2) 4.0 (2) 5.90 (5) 5.64 (5) 5.14(4) 3.50(1) 4.8(1) 4.2(1) 4.7 (1) 5.46 (4) 5.16(4) 5.43 (4) af l e q = (8/3)7t2ZEUijai*aj*(ai-aj) b For one of the two independent molecules. See text. 150 Appendix II Table II-l. Selected Crystallographic Data for [Re(dha)Cl2(OPPh3)(PPli3)]EtOH. complex [Re(dha)Cl2(OPPh3)(PPh3)]EtOH formula fw crystal system space group a, A b, A c, A a, deg P,deg y, deg V, A3 Z Pcfl/c g / c m 3 T, °C radiation Jt,A \i, cm-1 No. of reflections No. of variables R Rw C46H43Cl206P2Re 1010.90 monoclinic P2\/c 11.064(2) 24.167(2) 16.6687 (8) 90 98.779 (7) 90 4405.0 (6) 4 1.524 21 Mo 0.71069 30.00 13144 594 0.030 0.025 R = HIF0I-IFCII/EIF0I, Rw = (Zw(\F0\-\Fc\)2rLw\F0\2)m 151 Appendix III. Table III-l. Selected Crystallographic Data for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2). complex formula fw crystal system space group a, A b,k c,k a, deg P,deg y, deg V,A* z Pc«/c> g / c m 3 T, °C radiation I, A |i, cm-1 transm factors reflections / > 3a(/) No. of variables R Rw [Re(NPh)Cl(PPh3)(P02)]-2CHCl3(l) C44H35Cl7N02P2Re 1106.09 triclinic PT 11.997(1) 20.637(2) 10.511(1) 103.772(9) 113.504(8) 76.072(9) 2286.6(5) 2 1.606 21 Mo 0.71069 31.74 0.59-1.00 8203 541 0.034 0.034 [ReO(PO)(P02)] (2) C36H2704P2Re 771.76 monoclinic Flxln 10.132(2) 14.026(3) 22.046(2) 90 102.38(1) 90 3060.2(9) 4 1.675 21 Mo 0.71069 41.65 0.58-1.00 5957 388 0.030 0.026 R = IIIF0I-IFCII/X1F0I, Rw = (Zw(IF0l-IFcl)2/EwF0l2)1/2 152 Table III-2. Final Atomic Coordinates (Fractional) and Equivalent Isotropic Thermal Parameters Beq (A2)a for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2). atom x _y z #§q °_cc Re(l) Cl(l) Cl(2) Cl(3) Cl(4) Cl(5) Cl(6) Cl(7) Cl(8) Cl(9) Cl(10) P(l) P(2) O(l) 0(2) N(l) C(l) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(ll) C(12) C(13) C(14) C(15) 0.273264(15) 0.22275(11) 0.3020(2) 0.2094(2) 0.0432(2) -0.1117(6) -0.1795(3) -0.1731(6) -0.0980(9) -0.1963(9) -0.142(2) 0.24326(10) 0.38550(10) 0.1390(3) 0.1350(2) 0.4067(3) 0.1419(4) 0.0978(4) 0.0061(4) -0.0364(5) 0.0094(5) 0.0972(4) 0.1506(4) 0.1036(4) 0.0237(4) -0.0080(5) 0.0371(5) 0.1143(4) 0.3593(4) 0.4126(5) 0.5053(5) 0.238598(9) 0.24661(6) 0.06498(12) 0.01042(12) 0.08172(15) 0.3680(3) 0.2925(2) 0.2386(3) 0.2742(6) 0.2331(7) 0.3642(7) 0.22997(6) 0.33377(6) 0.1787(2) 0.31780(14) 0.1795(2) 0.1680(2) 0.1541(2) 0.1125(2) 0.0861(3) 0.0997(3) 0.1408(2) 0.3086(2) 0.3438(2) 0.4048(2) 0.4302(2) 0.3958(3) 0.3347(3) 0.2034(2) 0.1353(3) 0.1131(4) 153 0.37528(2) 0.13299(12) 0.9596(2) 1.1130(3) 0.8919(2) 0.0601(8) 0.1940(4) -0.0744(6) -0.1170(9) 0.0443(14) 0.096(2) 0.58151(12) 0.44240(11) 0.2849(3) 0.3876(3) 0.4100(4) 0.5089(5) 0.3621(5) 0.2933(5) 0.3707(7) 0.5158(6) 0.5858(5) 0.6151(4) 0.5005(5) 0.5058(5) 0.6220(6) 0.7354(6) 0.7315(5) 0.7419(5) 0.7357(6) 0.8532(8) 3.085(7) 4.64(5) 10.3(1) 13.0(2) 12.5(1) 14.0(3) 0.70 11.1(2) 0.70 20.0(4) 0.70 12.4(6) 0.30 14.0(7) 0.30 19(1) 0.30 3.33(4) 3.35(4) 4.2(1) 3.7(1) 3.5(1) 3.6(2) 3.9(2) 4.7(2) 5.5(2) 5.4(3) 4.7(2) 3.6(2) 3.6(2) 4.6(2) 5.4(2) 5.2(2) 4.6(2) 4.0(2) 5.3(2) 7.2(3) en O O O O O O O O O O O O O O O O O O O O O O O O O O O O O £ '* 2 os U> * t O H - O s o O O ^ ] O s < x 4 ^ U > t O H - ' O S O O O ~ J O \ < X 4 i - U > t O H - O S O O O - J O s G o p p 5 o I—> 00 as ON tX o tX 4^ as so ix o Os IX -o 00 os o <i u> 00 tx IX o ^1 h—* l—» o tX o as o o (X •t± o <x h—* -J IX 4^ o IX 00 H ^ SO 4^ o 1^ o 4^ 4^ Os p ^1 so to o <X o ^) Os o -o Os o as U) so u> IX o Ut 4^ 00 Os -li. o 4* •—» H-l 00 Lh o U> Os Os Os -J o tO 4^ 4^ t—L -J p p p p p p H - tO U) UJ U> U> Os O U> H -U> tX t— - J 0 0 OS U) v o p 4* o 4^ p p 0 0 (X '4^ '4^ l x Os ^O os to u> oo so so U> 00 Os O 00 4*. - J o /—\ Os -J ~o IX U> *>• X ^ ^s ^ o <X OS OS N> to o o p p o p o p o p o p o p p o p o p p p p p N i O O O O ^ ^ ^ ^ ^ k > U i l o L o 4 i - 4 ^ 4 i . 4 ^ 4 ^ U ) 4 ^ 0 0 ^ J 0 0 U > < X O 4 ^ - U > 0 , l U > < | H - . p > . O t O < l 0 0 < X O S 0 < X U i ^ i - ' SO i — ' O s s 0 O s 0 0 V 0 ^ - ' S 0 O 0 0 t O O 4 i . U > < l H - 0 0 O s S O t O t O O ^ S O O s U > O U > s O ~ J O S ^ I 0 4 i . < X U > t O ^ ) A U U U W U W I s ) U U U l U i U l | O U U U U N ) K ) U U ^ ^ U I O U ^ U l p IX o H ^ (X o 4*. oo O U) o 4^ •—> <X u> o U> ^1 H ^ u> o OJ SO u> o o to 4^ -J •~J o to to 1>J OS o 1—k tx IX SO O o ON SO i—> 00 K -o to 4^ . Os Os P W as CX 4*. as o 4^ O 1^ SO P IX 4^ ^1 vO s—\ SO O as u> ^i ^ ^i o IX so to as •—\ Os O 4*. tx 00 t—k /—\ (X o u> ^1 H-t ^i Os O OJ 00 4^ o ^1 o 4*. SO O ^1 SO O tX 00 4^ 4^ SO O IX ^1 as o Os o 4^ Os 00 O IX o to SO tx 4^ Os O to o OJ SO oo O t — * to oo O ^1 O >—* 4^ UJ ex Os O to U) IX Os (X o OJ (—* OJ I—* 4^ o tx SO UJ SO IX o J^ H-* IX •—" <—-. ^ O oo 4^ O ^1 / - • ^ ^1 o 00 (X 1—> O Os O 1^ to SO LO Os O Os O O O Lfi o oo Os IX Os IX o SO 00 to IX Os p SO ~J 4^ <x 00 b i ^ i o b o w a M b b b o i i j ^ b i x i i o ^ ^ ^ b o ^ a u u i b b i o w b Q ^ W O J W ^ ^ ^ 1 O W ^ 1 X W ^ O J / 4 ^ / S O J to ' to to U> 4^ - 4^ . u> to to £>• 4^ . to ft O o o cr H to a* to s 0 ra H ^ S5 CD cr 03 3 I. £ O to g to o o o >-* CL t — » • 13 rs tO > ° , - v o j>> gi ~ 5?> 2 . ^ o o CD O o o to to g w c ^-CD -C3 o O o' Table III-2 (cont'd). Final Atomic Coordinates (Fractional) and Equivalent Isotropic Thermal Parameters Bcq (A2)a for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2). atom JC [ReO(PO)(P02)] (2) Re(l) P(l) P(2) O(l) 0(2) 0(3) 0(4) C(l) C(2) C(3) C(4) C(5) C(6) C(7) C(8) C(9) C(10) C(ll) C(12) C(13) C(14) C(15) C(16) C(17) C(18) C(19) C(20) C(21) C(22) 0.02119(2) 0.08785(11) 0.22266(10) -0.0229(3) -0.1510(3) 0.1101(3) -0.0459(3) -0.0767(4) -0.1773(4) -0.3107(5) -0.3412(5) -0.2431(5) -0.1112(5) 0.1766(4) 0.1758(4) 0.2406(5) 0.3020(5) 0.3008(5) 0.2377(5) 0.1790(4) 0.3154(5) 0.3845(6) 0.3196(7) 0.1860(7) 0.1166(5) 0.1616(4) 0.0298(4) -0.0289(4) 0.0445(5) y 0.235071(11) 0.38014(7) 0.16590(7) 0.2963(2) 0.2625(2) 0.1876(2) 0.1017(2) 0.4237(3) 0.3522(3) 0.3780(3) 0.4699(4) 0.5399(3) 0.5167(3) 0.3401(3) 0.2396(3) 0.1980(3) 0.2525(4) 0.3518(4) 0.3941(3) 0.4749(3) 0.4911(3) 0.5610(4) 0.6119(4) 0.5951(4) 0.5268(3) 0.0450(3) 0.0309(3) -0.0599(3) -0.1339(3) z 0.379712(8) 0.33784(5) 0.44439(5) 0.43847(13) 0.31409(13) 0.31139(12) 0.38152(13) 0.3014(2) 0.2944(2) 0.2669(2) 0.2484(2) 0.2557(2) 0.2817(2) 0.2809(2) 0.2763(2) 0.2333(2) 0.1962(2) 0.1990(2) 0.2417(2) 0.3839(2) 0.3852(2) 0.4249(3) 0.4618(3) 0.4619(3) 0.4233(2) 0.4468(2) 0.4131(2) 0.4113(2) 0.4424(2) #eq 2.290(5) 2.31(4) 2.26(4) 3.3(1) 3.3(1) 2.6(1) 3.0(1) 2.6(2) 2.8(2) 4.0(2) 4.2(2) 3.8(2) 3.3(2) 2.5(2) 2.7(2) 3.8(2) 4.4(2) 3.9(2) 3.2(2) 2.8(2) 3.8(2) 5.5(3) 6.5(3) 6.0(3) 4.1(2) 2.5(2) 2.5(2) 3.2(2) 3.8(2) 155 Table III-2 (cont'd). Final Atomic Coordinates (Fractional) and Equivalent Isotropic Thermal Parameters 5 e q (A2)a for [Re(NPh)Cl(PPh3)(P02)]-2CHCl3 (1) and [ReO(PO)(P02)] (2). atom C(23) C(24) C(25) C(26) C(27) C(28) C(29) C(30) C(31) C(32) C(33) C(34) C(35) C(36) X 0.1747(5) 0.2335(5) 0.3726(4) 0.4498(5) 0.5652(5) 0.6030(5) 0.5260(5) 0.4100(5) 0.2834(4) 0.2339(5) 0.2775(6) 0.3713(6) 0.4235(7) 0.3833(6) y -0.1207(3) -0.0311(3) 0.1600(3) 0.2430(3) 0.2404(4) 0.1583(4) 0.0793(4) 0.0784(3) 0.2116(3) 0.2951(4) 0.3282(5) 0.2791(4) 0.1989(4) 0.1649(4) z 0.4762(2) 0.4788(2) 0.4127(2) 0.4142(2) 0.3884(2) 0.3630(2) 0.3613(2) 0.3854(2) 0.5229(2) 0.5422(2) 0.6019(3) 0.6429(2) 0.6246(3) 0.5643(2) #eq 4.1(2) 3.5(2) 2.6(2) 3.8(2) 4.7(2) 4.6(2) 4.2(2) 3.5(2) 2.7(2) 5.0(2) 6.4(3) 5.4(3) 6.7(3) 5.6(3) a 5 e q = (8/3)7C2(Un(aa*)2 + U22(bb*)2 + U33(cc*)2 + 2Ui2aa*bb*cosy+ 2Ui3aa*cc*cosp + 2U23bb*cc*cosa) b occ = occupancy 156 Appendix IV. * • • • i 2.0 1.0 0.0 -10 -2.0 EIV vs.Ag/AgQ Figure IV-1. Cyclic voltammograms a.0.1MTEAPinCH3CN; b. 0.1 M TEAP in CH3CN, with 3 drops of (cone.) HCIO4 added; c. [Tc(HP204)(EtOH)J in 0.1 M TEAP in CH3CN; d. [Tc(HP204)(EtOH)] in 0.1 M TEAP in CH3CN, with 3 drops of (cone.) HCIO4 added; e. [Fe(HP204)(H20)] in 0.1 M TEAP in CH3CN. 157 Table IV-l. Selected Crystallographic Data for a partially oxidized trans-[ReC\4(PP^3)2]-complex formula fw crystal system space group a, A b,k c, A a, deg P,deg y, deg y,A3 z Pca/c g / c m 3 T, °C radiation X,A |i, cm"1 reflections / > 3a(7) No. of variables R Rw partially oxidizedtrans- [ReCl4(PPb.3)2] C36H30Cl3.860o.i4P2Re 849.80 triclinic PT(#2) 14.204 (2) 16.381 (3) 11.878(2) 96.13 (1) 92.41 (1) 64.94 (1) 2489.2 (7) 3 1.701 21 Mo 0.71069 40.96 8572 594 0.032 0.025 R = HIF0I-IFCII/HF0I, Rw = (Zw(IF0l-IFcl)2/IwlF0l2)1/2 158 Figure IV-2. ORTEP drawing of the molecule in a general position in the asymmetric unit* of the partially oxidized fra/w-[ReCl4(PPh3>2]. The asymmetric unit contains 1.5 molecules - one in a general position and the other at a center of symmetry. A single CI site of the molecule in the general position is partially occupied by an oxo ligand. 159 C41 Figure IV-3. ORTEP drawing of the molecule in a center of symmetry in the asymmetric unit of the partially oxidized trans-[ReCU(PPhj)2]-160 Table IV-2. Selected Bond Lengths (A) for two molecules of the partially oxidized [ReCl4(PPh3)2]: (1) at a general position; (2) at a center of symmetry. Re(l)-Cl(l) Re(l)-Cl(2) Re(l)-Cl(3) Re(l)-Cl(4) Re(D-Pd) Re(l)-P(2) Re(l)-0(1) P(l)-C(l) P(l)-C(7) P(l)-C(13) P(2)-C(19) P(2)-C(25) P(2)-C(31) (1) 2.331(1) 2.339(1) 2.291(3) 2.338(1) 2.580(1) 2.568(1) 1.71(3) 1.834(4) 1.832(5) 1.839(5) 1.828(5) 1.830(4) 1.822(5) Re(2)-Cl(5) Re(2)-Cl(6) Re(2)-P(3) P(3)-C(37) P(3)-C(43) P(3)-C(49) (2) 2.334(1) 2.301(1) 2.564(1) 1.828(5) 1.822(5) 1.833(5) 161 Table IV-3. Selected Bond Angles (deg) for for two molecules of the partially oxidized [ReCl4(PPh3)2]: (1) at a general position; (2) at a center of symmetry. Cl(l)-Re(l)-Cl(2) Cl(l)-Re(l)-Cl(3) Cl(l)-Re(l)-Cl(4) Cl(l)-Re(l)-P(l) Cl(l)-Re(l)-P(2) Cl(l)-Re(l)-0(1) Cl(2)-Re(l)-Cl(3) Cl(2)-Re(l)-Cl(4) Cl(2)-Re(l)-P(l) Cl(2)-Re(l)-P(2) Cl(l)-Re(l)-0(1) Cl(3)-Re(l)-Cl(4) Cl(3)-Re(l)-P(l) Cl(3)-Re(l)-P(2) Cl(4)-Re(l)-P(l) Cl(4)-Re(l)-P(2) Cl(4)-Re(l)-0(1) P(l)-Re(l)-P(2) P(l)-Re(l)-0(1) P(2)-Re(l)-0(1) (1) 88.43(5) 177.69(8) 88.51(5) 88.59(4) 92.22(4) 174.1(7) 93.86(7) 176.54(5) 96.88(4) 84.33(4) 86.5(7) 89.20(7) 90.90(6) 88.24(6) 84.67(4) 94.17(4) 96.7(7) 178.56(4) 89.2(6) 90.1(6) Cl(5)-Re(l)-Cl(5)* Cl(5)-Re(l)-Cl(6) Cl(5)-Re(l)-Cl(6)* Cl(5)-Re(l)-P(3) Cl(5)-Re(l)-P(3)* Cl(6)-Re(l)-Cl(6)* Cl(6)-Re(l)-P(3) Cl(6)-Re(l)-P(3)* P(3)-Re(l)-P(3)* (2) 180 91.34(5) 88.66(5) 93.50(4) 86.50(4) 180 93.02(4) 86.98(4) 180 162 Appendix V. Table V-1. The tissue distribution of the complex in mice (per organ, in % of injected dose). Organ Blood Liver Kidney Spleen Muscle Stomach Lungs Heart Bone Brain 15min 3.66 (0.66) 67.98 (6.01) 1.71 (0.26) 1.56(0.47) 7.24(1.99) 0.11(0.02) 0.98(0.16) 0.41 (0.08) 0.05 (0.02) 0.00 (0.00) 30min 2.61 (0.29) 73.65 (28.23) 1.91 (0.61) 1.10(0.59) 7.00(3.10) 0.14(0.06) 0.66(0.13) 0.39(0.12) 0.03 (0.00) 0.01 (0.01) 60min 1.29(0.09) 49.10(6.42) 1.25(0.13) 0.66(0.17) 5.20(1.23) 0.10(0.02) 0.38 (0.05) 0.23 (0.03) 0.03 (0.01) 0.01 (0.01) 4 h 3.25 (3.08) 74.91 (33.57) 1.74(1.14) 0.46(0.17) 7.98 (3.90) 0.30 (0.34) 0.37 (0.27) 0.29 (0.24) 0.35 (0.52) 0.12(0.14) 24 h 2.44 (2.46) 18.02 (2.61) 0.70(0.31) 0.33 (0.14) 8.90 (4.76) 0.19(0.11) 0.23(0.15) 0.19(0.08) 0.16(0.11) 0.10(0.08) Table V-2. The tissue distribution of the complex in mice (per gram, in % injected dose). Organ Blood Liver Kidney Spleen Muscle Stomach Lungs Heart Bone Brain 15 min 1.16(0.17) 31.51 (5.29) 2.26 (0.79) 12.95 (2.73) 0.40(0.11) 0.42(0.11) 4.41 (1.06) 2.47 (0.62) 0.68 (0.44) 0.01 (0.01) 30 min 0.86(0.11) 34.28 (13.51) 2.95 (0.94) 9.10(4.95) 0.40(0.18) 0.64 (0.25) 2.74 (0.48) 2.19(0.72) 0.41 (0.09) 0.02 (0.03) 60 min 0.43 (0.02) 24.26(1.86) 1.83 (0.23) 5.29(1.55) 0.30 (0.07) 0.41 (0.10) 1.12(0.17) 1.09(0.21) 0.26 (0.07) 0.03 (0.02) 4h 1.01 (0.97) 34.38 (14.97) 2.56(1.59) 3.66 (1.29) 0.43 (0.22) 0.98 (1.00) 1.48(1.13) 1.50(1.23) 3.35(5.13) 0.22 (0.26) 24 h 0.74 (0.75) 8.02 (1.08) 0.98 (0.25) 2.54 (1.38) 0.47 (0.22) 0.66 (0.44) 0.91 (0.73) 0.99 (0.39) 1.61(1.14) 0.21 (0.17) 163 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059490/manifest

Comment

Related Items