LIBRARY Michigan State nivarsity PLACE iN RETURN BOX to remove this ohookout from your record. TO AVOID FINES Mum on or before date duo. DATE DUE DATE DUE DATE DUE —‘ ¥ ‘ ¥ ‘ * * MSU I. An Affirmative Action/Equal Opponmiiy institution CWMMI THE DESIGN OF mm mm PHOSPHATES w Young-00mm A DISSERTATION Submitted“ Michigan State University in partial fulfillment of the requirements {orthodoyooof DOCTORG‘W Department of Chemistry 1991 mMOFWmmm w Young-comm Layered Integrated Photochemical Systems can be designed by introducing multielectron photoactive centers into a redox-active layered materials. Singly bonded dirhodium and quadruply bonded dimolybdenum complexes are employed in the intercalation reactions of layered metal phosphates. Based .on the studies of multielecton reactivity from my“ and M02“, bimetallic centers. a strategy for the synthesis of multielectron LIPS schemes is explored. Whereas quadruply bonded dimers are quite well understood. a paucity of information on excited state properties of singly bonded dimers necessitated the photophysical study of these complexes. Although the usual decay pathway of 0‘ type excited states is dissociative. the lowest energy state of a 3(1‘0‘) spin-orbit coupling manifold may provide a channel for radiative decay. A model to explain the emission dynamics from the 3(x‘o‘) state is constructed and generalized based on the temperature dependence of the radiative and nonradiative rates of KnPt2(HPO4)4L2 (n = 2. L 2 H20; n a: 4. L = Cl. Br) in the D41, point group and for a series of dirhodium fluorophosphine derivatives of lower symmetry point groups. Intercalation reactions of acetonitrile solvated ha 4+ or M02“ cores into layered vanadium or niobium phosphates produce interesting new layered materials. Structural characterizations including ESCA. X-ray powder difl'raction. and absorption spectroscopy of these materials show that the bimetallic cores inside the layer are in a transverse arrangement. The EPR and magnetic susceptibility studies show the electrons in the layers are antiferromagnetically coupled and the exchange interaction is in weak. coupling regime. However. the paramagnetic guest molecules show evidence for 3- dimensional coupling. The LIPS strategy is further elaborated with methods to introduce desirable coordination environments into the layers of phosphates. The study reported herein may provide a glimpse of the new generation of multielectron photoactive solid materials. tomyfamily ACKNOWLEDGMENTS Dr. Daniel G. Nocera deserves great appreciation for his inspiration and encouragement throughout my graduate study. I shall remember that each and every step of this work has been guided by his enthusiasm and scientific ingenuity. The greatest gratitude of mine is to my wonderful husband Uhyon and my little girl Mina who have endured much more than their share of trouble in all my hard times, especially while this dissertation was being prepared. Last, but not least, I wish to say many thanks to my 'friends' who have allowed me to be myself. Tam OF CONTENTS Page list of Tables ............................................................................ Ix List of Figures ........................................................................... xi List of Abbreviation ................................................................... xxii I. Introduction .................................................................... 1 II. Luminescence from Singly Bonded Bimetallic Complexes: Dynamics of o‘ Ercited State ........................................... 25 A. Background ........................................................... 25 B. Experimental ............... 29 l . Synthesis of Compounds .............................. 29 a. K2H20W4J4m20)2 ............................ 29 b. K4Pt2(HPO4)4C12 .................................. 30 K4Pt2(I-IPO4J4Br2 ................................. 30 d. K4Ptglfli029)20]4Br2 ........................... 31 e. Dirhodium Fluorophosphine Complexes 3 l 2. Spectroscopic Techniques ............................. 32 a. Electronic Absorption Spectroscopy ..... 32 b. Steady-State Luminescence Spectroscopy 32 Excitation Spectroscopy ...................... 33 d. Time-Resolved Luminescence Spectroscopy ....................................... 37 C. Results and Discussion .......................................... 37 l . Diplatinum Tetraphosphates ......................... 37 2. Diplatinum Pyrophosphites ........................... 69 3. Dirhodium Fluorophosphines ....................... 72 111. Synthesis and Characterization of Layered Metal Phosphates lntercalated with Metal Dimers ........................................ 104 A. Background .................................................. 104 B. Experimental ......................................................... 108 l . Synthesis of Guest Compounds .................... 108 a. General Procedure ............................... 109 a. Rh2(CH3COO)4-(CH3OH)2 .................... 109 b. RhgiCH3CN) 10(BF4J4 ............................ 109 c. M02(CH3CN)3(BF4J4 ............................. l 10 2. Synthesis of Host Layers ............................... l 10 ' a. Layered Vanadium phosphate Dihydrate ............................................ 1 10 b. Layered Niobium Phosphate 'lrihydrate ....................................... 1 l l 2. lntercalation Reaction Chemistry .................. l 1 1 a. Sodium lntercalated Layered Phosphate 1 l l b. lntercalation of Bimetallic Cores .......... 1 12 3. Instrumental Techniques .............................. l 12 a. Powder X—ray Dim'action ...................... l 13 b Elemental Analysis .............................. 1 13 c infrared Spectroscopy .......................... 1 14 (1 Electron Paramagnetic Resonance l 14 Page e. Magnetic Susceptibility Measurements .................................. l 15 f. Electronic Absorption Spectroscopy ...... 115 C. Results and Discussion .......................................... l l 5 1 . Synthesis and Characterization of Layered Phosphates ................................................... 1 1 5 2. Structural Characterization .......................... 138 3. Magnetic Properties of Mini lntercalated Layers ........................................................... 1 52 a. Electron Paramagnetic Resonance ....... 1 52 b. Magnetic Susceptibility ........................ 164 4. Conclusion ................................................... 182 IV. New Directions in LIPS .................................................... 184 A. Modification of Phosphate Interlayer Galleries ........ 185 B. Modification of the MM Ugation Spheres Adapted for Layered Phosphate Incorporation ............................... 204 C. Conclusion ............................................................. 208 References ................................................................................ 2 10 List of Tables . Absorption and Emission Spectral Data for PtgillIJIan Tetraphosphates .............................................................. . Calculated Radiative and Nonradiative Decay Rate Constants and Energy Gaps of PtgflIIJIDIq Tetraphosphates ............ . Temperature Dependent Quantum Yields. Lifetimes and Nonradiative and Radiative Rate Constants of Crystalline Kgptga'IPOd4a'IgOb ........................................................ . Temperature Dependent Quantum Yields. Lifetimes and Nonradiative and Radiative Rate Constants of Crystalline K4“2(HPO4)4C12 .............................................................. . Temperature Dependent Quantum Yields. Lifetimes and Nonradiative and Radiative Rate Constants of Crystalline K4Pt2(HPO4)4Brg ............................................................. . Spectral Information of Bimetach Rhodium Fluorophosphine Compounds ................................................................. Page 62 63 81 10. ll. 12. Calculated Decay Rate Constants and Energy Gaps of Bimetallic Rhodium Fluorophosphine Compounds ........... Temperature Dependent Lifetimes and Calculated Nonradiative Rate Constants of Crystalline Bimetallic Rhodium Fluorophosphine Compounds ........................... identities and Energies of Peaks Used to Identify Elements in ESCA Spectra of Layered Metal Phosphates ..................... Stoichiometries and d-Spacings of M‘2(CH3CN)n(BF4J4 (n =- 8. M'sMo:n- 10.M‘sRh)IntercalatchOP04(M=VorNb) Layers ............................................................................. Peak Positions and the Indices of Powder X-ray Diffraction Patterns of Layered Metal Phosphates .............................. Electron g Value from ’EPR. Currie-Constant (Cm). Weiss- Temperature (6 / '1‘). Spin Density (SD / 96) and Diamagnetic Correction Factor (DCF / emu-mol’l) Deduced from Temperature Dependent Magnetic Susceptibility Measurements of lntercalated Layered Materials .............. Page. 99 122 123 141 ListofPigures . Schematic diagam of the molecular architecture of the thylakoid membrane ........................................................ . Schematic diagam of metalloporphyrin based electron transport chain in Zeolite-L channel. The hexagonal tubular structure represents Zeolite-L main channel with a cavity opening of about 1 3 A. The sensitizer is ZnTMPyP‘“ or Rht'bpy)32+ and A1 / A2 are derivatized methyviologens . Qualitative schematic of solid state supported photochemical assembly utilizing multielectron photochemical center (MPC) and layered materials ...................................................... . Immediate coordination spheres of dirhodium fluorophosphine complexes: (a) Rh2[(PF9)2NCH3]3(PF3)2; (b) Rhgflpp flgNCH3l3a’F 3)C12; and (C) RthWFflzNCHdaCM . Qualitative energ diagam for M-i-M (D4h) species in accordance with a general valence bond model ................ . Relative energy diagam for Mn2(C0) 10 ............................ ll 15 19 27 10. 11. 12. Block diagam of instrumental set-up for excitation arperiment ...................................................................... Solution electronic absorption spectra of Pt2(HPO4)4(1-120)22' ( -------- ) in aqueous solution and Pt2(HPO4)4C124‘ (—-—l and mammary- ( ------- n in 50% saturated KCl(aq) and KBriaq) solutions. respectively. at 300 K .......................... Electronic absorption spectra of low energy absorption region of Pt201P0d4C124' in a 1: 1 mixture of saturated aqueous LiCland 1 MHCl at is) 300Kand (b) 77K ...................... Corrected electronic emission spectra at 10 K of crystalline Knptzmmnqm=n10m=2i -------- );L=Cl,n=4(—-); L-Br,n=4( ------- ) ....................................................... (an) Electronic emission from -Pt2(HPO4)4C124’ in. a low- temperature glass (77 K) of saturated LiCl(aq) and 1 M HCl. (b) Scan over the same spectral region of the same solution atatemperature (14310abovethcgassingtemperature. 'l‘hisscanwasrecordedattentimethesensitivityofthatof scan (a) ........................................................................... Unpolarized excitation spectra of Han-P043402" in low- temperature glass (77 K) of 1:1 saturated LiCl(aq) and 1 M HCl. Progessively more concentrated solution were used in xii 36 39 43 49 13. 14. 15. 16. order to obtain adequate signal to noise ratios for the spectral regionsA. BandC. Ineachcase the OD/cmwas maintained < 0.2 within the spectral region being studied Fits of variations of (a) observed and (b) radiative decay rate constants of KgPtgaiPOdiingOb to eq 5 iii the 10 - 290 K temperature range. The solid lines represent calculated values .............................................................................. Fits of variations of (a) observed and (b) radiative decay rate constants of K432000440: to eq 5 in the 10 - 290 K temperature range. The solid lines represent calculated values .............................................................................. Fits of variations of (a) observed and (b) radiative decay rate constants of K4Pt20‘lPO4)4Br2 to eq 5 in the 10 - 290 K temperature range. The solid lines represent calculated values .............................................................................. Proposed energ diagam of the lowest energ' excited states of the PtgflIIJIDIq phosphates. The state manifold is derived from the spin-orbit coupling perturbation of the 3Eu state arising from the one-electron 3(dx‘ —n do‘) promotion ....... Page 51 54 56 67 17. 18. 19. 20. 21. 22. Electronic absorption (———-) and emission ( --------- ) spectra of K4Pt2(pop)4Br2 in deoxygenated aqueous solution at 300 Fit of the variation of the observed emission decay rate constant of K4Pt2(pop)4Br2 to eq 5 in the 10 - 290 K temperature range ........................................................... Qualitative enery level diagams of (a) RthPF 212NCH313WF312. (bi halWFflzNCHal3(PF3)012 and (c) Rh2[(PF7)2NCH3)3Cl4 generated by proper fragnents (C3v Rh(0)P4 fragnent for Rh(0) center and C4” (Rh(ll)P3Clg fragnent for Rhill) center). The dx- and d5- symmetry orbitals are filled and indicated by the shaded box. Low symmetry splitting within the dx level of the C4,, Rh(II)PCl fragnent is not considered ..................................... Excitation spectrum at low temperature (77 K) of crystalline RhgupF flgNCHalaa’F 3)C12 complex .................................. Excitation spectrum at low temperature (77 K) of crystalline Rh2[(PF2)2NCH3)3Cl4 complex .......................................... Fit of the variation of observed emission decay rate constants Of halippzlgNCHfl3M3lg to eq 5 iii the 10 - 160 K xiv Page 71 74 77 24. 25. 26. 27. temperature range. Steady-state emission shows no significant intensity above 160 K ..................................... Fit of the variation of observed emission decay rate constants of Rh2[(PF2)2NCH3]3(PF3)Clg to eq 5 in the 10 - 190 K temperature range. Steady-state emission shows no significant intensity above 190 K ..................................... Fit of the variation of observed emission decay rate constants 0f RhalWFflgNCthCh to eq 5 in the 10 - 290 K temperature range ........................................................... Fit of the variation of observed emission decay rate constants 0f Rbguszthnglaa’FQBrg to eq 5 in the 10 - 190 K temperature range. Steady-state emission shows no significant intensity above 190 K ..................................... Fit ofthe variation ofobserved emission decay rate constants of Rh2[(PF9)2NCH3]3Br4 to eq 5 in the 10 - 250 K temperature range. Steady-state emission shows no significant intensity above 250 K ..................................... Proposed energr diagams of the lowest energy excited states of PtzillIJII) phosphates and of Rhg fiuorophosphine complexes modified from the energ diagam of PtgallJll) phosphates. The state manifolds are derived from the spin- XV Page 89 91 93 95 97 29. 31. 32. orbit coupling perturbation of the 1.33“ states arising from the one-electron dx‘ —) do‘ promotion. Point goups. D3. Cs and C2. represent Rh2(0.0). Rh2(ll.0)X2 and haflIJDX4. reapecflvely Schematic representation of (a) 3-dimensional anhydrous VOPO4 and (b) layered VOPO4 dihydrate. The coordinating water molecule is shown as shaded spheres and the interlayer water molecule was omitted for clarity .............. ESCA spectra of (a) VOPO4-2H20; and (b) Nat-intercalated VOPO4 host layers ........................................................... ESCA spectra of (a) NbOPO4-3H20: and (b) Nat-lntercalated NbOPO4 host layers ......................................................... ESCA spectra of solids obtained from the reaction of M02(CH3CN)8(PF4)4 With (a) VOPO4°2H20; and (D) Nat- intercalated VOPO4 ......................................................... ESCA spectra of solids obtained from the reaction of M02(CH3CN)8(PF4)4 with (a) NbOPO4~ 21120: and (1)) Nat- intercalated NbOPO4 ....................................................... . ESCA spectra of solids obtained from the reaction of xvi Page 101 107 119 121 126 128 35. 37. 39. Rh2(CH3CN)10(PF4)4 With (a) 170130431120: and (b) NEI- lntercalated VOPO4 ......................................................... ESCA spectra of solids obtained from the reaction of Rh2(CH3CN) 10(PF4J4 with (a) NbOPO4- 21120: and (b) Nat- lntercalated NbOPO4 ................................................... Electronic absorption spectra of M02(CH3CN)8(PF4)4 intercalated NbOPO4i --------- ). the same layer afier exposure to air ( ----------- l and M02(HPO4)43' in deoxygenated 2 M H3PO4 (—-———) .............................................................. Powder X-ray difi'raction patterns for host materials: (a) VOPO4- 21120: (b) Nat-lntercalated VOPO4: (c) NbOPO4- 31120: and (d) Nat—lntercalated NbOP04 layers .......................... Coordination environment determined by EXAFS for mono- or divalent cations (123] situated in VOP04- 21120 structure Powder X-ray difi'raction patterns of solids obtained from the reaction of M02(CH3CN)3(PF4)4 with: (a) VOPO4-2H20: (b) Nat—lntercalated VOPO4: (c) NbOPO4- 31120: and (d) Nat- lntercalated NbOPO4 layers ............................................. Powder X-ray diffraction patterns of solids obtained from the reaction of Rh2(C1-13CN) 10(PF4)4 with: (a) VOPO4- 21120: (b) xvii Page 130 132 137 140 144 146 40. 41. 42. 43. Nat-intercalated VOPO4: (c) NbOPO4- 31120: and (d) Nat- intercalated NbOPO4 layers ............................................. Views of the VOPO4-2H20 layer from (a) the 00 1 plane where the four closest neighbor oxygens are shown in shaded circles: (b) the dimensions of the oxygen coordination environment for M025* intercalated VOPO4 [124): and (c) the dimensions of the oxygen environment of M02[(C5H5O)2P02]4(BF4)4 as determined from crystallogaphic analysis [39) ........................................... Electron paramagnetic spectra of VOPO4o2H20 (a) observed at 7 K and (b) simulated using a 2-site model for the cll VUV) centers. The microwave frequency and the sweep range were 9.471 1 GHz and 2512 - 4582 G. respectively ................... EPR spectrum of (a) Nat-intercalated (9.4724 GHz. 2497- 4577 G) and (b) Rhft-intercalated (9.4712 GHz. 2294-4382 G) VOPO4 layer at 7 K ...................................................... EPR spectra of (a) M02(CH3CN)3(BF4J4 intercalated VOP04 (9.4726 GHz. 970-6068 G) and (b) the spectrum of the solid after exposure to air (9.4718 GHz. 2293-4387 G) ............. Page 146 151 154 157 45. 47. Page EPR spectra of (a) NbOPO4-3H20 (9.4712 GHz. 1270-5379 G) and (b) Na5t-intercalated NbOPO4 layers (9.4718 GHz, 1270-5383 C) at 7 K ........................................................ 163 EPR spectra Of (a) M03(CH3CN)8(BFd4 intercalated NbOPO4 (9.4742 GHz. 2293-4382 G) and (b) the spectrum of the solid after exposure to air (9.4728 Griz. 1380-4590 G) ............. 166 (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of VOPO4- 21120 layer. The solid line represent the best-fit lines in (a) ................................................................................ 169 (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of Na+- irntercalated VOPO4-2H20 layer. The solid line represent the best-fit lines in (a) and the calculated results from eq 14 using.) values obtained from (a) ........................... _.., ........ 172 (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of M029- intercalated VOPO; 21120 layer by Route 1. The solid line represent the best-fit lines in (a) and the calculated results fromeq 14 usingJ values obtained from (a) ..................... 177 49. 51. 52. Page (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of M02?”- intercalated VOPO4-2H30 layer by Route 11. The solid line represent the best-fit lines in (a) and the calculated results fromeq 14 usinngalues obtained from (a) ..................... 179 (a) Curie—Weiss plot for (a) NbOPO4-3H20 and (b) Nat- intercalated NbOPO4 layers. The solid lines and open circles represent the best-fit lines and experimental points that are omitted from fittings ......................................................... 18 1 Structures of (a) simple erHPO4)2-2H20 and (b) modified Zr(RP03)2 layer with a organic phosphonic acid ........... 187 Structure of modified vanadium phosphate. VORPO3 in its ac-plane and bc-plane ................................................ 190 Powder X-ray patterns of vomooccnchgPoa (a) forming a hydrogen bonded bilayer for acid concentration < 0.2: (b) interdigitated bilayer formed for acid concentration 0.3 < x < 0.5 (see text) ................................................................... 193 Proposed structures of VOfliOOCCHgCHg)PO3 forming (a) hydrogen bonded and (b) lnterdigitated bilayer ................. 195 55. 57. Page Infrared spectra of vomooccugcngpoa forming (a) hydrogen bonded and (b) interdigitated bilayer ................. 198 Powder X-ray patterns of VOGiOOCCHgCHzlPO3 (a) before and (b) after esterification with (CH3CH2)30(BF4) ......... 20 l POSSIDIC structures of VO(CH3CH200CCH2CH7)PO3 ........ 203 Electronic absorption spectrum of 'Moz(cytocine)4' in dried. deoxygenated CH3OH .................................................. 207 ha(0.0l hamfllxa hamsm LIST 0" ABBREVIATIONS diamagnetic correction factor bis-(dimetylphosphino)methane. [(CH3)2P)2CH2 ethylene diamine tetraacetate electron paramagietic resonance electron spectroscopy for chemical analysis extended X-ray absorption fine structure highest occupied molecular orbital layered integated photochemical system ligand-to-metal charge transfer lowest unoccupied molecular orbital 6-methyl-2-hydroxyopyridine metal-to—mctal charge transfer multielectron photochemical center nicotineamide adenine dinucleotide phosphate reduced nicotineamide adernine dinucleotide phosphate Pymphoephite. (11029120 Photosystem halfl“ 2912MCHsllsWF312 Rh2[(F2P)2N(CH3))3(PF3)x¢ (X a C1 or Br) Rh2l(F2P)2N(CH3))3x4 (X 2 C1 or Br) SD spin density 2.5-dimethylo2.5-diisocyanohexane ultraviolet visible X-ray photoelectron spectroscopy CHAPTERI INTRODUCTION The multielectron chemisz of the photosynthetic reaction center has been one of the most popular models in the design of structures capable of efi'ecting light energy to chemical energy conversion. The overall reaction chemistry occurring in the photosynthetic reaction center is predicated on consecutive one-electron transfer reactions facilitated by cooperative interactions of many reaction components [1]. including Photosystems (PS) I and 11, several Fe-S enzymes. a quinone pool. and many other relay components. These components are anisotropically arranged within the thyiakoid membrane. A schematic representation of the photosynthetic assembly is shown in Figure 1 . Photosystems I and II are the primary receptors of solar energr and create highly reducing electrons that are separated from holes across the membrane. The strongly reducing electrons created by PS I. which are primarily localized at a tightly bound Fe-S center. are eventually transferred to NADP” through Fe-S centers, ferredoxin and ferredoxin-NADP-reductase to produce NADPH. A plastoquinone pool. which plays a vital role in coupling PS I to PS 11, accepts electrons released by PS 11 through the Figure 1 . Schematic diagam of the molecular architecture of the thylakoid membrane. hr c: 203 85:35.: z. .nn 09 Q .20 uzqcozwz son 9935»; . on 4/ .o . .. .efl 3H0 a new. .NO 9. .i. 2285. + #20 +IN IQOQZ 4 specially bound plastoquinone molecule. X320‘. The hole created by the solar photon in PS I is filled by electrons relayed through an Fe-S center. cytochrome-f-complex and plastocyanine network from PS 11. Strongly oxidizing holes localized at the Chi at of PS 11 are neutralized with electrons originating from water. mediated by the oxygen evolving center, which contains a Mn4 cluster as the active site. The oxygen evolving center is believed to accumulate four electrons before actual water oxidation to molecular oxygen occurs [2]. Thus the overall multielectron process of photosynthesis is derived from coupling a series of one-electron charge separation processes along a vectorial electron transport chain. A feature crucial to the process is that back electron transfer rates found in the reaction center are always much slower than the corresponding forward electron transfer reactions thereby enabling the propagation of emcient charge separation [3] . It is well recognized that the success of the multielectron chemistry in photosynthesis and more generally other natural multielectron assemblies is due not only to the complexity of the reaction path but also to the intricate arrangement and cooperative functions of the surrounding structure. Of course. the intricacies of this system makes the design and synthesis of a structure that can carry out similar multielectron transformations through consecutive single electron transfer reactions dificult. One basic scheme that has been adopted to mimic the photosynthetic center is a multicomponent system constructed with a sensitizer. relay and reaction center. Typically one—electron sensitizers such as .Ru(bpy)32+. metalloporphyrins or semiconductors are coupled by derivatized relays such as methylviologen and/or chains of conjugated 5 double bonds to solid state substrates such as Ru02 or Pt. which catalyze multielectron transformations [4.5] . Simple addition of these substrates to homogeneous solution with the goal of effecting multielectron chemistry has been unsuccessful [6]. Some of the reasons for this failure are the interference of back electron transfer between photoreacted sensitizer and relay at the difiusionally controlled limit. ineffective separation of reactive intermediates from electron. transfer reactions. and low conversion emciencies of the solid state substrates. Most research to date has focussed on forestalling the back electron recombination by constructing vectorial assemblies of electron transfer carriers. The facile separation of charges over a large distance assisted by a redox potential gadient has been the most common approach to achieving efficient charge separation. The synthesis of biologically relevant charge separating networks was pioneered by Chang with the preparation of cofacial diporphyrins in which one of the two porphyrins is the free base and the ’second is the magnesium complex [7]. Photocxcitation induces an electron to be transferred from the magnesium porphyrin to the free base porphyrin center. The forward charge separation proceeds with k > 10 l 1 s’1 where the reverse reaction is retarded significantly (k a 5 x 109 s") even though the two rings are only ~ 4.3 A apart [8). More recently. the donor/ acceptor strateg' has been further elaborated with the development of porphyrin-based diads [9] and triads [10.1 1]. In the former. porphyrins covalently juxtaposed to other porphyrin or quinone acceptors in cofacial and edge-on geometries yield charge separated states upon excitation of the porphyrin. Introduction of an electron donor onto the porphyrin photoreceptor in 6 Gust and Moore's [10] and Wastelewski's [l 1] molecular triads permits the photogenerated hole localized on the porphyrin to be trapped by the following sequence of charge separation electron transfer: DP‘A —) DP‘A" —n D‘PA’. This photogenerated charge separated state may persist into the microsecond range: the overall quantum yield however is usually low. In an efiort to achieve better charge separation. the covalently bonded molecular approach to vectorial charge separation has been complemented by constructing reaction components in a vectorial fashion within organized assemblies. Systems designed about many support structures. such as micelles [12]. polymers (13). and solid state substrates (14- 1 6) have been investigated. One promising approach is that of Mallouk et al. [1 7] who claim the vectorial disposition of building blocks based on a metalloporphyrln/ methylviologen/ Pt system within the channel type cavity of Zeolite-L. A diagammatic representation of the assembly is shown in Figure 2. The stepwise construction of each component was performed with implantation of Pt aggegates in Zeolite- L. The relatively small. rod-shape methylviologen molecules intercalated in the cavity. constitute a relay system for charge separation. Due to the size constraint. introduction of the large Ru(bpy)32+ or modified metalloporphyrin sensitizer units would be expected to only exchange out methylviologens at cavity opernings: methylviologens inside of the channel are presumably unafi'ected. The overall operation of the solid state vectorial assembly is initiated with the harvesting of incident photons by the sensitizer at cavity openings. Electrons donated by sensitizers are transferred through several methylviologens to the Pt aggegates. The photooxidized sensitizers are regenerated by sacrificial donors such as EUI‘A. At the surface of the Pt. reduction of protons provided from an r—r Figure 2. Schematic diagam of metalloporphyrin based electron transport chain in Zeolite-L channel. The hexagonal tubular structure represents Zeolite-L main channel with a cavity opening of about 13 A. The sensitizer is ZnT'MPyP4+ or Rh(bpy)32+ and A1 / A2 are derivatized methylviologens. .250 E ._ o.__os~ (how 9 acidic medium completes the oxidation-reduction cycle. Despite the careful construction of the system. the reported emciency for hydrogen production is only < 3 x 10‘5. This low quantum yield can undoubtedly be attributed to fast and efficient competing back electron transfer as compared to forward electron transfer to the Pt centers. To this end. though this approach has considerably expanded the horizons in building organized assemblies. neither the tempering of back electron transfer nor control over the location of each component is completely satisfactory. Moreover the system continues to rely on a sacrificial donor to drive the chemistry. Even if charge separation in one-electron systems is ultimately successful. multielectron photochemistry is not ensured. The initial electron/hole pair must be stored at the terminus of the network. and the overall process must be repeated to build up the necessary multielectron hole and electron equivalents. Ultimately. the charge equivalents must then be coupled to a catalytic center capable of promoting the overall multielectron process. Indeed. the synthesis of such centers comprises a major research area in its own right (2a.2b. 18. 19]. An entirely difi‘erent approach to the design of multielect'ron systems is the development of assemblies that can efi'ect multielectron transfer at once. Incorporation of compounds that are capable of achieving discrete multielectron transformations into solid state substrates is an attractive choice. Structured assemblies containing multielectron photoactive centers can provide not only a direct pathway to products but also accessibility to complementary redox active metal centers from the solid state support. Figure 3 schematically diagams a 10 Figure 3. Qualitative schematic of solid state supported photochemical assembly utilizing multielectron photochemical centers (MPC) and layered materials. l2 multielectron assembly. In principle. the photoactive center will harvest the photon to efi'ect discrete multielectron transfer producing a reduced substrate and oxidized multielectron photoreaction center (MPC+). The regeneration of the oxidized MPC can be obtained by auxiliary redox- active metal centers in the solid state support. Subsequent reaction of substrate at the layers will re-establish the formal orddation state of metals in supporting solid state substrates thereby completing the catalytic cycle. In this approach. relay molecules are not required since there will be no high way one-electron intermediates that need to be propagated along a charge separation network. The exclusion of the relay system in design leads to an important advantage of this approach in that the intricate structural engineering of photoactive solids needed to prevent charge recombination is circumvented. The centerpiece of the multielectron strateg' is the existence of the multielectron photoactive centers. Recently. discrete multielectron photochemistry has been observed in our research progam based on the two-electron reactions of bimetallic Mm L—Mn' 1 systems [20]. Two- electron orddations may be promoted at the MI" 1 site whereas the Mn' 1 site is particularly susceptible to two-electron reductions. Access to M11+ L—Mn' 1 photochemistry can be accomplished by the two routes shown in eqs 1 and 2. nun—Mu ——h—"——-) (MmL—Mn'l)‘ (n) Mn+l—Mn-l —11‘-'—-) (MmL—Mn-I)‘ (2) In eq 1 . excitation of a metal-metal charge transfer transition of a Mil—Mn dimer will photogenerate a W l—Mn' 1)‘ excited state. 13 Alternatively. gound state Mm 1—-M11' 1 complexes can be synthesized and typical of most photochemical reactions. thermodynamic or kinetic barriers confronting the gound state can be circumvented with the may of the excited state. In regard to the strategy based on eq 2. gound state two-electron mixed valence compounds are for the most part rare. However. syntheses of these complexes can be rationally designed by employing ligands that can stabilize both high and low oxidation states of a bimetallic center. In fact. the singly bonded two-electron mixed valence complex Rh2[CH3N(PF9)2)3(PF3)C12 (Rh2(II,0)Clg) was recently synthesized [2 I] in our laboratory. The two-electron mixed valence nature of the Rh2(Il,0)C12 complex is unequivocally established by the coordination geometry about individual rhodium centers as reproduced in Figure 4b. Pseudooctahedral and trigonal bipyramidal geometries are structural benchmarks for metal-metal bonded rhodium in divalent and zero valent states. respectively. Multielectron gound state reactivities of these complexes are as predicted. The thermal reactivity of hafll.0)C12 is dominated by two- electron oxidative-addition reactions at M” 1 center. and two-electron reductive-elimination reactions at the Mn+1 center [20.2 1). Reducing agents such as borohydride. naphthalide and cobaltocene react with Rh2(ll.0)C12 under PF3 atmosphere to give two electron reduced Rh2[CH3N(PF2)2]3(PF3)2 (Rh2(0.0)) (Figure 4a). Conversely. addition of orddants such as halogens or dichloroiodobenzene results in oxidative addition of halides at the Rh(0) center to give a Rh2(II.II) core. The octahedral geometry of Rh(II) center is obtained by removing PF3 at axial position and adding two incoming ligands from the oxidant (Figure 4c). 14 Figure 4. Immediate coordination spheres of dirhodium fluorophosphine complexes: (a) Rh2[(PF2)2NCH3]3(PF3)2: (b) Rh2[(PF9)2NCH3]3(PF3)012; and (C) Rh2[(PF2)2NCH3]3C14. 15 (a) m i=2 2‘- v I P‘ an: ‘. am , A - ,9 . PI ~ 5:34 "it :A P? a <' S) P5 as (b) (C) 012 PT Cit m4 16 All three classes of dirhodium complexes show characteristic Rh—Rh single bond lengths with electron configurations of d9—d9. d7—d9 and d7—d7 for reduced. mixed-valence and oxidized complexes. respectively. The chemical reactivity of singly bonded metal complexes such as these multielectron dirhodium compounds in excited states is yet to be revealed. The elaboration of photoinduced multielectron transformations of singly bonded bimetallic compounds will require that the photodissociation of metal cores is circumvented. The electronic structure of singly bonded bimetallic complexes with bridging ligands has not been studied in detail. Luminescence from bimetallic M—M complexes is rare. even if the metal bond dissociation is prevented by bridging ligands. Thus before the solid state supported Mn+ L—Mn' 1 complexes are designed. the excited state dynamics of the singly bonded M—M complexes needs to be elucidated. Chapter II reports our oburvations of long-lived luminescence from solids and low temperature gasses of singly bonded diplatinum tetraphosphate complexes and the mixed-valence dirhodium complexes. These observations are in striking contrast to that typically observed for singly bonded M—M complexes [18-20). which heretofore have not shown luminescence. Detailed photophysical studies of highly symmetric diplatinum tetraphosphates are presented along with a working model of the excited state dynamics. An attempt to generalize this model using dirhodium fiuorophosphine complexes is also reported in Chapter 11. These results provides the underpinning for the development of solids displaying multielectron reactivity according to eq 2. Alterrnatively. multielectron photoactive solid state materials based oneq 1 mayberealizedwiththerecentdiscoveryinourgoupofthe 17 quadruply bonded metal-metal (M—4—M) dimer photochemistry. Our results are based on eq 1 in that two electron mixed-valence character may be established in excited state by MMCT excitation. The d4—d4 electron count of M-i-M renders a 02x452 gound state electronic configuration. The HOMO and LUMO with 5—syMetry are formed from the interaction of two atomic d,‘y orbitals from each metal center. The lowest energy absorption band of quadruply bonded dimers usually corresponds to 1(5 -n 8‘) transition with the retention of a strong metal- metal interaction [23,24]. However. a molecular orbital model is a poor quantitative representation of the 5-orbitals in M-iM complexes due to the poor overlap between two atomic dxy orbitals. The electrons residing in these orbitals are more appropriately described by general valence bond theory [24-26] as shown in Figure 5. In this model. the llAlg gound state corresponds to a 1 (5’) molecular orbital where one electron resides in each d,‘y orbital. The lowest energy excited state. 311.2u (corresponding to 3(85‘» is energetically proximate to the gound state with (1,, electrons in the triplet configuration. At much higher energy are two singlet excited states arising from the linear combination of both dxy electrons on one metal center. The antisymmetric 1A2“ state which correlates with 3(58‘). is the origin of the long-lived red-luminescence of Mi-M complexes. Since the 1(5 —r 8‘) transition correlates with two electrons residing in the same metal center. the excited state acquires a significant amount of two-electron mixed-valence character. Compelling evidence for this ionic character comes from a 4.0 Debye dipole moment of the 88‘ excited state in M02Cl4(PMe3)4 as compared to the zero dipole moment of the gound state [27] . Moreover the time-evolution of the emission maximum 18 Figure 5. Qualitative energy diagam for M (D411) species in accordance with a general valence bond model. 19 — 1A1g(8*8‘). { + —‘Aa (66*) { _ — 3A2. (66*) i i (as) (d...) 20 occurs within the time scale of microscopic solvent reorganization time. The importance of solvent dynamics in the evolution of the 85" excited state strongly supports a general valence bond description of excited state with ionic character. The generation of an ionic mixed-valence excited state is reflected in the recurrent theme in M—‘LM chemistry that oxidation of the binuclear core is typically accompanied by ligand rearrangement (28). N0- electron oxidized M2X4L4 (M a: Mom). W(II). Rh(lll): X = halide or pseudohalide: L = donor ligand including halide) species can be stabilized by a confacial (29.30] or edge-sharing bioctahedral configuration of ligands [31]. which ensures octahedral coordination about the oxidized metal core. Indeed the time-resolved absorption spectrum of a deoxygenated CH2C12 solution of W2Cl4(PBt13)4 gives evidence for the occurrence of such species. Generation of edge-sharing bioctahedral species upon 1(8 -) 8‘) excitation can be attributed to the fold over of two chloride ligands to bridging positions. As shown in eq 3. this type of ligand rearrangement provides cooperative stabilization of a twooelectron mixed valence intermediate by forming an octahedral geometry about oxidized center and by removing the x-donating halides from the reduced center. on P P P rials __.~v airmail ,4! Ci" ci' | \c.’ I (3) or P P P By opening the axial coordination environment of a low oxidation state metal. multielectron reactivity has been observed. The quantitative 21 photoaddition of CH31 to the electronically excited W2(dppm)2Cl4 has been reported [32). The visible irradiation (1 > 450 nm) of the quadruply bonded W2(dppm)2Cl4 in the presence of C1131 leads to the oxidative- addition product. W2(dppm)2Cl4(CH3)(I) where the CH3 and I are added at terminal positions of edge sharing bioctahedra. Most importantly. the reaction does not show any evidence of intermediates indicative of sequential one-electron transfer reactions. Terminal coordination of the methyl goup is consistent with oxidative addition of substrate to two- electron mixed valance W2(I,III) core. Thus the MM compounds are MPC's. which form the basis for the development of the layered integated photochemical systems (LIPS) proposed in Figure 3. The ultimate design of the LIPS structure requires that suitable supporting host structures must be compatible with M—M and M-4—M bimetallic complexes and possess structurally organized reaction sites that are accessible to incoming guest molecules. Layered materials are obvious supports because they readily incorporate guests by facile intercalation reactions [33). Also the two-dimensional structure ensures mobility of reactants while maintaining the structural organization of interlayer species. Although our studies have uncovered factors important in the perturbation of excited state properties of guest molecules in clay environments (34). we have realized that the utilization of clays for LIPS studies is hindered by two crucial factors: (1) clays are. for the most part. chemically irreproducible materials: and (ii) the basic properties of the clay are not compatible with many transition metal compounds and especially the MM species. which are susceptible to hydroxo and oxo formation at the clay/ solution interface. Conversely. layered phosphates are superb host compounds for bimetallic cores. 22 because they may be synthesized to yield pure crystalline materials and they are stable in acidic environments [35.36]. Additionally. overall two- electron photoreactions of Mali-Mo phosphate complexes [37 -39] have been observed as shown in Scheme 1. ”02(02P(0Y)2)X2 + RR MoaHPoinf' + H2 XRRX 2H” (Y =30". alkyl: n -0) (Y. 1/2H‘; n .-4) hv M°2(02P(°Y)2)4n 3‘. Scheme 1 ’ "‘ ‘ Thus the comparative study between interlayer and molecular photochemistry can be irnitiated with the layered phosphate. Finally. the interlayer gallery of layered phosphates can be modified to accept a variety of MiM MPC's (vide infra). Implantation of M3-M cores into the layered materials can. in principle. be performed by (1) simple ligand exchange reactions from M-E-M complexes with weakly bound ligands: and (it) by constructing metal-metal bonds from monomeric precursors inside the gallery. The latter approach has not been successful due to the relatively high charges. usually MN). on each metal center. The charge neutrality requirement prevents monomeric units from being close proximity: thus the formation of dimeric units is precluded. Alternatively. fully solvated 23 M—“—M compounds are attractive candidates for introducing the MM cores for the former approach. These solvated MAM cores have been known to be attractive starting material for many new compounds because the solvent ligands undergo extremely facile substitution. lntercalation reactions can proceed analogously where the layered material is considered as a substituting ligand. To this end. Chapter 111 reports the investigation of acetornitrile' complexes of dimolybdenum and dirhodium as precursors. Intercalation reactions. in this case. can be performed irn nonaqueous media where no interference from protons can be expected. The successful intercalation of dimolybdenum and dirhodium cores with M02(CH3CN)3(BF4)4 and Rh2(CHaCN) 10(BF4)4 starting materials into VOPO4 and NbOPO4 related layered materials. and the characterization of these newly synthesized materials are also described. The native structure of the layered phosphates are restrictive to LIPS design in that the two-electron chemistry of M-‘lM (rn . 1 or 4) complexes requires ligands other than phosphate in the gallery environment. To increase the versatility of LIPS design. either (i) the interlayer space of host structures or (ii) the strateg' of intercalation may be modified. The galleries of layered phosphates can be chemically modified by the method of Dine's [40] where the pendant ligands are covalently anchored to the layer. Reaction of functionalized phosphornic or arsonic acids (RX(=O)(OH)2: X a P or As: R s a pendant goup that will be protruding into the gallery) with M4+ oxides. oxychlorides and halides yields M(03XR)2 (M :- Zr. U, Th and Ce) species. Jacobson and Johnson [41] have further elaborated Dine's strategy to include vanadium and niobium phosphates by the reaction of MnOm with organic phosphonates r... 24 to yield MO(03XR). Alternatively. straightforward intercalation reactions can be used for binuclear cores displaying modified ligation spheres. Alkyl and aryl amines readily intercalate into Zr(I-IPO4)2 [42) and VOPO4 [43]. and their related structures to form bilayers. Ligands containing the N112 goup at a remote position from ligating goup such as nucleic acids or amino acids are very attractive. since amine intercalation is completely general for a variety of layered materials [33). Our innitial attempts at the synthesis of chemically tunable LIPS utilizing these strategies are reported in Chapter IV. The work described herein represents the first step of a new synthetic approach to the design of solid state multielectron assemblies. CHAPTERII We: FRO! SINGLY BONDED MAILIC COMPLEXES: DYNAMICS 0F 0" EXCITED STATES A. Backgound The lowest energr electronic transitions of many singly bonded metal-metal (M—M) complexes involve the population of the do‘ metal- metal level (44-47]. Typically. significant weakening of the metal-metal irnteraction results upon promotion of an electron to the do‘ orbital. and not surprisingly. photoinduced cleavage of the metal-metal bond has emerged as the dominant excited-state decay pathway of M—M species [48-59). The d7—d7 dimer. Mn2(C0) 10. is the cornerstone example of such photochemical processes. Figure 6 shows the orbital energy diagam of Mn2(CO)1o. Excitation into the dimer's absorption marnifold. which is dominated by the intense do -> do‘ absorption band and weaker dx -) do‘ band to lower enery (60), leads to cleavage of the Mn—Mn bond to yield neutral -Mn(CO)5 radical fragnents [61-67). Because do‘ deactivation pathways are erdremely emcient. the lifetimes of electronically excited M—M complexes are short [68]. and luminescence from M—M complexes is rare. To this end. coordination of M—M cores 25 26 Figure 6. Relative energy diagam for Mn2(CO)10. 27 dIZ-yz :> < may: m. waves as _\\/>m /\/\\ chm. m. Amara 28 by bridging ligands will prevent metal-metal dissociation and the expectation of luminescence from M—M complexes is a reasonable one. Indeed recent observations of luminescence from diplatinum(III) pyrophosphite complexes. Pt2(pop)4X2"" (pop = 0102P)2O. X = halide) [69). a birnuclear dirhodium(ll.0) fiuorophosphine complex. Rh2[(F2P)2N(CI-L3)]3(PF3)C12 [21]. and bldentate phosphine derivatives of Re2(CO)1o [70] represent the first examples of do‘ luminescence. These complexes contain bridging ligands with only one bridgehead atom and thus deactivation by metal-metal bond cleavage in the excited state is circumvented. Although retention of the M—M core in the excited state is a necessary condition of do‘ luminescence. it is not sumcient. The paucity of emissive M—M complexes. despite the synthesis of several bridged complexes during the past years. suggests that subtler. less well understood perturbations play an important role in the deactivation of do‘ excited states. Owing to our interest in the excited-state chemistry of binuclear metal phosphate complexes [2 1 .37-39]. we have undertaken photophysical investigations of the single metal-metal bonded Pt2(HPO4)4l/z“' (L 3 H20. n - 2L 2 C1 or Br. n a 4) complexes whose structure consists of the symmetrical disposition of phosphate ligands about the Ptz6+ core with the L donor ligands occupying axial coordirnation sites [7 1). Typical of most d7—d7 complexes. an intense band attributable to the population of the do‘ level dominates the electronic absorption spectrum of Pt2(HPO4)414“' phosphate complexes. Solids and low temperature glasses of Ptzi'HPOaMizzn‘ phosphate complexes exhibit intense red luminescence upon irradiation irnto the ultraviolet and visible absorption manifolds. and that luminescence 29 persists even at higher temperature. A striking observation is the marked temperature dependence of the luminescence intensity and emission lifetime. In an efi‘ort to identify the factors important in mediating this temperature dependence. the photophysical properties of the Pt2(HPO4)4bzn’ complexes have been studied over a wide temperature range. In addition to the Pt2(IIPO4)414n‘ complexes. studies have been undertaken on solid K4Pt2(pop)4Br2 in an efi'ort to generalize models for do‘ luminescence. Luminescence from the K4Pt2(pop)4X2 complexes has been assigned previously as do‘ emission [69). Moreover. lower symmetry M—M complexes have been investigated. In particular. intense red luminescence from a series of bimetallic singly bonded dirhodium complexes. Rh2[(F2P)2N(CHs)la(PF312 (Rh2(0.0)). Rh2[(FzP)2N(CH3)13(PF3)X2 (Rh(II.O)X21 and Rh2HF2P12N(CH3)13X4 (1211201de (X a C1 or Br) (20.21] have been investigated. Interpretation of the temperature dependence.- of these M—M complexes has provided us with the first insight into the deactivating pathways of emissive do‘ excited states. Our results establish discrete long-lived excited states for M—M compounds arnd set the stages for W L—Mn— 1 excited state m'uitielectron reactivity. These studies establish M—M compounds as valuable cores for the design of multielectron photocatalyttc architectural assemblies. 3. Experimental 1. Synthesis of Compounds. 30 a. Kth2(BPO4)4(H20)2. The potassium salt of Pt2(HPO4)4(H20)22’ was prepared with a slight modification of literature methods [72.73]. A suspension of Pt(NH3)2(N02)2 (0.5 g. 1.6 mmol) in 85% H3PO4 (15 ml) was heated to 100.0 :i: 1.0 °C. The progess of the reaction was followed by visually monitoring the brown fumes of N02 released from solution. Upon the cessation of N02 liberation. K2HPO4 (0.7 g. 4.0 mmol in 5 ml of H20) was added while the cooled solution was stirred. The yellow precipitate. which formed after 1 hr. was collected. washed with ethanol and ether. and air-dried. Crystalline K2Pt2(HPO4)4(HzO)2 was obtained from an aqueous solutions of the complex by slow evaporation of the solvent. b. K4Pt2(BPO4)4Clz. The chloride derivative was prepared by the addition of an aqueous solution saturated with potassium chloride to a solution of K2Pt2(‘HPO4)4(HzO)2 in water. Slow evaporation of water from solutions containing the aquo complex and a 100 molar excess of the potassium chloride yielded pure. crystalline compound. The orange crystals were washed with small portions of water to remove excess potassium chloride. and air dried. c. K4Pt2(HP04)¢Bra. Similar to the chloride derivative. slow evaporation of solvent from an aqueous solution of K2Pt2(HPO¢)4(I-120)2 and a 100 molar excess of potassium bromide produced deep red crystalline K4Pt2(HPO4)4Br2. Resulting crystals were washed with water and air-dried. Unlike the chloride derivative. mixtures of rod-shaped sirngle crystals and inter-twined clover-shaped crystals were obtained. The clover-shaped crystals were removed under 31 microscope examination. They were found to be an emitting impurity [74]. a possible decomposition product of K4Pt2(HPO4)4Br2. d. KiPtziPopuBrz- K4Pt2(90p)4(HzO)2 was used as a precursor for the synthesis of the dibromide derivative by oxidative- addition of bromine. The starting diplatinum compound was obtained by standard methods (75]. A mixture of K2PtCl4 (0.5 g. 1.2 mmol) and phosphorous acid (2.3 g. 28.0 mmol) in 10 ml of water were heated on a steam bath for 5 hrs while the volume was kept constant by adding water occasionally. The resulting pale—yellow solution was dried overrniglnt at 120 °C. The yellow-geen residue was suspended in methanol and filtered. The solid was then washed with several portions of methanol and ether. air-dried. and stored under vacuum. The oxidative-addition of bromine was carried out in rigorously deoxygenated solvents. An aqueous solution saturated with Bra (5 ml) was added to a solution of diplatinum pyrophOSPhite [761. K4Pt2(popltai20)2 (0.12 g. in 3 ml H20). The luminescent geen from K4Pt2(pop)4(H20)2 solution quickly disappeared to leave a brownish orange solution. The slow addition of an aqueous KBr solution (3.2 g in 20 ml H20) led to the precipitation of orange-yellow powder. The product was filtered. washed with portions of ethanol and ether. air-dried and stored under vacuum. Extended exposure of the compound to air resulted in slow decomposition accompanied by a change in color to geernish yellow. e. Dirhodium Fluorophosphine Complexes. A series of dirhodium fiuorophosphine complexes. Rh2[(F2P)2N (CI-13))3(PF3)2. Rha- linPizNicmilsiPFaiClz. Rh2ltF2PileCHslisiPFaiBr2. halinPilecfisils- 32 C14 and Rh2[(F2P)2N(CH:3)]3Br4 were kindly provided by Joel I. Dulebohn and Janice Kadis. Their synthesis are described elsewhere [21.77]. Complexes used for photophysical studies were crystalline materials chosen from a batch on which x-ray crystallogaphic studies were performed. 2. Spectroscopic Measurements. a. Electronic Absorption Spectroscopy. Electronic absorption spectra were recorded on either a Cary 1 7 or Varian 2300 UV- vis-nir spectrometer. Low temperature absorption measurements employed a 1:1 mixture of saturated aqueous LiCl and l M aqueous HCl solution using high-vacuum cells consisting of a l-cm x 1 -cm cuvette and 10 ml capacity side arm reservoir separated by two Kontes high- vacuum quick release Teflon stopcocks. Prior to all absorption experiments. the solvent mixture was placed into the side arm reservoir and subjected to five freeze-pump-thaw cycles. than transferred to a 1- cm x l-cm quartz cuvette that contained the compound. Low temperature absorption measurements were achieved by placing the cuvette into a small liquid nitrogen Dewar flask equipped with three quartz windows (1 .7 cm diameter). The LiCl:HCl solvent mixture forms excellent glasses at 77 K. h. Steady-State. Luminescence Spectroscopy. Steady- state luminescence spectra were recorded by using a high-resolution emission spectrometer constructed at Michigan State Urniversity [78]. Excitation light (lac s 365 nm) from a 200 W Hg-Xe lamp was selected 33 by the double monochromator of the emission spectrometer in conjunction with an Oriel 365 nm (model 56430) interference filter. The luminescence from the sample was directed through the single monochromator and onto a dry ice cooled PMT (R3 16. Hamamatsu). Slit widths for the excitation and emission monochromators were 5 mm / 5 mm and 3 mm / 3 mm. respectively. The enhance slit of the emission monochromator was equipped with a Schott OG-560 cutofi' filter. All spectra were corrected for the instrument response function of the spectrometer as previously described [78). Variable temperature luminescence was recorded on powdered samples cooled with an Air Products closed-cycle cryogenic system by methods described elsewhere [34]. Quantum yields of powdered samples were determined by the technique reported for solids which relies on intensity differences between the light reflected from the sample and that reflected from a nonabsorbing MgO standard [79). Absolute quantum yields were determined by using Ru(bpy)3(ClO4)2 as a luminescence standard (0., a 0.003 in solid state at room temperature) according to eq 4. was.) - was 2; “'01 x luggage) - 130“) x Dr (4) where x and r refer to unknown and reference. respectively. 10m) is the intensity of the reflected light and D is the integated area under the emission profile. We estimate the error associated with the measurement of solid state quantum yields by this method to be :t 30%. c. Excitation Spectroscopy. Unpolarized excitation spectra of the diplatinum complex were recorded on a Perkin-Elmer MPF- 34 66 emission spectrometer housed in the Jet Propulsion Laboratories (California Institute of Technolog. Pasadena. CA 9 l 109). Measurements were performed on optically dilute (OD / cm < 0.2 at the excitation wavelength) solutions. Excitation spectra were corrected by using manufacturer supplied software. The spectrometers slit width was 5 mm. and all excitation spectra employed a 610-nm cutofi' filter on the emission side of the spectrometer. Samples for these measurements were contained in quartz tubes that were immersed in liquid nitrogen contained in a finger Dewar flask. Excitation spectra of dirhodium complexes were recorded on the aforementioned emission spectrometer built at Michigan State University [78) with modifications (Figure 7). A 150 W Xe lamp ((JSHIO) powered by an Oriel 8500 arc-lamp power supply was used as the excitation source. The excitation light was wavelength selected by a Spex 1680 double monochromator. A quartz beam splitter was placed between two collimating lenses of the excitation optical train to direct ~5% of excitation light to a photodiode whose output was monitored by an EG&G 128A lock-in amplifier. The emission was detected by a PMI‘ (R316. Hamamatsu). which was fed into an EG&G 5209 lock-in amplifier. The emission monochromator was fixed at a set wavelength and the emission intensity was monitored as the excitation monochromator was scanned. A MetraByte DASH- 16 interface was used to send information of the simultaneously collected lamp and emission intensities to the Zenith AT microcomputer. Excitation profiles were corrected for the lamp output and the photodiode response. MeasurementswereoncrystalsorpowdersinEPRtubesthatwere immersed in liquid nitrogen in a finger Dewar. A 715-nm cutofl' (Oriel RG-7 l 5) filter was installed in front of the emission monochromator to 35 Figure 7. Block diagam of instrumental set-up for excitation experiment. dE<.9_a BEES—.0822 .Em coSano .95 2.53.5 -m .. w 5a m .Vm ., 28 d - .x -. sea; m fa uni cogaw Eoom mm 3 gm m i... use“. :25 u 9 ”H w ... . 9:90 n. m m _i_ mm as: .I scan ex |||_ 88.6 5086930068‘ .xm hon—=20 8E3 37 filter out stray light from high order excitation lines. Slit widths were 5 nm / 5 mm and 3 mm / 3 mm for excitation and emission monochromators. respectively. (1. Time-Resolved Luminescence Spectrum. A Nd:YAG pulsed laser system (lac = 355 nm. fwhm = 8 ns) (34] was employed for lifetime measurements. The time-resolved response of luminescence was measured at a selected wavelength (at the emission maximum unless noted otherwise) by an R924 PMT. which was connected to a I.eCroy transient digitizer interfaced to a Compac 386 computer. The excitation light from the laser was filtered by an OG-560 cut-ofi‘ filter situated in front of the entrance slit of the monochromator. Data were recorded by CATALYST. a manufacturer-provided software package. When the decay of the emission intensity was biphasic. data fitting was accomplished iteratively by using KINFII‘ (80) to resolve each component. Monophasic decay times were obtained by standard linear regession methods. Temperature dependent measurements were achieved with the Air- product cryogenic system as described for temperature dependent quantum yield measurements. C. Results and Discussion 1. Diplatinum Tetraphosphates The UV-vis absorption specua of the Pt2(HPO4)414“" complexes. shown in Figure 8. accord well with those previously reported (72). The spectraaredominatedbyanintensebandinthe220-350nmspectral Figure 8. Solution electronic absorption spectra of Pt2(HPO4)4(I-120)22' ( -------- ) in aqueous solution and Pt2(l-IPO4_)4C124' (———-l and Pt2(I-IP04)4Br24'( ------- n in 50% saturated KCl(aq) and KBr(aq) solutions. respectively. at 300 K. 39 500'. 600 400... 300 200 ..w9.-w .01 xa X/nm “SW83 40 region and a less intense. broader band in the 390 - 410 nm region. This absorption profile. characteristic of d7—d7 complexes [44a.46.81.82]. closely parallels that of the Pt2(SO4)4l/zn’ and Pt2(pop)414n' (L : H20. n. = 2: L = C1 or Br. n = 4) ions [82-84]. Spectroscopic studies combined with results of molecular structure determinations of several diplatinumflll) pyrophosphite complexes have led to a o -> do‘ assignment for the intense UV absorption band [85-87). The pronounced shift of the UV band toward lower energ' along the axial ligand series H20 > C1 > Br is indicative of the o orbital possessing significant ligand character. Accordingly. it has recently been proposed that the lowest energr o -9 do‘ system in diplatinumalll pyrophosphites is most appropriately designated o(L) -r do‘ [88). This assignment is especially satisfying for the intense UV absorption band of Pt2(HP04)4l/3n" complexes because. despite the significant dependence of this band on the nature of the axial ligand. the interaction between the diplatinumflll) core and the axial ligands is weak. X-ray crystallogaphic studies have shown that the Pt—Pt bond distance is not geatly influenced by ligands in axial coordination sites [7 1 .89.90]. Moreover. kinetic studies reveal the axial ligands to be coordinated weakly to the diplatinumflll) core (72.91). In thisregard.itisunreasonabletoexpectflnatardalligandmirdngintothe do orbitals could account for the dramatic shifts observed in the absorption spectra of Figure 8. Alternatively. the do -r do‘ transition of the Pt2(I-IPO4)4I4“' complexes most likely lies to higher energr. The do —n do‘ transition of Pt2(pop)414“'(L=H20. n-2: LsClorBrn-4) ionshasbcenobscrved at ~2l5 nm [86). Owing to the much shorter Pt—Pt bond distances of Pt2(HP04)4l4“' complexes as compared to the pyrophosphite complexes 41 (d(Pt—Pt) 3 2.695 to 2.782 A for pyrophosphites: d(Pt—Pt) a 2.486 to 2.592 A for tetraphosphates) [84.89—94]. a larger do -> do‘ splitting should result from better overlap of the Pt dzz orbitals. By using the energy of the do —+ do‘ absorption of diplatinumflll) pyrophosphite ions as a benchmark. the do -r do‘ transition of the Pt2(HPO4)4I/2“' complexes should lie in the vacuum ultraviolet spectral region. In contrast to the UV transition. the lower energy. less intense absorption system of the Pt2(I-IPO4)4l/zn’ (L 2 H20. Cl. Br) complexes. for the most part. is insensitive to axial ligation. The HOMO's of most d7—d7 D41, complexes are of it or 8 symmetries formed fiom the bonding and antibonding combinations of the (dn. dyz) and dxy orbitals. respectively. on each metal. Lower energr absorptions. analogous to those in Figure 9. in the spectra of diplatinumfllll nymphosphites [85). and dirhodium(II) acetates [44a] and isocyanides [44b] have been assigned to the allowed dx‘ -9 do‘ transition. For each of these systems. the d8 -> do‘ transition. in ageement with electronic structural calculations [951 of d7—d" systems. has been observed to lie within 1500-2500 cm"1 of the dx‘ -n do‘ transition [44.85.87]. Conversely. absorption spectra of low-temperature gasses of Pt2(HPO4)4C124' display a distinct band which. on the basis of energ and intensity considerations. is consistent with a d8 -n do‘ assignment lying ~2300 cm“1 to higher may of the dx‘ -r do‘ transition (Figure 9). This result is not surprising in view of the fact that the much shorter Pt—Pt distance of the phosphate complexes will be manifested in a large ”dx" splitting and consequently a more destabilized dx‘ level than that typically found for d7—d7 complexes. 42 Figure 9. Electrornic absorption spectra of low energr absorption region of Pt2(HPO4)4C124' in a 1:1 mixture of saturated aqueous LiCl and l M HCI at (a) 300 K and (b) 77 K. (0) V (m 400 woz do emission band of Re2(CO)6(dmpm)2 (70). Although luminescence from crystalline Pt2(I-IPO4)4I4“’ complexes is detected over a wide temperature range. solutions of these complexes do not luminesce. This behavior is most vividly demonstrated by the results reproduced in Figure 1 1 . Frozen gasses of K4Pt2(I-IP04)4C12 are intensely luminescent. However. luminescence is not detected from solutions at temperature just above the gassing transition. These observations are consistent with recent photophysical studies demonstrating the importance of medium rigidity as a crucial controlling factor of do‘ luminescence [21.69.70]. Figure 12 displays the unpolarized excitation spectrum of low- temperature gasses of K4Pt2(HPOa)4C12 at 77 K. In ageement with the observed absorption profile. excitation bands corresponding to the proposed lieu.) -. do‘). lids -9 do‘) and 1(dir‘ -1 do‘) transitions are observed at 301. 360. 403 nm. respectively. Additionally. several other features not readily discerned by absorption spectroscopy are clearly distinguished in Figure 12. Most notable is the presence of a weak multicomponent feature on the lower mm tail of the 1(dx‘ -> do‘) excitation band that resolves into 465-nm maximum and a 505-nm 45 Figure 10. Corrected electronic emission spectra at 10 K of crystalline KnPt2(I'IPO4)4Iq:L=HO.n=-2( -------- ):L=Cl.n=4(-—):L=Br.n= 4 ( ------- ). INTENSITY l l l 800 900 X/nm 700 Figurelo 47 J. ed nos «.3 8:. s8 2e e \ on so one one oz. men was s \ .o no new a; was nun .. 8s a \ on: a can be . a 2 as bolas .ostbe s. \ a n1 \ p EVER} on? E:\ssfi.:u« Baku—sud} Sofieguficx ecsssaeosaeboe 35:53 so. see abooam 5.835 one 8:982 .a case Figure 1 1. (a) Electronic emission from Pt2(HPO4)4C124' in a low— temperature gass (77 K) of saturated LiCl(aq) and 1 M HCl. (b) Sean is over the same spectral regon of the same solution at a temperature ( 1 43 K) above the gassing temperature. Scan (b) was recorded at ten time 1:116 sensitivity of that of scan (a). INTENSITY 49 (o) (b) l 1 1 600 700 . 800 50 Figure 12. Unpolarized excitation spectra of 91211110414024- In a low- temperature gass (77 K) of 1:1 saturated IJCl(aq) and 1 M HCI. Progessively more concentrated solution were used irn order to obtain adequate signal to noise ratios for the spectral regions A. B and C. In each case the OD/ cm was maintained < 0.2 within the spectral region being studied. 51 AIISNBLNI 52 shoulder in spectra recorded on concentrated gasses of K4Pt2(HP04)4C12. The 3000 - 5000 cm“1 red shift of these weak features from the 1(d1t‘ -) do‘) transition are of the appropriate magnitude for singet-triplet splittings. In the absence of spin-orbit coupling only one transition. 3E“ 4— 1A13. is expected to arise from the one-electron 3(dir‘ -) do‘) promotion. However. the 3E1. excited state in 04h symmetry is 6- fold degenerate and will decompose into A1“. A2“. B1“. 82“. and En symmetries on the 0411' double goup [96]. First-order spin-orbit coupling calculations show that the degeneracy of the (Alu. A2“) and (B 1“. Ban) pairs does not split and that the may ordering of the spin- orbit components irncrease along the series (Bin. 82“) < F;u < (A1“. A2“) [98]. Of these spin-orbit states. transitions to the Eu and A2“ are dipole- allowed and expected to carry intensity because of singet-triplet mixing with energetically proximate 1E“ (arising from the 1(dx‘ --1 do‘) one- electron promotion) and IA“ (arising from the l(o(d12.l.) —> do‘) one- electron promotion) states. To this end. our observation of two weak transitions to lower energ of the 1(dx‘ -+ do‘) band is in accordance with straight forward spin-orbit considerations. and we assign the 465-nm and 505-nm excitation bands to the A2u(3Eu) (- 1.4118 and Eu(3Eu) .— 1111g transitions. respectively. In order to further elucidate the dynamics of the Pt2(HPO.i)414n" complex do‘ excited state. a detailed study of the temperature dependence of the luminescence was irnitiated afier observing a pronounced attenuation of emission from crystalline Ptzll'IPOthI/zn' complexes with increasing temperature. Figures 13 - 15 display the temperature dependences of the excited state decay rate constants of the potassium salts of Permanence", Hamilton" and 53 Figure 13. Fits of variations of (a) ”observed and (b) radiative decay rate constants of K2Pt2(I-IPO4)4(I{20)2 to eq 5 in the 10 - 290 K temperature range. The solid lines represent calculated values. (a) b P P 3 2 4| re .9 \ €528 >88 8 100 150 200 250 300 50 T/K To No. \Eeseoo 2mm 2.33m 100 150 200 250 300 50 T/K Figurels 55 Figure 14. Fits of variations of (a) observed and (b) radiative decay rate constants of K4Pt2(HPO4J4C12 to eq 5 in the 10 - 290 K temperature range. The solid lines represent calculated values. mm D b P m m w w w r... .2 Easeoo >88 .8580 150 200 250 300 T/K 100 50 P P D P P b 8 6 ‘ 2 o 8 1 re «a. \emfioo sec 333. 200 250 300 150 TIK Figure“ 57 Figure 15. Fits of variations of (a) observed and (b) radiative decay rate constants of K4Pt2(I-IPO4J4Br2 to eq 5 in the 10 - 290 K temperature range. The solid lines represent calculated values. 200 250 300 150 100 50 T/K rm ~2 €928 cam 38% 300 T/K Figure“ 59 Pt2(HPO4)413r24'. respectively: for all experiments. monoexponential decay of the emission was observed. An issue of principal interest is the temperature dependence of the Pt2(HPO4)414“' emission lifetimes. In each figure, a temperature regime in which the decay rate constant exhibits little variance is proceeded by a sharp monotonic increase of the rate with increasing temperature. These data clearly demonstrate the presence of a thermally accessible decay channel. As indicated by the solid linesinFigures 13-15. theobservedratesarefitextremelywellby an expression for the decay constant based on a two-state Boltzmann distribution. k1 + k2 GIN-AE/RBTJ k0“ ' 1 + exp(-AE/k3'l‘) (5’ where k; and k2 are the decay constants for two states in thermal equilibrium that are separated by an energy gap AE. Calculated rate constants and energy gaps for the Pt2(HPO4)414n' complexes are summarized in Table 2. For each of the phosphate complexes. examination of the data shows that an extremely emcient decay pathway is accessed via the higher enery excited state. This observation is further quantitated by combining temperature-dependence lifetimes with measurements of the corresponding temperature-dependent quantum yields. At a given temperature. the quantum yield. (be. is related to the observed lifetime by the radiative rate constant kt. by the well known expression [98]. ¢e ' to 1‘r (6) m 8 Bee Banana a a: x as. «2 x a.” as: m2 x a; x: x ed :8 s in «2 x ad «2 x me an: a: x a." «S x as 88 e \ 6 a: x 3 «2 x 3 new 52 x «a so“ x we on: a 3m: Tm \ «x 73 5 Tee \ m< To}: To \ E 7:832 : \ a 8:888 3am 055mm 8:885 8am gas—e502 stgagsx 33298850... 38:53.. 05 so 85 925 one 3:888 sex .889 gasesaoz use 253% 833930 .a sea. 61 from which the nonradiative decay rate km- may directly be determined, to = (kr4'knrlnl (7) The nonradiative decay rate constants are obtained by subtracting radiative decay rate constants from the observed decay rate constants. Because the observed quantum yields are ~2 x 10'2. and the observed lifetimes are of the order of 10‘5 s'l. the nonradiative decay rate constants are always 2 - 3 orders of magnitude faster than radiative constants. For example. representative measurements of temperature dependent quantum yields. lifetimes and calculated radiative and nonradiative rate constants of K4Pt2(HPO4)4Clz are displayed in Table 4. In the temperature-independent lifetime regime. k;- is 6.6(4) x 102 3'1 whereas km- is 3.3(6) x 104 s"; and this difl'erence between the radiative and nonradiative rate increases to 103 at room temperature (it,- s l .8 x 103 8']. km . 1.4 x 106 3'1 at 290 K). As noted by the much larger difl’erence of the radiative and the nonradiative rate constant for the higher energ state. its assimwt as a nonemitting state is reasonable. The anally substituted water and bromide complexes exhibit behavior parallel to that of the chloride (Tables 3 and 5) Tire significant increase in the nonradiative rate at higher temperatures is indicative of the additional contribution of this higher enery excited state to the decay of electronically excited Pt2(HPO4)4I/z“" complexes. The radiative decay rate constants also show behavior parallel to the nonradiative rate constants. The fitting results for the radiative rate constants are shown in Table 2. The energ gap obtained from the fitting of radiative rate constants (1729 cm’l) is smaller than the energy gap 62 Table 3. Temperature Dependent Quantum Yields. Lifetimes and Calculated Nonradiative and Radiative Rate Constants of Crystalline K2Pt2(HPO4)4(HzOl2 'r / K «be/10'<3 t/us limp/10584 1r,/102s'1 8.0 1.9 2.0 5.0 9.3 17 1.9 2.0 5.1 9.5 27 1.8 1.9 5.2 9.4 38 1.8 1.9 5.3 9.5 48 1.7 1.8 5.4 9.1 58 1.6 1.8 5.5 9.0 69 1.6 1.7 - 3.3 9.4 79 1.4 1.6 6.3 8.7 90 1.4 1.6 6.5 8.9 101 1.4 1.5 6.8 9.3 122 1.2 1.3 7.5 9.1 143 1.1 1.1 8.8 9.6 164 0.99 1.0 10 10 185 0.87 0.82 12 10 206 0.79 0.79 14 11 227 0.70 0.69 17 12 248 0.63 0.52 19 11 269 0.51 0.43 23 12 290 0.40 0.33 30 12 Table 4. Temperature Dependent Quantum Yields. Lifetimes and Calculated N onradiative and Radiative Rate Constants of Crystalline K2PtziHP04l4Cl2 T / K cue/10'2 r/us lint/1048.1 k,/ 10%"1 8.4 2.4 35 2.8 7.1 17 2.5 34 2.8 7.3 27 2.4 36 2.7 6.6 38 2.5 36 2.7 6.7 48 2.3 32 2.8 6.9 58 2.2 32 3.0 6.9 69 2.0 31 3.1 6.4 79 1.8 28 3.4 6.2 90 1.6 24 4.1 6.7 101 1.5 21 4.7 7.0 122 1.2 16 5.9 7.2 143 0.91 12 8.2 7.5 164 0.63 8.3 12.0 7.5 185 0.46 5.5 18.0 8.3 206 0.36 3.8 26.6 9.5 227 0.27 2.5 40.4 10.5 248 0.20 1.6 62.4 12.7 269 0.14 1.1 95.1 13.2 290 0.13 0.7 140.7 18.1 Table 5. Temperature Dependent Quantum Yields. Lifetimes and Calculated Nonradiative and Radiative Rate Constants 0f Crystalline K2Pt2(HP04)4Br2 T / K 0.3/10'3 this ling/1048.1 1r,,/102srl 8.1 5.2 37 2.7 1.4 17 5.2 38 2.6 1.4 27 4.9 35 2.8 1.4 38 4.6 37 2.7 1.3 48 4.1 35 2.7 1.1 58 3.8 32 3.1 1.2 69 3.5 28 3.5 1.2 79 3.1 " '26 3.8 1.2 90 2.8 26 3.9 1.1 101 2.6 22 4.5 1.2 122 1.9 17 5.9 1.1 143 1.4 12 8.9 1.2 . 164 1.1 7.6 13 1.4 185 0.67 4.4 23 1.5 206 0.48 1.8 53 2.5 227 0.35 1.3 79 2.8 248 0.22 0.8 130 2.8 269 0.15 0.5 190 2.8 290 0. 12 0.4 260 3. l 65 from nonradiative rate constants (2256 cm'l) of the chloride derivative. A similar trend is observed for axially substituted water and bromide complexes. The discrepancy between the enery gaps as calculated from the radiative and the nonradiative rates can be explained by a vibrationally accessed nonradiative deactivation pathways that are not accounted in the simple two level Boltzmann scheme. An excited state model consistent with the spectroscopic and photophysical properties of Pt2(HPO4)41/z“’ complexes is illustrated in Figure 16. The lowest enery excited state manifold is derived from the 3(dir‘do‘) electronic configuration: luminescence originates from the (B 1“. 82“) spin-orbit pair. Because there are no near-lying 1B1“ and 1B2“ states to conflgurationally mix with this pair. the emissive excited state is predicted to be nearly pure triplet in character. To this end. the model accounts nicely for the very low radiative rate constants (k, s 102 ~ 103 8’1 l for the Pt2(HP04)4l/3n" series. We attribute the observed temperature dependence of the 32011300414” complex lifetimes and emission quantum yields to thermal population of the Edam excited state. It is noteworthy that the energ gaps determined from temperature-dependent lifeume measurements (Fable 2) conform well with the splitting of (Blu, B2“) and En determined from first-order spin- orbit coupling calculations [97]. The calculated 952 cm'1 splitting is in order of magnitude agreement with the observed enery gaps. The larger energy gaps of Ptza-IPOeul/zn’ complexes axially substituted by halide ligands can be accounted for by increased second-order contributions of the halides. as compared to water. to the spin-orbit coupling constant. Additionally. enhanced nonradiative decay from the higher enery deactivating state is also explained by this model because intersystem 66 Figure 16. Proposed energ diagram of the lowest energy excited states of the Pt2(III,III)L2 phosphates. The state manifold is derived from the spin-orbit coupling perturbation of the 3Eu state arising from the one- electron 3(dn‘ —-> do‘) promotion. ed 5835 from the the one- 1 A2u(LO d0*) 1Eu(d1t*d0*) assessor) 67 (BiuvBZU) 68 crossing from the Eu(3Eu) state with respect to the lowest energy (B “(38.1). B2u(3Eu)) level will be promoted by the increased singlet character of the former state arising from efficient mixing with its energetically proximate singlet. Moreover. the Eu(3Eu) state is possibly split by the J ahn-Teller efl'ect and may therefore exhibit a larger non totally symmetric distortion than the other spin-orbit states [99] . Such distortions are usually effective in promoting efficient nonradiative decay. Moreover the calculated radiative rates are also consistent with this model. The radiative rate constant (k,) in eq 8 is directly related to the oscillator strength (i) by eq 9. k, . 2.88 x 10‘9 n2 «‘9;1 fijfl‘fl (8) 8n 9 f = 4.32 x 10'9 civldv (9) where 9. Vf, e. n and g1 / gu represent the emery in cm‘l. the emission energr in cm'1 . the molar extinction coemcient. the refractive index of the medium and the ratio of the degeneracies in the lower and upper states, respectively [100]. The radiative transition from the Edam state to ground state is symmetry allowed. whereas that from (B1u(3Eu), 82338.,» is strictly forbidden. Thus the radiative rates of the 8338..) -) lag transition should be larger than that of the (131.438,). 132.438..» -r lAstransition. Thisisobservedin'l‘able2where the radiative rates of the former transition is 102 -103 larger than that of the latter. Thus the excited state dynamics of the Pt2(HP04)4I/3“’ complexes clearly reveal the existence of emcient deactivation pathways of do“ excited states even when metal-metal 0r metal-axial ligand dissociation is 69 precluded. Our results suggest that the spin-orbit components of the 3(dit‘dtr‘) configuration controls the dynamics of do‘ emission from the Ptza'IPO4)4I/2n- complexes. Specifically. the Eu(3Eu) excited state proximate to the lowest enery emissive (B1u(3Eu). B2u(3Eu)) level provides a facile decay channel to ground state. This particular pattern of energ level splitting arises from the specific molecular orbitals involved in the one-electron promotion. whereas the specific values for the splitting is controlled by the nature of ligation sphere. To this end. it is reasonable to hypothesize that the spin-orbit interaction scheme should be applicable to other singly bonded emissive metal dimers where the HOMO and LUMO are of dx‘ and do‘ parentage. 2. Diplatinum Pyrophoephitee The singly bonded tetrakispyrophosphite complexes. Pt2(P205H2)414“' (n s 2. L . H20; n a 4. L a Cl or Br) possess the highly symmetric D41, ligand arrangement about the bimetallic core as is found in the PtzfliPoslmn' complexes. Extensive spectroscopic studies have led to a complete assignment of the absorption manifold [82-85]. All transitions populate the do‘ orbital and the lowest energy excited state is dx‘do‘ (Figure l 7). In addition. recent spectroscopic studies have revealed that the emissive excited state of Pt2(P205H2)41/zn' complexes possesses typical do‘ characteristics [69). We now report that the temperature dependence of the lifetime of Pt2(P205H2)4Br24' accords well with the spin-orbit coupling scheme proposed in Section C. l . The decay constant of the higher energ excited state is six orders of magnitude larger than that of the lowest energy 70 Figure 17. Electronic absorption (———) and emission ( --------- ) spectra of K4Pt2(p0p)4Br2 in deoxygenated aqueous solution at 300 K. . 71 Emission Intensity I ’0 'o a " "e ’ -— -’ ’v " 1” I ------—’ hisuolul uogdiosqv 1000 800 Mnm Figure 17 72 excited state (In =1.0 x 105 8'1. k2 = 2.8 x 1011 s"). with an energy gap of 4300 cm‘1 (Figure l 8). The better overlap between diplatinum core and the phosphorus of the pyrophosphite ligand as compared to the oxygen in the phosphate ligand will facilitate greater electronic interaction and presumably greater spin-orbit interaction than observed for the Pt2(III.III)Iq tetraphosphates. in addition. the higher degree of covalency in the axial Pt—Br interaction [94] will also play an important role in increasing the spin-orbit coupling contributions. Therefore the larger enery gap between Eu(3Eu) state and emissive (B 1u(3Eu). Bgu(3Eu)) states in Pt2(IIl,lII)Lz tetrakispyrophosphites. and the large difference between decay constants of radiative and non-radiative states in these complexes is in accordance with the proposed spin-orbit coupling model. 3. Dirhodium Fluorophoephinee The spin-orbit coupling model for do‘ luminescence is not necessarily confined to singly bonded metal-metal dimers that belong to high molecular symmetries. such as the diplatinum phosphates and pyrophosphites. The related series of dirhodium bis(difiuorophosphino)- methyamlne compounds. Rh2l(F2P)2N(CH.3l13(PF3)2 (Rh2(0.0ll. halszPlleCHslhlPFlez (12112013an and - Rh2l(F2Pl2N(CH3)13X4 (hamJnxltl (X 2 Cl or Br). synthesized in our laboratory. display strong red luminescence (20.2 1]. Previously published crystallographic analysis of Rh2(0.0). Rh2(II.0)C12. haflI.II)Cl4 [21] and more recently. Rh2(ll.II)Br4 [10 1). reveals trigonal bipyramidal and pseudo-octahedral coordination in this series of compounds about Rh(0) and Rh(II) centers. respectively. 73 Figure 18. Fit of the variation of the observed emission decay rate constant of K4Pt2(pop)4Br2 to eq 5 in the 10 - 290 K temperature range. 74 § 1 l l 1 L g; g s 8 8 F P PS 901 Iluelsuoo K5060 pe~esqo 1 00 1 50 200 250 300 50 UK Figurels 75 Representative structures summarizing the X-ray results for the compounds are shown in Figure 4. Three phosphorus atoms fiom twisted fiuorophosphine bridges occupy the three equatorial sites of each center and one more phosphorus atom from axial PF3 for Rh(0). and two halides in equatorial and axial positions for Rh(II) complete the coordination spheres. Because of these difi'erent ligation geometries of metal centers. the molecular symmetries of D3 (Rh2(0.0)). C, (Rh2(ll,0)X2) and C2 (Rh2(II,II)X4) represent a significant lowering from the D4}, symmetry observed for many M—M complexes. All crystallographically assessed dirhodium complexes display characteristic bond lengths of a Rh—Rh single bond (2.707 A ~ 2.841 A). originating from the overlap of dfi orbitals from each metal center [19a]. The simplified molecular orbital diagrams of the dirhodium series are presented in Figure 19. Molecular orbital treatments show that eight electrons of the d9 Rh(0)P4 fragment reside in orbitals of awn. d,,) and 5(dxy. dx2.y2) symmetries. with the remaining electron occupying the o(dzz) orbital. In the case of the d7 Rh(IDP3X2 fragment. the d32.y2 orbital is displaced to very high energ owing to the destabilizing 0‘ interaction of the metal with the ligands in equatorial plane. Consequently. the odd electron of the Rh(lI)P3X2 fragment. as was the case for Rh(0)P4 fragment. resides in the o(d,,2) orbital. with the remaining six electrons residing in lower energ x(dn. dyz) and 5(dxy) orbitals. The enery level diagrams are constructed by the orbital mixing of the appropriate rhodium fragments. In each case. the single bond formation results fiom the pairing of the odd electron residing in the 0(d12) orbital. In this regard the (daldl—dlma) Rh2(0.0) and (dald 1— d1(d6) Rh2(ll.II)X4 compounds are isoelectronic with the M—M prototypes 76 Figure 19. Qualitative energr level diagrams of (a) 1012“?!“ 2)2NCH:313(PF 3’2. “3) hallpF 9)2NCH313(PF3)012 and (cl Rh2[(PF9)2NCH3]3Cl4 generated by proper fragnents (C3v Rh(0)P4 fragment for Rh(0) center and C4v (Rh(IDP3C12 fragment for Rh(II) center). The dii- and d8- symmetry orbitals are filled and indicated by the shaded box. Low symmetry splittings within the tilt level of the C4" Rh(IDP3C12 fragment are not considered. 78 002(00):; (102) and Mn2(CO)lo [46] respectively, with the (d9)d1—dl(d3) Rh2(0.II)X2 complex formally completing the Rh—Rh single bonded series. The electronic absorption spectrum of Rh2(0.0) provides a benchmark for interpreting the electronic absorption spectra of this series of singly bonded metal-metal compounds. The absorption spectrum is typical of M—M compounds with an intense 305-nm band attributable to the allowed do -l do‘ transition and the less intense band at 440 nm consistent with dx‘ —) do‘ excitation [8 l .82]. The high optical electronegativity of the terminal PF3 ligands diminishes configurational mixing of the do and o(L) orbitals. which has previously been observed for a variety of metal complexes containing halides or pseudohalides coordinated axially to the M—M core (44b.85.88]. Yet the importance of configurational miidng in this series is readily apparent with the comparison of the chloride and the bromide spectra of Rhgm,ll)x4 and Rh2(0.lI)X2 compounds to that of Rh2(0.0). The absorption profiles of the d7—ti7 dimers 18201.1an and ham.II)Br4 are more congested in the UV spectral region than their Rh2(0.0) counterpart. Though the Rh—Rh separation difi'ers by no more than 0.1 A (d(Rh-Rh) . 2.841 A. 2.701 A and 2.750 A for Rh2(0.0l. Rhgm.II)Cl4 and RhgfllJDBr). respectively) in this set of complexes. no absorption band comparable in enery to the 305-nm do -) do‘ transition of Rh2(0.0) is observed. However. intense transitions to higher and lower energy of the Rh2(0.0) do -> do‘ absorption are present (rm/nor (e/M'l era-1) = 265 (18700) and 355 (24500) for hamrnch; (rm/nor (c/M'l era-1) . 290 (19600) and 395 (I 1400) for Rh2(II.II)Br4). Notably. these transitions exhibit a ~3000 cm‘l red shift with the 79 replacement of chloride by bromide. In contrast. the energ of the lowest energy band is energetically much less sensitive to the halide substitution am / nm (e / M‘ 1 cm' 1) s 445 (9430) for Rh2(II.II)Cl4; lmax/nm (e /M'1 cm-l) = 462 (11700) for Rh2(ll.lI)Br4). A similar trend is observed for the Rh2(0.II)X2 compounds. The absorption profiles show bands in energ regions similar to Rh2(II.II)X4 and a red-shift is observed upon bromide substitution. This result is not surprising in view of the similarities of the frontier molecular orbitals of the d7—d7 and d7—d9 dirhodium compounds. These spectral trends are strikingly similar to those of the ham.II)T'MB4I4““ (T'MB = 2,5-dimethyl-2.5-diisocyanohexane); L 2 H20. CH3CN (n s 4): L a Cl. Br (11 a 2)) compounds. In the cases of L 2 H20 or CH3CN. a single intense absorption is observed at 308 nm and assigned to the do -) do‘ (44b). The considerable red-shin of the do -> do‘ transition upon substitution of L by axial chloride and further red-shift with bromide (~2200 cm’l) is consistent with extensive mixing of the metal do orbital with the axial o(L) orbitals of the halide substituted compounds. As expected from a configurational interaction scheme. the ligand-to-metal charge transfer transition. o(L) -) do‘. appears to higher energ of the do -) do‘ transition. The lowest energy dx -l do‘ transition shows only a marginal halide dependence. red—shifting only slightly along a halide series. Despitethesimilaritiesintheenergyofthespectralprofileof Rh2(0.II)X2 and Rhgtll.II)X4 to other M—M dimers. there are notable difi'erences. For a typical M—M compound the spectrum is dominated by do -) do‘ and the higher energy configurationally mixed 000 -> do‘ band exhibits weaker intensity: the weakest transition is usually dx —9 do‘. 80 Conversely. the intensity of the three bands in the absorption manifold of the Rh2(0.II)X2 and Rh2(II.II)X4 complexes are of comparable intensity. Undoubtedly this result is due in large part to the lower molecular symmetry of the dirhodium complexes as compared to D4}, symmetry of typical M—M dimers. In these lower symmetry molecules. configuration interaction between metal and ligand based orbitals will be more extensive. Hence it is dificult to make direct correlation between dx and d5 orbital symmetries in D411 dimers with those of the Rh2(0.II)X2 and Rh2(II.II)X4 compounds; and definitive assignments for the dirhodium fiuorophosphine series based on the spectral trends of higher symmetry D411 M—M complexes is tenuous. Nonetheless. as proposed in Table 6. the band shapes and energ trends of the absorption profile observed in this dirhodium series are consistent with transitions arising from the promotions of configurationally mixed o(L) and do orbitals to do‘ and 8 lowest energy transition of (dx‘.d5‘) -) do‘ parentage. The luminescence properties of the dirhodium compounds further support an electronic structure dominated by a M—M parentage. Table 6 lists the emission energies and lifetimes (l‘ a 77 K). The luminescence spectral features are commensurate with the emerging characteristics of do‘ luminescence. The full-width at half height of the emission bands exhibit a substantial temperature sensitivity. increasing by more than 1500 cm“1 from 10 K to the highest temperatures at which emission can be detected for each complex. Moreover. luminescence is not detected fiom solutions at temperatures equivalent to those at which the crystalline solids emit. which confirms the importance of medium rigidity as a crucial controlling factor of do‘ luminescence [21.22.69.70). 388 belt... 585 B 885 855.8 5.858 s 81 ms Sm con man «we Emma—=32 com can new man 33. 305.882 82 8a man can a: «58.593 855m. 98 2m can men can «68.898 2. new mom 2% 8.993 . bulge bellow Earlene an? E: x .88 89¢ 0:509:00 E: x 8853‘ 8.0558860 oefiamosnuoafi 82023 055885 be 555585 .359on .o 32 82 Several data suggest that the state parentage of the do‘ luminescence arises fiom the promotion of an electron fiom the [(dx‘.dx‘). do‘] manifold that is triplet paired. First. the nature of the halide afi'ects the emission energy only marginally across the series. This enery insensitivity is consistent with the expectation of only small configurational mirdng of the do‘ orbital owing to the large enery gap between o(L) and do‘ orbitals. The slight energy dependence of the lowest energy absorption manifold. as observed in absorption spectra. also supports this assignment. Second. excitation spectra of the dirhodium series suggests that the emission arises from the spin forbidden states of this lowest enery manifold. The excitation spectra for Rh2(0.lI)C12 and marina. are shown in Figure 20 and Figure 21. respectively. Whereas peaks are energetically coincident with absorptions in the high enery absorption manifolds of Rh2(0,ll)C12 and RhgflLIDCl4. additional features to lower enery in the excitation profile are observed. In the case of Rh2(0.II)Cl2. this peak dominates the excitation spectrum as shown in Figure 20. For each dirhodium compound. the tail of the low energy profile overlaps the higher enery tail of the emission profile. As is the case for the Pt2(III.III)I4 tetraphosphates. such features are energetically consistent with states arising from a 3(dit‘do‘) spin-orbit manifold. Perhaps the most compelling evidence in support of this supposition comes from the time resolved luminescence properties of the dirhodium series. All the time-resolved emission experiments show biphasic decay with at least two very difi'erent lifetimes. The shorter lifetime component persists up to room temperature. even when no appreciable emission is detected by steady state luminescence measurement. Scattered laser 83 Figure 20. Excitation spectrum at low temperature (77 K) of crystalline Rh2[(PF9)2NCH3]3(PF3)C12 complex . cam own an earn E: \ 59.2852, coamgoxm owe 8v own in: Misuoiul 85 Figure 2 1 . Excitation spetrum at low temperature (7 7 K) of crystalline Rh2[(PF2)2NCH3]3Cl4 complex. one can 3 2:3.— Ec \ 598.963 cowgoxm one owe own d hisualm 87 light and / or the fluorescence component superimposed over the phosphorescence decay curve can not be overlooked. In fact. the decay profiles of fluorescence from Rh2(0.0) and Rh2(II.II)Cl4 (Met a 550 nm for Rh2(0.0) and 580 nm for Rh2(II.Il)Cl4) have been observed and decay with very short lifetime (~ l (1880 or less). On the other hand. the longer lifetime component shows a monotonic decrease of lifetime upon raising the temperature until no appreciable intensity in steady state emission spectra is observed. Figures 22-26 depict the temperature dependence of the longer lifetime component of the Rh2(0.0). Rh2(0.II)X2 and Rh2(lI.II)X4 series. With increasing temperature the lifetime exhibits a temperature independent regime followed by a monotonic decrease with increasing temperatures. The temperature dependence accords well with a two- state Boltzmann distribution (eq 5) in which a state possessing facile decay to ground state is accessed with increasing temperature. Calculated rate constants and energ gape (the fits for which are indicated by the solid lines) for the dirhodium series are summarized in Table 7 where k1 and k2 are the decay constants for two states in thermal equilibrium and AF: is the energ gap. The lifetimes and observed rate constants for all five compounds are provided in Table 8. This Boltzmann temperature dependence is analogous to that observed for the singly bonded PtzflflJIDIq tetraphosphates [22]. The interesting issue here is that the important features of the spin-orbit model in describing do‘ luminescence for D41, complexes is preserved for the lower symmetry dirhodium series. Figure 27 shows the correlated state energ level diagrams in accordance with the less symmetric D3, Cs. and C2 molecular symmetry point groups of the Rh2(0.0). Rh2(0.lI)X2 88 Figure 22. Fit of the variation of observed emission decay rate constants of Rh2[(PF2)2NCH3](PF3)2 to eq 5 in the 10 - 160 K temperature range. Steady-state emission shows no significant intensity above 1 60 K. 89 200 150 e 100 50 t5s ,01/luelsuoo timed pemsqo T/K Figure22 90 Figure 23. Fit of the variation of observed decay rate constants of Rh2[(PF2)2NCH3]3(PF3)C12 to eq 5 in the 10 - 190 K temperature range. Steady-state emission shows no significant intensity above 190 K. 91 0. N l «a c. in P k:3 ,01 nueisuoo Kweo pemsqo 200 150 100 50 T/K W23 92 Figure 24. Fit of the variation of observed decay rate constants of Rh2[(PF9)2NCH3]3Cl4 to eq 5 in the 10 - 290 K temperature range. 93 b p )- t:5 ,01 llUBlSUOO £9990 96419qu 100 150 200 250 300 50 T/K We“ 94 Figure 25. Fit of the variation of observed decay rate constants of Rh2[(PF2)2NCH3]3(PF3)Br2 to eq 5 in the 10 - 190 K temperature range. Steady-state emission shows no sigiificant intensity above 190 K. 95 l l l O. V O. 0. 0. {'0 N v- t:5 ,01 numsuoo Ameo Demosqo 200 150- 100 50 T/K W25 96 Figure 26. Fit of the variation of observed decay rate constants of Rh2[(PF2)2NCH3]3Br4 to eq 5 in the 10 - 250 K temperature range. Steady-state emission shows no sigiificant intensity above 250 K. 97 N ’ O C a 1 a L 1 .— ‘ O O CO V’ N O P l.8 ,0) liueisuoo £3990 pemesqo 100 150 200 250 300 50 T/K Figure26 Table 7. Calculated Decay Rate Constants and Energy Gaps of Dirhodium Fluorophosphine Complexes.a Compound AE / cm'1 k1 / s-l k2 / s'l Rh2(0.0) 900 1.1 x 104 1.1 x 106 Rh2(II,0)C12 1060 2.0 x 103 3.7 x 105 Rh2(ll,0)Br2 1470 5.4 x 103 2.2 x 106 Rh2(II.II)Cl4 1070 3.0 x 103 2.8 x 105 Rh2(ll,ll)Br4 2190 1.3 x 104 1.6 x 108 ‘1 Calculated from eq 5 - - no on - - - - - - can - - on em - - - - - - men .. ...m on an - - - - - - 2.8 me m. an we - - - - - - Sn 8... mm m. 8 - - - - - - 8a 8.8 on n. as an mm m. 8 - - mm. 6.8 8... ms 8. m. me o. 8 an m. 8. 8.. an a... 6... .. mm 8.4. on. m... on me. n. me 6... can a... on. an 8. 2.8 on an. .4. as 8.... can 8... on. m... can 8.. 8 .o. .... 2. m... 88 ..m 8. ma Sn 8.. 8 on 8.. me. an 8.. en 8. on Eu 8... 3. ms 8.. ..m a... c... on 88 a... can ...... 2. me 8.. .m ..m can an om. ..n on... 8.. as an ... mm a... 6.... on om. .... can 8.. 8 me 8.. cm on o..." or... com on can 6.. 8 mm ... mm ad 98 a... 6.8 2.8 8.. ... 8 R ... .3 ha can 8... Sn m... can 8.3 8. ... ... cm 8.8 8... on con ...8 8.. 8.. mm 6. Theater. 2.} Tomatoes. 2.: Tennis... 95 Toasts... 9.} insets... 85 a: 39......9... s5......55. «58....«5. «.08....«5. 6.28.... 3558800 oSthonqouon—E 55:00.3 658255 2an .o 3:388 35. 3.38.3852 888.830 use 88.38... 5.3588 83.885... .3 8...... 100 Figure 27. Proposed energy diagrams of the lowest energy excited states of Pt2(lll,lll) tetraphosphates and of Rhg fluorophosphine complexes modified from the energy diagram of Pt2(III,lIl) tetraphosphates. The state manifolds are derived from the spin-orbit coupling perturbation of the 1-3Eu states arising from the one-electron dx‘ -) do‘ promotion. Point goups. D3. C. and C2. describe the molecular symmetries of Rh2(0.0). Rhg(ll.0)ngand Rh2(II.II)X4. respectively. 101 S can... (P .

5) < .< 3...: Games. . me n .< m ..m rocked. < .< m .8<.2$I.II m ..< m ..m lllll ropes am. < .< 8 8< IIIIII ..eo o... a... No 5 .o e. an e. ...... e. 102 and Rh2(ll.II)X4 compounds. respectively. An important result from the previous study of the Pt2(III.III) tetraphosphates is that the lowest energr (B1u(3Eu). Bzu(3Eu)) excited state remains unique with respect to the Eu(3Eu) excited state. Thus singet character from the 1 [(dit‘.dx‘). do‘] configuration can not be introduced into the lowest energy excited state by a spin-orbit mechanism. Hence the lowest energ' excited state in all molecular symmetries should remain pure in triplet character. Our experimental results supports this conclusion. The experimentally determined decay constant of the higher energ. thermally accessible excited state is 102-103 geater than that of the lowest energy state. Additionally. the calculated energy gap monotonically increases with increased halide substitution (AE(Rh2(0.0) < AE(Rh2(0,II)X2) < AE(Rh2(II.ll)X4)). which is expected to add covalent character of a heavy atom to the o manifold and in turn increase the efi'ective spin-orbit interaction. In this regard;—.. as has been showrr for Pt2(III.llI)I/z tetraphosphates. the enery gaps for the chloride complexes are less than that of their bromide counterparts. Similarly, this increased spin- orbit coupling for the bromide complexes is also plainly evident fi'om the fact that k2/k1 ~ 103-104 as compared to only 102 for the chloride compounds. ' ~’ ’ Thus luminescence from M—M compounds exhibits a characteristic temperature dependence that arises from spin-orbit splittings of lowest enery excited states arising from a do‘dx‘ configuration. The spin-orbit coupling mechanism for nonradiative decay of do‘ luminescence. originally proposed for D411 M—M complexes. is applicable to singly bonded bimetallic compounds with lower molecular symmetry owing to the fact that the triplet character of the 103 lowest energy spin-orbit state is preserved and can not configurationally interact with energetically proximate states of singlet character. It is not yet clear if spin-orbit mechanisms play an important role in the excited state dynamics of M—M compounds whose emitting state is not from a 3(dir‘do‘) parentage. For instance. the emitting state of the R82(CO)6(P Hg (P P = bridging phosphine) has been assigled to 3(0 -) do‘) [103]. The contribution of spin orbit effects in governing the dynamics of luminescence. if any. in this class of singly bonded M—M complexes remains to be defined. CHAPTER III M AND WITCH OP LAYERED METAL PHOSPHATES mamwrm mum A. Background The existence of two-electron quadruple (Mi-M) and single bonded (M—M) mixed-valence systems provides the underpinning to the development of a new generation of photoactive solid state materials. The multielectron activity of these compounds may be preserved inside of the solid state support as long as the. metal-metal interaction is not disturbed. The intercalation reaction of bimetallic cores into a layered material is particularly interesting not only structurally but also electronically. The interlayer gallery of a layered host material can provide a well organized space for bimetallic compounds to reside. with possible shape selectivity furnished by the host structure. The electronic band structure of the extended layer as an electron sink or source may also play an important role in stabilinng photoreduced or photooxidized bimetallic cores. As discussed in Chapter I. intercalation reactions of bimetallic cores into layered compounds containing phosphate goups. such as 104 105 vanadium and niobium phosphates. potentially provides coordination environments for bimetallic cores. These layered metal phosphates can be synthesized in highly pure and crystalline forms with many difierent metal ions composing the layer including V. Nb. Mo. Ta [35.36.104-106]. Layered metal phosphates can be easily obtained by reacting metal oxide. halide or oxychloride with phosphoric acid under reflux conditions. The resulting layers display high stability toward temperature. irradiation. acidic media and most organic environments. High acid stability is crucial to carrying out intercalation reactions of bimetallic compounds in aqueous solution in that decomposition of bimetallic core to high-valent oxo or hydroxo compounds [107] is prevented. The vanadium phosphate structure. shown in Figure 28. is best described as an array of alternating M(V)03 octahedra and P04 tetrahedra which share their corners. Specifically. four equatorial oxygens of the MMOB are provided by four different adjacent phosphate goups. The apical positions of the M(V)03 octahedron are completed withaterminalMuObondandacoordinatingoxygenfromone oftwo gallery waters. This corner sharing of phosphates to~the M(V) center is extended in 2-dimensions to produce sheets of (HgOMOPOQn-Hgo that are stacked in the crystallogaphic c-direction [108). The second water molecule resides in the gallery between phosphate groups of adjacent layers. Because the central metal ions are in their highest formal oxidation state M(V). the reduction of these centers is a facile chemical process. in fact. metal centers in. the M(IV) state are one of the most common defects in VOPO4 type layered compounds [109]. Since the layer does not carry any exchangeable ions. incorporation of compounds 106 Figure 28. Schematic representation of (a) 3-dimensional anhydrous VOPO4 and (b) layered VOPO4 dihydrate. The coordinating water molecule is shown as shaded spheres and the interlayer water molecule was omitted for clarity. 107 Figure28 108 into VOPO4 type layers proceeds by true intercalation where reduction of the layer to maintain the charge neutrality is accompanied by inclusion of positively charged compounds [1 10]. Intercalation can be achieved by the two different routes shown in Scheme 2. MOP04( Msv or Nb) MPC .. MOPOAMPC), NaBH ‘ > M0P04(Na),. MPC Scheme 2 Direct reduction of the layers can occur by the guest molecule with concomitant incorporation of the oxidized multielectron photoactive center (MPCt) shown in the top route. Alternatively. reduction may be promoted by an ancillary species such as Kl or Nam-l4 with introduction of Na+. which can subsequently be exchanged by the desired guest species. Introduction of guest molecules is thus accomplished by ion- exchange. Regardless of the insertion mechanism. the final product should be identical or. at least. very similar. Efi'orts to identify the resulting new layered materials by the intercalation of dimolybdenum and dirhodium cores ligated by acetonitrile into vanadium and niobium phosphates are reported. 8. Experimental 1 . Synthesis of Guest Compounds 109 a. General Procedures. Solvents used for syntheses were dried by refluxing at least overnight with the proper drying agent under nitrogen atmosphere unless specified otherwise. Starting materials were purchased from either Aldrich Chemical Co. or Strem Chemical Co. and used as received. b. Rh3(CfiacOO)4(Cfiaofi)3. Dirhodium tetraacetate. Rh2(CH3000)4. was prepared by gentle reflux of RhCl3-3H20 (l g. 2 mmol) and sodium acetate trihydrate (2 g. 15 mmol) in a mixture of glacial acetic acid (20 ml. 350 mmol) and absolute ethanol (20 ml) [1 l 1]. After 3 hrs of reflux under argon atmosphere. the reaction mixture was cooled to room temperature and filtered to collect a dark-geen powder. The powder was dissolved in a minimum amount of boiling methanol (~200 ml) and the insoluble residue was removed by filtering. The resulting solution was concentrated to l / 2 of its volume and stored in the refiigerator overnight to yield blue-geen crystalline Rh2(CH3COO)4-2CH30H. More compound is obtained by repeating the recrystallization step. The compound was stable enough to be kept on the bench top for extended periods of time. c. RhfiCHgCN) 10(BF‘)... All reactions were performed in a Schlenk-tube with finger tip condenser under slightly elevated pressures of argon by equipping an argon gas line with a mercury bubbler (l 12]. The methanol adduct of Rh2(CH3COO)4 (0.2 g. 0.5 mmol) was dissolved in dried. deoxygenated CH3CN (20 ml) to give a characteristic purple solution. The alkylating reagent (CH3CHzl3OBF4 (10 ml, 1 M solution in CH2C12) was added to this solution by syringe. The reaction mixture was 110 heated in an oil bath (68°C - 70°C) for 7 - 10 days. Rod-like crystals. which were deposited on the bottom of the reaction vessel over time. were separated from solution by decantation and washed with a 1:3 mixture of CH3CN and CH2C12. More Rh2(CH3CN) 10(BF4)4 was precipitated in powder form fiom cooled solution by direct addition of CH2C12. The precipitate was filtered and washed with three portions of 1:1 mixture of CH3CN and CHZClz. Both crystals and powder displayed identical uv/vis and ESCA spectra. The reaction scale could not be increased to more than 200 mg due to the lengthy reaction period. d. Mo¢(CfiscN)g(BF4)4. . . Similar to the synthesis of Rh2(CH3CN) 10(BF4J4. the acetate ligands of M02(CH3COO)4. which was synthesized by the standard methods (113).: were esterified by (CH3CH2)30BF4 [1 14]. A suspension of bright yellow. crystalline Moglcriacoolr (0.2 g. 0.5 mmol) in deoxygenated CH3CN (15 ml). was charged with 10 ml of a l M CH2C12 solution of (CH3CH9)3OBF4 to produce a wine-red colored solution. The reaction mixture was refluxed under argon atmosphere for three clayey The mixture was cooled in an ice-bath to yield the product. which was filtered under a blanket of argon. mshed with three. 10 ml portions of CH2C12. and dried under vacuum. The resulting blue powder was air-sensitive and decomposed within minutes on the bench top; decomposition of the compound was accelerated by the presence of moisture in air. 2. Synthesis of Phosphate Host Layers 111 a. Layered Vanadium Phosphate Dihydrate. Layered vanadium phosphate was obtained by J ander's procedure [36]. The light brown powder of V205 (12.6 g. 69 mmol) was added to diluted phosphoric acid (100 ml. 28 %, 460 mmol) to obtain a 7: 1 ratio of phosphorous to vanadium. Upon refluxing the mixture. the color changed from brown to greenish yellow indicating the formation of VOPO4-2H20. Over three days of reflux. the shiny yellow-geen microcrystals of VOPO4-2H20 precipitated from the reaction mixture. After filtration. the solid was washed with water. acetone and air-dried. b. Layered Niobium Phosphate Trihydrate. Metallic niobium (5g. 54mmol)wasdissolvedinamixtureofHF(50ml. 40%) and nitric acid (5 ml. concentrated). and phosphoric acid (30 ml. 85 %. 500 mmol) was subsequently added. The resulting clear solution was heated on a water bath to remove excess HF. until white solid appeared at the solution] air interface. The white. flaky solid was collected from the solution. cooled to room temperature. and washed by resuspending in nitric acid (300 ml. 5 M). After filtration. the product was washed by successive suspension and filtration in water (200 ml) and ethanol (200 ml). It is noteworthy that powder X-ray difi'raction patterns showed multiple phases when the solid was washed simply by suction filtration. 3. Interesiation Reaction Chemistry a. Sodium intercalated Layered Phosphates. in order to avoid contamination from 12. inherent in J acobson's iodide reduction [1 10). NaBH4 was employed in place of Na! as a reductant. Addition of 112 VOPO4-2H20 (2 g. 10 mmol) to a NaBH4 (0.4 g. 10 mmol) suspension in dried. deoxygenated THF led VOPO4-2H20 to acquire a dark-geen color. The reaction was accompanied by the weak evolution of hydrogen gas. After 1 hr of stirring. the dark-geen solid was filtered and washed with TH!" and several portions of water to remove unreacted NaBH4. The resulting crystalline powder was washed with acetone and air-dried. The same procedure was used to produce Na‘t intercalated NbOPO4. However. no color change is observed upon the reduction of NbOPO4-3H20. b. Interealation of Bimetallic Cores. The incorporation reactions of solvated bimetallic cores into MOPO4 by an intercalative process (Route I) and by ion-exchange of MOPOinax (Route 11) are performed employing identical procedures. Insertion reactions were carried out in dried. deoxygenated acetonitrile owing to the extreme sensitivity of the solvated cores toward orddation. In a typical reaction. acetonitrile solutions of M02(CH3CN)8(BF4)4 (0.4 g. 0.5 mol: 2 meq) were transferred onto the layered material (0.1 g. ~0.5 mmol) under argonandthemixturewasstirredthreetosevendaysunderAr atmosphere. Unfortunately. the progess of reaction was rather dificult to observe. Therefore the desirable reaction period. usually three days. was determined by taking x-ray difi'raction patterns at 24 hr intervals. Upon completion of the reaction. the solid was filtered and washed with acetonitrile. ethyl ether and air-dried. All the dimer intercalated compounds were stored in an Ar filled dry box. 4. Instrumental Techniques 113 a. Powder x-ray Diffraction. The powder X-ray diffraction patterns were recorded on a Rotaflex system from Rigaku. The Cu--K(,l line was obtained fiom a rotating Cu anode (45 kV. 50 mA) and directed toward the sample chamber using a l / 2° divergence slit and a l / 2° receiving slit. The difi‘racted X-ray beam was further refined by a curved gaphite single crystal monochromator (l.05° scatter slit and l / 6° monochromator receiving slit). which was set for detection of the secondary X-ray difi'raction line. The compounds were mounted by pressing dried powder on a piece of double-sided tape attached to a 1" x 2" slide glass. The resulting data were recorded and processed using the manufacturer-provided software DMAXB on a microVAX computing system. b. Elemental Analysis. The analyses of elements in layered materials were performed with a Perkin Elmer PHI 4500 ESCA System. which is housed in the Composite Materials and Structure Center at Michigan State University. The X-ray source was monochromated Al or Mg Kc lines (Alma) . 1486.6 eV. 600 W/ 15 kV; Mg(K¢) 3 1253.6 eV. 400 W/ 15 kV). Samples to be analyzed were pressed onto double-sided tape adhered to a stage. which was placed at an oblique angle to the incident X-ray beam (65° for the Al source and 45° for the Mg source) under ultra-high vacuum (10'8 - 10'9 torr). The sample stage was positioned so that the siglal intensity for oxygen (bindingenergof531 eWwasmanmized.sincetheoxygenwasthe most abundant element in layered metal phosphates. Elements were identified by survey scans from binding energies of 1200 eV to 0 eV. The 114 strongest photoelectron line from each element was scanned separately to obtain relative atomic concentrations by integrating the area of the peak and multiplying an appropriate response factor. All data manipulation was accomplished on an Apollo Workstation using XPS ESCA software provided by the manufacturer. c. Infrared Spectroscopy. Infrared spectra were recorded on a Nicolet 740 ”JR Spectrometer. A KBr beam splitter / D'l‘GS-KBr detector was employed for the mid-IR spectral region whereas the Far-IR Solid Substrate“ beam splitter / DIES-PE detector was employed for the far—IR spectral range. Solid KBr or CsI pellet samples were prepared. depending on the spectral region. and an average of 16 scans was used for data collection. (1. Electron Paramagnetic Resonance. EPR spectra of layered compounds were measured by using a Bruker ER 200D X-band spectrometer equipped with an Oxford ESR-9 liquid helium cryostat. Magietic fields were measured with a Bruker ER 035M gaussmeter. and w the microwave fi-equency was measured with a Hewlett-Packard 5245L frequency counter. The layered compound was placed inside a quartz tube equipped with a Kontes quick release stopcock and evacuated until the pressure was less than 5 x 10"5 torr. Variation of temperature was achieved by controlling the flow rate of a gas generated from liquid helium or placing sample tubes in a finger Dewar flask filled with liquid nitrogen. Spectral simulations were performed using PROGRAM POWD developed by Belford et al. at the University of Illinois (1 15). 115 e. Magnetic Susceptibility Measurements. Magnetic susceptibilities were measured on a SHE 800 series variable-temperature SQUID magletometer controlled by an IBM-PC microcomputer. A known quantity of layered compound was placed in a KEL-F'"I bucket. whose magnetic susceptibility was independently determined for its application as a correction factor to observed sample susceptibilities. Measurements weremadewithanascendingtemperature ramptoSOOKfollowedbya descending ramp in an efi'ort to observe hysteresis. if there was any. Actualreadingsweremade onlyafterthe temperaturewasobserved tobe stabilized within :t: 5%. Typically. 10 measurements were averaged for one temperature setting: all readings were within 3: 5% error limit. 1'. Electronic Absorption Spectroscopy. Absorption spectra. obtained with a Cary 1 7 spectrophotometer. were measured on KBr pellets. Regular accessories associated with cuvettes for liquid samples were removed to accommodate pellet mounts at the focal point of the excitation beam. The chamber was purged with high nitrogen flow rates to prevent frosting of the KBr pellets. C. Results and Discussion 1. Synthesis and Characterization of Layered Phosphates In the construction of LIPS. singly and quadruply bonded bimetallic compounds are the candidates of choice for MPC's. The introduction of bimetallic cores to layered materials is most easily accomplished by employing solvated (lilio-‘-‘-Mo)4+ or (Rh—Rh)4+ cores. 116 because solvent ligands can be readily substituted by phosphates from the layers. A clean and emcient synthesis of acetonitrile compounds of dimolybdenum and dirhodium cores has recently become available [1 12. l 14]. Facile ligand exchange reactions of the solvating acetonitrile is the pre-eminent reaction pathway. as is the case for other solvated ions [1 16]. The (Nio-‘l-liiio)4+ core is very air-sensitive and its decomposition by oxidation takes place in minutes in air. On the other hand. the (Rh—Rh)“ core is quite stable in air with only the two acetonitrile ligands at axial positions undergoing facile ligand exchange reaction [1 l 7]. The intercalation reactions of these bimetallic cores may be performed by two separate routes. Route 1 MOPOs-nI-Izo + M2'(CH3CN)m4"' —-) MOPO4°M2)x (10) Route 11 MOPO4°nH20 + NaBH4 ——§ MOPO4-Nay (1 1) MOPO4-(Naly + M2'(Cl-13CN)m4* ——-) MOPO4-(M‘2); (12) (Marv,n:2:M=Nb.n-3andM'=Mo.m=8:M‘th.ma 10) Route I represents the redox intercalation pathway of Scheme 2, whereas Route II is the ion-exchange reaction pathways following reduction of the layer by NaBH4. The (Moi-Mo)“ solvated core can react with layered phosphates according to Route I as well as to Route II. For the former. the facile oxidation of (Mo-4—Mo)4+ cores to mixed-valence (Moi-Mo)5+ cores (38.39. 1 18] provides the driving force for the intercalation reaction. The high charge of (M05—Mo)4+ or Mag—5M0)5+ cores. if subject to 117 oxidation by the layer. will promote the ion-exchange of these cores. Conversely. the redox stability of the (Rh—Rh)4t core suggests that the incorporation of this species should proceed only by Route 11. Figures 29a and 308 display survey scans by ESCA of VOPO4 and NbOPO4 layers. Peaks characteristic ofM (M a V. 523. 784. 1017 eV: M 3 Nb, 209. 364. 379 eV). P (133. 191 eV) and O (531 6") are prominent. Additionally. a peak for C at 289 eV is observed. which is a commonly observed impurity in ESCA studies [1 19). The atomic concentrations of M and P. obtained by normalizing peak areas in separate scans of the strongest photoelectron lines of each elements (Table 9) show that the metal to phosphorus ratio in the layer (Ml P) is slightly lower than 1 .0. which is predicted for the native structure. The extra phosphorus can be attributed to phosphate groups on the crystal edge or possible defect sites. The layer stability toward intercalation is in evidence by the fact that pronounced change of this ratio is not observed upon introduction of guest molecules into the gallery. ' Several changes are observed by ESCA upon reduction of the host layers by NaBHli. Most obviously. as shown in Figures 29b and 30b. the Na photoelectron peak at 1072 eV appears. From these spectra we calculate that the Nat content in VOPOii is 0.3 (VOPOa-Nao,3) whereas the NaOPO4 layer is completely reduced to yield a product with the composition of NbOPOe-Nal .0- The oxygen content. which we define as the number of oxygen atoms per one metal center in the layer. is slightly decreased with intercalation of Na+ (Table 10). This result is in accordance with the reduced interlayer space in the solid (vide supra). which presumably is accompanied by loss of interlayer water from the more restricted environment. 118 Figure 29. ESCA spectra of (a) VOPO4- 21120; and (b) Na+-intercalated VOPO4 host layers. 119 8: each»... >6 . seem 9.88 . 85 gm com 8.. E 3. emu Bugunoo 120 Figure 30. ESCA spectra of (a) NbOPO4-3H20: and (b) Na+—1ntercalated NbOPO4 host layers. 121 8: >m 193m asufim 85 00933..— 8m 09 E .9 emu fiugunoo 122 Table 9. Identifies and Energies of Peaks Used to Identify Elements in ESCA Spectra of Layered Metal Phosphates Auger line Element XPS line Al Mg C 287 (18) 993 (KLL) N 402 (ls) O 531 (Is) 976 (KLL) 743 (KLL) 23 (25) 997 (KLL) 764 (KLL) F 686 Us) 832 (KLL) 599 (KLL) Na 1072 (Is) 497 (KLL) P 191 (Is) 133 (2p3)a V 523 (2p1) 1017 (LMM) 784 (LMM) 515 (2p3) Nb 379 (3m) 364 (3p3) 209 (3d3) 206 (M5) M0 233 (3d3) 230 (3d5) Rh 314 (3d3) 309 (3:15) *3 2p3 represents that the origin of this line is 2p orbital with Its magnetic quantum number of 3/2. 123 Ros 324822 can? 9.55.32 «a: ‘5.552 cummdbeofiseoé 23. c._az.vo._0nz osmodédozsgofi 23 «80:2 osmofifigzvegz S. H s .. oumodbaazvosnz we: I osmo.m.vo.5nz 2 m8 «82.3.29, 038.62 5.25m? m2 .5 «95> osmodé 5.8429, S «.8 «.8242? oamm. _ .2 .82..o.5> $2.8 N95> of? hdczamg 888 u osmofihgzeeg 8: .. 0690.89, 4. \ 352396 .534 “mom 05.2530 .233 32 3 > u E .682 §§S Em a .2 .2 u a no: u .2 .m u 5 «Eacfionmo..uz so 358m-.. 23 eBoaoESSm .3 3a.... 124 The Na+ ions undergo complete exchange in the presence of M02(C1-13CN)34+. The ESCA spectra of molybdenum intercalate prepared by Route 11 are shown in Figures 31b and 32b. The broad peaks at 230 eV are attributable to the doublets of 3d electrons of Mo. The atomic concentrations of Mo are 0.19 and O. 16 for VOPO4 and NbOPO4 based layers to give VOPO4-Moo,19 and NbOPO4-Moo,13. respectively. The number of metal ions found in the layers accounts for 32% to 38% of the exchange capacity of the layered compounds since the redox capacities of the layers are one electron per unit cell and the formal oxidation state of each intercalated metal is +2. The complete exchange of Na+ ions is rather striking. especially in NbOPO4-Moo,15 where the Na+ content prior to ion-exchange was 1 .0. A Mo content of O. 16 for the intercalated NbOPO4 layer by Route 11 accounts for only 32% of the charge equivalent. Thus the restoration of 70% of neutral units occurs to produce the observed Mo content from Nat-intercalated material. In contrast to the molybdenum exchanged layers. complete exchange of Na+ is not observed when the layered Nat-intercalate is exposed to Rh2(CHaCN) 10(BF4)4. as indicated by the presence of the 1072 eV peak in Figures 33b and 34b. The residual concentrations of Na“ in the layers is accompanied by I". which is revealed by the peaks at 686 and 599 eV. The 1" most probably originates from the counter anion of the solvated dirhodium core. (BF4)’ (Figure 33b and 34b). The probability of finding individual 1" ions inside of the negatively charged layer is not very high. Therefore. one reasonable explanation for 1" peaks in the ESCA spectra of the rhodium exchanged layers is the adsorption of (BF4)' at the edge or the outer surface of the layers. Unfortunately. this possibility can not be discerned because the B peak 125 Figure 31. ESCA spectra of solids obtained from the reaction of M02(CH3CN)3(PF4J4 with (a) VOPO4-2H20: and (b) Nat-intercalated VOPO4. 126 8: >m $35 9.25 005 39.3.3 8m 8— - E 38 9123 Bugunoo 127 Figure 32. ESCA spectra of solids obtained from the reaction of M02(CH3CN)8(PF4J4 with (a) NbOPO4-2l-I20: and (b) Nat-intercalated NbOPO4. 128 8: >6 \ 38cm 9.55m och «moan; com 8' 3 § 91133 Bugunoo 129 Figure 33. ESCA spectra of solids obtained from the reaction of Rh2(CH3CN) 10(PF4)4 With (a) VOPO4°2H20§ and (b) Na+-intercalated V0P04. 130 8:. >o \ 635 33m 85 38.9.— 8m 8— - 3 3. area Bununoo 131 Figure 34. ESCA spectra of solids obtained from the reaction of R112(CH3CN)10(PF4)4 With (a) NbOPO4°2H202 and (b) Na+-intercalated NbOPO4. 132 39:63 >o \ 38cm 9.65m - _ u d E 3. emu Bugunoo 133 at 190 eV is obscured by the P1, peak at 191 eV. Nevertheless we believe that the 1" peaks arise from (BF4)" loosely associated with the residual Na” ions. This is supported by a Nazi" ratio of ~4 as expected in NaBF4. The variable ratios of haF found in difl‘erent batches suggest that the F' is not associated with unreacted Rh2(Cl-hCN)10('BF4)4 on the surface. The stoichiometries of the layered materials are presented in Table 10 after removing the contribution from Nth. ESCAscansoflayeredmaterials preparedbyRoutelare shownin Figures 31a-343. The peak at ~400 eV represents physisorbed CH3CN molecules on the surface of layers. The atomic concentrations of Mo and Rh are 0.17 and 0.10 for the VOP04 layer and 0.12 and 0.10 for the NbOPO4 layer. respectively. The great surprise here is that Rh is found in the layer despite the redox stability of the Rh—Rh cores. The reaction time required for the Rh—Rh intercalation by Route 1 was about two weeks as compared to 3 days for Route 11. The driving force for this reaction may be provided by formation of the monomeric Rh(II) which can only be prepared from Rh2(Cl-13CN) 10(BF4)4 and can be easily oxidized [120]. The formation of monomeric Rh(ll) is not a favorable process in Route 11 which is evidenced by reaction times. We believe that these observed metal contents are the highest possible loading for these layers because the same loadings are observed even when the guest complex is ~5 times the redox capacity of host layers. Of course. an obvious issue is the nature of the metal species inside of layers: Is the metal-metal bond preserved upon incorporation into the layer? The high stabilizations of these metal-metal bonds. especially the Moi-Mo cores. suggests that this would be the case. However. attempts to observe the symmetric M—Q-M vibration by IR is 134 prevented because these symmetric stretches are 1R inactive since they belong to the A18 representation of the 04;, group. Raman spectroscopy is also of little value in confirming the presence of a M—‘lM bond. Though the symmetric M-Q-M vibration is allowed in Raman spectroscopy. interference from strong Raman bands of layers and the low concentration of bimetallic cores make it dificult to obtain reliable spectra. Far-IR spectra of M-L vibrations (ME-M vibrational region: 1 50 ~ 450 cm“) are usually weak and accordingly no peaks with appreciable intensity are observed after intercalation. The corresponding phosphate vibrational spectral region is too congested to draw any convincing conclusions. However. slight shifts to lower energ are observed for bands related to vN—OH). v(V=O). v(P-O). 8(V-O-H). and 8(0-P—O). Thus the phosphate group has not been greatly disturbed by the introduction of guest molecules. The NbOPO4 system shows similar spectral features to the VOPO4 system. However. the Nb—O vibrations are shifted to lower energy compared to the corresponding V-O vibrational modes and isolated from the phosphate vibrational region. A slight shift of the v(NbaO) band to higher frequency is observed upon intercalation. Most likely. the water molecule coordinated to the V center from the position trans to the V-O bond is removed upon the intercalation of guest ions. Nevertheless retention of the binuclear core within layered phosphates is indicated by electronic absorption spectroscopy. The MoiMo core has characteristic absorption spectra for the quadruply bonded (n s 4) and mixed valence dimer species (n a 3.5) [38,39,118]. Unfortunately. the observation of absorption bands of guest molecules in VOPO4 layered materials is problematic because of the green color of the VOPO4 layers (arising from transitions at VHV) impurity centers). By 135 contrast. the NbOPO4 layers do not exhibit transitions in the visible spectral region and hence absorption of guest molecules is not spectrally obscured by the layers. The white layers of NbOPO4 turn blue-grey upon exposure of NbOPO4 to M02(CH3CN)3(BF4)4 solutions for either Route 1 or Route 11. This color change is noteworthy because the M0201.II) tetraphosphate is pink whereas the mixed-valence MozalJll) tetraphosphate is blue-grey [38.39]. Specifically Figure 35 shows the electronic absorption spectrum of NbOPO4°M00. 19: an identical spectrum is obtained for NbOPO4-Moo,19 prepared by Route 11. Overlaying a rapidly rising baseline. due to the scattered light and] or a very broad charge transfer band of layer itself. two shoulders at ~480 nm and ~550 nm are observed. This spectrum establishes that the quadruply bonded Mozt'HPOdf' is not incorporated in the layer. which would show an intense 88 -§ 88‘ transition at 516 nm. Alternatively. the two weak features in the absorption spectrum of Mo intercalated NbOPO4 can be correlated to the characteristic 1: -9 8‘ and x -) 8 transitions of Mog(1-IPO4)43' at 420 nm and 595 nm. respectively [38]. For comparison. the absorption spectrum of the mixed-valence M02(HPO4)43' molecular speeies is shown in Figure 35. Unfortunately. the strong 1(8 -> 8‘) transition of the M02(U.III) core in the near-IR [38.39] could not be obtained due to the severe interference of layer absorption in the near infrared. Notwithstanding. the absorption spectrum clearly establishes the presence of 140201.111) intercalated in NbOP04. The similarity in the synthetic conditions for Mo intercalated NbOPO4 and VOPO4 layers suggests that this core will also be retained in the VOPO4 layers. Interestingly. the fact that the MOszn) intercalated material is obtained 136 Figure 35. Electronic absorption spectra of M02(CH3CN)8(PF4)4 intercalated NbOPO4( --------- ). the same layer after arposure to air (--- ..) and M02(HP04)43' in deoxygenated 2 M H3P04 (—-——). 137 3 2.9a E: \ £05653 omm cam one cow 0mm. .13 Né eoueqrosqv 138 by Routes I and 11 suggests that the M0201.11) core is not thermodynamically stable in the strongly oxidizing layers. 2. Structural Characterization Consistent with the formulation of a Mo§'—5Mo dimer in the layered phosphates is the structural characterization of the M02 intercalated VOPO4 and NbOPO4 phosphates. Figure 36 displays X—ray difi'raction patterns of host layers. VOPO4 and NbOPO4. The door peaks at 7.49 A (20 s 11.80°) and 8.02 A (20 a 11.02°). respectively. along with several higher order peaks are in very good agreement with their calculated positions based on published cell parameters (Table 11) [121,122]. In both materials. reduction of the layers by Nam-l4 causes the door peak to shift to higher angles indicating contraction of the layer spacing. The placement of negative charges onto the layer with concomitant insertion of positively charged Na" ions increases the interaction between host layers and guest molecules. This compression of the layers about Na+ ions as compared to the loose and open network of water within the interlayer region of the original MOPO4-nHzO layers is attributed to the electrostatic interactions of the former being much stronger than the hydrogen bonding interactions of the latter. In support of this contention is a recent EXAFS study regarding the location of guest ions inside the vanadium phosphate layers [123]. lntercalation of metal ions Fe3+. C02+ and Ni” yields metal ion intercalated layers with contracted d-spacings. According to coordination geometries based on the number (n) of nearest neighbors (n a 4. M - Fe3+. Co”: n a 6. M 2 Ni”) and native VOPO4 structure. the most probable location of intercalated metal ions is 139 Figure 36. Powder X-ray difi'raction patterns for host materials: (a) VOPO4-2H20: (b) Nat-intercalated VOPO4: (c) NbOP04o3H20: and (d) Nat-intercalated NbOPO4 layers. -4 ' " 140 Intensity [usually 20/degrse 141 Table 1 1. The Peak Positions and the Indices of the X-ray Difi'raction Patterns of Layered Metal Phosphates VOPO4'2H20 Nb0P04°3H20 (hkl) dobslA) denim“ (th) debslA) denim” (00 1) 7.49 7.4 1 (002) 8.02 8.04 (101) 4.79 4.76 (01 1) 5.99 5.94 (1 10) - 4.39 (012) 5.03 5.00 (002) 3.72 3.71 (1 1 1) 4.4-1— 4.35 (102) 3. 19 3. 18 (013) 4.05 4. 1 1 (200) 3. 12 3. 1 1 (004) 3.96 4.02 (201) 2.86 2.86 (014) 3.38 3.40 (020) 3.24 3. 19 (1 14) 3.00 3.00 (12 1) 2.85 2.8 1 (024) 2.50 2.50 (220) 2.29 2.25 a Lattice constants are a - b . 6.21 A. c . 7.41 A in tetragonal structure and one layer consists of one unit cell. 1’ Latticeconstantsarea=b=6.39A.c= 16.08Aintetragonal structureandtwolayersinoneunitcell. 142 suggested to be a tetragonal oxygen pocket defined by four in-plane vanadium octahedra (Figure 37). The immediate coordination spheres of each metal ion consists of oxygens that are ~ 2 A apart from central metal atoms. When the ionic radii of interlayer ions (0.6 A - 0.8 A) and 02- (1.2 A) are considered. the calculated interlayer distance based on known structural factors of the hydrated ion is reasonable. Contraction of the host layer spacing is also observed for VOP04 and NbOPO4 layers obtained by Route I and Route 11. Figures 38 and 39 display the X-ray difi'raction patterns of Mo and Rh intercalated layers. respectively. and Table 10 lists the measured d—spacings. Average contractions of 1.05 A and 0.7 A are observed for the Mo and Rh intercalated vopo. layers and 0.63 A and 0.6 A for the Mo and Rh intercalated NbOPO4 layers. respectively. Discrepancies between Rh intercalated layers prepared by Route land Route 11 are larger than that between Mo intercalated layers. These results suggest that Rh intercalation is dependent on Route 1 as suggested in Section C. l . We believe that the redox intercalation proceeding directly from the Rh24+ core occurs by decomposition of the core. Thus. hereafter we will only be considered with Rh intercalation proceeding by Route 11 where we believe the binuclear core is retained. Conversely. nearly identical d-spacings are observed for Mo intercalated layers regardless of the synthetic method. This result is consistent with ESCA studies which suggest that the M025 core is the thermodynamically favorable species within the layered phosphate environment. The d—spacings of the intercalated materials are consistent with the binuclear core keying into the tetragonal cavities of the MOPO4 layers. We will focus here on M02 intercalated in V0PO4. The projection 143 Figure 37. Coordination environment determined by EXAFS for mono- or divalent cations [l 23] situated in VOPO4r2H20 structure. 145 Figure 38. Powder X-ray diffraction patterns of solids obtained from the reaction of M02(CH3CN)8(PF4)4 with: (a) VOPO4- 21120: (b) Na+- intercalated VOPO4: (c) NbOPO4-3H20: and (d) N at—intercalated NbOP04 layers. ' 146 (b) (a) £23... 20 25 201de 15 10 (d) _ AAA (c) .... L, A .‘o .‘s 25 20/degrss W38 Emcee. 147 Figure 39. Powder X-ray difi’raction patterns of solids obtained from the reaction of Rh2(CH3CN) 10(PF4J4 with: (a) VOPO4-2H20: (b) Nat intercalated VOPO4: (c) Nb0P04-3H20: and (d) Nat-intercalated NbOPO4 layers. 148 5 g \ L [L a. L k a. i *r 1 1 __—l 1 5 10 15 20 25 m 35 ze/degrso i a E \J W#_ k (d M (c) 1 149 of the layer onto the a.b-plane clearly shows that the four oxygens important for M025+ coordination are arranged in a square of 4. 12 A dimension (Figure 40a). Using the crystallographic data of T'ietzs [124]. the distance between the planes defined by these four oxygens is 2.88 A in the c-direction as determined on the basis of the observed 6.45 A d- spacing. The dimensions of the tetragonal cavity defined by these eight oxygens in neigthring interlayers is summarized in Figure 40b. This configuration of oxygens is in accordance with a transverse disposition of the M025+ core keyed into the oxygen cavity of the interlayer. In this arrangement. four oxygens from the layer serve as the coordination sphere for one molybdenum center of the bimetallic core. This 'solid state‘ coordination environment of the M029 core nicely accords with the molecular crystal structure of M02((C6H50)2P02)4(BF2). which has been prepared in our goup [39]. In this species. a tetragonal cavity is formed from the tetraphosphate ligation sphere about the Moz5+ core. The dimensions of the tetragonal oxygen cavity for the molecular species are summarized in Figure 40c. The distance of 2.5 A between oxygen planes for the molecular species is consistent with the distance of the oxygen planes in the c-direction determined from the d-spacing. The parallel disposition of the binuclear core would require a 4. 12 A intermetallic distance (Figure 40). which cannot sustain the short metal-metal interaction. Moreover. the d-spacing of a core situated parallel to the layers would be 8.12 A (dm - 5.86 A. dva(M025+) . 2.26 A). which is 1.67 A greater than the observed d-spacing. Thus the transverse arrangement of the M0254' core permits interlayer keying. which is a common guest-host interaction in the chemistry of layered oxide materials [34] . 150 Figure 40. Views of the VOP04-2H20 layer from (a) the 00 1 plane where the four closest neighbor oxygens are shown in shaded circles: (b) the dimensions of the oxygen coordination environment for M025+ intercalated VOPO4 ' [124]; and (c) the dimensions of the oxygen environment of Mozucensompoghmrm as determined from crystallogaphic analysis (39] . 152 3. Magnetic Properties of 113-u lntercalated Layers a. Electron Paramagnetie Resonance The VOPO4 layer is expected to display an EPR spectrum owing to the unpaired electrons at V(lV) (d1) impurity sites [109]. As expected in an isolated spin system. the low temperature (8 K) EPR spectrum of the VOPO4-2H20 layered material. shown in Figure 41a. exhibits a very well defined signal. The structure of the signal can be ascribed to hyperfine splittings arising from the interaction between an unpaired electron and the nuclear spin of the WW) center. Similar EPR signals have been observed by Jain on single crystals of Rb2Co(SeO4)2-6H20 doped with V(1V)02+ [125]. The EPR spectrum of the VOPO4 layer is that for an axial doublet. which is consistent with the four-fold symmetry of the WW) center in the layer. Additionally. two different but very closely related environments about vanadium centers is indicated by the clear separation of two peaks on the g, , component. The separation arises from small difi‘erences in splitting constants for different V(IV) sites. The slight difi'erence in splitting constants can be attributed to the existence of the cis and trans arrangement of the V(=O)O4 about the equatorial V- O-P-O-V chain. As shown by Bordes et al. [126] in their structural study of a-VOPO4. the EPR spectrum can be simulated for two structures by overlapping two spectra that are showing a simple 8 line pattern centered on g-1.98 (I-Tgure 41b). In the case of a single dl structure. as is the case for Jain's system. a straightforward 8-line spectrum is obtained. Thus the d 1 electrons in VOPO4 host layer's arise from impurity spins that are completely isolated from each other: 153 Figure 41 . Electron paramagnetic spectra of VOPO4- 21120 (a) observed at 7 K and (b) simulated using a 2-site model for the (11 WW) centers. The microwave frequency and the sweep range were 9.471 1 GHz and 2512 - 4582 G. respectively. 155 therefore the hyperfine splitting pattern is undisturbed. Intercalation into the layer causes significant reduction of WV) to V(IV). This is vividly displayed by the drastic change of color to dark green arising from the tag —> e8 transition of the (11 WW) center. Figure 42a shows that significant changes in the EPR spectrum also accompany layer reduction. The complete disappearance of the hyperfine splitting results from the high degree of 1 interaction among the (11 spin centers. The line shape is Dysonian shape which is characteristic of conducting electrons exhibiting metallic character [127). Moreover. the Dysonian peak exhibits saturation at very low power. known as the passage efiect [127.128]. The asymmetric shape of the Dysonian peak gradually becomes symmetric upon reduction of" particle size. Thus. the observation of Dysonian peak indicates the electron density in the reduced layer is not only high. but interacting over the extended structure. Indeed. the metallic character of the layer is indicative of electron delocalization. The Rh2(11.ll) intercalate shows similar behavior to the Na" intercalated layer. The EPR spectrum of the Rh2(II.II) intercalated VOPO4 layer is shown in Figure 42b. Due to the redox stability of the Rhgtlul) core. the guest species remains diamagnetic. and paramagletism is thus confined to the layer. The striking resemblance of the line shape with Na+ intercalated layer implies that the Rh2(II.II) cores do not introduce a channel for magnetic communication between layers. Conversely. the intercalation of the Mo§°—5Mo core into the VOP04 layer has serious consequences on the EPR spectrum. As shown in Figure 438. a featureless single peak with an extremely broad spectral 156 Figure 42. EPR spectrum of (a) Na+-intercalated (9.4724 GHz. 2497- 457 7 G) and (b) Rh24+-intercalated (9.4712 GHz. 2294—4382 G) VOPO4 layers at 7 K. 157 (a) (b) Figure42 158 Figure 43. EPR spectra of (a) M02(CH3CN)8(PF4J4 intercalated VOPO4 (9.4726 GHz. 970-6068 G) and (b) the spectrum of the solid afier exposure to air (9.4718 GHz. 2293—4387 G). 160 band width is observed. The Moi-5m intercalate is unique because the MoaéMo core is a paramagnet with a very well defined EPR spectrum of axial symmetry [39]. Thus this system is magnetically described as a weakly coupled spin system in a 2-dimensional geometry which is interconnected by paramagnetic centers between the layers. However. neither the narrow Dysonian peak shape from the conducting electrons in the reduced layernorthe axial peakofthebimetallic core is observed. Line broadening in EPR spectra has been observed for inhomogeneous spin systems or with very strong spin-spin interactions causing magnetic ordering [129]. The extreme breadth of the spectral line width (~750 G peak-to-peak separation) in Figure 43a suggests that inhomogeneous broadening is not solely operative. and the existence of coupled spins between the reduced layers must play a role in explaining the magnetism of these solids. Undoubtedly. the electrons with S s 1 / 2 on the reduced layer must primarily be responsible for the peak due to their high concentration: But a strong spin-spin interaction furnished by nearby paramagnetic dimolybdenum centers may be the origin of broadening. An important experiment supporting this hypothesis is that the width of the EPR line is drastically reduced to a peak-to-peak separation of ~100 Guponexposureofthelayeredintercalatetoair. ltisknownthatthe M029 tetraphsophate core rapidly oxidizes in air to the diamagnetic Mo;6t triple bonded species [38.39]. Thus the observation of line narrowing is consistent with the depletion of the paramagnetic Mo3' 0 core by oxidation to a generate a diamagnetic MolMo core. A spectrum consistent with a reduced layer containing weakly coupled spins in a 2- dimensional plane is observed. 161 Similar to the vanadium phosphate. the niobium phosphate also contains tetravalent metal centers as an impurity. But the EPR spectrum shows poorly developed hyperfine structure (Figure 44a). accompanied with a small peak near ~1500 G (g a 4.31) which persists upon intercalation. The nuclear spin of 9/ 2 should give rise to a hyperfine structure consisting of 10 lines. Typically. molecular Nb(lV) compounds exhibit spectra with g values of 1.98 - 1.99 [130]. Peaks with g values much greater than 2 have not been reported for Nb(1V) centers. but have been for metal centers with higher spin states and in environments with extreme axial symmetry [131]. The g a 4.3 signal shows dramatic enhancement of the intensity upon reduction of the layer by Nam-l4. A high axial symmetry in the NbOP04 environment can be created by removal ofwater and the axial oxo group. Inasmuch as this environmental argument is appealing. it cannot provide an explanation for the drastic increase in the intensity of the g a 4.3 peak upon reduction of the layer. without afi‘ecting the intensity of g a 2 peak (Figure 44). The g s 4.3 signal is dificult to rationalize. Higher spin states indicative of possible Nb(III) or Nb(ll) centers might be considered. While smallpeaksnearg-2maybeassignedtoAM- ltransitions. the associated AM a 2 transition should be weak. By introducing a Nbfll) centerintothelayer.bothofthepeaksatg=4.3andga2canbe explained when large differences in the perpendicular and parallel g values are assumed. While this interpretation is plausible within a magietic framework. it is unrealistic from a chemical standpoint. Niobiumal) molecular species are rare [132]. and co-existence of Nb(V) 162 Figure 44. EPR spectra of (a) NbOPO4-3H20 (9.4712 GHz. 1270-5379 G) and (b) Na+-intercalated NbOPO4 layers (9.4718 Gl-Iz. 1270-5383 G) at 7 K. 164 and Nb(11) centers in the layer should be assumed to obtain the observed degree of reduction of about one electron a unit. Upon introduction of MoszI) into the niobium based layers. the g = 4.3 peak disappears to an intensity consistent with the unintercalcated layer with concomitant sharpening of the g =- 2 lines for layers prepared eitherbyRouteIorIIiFigure45a). Infact. theshapeofthegs2peakis nearly a single Dysonian peak. Also the spectrum shows that at least two difi'erent electron environments exist as indicated by the fine structure of weak intensity feature. The intensity of the central line greatly decreases upon exposure to air along with the pronounced development of fine structure as presented in Figure 45b. All the peak positions. except the central peak. are well correlated with the unintercalated NbOP04 layer which contains isolated electrons on impurity sites. Thus as with the vanadium intercalate. the paramagnetic MofiMo core does perturb the magnetism of the layer. However. our inability to interpret the magnetism of the layered structure leaves us with little hope of understanding the EPR of the NbOPO4 Mo3+ o intercalate. b. Magnetic Susceptibility From EPR experiments. four distinctly different situations with regard to the coupling between spins are implied. The simplest case is that of completely isolated spins of the native layered vanadium phosphate structure. Simple paramagnetic behavior is expected. When the spin concentration in the layer is increased. the strength of the exchange interaction becomes important. as observed in Nat-intercalated 165 Figure 45. EPR spectra of (a) M02(CH3CN)8[BF4)4 intercalated NbOPO4 (9.4742 GHz. 2293-4382 G): and (b) the spectrum 'of the: solid after exposure to air (9.4728 GHz. 1380-1590 G). 166 (a) (b) Figure“! 167 layers. The interactions may be further modulated by the physical linkage between layers or a magnetically silent guest as is the case of Rh24+ intercalate or a magnetically active guest as is the case for the M05” intercalation chemistry. The native VOP04 layer displays clear paramagretism following simple Curie-Weiss law behavior. Figure 46a shows a linear 1 /XM vs. T plot over all temperatures as expected from a Curie-Weiss relation. X a C / (T-O) (13) where X is susceptibility and T is temperature. An electron density of 1.8 % is calculated from the slope of the UM; vs T plot (Table 12). As indicated by the hyperfine splitting pattern of the EPR spectrum. the low spin density confirms that the paramagretism in the VOPO4 layer is due to V(IV) centers as isolated impurity sites. As expected for a simple paramagnet. the XMT vs T graph provided in Figure 46b is constant over all T. The reduced layer exhibits magnetic properties that deviate from Curie-Weiss behavior at low temperatures. The 1 [1M vs T plot shown in Figure 47a. fits Curie-Weiss behavior with an electron density per molecular unit of 39 % and a Weiss temperature of 1.437 K. These results agree nicely with the degree of reduction calculated from a sodium content of 30%. However. the XMT vs T graph in Figure 47b exhibits a markedly decreased XMT value with decreasing temperature below ~20 K. It is evident that the decrease of 1m in the low temperature regme must be due to antiferromagnetic coupling owing to the d1 paramagnetic centers. The sign and strength of the exchange 168 Figure 46. (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of VOP04-2H20 layer. The solid line represent the best-fit lines in (a). 169 8.06+4 2' 6.06+4 5 e5. \ 4.06“ P) 5 R 2.0644 0.0640 ' l l l 1 0 50 100 150 200 250 300 T/ K 1.06-2 (b) 8.06-3 ’ "is g 6.06-3 ~°° , g ..‘sso .... .0 . o o e . . G O O Q g ‘ 406-3 " ° ° ° T P o 2.06-3 ' O.m+o l l l l I 0 50 100 150 200 250 300 T / K Figure46 170 Table 12. Electron g Values from EPR. Curie-Constant (Cm). Weiss- Temperature (0 l ‘1). Spin Density (SD / %) and Diamagnetic Correction Factor (DCF / emu. moi-1) Deduced from Temperature Dependent Magnetic Susceptibility Measurements of lntercalated Layered Materials. (3.) VOPO4 Related Systems Compound Cm 6 SD VOP04-2H20 1.962 2 19.5 1 .47 1 .8 VOPO4Nao,3-2.OH20 1 .964 8.58 - 1 .08 38.9 VOPOdMoo, 17' 1 .5H20 1 .947 14.23 -9.57 153 VOPO4M00, 19- 1 .51-120 1 .947 16.43 -6.62 133 VOPO4Rho, 10201-120 1 .964 8.02‘I - 1 .24a 30a a Determined from the fitting of measurements made at 5 - 180 K owing to the severe deviation from Curie-Weiss Behavior in higher temperatures. (b) NbOPO4 Related Systems Compound g Cm O , SD DCF moms-3.01120 1 .994 990.8 -2.30 0.4 63. 1 NbOP04N81,o-2.OH20 2.068 388.8 -4.05 ‘ 1.0 20. 1 NbOPOaMoo, 12201-120 1 .926 1433.3 -3.52 0.3 ' 20.7 NbOPO4Moo,192.OH20 1.926 1460.2 -6.88 0.3 20.4 171 Figure 47. (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of Nat-intercalated VOPO4-2H20 layer. The solid lines represent the best-fit line in (a) and the calculated result from eq 14 using the J value obtained from (a). 172 to 1° 8 8' 2’.» to SE 2:. 1 — I smu-1 ~mol Xu 1 .06+3 5.0e+2 0.0640 0 50 100 150 200 250 300 1b) . ’. s . 0 I 1.06-1 8.00-2 6.06-2 . ' XuT/ omu~K~mol-1 4.06-2 " 2.06-2 0.0e+0 ‘ ‘ ' ' ' 0 so 100 150 200 250 300 T/K W47 173 interaction in vanadyl phosphate intercalate suggests 2-dimensional couplings within the a-b basal plane by a superexchange mechanism via phosphate tetrahedra [133]. For an exchange interaction confined to a 2- dimensional space. the magnetic susceptibility is given by a high temperature expansion series for a quadratic layer Heisenberg antiferromaglet [134). 4” (1 ...; x2 x3 ... ...... ) where x s kT/J - 2T/6. and N. B and k are Avogadro's number. the Bohr magneton and the Boltzmarm constant. respectively. With the g values deduced from EPR. the XMT vs T plot can be fit iteratively to yield the exchange coupling constant. J. The solid line in Figure 47b is calculated from eq 14 by substituting g I 1.962 and J :- 2T/0 - 1.46'1‘ normalized to the high temperature plateau. The calculated values are an order of magnitude larger than the experimental values because the electron density is lower than unity. On the other hand. the J value obtained from the 1 / I“ vs T graph. which determines the shape of the curveisstillvalid. ThisresultisnotsurprisinginlightofVilleneuve and Lenzama's study of V0804 [135] where the strength of the exchange interactioninthefullyreducedlayerhacthenterinVOSOsisina formal oxidation state of +4. d1) is still in the weak coupling regime. Our results are entirely consistent with Ferey’s recent study [134] on the fully reduced vanadyi phosphate layer. which too fit a weakly coupled 2- dimensional model. Thus the 39 % spin density of the WW) centers in the Nat intercalate surely ensure a weak coupling regime. 174 The behavior of the Rh—Rh‘“ intercalate should parallel the Na+ intercalated layer in that the Rh—Rh‘“ guest is not magnetic. The expected electron content based on the composition of Rh (16% by ESCA) is 32 %. The slope of the 1 / XM vs T plot shows 30 % spin density per unit cell. The negative Weiss-temperature obtained from the l / XM vs T plot implies that this material is also antiferromagnetic. On the other hand. the - Mo‘fl‘Mo intercalated layers have the additional channel for the exchange interaction arising from the paramagnetic Moi-5M0 cores between the layers. The Mo content of the layers. 17 % by Route 1 and 19 % by Route 11. suggest that the layers should have spin densities of 51 % and 57 %. respectively. where 5/ 6 of the total spin concentration originates from the reduced layer and l / 6 from intercalated Moa-‘E'Mo cores. However. the electron concentrations from magnetic susceptibility measurements (Table 12) are much higher than this estimated value. The l / XM vs T plots show spin concentrations of 1 57 % and 133 % for layers prepared by intercalative and exchange pathways. respectively. No logical chemical species can rationalize such high spin densities. One-electron oxidation of the MofiMo core will lead to the formation of stable diamagnetic MoiMo. Further oxidation of guest molecules will reduce the number of spins for guest molecules and increase the layer charge so that the maximum number of spins is six spins per each intercalated Mo unit. In addition. the possibility of host layer degradation upon excessive reduction limits the highest possible spin density from layers to 68 % per each V [110]. And perhaps the most compelling evidence against any chemical explanation is that spectroscopic and X-ray studies only show the presence of the MoEMo core as the guest. 175 Alternatively. as indicated by the extremely broad EPR spectra. strong spin-spin interaction leading to ferro- or antiferromagnetism can explain deviations from simple Curie-Weiss law conclusions. The 0 value from the 1 [1M vs T plot is -9.57 K with an implication of large antiferromagnetic ordering. Moreover. the XMT vs T plot shown in Figures 48b and 49b shows significant deviation from the calculated curve shape when a single J value from the Curie-Weiss plot is used. The analogous magnetism and poor fit ofeq 14 lead us to believe that a strong exchange interaction of spin systems in 3-dimensions is present in this system. The added dimensionality may simply result from increased interlayer communication owing to the short d-spacing of the Moi-5M05+ intercalate (d s 6.4 A) as compared to the Rh—Rh‘“ and Nat (d s 6.7 A) layered species. Additionally. the anisotropic arrangement of the spin 1 / 2 guests between 2-dimensional magnetic layers may be crucial to the discovery of new magnetically-ordered systems [136]. Magnetic susceptibility measurements of NbOPO4 related layers gve negative susceptibility readings except at very low temperatures. However. the temperature dependent behavior of the magnetic moment displays paramagnetic behavior. Considering the large atomic number of niobium and the small experimental values. a diamagretic correction becomes necessary. Correction values are determined gaphically by taking intercept values from the 1 /XM vs T plot (Table 12). Figure 50a shows the Curie-Weiss plot. Fits were performed only over the low temperature regme. owing to the extremely small susceptibility values for readings at temperatures above ~200 K. The spin densities calculatedforNbOPO4areverylow.rangngfrom0.4%t02%per niobium center. The native structure displays 0.4 % spin density in good 176 Figure 48. (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of M025+-intercalated VOPO4-2H20 layer by Route 1. The solid lines represent the best-fit line in (a) and the calculated result from eq 1 4 using the J value obtained from (a). 177 100 150 200 250 300 50 T/K 300 250 200 150 'T/)( 100 Figure“ 50 8004 pkfichAxcoxhhda 178 Figure 49. (a) Curie-Weiss plot and (b) XMT vs T plot for temperature dependent magnetic susceptibility measurements of M025+-intercalated VOP04o2H20 layer by Route 11. The solid lines represent the best-fit line in (a) and the calculated result from eq 14 using the J value obtained from (a). 179 5.0s+2 4.06+2 50 100 150 200 6.00-1 5.00-1 4.00-1 3.00-1 m I 0W'K'MO1'1 2.00-1 1.00-1 0 50 100 1 50 200 T/K W49 250 300 180 Figure 50. Curie-Weiss plots for (a) NbOP04-3H20 and (b) Na- intercalated NbOP04 layers. Solid lines and open circles represent best- fit lines and experimental points that are omitted from fittings. 181 3.06+5 Q) \ 0.0640 0 50 100 150 200 250 300 T l K 20+5 E 1645 F: E O \ v-| >3 Se+4 Oe+0 182 ageement with paramagretism arising from impurity centers. as inferred from the EPR spectrum. Figure 50b displays a representative 1 / 1M vs T plot of Na+ intercalated NbOPO4 layer as a representative plot of NbOPO4 related layers. Despite a Na+ content of 1.0 as determined from ESCA. the spin-density obtained by the 1 [1M vs T plot is only 2 %. The Mo intercalated layers also show spin-densities of only 0.3 %. It is also obvious that these numbers are not consistent with the chemical compositions deduced from ESCA. These results establish that simple paramagnetic models will not sumce for the NbOP04 layers. The large coupling may be manifested in stronger exchange interactions. as compared to vanadium phosphate layers. Hence weakly-coupled 2- dimensional models such as that summarized by eq 14 are not applicable to the niobium phosphate system. The small spin densities. low susceptibilities. and anomalous EPR spectra certainly support a magnetic limit of large exchange interaction where all the spins are paired. 4. Conclusion WehaveshownthatsolvatedM—MandM—iMcorescanbe intercalated into layered metal phosphates. Intercalation reactions prowed by reduction of the host layer with concomitant introduction of guest molecules. Magnetic properties suggest that the electrons are delocalized over the 2-dimensional phosphate layers to gve rise to very interesting EPR and magnetic properties of the host structure. These layered properties are further modulated by the presence of an anisotropic paramagnetic guest molecule. Thus. not only do these 183 studies establish the validity of the LIPS strateg' but they have unveiled new synthetic strateges towards the design of magnetically interesting materials. MR“! NEW DMONS 1N LIPS Layered metal phosphates have been the major building block in our effort to assemble layered integated photochemical system (LIPS). However. the limitation of a phosphate ligation sphere is a considerable obstacle in the desig'i of LIPS structure. For the most part. the bimetallic phosphate compounds studied in our research group will participate only in successive one-electron photochemistry. The rich multielectron photochemistry as described in Chapter I is derived from M2X411L)2 D2,, compounds in which the LL bidentate ligand is phosphine. hydroxypyridine and isonitrile. These ligation spheres can in principle be introduced into the LIPS structure at the level of the MPC or the layered support. Modification of either the interlayer with ligand functionality or of the ligands on the bimetallic compound will permit implantation of various MPC's with the desired ligation sphere. We now present these two approaches in LIPS design and suggest future courses of action for the synthesis of multielectron photoactive solids. 184 185 A. Modification of Phosphate Inter-layer Galleries Synthetic methods to modify the interlayer of metal phosphates have been developed by Dine and coworkers [30]. The most studied systems are Zr(HPO4)2- 21120 and related structures [35]. The Zr(HPO4)2-2H20 structure presented in Figure 51a. consists of metal ions in the M(IV) oxidation state that are bridged by six difi'erent phosphate tetrahedra in a Dad arrangement [33]. All of the tripodal oxygens are bound to metal centers leaving an apical oxygen bearing an exchangeable proton directed into the gallery space. Due to the existence of this exchangeable proton. metal phosphates with the Zr(HPO4)2-2HzO structure are known to be very emcient ion exchangers. Reactions of organic phosphonic or arsonic acid with M oxides. oxychlorldes and halides will yield M(03XR)2 (X a P or As: M 2 Ti. Zr. U. Th and Ce) layered materials. {MCLt or MOClzl + (HohxlolR ——-> M(03XR)2 (14) (M 8T1. 21'. U. m. and CC) In this manner. new layered compounds that have interlayer spaces tailored with organic functionality may be defined [137] . The backbone of these new layers are similar to simple zirconium phosphate layers with M(IV) coordinated by six oxygens from six difi'erent phosphonates. However the apical oxygen that bears hydrogen in Zri‘HPO4)2-2H20 is replaced by alkyl or aromatic groups. which too are normal to the layer and directed into the gallery space. A structure of Zr(CsH5PO3)2 is shown in Figure 5 lb. If compounds with two phosphonate g'oups at 186 Figure 5 1 . Structures of (a) simple Zr(HPO4)2-2H20 and (b) modified Zr(RP03)2 layer with a organic phosphonic acid. 188 both ends are employed. various pillars can be introduced into the interlayer galleries to result in 3-dimensional structures featuring void spaces between the pillars [138). A similar strateg' can be employed to modify the V0P04-2H20 structure [41]. where the starting material is the oxide as shown below. M205 4» (H0)2X(0)R ' —) M0(04XR)-H20 (15) (M a- V or Nb) In the modified layer. V03 octahedra share axial oxygens to form puckered -V-02V-0- chains with alternating long and short V-O bonds as shown in Figure 52b [139]. Each chain is separated by phosphate groups whose three tripodal oxygens are shared with V03 octahedra and the apical oxygens bare R g‘oups. It is noteworthy that the R g-oup is directed to the gallery environment similar to Zr(RP03)2 structure (Figure 52a). Our initial attempt to synthesize modified layers were based on Johnson and J acobson’s methods [4 1]. We wished to demonstrate two important features: (1) that the ligating functionality can be introduced into phosphate layers and (ii) that these functionalities are accessible for reaction. For these reasons we chose a carboxylate which is a good ligand of bimetallic cores and an ideal organic functionality predisposed to reaction. Reaction 15 was performed with X a P and R a CH2CH2000H. The material is prepared by heating the mixture of V205 and H00CCH2CH2P0(0H)2 in benzyl alcohol in the presence of 0.2 molar equivalence of 1 M HCl aqueous solution. The specific synthetic conditions are described in reference 140. 189 Figure 52. Structure of modified vanadium phosphate. V0RP03 in its ac-plane and bc-plane. 190 8an 191 The X-ray difi'raction pattern of the modified layer. shown in Figure 53a. indicates door = 18.6 A (26 a 480°). The pattern also shows well resolved 2nd and 3rd order difi'raction peaks. By subtracting the thickness of the inorganic VOP03 layer ((1 =- ~6 A: d = 5.41 A in Zr(I-IP03)9) from the observed d-spacing of 18.6 A. we find a' thickness of organic layer in the modified voapoa (R . CH2CH2COOH) of ~12.5 A This observed spacing is consistent with the formation of an organic bilayer structure in which the propyl groups are in an all trans configuration (Figure 54a). This bilayer arrangement of alkyl chains has been reported in many other layers when guests are long chain amines and alcohols [33.141]. The void space between alkylchains is likely filled with solvent molecules [41]. The d-spacing indicates that carboxylic acid chains are not interdigtated. probably resulting from hydrogen bonding between acid groups from adjacent layers. Interestingly. when an HCl:phosphonate ratio of more than 0.3 is employed (higher ratios cause the V205 to be dissolved completely to gve a gem solution). the resulting compound bares a d-spacing of 13.0 A (20 a- 680°) with very well developed higher order peaks. The signal to noise ratio of the 13.0 A phase is much better than the 18.6 A phase thereby implying better crystallinity. Considering that the length of the propionic acid chain is ~6.8 A (142]. this interlayer spacing ag'ees nicely with a monolayer assembly of interdigtated alkylchains. figure 54 shows the schematic for the carboxylic acid modified layers prepared in the difi'erent acid regmes. This acid concentration dependence of the product (it < 0.3. bilayer: 0.3 < x < 0.5 monolayer: x > 6.5; no layers where x . [HCl] / [phosphonaten may be explained by hydrolysis of the layer structure. According to the Jacobson and 192 Figure 53. Powder X-ray patterns of V0(I-IOOCCH2CH9)P03 (a) forming a hydrogen bonded bilayer for acid concentration < 0.2; (b) interdigtated bilayer formed for acid concentration 0.3 < x < 0.5 (see text). 193 (a) IntenSilY to ."" . 29/d99'99 (b) lntenSW age-km“- . . . .--- ..fivrr‘w' 5 26 / degrec Figure63 194 Figure 53. Proposed structures of V0(I-IOOCCH2CH9)P03 forming (a) hydrogen bonded and (b) interdigtated bilayer. 195 3.56:— E. 3 196 Johnson's study. the area defined by a. b dimensions of the layer is too small for two alkyl chains. and can accommodate only one pendant group per unit. This space restriction forces pendant groups to arrange in a bilayer. In this bilayer assembly. the carboxylic acid pendant groups establish a hydrogen bonding network through terminal acid groups with solvent molecules that are associated in the void space between guest molecules. In acidic conditions. owing to the asymmetric arrangement of V-O bonds in the layer. the long V-0 bond is susceptible to hydrolysis [143]. The hydrolyzed layer possesses two O-H groups in the place of one oxygen. and the in plane cell dimensions are expanded. This increased area can accommodate two alkyl chains in one unit cell and the interdigtated monolayer arrangement is possible. We believe this is the case for the layer prepared from high acid concentration. In high acid concentrations (it > 0.5 eq). the long V-O bonds are totally disrupted and the layer structure can not by sustained. Infraredspectraoftwomodifiedlayersareverywell resolvedas shown in Figure 55. Absorption peaks attributable to 0:0 vibrational modes at ~1710 cm‘1 and of the alkyl chain of the pendant group are observed at ~3000 cm‘l . Interiayer aromatic goups from the benzyl alcohol. used as a solvent in the preparation. of the material. are also observed in the bilayer sample at 1450. 1455 and 746 cm'1 along with phosphate related peaks from the inorganic backbone. The monolayer structure does not show strong peaks for aromatic groups indicating less solvent within the layers. This observation is consistent with solvent occupying space of the bilayer assembly. The broad C=0 band at 1720 cm’1 in bilayer structure supports the existence of dimeric acid groups along with smaller intensities of free acid peaks in 1300 - 1400 cm"1 197 Figure 55. Infrared spectra of VOMOOCCHQCHflPO3 forming (a) hydrogen bonded and (b) interdigtated bilayer. 198 omfi om» owns omma .:.9=a;n mmmznzm>¢3 omoa ommm ommm omwm ommm 0m hr 3 ...: BONULLINSNUUL 199 spectral regon. The sharp peak at 1709 cm“1 observed in the monolyer structure is attributable to free acid groups. These alkyl acid modified layers are very stable toward intercalation reactions that utilize the acid goup as a ligand. An attempt to introduce solvated bimetallic cores to obtain the layer-analog of molecular M2(CH3C00)4 was not successful. Neither lengthening the reaction period nor raising the reaction temperature promoted the irntercalation of bimetallic cores into the layer. No molybdenum was detected in ESCA spectra and no color change was observed upon reaction. The organized assembly of hydrogen bonding between layers in the 18.6 A phase may be energetically stabilized with respect to disruption by the bimetallic cores. The 13.0 A phase may be even more stable toward intercalation because the van der Waals interactions of the interdigtated layers must be overcome before guest molecules can coordinate to the 0001-1 functionality in an all trans orientation. This exceptionally strong association of layers can be broken when the hydrogen bonding is disrupted. or when a reaction is carried out that involves a singular carboxylic group (the introduction of a bimetallic core requires two carboxylic acid groups from independent chains). Small and strong alkylating reagents readily react with the carboxylic acid modified layer. The reaction of V0(HOOCCH2CH2)P03 (18.6 A) and (CI~13CH2)30(BF4) produces a new layer with a d-spacing of 21. 1 A (26 :- 4.21°). X-ray difi'raction patterns of the acid layer and the ester layer are presented together in Figure 56 for clear comparison. The increment of 2.5 A in the d-spacing correlates well with the distance of interdigtated C113CH2 chains at the end of carboxylic acid groups as shown in Figure 57 . The X-ray diffraction pattern shows at least three 001 reflections and 200 Figure 56. Powder X-ray patterns of V0(HOOCCH2CH9)P03 (a) before and (b) after esterification with (CH3CH9)30(BF4J. 201 on can... 868 \ em «a 3 o o Misuslul 202 Figure 57. Possible structures of V0(CH3CH200CCH2CH9)P03 203 /°. '. /°. '. “/5 '. / - o . e /'..,/'./../ / ° /'° '= / °‘ / O “‘ o - [.../..-...” Figure57 204 the breadth of peaks indicate that crystallinity is preserved. This esterification reaction provides the first evidence that the interlayer carboxylic acid groups are available for reaction. Synthetic routes for a variety of phosphonate or arsonate compounds have to be realized for the continued development of modified phosphate layers. Molecular models and our experiments on the propionic acid modified layers suggest that organic functionalities protruding into the gallery should be available for M—"—M coordination. The commercial availability of several phosphonates coupled with Abruzov and Bart chemistry [144] should permit us to further tailor the layers with amines. pyrimidines. nitriles. isonitriles. phosphines. arsines. and carboxylic acids. Some potentially interesting target molecules in regard to multielectron M£M photochemistry are shown by structures 1- 4. I? 0H 0 HoX‘ ’ H0\£I,0H HO I? ,on ' O Ax— GAO" 3 ' 4 ”GE B. Modification of the u-E-u Ligation Spheres Adapted for Layered Phosphate incorporation Alkyl and aryl amines readily intercalate into Zr(HPO4)2-2H20 [42]. V0P04-2H20 and their related structures [43] to form bilayers. For the 205 former phosphate. the driving force for intercalation is the formation of alkylammonium ions whereas in the latter. strong intermolecular and ion-dipole interactions provide the driving force for intercalation of the amine. Thus an alternative strategy for introducing M—“—M cores into layered phosphates is the introduction of an amine or pyridine functionality directly onto the ligand of the M—‘-‘M core. The synthesis of Mo2(cytocine)4 was undertaken based on the synthetic methodologes associated with of Mo2(mhp)4 (mhp a 6-methyl- 2-hydroxypyridine) formation [145]. Deprotonated cytocine (5) was refluxed with (NI-I4)5M02Clg in dried methanol. The detailed synthetic procedure can be found in reference 146. The resulting brownish red precipitate displays absorption bands at 412 and 526 nm. shown in Figure 57 . which are energetically similar to the absorption bands of Mo2(mhp)4 (Mum (alM‘lcm’l) - 405 (13270): 495 (2450)) (145]. The absorption profile is characteristic of an M species with the lowest energy band assigned to 1 (82 -> 88‘). The emission properties of the compound are yet to be defined and the incorporation of the compound into ZrfliPOi)2-2H2O and VOPO4-2H20 remains to be investigated. Notwithstanding this compound is a promising M-n—M guest molecule for layered phosphates. Other ligands that lend themselves to this approach include the nuclear (5.6) and amino acids (7) presented below. NH2 NHZ | I N/ OH HO NA HO 0 206 Figure 56. Electronic absorption spectrum of 'M02(cytocine)4' in dried. deoxygenated CH3OH. N 1 ' ' at.“ m x 207 Absorbance 208 C. CONCLUSION The ability to modify layered phosphate structures with phosphonates and the possibility of syntlnesizing new bimetallic compounds coordinated by ligands containing intercalation-active groups unveil almost limitless possibilities in the design of LIPS structures containing photoactive binuclear metal cores. The extensive development of these approaches will present many interesting questions posed by these materials including: How are the excited state properties of the MPC system perturbed by the layers?: Can the MPC communicate with metals of the layers by may transfer? electron transfer?: Is photoactivity of core retained within the layers?: And if so. how does the mechanism compare to that observed for MPC species in homogeneous solution? In addition to the search for answers to these questions. new types of layered materials need to be investigated to introduce complementary redox centers to ensure ultimate photocatalytic activity. The results described herein gves the first glimpse to the design of a 'synthetic leaf. However. the difi'erence in our approach as compared to biolog' is noteworthy. Multielectron photoacitve centers are employed to avoid the intricate structural engneering needed in nature's consecutive one-electron transfer approach. Our initial attempts to engineer layered integated photochemical systems utilizing layered phosphates and bimetallic cores are encouragng. Functionalized phosphate layers and the M-iM compounds bearing modified ligation spheres opens new arenas for research. Extensive exploration of this multielectron photosynthetic assembly deserves attention not only for its possibilities in developing new photo catalysts but also for its importance 209 in defining multieleetron mechanism. 1. (a) Johnson. S. G.: Tang. D.: Tankowiak. R. J. Phys. Chem. 1990. 94. 5849. (b) Levi. B. G. Physics Today 1989. 42. 17. (6) Barber. J. Nature 1987. 325. 663. (d) Renger. G. In Biophysics: Hoppe. W.; Lohmann. W.; Markl. H: Ziegler. H. Eds.: Springer-Verlag: Berlin. Heidelberg. 1983: Chapter 13. p 515. (a) Christou. G. Acc. Chem. Res. 1989. 22. 328. (b) Wieghardt. K. Angew. Chem. Int. Ed. Engl. 1989. 28. 1153. (c) Pecoraro. V. L. Photoche. Photobiol. 1988. 49. 249. (d) Babcock. G. T. In New Comprehensive Biochemistry 15. Photosynthesis; Amasz. J. Ed.: Elsevier: Amsterdam. 1987. (e) Renger. G. Angew. Chem. Int. Ed. Engl. 1987. 26. 643. (i) Kok. B.: Forbush. B.: McGloin. M. Photochem. Photobiol. 1970. 11. 457. (a) Boxer. S. G.: Goldstein. R. A.; Lockhart. D. J.; Middendorf. T. R.: Takifi. L. J. Phys. Chem. 1989. 93. 8280. (b) van Grondelle. R. BiochimicaetBiophysicaActa 1985. 811. 147. Scharf. H.-D.: Fleischhauer. J .: Leismann. 11.: Ressler. 1.: Schleker. W.; Weitz. R. Angew. Chem. Int. Ed. Engl. 1979. 18. 652. Iiehn.J.M.InPhotochemicalCmuersionandStorageofSolar Energy. Connolly. J. S. Ed.: Academic Press: New York. 1981. Chapter 6. Zameraev. K. 1.: Paemon. V. N. In Energy Resources TTuuugh Photochemistry and Catalysis. Gratzel. M. Ed.: Academic Press: New York. 1983. Chapter 5. Kalyanasundaram. K.: Gratzel. M. Coord. Chem. Rev. 1986. 69. 57. Gratzel. M.: Kalyanasundaram. K.: Kiwi. J. Strucut. Bonding. 1982. 49. 37. Netzel. T. L.: Kroger, P.: Chang. C. K.: Fujita. 1.; Faier. J. Chem. Phys. Lett. 1979. 67. 223. Netzel. T. L.: Bergkamp. M. A.: Chang. 0. K. J. Am. Chem Soc. 1982. 104. 1952. 210 9. 10. 11. 12. 13. 14. 211 (a) Osuka. A.: Maruyama. K.: Mataga. N. Asahi. T.: Yamazaki. N. J. Am Chem. Soc. 1990.112. 4958. M Schmidt. J. A.: McIntosh. A. R.: Weedon. A. C.: Bolton. J. R.: Connolly. J. S.: Hurley. J. K.: Wesielewski. M. R. J. Am. Chem Soc. 1988. 110. 1733. (c) Wasielewski. M. R.: Niemczyk. M. P. ACS Sym Ser. 1986. 321. 154. (d) Leland. B. A.: Jordan. A. D.: Felker..P. M.: Hopfield. J. J.: Zwail. A.: Dervan. P. B. J. Phys. Chem 1985. 89, 5571. (e) Siemiarczuk. A.: McIntosh. A. R.: Ho. T.-F.: Stillman. M. J.: Roach. K. J.: Weedon. A. C.: Bolton. J. R.: Connolly. J. S. J. Am. Chem Soc. 1983. 105. 7224. (a) Hasharoni. K.; Levanon. 11.: Tang. J.: Bowman. M. K.; Norice. J. R.: Gust. D.: Moore. T. A.: Moore. A. L. J. Am Chem Soc. 1990. 112. 6477. (b) Gust. D.: Moore. T. A.; Moore. A. L. et. al. Science 1990. 248. 199. (c) Gust. D.: Moore. T. A.: Moore. A. L.: Makings. L. R.: Seely. G. R.: Ma. X.: Trier. T. T.: Gao. F. J. Am. Chem Soc. 1988. 110. 7567. (d) Gust. D.: Moore. T. A.: Moore. A. L.: Barrett. D.: Makings. L. R.: Liddell. P. A.: De Schryver. F. C.: Van der Airweraer. M.: Benssasson. R. V.: Rougee. M. J. Am. Chem Soc. 1988. 110. 321. (e) Gust. D.: Moore. T. A. In Suprarmlecular Photochemistry; Balzani. V. Ed.: Reidel: Boston. 1987. (a) Wasielewski. M. R.: Minsek. D. W.; Niemczyk. M. P.: Svec. W. A.: Yang. N. C. J. Am. Chem. Soc. 1990. 112. 2823. Wasielewski. M. R. In Photoinduced Mun Wen Fox. M. A.. Chanon. M. Eds: Elsevier: Amsterdam. 1988: Part A. p 161. Wasielewski. M. R.: Niemeeyk. M. P.: Svec. W. A.: Pewltt. E. B. J. Am. Chem. Soc. 1985. 107. 5562. (a) Yamazaki. 1.: Tamai. N.: Yamazaki. T. J. Phys. Chem 1990. 94. 516. (b) Thomas. J. K. Chem Rev. 1980. 80. 283. Krishnan. M.: White. J. M.: Fox. M. A.: Bard. A. J. J. Am Chem Soc. 1983. 105.7002. Fox. M. A.: Cardona. R.: Gaillard. E. J. Am. Chem Soc. 1987. 109. 6347. Sobczynski. A.; Bard. A. J.: Campion. A.: Fox. M. A.: 15. 16. 17. 18. 19. 20. 21. 212 Mallouk. T.; Webber. S. E.: WhitC. J. M. J. Phys. Chem. 1987. 91. 3316. FOX. M. A. New J. Chem 1987. 11. 129. Pettlt. T. L.: FOX. M. A. J. Phys. Chem. 1986. 90. 1354. Duonghong. D.: Serpone. N.: Gratzel. M. Helv. Chim Acta 1984. 67. 1012. Houlding. V. 11.: Gratzel. M. J. Am Chem Soc. 1983. 105. 5695. Humphry-Baker. R.: IJlie. J .: Gratzel. M. J. Am Chem Soc. 1982. 104. 442. Pichat. P. New J. Chem. 1987. 11. 135. Nijs. 11.: Fripiat. J. J.: Damme. H. V. J. Phys. Chem 1983. 87. 1279. Persaud. L.: Bard. A. J.: Campion. A.: Fox. M. A.: Mallouk. T. E.: Webber. S. E.: White. J. M. J. Am Chem Soc. 1987. 109. 7309. (a) Collman. J. P.: Hampton. P. D.: Brauman. J. I. J. Am Chem Soc. 1990. 112. 2977. 2986. (b) Panicucci. R.: Bruice. T. J. Am Chem. Soc. 1990. 112. 6063. 6072. (c) Menage. S.: Brennan. B. A.; Juarez-Garcia. C.: Munch. E.; Que. L. Jr. J. Am Chem. Soc. 1990. 112. 6423. (d) Tyeklar. 2.: Karlin. K. D. Acc. Chem Res. 1989. 22. 241. (e) Lippard. S. J. Angew. Chem Int. Ed. Engl. 1988. 27. 344. (f) Groves. J. T. Ann. N. Y. Acad. Sci. 1986. 4. 99. (a) Lancaster. J. R. Jr.. Ed. The Biotnovyanic Chemistry of Nickel: VCH: New York. 1988. (b) one. L. Jr.. Ed. Metal Clusters in Protains: American Chemical Society: Washington. DC. 1988: Series 372. (c) Sykes. A. 0.. Ed. Advances in Inorganic and Bioiriayanic Mechanisms: Academic Press: London. 1983. Partiganoni. C. P.: Turro. C.: Shin. Y.-g. K.; Motry. D. 11.: Kadis. J.: Dulebohn. J. 1.: Nocera. D. G. NATO ASI series. Reidel: Amsterdam: 1990. p 0000 Dulebohn. J. 1.: Ward. D. L.: Nocera. D. G. J. Am Chem. Soc. 1990. 112. 2969. Dulebohn. J. 1.: Ward. D. L.: Nocera. D. G. J. Am Chem Soc. 1988. 110. 4054. 22. 23. 24. 25. 26. 27. 29. 31. 213 Shin. Y.-g. K; Miskowski. V. M.: Nocera. D. G. 1mm. Chem 1990. 29. 2308. Hopkins. M. D.: Gray. H. B.: Miskowski. V. M. Polyhedron 1987. 6. 705. Manning. M. C.: Trogler. W. C. J. Am Chem Soc. 1983. 105. 5311 and references sited therein. Hall. M. B. Polyhedron 1987. 6. 679. Bernard. M.: Wiest. R. Chem Phys. Lett. 1985. 122. 447. Hay. P. J. J. Am Chem Soc. 1982. 104. 7007. Zhang. X.: Kosik. M.: Sutin. N.: Winkler. J. R. ACS Symposium Series. 1990. p. 0000. (a) Cotton. F. A. In Understanding Molecular Properties: J. Avery: Dahl. J. P.: Hansen. A. E. Eds. Reidel: Boston. 1987: p. 17. Cotton. F. A. Polyhedron. 1987. 6. 667. (b) Shaik. S.: Hoffmann. R.: Fisel..C. R.: Summerville. R. H. J. Am Chem Soc. 1980. 102. 4555. (c) Summerville. R. H.: Hofl'mann. R. J. Am Chem Soc. 1979. 101. 3821. (d) Trogler. W. C. Inorg. Chem 1980. 19. 697. (e) Ziegler. R. J.: Risen. W. M. Jr. Inay. Chem 1972. 1 1. 2796. Cotton. F. A.: Walton. R. A. Multiple Bonds between Metal Atoms: Wiley-Interscience: New York. 1982: p 221. (a) Moynihan. K. J.: Gao. X.: Boorman. P. M.: Fait. J. F.: Freeman. G. K. W.; Thornton. P.: Ironmonger. D. J. Inorg. Chem 1990. 29. 1648. (b) Bott. S. G.: Clark. D. L.: Green. M. L. H.: Mountford. P. J. Chem. Soc. Chem Commum. 1989. 418. (c) Cotton. F. A.: Luck. R. L. lnovy. Chem 1989. 28. 182. (d) Cotton. F. A.: Poli. R. Inorg. Chem. 1987. 26. 3310. (e) Bergs. D. J.: Chisholm. M. H.: Folting. K.: Hufi'man. J. C.:-Stahl. K. A. Inorg. Chem 1988. 27. 2950. (0 Nocera. D. G.: Gray. H. B. Inorg. Chem 1984. 23. 3686. (a) Cotton. F. A.: Eglin. J. L.: Luck. R. L.: Son. K. Inorg. Chem 1990. 29. 1802. (b) Mui. H. D.: Poli. R. Inorg. Chem 1989. 28. 32. 35. 37. 39. 41. 42. 214 3609. (c) Chacon. S. T.: Chisholm. M. H.: Streib, W. E.; Van der Sluys. W. Inog. Chem 1989. 28. 5. (d) Schrock. R. R: Sturgeofi'. L. G.: Sharp. P. R. Inorg. Chem 1983. 22. 2801. (e) Jackson. R. B.: Streib, W. E. Inorg. Chem 1971. 10. 1760. Partiganoni. C. P.: Nocera. D. G. Inory. Chem 1990. 29. 2033. Whittingham. M. S.: Jacobson. A. J. Intercalation Chemistry: Academic Press. New York. 1982. Newsham, M. D.: Giannelis. E. P.: Pinnavaia. T. J.: Nocera. D. G. J. Am Chem Soc. 1988. 110. 3885. Clearfield. A. In Inorganic Ion-Exchange Materials: Clearfield. A. Ed.: CRC Press: Boca Ranton. FL. 1982. (a) Iadwig. V. G. Z. Anogr. Ally. Chem 1965. 338. 266. (b) Jander. V. G.: Jahr. K. F.: Witzmann. H. Z. Anogr. Allg. Chem 1934. 65. 271. Chang. I.-J.: Nocera. D. G. J. Am Chem Soc. 1987. 109. 4901. Partiganoni. C. P.: Chang. I-J.; Nocera. D. G. Coord. Chem Rev. 1990. 97. 105. Chang. l-J.: Nocera. D. G. Inorg. Chem 1989. 28. 4309. . Dines. M.: DiGiacomo. P. M. Inorg. Chem 1981. 20. 92. Johnson. J. W.; Jacobson. A. J.: Butler. W. M.: Rosenthal. S. E.: Brody. J. F.; Lewandowski. J. T. J. Am Chem Soc. 1989. 111. 381. (a) Clearfield. A.: Twinda. R. M. J. Inorg. Nucl. Chem 1979. 41. 871. (b) Behrendt. D.: Beneke. K.: Lagaly. G. Angew. Chem. Int. Ed. Engl. 1976. 15. 544. (c) Alberti. G.: Galli. P. J. Inorg. Nucl. Chem 1966. 28. 658. Beneke. K.: Lagaly. G. Inorg. Chem 1983. 22. 1503. 45. 47. 49. 51. 52. 215 (a) Miskowski. V. M.: Schaefer. W. P.: Sadeghi. B.: Santarsiero. B. D.: Gray. H. B. Inorg. Clem 1984. 23. 1154. (b) Miskowski. V. M.: Smith. T. P.: Loehr. T. M.: Gray. H. B. J. Am Chem Soc. 1985. 107. 7925. Albright. T. A.; Burdett. J. K: Whangbo. M.-H. Orbital Interactions in Chemistry: Wiley-lnterscience: New York. 1985: Chapter 17. Levenson. R. A.: Gray. H. B. J. Am Clem Soc. 1976. 97. 6042. Harvey. P. D.: Butler. 1. S.: Barreto. M. C.: Coville. N. J.: Harris. G. W. Inay. Clem 1986. 27. 639. Meyer. T. J.: Casper. J. V. Chem Rev. 1966. 85. 187. Geoffroy. G. L.: Wrighton. M. S. Organometallic Photochemistry: Academic Press: New York. 1979. Wrighton. M. S.: Grafl'. J. L.: Luong. J. C.: Reichel. C. L.: Robbins. J. L. In Reactivity QIMetal-Metal Bonds: Chisholm. M. H. Ed.: ACS Symposium Series 155: American Chemical Society: Washington. DC. 1981: p 85. (a) Rushman. P.: Brown. T. L. J. Am Clem Soc. 1987. 109. 3632. (b) Lee. I'L-W.: Brown. T. L. J. Am Chem Soc. 1987. 109. 3269. (c) Lee. K-W.: Brown. T. L. Inory. Chem 1967. 26. 1852. (a) Vaida. V. In High-Energy Proxsses in Organonetallic Chemistry: Suslick. K. S. Ed.: ACS Symposium Series 333: American Chemical Society: Washington. DC. 1987: p 70. (b) Hollingsworth. W. E.: Vaida. V. J. Phys. Chem 1966. 90. 1235. Wuu. Y.-M.: Zou. C.: Wrighton. M. S. J. Am Chem Soc. 1987. 109. 5861. (a) Metcalf. P. A.: Kubiak. C. P. J. Am Chem Soc. 1986. 108. 4682. (b) Reinking. M. K: Kullberg. M. L.: Cutler. A.R.: Kubiak. C. P. J. Am Clem Soc. 1986. 107. 3517. 55. 57. 59. 61. 62. 65. 67. 216 Isobe. K.: Kimura. S.: Nakamura. Y. J. Organomet. Chem 1987. 331. 221. Yasui‘uku. K.: Hiraga. N.: Ichimura. K.: Kobayashi. T. In Photochemiso'y and Photophysics of Coordination Compounds: Yersin. H.: Vogler. A. Ed.: Springer-Verlag: New York. 1987: p. 271. Velllard. A: Dedleu. A. Noun. J. Chlm. 1983. 7. 683. (a) Herrick. R. S.: Brown. T. L. Inorg. Chem 1984. 23. 4550. (b) Walker. H. W.; Herrick. R. S.: Olsen. R. J. Brown. T. L. Inorg. Clem 1984. 23. 3748. (c) Wegman. R. W.; Brown. T. L. Inovy. Clem 1983. 22. 183. Bohling. D. A.: Gill. T. P.: Mann. K. R. Inorg. Chem 1981. 20. 194. Levenson. R. A.: Gray. H. B.; Ceasar. G. P. J. Am Clem Soc. 1970. 92. 3653. (a) Prinslow. D. A.: Vaida. V. J. Am Clem Soc. 1987. 109. 5097. (b) Leopold. D. G.: Vaida. V. J. Am Chem Soc. 1984. 106. 3720. Seder. T. A.: Church. S. P.: Weitz. E. J. Am Chem Soc. 1986. 108. 7518. (a) Kobayashl. T.: Ohtani. H.: Nada. H.: Teratani. S.: Yamazaki. H.: Yasufuku. K. Organometallics 1986. 5. 110. (b) Kobayashi. T.. Yasufuku. K: Iwai. J.: Yesaka. H.: Noda. H.: Ohtani. H. Coord. Clem Rev. 1986. 64. 1. (c) Yesaka. H.: Kobayashi. T.: Yasufuku. K.: Nagakura. S. J. Am. Clem Soc. 1983. 105. 6249. medman. A.: Bersohn. R. J. Am Chem Soc. 1978. 100. 4116. Church. S. P.: Hermann. H.: Grevels. F.-W.: Schafl'ner. K. J. Chem Soc. Chem Comrmm. 1984. 785. Coville. N. J.: Stolzenberg. A. M.: Muetterties. E. L. J. Am Chem Soc. 1988. 105. 2499. Wrighton. M. S.: Ginley. D. S. J. Am Clem Soc. 1975. 97. 2065. 69. 70. 71. 72. 73. 74. 75. 76. 77. 217 Rothberg. L. J.: Cooper. N. J.: Peters. K. S.: Vaida. V. J. Am Chem Soc. 1982. 104. 3536. Stiegman. A. E.: Miskowski. V. M.: Gray. H. B. J. Am Chem Soc. 1986. 108. 2781. Stiegman. A. E.: Miskowski. V. M. J. Am Clem Soc. 1988. 110. 4053. (a) Cotton. F. A.; Falvello. L. R.; Han. S. Inmy. Chem 1982. 21. 1709. (b) Bancroft. D. P.: Cotton. F. A.; Falvello. L. R.: Han. S.: Schwotzer. W. Inovy. Chim Acta. 1984. 87. 147. El-Mehdawi. R.: Bryan. S. A.: Roundhill. D. M. J. Am Chem Soc. 1985. 107. 6282. Muraveiskaya. G. S.: Abashkin. V. E.: Evstai‘eva. O. N.: Golovaneva. I. P.: Shchelokov. R. N. Sov. J. Coord. Clem Engl. Trans. 1981. 6. 218. When the clover-shaped crystals are dissolved in H20 with excess KBr. absorption spectrum displays peaks on 368 nm and 314 nm with a shoulder on 302 nm of nearly equal intensity. Additionally. maximum of the emission spectrum in phosphoric acid glass shifts to higher energy. 622 nm. with much shorter lifetime. 3.7 usec. This impurity is a likely candidate that caused the biphasic decay of the emission from K4Ptz(HPO4)4Br2. The crystallographic study of the clover-shaped crystals are not pursued. due to the severely inter-twined growth. Alexander. K. A.; Bryan. S. A.; Dickson. M. K.; Hedden. D.: Roundhill. D. M. Inay. Synth. 1986. 24. 211. Che. C.-M.: Schaefer. W. P.: Gray. H. B.: Dickson. M. K.: Stein. P. B.: Roundflll. D. M. J. Am Clem Soc. 1982. 104. 4253. Shin. Y.-g. K.: “Dulebohn. J. 1.: Kadis. J.: Nocera. D. G. Inog. Clemsubmitted for publication. 78. 79. 81. 82. 85. 87. 89. 91. 218 Mussell. R. D.: Nocera. D. G. J. Am Clem Soc. 1988. 110. 2764. Wrighton. M. S.: Ginley. D. S.: Morse. D. L. J. Phys. Chem 1974. 78. 2229. Dye. J. L.: Nicely. v. A. J. Clem Ed. 1971. 48. 443. Lever. A. B. P. Inorganic Electronic Spectroscopy: Elsevier: New York. 1984: Chapter 7. Miskowski. V. M.: Gray. H. B. In Understanding Molecular Properties: J. Avery; Dahl. J. P.: Hansen. A. E. Eds. Reidel: Boston. 1987: Part I. p 1. Bryan. S. A.; Schmehl. R. H.: Roundhill. D. M. J. Am Chem Soc. 1986. 108. 5408. Roundhill. D. M.: Gray. H. B.: Che. C.-M. Ace. Che. Res. 1989. 22. 55. Che. C.-M.: Butler. L. G.: Grunthaner. P. J.: Gray. H. B. Inory. Clem1985. 24. 4662. lsci. H.: Mason. W. R. Inay. Clem 1986. 24. 1761. Che. C.-M.: Lee. W.-M.; Mak. T. C. W.; Gray. H. B. J. Am Clem Soc. 1986. 108. 4446. . Che. C.-M.: Mak. T. C. W.; Miskowski. V. M.: Gray. H. B. J. Am Clem Soc. 1986. 108. 7840. ' (a) Conder. H. L.: Cotton. F. A.: Falvello. L. R.: Han. S.: Walton. R. A. Inorg. Clem 1988. 22. 1887. (b) Cotton. F. A.: Han. S.: Conder. H. L.: Walton. R. A. Inorg. Chim Acta. 1988. 72. 191. El-Mehdawi. R.: Frontzek. F. R.: Roundhill. D. M.: Inay. Clem 1986. 25. 3714. El:Mehdawi. R.: Fronczek. F. R.: Roundhill. D. M. Inorg. Chem 1986. 25. 1155. 93. 95. 97. 100. 101. 102. 219 Che. C.-M.; Schaefer. W. P.: Gray. H. B.; Dickson. M. K.; Stein. P. B.: Roundhill. D. M. J. Am Chem Soc. 1982. 104. 4253. Woollins. J. D.: Kelly. P. F. Coord. Clem Rev. 1986. 65. 115. (a) Hedden. D.: Roundhill. D. M.: Walkinshaw. M. D. Inorg. Chem 1985. 24. 3146. (b) Alexander. K. A.; Bryan. S. A.; Frontzek. F. R.; Fultz. W. C.: Rheingold. A. L.: Roundhill. D. M.: Stein. P.: Watkins. S. F. Inorg. Chem 1988. 24. 2803. (a) Norman. J. G.: Renzoni. G. E.: Case. D. A. J. Am Clem Soc. 1979. 101. 5256. (b) Norman. J. G.: Kolari. H. J. J. Am Chem Soc. 1978. 100. 791. Winkler. J. R.; Gray. H. B. Inory. Clem 1985. 24. 346. The three drt‘ -1 do‘ assignments. AguPEu) (— lAlg. EuPEu) (— 1A18. and EuPEu) «- lAlg. were fit to the spin-orbit coupling matrix to yield zero-order energies of 201 17 and 24499 cm"1 for the 3E and 1E states. respectively. and reasonable [96) values for the spin-orbit coupling parameter 1. a 2655 cm’1 was obtained. The energies of the spin-orbit manifold are as follows: e(B I“) 2 e182“) . 18850 cm": 8(Eu(3Eu)) . 19802 cm'l: em...) = 21.42..) = 21504 cm“: emuileun . 24875 cm“. Turro. N. J. Modern Molecular Photochemistry: Benjamin] Cumming: Menlo Park. CA: 1978. Hopkins. M. D.: Miskowski. V. M.: Gray. H. B. J. Am Chem Soc. 1986. 108. 6908. Adamson. A. W.; Fleischauser. P. D. Eds. Concept of Inorganic Photoclemistry: John Wiley and Sons. Inc.: New York. 1975. Kadis. J.: Ward. D. L.: Nocera. D. G. unpublished results. Abrahamson. H. B.; Frazier. C. C.: Ginley. D. S.: Gray. H. B.; IJlienthal. J.: Tyler. D. R.; Wrighton. M. S. Inorg. Clem 1977. 16. 1554. 103. 104. 105. 106. 107. 108. 109. l 10. 111. l 12. 220 Castellani. M. P.: Weakley. T. J. R.: Tyler. D. R.: Miskowski. V. M.: Stiegman. A. E. J. Phys. Clem 1990. 94. 6599. Bordes. E.: Courtine. P. J. Catal. 1979. 57. 236. Bordes. E.: Courtine. P. J. Clem Soc. Clem Commun. 1985. 294. (a) Chahboun, H.: Groult. D.: Hervieu. M.: Raveau. B. J. Solid State Chem 1986. 65. 331. (b) Stranford. G. T.: Condrate. R. A. Sr. J. Solid State Clem 1984. 52. 248. Garcia-Ponce. A. L.: Moreno-Real. L.: Jimenez-Lopez. A. Inorg. Clem 1988. 27. 3372. Dulebohn. J. 1.: Ward. D. L.: Nocera. D. G. Polyhedron submitted for publication. (a) Bruque. S.: Martinez-Lara. M.: Moreno-Real. L.: Jimenez-Lopez. A.: Casal. B.; Ruiz-Hitzky. E. Inory. Clem 1987. 26. 847. (b) Tacha. M.: Theobald. P.: Bernard. J.: Hewat. A. W. Rev. Chim Miter. 1982. 19. 291. (c) Jordan. 8.: Calve. C. Can. J. Chem 1973. 51. 2621. R'kha. C.: Vandenborre. M. T.: IJvage. J.: Prost. R.: Huard. E. J. Solid State Clem 1986. 63. 202. Yakovleva. T. N.: Tarasova. D. V.: Mikheikin. I. D.: Timoshenko. V. I. J. Inorg. Chem 1988. 33. 134. Jacobson. A. J.:Johnson. J. W.; Brody. J. F.: Scanlon. J. C.: Lewandowski. J. T.: Inorg. Chem 1986. 24. 1782. Casan. N.: Amoros. P.: Ibanez. R.; Tamayo. E. M.: Beltran-Porter. A.: Beltran- Porter. D. J. IrelusionPlenom 1988. 6. 195. Cotton. F. A.: DeBoer. B. G.: LaPrade. M. D.: Pipal. J. R.: Ucko. D. A. J. Am Clem Soc. 1070. 90. 2926. Dunbar. K. R.: Pence. L. E. Inory. Synth. 29. in press. (b) Dunbar. K. R. J.Am Chem Soc. 1988. 110. 8247. 1 13. 114. 115. 1 16. 117. l 18. 1 19. 120. 121. 122. 123. 124. 125. 221 (a) Lawton. D.: Mason. R. J. Am Clem Soc. 1965. 82. 921. (b) Stephenson. T. A.; Bannister. E.: Wilkinson. G. J. Chem Soc. 1964. 4150 Cotton. F. A.; Wiesinger. K. J. Inorg. Clem 1991. 30. 871. (a) Belford. R. L.: Nilges. M. J. ”Computer Simulation of Powder Spectra" EPR Symposium. 2 1st Rocky Mountain Conference. Denver. CO. 1979. Cotton. F. A.: Wilkinson. G. Advanced Inorganic Chemistry; Wiley: New York. 1980. (a) Aquino. M. A. S.: Macarlney. D. H. Inorg. Clem 1987. 26, 2696. (b) Chavan. M. Y.: Lin. X. 0.: Ahsan. M. 0.: Bernal. 1.; Beer. J. L.: Kadish. K. M. Inorg. Chem 1986. 25.1281. Hopkins. M. D.: Miskowski. V. M.: Gray. H. B. J Am Clem Soc. 1986. 108. 959. Swingle. R. S. Jr.: Riggs. W. M. In Critical Reviews in Analytical Clemistry; CRC Press: 1975. Dunbar. K. R.; Haefner. S. C.: Pence. L. E. J. Am Clem Soc. 1989. 111. 5504. Johnson. J. W.; Jacobson. A. J.: Brody. J. P.: Rich. S. M.: Inorg. Chem 1982. 21. 3820. Bordes. E.: Courtine. P.: Pannetier. G. Ann. Chim 1978. 8. 105. Chernoeukov. N. G.: Egorov. N. P.: Mochalova. I. R. Russ. J. Inorg. Clem 1978. 23. 1627. Antonio. M. R: Barbour. R. 1.: Blum. p. a; Inorg. Clem 1987. 26. 1235. Tietze. H. R. Aust. J. Clem 1981. 34. 2035. (a) Barklie. R. C.: Girard. 0.; Braddel. O. J. Phys. Chem 1988. 92. 1371. (b) Jain. V. K. J. Magn. Reson. 1986. 64. 512. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 222 Tachee. M.: Theobaald. F. J. Solid State Chem 1981. 40. 280. Feher, G.: Kip. A. F. Phys. Rev. 1955. 98. 337. Feher, G.: Kip. A. F. Phys. Rev. 1954. 95. 1343. Wertz. J. E.: Bolton. J. R. Electron Spin Resonance: Elerrentary Tleory and PracticalApplicaitons: McGraw-Hill. New York. 1972. Bencini. A.: Gattechi. D. EPR of Exchange Coupled Systems: Springer-VerIag: New York. 1990. (a) Sanches. C.: Vivien. D. IJvage. J. Sala-Pala. J.: Viard. B. Guerchais. J. E. J. Chem Soc. Dalton 1981. 64. (b) Viard. B.: Sala-Pala. J.: Amadurt. J. Guerchais. J. E. Inorg. Chim Acta 1980. 39. 99. (c) Eson. l. H.: Kochi. J. K.: Klabunde. U.: Manzer. L. E.: Parshall. G. W.; Tebbe. F. N. J. Am. Chem. Soc. 1974. 96. 7374. (a) Kim. D. H.: Britt. R. D.: Klein. M. P.: Sauer. K. J. Am Chem Soc. 1990. 112. 9389. (b) McMahon. D. 11.; Silbee. R. H. Phys. Rev. 1964. 135. A91. Hubert-Pfalzgraf. L. G.: Postel. M.: Riess. J. G. In Comprehensive Coordination Clemistry: Wilkinson. G.: Gillard. R. D.: McCleverty. J. A. Ed.: Pergamon Press: Chap 34. Oxford. 1987. Beltran-Porter. D.: Amoros. P.: Ibanez. R. Martinez. E.: Beltran- Proter. A. Solid State Ionics 1989. 32/33. 57. (a) Lines. M. E. J. Phys. Clem Solids 1970. 31. 101. (b) Rusbrooke. G. S.: Wood. P. J. MoL Phys. 1. 257. 1958. Villeneuve. G.: Lemma. L. Mol. Crys. Liq. Crys. 1989. 179. 495. Clement. R.: Lomas. L: Audiere. J. P. Clem Mater. 1990. 2. 641. (a) Burwell. D. A.: Thompmn. M. E. Clem Mater. 1991. 3. 14. (b) Colon. J. L.: Yang. C.-Y.: Clearfleld. A.: Martin. C. R. J. Phys. Clem 1988. 92. 5777. (b) Yamanaka. S.: Yamasaka. K.: Hattori. 138. 139. 140. 141. 142. 143. 144. 223 M. J. Inclu. Plenom 1984. 2. 297. (d) Alberti. G.: Costantino. U.: Alluli. S.: Tomassini. N. J. Inorg. Nucl. Chem 1978. 40. l 1 13 Ferragina. C.: Massucci. M. A.: Patrono. P. J. Chem Soc. Daton 1111s. 1986. 265. Huan. G.:Jacobson. A. J.:Johnson. J. W.; Corcoran. E. W. Jr. Clem Mai. 1990. 2. 91. Excess amount of alkylphosphonate (HO2CCH2CH2P0(0H)2. 0.6 g. 4.54 mmol) and V205 (0.3 g. 1.97 mmol) were suspended in benzyl alcohol (30 ml) with a catalytic amount of hydrochloric acid (1 ml of 1 M aqueous solution. 0. 1 mmol). While heating the mixture in an oil bath at ~85°C. a yellow brown V205 suspension changed to the fine greenish blue powder suspended in a geen solution. The product was filtered and washed with several portions of ethanol and ether. and air-dried. see for example (a) Alagna. L.: Tomlinson. A. A. G.: Rodriguez- Castellon. E.: Olivera. P. P.: Bruque. S. J. Clem Soc. Dalton Trans. 1990. 1183. (b) Choy. J.-H.: Kim. Y.-g.: Weiss. A. Mat. Res. Bull. 1985. 20. 1401. (e) Ferragina. C.: Massucci. M.: La Ginestra. A.: Patrono. P.: Tomlinson. A. A. G. J. Chem Soc. Chem Commun. 1984. 1024. (d) Costantino. U. J. Chem Soc. Dalton. 1979. 402. The length of a propionic acid group was estimated as l = 1 .26 n + 1.70 A (1.26 A a lengh of C-C bond. 1.70 A a estimated van der Waals radious of OH group in the acid: taken from van der Waals radius of CH3 goup) where n is the number of C-C and C—0 bonds. Johnson. J. W.; Jacobson. A. J.: Brody. J. F.: Lewandowski. J. T. Inorg. Chem 1984. 23. 3842. (a) Hudson. R. F. Structure and Mechanism in Organophosphorous Clemistry: Academic: New York. 1965. (b) Yale. H. Heterocyclic Compounds: lnterscience: New York. 1964: Part 4: 439. 224 145. (a) Cotton. F. A.: Ilsley. W. H.: Kaim. W. H. Inorg. Chem 1980. 19. 1453. (b) Bino. A.: Cotton. F. A. Inorg. Chem 1979. 18. 1381. 146. The mixture of cytocine (0.1 g) and CH30Na (0.02 g) was placed in a flask equipped with a side arm containing (NI-I4)5M02019 and a reflux condenser on a high-vacuum line and evacuated until the pressure was less than 10’5 torr. When the (n-Bu)4N(BF4) was used as the deprotonating reagent. CH2C12 solution was used and transferred into the reaction flask by a syringe. The methanol (~20 meastransferredthroughthelineandwasdeairatedbyatleast 3 cycles of freeze-pump-thaw. After the reaction vessel was filled with Ar. (NH4)5M02019 was added to the reaction mixture and resulted in reddish-brown solution. The color of solution changes nearly clear and brown-red solid was formed while 3 hrs of refluzdng. The product was filtered through a vacuum frit. and washed with three 10 ml portions of dried. deoxygenated metanol and vacuum dried on the int. The whole reaction was performed on the vacuum line. Dried solid product was kept in a Ar filled dry-box. since it decomposed within 30 minutes in air and even faster in moist air.